Thursday, 26 May 2011

The minimum wage, again

The effects of the minimum wage are not the same across all groups in society. The effects of labour market wage mandates fall disproportionally on minority groups. In the case of the U.S. this means the effects are felt mainly by blacks. Getting good estimates of this effect is difficult due to the lack of sufficiently comprehensive data.

A new study from the Employment Policies Institute, by economists William Even (Miami University) and David Macpherson (Trinity University), overcomes this problem by amassing a dataset from the years 1994 to 2010 that includes over 600,000 data observations —including a robust sample of minority young adults.

There was in the U.S. something of a “natural experiment” created by the substantial interstate variation in the minimum wage between 1994 and 2010. The focus of the study was on 16-to-24 year-old males without a high school diploma, a group that previous studies suggest are particularly susceptible to wage mandates. The authors control for labour market and demographic differences when estimating the effects of wage mandates on minority groups.

The finding of the study suggest that among white males in this group, each 10 percent increase in a federal or state minimum wage decreased employment by 2.5 percent; for Hispanic males, the figure is 1.2 percent. But among black males in this group, each 10 percent increase in the minimum wage decreased employment by 6.5 percent. So employment for all groups falls with the fall being greatest for blacks and least for Hispanics. Looking at the p-values for the estimated elasticities of employment with respect to changes in the minimum wage you will see that the elasticities for men are significantly different from 0 for whites and blacks, but not for Hispanics. So the effect on Hispanic males could be zero.

The substantial disemployment effects that emerge from the data raise an obvious question: Why do black males suffer more harm from wage mandates than their white or Hispanic counterparts? The study argues that blacks are more likely to be employed in eating and drinking places–nearly one out of three black young adults without a high school diploma works in the industry. Businesses in this industry generally have narrow profit margins and are more likely to be adversely impacted by a wage mandate. There’s also substantial variation in regional location, as black young adults are overwhelmingly located in the South and in urban areas. It's also likely that unobserved differences in skill level and job experience play a role. To the extent that these differences are concentrated among young men of a particular race or ethnicity, this group would have the greatest risk of losing jobs when the minimum wage is increased.


Eric Crampton said...

There is a tension here. Restaurants disproportionately hire min wage staff, but also are unable easily to substitute capital for labour as labour price increases. Long run, computer terminals at tables replace wait staff. But not short term.

Paul Walker said...

EC: True. In the long run as the minimum wage rises and the cost of technology falls, physical capital will replace human capital. So the lack of response to the minimum wage in such industries will be temporary.

Tim Worstall said...

And this is where the Card and Kreuger paper went awry.

They looked at the effect on hiring only in the chain fast food restaurants. Not the independents. And the chains will already (yes, I've worked in the industry) be using a higher capital to labour ratio.

So a rise in the min wage might well make employment at the chains rise: at the expense of a fall in the independents.