Estimation of Equilibrium Wage Distributions with Heterogeneity
Equilibrium search models generalize the one-sided models used earlier
by endogenizing the wage offer distribution. Attempts to estimate
versions of these equilibrium n search models have been unsuccessful, ironically
because the predicted wage distribution did not fit the wage data.
In this paper we extend the applicability of search models by introducing
firm heterogeneity to account for the shape of the wage density.
The resulting estimation problem is difficult, but we propose a solution
and illustrate its feasibility and performance with a Monte Carlo study
and an application to U.S. labor market data.
Back to my
CV page
Equilibrium Search Models and the Transition from School to
Work
This paper applies the Burdett-Mortensen (1998) equilibrium search model
to study the school to work transition of U.S. high school graduates. We
consider the case of discrete firm heterogeneity and provide a computational
method to obtain the MLE. Our results show that unemployed blacks
receive fewer offers than whites and employed blacks are more likely to
lose their jobs. Importantly, employed blacks and whites receive job offers
at the same rate. Assigning the whites' search paramters to the blacks
and re-solving reveals that 75 percent of the observed wage differential
is explained by the job destruction rate differences.
Back
to my CV page
A Search Interpretation of Male-Female Wage Differentials
A general equilibrium search framework is used to examine the role
of gender differences in labor market behavior patterns (e.g. quit rates
for personal reasons) in determining gender wage differentials. For samples
of high school and college graduates from the National Longitudinal Survey
of Youth (NLSY), these behavioral patterns are found to be significantly
different across the sexes and account for 20-30% of the wage differentials.
In particular, they play a key role in explaining the male-female wage
differential that remains after controlling for the gender composition
across occupations.
Back
to my CV page
Discrimination and Skill Differences in an Equilibrium Search
Model
In this paper we analyze an equilibrium search model with three sources
for wage and unemployment differentials among workers with the same (observed)
human capital but different appearance (race): unobserved productivity
(skill), search intensities, and discrimination (Becker 1957) due to an
appearance-based employer disutility factor. Because they affect
the earnings distributions differently, empirical identification of these
potential sources for the explanation of the wage and unemployment differentials
is possible. We show that the structural parameters of the model,
including the firm's disutility from certain workers, are identifiable
using standard labor market survey data. We deomonstrate identification
using data from the National Longitudinal Survery of Youth. Estimation
of these parameters by matching moments from a sample of black and white
high school graduates implies: a) blacks have a 3.3% lower productivity
level than whites; b) the disutility factor in employer's preferences is
31% of the white's productivity level; and c) 56% of firms have a disutility
factor in their utility toward blacks.
Back
to my CV page
U.S.-Canadian Unemployment Rate and Wage Differences
Among Young, Low Skilled Males in the 1980's
During the mid 1980's young, low skilled adults in Canada were much more
likely to be out of work than their U.S. counterparts. The unemployment
rate gap for this cohort was 7 percentage points. At the same time wage
inequality was higher in the U.S. Using panel data from the U.S. National
Longitudinal Survey of Youth and the Canadian Labour Market Activity Survey,
in this study a general equilibrium search model of the labour market is
employed to identify structural differences contributing to these gaps.
The results reveal that both wage and unemployment differences are driven
by a higher job destruction/separation rate in Canada and higher job offer
arrival rates in the U.S. In general, the model characterizes the U.S.
labour market as having less search frictions than that of Canada. That
is, Canadian firms are found to have more monopsony power than their U.S.
counterparts.
Back
to my CV page
A Panel Data Analysis of the U.S.-Canadian Nonemployment Rate
Gap
Among Young, Low Skilled Males
Evidence from the U.S. National Longitudinal Survey of Youth and the Canadian
Labour Market Activity Survey shows differences in both incidence and duration
give rise to the mid 1980 U.S.-Canadian nonemployment rate gap of young,
low skilled males. Canadians are more likely to experience a firm-initiated
job separation, to have been in a seasonal or temporary job, to transition
to nonemployment rather than another job, and to take-up UI than Americans.
Overall, a pattern emerges of more intermittent employment in Canada with
intervening spells of UI-sponsored nonemployment.
Back to
my CV page
Search Friction in the U.S. Labor Market:
Equilibrium Estimates from the PSID
Equilibrium search models have been used to study a variety of labor market
issues both within
and across countries. Many of these research questions require
representative panel data at the
national level to be answered. To that end we determine the feasibility
of using data from the
Panel Study of Income Dynamics to estimate the Burdett-Mortensen general
equilibrium search
model for the U.S. We find the data contain sufficient information
on wages, labor force states,
durations, and transitions to generate estimates of the model’s structural
parameters. Our
analysis compares the relative labor market search friction for black
and white male household
heads. In general we find blacks face greater search friction
while unemployed than whites, but
a similar level while employed. Within the model this finding
implies substantial productivity
differentials are needed to generate the black-white wage differentials
found in the data.
