Imagine an antipoverty program with the following elements: a value-added tax in which the effective rate increases as family income declines. The tax revenue is distributed to families regardless of their income. Families below twice the poverty level get only one-third of the revenue, with only half of this amount going to families with children.
Most Americans wouldn’t cheer this program, nor would most political leaders champion it. Yet that is what happens when Congress raises the minimum wage.
In a peer-reviewed study, “How Effective Is the Minimum Wage at Supporting the Poor?” (forthcoming in the Journal of Political Economy), I analyzed who won and who lost after Congress raised the minimum wage in 1996 to $5.15 from $4.25, a raise that occurred in phases over the period 1996-97. That would be comparable to raising the current minimum wage of $7.25 to nearly $8.80. The results show the failure of minimum-wage hikes as an antipoverty policy.
To be sure, companies on their own—such as Wal-Mart last week—do raise the wages of their lowest-paid workers, typically when it is necessary to retain a stable, productive workforce. But this isn’t the same as a government-mandated, economy wide raise. Still, most Americans favor such mandated increases because they believe it helps poor workers support their families.
One problem is that only about 5% of families have children and are supported by low-wage earnings; another is that higher minimum wages cause some workers to lose their jobs. Advocates of a higher minimum wage argue that the number of workers who gain far exceeds those who lose. Whatever the credibility of this calculus, there is yet another problem: If someone’s income is arbitrarily increased thanks to a legislatively mandated wage increase, someone else must pay for it.
Since economic evidence indicates that higher minimum wages don’t significantly affect employers’ profit rates, advocates instead say that employers will pass on these increased labor costs by raising the prices of their goods and services—and that “society,” or more affluent consumers, will pay these costs.
But will low-income families earn more from an increase in the minimum wage than they will pay as consumers of the now higher-priced goods? My research strongly suggests that they won’t.
The first step in understanding why they won’t is to recognize that minimum-wage workers are typically not in low-income families; instead they are dispersed evenly among families rich, middle-class and poor. About one in five families in the bottom fifth of the income distribution had a minimum-wage worker affected by the 1996 increase, the same share as for families in the top fifth.
Virtually as much of the additional earnings of minimum-wage workers went to the highest-income families as to the lowest. Moreover, only about $1 in $5 of the addition went to families with children supported by low-wage earnings. As many economists already have noted, raising the minimum wage is at best a scattershot approach to raising the income of poor families.
The second step is to consider who actually bears the burden of higher labor costs that are passed on through higher prices of goods and services.
My analysis, using the Bureau of Labor Statistics’ Consumer Expenditure Survey, showed that the 1996 minimum-wage hike raised prices on a broad variety of goods and services. Food purchased outside of the home bore the largest share of the increased consumption costs, accounting for 21% with an average price increase of slightly less than of 2%; the next highest shares were around 10% for such commodities as retail services, groceries and household personal services.
Overall, the extra costs attributable to higher prices equaled 0.63% of the nondurable goods purchased by the poorest fifth of families and 0.52% of the goods purchased by the top fifth—with the percentage falling as the income level rose.
The higher prices, in other words, resembled a regressive value-added, or sales, tax, with rates rising the lower a family’s income. This is sharply contrary to normal tax policy. A typical state sales tax has a uniform rate—but with necessities such as food excluded, and this exclusion (which exists as well in countries with a value-added tax) is adopted expressly to lower the effective tax rate on consumption by people with lower incomes.
My analysis concludes that more poor families were losers than winners from the 1996 hike in the minimum wage. Nearly one in five low-income families benefited, but all low-income families paid for the increase through higher prices.
Consider a McDonald’s restaurant, often cited as ground zero in minimum wage debates. To cover costs of a mandated increase in the earnings of McDonald’s lowest-paid workers, customers pay more for the company’s food. The distributional question becomes: Which group comes from the least well-off families: McDonald’s customers or its lowest-paid workers? Economy-wide evidence shows that the customers disproportionately come from low-income families.
Mr. MaCurdy is an economics professor at Stanford University and a senior fellow at the Hoover Institution.