Determinants of Health

1 Introduction

Economic researchers have long recognized the interplay between health and economic outcomes at both the macro and micro levels. Researchers have documented the associations between health and wealth of both individuals and states and have explored extensively how improvements in health affect several outcomes of economic interest such as human capital, labor market success, social programs, and aggregate GDP, to name a few. While the relationship between health and economic outcomes takes place over all periods of the human life-cycle, an increasing body of literature has focused on the economic determinants of health and economic consequences of health in childhood, based on an increasingly accepted notion that broad investments in children produce returns over the entire life-cycle.

Evidence from the epidemiology literature argues for strong linkages between early child health and adult health. The work by David Barker, commonly referred to as the “fetal origins hypothesis,” suggests that poor fetal health is related to higher adult risk of disease, particularly coronary heart disease and type 2 diabetes (Barker 1998; Gluckman and Hanson 2006) and has served as a starting point for much of the economic research in this area. In her recent review of the child health literature, Currie (2009) notes that this research can been seen in the broader context of research which explores the interactions between genes and their environment where individuals may be genetically predisposed to certain health problems, but interactions with various environmental factors lead to the manifestation of the health conditions. Work by Heckman and colleagues (2007) extends this argument beyond childhood health and lays out a dynamic model of the effects of past health and human capital stock on the ability to accumulate current health and human capital (outlined in Cunha, Heckman, and Schennach 2006) to the broader notion that early childhood investment lays a foundation for the success of future investments and therefore future economic payoffs.

The notion that health acts as an input into the human production of other valuable goods and that the “health stock” itself is a function of current and past investments is based on the well-known theoretical work of Grossman (1972). Grossman’s notion of health as an input into the production of both work and leisure, and investments in the health stock having long-run payoffs can easily be extended back to childhood to provide an organizing framework for much of the literature reviewed in this chapter.

Children are born with a stock of health, H0. Children receive insults to their health in the form of chronic conditions (both mental and physical), diseases, injuries, etc. The health stock can also be augmented with parental investments of both time and money so that the health stock next period is a function of the health stock in the previous period, investments made to health, and any realized insults to the child’s health. A simple representation of this function would then be:

Ht = f(Ht−1ItPtΩ),

where It represents insults to the child’s health at time tPt represents parental investments of time and money at time t, and Ω denotes parental endowments (such as IQ, education, income, genes, etc.).

This health stock, while providing utility to a child, will also serve as an input into a variety of other goods and stocks. Importantly, health will be a key input into human capital accumulation. At the most basic level, a child must be well enough to go to school, with a health stock of, say, H(min). However, beyond this, variation in the stock of physical and mental health will affect the child’s ability to learn and participate at school and acquire human capital, E. Human capital, therefore, is a function of, among other things, the health stock at time t, or Et(HtXtΩ), where Xt represents other inputs (parental and otherwise) into the ability to acquire human capital. This, in turn, will affect future economic outcomes such as labor market earnings. We would expect that children from families with more resources would have, on average, a higher level of health stock (∂H/∂I>0). We would also expect that insults to health are not completely exogenous, but rather depend in part on the child’s environment (housing stock, neighborhoods, etc) such that children from families with fewer resources would have lower health (∂H/∂Ω>0), and may receive more shocks to their health (∂I/∂Ω>0). It may be the case that children from families with more resources will be able to mitigate the effects of health shocks more than families without resources (through better information, or better medical treatment).

Heckman (2007) describes the notion of “dynamic complementarity” in the case of human capital accumulation as arising when “stocks of capabilities acquired in the previous period make investment in the [current] period more productive. Such complementarity explains why returns to educational investments are higher at later stages of the child’s life-cycle for more able, more healthy, and more motivated children” (Heckman 2007: 13253). In this simple representation, health stocks in previous periods contribute to the current health stock, which then contributes to current human capital accumulation.

The empirical studies reviewed in this chapter broaden the general ideas presented above to examine the validity of the notion that health in childhood has both short-term and longer-term economic consequences, and that childhood health is itself a function of a broader set of policies, investment decisions and parental choices, and a function of parental health stock and economic background. While these areas are not solely explored by economists, economists have made significant contributions over the past decade or so by focusing on large sample analysis, and pressing on the identification of causal relationships. This review focuses on examples of recent economic research that have made such contributions to this literature.

While Grossman left the definition of health to be broadly defined, the empirical research reviewed here uses a broad set of survey and administrative instruments to construct a set of health measures over the life of the child across several dimensions. Childhood health at birth is measured using birth weight, gestational length, and Apgar scores (which measure the overall health of the child both five minutes and ten minutes after birth). As children age, several surveys and screeners are used to capture the development of cognitive ability, behavioral development, and physical health and development. Test scores probe picture recognition and vocabulary in early years, advancing to more standard tests of math and reading comprehension. Behavioral screeners are used to capture symptoms of common mental illnesses such as attention deficit hyperactivity disorder, aggression disorders, depression, and others at various stages of the child’s development. Measures of illnesses, chronic conditions, injuries, and measure of height and weight are used to examine physical health problems, acute health shocks, and physical growth and development.

The literature reviewed here strongly suggests that influences on the child, from pre-birth to adolescence, have important implications for both health and other markers of success such as education. Socioeconomic indicators and environmental shifts influence these relationships, and various studies with plausible causal identification suggest that the effects can be quite large. The extent to which policy levers can and do have large impacts on these relationships varies. Some, such as maternal leave policies and tax and subsidy policy appear to affect various margins of behavior and health, while others have proven less effective. Overall, there remains considerable scope for research on the effects of child health on the life-course and the plausible policy levers that can improve well-being.

2 Infant and Early Childhood Health

2.1 Health at Birth and Childhood Well-being

It seems evident that fetal health and health in early infancy would have significant effects on later infant health outcomes. But, since fetal health and health in early infancy, as well as later health, depend on family environment, socioeconomic status (SES), and other factors that may not be observed by researchers, it has been challenging to estimate the causal elements in this link. Several recent papers have examined the role of early infant health on survival and early infant health outcomes.

Many of these studies use large administrative samples and track children from birth onward. The studies seek to estimate causal relationships between infant health and future outcomes using variation within families and within twins in order to control for omitted variables common across families, or between twins, that may be related to both poor health and poor outcomes such as various aspects of socioeconomic status. The most common measures of health at birth found in the literature are birth weight, Apgar scores, and gestational length. In general these measures are considered more objective than survey measures of infant health. Weight at birth is considered low if it is below 2500 grams, and very low below 1500 grams. Gestational periods are considered premature if they are below 37 weeks. Apgar scores are based on five items and scored out of 10. Scores below 7 are considered poor.

One set of economic studies have examined the effects of poor infant health on short-term survival. For example, Almond, Chay, and Lee (2005) examine the relationship between low birth weight, low Apgar scores, and mortality in the first year of life. Using a large sample of twin births from the National Center for Health Statistics (US data), they show that, while both birth weight and Apgar scores are strongly related to infant mortality across families, the relationship between birth weight and infant mortality significantly decreases when differences between twins are examined. In contrast, the relationship between Apgar scores and infant mortality remains strong both across families and within twin pairs.

A second stream of social science literature has used twin studies to examine the longer-term effects of birth weight on health and education. Behrman and Rosenzweig (2004) use twin data from the Minnesota Twins Registry to examine the effects of low birth weight on the educational attainment and adult health of women. They find that increasing birth weight increases schooling attainment by about one third of a year and that this effect is stronger within twins than across children of different families.

Conley, Strully, and Bennett (2003) examine the effects of low birth weight on high school graduation and placement in special education using the Panel Study of Income Dynamics. They find that the effects of low birth weight on timely high school graduation are more pronounced among siblings than across families. This suggests that the within-family differences, that is the differences in birth weight between siblings, account for much of the relationship between birth weight and educational attainment. Differences in birth weight between families account for less of this relationship. The study does not look at other measures of infant health (Apgar and gestation) nor does it explore the potential non-linear effects of low birth weight on infant health.