Back to my CV
page
Job Match Quality Over the Business Cycle
Does the business cycle have an impact on job-matching, specificaaly on
the quality of job matches? Useing the National Longitudinal Suvery of
Youth data to capture match quality at the individual level, I attempt
to answer this question. Job tenure is used as a quality indicator under
the Jovanovic (1979) model where matches are experience goods. Starting
wages are also examined. Both local and national unemployment rates are
used as cyclical indicators. The finding is a negative cyclical impact
on both job match quality indicators. Thus during recessions individuals
take jobs that are lower paying and dissolve quicker.
Back to my
CV page
Matching Workers and Jobs: Cyclical Fluctuations in Match Quality
Using NLS Youth data on tenure and wages this paper analyzes the extent
to which the level of job mismatching varies over the business cycle and
how is it dealt with by the labor market. I find significant cyclical variation
in job match quality and an internalization of the variation by the labor
market through wages. Mismatching occurs more during recessions but is
primarily captured in starting wages. The evidence suggests the cyclical
phenomenon is one of general mismatching rather than an increased number
of stop-gap jobs during recessions.
Back
to my CV page
The Economics of Abuse
Domestic abuse affects a surprising number of families. However,
little is known about the effects of abuse on the behavior of women.
In this paper, we develop and estimate an economic model of abuse, where
the determinants of domestic violence and its consequences for the employment
and marital status decisions of women are considered. Abuse affects
employment via effects on productivity (wages) and on the utility from
leisure. In turn employment decisions can influence the level of
abuse in the marriage. Domestic violence also influences marital
status decisions through reductions in the gains to marriage as the level
of abuse is revealed. We estimate the model using a representative
Canadian data set on domestic violence. Our analysis reveals three
main findings. First, domestic abuse is less likely to occur in households
where women and their spouses are employed, suggesting domestic abuse may
be a response, in part, to economic stress. Second, women abused
in current marriages are less likely to be employed, consistent with reductions
in labor market productivity. Divorced women abused in past marriages
are also less likely to work, indicating long term effects of domestic
violence on employment. Finally, domestic abuse is a dominant factor
in the divorce and remarriage decisions, which in turn are major determinants
of employment. Failing to control for the simultaneity of employment,
marriage and abuse results in misleading inferences regarding the effects
of employment on abuse propensities, the effects of domestic abuse on employment
and the relationship between remarriage and abuse.
Back to my CV
page
Household Structure and Labor Demand in Agriculture:
Testing for Separability in Rural China
China’s economic reforms have brought rapid growth in rural off-farm employment,
raising questions about the assumption that rural China is labor surplus
and has poorly functioning factor markets. We investigate this by
testing for separability between household labor demand and supply using
panel data. We find that separability (1) is rejected overall, indicating
that factor markets remain underdeveloped; (2) is associated with labor
surplus in some areas and labor shortage in others; and (3) holds only
where substantial employment opportunities exist at the township level,
suggesting that such employment induces competitive allocation in villages
and promotes inter-village movement of resources.
Back to my CV
page
Business Cycle Models, Aggregation and Real Wage Cyclicality
Predictions regarding the cyclicality of real wages are often used as a
litmus test for macroeconomic business cycle models. In response
a substantial literature has developed to estimate the ‘true’ cyclicality
of real wages, i.e. composition bias free. Two major issues are addressed
in this paper: aggregation of heterogeneous workers and potential bias
in the measurement of the labor input. A general analysis of the
biases is presented and alternative approaches in the literature are nested
in a single framework. CPS data are used to compare the alternative approaches
in a single data set and to provide “corrected” estimates based on an efficiency
units concept that avoids the usual aggregation problems. The empirical
results show robust evidence of highly pro-cyclical real wages in the U.S.
Composition bias underestimates the usual parameters of interest unless
both the price and the quantity of the labour input are adjusted appropriately.
If only the input price measure is corrected, some composition bias will
remain biasing the cyclicality estimates downward. The total composition
bias is estimated at around 25%. The estimates suggest that the implied
labor supply elasticity is smaller than many in the previous literature
- on the order of .65 to 1.14.
Back
to my CV page
Comments on.... No Abstract Available
Back to
my CV page
The Addition of Voting to a Growth Model with Public Investment
Representative agent growth models often yield optimal public investment
paths. With heterogeneous agents added structure is needed to generate
equilibrium paths. I examine the addition of a political process to a growth
model with a heterogeneous electorate. Two types of public investment,
intratemporal and intertemporal, are studied under fixed and flexible taxation
policies. The political outcomes range from unanimity to no majority core
and depend on the timing relationship between government investment and
taxation and on the flexibility the government has in setting policy. The
more restrictions placed on the government, the more likely existence can
be achieved.
Back to
my CV page