This article examines the #Articles_of_Agreement of the International Bank for Reconstruction and Development, the founding legal charter drafted at the #Bretton_Woods conference in July 1944 and brought into force in December 1945. The document created the #IBRD, the first institution of what would later grow into the #World_Bank_Group, and it fixed the purposes, capital structure, governance rules, and operating limits that still shape the institution today. The study traces the economic and political conditions that made a permanent international lender seem necessary after the Great Depression and the Second World War, reconstructs how the charter was negotiated between the British and American delegations, and reads the treaty article by article. Special attention is given to three features that continue to define the institution: the #weighted_voting system that ties influence to capital contributions, the reliance on callable capital and bond market borrowing rather than taxpayer transfers, and the #apolitical_mandate contained in Article IV, Section 10. The article then follows the widening of the charter's meaning from post-war reconstruction toward long-term #development, considers the main lines of criticism and defence, and closes with the reform debates that surrounded the institution's eightieth anniversary. The argument throughout is that the 1944 charter was a carefully engineered compromise whose strengths and its most criticised limits both flow from the same original design choices. Keywords: International Bank for Reconstruction and Development; Bretton Woods; Articles of Agreement; World Bank Group; international financial institutions; development finance; global economic governance; weighted voting 1. Introduction Few legal documents have shaped the modern global economy as quietly and as durably as the #Articles_of_Agreement of the #IBRD. Signed as part of the Final Act of the United Nations Monetary and Financial Conference in 1944, this charter created a permanent international lender that has since financed roads, dams, power stations, schools, health systems, and policy reforms across most of the world. Yet the charter itself is rarely read by the students, borrowers, and citizens whose lives it touches. It is treated as background, a settled fact of international life, rather than as a negotiated text with a history and an argument inside it. This article takes the charter seriously as a legal and political artefact. The aim is not to celebrate or to condemn the #World_Bank, but to explain what the founders wrote, why they wrote it that way, and how the words on the page have carried a heavy institutional load for eight decades. The 1944 document did several things at once. It defined the purposes of the new institution. It set the rules for membership and for how much money each country would contribute. It designed a governance system in which votes are not equal but weighted by financial stake. It placed strict limits on what the institution could do, including a rule that it must ignore the political character of its members. And it built a financing method that let the institution raise money from private investors rather than depend on the yearly goodwill of national parliaments. Each of these choices had consequences that the drafters could only partly foresee. The purposes clause, written mainly with a ruined Europe in mind, proved elastic enough to justify half a century of lending to the #Global_South. The weighted voting rule gave the United States a controlling voice that it has never surrendered, and this remains the sharpest point of dispute in reform debates. The financing model turned the institution into a highly rated borrower on Wall Street and in London, which made it powerful and self-sustaining but also cautious. The political prohibition freed the institution to work with governments of every kind, yet it also became a shield behind which uncomfortable questions about human rights and corruption could be deferred. The discussion proceeds in eleven further sections. Section two sketches the economic disorder of the interwar years that motivated the founders. Section three describes the conference and the drafting process. Section four explains how the bank was tied to its sister institution, the currency fund. Section five walks through the structure of the charter. Sections six, seven, eight, and nine examine, in turn, the purposes, the capital and financial model, the governance and voting design, and the #apolitical_mandate. Section ten follows the institution's growth into a group of five bodies and the shift from reconstruction toward #development. Section eleven weighs the main criticisms and defences. Section twelve turns to the reform debates of recent years. A short conclusion draws the threads together. 2. The World Before Bretton Woods: Why a Permanent Bank Was Needed To understand the charter one has to understand the fear that produced it. The generation that met in New Hampshire in 1944 had lived through two disasters that they blamed, in large part, on the absence of any orderly system of international money and credit. The first was the collapse of the 1920s and 1930s. After the First World War, countries tried to rebuild the pre-1914 gold standard, but the effort was fragile and short-lived. When the Great Depression struck, governments turned inward. They raised tariffs, devalued their currencies to gain a trade advantage over neighbours, and imposed controls that strangled the flow of capital across borders. Each country tried to export its unemployment to the others, and the result was a downward spiral that deepened the slump and poisoned international politics. The second disaster was the war itself. By 1944 much of Europe and East Asia lay in ruins. Factories, ports, railways, and housing had been destroyed. Even the countries that would end up on the winning side faced years of shortage and rebuilding. The planners in Washington and London were convinced that the economic chaos of the 1930s had helped bring the war about, and they were determined not to repeat the mistake of leaving the recovery to chance and to bilateral bargaining. They wanted rules, institutions, and a shared framework that would make cooperation the normal path and beggar-thy-neighbour policy the exception. Two intellectual currents met in this project. One was the belief, associated with the economist #John_Maynard_Keynes, that markets left alone could settle at a bad equilibrium of mass unemployment, and that active public management of demand and of international finance was needed to keep economies near full employment. The other was the American conviction, carried by Treasury officials such as #Harry_Dexter_White, that stable exchange rates and open trade served both prosperity and the national interest of a country that had emerged from the war as the world's dominant creditor. These currents did not always agree, and their differences would mark the final charters. But they shared a starting point: that permanent, rules-based cooperation was better than the improvised and hostile arrangements of the previous decades. Within this larger project the idea of a bank had a specific job. A currency fund could help countries keep their exchange rates stable and cover short-term gaps in their balance of payments. But stability alone would not rebuild a bombed city or bring electricity to a poor countryside. That required long-term investment finance, the kind that private lenders in the 1930s had been unwilling or unable to provide across borders on any reliable scale. Private markets had failed to move capital from where it was abundant to where it was scarce, especially after the wave of sovereign defaults during the Depression had frightened investors away from foreign lending. The founders therefore imagined an institution that would do two things. It would lend its own funds for productive projects where private money was not available on reasonable terms. And, just as importantly, it would use its credit to encourage #private_investment by guaranteeing or participating in loans made by others. The bank was not meant to replace private capital. It was meant to prime the pump, to reduce the risk that had kept private lenders at home, and to channel resources toward #reconstruction and, in a phrase that would grow in importance, the development of less developed regions. This double ambition, part lender and part catalyst, is written into the very first article of the charter, and it explains much of what the institution became. It is worth pausing on how novel this was. There had been international loans before, and there had been public lending bodies inside countries. There had not been a permanent, treaty-based, member-owned institution designed to borrow from private markets and to lend across the world for productive investment. The founders were inventing an institutional form, and they had to write its rules from a nearly blank page. The care and caution of the resulting document reflect the seriousness with which they approached that task. 3. The Bretton Woods Conference and the Making of the Charter The charter was written in a hurry, in a specific place, by delegations that had already done years of preparatory work. In July 1944, as fighting continued in Europe and the Pacific, representatives of forty-four allied nations travelled to the Mount Washington Hotel in Bretton Woods, New Hampshire, for the United Nations Monetary and Financial Conference. The gathering ran from the first to the twenty-second of July and brought together more than seven hundred delegates and advisers. A smaller preparatory meeting had been held at Atlantic City in the weeks before, where the outlines of the two agreements were shaped so that the main conference could focus on settling disputes and drafting final text. Two figures dominated the intellectual background. #John_Maynard_Keynes led the British delegation and chaired the commission that dealt with the bank, while #Harry_Dexter_White of the United States Treasury led the American effort and chaired the commission on the currency fund. Their competing blueprints had circulated for two years. Keynes had proposed an ambitious international clearing union with its own reserve currency, a design that would have placed much of the adjustment burden on creditor countries. White favoured a more limited fund, financed by a fixed pool of national currencies and gold, that kept tighter control over how much credit any member could draw and that placed the United States dollar at the centre of the system. The plan that emerged resembled White's more closely than Keynes's, a reflection of the simple fact that the United States held most of the world's gold and would be the main source of funds. It is worth remembering that the bank was, in a sense, the junior partner at the conference. Most of the fierce negotiation concerned the #IMF and the rules of the monetary system. The bank was discussed later and with less drama, and several of its features were settled quickly once the harder currency questions were resolved. This does not mean the bank charter was careless. Rather, it means that the drafters could borrow structural ideas that had already been worked out for the Fund, such as the pattern of weighted voting and the model of a governing board resting on national subscriptions. On the twenty-second of July 1944 the delegates signed the Final Act, which contained the #Articles_of_Agreement for both the currency fund and the bank. Signature at the conference did not by itself create the institutions. Each government had to take the charter home and ratify it through its own constitutional process, and this took time and political effort. The charter of the bank entered into force on the twenty-seventh of December 1945, once governments holding a sufficient share of the planned #capital_subscriptions had formally accepted the treaty and their representatives met in Washington. The institution opened its doors for business on the twenty-fifth of June 1946. Its first loans came in 1947, when it approved a reconstruction credit of two hundred and fifty million United States dollars to France, at that time by far the largest single loan it would make measured against the institution's resources. The drafting context left visible marks on the text. The charter is cautious, technical, and preoccupied with protecting the institution's creditworthiness. It reads like a document written by finance officials who expected the new body to borrow from private markets and who wanted to reassure future bondholders that their money would be safe. This tone, more than any single clause, tells us what kind of institution the founders had in mind: not a charity, not an aid agency in the modern sense, but a bank with a public purpose that would have to earn and keep the confidence of the #bond_market. Historians of the institution have emphasised that this financial conservatism was baked in from the beginning and shaped the institution's culture for decades. 4. The Twin Institutions: How the Bank Was Tied to the Fund One feature of the charter is easy to overlook but important to grasp: the bank was not designed to stand alone. It was one half of a pair, joined at birth to the currency fund created by the other agreement signed at #Bretton_Woods. The two bodies had different jobs. The fund would help members keep their exchange rates orderly and would lend for short periods to cover temporary gaps in the balance of payments. The bank would provide long-term finance for productive investment. Yet the founders wove them together in ways that still matter. The clearest link sits in Article II of the bank's charter, which provides that only members of the #IMF may become members of the bank. A country that wants to join the bank must first join the fund and accept its obligations, including its rules on exchange arrangements and on providing economic information. This requirement created a single gateway into the whole #Bretton_Woods system. It also meant that the bank could rely on the fund's surveillance of members' economies, and that a country's standing in one institution carried over into the other. The two bodies shared a city, an annual meeting cycle, and a habit of close consultation, and they came to be known together as the #twin_institutions. The link had a logic beyond administrative tidiness. The founders believed that sound investment finance depended on sound money. There was little point lending for a power station in a country whose currency was collapsing or whose external accounts were in permanent crisis. By tying #membership together, the charter tried to ensure that borrowers from the bank were also participants in the disciplines of the monetary system. In practice this meant that the health of a borrower's macroeconomy, which the fund watched, was never far from the bank's own decisions, even though the bank was supposed to focus on projects. Over time the relationship between the two institutions grew more complicated, and at moments more tense. As the bank moved into lending tied to a country's economic policies, its work began to overlap with the fund's concern for macroeconomic stability. The two developed procedures to divide responsibilities and to avoid sending borrowers contradictory advice, though the boundary was never perfectly clean. The point for present purposes is that the 1944 charter did not create an isolated lender. It created one pillar of a paired structure, and the pairing shaped how the bank could operate. Reform debates about the international financial system almost always treat the two together, because the founders built them to work together. 5. The Architecture of the Charter: An Article by Article Reading The Articles of Agreement are organised into eleven articles, each divided into numbered sections. Reading them in order gives a clear picture of how the founders thought about the institution, moving from its reasons for existing to its money, its operations, its management, and finally the legal machinery that keeps it running. Article I sets out the purposes. It lists the aims of assisting reconstruction and development by facilitating the investment of capital for productive purposes, promoting #private_investment through guarantees and participations, promoting the balanced growth of international trade, arranging loans so that the most useful and urgent projects come first, and conducting operations with attention to their effect on business conditions. These purposes are examined in the next section, but their placement at the very front of the charter is deliberate. Under Article I the institution is instructed to be guided in all its decisions by the purposes set forth, which makes the article a lens through which every other provision must be read. Article II governs membership and capital. It provides that the original members are the countries represented at the conference that accept the charter, and that other countries may be admitted on terms set by the institution. As already noted, it ties membership in the bank to membership in the currency fund. Article II also fixes the authorised capital stock and the way each member's subscription is divided between a portion that is paid in and a much larger portion that is callable only if needed. The financial logic behind this split is discussed in section seven. Article III lays down general provisions relating to loans and guarantees. It states the conditions under which the institution may lend or guarantee, including the requirement that a member, its central bank, or a comparable agency guarantee repayment when the borrower is not the government itself. It also contains the principle that the institution should not finance a project when the borrower could obtain the money on reasonable terms from private sources, a rule meant to keep the institution from competing with the private capital it was supposed to encourage. Article IV covers operations, and it is one of the most consequential parts of the charter. It sets out the different methods by which the institution may make or facilitate loans, the terms and conditions attached, the currencies involved, and the ways the institution may use the funds it holds. It is within this article, at Section 10, that the charter places the rule prohibiting political activity, a provision examined in section nine. Article IV also contains protections for the institution's financial position, including rules about how members must handle the local currency the institution accepts and holds. Article V concerns organisation and management. It describes the three tiers of governance: the #Board_of_Governors, in which every member is represented; the #Executive_Directors, a smaller resident board that handles day-to-day decisions; and the President, who leads the staff and conducts ordinary business. Article V also sets the rules for voting, providing that each member has a number of basic votes plus additional votes tied to the shares it holds. This voting design, discussed in section eight, is the heart of the institution's governance and the focus of most reform proposals. Article VI deals with withdrawal and suspension of membership and with the suspension of the institution's operations. It preserves the right of a member to withdraw, sets out how accounts are settled when a country leaves, and provides for what happens if a member ceases to belong to the currency fund. These provisions matter because they define the exit options and the obligations that survive departure, questions that became real when countries considered leaving over the decades. Article VII sets out the status, #immunities, and privileges of the institution. It grants the bank full juridical personality, meaning it can contract, acquire property, and take legal action in its own name. It provides immunities designed to let the institution function without interference from any single member. Notably, and in contrast to the currency fund, this article allows actions to be brought against the bank in courts of competent jurisdiction in territories where it has an office or has issued or guaranteed securities. This limited openness to suit reflects the founders' awareness that a body borrowing from private investors must be able to be held to its contracts. Legal scholars treat this careful balance between immunity and accountability as one of the more sophisticated features of the charter. Articles VIII, IX, and X handle amendments, interpretation, and a technical rule on approvals. Article VIII provides that the charter may be amended, with most changes requiring acceptance by members holding a large supermajority of the total voting power, while a few especially sensitive changes require the consent of every member. This high threshold, discussed further below, means that no #amendment can pass over the objection of the largest shareholder. Article IX assigns the power to settle questions of #interpretation to the Executive Directors, with an appeal to the Board of Governors, keeping such disputes inside the institution rather than before outside courts. Article X provides that when the charter requires a member's approval before the institution acts, approval is deemed given unless the member objects within a set period, a device that keeps the machinery moving. Article XI contains the final provisions, including the arrangements for the charter's entry into force and the transitional steps needed to bring the institution to life. Taken together, the eleven articles form a compact but complete constitution for an international lender, balancing ambition in the purposes clause against caution in almost everything else. 6. Purposes: Reconstruction, Development, and the Catalysing of Private Capital The purposes in Article I repay close reading because they contain, in compressed form, the tension that has run through the institution's whole history. The first listed purpose speaks of assisting in the reconstruction and development of the territories of members by facilitating the investment of capital for productive purposes. The word reconstruction came first because the immediate problem in 1944 was a shattered Europe and Asia. But the word development sat beside it from the start, and it referred to the encouragement of productive facilities and resources in less developed countries. The founders did not treat these as two separate missions. They wrote them into a single clause, which allowed later leaders to shift the institution's centre of gravity from one to the other without amending the charter. The second purpose concerns #private_investment. The institution was told to promote private foreign investment by means of guarantees or participations in loans and other investments made by private investors, and, when private capital was not available on reasonable terms, to supplement private investment by providing finance from its own resources. This is the catalytic idea in legal form. The founders did not want a public monopoly on cross-border lending. They wanted an institution that would reduce risk, share it, and stand behind private lenders so that private money would flow again after the collapse of the 1930s. This ambition to mobilise private capital never disappeared, and it returned with new force in twenty-first century debates about how to fund the enormous investment needs of poorer countries. The third purpose looks outward to the whole trading system. It directs the institution to promote the long-range balanced growth of international trade and the maintenance of equilibrium in balances of payments by encouraging international investment for the development of the productive resources of members. Here the bank is placed inside the larger Bretton Woods vision, in which stable money, open trade, and productive investment were supposed to reinforce one another. The bank's lending was not an end in itself but a contribution to a healthier world economy. The fourth and fifth purposes are about sequencing and prudence. The fourth instructs the institution to arrange its loans, and to influence international loans by other channels, so that more useful and urgent projects, whether large or small, are dealt with first. The fifth tells it to conduct operations with due regard to the effect of international investment on business conditions in members' territories and, in the years right after the war, to help bring about a smooth transition from a wartime to a peacetime economy. These clauses reveal a self-image of the institution as a careful allocator of scarce capital, mindful of macroeconomic side effects, rather than a simple disburser of funds. Recent scholarship on the institution's history stresses how much interpretive room these purposes left open. The charter did not define development, did not rank reconstruction above development in any permanent way, and did not fix the boundary between what counted as a productive purpose and what did not. As the ruins of Europe were rebuilt with help from other sources, above all the #Marshall_Plan, the institution turned toward the poorer and newly independent countries of Asia, Africa, and Latin America, and the meaning of its purposes migrated with it. Historians of the institution have shown that this drift from reconstruction toward #development was as much a matter of practical opportunity and staff conviction as of any formal decision, and that the flexible language of Article I made the migration possible without a new treaty. 7. Capital and the Financial Model: Paid-In, Callable, and the Bond Market The most technically ingenious part of the charter is the way it arranges money. The founders faced a problem. They wanted an institution large enough to matter, but member governments, especially those still recovering from war, could not hand over vast sums of cash. The solution was a capital structure that let the institution command a very large balance sheet while requiring members to pay in only a small fraction of it up front. Under Article II each member subscribes to shares in the institution's capital. Only a portion of that subscription, originally twenty percent, was to be paid in and made available for lending and operations, and even that was split, with a small part payable in gold or United States dollars and the larger part in the member's own currency. The remaining eighty percent was not paid at all. It became #callable_capital, a legally binding promise that members would pay if, and only if, the institution ever needed the money to meet its own obligations to those who had lent to it or bought its securities. In normal times this callable portion sits unused on the books. It has never had to be called. This arrangement is more powerful than it looks. The callable capital functions as a guarantee, a treaty-based commitment by the world's governments that stands behind everything the institution borrows. Because that guarantee comes from a large group of states, including the wealthiest, lenders to the institution enjoy a level of security that no single project or borrowing country could offer. The institution can therefore raise money in private #bond_market at low interest rates, lend that money to member countries at rates those countries could not obtain on their own, and cover its costs from the margin, all without asking taxpayers for regular contributions. The callable capital, an instrument written into the charter in 1944, is the quiet engine of the whole model. The result over time was an institution that earned and kept an #AAA_rating, the highest grade credit-rating agencies give. That rating rests on four pillars that trace directly to the charter and to prudent practice: a strong base of paid-in and callable capital, a record of getting repaid because loans usually carry a #sovereign_guarantee, conservative financial management, and the backing of powerful shareholders. The founders' caution, the many clauses in the charter that protect the institution's financial position, was not timidity for its own sake. It was the price of access to private markets. An institution that intended to fund itself by selling bonds had to look, to investors, like the safest of borrowers. This model has costs as well as benefits, and contemporary analysts have examined them closely. Because the institution guards its rating so carefully, it tends to lend well within its theoretical capacity, holding more capital in reserve than a purely commercial calculation might require. In the reform discussions of the last few years, expert reviews commissioned by groups of major economies argued that #multilateral_development_bank balance sheets, including that of the bank, could be used more boldly, giving fuller recognition to the strength of callable capital and to the institutions' long history of being repaid. Proponents estimated that such changes could unlock a great deal of additional lending without new money from shareholders and without endangering the rating. The debate is, in effect, an argument about how conservatively to interpret a financial design first sketched in the 1944 charter. It is a striking illustration of how a decision made by finance officials in a New Hampshire hotel still frames a live policy question eight decades later. 8. Governance and Weighted Voting: Influence Tied to Money If the capital structure is the charter's most ingenious feature, the governance design is its most contested. The founders had to decide how power inside the institution would be shared among members that differed enormously in wealth and in the size of their contributions. They rejected the principle of one country, one vote that governs the General Assembly of the United Nations. Instead they chose #weighted_voting, a system in which a member's influence rises with its financial stake. The mechanics work as follows. Under Article V each member receives a fixed number of #basic_votes, the same for every member, plus additional #share_votes, one for each share of capital stock the member holds. The basic votes give even the smallest member a floor of representation and express the idea that membership itself carries some voice. The share votes, which vastly outnumber the basic votes, tie real power to capital subscriptions. Over the institution's history the share of total votes made up by basic votes has been small, so the practical distribution of power closely follows the distribution of shares, and the distribution of shares reflects the relative economic weight of members. This design has one dominant consequence. The largest subscriber, the United States, has always held the single biggest block of votes, historically in the region of fifteen to sixteen percent of the total in the bank. That number matters because of the amendment rule in Article VIII and because certain major decisions require a supermajority of the total #voting_power, commonly set at eighty-five percent. A shareholder with more than fifteen percent of the votes can, by itself, block any decision that needs an eighty-five percent majority. In effect the charter gives the United States a #veto over the most important structural changes, including changes to the charter itself. No other member holds this power alone, though groups of members can combine to reach blocking thresholds. The governance structure has three levels, all set out in Article V. At the top sits the #Board_of_Governors, made up of one governor for each member, usually the country's finance minister or an equivalent official. The Board of Governors holds the ultimate authority but meets only once a year and delegates most of its powers. Below it sits a resident board of #Executive_Directors, a smaller group who work at the institution's headquarters and take the great majority of operational decisions, including approving loans and policies. The largest shareholders appoint or elect their own directors, while other members are grouped into constituencies that share a director among several countries. At the third level stands the #President, who chairs the board of directors, leads the staff, and runs ordinary business. By a long-standing convention that is not written into the charter, the President has always been a national of the United States, just as the head of the currency fund has by convention been a European. These arrangements have drawn steady criticism, and the criticism has grown louder as the world economy has changed. When the institution was founded, its shareholding roughly matched the economic map of 1944. Eight decades later, several large and populous countries, especially in Asia, are widely seen as underrepresented, while a number of smaller European economies hold shares and board influence out of proportion to their current weight. Because the amendment and supermajority rules make any redistribution difficult without the consent of the dominant shareholder, formal reform has moved slowly. Scholars of #global_economic_governance describe a system that adjusts through occasional, negotiated increases in shares rather than through any automatic rule, and they note that the mismatch between economic reality and voting power has become one of the central #legitimacy problems of the institution. 9. The Apolitical Mandate: Article IV, Section 10 Perhaps no single clause has generated more legal and ethical debate than Article IV, Section 10, the provision headed Political Activity Prohibited. In plain terms, this clause instructs the institution and its officers not to interfere in the political affairs of any member and not to be influenced in their decisions by the political character of the member concerned. Decisions are to be based only on economic considerations, and those considerations are to be weighed impartially in order to achieve the purposes in Article I. A related provision requires that the proceeds of a loan be used without regard to political or other non-economic influences. The founders had concrete reasons for writing this rule. Member governments, especially smaller and weaker ones, worried that a powerful international lender dominated by the great powers might use its money to meddle in their internal politics, to punish governments it disliked, or to reward friendly regimes. The apolitical mandate was a promise that the institution would treat members as economic partners rather than as political subjects, that a country would not be denied a road or a power plant because of the ideology of its government. The clause was, in this sense, a protection for members and a discipline on the institution. Over time, however, the clause became a source of difficulty rather than a simple safeguard. The trouble is that the line between the economic and the political is not fixed, and much of what determines whether a loan succeeds is bound up with how a country is governed. When the institution began, in later decades, to care about corruption, the quality of public administration, the rule of law, the treatment of people displaced by its projects, and the environmental damage its projects could cause, it found itself pressing on matters that looked political to some members and merely economic to the institution's own lawyers and staff. Legal scholarship has traced how the institution's general counsel gradually reinterpreted the clause, distinguishing forbidden interference in a member's politics from the legitimate economic concern with #governance factors that affect whether money is well spent. This reinterpretation allowed the institution to launch #anti_corruption efforts, to build a sanctions system for firms that engage in fraud on its projects, and to attach conditions about public financial management to its lending, all while insisting that it remained within the bounds of Article IV, Section 10. Critics from different directions have found this uncomfortable. Some argue that the clause is used selectively, invoked to avoid confronting powerful members or to duck questions about #human_rights, while quietly set aside when the institution wants to push a particular policy agenda. Others argue that the institution has already strayed too far from the apolitical promise and now shapes the domestic choices of borrowing countries in ways the founders never intended. Legal analyses of the clause, including recent work focused on its implications for the fight against corruption in Africa, show that the tension is not a mistake to be corrected but a structural feature of an institution that must operate inside sovereign states while claiming to stand above their politics. The deeper point is that Article IV, Section 10 tries to draw a boundary that development itself keeps erasing. Building institutions, reforming budgets, protecting the environment, and reducing poverty are economic tasks with unavoidable political content. The clause survives because it does useful work, reassuring members and disciplining the institution, but it cannot resolve the underlying question of how a public lender should relate to the governments it finances. That question is as alive today as it was in 1944. 10. From One Bank to a Group: The Evolution into the World Bank Group The institution created by the 1944 charter did not stay alone for long, and it did not stay fixed on reconstruction. Over the following decades it became the anchor of a family of five related bodies, collectively known as the #World_Bank_Group, and its mission shifted decisively toward the reduction of poverty in the developing world. Understanding this evolution is essential, because the charter of the bank is the legal ancestor of the whole group, and the group's later institutions were built to fill gaps that the original charter left open. The first gap concerned the private sector. The bank generally lends to governments or requires a government guarantee, which limits its ability to finance private firms directly. To reach private business, members created the #IFC in 1956, with its own charter modelled on the parent institution. The IFC invests in and lends to private companies in developing countries without demanding a sovereign guarantee, taking on risk that the parent body cannot. The second gap concerned the poorest countries. The bank lends on terms close to those of the market, which the poorest members could not afford. To serve them, members created the #IDA in 1960. The IDA provides #concessional_finance, meaning credits with very low or zero interest and long repayment periods, along with grants, to low-income countries. Together the parent bank and the IDA are what most people mean when they say the World Bank, and they share the same staff, leadership, and headquarters while resting on separate charters and separate financing. Three further institutions completed the group. The #ICSID, established through a convention that entered into force in 1966, provides a forum for settling investment disputes between states and foreign investors, which reduces one of the risks that discourages cross-border investment. The #MIGA, established in 1988, offers insurance against political and other non-commercial risks to encourage foreign direct investment in developing countries. Each of these bodies has its own founding agreement, but all trace their institutional lineage and much of their governance style to the 1944 charter. Membership in the group still flows through the parent institution, since joining the other bodies generally requires membership in the bank, which in turn requires membership in the currency fund. Alongside this institutional growth came a change in purpose that the flexible language of Article I permitted. The reconstruction of western Europe was largely financed from other sources, above all the #Marshall_Plan, and by the 1950s the institution had turned its attention to the newly independent and developing countries. Its lending moved from rebuilding bombed infrastructure toward dams, power, transport, agriculture, and, over time, education, health, environmental protection, and policy reform. Historians of the institution describe a long journey from a narrow post-war lender into a self-described global development body with a stated mission of ending extreme poverty. The parent institution today counts one hundred and eighty-nine members and describes itself as the largest #multilateral_development_bank in the world. This journey was not smooth or uncontested. The turn toward policy-based lending in the 1980s, in which loans came with conditions requiring borrowers to change their economic policies, produced the era of #structural_adjustment that remains one of the most criticised chapters in the institution's history. Critics argued that the conditions imposed austerity and market reforms that harmed the poor and narrowed the policy choices of borrowing governments. Defenders argued that the reforms addressed real problems of unsustainable spending and distorted prices. The debate over that period feeds directly into present arguments about the institution's proper role, and it shows how far the meaning of the 1944 purposes had travelled from the founders' immediate concern with rebuilding a ruined Europe. 11. Criticisms and Defences: Weighing the Charter's Legacy Any honest account of the 1944 charter has to sit with the sharp disagreements about what it produced. The institution it created is admired and resented, often for the same reasons, and the arguments tend to cluster around a few themes that connect directly back to the founding text. The first theme is power and #accountability. Critics argue that the weighted voting design hard-wired the dominance of a handful of wealthy states, above all the United States, and that this dominance has shaped the institution's advice and lending in ways that suit powerful members and their firms. On this view the charter dressed up a political hierarchy in the neutral language of shares and votes. Defenders reply that weighted voting was the price of getting the wealthy countries to put up the money and stand behind the guarantees, and that an institution run on one country, one vote would never have earned the market confidence that made it useful. Both sides are describing the same clause from opposite ends. The second theme is #conditionality and #ownership. The practice of attaching policy conditions to loans, which grew far beyond anything the founders discussed, drew fierce criticism during the #structural_adjustment years and after. Opponents said the institution used its money to override the choices of elected governments and to impose a narrow economic model. Supporters said that lending without conditions would waste resources on bad policies and that borrowers, in principle, agreed to the terms. Recent scholarship has moved toward a more nuanced position, stressing the importance of genuine local ownership of reforms and questioning the effectiveness of conditions imposed from outside, while recognising that lenders cannot ignore how their money will be used. The third theme is the tension inside the apolitical mandate, already discussed, between a promise of political neutrality and a practice that reaches deep into how countries are governed. The fourth theme is environmental and social harm, as large infrastructure projects displaced communities and damaged ecosystems, prompting the institution over time to build safeguard policies and an independent complaints mechanism. These developments were not foreseen in 1944, but they grew out of the gap between the charter's project-lending model and the human consequences of the projects. What is striking is that the strongest criticisms and the strongest defences both point back to the founding design. The institution is durable because the charter made it financially self-sustaining and gave the wealthy a stake in its success. The institution is criticised as unaccountable because the same design concentrated control. The institution has done real and lasting good, financing infrastructure and services that would not otherwise have been built, and it has also caused real harm and made real mistakes. A balanced reading of the charter's legacy does not require choosing one story. It requires seeing that the founders built a powerful instrument whose power cuts in more than one direction. 12. The Charter at Eighty: Legitimacy, Voice, and Reform The eightieth anniversary of the Bretton Woods conference, marked in 2024, arrived at a moment of unusual pressure on the institution and its charter. A cluster of overlapping crises, the aftermath of a pandemic, high debt levels across the developing world, the enormous investment needs of the climate transition, and rising geopolitical rivalry, has forced a fresh look at whether an institutional design from 1944 still fits the world it must serve. The reform debate touches almost every feature of the charter examined in this article, which is why the anniversary produced a wave of analysis and proposal rather than mere commemoration. The sharpest debates concern voice and #legitimacy. Because the charter ties #voting_power to capital subscriptions and protects the position of the largest shareholders through supermajority rules, the distribution of influence has changed only slowly even as the world economy has been transformed. Several large emerging economies argue that they carry a weight in global output and population that their shares and board seats do not reflect, while a number of smaller advanced economies hold influence that looks outsized by the same measure. Reform advocates have called for a rebalancing of shares, a review of the constituency system that groups many countries under a single director, and changes to the unwritten convention that reserves the presidency for a national of the largest shareholder. Because formal amendment requires a supermajority that the dominant shareholder can block, progress has depended on negotiated shareholding reviews rather than on any automatic rule, and these reviews have repeatedly stalled. A second cluster of debates concerns money and mandate. Analysts and shareholder governments have pressed the institution to lend more, both to meet development needs and to help finance the response to climate change. Much of this discussion returns to the financial design written into the charter in 1944, and in particular to the question of how conservatively the institution should manage its balance sheet and interpret the strength of its #callable_capital. Independent expert reviews commissioned by groups of major economies have argued that the institution and its peers could expand lending substantially by using existing capital more efficiently, while others caution that pushing too hard could threaten the high credit rating on which the whole model depends. At the same time, some borrowing countries worry that a broader mandate, especially one focused on global challenges such as #climate_finance, could pull scarce resources away from their own national development priorities. This is a modern replay of the founders' original balancing act between ambition and prudence. A third strand of debate revisits the purposes and the wider architecture. Some scholars and policymakers have called for what they describe as a #new_Bretton_Woods, a fundamental reworking of the rules and institutions of #international_financial_architecture to match twenty-first century realities of development finance, debt, taxation, and climate. Others argue for evolution rather than replacement, updating the existing charters and practices without discarding a system that, whatever its faults, has proved durable and has mobilised very large sums of finance over its lifetime. Recent book-length treatments make the case for a renewed multilateral bargain, while detailed policy analyses debate the specific mechanics of #reform. What unites these otherwise different positions is the recognition that the 1944 charter, and the institution it created, remain central to any serious plan for the global economy, so that reform must run through the charter rather than around it. It is striking how many of these contemporary arguments are, at bottom, arguments about the original design choices. The complaint about voice is a complaint about weighted voting. The argument over lending capacity is an argument about callable capital and the financing model. The unease about mandate creep and about conditions attached to loans is an argument about the purposes clause and about the apolitical mandate. The founders built an institution whose strengths and whose limits flow from the same decisions, and the reform debate at eighty is, in large part, a debate about which of those original choices to keep and which to revise. 13. Conclusion The Articles of Agreement of the #IBRD are a short document that has carried an extraordinary weight. Written quickly at #Bretton_Woods in July 1944, brought into force in December 1945, and put to work from 1946, the charter created the first institution of what became the #World_Bank_Group and set the terms on which that institution would raise money, choose projects, share power, and relate to the governments it finances. Its purposes were broad enough to let the institution travel from post-war reconstruction to global #development without a new treaty. Its capital structure, resting on #callable_capital and access to the bond market, gave it the means to lend at scale without depending on annual grants. Its governance, built on #weighted_voting, tied influence to financial stake and handed the largest shareholder a durable #veto. Its #apolitical_mandate promised members that the institution would judge them by economics rather than politics, a promise that has proved both useful and impossible to keep cleanly. Each of these features has been praised and attacked, and the same features sit at the centre of the reform debates that surrounded the institution's eightieth anniversary. That is the deepest lesson of a close reading of the charter. The document was not a neutral piece of administrative plumbing but a set of deliberate choices, each reflecting the fears, hopes, and power balances of 1944. The institution's later strengths, its financial durability and its ability to convene and to lend across the world, grew from those choices. So did its most criticised limits, the unequal voice of its members and the difficulty of squaring a promise of neutrality with the political reality of development. Students who want to understand the World Bank, and the many arguments about its future, will find that most of those arguments lead back to the eleven articles drafted in a New Hampshire hotel eight decades ago. The charter is not merely a historical curiosity. It is the living constitution of one of the most important #international_financial_institutions in the world, and it rewards being read closely, sceptically, and with an eye to how a few pages of careful legal drafting can shape the fortunes of nations for generations. #IBRD #World_Bank #Bretton_Woods_1944 #Articles_of_Agreement #World_Bank_Group #International_Financial_Institutions #Development_Finance #Global_Economic_Governance #Multilateral_Development_Banks #Keynes_and_White #Twin_Institutions #Governance_Reform #Weighted_Voting #Callable_Capital #Postwar_Economic_Order References Alacevich, M., and Hosman, M. T. (2024). History and histories of the World Bank. In A. Vetterlein and T. Schmidtke (Eds.), The Elgar Companion to the World Bank (pp. 22-37). Cheltenham: Edward Elgar Publishing. Bayliss, K., and Fine, B. (2024). Locating the World Bank: the unmaking and remaking of development economics in its shifting vision. In A. Vetterlein and T. Schmidtke (Eds.), The Elgar Companion to the World Bank (pp. 38-50). Cheltenham: Edward Elgar Publishing. Bradlow, D. D. (2023). The Law of International Financial Institutions. Elements of International Law. Oxford: Oxford University Press. https://doi.org/10.1093/law/9780192862822.001.0001 Deciancio, M., Nemina, P., and Tussie, D. (Eds.). (2022). Handbook on the Politics of International Development. Cheltenham: Edward Elgar Publishing. Gallagher, K. P., and Kozul-Wright, R. (2022). The Case for a New Bretton Woods. Cambridge: Polity Press. Gasbarri, L. (2024). The World Bank and legal studies. In A. Vetterlein and T. Schmidtke (Eds.), The Elgar Companion to the World Bank (pp. 62-72). Cheltenham: Edward Elgar Publishing. Jacobs, M. (2023). Bretton Woods 2.0 and a new Marshall Plan? Brazil's G20 presidency and reform of the international financial architecture. CEBRI-Journal, 2(8), 112-125. https://doi.org/10.54827/issn2764-7897.cebri2023.08.03.07.112-125.en Koch-Weser, M., and Guggenheim, S. (Eds.). (2021). Social Development in the World Bank: Essays in Honor of Michael M. Cernea. Cham: Springer. Lemdjo, F. M. Y. (2021). The political prohibition clause of the World Bank Charter and the legal implications for the fight against corruption in Africa. Journal of Sustainable Development Law and Policy, 11(2), 382-406. https://doi.org/10.4314/jsdlp.v11i2.5 Thornton, C. (2021). Revolution in Development: Mexico and the Governance of the Global Economy. Oakland: University of California Press. Vetterlein, A., and Schmidtke, T. (2024). The World Bank: a changing organization in a changing world. In A. Vetterlein and T. Schmidtke (Eds.), The Elgar Companion to the World Bank (pp. 2-20). Cheltenham: Edward Elgar Publishing. Vetterlein, A., and Schmidtke, T. (Eds.). (2024). The Elgar Companion to the World Bank. Cheltenham: Edward Elgar Publishing.

