«∗ Daron Acemoglu J¨rn-Steﬀen Pischke o MIT LSE First Version: April 1998 This Version: September 2001 ∗ We thank John Browning and especially ...»
Minimum Wages and On-the-job Training
Daron Acemoglu J¨rn-Steﬀen Pischke
First Version: April 1998
This Version: September 2001
We thank John Browning and especially Aimee Chin for excellent research assistance, and Joe
Altonji, Richard Carson, Ken Chay, Jinyong Hahn, Lisa Lynch, Paul Oyer, Chris Taber, and Yoram
Weiss for useful comments. The ﬁrst draft of this paper was written while Pischke was a visiting scholar at the Northwestern University/University of Chicago Joint Center for Poverty Research. He is grateful for their hospitality and research support. Acemoglu is grateful for ﬁnancial support under the National Science Foundation Grant SBR-9602116.
0 Minimum Wages and On-the-job Training Abstract Becker’s theory of human capital predicts that minimum wages should reduce training investments for aﬀected workers because they prevent these workers from taking wage cuts necessary to ﬁnance training. In contrast, in noncompetitive la- bor markets, minimum wages tend to increase training of aﬀected workers because they induce ﬁrms to train their unskilled employees. We provide new estimates on the impact of the state and federal increases in the minimum wage between 1987 and 1992 on the training of low wage workers. We ﬁnd no evidence that minimum wages reduce training, and little evidence that they tend to increase training. We therefore develop a hybrid model where minimum wages reduce the training investments of workers who were taking wage cuts to ﬁnance their training, while increasing the training of other workers. Finally, we provide some evidence consistent with this hybrid model.
Keywords: Imperfect Labor Markets, Low Wage Workers, General Human Capital, Firm Sponsored Training JEL Classiﬁcation: J24, J31, J41 1 Introduction Much of the recent debate on the minimum wage has focused on its employment impli- cations. The theory of human capital suggests that minimum wages should also have important adverse eﬀects on human capital accumulation. In the standard human capi- tal theory, as developed by Becker (1964), Ben-Porath (1967), and Mincer (1972), a large part of human capital is accumulated on the job, and workers often ﬁnance these invest- ments through lower wages. A binding minimum wage will therefore reduce workplace training, as it prevents low wage workers from accepting the necessary wage cuts (Rosen, 1972). The early empirical literature has conﬁrmed this prediction. The negative impact on human capital formation has been an important argument against minimum wages in the minds of many economists and policy-makers, and an important piece of evidence in support of the standard theory of human capital.
In this paper, we revisit the impact of minimum wages on training. We build on our previous work, Acemoglu and Pischke (1999b), which showed that a compression in the structure of wages can induce ﬁrm-sponsored training. We show that in noncompetitive labor markets minimum wages can increase–rather than decrease–training investments because they compress the wage structure.
The intuition for this result is that minimum wages make it less proﬁtable to employ unskilled workers. When there are no rents in the employment relationship, as in a competitive labor market, the ﬁrm has no option but to lay oﬀ workers who were previously paid below the new minimum wage. In contrast, in the presence of labor market rents, it may be more proﬁtable to increase the productivity of workers, who are already receiving high wages, rather than laying them oﬀ. Figure 1 illustrates this intuition diagrammatically. It draws the relation between worker skills, τ, productivity, f (τ ), and wages w(τ ). The gap between productivity and wages, ∆, is the rent that the ﬁrm obtains. A binding minimum wage, in the absence of such rents, forces the ﬁrm to lay oﬀ the worker. However, with ∆ suﬃciently high, the ﬁrm would like to retain the worker despite the higher wages dictated by the minimum wage. In this case, the ﬁrm would also like to increase the productivity of the worker. Without the minimum wage, the gap between f (τ ) and w(τ ) was constant, so there was no point in incurring costs of training. However, with a minimum wage, proﬁts are less at τ = 0 than at τ = 1. So if the ﬁrm can increase its employee’s skills to τ = 1 at a moderate cost, it will prefer to do so. In essence, the minimum wage has made the ﬁrm the de facto residual claimant of the increase in the worker’s productivity, whereas without the minimum wage, the worker was the residual claimant.
This reasoning suggests that a binding minimum wage may induce ﬁrms to invest
Figure 1: Training with a Minimum Wage and Employment Rents more in the skills of their employees. Since this implication diﬀers sharply from the prediction of the standard theory, empirical evidence on this point can shed light on whether non-competitive features aﬀecting training decisions are important. Although existing evidence points to a negative eﬀect of minimum wages on training, we argue that this evidence suﬀers from a number of problems. We therefore adopt a diﬀerent approach and provide new empirical estimates that are quite diﬀerent from those in the literature. We use the National Longitudinal Survey of Youth (NLSY) for the period 1987 to 1992. This period encompasses a number of state minimum wage increases as well as two federal increases in the minimum in 1990 and 1991. Our data therefore contain a large amount of within state variation in minimum wages. Furthermore, the NLSY is a panel of youths and oversamples those from disadvantaged backgrounds, so it contains a relatively high number of low wage workers directly aﬀected by minimum wage increases.
Our empirical results show almost no evidence of a reduction in training in response to minimum wages. But, they also do not provide strong support for our alternative model. Overall, the evidence suggests that minimum wages appear to have little eﬀect on training investments for low-wage workers. Although we cannot rule out modest positive or negative eﬀects, our two standard error conﬁdence bands exclude large negative eﬀects 3 of minimum wage increases on training–in fact, most of our most reliable estimates are positive.
