Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, ECONOMICS AND FINANCE (economics.oxfordre.com). (c) Oxford University Press USA, 2018. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 13 November 2018

The Economics of Families and Health

Summary and Keywords

An individual’s health is produced in large part by family investments that start before birth and continue to the end of life. The health of an individual is intertwined with practically every economic decision including education, marriage, fertility, labor market, and investments. These outcomes in turn affect income and wealth and hence have implications for intergenerational transfer of economic advantage or disadvantage. A rich body of theoretical and empirical work considers the role of the family in health production over the life cycle and the role of health in household economic decisions. This literature starts by considering family inputs regarding health at birth, then moves through adolescence and midlife, where relationship decisions affect health. After midlife, health, particularly the health of family members, becomes an input into retirement and investment decisions. The literature on family and health showcases economists’ skills in modeling complex family dynamics, deriving theoretical predictions, and using clever econometric strategies to identify causal effects.

Keywords: health, household, family, early childhood, marriage, retirement, informal care, end-of-life care

Introduction

An individual’s health is produced in large part by family investments that start before birth and continue to the end of life. Household resources and thereby household decision-making contribute to health on multiple levels, from access to adequate housing, food, education, and medical care to the avoidance of health “bads” including secondhand smoke and unhealthy foods that lead to obesity.1 Parents make decisions for their children that affect both their child’s health and their own. Spouses make decisions about formal and household labor, leisure time activities, and finances that may be affected by the health of their partner and/or children as well as their own health. Finally, providing informal care for illness at any stage of life affects the health and wealth of family members both giving and receiving care. Thus, understanding family decision-making regarding health is essential to fully inform not only health but nearly every social and economic policy.

This article begins with a brief overview of theoretical constructs underlying household decision-making. Household decision models build upon the individual rational choice models predominant in the economic literature by considering the preferences and strategic behavior of multiple decision-makers in a family. Consistent with the work of Heckman (2012), a life-cycle, developmental perspective identifies those places where families invest in health, and empirical research on the causal effects of these investments is reviewed. The article starts with health investments in utero, moves on to early childhood, and continues through relationship formation. After midlife health becomes important in retirement and investment decisions. The article ends with health and wealth decisions associated with the end of life. Each section contains a discussion of the primary research questions and empirical challenges and offers short synopses of papers that explore these issues. The literatures drawn from are vast; readers are encouraged to look at the literature cited for more complete reviews.

Theoretical Constructs of Household Decision-Making

The theoretical literature specifies three nested models of household decision-making: the unitary model, the bargaining model (cooperative and non-cooperative), and the collective model (for excellent reviews, see Browning, Chiappori, & Weiss, 2014; Himmelweit, Santos, Sevilla, & Sofer, 2013; Vermeulen, 2002; Wax, 2017). Briefly, the unitary model assumes that the family acts as a single decision-maker (e.g., family members have identical preferences) with a pooled budget constraint. Thus, the unitary model is only a minor extension of the predominant individual rational choice model with the family conceived as a single homogeneous entity standing in place of the individual. Two central assumptions necessary to model family decisions in the unitary framework are income pooling (the source of resources has no impact on decisions) and Slutsky symmetry (marginal changes have the same effect on all family members), and both have been soundly rejected in the literature.

The bargaining model relaxes the assumption of identical preferences among family members and explicitly models a bargaining process in a game theoretic framework. Bargaining models tend to focus on the effect of policy and observable characteristics (e.g., labor income) on “threat points” (e.g., divorce). However, non-cooperative bargaining models do not necessarily result in efficient (Pareto optimal) decisions. If decisions are not at least assumed to be optimal, econometricians cannot use observational data to evaluate the bargaining models. Cooperative bargaining models predict Pareto efficient allocations under certain bargaining schemes. However, cooperative models still require strong assumptions about the bargaining process that make the results of any empirical rejection of the model unclear as to what is rejected—the prediction or just the particular bargaining assumption.

