Minimum wages in 20 states are set to increase Thursday, raising pay for some 3.1 million workers, according to an estimate by the liberal Economic Policy Institute. That includes both those who are working at the minimum wage and those who make slightly more, but are likely to receive a raise anyway because their employers want to compensate them above minimum wage. Nine of those 20 states are increasing wages automatically, as they do every year, but the other 11 increases are a result of decisions by voters and lawmakers that take effect Jan. 1. That makes New Year's Day more than just an arbitrary holiday determined by the calendar for low-wage workers.

Megan McArdle, however, wonders whether New Year's really is a cause for celebration for them. After all, economic theory predicts that any increase in the minimum wage will put some low-wage workers out of a job. Whether the theory applies in practice has been long debated, but it would be a shame if a policy intended to help the worst off in fact made their lives harder.

Here's McArdle:

The people confidently proclaiming their ability to see the future are often what I like to call "one-study wonders": people who have gotten their hands on a single study that confirms what they already believe (or would very much like to) and then proceed to wave it around while ignoring the rest of the vast, conflicting, suggestive but hardly definitive economic literature on the subject.

It's true that a single study proves little and that the research on minimum wage is extensive, but it's not as though that research doesn't tell us anything. The best studies show that raising the minimum wage may put some people out of work, but that the effect is "between small and vanishingly small," as economists Dale Belman and Paul Wolfson wrote in a recent review. If these studies are correct, the moderate increases in minimum wages scheduled this week (South Dakota's hike of $1.25 an hour is the largest) won't have a major effect on employment.

It's also true that there are a lot of conflicting studies showing that the minimum wage creates or kills a large number of jobs, but that's partly because journals have a preference for studies with dramatic results, Belman and Wolfson write. Editors tend to accept and publish results that show that the minimum wage has large effects, even if the estimates aren't as precise, as shown in the chart below.

All in all, an increase in the minimum wage would likely help more than it would hurt, the Congressional Budget Office has concluded. Increasing the federal minimum wage to $10.10 an hour would raise total income across all groups by $2 billion. Although some people would likely lose their jobs, by the end of 2016, 900,000 fewer people would be living below the poverty line. The middle class would also enjoy a slight increase in income on average, while the very rich would earn less.