Excerpts from Side Effects and Complications: The Economic Consequences of Health Care Reform
by Casey B. Mulligan
(c) copyright 2015 by the University of Chicago Press

Chapter 7

Adam Smith’s Equalizing Differences: More Hidden Taxes and More Health Coverage

The Affordable Care Act’s effects extend well beyond the employment rate and the length of the workweek. The purpose of this chapter is to examine the law’s incentives for workers to change the sectors where they work and the type of coverage they get (if any) and to examine the law’s incentives for employers to adjust the amount and composition of their employee compensation. The theory and arithmetic of equalizing differences is essential to this analysis because it shows how workers having no direct contact with the ACA’s penalties and subsidies nonetheless experience the law’s provisions because they compete and produce with workers who do. The first results given in the chapter are for the wage effects of the law separately for “low-skill” and “high-skill” workers. The last part of the chapter demonstrates the strength of the law’s incentives to expand health insurance coverage and how its employer penalty is essential for determining the source of coverage.

A. The Theory of Equalizing Differences

Both the employer penalty and the exchange subsidies affect the structure of wages, by which I mean the average level of wages and wage patterns across types of workers. The theory of equalizing differences is the best way to understand the ACA’s impact on the structure of wages, so this chapter begins with a simple, everyday illustration of the theory in action.1 I then apply the theory to consider the employer penalty in isolation—as if it were the only tax in the ACA—and afterward analyze the exchange subsidies.

1. Introduction: A Tale of Two Occupations

Even after adjusting for inflation, workers today earn far more than they did decades ago. The primary reasons for this are the changes in workers and their work environments that have made them vastly more productive. Workers today typically have more and better machinery to accomplish their tasks. Take the farmer. An early twentieth-century farm tractor is shown on the left (Figure 7.1a), and a modern tractor is shown on the right (Figure 7.1b).


Today 100 bushels of corn can be produced with fewer than three labor hours and less than one acre of land (U.S. Department of Agriculture, National Institute of Food and Agriculture 2014). In 1945, it took fourteen labor hours and two acres of land to produce the same 100 bushels because farmers then did not have the farm equipment and education that is available today. Even more farm work was required to produce the same results at the beginning of the twentieth century. It’s no puzzle that farmers can earn more today than they did years ago.

But wage growth is not limited to occupations where technology has progressed, capital has accumulated, or education is essential. Barbers today cut hair almost exactly the way they did in the first half of the twentieth century. The barbershop photographed in Figure 7.2a is circa 1920, and the equipment looks about the same as it does now (Figure 7.2b): scissors, chairs, and a sink. Yet barbers now are paid much more than the barbers from the era of the photograph.2


The fact that both farmers and barbers shared wage growth but not technology and other sources of productivity growth would seem to defy productivity explanations for wage growth, but this impression ignores the theory of equalizing differences. Because men’s desired quantity (not styles!) of haircuts have been pretty constant over the years, men must pay barbers enough to induce people to pursue the occupation and not some other profession for which productivity has been significantly advanced by capital accumulation, modern technology, and advanced education. Through the market mechanism, wage growth for barbers is a consequence of productivity growth in other occupations.

2. The Employer Penalty, Sectoral Shifts, and Hidden Taxes

As shown in Chapter 3, the ACA’s employer penalty is significant. At first glance, the penalty would seem to be just the problem of workers employed at penalized employers, with their workers receiving less weekly pay or losing their jobs altogether. But this is the same mistake as concluding that the pay of barbers is unaffected by technological changes outside their profession. As workers leave penalized employers and compete for jobs at employers that offer coverage, their departures drive down wages at ESI employers and mitigate some of the penalty’s effect on wages at non-ESI employers.

Figure 7.3a illustrates. The overall height of the bars indicates compensation per employee in each of two sectors (no-ESI and ESI) without the ACA, adjusted for any sectoral differences in nonpecuniary job attributes.3 The two heights are equal to indicate the competition between the two sectors for employees, so that employees cannot get a better deal by leaving their sector. Now the ACA penalty comes along and takes part of the pay of the no-ESI (i.e., penalized) sector, as indicated by the red “penalty” part of the no-ESI bar.4 Employees leave the penalized sector to take advantage of the higher ESI-sector pay. The more employees who seek work in the ESI sector, the less ESI employers are willing to pay for them.5 At the same time, the more employees who leave the penalized sector, the more penalized employers are willing to pay the employees who remain. Just as barbers were partially compensated for the technological progress that occurred outside their occupation, non-ESI employees will be partially compensated for the penalty-free opportunities existing outside their sector.


