In a report issued last February, the Congressional Budget Office (CBO) estimated that raising the federal minimum wage from $7.25 to $10.10 an hour—a 39 percent increase advocated by President Obama—would reduce U.S. employment by hundreds of thousands of jobs. Just a few weeks earlier, more than 600 economists joined seven economics Nobel laureates in endorsing the proposal and claiming that it would have “little or no negative effect on employment.” Then, in March, a statement signed by more than 500 economists, including three Nobel laureates, asked the federal government to think twice before increasing the minimum wage. (Disclosure: I was among the signatories of that statement.) Which group of economists is correct, and why is there such a difference of opinion?

The CBO—a nonpartisan, credible research bureau of the U.S. Congress—offers a best-estimate that increasing the federal minimum wage to $10.10 in three annual steps would, by the second half of 2016, deprive 500,000 workers of their jobs. The range of the estimate extends from a very light decrease in employment to one million workers out of work. Those workers would, of course, be made much poorer, but the tradeoff for their losses would be higher compensation for as many as 33 million other workers who would keep their jobs. The CBO estimates that similar, but smaller, effects would follow an increase of the minimum wage to $9 over two years.

The CBO report is quite prudent, incorporating all factors that could benefit employment and incomes. For example, it includes an increase of “a few tens of thousands of workers” (in the $10.10 scenario) whose jobs would be created by a Keynesian boost of aggregate demand caused by the higher minimum wage. It recognizes that its estimates are full of uncertainties, as illustrated by its assessment of a one-third chance that the employment effect will be out of the forecasted range. Nonetheless, the CBO concludes that the proposed increases in the minimum wage would likely create unemployment.

The Weight of the Evidence

This conclusion is not surprising. There is no good theoretical reason why a minimum wage, if it is binding on some employers, would have anything but a negative effect on employment. If the labor market is competitive (which it certainly is, especially at the bottom of the employment ladder), then any employer will let go—or not hire—an employee whose marginal productivity is lower than the minimum wage (plus payroll taxes and other employment costs) that the employer is obliged to pay.

A large number of empirical studies have confirmed that a minimum wage, if set above the equilibrium wage level, will destroy jobs. In their seminal 2008 book Minimum Wages, David Neumark and William Wascher present a review of empirical research on minimum wages in the United States. They conclude that there is plenty of evidence that “minimum wages reduce employment opportunities for less-skilled workers,” and they admit that their own research changed their prior views on the weight of evidence regarding the effects of minimum wages.

Anecdotal evidence that raising the minimum wage hurts employment is not hard to find. Last year, the left-wing magazine The Nation started an online petition demanding that Walmart boost its workers’ average hourly wage from $8.18 an hour to $12. Walmart countered that The Nation had only recently increased its intern stipends to the equivalent of the federal minimum wage of $7.25 (previously taking advantage of an exception that young people who are hired for less than 90 days need only be paid $4.25 an hour). The magazine conceded, “We are not yet certain how this will work out long term, but for the fall we are anticipating hiring 10 interns rather than 12.”

Yet, some recent empirical research (summarized by John Schmitt in a 2013 paper for the Center for Economic and Policy Research) finds little or no negative employment effect from increasing the minimum wage. In a forthcoming article, Neumark, J. M. Ian Salas, and Wascher criticize the econometric methods used in that research. The academic debate seems to be turning into arcane econometric disagreements. The more arcane the debate, the less we should be swayed from standard economic theory, according to which the demand curve for labor (like all demand curves) has a negative slope. Perhaps the negative employment effect of minimum wages is smaller than was previously thought, but it is hard to deny that there is an effect.

Even Paul Krugman is prudent in his evaluation of the evidence, affixing some telling qualifiers to his bold policy recommendations. An increase in the minimum wage, he wrote in a 2013 New York Times column, “would have overwhelmingly positive effects; it would have “little if any negative effect … on unemployment”; “modest increases … don’t necessarily reduce the number of jobs”; and “the main effect of a rise in minimum wages is a rise in the incomes of hard-working but low-paid Americans” (my emphasis). Those qualifiers seem to imply that even Krugman recognizes that some effects will be negative and that even a modest increase (let alone a 39 percent increase) in the minimum wage may reduce the number of jobs, and perhaps that secondary effects would reduce the incomes of some hard-working but low-paid Americans.

Yet, Krugman concludes (speaking about Republican leaders opposed to raising the minimum wage), that “[t]hey say that they’re concerned about the people who might lose their jobs, never mind the evidence that this won’t actually happen.” So “we should raise the minimum wage, now.” Why so much certainty in the policy prescription advanced by Krugman and economists of his obedience?

How Can Economists Be Wrong?

