He also posited that “differences about economic policy among disinterested citizens … in principle can be eliminated by the progress of positive economics—rather than from fundamental differences in basic values, differences about which men can ultimately only fight” (emphasis added because both qualifications may partially shield Friedman from criticism). In essence, he believed that empirical studies on the effects of specific policy proposals allowed economists to don the mantle of science, as well as offered the best way for “eliminating”—or maybe just “narrowing”—disagreements among policy opponents.

Today, nearly three-quarters of a century after the article and a half-century after empirical work began accruing on the minimum wage, there remains a large disagreement between policymakers, and between economists themselves, over the desirability of this policy. So, was Friedman wrong about the power of positive economics? Why hasn’t the debate over the minimum wage been decided and retired?

The Political Combatants on Minimum Wage Hikes

In line with Friedman’s “ventured judgment,” minimum wage proponents’ and opponents’ disagreements have long maintained contrasting assessments of the hikes’ market effects.

Proponents have tended to support hikes on the grounds that they have little to no negative employment effects, while they enhance the welfare of many covered workers. The proponents are unswayed by concerns that employers will respond with price increases and reductions in other forms of worker compensation. Indeed, they argue that minimum wage hikes might increase employment because higher income can boost consumer demand for goods that low-wage workers produce. Then, hikes can induce monopsony employers to elevate their suppressed wage levels toward competitive levels (Kalenkoski, 2024). Some proponents also argue that, though minimum wage hikes may have adverse employment and other social effects, such effects do not close the case against minimum wage increases; alternative government welfare/anti-poverty policies—such as food stamps and rent controls—can also reduce low-wage employment, yet those programs are still deemed worthwhile (see Reich and Sosinskiy, 2024).

Opponents, on the other hand, criticize minimum wage hikes primarily because of their negative employment effects, as microeconomic theory predicts (see Brown et al., 1982). The hikes are also expected to have an array of unwanted subsidiary effects, such as fueling discrimination against young, minority workers, thereby reducing their chances of achieving livable wages in the long run. If empirical studies show minimal or no effects of hikes, opponents argue, it’s most likely because the effects of competitive market pressures on employers have caused them to offload their added labor costs by raising prices, reducing workers’ benefits, and increasing work demands. In addition, hikes can have perverse social consequences—for example, increasing would-be workers’ self-employment in crime industries. Fone et al. (2023), looking at workers 16–24 years of age, found that a 1 percent increase in the minimum wage led to a 0.2 percent increase in arrests for property crimes (related to larceny). Why? A minimum wage hike can lead to unemployment, causing some unemployed workers to turn to crime.

Finally, as Richardson and I (2021) have argued, minimum wage hikes can lower (in hidden ways) the disposable incomes of covered low-wage workers who also benefit from a variety of welfare programs. Why? Many welfare program benefits are conditioned on low earned income. Thus, wage hikes can lower worker welfare benefits from multiple programs by more than their earned income rises.

The Debate Continues Despite a Flood of Empirical Studies

My claim here is not that Friedman was wrong on the theory of minimum wage hikes and deduced predictions on the directional employment effects of hikes. He was a staunch opponent of all minimum wages, not just hikes in their levels. In a 1966 Newsweek column, he opposed minimum wage hikes for the usual reason: They reduce employment and therefore national income. But he also noted that they were a major driver of the “shockingly” higher unemployment rates of “women, teenagers, Negroes and particularly Negro teen-agers.” He wrote, “Within two years after the legal minimum was raised from 75 cents to $1 an hour in 1956, unemployment among Negro boys shot up to 24 per cent and among white boys to 14 per cent.” Having a libertarian bent, he also lamented the infringement of mandated wages on both workers’ and employers’ freedom to strike mutually beneficial contracts. In short, he deduced, “The rise in the legal minimum wage rate is a monument to the power of superficial thinking.” In my view, all these arguments are sound.