Evidence using data from other nations extends many of the findings in the United States. Currie and Hyson (1999) show that, conditional on many measures of family background and circumstances, low birth weight children from the 1958 British birth cohort have lower test scores, educational attainments, wages, and probabilities of being employed as of age 33. Black et al. (2007) use a sample of Norwegian twins to examine the long-run consequences of low birth weight. Their evidence confirms that low birth weight is not a good predictor of infant death within twin pairs. However, they do find long-term effects of low birth weight on cognitive outcomes, educational outcomes, and on earnings. In particular, birth weight has long-term effects on height, IQ, earnings, and education. Oreopoulos et al. (2008) use administrative data from the Canadian province of Manitoba and find both low birth weight and low Apgar scores to be strong predictors of both high school completion and welfare take-up and length.

The evidence that early infant health matters in both the short and longer term is large and fairly consistent. What is less clear is how these early measures of health affect longer term health—whether they are early indicators of future health problems, whether they trigger environmental responses within or outside the family that have longer-term consequences, or whether there is some other explanation for the observed linkage between health at birth and outcomes throughout the life-course. Understanding these mechanisms, and further research into the role of economic inputs into child health over the life-course are both important next steps for research and key elements for policymakers seeking to improve population outcomes.

2.2 Environmental Factors and Infant Health

Environmental contamination such as pollution is commonly thought to have particularly strong effects on the old and very young. Some recent research attempts to investigate a relationship between air pollution and health in infancy and childhood.

Chay and Greenstone (2003a) use variation in pollution induced by the 1980–82 US recession to examine its effect on child hood health. They report that reductions in total suspended particulates lead to reductions in infant mortality, primarily in the neonatal period. The recession is estimated to have reduced infant deaths by 2500 in this period. Supporting evidence that uses variation in pollution induced by the 1970 US Clean Air Act is provided by Chay and Greenstone (2003b).

Neidell (2004) examines the impact of pollution on childhood asthma exploiting seasonal variation in pollution levels within place. Carbon monoxide is found to affect asthma, with a greater impact for children of lower SES. Clearly environmental factors play a contributing role in development of childhood health, both on their own, and through interactions with SES. These relationships, with plausible causal pathways, are informative to policymakers looking to prevent longer term health problems through focusing on the socioeconomic circumstances and living environment of the child.

2.3 Maternal Employment

A prominent focus of economic research on child health and development is the roles of maternal employment and parental and non-parental care, including interventions that provide early childhood education. An important economic dimension of this research is the mother’s decision to work in the period following giving birth.

Klerman and Leibowitz (1997) outline a static model of the decision to work post-birth. Employers may voluntarily choose to offer a maternity leave out of a desire to preserve match or job specific human capital. Mothers choose between whatever leave is offered and the option of quitting. An optimal duration of post-birth absence from the labor market is chosen in light of the difference in wages between the current and best alternative jobs and a reservation wage that declines with each month post-birth.

In most developed countries governments intervene in this private transaction by mandating a minimum period of job protected, and perhaps compensated, maternity leave. Interestingly, Klerman and Leibowitz point out that the predicted impact of such legislation on time at home post-birth is ambiguous, as some mothers choose to stay home longer in the presence of a mandate while others will stay home a shorter period.

An important issue in empirical research investigating the impact of maternal employment on children’s health is omitted variables bias. The problem is that variables that are not included as covariates in the estimating equation that are correlated both with a mother’s decision to work and the outcomes of her children. These variables may not be available because they simply were not collected in the survey data, or they may be fundamentally unmeasurable. There are two common approaches to this problem in this literature. The first is to seek out data sets with extensive arrays of information about mothers and their families, and to include these variables as controls in the estimating equation. The hope here is, in effect, that there are no omitted variables once all these covariates have been included. The second is to exploit variation in maternal employment induced by maternity leave mandates, or other public policies that affect the labor market decisions of mothers with young children. In this case the assumption is that this variation in employment is not correlated with the omitted variables in the error term of the estimating equation. A noted limitation of this approach is that the result is usually a local average treatment effect specific to the sub population whose behavior is affected by the policy.

At the very beginning of a child’s life maternal employment will interrupt the first months of maternal care. During this period mothers recover from giving birth, bond with their children and importantly have the opportunity to breastfeed. Perhaps the strongest advice for childhood nutrition is that babies initially be breastfed. The World Health Organization recommends 6 months of exclusive breastfeeding, counsel echoed by many public health and medical associations. In many countries there is further guidance to include breast milk in children’s diets up to age 2.

Many benefits are attributed to breast milk including reduced rates of mortality, respiratory ailments, gastro-intestinal diseases and allergies for children and lower incidence of ovarian and premenopausal breast cancer for mothers. Many reviews of the literature note, however, that the supporting evidence is almost exclusively observational and the contribution of confounding factors may be important (e.g. Horta et al. 2007). There appears to be only one study of the benefits of breastfeeding that adopts a randomized design (Kramer et al. 2001).

Although breastfeeding data is not collected systematically in many countries, there appears to be a consensus that practice falls well short of the desired behavior (e.g. UNICEF 2008). Promotion of breastfeeding targets both initiation and duration. Surveys of mothers in developed countries reveal physical and technical difficulties lead the reasons for ending breastfeeding at short durations while work related issues are the leading reason at longer durations (Hamlyn et al. 2002, Lansinoh Laboratories 2005; see also Schwartz et al. 2002). It is therefore not surprising that the very small economic literature on breastfeeding focuses on the mother’s decision to return to work post-birth.

A number of studies have documented the association between shorter breastfeeding duration (although not incidence) and the return to work (surveyed in Dennis 2002). A caution for much of this evidence is that some omitted factor may drive both breastfeeding and work behavior.

Roe et al. (1999) use an instrumental variables (IV) approach to sort out the causality assuming that mother’s occupation affects the return to work decision but not breastfeeding decisions. They conclude that causality flows from work to breastfeeding and longer periods at home post-birth are associated with longer breastfeeding durations.

Chatterji and Frick (2005) use a family fixed effects framework to control for unobserved family background factors. They report that return to work within three months of birth reduces the probability of initiating breastfeeding by 16–18 percent and a reduction in duration (among those who initiate) of 4 to 5 weeks. Complementary evidence is reported by Berger et al. (2005) who, using propensity score matching, report return to work within the first three months reduces the incidence of breastfeeding by 13 percent and the duration by 4.5 weeks.

An alternative approach to the omitted variables problem is to use policy induced changes in behavior. Haider et al. (2003) examine the impact of the work requirements of recent welfare reform in the US on breastfeeding rates. They estimate that these reforms reduced national breastfeeding rates at six months post-birth by over 5 percent.

Baker and Milligan (2008a) use a recent legislated increase in job protected maternity leave (in Canada) from six months to one year to explore the relationship between time at home post-birth and breastfeeding. They report little impact on the incidence of breastfeeding, but an increase in the number of months babies were breastfed in their first year of life of one month. This can be compared to the increase of over three months in the amount of time mothers were at home post-birth as a result of the reform. The increase in the proportion of mothers exclusively breastfeeding at six months was over 39 percent. They also examine a collection of parent reported measures of their children’s health, focusing on respiratory ailments, finding the increases in breastfeeding had little impact.

It is clear that the recent economic studies in this area depart from most of the breastfeeding literature in that the question of causality if not answered is at least actively pursued. This is a welcome development given the central role breastfeeding is thought to play in infant health.

What other outcomes might be affected by a mother’s decision to work? One focus of research is childrens’s cognitive development. Older studies in this area offer mixed results and more recent contributions do not completely resolve the uncertainty. A notable feature of this research is the number of studies that use the same data set (the National Longitudinal Survey of Youth, NLSY).

A starting point is Ruhm’s (2004) study using NLSY data that measures cognitive development using the Peabody vocabulary, reading and math tests. His empirical strategy is to regress child outcomes on an indicator of mother’s employment controlling for a large number of observable child, mother and household characteristics to account for differences across mothers who do and do not choose to work. He finds maternal employment in the first three years has a small negative effect on the verbal abilities of 3–4-year-olds and a larger negative effect on the math and reading abilities of 5–6-year-olds. For example, an additional twenty hours of work in the first three years is estimated to reduce PIAT reading scores by 0.11 standard deviations and PIAT math scores by 0.08. Any work in the first year is estimated to reduce these scores by 0.08 standard deviations.

Complementary evidence for work in the first year, using the same data and test scores, is provided by Baum (2003c). Berger et al. (2005), using the same data, estimate a negative impact on PPVT of returning to work with twelve weeks of birth but most of the estimates are not statistically significant. Using a similar empirical approach, Hill et al. (2005) report full time work before eighteen months has small negative effects on cognitive development measured at ages 4–7 in the UK.