The #Articles_of_Agreement of the #International_Monetary_Fund, signed at #Bretton_Woods, New Hampshire, on 22 July 1944, remain one of the most important legal and economic documents of the twentieth century. This article examines how and why the treaty was written, what it actually says, and how it built an institution meant to watch over the #global_financial_system and encourage #monetary_cooperation among nations. It traces the intellectual and political struggle behind the treaty, especially the contest between the British plan associated with John Maynard Keynes and the American plan associated with Harry Dexter White. It then explains the core commitments the Articles created: a system of fixed but adjustable exchange rates, a pool of national contributions known as quotas, a set of rules on payments and currency practices, and a governing structure built on weighted voting. The article follows the treaty through its seven amendments, showing how the Fund survived the collapse of the original fixed exchange rate arrangement in the early 1970s and reinvented itself around surveillance, lending, and technical support. Drawing on recent scholarship, it also reviews the main criticisms directed at the Fund, including debates about conditionality, austerity, legitimacy, and the under representation of developing economies. The article argues that the 1944 charter succeeded not because it predicted the future correctly, but because it was flexible enough to be rewritten as the world changed, while keeping its founding purpose of cooperative management of international money. The study is intended as a clear, student friendly reference that still meets the standards of a scholarly review. Keywords: #International_Monetary_Fund; #Articles_of_Agreement; #Bretton_Woods; #exchange_rate_stability; #global_economic_governance; #Special_Drawing_Rights INTRODUCTION Every day, money moves across borders in enormous amounts. Traders pay for goods in foreign currencies, governments borrow from abroad, tourists exchange cash, and central banks hold reserves in currencies that are not their own. For all of this to work, countries need some shared understanding of how currencies relate to one another and what happens when a nation runs out of the foreign money it needs to pay its bills. Before 1944, no permanent global body existed to manage these questions. When trouble came, as it did badly in the 1930s, nations acted alone, and their separate choices often made the shared problem worse. The #Articles_of_Agreement of the #International_Monetary_Fund were the world's first serious attempt to fix this gap through a permanent, treaty based institution. Signed in the small resort town of #Bretton_Woods in July 1944, while the Second World War was still being fought, the document created an organization whose job was to oversee international money and to push governments toward cooperation rather than conflict. The #International_Monetary_Fund, usually shortened to the #IMF or simply the Fund, opened for business a few years later and has operated continuously ever since. It began with twenty nine ratifying governments and today counts one hundred and ninety members, which makes it close to universal. This article has three aims. The first is to explain the historical setting that made the treaty necessary, so that readers understand the fears and hopes of the people who wrote it. The second is to describe, in plain terms, what the Articles actually established: the purposes of the Fund, the exchange rate rules, the quota system, and the governing organs. The third is to trace how the treaty changed over eight decades and how the modern Fund fits into the present global economy. Throughout, the article keeps a balance between describing the institution as its founders intended it and reporting the serious debates that surround it today. The Fund is admired by some as a stabilizing anchor and criticized by others as an instrument of powerful states. Both views deserve a fair hearing, and recent research gives us better evidence than ever before to weigh them. The discussion is organized as follows. Section two sets out the historical background, especially the economic disasters of the interwar years. Section three describes the #Bretton_Woods conference and the drafting contest. Section four analyzes the structure and content of the treaty. Section five focuses on #global_economic_governance and the voting system. Section six explains the three main functions of the modern Fund. Section seven follows the amendments and the collapse of the original exchange rate system. Section eight places the Fund in the contemporary financial order. Section nine reviews criticisms and reform debates. The conclusion draws the threads together. THE ROAD TO BRETTON WOODS: WHY THE WORLD NEEDED A MONETARY INSTITUTION To understand why forty four nations sent delegates to a New Hampshire hotel in the middle of a war, we have to look back at the two decades before it. The interwar period, roughly 1919 to 1939, was a case study in how badly international money can break down when there are no shared rules and no trusted referee. Before the First World War, most major economies operated on the classical #gold_standard. Each currency was defined as a fixed weight of gold, which meant that exchange rates between currencies were effectively fixed as well. The system had real advantages. It made international prices predictable and discouraged governments from printing money recklessly, because they had to keep enough gold to back their currency. But it was also rigid. A country facing an economic slump could not easily loosen policy to help its own people, because doing so risked losing gold. The system worked partly because Britain and the City of London managed it with a light but steady hand. The war shattered this arrangement. Governments printed money to pay for the fighting, gold links were suspended, and by the 1920s the world tried to rebuild a version of the old standard on unstable foundations. The attempt failed. Reparations from the defeated powers, large war debts among the victors, and mismatched exchange rates created constant strain. When the #Great_Depression struck after 1929, the whole structure fell apart in a way that scarred everyone who lived through it. The 1930s produced a set of behaviors that the #Bretton_Woods planners were determined never to repeat. Countries devalued their currencies one after another, each hoping to make its exports cheaper and grab a larger share of shrinking world trade. This tactic is sometimes called competitive devaluation, and because one nation's gain came at another's expense, economists describe it as a beggar thy neighbor policy. Nations also threw up trade barriers, imposed strict controls on moving money across borders, and split into rival currency blocs. World trade collapsed, unemployment soared, and the economic bitterness fed political extremism. Many of the planners believed, with good reason, that the financial chaos of the 1930s had helped pave the road to war. Two lessons stood out. First, the #balance_of_payments problem needed a shared solution. A country's balance of payments is, in simple terms, the record of its money coming in and going out through trade and finance. When a nation persistently spends more abroad than it earns, it runs short of foreign currency and gold, and it faces painful choices: cut spending hard, devalue, or block payments. Left alone, a government under this pressure tended to choose options that hurt its trading partners. The planners wanted an institution that could lend a country foreign currency for a while, giving it breathing room to adjust in an orderly way instead of lashing out. Second, exchange rate changes needed rules and oversight. Sudden, secret, or purely selfish devaluations poisoned trust between nations. If countries were going to change the value of their currencies, they should do so openly, for genuine reasons, and after consulting others. The idea of exchange rate stability, backed by an agreed procedure for change, became a central goal. It is worth pausing on how unusual the planning was. Governments were designing the peacetime economic order while their soldiers were still fighting and dying, and while the outcome of the war was not yet certain. This tells us something about how deeply the lesson of the 1930s had sunk in. The people drafting these plans believed that winning the war would not be enough. If the victors repeated the economic mistakes that followed the First World War, they feared the peace would again give way to depression, resentment, and conflict. The design of the monetary system was treated as part of the war effort itself, a way of securing a durable peace rather than a mere technical afterthought. Preparatory talks were held in the months before the main conference, including a drafting meeting at Atlantic City in June 1944, so that delegates arrived at the mountain hotel with much of the groundwork already laid. By the early 1940s, even while the war raged, officials in Washington and London were already sketching a better system for the peace that would follow. The Atlantic Charter of 1941 and later wartime agreements pointed toward a cooperative economic order. The two great architects of this vision were John Maynard Keynes, the most famous economist of the age, working for the British Treasury, and Harry Dexter White, a senior official in the United States Treasury. Their competing blueprints would shape the treaty signed at #Bretton_Woods. THE BRETTON WOODS CONFERENCE AND THE DRAFTING CONTEST The formal meeting that produced the treaty was the United Nations Monetary and Financial Conference, held from 1 to 22 July 1944 at the Mount Washington Hotel in #Bretton_Woods, New Hampshire. Around seven hundred delegates from forty four nations attended. The location was remote and quiet, which suited the enormous task of negotiating two new international institutions at once: the #IMF and its sister body, the International Bank for Reconstruction and Development, which grew into the #World_Bank. The Soviet Union sent a delegation and took part in the talks, though it later chose not to ratify the agreements. The intellectual heart of the conference was a disagreement between the British and American plans. The #Keynes plan proposed a bold new institution called the International Clearing Union. Under this design, an international body would create a new reserve asset, which Keynes called bancor, and would run a global system of accounts. The plan was generous to debtor nations and, importantly, tried to put pressure on creditor nations too. In Keynes's view, if a country ran a large and lasting surplus, hoarding the world's money while others struggled, that surplus country was part of the problem and should share the burden of adjustment. This reflected Britain's expectation that it would emerge from the war as a debtor needing support. The American plan, associated with #Harry_Dexter_White, was more cautious and more protective of creditor interests, which suited the position of the United States as the coming financial superpower. Instead of a powerful clearing union that could create money on a large scale, the White plan envisioned a stabilization fund built from national contributions. Countries would pay in money according to their economic size, and they could draw on this common pool when they hit balance of payments trouble. Access to the pool would be limited and conditional, not automatic and unlimited. The burden of adjustment would fall mainly on the deficit country that needed to borrow, not on the surplus country. Because the United States held most of the world's gold and would be the main source of funds, the American vision largely prevailed. The final #Articles_of_Agreement created a quota based fund much closer to White's design than to Keynes's clearing union. Keynes did win some points and shaped the language and spirit of the treaty, and his warnings about the dangers of putting all the adjustment burden on debtors would echo in criticisms of the Fund for decades. But the core machinery was American. The delegates signed the #Articles_of_Agreement on 22 July 1944. The treaty then had to be ratified by national governments. It entered into force on 27 December 1945, once countries holding a large enough share of the planned quotas had formally accepted it. The Fund began financial operations in 1947. What the delegates had produced was not a vague statement of good intentions but a working charter with detailed rules, financial commitments, and a permanent organization to enforce them. The next section examines what those rules said. STRUCTURE AND CONTENT OF THE ARTICLES OF AGREEMENT The #Articles_of_Agreement function as the constitution of the Fund. They set out its purposes, define who can join and on what terms, establish the exchange rate system, create the financial pool of quotas, lay down rules on currency practices, and describe the organs that run the institution. It helps to take these in turn. 4.1 The Purposes of the Fund The very first article states the purposes of the Fund, and everything else in the treaty is meant to serve them. The founders listed several goals that fit together. The Fund was to promote #international_cooperation on monetary matters through a permanent body where governments could consult and work together. It was to help expand and balance the growth of world trade, and through that to support high employment, rising incomes, and the development of productive resources. It was to promote #exchange_rate_stability and orderly currency arrangements, and to discourage the competitive devaluations that had done such damage in the 1930s. Two further purposes deal directly with the balance of payments problem. The Fund was to help build a multilateral system of payments, meaning a world in which money for trade could move freely between members rather than being trapped by restrictions and rival blocs. And it was to give members confidence by making the Fund's resources temporarily available to them, under adequate safeguards, so that they could correct problems in their balance of payments without resorting to measures that damage national or international prosperity. The final purpose ties these together: to shorten the duration and lessen the degree of imbalance in members' international payments. These purposes still guide the institution today, and they are quoted in almost every serious study of the Fund. They matter because the Fund's later actions, including its lending conditions and its policy advice, are supposed to be judged against them. 4.2 The Par Value System: Fixed but Adjustable Exchange Rates The original treaty built a specific kind of exchange rate arrangement, often called the #par_value_system. Each member agreed to set an official value, or par value, for its currency. This value was expressed in terms of gold or, in practice, in terms of the United States dollar, which was itself tied to gold at a fixed rate of thirty five dollars per ounce. Members promised to keep the actual market value of their currency close to its par value, within a narrow band of about one percent on either side. Central banks did this by buying and selling currencies to steady the rate. The genius of the design was in the word adjustable. Unlike the old #gold_standard, the #Bretton_Woods system allowed a country to change its par value if it faced what the treaty called a fundamental disequilibrium, meaning a deep and lasting imbalance that could not be fixed by ordinary means. But a country could not do this secretly or on a whim. Small changes were allowed with notice, and larger changes required consultation with the Fund. This was the answer to the competitive devaluations of the 1930s. Exchange rates would be stable most of the time, giving traders and investors confidence, but they could be adjusted in an orderly, supervised way when a genuine problem demanded it. At the center of this arrangement sat the United States dollar, freely convertible into gold for foreign governments. Other currencies were pegged to the dollar, and the dollar was pegged to gold. This made the dollar the anchor of the whole structure. That central role brought the United States great influence, but it also planted a weakness that would eventually break the system, as later sections explain. 4.3 Quotas: The Financial and Political Core If the par value system was the treaty's exchange rate rule, the #quota system was its financial engine and, quietly, its political heart. When a country joins the Fund, it is assigned a quota, which is a sum broadly reflecting its size and importance in the world economy. The member pays this quota as its subscription into the Fund. Historically part was paid in gold or hard currency and part in the member's own currency. Quotas are measured in the Fund's own unit of account, the #Special_Drawing_Rights, which later sections explain. Quotas do several jobs at once, and this is why they matter so much. First, they determine how much money a country contributes to the common pool, and therefore how large that pool is. Second, they set a rough limit on how much a country can normally borrow from the Fund when it needs help. A larger quota means larger access. Third, and most sensitive of all, quotas largely determine #voting_power. A member's votes are made up of a set of equal basic votes given to every country plus additional votes tied to the size of its quota. Because quotas track economic weight, the largest economies hold the most votes. This design was deliberate. The founders rejected the one country, one vote model used in bodies like the United Nations General Assembly. They chose weighted voting so that the states putting the most money at risk would have the most say, a model sometimes called shareholder governance. It made the institution attractive to the powerful countries whose participation and money were essential. But it also built in a permanent tension. As the world economy changed and new powers rose, the distribution of quotas often lagged behind reality, leaving fast growing economies under represented. That tension drives the #quota_reform debates that continue to this day, discussed in section nine. 4.4 Rules on Currency Practices and Convertibility The treaty did more than set exchange rates and pool money. It also laid down obligations about how members should treat currency and payments. A central goal was current account convertibility, meaning that members should not block payments and transfers for ordinary trade in goods and services. If a merchant in one country sold goods to a buyer in another, the seller should be able to convert and repatriate the payment. Restricting such payments was the kind of behavior that had strangled trade in the 1930s, so the treaty pushed members toward removing these barriers, while allowing time and exceptions for countries not yet ready. At the same time, the founders were more relaxed about controls on capital movements, meaning flows of investment money and short term speculative funds rather than payments for trade. In the 1940s, memories of destabilizing speculation were fresh, and the treaty permitted members to regulate capital flows. This distinction between current payments, which should be free, and capital flows, which could be controlled, is a subtle but important feature of the original design. Debates about how far countries should open themselves to volatile capital flows remain lively in economics, and the Fund's own thinking on the matter has shifted over the years. 4.5 The Organs of the Fund Finally, the treaty created the machinery to run the institution. Three organs stand out, and a fourth advisory body developed later. At the top sits the #Board_of_Governors, made up of one governor from each member country, usually a finance minister or central bank head. The Board of Governors holds ultimate authority and decides the biggest questions, such as admitting new members, approving quota changes, and amending the Articles. But it meets only occasionally, so it delegates most day to day power. That daily power rests with the #Executive_Board, a smaller group of executive directors who work full time at the Fund's headquarters in Washington. The Executive Board handles the constant business of the institution: approving loans, reviewing members' economies, and setting policy within the limits the treaty allows. The largest economies have historically appointed their own directors, while groups of smaller countries share directors through a constituency system. Over time the size of the board and the mix of appointed and elected directors changed through amendment and agreement. Leading the staff is the #Managing_Director, who chairs the Executive Board and runs the organization day to day. By a long standing informal understanding, the Managing Director has always been a European, while the head of the World Bank has been an American. This convention is not written in the treaty, and it is increasingly criticized as outdated, but it has proved remarkably durable. A fourth body, the International Monetary and Financial Committee, developed as a high level advisory group that guides the Fund's direction, though it does not have formal decision making power in the way the boards do. Together these organs turn the words of the treaty into the actions of a living institution. GLOBAL ECONOMIC GOVERNANCE AND THE POLITICS OF VOTING No feature of the #Articles_of_Agreement has generated more study and more argument than the way power is distributed inside the Fund. Because so much flows from the quota and #voting_power system, it deserves a section of its own. The heart of the matter is weighted voting. Most decisions can be made by a simple majority of the votes cast, but the treaty reserves the most important decisions for special supermajorities. Many major matters, including amendments to the Articles and major quota decisions, require an eighty five percent majority of total voting power. This threshold has a striking consequence. Because a single member, the United States, has historically held a share of the votes just above the level needed to block an eighty five percent majority, that member can veto the most important changes on its own. No other single country has this power. Other big decisions require a seventy percent majority, which shifts influence toward coalitions of large shareholders. Scholars describe two layers of power inside the Fund. The first is the formal layer written into the treaty: quotas, votes, and the thresholds just described. The second is an informal layer of influence that operates through negotiation, agenda setting, and quiet pressure inside the boards. Recent research using large collections of Executive Board records has shown that the most powerful members do not simply cast votes; they shape discussions and outcomes in ways that reflect their trade and strategic interests, and they coordinate with one another to steer decisions (Forster, Honig and Kentikelenis, 2025). This work challenges the older assumption that the boards rarely matter because votes are seldom taken. In fact, the deliberations themselves carry weight, and the influence of the largest members shows up in the policy conditions later attached to loans. The politics of quotas also drives a long running reform debate. The treaty requires the Fund to review quotas at regular intervals so that they can keep pace with the changing world economy. In practice, these reviews are slow and highly political, because raising one country's share usually means lowering another's, and no powerful state wants to give up influence. The result is that the voting shares of major emerging economies have often lagged well behind their real economic weight, while some smaller advanced economies remain over represented. For many observers, especially those speaking for the #Global_South, this gap between economic reality and formal power is the central weakness of the Fund's #global_economic_governance. It raises questions of #legitimacy that reach into every part of the institution's work, because a body that claims to speak for the world economy is on weaker ground when its own voting map is decades out of date. Understanding this governance structure is essential for understanding the criticisms discussed later. Many complaints about the Fund are, at bottom, complaints about who controls it and whose interests it serves. The weighted voting model, chosen in 1944 to attract the powerful, is both the source of the Fund's strength and the focus of its critics. THE CORE FUNCTIONS OF THE MODERN FUND The treaty gave the Fund several jobs, and over time these settled into three main functions that define the institution today: surveillance, lending, and capacity development. Each grows out of the original purposes, even though the tools have changed a great deal since 1944. 6.1 Surveillance The first function is #surveillance, meaning the ongoing monitoring of the world economy and of members' individual policies. The legal basis for much of this work is #Article_IV of the treaty, and the annual health checks the Fund performs are still known as Article IV consultations. During a consultation, a team of Fund economists visits a member country, studies its economy, and holds detailed discussions with officials on exchange rates, budgets, monetary policy, financial regulation, and structural reforms. The team also meets people outside government, such as business groups, labor unions, and civil society. The Fund then produces a report with its assessment and advice. Surveillance comes in two forms. Bilateral surveillance looks at one country at a time, as in an Article IV consultation. Multilateral surveillance looks at the whole system and at how countries affect one another, and it is published in flagship reports such as the World Economic Outlook and the Global Financial Stability Report. The idea behind surveillance is prevention. By spotting dangerous imbalances and risky policies early, the Fund hopes to warn members and steer them toward safer choices before a crisis erupts. In a recent financial year the Fund carried out well over one hundred Article IV consultations, which shows how central this routine work has become (International Monetary Fund, 2024). Surveillance is also the least coercive of the Fund's tools. When it advises a country that has not borrowed money, the Fund can persuade but cannot compel. Its influence rests on the quality of its analysis and on the weight of its opinion in the eyes of markets and governments. Critics sometimes argue that the Fund's advice reflects a particular economic philosophy rather than neutral technical truth, a point taken up in section nine. 6.2 Lending The second function is lending, and it is the one the public knows best. This is the modern form of the original idea of making the Fund's resources temporarily available to members in balance of payments trouble. When a country cannot pay for its imports, service its foreign debts, or defend its currency, it can turn to the Fund for financial support. The Fund draws on its pool of quota resources, supplemented by borrowing arrangements, to provide the foreign currency the country needs. Fund lending almost always comes with #conditionality, meaning a set of policy commitments the borrowing government agrees to carry out in exchange for the money. The stated logic is straightforward. The Fund is lending members' pooled money and must be reasonably sure it will be repaid, and it wants the borrower to fix the underlying problem so that it does not return again and again. Conditions have historically included measures such as reducing budget deficits, tightening monetary policy, reforming exchange rate arrangements, and, in many cases, structural changes to the economy like privatization or trade liberalization. Conditionality is the single most controversial output of the Fund, and a large body of careful research now examines its effects. Studies show that the number and reach of conditions grew substantially from the 1980s onward, moving beyond simple macroeconomic targets into detailed structural reforms (Kentikelenis and Stubbs, 2023). Some scholars argue that heavy conditionality can be counterproductive, because programs loaded with too many demands become hard to carry out and are frequently left unfinished, a pattern one influential study described as being unimplementable by design (Reinsberg, Stubbs and Kentikelenis, 2022). The Fund has repeatedly promised to streamline and soften its conditions, and it now speaks about protecting social spending during adjustment. Yet detailed reviews of recent loans find that austerity remains a core feature of many programs, and that promised protections for the most vulnerable are applied unevenly (Kentikelenis and Stubbs, 2024). This gap between rhetoric and practice is one of the most important open questions about the modern Fund, and section nine returns to it. The Fund runs different lending windows for different situations. General resources are available to all members on standard terms. Separate arrangements, funded by contributions and managed as trusts, offer cheaper loans to low income countries. In 2022 the Fund created the Resilience and Sustainability Trust to help vulnerable members deal with longer term challenges such as climate change and pandemic preparedness, a sign that the institution keeps stretching its mandate to meet new kinds of risk. 6.3 Capacity Development The third function is #capacity_development, sometimes called technical assistance and training. This work is quieter and less controversial than lending, but it reaches deep into the daily running of many governments. Fund staff help countries build the practical skills to manage their economies well: designing tax systems, running central banks, gathering reliable statistics, managing public debt, and supervising banks. The reasoning is that good policy needs capable institutions, and many members, especially poorer ones, lack the trained people and systems to do this work alone. Capacity development connects to the treaty's original purposes in a simple way. The Fund wants members to keep their economies stable and their payments in order, and countries that can manage their own finances competently are less likely to fall into the crises that require rescue lending. In this sense, technical support is preventive, like surveillance. It is also a source of the Fund's soft influence, since the training it provides tends to spread its way of thinking about economic policy. EVOLUTION AND AMENDMENT: HOW THE TREATY SURVIVED THE COLLAPSE OF ITS OWN SYSTEM One of the most remarkable facts about the #Articles_of_Agreement is that the exchange rate system they created, the fixed but adjustable #par_value_system, no longer exists, and yet the institution built to run it is stronger than ever. Understanding how this happened is key to understanding the Fund's staying power. The treaty has been amended seven times, and the two most consequential changes came in response to the greatest crisis in the Fund's history. 7.1 The First Amendment and the Birth of Special Drawing Rights By the 1960s a strange problem had appeared. World trade was growing fast, and countries needed ever more reserves to support it. Most of those reserves were held in United States dollars. But there was a catch, sometimes named the Triffin dilemma after the economist who described it. For the world to get more dollar reserves, the United States had to keep running deficits and sending dollars abroad, yet the more dollars piled up overseas, the more people doubted that the United States could really convert them all into gold at the promised rate. The system's need for liquidity was undermining confidence in its anchor. The Fund's answer was to create a brand new reserve asset that did not depend on any single country's deficits. This was the #Special_Drawing_Rights, introduced through the First Amendment to the treaty, which took effect in 1969. The Special Drawing Rights, usually shortened to SDRs, are not a currency in the ordinary sense. You cannot buy coffee with them. They are an international reserve asset created by the Fund and allocated to members in proportion to their quotas. A member holding SDRs can exchange them for usable currencies with other members. The value of an SDR is set by a basket of major world currencies. Today that basket contains the United States dollar, the euro, the Chinese renminbi, the Japanese yen, and the British pound, with the renminbi added in 2016 in recognition of China's rising weight. SDRs also serve as the Fund's unit of account, which is why quotas and loans are measured in them. 7.2 The Nixon Shock and the Second Amendment The deeper crisis broke in the early 1970s. The strains described above became unbearable. Foreign holdings of dollars far exceeded American gold, and pressure built for the United States to convert dollars into gold faster than it could sustain. On 15 August 1971, President Richard Nixon suspended the convertibility of the dollar into gold. This event, often called the Nixon shock, pulled the anchor out of the whole #Bretton_Woods system. Attempts to patch the fixed rate arrangement followed, but by 1973 the major currencies were floating, meaning their values were set by markets rather than by official pegs. The exchange rate system the treaty had created was effectively dead. This was an existential moment. The Fund had been designed to police a fixed rate system that no longer existed. It could have faded into irrelevance. Instead, the members chose to rewrite the treaty to fit the new reality. The Second Amendment, which took effect in 1978, is arguably the most important change in the Fund's history. It formally accepted floating exchange rates, giving members broad freedom to choose their own currency arrangements rather than forcing them into a single fixed system. It reduced the formal role of gold in the international system. And crucially, it strengthened the Fund's #surveillance role under #Article_IV. Since the Fund could no longer enforce fixed rates, it was given a broad mandate to oversee members' policies and to promote a stable system through monitoring and advice. In effect, the Fund traded a narrow enforcement job for a wider watchdog role, and that role has grown ever since. Later amendments dealt with other matters. One allowed the Fund to suspend the voting rights of a member that fails to meet its obligations, giving the institution a sharper disciplinary tool. Another provided for a special one time allocation of SDRs to correct the fact that many newer members had never received an allocation. The most recent amendment, approved by the Board of Governors in 2010 and brought into force in 2016, reformed the Executive Board as part of a wider package of quota and governance changes meant to give emerging economies a larger voice. Each amendment shows the same pattern: rather than replace the institution, the members repaired and updated its charter. 7.3 The Lesson of Flexibility The larger lesson is about design. The founders of 1944 did not foresee floating exchange rates, the creation of SDRs, or the rise of China. What they did build was a treaty that could be amended and an institution that could reinvent its purpose. When the original system collapsed, the Fund did not collapse with it. This adaptability, more than any single rule, explains why the #Articles_of_Agreement remain in force after eighty years while so many other structures of the 1940s have vanished. A charter that can be rewritten is a charter that can survive. THE FUND IN THE CONTEMPORARY GLOBAL FINANCIAL SYSTEM The Fund of today would be recognizable to its founders in purpose but not in scale or scope. It sits at the center of what is now called the #global_financial_safety_net, the layered set of arrangements that catch countries falling into financial trouble. That net has several layers: a country's own reserves, currency swap lines between central banks, regional funds that pool resources among neighbors, and, at the center, the Fund with its near universal membership and its ability to lend to almost any country. When a crisis is severe or a country has no other backstop, the Fund is the lender of last resort for governments. Several recent developments show the Fund stretching to meet new demands. The most dramatic was the response to the pandemic. In August 2021 the Fund carried out the largest SDR allocation in its history, distributing the equivalent of about six hundred and fifty billion United States dollars to members to boost global reserves during the crisis. Because SDRs are allocated by quota, most of this new liquidity went to wealthy countries that did not urgently need it, while poorer countries received only a small share. This sparked a lively debate about rechanneling, meaning arrangements through which rich countries lend their unused SDRs to help poorer ones. The debate over how to make SDRs work better for the countries that need them most is one of the liveliest in current international finance. A second development is the long running effort to reform quotas. In December 2023 the Board of Governors concluded the Sixteenth General Review of Quotas and agreed to increase total quotas by fifty percent. This was presented as a way to strengthen the Fund's permanent resource base and reduce its reliance on temporary borrowing. Yet the increase was designed to be roughly proportional, which means it did not do much to shift voting shares toward under represented emerging economies. As a result, many observers welcomed the extra resources but criticized the missed chance to modernize the balance of power, and attention has already turned to the next review and to the difficult question of realigning shares to match today's economy (International Monetary Fund, 2023). A third development is the widening of the Fund's concerns beyond traditional macroeconomics. Through instruments like the Resilience and Sustainability Trust, and through the growing attention its surveillance pays to climate change, digital finance, and financial integrity, the Fund has stretched its mandate to cover risks that the 1944 delegates never imagined. Supporters see this as sensible evolution, since these risks can now threaten the very economic and financial stability the Fund exists to protect. Skeptics worry that a body designed for balance of payments problems is spreading itself too thin and drifting beyond its expertise and its legitimate authority. This tension between focus and breadth is likely to define the Fund's next decade. It is also useful to remember that the Fund never worked alone. It was born as one half of a pair, alongside the World Bank, and the two institutions have divided the labor of postwar economic management between them, with the Fund focused on short term stability and payments crises and the Bank focused on longer term development and reconstruction lending. In recent years the two have been drawn closer together again by the return of severe debt problems in many low and middle income countries. When a government owes more than it can repay, orderly relief usually requires the Fund, the Bank, national creditors, and increasingly private bondholders to coordinate, and the Fund often plays the role of honest broker by judging what level of debt a country can realistically carry. This debt work sits squarely within the treaty's original concern with balance of payments trouble, even though the creditors and instruments involved look very different from those of 1944. What has not changed is the Fund's central position. Whatever its critics say, when a major financial crisis hits, governments and markets still look to the Fund. Its combination of money, near universal membership, and analytical firepower gives it a role that no regional body or private lender can fully replace. The safety net has many threads, but the Fund is the knot that holds them together. CRITICISMS, CONTROVERSIES, AND REFORM DEBATES An honest account of the #Articles_of_Agreement and the institution they created must give serious space to criticism, because the Fund is one of the most contested organizations in the world. The debates cluster around three themes: the effects of its lending conditions, the fairness of its governance, and the philosophy behind its advice. 9.1 Conditionality and Austerity The sharpest criticism concerns #conditionality and, in particular, #austerity. Austerity means cutting government spending and raising revenue to shrink budget deficits, often quickly and deeply. For decades, the policy packages attached to Fund loans have leaned heavily on austerity, along with measures like privatizing state enterprises, liberalizing trade, and deregulating markets. Critics, especially those focused on the #Global_South, argue that these programs have often harmed the very people they were supposed to help by forcing cuts to health, education, and social protection at the worst possible moment, when economies were already in crisis. The strength of the modern debate is that it now rests on large data sets rather than only on anecdotes. Detailed research examining thousands of loan documents over four decades finds that, despite reform promises, Fund programs continue to place a heavy burden of adjustment on borrowing countries, and that this burden is linked to reduced social spending, higher income inequality, and worse health outcomes (Kentikelenis and Stubbs, 2023). A close look at loans signed during the recent period found that most borrowers experienced a shrinking of their fiscal space over the life of their programs, and that the social spending protections the Fund advertises were applied inconsistently and lacked a clear common definition (Kentikelenis and Stubbs, 2024). Other work shows that overloaded programs are frequently left incomplete, raising doubts about whether such demanding conditionality even achieves its own goals (Reinsberg, Stubbs and Kentikelenis, 2022). The Fund's defenders respond that countries usually come to it only when they are already in deep trouble, that painful adjustment is often unavoidable, and that the alternative to a Fund program may be a disorderly default that hurts people even more. They also point out that the Fund has changed, giving more attention to protecting the poor and to countercyclical spending during downturns. The evidence suggests that both sides capture part of the truth. The Fund has genuinely shifted its language and some of its practice, yet the core pattern of demanding, front loaded adjustment has proven remarkably persistent. This is why the phrase promise versus performance appears so often in recent studies. 9.2 Legitimacy and the Balance of Power The second cluster of criticism concerns #legitimacy, which brings us back to the governance structure examined in section five. Because #voting_power is tied to quotas, and because quota reviews lag behind economic change, the Fund's decision making has long been dominated by a handful of advanced economies, with the United States holding an effective veto over the biggest decisions. Meanwhile, large and fast growing economies remain under represented, and the poorest members have very little say at all. The convention that the Managing Director is always European adds to the sense that the top of the institution is reserved for a privileged group. For critics, this imbalance undermines the Fund's claim to act on behalf of the whole membership. How can an institution credibly tell developing countries to reform their governance when its own governance is decades out of date? Recent research reinforces the concern by showing that the preferences of the most powerful members, shaped by their trade and strategic interests, feed into the conditions attached to loans for countries in the #Global_South (Forster, Honig and Kentikelenis, 2025). This connects the two debates: unequal power at the top may help explain the persistence of demanding conditionality at the bottom. Reform proposals abound. They include realigning quotas to match today's economy, boosting the equal basic votes that favor smaller members, opening the selection of the Managing Director to candidates of any nationality, and strengthening the voice of borrowing countries on the Executive Board. Progress has been slow, because meaningful reform requires powerful states to give up influence, and the same weighted voting rules that critics want to change are the rules that protect the status quo. The fifty percent quota increase of 2023 delivered more money without much shift in power, which many saw as a telling illustration of how hard genuine reform is. 9.3 The Question of Economic Philosophy A third, deeper criticism is that the Fund's advice reflects a particular economic worldview rather than neutral technical truth. For much of the late twentieth century, the Fund was closely associated with a set of market oriented ideas sometimes labeled the Washington Consensus, emphasizing balanced budgets, free trade, open capital markets, and a smaller role for the state. Critics argue that this philosophy was applied too uniformly, without enough regard for the different circumstances of different countries, and that it sometimes ignored the human costs. In fairness, the Fund's thinking has evolved. It has become more open about the risks of unregulated capital flows, more willing to consider controls in some situations, and more attentive to inequality and social outcomes than it once was. Its research arm regularly publishes work that questions old certainties. Yet skeptics contend that the underlying orientation has shifted less than the language suggests, and that the institution still tends to favor a certain style of reform. This debate matters because the Fund's advice, backed by its money and its prestige, shapes the lives of hundreds of millions of people. When the Fund is right, it can steer countries away from disaster. When it is wrong, or when it applies a general formula to a situation that needs something different, the consequences can be severe. Taken together, these three debates do not amount to a simple verdict. The Fund is neither the neutral technical body its defenders sometimes describe nor the villain its harshest critics portray. It is a powerful, imperfect institution, shaped by the states that control it, doing genuinely useful work in crisis management and economic monitoring while carrying real and well documented flaws in its lending practices and its distribution of power. A mature understanding holds both of these truths at once. CONCLUSION The #Articles_of_Agreement of the #International_Monetary_Fund were written by people who had watched the world economy fall apart and had seen where that collapse led. Their central insight was simple and profound: international money is a shared problem, and shared problems need cooperative solutions and a permanent institution to manage them. Out of the contest between the #Keynes plan and the #Harry_Dexter_White plan came a treaty that built a quota funded pool, a fixed but adjustable exchange rate system, a set of rules on currency practices, and a governing structure based on weighted voting. The document opened for business a system meant to replace the beggar thy neighbor chaos of the 1930s with orderly and rule based cooperation among nations. The most impressive thing about the treaty is not that it got everything right. It did not. The exchange rate system it created lasted less than thirty years before collapsing under strains its founders only partly foresaw. What made the #Articles_of_Agreement a lasting success was their flexibility. When the par value system died in the 1970s, the members did not abandon the Fund. They rewrote its charter, invented the #Special_Drawing_Rights, and reoriented the institution around surveillance, lending, and capacity development. A treaty that could be amended became an institution that could survive. Eight decades on, the Fund sits at the center of the #global_financial_safety_net, still doing the core work its founders imagined: watching the world economy, lending to countries in trouble, and helping members manage their finances. It also carries forward the tensions built into it from the start. Its weighted voting attracted the powerful and still favors them, fueling deep debates about #legitimacy and #quota_reform. Its lending conditions provide vital rescue money but have imposed heavy and unevenly shared costs, fueling debates about #conditionality and #austerity that recent research has sharpened rather than settled. These are not signs that the institution has failed. They are the ordinary strains of a body that must serve nearly two hundred members with very different interests and very unequal power. For students of economics, law, and international relations, the #Articles_of_Agreement repay close study precisely because they sit at the meeting point of ideals and power. They show how good ideas about cooperation get filtered through the interests of strong states, how institutions adapt to survive shocks, and how the rules written in one era keep shaping outcomes in the next. Whether the Fund can reform itself enough to match the economic map of the twenty first century, especially by giving a fairer voice to the #Global_South, is one of the central questions of #global_economic_governance today. The answer will help decide whether the vision born at #Bretton_Woods in 1944 remains a living framework for cooperation or slowly hardens into a relic of a world that no longer exists. HASHTAGS #International_Monetary_Fund #IMF #Bretton_Woods_1944 #Articles_of_Agreement #global_financial_system #monetary_cooperation #exchange_rate_stability #Special_Drawing_Rights #IMF_quotas #IMF_conditionality #global_economic_governance #balance_of_payments #IMF_surveillance #international_finance #economics_students REFERENCES Forster, T., Honig, D., and Kentikelenis, A. (2025). Formal governance matters: when, how, and why states act on the IMF Executive Board. Review of International Political Economy, 32(3), 818 to 846. https://doi.org/10.1080/09692290.2024.2441136 International Monetary Fund. (2020). Articles of Agreement of the International Monetary Fund. Washington, DC: International Monetary Fund. International Monetary Fund. (2022). IMF Quotas (Factsheet). Washington, DC: International Monetary Fund. International Monetary Fund. (2023). 2021 Special Drawing Rights Allocation, Ex Post Assessment Report (IMF Policy Paper). Washington, DC: International Monetary Fund. International Monetary Fund. (2023). Sixteenth General Review of Quotas, Report of the Executive Board to the Board of Governors (IMF Policy Paper). Washington, DC: International Monetary Fund. International Monetary Fund. (2024). IMF Annual Report 2024. Washington, DC: International Monetary Fund. International Monetary Fund. (2025). The Global Financial Safety Net, A Review (IMF Policy Paper). Washington, DC: International Monetary Fund. Kentikelenis, A. E., and Stubbs, T. H. (2023). A Thousand Cuts: Social Protection in the Age of Austerity. New York: Oxford University Press. https://doi.org/10.1093/oso/9780190637736.001.0001 Kentikelenis, A. E., and Stubbs, T. H. (2024). Social protection and the International Monetary Fund: promise versus performance. Globalization and Health, 20(1), Article 41. https://doi.org/10.1186/s12992-024-01045-9 Reinsberg, B., Stubbs, T., and Kentikelenis, A. (2022). Unimplementable by design? Understanding (non) compliance with International Monetary Fund policy conditionality. Governance, 35(3), 689 to 715. https://doi.org/10.1111/gove.12605 Turan, T., Evrensel, A. Y., and Yanikkaya, H. (2023). Economic growth effects of IMF programs. International Conference on Eurasian Economies, 332 to 339. https://doi.org/10.36880/c15.02756

The Worst Forms of Child Labour Convention, adopted by the International Labour Organization in 1999 as Convention No. 182, is one of the most widely accepted human rights instruments in the world and the first ILO Convention to reach universal ratification. It commits every ratifying State to take immediate and time-bound measures to prohibit and eliminate the most severe forms of child exploitation, including slavery and slavery-like practices, the commercial sexual exploitation of children, the use of children in illicit activities such as drug trafficking, and hazardous work. This article reviews the legal architecture of the Convention, explains how each of the four categories of worst forms is defined and understood in international law, and examines how the instrument has been implemented in practice more than two decades after its adoption. Drawing on recent scholarship and the latest global estimates, the study argues that legal universality has not translated into practical elimination. In 2024, roughly 138 million children remained in child labour, and the 2025 target for ending child labour was missed. The article identifies the main barriers to progress, including poverty, the informal economy, weak enforcement, armed conflict, data gaps, and the hidden nature of exploitation in global supply chains. It concludes that Convention No. 182 remains a powerful normative anchor, but that its promise depends on sustained investment in social protection, education, labour inspection, and international cooperation. The analysis is written for students and early researchers who need a clear, evidence-based account of the Convention and the debates surrounding it. Keywords: worst forms of child labour, ILO Convention No. 182, child protection, human rights, forced labour, child trafficking, Sustainable Development Goal 8.7 Introduction #child_labour is one of the oldest and most persistent problems in the history of work. For most of human history, children laboured in fields, workshops, and homes as a normal part of family survival. The idea that childhood should be protected from economic exploitation is comparatively recent, and the idea that the international community should agree on binding rules to defend that protection is more recent still. The Worst Forms of Child Labour Convention, known formally as #Convention_182 and adopted by the International Labour Organization in June 1999, represents a high point in that long effort. It is the instrument through which the world agreed that some forms of child work are so damaging that they must be stopped at once, without waiting for economies to grow or for poverty to disappear. This article examines that Convention in depth. It asks three questions. First, what exactly does #Convention_182 require, and how does it define the #worst_forms of child labour it seeks to abolish. Second, how has the Convention been put into practice in the years since 1999, and what has been achieved. Third, why, despite universal legal acceptance, do the worst forms of child labour persist on a large scale. These questions matter because the gap between law and reality in this field is unusually wide. Convention No. 182 has been ratified by every member State of the #ILO, a level of agreement almost unheard of in international law, and yet tens of millions of children continue to work in conditions the Convention was designed to end. The distinction that sits at the centre of the Convention is the difference between child labour in general and the worst forms of it. Not all work done by children is harmful. A teenager who helps in a family shop after school, or who does light chores at home, is not necessarily a victim of exploitation. International standards recognise this by drawing a line based on age, hours, and the nature of the task. What Convention No. 182 targets is the sharp end of the problem, the situations in which children are enslaved, trafficked, sold for sex, used to carry drugs or weapons, or forced into work that destroys their health. These are not grey areas. They are, in the language of the Convention, forms of exploitation that must be prohibited and eliminated as a matter of #urgency. The Convention was born from a sense that earlier standards, while important, had not moved fast enough. The ILO had adopted the Minimum Age Convention, #Convention_138, in 1973, setting general rules about how old a child must be before entering different kinds of employment. That instrument aimed at the long-term goal of abolishing all child labour. But by the 1990s it was clear that a more focused and immediate response was needed for the most extreme cases. Rather than asking States to work gradually toward a distant goal, Convention No. 182 asked them to act at once against the worst abuses, while keeping the wider objective of ending all child labour in view. This combination of immediacy and focus helps explain why the Convention was ratified so quickly. For students, the Convention is a useful case study in how international law tries to change behaviour. It shows the strengths of a clear, morally compelling standard that almost every government can accept. It also shows the limits of law when the underlying causes of a problem, especially #poverty and inequality, remain unaddressed. Understanding both sides of this picture is essential for anyone who wants to work in human rights, development, labour policy, or child protection. The article proceeds as follows. It first sets out the historical and normative background of international efforts against child labour. It then explains the method used to review the evidence. The main body analyses the legal structure of the Convention, defines each category of worst forms, and describes the mechanisms for implementation and monitoring. Later sections present the latest global data, discuss the persistent barriers to progress, offer regional perspectives, and connect the Convention to the Sustainable Development Goals. The article closes with a discussion of what the evidence means and a set of recommendations for strengthening the response. Historical and Normative Background The regulation of children's work has deep roots in the industrial age. During the industrial revolution in Europe, young children worked long hours in factories, mines, and mills, often in dangerous conditions and for very low pay. Public concern about their suffering led to some of the earliest labour laws, which limited working hours and set minimum ages for certain jobs. These national reforms established an important principle, that the state has a legitimate interest in protecting children from economic exploitation, even against the wishes of employers or, at times, families. When the ILO was founded in 1919, child labour was among the first issues it addressed. Over the following decades, the organisation adopted a series of minimum age conventions covering specific sectors such as industry, agriculture, and maritime work. These instruments were gradually consolidated into the Minimum Age Convention of 1973, which brought together the earlier standards into a single framework. #Convention_138 set the general minimum age for admission to employment, normally fifteen years, with a higher age of eighteen for hazardous work and provisions allowing developing countries some flexibility. Its ambition was broad, the eventual abolition of all child labour. Alongside this labour-standards tradition, a separate strand of international law developed around the rights of the child. The most important expression of this is the United Nations Convention on the Rights of the Child, adopted in 1989 and now the most widely ratified human rights treaty in the world. It recognises the child's right to protection from economic exploitation and from work that is hazardous or harmful to development. Other instruments reinforced particular aspects of the problem. The Palermo Protocol against trafficking in persons, adopted in 2000, addressed the trade in human beings, including children, while an optional protocol to the Convention on the Rights of the Child addressed the involvement of children in armed conflict. Recent legal scholarship stresses that these instruments together create the framework within which national action against child exploitation takes place, and that #Convention_182 is best understood as part of this wider web rather than in isolation. By the 1990s, a consensus had formed that the existing standards, though valuable, were not producing fast enough change for the most severe cases. Research commissioned by the ILO documented the scale and brutality of the worst abuses, from bonded labour to the sexual exploitation of children. The response was to create a new instrument that would not replace the minimum age standard but would sit beside it, sharpening the focus on the situations that demanded immediate action. The result was Convention No. 182, adopted at the International Labour Conference in June 1999 and accompanied by the Worst Forms of Child Labour Recommendation, #Recommendation_190, which provides detailed guidance on how the Convention should be applied. The speed with which the new Convention was accepted was remarkable. It entered into force in November 2000, and by 2020 every member State of the ILO had ratified it, making it the first ILO Convention to achieve #universal_ratification and one of the most rapidly ratified international agreements ever concluded. This near-unanimous acceptance reflects a genuine moral consensus. Very few governments are willing to defend the enslavement of children or their use in prostitution. The difficulty, as later sections show, lies not in agreeing that these practices are wrong but in ending them in practice. It is worth pausing on why Convention No. 182 succeeded where broader efforts had struggled. Part of the answer is its framing. By singling out the worst forms rather than all #child_labour, the Convention avoided a debate that had long divided rich and poor countries. Governments that could not realistically end all child work overnight, given the poverty of their populations, could still commit without embarrassment to ending slavery, trafficking, and the sexual exploitation of children. The Convention thus found common ground that a more sweeping instrument could not. Another part of the answer is that it aligned with a growing public awareness, spread by media coverage and civil society campaigns during the 1990s, of the trafficking and abuse of children across borders. The Convention gave that concern a legal form. This history carries a lesson for students of international law, that carefully chosen scope, matched to what States are actually willing to do, can achieve broad agreement that a more ambitious text would forfeit. Method and Approach This article is a structured review of legal texts, scholarly literature, and authoritative statistical reports concerning Convention No. 182 and the worst forms of child labour. It does not present new empirical fieldwork. Instead, it synthesises existing knowledge to give students and early researchers a clear and reliable account of the Convention and the debates around it. Three kinds of source were used. The first is the primary legal material, principally the text of Convention No. 182 itself, together with Recommendation No. 190 and related instruments such as Convention No. 138 and the Convention on the Rights of the Child. These define what the law actually requires. The second is recent academic scholarship, drawn from peer-reviewed journals in law, development studies, criminology, and public health, with a focus on work published within the last five years so that the analysis reflects current thinking. The third is statistical and policy material produced by international organisations, especially the joint global estimates of child labour prepared by the ILO and the United Nations Children's Fund, which provide the most authoritative numbers available. Sources were selected for relevance, credibility, and recency. Priority was given to studies that address the specific categories of worst forms covered by the Convention, namely slavery-like practices, sexual exploitation, illicit activities, and hazardous work, and to material that speaks to implementation and enforcement rather than to child labour in the abstract. Where figures are cited, the year and origin of the estimate are stated, because child labour data change over time and are sensitive to definitions and methods. The analysis is qualitative and interpretive. It reads the evidence critically, noting areas of agreement, points of dispute, and gaps that future research should address. A limitation of this approach should be acknowledged. #Child_labour, and especially its worst forms, is by nature hidden. Much of it takes place in the informal economy, in private homes, in conflict zones, or in criminal settings where measurement is difficult and dangerous. Global estimates are careful and well-constructed, but they rely on surveys and modelling and cannot capture the full reality. This uncertainty should be kept in mind throughout, and it is itself an important finding, because a problem that cannot be seen clearly is hard to eliminate. The Legal Architecture of Convention No. 182 Convention No. 182 is short, but each of its provisions carries weight. Understanding the structure of the instrument is the key to understanding what ratifying States have actually promised to do. The Convention opens by defining the child. For its purposes, a child is any person under the age of eighteen. This is a deliberate and important choice. Whereas the minimum age for ordinary employment can be lower, the protections against the #worst_forms apply to everyone below eighteen without exception. A seventeen-year-old sold into prostitution or forced to carry weapons is as much a victim under this Convention as a much younger child. This uniform age threshold removes the temptation to treat older adolescents as adults for the purpose of the most serious abuses. The heart of the Convention is its definition of the worst forms, set out in Article 3. It identifies four categories. The first covers all forms of #child_slavery and practices similar to slavery, including the sale and #child_trafficking of children, #debt_bondage and serfdom, and forced or compulsory labour, with the forced or compulsory recruitment of children for use in armed conflict expressly included. The second covers the use, procuring, or offering of a child for prostitution, for the production of pornography, or for pornographic performances. The third covers the use, procuring, or offering of a child for illicit activities, in particular for the production and #drug_trafficking of drugs as defined in the relevant international treaties. The fourth covers work which, by its nature or the circumstances in which it is carried out, is likely to harm the health, safety, or morals of children, commonly called #hazardous_work. A useful distinction runs through these categories. The first three are often described as the unconditional worst forms. They are wrong in themselves, regardless of context, and no country may permit them. The fourth, hazardous work, is different in that it depends on national determination. Because the tasks and conditions that are dangerous vary from one setting to another, the Convention leaves it to each State, in consultation with employers and workers, to draw up a list of the types of work considered hazardous within its territory. Recommendation No. 190 offers guidance on what such lists should include, for example work that exposes children to physical, psychological, or sexual abuse, work underground or under water, work with dangerous machinery, and work in unhealthy environments. The accompanying #Recommendation_190 does more than list hazards. It advises States on how to give effect to the Convention as a whole, and it repays close reading because it turns broad obligations into practical steps. It recommends that governments compile detailed information and statistics on the worst forms, that they establish or designate mechanisms to monitor the Convention's application, and that certain of the worst forms be treated as criminal offences. It suggests measures to reach children who are especially exposed, including very young children, girls, and children in hidden work situations such as domestic service in private households. It also stresses coordination, urging States to bring together the many agencies whose cooperation is needed, from labour ministries and police to schools and social services. In short, the Recommendation acknowledges that a prohibition is only as good as the machinery built to apply it, and it sketches what that machinery should look like. Although a recommendation is not binding in the way a convention is, it carries real weight as an authoritative guide to good practice, and supervisory bodies draw on it when they assess how well a country is meeting its obligations. The remaining provisions turn from definition to action. States are required to take immediate and effective measures to secure the prohibition and elimination of the worst forms as a matter of urgency. They must design and implement programmes of action to eliminate these forms as a priority, in consultation with relevant institutions and social partners. They must put in place appropriate mechanisms to monitor the application of the Convention. Crucially, the Convention does not treat the problem only as a matter of #enforcement. It requires States to take account of the importance of education and to provide the necessary and appropriate direct assistance for the removal of children from the worst forms, together with their rehabilitation and social integration. It calls for free basic education and, where possible and appropriate, vocational training for children removed from these forms of work. It gives particular attention to the situation of girls, who face specific risks, and to children at special risk. The Convention also recognises that penalties matter. It requires States to take steps to ensure the effective implementation and enforcement of its provisions, including the provision and application of penal sanctions or, where appropriate, other sanctions. And because many of the worst forms cross borders, it calls for enhanced #international_cooperation and assistance, including support for social and economic development, poverty eradication programmes, and universal education. In this way the Convention combines prohibition with prevention and rehabilitation, treating the child not only as someone to be rescued but as someone whose future must be secured. Read as a whole, the legal architecture of Convention No. 182 is coherent and demanding. It is not merely a statement of principle. It obliges States to legislate, to enforce, to monitor, to rehabilitate, and to cooperate. The measure of its success, therefore, is not ratification but the extent to which these obligations are met in practice. Defining and Understanding the Worst Forms The four categories in Article 3 are not abstract. Each describes real and widespread patterns of harm. This section examines them in turn, drawing on recent research to show what the worst forms look like in practice and why they are so difficult to end. 5.1 Slavery, Trafficking, and Slavery-like Practices The first category is the gravest. It covers situations in which a child is owned, sold, or controlled as if they were property. Modern slavery rarely takes the open form of historical chattel slavery, but its substance survives in #debt_bondage, in which a child works to repay a debt, often one incurred by a parent, under conditions that make repayment impossible. It survives in the sale and #child_trafficking of children, in which children are moved within or across borders and handed over for exploitation. And it survives in forced labour, in which children are made to work through threats, deception, or violence. Trafficking is central to this category. Research on child trafficking describes it as one of the most profitable forms of organised crime, feeding both labour and sexual exploitation, and stresses how difficult it is to detect because victims are hidden, frightened, and often lack documents that would prove their age or identity. Studies of trafficking in specific national contexts show that poverty, weak birth registration, and gaps between the law on paper and its enforcement combine to leave children exposed. They also show that effective responses require more than criminal law. They demand prevention, protection of survivors, and cooperation between agencies and across borders. A distinct and especially disturbing sub-category is the recruitment of children for armed conflict. The Convention expressly includes the forced or compulsory recruitment of children for use in armed conflict within its definition of slavery-like practices. So-called #child_soldiers are used not only as fighters but as scouts, porters, cooks, messengers, and, in the case of many girls, as targets of sexual violence. Recent empirical work has tried to estimate how many children live in areas where recruitment by state or non-state armed groups is a real risk, and finds that the danger is concentrated in particular conflict zones and shaped by local conditions. This research emphasises a situational view. The risk to a child is not the same everywhere but depends on the presence of armed groups, the intensity of conflict, and the breakdown of the social structures that would normally protect them. Verified cases of recruitment number in the tens of thousands over recent years, and the true figure is believed to be far higher, because conflict settings are exactly where monitoring is weakest. A quieter but widespread slavery-like practice is child domestic servitude. Children, most often girls, are placed in the homes of others to cook, clean, and care for younger children, sometimes far from their own families. In its milder forms this may resemble a family arrangement, but at its worst it becomes servitude, in which the child works long hours for no pay, cannot leave, is cut off from schooling, and is exposed to physical and sexual abuse behind closed doors. Because it happens in private homes, it is one of the hardest forms of exploitation to detect and one of the least reached by inspection. Its invisibility is precisely what makes it dangerous, and it illustrates why the Convention and its Recommendation give special attention to hidden work situations. The persistence of slavery-like practices reflects the fact that they are driven by profit and enabled by impunity. Where children can be exploited cheaply and where those who exploit them face little risk of punishment, the practices continue. Ending them therefore requires attacking both the demand that makes them profitable and the #impunity that makes them safe. 5.2 Sexual Exploitation of Children The second category covers the use of children for prostitution, for the production of pornography, and for pornographic performances. This is the #commercial_sexual_exploitation of children, and it is among the most severe violations a child can suffer. Research in public health and criminology documents the lasting damage it causes, including trauma, physical injury, sexually transmitted infections, and long-term harm to mental health and social development. Scholarly reviews of services for children who have been commercially sexually exploited highlight several themes. Exploited children are frequently misidentified, treated as offenders rather than victims, or overlooked entirely by systems that are not designed to recognise them. Effective help requires trauma-informed care, coordination between health, social, and justice services, and long-term support rather than one-off intervention. Longitudinal work following young people who have experienced this exploitation finds that its effects can persist well into adulthood, shaping health and life outcomes for years. These findings reinforce a point the Convention itself makes, that rescue is only the beginning, and that #rehabilitation and reintegration must follow. The character of this exploitation has changed with technology. The internet has created new channels for the production and distribution of abuse material and for the grooming and exploitation of children online. Recent policy discussions place online exploitation alongside the traditional forms as an emerging challenge that legal frameworks are still adapting to meet. This does not change the underlying wrong, but it makes detection and enforcement harder, because exploitation can now cross borders instantly and hide in encrypted and anonymous spaces. Girls remain disproportionately affected by commercial sexual exploitation, a pattern that runs through the research and that gives particular urgency to the Convention's attention to the situation of girls. 