The collective model further relaxes the assumptions about the bargaining processes by minimally assuming “collective rationality”—that observed household decisions are Pareto efficient. Collective models result in “share rules,” or “Pareto weights” that combine presumed heterogeneous individual utility functions in the collective household decision problem. Objects of decision-making can include formal and household labor, leisure, consumption of market goods, and consumption of public household goods, which can include children. Even assuming collective rationality, the collective model still faces significant empirical challenges to identify the sharing rule (Cherchye, DeRock, Lewbel, & Vermuelen, 2015). If the collective model includes household production, which would be necessary to consider the production of health, then time is no longer simply allocated between labor and leisure. As a result, the collective model with household production requires even more assumptions about household preferences and/or more detailed data on individual consumption of household goods to identify the sharing rule (see Apps & Rees, 1996, 1997; Chiappori, 1997).

Jacobson, Bolin, and Lindgren undertake a significant theoretical exploration of household decision-making about health over three consecutive papers. Jacobson (2000) first extended the Grossman (1972) model of individual investment in health capital to include individual contributions to other family members’ health production. Jacobson (2000) concluded, consistent with Grossman, that the family will invest in the health of each family member to the point where the marginal (lifetime) utility of health capital equals the marginal cost of health capital. The key observation is that this equilibrium condition does not result in equal investments in health among family members in part because of differential health benefits and rates of health depreciation. However, Jacobson (2000) assumed “common preferences,” the unitary model approach that has been decisively rejected in the empirical literature.

Bolin, Jacobson, and Lindgren (2001) moved from the unitary model to a cooperative bargaining model. While spouses have identical utility functions, differences in the elements of their utility functions result in different demands for health. Consistent with the focus on threat points in bargaining models, Bolin et al. (2001) concluded that the risk of divorce causes spouses to underinvest in each other’s health because health capital cannot be divided upon divorce. Similarly the risk of divorce causes parents to underinvest in child health, which is explicitly modeled as a public good for which parents can have different demands and make different contributions. Finally, Bolin, Jacobson, and Lindgren (2002) posit a non-cooperative bargaining model by relaxing the assumption that spouses can enforce their strategic decisions—in other words, each can renege on any agreement to allocate resources to each other’s or to a child’s health. The non-cooperative model exposes a “free-rider” problem where the family underinvests in all family members’ health. While Bolin et al. (2001, 2002) do not offer empirical estimates, their predictions are consistent with observations of divorced parents underinvesting in child health.

Basu and Meltzer (2005) were the first to offer a collective model to theoretically and empirically demonstrate “spillover effects” of illness on family members. The Basu and Meltzer model is flexible enough to model the central trade-off between quantity (longevity or survival) and quality (side effects, caregiving burden) of life. Their equation separates the effect of illness into individual and family effects:

Totaleffectofanadversehealthstate=Directeffectonpatient’sutility+indirecteffectonpatient’sutilitythroughfamilymembers’utility+directeffectonfamilymembersutility

The italicized elements (added) adjust the individual’s utility by the utility value associated with the family.

The Basu and Meltzer (2005) paper spawned a mini-industry of empirical estimates of spillover effects for various types of illness and family members. Basu, Elstein, and Meltzer (2010) innovate empirical methods to directly estimate caregiver (dis)utility using time trade-offs (TTO) and find non-trivial spillover effects of prostate cancer treatment on spouses. Many other empirical papers in this space have no theoretical component, but in effect they maintain the collective model assumption of collective rationality when they draw conclusions based on observed behavior. Wittenberg and Prosser (2013) offer a review of spillover effects of illness and find small negative effects of caregiving on caregiver well-being but no consensus in the literature on measurement methods. Similarly, Krol, Papenberg, and van Excel (2015) review the literature on the spillover effect of providing informal care for several specific diseases for which informal care is prevalent (e.g., Alzheimer’s disease). They find that most cost-effectiveness analysis (CEA) studies do not include informal care (77 out of 100), and among the studies that include spillover effects, there is substantial variation in how spillover is measured (e.g., time costs only, time and health costs). Bobinac, van Excel, Rutten, and Brouwer (2010) disentangle a “caregiving effect” associated with actually providing care to an ill family member with a “family effect” reflecting an individual’s utility from a family member, which is presumably higher when that family member is healthy. By directly eliciting well-being measures from a Dutch sample of caregivers, Bobinac et al. (2010) find that both caregiving and family effects are statistically significant and of comparable magnitude, leading to their conclusion that caregiving effects are overstated if family effects are ignored. Lavelle, Wittenberg, Lamarand, and Prosser (2014) explore the variation in spillover effects of chronic illness among different family members using an online survey in the United States. They find substantial variation in spillover effects on different family members: the effects on parents with ill children and on adult children with ill parents are greater than those on spouses with ill partners. That illness affects different family members differently is perhaps not surprising, but this finding should prompt broader definitions of family beyond just spouses and parents and children to include siblings, grandparents, and other members of non-traditional families who are also likely to be influenced by and influence health and household well-being.

