The History of Minimum Wage Laws

The economic textbook case against minimum wages is elegant and wrong. In a perfectly competitive labor market, wages are set at the marginal product of labor, meaning that the wage a worker receives equals the value of what they add to production at the margin. If a government mandates a wage floor above this equilibrium, employers will hire fewer workers — the supply and demand curves cross at a quantity below what would prevail at the floor price, creating unemployment. This argument is logically valid given its assumptions. The assumptions do not describe the labor markets in which most low-wage workers actually work. This gap between the theoretical model and the institutional reality it is supposed to represent explains why a century of minimum wage experience has produced outcomes that the standard model consistently failed to predict, and why the political economy of minimum wages has been systematically underestimated by economists who trusted their models more than the evidence.

New Zealand introduced the world’s first statutory minimum wage in 1894, embedded in the Industrial Conciliation and Arbitration Act. The context matters: New Zealand in the 1890s was a politically radical colony in the middle of a severe depression, governed by a Liberal party that was attempting to institutionalize wage determination through compulsory arbitration rather than direct legislation. The minimum wage was not set as a fixed floor for all workers but emerged from a system of wage boards and arbitration tribunals that set minimum rates by industry and occupation. Australia followed a similar path, developing a compulsory arbitration system that set minimum wages through judicial determination rather than legislative mandate. The Harvester Case of 1907 is the foundational document of Australian minimum wage law: Justice H.B. Higgins of the Conciliation and Arbitration Court ruled that the Sunshine Harvester Company must pay its workers a “fair and reasonable” wage, which he defined as enough to support “a human being in a civilized community” — not the employer’s ability to pay, not the market clearing wage, but the subsistence floor required for dignity.

The philosophical distance between Higgins’s reasoning and the neoclassical economic framework could not be greater. Higgins was not asking what wage the market would produce and declaring it fair. He was asking what wage a human being required to live decently and declaring that no employer had the right to pay less, regardless of what a competitive market might produce. This is a claim about rights, not about efficiency. It implies that labor markets — like other markets for things that matter deeply to human welfare, such as housing, healthcare, and food — produce outcomes that are not automatically acceptable simply because they are the results of voluntary exchange. A worker who accepts a starvation wage because the alternative is starvation has not freely consented to a just arrangement. The exchange is voluntary in the narrow contractual sense and coerced in the broader social sense.

American labor politics resisted statutory minimum wages far longer than Antipodean politics did, partly because of the power of the Supreme Court, which in Lochner v. New York (1905) and subsequent decisions struck down labor regulations as unconstitutional violations of freedom of contract. The Lochner era lasted until 1937, when the Court abruptly shifted position and upheld Washington State’s minimum wage law for women in West Coast Hotel v. Parrish. The Fair Labor Standards Act of 1938 then established the first federal minimum wage of 25 cents per hour — a rate that was set at about 40% of average manufacturing wages at the time, which is the range that has historically characterized American minimum wage policy before subsequent erosion by inflation.

The FLSA’s passage was not a smooth demonstration of progressive political power. It was the product of a complicated coalition that included southern Democrats who supported a federal minimum only because they believed — correctly, given the political economy of the era — that the rate would be set low enough and the coverage narrow enough to protect the low-wage southern economy from real disruption. Agricultural workers and domestic servants were excluded from coverage, a carve-out that had the predictable effect — and, for the southern legislators who insisted on it, the intended effect — of excluding the majority of Black workers from protection. The minimum wage was a labor standards victory that was deliberately structured to provide the least protection to the most exploited workers. This pattern — labor market standards that protect the organized and relatively privileged while excluding the most vulnerable — has recurred in minimum wage policy across multiple countries and decades.

The empirical debate over minimum wage effects became substantially more sophisticated after 1994, when David Card and Alan Krueger published their study of the 1992 New Jersey minimum wage increase. New Jersey had raised its minimum wage; neighboring Pennsylvania had not. Card and Krueger compared employment in fast food restaurants on both sides of the border before and after the increase, and found no evidence that employment fell in New Jersey relative to Pennsylvania. The study was methodologically innovative — the state border provided something close to a natural experiment — and its conclusions were explosive. Standard economic models predicted that a minimum wage increase would reduce employment. The evidence said it had not. This was not a minor discrepancy. It was a challenge to the fundamental theoretical framework.

