Increase of FMW expands unions beyond direct beneficiaries – raises public and private sector unions
Clemens 20 [Jeffrey & Michael R. Strain; October 2020; associate professor of economics at UC San Diego; director of Economic Policy Studies and the Arthur F. Burns Scholar in Political Economy at the American Enterprise Institute; National Bureau of Economic Research; “Public Policy and Participation in Political Interest Groups: An Analysis of Minimum Wages, Labor Unions, and Effective Advocacy”; 10.3386/w27902 KF]
In this section, we present evidence that speaks to the plausibility of alternative theories of interest-group membership. We do this by presenting analyses that divide the population into subsamples that (1) are directly affected by the minimum wage, (2) may be indirectly affected by 11 These variables come from the State and Legislative Partisan Composition Database from the National Conference of State Legislatures (NCSL, 2020). 28 the minimum wage, or (3) are not plausibly affected by the minimum wage. We also present evidence on the relationship between minimum wage increases and union wage differentials. Analyses of Subgroups of Workers In this section, we provide evidence on which types of workers and demographic groups are more likely to become members of unions in the wake of minimum wage increases. The subgroups we analyze differentiate between individuals who might be directly impacted by the minimum wage, who might benefit indirectly from the minimum wage, and whose employment and wages have no plausible connection to the minimum wage. To be more specific, we distinguish between public- and private-sector workers, between high- and low-education workers, and between workers employed in high- and low-wage industries. We first explore the relationship between minimum wage increases and public- versus private-sector union membership. This distinction is relevant for two reasons. First, public- and private-sector unions have been on different paths for decades; private-sector unions have declined, while public-sector unions have retained their strength. Second, minimum wage increases tend to affect neither the wages nor the employment of public-sector workers. We estimate our full sample regression using dependent variables that differentiate between public- and private-sector union membership. These specifications are designed so that our overall estimate can be readily decomposed into public and private components. The estimates, which appear in Table 6, imply that public-sector union membership accounts for roughly half the overall increase we observe. For further context, note that public-sector union membership accounts for roughly half of all union membership among individuals age 16–40 at 29 baseline, but less than 15 percent of this group’s employment. On average across these broad groups of workers, the minimum wage has a weak correlation with wages (Panel B). We next add an education dimension to our analysis of both public- and private-sector union membership. Specifically, we analyze the relationship between minimum wage increases and public- versus private-sector union membership among individuals with at least some college education. We analyze this sample to further probe the evidence that minimum wage increases cause increases in union membership among groups that are not directly affected by the minimum wage. The estimates in Panel B of Table 7 reveal, as should be expected, that minimum wage increases have no impact on these groups’ wage rates. The estimates in Panel A reveal that minimum wage increases quite strongly predict increases in these groups’ rates of union membership. Together, these results provide evidence that a comprehensive theory of union membership will need to include nonmaterial factors. Our next piece of analysis focuses on individuals employed in minimum-wage-intensive industries. Specifically, we focus on the food-service and retail industries. We present this analysis in Table 8. Panel A presents estimates of the impact of minimum wage increases on these individuals’ union membership rates, while Panel B presents estimates of impacts on wages. Each column presents an analysis of a different subsample, with individuals differentiated based on age and education. The estimates in columns 1 and 2 focus on individuals who are employed in food-service and retail industries and who are between ages 16 and 21. The estimates in columns 1 and 2 of Panel B reveal that these individuals’ wages are quite strongly impacted. On average, a $1 increase in the minimum wage predicts a wage increase of around 32 cents for these individuals. 30 The estimate is modestly larger in column 1, where we do not include controls for macroeconomic covariates, relative to column 2. The estimates in columns 1 and 2 of Panel A reveal that young individuals in low-wage occupations are less likely to belong to unions following minimum wage increases. Note that this sample consists of individuals who are employed and who experienced wage gains due to minimum wage increases. The sample thus contains the minimum wage’s direct beneficiaries. The decline in this group’s union membership thus appears inconsistent with theories that view membership and payment of dues as a form of reciprocity. While the evidence in columns 1 and 2 is inconsistent with a reciprocity-oriented theory of union membership, it is consistent with “free riding.” The free-rider problem has received attention in research on the effects of Right to Work laws on union membership. Right to Work laws tend to reduce union membership because they enable workers to benefit from unionnegotiated wages without paying dues (Baird, 1998). More generally, these laws allow workers to sidestep union restrictions on who can work. For young retail and food-service workers, a minimum wage increase and union membership can be viewed as substitutable sources of bargaining leverage. Consequently, the minimum wage increase reduces the direct material benefit these individuals might obtain from joining a union. Among moderately skilled individuals in minimum-wage-intensive industries, we find no evidence of changes in union membership. Interestingly, while the estimates lack statistical significance, we tend to find a positive relationship between minimum wages and union membership for older workers in these industries. This is of interest because these individuals may benefit indirectly if minimum wage increases result in increases in employers’ skill 31 requirements (Clemens, Kahn, and Meer, 2020). Because the estimates are imprecise, however, the evidence is less strong than the evidence we see elsewhere. Estimates of the Relationship Between Minimum Wage Increases and Union Wage Differentials In this section, we present estimates of the relationship between minimum wage increases and union wage differentials. Effects of minimum wage increases on union wage differentials may shed light on the relevance of direct material interests for individuals’ decisions to join unions. That said, our estimates must be interpreted with caution due to concerns linked to selection. That is, we have shown that minimum wage increases predict increases in union membership rates. Observed changes in union wage differentials may thus be driven by changes in the composition of which workers are in unions. Our estimates of the relationship between minimum wages and union wage differentials are in Table 9. In column 1, we estimate that the relationship between minimum wages and the union wage differential is negative across all employed workers. In columns 2 and 3, we estimate the relationship separately for private- and public-sector workers. Columns 4 and 5 focus on workers in low-wage industries. Finally, in columns 6, 7, and 8 we analyze sub-samples that are differentiated by educational attainment. The relationship between minimum wage increases and the union wage differential is positive for workers in the food-service and retail industries, but negative in all other groupings we consider. In most specifications, we observe a negative relationship between minimum wages and union wage differentials. This provides additional evidence that a narrow notion of material interest is unlikely to be the primary driver of changes in union membership. A key caveat to this 32 interpretation is that union wage premiums may shift due to changes in the composition of union workers. This may explain the positive relationship we observe in column 4 of Table 9, for example, since the analysis in Table 8 revealed that young workers in the food-service and retail industries become less likely to be members of unions following minimum wage increases. That said, our evidence on union wage premiums has implications that are consistent with our earlier subsample analyses. That is, we see no evidence to suggest that increases in union membership among high-education individuals are driven by material benefit in the form of higher wage.