A lot of people are asking me about this David Cay Johnston article, the title of which bizarrely intimates that the EITC (earned income tax credit) raises poverty. It does not. In fact, this robust, refundable tax credit for low-wage workers (refundable means that if their tax credit exceeds their tax liability, the IRS sends workers a check for the difference) can add more than $5K to the earnings of low-income, working families with kids. As work by my CBPP colleagues shows in great detail, it reduces poverty very significantly.
So where is David CJ—a stalwart fighter for good causes; his book Perfectly Legal on the extent to which the rich game the tax system is perhaps the most important book on the subject—coming from?
He starts from a similar place as I do here, from a piece I wrote over a decade ago, based on the simple point that a policy that induces greater labor supply has the potential to place downward on wages for the affected group.
That piece was part of the inspiration for a then-young (ha!) labor economist named Jesse Rothstein to look into the impact (here’s his latest paper on it). He finds some of the leakage Johnston talks about, but about three-quarters of the water reaches the fire, and Jesse, now a Berkeley prof and a young lion in the profession, views the wage credit as highly cost effective.
I asked him to take a look at the Johnston piece; here’s what he said:
I think there are a number of problems with the analysis. First, Johnston emphasizes the potential for individuals to reduce their work hours in order to qualify for larger Earned Income Credits. A great deal of research has conclusively demonstrated that this kind of “intensive margin” (changes in hours or weeks worked among those who would work in any case) response has been minimal, at best.The overwhelming effect of the EITC has been to increase the share of low-skill parents who work at all, shifting them out of poverty.
A second problem is that the OECD report that he cites as evidence of increased hours of work among low-wage Americans actually focuses on individuals with low annual earnings, not those with low hourly wages. In other words, the change in hours is driven by rising wage inequality, which ensures that increasing shares of the bottom of the annual earnings distribution are low-wage workers with high hours (instead of higher-skilled workers with part-time or part-year schedules).
These problems aside, Johnston has identified a real concern. U.S. antipoverty policy has taken on an increasingly prescriptive tone in recent decades, putting ever more pressure on low-skill workers to find jobs, even at sub-poverty wages and with poor working conditions. At the same time, technological and other changes have eroded the demand for such workers — there is no shortage of low-skill workers, but rather a shortage of jobs for them to take. These trends do not sit well with each other, as every former welfare recipient who is pushed into the labor market by welfare reform and the EITC serves to bid down the wage for all of the other low-skill workers. To reduce poverty, we need to tackle rising wage inequality directly. It isn’t enough simply to encourage work; in President Clinton’s phrase, we need to make work pay.
What Jesse said. Not what David said.