How does the minimum wage affect employment?

Economists these days tend to think that a minimum wage can be a useful policy tool for reducing poverty while leaving employment numbers intact, as long as the policy is well designed. This column looks at recent evidence from a new perspective and claims that much of the recent research suggesting that minimum wage hikes barely reduce the number of jobs in the short run should be taken with caution. The longer-run disemployment effects are potentially large.

When people who work full time still live in poverty, there is a strong urge among the public to demand that employers pay a minimum or ‘living’ wage. Consequently, such legislation has now been in force in the US and UK for over 75 and 100 years, respectively, and nearly every employee in the developed world today is guaranteed a minimum wage.

However, for families that rely on the earnings of a single full-time minimum wage worker, the current federal rate in the US ($7.25) and UK (£6.70 for those aged 21 and older, and lower levels for those under 21) are not sufficient to lift a family of four out of poverty. For this reason, a number of proposals have passed that will significantly raise the minimum wage in both countries. The UK’s minimum wage will rise to £7.20 in April 2016 for those aged 21 and older, with more increases thereafter. In the US, federal minimum wage legislation failed to pass Congress in 2013 and 2014 but several US states have raised their minimum wage. And in a new development, many US cities have passed minimum wage laws, with some going as high as $15 per hour over the next few years.

Economists and the minimum wage debate

While recognising that the alleviation of poverty is a primary goal of policy, economists almost uniformly opposed minimum wage legislation prior to the early 1990s. The standard rationale was that raising wages, while improving the lot of many low-wage workers, also led to lower employment, potentially causing significant earnings losses among those who lost work opportunities. This argument seemed to be supported by the best research, summarised by Brown et al. (1982), which consistently found a statistically and economically significant loss in employment after a minimum-wage hike.

However, research in the early 1990s created a heated debate about the size and the direction of the employment response.[1] At a minimum, that research argued that the disemployment effect of minimum wages was smaller than what conventional economic theory suggested. Advocates of raising the minimum wage often refer to these findings to support the minimum wage as a poverty-fighting tool with few negative side effects.

Much of the newer US research exploits the fact that states often raise the minimum wage above the Federal minimum. Therefore, employment effects can be uncovered by comparing states that raised the minimum wage with states that did not, both before and after the hike. More recently, some researchers (e.g. Dube et al. 2010) exploit differences between counties sitting on each side of a state border where one state raised their minimum wage and another did not. Many, but certainly not all, studies that use these approaches deliver small disemployment effects. An important exception is Clemens and Wither (2015), although their study is focused on the Great Recession and its immediate aftermath, when labour market conditions were historically dire.

Distinguishing the short run from the long run

A great majority of studies focus on estimating the employment response in the months, or at most a year, after a minimum-wage hike. This is for good reason. When a US state raises the minimum wage, other states typically follow soon thereafter. Moreover, these minimum wage studies rely on the assumption that the only factor driving differences in employment growth across areas that hiked their minimum wage and others that did not is the minimum wage itself. This assumption likely becomes more tenuous over time. For these reasons, it is extremely difficult to study the longer-run effects of the minimum wage.

However, it might take a while for an establishment to react to higher legislated wages, especially in cases where the firm’s production processes are hard to change. Thus replacing workers with other factors of production may take a while. In such cases, the long-run disemployment effect might be bigger than the short-run effect often estimated in the literature.

Our findings

Our recent research (Aaronson et al. 2015, Sorkin 2015) presents new evidence on how the restaurant industry, the largest US employer of low-wage labour, responds to minimum-wage hikes. We document three new findings. First, fast food restaurants are more likely to shut-down (exit) and open up (enter) after a minimum-wage hike. Second, the rise in entry is higher among large chains, which use less labour than the rest of the fast food industry. Third, there is no change in employment among fast food restaurants that remain open throughout (neither exit nor enter). Together, these results imply a small decline in employment two years after a minimum-wage hike.