This article will next explore the empirical literature more specifically along the life course from parents’ investments in child health through adolescence to marriage, retirement, and end-of-life care. The theoretical connections and implicit theoretical assumptions used to tie together disparate literatures and identify areas in need of additional theoretical development are discussed along the way.

Family Investments in Child Health

The economic focus in the child-health literature is on estimating the causal effects of maternal choices using an infant health production function as the underlying theoretical framework. The infant health production function can be thought of as one element of a household production function conceptualized in the theoretical literature (e.g., Apps & Rees, 1996, 1997;Chiappori, 1997). The underlying assumption in much of this literature is that altruistic parents use their household production time and market goods to invest in their child’s health.

Maternal Investments in Infant Health

Pregnant women make many decisions that potentially affect the health of their unborn children including what and how much to eat, whether and how much prenatal care to get, as well as decisions about negative behaviors such as smoking, drinking, and using illicit drugs. The “fetal origins hypothesis” in the clinical literature brought attention to the idea that early-life conditions can have lasting consequences (Barker, 1990, 1998), and this hypothesis has been explored by numerous economists (e.g., Almond & Currie, 2011). While older children can certainly invest in their own health, for the first few years of life, child-health investment is clearly in the hands of the family, often the mother. One complication of estimating infant health production functions is the endogeneity of the mother’s choices, which may be correlated with unobservable factors that affect both her choice and the infant health outcome. Researchers have addressed this endogeneity in a variety of ways, as noted below.

This section focuses on mothers’ choices associated with maternal weight, prenatal care, the use of substances in pregnancy (smoking, alcohol use), breastfeeding, and the variation in maternal investments associated with birth order because there has been considerable economic work in these areas. The most often examined infant health outcome is birthweight. Low birthweight (defined as birthweight less than 2500 grams) is a strong risk factor for infant mortality and morbidity, is easily measured, and is often available in social science data sets. However, it is worth keeping in mind that many low-birthweight children (even the very smallest) have no serious health problems (Reichman, Corman, Noonan, & Dave, 2009). Thus, using birthweight to proxy infant health can lead to incorrect inferences about the production of infant health. Other outcomes are noted where appropriate.

Pre-Pregnancy Obesity and Gestational Weight Gain

Given the obesity epidemic present in nearly all developed countries, there has been interest in how pre-pregnancy obesity and/or gestational weight gain (GWG) might affect outcomes at birth. Early work using ordinary least squares (OLS) models to compare a cross-section of mothers tended to find a negative effect on birth outcomes. For example, obese mothers were found to have a higher probability of cesarean section births, were less likely to breastfeed, and had babies who were particularly large (macrosomic) (see Averett & Fletcher, 2016, for a review of this literature). However, models that use mother fixed effects to control for endogeneity reach somewhat different conclusions. For example, Ludwig and Currie (2010), using a sample containing all births in Michigan and New Jersey from 1989 to 2003, report that maternal weight gain during pregnancy increases birthweight independently of genetic factors. However, they do not have information to control for the mother’s pre-pregnancy weight. A study by Yan (2015) using both maternal fixed effects and controlling for pre-pregnancy weight also reports adverse effects of both excess maternal weight gain and low maternal weight gain with respect to infant birthweight. Averett and Fletcher (2016) both control for pre-pregnancy weight and examine a broader set of outcomes. In maternal fixed-effects models using data from the National Longitudinal Survey of Youth 1979 cohort, they find that pre-pregnancy obesity increases weeks of gestation, lowers the probability of low birthweight and preterm birth, lowers the probability that the infant is small for gestational age, but raises the probability that the newborn is large for gestational age. They find no effect on the probability of a cesarean section birth.