An appealing way to explain the empirical ﬁndings is a hybrid model in which minimum wages increase training for some workers while reducing it for others. In this model, as in the standard theory, the wages of some workers are low because they are compensating their employers for investments in general skills. The minimum wage laws prevent this. This approach therefore suggests that minimum wages reduce the training of workers taking wage cuts to ﬁnance their training, while inducing further training for those who were constrained in their human capital investments. This hybrid model predicts that the impact of minimum wages should depend on the presence and size of labor market rents. We end the paper with some evidence pointing in this direction.
The rest of the paper is organized as follows. The following section discusses the previous empirical literature. Section 3 presents a simple theoretical setup where, contrary to the predictions of the standard theory, minimum wages increase training investments.
In Section 4, we describe our data set, and discuss the empirical strategy to estimate the eﬀects of minimum wages on training. Our results are presented in Section 5. Since the empirical evidence supports neither the standard Becker theory nor our model, in Section 6, we consider a hybrid model where minimum wages increase training for some workers, while reducing it for others. In Section 7, we provide some additional empirical evidence on this hybrid model. Section 8 concludes.
2 A Critique of the Previous Empirical Literature There is a small empirical literature investigating the impact of minimum wages on training. Part of this literature focuses on whether minimum wage laws lead to slower observed wage growth in micro data. Both Leighton and Mincer (1981) and Hashimoto (1982) have found this to be the case and concluded that minimum wage laws lead to less training. But, since a minimum wage increases the wages of low paid workers, it can reduce wage growth without aﬀecting training. Therefore, it is unsatisfactory to interpret the decline in age-earnings proﬁles as evidence of reduced investment in general training.
Consistent with this view, Grossberg and Sicilian (1999) ﬁnd no eﬀect of minimum wages on training, but still ﬁnd lower wage growth for minimum wage workers. Furthermore, Card and Krueger (1995) compared cross sectional wage proﬁles in California before and after the 1988 minimum wage increase with a number of comparison states. They also found ﬂatter proﬁles in California after the minimum wage increase. However, they point out that the Californian proﬁle also shifts up and does not cross the previous age-wage proﬁle. This pattern contradicts the standard theory, but is consistent with 4 the predictions of our model.
Given the diﬃculty of interpreting changes in the slope of wage proﬁles, we ﬁnd it more compelling to look at the impact of minimum wages on training directly, but we are only aware of four previous studies doing this for the US. Leighton and Mincer (1981) use worker reported data on the receipt of training from the Panel Study of Income Dynamic (PSID) and the National Longitudinal Survey and ﬁnd that workers in states with lower wages and therefore a more binding federal minimum wage receive signiﬁcantly less training. Cross state comparisons may be confounded by the presence of other state eﬀects, however. For example, industrial and occupational composition of employment varies substantially across states, and diﬀerent industries and occupations have diﬀerent skill requirements. These considerations suggest that across state comparisons are hard to interpret.
Schiller (1994) reports a similar ﬁnding using later data from the NLSY by comparing the training incidence of minimum wage workers with those earning higher wages.
The evidence from this study is even harder to interpret because worker traits which lead to higher pay are typically also associated with more training. Grossberg and Sicilian (1999) use data from the Employment Opportunity Pilot Project (EOPP) and compare minimum wage workers both to workers earning slightly less and slightly more, ameliorating the problem of worker heterogeneity somewhat. They ﬁnd insigniﬁcant negative eﬀects on training for male minimum wage workers and insigniﬁcant positive eﬀects for women. Leighton and Mincer only analyzed men, although women make up the majority of minimum wage workers.
Some of these problems are overcome in a more recent study by Neumark and Wascher (1998), who use Current Population Survey (CPS) supplements to compare the impact of minimum wages on training within states using comparisons of young workers in 1991 with older workers (who are less likely to be aﬀected by the minimum wage) and with young workers in 1983. These comparisons assume that state diﬀerences in training levels are the same for younger and older workers and remain so over long time periods, which are stringent requirements. They also ﬁnd negative eﬀects of minimum wages on training, but these eﬀects seem to be too large to be sensible.
To see why the eﬀects implied Neumark and Washer’s paper are implausibly large, note that their treatment group consists of all young workers. Not all of these workers are aﬀected by the minimum wage, however. Let us assume, quite generously, that all workers earning less than 160 percent of the minimum are “aﬀected” by the minimum wage. The 160 percent of the average federal minimum over the period they study is $5.60, and 40 percent of workers aged 20-24 are paid below this wage in 1991.
Neumark and Washer’s estimates imply that formal training among workers aged 20-24 5 in California (a high minimum wage state) was 3.2 percentage points lower than in states which were subject to the lower federal minimum. This point estimate, then, implies that among aﬀected workers, training will be lower by approximately 8 percentage points (i.e. 3.2 percentage points divided by 0.40). The average incidence of training among aﬀected workers in low minimum wage states is 3.0 percent (much lower than among all workers aged 20-24 for whom the incidence is 10 percent). So this estimate implies that introducing California’s minimum wage to low minimum wage states should have wiped out all training two and a half times among aﬀected workers in these states! Clearly, an implausibly large eﬀect.1 3 Minimum Wages and Training In Noncompetitive Labor Markets In this section, we use a two-period model to analyze the impact of minimum wages on training. The main result of this analysis is that plausible deviations from perfectly competitive labor markets, which introduce ﬁrm-speciﬁc rents and prevent workers from ﬁnancing their own training, change the conclusions of Becker’s theory. Namely, we ﬁnd that minimum wages can increase investments in general training.
3.1 Environment The world lasts for two periods, 1 and 2. There is no discounting, and all agents are riskneutral. There is a continuum of workers with mass 1, who supply labor inelastically.
These workers diﬀer by ability. More speciﬁcally, there is a distribution of abilities across h i workers denoted by G(η) with support η, η. We introduce heterogeneity in abilities to capture the disemployment eﬀects of the minimum wage. This feature will also be useful later in Section 6 when we discuss the possibility of workers paying for their own training.