Alternative approaches to the omitted variable problem lead to both supportive and dissenting conclusions. James-Burdumy (2005), again using NLSY data and a family fixed effects estimator, finds that first year employment only negatively affects PIAT reading scores, and work in the second or third years has either no or positive impacts on test scores.6 Bernal (2008) estimates a dynamic model of women’s labor supply and child care decisions post-birth and finds a year of maternal full time employment in the first five years of life leads to a reduction in the NLSY cognitive scores of 1.8 percent (0.13 standard deviations).

It is important to note that there is a large complementary body of research by sociologists and developmental psychologists among others that is frequently cited in economic studies and often uses similar methods. Waldfogel (2006) provides a summary of the results of much of this research concluding maternal employment in the first year of life negatively affects cognitive development while work at later ages does not.

A majority of this research adopts an observational approach, relying on a battery of control variables to account for mothers’ decisions to work. Given the persistent disagreement in the conclusions of these studies a priority for future research should be the development of alternative empirical strategies.

A very related although not identical literature investigates the impacts of different types of non-parental care on child development. For many families, however, maternal employment and non-parental care go hand-in-hand, and in many data sets the detail on the type of non-parental care is limited, so the difference between the two literatures is practically (although not conceptually) of small significance.

An ongoing area of research here is the impact of the US Head Start program for disadvantaged children.7 Currie and Neidell (2007) attempt to discover what characteristics of the program have the strongest beneficial effects. They find that children in areas with higher Head Start spending have better reading and vocabulary scores and grade retention is lower in areas where a higher proportion of expenditures are on child centered (health and education) activities.

There are also studies of various types of pre-school programs. Loeb et al. (2007), using an array of observable characteristics to control for selection into these centers, report that they raise language, pre-reading, and math skills by 10 percent of a standard deviation, with the largest gains for children from the lowest income families. Magnuson et al. (2007) report on the impacts of pre-kindergarten, pre-school, Head Start, and other non-parental care. They find that both pre-kindergarten and pre-school raise kindergarten reading and math scores although the effects largely dissipate by grade 1.

A study of more generic non-parental care turns up negative impacts. Bernal and Keane (2008) exploit US welfare reform in the 1990s for identification, and find for the children of single mothers the use of childcare leads to a 2.9 percent reduction in the NLSY cognitive scores, a result that is driven by informal (e.g. family childcare) rather than formal childcare. Informal daycare, while used by many parents, has not been a focus of past research and deserves greater attention in the future.

As noted by Ruhm (2004) the mechanisms for these cognitive effects are not well understood. In a recent study, Cawley and Liu (2007) use time use data to shed light on this question. Using state unemployment rates as an instrument for mothers’ employment, they report working mothers are less likely to read to their children or help them with their homework, or spend less time in these activities if they do them. Likewise they expend less time playing with and supervising their children. The average age of the children in this study is 7. Baker and Milligan (2010) provide some evidence of the impact of mothers’ time at home in the first year of life based on an expansion of Canada’s mandated maternity leave from six to twelve months. They report a large increase (50%) in the amount of time mothers were at home in the first year, but no corresponding impacts on children’s temperament, motor and social skills or the amount of activities (e.g. reading, play) with their mothers, up to 29 months of age.

Another child outcome that has been associated with mothers’ employment is children’s mental health and behavioral development. This literature took a turn with the publication of findings from the National Institute of Child Health and Human Development (NICHD) Study of Early Child Care (e.g. NICHD 2003). This observational analysis indicated that the amount of time through the first 4.5 years of life that a child spends away from his or her mother is a predictor of assertiveness, disobedience, and aggression.

Confirmations of these negative effects are offered in a number of economic studies in this area. In a nationally representative US sample, Loeb et al. (2007) finds that detrimental effect of centre based non-parental care on behavior is increasing in the hours of exposure, and is particularly pronounced when entry into care is before the age of 1. Using the NLSY, Berger et al. (2005) find that the children of mothers who returned to work within twelve weeks of birth displayed more externalizing behavior problems at age 4 as measured by the Behavior Problems Index.

Baker et al. (2008b) study the introduction of subsidized universal childcare in the Canadian province of Quebec. Using comparisons of Quebec children to children in other provinces, they find a sharp increase in the use of non-parental care is matched by negative impacts on measures of behavior (aggressiveness, anxiety) as well as motor social development.

Another source of evidence is from studies of the effects of pre-kindergarten and pre-school programs. Magnuson et al. (2007) report that attendance in either a pre-kindergarten program or a pre-school is associated with elevated levels of externalizing behavior and lower levels of self control in kindergarten, with larger impacts for pre-kindergarten. Unlike the impacts on cognitive development reported above, these impacts on behavior persist into first grade.

The increasing number of studies that find these negative behavioral effects of maternal employment/non-parental care provide a challenge for future research in this area. Integrating these findings with evidence of the positive cognitive effects of some programs is one priority. Perhaps more important is that they highlight the limitations of focusing on only a single dimension of child health and development in program evaluations.

A final focus of recent research in this area are effects of maternal employment on the medical care of children and the incidence of specific ailments. Gordon et al. (2007) examine the impact of maternal employment and non-maternal care on the health of children aged 12–36 months using NICHD-SECC data and a mother–child fixed effects specification. They find little effect on the incidence of either infectious disease or injury. However, time in centre based care increases the rates of ear infections for children aged 12–24 months and the rates of respiratory illness for children aged 12–36 months. Baker et al. (2008) provide evidence that non-parental care increases the odds of both ear infections and nose/throat infections among Canadian children, and their results by age (0–2 years, 3–4 years) suggest the effect is present in both centre based and family child care. Berger et al.’s (2005) study finds a link between early return to work post-birth (within twelve weeks) and lower rates of medical checkups and lower rates of immunizations for diptheria, pertussis, tetanus and (oral) polio. Ruhm (1998) and Tanaka (2005) report that maternity leave mandates, presumably acting through an effect on maternal employment, lower post-neonatal mortality in a sample of primarily European countries.

2.4 Socioeconomic Status

Whether money matters for children is a wide-ranging question, but certainly one important dimension is whether it matters for children’s health. An effect of socioeconomic status (SES) on child health could arise because wealthier families can afford better health care and healthier environments for their children to be born and grow up in. A separate chapter of this volume is dedicated to the relationship between SES and health, but it’s worth briefly highlighting here some of the contributions of economists to the study of this issue for children.

Recent research documents a correlation between parents’ SES and children’s health outcomes. While the measures of SES and health vary by study, the message of this research is that lower SES is associated with lower health outcomes. This association is present at young ages. Evidence presented in Currie’s (2009) review of the subject, drawing on studies by Case et al. (2002), Currie and Stabile (2003), and Currie, Shields, and Price (2004), shows a negative correlation between the reported health of children under 4 and family income in the US, Canada, and the UK.

Further support is offered by studies of health shocks and specific illnesses and ailments. Case et al. (2002), Currie and Stabile (2003), Currie, Shields, and Price (2004), and Currie and Lin (2007) all find higher rates of chronic conditions (e.g. asthma, poor mental health) among poor children. Currie (2009) provides comparisons of the incidence of a substantial list of chronic conditions, illnesses, birth outcomes, and activity limitations for US children aged 2–17 living in poor and non-poor families. In almost all instances those in poor families have worse outcomes. Berger et al. (2005) report low-income children in the US have lower cognitive scores and worse behavior. Interestingly, Dooley and Stewart (2007) present dissenting evidence on this specific point for Canada, reporting little relationship between income and children’s emotional behavioral outcomes. Finally, Propper et al. (2004) present related evidence for children in the UK. They report that low-income children exhibit worse health by a maternal reported summary measure, and a high number of symptoms of poor health.

What it is about low SES that has a negative impact on children’s health is an area of continuing research? An important alternative hypothesis is that some other, typically unobserved, factor determines both SES and children’s health outcomes. The number of studies associating poor child outcomes with low SES far exceeds the number that make substantive progress on this difficult question of causality.

One area in which there has been progress is a good example of economists’ contribution to this literature. One common marker of families’ SES is the educational attainment of parents. There is burgeoning literature in economics estimating the returns to education that provides an array of instruments for educational attainment that could be used in this context.