Maternal Cigarette Smoking During Pregnancy

A key goal of public health policies is to reduce costly adverse birth outcomes to which prenatal smoking is thought to be a crucial contributor. Women’s harmful health behaviors such as smoking appear inconsistent with the assumption of altruism or the unitary model of household utility, but they could reflect a trade-off between her utility from smoking and her concern for her unborn child. Consistent with a “rational addiction,” which suggests a trade-off between happiness today and happiness tomorrow, these bad health behaviors suggest a possible trade-off between the mother’s utility today and the child’s utility tomorrow. Her smoking may also result from underestimating the harm to the infant. The literature frames prenatal smoking as another input into the infant health production function without always directly addressing the intra-household trade-offs.

There is a debate about the extent to which the effect of prenatal smoking on infant health outcomes is causal in OLS models because a woman’s choice to smoke is likely not independent of her other choices that affect her child’s health. Fertig (2010) compares year-by-year OLS estimates of the impact of smoking on birthweight over time to assess the degree of selection into smoking. She reports that OLS estimates based on 1958 data from the United Kingdom are considerably smaller than those based on 2000 data, suggesting that selection can explain a substantial portion of OLS estimates.

Scholars have used various methods to address the endogeneity of cigarette smoking in the infant health production function. One study of about 1.5 million births in New Jersey that compared mothers who smoked during one pregnancy but not the other finds that smoking increases the probability of low birthweight (Currie, Neidell, & Schmieder, 2009). Other scholars leverage policies designed to reduce smoking as natural experiments to address the endogeneity of maternal smoking. For example, Yan (2014) evaluates the impacts of a minimum cigarette purchase age of 21 implemented in the state of Pennsylvania on prenatal smoking and infant health. Using a regression discontinuity method, Yan (2014) shows that a smoking age of 21 reduces the prenatal daily cigarette consumption by 15% and lowers the incidence of low-birthweight infants by 19%. In contrast, taking advantage of county-level variation in smoking bans and using a differences in differences research design, Gao and Baughman (2017), report little effect of smoking bans on birthweight, length of gestation, Apgar scores, or incidence of cleft palate, but they do find a small effect on the probability of low birthweight. Using a difference-in-difference approach, Simon (2016) uses state and year variation in state cigarette excise taxes as a proxy for exposure to cigarette smoke in utero. He finds significant reductions in days missed from school and in doctor visits for children potentially exposed to cigarette smoke. Finally, Yan (2013) focuses on the timing of smoking cessation. Using mother fixed effects, he finds quitting during the first trimester essential as smoking later in the pregnancy leads to adverse infant outcomes.

Prenatal Care

The goal of prenatal care is to promote the health of the mother, child, and family throughout the pregnancy and delivery to foster the child’s development. Although the vast majority of mothers giving birth in developed countries receive some prenatal care (an estimated 95% of pregnant women in the United States receive some prenatal care; Currie & Rossin-Slater, 2015), past research has not always found compelling evidence that early or adequate prenatal care has favorable effects on birth outcomes. Because prenatal care is clearly endogenous, scholars have had to be clever with their identification strategies. For example, Reichman et al. (2009) use an unusually rich data set on economically disadvantaged families (Fragile Families) to control for typically unobserved variables, including whether the pregnancy was wanted and maternal health endowments. They find that these additional variables are important inputs to infant health but that their exclusion does not markedly bias the effect of prenatal care on infant outcomes, including birthweight. They report small, insignificant effects of prenatal care on infant birthweight. However, Abrevaya and Dahl (2008) find positive effects of prenatal care on birthweight using data on all sibling births in Arizona and Washington from 1992 to 2002 and a mother fixed effects framework. Finally, Sonchak (2015) reports mixed findings after instrumenting prenatal care with Medicaid reimbursement rates, which vary by state. Women on Medicaid are economically disadvantaged and more at risk for adverse birth outcomes. She finds that women living in states with higher Medicaid reimbursement rates are more likely to have prenatal care. However, the effect of prenatal care on birthweight varies by race, with no effect for black disadvantaged mothers but a positive and significant effect for white disadvantaged mothers.