The labor market equilibrium in the presence of the penalty also equalizes compensation between sectors, as shown in Figure 7.3b. The figure includes a dashed outline of the ESI bar to indicate its height without the penalty (see Figure 7.3a) and thereby the amount by which the penalty depresses pay among ESI employees. This amount can be interpreted as a hidden tax on ESI employees: the employer penalty reduces their pay even though their employers do not pay it! In effect, penalized employees escape part of their penalty by passing it on to ESI employees. Ultimately employees in both sectors pay the same amount, which is the degree to which the two green bars in Figure 7.3b are less tall than the bars in Figure 7.3a. Here is how the father of economics, Adam Smith, would put it:

The whole of the advantages and disadvantages of the different employments of labour and stock must, in the same neighbourhood, be either perfectly equal or continually tending to equality. If in the same neighbourhood, there was any employment evidently either more or less advantageous than the rest, so many people would crowd into it in the one case, and so many would desert it in the other [think “employer penalty”], that its advantages would soon return to the level of other employments. (Smith 1776/1904, chapter I.10.1, brackets added)

The gross wage gap for ESI employees shown in figure 7.3b as the red box is known as a “compensating” or “equalizing” difference because it equalizes wages net of the other advantages of ESI employment (in this case, the advantage is tax avoidance), exactly as Adam Smith observed.6

The workers in Figure 7.3 just move from one sector to another and do not leave work entirely. If some of them did leave work entirely, then wages might rise in both sectors as workers became somewhat more scarce. For this reason, the sectoral shifts shown in Figures 7.3a and 7.3b describe the effect of a sector-specific penalty on overall labor demand—even the demand for workers who work in untaxed sectors—and not necessarily the effect on overall wages. The overall wage impact and the labor demand impact would coincide if the total supply of labor were held fixed.

Figure 7.3 is a more elaborate model of the penalty’s effects on wages than the model shown in Chapter 6. Both models are consistent with the long-term hypothesis that the employer penalty comes out of worker pay rather than employer profits, but Chapter 6 had the penalty subtracted from worker pay even without any explicit adjustments by labor or capital. Figure 7.3 shows how the employer penalty causes penalized sectors to contract and other sectors to expand, which in the short run may reduce profits in the penalized sectors and increase profits in the others. Figure 7.3 also shows how some of the wage reductions from the penalty will be experienced by ESI workers, which also suggests that some of the effects of the penalty on aggregate work hours featured in Chapter 6 will be experienced by ESI workers rather than non-ESI workers.

To read the rest ... buy the entire book
  1. See Rosen (1986) for an overview of the theory. Many of the advances in the theory could be applied to the ACA’s various taxes more fully than I have done in this book.

  2. For example, the Chicago Barbers Association (2011) reports that the inflation-adjusted price of haircuts more than doubled from the 1920s to the 1960s, and more than doubled again in the subsequent decades. Also note that the average price of a men’s haircut today is about $28 (http://www.usnews.com/news/blogs/data-mine/2014/02/28/what-america-pays-for-a-haircut), which would be $2.11 in 1925 prices (http://research.stlouisfed.org/fred2/series/CPIAUCNS). Legend has it that haircuts cost about $0.25 back then (the CBA data suggest closer to $0.50); we can be sure that 1925 haircut prices were far less than $2.11, and therefore sure that haircut prices far outpaced inflation 1925 to present, because $2.11 was about a day’s wages for a laborer (U.S. Department of Labor 1939, and Wolman 1935) and regular people would not pay that much for a haircut and would seriously consider getting into the barber business if they could earn a day’s wages in less than an hour.

  3. For the sake of illustration, I do not consider the possibility that the two sectors have different weekly work schedules, so equalized compensation per employee is the same as equalized compensation per hour.

  4. An alternative way to draw Figure 7.3a would be to have the penalty paid by the non-ESI employer (that is, stacked above the dashed horizontal line), with penalized employers responding by letting employees go and lowering wages. The important point is that the sector-specific penalty is a force moving employees in the direction of the arrow.

  5. In technical terms, the marginal revenue product of labor in ESI sector falls as the sector absorbs additional employees. This may happen because the marginal product of labor is diminishing in the amount of labor in the sector or because the goods produced in that sector become less scarce and thereby less expensive.

  6. Economics uses the phrases “compensating difference” or “equalizing difference” to describe the relationship between wage and nonwage attributes of a job as described by Adam Smith.

(c) copyright 2015 by the University of Chicago Press