5.3 Use of Children in Illicit Activities The third category covers the use, procuring, or offering of a child for illicit activities, and in particular for the production and trafficking of drugs. This is a form of exploitation in which criminal organisations use children as instruments of crime. Children may be used to carry, sell, or produce drugs, to act as lookouts, or to perform other tasks in criminal enterprises, precisely because they are less likely to attract suspicion and, if caught, may face lighter penalties than adults. This category illustrates a difficult tension in the response to the worst forms. A child used to transport drugs is, under Convention No. 182, a victim of exploitation. Yet the same child may be treated by the criminal justice system as an offender. The Convention's logic is that the child's involvement is itself a form of exploitation to be prevented and remedied, not simply a crime to be punished. Handling these cases well requires justice systems that can see the child behind the offence and that respond with protection and support rather than only with punishment. In regions where organised crime is powerful and where poverty pushes children toward whatever income is available, breaking the link between children and criminal enterprises is a serious and continuing challenge. 5.4 Hazardous Work The fourth category, #hazardous_work, is the largest in numerical terms and in some ways the most complex. Unlike the first three, it is defined nationally rather than universally, because what counts as dangerous depends on the task, the tools, the environment, and the age and condition of the child. Hazardous work includes labour that exposes children to toxic substances, dangerous machinery, extreme heat or cold, heavy loads, long hours, or unhealthy conditions. The single largest setting for hazardous child labour is #agriculture. Around seven in ten children in child labour work in agriculture, on family farms and plantations, where they may handle pesticides, use sharp tools, and carry loads far beyond their strength. Mining, especially small-scale and artisanal mining, is among the most dangerous sectors, exposing children to collapse, dust, chemicals such as mercury, and exhausting physical labour. Manufacturing, construction, and domestic work also carry serious risks. In every case the harm is not only immediate injury but long-term damage to growth, health, and schooling. The harm of hazardous work is easy to underestimate because much of it looks ordinary. A child spraying pesticide without protection, or spending hours bent over in a field, or breathing dust in a small workshop, may not appear to be in obvious danger at any single moment. The damage accumulates. Exposure to chemicals can impair development, repetitive strain and heavy loads can deform growing bodies, and the long hours steal the time and energy that should go to school. Because children are still growing, they absorb toxins more readily than adults and are more easily injured, so the same task that is merely tiring for an adult can cause lasting harm to a child. This is the reasoning behind the Convention's firm rule that hazardous work is off-limits for everyone under eighteen, even though the general minimum age for ordinary employment is lower. Much hazardous child labour is now understood to be embedded in global #supply_chains. Studies of sectors such as cocoa, coffee, cotton, tobacco, garments, and electronics find that the risk of child labour is greatest at the lower tiers of production, in the fields and small workshops where raw materials are grown and first processed, far from the visibility of the brands that ultimately sell the finished goods. Investigations of these supply chains repeatedly find that a zero-tolerance policy on paper is not enough, because the exploitation is hidden below the surface and beyond the reach of ordinary monitoring. This has led, in several jurisdictions, to new laws requiring large companies to carry out human rights due diligence across their supply chains, a development that connects the Convention to contemporary debates about corporate responsibility. The academic literature on child labour in supply chains argues that responsibility is shared, and that firms which source from high-risk regions cannot treat exploitation as someone else's problem. Implementation, Ratification, and Monitoring The near-universal ratification of Convention No. 182 is a genuine achievement, but ratification is a promise, not a result. This section examines how the promise is meant to be kept and how compliance is monitored. Once a State ratifies, it is obliged to bring its law and practice into line with the Convention. In practice this means several things. It means enacting or strengthening legislation that criminalises the unconditional worst forms and that prohibits hazardous work for children under eighteen. It means drawing up, in consultation with employers and workers, a national list of the types of work considered hazardous. It means building the institutional capacity to enforce these rules, including trained labour inspectors, police, prosecutors, and child protection services. And it means putting in place programmes to withdraw children from the worst forms, to rehabilitate them, and to ensure they can go to school. The ILO supports these efforts through technical assistance. Its programme on the elimination of child labour, established in 1992 and now operating in an expanded form, helps countries build the systems the Convention requires. One tool developed for this purpose is the time-bound programme, in which a country commits to eliminating the worst forms within a defined period through a coordinated package of measures. These programmes bring together law reform, enforcement, education, and social protection in a single national strategy, on the understanding that no single measure is enough on its own. Monitoring compliance is the task of the ILO's regular supervisory system. States that have ratified must report periodically on how they are applying the Convention. A committee of independent legal experts examines these reports, together with information from workers' and employers' organisations, and issues observations and requests where it finds shortcomings. This process does not carry the power to impose sanctions, but it creates a continuing, public record of each country's performance and applies steady pressure to close the gap between commitment and practice. Legal scholars note that this supervisory dialogue, though sometimes slow, is one of the Convention's most valuable features, because it keeps the obligation alive long after the moment of ratification. Yet the same scholars caution that the impact of the Convention on the ground is not straightforward. Universal ratification tells us that governments accept the norm. It does not tell us that children are protected. In many countries the law now prohibits the worst forms, but enforcement is weak, resources are scarce, and the underlying pressures that push children into exploitation remain strong. The distance between the legal and the real is the central problem of implementation, and it is to the evidence of that distance that the article now turns. The Global Picture: Statistics and Trends The most reliable numbers on child labour come from the joint global estimates produced by the ILO and the United Nations Children's Fund. These estimates, updated periodically, are built from national surveys and careful modelling, and they provide the clearest available picture of the scale of the problem. The long-term trend has two faces. Over the first two decades of this century, #child_labour fell substantially. From around 246 million children at the start of the century, the number declined as economies grew, schooling expanded, and legal protections spread. This decline was real and significant. But progress then stalled, and between 2016 and 2020 the number of children in child labour rose for the first time in decades, reaching about 160 million in 2020, of whom roughly 79 million were in #hazardous_work. The COVID-19 pandemic, with its shocks to household income and its disruption of schooling, threatened to push millions more children into work. The most recent estimates, reflecting the situation in 2024, show that progress has resumed. Around 138 million children were in child labour, some 59 million girls and 78 million boys, accounting for almost eight percent of all children. This represents more than twenty million fewer child labourers than in 2020, a welcome sign that the reversal of the late 2010s has been halted and turned around. Agriculture remains the dominant sector, and a large share of the total continues to be in hazardous work. Recent analyses stress that these figures come from surveys conducted over several years and adjusted to a common reference point, and that they should be read with care, because the hidden nature of the worst forms means the true numbers may be higher. Despite the renewed decline, the headline finding of the recent estimates is one of failure against a target. The international community, through Sustainable Development Goal Target 8.7, had committed to ending child labour in all its forms by 2025. That target has not been met. Analysts calculate that reaching elimination even by 2030 would require the pace of decline to accelerate many times over, and that on current trends the goal will remain out of reach for decades. This sobering assessment frames the debate about the Convention. The norm is universal, and the trend is again moving in the right direction, but the speed of change falls far short of what was promised, and the worst forms in particular remain stubbornly resistant because they thrive in the settings hardest to reach. Persistent Challenges Why, more than two decades after a Convention accepted by every government, do the worst forms of child labour persist. The answer lies in a set of interlocking challenges that law alone cannot solve. The deepest of these is #poverty. Families send children to work primarily because they need the income or the labour to survive. Where adult wages are too low, where social protection is absent, and where a shock such as illness, drought, or economic crisis strikes, a child's earnings can become the difference between eating and going hungry. Research on the political economy of child labour consistently finds that poverty, and the vulnerability that comes with it, is the strongest driver, and that measures which raise household incomes and cushion shocks are among the most effective responses. Evidence on interventions points in the same direction, with cash transfer programmes standing out as one of the more reliable ways to reduce child labour, precisely because they ease the economic pressure that pushes children into work. A second challenge is the #informal_economy. Most child labour, and most of its worst forms, takes place outside the reach of formal regulation, on small farms, in home-based workshops, in private households, and in illegal enterprises. Labour inspectors, whose job is to enforce workplace standards, rarely have the mandate, the numbers, or the access to reach these settings. A child working in a stranger's home as a domestic worker, or on a remote farm, or in a criminal operation, is almost invisible to the systems designed to protect them. Recent work extending worker rights and protections to informal and younger workers suggests that closing this gap can make a real difference, but doing so is difficult and slow. A third challenge is weak #enforcement and the gap between law and practice. Ratifying a Convention and passing a law is one thing. Building the police capacity, the prosecutorial will, the specialised courts, and the child protection services needed to enforce it is another, and it requires sustained resources that many States lack. Scholarship on the Convention's application in specific regions finds that the instrument's real influence depends heavily on national capacity and political commitment, and that where these are weak, universal ratification changes little on the ground. A fourth challenge is #armed_conflict and instability. Conflict destroys the institutions that protect children, displaces families, closes schools, and creates conditions in which recruitment, trafficking, and forced labour flourish. Children in conflict zones are exposed to the most severe worst forms and are the hardest to reach with any form of protection. As conflicts multiply, this challenge grows rather than recedes. A fifth challenge concerns #supply_chains and the global economy. As earlier sections noted, much hazardous and forced child labour is hidden in the deeper tiers of the supply chains that produce everyday consumer goods. The distance between the child in the field and the shopper in the store makes the exploitation easy to ignore and hard to trace. New due diligence and import laws attempt to bridge this distance by placing responsibility on companies, but their effectiveness is still being tested, and enforcement remains uneven. A further challenge is social and cultural. In some settings, sending children to work is not seen as exploitation but as a normal part of upbringing, a way of teaching skills, contributing to the family, and preparing for adult life. Where this view is strong, laws that prohibit child labour can feel remote or even unfair, and enforcement can meet quiet resistance. This does not justify the worst forms, which cause harm no tradition can excuse, but it does mean that lasting change depends on persuasion and on offering families a better alternative, not on prohibition alone. Programmes that work with communities, that explain the long-term costs of lost schooling, and that make education genuinely worthwhile tend to succeed where top-down bans fail. Recent commentary on the Convention stresses this point, arguing that standards are applied best when they take account of the circumstances in which children live and when the children themselves are heard. A final challenge is the problem of #data and visibility. Because the worst forms are hidden, illegal, and often bound up with crime and conflict, they are extremely hard to measure. Global estimates are careful, but they acknowledge significant gaps, especially for the categories that are most severe. A problem that cannot be counted accurately is a problem that is hard to target, to fund, and to hold governments accountable for. Improving data, while protecting the children involved, is therefore not a technical footnote but a central part of the response. Regional Perspectives Child labour is a global problem, but it is not evenly distributed, and the shape it takes differs from region to region. A brief regional view helps to ground the analysis. Sub-Saharan Africa carries the heaviest burden, both in the number of children in child labour and in the share of children affected. The drivers there are familiar but intense, including widespread poverty, large agricultural economies, rapid population growth, gaps in schooling, and, in several countries, armed conflict. Much of the region's child labour is in subsistence and commercial agriculture, and the region is also where conflict-related worst forms, including the recruitment of children by armed groups, are most concentrated. Progress has been slower here than elsewhere, and without stronger investment in social protection and education the region will continue to account for a large part of the global total. Asia and the Pacific has seen substantial long-term declines, driven by economic growth and expanding education, but the region remains home to very large numbers of working children because of its population size. Here the concerns include hazardous work in agriculture and manufacturing, bonded labour in some sectors, trafficking, and the exploitation of children in the informal economy. The region's deep integration into global supply chains also makes it a focus of concern for the hidden child labour embedded in the production of goods for export. In Latin America and the Caribbean, child labour had been declining, but recent analysis notes that the numbers of working children rose in the period around the pandemic, interrupting earlier gains. Scholarship on the application of Convention No. 182 in the region highlights a recurring theme, that ratification and even good legislation have not translated smoothly into protection, because vulnerable children remain exposed to poverty, informality, and, in some areas, the reach of organised crime and drug trafficking. The regional experience underlines the article's central argument, that legal commitment is necessary but not sufficient, and that the outcome depends on the strength of the systems built to give the law effect. Across all regions, two patterns recur. The first is that agriculture and the informal economy dominate, which means that strategies focused only on formal workplaces will miss most of the problem. The second is that the worst forms cluster where poverty, weak governance, and instability overlap, which means that ending them requires action on development and security as much as on law. Convention No. 182 and the Sustainable Development Goals Convention No. 182 does not stand alone in the international system. It is closely tied to the Sustainable Development Goals adopted in 2015, and in particular to Target 8.7, which commits States to take immediate and effective measures to eradicate forced labour, end modern slavery and human trafficking, secure the prohibition and elimination of the worst forms of child labour including the recruitment and use of child soldiers, and end child labour in all its forms. The Convention and the target reinforce each other. The Convention provides the binding legal norm, while the target provides a shared political goal and a measurable indicator. The connection runs in both directions and reaches well beyond Goal 8. Child labour keeps children out of school, undermining the goal of quality education. It traps families in poverty across generations, working against the goal of ending poverty. It harms health and well-being, both in childhood and later life. And it corrodes the fairness and sustainability of production, complicating the goal of responsible consumption. In this sense the elimination of the worst forms is not an isolated aim but a thread woven through the wider development agenda. Progress on child labour supports progress on many other goals, and failure on child labour drags them down. The awkward fact, already noted, is that the specific deadline attached to Target 8.7, the elimination of child labour by 2025, has passed without being met. This does not diminish the value of the goal, but it does force a reckoning. Recent policy analysis argues that meeting the objective, even on a later timetable, will require moving beyond prohibition to tackle the root causes, through social protection, decent work for adults, accessible education, and childcare, supported by genuine dialogue with affected communities and with children themselves. The lesson drawn from the missed deadline is not that the goal was wrong, but that the means devoted to it have been too limited, and that Convention No. 182 will only deliver on its promise if it is backed by the investment its obligations imply. Discussion The evidence reviewed here supports a clear central claim. Convention No. 182 is a normative success and an implementation challenge. Its success lies in the fact that it expresses, in binding legal form, a moral consensus that almost no government is willing to reject. The unconditional prohibition of #child_slavery, of the sexual exploitation of children, and of their use in crime and armed conflict is now part of the accepted architecture of international law. That achievement should not be underestimated, because it gives advocates, survivors, and reformers a firm standard to invoke and a continuing process through which to hold governments to account. Yet the same evidence shows the limits of law when it is not matched by capacity and investment. Universal ratification has coincided with the continued existence of the very practices the Convention condemns. The recent resumption of decline in overall child labour is encouraging, but the worst forms are concentrated in exactly the settings, deep poverty, the informal economy, conflict, and organised crime, where law is hardest to enforce and where progress is slowest. The missed 2025 target is the clearest sign of this gap. A norm can be universal and still be routinely violated if the conditions that produce the violation are left in place. Several tensions run through the field. One is between prohibition and prevention. Banning the worst forms is essential, but bans do not by themselves remove the poverty that drives children into work, and where enforcement is harsh but support is absent, children may be pushed further into hiding rather than into safety. A second tension is between the universal and the local. The Convention rightly sets absolute standards for the gravest abuses, but recent commentary reminds us that applying these standards well requires attention to the social and economic circumstances in which children live, and, where appropriate, to the voices of the children affected. A third tension is between visibility and hiddenness. The problem the Convention targets is by its nature concealed, which makes measurement, enforcement, and accountability perpetually difficult. For students and researchers, these tensions define an agenda. They point to the need for better data on the worst forms specifically, for evaluation of which interventions actually work and at what cost, for study of how supply chain due diligence laws perform in practice, and for careful analysis of how justice systems can treat exploited children as victims rather than offenders. The Convention gives the field its normative foundation. The task now is empirical and practical, to learn what closes the gap between the law's promise and children's lives. Recommendations Several conclusions for policy follow from the analysis. They are offered not as novel discoveries but as the settled implications of the evidence. First, invest in #social_protection. The strongest and most consistent finding in the literature is that easing the economic pressure on poor households reduces child labour. Cash transfers, in particular, have a measurable effect, because they address the reason most children work in the first place. Social protection should be treated not as charity but as a core instrument for realising the Convention. Second, guarantee accessible and quality #education. Every child removed from the worst forms needs somewhere to go, and school is that place. Free, safe, and relevant education, close to where children live, both prevents child labour and gives rescued children a future. The Convention's own emphasis on education reflects this, and it should be honoured with real resources. Third, strengthen #enforcement in a way that protects children. This means more and better-trained labour inspectors with the mandate to reach the informal economy, police and prosecutors equipped to handle trafficking and exploitation, and justice systems that treat exploited children as victims. Enforcement should be paired with support, so that removal from the worst forms leads to rehabilitation rather than to destitution or criminalisation. Fourth, extend responsibility along #supply_chains. Governments should require and enforce human rights due diligence by companies, and firms should accept genuine responsibility for conditions at the lower tiers of their supply chains, where the risk is highest. Consumers, investors, and civil society all have roles in sustaining this pressure. Fifth, protect children in conflict and crisis. Because the most severe worst forms cluster in conflict zones, protecting children there, through demobilisation, reintegration, and the restoration of schooling and services, must be a priority, difficult as it is. Sixth, improve #data and cooperation. Better measurement of the worst forms, undertaken with care for the children involved, and stronger international cooperation across borders, are essential for targeting effort and holding governments to account. The Convention's call for international assistance and cooperation should be met in practice, not only in principle. Taken together, these recommendations share a single premise. Convention No. 182 provides the law. Ending the worst forms requires the sustained investment, capacity, and cooperation that turn law into protection. Conclusion The Worst Forms of Child Labour Convention stands as one of the clearest moral statements in international law. In agreeing to it, every government in the world accepted that the enslavement, trafficking, sexual exploitation, criminal use, and dangerous labour of children must be prohibited and eliminated at once. That consensus is real and valuable, and the recent renewal of progress against child labour shows that determined action can move the numbers in the right direction. But the record since 1999 also teaches a hard lesson. A universally accepted norm is not the same as a solved problem. Roughly 138 million children remained in child labour in 2024, the worst forms persist in the shadows of poverty, conflict, and crime, and the promised deadline for ending child labour has come and gone. The gap between the Convention on paper and children's lives in practice is the defining challenge of this field. Closing that gap does not require a new law. The legal foundation already exists, and it is strong. What it requires is the will and the resources to act on the obligations already accepted, to lift families out of the poverty that drives exploitation, to keep children in school, to enforce the rules where they are hardest to enforce, and to cooperate across borders against a problem that respects none. Convention No. 182 has given the world a shared standard and a shared goal. Whether that standard becomes a reality for the tens of millions of children still at risk depends not on the text, but on what governments, businesses, and societies now choose to do. The immediate prohibition the Convention demands remains, for far too many children, still to be delivered. References International Labour Organization and United Nations Children's Fund. (2025). Child Labour: Global Estimates 2024, Trends and the Road Forward. Geneva and New York: ILO and UNICEF. International Labour Organization. (2023). Framework of Action to Accelerate the Elimination of Child Labour 2023-2025. Geneva: International Labour Office. Zsogon, X. (2024). ILO's Convention 182 on the Worst Forms of Child Labour: challenges for its application in Latin America. Studia Iuridica Lublinensia, 33(3), 157-172. https://doi.org/10.17951/sil.2024.33.3.157-172 Papatheophilou, S., and Troster, B. (2025). Ending Child Labour: Implementation of SDG 8.7. Research Report 24/2025. Vienna: Austrian Foundation for Development Research. https://doi.org/10.60637/2025-rr24 Ostby, G., Rustad, S. A., Haer, R., and Arasmith, A. (2023). Children at risk of being recruited for armed conflict, 1990-2020. Children and Society, 37(2), 524-543. https://doi.org/10.1111/chso.12609 Lynam, T., Johnson, D., and Baillie Abidi, C. (2025). Reliable knowledge claims on the recruitment and use of children: an empirical perspective. Journal of Peace Research, 62(6), 1889-1907. https://doi.org/10.1177/00223433251318862 Kim, B. E., Dierkhising, C. B., De Leon, J., Sandoval, J., Brissett, A., and Bounds, D. (2023). Evaluation of services for the commercial sexual exploitation of children and youth: a scoping review. Trauma, Violence, and Abuse, 24(5), 3236-3250. https://doi.org/10.1177/15248380221126185 O'Brien, J. E., Jensen, T., Mitchell, K. J., and White, K. (2023). Commercial sexual exploitation outcomes in a community sample of youth. Journal of Adolescent Health, 72(1), 51-58. https://doi.org/10.1016/j.jadohealth.2022.08.028 Neha, and Kumar, R. (2023). Child trafficking in India and procedures for prevention and protection of children's rights: an analysis. Journal of Victimology and Victim Justice. https://doi.org/10.1177/23493003221110731 Tolbaru, C.-E. (2023). Sex trafficking and commercial sexual exploitation of children within the context of globalisation. Bulletin of the Transilvania University of Brasov, Series VII: Social Sciences and Law, 16(65). Lakdawala, L. K., Martinez Heredia, D., and Vera-Cossio, D. A. (2025). The effects of expanding worker rights to children. Journal of Development Economics, 172. Dimova, R. (2021). The Political Economy of Child Labor. GLO Discussion Paper No. 816. Essen: Global Labor Organization. Hashtags #Worst_Forms_of_Child_Labour #ILO_Convention_182 #Child_Labour #Child_Protection #Child_Rights #Human_Rights #Child_Trafficking #Child_Slavery #Forced_Labour #Commercial_Sexual_Exploitation #Child_Soldiers #Hazardous_Work #SDG_8_7 #End_Child_Labour #Recommendation_190

This article examines the Minimum Age Convention, 1973 (No. 138) of the International Labour Organization, with a focus on the duty it places on ratifying states to pursue a national policy designed to secure the #effective_abolition of #child_labour and to raise the #minimum_age for work over time. The Convention is one of the core labour standards of the #ILO and sits at the centre of the global legal framework against child labour, alongside the Worst Forms of Child Labour Convention, 1999 (No. 182). Using a doctrinal and narrative review of the treaty text, the reports of the ILO supervisory bodies, and peer reviewed studies published between 2020 and 2025, the paper does three things. First, it explains the structure of #Convention_138, including its age tiers, its treatment of #hazardous_work and #light_work, and the flexibility clauses that allow poorer states to adjust the standard to local conditions. Second, it studies the meaning of the #national_policy obligation in Article 1 and argues that this duty, rather than the age numbers alone, is the part of the Convention that carries the heaviest weight. Third, it reviews the recent evidence on how far the Convention has helped in practice, drawing on the 2024 global estimates that place roughly 138 million children in child labour worldwide. The paper finds that the Convention has shaped national law across most of the world and has helped push child labour down over two decades, yet the target of ending child labour by 2025 was missed. It concludes that the Convention works best when treated not as a single ban but as a call for a joined up policy that links #social_protection, free #compulsory_education, #decent_work for adults, and steady #enforcement. Keywords: child labour; minimum age; International Labour Organization; Convention No. 138; national policy; social protection; compulsory education; hazardous work; ratification; Sustainable Development Goal 8.7 1. Introduction Few social problems have stayed on the world agenda for as long as child labour. It was one of the first concerns of the ILO when the organization was founded in 1919, and it remains a live problem today. The reason is simple. When a child works long hours, or does dangerous work, or works instead of going to school, the harm is not only immediate. It follows the child into adult life through lost schooling, poor health, and lower earnings. It also passes down to the next generation, because a child who never learned to read is more likely to raise children who work as well. This is why the fight against #child_labour is treated in international law as both a human rights matter and a development matter. The Minimum Age Convention, 1973 (No. 138) is the main legal tool that the ILO uses to attack the problem at its root. The logic behind the Convention is that one of the clearest ways to protect children is to fix an age below which they cannot legally be employed or put to work, and to tie that age to the end of compulsory education. If a country sets a firm floor and enforces it, children stay in school during the years that matter most for their growth. The Convention therefore requires each ratifying state to declare a minimum age for admission to employment and to make sure that no one below that age is admitted to work in any occupation, subject to a small number of exceptions. Yet the Convention is more than a rule about age. Its opening article contains a broader promise. Each state that ratifies undertakes to pursue a national policy designed to ensure the effective abolition of child labour and to raise the minimum age step by step to a level that fits the full physical and mental development of young people. This is the heart of the instrument. The age numbers are the visible part, but the national policy duty is the engine. A country can copy a minimum age into its statute book in a single afternoon. Building a policy that actually keeps children out of work and in classrooms takes years of coordinated effort across many ministries. This article is written for students and general readers who want a clear account of what Convention 138 says, why it matters, and how well it has worked. It treats the Convention as a case study in how international standards try to change behaviour inside sovereign states. The argument runs as follows. The Convention has been a success as a norm setting device. It has been ratified by the large majority of ILO members and has shaped national law across every region. It has also coincided with a long fall in the number of working children. At the same time, the Convention has clear limits. A treaty cannot remove #poverty, cannot build schools, and cannot by itself fund the #labour_inspection systems needed to make the law real. The distance between the law on paper and the situation on the ground remains wide in many places, and the world has missed the deadline it set for itself. The paper proceeds in twelve further sections. Section 2 places the Convention in its historical setting. Section 3 sets out its legal architecture in plain terms. Section 4 studies the national policy obligation in depth. Section 5 states the method used. Section 6 covers #ratification and the supervisory machinery. Section 7 presents the current global picture. Section 8 examines the root causes that keep child labour alive. Section 9 turns to implementation and enforcement gaps. Section 10 offers regional and country perspectives. Section 11 sets out policy pathways. Section 12 reviews the main debates and criticisms surrounding the Convention. Section 13 discusses whether the Convention is working, before a short conclusion. 2. Historical background: from 1919 to 1973 The #ILO adopted its first minimum age instrument in 1919, its founding year, and it dealt with industry. Over the following decades the organization adopted a series of separate conventions, each covering a single sector such as industry, agriculture, sea work, trimmers and stokers, non industrial employment, fishing, and underground work. This piecemeal method had a logic in its time, because working conditions differed sharply between sectors. But it also produced a patchwork. A child might be protected in a factory yet unprotected on a farm, even though the harm was the same. By the early 1970s the ILO decided that a single, general standard was needed to replace the scattered older texts and to close the gaps between them. The result was #Convention_138, adopted by the International Labour Conference in Geneva on 26 June 1973. It entered into force on 19 June 1976, twelve months after the first two states, Cuba and Libya, deposited their ratifications. The new Convention was written to be broad. It applies to all sectors of economic activity and to all forms of work, whether or not the work is done under a formal contract of employment. This was a deliberate choice. Much of the world's child labour happens outside formal jobs, on family farms, in small workshops, in street trades, and in domestic service. A standard that only covered formal employment would have missed most working children. By covering work as well as employment, the Convention reached into the #informal_economy where the problem is most severe. The Convention was designed to absorb and replace the older sector specific texts. When a state ratifies Convention 138 and accepts certain obligations, the older conventions it had ratified are treated as denounced under set conditions. In this way the drafters tried to tidy up the field and give the world one reference point rather than ten. Convention 138 is accompanied by a non binding guide, #Recommendation_146, which is not open for ratification but which sets out the wider policy measures that give the Convention its full meaning. The Recommendation stresses poverty reduction, decent work for adults, free and compulsory education, vocational training, the extension of social protection, and reliable #birth_registration so that a child's age can be proven. It is worth pausing on the slow early pace of ratification. For many years states were reluctant to bind themselves to a firm age floor, in part because they feared that a strict rule would clash with the reality of poverty in rural economies. This reluctance is one reason the ILO later adopted a narrower and more urgent instrument, #Convention_182 on the worst forms of child labour, in 1999. That later Convention focused world attention on the most extreme abuses, such as slavery, trafficking, the use of children in armed conflict, and clearly dangerous work, while keeping the long term goal of ending all child labour. The two conventions are meant to work together. #Convention_138 sets the general floor, and Convention 182 marks the abuses that must be attacked first and fastest. Both are counted among the #fundamental_conventions of the ILO, which means that all member states are expected to respect the principles behind them even without ratification. 3. The legal architecture of Convention No. 138 To understand the Convention, it helps to walk through its main working parts in order. The instrument is short, but each article carries weight, and the age rules only make sense once the flexibility clauses are placed beside them. 3.1 The general obligation Article 1 is the anchor. It commits each ratifying state to pursue a #national_policy designed to ensure the #effective_abolition of #child_labour and to raise the minimum age for admission to employment progressively to a level consistent with the fullest development of young persons. Two ideas are packed into this sentence. The first is abolition as the ultimate aim, not mere regulation. The Convention does not seek to make child labour tidy or safe. It seeks to end it. The second idea is progression. States are not told to reach a high age overnight. They are told to keep moving upward over time. This recognizes that a very poor country may need to start lower and climb as its economy and schools improve. 3.2 The minimum age Article 2 requires each state to declare a #minimum_age in a formal statement attached to its #ratification. Below that age, no one may be admitted to work in any occupation, subject to the exceptions described later. The Convention then sets the rules for how low that age may be. As a baseline, the declared minimum age must not be lower than the age at which #compulsory_education ends, and in any case must not be below 15 years. This link between the work age and the school leaving age is one of the most important features of the whole instrument. It ties two policies together so that a child leaving school does not fall into a gap where he or she is too young to be protected as a worker but no longer required to be in a classroom. The Convention allows one important easing of this baseline. A state whose economy and educational facilities are not sufficiently developed may, after consulting employers and workers, declare a minimum age of 14 for a period of time, as a transitional step. A large share of ratifying states used this option. The idea is that the age can be raised later as conditions allow, in keeping with the progression principle in Article 1. 3.3 Hazardous work Article 3 deals with the most dangerous category. Work that is likely to harm the health, safety, or morals of young persons may not be done by anyone under 18. This higher age reflects the seriousness of the risk. Hazardous work includes tasks such as mining, work with dangerous machinery, exposure to harmful chemicals, work at unsafe heights, and other activities that can cause lasting physical or psychological damage. The Convention allows a narrow exception. National law may permit such work from the age of 16 under strict conditions, provided the health, safety, and morals of the young workers are fully protected and they have received adequate instruction or training in the relevant field. Each country decides, through consultation, which specific tasks count as hazardous work and lists them, so that inspectors and employers know where the line falls. 3.4 Light work Not all work by children is harmful, and the Convention does not treat it as such. Article 7 allows light work for young people below the general minimum age within a defined band. Where the minimum age is 15, children aged 13 to 15 may do #light_work. Where the state has set the age at 14 under the transitional rule, the band is 12 to 14. Light work is work that does not harm a child's health or development and does not get in the way of schooling, attendance, or the benefit that a child gets from education. States that allow it must define which activities count, and must set the hours and conditions under which it may be done. About half of the states that ratified the Convention chose to regulate light work in this way. The purpose is to draw a clear line between harmful child labour, which must end, and modest age appropriate tasks that can support a child's growth and a household's wellbeing. 3.5 Flexibility clauses The Convention is famous for its flexibility, and this flexibility is one reason it reached such wide ratification. Beyond the transitional age of 14 and the light work band, the instrument contains several clauses that let states adjust the standard. Article 4 permits a state, after consultation, to exclude limited categories of employment or work from the Convention where special and substantial problems of application arise, although this cannot be used to escape the rules on #hazardous_work. Article 5 lets a state whose economy and administrative facilities are not sufficiently developed initially limit the scope of application to certain branches of activity, again after consultation, while committing to extend coverage over time. Article 8 allows exceptions for individual cases such as participation in artistic performances, granted by permit and subject to limits on hours and conditions. These clauses have been read by some as loopholes and by others as the price of near universal acceptance. The honest view is that they are both. The flexibility made it possible for states at very different levels of development to accept a common norm, which is a real achievement. It also created spaces where practice can drift away from the spirit of the Convention if oversight is weak. The #ILO supervisory bodies spend much of their effort examining how states use these clauses and pressing them to narrow exclusions and raise ages as conditions improve. 3.6 The place of Recommendation No. 146 Recommendation 146 is not binding, but it explains what a serious national policy looks like. It links the #minimum_age to a wider package that includes poverty reduction, the creation of decent work for adults, the steady extension of social protection, free and compulsory education up to at least the minimum age, vocational guidance and training, and birth registration so that ages can be verified. Read together with Article 1, the Recommendation makes clear that the Convention was never meant to be a single ban enforced by inspectors alone. It was meant to sit at the centre of a broad social policy that removes the reasons families send children to work in the first place. 4. The meaning of the national policy obligation The phrase at the centre of this paper, and at the centre of the Convention, is the duty to pursue a national policy designed to ensure the effective abolition of child labour. It is worth unpacking each part of that phrase, because a great deal turns on it. The word policy signals that the obligation is about more than a statute. A minimum age law is necessary, but a law is a single instrument. A policy is a coordinated set of measures pursued over time toward a stated goal. The Convention therefore asks states to do something harder than pass a law. It asks them to organize the machinery of government around the goal of ending child labour, and to keep that machinery pointed at the goal even when attention drifts to other issues. The word national signals scope. The policy must cover the whole country and all sectors, including the parts of the economy where #child_labour is hardest to reach. This is a demanding requirement in states where much work happens on family farms and in the informal economy, far from any inspector. It means that the policy cannot be limited to the formal, visible economy. It must find ways to reach the fields, the workshops, the markets, and the homes where children actually work. The word effective signals that results matter. The Convention does not ask for a policy that looks good in a report. It asks for a policy that works, that actually reduces the number of children in child labour. This is why the ILO supervisory bodies press states not only to describe their laws but to show what the laws achieve. A country that has a fine statute and rising numbers of working children has not met the obligation, whatever its paperwork says. The word abolition signals the end goal. The Convention treats child labour as something to be removed, not merely reduced or managed. This is an ambitious aim, and it is deliberately so. The point of naming abolition as the target is to keep pressure on states to move forward rather than settle for a comfortable level of the problem. The progression principle in Article 1, which requires the minimum age to rise over time, is the practical expression of this aim. Read as a whole, then, the #national_policy obligation asks each state to treat the ending of child labour as a standing goal of government, to organize policy across ministries around that goal, to reach every sector, and to show real results. This is a much larger commitment than the age numbers alone suggest. It is also the part of the Convention that best explains why some countries have made rapid progress while others have not. The countries that have done best are those that treated Convention 138 as a call to build a joined up policy, linking the #minimum_age with social protection, free schooling, and #decent_work. The countries that have done worst are those that copied the age into law and did little else. It also explains why the Convention pairs so naturally with the wider development agenda. The commitment to end child labour by 2025 that the world adopted under Sustainable Development Goal target 8.7, known here as #SDG_8_7, restates in modern language the same aim that Article 1 set out in 1973. The Convention gave the goal its legal shape. The development agenda gave it a deadline. Both rest on the same insight, that child labour cannot be ended by a rule alone but only by a policy that changes the conditions that produce it. 5. Methodology This paper is a doctrinal and narrative review rather than an empirical study. It draws on three kinds of material. The first is primary legal material, meaning the text of #Convention_138 itself, its companion #Recommendation_146, and the related instrument #Convention_182, together with the interpretive comments of the ILO supervisory bodies that examine how states apply the Convention. The second is data material, chiefly the joint global estimates on #child_labour produced by the #ILO and UNICEF, which provide the numbers used in the analysis. The third is scholarly material, meaning peer reviewed journal articles and academic books published between 2020 and 2025 that study the causes of child labour, the effects of minimum age laws, and the policies that reduce the problem. The review method has strengths and limits that readers should keep in mind. Its strength is breadth. By combining the legal text with recent evidence, it can show both what the Convention requires and how far practice matches the requirement. Its limit is that it does not test a single hypothesis with new data. It synthesizes existing knowledge rather than generating new measurements. Where the paper reports figures, those figures come from the cited sources and carry the uncertainties that the sources themselves acknowledge. Estimates of child labour are especially hard to pin down, because much of the work is hidden, definitions vary, and data are missing for some regions. The paper treats all such numbers as best available estimates rather than exact counts. 6. Ratification and the supervisory system A treaty only binds the states that accept it, so ratification matters. Convention 138 was ratified slowly at first but steadily gathered support, and by the mid 2020s more than 175 of the ILO's 187 member states had ratified it. This makes it one of the most widely accepted labour standards in the world. A small number of states, including the United States and New Zealand, had not ratified it as of the middle of the decade, though non ratifying states are still expected to respect the principle of the effective abolition of child labour because the Convention is one of the fundamental conventions. Its companion, Convention 182 on the worst forms of child labour, reached universal ratification in 2020, when Tonga became the final state to ratify, the first time in the ILO's history that any convention was accepted by every member. Ratification is not the end of the story. It is the start of an ongoing relationship between the state and the ILO's supervisory system. States that have ratified must report at regular intervals on how they apply the Convention in law and practice. A committee of independent legal experts examines these reports, along with information from employers' and workers' organizations, and issues comments. Where it finds gaps, it asks the state to explain and to take action. In more serious cases the matter can be taken up by a tripartite committee at the annual International Labour Conference, which brings governments, employers, and workers together. This machinery has no power to fine or punish. Its power is the power of scrutiny, publicity, and steady pressure. Over time, the accumulated comments of the supervisory bodies build a body of interpretation that tells states what the Convention actually demands in concrete situations. The value of this system is that it keeps the Convention alive after #ratification. Without supervision, a treaty can become a dead letter, ratified and forgotten. With supervision, states are held to account year after year, and the meaning of the Convention is worked out through a long conversation between the experts and the states. The weakness of the system is equally clear. It depends on states reporting honestly, on inspectors gathering accurate information, and on the willingness of states to act on the comments they receive. Where a government lacks the capacity or the will to reach into the #informal_economy, the supervisory system can identify the gap but cannot close it on its own. 7. The global picture: what the data show Numbers give the Convention its context, and the most recent global figures come from the joint estimates produced by the #ILO and UNICEF. According to the 2024 estimates, released in 2025, nearly 138 million children aged 5 to 17 were in child labour worldwide, which is close to 8 percent of all children. Of these, about 54 million were doing #hazardous_work likely to harm their health, safety, or morals. The gender split showed more boys than girls in #child_labour when measured by economic activity, though the gap narrows or reverses once long hours of unpaid household chores are counted. The direction of change is encouraging over the long run. Since 2000, the number of children in child labour has almost halved, falling from around 246 million to 138 million. The most recent four year period, from 2020 to 2024, saw a fall of more than 22 million, which reversed a worrying rise recorded between 2016 and 2020 when #poverty, weak #social_protection, and the shock of the COVID-19 pandemic pushed the numbers up. This return to progress is real and worth noting. It shows that child labour is not a fixed feature of the world but a problem that responds to policy. At the same time, the data carry a hard message. The world set itself the goal of ending all child labour by 2025 under SDG target 8.7, and that deadline passed with the problem far from solved. Analysts estimate that to end child labour within the following five years, progress would have to run many times faster than it does now, on the order of eleven times faster. The gap between ambition and reality is therefore very wide. The data also show where the problem sits. Agriculture accounts for the majority of #child_labour, roughly 61 percent of cases, followed by services such as domestic work and street selling at around 27 percent, and industry, including mining and manufacturing, at around 13 percent. This distribution matters for policy. A strategy that focuses only on factories will miss most working children, because most of them are on farms. It also explains why the informal economy is so central to the problem. Farm work and household service are largely informal, beyond the reach of ordinary labour inspection, which is why the Convention's demand for a broad national policy is so important. Regionally, progress has been uneven. Asia and the Pacific recorded the largest fall in prevalence in the most recent period, while sub Saharan Africa continues to carry the heaviest burden, driven by high poverty, fast population growth, conflict, and thin social protection. 8. Why child labour persists: the root causes If the law is so widely accepted, why does child labour persist on such a scale. The answer is that child labour is a symptom of deeper conditions, and a minimum age law addresses the symptom more than the cause. Recent research points to a set of linked drivers that a #national_policy must tackle together. 8.1 Poverty and household shocks The single most important driver is poverty. When a household cannot meet its basic needs, the labour of its children can seem like a necessity rather than a choice. Studies of poor rural households show that families often weigh the small income a child brings against the cost of schooling, and that when money is short the balance tips toward work. Sudden shocks make this worse. A failed harvest, an illness, the loss of a parent's job, or a wider economic crisis can push a household to send children to work as a coping strategy. Research from Ghana, for example, has found that child labour is linked to longer term household #poverty and food insecurity, which suggests a trap in which poverty leads to child labour and child labour then helps to keep the household poor. Breaking this trap is the core task of any serious policy. 8.2 The cost and quality of education The second driver is the state of schooling. The Convention ties the minimum age to the end of #compulsory_education for good reason, but the link only helps if school is genuinely available, affordable, and worth attending. In many places formal fees have been abolished yet indirect costs remain, such as uniforms, books, transport, and informal charges, and these costs can be enough to keep a poor child out of class. Where schools are far away, overcrowded, or of low quality, parents may see little point in the sacrifice, and children drift into work. Evidence from cocoa growing regions shows that better school quality is linked to less #child_labour, which supports the idea that education is not just an alternative to work but an active tool against it. A national policy that raises the minimum age without investing in free, close, and decent schools is likely to fail. 8.3 Weak social protection The third driver is the absence of social protection. Where families have no safety net, the labour of children becomes the safety net. This is why #cash_transfers to poor households have become one of the most studied and most effective tools against child labour. When a family receives a regular payment, often on the condition that children attend school, the pressure to send children to work eases, and school attendance tends to rise. Programmes of this kind in Latin America and elsewhere have shown real results. The lesson is that social protection is not a soft extra but a central plank of an effective policy. The recent fall in child labour has coincided with the spread of child focused #social_protection, and the recent rise in some regions has coincided with its absence or withdrawal. 8.4 The informal economy and enforcement reach The fourth driver is the structure of the economy. Most child labour happens in the informal economy, on family farms, in small unregistered workshops, in street trades, and in private homes as domestic workers. These are exactly the places that ordinary labour inspection cannot easily reach. A country can have a strong #minimum_age law and a working inspection service for factories and still see widespread child labour on farms and in homes, because the law never touches those settings in practice. Extending the reach of policy into the informal sector is one of the hardest tasks in the whole field, and it is where many national policies fall short. 8.5 Social norms and expectations The fifth driver is less visible but powerful. In some communities, work by children is seen as normal, even as good training, and not as a harm. A systematic review of the social norms behind family child labour found that ideas about obedience, apprenticeship, gender roles, and the passing of trades from parent to child all shape how families treat children's work. Where these norms are strong, a law that clashes with them will meet quiet resistance. Changing norms takes time and dialogue and cannot be done by decree. This is why awareness raising, the involvement of communities, and even the voices of working children themselves are increasingly seen as part of a serious policy. 