While looking for something else, I came across this post on the right-wing blog Power Line, attacking a groundbreaking study of the minimum wage by economists David Card and Alan Krueger. From a summary of the study by John Schmitt (pdf link):
The study received widespread attention, not because its findings were unusual (most studies since the mid-1980s have found that moderate increases in the minimum wage have little or no impact on employments), but because its methodology was so careful and convincing. Card and Krueger surveyed 33 1 fast-food restaurants in New Jersey and 79 restaurants in eastern Pennsylvania in the two months before the April 1, 1992 increase. They then reinterviewed the same restaurants about eight months later. The study’s unique design allowed them to use the restaurants in eastern Pennsylvania, where the minimum wage did not increase, as a control group to gauge the response of the New Jersey restaurants.
By comparing changes in employment in the two states (a procedure they referred to as “differences-indifferences”), Card and Krueger were able to estimate the employment effects of the New Jersey increase. Using this approach, they found that the differences in employment growth between the two states were not statistically significant. They concluded, therefore, that the minimum-wage increase did not lower employment in New Jersey.
Tihs is important because the main argument against the minimum wage is that raising the minimum wage hurts low-income workers by raising unemployment among low-wage employers (such as fast food restaurants). If this isn’t true – if small increases in the minimum wage don’t actually hurt employment significantly – then the case against the minimum wage becomes much, much weaker.
(The minimum wage is also especially important to feminists because women are more likely to earn the minimum wage than men. Raising the minimum wage probably has the effect of reducing the wage gap between men and women, and also between some people of color and whites).
To refute Card and Krueger’s study, Power Line quotes a 1995 Reason Magazine article by Benjamin Zychler, which accuses Card and Krueger of using bad data:
Moreover, the Card/Krueger study turns out to have a major flaw: The survey data upon which it depends are lousy.
Suspicious of the Card/Krueger data and findings, the Employment Policies Institute [EPI] gathered the actual payroll records from the Burger King franchises in the Card/Krueger zip codes and compared them to franchises surveyed in those zip codes. The survey data were wildly inconsistent with the payroll records.[...]
So Neumark and Wascher [two right-wing economists working with EPI] reviewed the payroll employment data gathered by EPI. When they applied the payroll data to the same econometric model used by Card and Krueger, they got completely different results. The variation in employment changes declined markedly, and analysis of the new data yields an estimated 4.8-percent decline in New Jersey employment relative to the Pennsylvania sample as a result of the higher minimum wage. Where payroll data could be compared with survey data for specific restaurants, Neumark and Wascher also found numerous errors in the Card/ Krueger data. [...]
In short, using the actual payroll data instead of the survey “guesstimates” effectively refutes the Card/Krueger findings yielded by the New Jersey/Pennsylvania “natural experiment.”
That seems pretty damning, doesn’t it? The problem is, Power Line has quoted an article which was written when the story was only halfway through. It turns out that there is “lousy data” here – but the lousy data wasn’t Card and Krueger’s.
The attack on Card and Krueger relies on payroll data gathered by EPI – an anti-minimum-wage think tank funded to a significant extent by the fast food industry. Even Neumark and Wascher had to admit that the EPI was not a neutral party. Normally this sort of problem is solved by transparency; researchers make their data and methods available to the public (or at least to other researchers) to prove that they haven’t cooked the data.
However, EPI has refused to do that. Essentially, the attack on Card and Krueger consists of a biased think tank saying “we have data that proves Card and Krueger are wrong. But it’s super-secret data that we won’t let anyone check, so you’ll just have to take our word for it.”
To get around the “super-secret data” problem, the economists the EPI was working with, Neumark and Wascher, gathered their own data. The results didn’t make the EPI’s secret data look good. From The American Prospect:
When Neumark and Wascher analyzed the Employment Policies Institute sample and their own sample separately, a funny thing happened. The slightly expanded Employment Policies Institute sample indicated that the minimum wage did have a significant negative impact on employment in New Jersey (at least in one of the two statistical tests). But Neumark and Wascher’s own data found no statistical difference in employment growth in the two states. Quite unintentionally, Neumark and Wascher had vindicated Card and Krueger.
Neumark and Wascher scrambled to explain their results. They maintained that the combined data provided the best basis for determining the employment effects. But their position was undermined by major differences in the two samples. The Employment Policies Institute restaurants showed much more uniform employment changes than those in the Neumark and Wascher sample. In fact, basic statistical tests demonstrated convincingly that the Neumark and Wascher sample showed so much more variation in employment changes across restaurants that it was highly unlikely that the two samples were chosen randomly from the same population of restaurants.
In response to the controversy, in 1997 the Federal Bureau of Labor Statistics made relevant payroll data for New Jersey and Pennsylvania available to researchers. The BLS payroll data is the most authoritative data available; it cannot credibly be accused of being gathered in a biased manner, or of being a “guesstimate.” The results? Not surprisingly, the actual payroll data contradicted the “secret methodology” data right-wingers prefer; raising the minimum wage in New Jersey did not lead to decreased employment.
Furthermore, researchers were now able to examine BLS data from 1996 – when an increase in the national minimum wage caused the minimum wage to go up in Pennsylvania, but not in New Jersey (because NJ had already raised their minimum in 1992). Once again, the data showed no increase in unemployment in the state raising its minimum wage, compared to the state that didn’t.
In short, Power Line’s attempted refutation of Card and Krueger’s study is based on a fraud. The best evidence available shows that there are no significant effects on employment caused by reasonable increases in the minimum wage; and the only way the right has been able to refute this is by using data so embarrassing and dishonest that they don’t dare let other researchers examine it.
The right’s cooked “super secret” data aside, the best evidence available shows that low-wage workers benefit from raising the minimum wage. (Well, duh.)
- Trends in the Wage Gap (wage gap series, part 2)
- Wage Gap Myth: The pay gap only exists because men work so many more hours than women. (wage gap series, part 4)
- Myth: The Wage Gap is Caused by Men's Higher Pay for Dangerous Jobs (wage gap series, part 10)
- The Wage Gap Series, so far
- Some Evidence of Discrimination (wage gap series, part 9)