Volume 12, Issue 2

This study examines the changing relationship between spousal health insurance coverage and labor market outcomes for married women over time as healthcare costs have increased. In particular, I investigate how husbands' health insurance coverage offers affect wives' decisions to enter the labor force and work full-time and how this has changed over time. I endeavor to correct for potential biases of these effects by 1) using an instrumental variables model to deal with endogeneity and 2) estimating and netting out likely unobserved heterogeneity biases, such as assortative mating or income effects. Using Current Population Survey data from 1995 to 2005, I find that husbands' employer-provided health insurance coverage has a negative effect on wives' labor supply that has increased (become more negative) over time.

Additional work hours may lead to weight gain by decreasing exercise, causing substitution from meals prepared at home to fast food and pre-prepared processed food, or reducing sleep. Substitution toward unhealthy convenience foods could also influence the weight of one's spouse and children, while longer work hours for adults may further impact child weight by reducing parental supervision. I examine the effects of adult work hours on the body mass index (BMI) and obesity status of adults as well as the overweight status of children. Longer hours increase one's own BMI and probability of being obese, but have a smaller and statistically insignificant effect on these outcomes for one's spouse. Mothers', but not mother's spouse's, work hours affect children's probability of being overweight. My estimates imply that changes in labor force participation account for only 1.4% of the rise in adult obesity in recent decades, but a more substantial 10.4% of the growth in childhood overweight.

This paper explores a central question in health economics: How sensitive is worker demand for health insurance? After controlling for variables omitted in other analyses, such as the generosity of plan coverage and aspects of worker demand that are constant within firms over time, I estimate a price elasticity (between -0.014 and -0.017) which is smaller than previous estimates. The analysis also finds that employees are more likely to take-up policies with greater insurance protection from hospital expenses, but not for increased coverage for prescription drug or provider office visit expenses. Taken together, increases in worker-paid premiums explain about 60 percent of the fall in take-up of employer policies over time, whereas increases in insurance cost-sharing explain about 10 percent of that change. Changes in employer contributions for health insurance had a limited effect on take-up compared with the amounts employees paid out-of-pocket for premiums. An implication of these findings is that policies which attempt to subsidize employee-paid portions of the premium would be an expensive and potentially ineffective strategy for achieving greater coverage, particularly if the quality of that coverage is not perceived as worthwhile.

An important, unresolved question for health policymakers and consumers is whether cigarette smoking in young adulthood has 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.

Something about being poor makes people fat. Though there are many possible explanations for the income-body weight gradient, we investigate a promising but little-studied hypothesis: that changes in body weight can—at least in part—be explained as an optimal response to economic insecurity. We use data on working-age men from the 1979 National Longitudinal Survey of Youth (NLSY79) to identify the effects of various measures of economic insecurity on weight gain. We find in particular that over the 12-year period between 1988 and 2000, the average man gained about 21 pounds. A one percentage point (0.01) increase in the probability of becoming unemployed causes weight gain over this period to increase by about 0.6 pounds, and each realized 50% drop in annual income results in an increase of about 5 pounds. The mechanism also appears to work in reverse, with health insurance and intrafamily transfers protecting against weight gain.

In the contingent valuation literature, anchoring bias poses problems when using an iterative bidding game to infer willingness to pay. This bias occurs when the willingness to pay estimate is sensitive to the initially presented starting value. More generally, whenever a survey format is used and not all of those contacted participate, selection bias raises concerns about the representativeness of the sample. In this paper, we estimate students' willingness to pay for student health care at Stanford University while accounting for both of these biases. As there is no cost sharing for students, we assess willingness to pay by having a random sample of students play an on line iterative bidding game. Our main results are that (1) demand for student health care is elastic by conventional standards; (2) ignoring anchoring bias would lead to a substantially biased measure of the demand elasticity; and (3) standard selection correction methods indicate no bias from selective non-response and newer bounding methods support this conclusion of elastic demand.

Background: Heart disease is among the leading causes of death in the U.S. and Canada. Despite the U.S.'s higher spending on health care, it is unclear whether persons with heart disease fare better in one country or the other.Methods: To evaluate and compare the health of people aged 45 and older in the U.S. and Canada, we drew upon the Joint Canada-U.S. Survey of Health (JCUSH), a random telephone interview conducted from 2002 to 2003. We used self-reported fair or poor health, disability, and functional impairment as dependent variables in logistic regressions, which controlled for demographic variables and other risk factors.Results: Adjusting for covariates, Canadian respondents with heart disease reported better health as measured by disability, but there was no difference for functional impairment or self-reported fair or poor health. The odds ratios (Canada:U.S.) were 1.10 (p=0.69) for fair or poor health, 0.56 (p=0.06) for disability, and 0.78 (p=0.32) for functional impairment.Conclusions: Our results indicate that people with heart disease are in better health in Canada as measured by disability, but there is no difference for overall self-reported health or functional impairment. Further research must be done to determine the cause of outcomes differences among heart disease patients.

There has been growing interest in using specific genetic markers as instrumental variables in attempts to assess causal relationships between health status and socioeconomic outcomes, including human capital accumulation. In this paper, we use a combination of family fixed effects and genetic marker instruments to estimate the causal effects of poor adolescent mental and physical health status on years of completed schooling. By exploiting the genetic variation in inheritance within families, this empirical strategy presents a unique opportunity to isolate the variation in genetic factors from other dynastic and family characteristics. We present evidence that inattentive symptoms in early childhood have large lasting effects in reducing completed schooling. We also find little consistent evidence that adolescent overweight status influences years of schooling completed.