8.6 Shocks: conflict, climate, and pandemic A final set of drivers are the large shocks that can undo years of progress. The COVID-19 pandemic pushed millions of children toward work as households lost income and schools closed. Armed conflict drives #child_labour by destroying livelihoods and schooling, and studies of the Middle East and North Africa show how cascading crises raise the risk. Climate change is emerging as a further threat, as failed harvests and disasters strip households of income. These shocks explain why progress is not steady and why gains can reverse. They also strengthen the case for social protection, which acts as a buffer when shocks hit. 9. Implementation gaps and enforcement challenges Even where the law is sound and the policy is well designed, the step from paper to practice is where most of the difficulty lies. Several gaps recur across countries. The first gap is #labour_inspection capacity. Many states have too few inspectors, too little funding, and too little authority to reach the places where children work. Inspectors are often concentrated in cities and formal workplaces, while child labour is scattered across rural areas and private homes. Even a committed inspection service cannot cover a vast informal sector with a small budget. Strengthening inspection, and giving it the reach and resources to enter agriculture and domestic settings, is a constant theme in the comments of the ILO supervisory bodies. The second gap is birth registration. A minimum age law depends on knowing how old a child is. Where births are not registered, a child's age cannot be proven, and both employers and inspectors can claim uncertainty. This is why the Convention's supporting materials stress reliable #birth_registration as part of the policy. Without it, the age floor floats free of any way to apply it. The third gap is the design of the law itself. Some national laws contain wide exclusions, weak penalties, or definitions of hazardous work that leave out common dangers. Reviews of national frameworks in South Asia, for example, have shown how loopholes and gaps between different laws can leave many working children unprotected in practice, even where a minimum age exists on paper. A law with holes invites evasion. Closing these gaps, narrowing exclusions, and aligning the different laws that touch children's work is slow but necessary. The fourth gap is coordination. The national policy that the Convention requires cannot be run by a labour ministry alone. It needs education, social welfare, health, justice, and finance to pull together. Where these bodies work in silos, policy fragments, and children fall between the cracks. The most effective responses treat child labour as a cross cutting issue owned by the whole of government, not the project of a single department. The fifth gap is funding. All of the above costs money, for inspectors, for schools, for cash transfers, and for the systems that tie them together. When budgets are cut, whether by domestic austerity or by falls in international aid, the machinery weakens and progress stalls. Recent warnings from the ILO and UNICEF stress that reductions in support for education and social protection can push vulnerable families back toward child labour, which shows how fragile the gains can be. 10. Country and regional perspectives The general picture becomes clearer through regional and national examples, drawn from the recent literature and the global estimates. These are offered as illustrations of the wider patterns, not as full case studies. In sub Saharan Africa, the burden of child labour remains the heaviest in the world. High rates of poverty, rapid growth in the number of children, conflict in several regions, and thin #social_protection combine to keep prevalence high. Agriculture dominates, including cocoa farming in West Africa, where research links school quality to reduced child labour. The region shows both the scale of the challenge and the promise of education focused responses. In Asia and the Pacific, the most recent estimates record the largest fall in prevalence of any region, a sign that sustained growth, the spread of schooling, and expanding social protection can move the numbers quickly. India, the most populous country in the region, ratified both #child_labour conventions in 2017, a step that brought a very large share of the world's children under the coverage of the standard. South Asian countries also illustrate the enforcement problem, with reviews of national laws in Pakistan and Bangladesh pointing to persistent gaps between legislation and practice, especially in informal and hazardous work such as brick kilns and domestic service. In Latin America, the region has been a testing ground for #cash_transfers, and conditional transfer programmes there are among the most cited examples of effective policy against child labour. By giving poor families a regular payment tied to school attendance, these programmes have helped keep children in class and out of work, showing how social protection and education reinforce each other. In the Middle East and North Africa, cascading crises of conflict, displacement, and economic collapse have raised the risk of child labour, and research warns that without strong social protection the numbers there could climb. This region shows how quickly shocks can undo progress and how central protection systems are to holding the line. Even high income countries are not free of the problem, which is a point often missed. In wealthy states, child labour tends to take different forms, such as underage work in agriculture, family businesses, or among migrant and refugee children who fall outside the reach of ordinary protection. The fact that a small number of high income states, including the United States, have not ratified the Convention shows that acceptance of the norm is not automatic even where resources are ample. It also reminds us that the Convention is not only a standard for poorer nations. Its principle, that no child should be admitted to work below a set age tied to schooling, applies everywhere, and gaps in richer states deserve the same scrutiny as gaps in poorer ones. Across all regions, the common thread is that legal ratification is necessary but not sufficient. The countries that have moved fastest are those that used the Convention as a frame for a wider policy, combining the minimum age with schooling, protection, and decent work for adults. The countries that lag are those where the law stands alone. 11. Policy pathways: making the national policy work The evidence points toward a fairly clear set of measures that turn the Convention's promise into results. None is new, but together they form the joined up policy that Article 1 envisions. The first pathway is #social_protection. Regular support to poor households, including child benefits and conditional or unconditional cash transfers, reduces the economic pressure that drives families to rely on children's work. This is the measure with the strongest evidence behind it, and it should sit at the centre of any policy. Universal child benefits, where affordable, remove the need for families to prove eligibility and reach children before they fall into work. The second pathway is free, compulsory education of decent quality, available close to where children live. Removing not only formal fees but also the indirect costs of uniforms, books, and transport helps keep poor children in school. Improving the quality of schooling matters as much as access, because a school that teaches little gives families little reason to choose it over work. Tying the school leaving age to the minimum age, as the Convention requires, closes the gap in which children are too old for school but too young for legal work. The third pathway is decent work for adults. Child labour often reflects the failure of adult labour markets. When parents cannot earn enough through decent jobs, the household turns to the labour of its children. Policies that raise adult wages, protect the right to organize, and create stable employment reduce the demand for child labour at its source. This is why the Convention's supporting materials place #decent_work for adults alongside the #minimum_age. The fourth pathway is stronger #enforcement, meaning labour inspection with the reach, funding, and authority to enter agriculture and domestic settings, backed by reliable birth registration so that ages can be verified. Enforcement alone will not end child labour, and heavy handed enforcement that punishes poor parents can even backfire by driving work underground or stigmatizing children. But without any enforcement, the law is empty. The aim is enforcement that is firm on employers and supportive toward families. The fifth pathway is action on #supply_chains. As global trade has grown, child labour in the production of goods such as cocoa, cotton, and minerals has drawn attention from consumers, companies, and governments. Human rights due diligence rules, which require firms to check their supply chains for #child_labour and to act on what they find, are becoming more common. These measures extend the reach of policy beyond national borders and can bring pressure to bear on informal producers who are otherwise hard to reach. They work best when paired with support for the affected families, so that removing a child from work does not simply deepen the household's #poverty. The sixth pathway is the involvement of communities and of working children themselves. Because #social_norms shape how families treat children's work, lasting change needs dialogue, not just rules. Programmes that engage local leaders, listen to working children, and address the reasons behind the norms tend to be more durable than top down bans. Recent gatherings of organizations of working children have argued that policy made without consulting them can be ineffective or even harmful, a point that serious policy makers now take seriously. Taken together, these pathways describe the #national_policy that the Convention asks for. The minimum age is the legal spine. Social protection, education, decent work, enforcement, supply chain action, and community engagement are the muscles that make it move. A policy that has the spine but none of the muscles cannot walk. 12. Debates and criticisms The Convention is widely respected, but it is not beyond argument. Several debates run through the academic and policy literature, and a fair account of the instrument should set them out. The first is the debate between abolition and protection. The Convention takes an abolitionist position. It treats child labour as a wrong to be ended, and it draws a firm line at a set age. A different school of thought, associated with some movements of working children, argues that this line can be too blunt. In this view, work is a reality of many children's lives, especially in poor communities, and the more useful aim is to protect working children from harm and exploitation rather than to ban their work outright. Supporters of this position point out that organizations of working children in Africa, Asia, and Latin America have asked to be consulted and have warned that bans made without their voice can push them into worse conditions. Defenders of the Convention reply that its careful distinctions already separate harmful #child_labour from acceptable light work and age appropriate tasks, and that abolition as a long term aim does not mean ignoring the needs of children who work today. This debate is not fully settled, and it shapes how different actors read the same treaty. The second debate concerns unintended effects. Critics warn that a strict age rule, if enforced without support, can drive child labour underground into settings that are even harder to see and even more dangerous. A ban on visible factory work may simply move children into hidden workshops or home based labour beyond any inspector's reach. There is also a risk that enforcement falls on poor parents, who may be fined or shamed for a situation that is itself a product of poverty. Some field observers report that punitive approaches can stigmatize affected children and raise new barriers to school. The lesson drawn from these concerns is not that the age floor is wrong, but that it must be paired with support, so that a child removed from work is offered school and the family is offered help, rather than left worse off. The third debate is empirical and asks a hard question. Does the Convention actually reduce child labour, or does the fall in child labour come mainly from economic growth, rising incomes, and the spread of schooling, with the law simply following behind. Earlier studies raised doubts about whether ratification by itself changed outcomes, since a country can ratify and still see little change. Recent work is more nuanced. It suggests that the law matters most when it is embedded in a wider policy and backed by enforcement and social protection, and matters least when it stands alone. In other words, the Convention is neither a magic cure nor an empty gesture. It is a lever whose effect depends on what else is pulled with it. This fits the argument running through this paper, that the #national_policy obligation, not the age number, is the part that carries the weight. The fourth debate concerns the flexibility clauses. The transitional age of 14, the power to exclude categories, and the option to limit scope were the price of wide acceptance, but critics argue they leave too much room for weak application. A state can hide behind a clause and delay progress for years. Defenders answer that without this flexibility many states would never have ratified at all, and that the supervisory system exists precisely to press states to narrow their exceptions and raise their ages over time. Both sides agree that the clauses need close watching. A final debate is about measurement. How child labour is counted shapes the picture. When only economic activity is measured, more boys appear in child labour than girls. When long hours of unpaid household chores are added, the gender gap narrows or reverses, because girls carry much of the hidden domestic burden. What counts as hazardous, how old a child must be, and where the line sits between acceptable and harmful work all affect the totals. These are not dry technical points. They decide which children the law and the data see, and which remain invisible. A policy can only protect the children it can count, which is one more reason the Convention's supporting materials stress reliable data and birth registration. 13. Discussion: is the Convention working? So has Convention 138 succeeded. The honest answer is that it depends on the standard used to judge it. As a norm setting instrument, the Convention has clearly succeeded. It has been ratified by the large majority of the world's states. It has shaped national law across every region, giving almost all countries a legal minimum age tied to schooling. It has provided the legal foundation for the global commitment to end child labour, later restated as #SDG_8_7. Its companion, Convention 182, achieved universal ratification, a first in the history of the ILO. By the measure of building a shared world norm, the achievement is remarkable. Two generations ago the idea that every country should have a legal floor on the age of work was contested. Today it is close to universal. As a tool for changing conditions on the ground, the record is mixed. The long fall in child labour since 2000 shows that progress is possible, and the recent return to progress after the setback around 2020 shows that the problem responds to policy. Where states treated the Convention as a call to build a joined up policy, results followed. But the missed 2025 deadline shows the limits. A treaty cannot remove poverty, cannot build schools, and cannot fund inspection. It can set the goal, shape the law, and hold states to account through supervision, but it cannot do the work of development that ending child labour requires. The 138 million children still in child labour are a measure of that gap. This mixed record is not a failure of the Convention so much as a lesson about what treaties can and cannot do. #Convention_138 was always meant to be one part of a wider effort. Its own text says so, in the demand for a national policy rather than a mere ban. The instrument works when it is used as intended, as the frame for a broad policy, and disappoints when it is treated as a stand alone rule. The flexibility clauses that made wide ratification possible also created room for practice to drift, and the supervisory system, powerful as scrutiny, cannot compel a poor or unwilling state to build what it lacks. There is also a deeper point about the relationship between law and poverty. The Convention rests on the belief that setting a floor protects children, and this is true. But a floor only helps children who would otherwise fall below it and who have somewhere better to go. For a child whose family will starve without the child's income, and whose local school is distant or empty, a minimum age law by itself offers little. This is why the evidence keeps returning to social protection and education. The law names the wrong and sets the target. The policy, funded and coordinated, does the actual protecting. The genius of Article 1 is that it saw this in 1973, when it asked not only for an age but for a policy aimed at #effective_abolition. 14. Conclusion The Minimum Age Convention, 1973 (No. 138) is one of the most important instruments in the international effort against #child_labour. Its central demand is not simply that states set an age below which children may not work, though it does require that. Its central demand is that each state pursue a national policy designed to ensure the effective abolition of child labour and to raise the minimum age over time. This national policy obligation, more than the age numbers, is the heart of the Convention and the key to its success or failure. The evidence reviewed here supports a clear conclusion. The Convention has been a triumph as a world norm and a partial success as a practical tool. It has shaped the law of almost every country and has coincided with a long fall in child labour, from around 246 million children in 2000 to roughly 138 million in 2024. Yet the world missed its own deadline to end child labour by 2025, and progress remains far too slow to reach the goal soon. The reason is that the deep causes of child labour, above all poverty, weak social protection, and the cost and quality of schooling, lie beyond the reach of any single law. The path forward is the one the Convention itself pointed to. It runs through social protection that shields poor families from the shocks that drive children to work, through free and compulsory education of decent quality, through decent work for adults, through labour inspection and birth registration that give the law teeth, through action on #supply_chains, and through respect for the communities and children affected. Convention 138 provides the legal spine for this effort and the standard against which to measure it. Whether the world finally ends child labour will depend not on adopting new rules but on funding and coordinating the policy that the Convention has asked for since 1973. The law has done its part in naming the goal. The remaining work belongs to policy, to budgets, and to political will. References Abdullah, A., Cudjoe, E., & Frederico, M. (2022). Social norms and family child labor: A systematic literature review. International Journal of Environmental Research and Public Health, 19(7), 4082. https://doi.org/10.3390/ijerph19074082 Adami, R., Kaldal, A., & Aspan, M. (Eds.). (2023). Children's rights from an international perspective. Brill. Ahmady, K. (2024). Gender differences in child labour: A systematic review of causes, forms, features and consequences. International Journal of Multidisciplinary Research and Growth Evaluation, 5(1), 947-956. Alam, A. (2023). The impact of Pakistan's child labour laws on the country. Journal of Regional Studies Review, 2(1), 10-25. https://doi.org/10.62843/jrsr/2023.79302 Close, P. (2025). Child labour in the global human rights regime: A political economy perspective. Emerald Publishing. European Union Agency for Fundamental Rights, & Council of Europe. (2022). Handbook on European law relating to the rights of the child (2022 ed.). Publications Office of the European Union. Gunnlaugsson, G. (2022). Work-free childhood to child labour and street life. Acta Paediatrica, 111(7), 1298-1300. https://doi.org/10.1111/apa.16347 Habib, R. R., El Khayat, M., Ghanawi, J., Katrib, R. S., Hneiny, L., & Halwani, D. A. (2024). Child labor and associated risk factors in the wake of the COVID-19 pandemic: A scoping review. Frontiers in Public Health, 11, 1240988. https://doi.org/10.3389/fpubh.2023.1240988 Habib, R. R., Katrib, R. S., Mikati, D., & Abouzeid, M. (2025). Growing risk of child labor in the Middle East and North Africa amid cascading crises. Frontiers in Public Health, 13, 1502687. https://doi.org/10.3389/fpubh.2025.1502687 International Labour Organization, & United Nations Children's Fund. (2021). Child labour: Global estimates 2020, trends and the road forward. ILO and UNICEF. International Labour Organization, & United Nations Children's Fund. (2025). Child labour: Global estimates 2024, trends and the road forward. ILO and UNICEF. Lakdawala, L. K., Martinez Heredia, D., & Vera-Cossio, D. A. (2025). The effects of expanding worker rights to children. Journal of Development Economics, 172, 103378. Lambon-Quayefio, M. P., & Owoo, N. S. (2021). Investigating the long-term effects of child labor on household poverty and food insecurity in Ghana. Journal of Demographic Economics, 87(4), 561-587. https://doi.org/10.1017/dem.2020.29 Mesfin, H., Cecchi, F., Nillesen, E., & Tirivayi, N. (2022). The effect of siblings' sex ratio on physical capital, human capital, and gendered time use among adolescents in Ethiopia. Economics and Human Biology, 47, 101182. https://doi.org/10.1016/j.ehb.2022.101182 Stark, O., & Budzinski, W. (2023). The demand for gratitude as a restraint on the use of child labor: A hypothesis. Journal of Demographic Economics, 89(1), 137-147. Wortsman, B., Bhattacharya, J., Lim, J., Tanoh, F., Shaheen, S., Ogan, A., & Jasinska, K. (2024). Accelerating progress towards eradicating child labour (SDG 8.7) with quality education (SDG 4): School quality is linked to reduced child cocoa labour in Cote d'Ivoire. Global Studies of Childhood, 14(3), 245-263. https://doi.org/10.1177/17454999241255182 #child_labour #Convention_138 #minimum_age #ILO #effective_abolition #national_policy #child_rights #SDG_8_7 #decent_work #social_protection #hazardous_work #compulsory_education #child_labour_law #end_child_labour #ILO_Convention_138

The Discrimination (Employment and Occupation) Convention, 1958 (No. 111) is one of the oldest and most widely ratified human rights instruments in the field of labour. It asks every ratifying State to declare and pursue a national policy that promotes equality of opportunity and treatment at work, and to remove laws and practices that treat people unfairly because of who they are rather than what they can do. This article offers a student friendly but rigorous study of the Convention. It uses a doctrinal method to explain the text of the instrument, the meaning of #discrimination under it, the duties it places on States, and the machinery the #International_Labour_Organization uses to check whether States keep their promises. The article then moves from law on paper to law in practice. Drawing on recent comparative research, it shows that large gaps remain between the promise of #equality_of_opportunity and the reality of the labour market for women, ethnic minorities, persons with disabilities, migrants, and lesbian, gay, bisexual, transgender, queer and intersex workers. It also examines newer threats that the drafters of 1958 could not have foreseen, above all #algorithmic_discrimination produced by automated hiring tools. The study argues that Convention No. 111 remains a living and flexible standard rather than a museum piece, mainly because its definition of discrimination can grow to cover new grounds and new forms of unfair treatment. At the same time, the article contends that the Convention's promise depends on active #enforcement, honest reporting, and the willingness of States to treat #workplace_equality as a matter of human rights rather than administrative box ticking. It closes with practical recommendations for legislators, employers, workers' organisations, and students of labour law. Keywords: ILO Convention No. 111; employment discrimination; equality of opportunity; international labour standards; non-discrimination law; algorithmic bias; decent work. 1. Introduction Work is more than a way to earn money. It shapes a person's dignity, social standing, and sense of belonging. When someone is refused a job, denied training, paid less, or passed over for promotion because of their skin colour, their religion, their sex, or their political views, the harm reaches far beyond the pay slip. It touches self respect, family security, and the ability to take part fully in society. For this reason, the fight against unfair treatment at work sits at the centre of modern labour law and of the wider human rights project. The main global instrument on this subject is the Discrimination (Employment and Occupation) Convention, 1958 (No. 111), adopted by the #International_Labour_Organization (ILO). In plain terms, the Convention requires ratifying States to put in place a national policy and, where needed, #legislation that prohibits unfair distinctions and exclusions in the world of work, and to repeal rules that are not built on the idea of equal chances (Carby-Hall and Tyc, 2022). It is one of the ILO's #fundamental_conventions, which means it carries special weight in the international system. By early 2023 it had been ratified by 175 of the ILO's 187 member States, making it one of the most accepted labour treaties ever drafted. This article studies #ILO_Convention_111 in a way that is useful for students yet faithful to the standards of a peer reviewed journal. It has three goals. The first is descriptive: to explain clearly what the Convention says, whom it protects, and what it demands of States. The second is analytical: to explore how the concept of #discrimination has been understood and stretched over time, and how the ILO checks compliance. The third is evaluative: to ask whether an instrument written in 1958 can still do useful work in a labour market reshaped by migration, automation, and the platform economy. The argument developed across the article is that Convention No. 111 has aged well, not because its authors predicted the future, but because they wrote an open ended standard. The Convention lists seven specific grounds of unfair treatment, yet it also builds in a mechanism for States to add more. This design has allowed national systems and ILO supervisory bodies to bring newer concerns, such as #disability, age, HIV status, and #sexual_orientation, within the reach of the equality principle. The weakness of the instrument lies not in its text but in the distance between formal commitment and daily practice, a distance that recent comparative studies measure in stark numbers (Heymann and others, 2023). The article proceeds as follows. Section 2 sets out the historical and institutional background. Section 3 explains the method. Section 4 analyses the normative content of the Convention article by article. Section 5 examines the duties of States. Section 6 unpacks the different forms that discrimination can take. Section 7 describes the ILO #supervisory_system. Section 8 places the Convention inside the framework of fundamental principles and rights at work. Section 9 reviews implementation in practice. Section 10 turns to regional and national approaches. Section 11 addresses contemporary challenges. Section 12 links the Convention to the Sustainable Development Goals. Section 13 discusses the findings, and Section 14 concludes with recommendations. 2. Historical and Institutional Background 2.1 The ILO and the making of international labour standards The #International_Labour_Organization was created in 1919 in the aftermath of the First World War, on the belief that lasting peace could only rest on social justice. The Preamble to its Constitution warned that conditions of labour involving injustice, hardship, and privation for large numbers of people produce unrest so great that the peace and harmony of the world are placed in danger. From the start the Organization was built on a distinctive feature known as #tripartism: governments, employers, and workers each hold seats and votes. This means that the standards the ILO produces are not handed down by diplomats alone but negotiated among the three groups whose lives the rules will touch. Standards take two main forms. Conventions are treaties that become binding on a State once it ratifies them. Recommendations are non binding texts that offer guidance and detail. Together they make up the body of international labour standards, covering an enormous range of subjects from hours of work and wages to safety, social security, and freedom of association. A Convention typically begins life as an item on the agenda of the annual International Labour Conference, is discussed over two successive years, and is then adopted if it wins a two thirds majority. This slow and deliberate process is meant to produce texts that command broad agreement. A guiding value of the Organization was set out in the #Declaration_of_Philadelphia of 1944, which reaffirmed the aims of the ILO at the end of the Second World War. It affirmed that all human beings, whatever their race, creed, or sex, have the right to pursue both their material well being and their spiritual development in conditions of freedom, dignity, economic security, and equal opportunity. That sentence, echoing the language of the emerging human rights movement, became the moral seed from which the equality Conventions grew. It tied the fight against discrimination directly to the founding purpose of the Organization. 2.2 The road to 1958 In the years after the Second World War, the international community turned its attention to equality with new energy. The Universal Declaration of Human Rights of 1948 proclaimed that all people are born free and equal in dignity and rights, that everyone has the right to work and to free choice of employment, and that all are entitled to these rights without distinction of any kind. This universal statement created pressure to translate broad principle into concrete labour standards. Within the ILO, the first concrete step in the field of pay was the Equal Remuneration Convention, 1951 (No. 100), which set out the principle of equal pay for men and women workers for work of equal value. Attention then broadened from pay to the whole employment relationship. It became clear that a woman could be paid the same rate as a man for a given job and still be shut out of that job in the first place, or denied the training needed to reach it. A wider instrument was required. The General Conference of the ILO met for its forty second session in Geneva in June 1958 and, on 25 June, adopted the Discrimination (Employment and Occupation) Convention, No. 111, together with an accompanying #Recommendation_111. The Convention entered into force on 15 June 1960 once the required ratifications had been deposited. Its purpose was to move beyond pay and to attack unfair treatment across the entire span of working life, from the training that opens the door to a trade, through hiring and terms of employment, to promotion and dismissal. It is worth remembering that this happened during a period of decolonisation and of sharp ideological division, when questions of race and social class were high on the world agenda. The Convention was, in part, a response to that moment. 2.3 The companion instruments Convention No. 111 does not stand alone. It works as a pair with Convention No. 100 on equal pay, and the two are usually treated together as the ILO's core equality standards. #Recommendation_111 gives fuller guidance on how a national policy against unfair treatment might be built, addressing matters such as access to training, the role of public employment services, and cooperation with employers' and workers' organisations. Later instruments extended the same spirit to particular groups, including the Workers with Family Responsibilities Convention, 1981 (No. 156), the Indigenous and Tribal Peoples Convention, 1989 (No. 169), and, more recently, the Violence and Harassment Convention, 2019 (No. 190). Convention No. 111 was also drawn upon during the drafting of the International Convention on the Elimination of All Forms of Racial Discrimination, showing how ILO standards fed into the wider United Nations human rights system. 2.4 A near universal standard Few labour treaties can claim the reach of Convention No. 111. Its #ratification by 175 States means that the great majority of the world's governments have accepted a binding duty to pursue #equality_of_opportunity at work. Ratification, however, is only the beginning. A treaty that is widely ratified but weakly applied can create a false sense of progress. Much of the analysis in later sections concerns exactly this gap between the impressive map of ratifications and the uneven reality on the ground. 3. Methodology This article uses doctrinal legal analysis, the standard method for the study of a treaty. Doctrinal analysis reads the primary legal text closely, in this case the Convention itself and its Recommendation, and interprets each provision in the light of its ordinary meaning, its context, and its purpose. This approach follows the general rules of treaty interpretation and pays particular attention to the way the ILO's own supervisory bodies have read the instrument over decades of practice. In international labour law the interpretations of the Committee of Experts carry special authority, because they represent a consistent and independent reading built up over nearly a century. The study draws on three layers of material. The first layer is the primary instruments: Convention No. 111, Recommendation No. 111, the ILO Constitution, and the Declaration on Fundamental Principles and Rights at Work of 1998 as amended in 2022. The second layer is the interpretive output of the #supervisory_system, in particular the observations and #General_Survey reports of the Committee of Experts on the Application of Conventions and Recommendations. The third layer is recent secondary scholarship published within roughly the last five years, which supplies comparative data and critical perspectives on how the equality principle operates in practice and how it is being tested by new technologies. The article does not carry out original empirical fieldwork. Its contribution is synthetic and interpretive: it gathers the legal architecture and the most recent evidence into a single accessible account, and it evaluates the strengths and limits of the Convention against that evidence. Where numerical findings are reported, they come from published comparative studies that are cited in full. This method suits the aim of the article, which is to give students a clear map of the law while showing where the law meets its hardest tests. A limitation of the approach is that it depends on the quality and coverage of existing studies, which are richer for some regions and some grounds than for others. 4. The Normative Architecture of Convention No. 111 4.1 The definition of discrimination The heart of the Convention is its definition in Article 1. In substance, the Convention treats as discrimination any distinction, exclusion, or preference that is based on certain listed grounds and that has the effect of nullifying or impairing equality of opportunity or #equal_treatment in #employment_and_occupation. Two features of this wording deserve close attention. First, the definition looks at effect, not only at intention. A rule or practice can amount to discrimination even if no one set out to harm anyone, provided that its result is to cancel out or weaken the equal chances of a protected group. This effect based approach is what allows the Convention to reach hidden and structural forms of unfair treatment, not just open hostility. It is one of the most far sighted features of the 1958 text, and, as later sections show, it turns out to be well suited to problems the drafters never imagined. Second, the definition uses three verbs, distinction, exclusion, and preference, which together cover both negative acts that shut people out and positive acts that favour one group over another. A preference for one religion in hiring is as much discrimination as an exclusion of another. The breadth of these words leaves little room for an employer to argue that a practice, because it takes the form of a preference rather than an outright bar, escapes the reach of the Convention. 4.2 The seven listed grounds Article 1 sets out seven grounds on which distinctions are prohibited: race, colour, sex, religion, #political_opinion, #national_extraction, and #social_origin. Each ground carries a rich meaning developed through decades of supervisory practice. Race and colour cover unfair treatment linked to descent, ethnic origin, and skin tone. In practice the two often overlap, and supervisory bodies read them broadly to capture the many ways in which ethnicity affects a person's chances at work. Sex covers both women and men and has been read to include pregnancy, childbirth, maternity, and marital status, and, in the reading of many bodies, matters connected to gender roles and expectations. Religion protects both belief and the outward practice of faith, and it requires reasonable room for religious observance at work, such as adjustments to dress rules or working time where these can be made without undue hardship. #Political_opinion protects the holding and expression of political views, subject to narrow limits where a particular view would genuinely conflict with the inherent demands of a sensitive post. #National_extraction refers to a person's place of birth, ancestry, or foreign origin, and is distinct from present nationality; it protects, for example, a citizen who is treated unfairly because of the country their family came from. #Social_origin concerns a person's class, #caste, or social background, a ground of great importance in societies where birth still shapes access to work and where deeply rooted hierarchies continue to limit opportunity. 4.3 The door to additional grounds The list of seven is not the end of the story. Article 1 also allows a ratifying State, after consulting representative employers' and workers' organisations and other appropriate bodies, to treat other distinctions as prohibited discrimination. This is the mechanism that keeps the Convention alive. Through it, national systems have added grounds such as age, #disability, HIV status, family responsibilities, #sexual_orientation, #gender_identity, and trade union membership. The Convention therefore sets a floor of seven grounds while leaving the ceiling open. This is why the popular summary that the Convention prohibits unfair treatment on any basis is broadly true in spirit, even though the binding core is the seven named grounds plus whatever a State chooses to add through the proper procedure. 4.4 The exception for inherent job requirements Not every distinction is unlawful. Article 1 provides that any distinction, exclusion, or preference for a particular job that is based on the inherent requirements of that job is not discrimination. The idea is narrow and must be read strictly. It allows an employer to require, for example, that an actor playing a specific role be of a particular sex, or that a teacher of a faith share that faith in a genuinely religious school. The exception cannot be stretched into a general licence to discriminate. The requirement must belong to the job itself, not to the prejudices of the employer, the preferences of customers, or general assumptions about a group. Supervisory bodies have repeatedly warned against loose use of #inherent_job_requirements as a cover for exclusion, insisting that the requirement be examined for each specific post rather than applied to a whole category of work. 4.5 The scope: employment and occupation Article 1 also defines how far the Convention reaches. The terms employment and occupation include access to #vocational_training, access to employment and to particular occupations, and the terms and conditions of employment. This is a wide span. It covers the training that prepares a person for work, the moment of hiring, the choice of trade or profession, and the whole life of the job, including pay, hours, benefits, promotion, discipline, and dismissal. The Convention thus protects a person from the classroom door to the exit interview. The inclusion of access to training is especially important, because unequal access to skills at an early stage locks in disadvantage for the rest of a working life. 4.6 State security and special measures Two further provisions define what is not discrimination. Article 4 states that measures affecting a person who is justifiably suspected of, or engaged in, activities prejudicial to the security of the State are not discrimination, provided that the person has a right of appeal to a competent body. This clause must be handled with care, because a broad reading could be used to punish people for lawful #political_opinion or for trade union activity. Supervisory bodies insist that it be confined to genuine and individual security concerns, and that the right of appeal be real and effective. Article 5 makes clear that #special_measures of protection or assistance are not discrimination. This covers protective measures recognised in other ILO instruments and, importantly, measures designed to meet the particular needs of groups that face disadvantage. In modern language this is the space for #affirmative_action or positive action: steps that help under represented groups reach genuine equality, such as targeted training, outreach, or, in some systems, temporary preferences in hiring. Far from breaching the equality principle, such measures serve it, because formal equal treatment on its own can leave existing gaps untouched. 5. State Obligations: National Policy and the Legislative Duty 5.1 The core duty in Article 2 The operative promise of the Convention lies in Article 2. Each ratifying State undertakes to declare and pursue a #national_policy designed to promote, by methods suited to national conditions and practice, equality of opportunity and treatment in #employment_and_occupation, with a view to eliminating discrimination. Two words in that sentence carry great weight. To declare means the State must make a clear public commitment, so that the goal of equality is written into national policy rather than left to chance. To pursue means it must actually take steps, and keep taking them, rather than resting on a paper declaration. The duty is ongoing and cannot be discharged once and forgotten. 5.2 The measures in Article 3 Article 3 spells out the kinds of steps a State must take to give effect to its policy. These include seeking the cooperation of employers' and workers' organisations, enacting #legislation and promoting educational programmes, repealing statutory provisions and modifying administrative instructions that are inconsistent with the policy, and pursuing the policy in respect of employment under the direct control of a national authority, among other things. The legislative duty is central. States are expected to prohibit discrimination in law and to remove existing rules that conflict with equality, including old laws that once barred women from certain occupations or reserved posts for particular groups. Yet the Convention is careful not to reduce the fight against discrimination to legislation alone. Education, awareness raising, #social_dialogue, and the practical conduct of public employment services all form part of the required national effort. A law that sits unenforced on the statute book changes little; the Convention envisages a whole policy, not a single statute. 5.3 The nature of the obligation An important question for students is how demanding this obligation is. The Convention does not require a State to abolish all discrimination overnight, which no State on earth has achieved. Instead it requires a genuine, active, and continuing national policy backed by concrete measures. The ILO's approach starts from the realistic premise that no country is fully free of discrimination, because prejudice is rooted in culture and politics and shifts only slowly, and because the very idea of what counts as unacceptable discrimination keeps expanding as social understanding grows. For this reason the supervisory bodies focus on the direction of travel: is the State moving forward, taking proactive steps, and closing gaps, or is it standing still or sliding backward. This progressive but purposeful reading gives the Convention both flexibility and bite. It respects national differences while refusing to accept inaction as compliance. 6. The Many Faces of Discrimination To apply the Convention well, one must understand that unfair treatment wears several faces. The instrument's effect based definition is broad enough to capture all of them, and supervisory practice has drawn out the distinctions. 6.1 Direct discrimination #Direct_discrimination is the most visible form. It occurs when a rule or act treats a person less favourably, openly and explicitly, because of a protected ground. A job advertisement that says only men may apply, or a pay scale that pays members of one ethnic group less for the same work, is direct discrimination. Because it is open, it is often the easiest form to identify, though not always the easiest to prove, since employers rarely admit the true reason for a decision and may offer a neutral explanation to disguise a biased one. 6.2 Indirect discrimination #Indirect_discrimination is subtler and, in modern economies, more common. It occurs when a rule, condition, or practice looks neutral on its face but in fact places a protected group at a particular disadvantage, without a good and proportionate justification. A minimum height requirement for a job may seem neutral, yet it can screen out most women and members of some ethnic groups even though height has nothing to do with the ability to do the work. A requirement of continuous full time service for promotion can disadvantage those, mostly women, who have taken time out to care for children. The effect based definition of the Convention is well suited to catching indirect discrimination, because it looks at outcomes rather than labels. 6.3 Structural and systemic discrimination Beyond individual acts lies #structural_discrimination, sometimes called systemic discrimination. This is the pattern of disadvantage built into the ordinary workings of institutions, culture, and history. It shows up as the crowding of women into low paid sectors, the near absence of certain minorities from senior posts, or the way informal recruitment through personal networks quietly reproduces an existing workforce. Recent empirical work confirms that systemic patterns of unequal treatment persist even among large and sophisticated employers, and that these patterns are stable and measurable rather than random (Kline, Rose and Walters, 2022). Tackling structural discrimination usually requires more than a prohibition. It calls for positive measures, monitoring of outcomes, and sustained effort of the kind Article 2 demands. 6.4 Harassment as discrimination Supervisory practice has also confirmed that #harassment, including sexual harassment, is a form of discrimination within the meaning of the Convention. A hostile working environment that targets a person because of sex, race, or another protected ground impairs their #equal_treatment as surely as a refusal to hire. This reading connects Convention No. 111 to the newer Violence and Harassment Convention, 2019 (No. 190), which addresses violence and harassment in the world of work directly and recognises that such conduct is both a human rights violation and a threat to equal opportunity. 6.5 The problem of proof A recurring difficulty in all of these forms is proof. The person who suffers discrimination rarely has access to the evidence needed to establish it, because the reasons for a decision lie with the employer. For this reason many legal systems shift the #burden_of_proof once the complainant has shown facts from which discrimination may be presumed, requiring the employer to prove that the decision had a lawful and non discriminatory basis. This shift does not reverse the ordinary rules lightly, but it recognises the imbalance of information between the two sides. The question of proof becomes even sharper when decisions are made by automated systems, as later sections explain. 7. The ILO Supervisory System A treaty is only as strong as the machinery that checks compliance. The ILO has one of the most developed supervisory systems in international law, and Convention No. 111, as a fundamental instrument, receives close attention within it. 7.1 The regular system and the Committee of Experts Once a State ratifies a Convention it must report regularly on the measures it has taken to apply it. For the #fundamental_conventions these reports are due at relatively short intervals. The reports are examined by the Committee of Experts on the Application of Conventions and Recommendations, known as the #CEACR. Set up in 1926, the Committee is composed of twenty independent legal experts drawn from different regions and legal traditions, appointed in their personal capacity, who provide an impartial technical assessment of how far national law and practice conform to ratified Conventions. Because the members come from many different legal, economic, and social systems, the Committee can read a Convention against a wide range of national realities. The Committee expresses itself in two main ways. Observations are published comments on more serious or long standing matters. Direct requests are more technical points sent to the government concerned. Employers' and workers' organisations may submit their own comments on how a Convention is applied, and these often bring problems to light that a government report would not mention. Although the Committee cannot impose sanctions, its careful and consistent readings carry great authority and shape how the Convention is understood worldwide. Over time it has functioned less like a simple advisory body and more like a quasi judicial one, whose interpretations guide States, courts, and scholars alike. 7.2 The Conference Committee The Committee of Experts reports each year to the International Labour Conference, where a tripartite body known as the Conference Committee on the Application of Standards selects a number of cases for public discussion. Governments named in these cases are invited to explain themselves before representatives of governments, employers, and workers. This public scrutiny adds a political layer to the technical work of the experts and can generate real pressure for change, because no government wishes to be named repeatedly among the most serious cases. The interplay between the technical assessment of the experts and the political forum of the Conference is a defining feature of the ILO system. 7.3 General Surveys Under Article 19 of the ILO Constitution the Committee of Experts also prepares an in depth #General_Survey on a chosen theme, covering the law and practice of member States whether or not they have ratified the relevant Convention. In 2023 the Committee published a General Survey on achieving #gender_equality at work, which examined the equality and non discrimination instruments and gave an up to date picture of progress and obstacles across the world (International Labour Office, 2023). General Surveys are among the most valuable sources for anyone studying how an equality standard actually operates, because they gather national experiences into a single comparative view and identify the recurring barriers to equality. 7.4 Complaints and representations Alongside the regular system, the ILO Constitution provides special procedures. Under Article 24, an employers' or workers' organisation may make a representation alleging that a State has failed to observe a Convention it has ratified. Under Article 26, a member State, a Conference delegate, or the Governing Body may lodge a complaint, which can lead to a Commission of Inquiry, the most serious investigative step in the system. These procedures are used sparingly, and most supervision runs through the regular reporting cycle, but they exist as a backstop when ordinary supervision is not enough. Together, the regular and special mechanisms form a layered system that relies more on transparency, dialogue, and reputation than on coercion. 8. Convention No. 111 within the Fundamental Rights Framework Convention No. 111 gained added significance in 1998, when the ILO adopted the Declaration on Fundamental Principles and Rights at Work. The #1998_Declaration singled out a small set of principles as so basic that every member State is bound to respect, promote, and realise them simply by virtue of belonging to the Organization, whether or not it has ratified the specific Conventions. The elimination of discrimination in respect of employment and occupation is one of these principles, giving Convention No. 111 a reach beyond its already large list of ratifications (Trebilcock, 2023). In effect, even a State that has not ratified the Convention carries an obligation to move toward its underlying principle. The original framework recognised four categories of fundamental principles: freedom of association and collective bargaining, the elimination of forced labour, the abolition of child labour, and the elimination of discrimination in employment and occupation. These were expressed through eight core Conventions, of which No. 111 and No. 100 formed the equality pillar. The 1998 Declaration was accompanied by a follow up mechanism, under which the Director General prepares periodic global reports and States report on efforts to realise the principles they have not ratified. The framework was widened in 2022. At its 110th Session, the International Labour Conference amended the #1998_Declaration to add a fifth principle, a safe and healthy working environment, and designated two occupational safety and health Conventions as fundamental (Politakis, 2023; Trebilcock, 2023). This brought the number of core Conventions to ten and reflected a growing understanding that the right to work in safety is itself a basic right. The process had begun in 2019, when the ILO Centenary Declaration for the Future of Work recognised that safe and healthy working conditions are fundamental to #decent_work. The 2022 amendment did not change Convention No. 111, but it confirmed the direction in which the whole framework is moving, treating an expanding set of labour protections as fundamental #human_rights rather than optional extras. The connection between decent work and equality runs through all of it: work cannot be decent if it is distributed unfairly, and a workplace cannot be safe for everyone if some groups are exposed to greater risk or excluded from protection. Placing freedom from discrimination among these fundamental principles keeps it out of the ordinary bargaining over economic policy, where social rights are often the first to be sacrificed. 9. Implementation in Practice: Progress and Persistent Gaps The true test of a Convention is not the elegance of its text but the condition of the workers it is meant to protect. Recent comparative research, much of it built on the ILO's own legal databases, paints a mixed picture of real progress and stubborn gaps. 9.1 Race and ethnicity A systematic study of legal protections across all 193 United Nations member States found that formal protection against #racial_discrimination in employment is still far from universal. More than one in five countries do not explicitly prohibit racial discrimination at work, dozens fail to guarantee equal pay regardless of race, and a large majority of those that do prohibit discrimination stop short of requiring employers to take active steps to prevent it (Heymann and others, 2023). The lesson is sobering. A prohibition on paper, without duties of prevention and without #enforcement, leaves much of the promise of Convention No. 111 unfulfilled. It also shows that ratification alone is a weak measure of protection, since many States that have ratified the Convention still lack detailed and effective national rules. 9.2 Gender Sex remains one of the most persistent grounds of unfair treatment worldwide. Women continue to be concentrated in lower paid sectors, to carry the larger share of unpaid care work, to face a pay gap for work of equal value, and to hit barriers to senior positions often described as a glass ceiling. The 2023 General Survey on #gender_equality at work confirmed that, despite decades of law reform, equality of opportunity between women and men is still a goal rather than an achievement in most of the world (International Labour Office, 2023). Pregnancy and maternity remain frequent triggers of unfair treatment, from refusal to hire to dismissal, and the design of many workplaces still assumes a worker with no caring responsibilities. Progress has been real but slow, and gains can be reversed in times of economic crisis. 9.3 Religion Religion sits among the seven named grounds, yet it raises distinctive questions. Protecting religious freedom at work sometimes calls for adjustments, such as flexibility over dress, prayer time, or religious holidays. At the same time, employers and States must balance the beliefs of one worker against the rights of others and against the genuine needs of the job. Supervisory practice asks whether a rule that burdens religious practice is truly necessary or whether a reasonable accommodation could remove the conflict. The ground of religion therefore illustrates how the equality principle often requires not identical treatment but sensitivity to difference. 9.4 Disability Persons with disabilities face some of the widest employment gaps of any group. International understanding has shifted from a charity model to a rights model, under which the failure to provide #reasonable_accommodation is itself a form of discrimination. Guidance from United Nations human rights bodies on the right of persons with disabilities to work has reinforced this reading. Convention No. 111, through the mechanism for additional grounds and through the concept of #indirect_discrimination, provides a solid legal basis for protecting disabled workers, even though disability is not one of the seven named grounds. Reasonable accommodation, such as adjusting equipment, tasks, or hours, allows a worker to compete on genuinely equal terms, and denying it without good reason places the worker at a disadvantage that the Convention's effect based logic captures. 9.5 Sexual orientation and gender identity Protection for lesbian, gay, bisexual, transgender, queer, and intersex workers has expanded rapidly in some regions and stalled in others. A comparative study of national laws found that by 2023, 71 countries prohibited at least some workplace discrimination based on sexual orientation, 41 did so based on #gender_identity, and far fewer addressed gender expression or sex characteristics (Weistroffer and others, 2025). The trend line points upward, but coverage remains patchy, and in many States such workers enjoy no explicit protection at all, or even face laws that criminalise them. Because these grounds are not named in Article 1, their protection depends on States choosing to add them, or on courts and supervisory bodies reading them into broader concepts such as sex. Field experiments in several countries continue to show that applicants perceived as gay, lesbian, or transgender receive fewer interview invitations, which confirms that the problem is practical and not merely theoretical. 9.6 Migrant workers, the informal economy, and the global South Two structural features of the world economy strain the equality principle. The first is migration. #Migrant_workers often face discrimination on grounds of #national_extraction, colour, and religion, and they are frequently concentrated in the most precarious and least protected jobs. Language barriers, unfamiliarity with local rights, and dependence on employers for visa status all deepen their vulnerability. The second is informality. In much of the world, most workers are in the #informal_economy, without written contracts, legal protection, or social security. Labour law, including anti discrimination law, was designed with formal employment in mind, and it reaches informal workers only weakly. Scholars have long noted a mismatch between the assumptions of labour law and the realities of the developing world, where the effectiveness of equality legislation in improving the lives of the most disadvantaged is far from guaranteed (Carby-Hall and Tyc, 2022). Women are over represented in informal work, so the failure to reach the informal economy also deepens gender inequality. Closing this gap is one of the central challenges for the future of the Convention, and it requires thinking beyond the standard employment relationship toward protections that follow the worker rather than the contract. 10. Regional and National Approaches The global standard set by Convention No. 111 does not operate in a vacuum. It is put into effect through regional systems and national laws, which vary widely in detail and strength. At the regional level, the European Union has built one of the most developed bodies of #non_discrimination_law in the world. A series of directives requires equal treatment in employment and occupation across grounds including sex, racial or ethnic origin, religion or belief, disability, age, and sexual orientation, and they define both direct and indirect discrimination in binding terms. More recent measures address pay transparency and the balance of women and men on company boards. These directives give the broad principle of Convention No. 111 sharp and enforceable content within the Union, and they are backed by a court that can rule on their meaning. Regional human rights courts and commissions in Europe, the Americas, and Africa have also contributed, treating freedom from discrimination as a fundamental right and, in some cases, hearing complaints about unequal treatment at work. National constitutions in many countries guarantee equality and prohibit discrimination, and dedicated equality bodies or ombudsman offices have been created to receive complaints, conduct inquiries, and support victims. The strength of these institutions varies. Some have real powers to investigate and to bring cases; others are advisory and under resourced. The comparison across systems yields a clear lesson. The countries that make the most progress tend to combine several elements: a clear prohibition covering a wide range of grounds, positive duties on employers to prevent discrimination and to accommodate difference, an accessible route to remedy with fair rules on proof, and an independent body with the power and resources to act. Where one of these elements is missing, protection weakens. Convention No. 111 sets the direction, but the detailed design of national systems determines whether the promise is kept. This is why the supervisory bodies pay such close attention not only to whether a State prohibits discrimination but to how, and with what practical effect. 11. Contemporary Challenges The drafters of 1958 worked in a world of factories, offices, and printed job advertisements. Today the equality principle must operate in a labour market transformed by digital technology and by new forms of work. Several developments in particular test the Convention. 11.1 Algorithmic and AI driven discrimination Employers increasingly use automated tools to screen applications, rank candidates, and even analyse video interviews. These #artificial_intelligence hiring systems are often sold as neutral and efficient. In practice they can reproduce and even magnify existing bias, because they learn from historical data that reflects past discrimination. A model trained on the record of a workforce that was mostly male may quietly downgrade the applications of women, and a model that uses proxies such as postcode or membership of a particular group may end up sorting candidates by race or another protected ground (Gerards and Xenidis, 2021; Fabris and others, 2024). Empirical study of how employers actually deploy these systems shows that the risks are not hypothetical. A recent socio legal analysis of AI hiring tools found that the way they are used in practice can create real and hard to detect discrimination, and that many protected groups lack the digital access needed even to reach the opportunities these systems control (Sheard, 2025). Legal scholars have highlighted a further problem: proving #algorithmic_discrimination is unusually difficult, because the decision is buried in code and data that the applicant cannot see, which weakens the ordinary tools of #non_discrimination_law and strains the usual rules on the burden of proof (Kelly-Lyth, 2023; Carter, 2024; Gerards and Zuiderveen Borgesius, 2022). High profile disputes, including litigation over automated screening tools that allegedly filtered out older applicants and members of other protected groups, show that these questions are already before the courts. For Convention No. 111 the message is twofold. On one hand, its effect based definition of discrimination is well suited to the digital age, because it looks at outcomes rather than intentions, and an algorithm has no intentions in the human sense. On the other hand, the Convention's promise cannot be kept unless States update their enforcement tools, require transparency and independent auditing of automated systems, and place the burden of explanation on those who deploy them. Without such steps, a technology sold as objective may entrench old prejudices behind a screen of apparent neutrality. 11.2 Platform and gig work The rise of #platform_work has produced millions of jobs that sit uneasily between employment and self employment. Workers who find tasks through apps may be classified as independent contractors and so fall outside much protective legislation, including anti discrimination rules that apply only to employees. Ratings systems and task allocation algorithms can also transmit customer prejudice directly to the worker, for example when biased ratings reduce a worker's access to future jobs. A driver or delivery worker who is rated poorly because of their accent or appearance may lose income without ever learning why. Extending the equality principle to these new arrangements is a live question for legislators and for the ILO, and it connects to a wider debate about how to classify and protect platform workers. 11.3 Intersectionality Real people do not experience discrimination one ground at a time. A migrant woman may face disadvantage that is not simply the sum of discrimination based on sex and discrimination based on national extraction, but a distinct combination that neither ground captures alone. This idea of #intersectionality has become central to modern equality thinking. The open texture of Convention No. 111, and its focus on the effect of a distinction, gives supervisory bodies room to recognise intersecting forms of disadvantage, though national laws often lag behind, treating each ground in a separate silo and missing the combined harm. 11.4 Emerging grounds The list of grounds continues to grow at national level. Age, genetic information, health status including HIV, family and care responsibilities, and criminal record are all becoming grounds of protection in various systems. The mechanism in Article 1 that lets States add grounds after consultation with the social partners means the Convention can absorb these developments without needing to be rewritten. This adaptability is the quiet genius of the 1958 text, and it explains why an instrument of that age can still speak to concerns that did not exist when it was drafted. 11.5 Crisis, remote work, and the changing workplace Recent global shocks have reshaped where and how people work. The spread of remote and hybrid work has opened opportunities for some, including certain workers with disabilities and those with caring duties, while creating new risks of unequal treatment, for example when workers who are not physically present are overlooked for promotion. Economic crises tend to hit already disadvantaged groups hardest and can widen existing gaps. These shifts do not require a new Convention, but they do require States and employers to apply the old principle with fresh attention, watching for the ways in which new working patterns can quietly reproduce old inequalities. 