One strategy from this literature is to exploit exogenous variation in the costs of attending university, proxied by distance to a post-secondary institution.

Currie and Moretti (2003) use the geographical expansion of colleges in the US as an instrument for university attendance. They find that the increase in educational attainment resulting from the opening of a local college improves pre-natal factors (less smoking, greater use of pre-natal care), birth weight and gestational age. Carneiro et al. (2007) use a related strategy instrumenting maternal education with the presence (rather than the opening) of a college locally at age 14. They report consequent improvements on children’s cognitive scores and reductions in their behavioral problems.

Another strategy adapted from this literature is geographic variation in school entry ages and compulsory schooling laws.11 McCrary and Royer (2006) use this approach finding an impact on educational attainment that has little consequence for the health of offspring. In a related paper Chou et al. (2010) study the impact of an increase in the length of compulsory schooling in Taiwan. In this case the length of compulsory schooling was increased (by three years). The resulting increase in educational attainment is found to be linked to lower rates of low birth weight and infant mortality (both neonatal and post-neonatal).

These approaches have limitations. First, as in any instrumental variables strategy we must convince ourselves that the instruments are legitimate. Second, if an increment in education has different effects on different people, we must recognize that we are estimating local average treatment effects. Most properly this is evidence for the types of individuals whose educational investment decisions react to college proximity or compulsory schooling laws. Nevertheless, in a literature that is much stronger on observation than on causality, these types of studies represent a step forward.

3 Access to Medical Care

Researchers have used expansions and contractions in public insurance coverage to examine the relationship between improved access to health care, the utilization of health care, and health status in several contexts. In the US there is a considerable body of evidence examining the Medicaid expansions through the 1980s and 1990s. Currie and Grogger (2002) note that an important goal of these changes is to improve health by encouraging low income and children to seek appropriate care.

Beginning in 1984 states were first permitted and then required to extend Medicaid coverage to other groups of children. By 1992 States were required to cover children below age 6 in families with incomes up to 133 percent of the poverty line and had the option of covering families up to 185 percent of the poverty line. For further review of the expansions see Cutler and Gruber (1996).

Several studies have examined the effects of the expansions on health care utilization and child health. Currie and Gruber (1996a, 1996b) found that many eligible women did not take up the coverage that was made available to them. However, their findings do indicate that despite these low take-up rates, the expansions targeted at low income families with children did increase the probability of going to a doctor for preventative care (measured as yearly check-ups as a measure of preventative care) by 9.6 percent. They also find large effects on the relationship between increased access to care through insurance expansions and child mortality. Their results suggest that for every ten percentage point increase in the fraction of children eligible for Medicaid, child mortality drops by 0.128 percentage points or 3.4 percent.

Currie and Gruber (2001) examine the effects of the Medicaid expansions on the medical treatment at child birth and on infant mortality. Among teen mothers and mothers with less than a high school education, enhanced eligibility resulted in increased utilization of a variety of obstetric procedures. The authors find less evidence of reductions in infant mortality (conditional on infant health) but do find significant reductions for mothers for whom the nearest hospital had a NICU.

Kaestner, Joyce, and Racine (1999) and Dafny and Gruber (2005) examine the nature of the increase in utilization. Their findings suggest that eligibility leads to fewer “avoidable” hospitalizations. Kaestner, Joyce, and Racine (1999) find no evidence of Medicaid expansions and increases in utilization on improvements in self-reported health status.

Research examining the Canadian experience of moving to a single, publicly funded payer also provides evidence on the relationship between increased access to care and utilization. In her 1996 paper, Hanratty investigates the effects of Canada’s move to national health insurance through the 1960s on infant health. Her findings suggest that the introduction of national health insurance in Canada is associated with a 4 percent decline in the infant mortality rate and a decrease in the incidence of low birth weight of, on average, 1.3 percent for all births and by 8.9 percent for births to single, lower income mothers.

Overall, evidence on the relationship between increased access to care, primarily through expansions in public health insurance and health care utilization suggests that improved access to medical care does result in increase in the use of preventative care, particularly among expectant mothers, with small but significant improvements in child health. The magnitude of the relationship between increased access to care and improved health is a function of two things: first, whether individuals and families use the increased access made available to them, and, second, whether the increased utilization is effective in improving health. Clearly, simply improving access to care is not sufficient to ensure take-up of care. Information barriers and SES gradients remain. A second question is whether, conditional on accessing the care, the care is effective in improving health. Here, the evidence reviewed suggests that access to preventative care, particularly among expectant mothers and children, can be effective in improving health. There is little evidence beyond these groups however, to draw broader conclusions on the role of increased utilization on health.

4 Health and Behavior Post-Early-Years

The last decade or so has seen a considerable amount of research on the economics of health related behaviors at older ages, with considerable focus specifically on the economic factors that affect health at young ages. We focus here on the key literature regarding smoking, drinking, and obesity. We then examine some of the new research on the economics of mental health and childhood behavior.

4.1 Smoking

Several studies have examined the relationship between prices, regulation and smoking across both adults and youths. The literature up to the last decade is reviewed in a paper by Chaloupka and Warner (2000) provides a comprehensive review of the economics of “bads.” Here we provide a brief update on the literature on youth smoking over the past few years. Chaloupka and Warner review several studies of smoking that rely mainly on cross sectional estimates of the effects of prices and regulation on smoking behavior. As more longitudinal data have become available in the US and elsewhere, several recent studies have re-estimated the relationship between price, regulation, and smoking behavior, using variation across jurisdictions over time to identify this relationship.

Gruber and Zinman (2000) and Gruber (2001) use variation in state cigarette taxes as instruments for state prices and state laws over time to estimate models of youth smoking, and in particular to try and explain the rise in youth smoking in the US over the 1990s. They show that prices are “powerful determinants” of smoking for high school seniors (Gruber and Zinman 2000) although price does not appear to be as important for younger teens. Gruber (2001) finds less evidence of the efficacy of laws that make it more difficult to smoke, other than small effects of age restrictions for the purchase of cigarettes.

Glied (2003) examines whether changes in prices which lead to declines in youth smoking result in fewer youths smoking when they are older by tracking individuals using panel data from the NLSY. Glied examines the relationship between taxes and smoking behavior when individuals are young and then tracks their smoking behavior into adulthood. Understanding whether helping price sensitive youths avoid smoking during their teenage years results in fewer adult smokers, or whether it simply delays the onset of smoking is an important element in understanding the efficacy of tax policy in reducing the number of long term smokers in the population and therefore the significant health consequences of long-term smoking. Glied finds that the reductions in youth smoking are partially offset in adulthood, suggesting that estimates of the effects of price on youth smoking overstate the life-course effects of price as a tobacco control strategy.

DeCicca et al. (2002) and DeCicca et al. (2008) use a complementary US data set longitudinal data set, the National Educational Longitudinal Study, to re-examine the relationship between youth smoking participation and state taxes and find little evidence that taxes affect smoking behavior once state fixed effects and differences in smoking attitudes across states over time are properly accounted for. They do, however, find some evidence on the responsiveness to price among youths conditional on being a smoker.

Carpenter and Cook (2008) use repeated cross-sections from the Youth Risk Behavior Surveys, to re-examine the relationship between cigarette taxes and smoking behavior among youths. These data survey nearly 750,000 young adults and are designed to be representative of the local area sampled. The authors find evidence consistent with earlier studies (cf. Gruber 2001) that found negative effects of higher state taxes on youth smoking behavior.

In sum, there has been considerable new research on the role for public policy on youth smoking behavior that has resulted in a variety of results. The bulk of the evidence suggests that prices do affect youth behavior and have stronger effects on cessation than initiation but some recent studies call into question whether these effects are permanent or simply delay smoking initiation until later in life.

4.2 Obesity

Research examining the increase in obesity in North America, and more recently across the OECD has grown considerably over the past decade. The literature has focused on two broad areas, the causes of obesity, both at a point in time and over longer periods of time, and the consequences of obesity. While the medical literature has examined the genetic and metabolic determinants of obesity, the economic literature has focused on external and environmental causes. One strand of the literature argues that technological change has changed the nature of work, making market work more sedentary. Technological change has also resulted in lowered costs of calories through more efficient food production (cf. Philipson and Poser 2003; Lakdawalla and Philipson 2009). A second strand of the literature claims that the change in obesity is due to caloric intake (increased eating) and not a decrease in the number of calories burned (cf. Cutler, Glaeser, and Shapiro 2003; Bleich, Cutler, Murray, and Adams 2008). Declines in the time cost of food production and technological advances in firm level food preparation are cited as the primary causes of increased caloric intake. A third strand of literature examines household work decisions and childhood obesity and find evidence that increased labor participation among women is correlated with increases in childhood obesity (Anderson, Butcher, and Levin 2003; Chia 2008).