Prenatal care may also affect maternal outcomes. A study by Yan (2017) using a unique data set of over 100,000 mothers using maternal fixed effects shows that poor prenatal care leads to a host of adverse maternal outcomes such as prenatal smoking, complications during labor, and smoking both pre- and postpartum. However, prenatal care may also lead to better outcomes postpartum by introducing the mother to healthier behaviors, helping her set up well-baby visits, and encouraging breastfeeding. Using data from the Fragile Families study, Reichman, Corman Noonan, and Schwartz-Soicher (2010) estimate the effect of prenatal care on maternal postpartum smoking, preventive health care visits for the child, and breastfeeding. They find that first-trimester prenatal care appears to decrease maternal postpartum smoking and increases the likelihood of well-baby visits. There is also suggestive evidence that prenatal care may increase the probability of breastfeeding.

Breastfeeding

Public health campaigns exhort mothers to breastfeed because breastfeeding is believed to be associated with many health benefits for both infants and mothers. Economists have sought to determine if any positive correlation is causal. Using a recursive bivariate probit model, Green (2011) uses U.S. data from the Fragile Families data set to examine whether differences in asthma diagnosis can be attributed to differences in breastfeeding incidence and duration. She finds that breastfeeding may exert a protective effect against asthma for young children but not for older children. Using a sample of births from the United Kingdom, Fitzsimons and Vera-Hernández (2013) instrument breastfeeding with whether the birth occurred on a weekend (making it less likely that the mother received training regarding breastfeeding) and find no effect on various measures of child health starting at nine months, although they find positive effects on child cognitive ability. Wehby (2014) uses a mother fixed-effects model to account for family-level unobservable confounders when examining the effect of breastfeeding on child disability. He finds that prolonged breastfeeding is associated with reduced risk of child disability. Breastfeeding may also protect against childhood obesity, as Modrek et al. (2017) discover in their analysis using an instrumental variable approach where the instrument for breastfeeding is whether the hospital’s policies are supportive of breastfeeding.

Birth Order and Maternal Investment in Child Health

There is reason to believe that women may not invest equally in children of different birth order. Previous work has shown that within families, time spent reading to or playing with a child is greater for first-born children (Price, 2008). Despite a large literature on education and birth order, which has tended to find a clear advantage to being born first, relatively little is known about how maternal health investments vary with birth order. To investigate if mothers make differential investment in children by parity, Buckles and Kolka (2014) estimate within-mother differences in pre- and post-natal behaviors and find that mothers are less likely to breastfeed higher-parity children. Research using Danish register data (Brenøe & Molitor, 2018) and family fixed effects finds large and positive differences in health at birth that are robust to various definitions of health and hold irrespective of family size: lower-parity children are less healthy. These findings suggest a role for early maternal investment in determining birth-order effects on health.2 Much of this work is in its infancy, and there is more to do to fully understand the mechanisms behind these findings.

The research presented in this section offers ample evidence that maternal investments in child health are important. What women choose to do while pregnant and in the postpartum period matters to child health across various investments and numerous outcomes using data from many different countries.

Childhood and Adolescence

As children age, both parents have the opportunity to invest in the health of their child, and children have more control over their own health. Cognitive and non-cognitive capabilities such as self-regulation, motivation, time preference, farsightedness, penchant for risk taking, and other characteristics can affect the evolution of health capital by influencing choices made by both parents and children (Heckman, 2012).

This section considers how family structure and time allocation of parents affects childhood obesity as a specific child health outcome. This section then explores adolescence, which brings concerns over the health-related decisions that children may make as they explore boundaries. There is a large literature that examines the determinants of adolescent risky behavior (drinking alcohol, smoking cigarettes, and sexual intercourse). Some of this literature focuses specifically on how family inputs might affect these behaviors.