12. Convention No. 111 and the Sustainable Development Goals The equality principle also connects to the broader development agenda. The Sustainable Development Goals adopted by the international community place #decent_work and reduced inequality at the centre of a shared vision for the future. Goal 8 calls for decent work and economic growth for all, including full and productive employment and equal pay for work of equal value. Goal 5 seeks gender equality and the empowerment of all women and girls, including in economic life. Goal 10 aims to reduce inequality within and among countries, in part by ensuring equal opportunity and eliminating discriminatory laws and practices. Convention No. 111 supplies the legal backbone for these commitments in the field of work. Progress toward the #Sustainable_Development_Goals therefore depends, in part, on the effective application of the Convention, and the data gathered to track the Goals in turn helps to reveal where the equality principle is failing in practice. The two agendas reinforce one another: the Goals give political momentum to the Convention, and the Convention gives legal precision to the Goals. 13. Discussion Bringing the threads together, several points stand out for evaluation. The first is that Convention No. 111 is stronger as a standard than as a guarantee. Its normative content is excellent. The effect based definition, the wide scope covering training, hiring, and conditions, the narrow reading of exceptions, and above all the open mechanism for adding grounds, together make it one of the most durable equality instruments ever drafted. Its weakness lies downstream, in implementation. The comparative evidence shows that formal prohibition is common but preventive duties and effective enforcement are rare, and that whole categories of workers, in the #informal_economy, in migration, and now in #platform_work, remain beyond the practical reach of the law (Heymann and others, 2023). The second point concerns the relationship between the seven named grounds and the wider promise of equality. The popular description of the Convention, as requiring States to prohibit unfair treatment on any basis, is inspiring but slightly misleading. The binding core is the seven grounds; everything beyond depends on States choosing to expand protection through the proper procedure. This is both a strength, because it respects national #social_dialogue and local conditions, and a limit, because it allows some States to leave vulnerable groups such as LGBTQI workers without explicit protection. The uneven global map of protection for sexual orientation and gender identity illustrates the point (Weistroffer and others, 2025). The third point is that the #supervisory_system, though it lacks hard sanctions, is the engine that keeps the Convention meaningful. The steady work of the #CEACR, the public exposure of the Conference Committee, and the periodic General Survey reports turn a static text into a living body of interpretation. The value of this system depends, however, on honest reporting by States and on the active participation of workers' and employers' organisations. Where social dialogue is weak or where trade unions are suppressed, the system loses much of its power, and problems can remain hidden from view. The fourth point looks to the future. The greatest new test facing the equality principle is technological. #Algorithmic_discrimination threatens to make unfair treatment more widespread and less visible at the same time. The effect based logic of Convention No. 111 is a good match for this problem in principle, but principle is not enough. States will need new rules on transparency, auditing, and liability, and they will need to adjust the rules on proof so that an applicant harmed by an opaque system is not left with an impossible task (Kelly-Lyth, 2023; Carter, 2024). The ILO, with its tripartite structure, is well placed to help build a shared global response, but the pace of technology is fast and the pace of standard setting is slow, and this mismatch is itself a challenge. A final reflection concerns the identity of the Convention as a human rights instrument. Its inclusion among the ILO's fundamental principles, reaffirmed and extended through the 1998 Declaration and the 2022 amendment, signals that freedom from discrimination at work is not a matter of economic policy to be traded away, but a basic entitlement of every worker (Politakis, 2023; Trebilcock, 2023). Keeping this framing at the centre of national debate is perhaps the most important task of all, because it is the framing that resists the temptation to treat equality as a luxury that can wait for better times. 14. Conclusion and Recommendations The Discrimination (Employment and Occupation) Convention, 1958 (No. 111) has proved to be one of the most enduring achievements of international labour law. Written more than six decades ago, it still supplies the basic grammar of #workplace_equality: a broad, effect based idea of discrimination, a wide field of application, a duty on States to declare and pursue an active #national_policy, and a flexible mechanism that lets protection grow as social understanding grows. Its near universal #ratification and its status as a fundamental instrument give it a moral and legal weight that few treaties can match. Yet the study also shows the limits of law on paper. Deep gaps remain for women, ethnic minorities, persons with disability, LGBTQI workers, migrants, and the vast population of the informal economy. New technologies threaten to open fresh fronts of unfair treatment that are harder to see and harder to prove. The Convention's promise will be realised only if States move from prohibition to prevention, from declaration to sustained action, and from paper equality to measured outcomes. Several practical recommendations follow from the analysis. First, States that have not yet done so should expand their national grounds of protection to include disability, age, HIV status, sexual orientation, and gender identity, using the mechanism the Convention already provides. Second, legislation should impose positive duties of prevention on employers, not merely prohibitions, and should support them with clear guidance and effective enforcement. Third, national systems should modernise their non-discrimination law to address algorithmic discrimination, requiring transparency and independent auditing of automated hiring tools and easing the burden of proof for those affected. Fourth, protection should be extended in practical terms to #platform_work and to the #informal_economy, so that the equality principle reaches the workers who need it most, including through protections that attach to the person rather than to a particular contract. Fifth, States should strengthen #social_dialogue and protect the ability of workers' organisations to take part in the supervisory system, since honest reporting and independent scrutiny are what keep the Convention alive. Sixth, equality bodies should be given real powers and adequate resources, so that a worker who suffers discrimination has a genuine route to remedy rather than a right that exists only on paper. For students of labour law, Convention No. 111 offers a rich case study in how a well drafted standard can outlast the world that produced it. It shows that the strength of a human rights instrument lies not only in what it forbids today but in its capacity to keep learning. The task for the next generation of lawmakers, judges, employers, workers, and scholars is to make sure that this capacity is used, so that the simple and powerful idea at the heart of the Convention, that people should be judged at work by what they can do and not by who they are, becomes a lived reality rather than a distant aspiration. References Carby-Hall, J., and Tyc, A. (eds.) (2022). Discrimination and Employment Law: International Legal Perspectives. London: Routledge. Carter, C. (2024). Why the algorithmic recruiter discriminates: The causal challenges of data-driven discrimination. Maastricht Journal of European and Comparative Law. https://doi.org/10.1177/1023263X241248474 Fabris, A., et al. (2024). Fairness and bias in algorithmic hiring: A multidisciplinary survey. ACM Transactions on Intelligent Systems and Technology, 15(4). Gerards, J., and Xenidis, R. (2021). Algorithmic Discrimination in Europe: Challenges and Opportunities for Gender Equality and Non-Discrimination Law. Luxembourg: Publications Office of the European Union. Gerards, J., and Zuiderveen Borgesius, F. (2022). Protected grounds and the system of non-discrimination law in the context of algorithmic decision-making and artificial intelligence. Colorado Technology Law Journal, 20(1). Heymann, J., Varvaro-Toney, S., Raub, A., Kabir, F., and Sprague, A. (2023). Race, ethnicity, and discrimination at work: A new analysis of legal protections and gaps in all 193 UN countries. Equality, Diversity and Inclusion: An International Journal, 42(9), pp. 16-34. https://doi.org/10.1108/EDI-01-2022-0027 International Labour Office (2023). Achieving Gender Equality at Work. General Survey concerning the equality and non-discrimination instruments. International Labour Conference, 111th Session. Geneva: International Labour Office. Kelly-Lyth, A. (2023). Algorithmic discrimination at work. European Labour Law Journal, 14(2). https://doi.org/10.1177/20319525231167300 Kline, P., Rose, E. K., and Walters, C. R. (2022). Systemic discrimination among large US employers. The Quarterly Journal of Economics, 137(4), pp. 1963-2036. Politakis, G. P. (2023). The recognition of occupational safety and health as a fundamental principle and right at work. International and Comparative Law Quarterly, 72(1), pp. 217-226. Sheard, N. (2025). Algorithm-facilitated discrimination: A socio-legal study of the use by employers of artificial intelligence hiring systems. Journal of Law and Society. https://doi.org/10.1111/jols.12535 Trebilcock, A. (2023). Introductory note to the 2022 amendments to the ILO Declaration on Fundamental Principles and Rights at Work. International Legal Materials, 62(4). Weistroffer, R. L., Raub, A., Sprague, A., and Heymann, J. (2025). Mapping progress in workplace protections: A study of global LGBTQI+ workplace antidiscrimination legislation in 2016 and 2023. Sexuality Research and Social Policy. https://doi.org/10.1007/s13178-024-01074-4 #ILO_Convention_111 #employment_discrimination equality of opportunity non-discrimination law #international_labour_standards #decent_work workplace equality #fundamental_rights_at_work #CEACR #gender_equality algorithmic discrimination #labour_law #human_rights_at_work #social_justice #No_111_1958

The Abolition of Forced Labour Convention, 1957 (No. 105), is one of the core standards of the International Labour Organization. Its job is narrow but powerful. It bans states from using #forced_labour for certain political and economic ends, above all as a means of #political_coercion and as a punishment for holding or expressing political views that go against the established order. This article explains what Convention No. 105 says, why it was written the way it was, and how it still matters today. Using a doctrinal reading of the text together with recent scholarship and ILO supervisory materials from 2021 to 2025, the study sets out the five specific purposes for which #compulsory_labour is prohibited, and it shows how the Convention builds on, rather than replaces, the older Forced Labour Convention, 1930 (No. 29). The article then examines how compliance is watched through the ILO supervisory system, and how the Convention connects to wider #human_rights protections such as freedom of expression and freedom of association. It closes by discussing live problems: state-imposed forced labour tied to political dissent, the treatment of #prison_labour, and the growing use of trade and supply-chain rules to fight #modern_slavery. The article argues that Convention No. 105 remains a clear moral and legal line against the use of work as a weapon of the state, even as new forms of coercion test its limits. The analysis is written for students and early researchers who need a plain but rigorous entry point into the topic. Keywords: forced labour; International Labour Organization; Convention No. 105; political coercion; fundamental rights; prison labour; supervisory system; modern slavery 1. Introduction Work is normally something people choose. They agree to it, they are paid for it, and they can leave it. #Forced_labour is the opposite of this. It is work taken from a person by threat or pressure, work that the person never freely agreed to do. When a state itself organises this kind of labour, and does so to silence critics, punish strikers, or push through economic plans, the harm reaches beyond the individual worker. It strikes at basic freedoms that a fair society is supposed to protect. The Abolition of Forced Labour Convention, 1957 (No. 105), speaks directly to this problem. It is a short treaty of ten articles, but its message is sharp. It tells governments that certain uses of compulsory labour are never acceptable, no matter the reason given. Among these, the first and most famous is the ban on forced labour as a means of #political_coercion or education, and as a punishment for people who hold or express #political_views opposed to the established political, social, or economic system (ILO, 2024a). Convention No. 105 did not appear out of nowhere. It was adopted in Geneva on 27 June 1957, during the fortieth session of the International Labour Conference, and it entered into force on 17 January 1959. It came at a tense moment in world history. The Cold War was reshaping global politics, decolonisation was under way across Asia and Africa, and detailed reports of large forced-labour systems in several states had pushed the international community to act. The Convention was designed to close gaps left by the earlier Forced Labour Convention, 1930 (No. 29), and to name clearly the political and economic purposes for which forced labour must be abolished. Today, Convention No. 105 sits among the small group of #fundamental_rights instruments of the ILO. These are the standards that every member of the Organization is expected to respect because of the values they carry, whether or not the state has formally ratified them. Convention No. 105 is also one of the most widely ratified treaties in the whole ILO system, accepted by around 180 of the Organization's 187 member states (Jarman and Heard, 2023). That near-universal acceptance tells us something. On paper at least, almost every government in the world agrees that people should not be forced to work as a punishment for what they think or say. Yet the gap between paper and practice is the reason the Convention still needs study. Modern slavery has not disappeared. The most recent global estimates suggest that around 28 million people were in forced labour in 2021, an increase of 2.7 million compared with 2016 (ILO, Walk Free and IOM, 2022). A share of that total is imposed directly by states, which is exactly the kind of coercion the 1957 Convention targets. New pressures, from prison work programmes to global supply chains, keep raising fresh questions about where the line should fall. This article has three aims. The first is to explain the content and legal structure of Convention No. 105 in clear language. The second is to place the Convention in its historical and human-rights context, so that students understand not only what it says but why it says it. The third is to examine the modern debates that surround it, so that readers can see the instrument as a living part of international law rather than a museum piece. The discussion is organised to move from text, to system, to context, and finally to open problems. Throughout, the aim is to keep the language simple while respecting the legal detail that makes the topic worth studying. 2. Method and Approach This study uses a doctrinal method combined with a focused review of recent literature. A doctrinal method means reading the legal text closely, working out what each provision requires, and setting the provisions against one another to see how they fit together. For an international treaty, this reading is guided by the ordinary meaning of the words, the purpose of the instrument, and the way expert bodies have interpreted it over time. In the case of the ILO, those expert bodies produce a large body of interpretation, and their comments give the treaty text a settled meaning that goes well beyond the bare words. Alongside the text, the article draws on secondary sources published mainly between 2021 and 2025, so that the analysis reflects current thinking. These include scholarly books and journal articles on forced labour, prison labour, and labour rights, as well as official ILO reports and supervisory documents. Where figures on the scale of forced labour appear, they come from the joint global estimates produced by the ILO and its partners, which are the standard reference in the field (ILO, Walk Free and IOM, 2022). Where the article discusses the meaning of the Convention itself, it relies on the recent general observation of the ILO expert committee, which restates the current interpretation (ILO, 2024a). The article does not carry out empirical fieldwork, and it does not test hypotheses with data. Its contribution is interpretive and educational. It aims to gather what is known about Convention No. 105 into a single, readable account that respects the legal detail while remaining open to a wide student audience. Readers who wish to go deeper can follow the references, all of which are recent and widely available in university libraries. This design suits a topic where the primary sources are treaty texts and official reports rather than surveys or experiments, and where the main task is to make a complex legal framework understandable. 3. Historical Background and Drafting Context To understand Convention No. 105, we have to start earlier, with the ILO itself. The Organization was created in 1919, at the close of the First World War, on the belief that lasting peace depends on social justice. From the beginning, one of its concerns was forced labour. In fact, forced labour was among the first labour issues to be treated at the international level, with roots that reach back to nineteenth-century campaigns against slavery and to colonial labour practices that troubled reformers well before the ILO existed (ILO, 2024a). The first major ILO answer to the problem was the Forced Labour Convention, 1930 (No. 29). This instrument set out a general definition of forced labour and required states to suppress its use. It defined #compulsory_labour as all work or service exacted from a person under the menace of a penalty, and for which the person has not offered themselves voluntarily. That definition has two key parts: the threat of a penalty, and the absence of free consent. It remains the anchor definition in international law today, and every later instrument, including Convention No. 105, is read against it. Convention No. 29, however, was a product of its time. It allowed several exceptions. Compulsory military service, normal civic obligations, work done as part of a criminal sentence under public control, genuine emergencies, and minor communal services were all placed outside the ban. Some of these exceptions made practical sense. A country facing a flood or an earthquake may need to call on its citizens; a lawful court sentence may include work. Others left room for abuse. In particular, the exception for prison work, and the loose idea of public purposes, could be stretched to cover systems of state labour that had little to do with justice and much to do with control. By the early 1950s, the international mood had shifted. Two developments pushed states toward a new instrument. The first was a wave of detailed reports about very large systems of forced labour operated by some governments, used both to punish opponents and to drive economic output. A joint inquiry set up by the United Nations and the ILO, often described as the Ad Hoc Committee on Forced Labour, gathered evidence during the early 1950s and drew wide attention to these practices. Its work showed that forced labour was not a relic of the past but a live method of political and economic control in the present. The second development was the rise of the human-rights movement. The Universal Declaration of Human Rights had been adopted in 1948, and its language of freedom and dignity created pressure to name and forbid the political abuse of labour. Out of this context came the Abolition of Forced Labour Convention, 1957 (No. 105). Its drafters made a deliberate choice. Rather than rewrite the definition of forced labour in Convention No. 29, they left that definition standing and added a new layer on top. The new layer did not ask what forced labour is. It asked what forced labour is for. It identified specific political and economic purposes and declared that forced labour serving those purposes must be abolished immediately and completely. This shift, from the elements of coercion and consent to the purpose behind the labour, is the intellectual heart of the Convention. The change of focus was significant. Convention No. 29 targets the practice in general. Convention No. 105 targets the motive. This means that even where a state might argue that some compulsory work is allowed under the older exceptions, that argument cannot save the labour if it is being used for one of the forbidden purposes in the 1957 text (ILO, 2024a). The two conventions therefore work as a pair. The 1930 instrument gives the definition and the general ban; the 1957 instrument closes political and economic loopholes that the definition alone could not reach. It is worth pausing on the Cold War background, because it explains the exact wording. The reference to political coercion, and to views opposed to the established political, social, or economic system, reflects concern about states that jailed dissidents and put them to work. The ban on forced labour as a method of mobilising labour for economic development responds to schemes in which whole populations were compelled to work on state projects in the name of progress. These clauses were not abstract. They were written with real practices in mind, and they were meant to draw a firm line that no ideology, of the left or of the right, could cross. The debates of the 1957 Conference were sharp, precisely because the new standard touched the way governments of very different kinds ran their economies and treated their critics. 4. The Legal Architecture of Convention No. 105 4.1 Structure of the instrument Convention No. 105 contains ten articles, but only the first two carry the substantive obligations. The rest deal with procedure: how ratifications are registered, when the Convention enters into force, how a state may denounce it, and how any future revision would work. This lean design is typical of ILO conventions. The obligations are stated plainly so that they can be written into national law and checked by the supervisory bodies. Article 1 lists the five prohibited purposes. Article 2 sets the duty. Each ratifying state undertakes to take effective measures to secure the immediate and complete abolition of forced or compulsory labour as described in Article 1 (ILO, 2024a). The word immediate matters. Unlike some treaties that allow gradual progress toward a goal, Convention No. 105 asks for prompt action. The prohibited uses of labour are treated as wrongs to be stopped, not problems to be managed over decades. This is a demanding standard, and it reflects the seriousness with which the drafters viewed the abuses in front of them. 4.2 The five prohibited purposes The core of the Convention is the list in Article 1. It forbids the use of any form of forced or #compulsory_labour in five situations. Understanding each one is essential, so we take them in turn, with examples to make the meaning concrete. The first purpose is the ban on forced labour as a means of #political_coercion or education, or as a punishment for holding or expressing #political_views, or views ideologically opposed to the established political, social, or economic system. This is the clause the article title highlights. It protects the freedom to think, to speak, and to disagree. A government that imprisons a person for a peaceful political opinion and then compels that person to work breaks this rule. So does a government that uses forced labour as a form of re-education aimed at changing political or ideological beliefs. The clause reaches both the punishment of dissent and the attempt to reshape it through compulsory work. It does not matter whether the state calls the labour a sentence, a training programme, or a civic duty. If the aim is to coerce or punish opinion, the labour is forbidden. The second purpose is the ban on forced labour as a method of mobilising and using labour for purposes of #economic_development. This clause responds to programmes in which people are compelled to work on state projects, such as roads, dams, farms, or factories, in the name of national growth. The point is not that development is bad. The point is that development can never justify forcing people to labour against their will. Economic goals, however large, do not override the freedom not to be enslaved. This clause was aimed at real schemes in which states treated their own citizens as a labour reserve to be drafted for construction and production. The third purpose is the ban on forced labour as a means of #labour_discipline. This concerns situations where workers are forced to labour, or punished with compulsory work, as a way of enforcing discipline in employment. For example, a rule that punishes a worker who leaves a job, arrives late, or breaks a workplace rule by forcing them into labour would fall under this ban. The clause protects the basic idea that employment is a relationship people can enter and leave, not a form of bondage where mistakes are met with forced work. The fourth purpose is the ban on forced labour as a punishment for having participated in #strikes. This clause does not, by itself, create a general right to strike. What it does is forbid one particular response to strikes: forcing the strikers to work as a penalty. Read together with the ILO standards on freedom of association, it forms part of a wider protection for collective action. A state that jails strikers and sentences them to hard labour for the act of striking violates this provision. The clause recognises that the strike is a legitimate tool of workers, and that turning it into a crime punishable by forced labour is a serious abuse. The fifth purpose is the ban on forced labour as a means of racial, social, national, or religious #discrimination. Here the Convention meets anti-discrimination law. It forbids the use of compulsory labour to target or oppress people because of who they are. This clause recognises that forced labour has often been used as a tool of group oppression, and it shuts that door. A scheme that singles out an ethnic or religious minority for compulsory work is caught by this provision even if the state describes it in neutral terms. Taken together, these five purposes cover the main ways in which states have historically abused labour for political and social control. They are not a random list. They map onto real patterns of abuse: silencing opponents, driving development by force, disciplining workers, crushing strikes, and oppressing minorities. By naming each one, the Convention makes it harder for any government to claim that its particular scheme is somehow different or exempt. The precision of the list is a deliberate strength. It gives the supervisory bodies clear categories to work with, and it gives states clear rules to follow. 4.3 The relationship with Convention No. 29 A common point of confusion for students is how Convention No. 105 relates to Convention No. 29. The simplest way to see it is this. Convention No. 29 answers the question what is forced labour, and it bans the practice in general. Convention No. 105 answers the question for what purposes must forced labour always be abolished, and it names five of them. The 1957 instrument supplements the 1930 one. It does not replace it, and it does not change the underlying definition (Sitzia, 2023). This relationship has a practical edge. The exceptions in Convention No. 29, such as the one for prison work, do not automatically carry over to the five situations in Convention No. 105 (ILO, 2024a). Suppose a person is convicted for a peaceful political protest and then required to work in prison. Under the general rules, prison labour after a lawful conviction can fall outside the definition of forced labour. But because the underlying conviction is a punishment for political views, the compelled work runs into the first prohibited purpose of Convention No. 105. The exception cannot rescue it. This is one of the most important points in the whole area, and it is why the two conventions must be read side by side rather than in isolation. Because both instruments deal with the same broad evil from different angles, the ILO often treats them as a set. Its supervisory bodies review them together, and states report on both. When people speak of the ILO forced-labour standards, they usually mean this pair, together with the 2014 Protocol to Convention No. 29 and its accompanying Recommendation, which added modern tools for prevention, protection, and access to justice. 4.4 The 2014 Protocol and the modern package The forced-labour standards did not stop developing in 1957. In 2014, the ILO adopted a Protocol to Convention No. 29, along with the Forced Labour (Supplementary Measures) Recommendation, 2014 (No. 203). The Protocol was a response to the changing shape of forced labour. Much of the problem today is not the state marching people to a work site. It is private exploitation hidden inside ordinary economic activity, often connected to migration and trafficking. The Protocol asks ratifying states to take effective measures to prevent forced labour, to protect victims, and to give them access to remedies, including compensation. It complements the older conventions by focusing on the practical steps needed to root out modern forms of the abuse. For Convention No. 105, the Protocol is a useful companion rather than a competitor. The 1957 Convention stays focused on the political and economic purposes that governments must never pursue. The Protocol widens the effort against forced labour in the private economy. Together, the three instruments, the 1930 Convention, the 1957 Convention, and the 2014 Protocol, form the backbone of the international legal response. A student who grasps how they divide the work will find the whole field much clearer. 5. Key Concepts and Definitional Boundaries 5.1 The two elements of forced labour Since Convention No. 105 relies on the definition in Convention No. 29, it helps to look closely at that definition. Forced labour has two elements. The first is the menace of a penalty. This does not have to mean a criminal fine or a prison term. The penalty can be the loss of rights or benefits, the threat of violence, the confiscation of identity documents, the withholding of wages, or pressure placed on a person's family. Courts and expert bodies read this element broadly, because coercion comes in many forms. The second element is the absence of free and informed consent. A person may appear to agree to work, but if that agreement was obtained through deception, coercion, or the abuse of a position of weakness, it is not genuine consent. A worker who signs a contract without understanding it, or who cannot refuse because their passport has been taken, has not truly volunteered. Both elements, the menace and the lack of consent, must be present for a situation to count as forced labour. This matters for the five prohibited purposes. A state that pressures citizens to join a public works scheme, under threat of losing benefits or facing punishment, may be imposing forced labour for economic development, which the Convention forbids. The presence of the menace of a penalty turns what might look like ordinary public service into something the law does not allow. Understanding these two elements gives students a reliable test to apply to any situation they meet. 5.2 The penal exception and its limits The clearest boundary problem in this field is #prison_labour. International law does not treat all work by prisoners as forced labour. Under Convention No. 29, work required of a person as a result of a conviction in a court of law is excluded from the definition, provided the work is carried out under the supervision and control of a public authority, and the person is not hired to, or placed at the disposal of, private individuals or companies. In other words, ordinary prison work, supervised by the state and not sold off for private profit, is generally permitted (Jarman and Heard, 2023). Convention No. 105 does not remove this exception, but it adds a vital limit. If the conviction itself is a punishment for one of the prohibited purposes, the work cannot be justified. A person jailed for holding certain political views cannot lawfully be compelled to work, because the whole chain, from arrest to sentence to labour, serves a forbidden aim. The penal exception is meant for genuine criminal justice, not for the punishment of dissent dressed up as a criminal sentence. The exception also has conditions of its own. The work must be public, supervised, and not handed to private profit-makers. When any of these conditions fail, even ordinary prison work can slide into forced labour. This limit has become more important as scholars examine the reality of prison work around the world. Recent studies point out that the line between rehabilitation and exploitation can be thin, and that the involvement of private firms in prison labour raises hard questions (Sitzia, 2023). A country-level assessment of prison conditions carried out with the ILO, for example, examined whether prison work programmes actually met the standards of Conventions Nos. 29 and 105, and it found both progress and remaining gaps in the legal framework (National Human Rights Commission of Mongolia and ILO, 2025). These studies show that the penal exception is not a free pass. It is a narrow gate with clear conditions, and it must be checked against the facts of each system. 5.3 What the Convention does not do It is useful to be clear about the limits of Convention No. 105. It does not ban all compulsory work. Normal civic duties, genuine emergency service, and lawful prison work outside the five prohibited purposes remain permitted under the general framework. It does not create a detailed right to strike; that protection comes mainly from the freedom-of-association standards. And it does not, by itself, regulate private forced labour in supply chains, which is addressed through other tools, including the 2014 Protocol and a growing set of national and regional laws. Its focus is precise: the abolition of forced labour used for the five named political and economic purposes. Understanding this focus prevents students from expecting the Convention to do more than it was built to do. A common mistake is to treat Convention No. 105 as a general anti-slavery law. It is better understood as a targeted rule against the political and economic misuse of labour by the state. Its power comes from that precision, not from breadth. 6. The Supervisory System: How Compliance Is Watched A treaty is only as strong as the system that checks whether states keep their word. The ILO has one of the oldest and most developed supervisory systems in international law, and it applies to Convention No. 105 like any other ratified standard. 6.1 Regular reporting and the Committee of Experts At the centre of the regular system sits the Committee of Experts on the Application of Conventions and Recommendations, usually shortened to #CEACR. This body is made up of twenty independent legal experts drawn from different regions, legal systems, and cultures. They are not government representatives. Their job is to give an impartial, technical assessment of how well states are applying the standards they have ratified (ILO, 2024a). States that have ratified Convention No. 105 must report at regular intervals on how they apply it in law and in practice. The Committee of Experts studies these reports, along with comments from workers' and employers' organisations, and it issues observations and direct requests. Observations are the more formal comments, often raising serious or long-standing issues. Direct requests deal with more technical or detailed points. From time to time, the Committee also issues a general observation, which sets out its thinking on a whole theme across many countries. At its late-2023 session, the Committee adopted such a general observation on Convention No. 105, later published for wide use, which shows that the standard is still an active subject of expert attention (ILO, 2024a). Over the decades, the accumulated comments of the Committee have become the main guide to what the Convention actually requires in concrete cases. 6.2 The tripartite dimension What makes the ILO unusual is its tripartite structure. Governments, employers, and workers all sit at the table. This #tripartism runs through the supervisory system. Workers' and employers' organisations can submit comments on how a government applies a convention, and these comments feed directly into the review. This design means that the people most affected by forced labour, and those who might be tempted to use it, both have a voice in the process. It also makes it harder for a government to hide abuses behind an official report, because independent organisations can challenge that report with evidence of their own. Each year, a tripartite committee at the International Labour Conference, the Conference Committee on the Application of Standards, examines a selection of the most serious cases identified by the experts. This public discussion adds political weight. A government named in this setting faces the attention of its peers and of the wider public. While the ILO cannot impose sanctions in the way a national court might, the pressure of reasoned criticism, repeated year after year, has real influence. States generally prefer to avoid the reputational cost of appearing on the list of serious cases, and this preference can move them toward reform. 6.3 Complaints and representations Beyond the regular reporting cycle, the ILO Constitution allows for complaints and representations. A workers' or employers' organisation can make a representation claiming that a state is not observing a convention it has ratified. A member state, or a delegate to the Conference, can file a more serious complaint, which may lead to a commission of inquiry, the most solemn investigative procedure the ILO has. These procedures are used sparingly, but they exist as a backstop for grave and persistent violations. For a standard as fundamental as the ban on forced labour, they provide an important route when ordinary supervision is not enough, and their very existence encourages states to take the reporting process seriously. 6.4 Strengths and weaknesses of the system The ILO system has clear strengths. It is expert, it is regular, and it involves the affected parties. Over decades, it has built a rich body of interpretation that guides states on what the standards actually require. This body of comment gives the short treaty texts a depth and detail they could never carry on their own. At the same time, the system has limits. It depends heavily on state cooperation and on honest reporting. It moves slowly, sometimes taking years to resolve a single issue. And it lacks the power to enforce its findings directly. Scholars have noted that the ILO's authority is rooted in a national, territorial model that struggles with problems that cross borders, such as forced labour in international migration and global supply chains (Fudge, 2025). These weaknesses do not make the system useless. They simply mark the edges of what a promotional, dialogue-based approach can achieve on its own, and they help explain why other tools, including trade measures, have grown up alongside it. 7. Convention No. 105 in the Wider Human Rights Framework Convention No. 105 does not stand alone. It belongs to a web of standards that protect human freedom, and it gains much of its force from those connections. 7.1 A fundamental principle at work Since 1998, the ILO has recognised a set of #fundamental_rights that every member state must respect simply by virtue of membership, whether or not it has ratified the relevant conventions. The elimination of all forms of forced or compulsory labour is one of these principles, and Convention No. 105 is one of the instruments that gives it shape. In 2022, the ILO amended its Declaration on Fundamental Principles and Rights at Work to add a safe and healthy working environment as a fifth category, alongside freedom of association, the elimination of forced labour, the abolition of child labour, and the elimination of discrimination (ILO, 2022). This amendment confirmed the central place of the forced-labour standards in the modern rights framework, and it showed that the framework can grow to meet new concerns. Because forced labour is a fundamental principle, even states that have not ratified Convention No. 105 carry an obligation to respect the underlying value. This gives the standard a reach beyond its formal ratifications. It becomes part of the baseline that the international community expects of every government, and states that have not ratified must still report on their efforts under the Declaration's follow-up. 7.2 Links to civil and political rights The first prohibited purpose of Convention No. 105 sits close to the core of civil and political rights. The freedom to hold and express #political_views is protected by the Universal Declaration of Human Rights and by the International Covenant on Civil and Political Rights. That same Covenant contains its own ban on slavery and forced labour. When a government forces political prisoners to work, it breaks more than one rule at once. It violates the freedom of expression, and it violates the freedom from forced labour. Convention No. 105 makes the labour dimension explicit and gives the ILO a clear basis to act. The connection to #freedom_of_association is just as strong. The fourth prohibited purpose, the ban on forced labour as a punishment for strikes, only makes full sense when read together with the right of workers to organise and to take collective action. A state that punishes strikers with forced labour attacks both the freedom from forced labour and the freedom to associate. The standards reinforce one another, and the ILO reads them as a connected whole. Regional systems add another layer of protection. In Europe, for example, the European Convention on Human Rights contains a prohibition of slavery and forced labour, overseen by a regional court, so that a person may find protection under both the ILO standards and the regional treaty. 7.3 Structural approaches and the role of the state Recent scholarship has pushed the analysis further by asking how the law itself can create the conditions for exploitation. One influential line of thought describes what it calls state-mediated structural injustice, where laws that look legitimate in fact trap groups of workers in vulnerability, including migrants, welfare recipients, and prison workers (Mantouvalou, 2023). This work reminds us that the abuse Convention No. 105 targets is not always crude or obvious. Sometimes the machinery of coercion is built quietly into ordinary rules. Reading the Convention with this lens encourages students to look past the label and ask who really controls a person's labour, and under what pressure. A related body of writing traces how the very idea of #modern_slavery has been constructed in law and politics, and it warns that anti-slavery efforts can sometimes serve narrower agendas than they claim (Fudge, 2025; Kenway, 2021). For Convention No. 105, this critical view is a useful check. It keeps the focus on the actual freedom of the worker, rather than on symbolic declarations or slogans. It also underlines that the ILO's forced-labour standards were shaped by history and politics, and that their meaning continues to be argued over. A standard is not a fixed object. It is interpreted, contested, and applied by many hands, and the critical literature helps students see that process clearly. 8. Contemporary Challenges and Debates The Convention was written in 1957, but the problems it addresses are far from settled. This section examines the main debates that keep it relevant. 8.1 State-imposed forced labour and political dissent The most direct challenge to Convention No. 105 comes from #state_imposed_forced_labour. According to the most recent global estimates, most forced labour is imposed by private actors, but a significant share, around fourteen per cent, is imposed by states (ILO, Walk Free and IOM, 2022). This state-imposed category is exactly what the Convention was built to stop. It includes forced labour used for political coercion, for pushing economic programmes, and for the oppression of particular groups. The hardest cases involve labour tied to the punishment of dissent. When a government detains people for their beliefs, their ethnicity, or their peaceful activism, and then puts them to work, it engages the first and fifth prohibited purposes at once. These situations are often difficult to investigate, because the states involved control access to information and may frame the labour as re-education, vocational training, or lawful punishment. The ILO supervisory bodies press for facts, but they depend on cooperation that is not always given. Even so, the Convention gives the international community a clear standard to invoke, even when direct enforcement is hard. A clear rule allows other actors, from foreign governments to civil society groups, to name a practice as a violation and to build pressure around that judgment. 8.2 Prison labour and the penal exception The debate over #prison_labour has grown louder in recent years. Millions of people are held in prisons worldwide, and many of them work. The core legal question is whether that work respects the conditions of the penal exception, or whether it crosses into forced labour. Two pressures are especially important. The first is the involvement of private companies. When prison labour is placed at the disposal of private firms for profit, it can fall outside the exception and become forced labour under the standards (Jarman and Heard, 2023). The second is the risk that criminal law itself is used to punish dissent, which, as we have seen, brings Convention No. 105 into play regardless of the penal setting. Scholars have argued that prison work sits on a spectrum. At one end lies genuine rehabilitation, where work builds skills and eases the return to society. At the other end lies exploitation, where prisoners are a cheap and captive workforce with little protection (Mantouvalou, 2023; Sitzia, 2023). Trade rules add another layer, since some customs schemes treat prison-made goods as unfair competition and restrict their import (Sitzia, 2023). The result is a rich but unsettled area, where the abstract standards of Conventions Nos. 29 and 105 meet the messy reality of penal systems. For students, it is one of the best places to see how a general principle must be applied to concrete facts, and how honest people can disagree about where the line falls. 8.3 Global supply chains and trade measures A newer front concerns #global_supply_chains. Forced labour is not confined to prisons or to political detention. It appears in fields, factories, fishing boats, and building sites, often far from the consumers who buy the products. The financial scale is enormous. A recent ILO study estimated that forced labour generates illegal profits of around 236 billion United States dollars each year (ILO, 2024b). Numbers like this have pushed governments to act through trade and corporate rules, rather than through the ILO alone. Several major measures have emerged. The United Kingdom and Australia adopted modern slavery acts that require large companies to report on the steps they take to root out forced labour. The United States has used import bans on goods linked to forced labour. The European Union adopted a regulation in 2024 that aims to keep products made with forced labour off its market (European Union, 2024). These tools do not replace Convention No. 105. They draw on the same definitions and the same values, and they extend the fight into the marketplace, where the pressure of lost sales can move companies and governments alike. These developments also raise fresh questions. Import bans and reporting duties depend on evidence that can be hard to gather in distant supply chains. There is a risk that trade measures may be used for protectionist ends, dressed up as human-rights action. And there is a real danger that the burden falls on workers themselves, if a ban simply cuts them off from income without offering an alternative. Yet the ILO's forced-labour standards remain the shared reference point that keeps these different national and regional rules aligned around a common meaning of forced labour. Without that common standard, each country would define the abuse in its own way, and coordinated action would fall apart. 8.4 Ratification, denunciation, and the question of universality Convention No. 105 is one of the most widely ratified instruments in the ILO system, and recent years have seen important additions. Notably, some large states that had long stayed outside the forced-labour conventions moved to ratify them, bringing the standard closer to universal coverage (Jarman and Heard, 2023). This trend strengthens the claim that the ban on politically motivated forced labour is now a settled part of international law, accepted in principle by almost every government. At the same time, #ratification is not the end of the story. A small number of states have not ratified, and at least two states that once ratified later denounced the Convention. Denunciation is a lawful option under the treaty, available at set intervals, but it is rare, and it carries a cost in reputation. These movements remind us that acceptance of a standard is not fixed forever. It has to be maintained through continued political will and through the steady work of the supervisory system. For students, the pattern of ratification and #denunciation is a useful window into how states balance international commitments against domestic pressures, and into how a near-universal norm can still be contested at the edges. 8.5 National implementation and the role of civil society A standard means little until it is written into national law and enforced on the ground. States that ratify Convention No. 105 are expected to bring their laws into line, which may mean repealing old provisions that allow forced labour for the prohibited purposes, such as penalties for strikes that include compulsory work. The supervisory bodies often focus on exactly these details, asking a state to amend a specific law or to change a specific practice. This slow, technical work is where much of the real progress happens. Civil society and trade unions play a large part in this process. They gather evidence, they file comments with the supervisory bodies, and they campaign at home for reform. Independent writers and researchers also shape the debate by exposing abuses and by questioning easy answers (Kenway, 2021). The result is a system that does not rely on governments alone. It draws on many actors, each pressing in its own way for the principle to become practice. For a student, understanding this wider ecosystem is as important as understanding the treaty text, because it shows how international law actually takes effect. 8.6 The limits of a promotional model A final debate concerns the ILO's own approach. The Organization relies mainly on dialogue, expert review, and public pressure, rather than on hard enforcement. Critics argue that this promotional model is too weak for the gravest abuses, especially those committed by powerful states behind closed doors. Supporters reply that persuasion and reasoned criticism, backed by tripartite legitimacy, can shift behaviour over time in ways that sanctions often cannot. The truth likely lies in between. The model has clear limits, yet it also builds a durable, shared understanding of what forced labour is and why it is wrong. That shared understanding is what makes the newer trade and corporate measures possible in the first place. It is also what gives smaller actors a language to use when they challenge a powerful state. A promotional system cannot force a government to change overnight, but it can make the wrong harder to deny, and it can keep the standard alive until the political conditions for change arrive. 9. Discussion: Why Convention No. 105 Still Matters Pulling these threads together, several conclusions stand out for students and researchers. First, Convention No. 105 draws a moral and legal line that has proven remarkably durable. The idea that no state may force people to work because of their politics, their group identity, or a development plan has moved from a contested proposal in 1957 to a near-universal norm today. That journey is itself a lesson in how international standards can shape expectations across very different societies, and how a rule born in one era can outlast the specific systems that inspired it. Second, the Convention is best understood as one piece of a larger structure. It relies on the definition in Convention No. 29, it gains force from its status as a fundamental principle, and it connects to civil and political rights, to freedom of association, and to modern supply-chain rules. Studying it in isolation gives only half the picture. Studying it in context reveals a network of mutually supporting protections, in which each part strengthens the others. Third, the hardest problems today are ones of application, not of principle. Few governments openly defend forced labour as a means of #political_coercion. The disputes are about facts, about how to classify particular schemes, and about how to reach abuses that hide behind closed borders or private contracts. This is why the supervisory system, the trade measures, and the critical scholarship all matter. They are the tools that translate a clear principle into pressure on real conduct, and they are where the action lies for anyone who wants to make a difference. Fourth, the Convention rewards careful reading. Its precision is a strength. By naming five specific purposes, it resists the temptation to treat every hardship as forced labour, while still catching the political and economic abuses that history has shown to be most dangerous. For researchers, this precision offers a model of how to write a standard that is both firm and workable, neither so vague that it means nothing nor so narrow that it misses the point. The article also points to open questions worth further study. How should the penal exception be redrawn, if at all, in an age of private prisons and prison-made exports? How can the international community verify claims about state labour when access is denied? How should the ILO's promotional model relate to the harder-edged trade and corporate rules now spreading across the world? And how can new laws against forced labour in supply chains protect workers without simply cutting off their livelihoods? These are live questions, and Convention No. 105 sits at the centre of each. They are good starting points for a dissertation or a research project, because they combine settled principle with unsettled practice. 10. Conclusion The Abolition of Forced Labour Convention, 1957 (No. 105), is short, sharp, and still important. It forbids the use of #forced_labour for five specific purposes, and above all as a means of political coercion and as a punishment for holding or expressing political views opposed to the established order. It was born in a tense moment of world history, designed to close the political and economic loopholes left by the earlier Forced Labour Convention, 1930 (No. 29). It does not redefine forced labour. Instead, it asks a different question, not what forced labour is, but what it is for, and it rules out the most dangerous answers. Almost seventy years after its adoption, the Convention remains one of the most widely accepted treaties in the world, and one of the ILO's fundamental-rights standards. Its message has outlived the specific systems that inspired it. Yet the work is not finished. Millions of people are still in forced labour, some states still impose labour for political and economic ends, and new debates over prison labour and global supply chains keep testing where the line should fall. The Convention gives the world a clear standard to hold up against these abuses. Its future depends less on the text, which is strong, than on the will to apply it. For students entering this field, that gap between principle and practice is the most important thing to understand, and the most useful place to begin. References European Union (2024) Regulation (EU) 2024/3015 of the European Parliament and of the Council on prohibiting products made with forced labour on the Union market. Official Journal of the European Union, L series, 12 December 2024. Fudge, J. (2025) Constructing Modern Slavery: Law, Capitalism, and Unfree Labour. Cambridge: Cambridge University Press. doi:10.1017/9781108562058. International Labour Organization (2022) ILO Declaration on Fundamental Principles and Rights at Work (as amended 2022). Geneva: International Labour Office. International Labour Organization (2024a) Abolition of Forced Labour Convention, 1957 (No. 105): General Observation. Geneva: International Labour Office. International Labour Organization (2024b) Profits and Poverty: The Economics of Forced Labour. Geneva: International Labour Office. International Labour Organization, Walk Free and International Organization for Migration (2022) Global Estimates of Modern Slavery: Forced Labour and Forced Marriage. Geneva: International Labour Office. Jarman, B. and Heard, C. (2023) Labouring Behind Bars: Assessing International Law on Working Prisoners. London: Institute for Crime and Justice Policy Research. Kenway, E. (2021) The Truth about Modern Slavery. London: Pluto Press. Mantouvalou, V. (2023) Structural Injustice and Workers' Rights. Oxford: Oxford University Press. doi:10.1093/oso/9780192857156.001.0001. National Human Rights Commission of Mongolia and International Labour Organization (2025) Assessing Compliance of the Mongolian Prisons with ILO Conventions Nos. 29 and 105 on Forced Labour. Geneva: International Labour Office. Sitzia, A. (2023) Prison labour, customs preference schemes and decent work: Critical analysis and outlook. International Labour Review. Advance online publication. doi:10.1111/ilr.12399. #Abolition_of_Forced_Labour_Convention #ILO_Convention_105 #ForcedLabour #ForcedLabourAbolition #PoliticalCoercion #Convention_No_105_1957 #InternationalLabourStandards #Freedom_of_Expression #Prison_Labour #Modern_Slavery #Human_Rights_Law #ILO_Standards #No105 #Labour_Rights #EndForcedLabour

