Understanding the CBO on Jobs and the Minimum Wage, and the Real Reason GOP Hates It

Understanding the CBO on Jobs and the Minimum Wage, and the Real Reason GOP Hates It

Last week, the Congressional Budget Office released its estimates on the president's proposal to raise the federal minimum wage to $10.10 an hour. Among its key findings:

  • 900,000 Americans would be lifted out of poverty.
  • 16.5 million low-wage Americans would see an increase in their income.
  • Income of low-wage workers would increase a collective $31 billion.
  • Employment would decrease by 500,000 compared to current law estimates.

Republicans and other opponents of the minimum wage increase grabbed onto on the last statistic, doing what they do best: scaring Americans that paying the poorest workers more would cost jobs ... because job creators, or something. The White House and the President's Council of Economic Advisers disputed the number, however, saying increase in the minimum wage would have no effect on employment, while Democrats seized on near 1 million people being lifted out of poverty.

While those are the headline numbers and the headline arguments, our role at TPV has often been to explain rather than entertain. I plan to do that today by showing that by CBO's own admission, the estimate on the effect on employment is unreliable, and that far better estimates exist of peer reviewed studies.

The CBO explains its methodology in its report. The CBO's data is based on a plethora of studies that estimate effects of the minimum wage on employment, and the prevailing but somewhat misleading average that a 10% increase in the minimum wage reduces employment among low-wage workers (mostly teenagers) by between 1 and 3 percent, and while most of the studies referred to by the report's reference section CBO to its credit kept its estimate on the lower end.

Still, that data is outdated - and that is not the CBO's fault. Consortia of studies conducted in this century find that there is essentially no effect on employment from a minimum wage increase. One such consortium of studies plot the data points on the effects of minimum wage increase on employment. Notice the concentration of the data points right around the zero mark:

What you see above is a collection of the conclusions of 64 studies between 1964 and 2007 that peer-reviewed research considers to be the best estimates of the respective researchers representing 1,492 employment estimates. The economists conducting the "study of the studies", Hristos Doucouliagos and T. D. Stanley, note


Two scenarios are consistent with this empirical research record. First, minimum wages may simply have no effect on employment... Second, minimum-wage effects might exist, but they may be too difficult to detect and/or are very small.

Perhaps it is because of this lack of discernible effect of the minimum wage on employment that even the CBO noted a rather high degree of uncertainty in


 estimate. The CBO explicitly states that their estimated range of almost zero reduction in employment to as high as a million less jobs can only be given a two-thirds confidence level - meaning that the

CBO is projects that there is a 67% probability that the job losses will be between very low to 1 million.

67% probability is a lot, right? Wrong. In statistical analysis, a 67% confidence level is rarely relevant, let alone significant. Estimates that fall at a 67% confidence level is rarely statistically reliable data. Most statistical studies, in order to be considered valid, need to have a 95% confidence level.

Those who know a little bit about statistics are probably thinking that the CBO could simply give a wider estimate - say from 1 million added jobs to 1.8 million lost jobs (just my guesses since the CBO did not release the entire range of their data) - and be 95% confident of that (in other words, one could say that there is a 95% chance that raising the minimum wage to $10.10 will have an impact on employment in the range of increasing employment by 1 million to reducing employment by 1.8 million). That is correct, and that is exactly the point. Getting to a 95% probability model on this would have include such a large range - and the range would include zero - that essentially no conclusion could be drawn on the effects of a minimum wage hike on employment.

And that, of course, is the conclusion that economists that have reviewed thousands of data points on this have come to: there is no conclusive evidence that increasing the minimum wage reduces employment, or has any effect on employment levels at all. But the real part of the CBO report that is concerning to conservatives isn't the weak part about job losses. It is this part:

That is the chart in the CBO report showing that an increase in the minimum wage would raise the aggregate real family income for all families for those at the very top. It shows that the aggregate income of all families making more than 150,000 in today's dollars (assuming a family of 4) - or more than six times the federal poverty threshold - would drop by roughly $17 billion, not from today's levels but from what is is expected to be in 2016. Of course, that drop isn't going to mostly come from families of 4 making $150,000 but from the super-wealthy who make most of the income.

And therein lies the Republican concern about the minimum wage. It isn't the loss of jobs. It is the loss of income that millionaires and billionaires would suffer because, Heaven forbid, they'd have to pay the workers responsible for all of their profits a fair wage.

Like what you read? Chip in, keep us going.

Of subhuman mongrels and keeping away the gay

The Problem with "Earned Benefits"