New research has also examined the relationship between obesity and school outcomes. Evidence suggests that obese children in US miss more school than non-obese children but there is little evidence that obesity results in poorer outcomes on test scores (Schwimmer et al. 2003; Chia 2007). On the other hand, evidence from Canada suggests that overweight children (not included in the obese category) fare poorer on math test scores. These differences are not driven by observable differences in the populations, leaving open the question of why health consequences differ across populations (Chia 2007).

4.3 Mental Health

Currie and Stabile (2009, 2006) note the large numbers of children who have some form of mental health problem. The MECA Study cited in the 1990 US Surgeon General’s Report on Mental Health states that 20 percent of children have some form of impairment from a mental or behavioral disorder and 5 percent suffer extreme functional impairment. Given these large numbers of children, researchers have recently begun to investigate the relationship between mental health in childhood and a variety of human capital and labor market outcomes later in life.

Economic research on the relationship between child mental health and economic outcomes falls into three categories. There are a series of studies that look at the longer term consequences of behavior problems in large samples (cf. Farmer 1993, 1995; Kessler et al. 1995; Gregg and Machin 1998; Caspi et al. 1998; Miech et al. 1999; and McLeod and Kaiser 2004). Findings here suggest that children with either a mental health diagnosis, or high survey scores of aggregated mental health problems are less likely to complete schooling, lower earnings and lower probabilities of employment at early stages of their working careers.

A second stream of the literature focuses on particular “externalizing” mental health conditions such as attention deficient hyperactivity disorders (ADHD) and behavior problems. Mannuzza and Klein (2000) review three studies of the long-term outcomes of children with ADHD and find that the ADHD children consistently have worse outcomes in adolescence and young adulthood than control children. Also, the studies do not address the possibility that the negative outcomes might be caused by other factors related to a diagnosis of ADHD, such as poverty, the presence of other learning disabilities, or the fact that many people diagnosed with ADHD end up in special education.

Currie and Stabile (2006) address these problems by examining the effects of ADHD in sibling fixed effects models using longitudinal data from both the United States and Canada. In a follow-up paper (Currie and Stabile 2008) they consider several other mental health conditions as well as aggregate measure of child mental health. The authors find that behavior problems have a large negative effect on future educational outcomes. The most consistent effects across the two countries are found for ADHD. In models that include sibling fixed effects, anxiety/depression is found to increase grade repetition but has no effect on the other outcomes we examine (such as test scores), suggesting that depression acts through a mechanism other than decreasing cognitive performance. Conduct disorders are also found to have broadly negative effects in the US, while in Canada, they reduce the probability that 16–19-year-old youths are in school but do not have significant effects on other outcomes. They find little evidence that these effects are modified by socioeconomic status.

A third recent strand of research examines the importance of “non-cognitive skills” such as hyperactivity, anxiousness, and self-esteem (cf. Blanden, Gregg, and Macmillan 2006; Heckman et al. 2006) in human capital formation, and later on income and income inequality. The findings suggest that such non-cognitive skills are important determinants of academic and economic success.

4.4 Maternal Behavior and Child Health and Behavior

A related area of research is the role that maternal behavior plays in the development of child health and behavior. A number of papers have approached these issues from various angles. We focus on a handful of representative papers here.

Chatterji and Markowitz (2001) examine the relationship between maternal substance abuse and child mental health as measured by a Behavioral Problems Index. The Behavioral Problems Index (also used in some of the mental health studies cited above) consists of a series of questions, answered by a child’s mother, on the child’s behavior and mental health, including hyperactivity, anxiety, anti-social behavior, and depression. The authors use data from the National Longitudinal Survey of Youth to examine whether there is evidence of a causal relationship between maternal use of alcohol and drugs and concurrent child behavioral problems for children ages 4 through 15. Fixed effects results suggest that there is some evidence of a link between maternal risky behavior and higher scores on the behavioral problems index. Results are stronger for drug use than for alcohol use.

Perry (2008) investigates the link between maternal depression and the management of childhood health—in particular childhood asthma. She analyzes how treatment of maternal depression affects child outcomes. To address the possibility of some unobserved factor that drives both the decision to seek treatment and the management of the child’s asthma, treatment is instrumented with a measure of the variation in the propensity of primary care physicians to treat depression. The results indicate the treatment of mothers’ depression leads to better management of the child’s condition and thus asthma care related costs.

A study by Hango and Houseknecht (2005) again uses the NLSY to examine the role that marital disruptions play on child health. In this case, the authors look at the effects of divorce/separation on children’s likelihood of suffering medically treated injuries. The measure of injuries is from a question in the NLSY that asks about whether the child had an injury in the past year that required medical attention. They distinguish between the direct effects of the marital disruption on injury—direct harm, usually emotional, caused by the parent—and potential mediating effects which might come through altering the relationship between the parent and child, or secondary effects such as financial changes, which could then lead to an increase in injuries. Overall there is little evidence which shows that marital disruption affects the likelihood of child injury for boys, either directly or indirectly. They find some evidence that marital disruption reduces the likelihood of injuries to girls. Overall, the authors conclude that there does not appear to be any significant mediating effect at work and only weak evidence of direct effects for girls.

8.5 Conclusions

Empirical research on child health spans many disciplines. What distinguishes economists’ contributions to the field? At one time it may have been econometric tools, but anyone who reads widely in this area realizes that these methods are quickly gaining currency in other disciplines.

Some of the research reviewed in this chapter identifies another possibility. Looking behind the economic correlates of child health to the individual choices that determine them has provided new insights and offered new empirical strategies. A good example is use of the instrumental variables strategies from the literature on the returns to education in research on maternal education and child health. Relating breastfeeding and the non-parental care of children to mothers’ decisions to work has identified labor market policies as a source of identifying variation for investigating child outcomes. This economic approach has in some areas caused researchers to revisit conventional wisdom. In others, it has helped to gauge the magnitude of long-understood associations. More generally it brings a heightened appreciation of the importance of causality to a literature that seems too often content with correlation.

New data sets are providing opportunities for wider application of this approach. Particularly exciting are administrative data form birth records, hospital admissions and medical records that paint a much richer portrait of the health outcomes of people of all ages, and can permit analyses of how events in childhood are related to later outcomes. The burgeoning literature on the long run consequences of low birth is a good example of this point. This relationship between health in childhood and at older ages is an important contribution of health economics to the larger research project on how investments in children pay off across the life-cycle. New findings from this wider field are transforming how we view the importance of childhood.

Our review of this literature also identifies some priorities for future research. First, there are many areas in which we are growing more confident that causal relationships exist, but remain ignorant of the mechanisms that underlie them. For example, what exactly is it about being born low birth weight, or its correlates, that leads to poor labor market outcomes in later life and what role do economic inputs play along the way? Second, attracted by the empirical design of social experiments for disadvantaged children, we sometimes neglect the health of more heterogeneous populations of children. More advantaged children, however, will be the majority clients of universal public policies. Third, as noted above follow up study of findings from infancy and childhood will make a fundamental contribution to the portrait of life-cycle health. Finally, while the field of child health is very broad, economists’ contributions are not, and there would appear to be many areas that might benefit from attention from economists.