This article examines the International Labour Organization (ILO) Equal Remuneration Convention, 1951 (No. 100), the first binding international instrument to require #equal_pay for men and women workers for work of #equal_value. The study traces the historical origins of the Convention, sets out its legal architecture, and explains the distinctive concept of "value" that separates it from the narrower idea of equal pay for identical work. It then evaluates how the Convention is monitored, how it interacts with later instruments such as the Convention on the Elimination of All Forms of Discrimination against Women, the Sustainable Development Goals, and the European Union pay transparency framework, and why, more than seventy years after adoption, the global #gender_pay_gap remains close to twenty per cent. Using a doctrinal and comparative method, the article argues that the Convention supplies a sound normative foundation but that its promise is repeatedly weakened by narrow national transposition, weak enforcement, occupational segregation, and the persistent undervaluation of female-dominated work. The article concludes with practical recommendations for students, policymakers, and practitioners, including the wider use of gender-neutral #job_evaluation, mandatory pay reporting, and stronger remedies. The findings are intended to help students understand both the strengths and the limits of international labour standards in the field of pay equity. Keywords: equal remuneration; work of equal value; gender pay gap; international labour standards; pay equity; non-discrimination. 1. Introduction Pay is one of the clearest measures of how a society values work. When two people perform work of the same worth but receive different wages simply because one is a woman and the other a man, the wage packet becomes a signal of inequality rather than of contribution. The idea that this should not happen is old, but the first time it took the form of a binding international rule was in 1951, when the International Labour Conference adopted the #Equal_Remuneration_Convention, known formally as Convention No. 100 (ILO, 1951). The Convention requires member States to promote and, so far as consistent with their methods of wage setting, to ensure equal remuneration for men and women workers for work of equal value. At first reading the principle looks simple. In practice it has proved to be one of the most demanding standards in the whole body of international labour law. The difficulty lies in three words: "work of equal value." The Convention does not merely ask that a woman and a man doing the identical job be paid the same. It asks something more ambitious, namely that jobs which are different in content but equal in demands, such as a cleaner and a gardener, or a nurse and a mechanic, be compared and, if they are found to be of equal worth, be paid on an equal basis. This is the concept of #comparable_worth, and it strikes directly at the structural undervaluation of jobs that are mainly done by women. This article offers a student-friendly but rigorous analysis of Convention No. 100. It is written for readers who are studying labour law, human rights, gender studies, public policy, or industrial relations, and who need a clear account of what the Convention says, why it was adopted, how it is supervised, and how far it has succeeded. The article does not treat the Convention as a museum piece. On the contrary, it links the 1951 text to very recent developments, including the European Union Pay Transparency #Directive of 2023 and the most recent global wage data, in order to show that the questions raised in 1951 are still unresolved in the mid-2020s. The central argument is that Convention No. 100 provides a strong and durable #normative_foundation, but that its practical effect depends almost entirely on how seriously States translate the principle into national law, wage-setting machinery, and enforcement. Where the principle is reduced to "equal pay for the same job," it leaves most of the pay gap untouched, because women and men are frequently not in the same jobs at all. Where it is applied in its full sense, through objective job evaluation and effective remedies, it becomes a powerful tool. The gap between these two outcomes explains why a Convention that has been ratified by the great majority of the world's States coexists with a global gender pay gap that has barely moved in a generation. The remainder of the article proceeds as follows. Section 2 sets out the historical background. Section 3 explains the method used. Section 4 analyses the legal architecture of the Convention. Section 5 examines the concept of value and the role of job evaluation. Section 6 describes the supervisory machinery. Section 7 places the Convention within the wider international framework. Section 8 presents the empirical picture of the pay gap. Section 9 discusses the main challenges and criticisms. Section 10 offers comparative national perspectives. Section 11 discusses the findings, and Section 12 concludes with recommendations. 2. Historical Background and Origins The demand for equal pay did not begin with the ILO. Women workers, trade unions, and reformers had pressed for it since the nineteenth century, and the principle appeared in the very Constitution of the ILO when the Organization was founded in 1919. The Preamble to that Constitution listed the principle of equal remuneration for work of #equal_value among the matters requiring urgent improvement. In other words, the idea was written into the ILO's founding text three decades before it became a stand-alone Convention. Two forces pushed the principle from aspiration to binding standard. The first was the experience of the two world wars. During both conflicts, women entered factories, transport, and services in very large numbers to replace men who had gone to fight. They frequently did the same work as men but were paid less, and the injustice of this became visible on a mass scale. The second force was the human rights movement that followed the Second World War. The Universal Declaration of Human Rights of 1948 proclaimed that everyone, without any discrimination, has the right to equal pay for equal work. The ILO, as the specialized agency of the United Nations responsible for the world of work, was the natural body to convert this declaration into an enforceable labour standard. The drafting took place at the thirty-fourth session of the International Labour Conference in Geneva. On 29 June 1951 the Conference adopted the Convention, which entered into force on 23 May 1953 once the required number of ratifications had been registered (ILO, 1951). Alongside the Convention, the Conference adopted the #Equal_Remuneration_Recommendation, 1951 (No. 90), a non-binding text that gives more detailed guidance on how the principle should be applied, for example in wage rates fixed by public authority. The pairing of a Convention with a Recommendation is a standard ILO technique: the Convention sets the binding obligation, and the Recommendation offers practical direction. The most important drafting decision concerned the choice of words. Early proposals spoke of equal pay for equal work, a formula that would have limited comparison to identical or nearly identical jobs. Delegates recognized that this would leave a large loophole, because labour markets are heavily segregated by sex. If women are concentrated in a small number of occupations, and if those occupations are systematically underpaid, then a rule confined to identical work would achieve very little. The Conference therefore adopted the broader phrase "work of #equal_value." This deliberate choice is the intellectual heart of the Convention and the reason it remains relevant today (McIsaac and Knaul, 2024). The debates during the drafting also revealed tensions that still echo today. Some government and employer delegates worried that a binding obligation to ensure equal value across an entire economy would be impractical, particularly in countries where wages were set by decentralized bargaining or by market forces rather than by statute. Others feared that a strict rule might, in the short term, reduce the number of jobs open to women if employers responded to higher wage bills by hiring fewer of them. Worker delegates and many governments countered that these fears were exactly the arguments that had always been used to justify paying women less, and that the principle of #equal_value was a matter of basic justice that could not be traded away for supposed efficiency. The compromise that emerged, allowing States to choose their own means of application, reflected these tensions. It secured broad support at the cost of leaving the strength of application to national discretion, a choice whose consequences are still visible in the uneven results discussed later in this article. It is worth stressing how radical this was for its time. In 1951 the male breadwinner model was dominant, women's earnings were widely regarded as secondary, and many countries still operated formal or informal marriage bars that pushed women out of paid work after they married. Against this background, an international treaty asserting that the value of work, not the sex of the worker, should determine pay was a significant statement. Convention No. 100 later became one of the ILO's fundamental Conventions, meaning that all member States, even those that have not ratified it, are considered to have an obligation arising from their membership of the Organization to respect and promote the principle it embodies. In 2022 the framework of fundamental principles and rights at work was itself broadened to include occupational safety and health, expanding the set of fundamental Conventions, but the status of Convention No. 100 within that core group was never in doubt. By the early 2020s the Convention had been ratified by 174 of the 187 member States, making it one of the most widely accepted labour standards in existence (ILO, 2022). 3. Methodology This article uses a combined doctrinal and comparative approach, supported by descriptive empirical evidence. The doctrinal element involves close reading of the primary legal texts, namely the Convention itself, its accompanying Recommendation, and the interpretive comments of the ILO supervisory bodies. The comparative element places Convention No. 100 alongside other international and regional instruments, in particular the Convention on the Elimination of All Forms of Discrimination against Women, the relevant Sustainable Development Goal, and the European Union framework, in order to identify convergences and gaps. The empirical element draws on recent statistical reporting from the ILO and other bodies to assess how far the principle has been realized in practice. The sources used are of three kinds. The first kind is primary legal material: treaty texts and official reports of the supervisory machinery. The second kind is recent peer-reviewed scholarship on pay equity, #occupational_segregation, and pay transparency, selected for its currency and relevance. The third kind is authoritative statistical output on wages and the gender pay gap. Priority has been given to material published within the last five years so that the analysis reflects the current state of debate, although the primary legal instruments are cited in their original form because they remain in force. The study has limits that students should keep in mind. It is a conceptual and comparative analysis, not an econometric one, and it does not attempt to measure the precise causal effect of ratification on wages. The relationship between a ratified Convention and observed wage outcomes is mediated by many national factors, including the strength of collective bargaining, the design of minimum wage systems, and the vigour of enforcement, and disentangling these is beyond the scope of a single article. What the study offers instead is a structured framework for understanding why a widely accepted standard has produced uneven results. 4. The Legal Architecture of Convention No. 100 The Convention is short, with only fourteen articles, and the substantive obligations are concentrated in the first four. Understanding these four articles is essential, so this section examines each in turn. Article 1 contains the definitions. It defines #remuneration broadly to include the ordinary, basic, or minimum wage or salary, together with any additional emoluments payable directly or indirectly, whether in cash or in kind, by the employer to the worker and arising out of the worker's employment (ILO, 1951). This wide definition matters a great deal. If remuneration meant only the basic wage, employers could comply on paper while creating inequality through bonuses, allowances, housing, transport, pension contributions, or benefits in kind. By capturing all these components, the Convention prevents the principle from being evaded through the structure of the pay package. Article 1 then defines the key phrase, stating that equal remuneration for men and women workers for work of equal value refers to rates of remuneration established without #discrimination based on sex. The comparison is therefore between rates, not between the total earnings of individuals, which may differ for lawful reasons such as seniority or hours worked. Article 2 sets out the obligation and the means of applying it. Each ratifying State must promote and, in so far as is consistent with the methods in operation for determining rates of remuneration, ensure the application to all workers of the principle of equal pay for work of equal value. Crucially, the Convention leaves States free to choose the method. The principle may be applied through national laws or regulations, through legally established or recognized machinery for wage determination, through #collective_agreements between employers and workers, or through a combination of these means. This flexibility was a compromise. It respected the very different wage-setting traditions of member States, from countries with statutory minimum wages to countries where pay is set mainly by collective bargaining. The price of the flexibility is that the strength of application varies enormously from one country to another. Article 3 addresses measurement. It provides that, where such action will assist in giving effect to the Convention, measures shall be taken to promote the objective appraisal of jobs on the basis of the work to be performed. This is the legal basis for gender-neutral job evaluation, the technical exercise of comparing different jobs according to factors such as skill, effort, responsibility, and working conditions. Article 3 also makes clear that differences in rates which correspond, without regard to sex, to differences in the value of the work are not contrary to the principle. In other words, the Convention does not require identical pay for everyone. It requires that any pay difference be explained by the value of the work rather than by the sex of the worker. Article 4 requires each State to cooperate as appropriate with the employers' and workers' organizations concerned in giving effect to the Convention. This reflects the ILO's #tripartite character, under which governments, employers, and workers all participate in shaping and applying standards. The remaining articles are procedural, dealing with ratification, entry into force, denunciation, and reporting. Several features of this architecture deserve emphasis. First, the obligation is one of both promotion and, where consistent with national methods, of result, which gives States a genuine duty rather than a mere aspiration. Second, the broad definition of remuneration closes obvious loopholes. Third, and most importantly, the choice of "value" over "equal work" transforms the Convention from a modest anti-favouritism rule into a structural tool capable of challenging the undervaluation of women's work. The next section explores that concept in detail. 5. The Concept of "Work of Equal Value" and the Role of Job Evaluation The phrase "work of #equal_value" is easy to state and hard to apply. To understand it, it helps to contrast it with two narrower ideas that are often confused with it. The narrowest idea is "equal pay for equal work," meaning the same pay for people doing the identical job. Almost everyone accepts this, and most legal systems now prohibit paying a woman less than a man for exactly the same work. But this idea catches very little of the real problem, because women and men are frequently not doing the same job. Labour markets are marked by #occupational_segregation, in which certain occupations are dominated by women and others by men (Valencia-Contrera and Rivera-Rojas, 2023). If a rule reaches only identical jobs, it never touches the pay difference between, for example, a female-dominated caring occupation and a male-dominated technical one. A slightly broader idea is "equal pay for similar work," meaning the same pay for jobs that are broadly comparable in content. This reaches a little further but still fails where the jobs being compared look quite different on the surface. Convention No. 100 goes further than both. It requires equal pay for work of equal value, which means that jobs may be compared even when their content is entirely different, provided that, when assessed against objective criteria, they are of equal worth. A classic illustration is the comparison between a hospital cleaner and a hospital groundskeeper, or between a school cook and a school caretaker. The tasks are different, but if a fair evaluation shows that they demand a similar combination of skill, effort, responsibility, and working conditions, then the Convention requires that they be paid on an equal footing regardless of whether one job is done mainly by women and the other mainly by men. This is where #job_evaluation becomes central. A gender-neutral job evaluation scheme breaks each job down into measurable factors and assigns points or weightings to them, allowing very different jobs to be placed on a common scale. Done well, such schemes expose the hidden undervaluation of female-dominated work, in which skills that are stereotypically associated with women, such as caring, communication, and dexterity, are treated as natural attributes rather than as skills that merit reward (Chordiya and Hubbell, 2023). The Convention's supervisory bodies have repeatedly stressed that job evaluation must be free of gender bias in the choice and weighting of factors, because a scheme that quietly rewards physical strength while ignoring emotional labour will simply reproduce the existing gap under a veneer of objectivity. A simple worked example helps to make the method concrete. Suppose an organization employs both storeroom handlers, a role held mostly by men, and childcare assistants, a role held mostly by women, and pays the handlers more. Under a narrow equal work rule these jobs would never be compared, because their tasks are entirely different. Under a gender-neutral evaluation scheme, each job is scored against agreed factors. Skill might include the training, knowledge, and interpersonal ability the job requires. Effort might include both physical and mental or emotional demands, so that the concentration and patience needed to care for young children is counted alongside the physical demand of lifting stock. Responsibility might include responsibility for people, for money, or for equipment, so that responsibility for the safety of children is weighed against responsibility for goods. Working conditions might include noise, risk, and unsocial hours. When the two jobs are scored in this way, it is quite possible that the childcare role turns out to be of equal or greater value, and yet had been paid less purely because caring skills had been treated as natural rather than as skills. This is exactly the kind of hidden undervaluation that the Convention's concept of #equal_value is meant to expose and correct. The example also shows why the choice and weighting of factors is so sensitive: a scheme that gave heavy weight to physical strength and none to emotional labour would quietly rebuild the very bias it claimed to remove. The obstacles to proper evaluation are real. Designing a scheme requires expertise and resources that small employers may lack. Comparisons across an entire sector or economy, rather than within a single firm, are analytically and politically difficult. And the exercise inevitably involves value judgments about which job characteristics matter, judgments that can smuggle bias back in. For these reasons the OECD has observed that gender-neutral job evaluation, although increasingly required by law, is still applied unevenly, and that many countries are only now building the analytical tools needed to compare the value of different jobs at the level of the employer (OECD, 2026). Despite these difficulties, job evaluation remains the practical engine through which the abstract idea of #equal_value is turned into concrete wage decisions. Without it, "work of equal value" risks remaining a slogan. 6. Supervisory Machinery and Implementation A Convention is only as strong as the system that watches over its application. The ILO possesses one of the oldest and most developed supervisory systems in international law, and Convention No. 100 is subject to it in full. The cornerstone of regular supervision is the Committee of Experts on the Application of Conventions and Recommendations, usually abbreviated to #CEACR. Established in 1926, the Committee is composed of twenty independent jurists drawn from different regions, legal systems, and cultures, appointed in their personal capacity to give impartial and technical assessments of how States apply the standards they have ratified (ILO, 2025). Under the ILO Constitution, ratifying States must report periodically on the measures they have taken. For fundamental Conventions such as No. 100, reports are examined on a regular cycle. The Committee then issues two kinds of comment. #Observations deal with fundamental questions about the application of a Convention and are published in the Committee's annual report. Direct requests deal with more technical matters or ask for further information and are communicated directly to the government concerned. The comments of the Committee of Experts on Convention No. 100 have, over the decades, done much to give the principle real content. It is largely through these comments that the meaning of "work of equal value" has been clarified, that States have been pressed to move beyond a narrow "equal work" formula, and that the need for gender-neutral job evaluation and for comparisons reaching beyond the single establishment has been established. Because these comments are public and are repeated year after year until a problem is resolved, they create a slow but persistent pressure on governments. Alongside the Committee of Experts sits the Conference Committee on the Application of Standards, a tripartite body that meets during the annual International Labour Conference and examines a selection of the most serious cases. Being called before this Committee carries a reputational cost, and the resulting conclusions add political weight to the technical findings of the experts. Beyond this regular supervision, the ILO also has complaint and representation procedures that can be triggered in specific situations. The strength of this machinery is that it is expert, continuous, and public. Its weakness is that it depends heavily on #state_reporting and on moral and reputational pressure rather than on hard sanctions. The ILO cannot fine a government or order it to raise wages. If a State chooses to ignore repeated observations, the ultimate consequences are limited. This is the familiar dilemma of international law, where obligations are real but enforcement is soft. In the case of equal remuneration, the softness of enforcement helps to explain why formal ratification has not translated into the disappearance of the pay gap. Supervision can identify and name the problem with great precision, but closing the gap requires action at the national level, where the political will is often lacking. 7. Convention No. 100 in the Wider International and Regional Framework Convention No. 100 does not stand alone. It sits within a dense network of international and regional instruments that reinforce and extend the principle of #equal_pay. Understanding these connections helps students see how international law on pay equity has developed as a system rather than as a set of isolated rules. The closest companion within the ILO is the Discrimination (Employment and Occupation) Convention, 1958 (No. 111), which prohibits discrimination in employment and occupation on grounds including sex. Convention No. 100 deals specifically with pay, while Convention No. 111 deals with the broader field of access to jobs, training, and conditions. Together they form the core of the ILO's #non_discrimination agenda, and the supervisory bodies routinely examine them side by side, since unequal pay is both a cause and a consequence of wider discrimination. Outside the ILO, the most important instrument is the Convention on the Elimination of All Forms of Discrimination against Women of 1979, often called the international bill of rights for women. Its Article 11 obliges States to ensure the right to equal remuneration, including benefits, and to equal treatment for work of equal value. This provision echoes Convention No. 100 almost word for word and shows how the ILO standard fed into the general human rights framework of the United Nations. A State that has ratified both instruments is subject to two overlapping monitoring systems, which increases the pressure to comply. The principle also appears in the 2030 Agenda for Sustainable Development. #Sustainable_Development_Goal 8, which concerns decent work and economic growth, includes a target to achieve, by 2030, full and productive employment and decent work for all women and men, and equal pay for work of equal value. This inclusion is significant because it embeds the seventy-year-old labour standard in the most prominent global policy framework of our era, and because progress is measured through indicators that keep the issue visible. At the regional level, the European Union offers the most advanced example. The principle of equal pay for equal work or work of equal value has been part of European law since the Treaty of Rome in 1957 and is now enshrined in Article 157 of the Treaty on the Functioning of the European Union. The recast Directive of 2006 consolidated earlier rules on equal treatment. Most recently, the European Union adopted Directive 2023/970 on #pay_transparency, which strengthens the application of the equal pay principle through disclosure and enforcement (European Parliament and Council of the European Union, 2023). The Directive gives workers the right to information about pay levels for work of equal value, requires employers above certain size thresholds to report their gender pay gaps, and obliges them to conduct a joint pay assessment with worker representatives where an unexplained gap of at least five per cent is found and is not corrected within a set period. Member States have until mid-2026 to bring the Directive into national law. This regional development is essentially an attempt to give the old principle the teeth that Convention No. 100 alone could not provide, by converting a general duty into concrete #transparency and reporting obligations backed by remedies (Bennedsen et al., 2023). Seen together, these instruments show a clear line of descent. The ILO Constitution of 1919 planted the principle, Convention No. 100 in 1951 made it a binding labour standard, the human rights framework of 1979 gave it universal reach, the Sustainable Development Goals of 2015 gave it global visibility, and the European pay transparency framework of 2023 gave a group of States a stronger enforcement model. The principle has been remarkably stable across a century, even as the tools for applying it have grown sharper. 8. The Persistent Gender Pay Gap: The Empirical Picture If the legal framework is so extensive, why does the problem persist? The empirical picture answers this question and provides the strongest argument for continued attention to Convention No. 100. The most striking figure is the size of the global #gender_pay_gap. The ILO estimates that women across the world continue to earn, on average, about twenty per cent less than men (ILO, 2024). United Nations bodies express the same idea by saying that women earn around seventy-seven cents for every dollar earned by men for work of equal value, with an even wider gap for women who are mothers. More recent labour income analysis finds that, among those in employment, female workers earn roughly seventy-eight cents for every dollar earned by male workers over a year, and that once differences in employment participation are added, women collectively receive far less than half of total labour income. These figures vary a great deal between countries and regions, but the direction is the same everywhere: women are paid less. The gap is not primarily explained by education. In many countries women now match or exceed men in educational attainment, yet the pay gap persists and, in a number of occupations, has widened rather than narrowed over the past two decades (ILO, 2024). This finding is important because it undermines the comfortable assumption that the gap will disappear on its own as women's qualifications catch up. The evidence points instead to structural causes. Three structural causes stand out. The first is #occupational_segregation. Women are concentrated in a limited range of occupations and sectors, such as care, health, education, cleaning, and retail, and these sectors tend to be paid less than male-dominated sectors of comparable skill (Valencia-Contrera and Rivera-Rojas, 2023). Because the undervaluation is built into the sector rather than into the treatment of individuals, a rule confined to identical jobs cannot reach it. The second cause is the #motherhood_penalty, the wage loss that women experience after having children, driven by career interruptions, part-time work, and the unequal distribution of unpaid care. Women perform a large share of unpaid household and care work, which reduces their hours and continuity in paid employment. The third cause is direct and indirect pay #discrimination, which persists even after all measurable factors are controlled for, and which the most careful studies attribute to bias in how women's work and women's requests for pay are received (Roussille, 2024). The care sector illustrates the problem sharply. In health and care work, where women are heavily overrepresented, the gender pay gap is larger than in the economy as a whole, with women earning on average around a quarter less than their male counterparts (McIsaac and Knaul, 2024). This is a sector that societies depend upon, and its persistent undervaluation is a direct example of the phenomenon Convention No. 100 was designed to address. The picture also changes when the gap is viewed through more than one lens at a time. The average gender pay gap conceals much larger gaps for particular groups of women. Women who face disadvantage on more than one axis, such as women of colour, migrant women, women with disabilities, and mothers, typically confront a pay gap far wider than the headline figure. Analysis of national data shows, for example, that in some high-income countries certain groups of women earn barely two thirds of what comparable men earn, and that a large part of that difference cannot be explained by education, experience, or occupation, pointing to the role of discrimination in the labour market (Clark and Zuccala, 2025). For students this is an important lesson in method: relying only on the average can hide the experience of the women who are worst affected, and a serious analysis of pay equity must disaggregate the data by more than sex alone. Convention No. 100 is framed around the comparison between men and women, but its underlying logic, that pay should track the value of work rather than the identity of the worker, extends naturally to these intersecting inequalities. The most recent wage data offer little comfort. The ILO's Global Wage Report for 2024 to 2025 found that real wages grew globally by about 2.7 per cent in 2024, the largest rise in fifteen years, but that the recovery was uneven and that wage inequality remained high, with women, especially in lower-middle-income countries, disproportionately affected because they are overrepresented in informal and precarious work (ILO, 2024). In several high-income countries the gap has recently widened rather than narrowed, and gender pay reporting from individual large employers regularly reveals double-digit gaps years after such reporting became mandatory (Clark and Zuccala, 2025). At current rates of change, projections for reaching global pay parity are measured in decades or even generations. The empirical picture, in short, is one of a principle that is almost universally accepted in law and almost universally violated in practice. 9. Challenges, Barriers, and Critiques Why does such a wide gap remain between the legal ideal and the economic reality? This section examines the main challenges and the principal criticisms of the Convention and its application. The first and deepest challenge is #narrow_transposition. Many States that have ratified Convention No. 100 have written into their national law only the narrow principle of equal pay for the same or similar work, rather than the broad principle of work of equal value. The Committee of Experts has repeatedly pointed this out. Where the national law is narrow, the comparison is confined to workers doing the same job, and the structural undervaluation of female-dominated occupations escapes entirely. This is perhaps the single most important reason why ratification has not closed the gap. The Convention's most powerful concept is lost in the act of implementation. The second challenge is the difficulty of comparison. Applying the concept of equal value requires comparing different jobs, and this in turn requires gender-neutral #job_evaluation, access to pay data, and a permissible comparator. In practice, workers often cannot identify a suitable comparator, especially in a segregated workplace where there may be no man doing a comparable job, or cannot obtain the pay information needed to build a case. Pay secrecy has historically been a formidable obstacle, which is precisely why the recent policy trend, exemplified by the European pay transparency Directive, focuses on disclosure (Bennedsen et al., 2023). The third challenge is weak enforcement and weak remedies. As noted above, the international supervisory system relies on reputational pressure rather than sanctions. At the national level, remedies are frequently limited, litigation is slow and expensive, and the burden of proof often falls on the worker who has the least access to the relevant evidence. Where an individual must fund years of litigation to establish that her job is of equal value to a differently titled male job, the theoretical right becomes practically unusable for most workers. The fourth challenge is the informal economy. A very large share of the world's women work informally, without written contracts, formal wage structures, or effective legal protection. Convention No. 100, like most labour standards, was designed with the formal employment relationship in mind, and its machinery has limited reach where that relationship does not exist. For hundreds of millions of women, the debate about job evaluation and comparators is remote from a working life governed by daily bargaining and insecurity. This is not a marginal exception but, in many economies, the majority experience. Domestic workers, home-based workers, market traders, and casual agricultural labourers rarely have an identifiable employer against whom a claim could be brought, still less access to the pay records of a male comparator. In these settings the principle of #equal_value can only be advanced indirectly, through the extension of formal protection, the strengthening of minimum wage floors that lift feminized low-wage sectors, and the organization of informal workers so that they gain a collective voice. The recent attention paid by the ILO to domestic work and to violence and harassment in the world of work reflects a growing recognition that the pay and conditions of the most vulnerable women cannot be addressed by the equal pay principle alone. Any honest account of Convention No. 100 must therefore acknowledge that its most sophisticated tools are least available precisely where the need is greatest. There are also critiques of the underlying concept. Some economists argue that wages should be left to the market and that comparable worth exercises substitute the judgment of evaluators for the judgment of supply and demand, potentially creating distortions. Defenders respond that the market itself is shaped by the historical undervaluation of women's work, so that leaving matters to the market simply preserves an inherited bias, and that job evaluation makes explicit and challengeable the value judgments that markets make silently. A further critique, from within feminist scholarship, is that focusing on comparisons between jobs can distract from the larger project of revaluing care and reproductive labour as a whole. These debates are healthy and remind students that the principle of equal value, though widely endorsed, rests on contested foundations. Finally, the framework has been criticized for being built around a binary conception of sex, and contemporary scholarship increasingly asks how pay equity intersects with race, migration status, disability, and other dimensions of disadvantage, since the gap facing women of colour or migrant women is typically far larger than the average (Clark and Zuccala, 2025). 10. Comparative National and Sectoral Perspectives Because the Convention leaves the method of application to States, national experiences differ widely, and comparing them is instructive. This section sketches several patterns rather than offering an exhaustive survey. A first pattern is found in countries with strong, coordinated #collective_bargaining, particularly in parts of northern Europe. Where wages across a whole sector are set by negotiation between unions and employers, the principle of equal value can be built into the wage grid itself, and gender-neutral job evaluation can be applied at sectoral level. These countries tend to have smaller pay gaps, although even there the gap is far from closed, partly because segregation channels women into lower-paid grids. The lesson is that the machinery of wage setting matters at least as much as the letter of the law. A second pattern is found in countries that rely mainly on legislation and individual enforcement. Here the principle is written into an equal pay statute, and workers who believe they are underpaid must assert their rights, often through a tribunal or court. This model can produce important precedents, but it places a heavy burden on individuals and tends to reach only those with the knowledge, resources, and support to litigate. The recent shift toward mandatory pay reporting and transparency in the European Union is in part an attempt to lighten this burden by making the gap visible without requiring each worker to prove her own case (European Parliament and Council of the European Union, 2023). A third pattern is found in many developing and emerging economies, where the formal law may faithfully reproduce the equal value principle but where the large informal sector, weak labour inspection, and limited institutional capacity blunt its effect. In such contexts, the binding character of Convention No. 100 provides an important benchmark for advocacy and law reform, but the distance between the statute book and the workplace is wide. Sectoral comparison is equally revealing. The care and health sector, discussed above, shows how an entire field can be systematically undervalued because it is coded as women's work (McIsaac and Knaul, 2024). The academic and professional sectors show that even highly educated women face a stubborn residual gap that education alone does not remove, and that the gap tends to open up around the years of family formation. The technology sector shows large unadjusted gaps driven by the concentration of men in the highest-paid roles. Across all these sectors, the common thread is that formal equal pay rules leave untouched the deeper structures of segregation and undervaluation that Convention No. 100, properly applied, is meant to address. These comparisons carry a clear message for students and policymakers. Ratifying the Convention is necessary but not sufficient. What determines outcomes is the interaction between the legal principle, the wage-setting institutions, the availability of pay information, the strength of enforcement, and the wider social distribution of care. Where these align, the gap narrows. Where they do not, a ratified Convention coexists with entrenched inequality. 11. Discussion The analysis so far reveals a paradox. Convention No. 100 is among the most widely ratified and most clearly reasoned of all international labour standards, yet the inequality it targets remains stubbornly large. How should this paradox be understood? The first point is that the paradox is not evidence that the Convention has failed in the way a poorly drafted instrument fails. The concept of #equal_value is sound, and where it is applied in full it works. The problem lies in the repeated shrinking of the principle at the point of national implementation, in the difficulty of comparison, and in the softness of enforcement. In other words, the Convention is doing the intellectual work correctly, but the surrounding machinery is too weak to deliver the result. This diagnosis matters, because it points toward reform of implementation rather than abandonment of the principle. The second point is that pay is downstream of structure. The persistence of the #gender_pay_gap reflects the persistence of occupational segregation, the unequal distribution of unpaid care, and the undervaluation of skills associated with women. A pay equity rule can challenge these structures, but it cannot dismantle them on its own. Closing the gap therefore requires action on several fronts at once, including care policy, education and training that opens male-dominated fields to women, and support for women's continuous participation in paid work. Convention No. 100 is one instrument in a larger toolkit, and expecting it to close the gap by itself sets it up to look like a failure when it is in fact doing its part. The third point concerns the direction of recent policy. The move toward #pay_transparency represents a shift in strategy. For decades the dominant model was individual enforcement, in which an underpaid worker had to discover the gap, find a comparator, and litigate. The new model shifts responsibility toward employers and toward systemic disclosure, requiring organizations to measure and publish their gaps and to act when unexplained gaps appear. The evidence on transparency is still developing, but it suggests that disclosure can narrow gaps, partly by restraining pay growth at the top and partly by giving workers and their representatives the information they need (Bennedsen et al., 2023). This shift can be read as the international community learning from seventy years of experience with Convention No. 100 and building the enforcement architecture that the original instrument lacked. The fourth point is about interpretation over time. The Convention has proved to be a living instrument. Its meaning has been developed and sharpened by the comments of the supervisory bodies, which have insisted on the broad reading of value, on gender-neutral evaluation, and on comparisons beyond the single establishment. This shows that an old standard can remain effective if it is accompanied by an active interpretive practice. The text of 1951 is short, but the body of interpretation built around it is substantial, and it is that interpretation, as much as the text, that gives the Convention its contemporary force. A fifth point concerns the economic case, which is often overlooked in debates that treat pay equity purely as a question of fairness. Persistent pay inequality is not only unjust; it is also inefficient. When the skills of half the workforce are undervalued and underused, an economy forgoes output and productivity. The concentration of women in a narrow band of low-paid occupations wastes talent, weakens household incomes, deepens poverty, and reduces the tax base and pension contributions on which societies depend. Studies of the wider economic consequences of gender gaps in employment and pay find that closing these gaps would produce substantial gains at national and global level, and that the losses from segregation and undervaluation are measured in very large sums (McIsaac and Knaul, 2024). This means that the equal value principle can be defended not only on grounds of rights but also on grounds of sound economic policy, a framing that may carry weight with governments and employers who are unmoved by the language of justice alone. Recognizing the business and economic case does not displace the rights case; rather, the two reinforce each other and widen the coalition that can be assembled behind reform. The overall assessment, then, is cautiously constructive. Convention No. 100 has not solved the problem, but it has defined it correctly, kept it on the international agenda, and provided the conceptual foundation on which more recent and more forceful measures are being built. The task now is less to reinvent the principle than to equip it with the transparency, remedies, and complementary policies needed to make it bite. 12. Conclusion and Recommendations This article has examined the ILO Equal Remuneration Convention, 1951 (No. 100), from its origins to its contemporary significance. It has shown that the Convention's central achievement was the choice of the phrase "work of #equal_value," which allows the comparison of different jobs and thereby reaches the structural undervaluation of women's work that a narrower rule would miss. It has set out the Convention's legal architecture, explained the role of gender-neutral job evaluation, described the supervisory machinery, and placed the instrument within a wider framework that runs from the human rights conventions to the Sustainable Development Goals and the European pay transparency regime. It has also confronted the uncomfortable fact that, despite near-universal ratification, the global #gender_pay_gap remains close to twenty per cent and is closing only slowly, if at all. The core conclusion is that the gap between the legal ideal and the economic reality is caused less by any defect in the Convention than by weaknesses in how it is implemented and enforced. Narrow national transposition, the practical difficulty of comparison, weak remedies, the reach of the informal economy, and the deeper structures of segregation and unpaid care all conspire to blunt a sound principle. Several recommendations follow for students, researchers, and policymakers. First, States should transpose the full principle of #equal_value into national law, not merely the narrow principle of equal pay for the same work, and should ensure that comparison can extend beyond a single employer. Second, governments and employers should invest in gender-neutral #job_evaluation and make the analytical tools available to smaller employers who cannot build them alone. Third, #pay_transparency should be strengthened, with mandatory reporting of pay gaps, a right for workers to obtain relevant pay information, and a duty to act where unexplained gaps appear. Fourth, remedies should be made effective, with a fair allocation of the burden of proof, accessible procedures, and adequate compensation, so that the right does not remain theoretical. Fifth, pay equity policy should be joined to care policy, education, and measures against occupational segregation, because pay outcomes are shaped by all of these. Sixth, the supervisory comments of the ILO should be used actively at national level as a guide to reform rather than treated as distant technical documents. For students, the enduring lesson of Convention No. 100 is that a well-drafted international standard is a necessary foundation but never a sufficient one. Rights on paper become real only through institutions, information, and enforcement, and through the political will to challenge structures that have long been taken for granted. The principle that a person's pay should reflect the value of their work and not their sex is now more than a century old as an idea and more than seventy years old as a binding standard. Whether it becomes a lived reality within the working lives of today's students will depend not on the wording of the Convention, which is already sound, but on the seriousness with which the world chooses to apply it. References Bennedsen, M., Larsen, B., Simintzi, E., and Scur, D. (2023). Gender wage transparency and the gender pay gap: A survey. Journal of Economic Surveys. https://doi.org/10.1111/joes.12545 Chordiya, R., and Hubbell, L. (2023). Fostering internal pay equity through gender-neutral job evaluations: A case study of the Federal Job Evaluation System. Public Personnel Management. https://doi.org/10.1177/00910260221124866 Clark, J., and Zuccala, E. (2025). Gender pay gaps and inequity at science publishers. PLOS Global Public Health, 5(6). https://doi.org/10.1371/journal.pgph.0004673 European Parliament and Council of the European Union. (2023). Directive (EU) 2023/970 of 10 May 2023 to strengthen the application of the principle of equal pay for equal work or work of equal value between men and women through pay transparency and enforcement mechanisms. Official Journal of the European Union, L 132, 21 to 44. International Labour Organization. (1951). Equal Remuneration Convention, 1951 (No. 100). Geneva: International Labour Office. International Labour Organization. (1951). Equal Remuneration Recommendation, 1951 (No. 90). Geneva: International Labour Office. International Labour Organization. (2022). ILO Action Plan for Gender Equality 2022 to 2025. Geneva: International Labour Office. International Labour Organization. (2024). Global Wage Report 2024 to 2025: Is wage inequality decreasing globally? Geneva: International Labour Office. International Labour Organization. (2025). Report of the Committee of Experts on the Application of Conventions and Recommendations (Report III, Part A). Geneva: International Labour Office. McIsaac, M., and Knaul, F. M. (2024). How economic implications of gender gaps in employment affect global health equity. Bulletin of the World Health Organization, 102(2). https://doi.org/10.2471/BLT.23.291271 Organisation for Economic Co-operation and Development. (2026). Pay Transparency in Progress: Gender-neutral job evaluation and classification. Paris: OECD Publishing. Roussille, N. (2024). The role of the ask gap in gender pay inequality. The Quarterly Journal of Economics. https://doi.org/10.1093/qje/qjae004 United Nations. (1979). Convention on the Elimination of All Forms of Discrimination against Women. New York: United Nations. Valencia-Contrera, M., and Rivera-Rojas, F. (2023). Gender job segregation. Revista Brasileira de Medicina do Trabalho, 21(3). https://doi.org/10.47626/1679-4435-2023-1108 #Equal_Remuneration_Convention #Convention_No_100 #equal_pay_for_equal_value #ILO #international_labour_standards #gender_pay_gap #pay_equity #work_of_equal_value #equal_remuneration #decent_work #women_at_work #labour_law #non_discrimination #job_evaluation #pay_transparency #comparable_worth #gender_equality #SDG8 #fair_wages #closing_the_pay_gap

The Right to Organise and Collective Bargaining Convention, 1949 (No. 98) is one of the oldest and most widely ratified instruments in the body of international labour law. It sets out two connected guarantees: that workers must be shielded from acts that punish them for belonging to or acting through a union, and that the state must actively encourage free and voluntary negotiation between employers and workers over pay and conditions. This article examines the legal structure of the Convention article by article, situates it within the wider framework of the International Labour Organization, and analyses how the supervisory bodies of that organisation have given the text practical meaning over more than seven decades. It argues that the Convention's strength lies not only in its formal wording but in the accumulated interpretation produced by expert committees and the freedom of association machinery. The study also considers where the instrument is under strain, including the rise of platform work, the persistence of the informal economy, falling union density in many industrialised states, and the difficulty of enforcing labour standards across long global supply chains. The article concludes that Convention No. 98 remains a durable reference point for workers' rights, but that its real-world effect depends heavily on national enforcement, credible remedies, and the political willingness of governments to treat labour rights as human rights rather than as obstacles to competition. Keywords: anti-union discrimination; collective bargaining; freedom of association; International Labour Organization; Convention No. 98; trade union rights; labour law; social dialogue 1. Introduction Few areas of law sit closer to the daily reality of working life than the rules that decide whether a person can join a union without losing their job, and whether that union can sit across a table from an employer and negotiate. These questions are the core of the Right to Organise and Collective Bargaining Convention, 1949, known within the #International_Labour_Organization simply as #Convention_98. The instrument protects #workers against acts of #anti_union_discrimination and requires member states to promote genuine #collective_bargaining. Its opening article states, in plain terms, that workers must enjoy adequate protection against acts of anti-union discrimination in respect of their employment. The importance of this guarantee is easy to underestimate until it is absent. Where a worker can be dismissed simply for signing a union card, or where an employer can quietly fund a compliant in-house union to keep an independent one out, the formal right to organise means very little. Convention No. 98 was drafted to close exactly these gaps. It complements the earlier Freedom of Association and Protection of the Right to Organise Convention, 1948, or #Convention_87, which guarantees the right of workers and employers to form and join organisations of their own choosing. Together the two instruments form the backbone of what the ILO treats as a single #fundamental_principle: freedom of association and the effective recognition of the right to bargain collectively. Convention No. 98 was adopted on 1 July 1949 at the thirty-second session of the International Labour Conference in Geneva, and it entered into force on 18 July 1951. It has since become one of the most ratified conventions the ILO has ever produced, binding 168 member states. Only a handful of countries remain outside it, the most prominent being the United States, whose domestic law and federal structure have long complicated ratification. This near-universal acceptance gives the Convention unusual authority. When almost every state in the world has formally accepted a legal standard, that standard becomes hard to dismiss as the preference of one region or economic bloc. This article has three aims. First, it explains what the Convention actually says and requires, reading its articles closely and drawing out the meaning that the ILO's own bodies have attached to them. Second, it describes how the Convention is supervised and enforced, since a treaty without a mechanism to check compliance is little more than a statement of hope. Third, it looks honestly at the pressures the Convention now faces, from the growth of #platform_work to the erosion of #trade_unions in many wealthy economies. The overall claim is that Convention No. 98 has proven remarkably durable, but that durability on paper should not be confused with protection in practice. 2. Historical and Institutional Background The International Labour Organization was created in 1919, in the aftermath of the First World War, on the conviction that lasting peace could not rest on social injustice. Its structure is unusual among international bodies because it is tripartite: governments, employers, and workers each hold seats and votes. This design was deliberate. Labour standards drafted only by governments risk ignoring the people who must live under them, so the ILO built worker and employer voices directly into its law-making. For the first three decades of its existence, however, the ILO had no binding standard that squarely protected the independence of unions from both the state and the employer. The gap became urgent after the Second World War, as many countries rebuilt their industrial relations systems and the newly formed United Nations placed human rights at the centre of the international agenda. In 1948 the Conference adopted Convention No. 87, which secured the right to establish and join organisations free from government interference. Yet Convention No. 87 said little about the employer relationship or about bargaining itself. A worker could, in theory, form a union under Convention No. 87 and still be lawfully dismissed by an employer for doing so. Convention No. 98 was written the following year precisely to answer that problem. The two conventions were understood at the time as a pair. With their adoption, the ILO could claim two #fundamental instruments guaranteeing the autonomy of unions against public authorities on one side and against employers on the other. Contemporary commentary from within the ILO records the anxiety that surrounded them: many feared that only a few states would ratify. That fear proved largely unfounded, particularly for Convention No. 98, which was taken up by the large majority of member states within a few decades. The status of the Convention was reinforced in 1998, when the Conference adopted the Declaration on Fundamental Principles and Rights at Work. The Declaration identified a small set of principles so basic that all ILO members are bound to respect them simply by virtue of membership, whether or not they have ratified the relevant conventions. Freedom of association and the effective recognition of the right to #collective_bargaining were placed at the top of that list. This move changed the legal character of Convention No. 98. Even a state that had not ratified it could no longer claim to be free of the underlying obligation, because the principle itself had become a condition of belonging to the organisation. The framework has continued to evolve. Following the adoption of the 2014 Protocol to the Forced Labour Convention, a further instrument joined the fundamental category. Then, at the 110th session of the Conference in June 2022, the ILO added a safe and healthy working environment to its list of fundamental principles, which brought the occupational safety and health conventions into the fundamental group. What began as eight fundamental conventions has grown, but Convention No. 98 has remained a fixed point throughout. Its place in the architecture of #labour_rights has never been in doubt. 3. Research Approach This article uses a doctrinal method, which is the standard approach for legal analysis. Doctrinal work treats authoritative legal texts as its primary material and asks what those texts mean, how they fit together, and how the bodies charged with applying them have interpreted them over time. The primary sources here are the Convention itself, its companion Convention No. 87, and the 1998 Declaration. To these are added the interpretive output of the ILO supervisory bodies, whose accumulated observations and decisions function much like a body of case law even though they are not, strictly speaking, judgments of a court. The analysis is also comparative in a limited way. Because Convention No. 98 operates through national implementation, its practical meaning can only be understood by looking at how states have translated it into domestic law and where they have fallen short. The article draws on examples from several jurisdictions to illustrate both good practice and recurring failures. Finally, the study is informed by recent scholarship in labour law and industrial relations, which has increasingly examined how classic guarantees such as those in Convention No. 98 apply to new forms of work that the drafters in 1949 could not have imagined. The intention throughout is to keep the discussion accessible to students while preserving the analytical care expected of a peer-reviewed article. 4. The Normative Architecture of Convention No. 98 The Convention is short. Its substantive obligations are contained in the first six articles, and the remaining provisions deal with technical matters such as ratification, entry into force, and revision. This brevity is a feature rather than a defect. The drafters set out broad principles and left the detail to national law and to interpretation, which has allowed the instrument to remain relevant across very different legal systems and across many decades. 4.1 Article 1: Protection against anti-union discrimination The heart of the Convention is Article 1. It provides that #workers shall enjoy adequate protection against acts of anti-union discrimination in respect of their employment. The article then gives two concrete examples of the acts it targets. The first is making a person's employment conditional on not joining a union or on giving up union membership. This captures the so-called yellow-dog contract, in which a worker must promise never to organise as the price of being hired. The second is dismissing or otherwise prejudicing a worker because of union membership or because of taking part in union activities, whether outside working hours or, with the employer's consent, during working hours. Two points about this wording deserve emphasis. First, the protection is described as an individual right held by the worker. It is the person who is shielded, not merely the organisation. Second, the phrase "or otherwise prejudice" is deliberately open. Anti-union #discrimination is not limited to dismissal. It reaches demotion, transfer to an undesirable post, denial of promotion, cuts to hours, exclusion from training, and the many quieter penalties that employers can impose on someone who has become inconvenient. The ILO bodies have read the article to cover harassment and intimidation as well, on the reasoning that acts designed to discourage union membership violate the right to organise even when they do not formally change the terms of employment. The word "adequate" carries real weight. The supervisory bodies have insisted that protection is adequate only when it is backed by remedies that genuinely deter. A legal rule that forbids anti-union dismissal but offers only token compensation is not adequate protection, because a well-resourced employer may treat the payout as an acceptable cost of removing an organiser. Effective remedies typically include reinstatement of the dismissed worker, back pay, and penalties sufficient to make the employer think twice. Where reinstatement is impossible or refused, compensation must be high enough to be dissuasive rather than nominal. 4.2 Article 2: Protection against acts of interference Article 2 shifts the focus from the individual worker to the organisation. It provides that workers' and employers' organisations must be protected against acts of interference by one another in their establishment, functioning, or administration. The provision is symmetrical in form, protecting both sides, but in practice its main target is the domination of workers' organisations by employers. The second paragraph of the article makes this concrete. It treats as interference any act designed to promote the creation of workers' organisations under the control of employers, or to support such organisations through financial or other means with the aim of placing them under employer control. This is the classic problem of the company union: an organisation that looks like a union and even calls itself one, but that exists to serve the employer and to keep an independent union out. By naming and prohibiting this tactic, Article 2 protects the very independence that makes #collective_bargaining meaningful. A union that is funded and steered by management cannot bargain at arm's length, and a bargaining process between an employer and its own creature is a hollow ritual. 4.3 Article 3: Machinery to ensure respect Rights on paper are worth little without institutions to give them effect. Article 3 addresses this directly by requiring that machinery appropriate to national conditions be established, where necessary, to ensure respect for the right to organise as set out in the previous articles. The phrase "appropriate to national conditions" gives states latitude to design the mechanism that fits their legal culture. In some countries this means specialised labour courts; in others it means labour inspectorates, tribunals, or administrative bodies with the power to investigate complaints and order remedies. What Article 3 does not permit is the absence of any effective mechanism at all. A state that guarantees the right to organise but provides no forum to vindicate it has not met its obligation. 4.4 Article 4: The promotion of voluntary collective bargaining Article 4 turns to the second great theme of the Convention. It requires states to take measures appropriate to national conditions, where necessary, to encourage and promote the full development and use of machinery for voluntary negotiation between employers or employers' organisations and workers' organisations, with the aim of regulating terms and conditions of employment through #collective_agreements. Several ideas are packed into this sentence. The first is that bargaining should be voluntary. The state is to encourage and promote negotiation, not to impose the content of agreements or to force parties into compulsory arbitration as a matter of course. The supervisory bodies have consistently held that free and voluntary bargaining is a core principle, and that measures which effectively replace negotiation with state-imposed outcomes are generally incompatible with the article. The second idea is that the obligation is one of promotion. The state must create a favourable environment, remove obstacles, and put in place the procedures through which bargaining can happen. It is not enough for a government to stand back and declare that parties are free to bargain if they wish; the language of "encourage and promote" imposes a positive duty. The third idea is that the purpose of bargaining is the regulation of employment terms through agreements that have real effect. Bargaining that cannot produce binding results is not the bargaining the Convention has in mind. The reference to "workers' organisations" also carries an important implication about who may bargain. The supervisory bodies have taken the view that the right to negotiate belongs to genuine, independent trade unions rather than to ad hoc groups assembled by an employer. Where an employer bypasses a representative union to strike deals with non-union groups of workers, the effect can be to undermine the union and to weaken the bargaining right the Convention protects. 4.5 Articles 5 and 6: The permitted exclusions The Convention does not extend to everyone. Article 5 allows national law to determine the extent to which the guarantees apply to the armed forces and the police, and it provides that ratification does not affect existing laws or arrangements under which members of the armed forces or police enjoy rights guaranteed by the Convention. This exclusion reflects a widely shared, if contested, view that the special functions of the military and police justify particular rules on organisation. Article 6 excludes public servants engaged in the administration of the state, while adding that the Convention should not be read as prejudicing their rights or status in any way. The scope of this exclusion has generated considerable interpretation. The supervisory bodies have read it narrowly, confining it to officials who genuinely exercise authority in the name of the state, such as senior civil servants in ministries. Ordinary public sector workers, such as teachers, nurses, postal workers, and employees of public enterprises, are not excluded and are entitled to the protections of the Convention, including the right to bargain collectively. Reading Article 6 too broadly would allow a state to strip millions of public employees of their bargaining rights simply by classifying them as administrators, which would defeat the purpose of the instrument. 5. Convention No. 98 within the Wider ILO Framework Convention No. 98 does not stand alone. It works alongside Convention No. 87 and a number of other instruments that together form the ILO's law on freedom of association and #collective_bargaining. Understanding this network helps explain what Convention No. 98 does and does not cover. Convention No. 87 secures the right to organise as against the state. It protects the freedom to form and join organisations, the right of those organisations to draw up their own rules and elect their own officers, and their right to function without dissolution or suspension by administrative authority. Convention No. 98 secures the right to organise as against the employer, and it adds the positive promotion of bargaining. The division of labour is clean: Convention No. 87 concerns the relationship between organisations and the government, while Convention No. 98 concerns the relationship between organisations, employers, and the machinery of negotiation. Neither is complete without the other. Beyond these two, the ILO has adopted further instruments that build on the same foundation. The Workers' Representatives Convention, 1971 protects elected representatives at the workplace from prejudice and gives them facilities to carry out their functions. The Collective Bargaining Convention, 1981 develops the promotion of bargaining in more detail, and the accompanying recommendation encourages measures such as the extension of collective agreements to cover workers beyond the immediate signatories. The Labour Relations (Public Service) Convention, 1978 addresses bargaining in the public sector, where the exclusions in Article 6 of Convention No. 98 leave gaps. These instruments show that Convention No. 98 is a starting point rather than the last word. It sets the fundamental guarantee, and later conventions refine and extend it. The ILO also classifies its conventions into categories. Convention No. 98 sits among the fundamental conventions, those regarded as expressing basic principles and rights at work. A separate group of governance or priority conventions concerns the machinery of the system, such as labour inspection and tripartite consultation, while a large body of technical conventions addresses specific subjects. Placing Convention No. 98 in the fundamental group signals that it is not merely one standard among hundreds but part of the irreducible core of #labour_rights that the organisation exists to defend. 6. The Concept of Anti-Union Discrimination in Practice The phrase "anti-union discrimination" can sound abstract, but the conduct it describes is concrete and, in many workplaces, familiar. It is worth setting out the main forms this conduct takes, because the effectiveness of Convention No. 98 depends on whether national systems recognise and punish them. The clearest form is dismissal for union activity. A worker who helps organise a union, who stands for election as a representative, or who leads a grievance is fired on a pretext, or with no reason given where the law allows dismissal at will. This is the tactic the Convention most directly targets, and it remains common. Reports from workers' organisations and human rights monitors regularly document the dismissal of union organisers as a first response by hostile employers, precisely because removing the organiser can stop a campaign before it starts. A second form is discrimination in hiring. Employers may screen out applicants known to be union supporters, or require new hires to sign statements renouncing union membership. Because it happens before the employment relationship begins, this form is harder to detect and harder to remedy, yet it is squarely within Article 1. A third form is the subtler penalty imposed on those who remain employed. Union activists may find themselves passed over for promotion, moved to less desirable shifts or locations, denied overtime, excluded from training that others receive, or subjected to heightened scrutiny and discipline. Each act may be small, and each may be dressed in a neutral business justification, but the pattern communicates a clear message to the workforce. The supervisory bodies have recognised that such acts, taken together, can be as effective as dismissal in discouraging organisation. A fourth form is intimidation and surveillance. Threats to close a workplace if workers unionise, monitoring of union meetings, and the use of anti-union consultants to run campaigns against organising drives all fall within the mischief the Convention addresses. Recent monitoring has drawn attention to a growing use of digital surveillance of workers, which raises the prospect that anti-union conduct can now be carried out through data and algorithms rather than through open confrontation. The response required by the Convention is not merely a prohibition but a system of deterrence and repair. The ILO bodies have repeatedly stressed that protection is meaningful only where the law provides for rapid procedures, effective remedies, and penalties that are sufficiently dissuasive. Where a worker is dismissed for union activity, the preferred remedy is reinstatement together with payment for lost wages. Where reinstatement is genuinely impossible, compensation must be set at a level that removes any incentive for the employer to repeat the conduct. A regime in which the cost of unlawful dismissal is trivial fails the test of adequacy, however elegant its prohibitions may look on paper. 7. Collective Bargaining as a Fundamental Right If Article 1 answers the question of whether workers may organise safely, Article 4 answers the question of what that organisation is for. The purpose of a union, in the scheme of Convention No. 98, is above all to bargain: to negotiate with the employer over wages, hours, and conditions and to reach agreements that regulate the employment relationship. #Collective_bargaining is therefore not a peripheral benefit of unionisation but its central function, and the Convention treats the promotion of bargaining as a duty of the state. Several principles have grown up around Article 4 through decades of interpretation. The principle of free and voluntary negotiation holds that the parties, not the state, should determine the content of their agreements. Governments may set a framework, provide conciliation, and encourage settlement, but they should not routinely substitute compulsory arbitration or legislative fiat for genuine negotiation. There are recognised exceptions, such as essential services where interruption would endanger life or public safety, but these are read narrowly. A second principle concerns good faith. Bargaining is not a genuine process if one party goes through the motions without any intention of reaching agreement. The supervisory bodies have emphasised that both sides should make efforts to reach a settlement, should engage in real dialogue, and should avoid unjustified delays. Good faith does not require either side to concede, but it does require honest engagement. A third principle concerns representativeness. Systems must have a fair way of determining which organisation is entitled to bargain, and the criteria used should be objective and known in advance. Where a single union represents a clear majority, granting it bargaining rights is straightforward. Where representation is fragmented, the system must still allow bargaining to proceed without allowing the employer to pick a favoured partner or to use fragmentation as an excuse to avoid negotiation altogether. A fourth principle concerns the level at which bargaining takes place. The Convention does not prescribe a single model. In some countries bargaining happens mainly at the level of the individual enterprise, while in others it is organised by sector or even at the national level through broad social pacts. The supervisory bodies have taken the position that the choice of level should be left to the parties and should not be dictated or restricted by the state. Systems that force bargaining down to the enterprise level, or that block sector-wide negotiation, can weaken the bargaining power of #workers, since a single small workplace often has little leverage on its own. The promotion duty in Article 4 therefore includes not obstructing the level of bargaining that #trade_unions and employers freely choose. A related question is the extension of #collective_agreements. In many systems, an agreement reached between representative parties in a sector can be extended by the authorities to cover all employers and workers in that sector, including those who were not signatories. The accompanying ILO recommendation encourages such extension where appropriate to national practice. Extension mechanisms are one of the main reasons why bargaining coverage in some countries remains high even where union membership is modest, because coverage is decoupled from individual membership. Where these mechanisms are dismantled, coverage tends to collapse toward the level of membership, which helps explain the sharp falls recorded in several states that suspended or restricted extension during periods of reform. The wider value of #collective_bargaining is not only individual but social. Comparative evidence gathered by international bodies indicates that strong bargaining institutions are associated with lower wage inequality, more stable industrial relations, and greater resilience during economic shocks. At the same time, the reach of bargaining has narrowed in many countries. Data compiled by the Organisation for Economic Co-operation and Development show that the share of employees covered by collective agreements across its member countries fell substantially over recent decades, and that trade union density declined over the same period in most of these economies. This decline does not lessen the legal obligation in Article 4; if anything, it sharpens the question of whether states are truly meeting their duty to promote bargaining, or merely tolerating its slow retreat. 