Among the general public, opinion surveys show a massive support—around 75 percent—for increasing the minimum wage. This stance among people with no economic background is not surprising. Only a small minority is hurt by minimum wages. Around 5 percent of employed workers receive the current minimum wage or (given various legal exceptions) less. (The proportion would be a bit higher if we consider state laws; as of the beginning of this year, 21 states had a higher minimum wage than the federal wage.) The half-million workers that the CBO estimates would lose their jobs if the federal minimum wage were raised to $10.10 an hour represent only one-third of 1 percent of the employed work force. Moreover, as generally is the case in politics, voters remain rationally ignorant of the issue and the relevant research.

On minimum wages (like on many other issues), economists take a different stance than the general public, although economists are not as free-market as many people assume. Most surveys suggest that American economists are less favorably disposed to the minimum wage than the general public. They are split about half and half on its desirability, having perhaps become more supportive than they used to be.

To explain why so many economists support the minimum wage, Neumark and Wascher suggest that the average economist is simply not familiar with research in a field different from his primary specialization. There might be something to this explanation, but one wonders why, before going public, economists would not bother to review the literature.

In late 2006, Daniel Klein and Stewart Dompe surveyed a group of economists who had signed a public statement endorsing a minimum wage increase. They found that roughly two-thirds of the respondents conceded that raising the minimum wage would have some disemployment effects (though most of them characterized the effects as “minor”). Among other questions, Klein and Dompe asked the respondents if they believed minimum wages to be coercive and thus contrary to liberty. About three-fourths did not believe so. Although coercion is a more complex notion than many think, some write-in comments (which respondents were encouraged to add) seem remarkably naive—like, for example, these two from two different respondents:

  • “I do not subscribe to this definition of liberty. For me, it has more to do with free speech and freedom of association and other civil liberties.”
  • “[To me, the primary meaning of liberty is] Freedom of expression.”

In other words, those respondents seem to believe that liberty is only important if it involves a field of activity they personally value.

Is this naiveté or something else? That question leads us to another reason why some economists favor minimum wages: they may simply espouse values about distribution that are different from those of economists on the other side of the issue. Economists promoting minimum wages may be willing to make a different tradeoff between winners and losers, between those who lose their jobs and those who stay employed at higher wages. From the Klein and Dompe survey, we can conclude that two-thirds of respondents were conscious of a tradeoff against the unemployed, and were willing to make it. They prefer that some people lose their jobs in order that others benefit from redistribution (mainly from business owners, according to many respondents). This sort of value-laden distributive judgment is the essence of politics.

Economists promoting minimum wages may be willing to make a different tradeoff between winners and losers, between those who lose their jobs and those who stay employed at higher wages.

Another reason why many economists have abandoned their skepticism toward the minimum wage may simply be that they follow their fans—that is, they feel obliged to satisfy their fans’ expectations. Like other intellectuals, economists are not only influenced by their political or academic bosses, they are also subject to what their admirers expect from them. This dynamic may have a stronger effect on more famous economists for the simple reasons that they have more fans. Krugman’s views, for example, have become less scholarly as he’s become more popular and politically active.

All this does not bode well for the credibility of economics as a science. When acting as a policy adviser, the economist cannot avoid having skin in the game—contrary to the physicist, whose deep personal values are probably not challenged much by the fate of Shrödinger’s cat, whether dead or alive or both. That most economists know little about welfare economics does not help them distinguish between positive analysis and value judgments.

A Problem in Welfare Economics

However you interpret the data, a minimum wage increase will harm some people and benefit others. It is inconceivable that such a policy would benefit everyone and harm no one. That leaves us with the standard problem in welfare economics: how to evaluate a policy measure that is not a Pareto improvement. Such an evaluation cannot be done without making extra-economic value judgments.

Imagine a situation where, before establishing a minimum wage, 5 percent of employed workers earned a wage lower than the proposed minimum. Many of those low-wage people would be poor, although virtually all who wanted a job could get one. A certain distribution of income (and utility) would thereby be generated. At the bottom of the distribution, poverty would be a consequence of everybody’s liberty to enter into employment contracts at mutually agreed upon wages.

Under a minimum wage law or an increase in that wage, the distribution of income (and utility) is changed. Some of the poor, as well as some of the non-poor, get higher wages. But it is very likely that others will lose their jobs and become poorer. Whether the new poor number one million, 500,000 or just one, the welfare economics problem is the same. The distribution has been changed by coercive government regulation, “coercive” meaning that the new poor would not comply with the law if they, and their potential employers, were not forced to do so. At the bottom of the distribution, some poverty results from political diktats and bureaucratic enforcement or, in other terms, from the coercive power of some individuals over others.

Answering the liberty-coercion question in the Klein and Dompe survey, one economist commented: “Have you ever had to sit and listen to children crying for want of food?” That comment is the stuff of public relations dreams, but it also illustrates our welfare problem. Let’s pursue the story in those terms.