The response from orthodox economists was intense and not always intellectually generous. Multiple studies attempted to find methodological flaws in Card and Krueger’s work. Some succeeded in identifying issues with the data. None succeeded in establishing that minimum wage increases consistently reduced employment by the magnitude the competitive model predicted. Over the following three decades, the empirical literature expanded enormously — covering dozens of countries, hundreds of minimum wage changes, and a wide variety of methodological approaches — and its conclusions have been remarkably consistent. Minimum wage increases in the range of policy relevance — meaning increases that bring the minimum to something between 40% and 60% of median wages — produce little or no detectable employment reduction in the short to medium term. Increases that push the minimum wage above roughly 60-70% of median wages begin to show clearer employment effects, though even here the evidence is mixed and context-dependent.

The theoretical explanation for these findings centers on monopsony — a concept that the competitive labor market model excludes by assumption but that describes the actual structure of many low-wage labor markets with considerable accuracy. In a monopsony, one employer (or a small number of employers) dominates the local labor market and faces an upward-sloping labor supply curve rather than the flat supply curve implied by perfect competition. A monopsonistic employer will hire fewer workers than the competitive equilibrium would produce and pay them less — both because it can, given its market power, and because hiring an additional worker requires raising wages for all existing workers. In this setting, a minimum wage increase can actually increase both wages and employment by pushing the monopsonistic employer toward the competitive optimum. The minimum wage is a corrective for market power, not an interference with competitive equilibrium.

The sectors that employ the most minimum wage workers — retail, food service, hospitality, domestic care — share structural features consistent with monopsony or oligopsony. Workers face significant job search costs, geographic constraints, family obligations, and information asymmetries that reduce their ability to shop among employers. Employers in these sectors often engage in implicit or explicit wage coordination through industry associations, management consulting firms that benchmark compensation, and shared ownership structures. The assumption that minimum wage workers operate in thick, competitive labor markets where any individual employer’s wage offer is disciplined by constant worker mobility and multiple competing offers is inconsistent with the empirical literature on labor market structure in these sectors. Minimum wage workers are, almost by definition, the workers with the least labor market power. Modeling their wages as the outcome of competitive equilibrium is not an empirical description. It is an ideological commitment.

The political economy of minimum wages has consistently surprised economists who expected employers to mobilize against increases. In practice, minimum wage increases often face ambivalent business opposition, particularly from larger employers who have already implemented wage floors above the statutory minimum and who benefit from having competitors forced to pay similar rates. Walmart’s public support for minimum wage increases in the 2010s was not altruistic — it reflected the calculation that a higher federal minimum would squeeze the small competitors who were undercutting Walmart’s prices partly by paying lower wages. The business community’s relationship to minimum wage policy is not simply oppositional. It is complicated by the competitive dynamics within industries and the distributional effects of different wage floor levels.

Cross-country evidence adds another dimension that the exclusively American literature cannot capture. Countries with minimum wages set at 60-70% of median wages — France, Australia, the United Kingdom post-2015 — have not experienced the employment catastrophes that standard models predicted. Britain’s Low Pay Commission, established in 1998 to set the National Minimum Wage at levels informed by evidence rather than political bargaining, has overseen twenty-five years of minimum wage increases that have substantially compressed the bottom of the wage distribution without producing measurable employment reductions. The experience is imperfect evidence — Britain changed in many other ways simultaneously, and the counterfactual is unknowable — but it is evidence, and it is inconsistent with the simplest version of the competitive model’s predictions. The economists who confidently predicted employment disaster from British minimum wage increases were wrong, and their being wrong has not fundamentally changed the prior beliefs they bring to the next round of predictions.

The history of minimum wage policy is ultimately a history of the tension between institutional economics and neoclassical economics — between a tradition that takes seriously the power relationships embedded in labor market institutions and a tradition that abstracts from those relationships to derive clean theoretical results. The institutional tradition, which produced the New Zealand arbitration system, the Harvester Case, and the FLSA, started from the observation that unregulated labor markets produce outcomes that societies find unacceptable and that workers with limited bargaining power cannot remedy through individual negotiation. The neoclassical tradition started from a model of markets as self-correcting equilibrating mechanisms and treated minimum wages as distortions of an otherwise efficient system. A century of evidence has not definitively resolved this debate, because the counterfactuals are never available and the evidence is always contingent. But the weight of the evidence is clearly on one side of it, and that side is not the one that has dominated economics departments.