Brendan Epstein by gjjur4356


									                             Brendan Epstein
      Department of Economics – University of Michigan, Ann Arbor
                        Dissertation Abstract

             Essays on Labor Heterogeneity and Macroeconomics

                                 Chapter 1
  “Heterogeneous Workers, Optimal Job Seeking, and Aggregate Unemployment”
Why do vacancies and unemployment coexist? The standard model of equilibrium
unemployment theory is cast in a homogeneous-agent framework, and explains involuntary
unemployment as the result of search frictions. This paper develops an understanding of
how, in addition to search frictions, heterogeneity can influence aggregate labor-market
fluctuations, and, in particular, the cyclical behavior of aggregate unemployment. I capture
heterogeneity by considering a labor force composed of individuals who have a
comparative advantage in a particular job, yet are still able to work in jobs in which they
are at a comparative disadvantage. I assume no worker has an absolute advantage in
production. In addition, I endogenize the optimal job-seeking behavior across job
opportunities of all searchers, both those unemployed and those searching on the job. On-
the-job search results from workers who are employed in jobs in which they are at a
comparative disadvantage, but search for comparative advantage employment.

The extent to which vacancies and unemployment coexist is summarized by the aggregate
vacancy-unemployment (V/U) ratio. Empirically, in the U.S., the V/U ratio is strongly
procyclical, and part of its adjustment in response to changes in productivity is sluggish.
Under standard calibrations the benchmark, homogeneous-agent model of equilibrium
unemployment theory can account for slightly less than half of the empirical elasticity of
the V/U ratio with respect to productivity. However, contrary to the data, all adjustments
occur instantaneously. Results from the model developed in the present paper suggest that
the impact of worker-side heterogeneity and optimal job-seeking behavior can be
important. Quantitative analysis reveals that the model can account for the majority of the
empirical elasticity of the V/U ratio with respect to productivity in the United States. In
addition, the theory uncovers a natural channel through which adjustments in the V/U ratio
can be slow moving. Given heterogeneity, vacancy-posting decisions are based on firms’
expectations regarding match quality. These expectations depend on worker-side job-
seeking behavior. In addition, they also depend on the (slow-moving) masses of
unemployed and on-the-job searchers. Ultimately, this results in slow-moving expected
gains from posting vacancies, which contributes to sluggish adjustment in the V/U ratio in
response to changes in productivity.

                                     Chapter 2
                  “The Decline of Drudgery” (with Miles S. Kimball)
Economists have long understood that cross-sectional differences in on-the-job utility at a
particular time give rise to compensating differentials. This paper develops a theory that
focuses on a less-studied topic: understanding the long-run macroeconomic consequences
of trends in on-the-job utility. One important implication of secular improvements in on-
                                                                                 Page 1 of 2
the-job utility is that it is possible for hours per worker to remain approximately constant
over time even if the income effect of higher wages on labor supply exceeds the
substitution effect of higher wages. Another important implication is that secular
improvements in on-the-job utility can themselves be a substantial component of the
welfare gains from technological progress. These two implications are connected by an
identity: improvements in on-the-job utility that have a significant effect on labor supply
tend to have large welfare effects.

A typology of different sources of improvement in on-the-job utility elucidates how these
improvements can be endogenized. Amenities, such as air-conditioning, increase due to
both income effects and ordinary technological progress. Effort - which can be defined
broadly as undesirable dimensions of a job that lead to higher output - behaves in a way
qualitatively similar to hours per worker. However, unlike work hours, effort is
unmeasured in standard data series. Unobserved movements in effort can act as substitutes
for movements in hours per worker. A decline in drudgery represents an increase in on-
the-job utility for a given level of effort and amenities. Such increases in on-the-job utility
lead to firms gaining a competitive advantage. Therefore, profit-driven innovation is a
natural channel through which declines in drudgery can occur. These declines, along with
endogenous increases in amenities, tend to increase hours per worker and effort. This
serves to counteract decreases in work hours and effort that would otherwise take place
when income effects exceed substitution effects.

                                   Chapter 3
    “Beyond Taxes: Understanding the Labor Wedge” (with Shanthi P. Ramnath)
A recent literature has shown that the standard neoclassical macroeconomic model falls
short of explaining the trend behavior of hours per population within countries. This is
captured by the extent to which the model’s first-order conditions for equilibrium hours per
population fail to hold: the labor wedge. The literature argues that across countries a
substantial fraction of the labor wedge can be explained when the standard model is
enhanced to account for taxes. However, the model's predictions regarding Canada and the
U.S. remain at odds with the data. This paper focuses on understanding why taxes provide
a good explanation for the labor wedge across countries, but not as much for the U.S. and

Hours per population are equal to the product of hours per worker and the employment to
population ratio. We show that, empirically, trends in hours per population in the U.S. and
Canada over the last several decades have been driven by changes in the employment to
population ratio. The standard model implicitly assumes that all household members are
employed and work the same amount of hours. Therefore, it has no channels through
which adjustments in the employment to population ratio can be captured. We develop a
model that allows for a representative household to have an explicit choice of labor supply
at both the intensive and extensive margins. The model nests the standard theory, and
helps explain its shortcomings regarding the U.S. and Canada. We use our model to
examine the macroeconomic implications of tax policy on hours per population through its
disaggregate components: hours per worker and the employment to population ratio.

                                                                                    Page 2 of 2

To top