Suppose that, in contemporary America, some children cry because they are hungry. They are probably few in number, but even if there is just one single hungry child, our welfare calculation must count him. (Note in passing how strange such a situation is when all levels of U.S. government devote 40 percent of their expenditures, or around 15 percent of gross national product, to social transfers and social programs. Perhaps the state is not that great after all.)

If the CBO’s estimates are correct, there probably will be fewer crying children after the minimum wage has been increased to $10.10 because the measure will have pulled 900,000 individuals above the official poverty level. Those individuals include children in poor families, but they are not net crying children. Some previously crying children will be fed, and stop crying; but other children, in families with one or more of the 500,000 new unemployed, will lose their food and start crying or cry all the harder.

The plights of those two groups of children (and their parents) have very different sources. The ones crying before the increase in the minimum wage owed their impoverished condition to the impersonal result of everybody’s equal liberty. The plight of the new crying babes, after the minimum wage enactment or increase, will have been directly caused by the people holding political power.

Some economists argue that there will be little or no disemployment effect because of efficiency gains—in other words, employers will make their employees work harder in order to justify the higher wages. The question is, why didn’t the employers achieve those gains before a new minimum wage was imposed? Perhaps those employers weren’t actively trying to maximize profits until the politicians changed the minimum wage—but that seems unlikely. The more likely reason is that it was not profitable for the employers to do so, perhaps because the low-productivity employees did not want to—or could not—work harder. If, after a minimum wage is imposed, employers find ways to work their employees harder instead of firing them, then you may have fewer crying babies, but you will also have some crying, exploited workers. There is no such thing as a free lunch.

Feeding crying children, even if there is only a small number of them, is a worthy objective. But trying to achieve it through political coercion not only implies making tradeoffs among individuals, but it also increases the burden of regulation and dampens economic growth and prosperity. Econometric estimates by John Dawson and John Seater suggest that if federal regulation had remained at its 1949 level, the average American would now earn more than three times his current income. Even more gains would have been made in the absence of state and local regulation. Admittedly, the minimum wage is just a small part of that flood of government intervention, but it does not help to stop it.

A policymaker and his economic adviser face two alternatives: Will they let some babies cry, and wait for mounting prosperity to feed them? Or will they feed some crying babies now with the food taken from other crying mouths? A related question is how many crying babies on one side justify taking food from the mouth of a baby on the other? (Make sure your numbers are correct!) Those are the value judgments that economists ultimately have to make when they take a policy stance in the minimum wage debate.

I suggest that leaving the minimum wage alone—or better, abandoning it—is both more conducive to prosperity and more easily defensible from a moral viewpoint.

Readings

  • “America’s ‘Top Economists’: What Explains Their Low Opposition to a Minimum Wage Increase in 2013?” by Jason Briggeman. Social Science Research Network Paper no. 2249040, April 11, 2013.
  • “Federal Regulation and Aggregate Economic Growth,” by John W. Dawson and John J. Seater. Journal of Economic Growth, Vol. 18 (2013).
  • “Is There a Free-Market Economist in the House?” by Daniel B. Klein and Charlotta Stern. American Journal of Economics and Sociology, Vol. 66, No. 2 (April 2007).
  • Minimum Wages, by David Neumark and William L. Wascher. MIT Press, 2008.
  • “Minimum Wages: A Poor Way to Reduce Poverty,” by Joseph J. Sabia. Cato Institute Tax and Budget Bulletin no. 70, February 27, 2014.
  • “Raise That Wage,” by Paul Krugman. New York Times, February 17, 2013.
  • “Reasons for Supporting the Minimum Wage: Asking Signatories of the ‘Raise the Minimum Wage’ Statement,” by Daniel B. Klein and Stewart Dompe. Econ Watch Journal, Vol. 4, No. 1 (January 2007).
  • “Revisiting the Minimum Wage-Employment Debate: Throwing Out the Baby with the Bathwater,” by David Neumark, J. M. Ian Salas, and William Wascher. Industrial and Labor Relations Review, forthcoming.
  • The Effects of a Minimum Wage Increase on Employment and Family Income, published by the Congressional Budget Office. Washington, D.C., February 2014.
  • “The Negative Effects of Minimum Wage Laws,” by Mark Wilson. Cato Institute Policy Analysis no. 701, June 21, 2012.
  • “Walmart Needles The Nation about Low-Wage Intern Jobs, Hilarity Ensues,” by Katherine Mangu-Ward. Reason, August 8, 2013.
  • Why Does the Minimum Wage Have No Discernible Effect on Employment? by John Schmitt. Center for Economic and Policy Research, February 2013.