Human Capital and Health Behavior: Volume 25

Virtually all parents want their children to succeed academically. How to achieve this goal, though, is far from clear. Specifically, the temporal spacing between adjacent births has been shown to affect educational outcomes. While many of these studies have produced substantial and statistically significant results, these results have been relatively narrow in their application due to data limitations. Using Colorado birth certificates matched to schooling outcomes, we investigate the relationship between birth spacing and educational attainment. We instrument birth spacing with a previous pregnancy that did not result in a live birth. We find no overall effect of spacing on either the first or second children’s grade 3–10 test scores. Stratifying by the sexes of the children, we find that when the first child is a boy and the second a girl, an extra year of spacing increases the first child’s math, reading, and writing test scores by 0.07–0.08 SD, while there is no impact on the second child. This is the first study to do such an analysis using matched large-scale birth and elementary to high school administrative data, and to leverage a very large dataset to stratify our results by the sexes of the children.

Empirical studies show that years of schooling are positively correlated with good health. The implication may go from education to health, from health to education, or from factors that influence both variables. We formalize a model that determines an individual’s demand for knowledge and health based on the causal effects, and study the impacts on the individual’s decisions of policy instruments such as subsidies on medical care, subsidizing schooling, income tax reduction, lump-sum transfers, and improving health at young age. Our results indicate that income redistribution policies may be the best instrument to improve welfare, while a medical care subsidy is the best instrument for longevity. Subsidies to medical care or education would require large imperfections in these markets to be more welfare improving than distributional policies.

Child immunization is widely recognized as a cost-effective preventive medicine. Unfortunately, in India about 50% of the eligible children aged 12–23 months miss some essential vaccination. Though a positive association between maternal education and markers of child health like immunization has been long established, the literature has struggled to find a causal relationship, mainly because education is inextricably correlated with other socioeconomic variables like income. In this chapter, I propose a new instrument for women’s education in India using the following facts. First, due to lack of sanitary facilities in schools, particularly rural schools, large number of girls drop out of school once they reach puberty. Second, age at menarche is largely determined by biological factors and not social factors. Together, age at menarche can explain variations in schooling, yet be independent of outcome variables like child immunization. I find that additional years of maternal schooling (conditional on strictly positive years of schooling) do increase the probability of complete immunization of children.

This study investigates the effects of a broad-based policy change that altered maternal employment, family income, and other family characteristics on drug-related crime among youth. Specifically, we exploit differences in the implementation of welfare reform in the United States across states and over time in the attempt to identify causal effects of welfare reform on youth arrests for drug-related crimes between 1990 and 2005, the period during which welfare reform unfolded. We use monthly arrest data from the U.S. Federal Bureau of Investigation’s Uniform Crime Reports to estimate the effects of welfare reform implementation on drug-related arrests among 15- to 17-year-old teens exposed to welfare reform. The findings, based on numerous different model specifications, suggest that welfare reform had no statistically significant effect on teen drug arrests. Most estimates were positive and suggestive of a small (3%) increase in arrests.

The U.S. 2009 Tobacco Control Act opened the door for new antismoking policies by giving the Food and Drug Administration broad regulatory authority over the tobacco industry. We develop a behavioral welfare economics approach to conduct cost-benefit analysis of FDA tobacco regulations. We use a simple two-period model to develop expressions for the impact of tobacco control policies on social welfare. Our model includes: nudge and paternalistic regulations; an excise tax on cigarettes; internalities created by period 1 versus period 2 consumption; and externalities from cigarette consumption. Our analytical expressions show that in the presence of uncorrected internalities and externalities, a nudge or a tax to reduce cigarette consumption improves social welfare. In sharp contrast, a paternalistic regulation might either improve or worsen social welfare. Another important result is that the social welfare gains from new policies do not only depend on the size of the internalities and externalities, but also depend on the extent to which current policies already correct the problems. We link our analytical expressions to the graphical approach used in most previous studies and discuss the information needed to complete cost-benefit analysis of tobacco regulations. We use our model as a framework to reexamine the evidence base for strong conclusions about the size of the internalities, which is the key information needed.

I investigate the well-known educational gradient in smoking. It is well established that, at least in recent decades, people with higher levels of education are less likely to smoke and, conditional on being a smoker, are more likely to quit than are people with less education. Using longitudinal data on lifetime smoking histories, I explore whether the educational gradient changes when one accounts for differences in the amount of information smokers have about the health risks associated with smoking. At the core of the analysis is a new way to measure not only the flow of information a person receives but also a person’s stock of information in any year. I construct measures of the stock and flow of information with consumer magazine articles that discuss cigarette smoking and health. To calculate exposure, I predict individuals’ reading of particular magazines and link predicted exposure to data on individual smoking status in every year of life. The analysis sample includes many individuals who started smoking in the 1930s and 1940s – well before scientific evidence had accumulated. After replicating the education gradient in terms of smoking cessation, I show that it is mostly explained by the interaction between educational attainment and the stock of knowledge individuals possess. The findings suggest that education affects whether and how a stock of health risk information induces people to quit smoking.

Misinterpretation of a negative test results in health screening may initiate less preventive effort and more future lifestyle-related disease. We predict that misinterpretation occurs more frequently among individuals with a low level of education compared with individuals with a high level of education.

The empirical analyses are based on unique data from a randomized controlled screening experiment in Norway, NORCCAP (NORwegian Colorectal Cancer Prevention). The dataset consists of approximately 50,000 individuals, of whom 21,000 were invited to participate in a once only screening with sigmoidoscopy. For all individuals, we also have information on outpatient consultations and inpatient stays and education. The result of health behaviour is mainly measured by lifestyle-related diseases, such as COPD, hypertension and diabetes type 2, identified by ICD-10 codes.

The results according to intention-to-treat indicate that screening does not increase the occurrence of lifestyle related diseases among individuals with a high level of education, while there is an increase for individuals with low levels of education. These results are supported by the further analyses among individuals with a negative screening test.

Unemployment insurance (UI) reduces the opportunity cost of leisure, but it is unknown whether this additional leisure time is physically active. To obtain unbiased estimates of the effect of UI on physically active leisure participation, I exploit changes in UI program legislation across US states and time. Using nationally representative monthly data between 2003 and 2010 from the Behavioral Risk Factor Surveillance System (BRFSS) and the American Time Use Survey (ATUS), I find evidence that both state UI eligibility expansions and increases in maximum allowable state UI benefits coincide with greater probability of physical activity among the recently unemployed. Based on point estimates, state UI eligibility expansions increased the probability of physical activity participation by 8–10 percentage points among the unemployed with less than a high school education, while a 10% increase in the maximum allowable state UI benefit increased the probability of physical activity by 0.3 to 0.6 percentage points among the unemployed who have completed high school or some college.