At the same time, Friedman was dead wrong—meaning overly optimistic—on the extent to which empirical studies would (and maybe could) “eliminate,” or just reduce, the perennial disagreements over proposed hikes. The political controversy over minimum wage hikes rages today, perhaps even more fiercely and with greater media attention, than at the time Friedman published his 1953 article. After nearly three-quarters of a century of a steady (and maybe increasing) flow of published studies on an ever-widening array of empirically validated effects of minimum wage hikes, there may be no more policy unity on proposed hikes today than before the early 1950s—perhaps partially because of (not despite) the ongoing stream of empirical studies done on hikes made at all levels of government. Proponents and opponents are now armed with portfolios of empirical studies they regularly cite to validate their positions.

Today, most Americans favor more than doubling the federal minimum wage (Dunn, 2021). It’s hardly unreasonable to suggest there is no consensus among policymakers on the minimum wage. That can also be said for economists, at least as of a decade ago (Vail, 2015).

Friedman’s Minimum Wage Hypothesis

Friedman’s proposed hypothesis, grounded in his positive economics, was straightforward: Since policy disagreements tended to be on the economic and social effects of minimum wage hikes, those divides could be narrowed, if not eliminated, by empirical studies of the effects on affected labor markets. The studies would consider employment and crime rates, fringe benefits, and work tenures, as well as a host of more amorphous effects such as those on health, education, and happiness.

Since Friedman published his methodology essay and mused that minimum wage disputes could be replaced by a “consensus,” there have been hundreds of published empirical studies on various labor-market effects of many federal, state, and municipal minimum wage increases. The record strongly supports his analytical deduction and empirical prediction that minimum wage increases reduce employment of the targeted labor market group. However, he has been proven overly optimistic—maybe dead wrong—on his “ventured judgment” that empirical findings would narrow, and even resolve, the policy debate between proponents and opponents, especially if the effects of hikes are replicated across studies undertaken across the more than seven decades.

Friedman supporters can be pleased that federal policymakers, no matter which party held power in Washington, have not increased the federal minimum wage since 2009, and that increases before then, going back to the late 1970s (when the federal Minimum Wage Study Commission—whose staff included Brown et al.—was at work), have been small. The purchasing power of the federal minimum wage has been in precipitous decline since 1968, falling 52 percent.

Those supporters would be much less happy with policy at the state and local levels, where minimum wage laws and increases have proliferated. Thirty states and the District of Columbia now have minimum wage rates above the federal minimum, ranging from $8.75 in West Virginia to $17.50 in DC, with 11 of the states and DC at $15 or more. California’s minimum wage for industries besides fast food is $16.50, and since April 2024 the state has required fast-food restaurant chains with 60 or more locations nationwide to pay a minimum wage of $20, to be adjusted annually for inflation.

The Empirical Minimum Wage Record

Testing the employment effects of minimum wage hikes since the early 1950s has become something of a cottage industry for the academy and business and labor interest groups (both wage supporters and opponents). A substantial majority—I’d estimate upward of 95 percent—of published studies have found confirmation of Friedman’s and almost all neoclassical economists’ usual predictions that increases have led to fewer jobs and higher unemployment rates for the covered workers. Minimum wage hikes have also been shown to reduce employment benefits and increase crime rates. Name a potential work benefit, and some study has shown that minimum wage hikes have resulted in less of it, making covered workers worse off (McKenzie, 2021).

Brown et al. (1982b) were the first to review previously published empirical studies on the employment effects of minimum wage hikes. They observed that teenagers had been the most frequently studied segment of the low-wage labor force, mainly because they had jobs most sensitive to minimum wage hikes. They concluded that “time-series studies typically find that a 10 percent increase in the minimum wage reduces teenage employment by 1 to 3 percent.” All studies they reviewed found negative employment effects, or negative coefficients on the employment variables used, though only about half the coefficients were statistically significant. This finding has been used as fodder by proponents of minimum wage hikes, who point to the lack of significance to cast doubt on the reliability of claims of negative employment effects.

