I estimate the long term effect of military service on quantiles of earnings and education using the Vietnam draft lottery eligibility status as an instrument. I compare the local quantile treatment effect estimator studied by Abadie, Angrist, and Imbens (2002) to the instrumental variables quantile regression technique developed by Chernozhukov and Hansen (2008). Ordinary quantile regression shows a large negative association between service in Vietnam and earnings of white men, with the effect increasing in magnitude for the upper quantiles. Quantile treatment effects estimates show the opposite pattern, although much smaller in magnitude, with a small negative effect at the lower end of the distribution, and a small positive effect at the upper end. This suggests the ordinary quantile result is due to heterogeneous selection effects. The two methods of quantile treatment effects estimation give similar results.
-
Did Vietnam Veterans Get Sicker in the 1990s? The Complicated Effects of Military Service on Self-Reported Health
August 2009
Working Paper Number:
CES-09-19
The veterans disability compensation (VDC) program, which provides a monthly stipend to disabled veterans, is the third largest American disability insurance program. Since the late 1990s, VDC growth has been driven primarily by an increase in claims from Vietnam veterans, raising concerns about costs as well as health. We use the draft lottery to study the long-term effects of Vietnam-era military service on health and work in the 2000 Census. These estimates show no significant overall effects on employment or work-related disability status, with a small effect on non-work-related disability for whites. On the other hand, estimates for white men with low earnings potential show a large negative impact on employment and a marked increase in non-work-related disability rates. The differential impact of Vietnam-era service on low-skill men cannot be explained by more combat or war-theatre exposure for the least educated, leaving the relative attractiveness of VDC for less skilled men and the work disincentives embedded in the VDC system as a likely explanation.
View Full
Paper PDF
-
The Impact of College Education on Old-Age Mortality: A Study of Marginal Treatment Effects
January 2017
Working Paper Number:
CES-17-30
Using a newly constructed dataset that links 2000 U.S. Census long-form records to Social Security Administration data files, I evaluate the effect of college education on mortality. In an OLS regression, women and men who have at least some college education have 20% lower mortality rates than those with a high school degree or less. I proceed with an empirical design intended to illuminate the extent to which this relationship is causal, estimating marginal treatment effects (MTEs) using the proximity of the nearest college to individuals' birthplace as an instrument. Results indicate positive selection into college education (in terms of longevity) for both women and men. Selection drives almost all of the mortality gap for women. For men, longevity gains from college attendance are concentrated among individuals with unobserved variables that make them unlikely attend college. This suggests that men who would benefit most from receiving college education in terms of mortality reductions are those who are not attending.
View Full
Paper PDF
-
Long-Term Effects of Vietnam-Era Conscription: Schooling, Experience and Earnings
August 2007
Working Paper Number:
CES-07-23
Instrumental variables (IV) estimates using the draft lottery show that white Vietnam-era draftees suffered substantial post-service earnings losses in the 1970s and 1980s. Angrist (1990) explains these losses as due primarily to lost labor market experience. Non-public use data from the 2000 Census allow the first longerterm follow-up for a large sample from the draft-lottery cohorts. We use these data to estimate the effects of military service on earnings, schooling, and a number of other variables. Consistent with the loss-of -experience model, IV estimates of the effects of Vietnam-era service on earnings are close to zero in 2000, when the draft-lottery cohorts were middle-aged and experience profiles relatively flat. On the other hand, draft-lottery estimates show a marked increase in schooling for Vietnam-era veterans. A variety of evidence suggests this increase reflects the impact of the Vietnam-era GI Bill more than draft-avoidance behavior. The economic return to the increased schooling generated by Vietnam-era service, estimated in a wage equation that constrains the impact of Vietnam-era military service to run solely through the experience and schooling channels, appears to be less than the OLS return. Finally, we look at measures of disability. The IV estimates point to an increase in non-workrelated disability rates and non-SSA disability income, but the fact that there is no corresponding effect on employment, hours worked, or work-related disability rates suggests health was affected little by Vietnam-era service. Allowing for excess disability among veterans raises the estimated returns to GI-Bill schooling slightly.
View Full
Paper PDF
-
The Effects of Smoking in Young Adulthood on Smoking and Health Later in Life: Evidence Based on the Vietnam Era Draft Lottery
September 2008
Working Paper Number:
CES-08-35
An important, unresolved question for health policymakers and consumers is whether cigarette smoking in young adulthood has significant lasting effects into later adulthood. The Vietnam era draft lottery offers an opportunity to address this question, because it randomly assigned young men to be more likely to experience conditions favoring cigarette consumption, including highly subsidized prices. Using this natural experiment, we find that military service increased the probability of smoking by 35 percentage points as of 1978-80, when men in the relevant cohorts were aged 25-30, but later in adulthood this effect was substantially attenuated and did not lead to large negative health effects.
View Full
Paper PDF
-
Wage Dispersion, Compensation Policy and the Role of Firms
November 2005
Working Paper Number:
tp-2005-04
Empirical work in economics stresses the importance of unobserved firm- and person-level characteristics
in the determination of wages, finding that these unobserved components account for the overwhelming
majority of variation in wages. However, little is known about the mechanisms sustaining these wage di'er-
entials. This paper attempts to demystify the firm-side of the puzzle by developing a statistical model that
enriches the role that firms play in wage determination, allowing firms to influence both average wages as
well as the returns to observable worker characteristics.