8. The Supervisory and Enforcement Machinery A treaty is only as strong as the system that watches over it. The ILO has built one of the most developed #supervisory_systems in international law, and it is through this machinery that Convention No. 98 acquires practical force. The system has two broad arms: a regular arm that reviews reports from every ratifying state, and a set of complaint-based procedures that respond to specific allegations. 8.1 The regular supervisory system At the centre of the regular system is the Committee of Experts on the Application of Conventions and Recommendations. Established in 1926, it is composed of twenty independent jurists drawn from different regions, legal traditions, and cultures. Ratifying states are required to report periodically on how they are applying the conventions they have accepted, and the Committee of Experts examines these reports. It produces observations, which are published comments on particular countries, and direct requests, which raise more technical points. Although its conclusions are not binding judgments, they carry great weight because they represent an impartial technical assessment built up over many years and applied consistently across states. The observations of the Committee of Experts feed into the Conference Committee on the Application of Standards, a tripartite body that meets during the annual International Labour Conference. This committee selects a number of cases each year for public examination, questions government representatives, and adopts conclusions. The process is often described as the point where the technical work of the experts meets the political weight of the tripartite Conference. A government summoned to explain a serious failure to apply Convention No. 98 before an audience of its peers, together with employer and worker delegates from around the world, faces a form of reputational pressure that few other international systems can match. 8.2 The freedom of association procedures Alongside the regular system sits a special mechanism devoted to freedom of association. The Committee on Freedom of Association was created in 1951, soon after the adoption of Conventions Nos. 87 and 98, when the ILO concluded that these principles needed a dedicated procedure. The committee is a tripartite body of the Governing Body, with an independent chairperson and representatives of governments, employers, and workers. Its most striking feature is that it can examine complaints whether or not the state concerned has ratified the relevant conventions, because the underlying principle of freedom of association is treated as flowing from ILO membership itself. Over its history the committee has examined more than three thousand cases, and many governments have acted on its recommendations and reported changes in law or practice as a result. The complaints procedure of the Committee on Freedom of Association allows workers' and employers' organisations to bring allegations directly. This gives unions a route to Geneva that does not depend on their own government's willingness to raise a matter. When a union alleges, for example, the mass dismissal of organisers or the refusal of an employer or state to bargain, the committee examines the facts, sets out the applicable principles, and makes recommendations. Its accumulated decisions have been compiled and published, and this compilation functions as a practical guide to what freedom of association and #collective_bargaining require in concrete situations. 8.3 Representations and complaints under the Constitution Two further procedures arise directly from the ILO Constitution. Under Article 24, an industrial association of workers or employers may make a representation alleging that a state has failed to observe a convention it has ratified. A tripartite committee examines the representation and reports to the Governing Body. Under Article 26, a member state, a Conference delegate, or the Governing Body itself may file a complaint against a ratifying state, which can lead to the establishment of a Commission of Inquiry. The commission is the most solemn of the ILO's investigative mechanisms, capable of conducting a full examination and issuing recommendations. Although used sparingly, its existence signals that persistent and serious violations can attract the organisation's heaviest scrutiny. It is worth asking how findings by these bodies actually produce change, since none of them can hand down a binding, enforceable judgment in the way a national court can. The answer lies in a combination of pressures. A published observation that a state is failing to protect #workers against #anti_union_discrimination becomes part of a permanent record that other governments, trade partners, investors, and civil society can read. A case examined before the Conference Committee exposes the government to direct questioning in a public tripartite forum. Recommendations of the freedom of association machinery are followed up over successive sessions, so a government that ignores them faces repeated and escalating scrutiny. Over time, many states have amended their laws, reinstated dismissed unionists, or dropped restrictions on bargaining in response. The record of the Committee on Freedom of Association shows that a large number of countries have acted on its recommendations and reported concrete improvements. The limits of the system are equally real. Compliance ultimately depends on the state's own willingness to act. A determined government can absorb criticism, delay, or offer formal changes that are not implemented in practice. There is no international police force for #labour_rights, and the ILO cannot override national sovereignty. What the system offers instead is legitimacy, transparency, and persistence. It converts scattered complaints into a coherent and authoritative body of interpretation, and it keeps the spotlight on violations long after the initial headlines have faded. For a #supervisory_system operating among sovereign states without coercive power, this is a considerable achievement, even if it falls short of guaranteed enforcement. Taken together, these procedures form a layered system. Most compliance work happens quietly through the regular reporting cycle. Where problems are more acute, the freedom of association machinery and the constitutional procedures provide escalating responses. The system has no power to impose fines or to send anyone to prison. Its authority is persuasive and reputational, resting on publicity, peer pressure, technical credibility, and the slow accumulation of an interpretive record that governments find difficult to ignore. 9. National Implementation and Comparative Perspectives Because Convention No. 98 operates through domestic law, its meaning on the ground varies from country to country. Ratification is a promise to bring national law and practice into line with the Convention, but the distance between promise and performance can be wide. A few examples illustrate the range. Some states have ratified relatively recently and used the moment to reform their systems. Vietnam ratified Convention No. 98 in 2019, with the instrument taking effect in 2020, as part of broader labour reforms linked to its trade commitments. The ratification was accompanied by changes to the labour code intended to allow more independent worker representation and to strengthen bargaining, and it was supported by ILO technical cooperation. Whether the reforms deliver genuine independence in practice is a question that the supervisory bodies continue to follow, but the ratification marked a significant shift in a system that had long channelled representation through a single official structure. Mexico offers a related example. Its ratification of Convention No. 98 in 2018 was tied to a reckoning with so-called protection contracts, agreements negotiated between employers and unions that workers had no real part in choosing and that served to pre-empt genuine bargaining. Ratification brought the Mexican system under the full scrutiny of the ILO's supervisory machinery and reinforced domestic reforms aimed at ensuring that workers can vote on their contracts and choose their representatives freely. The episode shows how ratification can be used as a lever to dismantle structures that undermine authentic #collective_bargaining. Canada ratified the Convention in 2017, comparatively late for an industrialised democracy, reflecting the fact that formal ratification sometimes lags behind substantive protection. Many states protect union rights through constitutional guarantees, statutes, and judicial decisions well before they formally ratify, and the act of ratification then confirms and locks in protections that already exist. The most discussed non-ratifying state is the United States. It has not ratified Convention No. 98, and its position is often explained by reference to its federal structure, in which labour matters are divided between national and state authority, and by the practice of ratifying only those conventions that already conform to existing domestic law. Critics, including workers' organisations and human rights monitors, have long argued that United States labour law falls short of the Convention's standard in several respects, pointing to weak remedies for anti-union dismissal, the permanent replacement of striking workers, and the exclusion of significant categories of workers from bargaining rights. The debate illustrates that non-ratification does not place a state entirely outside the reach of the principle, since the 1998 Declaration binds all members to respect freedom of association and #collective_bargaining regardless of ratification, and the freedom of association machinery can examine complaints against non-ratifying states. A further dimension of national implementation is the growing link between #labour_rights and trade. In recent decades many trade agreements have included labour chapters that commit the parties to respect fundamental principles, including freedom of association and the effective recognition of #collective_bargaining. The trade dimension helps explain several ratifications. Vietnam's decision to ratify Convention No. 98 was closely connected to its trade relationship with the European Union, whose agreement encouraged the adoption of ILO fundamental conventions. Mexico's labour reforms were reinforced by commitments in its regional trade arrangement, which included mechanisms allowing complaints about the denial of bargaining rights at particular workplaces. These developments show how the standards of Convention No. 98 can be given additional force through economic leverage that the ILO's own machinery lacks. The trade link is not without difficulty. Critics point out that labour clauses are unevenly enforced, that they may be used selectively for competitive or political reasons, and that they can place the burden of compliance on exporting states while doing little for #workers in the informal economy who produce nothing for export. There is also a risk that trade-based enforcement treats labour rights as a bargaining chip rather than as fundamental rights that deserve protection in their own right. Even so, the incorporation of the Convention's principles into trade instruments has given them reach into national systems that might otherwise have resisted change, and it reflects a wider recognition that the suppression of bargaining can distort competition by holding wages and conditions artificially low. Across these examples a common pattern appears. Ratification is necessary but not sufficient. What matters is whether national law provides real protection against #anti_union_discrimination, whether remedies deter, whether the state actively promotes bargaining, and whether independent unions can function without capture by employers or governments. The supervisory bodies devote much of their work to precisely these questions, and their country observations record both progress and persistent failures. 10. Contemporary Challenges Convention No. 98 was drafted for a world of factories, established employers, and stable full-time jobs. That world has not disappeared, but it has been joined by new forms of work and new pressures that test the Convention's reach. Several challenges stand out. The first is the rise of #platform_work and the gig economy. Digital platforms often classify the people who work through them as independent contractors rather than employees. This classification, whatever its merits in other respects, can place workers outside the protective scope of labour law, including the right to organise and bargain. If a rider or driver is not an employee, national law may deny them the machinery of #collective_bargaining altogether, even though their economic position closely resembles that of an employee. Recent scholarship has examined how platform workers have nonetheless mobilised and pressed for recognition, drawing on a human rights framing of freedom of association to argue that the classification of a worker should not be allowed to extinguish a fundamental right. The tension between formal legal categories and the substance of the working relationship is one of the sharpest issues now facing the Convention. The second challenge is the informal economy. In large parts of the world, most workers are outside the formal system entirely, without written contracts, registration, or the institutions through which bargaining ordinarily happens. Street vendors, domestic workers, and casual labourers may have a formal right to organise but no practical channel to exercise it. Recent work in labour law has explored how the guarantees of #collective_bargaining might be extended to these workers, including through arrangements that do not depend on a conventional employer-employee relationship. The Convention's language is broad enough to accommodate such extensions, but doing so requires imagination and political will that many states have not shown. The third challenge is the long-term decline of #trade_unions in many industrialised countries. As union density falls and bargaining coverage shrinks, the machinery that Article 4 requires states to promote weakens from within. A right to bargain is of limited value if few workers belong to organisations capable of bargaining. The data assembled by international bodies show a marked drop in both membership and coverage across many advanced economies over recent decades. This decline is driven by many factors, including changes in industrial structure, the growth of small and dispersed workplaces, hostile legal environments in some countries, and the individualisation of employment relations. Whatever its causes, it raises a pointed question about whether states are meeting their promotional duty under the Convention or merely watching bargaining institutions erode. The fourth challenge is the global supply chain. Much production now takes place through networks of suppliers spread across many countries, and the workers who make goods for a global brand are often employed by distant subcontractors rather than by the brand itself. Anti-union #discrimination and the suppression of bargaining in supplier factories are widely documented, yet the legal responsibility of the lead firm is often unclear. Recent research on mandatory human rights due diligence examines how law might require companies to identify and address violations of freedom of association in their supply chains, moving beyond voluntary codes of conduct. This is an area where the principle of Convention No. 98 is increasingly invoked but where enforcement remains difficult, because the worker, the immediate employer, and the firm with real economic power may sit in three different jurisdictions. A fifth challenge concerns migrant workers, who in some regions are excluded from union membership by law or in practice, and whose dependent immigration status can make organising perilous. Monitoring bodies have found that migrant workers frequently face both legal and practical barriers to exercising their rights to freedom of association and #collective_bargaining. Systems that tie a worker's legal presence to a single employer are especially corrosive, because the threat of losing that sponsorship can silence any attempt to organise. Finally, there is the long-running debate over the right to strike. The Convention does not mention strikes expressly, yet the supervisory bodies have long treated the right to strike as an essential corollary of freedom of association and #collective_bargaining, since bargaining without any capacity for collective action tends to be bargaining on unequal terms. This interpretation has been contested within the ILO, particularly by the employers' group, and the disagreement has become one of the most significant institutional debates in recent years. Recent scholarship has revisited the international legal basis of the right to strike and its relationship to the freedom of association conventions, and the question of how far collective action is protected remains a live and important one for the future application of Convention No. 98. 11. Convention No. 98 and the Human Rights Framework One of the most important developments in the life of Convention No. 98 has been its absorption into the broader body of international human rights law. When the Convention was adopted in 1949, labour standards and human rights were often treated as separate fields. Over the following decades that separation broke down, and freedom of association and #collective_bargaining came to be understood as human rights in their own right. The International Covenant on Civil and Political Rights, adopted in 1966, protects freedom of association, including the right to form and join trade unions. The International Covenant on Economic, Social and Cultural Rights of the same year protects the right to form and join unions and the right to strike, exercised in conformity with national law. Regional human rights systems have gone further still, with courts interpreting general guarantees of freedom of association to include a right to bargain collectively. Through these instruments and decisions, the principle expressed in Convention No. 98 has been woven into the fabric of human rights law, so that a violation of the right to organise is now widely understood as a violation of human rights and not merely of a technical labour standard. This convergence matters for two reasons. First, it strengthens the normative claim of the Convention. A right recognised across multiple treaties and by multiple bodies is harder to dismiss and easier to invoke. Second, it multiplies the forums in which the right can be asserted. A worker or union denied the right to bargain may be able to rely not only on the ILO machinery but also on human rights bodies, regional courts, and national constitutional guarantees that reflect the same principle. The 1998 Declaration reinforced this integration within the ILO by treating freedom of association and the effective recognition of #collective_bargaining as obligations flowing from membership itself. The overall effect is that Convention No. 98 sits at the centre of a wider web of protection rather than standing alone as a single treaty. 12. Discussion Bringing these threads together, several observations can be made about the significance and the limits of Convention No. 98. The first is that the Convention's authority rests on a combination of near-universal ratification, integration with human rights law, and a sophisticated supervisory system. These features give it a legitimacy that few international instruments enjoy. When 168 states have accepted a standard, when that standard is echoed across the major human rights treaties, and when an independent expert body has interpreted it consistently for decades, the resulting norm is genuinely authoritative. This is why even a non-ratifying state cannot easily claim to be outside the principle. The second observation is that the Convention's real effect depends on national enforcement, and here the record is mixed. The gap between formal ratification and practical protection is the central problem of the whole system. A state may ratify and yet tolerate the routine dismissal of organisers, provide remedies too weak to deter, allow employers to dominate compliant unions, or decline to promote bargaining in any meaningful way. The supervisory bodies can expose these failures and press for change, and they often succeed, but they cannot compel a reluctant government. The system relies on persuasion, publicity, and the slow force of accumulated interpretation. This is both its weakness and, in a world of sovereign states, its realistic strength. The third observation concerns remedies. Much of the practical value of Article 1 turns on whether the sanctions for anti-union #discrimination are dissuasive. The insistence of the ILO bodies on reinstatement and on compensation set high enough to remove any incentive to repeat the conduct is not a technicality. It is the difference between a right that protects and a right that merely records disapproval. National systems that treat unlawful anti-union dismissal as a minor matter, remediable by a small payment, hollow out the Convention regardless of how firmly their statutes prohibit the conduct. The fourth observation is that the Convention faces a structural challenge from the changing shape of work. The instrument assumes a recognisable employer, an identifiable workplace, and a stable relationship within which bargaining can occur. Where work is mediated by platforms, fragmented across supply chains, pushed into the informal economy, or performed by workers whose legal status is precarious, these assumptions weaken. The principle of the Convention remains valid, but its application requires interpretation and, in some cases, new instruments or new legal techniques such as due diligence obligations on lead firms. The scholarship of the last few years has been preoccupied with exactly this task of adapting classic guarantees to new realities, and its conclusions suggest that the future relevance of Convention No. 98 depends on the willingness of states and courts to look at the substance of a working relationship rather than its formal label. The fifth and final observation is that the Convention should be understood as a floor rather than a ceiling. It sets a minimum below which no ratifying state should fall, but it does not prevent states from providing stronger protection, wider coverage, or more generous machinery. The best systems go well beyond the minimum. The danger lies in treating the floor as an aspiration rather than an obligation, and in allowing bargaining institutions to decay while pointing to formal ratification as evidence of compliance. 13. Conclusion The Right to Organise and Collective Bargaining Convention, 1949, is a short instrument that carries a large responsibility. In six substantive articles it protects #workers against #anti_union_discrimination, shields their organisations from capture by employers, requires machinery to enforce these guarantees, and obliges states to promote free and voluntary #collective_bargaining. Adopted alongside its companion Convention No. 87, elevated to fundamental status by the 1998 Declaration, and woven into the wider human rights framework, it has become one of the most widely accepted and authoritative standards in international law, binding 168 states. Yet its value cannot be read off the ratification tables alone. The Convention lives or dies in national practice, in whether a dismissed organiser is reinstated, whether a company union is prohibited, whether a genuine union can reach a binding agreement, and whether remedies bite hard enough to deter. The ILO's supervisory machinery, from the Committee of Experts to the Committee on Freedom of Association, has given the text meaning and has held governments to account for more than seventy years, but its power is persuasive rather than coercive. The pressures of platform work, the informal economy, declining union density, global supply chains, and precarious migration test the Convention as never before. The sensible conclusion is neither triumphant nor despairing. Convention No. 98 has proven durable and adaptable, and its core principle, that people should be able to join together and bargain over the terms of their working lives without fear of reprisal, has lost none of its force. The task now is to make that principle real in forms of work its drafters never anticipated, and to ensure that the promise of protection is matched by the reality of enforcement. For students of #labour_rights, the Convention is both a settled foundation and an unfinished project, a standard whose text has held firm while the world of work around it continues to change. References Primary legal instruments and authoritative institutional sources International Labour Organization. (1948). Freedom of Association and Protection of the Right to Organise Convention (No. 87). Geneva: International Labour Office. International Labour Organization. (1949). Right to Organise and Collective Bargaining Convention (No. 98). Geneva: International Labour Office. International Labour Organization. (1998, amended 2022). ILO Declaration on Fundamental Principles and Rights at Work. Geneva: International Labour Office. International Labour Organization. (2018). Freedom of Association: Compilation of decisions of the Committee on Freedom of Association (6th ed.). Geneva: International Labour Office. International Labour Organization. (2022). Social Dialogue Report 2022: Collective bargaining for an inclusive, sustainable and resilient recovery. Geneva: International Labour Office. Secondary scholarly literature International Trade Union Confederation. (2025). ITUC Global Rights Index 2025. Brussels: International Trade Union Confederation. Lopez Lopez, J. (Ed.). (2022). Inscribing Solidarity: Debates in Labour Law and Beyond. Cambridge: Cambridge University Press. Marshall, S., Landau, I., Shamir, H., Barkay, T., Fudge, J., & van Heerden, A. (2023). Mandatory human rights due diligence: Risks and opportunities for workers and unions. Research Paper, RMIT University and Monash University. Organisation for Economic Co-operation and Development. (2025). Membership of unions and employers' organisations, and bargaining coverage. Paris: OECD Publishing. https://doi.org/10.1787/fe47107c-en Simutina, Y. (2024). The impact of war and European accession on the labour law reform in Ukraine. European Studies: The Review of European Law, Economics and Politics. Stylogiannis, C. (2023). Freedom of association and collective bargaining in the platform economy: A human rights-based approach and an ever-increasing mobilization of workers. International Labour Review. https://doi.org/10.1111/ilr.12340 Vogt, J., Novitz, T., & Fudge, J. (Eds.). (2026). The Right to Strike Reimagined. Oxford: Hart Publishing. Vogt, J., & Subasinghe, R. (2025). Turning up the heat: The right to strike and the climate crisis. Comparative Labor Law & Policy Journal. Von Broembsen, M. (2025). Collective bargaining for the "new" working class: Putting "personal work relations" to work for street vendors. Industrial Law Journal. #ILO_Convention_98 #RightToOrganise #CollectiveBargaining #AntiUnionDiscrimination #FreedomOfAssociation #WorkersRights #LabourLaw #TradeUnionRights #InternationalLabourStandards #DecentWork #SocialDialogue #C98 #LabourRights #IndustrialRelations #Convention87

The Right to Organise and Collective Bargaining Convention, 1949 (No. 98) is one of the oldest and most widely ratified instruments in the body of international labour law. It sets out two connected guarantees: that workers must be shielded from acts that punish them for belonging to or acting through a union, and that the state must actively encourage free and voluntary negotiation between employers and workers over pay and conditions. This article examines the legal structure of the Convention article by article, situates it within the wider framework of the International Labour Organization, and analyses how the supervisory bodies of that organisation have given the text practical meaning over more than seven decades. It argues that the Convention's strength lies not only in its formal wording but in the accumulated interpretation produced by expert committees and the freedom of association machinery. The study also considers where the instrument is under strain, including the rise of platform work, the persistence of the informal economy, falling union density in many industrialised states, and the difficulty of enforcing labour standards across long global supply chains. The article concludes that Convention No. 98 remains a durable reference point for workers' rights, but that its real-world effect depends heavily on national enforcement, credible remedies, and the political willingness of governments to treat labour rights as human rights rather than as obstacles to competition. Keywords: anti-union discrimination; collective bargaining; freedom of association; International Labour Organization; Convention No. 98; trade union rights; labour law; social dialogue 1. Introduction Few areas of law sit closer to the daily reality of working life than the rules that decide whether a person can join a union without losing their job, and whether that union can sit across a table from an employer and negotiate. These questions are the core of the Right to Organise and Collective Bargaining Convention, 1949, known within the #International_Labour_Organization simply as #Convention_98. The instrument protects #workers against acts of #anti_union_discrimination and requires member states to promote genuine #collective_bargaining. Its opening article states, in plain terms, that workers must enjoy adequate protection against acts of anti-union discrimination in respect of their employment. The importance of this guarantee is easy to underestimate until it is absent. Where a worker can be dismissed simply for signing a union card, or where an employer can quietly fund a compliant in-house union to keep an independent one out, the formal right to organise means very little. Convention No. 98 was drafted to close exactly these gaps. It complements the earlier Freedom of Association and Protection of the Right to Organise Convention, 1948, or #Convention_87, which guarantees the right of workers and employers to form and join organisations of their own choosing. Together the two instruments form the backbone of what the ILO treats as a single #fundamental_principle: freedom of association and the effective recognition of the right to bargain collectively. Convention No. 98 was adopted on 1 July 1949 at the thirty-second session of the International Labour Conference in Geneva, and it entered into force on 18 July 1951. It has since become one of the most ratified conventions the ILO has ever produced, binding 168 member states. Only a handful of countries remain outside it, the most prominent being the United States, whose domestic law and federal structure have long complicated ratification. This near-universal acceptance gives the Convention unusual authority. When almost every state in the world has formally accepted a legal standard, that standard becomes hard to dismiss as the preference of one region or economic bloc. This article has three aims. First, it explains what the Convention actually says and requires, reading its articles closely and drawing out the meaning that the ILO's own bodies have attached to them. Second, it describes how the Convention is supervised and enforced, since a treaty without a mechanism to check compliance is little more than a statement of hope. Third, it looks honestly at the pressures the Convention now faces, from the growth of #platform_work to the erosion of #trade_unions in many wealthy economies. The overall claim is that Convention No. 98 has proven remarkably durable, but that durability on paper should not be confused with protection in practice. 2. Historical and Institutional Background The International Labour Organization was created in 1919, in the aftermath of the First World War, on the conviction that lasting peace could not rest on social injustice. Its structure is unusual among international bodies because it is tripartite: governments, employers, and workers each hold seats and votes. This design was deliberate. Labour standards drafted only by governments risk ignoring the people who must live under them, so the ILO built worker and employer voices directly into its law-making. For the first three decades of its existence, however, the ILO had no binding standard that squarely protected the independence of unions from both the state and the employer. The gap became urgent after the Second World War, as many countries rebuilt their industrial relations systems and the newly formed United Nations placed human rights at the centre of the international agenda. In 1948 the Conference adopted Convention No. 87, which secured the right to establish and join organisations free from government interference. Yet Convention No. 87 said little about the employer relationship or about bargaining itself. A worker could, in theory, form a union under Convention No. 87 and still be lawfully dismissed by an employer for doing so. Convention No. 98 was written the following year precisely to answer that problem. The two conventions were understood at the time as a pair. With their adoption, the ILO could claim two #fundamental instruments guaranteeing the autonomy of unions against public authorities on one side and against employers on the other. Contemporary commentary from within the ILO records the anxiety that surrounded them: many feared that only a few states would ratify. That fear proved largely unfounded, particularly for Convention No. 98, which was taken up by the large majority of member states within a few decades. The status of the Convention was reinforced in 1998, when the Conference adopted the Declaration on Fundamental Principles and Rights at Work. The Declaration identified a small set of principles so basic that all ILO members are bound to respect them simply by virtue of membership, whether or not they have ratified the relevant conventions. Freedom of association and the effective recognition of the right to #collective_bargaining were placed at the top of that list. This move changed the legal character of Convention No. 98. Even a state that had not ratified it could no longer claim to be free of the underlying obligation, because the principle itself had become a condition of belonging to the organisation. The framework has continued to evolve. Following the adoption of the 2014 Protocol to the Forced Labour Convention, a further instrument joined the fundamental category. Then, at the 110th session of the Conference in June 2022, the ILO added a safe and healthy working environment to its list of fundamental principles, which brought the occupational safety and health conventions into the fundamental group. What began as eight fundamental conventions has grown, but Convention No. 98 has remained a fixed point throughout. Its place in the architecture of #labour_rights has never been in doubt. 3. Research Approach This article uses a doctrinal method, which is the standard approach for legal analysis. Doctrinal work treats authoritative legal texts as its primary material and asks what those texts mean, how they fit together, and how the bodies charged with applying them have interpreted them over time. The primary sources here are the Convention itself, its companion Convention No. 87, and the 1998 Declaration. To these are added the interpretive output of the ILO supervisory bodies, whose accumulated observations and decisions function much like a body of case law even though they are not, strictly speaking, judgments of a court. The analysis is also comparative in a limited way. Because Convention No. 98 operates through national implementation, its practical meaning can only be understood by looking at how states have translated it into domestic law and where they have fallen short. The article draws on examples from several jurisdictions to illustrate both good practice and recurring failures. Finally, the study is informed by recent scholarship in labour law and industrial relations, which has increasingly examined how classic guarantees such as those in Convention No. 98 apply to new forms of work that the drafters in 1949 could not have imagined. The intention throughout is to keep the discussion accessible to students while preserving the analytical care expected of a peer-reviewed article. 4. The Normative Architecture of Convention No. 98 The Convention is short. Its substantive obligations are contained in the first six articles, and the remaining provisions deal with technical matters such as ratification, entry into force, and revision. This brevity is a feature rather than a defect. The drafters set out broad principles and left the detail to national law and to interpretation, which has allowed the instrument to remain relevant across very different legal systems and across many decades. 4.1 Article 1: Protection against anti-union discrimination The heart of the Convention is Article 1. It provides that #workers shall enjoy adequate protection against acts of anti-union discrimination in respect of their employment. The article then gives two concrete examples of the acts it targets. The first is making a person's employment conditional on not joining a union or on giving up union membership. This captures the so-called yellow-dog contract, in which a worker must promise never to organise as the price of being hired. The second is dismissing or otherwise prejudicing a worker because of union membership or because of taking part in union activities, whether outside working hours or, with the employer's consent, during working hours. Two points about this wording deserve emphasis. First, the protection is described as an individual right held by the worker. It is the person who is shielded, not merely the organisation. Second, the phrase "or otherwise prejudice" is deliberately open. Anti-union #discrimination is not limited to dismissal. It reaches demotion, transfer to an undesirable post, denial of promotion, cuts to hours, exclusion from training, and the many quieter penalties that employers can impose on someone who has become inconvenient. The ILO bodies have read the article to cover harassment and intimidation as well, on the reasoning that acts designed to discourage union membership violate the right to organise even when they do not formally change the terms of employment. The word "adequate" carries real weight. The supervisory bodies have insisted that protection is adequate only when it is backed by remedies that genuinely deter. A legal rule that forbids anti-union dismissal but offers only token compensation is not adequate protection, because a well-resourced employer may treat the payout as an acceptable cost of removing an organiser. Effective remedies typically include reinstatement of the dismissed worker, back pay, and penalties sufficient to make the employer think twice. Where reinstatement is impossible or refused, compensation must be high enough to be dissuasive rather than nominal. 4.2 Article 2: Protection against acts of interference Article 2 shifts the focus from the individual worker to the organisation. It provides that workers' and employers' organisations must be protected against acts of interference by one another in their establishment, functioning, or administration. The provision is symmetrical in form, protecting both sides, but in practice its main target is the domination of workers' organisations by employers. The second paragraph of the article makes this concrete. It treats as interference any act designed to promote the creation of workers' organisations under the control of employers, or to support such organisations through financial or other means with the aim of placing them under employer control. This is the classic problem of the company union: an organisation that looks like a union and even calls itself one, but that exists to serve the employer and to keep an independent union out. By naming and prohibiting this tactic, Article 2 protects the very independence that makes #collective_bargaining meaningful. A union that is funded and steered by management cannot bargain at arm's length, and a bargaining process between an employer and its own creature is a hollow ritual. 4.3 Article 3: Machinery to ensure respect Rights on paper are worth little without institutions to give them effect. Article 3 addresses this directly by requiring that machinery appropriate to national conditions be established, where necessary, to ensure respect for the right to organise as set out in the previous articles. The phrase "appropriate to national conditions" gives states latitude to design the mechanism that fits their legal culture. In some countries this means specialised labour courts; in others it means labour inspectorates, tribunals, or administrative bodies with the power to investigate complaints and order remedies. What Article 3 does not permit is the absence of any effective mechanism at all. A state that guarantees the right to organise but provides no forum to vindicate it has not met its obligation. 4.4 Article 4: The promotion of voluntary collective bargaining Article 4 turns to the second great theme of the Convention. It requires states to take measures appropriate to national conditions, where necessary, to encourage and promote the full development and use of machinery for voluntary negotiation between employers or employers' organisations and workers' organisations, with the aim of regulating terms and conditions of employment through #collective_agreements. Several ideas are packed into this sentence. The first is that bargaining should be voluntary. The state is to encourage and promote negotiation, not to impose the content of agreements or to force parties into compulsory arbitration as a matter of course. The supervisory bodies have consistently held that free and voluntary bargaining is a core principle, and that measures which effectively replace negotiation with state-imposed outcomes are generally incompatible with the article. The second idea is that the obligation is one of promotion. The state must create a favourable environment, remove obstacles, and put in place the procedures through which bargaining can happen. It is not enough for a government to stand back and declare that parties are free to bargain if they wish; the language of "encourage and promote" imposes a positive duty. The third idea is that the purpose of bargaining is the regulation of employment terms through agreements that have real effect. Bargaining that cannot produce binding results is not the bargaining the Convention has in mind. The reference to "workers' organisations" also carries an important implication about who may bargain. The supervisory bodies have taken the view that the right to negotiate belongs to genuine, independent trade unions rather than to ad hoc groups assembled by an employer. Where an employer bypasses a representative union to strike deals with non-union groups of workers, the effect can be to undermine the union and to weaken the bargaining right the Convention protects. 4.5 Articles 5 and 6: The permitted exclusions The Convention does not extend to everyone. Article 5 allows national law to determine the extent to which the guarantees apply to the armed forces and the police, and it provides that ratification does not affect existing laws or arrangements under which members of the armed forces or police enjoy rights guaranteed by the Convention. This exclusion reflects a widely shared, if contested, view that the special functions of the military and police justify particular rules on organisation. Article 6 excludes public servants engaged in the administration of the state, while adding that the Convention should not be read as prejudicing their rights or status in any way. The scope of this exclusion has generated considerable interpretation. The supervisory bodies have read it narrowly, confining it to officials who genuinely exercise authority in the name of the state, such as senior civil servants in ministries. Ordinary public sector workers, such as teachers, nurses, postal workers, and employees of public enterprises, are not excluded and are entitled to the protections of the Convention, including the right to bargain collectively. Reading Article 6 too broadly would allow a state to strip millions of public employees of their bargaining rights simply by classifying them as administrators, which would defeat the purpose of the instrument. 5. Convention No. 98 within the Wider ILO Framework Convention No. 98 does not stand alone. It works alongside Convention No. 87 and a number of other instruments that together form the ILO's law on freedom of association and #collective_bargaining. Understanding this network helps explain what Convention No. 98 does and does not cover. Convention No. 87 secures the right to organise as against the state. It protects the freedom to form and join organisations, the right of those organisations to draw up their own rules and elect their own officers, and their right to function without dissolution or suspension by administrative authority. Convention No. 98 secures the right to organise as against the employer, and it adds the positive promotion of bargaining. The division of labour is clean: Convention No. 87 concerns the relationship between organisations and the government, while Convention No. 98 concerns the relationship between organisations, employers, and the machinery of negotiation. Neither is complete without the other. Beyond these two, the ILO has adopted further instruments that build on the same foundation. The Workers' Representatives Convention, 1971 protects elected representatives at the workplace from prejudice and gives them facilities to carry out their functions. The Collective Bargaining Convention, 1981 develops the promotion of bargaining in more detail, and the accompanying recommendation encourages measures such as the extension of collective agreements to cover workers beyond the immediate signatories. The Labour Relations (Public Service) Convention, 1978 addresses bargaining in the public sector, where the exclusions in Article 6 of Convention No. 98 leave gaps. These instruments show that Convention No. 98 is a starting point rather than the last word. It sets the fundamental guarantee, and later conventions refine and extend it. The ILO also classifies its conventions into categories. Convention No. 98 sits among the fundamental conventions, those regarded as expressing basic principles and rights at work. A separate group of governance or priority conventions concerns the machinery of the system, such as labour inspection and tripartite consultation, while a large body of technical conventions addresses specific subjects. Placing Convention No. 98 in the fundamental group signals that it is not merely one standard among hundreds but part of the irreducible core of #labour_rights that the organisation exists to defend. 6. The Concept of Anti-Union Discrimination in Practice The phrase "anti-union discrimination" can sound abstract, but the conduct it describes is concrete and, in many workplaces, familiar. It is worth setting out the main forms this conduct takes, because the effectiveness of Convention No. 98 depends on whether national systems recognise and punish them. The clearest form is dismissal for union activity. A worker who helps organise a union, who stands for election as a representative, or who leads a grievance is fired on a pretext, or with no reason given where the law allows dismissal at will. This is the tactic the Convention most directly targets, and it remains common. Reports from workers' organisations and human rights monitors regularly document the dismissal of union organisers as a first response by hostile employers, precisely because removing the organiser can stop a campaign before it starts. A second form is discrimination in hiring. Employers may screen out applicants known to be union supporters, or require new hires to sign statements renouncing union membership. Because it happens before the employment relationship begins, this form is harder to detect and harder to remedy, yet it is squarely within Article 1. A third form is the subtler penalty imposed on those who remain employed. Union activists may find themselves passed over for promotion, moved to less desirable shifts or locations, denied overtime, excluded from training that others receive, or subjected to heightened scrutiny and discipline. Each act may be small, and each may be dressed in a neutral business justification, but the pattern communicates a clear message to the workforce. The supervisory bodies have recognised that such acts, taken together, can be as effective as dismissal in discouraging organisation. A fourth form is intimidation and surveillance. Threats to close a workplace if workers unionise, monitoring of union meetings, and the use of anti-union consultants to run campaigns against organising drives all fall within the mischief the Convention addresses. Recent monitoring has drawn attention to a growing use of digital surveillance of workers, which raises the prospect that anti-union conduct can now be carried out through data and algorithms rather than through open confrontation. The response required by the Convention is not merely a prohibition but a system of deterrence and repair. The ILO bodies have repeatedly stressed that protection is meaningful only where the law provides for rapid procedures, effective remedies, and penalties that are sufficiently dissuasive. Where a worker is dismissed for union activity, the preferred remedy is reinstatement together with payment for lost wages. Where reinstatement is genuinely impossible, compensation must be set at a level that removes any incentive for the employer to repeat the conduct. A regime in which the cost of unlawful dismissal is trivial fails the test of adequacy, however elegant its prohibitions may look on paper. 7. Collective Bargaining as a Fundamental Right If Article 1 answers the question of whether workers may organise safely, Article 4 answers the question of what that organisation is for. The purpose of a union, in the scheme of Convention No. 98, is above all to bargain: to negotiate with the employer over wages, hours, and conditions and to reach agreements that regulate the employment relationship. #Collective_bargaining is therefore not a peripheral benefit of unionisation but its central function, and the Convention treats the promotion of bargaining as a duty of the state. Several principles have grown up around Article 4 through decades of interpretation. The principle of free and voluntary negotiation holds that the parties, not the state, should determine the content of their agreements. Governments may set a framework, provide conciliation, and encourage settlement, but they should not routinely substitute compulsory arbitration or legislative fiat for genuine negotiation. There are recognised exceptions, such as essential services where interruption would endanger life or public safety, but these are read narrowly. A second principle concerns good faith. Bargaining is not a genuine process if one party goes through the motions without any intention of reaching agreement. The supervisory bodies have emphasised that both sides should make efforts to reach a settlement, should engage in real dialogue, and should avoid unjustified delays. Good faith does not require either side to concede, but it does require honest engagement. A third principle concerns representativeness. Systems must have a fair way of determining which organisation is entitled to bargain, and the criteria used should be objective and known in advance. Where a single union represents a clear majority, granting it bargaining rights is straightforward. Where representation is fragmented, the system must still allow bargaining to proceed without allowing the employer to pick a favoured partner or to use fragmentation as an excuse to avoid negotiation altogether. A fourth principle concerns the level at which bargaining takes place. The Convention does not prescribe a single model. In some countries bargaining happens mainly at the level of the individual enterprise, while in others it is organised by sector or even at the national level through broad social pacts. The supervisory bodies have taken the position that the choice of level should be left to the parties and should not be dictated or restricted by the state. Systems that force bargaining down to the enterprise level, or that block sector-wide negotiation, can weaken the bargaining power of #workers, since a single small workplace often has little leverage on its own. The promotion duty in Article 4 therefore includes not obstructing the level of bargaining that #trade_unions and employers freely choose. A related question is the extension of #collective_agreements. In many systems, an agreement reached between representative parties in a sector can be extended by the authorities to cover all employers and workers in that sector, including those who were not signatories. The accompanying ILO recommendation encourages such extension where appropriate to national practice. Extension mechanisms are one of the main reasons why bargaining coverage in some countries remains high even where union membership is modest, because coverage is decoupled from individual membership. Where these mechanisms are dismantled, coverage tends to collapse toward the level of membership, which helps explain the sharp falls recorded in several states that suspended or restricted extension during periods of reform. The wider value of #collective_bargaining is not only individual but social. Comparative evidence gathered by international bodies indicates that strong bargaining institutions are associated with lower wage inequality, more stable industrial relations, and greater resilience during economic shocks. At the same time, the reach of bargaining has narrowed in many countries. Data compiled by the Organisation for Economic Co-operation and Development show that the share of employees covered by collective agreements across its member countries fell substantially over recent decades, and that trade union density declined over the same period in most of these economies. This decline does not lessen the legal obligation in Article 4; if anything, it sharpens the question of whether states are truly meeting their duty to promote bargaining, or merely tolerating its slow retreat. 8. The Supervisory and Enforcement Machinery A treaty is only as strong as the system that watches over it. The ILO has built one of the most developed #supervisory_systems in international law, and it is through this machinery that Convention No. 98 acquires practical force. The system has two broad arms: a regular arm that reviews reports from every ratifying state, and a set of complaint-based procedures that respond to specific allegations. 8.1 The regular supervisory system At the centre of the regular system is the Committee of Experts on the Application of Conventions and Recommendations. Established in 1926, it is composed of twenty independent jurists drawn from different regions, legal traditions, and cultures. Ratifying states are required to report periodically on how they are applying the conventions they have accepted, and the Committee of Experts examines these reports. It produces observations, which are published comments on particular countries, and direct requests, which raise more technical points. Although its conclusions are not binding judgments, they carry great weight because they represent an impartial technical assessment built up over many years and applied consistently across states. The observations of the Committee of Experts feed into the Conference Committee on the Application of Standards, a tripartite body that meets during the annual International Labour Conference. This committee selects a number of cases each year for public examination, questions government representatives, and adopts conclusions. The process is often described as the point where the technical work of the experts meets the political weight of the tripartite Conference. A government summoned to explain a serious failure to apply Convention No. 98 before an audience of its peers, together with employer and worker delegates from around the world, faces a form of reputational pressure that few other international systems can match. 8.2 The freedom of association procedures Alongside the regular system sits a special mechanism devoted to freedom of association. The Committee on Freedom of Association was created in 1951, soon after the adoption of Conventions Nos. 87 and 98, when the ILO concluded that these principles needed a dedicated procedure. The committee is a tripartite body of the Governing Body, with an independent chairperson and representatives of governments, employers, and workers. Its most striking feature is that it can examine complaints whether or not the state concerned has ratified the relevant conventions, because the underlying principle of freedom of association is treated as flowing from ILO membership itself. Over its history the committee has examined more than three thousand cases, and many governments have acted on its recommendations and reported changes in law or practice as a result. The complaints procedure of the Committee on Freedom of Association allows workers' and employers' organisations to bring allegations directly. This gives unions a route to Geneva that does not depend on their own government's willingness to raise a matter. When a union alleges, for example, the mass dismissal of organisers or the refusal of an employer or state to bargain, the committee examines the facts, sets out the applicable principles, and makes recommendations. Its accumulated decisions have been compiled and published, and this compilation functions as a practical guide to what freedom of association and #collective_bargaining require in concrete situations. 8.3 Representations and complaints under the Constitution Two further procedures arise directly from the ILO Constitution. Under Article 24, an industrial association of workers or employers may make a representation alleging that a state has failed to observe a convention it has ratified. A tripartite committee examines the representation and reports to the Governing Body. Under Article 26, a member state, a Conference delegate, or the Governing Body itself may file a complaint against a ratifying state, which can lead to the establishment of a Commission of Inquiry. The commission is the most solemn of the ILO's investigative mechanisms, capable of conducting a full examination and issuing recommendations. Although used sparingly, its existence signals that persistent and serious violations can attract the organisation's heaviest scrutiny. It is worth asking how findings by these bodies actually produce change, since none of them can hand down a binding, enforceable judgment in the way a national court can. The answer lies in a combination of pressures. A published observation that a state is failing to protect #workers against #anti_union_discrimination becomes part of a permanent record that other governments, trade partners, investors, and civil society can read. A case examined before the Conference Committee exposes the government to direct questioning in a public tripartite forum. Recommendations of the freedom of association machinery are followed up over successive sessions, so a government that ignores them faces repeated and escalating scrutiny. Over time, many states have amended their laws, reinstated dismissed unionists, or dropped restrictions on bargaining in response. The record of the Committee on Freedom of Association shows that a large number of countries have acted on its recommendations and reported concrete improvements. The limits of the system are equally real. Compliance ultimately depends on the state's own willingness to act. A determined government can absorb criticism, delay, or offer formal changes that are not implemented in practice. There is no international police force for #labour_rights, and the ILO cannot override national sovereignty. What the system offers instead is legitimacy, transparency, and persistence. It converts scattered complaints into a coherent and authoritative body of interpretation, and it keeps the spotlight on violations long after the initial headlines have faded. For a #supervisory_system operating among sovereign states without coercive power, this is a considerable achievement, even if it falls short of guaranteed enforcement. Taken together, these procedures form a layered system. Most compliance work happens quietly through the regular reporting cycle. Where problems are more acute, the freedom of association machinery and the constitutional procedures provide escalating responses. The system has no power to impose fines or to send anyone to prison. Its authority is persuasive and reputational, resting on publicity, peer pressure, technical credibility, and the slow accumulation of an interpretive record that governments find difficult to ignore. 9. National Implementation and Comparative Perspectives Because Convention No. 98 operates through domestic law, its meaning on the ground varies from country to country. Ratification is a promise to bring national law and practice into line with the Convention, but the distance between promise and performance can be wide. A few examples illustrate the range. Some states have ratified relatively recently and used the moment to reform their systems. Vietnam ratified Convention No. 98 in 2019, with the instrument taking effect in 2020, as part of broader labour reforms linked to its trade commitments. The ratification was accompanied by changes to the labour code intended to allow more independent worker representation and to strengthen bargaining, and it was supported by ILO technical cooperation. Whether the reforms deliver genuine independence in practice is a question that the supervisory bodies continue to follow, but the ratification marked a significant shift in a system that had long channelled representation through a single official structure. Mexico offers a related example. Its ratification of Convention No. 98 in 2018 was tied to a reckoning with so-called protection contracts, agreements negotiated between employers and unions that workers had no real part in choosing and that served to pre-empt genuine bargaining. Ratification brought the Mexican system under the full scrutiny of the ILO's supervisory machinery and reinforced domestic reforms aimed at ensuring that workers can vote on their contracts and choose their representatives freely. The episode shows how ratification can be used as a lever to dismantle structures that undermine authentic #collective_bargaining. Canada ratified the Convention in 2017, comparatively late for an industrialised democracy, reflecting the fact that formal ratification sometimes lags behind substantive protection. Many states protect union rights through constitutional guarantees, statutes, and judicial decisions well before they formally ratify, and the act of ratification then confirms and locks in protections that already exist. The most discussed non-ratifying state is the United States. It has not ratified Convention No. 98, and its position is often explained by reference to its federal structure, in which labour matters are divided between national and state authority, and by the practice of ratifying only those conventions that already conform to existing domestic law. Critics, including workers' organisations and human rights monitors, have long argued that United States labour law falls short of the Convention's standard in several respects, pointing to weak remedies for anti-union dismissal, the permanent replacement of striking workers, and the exclusion of significant categories of workers from bargaining rights. The debate illustrates that non-ratification does not place a state entirely outside the reach of the principle, since the 1998 Declaration binds all members to respect freedom of association and #collective_bargaining regardless of ratification, and the freedom of association machinery can examine complaints against non-ratifying states. A further dimension of national implementation is the growing link between #labour_rights and trade. In recent decades many trade agreements have included labour chapters that commit the parties to respect fundamental principles, including freedom of association and the effective recognition of #collective_bargaining. The trade dimension helps explain several ratifications. Vietnam's decision to ratify Convention No. 98 was closely connected to its trade relationship with the European Union, whose agreement encouraged the adoption of ILO fundamental conventions. Mexico's labour reforms were reinforced by commitments in its regional trade arrangement, which included mechanisms allowing complaints about the denial of bargaining rights at particular workplaces. These developments show how the standards of Convention No. 98 can be given additional force through economic leverage that the ILO's own machinery lacks. The trade link is not without difficulty. Critics point out that labour clauses are unevenly enforced, that they may be used selectively for competitive or political reasons, and that they can place the burden of compliance on exporting states while doing little for #workers in the informal economy who produce nothing for export. There is also a risk that trade-based enforcement treats labour rights as a bargaining chip rather than as fundamental rights that deserve protection in their own right. Even so, the incorporation of the Convention's principles into trade instruments has given them reach into national systems that might otherwise have resisted change, and it reflects a wider recognition that the suppression of bargaining can distort competition by holding wages and conditions artificially low. Across these examples a common pattern appears. Ratification is necessary but not sufficient. What matters is whether national law provides real protection against #anti_union_discrimination, whether remedies deter, whether the state actively promotes bargaining, and whether independent unions can function without capture by employers or governments. The supervisory bodies devote much of their work to precisely these questions, and their country observations record both progress and persistent failures. 