References:

  1. ALMOND, DOUGLAS, CHAY, KENNETH, and LEE, DAVIDS. (2005), “The Costs of Low Birth Weight,” Quarterly Journal of Economics, 120(3): 1031–83.
  2. —— EDLUND, LENA, LI, HONGBIN, ZHANG, JUNSEN(2007), “Long-term Effect of the 1959–1961 China Famine: Mainland China and Hong Kong.” NBER Working Paper 13384.
  3. ANDERSON, PATRICIA, LEVINE, PHILLIP, and BUTCHER, KRISTIN(2003), “Maternal Employment and Overweight Children,” Journal of Health Economics, 22(3): 477–504.
  4. ANGRIST, JOSHUA and KRUEGER, ALAN(1991), “Does Compulsory School Attendance Affect Schooling and Earnings?” Quarterly Journal of Economics, 106(4): 979–1014.
  5. BAKER, MICHAELand MILLIGAN, KEVIN(2008a), “How Does Job-Protected Maternity Leave Affect Mothers’ Employment and Infant Health?” Journal of Labor Economics, 26: 655–92.
  6. —— —— (2008b), “Maternal Employment, Breastfeeding and Health: Evidence from Maternity Leave Mandates,” Journal of Health Economics, 27: 871–87.
  7. —— —— (2010), “Evidence From Maternity Leave Expansions of the Impact of Maternal Care on Early Child Development,” Journal of Human Resources, 45(1): 1–32.
  8. —— GRUBER, J., and MILLIGAN, K. (2008), “Universal Childcare, Maternal Labour Supply and Child Well-being,” Journal of Political Economy, 116: 709–45.
  9. BARKER, DAVID(1998), Mothers, Babies and Health in Later Life, 2nd edn. (Edinburgh: Churchill Livingstone).
  10. BAUMII, CHARLES (2003a), “The Effect of State Maternity Leave Legislation and the 1993 Family and Medical Leave Act on Employment and Wages,” Labour Economics, 10(5): 573–96.
  11. —— (2003b), “The Effects of Maternity Leave Legislation on Mothers’ Labor Supply after Childbirth,” Southern Economic Journal, 69(4): 772–99.
  12. —— (2003c). “Does Early Maternal Employment Harm Child Development? An Analysis of the Potential Benefits of Leave-Taking,” Journal of Labor Economics, 21(2): 409–48.
  13. —— and FORD, WILLIAM (2004), “The Wage Effects of Obesity: A Longitudinal Study,” Health Economics, 13(9): 885–99.
  14. BEHRMAN, JERE and ROSENZWEIG, MARKR. (2004), “Returns to Birthweight,” Review of Economics and Statistics, 86(2): 586–601.
  15. BELSKY, JAY(2001), “Developmental Risks (Still) Associated with Early Child Care,” Journal of Child Psychology and Psychiatry, 42: 845–59.
  16. BERGER, LAWRENCE, HILL, JENNIFER, and WALDFOGEL, JANE(2005), “Maternity Leave, Early Maternal Employment and Child Health and Development in the US,” The Economic Journal, 115(501): F29–F47.
  17. BERNAL, RAQUEL(2008), “The Effect of Maternal Employment and Child Care on Children’s Cognitive Development,” International Economic Review, 49(4): 1173–209.
  18. —— and KEANE, MICHAEL(2010), “Quasi-Structural Estimation of a Model of Child Care Choices and Child Cognitive Ability Production,” Journal of Econometrics, 156(1): 164–89.
  19. BLACK, SANDRA, DEVEREUX, PAULJ., and SALVANES, KJELL G. (2007), “From the Cradle to the Labor Market? The Effect of Birth Weight on Adult Outcomes,” Quarterly Journal of Economics, 122(1): 409–39.
  20. BLANDEN, JO, GREGG, PAUL, and MACMILLAN, LINDSEY(2006), “Explaining Intergenerational Income Persistence: Non-cognitive Skills, Ability and Education.” Centre for Market and Public Organisation Working Paper 06/146.
  21. BLEICH, SARA, CUTLER, DAVID, MURRAY, CHRISTOPHER, and ADAMS, ALYCE(2008), “Why is the Developed World Obese?” Public Health, 29: 273–95.
  22. BROOKS-GUNN, JEANNEand DUNCAN, GREG (1997), “The Effects of Poverty on Children,” The Future of Children, 7(2): 55–71.
  23. CARD, DAVID(1995), “Using Geographic Variation in College Proximity to Estimate the Return to Schooling,” in Louis N. Christofides, E. Kenneth Grant and Robert Swindinsky (eds.), Aspects of Labour Market Behaviour (Toronto: University of Toronto Press).
  24. CARNEIRO, PEDRO, MEGHIR, COSTAS, and PAREY, MATTHIAS(2008), “Maternal Education, Home Environments and the Development of Children and Adolescents.” Institute for Fiscal Studies Working Paper W07/15.
  25. CARPENTER, CHRISTOPHERand COOK, PHILIP (2008), “Cigarette Taxes and Youth Smoking: New Evidence from National, State, & Local Youth Risk Behavior Surveys.” Journal of Health Economics, 27(2): 287–99.
  26. CASE, ANNE, LUBOTSKY, DARREN, and PAXSON, CHRISTINA(2002), “Economic Status and Health in Childhood: The Origins of the Gradient,” American Economic Review, 92(5): 1308–34.
  27. CASPI, AVSHALOM, WRIGHT, BRADLEY, MOFFITT, TERRIE, and SILVA, PHIL(1998), “Early Failure in the Labor Market: Childhood and Adolescent Predictors of Unemployment in the Transition to Adulthood,” American Sociological Review, 63: 424–51.
  28. CAWLEY, JOHN(2004), “The Impact of Obesity on Wages,” Journal of Human Resources, 39(2): 451–74.
  29. —— and LIU, FENG(2007), “Mechanisms for the Association Between Maternal Employment and Child Cognitive Development.” NBER Working Paper 13609.
  30. —— and DANZIGER, SHELDON(2005), “Obesity as a Barrier to the Transition from Welfare to Work.” Journal of Policy Analysis and Management, 24(4): 727–43.
  31. CHALOUPKA, FRANK and WARNER, KENNETHE. (2000), “The Economics of Smoking,” Handbook of Health Economics, 1st edn., Vol. 1, No. 1 (Amsterdam: Elsevier), 1539–627.
  32. CHATTERJI, PINKAand FRICK, KEVIN (2005), “Does Returning to Work After Childbirth Affect Breastfeeding Practices?” Review of Economics of the Household, 3(3): 315–35.
  33. —— and MARKOWITZ, SARA(2001), “The Impact of Maternal Alcohol and Marijuana Use on Children’s Behavioral Problems,” Journal of Health Economics, 20(5): 703–31.
  34. CHAY, KENNETHand GREENSTONE, MICHAEL (2003a), “The Impact of Air Pollution on Infant Mortality: Evidence from Geographic Variation in Pollution Shocks Induced by a Recession,” Quarterly Journal of Economics, 118(3): 1121–67.
  35. —— (2003b), “Air Quality, Infant Mortality, and the Clean Air Act of 1970.” NBER Working Paper No. 10053.
  36. CHIA, YEEFEI (2007), “Weighty Issues: An Examination of Childhood Weight and School Outcomes During Puberty.” Mimeo, University of Toronto.
  37. —— (2008), “Maternal Labour Supply and Childhood Obesity in Canada: Evidence from the NLSCY,” Canadian Journal of Economics, 41(1): 217–42.
  38. CHOU, SHIN-YI, LIU, JIN-TAN, GROSSMAN, MICHAELand JOYCE, THEODORE (2007), “Parental Education and Child Health: Evidence from a Natural Experiment in Taiwan.” American Economic Journal: Applied Economics, 2(1): 33–61.
  39. CONLEY, D., STRULLLY, K., and BENNETT, NEIL(2003), The Starting Gate: Birth Weight and Life Chances (Berkeley, CA: University of California Press).
  40. CUNHA, FLAVIO, HECKMAN, JAMES, and SCHENNACH, SUSANNE(2008), “Estimating the Elasticity of Intertemporal Substitution in the Formation of Cognitive and Non-Cognitive Skills.” Mimeo, University of Chicago.
  41. CURRIE, ALISON, SHIELDS, MICHAEL, and PRICE, STEPHANWHEATLEY (2004), “Is the Child Health/Family Income Gradient Universal? Evidence from England.” IZA Discussion Paper 1328.
