Minimum-wage laws are a traditional bête noire of free-market economists because a government-mandated wage floor interferes with freedom of contract between employers and employees. Moreover, a law that increases wages is expected to reduce employment, because the marginal product of some workers, though equal to or above the free-market wage, may be below the minimum wage.
Despite this, there is very little evidence that minimum wage laws have a significant disemployment effect. Studies have found, for example, that when one state raises its minimum wage, and an adjacent state does not, a comparison of employment in adjacent counties consisting of one in one state and one in the other state reveals no drop in employment in the county located in the state that increased its minimum wage.
One reason the federal minimum wage seems not to have a significant disemployment effect is that it has been falling in real terms for many years. In current (2013) dollars it was $10.60 in 1968; today it is $7.25. That is a poverty-level wage, as it translates into an annual income (assuming a worker works 50 40-hour weeks in a year) of only $14,500. And I am not aware that the real (that is, inflation-adjusted) reduction in the minimum wage resulted in increased employment.
With the federal minimum wage so low, it strains credulity to think that increasing it would cause significant disemployment (though this depends of course on how much it’s increased—I’ll come back to that critical issue shortly). Many persons who can’t command a wage higher than the federal minimum will prefer not to work, but instead to live on public benefits, engage in petty crime, switch to household work, engage in protracted search for better employment (and not work during the search), or even beg. An increase in the minimum wage will increase their incentive to seek productive work. In addition, forced to pay a higher wage, employers may invest more in their workers’ human capital—may give them better training, for example, to make them more productive and thus worth their higher wage to the employer. If the percentage increase in the minimum wage exceeds the percentage drop in employment that results from that increase, the aggregate income of persons at the bottom of the income distribution will rise, reducing inequality of income—a growing phenomenon, and serious social problem, in the United States. Notice too that to the extent that the minimum wage deflects workers to dependence on public benefits, or to crime, it imposes costs on the society at large that may offset the costs resulting from any disemployment caused by an increase in the mininum wage, provided that disemployment effect is modest.
That proviso is critical, and a reminder that at some level an increase in the minimum wage is bound to produce significant disemployment effects, causing some businesses to shrink and others to automate. There is much talk about increasing the minimum wage to $10 or $13 an hour. It seems both imprudent and unnecessary to consider such steep, sudden jumps. I would favor increasing the federal minimum wage by 20 percent, to $8.70 an hour. That would yield a minimum-wage worker an annual income (assuming he or she worked 2000 hours per year) of $17,400—still very modest; but if he disemployment effect proves to be slight, as I would guess it would be, a furher increase could be considered. At the very least, the 20 percent increase would yield valuable information on the elasticity of unemployment to changes in the minimum wage.
If the disemployment effect of a 20 percent increase in the federal minimum wage were slight, the redistributive effect of the increase would in my opinion justify it as a measure for reducing the growing inequality of income in the United States.