Brown et al. also concluded that the “preferred” portion of the 1–3 percent range for assessing employment effects was the “lower half,” meaning that a 10 percent minimum wage hike would most likely reduce teenage employment by 1.5 percent or less. The authors, in another article, refined their point when they suggested that the percentage of teenage job losses from 0 to 0.75 percentage points was the “most plausible” (Brown et al., 1982a, p. 73).

Minimum wage studies undertaken since 1982 have tended to fortify Brown et al.‘s central conclusions for teenage workers, finding negative but small employment effects. In a later literature review, Neumark and Wascher (2006) came to much the same conclusion: Minimum wage hikes reduced employment, but the effects were small as a percentage of covered workers.

A small number of other researchers have found positive, albeit small, employment effects or no detectible effects. The most prominent of these is Card and Krueger (1994), which the authors expanded upon in their 1995 book Myth and Measurement. They used something of a natural experiment to assess the employment effects of a minimum wage hike imposed in New Jersey but not neighboring Pennsylvania. Their central finding was that the 10 percent hike in New Jersey did not suppress fast-food employment relative to Pennsylvania.

This conclusion was challenged by Neumark and Wascher. Looking across the breadth of the minimum wage research, they concluded, “The oft-stated assertion that recent research fails to support the traditional view that the minimum wage reduces the employment of low-wage workers is clearly incorrect.” In a 2012 interview for National Public Radio, Neumark further explained, “The consensus from a lot of studies I’ve surveyed—including my own—says that a 10 percent increase in the minimum reduces employment of those very low-skilled groups by about 1 to 2 percent,” which the NPR interviewer described as a “tiny” percentage of the total labor force (Kaste, 2012).

California / As noted above, California has instituted a minimum wage of $20 per hour for employees of fast-food restaurants with 60 or more locations across the country. At the time, the state’s minimum wage was $16.50, meaning that the wage for fast-food restaurants underwent a 21 percent increase. Most critics expected that the new fast-food minimum would likely have a larger negative employment effect than found in previous studies that largely examined the employment effects of only 10 percent increases. This is because the hike would exceed the wage of many fast-food workers, and employers would be hard-pressed to offset the new, substantially higher minimum by, say, reducing benefits and raising work demands.

Sure enough, the new fast-food minimum did have higher disemployment effects than prior 10 percent increases, but not by much: The percentage drop in fast-food employment was still pretty modest. Clemens et al. (2025) found:

that employment in California’s fast-food sector declined by 2.7 percent relative to employment in the fast-food sector elsewhere in the United States from September 2023 through September 2024…. Our median estimate translates into a loss of 18,000 jobs in California’s fast-food sector.

Battlelines / Proponents and opponents of minimum wage hikes have found new grounds for making their policy cases, magnifying (so it seems) the political influence of estimated disemployment statistics once they start addressing public—not academic—audiences. Proponents (including labor groups) emphasize the percentage reduction in job losses because the percentage reduction (2.7 percent of fast-food workers) may seem “small” or even “tiny” to non-academics. To such an audience, that number might be small enough that the estimated negative effect could be a random statistical error, one that leaves open the prospects of the “true” disemployment effect being zero or even positive. Presidents Barack Obama and Lyndon Johnson both pushed for higher minimum wages by stressing the “small” measured disemployment effects or even the reported employment increases. In his 1995 State of the Union address, President Bill Clinton even noted Card and Krueger’s work to support his 1995 proposed increase to $5.15 from $4.25 an hour (Kosters, 1996, p.73).

Opponents stress the number of job losses: 18,000 in the case of California’s imposition of its fast-food minimum. Taken by itself, that sounds “large” to general audiences. These opponents never mention that the total labor force in California’s fast-food sector is nearly three-quarters of a million workers.