  42. CURRIE, J. (2009), “Healthy, Wealthy, and Wise: Socioeconomic Status, Poor Health in Childhood, and Human Capital Development.” Journal of Economics Literature, 47(1): 87–122.
  43. —— and HYSON, ROSEMARY(1999), “Is the Impact of Health Shocks Cushioned by Socioeconomic Status? The Case of Low Birth Weight,” American Economic Review 2 (May): 19–22.
  44. —— and MORETTI, ENRICO(2003), “Mother’s Education and the Intergenerational Transmission of Human Capital: Evidence From College Openings,” Quarterly Journal of Economics, 118(4): 1495–532.
  45. —— and GROGGER, JEFFREY(2002), “Medicaid Expansions and Welfare Contractions: Offsetting Effects on Prenatal Care and Infant Health?” Journal of Health Economics, 21(2): 313–35.
  46. —— and GRUBER, JONATHAN(1996a), “Health Insurance Eligibility, Utilization of Medical Care, and Child Health,” Quarterly Journal of Economics, 111(2): 431–66.
  47. —— —— (1996b), “Saving Babies: The Efficacy and Cost of Recent Changes in the Medicaid Eligibility of Pregnant Women,” Journal of Political Economy, 104(6): 1263–96.
  48. —— (2001), “Public Health Insurance and Medical Treatment: The Equalizing Impact of the Medicaid Expansions,” Journal of Public Economics, 82(1): 63–89.
  49. —— and STABILE, MARK(2003), “Socioeconomic Status and Health: Why is the Relationship Stronger for Older Children?” American Economic Review, 93(5): 1813–23.
  50. —— (2006), “Child Mental Health and Human Capital Accumulation: The Case of ADHD,” Journal of Health Economics, 25(6): 1094–118.
  51. —— (2009), “Mental Health in Childhood and Human Capital.” In Jonathan Gruber (ed.), An Economic Perspective on the Problems of Disadvantaged Youth(Chicago: University of Chicago Press for NBER).
  52. —— and NEIDELL, MATTHEW(2007), “Getting Inside the ‘Black Box’ of Head Start Quality: What Matters and What Doesn’t,” Economics of Education Review, 26(1): 83–99.
  53. —— and HYSON, ROSEMARY(1999), “Is the Impact of Health Shocks Cushioned by Socioeconomic Status? The Case of Low Birthweight,” American Economic Review, 89(2): 245–50.
  54. —— and LIN, WANCHUAN(2007), “Chipping Away at Health: More on the Relationship Between Income and Child Health,” Health Affairs, 26(2).
  55. CUTLER, DAVID and GRUBER, JONATHAN(1996), “The Effect of Medicaid Expansions on Public Insurance, Private Insurance, and Redistribution,” American Economic Review, 86(2): 378–83.
  56. —— GLAESER, EDWARD, and SHAPIRO, JESSEM. (2005), “Public Insurance and Child Hospitalizations: Access and Efficiency Effects,” Journal of Public Economics, 89(1): 109-29.
  57. DAFNY, LEEMOREand GRUBER, JONATHAN (2005), “Public Insurance and Child Hospitalizations: Access and Efficiency Effects,” Journal of Public Economics, 89(1): 109–29.
  58. DECICCA, PHILIP, KENKEL, DONALD, and MATHIOS, ALAN(2002), “Putting Out the Fires: Will Higher Taxes Reduce the Onset of Youth Smoking?” Journal of Political Economy, 110(1): 144–69.
  59. —— and SHIN, JUSTINE(2008), “Youth Smoking, Cigarette Prices and Anti-smoking Sentiment,” Health Economics, 17(6): 733–49.
  60. DENNIS, CINDY-LEE(2002), “Breastfeeding Initiation and Duration: A 1990–2000 Literature Review,” Journal of Obstetric, Gynecologic, & Neonatal Nursing, 31(1).
  61. DOOLEY, MARTINand STEWART, JENNIFER (2007), “Family Income, Parenting Styles and Child Behavior: Emotional Outcomes,” Health Economics, 16(2).
  62. FARMER, ELIZABETH Z. (1993), “Externalizing Behavior in the Life Course: The Transition from School to Work,” Journal of Emotional and Behavioral Disorders, 1: 138–48.
  63. —— (1995), “Extremity of Externalizing Behavior and Young Adult Outcomes,” Journal of Child Psychology and Psychiatry, 36. (Fiscal Studies Working Paper No. 15/07. Sept. 2007.)
  64. GLIED, SHERRY(2003), “Is Smoking Delayed Smoking Averted?” American Journal of Public Health, 93(3): 412–16.
  65. GLUCKMAN, P. D. and HANSON, M. A. (2006), “Adult Disease: Echoes of the Past,” European Journal of Endocrinology, 155(Suppl. 1): 47–50.
  66. GORDON, RACHEL, KAESTNER, ROBERT, and KORENMAN, SANDERS(2007), “The Effects of Maternal Employment on Child Injuries and Infectious Disease,” Demography, 44(2): 405–26.
  67. GREGG, PAULand MACHIN, STEVEN (1998) “Child Development and Success or Failure in the Youth Labour Market.” Center for Economic Performance, London School of Economics Discussion Paper 0397, July.
  68. GRUBER, JONATHAN(2001), “Youth Smoking in the 1990s: Why Did It Rise and What Are the Long-Run Implications?” American Economic Review, 91(2): 85–90.
  69. —— and ZINMAN, JONATHAN(2000), “Youth Smoking in the US: Evidence and Implications.” NBER Working Paper 7780.
  70. HAIDER, STEVEN, JACKNOWITZ, ALISON, and SCHOENI, ROBERTF. (2003), “Welfare Work Requirements and Child Well-Being: Evidence from the Effects on Breast-Feeding,” Demography, 40(3): 479–97.
  71. HAMLYN, BECKY, BROOKER, SUE, OLEINIKOVA, KARIN, and WANDS, SARAH(2002), Infant Feeding 2000 (Norwich: The Stationery Office).
  72. HAN, W.-J., RUHM, C., and WALDFOGEL, J. (2007), “Parental Leave Policies and Parents’ Employment and Leave-Taking,” Journal of Policy Analysis and Mangement, 28(1): 29–54.
  73. HANGO, DARCY and HOUSEKNECHT, SHARONK. (2005), “Marital Disruption and Accidents/Injuries Among Children,” Journal of Family Issues, 26(1): 3–31.
  74. HANRATTY, MARIA (1996) “Canadian National Health Insurance and Infant Health,” American Economic Review, 86(1): 276–84.
  75. HECKMAN, JAMES(2007), “The Economics, Technology and Neuroscience of Human Capability Formation,” Proceedings of the National Academy of Sciences, 104(33): 13250–5.
  76. —— STIXRUD, JORA, and URZUA, SERGIO(2006), “The Effects of Cognitive and Noncognitive Abilities on Labor Market Outcomes and Social Behavior,” Journal of Labor Economics, 24(3): 365–89.
  77. HILL, JENNIFER, WALDFOGEL, JANE, BROOKS-GUNN, JEANNE, and HAN, WEN-JUI(2005), “Maternal Employment and Child Development: A Fresh Look Using Newer Methods,” Developmental Psychology, 41(6): 833–50.
  78. HORTA, B., BAHL, R., MARTINES, J. C. and VICTORIA, C. G. (2007), Evidence on the Long-Term Effects of Breastfeeding, Systematic Reviews and Meta-Analysis(Geneva: World Health Organization).
  79. IMBENS, GUIDO and ANGRIST, JOSHUAD. (1994), “Identification and Estimation of Local Average Treatment Effects,” Econometrica, 62(2): 467–75.
  80. JAMES-BURDUMY, SUSANNE(2005), “The Effect of Maternal Labor Force Participation on Child Development,” Journal of Labor Economics, 23(1): 176–7.
  81. KAESTNER, ROBERT, JOYCE, THEODORE, and RACINE, ANDREW(1999), “Does Publicly Provided Health Insurance Improve the Health of Low-Income Children in the United States.” NBER Working Paper 6887.
  82. KESSLER, R. C., FOSTER, C. L., SAUNDERS, W. B., and STANG, P. E. (1995), “Social Consequences of Psychiatric Disorders, I: Educational Attainment,” American Journal of Psychiatry, 152(7): 1026–32.
  83. KLERMAN, JACOBALEX and LEIBOWITZ, ARLEEN (1997), “Labor Supply Effects of State Maternity Leave Legislation,” in Francine D. Blau and Ronald G. Ehrenberg (eds.), Gender and Family Issues in the Workplace (New York: Russell Sage Foundation).
