Home -> Abstracts
Abstracts of Recent Work
This paper tests several leading bequest motive theories using a
uniquely appropriate longitudinal data set, the Mexican Health
and Aging Study. These data include a population-representative
sample of bequests and bequest intentions of parents that is
matched with rich measures of child characteristics and
behavior. Our results are consistent with the theory of
Bernheim et al. that parents use their bequest strategically to
induce children to provide services. We also find evidence that
contradicts predictions of pure theories of altruism and
suggestive evidence that when business assets are at stake,
parents favor those children who are most qualified to manage
those assets.
Recent evaluations of the Oportunidades schooling and health subsidy
program in Mexico have demonstrated statistically significant positive
impacts on schooling and health outcomes. This paper adapts methods
developed in Dinardo, Fortin and Lemieux (1996) for use in studying
how these schooling and health impacts will affect the future earnings
distributions of cohorts recently exposed to the program. Our approach
nonparametrically simulates earnings distributions, with and without the
program, and quantifies resulting changes in mean earnings,
poverty rates, and earnings inequality. It is well recognized that the
Oportunidades program has reduced poverty and inequality of the
current generation through its targeted cash transfers. This paper finds
that by enriching human capital, as measured by schooling and height,
the program will also generate increases in future earnings. However, it
will achieve only modest reductions in poverty and earnings inequality.
In this project, we propose a new method to estimate rich
dynamic models of health that exploits longitudinal observations of
multiple health measures. Our method adapts and combines two techniques
first developed in different contexts. In a first step, we use factor
analytic methods to estimate a series of age-specific static measurement
models that determine how underlying latent health is related to
observed discrete and continuous measures. This step also reveals the
unconditional nonparametric distribution of latent health at each age.
We then model how latent health evolves stochastically over time. We
estimate the parameters of this dynamic model using the method of
simulated moments. Specifically, we simulate the dynamic health process
and use the previously estimated measurement process to derive an
implied set of moments that can be compared with moments observed in the
data. We demonstrate the method by estimating health processes for men
and women using data from the Health and Retirement Study.
It is well known that among elderly Americans, education
and good health are positively correlated. We contribute to this
literature both methodologically and empirically. We develop a
sophisticated model of the measurement and evolution of health and
estimate it using a new method. Our measurement model allows us to
exploit continuous measures of physical health including grip strength
and lung capacity, as well as more subjective ordinal measures such as
self-rated health status and reported difficulties with activities of
daily living. Our dynamic model allows us to explore how endogenous
mortality, severe health crises (such as heart attacks or cancers), and
more gradual processes affect aging. We estimate both models separately
for low and high education groups in the Health and Retirement Study.
Our results reveal how much inequality in old age is driven by
differences in health at age 50 (the beginning of our sample) and how
much stems from differences in the aging process.
There is little consensus about the optimal role of
technology in education, and unsurprisingly, most classrooms around the
world don't look very different from classrooms one hundred years ago.
Intel Teach for the Future is a program that trains teachers how to
incorporate information and communications technology into their
classrooms. Over the last ten years, the program has been implemented
in more than 40 countries and in Costa Rica has reached more than 30% of
the public elementary and high school teachers. This high level of
penetration provides a unique opportunity to evaluate the program's
effect on student outcomes. In this project, we match data on program
roll out over time across the country to nationally representative
household survey data. We use both temporal and spatial variation to
estimate causal effects of the program on student retention,
college-going, employment, and wages.
Across birth cohorts of Americans, education and smoking
status in families of origin have become more aligned. Over time, men
who smoke became more likely to marry women who smoke, especially among
couples with less schooling. We examine how much this alignment of
smoking and education matters for children's life chances. We use a
two-sex demographic projection model, which accounts for the statuses of
both men and women, combined with simulations to examine how changes in
assortative mating affect the distribution of smoking and education in
future generations. The model incorporates assortative mating,
differential fertility, and the transmission of status. Preliminary
findings show that intergenerational results depend on which cohort's
marriage and fertility rules are applied and whether we consider smoking
status as a binary status or a quantitative one. But these differences,
though descriptively important, are modest. Mobility across generations
dominates the effects of assortative mating in the
population.
In developed countries, most workers depend on government-based social
security systems and private company-based pensions to smooth
their consumption. In the developing world, the mechanisms are
quite different. In the absence of significant formal institutional
support, older individuals rely on their own labor income and on family
support in the form of transfer payments, coresidence, and participation
in family businesses. However, the dramatic gains in life expectancy and
declines in family size and coresidence that accompany development
suggest that these traditional forms of support may break down. In this
paper I build and estimate a structural dynamic model of labor supply
for older men in Indonesia that incorporates these mechanisms. I
use this model to simulate the effects of demographic change on labor
supply and to evaluate the labor supply and welfare effects of a broad
public pension reform, which many believe may be a way to address the
growing needs of aging populations in developing settings. The
estimation results show that families and health play a key role in
labor supply choices in old age. When faced with a 50% reduction in
family support, simulations show that older men, especially those
in the informal sector, stay in the labor force at ages when they would
otherwise exit and experience a substantial decline in welfare.
Simulations also show that a unified defined-contribution pension
program for government and private sector workers would provide modest
welfare gains at a reasonable cost, but would not offset the potential
welfare losses brought by declines in family support. These results
highlight the complexities of old age labor supply and pension reform in
the context of rapidly developing societies.
In 1997 the Mexican government instituted a major reform that dismantled
the existing defined benefit pension program and created one of the
world's largest systems of privately managed individual retirement
accounts for all formal sector workers. In this paper I develop and
estimate a dynamic structural model of labor supply and saving that
addresses several key questions. First, how have individuals changed
their retirement and saving behavior under the new system? Second, how
has the new incentive structure changed the flows of workers between the
formal and informal sectors? And third, how have informal family
systems of support responded to the reform? For this project, I pool
data from two rich longitudinal data sets. The Mexican Health and Aging
Study (MHAS) interviews a sample of individuals over age 50 in 2001 and
re-interviews them in 2003. The Mexican Family Life Survey (MxFLS),
conducted in 2002 and 2005, samples from the entire Mexican population
and allows observation of the labor supply and saving responses of
younger individuals.
We combined data from a population-based longitudinal survey with
satellite measures of aerosol levels to assess the impact of smoke from
forest fires that blanketed the Indonesian islands of Kalimantan and
Sumatra in late 1997 on adult health. To account for unobserved
differences between haze and non-haze areas, we compared changes in the
health of individual respondents. Between 1993 and 1997, individuals who
were exposed to haze experienced greater increases in difficulty with
activities of daily living than did their counterparts in non-haze
areas. The results for respiratory and general health, although more
complicated to interpret, suggest that haze had a negative impact on
these dimensions of health as well.
This study examines how the wage returns to schooling, early work experience,
and unobserved skills changed for young men and women in the
United States over the latter half of the twentieth century. Our
analysis focuses on the experiences of young men and women from two
different birth cohorts--one group that was of high school age during
the second half of the 1960s and a second that began their transition
from school to work in the late 1970s and early 1980s. We pay particular
attention to how differences across cohorts in these transitions
vary by gender and race/ethnicity and how these differences affect
subsequent wage attainment. We examine returns based on several
different estimation strategies, including one that attempts to deal
with both the endogeneity of accumulated work and schooling experiences
and selection bias in our wage data.