Minimum wage increases in a recessionary environment
by John Addison, McKinley Blackburn and Chad Cotti
What happens when the minimum wage is increased? Does it harm those in low- wage sectors and low wage earners such as teenagers or, as much recent research seems to indicate, leave their employment largely unaffected while improving their earnings? The large number of factors at play in the employment relationship underscored by minimum wage increases that are often modest in magnitude and implemented in good economic times has complicated the task of measurement. Now a team of academics have addressed the question: ‘do seemingly large minimum wage increases in an environment of deep recession produce clearer evidence of disemployment than is often observed in the modern minimum wage research?’
There is no consensus in the literature. Early studies suggested that increasing the minimum wage had a negative impact upon employment: for example, a ballpark estimate was that a 10% increase in the minimum wage reduced teenage employment by 3%. Gradually, estimates of this disemployment effect were downplayed with some researchers even suggesting that minimum wage hikes might be good for employment as well as wages. These results have in turn given way to disagreement as to whether employment measurably falls in the wake of increases in the minimum wage. Addison and his colleagues argue that if minimum wages have adverse effects these should be most discernible when the economy is doing badly and in circumstances where minimum wage increases are above normal. This is the very economic scenario in which they conduct their investigation of the most recent increases in the U.S. minimum wage.
Modelling the Problem
The research team chose a modelling approach, drawing upon three data sources and exploiting geographic variation in the setting of minimum wages across U.S. states. Not only do states with minimum wage laws pay wages that exceed the federal minimum wage but, critically, their employment trends also differ independently of their higher minimum wages. An added complication is that these geographic specific trends can be sensitive to the particular time period examined. The authors therefore deliberately select a short timescale, namely 2005-2010, where the subperiod 2007-2010 is an interval of recession. They focus on the employment impact of the minimum wage when unemployment is at its average level during the pre-recession years and whether this effect varies with the deviation of the unemployment rate around the former value. The model is applied to each of the three data sources separately: two of them having been deployed in prior work while the third is used for the first time in minimum wage research. Given the datasets and that minimum wages should have most impact on low-wage sectors and among low-wage earners the team focused upon the restaurant sector and teenagers (16- to 19-year-olds).
Answering the Minimum Wage Problem
So what was the result of the modelling? At the broadest level, minimum wages had no statistically significant effect on employment in the restaurant sector as a whole, although in one case a small disemployment effect was found for full-service restaurants, and a more substantial negative effect for teenagers was reported for one of the two datasets allowing investigation of this group. But these results are for specifications that abstract from the recession per se. When the minimum wage effect is allowed to vary with the unemployment rate, much sharper disemployment effects are evident. First, in states with high unemployment rates there is evidence of a well determined negative employment effect for the restaurant sector as whole and also for fast-food restaurants - which result the authors conjecture reflects this subsector’s more limited ability to pass on wage increases. Yet the magnitudes are still small. For teenagers, however, there is much stronger evidence that the recent recession enhanced the adverse effect of the minimum wage. Here the estimated responsiveness of employment to the wage during the recession years is at the higher range of results published in modern research. Further examination of the implications of recession for such high risk groups is therefore required.
[The full paper was published in Labour Economics, vol. 23, pp. 30-39.]