10. Contemporary Challenges Convention No. 98 was drafted for a world of factories, established employers, and stable full-time jobs. That world has not disappeared, but it has been joined by new forms of work and new pressures that test the Convention's reach. Several challenges stand out. The first is the rise of #platform_work and the gig economy. Digital platforms often classify the people who work through them as independent contractors rather than employees. This classification, whatever its merits in other respects, can place workers outside the protective scope of labour law, including the right to organise and bargain. If a rider or driver is not an employee, national law may deny them the machinery of #collective_bargaining altogether, even though their economic position closely resembles that of an employee. Recent scholarship has examined how platform workers have nonetheless mobilised and pressed for recognition, drawing on a human rights framing of freedom of association to argue that the classification of a worker should not be allowed to extinguish a fundamental right. The tension between formal legal categories and the substance of the working relationship is one of the sharpest issues now facing the Convention. The second challenge is the informal economy. In large parts of the world, most workers are outside the formal system entirely, without written contracts, registration, or the institutions through which bargaining ordinarily happens. Street vendors, domestic workers, and casual labourers may have a formal right to organise but no practical channel to exercise it. Recent work in labour law has explored how the guarantees of #collective_bargaining might be extended to these workers, including through arrangements that do not depend on a conventional employer-employee relationship. The Convention's language is broad enough to accommodate such extensions, but doing so requires imagination and political will that many states have not shown. The third challenge is the long-term decline of #trade_unions in many industrialised countries. As union density falls and bargaining coverage shrinks, the machinery that Article 4 requires states to promote weakens from within. A right to bargain is of limited value if few workers belong to organisations capable of bargaining. The data assembled by international bodies show a marked drop in both membership and coverage across many advanced economies over recent decades. This decline is driven by many factors, including changes in industrial structure, the growth of small and dispersed workplaces, hostile legal environments in some countries, and the individualisation of employment relations. Whatever its causes, it raises a pointed question about whether states are meeting their promotional duty under the Convention or merely watching bargaining institutions erode. The fourth challenge is the global supply chain. Much production now takes place through networks of suppliers spread across many countries, and the workers who make goods for a global brand are often employed by distant subcontractors rather than by the brand itself. Anti-union #discrimination and the suppression of bargaining in supplier factories are widely documented, yet the legal responsibility of the lead firm is often unclear. Recent research on mandatory human rights due diligence examines how law might require companies to identify and address violations of freedom of association in their supply chains, moving beyond voluntary codes of conduct. This is an area where the principle of Convention No. 98 is increasingly invoked but where enforcement remains difficult, because the worker, the immediate employer, and the firm with real economic power may sit in three different jurisdictions. A fifth challenge concerns migrant workers, who in some regions are excluded from union membership by law or in practice, and whose dependent immigration status can make organising perilous. Monitoring bodies have found that migrant workers frequently face both legal and practical barriers to exercising their rights to freedom of association and #collective_bargaining. Systems that tie a worker's legal presence to a single employer are especially corrosive, because the threat of losing that sponsorship can silence any attempt to organise. Finally, there is the long-running debate over the right to strike. The Convention does not mention strikes expressly, yet the supervisory bodies have long treated the right to strike as an essential corollary of freedom of association and #collective_bargaining, since bargaining without any capacity for collective action tends to be bargaining on unequal terms. This interpretation has been contested within the ILO, particularly by the employers' group, and the disagreement has become one of the most significant institutional debates in recent years. Recent scholarship has revisited the international legal basis of the right to strike and its relationship to the freedom of association conventions, and the question of how far collective action is protected remains a live and important one for the future application of Convention No. 98. 11. Convention No. 98 and the Human Rights Framework One of the most important developments in the life of Convention No. 98 has been its absorption into the broader body of international human rights law. When the Convention was adopted in 1949, labour standards and human rights were often treated as separate fields. Over the following decades that separation broke down, and freedom of association and #collective_bargaining came to be understood as human rights in their own right. The International Covenant on Civil and Political Rights, adopted in 1966, protects freedom of association, including the right to form and join trade unions. The International Covenant on Economic, Social and Cultural Rights of the same year protects the right to form and join unions and the right to strike, exercised in conformity with national law. Regional human rights systems have gone further still, with courts interpreting general guarantees of freedom of association to include a right to bargain collectively. Through these instruments and decisions, the principle expressed in Convention No. 98 has been woven into the fabric of human rights law, so that a violation of the right to organise is now widely understood as a violation of human rights and not merely of a technical labour standard. This convergence matters for two reasons. First, it strengthens the normative claim of the Convention. A right recognised across multiple treaties and by multiple bodies is harder to dismiss and easier to invoke. Second, it multiplies the forums in which the right can be asserted. A worker or union denied the right to bargain may be able to rely not only on the ILO machinery but also on human rights bodies, regional courts, and national constitutional guarantees that reflect the same principle. The 1998 Declaration reinforced this integration within the ILO by treating freedom of association and the effective recognition of #collective_bargaining as obligations flowing from membership itself. The overall effect is that Convention No. 98 sits at the centre of a wider web of protection rather than standing alone as a single treaty. 12. Discussion Bringing these threads together, several observations can be made about the significance and the limits of Convention No. 98. The first is that the Convention's authority rests on a combination of near-universal ratification, integration with human rights law, and a sophisticated supervisory system. These features give it a legitimacy that few international instruments enjoy. When 168 states have accepted a standard, when that standard is echoed across the major human rights treaties, and when an independent expert body has interpreted it consistently for decades, the resulting norm is genuinely authoritative. This is why even a non-ratifying state cannot easily claim to be outside the principle. The second observation is that the Convention's real effect depends on national enforcement, and here the record is mixed. The gap between formal ratification and practical protection is the central problem of the whole system. A state may ratify and yet tolerate the routine dismissal of organisers, provide remedies too weak to deter, allow employers to dominate compliant unions, or decline to promote bargaining in any meaningful way. The supervisory bodies can expose these failures and press for change, and they often succeed, but they cannot compel a reluctant government. The system relies on persuasion, publicity, and the slow force of accumulated interpretation. This is both its weakness and, in a world of sovereign states, its realistic strength. The third observation concerns remedies. Much of the practical value of Article 1 turns on whether the sanctions for anti-union #discrimination are dissuasive. The insistence of the ILO bodies on reinstatement and on compensation set high enough to remove any incentive to repeat the conduct is not a technicality. It is the difference between a right that protects and a right that merely records disapproval. National systems that treat unlawful anti-union dismissal as a minor matter, remediable by a small payment, hollow out the Convention regardless of how firmly their statutes prohibit the conduct. The fourth observation is that the Convention faces a structural challenge from the changing shape of work. The instrument assumes a recognisable employer, an identifiable workplace, and a stable relationship within which bargaining can occur. Where work is mediated by platforms, fragmented across supply chains, pushed into the informal economy, or performed by workers whose legal status is precarious, these assumptions weaken. The principle of the Convention remains valid, but its application requires interpretation and, in some cases, new instruments or new legal techniques such as due diligence obligations on lead firms. The scholarship of the last few years has been preoccupied with exactly this task of adapting classic guarantees to new realities, and its conclusions suggest that the future relevance of Convention No. 98 depends on the willingness of states and courts to look at the substance of a working relationship rather than its formal label. The fifth and final observation is that the Convention should be understood as a floor rather than a ceiling. It sets a minimum below which no ratifying state should fall, but it does not prevent states from providing stronger protection, wider coverage, or more generous machinery. The best systems go well beyond the minimum. The danger lies in treating the floor as an aspiration rather than an obligation, and in allowing bargaining institutions to decay while pointing to formal ratification as evidence of compliance. 13. Conclusion The Right to Organise and Collective Bargaining Convention, 1949, is a short instrument that carries a large responsibility. In six substantive articles it protects #workers against #anti_union_discrimination, shields their organisations from capture by employers, requires machinery to enforce these guarantees, and obliges states to promote free and voluntary #collective_bargaining. Adopted alongside its companion Convention No. 87, elevated to fundamental status by the 1998 Declaration, and woven into the wider human rights framework, it has become one of the most widely accepted and authoritative standards in international law, binding 168 states. Yet its value cannot be read off the ratification tables alone. The Convention lives or dies in national practice, in whether a dismissed organiser is reinstated, whether a company union is prohibited, whether a genuine union can reach a binding agreement, and whether remedies bite hard enough to deter. The ILO's supervisory machinery, from the Committee of Experts to the Committee on Freedom of Association, has given the text meaning and has held governments to account for more than seventy years, but its power is persuasive rather than coercive. The pressures of platform work, the informal economy, declining union density, global supply chains, and precarious migration test the Convention as never before. The sensible conclusion is neither triumphant nor despairing. Convention No. 98 has proven durable and adaptable, and its core principle, that people should be able to join together and bargain over the terms of their working lives without fear of reprisal, has lost none of its force. The task now is to make that principle real in forms of work its drafters never anticipated, and to ensure that the promise of protection is matched by the reality of enforcement. For students of #labour_rights, the Convention is both a settled foundation and an unfinished project, a standard whose text has held firm while the world of work around it continues to change. References Primary legal instruments and authoritative institutional sources International Labour Organization. (1948). Freedom of Association and Protection of the Right to Organise Convention (No. 87). Geneva: International Labour Office. International Labour Organization. (1949). Right to Organise and Collective Bargaining Convention (No. 98). Geneva: International Labour Office. International Labour Organization. (1998, amended 2022). ILO Declaration on Fundamental Principles and Rights at Work. Geneva: International Labour Office. International Labour Organization. (2018). Freedom of Association: Compilation of decisions of the Committee on Freedom of Association (6th ed.). Geneva: International Labour Office. International Labour Organization. (2022). Social Dialogue Report 2022: Collective bargaining for an inclusive, sustainable and resilient recovery. Geneva: International Labour Office. Secondary scholarly literature International Trade Union Confederation. (2025). ITUC Global Rights Index 2025. Brussels: International Trade Union Confederation. Lopez Lopez, J. (Ed.). (2022). Inscribing Solidarity: Debates in Labour Law and Beyond. Cambridge: Cambridge University Press. Marshall, S., Landau, I., Shamir, H., Barkay, T., Fudge, J., & van Heerden, A. (2023). Mandatory human rights due diligence: Risks and opportunities for workers and unions. Research Paper, RMIT University and Monash University. Organisation for Economic Co-operation and Development. (2025). Membership of unions and employers' organisations, and bargaining coverage. Paris: OECD Publishing. https://doi.org/10.1787/fe47107c-en Simutina, Y. (2024). The impact of war and European accession on the labour law reform in Ukraine. European Studies: The Review of European Law, Economics and Politics. Stylogiannis, C. (2023). Freedom of association and collective bargaining in the platform economy: A human rights-based approach and an ever-increasing mobilization of workers. International Labour Review. https://doi.org/10.1111/ilr.12340 Vogt, J., Novitz, T., & Fudge, J. (Eds.). (2026). The Right to Strike Reimagined. Oxford: Hart Publishing. Vogt, J., & Subasinghe, R. (2025). Turning up the heat: The right to strike and the climate crisis. Comparative Labor Law & Policy Journal. Von Broembsen, M. (2025). Collective bargaining for the "new" working class: Putting "personal work relations" to work for street vendors. Industrial Law Journal. #ILO_Convention_98 #RightToOrganise #CollectiveBargaining #AntiUnionDiscrimination #FreedomOfAssociation #WorkersRights #LabourLaw #TradeUnionRights #InternationalLabourStandards #DecentWork #SocialDialogue #C98 #LabourRights #IndustrialRelations #Convention87

The Freedom of Association and Protection of the Right to Organise Convention, 1948 (No. 87) is one of the oldest and most widely ratified instruments of the International Labour Organization. It protects the right of workers and of employers, without distinction, to set up and join organizations of their own choosing, to run those organizations free from state interference, and to build federations and international alliances. This article explains what the Convention says, why it was written, and how it works in practice. It traces the historical roots of the instrument, sets out its provisions article by article in plain language, and locates it within the wider system of international labour standards, including its close partner, the Right to Organise and Collective Bargaining Convention, 1949 (No. 98), and the 1998 Declaration on Fundamental Principles and Rights at Work as amended in 2022. The article then describes the supervisory machinery that gives the Convention life, including the Committee of Experts, the Conference Committee on the Application of Standards, and the Committee on Freedom of Association. Particular attention is given to the long dispute over whether the Convention protects the right to strike, a dispute that paralyzed part of the supervisory system after 2012 and was finally resolved by the advisory opinion of the International Court of Justice delivered on 21 May 2026. The study closes by examining contemporary pressures on freedom of association, including platform work, the informal economy, migrant labour, and shrinking civic space, and it argues that the value of Convention No. 87 lies not only in its text but in the institutions and interpretive practice built around it. The article is written for students of labour law, international relations, and human rights. Keywords: freedom of association; International Labour Organization; Convention No. 87; right to organise; right to strike; collective bargaining; workers rights; supervisory system. 1. Introduction Few ideas in the modern world of work are as basic as the freedom to join together with others. When workers form a #trade_union, or when employers create a business association, they are doing something that most people take for granted in daily life: acting as a group rather than as isolated individuals. Yet for much of history this simple act was treated as a crime. Workers who tried to combine were prosecuted, imprisoned, or dismissed. The recognition that combining is a right, and not an offence, was won slowly and at great cost. The #Freedom_of_Association and Protection of the Right to Organise Convention, 1948 (No. 87), is the international treaty that captures this recognition and turns it into a binding legal standard. Adopted by the #International_Labour_Organization in 1948, it is a short document, but its influence is very large. It is one of the ten instruments that the ILO now treats as fundamental, and as of 2026 it has been ratified by roughly 158 member States, which makes it one of the most accepted labour treaties ever written. Alongside its sister instrument, the Right to Organise and Collective Bargaining Convention, 1949 (No. 98), it forms the legal backbone of what the ILO calls #freedom_of_association. This article has three aims. The first is descriptive. It sets out clearly what #Convention_No_87 actually protects, so that a reader with no legal background can understand the rights at stake and the duties that States take on when they ratify. The second aim is institutional. A treaty on paper means little without machinery to apply it, and the ILO has built one of the most developed monitoring systems in international law. Understanding that machinery is the only way to understand how the Convention operates in real disputes. The third aim is critical. The Convention has been the subject of one of the sharpest interpretive battles in the history of international labour law, the fight over whether it protects the #right_to_strike. That battle ran for more than a decade, damaged the credibility of the ILO supervisory bodies, and was only settled in 2026 by the International Court of Justice. Studying that episode teaches something important about how legal texts acquire meaning over time. The structure of the article follows these aims. Section 2 describes the historical background and the drafting of the Convention. Section 3 explains the provisions in ordinary language. Section 4 places the Convention inside the wider family of ILO standards. Section 5 describes the supervisory bodies. Section 6 examines the right to strike controversy and the 2026 advisory opinion of the #ICJ. Section 7 turns to modern challenges such as the #platform_economy and the informal sector. Section 8 looks at regional and national variation and at the gap between ratification and reality. Section 9 sets out the main criticisms and debates. Section 10 offers a discussion, and Section 11 concludes. A note on method is useful here. This is a doctrinal and analytical study rather than an empirical one. It relies on the text of the Convention, on the practice of the ILO supervisory bodies as reported in their published decisions, on the recent advisory opinion of the International Court of Justice, and on secondary scholarship in labour law and international law. The goal is to synthesize existing knowledge in a form that is accurate and accessible, not to present new field data. 2. Historical Background and the Origins of Convention No. 87 To understand why Convention No. 87 exists, it helps to remember why the ILO exists. The Organization was created in 1919, in the aftermath of the First World War, as part of the peace settlement. Its founders believed that lasting peace could not be built on social injustice, and that poor working conditions in one country would place unfair pressure on others. The tripartite structure of the ILO, in which governments, employers, and workers all sit at the table, was itself a recognition that #workers_organizations and #employers_organizations are legitimate actors in shaping the rules of work. The very Constitution of the ILO named freedom of association as one of the principles capable of improving the conditions of labour. In other words, the freedom to combine was written into the foundations of the Organization from the start. What was missing in the early decades was a specific, detailed treaty that spelled out the content of that freedom and made it enforceable through ratification. The road to such a treaty was shaped by hard experience. In the nineteenth century, many States had laws that made worker combinations illegal. In England, for example, the Combination Acts kept combinations unlawful for a quarter of a century, and workers had to meet in secret to discuss their common interests. Across Europe and beyond, the twentieth century brought waves of both progress and repression. The rise of authoritarian regimes in the years between the two World Wars showed how quickly independent unions could be crushed, absorbed into the State, or replaced by official bodies that answered to the government rather than to their members. This lesson was central. A freedom of association standard would have to protect not just the existence of organizations, but their independence from the State. The ILO had not been silent on the question before 1948. As early as 1921 it adopted the Right of Association (Agriculture) Convention, No. 11, which recognized the right of agricultural workers to associate on the same terms as workers in industry. That early instrument was narrow, since it dealt with a single sector and did not spell out the full content of the freedom, but it showed that the Organization understood association as a matter of principle from its first years. Through the 1920s and 1930s, however, repeated efforts to adopt a general freedom of association Convention failed, partly because governments were reluctant to accept binding limits on their power over unions, and partly because the political turmoil of the period made agreement difficult. The experience of the 1930s, when independent unions were destroyed in several countries, made the need for a strong general standard impossible to ignore. After the Second World War the moment for a strong instrument arrived. The 1944 Declaration of Philadelphia reaffirmed that freedom of expression and of association were essential to sustained progress. In this climate the ILO turned to drafting. The Convention was discussed and adopted at the 31st session of the International Labour Conference, held in San Francisco, and it was adopted on 9 July 1948. It came into force on 4 July 1950 after the required number of ratifications. The timing matters. The Convention was born in the same period as the Universal Declaration of Human Rights, and it shares that document's conviction that certain rights belong to people simply because they are human. Freedom of association sits at the meeting point of labour rights and civil liberties, because a person cannot form a union freely if they can be jailed for speaking, meeting, or marching. Two features of the drafting deserve emphasis because they explain later debates. First, the Convention was deliberately written to protect both workers and employers. It is not a union charter alone. Employers also have the protected right to organize and to run their associations without interference. Second, the drafters chose to state broad principles rather than to list every detail. They wrote of the right to draw up rules, to elect representatives, and to organize activities, but they did not attach a long code of specific permitted and forbidden acts. This choice made the Convention flexible and durable, but it also left room for disagreement about what the principles include. The most famous of those disagreements, over the right to strike, is discussed in Section 6. 3. The Legal Architecture: Provisions of Convention No. 87 in Plain Language Convention No. 87 contains twenty one articles. Some set out substantive rights, and the rest deal with technical matters such as ratification and territorial application. This section walks through the substantive core so that the reader can see exactly what is protected. The language below is a paraphrase written for clarity, not a substitute for the official text. 3.1 The right to establish and join organizations (Article 2) Article 2 is the heart of the treaty. It provides that workers and employers, without distinction whatsoever, have the right to establish and to join organizations of their own choosing, subject only to the rules of the organization concerned, and to do so without previous authorization. Several words in this sentence carry great weight. The phrase "without distinction whatsoever" is expansive. It means that the right is not limited by race, sex, nationality, political opinion, occupation, or similar grounds. The right belongs to all workers and all employers. There is only one narrow group that States may treat differently, the armed forces and the police, and even that exception is limited, as explained below. The words "of their own choosing" protect a genuine freedom to decide which organization to belong to, or to create a new one. This is a defence against so called trade union monopoly imposed by law, where the State forces all workers into a single official union. It is also a defence against employer controlled organizations that only pretend to be independent. The phrase "without previous authorization" means that a government cannot require groups to obtain permission before they may exist. Registration formalities are allowed, but they must not become a hidden system of political approval in which the authorities can refuse to register a union they dislike. #Ratification of the Convention therefore commits a State to letting organizations form on their own initiative. 3.2 The right to function freely (Article 3) Article 3 protects the internal life of organizations. It provides that they have the right to draw up their own constitutions and rules, to elect their representatives in full freedom, to organize their administration and activities, and to formulate their programmes. Public authorities are required to refrain from any interference that would restrict this right or impede its lawful exercise. This article is where much of the day to day work of the supervisory bodies takes place. Governments have tried many indirect ways to control unions: writing the union's rules for it, disqualifying certain candidates from union office, dictating how often meetings may be held, or freezing a union's funds. Article 3 treats all of these as interference. The right to elect leaders "in full freedom" is especially important, because a union whose leaders are chosen or vetted by the government is not really independent. The right to "organize activities" is also the textual home of the argument about the right to strike, because a strike is a form of collective activity, a point taken up in Section 6. 3.3 Protection against dissolution or suspension by administrative authority (Article 4) Article 4 provides that organizations of workers and of employers shall not be liable to be dissolved or suspended by administrative authority. This is a strong safeguard. It does not say that a union can never be dissolved. It says that dissolution or suspension cannot be ordered by a minister or official acting alone. If a State wishes to dissolve an organization, it must do so through proper judicial process, with the safeguards that courts provide. The reason is obvious from history. A government that can shut down a union by decree can silence every union that criticizes it. 3.4 The right to federate and to affiliate internationally (Articles 5 and 6) Article 5 gives organizations the right to establish and join federations and confederations, and to affiliate with international organizations of workers and employers. Article 6 confirms that the guarantees in Articles 2, 3, and 4 apply equally to these higher level bodies. The practical meaning is that #solidarity is protected across enterprises, across sectors, and across borders. A local union may join a national federation, and that federation may join a global body. This international dimension is one reason the Convention matters so much to the wider #labour_movement. 3.5 Legal personality and the rule of law (Articles 7 and 8) Article 7 provides that the acquisition of legal personality by organizations, federations, and confederations shall not be made subject to conditions that would restrict the rights in Articles 2, 3, and 4. In plain terms, the formalities of gaining legal status must not be used to strip away the substance of the freedom. Article 8 addresses the relationship between organizations and the law. It has two parts. First, workers, employers, and their organizations, like everyone else, must respect the law of the land. Second, and crucially, the law of the land must not be framed or applied in a way that impairs the guarantees of the Convention. This is a careful balance. Unions are not above the law, but the law cannot be turned into a tool for defeating the very rights the Convention protects. 3.6 The armed forces and police exception (Article 9) Article 9 allows national law to determine the extent to which the guarantees of the Convention apply to the armed forces and the police. This is the single permitted carve out, and it is narrow. It covers only these two groups, and it is discretionary, meaning that a State may choose to extend full rights to them. The supervisory bodies read this exception strictly, so that governments cannot smuggle other categories of workers, such as civil servants generally, into the exception. 3.7 Definitions and general obligations (Articles 10 and 11) Article 10 defines the term organization broadly, as any organization of workers or of employers for furthering and defending their interests. Article 11, which opens Part II of the Convention, states the general duty: each member for which the Convention is in force undertakes to take all necessary and appropriate measures to ensure that workers and employers may exercise freely the right to organize. This is the positive obligation. The State does not merely promise to leave organizations alone; it promises to create the conditions in which the freedom can actually be used. 3.8 Final provisions (Articles 12 to 21) The remaining articles are technical. They deal with the application of the Convention to non metropolitan territories, the process of ratification and registration, entry into force, the conditions and timing of denunciation, and the ordinary machinery of a multilateral treaty. They are important for lawyers but they do not change the substance of the rights described above. Taken together, these provisions describe a coherent scheme. Workers and employers may form and join organizations freely (Articles 2, 5, 7), run them independently (Articles 3, 6), and rely on protection against arbitrary destruction (Article 4), all within a framework where they respect the law but the law respects their rights (Article 8), subject only to a narrow exception (Article 9), and backed by a positive State duty (Article 11). 4. Convention No. 87 within the ILO Normative Framework Convention No. 87 does not stand alone. It is the first pillar of a two pillar structure, and it is embedded in a much larger body of standards and declarations. The second pillar is the Right to Organise and Collective Bargaining Convention, 1949 (No. 98). If Convention No. 87 protects the freedom to form and run organizations, Convention No. 98 protects those organizations and their members once they exist and try to act. It guards workers against #anti_union_discrimination in employment, such as being dismissed or passed over because of union membership or activity. It protects unions and employer organizations against interference by one another. And it promotes voluntary #collective_bargaining, the process by which unions and employers negotiate the terms of work. The two Conventions are usually read as a pair, and States are encouraged to ratify both. Convention No. 98 is even more widely ratified than Convention No. 87. Both instruments sit inside the ILO Declaration on Fundamental Principles and Rights at Work. First adopted in 1998, this Declaration states that all member States, by the very fact of belonging to the ILO, have an obligation to respect, to promote, and to realize the principles behind the fundamental Conventions, whether or not they have ratified them. Freedom of association and the effective recognition of the right to collective bargaining form the first of these categories. In June 2022 the International Labour Conference amended the Declaration to add a fifth category, a safe and healthy working environment, so that the fundamental principles now number five (2022 Amendments, 2023). All 187 member States are expected to observe these principles regardless of ratification, and this is why the ILO can examine freedom of association complaints even against countries that have never ratified Convention No. 87. This point about universal reach is worth dwelling on, because it makes freedom of association different from most treaty rights. Normally a treaty binds only the States that have agreed to it. Freedom of association is different, because the ILO treats it as flowing directly from membership in the Organization and from the ILO Constitution itself. A State cannot escape scrutiny simply by refusing to ratify. This constitutional character gives the principle unusual strength. The Convention is also linked to the broader human rights system. Freedom of association appears in the Universal Declaration of Human Rights, in the International Covenant on Civil and Political Rights, and in the International Covenant on Economic, Social and Cultural Rights, as well as in regional instruments such as the European Convention on Human Rights and the American Convention on Human Rights. The ILO standards and the human rights treaties reinforce one another, and courts in many regions cite the ILO supervisory bodies when they interpret the association guarantees in general human rights law. In this sense Convention No. 87 is both a labour instrument and a #human_rights_at_work instrument, and it belongs to two legal worlds at once. 5. The Supervisory and Complaints Machinery A treaty is only as strong as the system that applies it. The ILO has developed one of the most detailed supervisory systems in international law, and freedom of association benefits from a special complaints procedure that no other subject enjoys. Four bodies deserve description. 5.1 The Committee of Experts on the Application of Conventions and Recommendations The Committee of Experts, often shortened to the #Committee_of_Experts or CEACR, is a group of independent legal specialists, not government representatives. Each year it reviews reports that States are required to submit on the Conventions they have ratified, along with comments from workers and employers organizations. When it finds that national law or practice does not match a Convention, it publishes an observation or a direct request. These comments are technical and reasoned, and over many decades they have built up a body of interpretation that guides governments on what compliance actually requires. The Committee of Experts is the main engine of the regular #supervisory_system, and its long standing view on the right to strike was at the centre of the controversy discussed below. 5.2 The Conference Committee on the Application of Standards The Conference Committee on the Application of Standards is a tripartite body that meets each year during the International Labour Conference. Unlike the Committee of Experts, it is made up of government, employer, and worker delegates. It selects a number of individual cases for public examination, questions the governments concerned, and adopts conclusions. Because it is tripartite and public, it adds political weight and visibility to the technical findings of the experts. The interaction between these two committees, one expert and one tripartite, is a defining feature of ILO supervision, and it was the friction between them that produced the 2012 crisis. 5.3 The Committee on Freedom of Association The #Committee_on_Freedom_of_Association, or CFA, is the special body for this subject. It was set up by the Governing Body in November 1951, and it is tripartite. Its unique feature is that it can examine complaints of violations of freedom of association even where the State concerned has not ratified Convention No. 87 or No. 98, because the principle flows from the Constitution. Workers organizations, employers organizations, and governments may bring complaints. The Committee does not act as a criminal court. Its aim, as it often explains, is not to blame governments but to promote respect for freedom of association through dialogue and specific recommendations. Over more than seven decades it has examined thousands of cases and has developed a rich and detailed body of principles, described in a celebratory study published for its seventieth anniversary (Curtis and Wolfson, 2022). One recurring theme in that work is the link the Committee has always drawn between freedom of association and basic civil liberties such as freedom of expression, assembly, and protection against arbitrary arrest. 5.4 Commissions of inquiry and Article 26 complaints For the most serious and persistent violations, the ILO Constitution provides a heavier procedure. Under Article 26, a complaint may lead to the appointment of a Commission of Inquiry, the Organization's highest level investigative body. Commissions of inquiry are rare and are reserved for grave situations. They produce detailed findings and recommendations, and a State that refuses to comply may in principle face further measures. This procedure signals that freedom of association is not treated as a minor administrative matter but as a fundamental commitment. Together these bodies form a layered system. Routine problems are handled through the Committee of Experts and the Conference Committee. Specific freedom of association complaints go to the CFA. The gravest cases can reach a Commission of Inquiry. This layered design allows the system to respond in proportion to the seriousness of the violation, and it explains why the ILO has been able to say so much, in such detail, about the meaning of freedom of association. As one commentator has observed, the meaning of this right has been defined more fully by the ILO machinery than almost any other labour right. 6. The Right to Strike Controversy and the 2026 Advisory Opinion No account of Convention No. 87 would be complete without the story of the right to strike, because it is the clearest example of how a short treaty text can generate a long and bitter dispute, and of how that dispute can finally be resolved. 6.1 The interpretive question The text of Convention No. 87 does not mention strikes. It protects the right of organizations to organize their activities and to formulate their programmes (Article 3), and it protects the freedom to associate more generally. For many years the Committee of Experts and the Committee on Freedom of Association took the view that the #right_to_strike is a necessary corollary of freedom of association. Their reasoning was straightforward. If workers may form unions and bargain collectively, but may never withhold their labour, then their bargaining power is hollow, because the employer has no reason to reach agreement. On this view, the right to strike is implied by the right to organize, even though it is not spelled out. For decades this interpretation was broadly accepted in practice, and the supervisory bodies built up a detailed set of principles about it: which workers may be restricted, what limits are permissible in essential services, when notice may be required, and so on. 6.2 The 2012 crisis The consensus broke in 2012. At the International Labour Conference that year, the Employers group challenged the whole approach. They argued that the Committee of Experts had exceeded its mandate by reading a right to strike into a Convention that says nothing about strikes. On their view, the experts were legislating rather than interpreting, and States had never agreed to be bound on this point. The Workers group and many governments disagreed, insisting that the corollary reading was correct and long established. The consequences were serious. The Conference Committee on the Application of Standards was unable to agree on its conclusions, and for a period the regular supervisory system was thrown into difficulty. A disagreement about interpretation had become a threat to the functioning of the institution itself. The dispute simmered for years, with various attempts at compromise, and it exposed a deeper question about who has the authority to say what an ILO Convention means. The scholarly response was significant. A detailed study by a group of leading labour lawyers examined the sources of international law and concluded that the right to strike is in fact protected, and that there was no credible basis for the claim that it enjoys no international protection (Vogt et al., 2020). But scholarship, however careful, could not by itself settle a dispute among the ILO's own constituents. 6.3 The referral to the International Court of Justice The ILO Constitution provides a way out of exactly this kind of deadlock. Article 37, paragraph 1, allows questions about the interpretation of a Convention to be referred to the International Court of Justice. This mechanism is almost never used, and referring the strike question to the Court would be only the second such referral in the Organization's history. At a special session of the Governing Body on 10 November 2023, following a request from the Workers group supported by a large number of governments, the ILO decided by vote to ask the Court to give an urgent advisory opinion on a single question: is the right to strike of workers and their organizations protected under Convention No. 87? The vote was 33 in favour, 21 against, and 2 abstentions, which shows how divided the constituents remained. The request was transmitted to the Court, which opened proceedings as case number 191 on its General List and invited written and oral submissions. A striking number of States and organizations took part, including many governments, several international organizations, and both the workers and employers international bodies, and public hearings were held in October 2025. 6.4 The opinion of 21 May 2026 On 21 May 2026 the International Court of Justice delivered its advisory opinion. By ten votes to four, the Court held that the right to strike of workers and their organizations is protected under Convention No. 87 (ICJ, 2026). The Court acknowledged that the Convention contains no explicit reference to strikes, but it reasoned that the absence of an express provision does not necessarily mean that the matter is excluded from the treaty. Interpreting the Convention in good faith, in accordance with the ordinary meaning of its terms and in light of its object and purpose, the Court found that the protection of the right to strike follows from the freedom of workers and their organizations to organize their activities in defence of their interests. Two features of the opinion are important for students. First, it is an authoritative interpretation. Advisory opinions are not binding judgments in the way that a decision between two States would be, but under the ILO Constitution the Court is the body competent to give authoritative interpretations of ILO Conventions, and its conclusion carries very great legal and political weight for the roughly 158 States bound by Convention No. 87. Second, the Court was deliberately careful. It answered the narrow question put to it, whether the right is protected, but it declined to define the exact scope, content, or permissible limits of the right. In other words, it confirmed that the right exists under the Convention, while leaving the detailed rules about how far it extends to be worked out through the ordinary processes of the Organization. The ILO Governing Body was expected to consider the opinion and any follow up at its session in November 2026. 6.5 Why this episode matters The strike controversy is not just a technical footnote. It teaches several lessons that reach beyond labour law. It shows that legal texts do not interpret themselves, and that the meaning of a treaty is settled through institutions and practice over time. It shows that a dispute about interpretation can become a dispute about legitimacy, threatening the authority of the very bodies charged with applying the law. And it shows the value of having an ultimate arbiter. For more than a decade the ILO could not resolve the question on its own, precisely because its tripartite structure gave employers, workers, and governments each a veto over consensus. The referral to a court broke the deadlock, at the cost of taking the question outside the tripartite family. Scholars will debate for years whether this was a triumph of the rule of law or a sign that #tripartism had reached its limits, and both readings contain truth. 6.6 What the opinion left open: limits and essential services Because the Court confirmed the right but did not define its boundaries, it is worth understanding the framework that the supervisory bodies had already developed over many years, since that framework is likely to remain the reference point for the detailed rules. The Committee of Experts and the Committee on Freedom of Association never treated the right to strike as unlimited. They accepted that reasonable conditions may be attached, such as a requirement to give prior notice, to hold a ballot, or to exhaust conciliation before action begins, provided these conditions do not become so heavy that they make lawful action almost impossible in practice. They also recognized categories where the right may be restricted or even removed. The clearest is the essential services category, understood in the strict sense of services whose interruption would endanger the life, personal safety, or health of the population, such as hospitals, water supply, or air traffic control. Where a genuine essential service is at stake, a State may limit strike action, but the supervisory bodies insisted on compensation for the affected workers, for example through fair and rapid arbitration, so that people who lose the right to strike are not simply left without a remedy. A further recognized limit applies to public servants exercising authority in the name of the State, as distinct from ordinary public sector employees who do provide essential services. These distinctions are technical, but they show that the debate was never about whether limits are allowed. It was about the source of the right and about who has authority to define it. The 2026 opinion answered the first question, and the ordinary work of the Organization will now continue on the second. 7. Contemporary Challenges to Freedom of Association The world of work in the 2020s looks very different from the world of 1948. Convention No. 87 was written for a labour market dominated by stable, full time, factory based employment. Many of the hardest questions today arise precisely where that model no longer fits. 7.1 The platform economy The most visible challenge comes from the #platform_economy, the growing sector in which people find work through digital applications: driving passengers, delivering food, completing online tasks. Platform businesses often classify the people who work for them as independent contractors rather than employees. This classification has large consequences. In many legal systems, only employees enjoy the right to organize and bargain collectively, so treating platform workers as self employed can place them outside the protection of freedom of association altogether. It can also expose collective action by such workers to challenge under competition law, which forbids independent businesses from fixing prices. The ILO has been studying these issues closely. Research has mapped how freedom of association and collective bargaining might reach platform workers, both through new law and through voluntary agreements (Hadwiger, 2022), and later work has tracked concrete developments in legislation, case law, and collective agreements across several countries (Forsyth and McCrystal, 2025). Some jurisdictions have created special regimes that allow certain platform workers in transport or delivery to bargain collectively and even to take collective action, showing that the classification barrier is not insurmountable. The most important recent development came in 2026. At its session that year, the International Labour Conference adopted a new Convention on decent work in the platform economy, Convention No. 193, by a large majority. That instrument extends core protections, including freedom of association and the effective recognition of the right to collective bargaining, to platform workers regardless of how they are classified. This is a direct response to the classification problem, and it shows the ILO adapting its oldest principle to its newest sector. For freedom of association, the message is clear: the right is meant to follow the worker, not the label attached to the worker. 7.2 The informal economy A second challenge is the #informal_economy, which employs a very large share of the world's workers, especially in developing countries. Informal workers often lack a recognized employer, a fixed workplace, or a written contract. They may be street vendors, home based producers, or day labourers. For such workers the machinery of union recognition and collective bargaining, designed for the formal enterprise, can be difficult to use. Extending genuine freedom of association to the informal economy requires creative forms of organization, such as associations based on trade or locality rather than on a single employer. The supervisory bodies have increasingly recognized that the Convention protects these workers too, but turning that recognition into practice remains a major task. 7.3 Migrant workers Third, #migrant_workers face particular obstacles. Their right to freedom of association is protected in principle, but in practice their immigration status, their dependence on a single employer, and their unfamiliarity with local law can make it dangerous to organize. In some regions, sponsorship arrangements tie a worker's legal presence to a specific employer, which can effectively exclude large numbers of migrant workers from union activity. Recent ILO research has found that migrant workers face both legal and practical barriers to realizing their rights to organize and to bargain collectively. Because migrants often work in the sectors most in need of collective protection, this is a serious gap. 7.4 Shrinking civic space Fourth, freedom of association is threatened wherever civic space shrinks. Because the right is closely tied to freedom of expression, assembly, and protection from arbitrary arrest, it suffers when those broader liberties suffer. In many countries trade unionists face harassment, criminal charges, violence, or worse for ordinary union activity. Annual monitoring by the international trade union movement has reported a steady deterioration in respect for workers rights across many regions, including a rising share of countries where workers are denied access to justice (ITUC, 2025). The ILO itself has estimated that more than forty per cent of the world's population lives in countries that offer no protection for freedom of association and the right to collective bargaining (International Labour Office, 2026). These figures are a reminder that ratification is not the same as respect, a point developed in the next section. 7.5 New forms of employer interference Finally, the Convention protects employers organizations as well, but the more common problem in practice is employer interference with workers organizations. Modern forms of interference can be subtle. They include surveillance of organizing activity, the use of temporary and outsourced contracts to fragment the workforce, and the replacement of genuine collective bargaining with individual arrangements. A recent and widely discussed example in the United Kingdom, involving mass dismissals in the maritime sector, prompted a complaint to the Committee on Freedom of Association and a significant scholarly debate about the effectiveness of the international guarantees (Bogg, 2025; Katsaroumpas, 2024). Cases like this show that the questions Convention No. 87 was written to address in 1948 are still very much alive. 8. Regional and National Dimensions Convention No. 87 sets a single international standard, but it operates in very different national contexts. The gap between the text and the reality on the ground is one of the central problems of the whole system. Ratification is widespread but not universal. Roughly 158 States have ratified the Convention as of 2026, which is a remarkable figure for any treaty. Yet several large and economically important countries have not ratified it, sometimes because their domestic labour law model does not fit neatly with the Convention's requirements, and sometimes for political reasons. Non ratification does not remove all scrutiny, because the constitutional principle and the Committee on Freedom of Association still reach non ratifying States, but it does reduce the formal legal obligation. Even among ratifying States, implementation varies enormously. In some countries the Convention is fully reflected in national law, with strong protection for the independence of unions, easy registration, and effective remedies against anti union discrimination. In others, the law on paper looks acceptable but is undermined in practice by weak enforcement, slow courts, or hostile administration. In still others, the law itself falls short, for example by imposing single union systems, by giving officials the power to dissolve organizations, or by restricting who may hold union office. The Committee of Experts documents these shortcomings year after year in its observations, and the pattern shows that legal reform is necessary but not sufficient. What matters is whether workers and employers can actually use the freedom in daily life. Regional human rights systems add another layer. In Europe, the European Court of Human Rights has read collective labour rights into the freedom of association guarantee of the European Convention on Human Rights, often drawing on ILO standards and on the decisions of the ILO supervisory bodies. In the Americas, the Inter American human rights system has done similar work. This cross referencing means that a State's compliance with Convention No. 87 can be tested not only in Geneva but also before regional courts, which increases the pressure to comply and helps to harmonize interpretations across systems. A national labour law scholar examining a specific country, such as the analysis of protection against victimisation in the South African context, will typically read the domestic constitution, the relevant ILO Conventions, and the regional case law together (Manamela, 2023). The lesson of this section is that Convention No. 87 is best understood as a floor rather than a ceiling. It sets a minimum that all ratifying States must meet, but the real content of freedom of association in any country is shaped by national institutions, by the strength of civil society, and by the willingness of courts and administrations to give the standard practical effect. 9. Critiques and Debates Convention No. 87 is widely admired, but it is not beyond criticism, and a serious study should set out the main debates. The first debate concerns interpretation and legitimacy. As the strike controversy showed, some actors, especially in the employers group, have long argued that the supervisory bodies read too much into a short text and effectively create new obligations that States never accepted. Defenders reply that treaty interpretation always involves working out the implications of general language, and that decades of consistent practice give the corollary reading strong authority. The 2026 advisory opinion has now confirmed the substance of the workers position on the strike question, but the deeper debate about the proper limits of expert interpretation will continue, and it applies to many other areas of ILO law. A second debate concerns the balance between workers and employers. The Convention protects both, but critics on different sides accuse the system of tilting. Some employers feel that the machinery focuses too heavily on workers rights and treats employer organizations as an afterthought. Some worker advocates argue that in practice the system does too little to stop powerful employers from undermining unions through lawful looking means. Both criticisms contain some truth, and they reflect the underlying reality that freedom of association is exercised in a relationship of unequal power. A third debate concerns effectiveness. The ILO has no police force and cannot impose binding sanctions in the way that, for example, a trade body might. Its main tools are dialogue, publicity, and the moral and political weight of its findings. Critics say this is too weak, and that States which are determined to repress unions can absorb the reputational cost. Defenders point out that persuasion and publicity have in fact produced real change in many cases, and that a system built on cooperation reaches further, over the long run, than one built on punishment. This is a genuine trade off with no easy answer. A fourth debate concerns scope and modern work. As Section 7 explained, the Convention was written for a world of stable employment relationships. Some scholars argue that the whole architecture of freedom of association, built around the employer and the enterprise, is poorly suited to platform work, self employment, and the informal economy, and that deeper rethinking is needed rather than the mere extension of old categories. Others argue that the principles are flexible enough to be stretched to new forms of work, and that the 2026 platform Convention proves the point. This debate is likely to define the next generation of scholarship. Finally, there is a debate about universality and cultural difference. Freedom of association is presented as a universal human right, yet its practical shape differs across legal traditions and political systems. Some governments argue that international bodies apply a Western model of pluralist unionism that does not fit their own arrangements. The ILO's answer is that the core of the right, the freedom to form and run independent organizations without state control, is genuinely universal, even if the details of union structure legitimately vary. Where the line falls between legitimate variation and unacceptable restriction is a recurring source of argument. 10. Discussion Stepping back from the detail, several broader observations emerge from this study. The first is that the enduring strength of Convention No. 87 lies less in its words than in the institutions that surround it. The text is short and general. What has made it powerful is the accumulated practice of the Committee of Experts, the Committee on Freedom of Association, and the Conference Committee, together with the willingness of courts and other bodies to take that practice seriously. When students study this Convention, they are really studying a living system of interpretation, not a static document. The strike controversy is the clearest proof of this. The words did not change between 2012 and 2026, but the authoritative understanding of what they require was contested, damaged, and finally reaffirmed through institutional processes. The second observation is that freedom of association is both a labour right and a civil liberty, and that it cannot be secured in one dimension while being denied in the other. A country that arrests demonstrators, censors speech, and jails critics cannot deliver genuine freedom of association at work, however good its labour code looks on paper. This is why the supervisory bodies have always insisted on the link between union rights and basic liberties. It is also why the health of freedom of association is a useful indicator of the broader health of a democracy. The third observation is about power. Freedom of association exists because individual workers are weak against employers when they act alone and strong when they act together. The right is, at bottom, a legal recognition of the value of collective power as a counterweight to economic power. Every debate examined in this article, from the right to strike to the classification of platform workers, is in the end a debate about how much collective power the law will allow workers to build. Reading the Convention with this in mind makes its provisions easier to understand and its controversies easier to follow. The fourth observation is that the system is adapting, though slowly. The addition of a safe and healthy working environment to the fundamental principles in 2022, the 2026 advisory opinion on the right to strike, and the 2026 platform economy Convention all show an Organization trying to keep a 1948 principle relevant to twenty first century conditions. Whether this adaptation is fast enough to match the pace of change in the labour market is an open question, and it is one that the next wave of research and practice will have to answer. Finally, the study underlines the gap between formal ratification and lived reality. With around 158 ratifications, Convention No. 87 is a near universal legal standard. Yet a large share of the world's people live in places where freedom of association is not genuinely respected. This gap is not a reason to dismiss the Convention. It is a reason to focus attention on implementation, enforcement, and the everyday conditions that decide whether a legal right becomes a real one. 11. Conclusion The Freedom of Association and Protection of the Right to Organise Convention, 1948 (No. 87), is a compact treaty with an outsized role in international labour law. It guarantees that workers and employers may form and join the organizations of their choice, run them free from state control, protect them against arbitrary dissolution, and build them into federations and international alliances. It sits alongside Convention No. 98 as one of the twin pillars of #freedom_of_association, and it draws additional force from the ILO Constitution and the 1998 Declaration, which make its core principle binding on every member State regardless of ratification. The Convention's meaning has been developed over more than seven decades by a sophisticated supervisory system, and that system faced its greatest test in the long dispute over the right to strike. The advisory opinion of the International Court of Justice on 21 May 2026, confirming that the right to strike is protected under the Convention while leaving its precise limits to be worked out, marks a turning point in that story and reaffirms the authority of the standard. At the same time, the Convention faces real pressures. The rise of platform work, the size of the informal economy, the vulnerability of migrant workers, and the shrinking of civic space in many countries all test whether a principle written in 1948 can protect the workers of today. The recent adoption of a Convention on decent work in the platform economy suggests that the ILO is willing to extend the old principle to new forms of work, but the wider gap between ratification and respect remains large. For students of #labour_law and #human_rights_at_work, the deepest lesson of Convention No. 87 is that rights are not self executing. A short and elegant text becomes a living guarantee only through the patient work of institutions, courts, unions, employers, and citizens who insist that the freedom to combine is not a privilege granted by the State but a right that belongs to working people everywhere. References 2022 Amendments to the ILO Declaration on Fundamental Principles and Rights at Work. (2023). International Legal Materials, 62(4), 649-655. Cambridge University Press. https://doi.org/10.1017/ilm.2023.14 Bogg, A. (2025). Rhetoric and reality: The P&O Ferries scandal and the right to strike. Comparative Labor Law & Policy Journal, 45(2). Curtis, K., & Wolfson, O. (Eds.). (2022). 70 years of the ILO Committee on Freedom of Association: A reliable compass in any weather. International Labour Office. Forsyth, A., & McCrystal, S. (2025). Collective bargaining and collective action in the platform economy: A 2024 update of developments in legislation, case law and collective agreements (ILO Working Paper No. 143). International Labour Office. Hadwiger, F. (2022). Realizing the opportunities of the platform economy through freedom of association and collective bargaining (ILO Working Paper No. 80). International Labour Office. International Court of Justice. (2026). Right to strike under ILO Convention No. 87 (Request for advisory opinion), Advisory Opinion of 21 May 2026. I.C.J. Reports 2026. International Labour Office. (2026). Fundamental principles and rights at work: A critical crossroads. International Labour Office. International Trade Union Confederation. (2025). ITUC Global Rights Index 2025. International Trade Union Confederation. Katsaroumpas, I. (2024). Authoritarian liquid transgressions: The case of P&O Ferries. Journal of Law and Society, 51(2), 263-289. https://doi.org/10.1111/jols.12477 Manamela, M. E. (2023). The right to freedom of association and the protection of employees against victimisation in the workplace. Potchefstroom Electronic Law Journal, 26, 1-27. https://doi.org/10.17159/1727-3781/2023/v26i0a13939 Mantouvalou, V. (2023). Structural injustice and workers' rights. Oxford University Press. https://doi.org/10.1093/oso/9780192857156.001.0001 Vogt, J., Bellace, J., Compa, L., Ewing, K. D., Hendy, J., Lorcher, K., & Novitz, T. (2020). The right to strike in international law. Hart Publishing. https://doi.org/10.5040/9781509933587 #Freedom_of_Association #ILO_Convention_87 #Convention_No_87 #Right_to_Organise #Right_to_Strike #Collective_Bargaining #International_Labour_Organization #Workers_Rights #Trade_Union_Rights #Labour_Law #Human_Rights_at_Work #Committee_on_Freedom_of_Association #ILO_Supervisory_System #Convention_87_1948 #Freedom_Of_Association_And_Protection_Of_The_Right_To_Organise #C087 #ICJ_Advisory_Opinion_2026 #Platform_Economy_Labour_Rights #Fundamental_Principles_And_Rights_At_Work #FreedomOfAssociation #RightToOrganise #ILOConvention87