I exploit the hierarchical nature of a unique employer-employee linked dataset for the United States,
estimating a multilevel statistical model of earnings that accounts for firm-specific deviations in average
wages as well as the returns to components of human capital - race, gender, education, and experience -
while also controlling for person-level heterogeneity in earnings. These idiosyncratic prices reflect one aspect
of firm compensation policy; another, and more novel aspect, is the unstructured characterization of the
covariance of these prices across firms.
I estimate the model's variance parameters using Restricted (or Residual) Maximum Likelihood tech-
niques. Results suggest that there is significant variation in the returns to worker characteristics across
firms. First, estimates of the parameters of the covariance matrix of firm-specific returns are statistically
significant. Firms that tend to pay higher average wages also tend to pay higher than average returns to
worker characteristics; firms that tend to reward highly the human capital of men also highly reward the
human capital of women. For instance, the correlation between the firm-specific returns to education for
men and women is 0.57. Second, the firm-specific returns account for roughly 9% of the variation in wages
- approximately 50% of the variation in wages explained by firm-specific intercepts alone. The inclusion of
firm-specific returns ties variation in wages, otherwise attributable to firm-specific intercepts, to observable
components of human capital.
View Full
Paper PDF
-
Distribution Preserving Statistical Disclosure Limitation
September 2006
Working Paper Number:
tp-2006-04
One approach to limiting disclosure risk in public-use microdata is to release multiply-imputed,
partially synthetic data sets. These are data on actual respondents, but with confidential data
replaced by multiply-imputed synthetic values. A mis-specified imputation model can invalidate
inferences because the distribution of synthetic data is completely determined by the model used
to generate them. We present two practical methods of generating synthetic values when the imputer
has only limited information about the true data generating process. One is applicable when
the true likelihood is known up to a monotone transformation. The second requires only limited
knowledge of the true likelihood, but nevertheless preserves the conditional distribution of the confidential
data, up to sampling error, on arbitrary subdomains. Our method maximizes data utility
and minimizes incremental disclosure risk up to posterior uncertainty in the imputation model and
sampling error in the estimated transformation. We validate the approach with a simulation and
application to a large linked employer-employee database.
View Full
Paper PDF
-
Family Formation and the Great Recession
December 2020
Working Paper Number:
CES-20-42R
This paper studies how exposure to recessions as a young adult impacts long-term family formation in the context of the Great Recession. Using confidential linked survey data from U.S. Census, I document that exposure to a 1 pp larger unemployment shock in the Great Recession in one's early 20s is associated with a 0.8 pp decline in likelihood of marriage by their early 30s. These effects are not explained by substitution toward cohabitation with unmarried partners, are concentrated among whites, and are notably absent for individuals from high-income families. The estimated effects on fertility are also negative but imprecisely estimated. A back-of-the-envelope exercise suggests that these reductions in family formation may have increased the long-run impact of the Recession on consumption relative to its impact on individual earnings by a considerable extent.
View Full
Paper PDF
-
Run Effects of Military Service: Evidence from the 911 Attacks
November 2021
Working Paper Number:
CES-21-36
We investigate the effect of military service on labor market, health and family formation outcomes, leveraging differential changes in enlistment rates brought about by the September 11th attacks (911). Using restricted microdata, we identify hundreds of 'high service" counties in which certain birth-county cohorts exhibit large enlistment responses to 911. We find that individuals born into high service counties between 1977 and 1983 (aged 18-24 at the time of the attack), enlisted at nearly twice the rate of earlier birth cohorts (older than 24 at the time of the attack). These high service birth-county cohorts experienced a 10% increase in wages, decreased unemployment and impacts on other labor market measures as well as key household formation measures including marriage and fertility. We also find increases in the hospitalization and mortality rates. Labor market benefits outweigh mortality costs at standard discount rates.
View Full
Paper PDF
-
The Timing of Teenage Births: Estimating the Effect on High School Graduation and Later Life Outcomes
January 2016
Working Paper Number:
CES-16-39R
We examine the long-term outcomes for a population of teenage mothers who give birth to their children around the end of their high school year. We compare the mothers whose high school education was interrupted by childbirth, because the child was born before her expected graduation date to mothers who did not experience the same disruption to their education. We find that mothers who give birth during the school year are seven percent less likely to graduate from high school, are less likely to be married, and have more children than their counterparts who gave birth just a few months later. The labor market outcomes for these two sets of teenage mothers are not statistically different, but with a lower likelihood of marriage and more children, the households of the treated mothers are more likely to fall below the poverty threshold. While differences in educational attainment have narrowed over time, the differences in labor market outcomes and family structure have remained stable.
View Full
Paper PDF
-
Every Breath You Take, Every Dollar You'll Make: The Long-Term Consequences of the Clean Air Act of 1970
September 2013
Working Paper Number:
CES-13-52
This paper examines the long-term impacts of in-utero and early childhood exposure to ambient air pollution on adult labor market outcomes. We take advantage of a new administrative data set that is uniquely suited for addressing this question because it combines information on individuals' quarterly earnings together with their counties and dates of birth. We use the sharp changes in ambient air pollution concentrations driven by the implementation of the 1970 Clean Air Act Amendments as a source of identifying variation, and we compare cohorts born in counties that experienced large changes in total suspended particulate (TSP) exposure to cohorts born in counties that had minimal or no changes. We nd a signi cant relationship between TSP exposure in the year of birth and adult labor market outcomes. A 10 unit decrease in TSP in the year of birth is associated with a 1 percent increase in annual earnings for workers aged 29-31. Most, but not all, of this effect is driven by an increase in labor force participation. In present value, the gains from being born into a county affected by the 1970 Clean Air Act amount to about $4,300 in lifetime income for the 1.5 million individuals born into
these counties each year.
View Full
Paper PDF