PERI Living Wage Study

Comments of Dr. Richard B. Freeman
Harvard University

Report on “Economic Analysis of Santa Monica Living Wage” Study 

This is a well-done empirical investigation of the potential impacts of several related living wage proposals for Santa Monica. It uses a variety of data – interviews with businesses to assess the likely business responses to the living wage proposal, econometric analysis of time series, tabulations of the Current Population Survey, and a survey of workers in the potential beneficiary group – to explore the potential effects of a Coastal Zone living wage and alternatives. It deals in a straightforward way with the economic issues involved in applying a living wage to the Coastal Zone. It presents the empirical calculations clearly, and it reaches what seems to me to be a measured assessment. The analysis shows sensitivity to the fact that the living wage will affect different firms differently and different workers differently. 

There is a basic tension in a living wage policy. The point of the policy is to increase substantially the well-being of the covered workers – if not, living wages are useless. But wage increases necessarily lower the profits of firms and could backfire by harming many affected workers or unduly affecting the economic operation of the firms. When firms respond to wage increases by improving productivity, or when the increases reduce turnover, or when the firms pass on the wage increases to consumers with little loss of business, living wages can succeed. Conversely, when firms respond to living wages largely by laying people off or by relocating or by replacing existing workers with more skilled workers, the policy is unlikely to help the persons it is meant to help. The greater the shift in money to the covered workers the greater the benefit to them but the greater the risk that the policy will self-destruct. A balance must be struck between the extent of the living wage – its level and coverage – and the risk that the policy will have unintended adverse consequences on workers and firms. 

This evidence in this report shows that the Santa Monica Coastal Zone proposal strikes a defensible balance. The key argument is on the cost absorption front: that the hotels who will bear much of the cost have benefitted from the growth restriction policies of the city, and thus earn “economic rents” which can be transferred to low wage workers at little risk to business viability. Put differently, city policies restricting growth in the Coast Zone create extra profits for hotels - profits associated solely with the location that they can afford to spend on wage increases without altering much of their operations. The rents ultimately come from Proposition S, and can be distributed to workers through the living wage rather than to the hotels. The case that restaurants can readily absorb the costs is less powerful, but the exemption of workers with sizable tips arguably protects the restaurants from excessive cost increases. The study also provides some evidence that lower turnover induced by higher wages will save the firms plausible amounts. 

In the first draft of this report, I found several places where the evidence was less than convincing, and I so told the authors. They have been very responsive on the bulk of these points, so my critical comments here will be limited. I focus my discussion on two issues: whether the low paid workers truly benefit from the living wage, which depends upon who is covered and the responses of business to higher costs; and who bears the cost of the transfer to the low wage workers, which also depends in part on the responses of business. 

The raison d’etre of a living wage is to change in living standards for affected workers. The key tables that estimate how much the living standards of these workers will be increased are in chapter 8 (8.16- 8.18), which I would have brought into the summary. These tables simulate the gain in disposable income to families where workers gain from the legislated living wage. Critics of living wages often stress that because of federal and state tax and benefit policies, the benefits to workers are considerably less than the increase in wage would suggest. The reason is that there is a “leakage” of money from the workers due to the income tax, payroll tax, and EITC. These tables show that the leakage in Santa Monica is on the order of 43%. Hence, for every dollar increase in the living wage, workers’ disposable income increases by about 57% (averaging the two cases in the table). After taking account of the leakages, the families of workers gain $2000-2500. But it is critical to realize that the leakages are not waste, but transfers to the federal and state governments. This is an important point, because leakages are not an economic cost that reduces the efficiency of the Santa Monica economy. They are a shift in money from persons paying the higher wage (the firm or consumers) to the government rather than to the low wage workers. 

The study considers the case where workers gain 15 paid days off. Since the government does not tax time off, this transfer goes fully to the workers: at the same total annual income, they enjoy more time with their families. When I first reviewed the report, I thought the 15 days was excessive – a near doubling over the 8 days the average worker has – but I now see this as a useful way to improve the living standards of persons while avoiding the leakages due to higher tax rates and loss of EITC. 