Figure 1. Share of firms exiting and entering per year after 10% minimum-wage hike

In addition, because restaurants that remain open cannot change how mechanised they are, the model predicts a small short-run employment response, consistent with the previous literature.

Finally, our model predicts that restaurant meal prices rise with the minimum wage, since even those restaurants that are capital-intensive still employ workers at or near the minimum wage. Previous work has shown that 100% of the higher labour costs of the minimum wage are pushed on to consumers in the form of higher prices (Aaronson 2001, Aaronson et al. 2008).

Putty-clay and the long-run effects of the minimum wage

Importantly, in the putty-clay model, as labour-intensive restaurants are slowly replaced with more capital intensive restaurants, the disemployment effect of the minimum wage grows over time. To be clear, restaurants only shut down when they are no longer productive. Indeed, our results suggest that a typical minimum-wage hike causes an older fast food restaurant to shut down one year earlier than it otherwise would have. Since fast food restaurants last for many years, the employment adjustment process is likely to be slow. Figure 2 presents model predicted consumer price, sales, and employment responses to minimum-wage hikes. Note that the full price and sales responses is immediate but the employment response is spread over 17 years. In particular, as shown in Table 1, the short run disemployment elasticity of the minimum wage is -0.08, consistent with much of the evidence in the literature, but the long-run disemployment effect is -0.40. In words, a 10% minimum-wage hike leads to a 0.8% decline in restaurant employment within a year but a 4% decline over the long-run.

Figure 2. Price, sales, and employment following 10% minimum-wage hike, dates measured in years

Conclusion

Our results suggesting a long-run disemployment effect that is five times larger than the short-run effect is based on an economic model and is not directly measured in the data. However, the key mechanism behind the model is supported by empirical evidence. In particular, we find an increase in restaurant exit and entry within two years of minimum-wage hikes and suggestive evidence that the spike in entry occurs primarily from chain restaurants that are relatively more capital intensive than the rest of the fast food industry. Therefore, we believe that that much of the recent research suggesting that minimum-wage hikes barely reduce the number of jobs in the short run should be taken with caution. The longer-run disemployment effects are potentially large.

References

Aaronson, D (2001), “Price Pass-Through and the Minimum Wage", Review of Economics and Statistics 83(1):158-169.

Aaronson, D and E French (2007), “Product Market Evidence on the Employment Effects of the Minimum Wage", Journal of Labor Economics 25(1): 167-200.

Aaronson, D, E French, and J MacDonald (2008), “The minimum wage, restaurant prices, and labor market structure", Journal of Human Resources 43(3): 688-720.

Aaronson, D, E French, and I Sorkin (2015), “Industry Dynamics and the Minimum Wage: A Putty-Clay Approach”, unpublished manuscript.

Brown, C, C Gilroy, and A Kohen (1982), “The Effect of the Minimum Wage on Employment and Unemployment”, Journal of Economic Literature 20(2): 487-528.

Card, D and A B Krueger (1995), Myth and Measurement: The New Economics of the Minimum Wage, Princeton, NJ: Princeton University Press.

Dickens, R, S Machin, and A Manning (1999), “The Effects of Minimum Wages on Employment: Theory and Evidence from Britain”, Journal of Labor Economics 17(1): 1-22.

Dube, A, T William Lester, and M Reich (2010), “Minimum Wage Effects Across State Borders: Estimates Using Contiguous Counties", Review of Economics and Statistics 92 (4):945-964.

Neumark, D and W L Wascher (2008), Minimum Wages Cambridge, MA: MIT Press.

Kennan, J, “Minimum Wage Regulation”, The New Palgrave Dictionary of Economics and the Law, 1998.

Sorkin, I (2015), “Are There Long-Run Effects of the Minimum Wage?", Review of Economic Dynamics 18(2): 306-333.

Endnotes

1 For excellent summaries of the empirical literature, see Card and Krueger (1995), Kennan (199x), and Neumark and Wascher (2008). British evidence includes Dickens et al. (1999). See Aaronson and French (2007) for a discussion of the disemployment effects implied by the benchmark model of labor demand.