A Minimum Wage Study With Minimum Credibility

If newspaper accounts can be believed, the American Economic Assn. intends to award its top prize to an economist who does not believe in the law of demand, the cornerstone of economic science.

Princeton University economist David Card is apparently due to receive the John Bates Clark Medal next January for research concluding that an increase in the minimum wage does not cause unemployment and may even produce an increase in jobs. President Clinton has based his case for a hike in the minimum wage on Card's conclusion, which is contradicted by economic theory and a vast accumulation of empirical evidence.

Normally, when confronted with research results that challenge a basic tenet of a science, scientists look at the data before they proclaim that the sky is falling. But not Professor Card. Even more curious, neither the reviewers and editors of The American Economic Review, who rushed Card's results to print, nor the economists who selected Card for the medal examined the data that produced the counterintuitive result.

If anyone in the peer review process had peeked at the data, it would have been obvious that something was seriously amiss. The association would have spared itself and Professor Card the embarrassment that comes from awarding a prestigious prize for flawed results.

Card and his Princeton colleague Alan B. Krueger, currently chief economist for Labor Secretary Robert B. Reich, drew their iconoclastic conclusion from data they collected from telephone interviews. They surveyed fast-food establishments in New Jersey and Pennsylvania before and after an 80 cents hike in New Jersey's minimum wage in 1992 to see if employment in the Garden State suffered by comparison.

MEDIA CHORUS. The data they collected are characterized by implausibly large employment variations in the same establishments over the eight-month period between their surveys in February and November, 1992. For example, one Wendy's went from zero full-time and 30 part-time employees (0/30) in February to 35 full-time and 30 part-time employees (35/30) in November. Another Wendy's went from 30/10 to 0/30. One Burger King went from 6.5/20 to 30/25, while another went from 50/35 to 15/18.

The anomalies appear to result from different people answering the before-and-after telephone inquiries: In each case, respondents read into the question their own definitions of employment and the time period for which it was being reported. The replies may have ranged from employment on the shift during which the telephone survey took place to an entire payroll period. Moreover, the survey asked about the number of workers but not the hours worked. The imprecisions in data-collection would have warned any scientist against putting stock in the results.

DUNCE CAPS. Instead, the Clinton Administration and its media allies spoke of this irreparably flawed study as "Nobel prize material," to cite The New York Times, as if the empirical evidence accumulated over decades meant nothing. Democrats used the study at a Feb. 22, 1995, Joint Economic Committee (JEC) hearing to impugn the integrity of witnesses who testified that a higher minimum wage would price some workers out of jobs. Representative Pete Stark (D-Calif.) was so abusive of witnesses who disagreed with Card's findings that he had to be publicly rebuked in front of TV cameras by JEC Vice-Chairman Jim Saxton (R-N.J.).

The media blitz aroused skepticism among many economists, including my fellow columnist Gary S. Becker (BW--Mar. 6). Some of them put the Card-Krueger "study" to the test. Michigan State University Professor David Neumark and Federal Reserve economist William Wascher acquired the actual payroll data from fast-food establishments in New Jersey and Pennsylvania. The payroll data show that fast-food employment did not increase in New Jersey after the minimum-wage increase. Instead, it declined 4.8% relative to the control group in neighboring Pennsylvania. Thus, unsurprisingly, the law of demand still holds.

Since few in the field of economics can believe that the cornerstone of their edifice could be undermined by such flawed data as Card's, serious questions arise about the quality of the review process at The American Economic Review and the rigor of the selection process for the awarding of the John Bates Clark Medal. Political correctness seems to have crept into the inner sanctum of the AEA, discrediting its scholarly journal and debasing its top prize. Unless the association cleans up its act, it can kiss its credibility good-bye.

Before it's here, it's on the Bloomberg Terminal.