The decision to pursue the sales threshold seems correct, and the discussion of how to deal with the problem of variability in sales and possible incentives for firms to operate so that they fall below the threshold shows considerable sensitivity to the way such limits may affect behavior or accounting. Still, I think that firms may have more ways to avoid this than the study recognizes. A firm that increases sales from below to above the threshold will suddenly face a rise in wages for its below-living wage work force. This seems to call out for subcontracting or some other evasive scheme. Thresholds are discontinuous, so why not do what the market does and relate living wages to the size of firm through a continuous transformation – for instance applying a lower graduated living wage for smaller firms? Graduated income taxes are a main feature of almost all income tax systems. In the pre-IT world this might be an administrative nightmare, but perhaps it is worth considering given new computer technologies. 

The report points out that the living wage will reallocate who gets Coastal Zone jobs. With higher wages, the jobs will become more attractive to more skilled workers and the firms will select those workers over less skilled workers. The report examines carefully data on substitution of skills within occupations, using Current Population Survey files, on the qualifications of workers in affected occupations with higher and lower pay in various geographic areas. This evidence suggests that some substitution is likely. The authors regard the degree of substitution as modest. The proposed Santa Monica hiring hall might provide a way to deal with it. I think that even if substitution is a more serious issue, this does not affect the overall conclusion that the living wage will benefit low wage workers. If the Coastal firms hire more skilled workers attracted by the better wages, the less skilled who might have gotten the Coastal jobs could find work with the firms that would otherwise have hired the more skilled. In this case the unskilled workers still gain, though by less than if there was no substitution of more skilled for less skilled workers, while more skilled workers (also relatively low paid) also gain. 

On the issue of who pays for the living wage, I accept the main point that most of the cost will come out of economic rent. Table S.2 shows that 40-50% of hotels and restaurants expect to respond with layoffs and 40-70% to hire fewer workers in the future. While the authors found no evidence that there are price effects on demand for Santa Monica rooms (p 72 and Appendix 5), this data is limited and they are right to be dubious of the value of this exercise. I endorse their alternative mode of analysis of applying an estimated elasticity to the increase in wages, to assess the potential loss of jobs. Since estimated elasticities tend to be small, their leaning in the direction of picking a larger estimate from Neumark and Wachtel made good sense to me. The worst case scenario shows only modest job losses, so I think their overall conclusion is correct: in the band of wages considered for the Coastal Zone firms, this should not a major problem is correct. It fits with the entire literature on the minimum wage, that employment losses are modest (given that the wage increases are invariably set to avoid an employment disaster). 

On some of the other issues, I accept the argument that minimum wages compress differentials so that spillovers within firms are unlikely to be that great. I cannot judge the assessment of the likely costs of administering the ordinance, but if they are even roughly on target, those costs are dwarfed by the amount of the transfers, making this a relatively efficient way to move money to the lower paid part of the population. I was not worried over the danger that higher wages will create greater cyclic problems nor with some of the other technical issues that the report dealt with, though presumably others are so worried, justifying these analyses. 

CONCLUSION 

The study does not conclude with a benefit-cost analysis of the Santa Monica proposal, nor does it make any recommendation, but the analysis suggests that a reasonable benefit-cost bottom-line would favor the Coastal Zone living wage proposal. One might criticize particular statistics or calculations as being imperfect, but the fact that much of the cost of the living wage will be paid out of economic rents from hotels and other businesses that benefit from the restricted growth policy comes across clearly. 

As for alternatives, the study disposes of the notion that getting families into EITC will help them much (since most are already covered. The study rejects the contractors only ordinance because too few will be covered and because it offers lower wages and benefits. I have no problem with this. In addition, residents will pay for at least part of a contractors only living wage while the Coastal Zone ordinance will largely tax non-residents, so it seems to be better on both the benefit and cost side to residents. Since businesses outside the Coastal Zone do not have economic rents from restrictive growth policies, applying the living wage to them is likely to create greater economic costs, so that the level of the living wage would have to be lower to obtain the same benefit-cost ratio. 

 

This page was last updated on 10/25/06.