Liveblogging World War II: February 25, 1944
Noted for Your Morning Procrastination for February 25, 2014

Over at the Washington Center for Equitable Growth : What Effect Will a Minimum Wage Increase Have?: Tuesday Focus: February 25, 2014

What Effect Will a Minimum Wage Increase Have?: Tuesday Focus: February 25, 2014: First of all, it seems very clear to me that whatever disemployment effects a minimum wage increase would have are swamped for any reasonable greatest-good-of-the-greatest number calculation by the positive effects on income distribution and on the effectiveness of the EITC as an anti-poverty program that the minimum wage increase would have.

Second, I agree with my colleague Michael Reich that the CBO’s relatively high–but still low in both absolute and relative to the size of the economy terms, for it is high only relative to other studies and to the consensus view around here–estimate of the disemployment effect of the minimum wage increase is puzzling. READ MORE

It seems to me to be more likely to be wrong than right:

Michael Reich: No, a Minimum-Wage Boost Won’t Kill Jobs: "The CBO examines the effects of a bill to raise the federal minimum wage from its current $7.25 to $10.10 by 2016, an increase of 39 percent.

Based on its own research, the CBO report estimates some stunning benefits: about 23 million people would receive pay increases and 900,000 people would be lifted above the federal poverty level. Pretty good news for a White House that has been touting the virtues of a minimum-wage increase.... The CBO also reported a definite cost: Employment would fall by 500,000.... These estimated employment losses have become the subject of considerable disagreement among wonks and economists. Jason Furman, chair of the White House’s Council of Economic Advisers, and his colleague Betsey Stevenson argue that the CBO could easily have picked a much lower job-loss number, including one that would be so small as to be negligible. Conservative groups... have said that the CBO’s job loss figures are consistent with their own estimates. So who’s right?...

The CBO report’s appendix describes, but not very clearly, how it estimated the likely job losses.... It reviewed a number of recent research papers... [and] constructed its own “synthesis” estimate.... Furman and Stevenson responded that CBO’s chosen estimate is much too high and does not reflect the consensus of the research literature. Douglas Elmendorf, the CBO’s director, has replied that it does.... As it happens, the [CBO's] 0.075 figure for teens lies halfway between the two most widely cited estimates in the research literature. One... by David Neumark and William Wascher.... The other... from more recent studies, authored primarily by me and Arindrajit Dube....

The CBO’s apparent strategy—to adopt the average of these studies—might seem reasonable.... [But] my co-authors and I show that comparisons between states that have increased minimum wages are not a random sample of all the states. They differ from the other states in ways that affect their low-wage employment trends, but which are unrelated to minimum-wage policy. We then show that teen employment was already growing more slowly in the higher minimum-wage states than in the other states, even as much as two years before a minimum wage increase is introduced. In other words, higher minimum wages are correlated with less employment growth, but it doesn’t follow that the higher minimum wages caused it. The negative correlation implies that comparing employment trends in states that are far apart—essentially Neumark and Wascher’s method—will generate misleading negative employment effects....

We conclude, and many other labor economists agree, that our studies invalidate the previous approach used in many studies by Neumark and Wascher and others. It makes no sense to take an average between a rigorous study and one that has been shown to be flawed. That is why Furman and Stevenson’s response to the CBO report cited my research with Dube as reflecting the current research consensus and why former CEA Chair Alan Krueger, now back at Princeton, called it “particularly compelling.” If the CBO had done more homework, they would have agreed.

But I would not say that I have delved deeply enough into the points at issue between Reich and Dube and Newmark and Wachter to understand why Newmark and Wachter reject Reich and Dube's methods for controlling for employment-growth trends driven by factors other than the minimum wage (i.e., proximity to Mexico, air conditioning, etc.).

However, there is somebody who has delved deeply enough: Daniel Kuehn of American University. So I would urge anyone wanting to dig deeper to read him: he is very smart, very thoughtful, and overwhelmingly interested in understanding the world and marking his beliefs to market:


Daniel Kuehn: Facts & other stubborn things: New Upjohn book on the minimum wage: "I noticed an announcement of a forthcoming meta-analysis on the minimum wage.

It looks great. Here's the summary (it's just on the main page - doesn't even have an Upjohn Press page yet:

Institute to publish comprehensive study on the effects of the minimum wage: In his State of the Union address, President Obama called on Congress to act on a proposal to increase the minimum wage to $10.10 per hour. Raising wages, he said, is "Good for the economy; it's good for America." To contribute to the discussion the Upjohn Institute is providing—prior to publication this spring—the findings of an exhaustive new study of the wage, employment, and poverty impacts of the minimum wage. Supported by the Upjohn Institute, Dale Belman and Paul Wolfson, authors of the forthcoming book titled What Does the Minimum Wage Do? (Upjohn Press), performed a meta-analysis on scores of published studies examining the effects of the minimum wage.  The studies were based on data mostly from the United States but also from other countries. The authors’ comprehensive analytical efforts allow them to conclude the following:

  • Moderate increases in the minimum wage, characteristic of the United States over the last half of the twentieth century, have the effect that was intended by original supporters of the legislation; increasing the minimum wage substantially increases the earnings of those at the bottom of the income distribution and reduces wage inequality.

  • Negative effects on employment resulting from increases in the minimum wage were too small to be statistically detectable in the meta-analsis. The authors conclude thatemployment effects are too modest to have meaningful consequences for public policy in the dynamically changing labor markets of the United States.

  • Evidence of positive spillover effects on the wages of those earning slightly more than the new minimum wage is mixed, but generally supports their existence, particularly for women.

The bottom line for Belman and Wolfson is that the minimum wage should be seen as one of a set of public policy tools aimed at improving the standard of living of the less well-off, and moderate increases in the minimum wage would likely aid low income individuals and families with acceptable costs to the nation. What Does the Minimum Wage Do? is to be published in the spring of 2014. Click here to request an email letting you know as soon as the book is available, or pre-order your copy here.

Daniel Kuehn: Facts & other stubborn things: Brief, and entirely unsatisfying post on Ozimek and the minimum wage: "I look at Ozimek's post on DLR here.

It's very good, although I don't know how much it resolves. It goes over a lot of the time-trend issues we've been over here.... The difficult task isn't identifying a trend change (which requires a flow change and results in a level change), the difficulty is identifying the right counterfactual trend. It ultimately boils down to whether we think the time-trends are appropriate or not and if there's an obvious econometric test for it I'm not sure what it is. Time-tends may be wrong, but Occam's razor seems to suggest we should include them (as in DLR). Spatially heterogeneous time trends seem more reasonable than just the right circumstances that would actually introduce bias by including time trends.... I feel like many of the same points are made here that I made the other day, namely: (1) time trends should help to reduce bias in most cases, but (2) you can imagine specific scenarios where the opposite would be the case.

Entirely unsatisfying, eh?

I still think DLR offers the most sensible default - just at first appearances. That doesn't mean there isn't something else going on, but I think it needs to be demonstrated.