Market Explanation for the “Minor” Minimum Wage Job Losses

In the early 1980s, seeking fresh explanations for the “small” effects of minimum wage hikes, Wessels (1987) and I (1980) separately devised similar market explanations for the “low” or “small” estimated job losses. Conventional analyses of the employment effects of minimum wage hikes presumed that a mandated wage rate set above market (or equilibrium) would result in a market surplus of labor equal to the sum of fewer workers hired and more workers seeking employment. That’s where typical conceptual discussions left the analytics: They implicitly presumed that employers had no way to respond to the hikes to reduce their labor costs and/​or didn’t have sufficient incentives to discover ways of reducing those costs. That’s hardly what market economists should have expected of profit-maximizing employers.

Instead of considering employer offsets to forced wage hikes, economists generally explained the limited job losses by deducing that employers’ demand for low-wage labor was “inelastic,” meaning the amount used did not change much relative to the change in price. Frankly, that presumption doesn’t square with typical market arguments that the elasticity of labor demand is a function of the number of substitutes for workers. Employers of low-wage workers, then and now, are generally thought to have a number of available major substitutes for low-skilled, low-wage workers, presumably increasing the demand elasticity for labor: They can, for example, release their lower-skilled workers and keep their higher-skilled ones, automate work processes, hire illegal aliens, seek workers from higher-skilled labor markets, and use import substitutes, just to name a few options.

Wessels and I each concluded that the low-demand-elasticity argument of low-wage labor is a weak point on which to explain the low employment effects of minimum wage hikes. And we both explored how profit-maximizing employers could be expected to reduce the added labor-cost burden of minimum wage hikes by reducing—to the extent feasible—employee benefits and increasing the work demands on employees. Employers had two motivations to do so: First, they could increase their profits or reduce their losses by such measures. Second, they could be pressed to do so by other employers who use the strategies to gain a cost advantage in production and a price advantage in their product markets. And with the labor-market surplus, employers need not worry about objections to the new work conditions because of the threat of existing workers being let go.

More importantly, our analytics led to conclusions at substantial odds with conventional analysis. Economists have traditionally deduced that a minimum wage set above the market-clearing wage would lead to a market surplus of workers (the counterpart of which is a market shortage of jobs). The result? Fewer workers would be hired than before the wage hike, and some workers would be displaced and have to seek lower-paying jobs in labor markets not covered by the minimum wage. Alternatively, the displaced workers would go on unemployment and/​or welfare.

So, the conventional wisdom is that many covered workers—maybe a substantial majority—would be made better off with higher incomes, while a much smaller share of covered workers would be made worse off. Wessels’ and my analytics point out dramatically different effects: reductions in benefits and increases in work demands, resulting in covered workers being made worse off. That is mainly because the value of the benefits taken away and the negative consequences of the added work demands are bound to be more valued by workers than the value of the higher money wage or else employers would not have provided the benefits and reduced work demands in the first place.

Several researchers have undertaken studies that support our hypotheses. They have shown that minimum wage increases reduce what limited benefits these workers receive and increase their work demands (McKenzie, 2021).

In short, our analytics may help explain findings of “small” or “tiny” job losses from minimum wage hikes. If we are right, then the magnitude of the direct effects of minimum wages are generally understated, maybe seriously so. Yet, much of the empirical literature on raising the minimum wage overlooks the effects along these dimensions. Wessels’ and my arguments have not received much attention from either proponents or opponents of minimum wage increases, maybe because neither side finds them helpful in fortifying their political positions. (See Wall Street Journal Editorial Board, 2025, for an example of an opponent that continues to stick with the old arguments.)

Concluding Comments

The vast published empirical literature that has accumulated on the labor-market effects of minimum wage increases since Friedman’s 1953 article has strongly supported his (and his followers) on his most important economic prediction: Minimum wage increases reduce the count of low-wage jobs in the economy and have other undesirable effects such as reductions in job-related benefits. However, the empirical evidence that has emerged in the latter half of the 20th century must have left him disappointed because it showed the disemployment effects have been “small.” Indeed, they likely have been smaller than Friedman imagined when he expressed confidence that empirical work, guided by his positive economics, would relieve at least some of the policy conflicts over proposed hikes.