  84. KRAMER, MICHAEL, CHALMERS, BEVERLEY, HODNETT, ELLEND., et al. (2001), “Promotion of Breastfeeding Intervention Trial (PROBIT): A Randomized Trial in the Republic of Belarus,” Journal of the American Medical Association, 285(4): 463–4.
  85. LAKDAWALLA, DARIUSand PHILIPSON, TOMAS (2009), “The Growth of Obesity and Technological Change,” Economics and Human Biology, 7(3): 283–93.
  86. LANSINOHLABORATORIES (2003), “Moms Want to Breastfeed But Feel Unprepared, Unsupported.” https://lansinoh.com/blogs/breastfeeding/moms-want-to-breastfeed-but-feel-unprepared-unsupported, accessed June 2, 2022.
  87. LOEB, SUSANNA, BRIDGES, MARGARET, BASSOK, DAPHNA, FULLER, BRUCE, and RUMBERGER, RUSSEL (2007), “How Much is Too Much? The Influence of Preschool Centers on Children’s Social and Cognitive Development,” Economics of Education Review, 26(1): 52–66.
  88. MAGNUSON, KATHERINE, RUHM, CHRISTOPHER, and WALDFOGEL, JANE(2007), “Does Prekindergarten Improve School Preparation and Performance?” Economics of Education Review, 26(1): 33–51.
  89. MANNUZZA, SALVATOREand KLEIN, RACHEL (2000), “Long-term Prognosis in Attention-Deficit/Hyperactivity Disorder,” Child and Adolescent Psychiatric Clinics of North America, 9(3): 711–26.
  90. MCCRARY, JUSTINand ROYER, HEATHER (2006), “The Effect of Female Education on Fertility and Infant Health: Evidence from School Entry Policies Using Exact Data of Birth.” NBER Working Paper 12329.
  91. MCLEOD, JANE and KAISER, KAREN(2004), “Childhood Emotional and Behavioral Problems and Educational Attainment,” American Sociological Review, 69(5): 636–58.
  92. MIECH, RICHARD, CASPI, AVSHALOM, MOFFITT, TERRIE., ENTNER WRIGHT, BRADLEY R. and SILVA, PHIL A. (1999), “Low Socioeconomic Status and Mental Disorders: A Longitudinal Study of Selection and Causation during Young Adulthood,” American Journal of Sociology, 104(4): 1096–131.
  93. NEIDELL, MATTHEW (2004), “Air Pollution, Health, and Socio-economic Status: The Effect of Outdoor Air Quality on Childhood Asthma,” Journal of Health Economics, 23(6): 1209–36.
  94. NICHD (NATIONALINSTITUTE OF CHILD HEALTH AND HUMAN DEVELOPMENT; EARLY CHILDCARE RESEARCH NETWORK) (2003), “Does Amount of Time Spent in Childcare Predict Socioemotional Adjustment During the Transition to Kindergarten?” Child Development, 74(4): 976–1005.
  95. OREOPOULOS, PHILIP, STAIBLE, MARK, WALLD, RANDY, and ROOS, LESLIE (2008), “Short-, Medium-, and Long-Term Consequences of Poor Infant Health: An Analysis Using Siblings and Twins,” Journal of Human Resources, 43(1): 88–138.
  96. PERRY, CYNTHIA (2008), “Does Treating Maternal Depression Improve Child Health Management? The Case of Pediatric Asthma,” Journal of Health Economics, 27(1): 157–73.
  97. PHILIPSON, TOMAS and POSNER, RICHARDA. (2003), “The Long-Run Growth in Obesity as a Function of Technological Change,” Perspectives in Biology and Medicine, 46(3): 87–108.
  98. PROPPER, CAROL, BURGESS, SIMON, and RIGG, JOHN(2004), “The Impact of Low Income on Child Health: Evidence from a Birth Cohort Study.” Leverhulme Centre for Market and Public Organization, Working Paper 04/098.
  99. ROE, BRIAN, WHITTINGTON, LESLIE, FEIN, SARABECK, and TEISL, MARIO F. (1999), “Is there Competition between Breast-Feeding and Maternal Employment?” Demography, 36(2): 157–71.
  100. RUHM, CHRISTOPHER (1998), “The Economic Consequences of Parental Leave Mandates: Lessons From Europe,” Quarterly Journal of Economics, 113(1): 285–317.
  101. —— (2004), “Parental Employment and Child Cognitive Development,” Journal of Human Resources, 39(1): 155–92.
  102. SCHONBERG, UTAand LUDSTECK, JOHANNES (2007), “Maternity Leave Legislation, Female Labor Supply, and the Family Wage Gap.” IZA Discussion Paper 2699.
  103. SCHWARTZ, KENDRA, D’ARCY, HANNAH S., GILLESPIE, BRENDA, BOBO, JANET, LONGEWAY, MARYLOU, and FOXMAN, BETSY(2002), “Factors Associated with Weaning in the First 3 Months Postpartum,” Journal of Family Practice, 51(5): 439–44.
  104. MIECH, RICHARD, CASPI, AVSHALOM, MOFFITT, TERRIE., ENTNER WRIGHT, BRADLEY R. and SILVA, PHIL A. (1999), “Low Socioeconomic Status and Mental Disorders: A Longitudinal Study of Selection and Causation during Young Adulthood,” American Journal of Sociology, 104(4): 1096–131.
  105. NEIDELL, MATTHEW (2004), “Air Pollution, Health, and Socio-economic Status: The Effect of Outdoor Air Quality on Childhood Asthma,” Journal of Health Economics, 23(6): 1209–36.
  106. NICHD (NATIONALINSTITUTE OF CHILD HEALTH AND HUMAN DEVELOPMENT; EARLY CHILDCARE RESEARCH NETWORK) (2003), “Does Amount of Time Spent in Childcare Predict Socioemotional Adjustment During the Transition to Kindergarten?” Child Development, 74(4): 976–1005.
  107. OREOPOULOS, PHILIP, STAIBLE, MARK, WALLD, RANDY, and ROOS, LESLIE (2008), “Short-, Medium-, and Long-Term Consequences of Poor Infant Health: An Analysis Using Siblings and Twins,” Journal of Human Resources, 43(1): 88–138.
  108. PERRY, CYNTHIA (2008), “Does Treating Maternal Depression Improve Child Health Management? The Case of Pediatric Asthma,” Journal of Health Economics, 27(1): 157–73.
  109. PHILIPSON, TOMAS and POSNER, RICHARDA. (2003), “The Long-Run Growth in Obesity as a Function of Technological Change,” Perspectives in Biology and Medicine, 46(3): 87–108.
  110. PROPPER, CAROL, BURGESS, SIMON, and RIGG, JOHN(2004), “The Impact of Low Income on Child Health: Evidence from a Birth Cohort Study.” Leverhulme Centre for Market and Public Organization, Working Paper 04/098.
  111. ROE, BRIAN, WHITTINGTON, LESLIE, FEIN, SARABECK, and TEISL, MARIO F. (1999), “Is there Competition between Breast-Feeding and Maternal Employment?” Demography, 36(2): 157–71.
  112. RUHM, CHRISTOPHER (1998), “The Economic Consequences of Parental Leave Mandates: Lessons From Europe,” Quarterly Journal of Economics, 113(1): 285–317.
  113. —— (2004), “Parental Employment and Child Cognitive Development,” Journal of Human Resources, 39(1): 155–92.
  114. SCHONBERG, UTAand LUDSTECK, JOHANNES (2007), “Maternity Leave Legislation, Female Labor Supply, and the Family Wage Gap.” IZA Discussion Paper 2699.
  115. SCHWARTZ, KENDRA, D’ARCY, HANNAH S., GILLESPIE, BRENDA, BOBO, JANET, LONGEWAY, MARYLOU, and FOXMAN, BETSY(2002), “Factors Associated with Weaning in the First 3 Months Postpartum,” Journal of Family Practice, 51(5): 439–44.
  116. SCHWIMMER, JEFFREY, BURWINKLE, TASHAM., and VARNI, JAMES W. (2003), “Health-Related Quality of Life of Severely Obese Children and Adolescents,” Journal of the American Medical Association, 289(14): 1851–3.
  117. TANAKA, SAKIKO(2005), “Parental Leave and Child Health across OECD Countries,” The Economic Journal, February, 115(501): F7–F28.
  118. UNICEF (2008) “Breastfeeding,” https://www.unicef.org/reports/breastfeeding, accessed June 2, 2022.
  119. WALDFOGEL, JANE(1999), “The Impact of the Family and Medical Leave Act,” Journal of Policy Analysis and Management, 18(2): 281–302.
  120. —— (2006), What Children Need(Cambridge: Harvard University Press).