Harry J. Holzer is a professor of public policy at Georgetown University. He was chief economist of the Labor Department in the Clinton administration.

Three jurisdictions in the Washington metro area — the District, Montgomery County and Prince George’s County — have voted recently to raise their minimum wages to $11.50 over a three-year period. At least 20 states and some major localities have already increased their minimum wages above the federal level of $7.25. A handful of states have chosen to index their statutory minimum wages to measures of inflation over time.

States and localities are motivated, at least in part, by federal gridlock: President Obama proposed increasing the minimum wage this year, but Washington’s paralysis has eliminated the prospect of a reasonable federal increase anytime soon. Minimum-wage increases seem to combat the disturbingly high earnings inequality that has arisen in recent decades. They seem to help low-wage workers in particular. But these efforts, while often popular, can backfire.

As a labor economist, I generally support modest and periodic increases in the federal minimum wage, along with some state increases above that level. The biggest concern among economists is that imposing pay increases on employers will reduce the hiring of low-wage workers and raise unemployment. But in four decades of research by economists, this appears to be a small or nonexistent effect. And, under some circumstances, indexing the statutory minimum to the rate of inflation is not a bad idea.

Some of the recent actions to raise wages, however, give cause for concern. Many proposed increases are very large and very local. For instance, the increases in the Washington region will raise the minimum wage more than 50 percent in the two Maryland counties and nearly 40 percent in the District, even after adjusting for inflation. This comes on top of the sizable federal increases — about 35 percent after inflation — that occurred between 2007 and 2009.

Increases this large could generate larger employment losses than in the past as employers seek to minimize payroll costs. And the smaller the geographic area in which such increases occur, the greater the risk that employment growth will shift across a municipal or state border as employers relocate to areas with lower minimum wages. It’s noteworthy that the previous state and federal increases on which most research is based generally do not share these characteristics, so that body of research is a less useful guide to the changes in employment that may soon occur.

There are a few other reasons to be cautious about these increases. Most employees working at or near the minimum wage are not the heads of poor households. They are typically either young (up to about 25) or are second-earners, in which case their households do not rely exclusively on them for income. Although Americans might be happy to see all of these workers get a raise, we should perhaps be concerned that any loss of employment might be most concentrated among the small fraction of these workers who are poor adults and who most need the jobs, as some research suggests.

Indexing increases in the minimum wage to a measure of price inflation makes sense in some cases. But indexing should begin closer to the lower end of the ranges recently set. One reason employers might not reduce hiring in response to wage increases is that, traditionally, these increases tend to erode over time with inflation. If it is costly to change production methods in ways that reduce employment, many employers might judge that it is not worth doing so, since the increased wages would be normalized with time anyway. But indexing might change those calculations, because the newly implemented increases would not erode over time. And, since little indexing has been done in this country, there is little research to explain how employers would respond to minimum-wage increases in that context.

The biggest reason to be cautious in raising minimum wages is the weak job market. Employment opportunities nationwide continue to be limited. In a strong or rapidly improving job market — such as that of 1996-98, when a federal minimum increase took effect — there is little reason to worry about such increases reducing employment levels. But the job market’s recovery from the recession has been agonizingly slow, and two of the groups whose wages stand to be increased — the young and least-educated — already have the hardest time finding work. Many are experiencing long-term unemployment from which it is increasingly difficult to escape.

As localities and states consider appropriate minimum-wage increases and where to apply them, they should try to avoid increases that might make it even harder for the youngest or least-educated among us to find work.