He surely didn’t expect that repeated findings of “small” employment effects would be used before and, to a growing extent, after his death to support the position that minimum wage increases raise the incomes of a large majority (maybe over 95 percent) of low-wage workers. Proponents have used the accumulated findings to argue that the limited disemployment effects show that minimum wage hikes improve the welfare of more low-wage workers than other available politically designed welfare policies, which have also been shown to have negative effects on low-wage labor markets.

No doubt, Friedman would have shaken his head in disbelief at California Gov. Gavin Newsom, who said of a University of California, Berkeley, study (Reich and Sosinkiy, 2024) of the 21 percent raise in the fast-food minimum wage, “This study reaffirms that our commitment to fair wages for fast-food workers is not only lifting working families economically, but also strengthening our economy.”

If Friedman were alive today, I am confident that he would want to revise his 1953 paper. He was far too optimistic about empirical studies settling policy debates. I suspect he would lament the extent to which empirical studies, by overlooking some effects that emerge in a dynamic market, could widen the gap between policy partisans and make their differences more intense.

Readings

• Brown, Charles, et al., 1982a, “The Effects of the Minimum Wage on Employment and Unemployment, A Survey,” Working Paper No. 846, National Bureau of Economic Research, January.
• Brown, Charles, et al., 1982b, “The Effects of the Minimum Wage on Employment and Unemployment,” Journal of Economic Literature 20(2): 487–528.
• Card, David, and Alan B. Krueger, 1994, “Minimum Wages and Employment: A Case Study of the Fast-Food Industry in New Jersey and Pennsylvania,” American Economic Review 84(4): 772–793.
• Card, David, and Alan B. Krueger, 1995, Myth and Measurement: The New Economics of the Minimum Wage, Princeton University Press.
• Clemens, Jeffrey, et al., 2025, “Did California’s Fast Food Minimum Wage Reduce Employment?” Working Paper 34033, National Bureau of Economic Research, July.
• Dunn, Amina, 2021, “Most Americans Support a $15 Federal Minimum Wage,” Pew Research Center, April 22.
• Fone, Zachary S., et al., 2023, “The Unintended Effects of Minimum Wage Increases on Crime,” Journal of Public Economics 219: 104780.
• Friedman, Milton, 1953. “The Methodology of Positive Economics,” Essays in Positive Economics, University of Chicago Press, pp. 3–43.
• Friedman, Milton, 1966, “Minimum Wage Rates,” Newsweek 68(13): 82–83 (September 26).
• Kalenkoski, Charlene Marie, 2024, “The Effects of Minimum Wages on Youth Employment,” IZA World of Labor.
• Kaste, Martin, 2012, “Raising the Minimum Wage: Whom Does It Help?” Morning Edition, National Public Radio, January 3.
• Kosters, Marvin, 1996. The Effects of the Minimum Wage on Employment, AEI Press.
• McKenzie, Richard B., 1980, “The Labor Market Effects of Minimum Wage Laws: A New Perspective,” Journal of Labor Research 1(2): 255–264.
• McKenzie, Richard B., 2021, “On the Minimum Wage, Both Sides Have Their Economics Wrong,” Regulation 44(2): 9–11.
• Neumark, David, and William Wascher, 2006, “Minimum Wages and Employment: A Review of Evidence from the New Minimum Wage Research,” Working Paper No. 12663, National Bureau of Economic Research, November.
• Reich, Michael, and Denis Sosinskiy, 2024, “Sectoral Wage-Setting in California,” IRLE Working Paper No. 104–24, Institute for Research on Labor and Employment, University of California, Berkeley, September.
• Richardson, Craig J., and Richard B. McKenzie, 2021, “Progressives’ Desires to Help the Poor Will End Up Hurting Them Instead,” EconLib, September 6.
• Vail, Chelsea, 2015, “What Economist Think about a $15 Minimum Wage,” Chicago Booth Review, September 22.
• Wall Street Journal Editorial Board, 2025, “The Wages of Teen Unemployment,” Wall Street Journal, August 29.
• Wessels, Walter J., 1987. Minimum Wages, Fringe Benefits, and Working Conditions, AEI Press.