Childhood Obesity

Rising maternal employment rates have occurred at roughly the same time that childhood obesity started to rise, hence scholars have explored whether there is a causal link. In theory, parental employment may have two effects—the higher income from employment may allow the family to purchase healthier but more expensive foods such as fresh fruits and vegetables. On the other hand, if both parents are working, the family may rely more on fast food or pre-packaged foods, which save time but are calorically denser than fresh foods. In one of the first studies to examine this, Anderson, Butcher, and Levine (2003) report evidence that a child is more likely to be overweight if his or her mother worked more hours per week over the child’s life. Subgroup analysis reveals that children of working mothers with higher socioeconomic status are particularly affected by their mother’s work intensity. Extending their work, Cawley and Liu (2012), frame their analysis of parental time use and childhood obesity based on the theoretical work by Apps and Rees (1996, 1997) by including investments in child health as an explicit element of household production. Using data from the American Time Use Survey, they explore working women’s time use and report that a plausible mechanism for why maternal work may impact obesity is that working mothers spend less time investing in child diet and physical activity.

In related work Anderson (2012) considers how parental employment may affect obesity by disrupting family routines including shared mealtimes. For example, dual-career families typically have to make more use of child care including before and/or afterschool care. The stability of these arrangements may cause families to deviate from routines that may be healthier for their children. However, they conclude that family routines cannot explain the deleterious effect of maternal employment intensity on children’s obesity.

Family Structure, Adolescent Health, and Adolescent Risky Behavior

The incidence of the intact two–biological-parent (traditional) family has been steadily decreasing since the 1960s. The effects of growing up with a single parent on cognitive and educational outcomes are fairly well established in the literature, but less is known about the effects of living in a single-parent family on the health of children. Families with non-traditional structures have, on average, less income and less time available, which can lower their investment in child health. Identifying the effect of family structure can be challenging since child health itself can be a predictor of family structure. For example, a sick child may be a precursor to a divorce (Reichman, Corman, & Noonan 2004), and unobservables might affect both the health outcome and the family structure. Exogenous changes in family structure are hard to come by, making identification of any effect of family structure on health challenging. Conway and Li (2012) look broadly at the effect of family structure on child health outcomes acknowledging the difficulty of addressing the endogeneity of family structure. They find that some non-traditional family structures may have detrimental impacts on child health, although they cannot claim a causal relationship.

Slade, Beller, and Powers (2017) examine how parental absence affects two health outcomes, self-rated health and mental health, as well as smoking behavior of adolescents., They frame their empirical work in a theoretical model of parental investment in child health which occurs over the course of child’s life consistent with the Bolin et al. (2001, 2002) extensions of the Grossman (1972) health capital models. While they recognize the potential for reverse causality, they have a rich set of covariates and focus on child health at different points in childhood, arguing that certain periods in childhood are less susceptible to reverse causality. Using data from the National Longitudinal Survey of Adolescent Health (AdHealth), they report lasting effects of parental absence on health, particularly evident for girls.

Smoking and drinking are critical problems in adolescence that can have long-term adverse impacts on health and socioeconomic factors. Fletcher and Sindelar (2012) focus on family stressors including divorce and how they affect adolescent substance initiation. Using data from the National Educational Longitudinal Study (NELS) and a school-level fixed-effects approach, they report that divorce significantly increases the probability of smoking and drinking, although the results for drinking are less precisely estimated.

Within a family, parental supervision may lessen the likelihood that an adolescent will engage in risky health-related behaviors as established in work by Averett, Argys, and Rees (2011). However, See (2016) presents evidence that unobservables may drive part of that relationship. Related to supervision is how parents talk with their children and what behaviors they model. Averett and Estelle (2014) ask whether parental communication regarding risky sexual behavior (engaging in sex, having unprotected sex) can deter this behavior. However, communication itself is endogenous with adolescent behaviors. In other words, parents may communicate with their kids precisely because they believe their adolescent is engaged in risky behaviors. OLS models in this case are clearly biased, so they use an instrumental variables (IV) approach and find smaller and less precise estimates of the effect of parental talk on risky sexual behavior compared to OLS estimates.

Another aspect of family structure is the presence and number of siblings. Older siblings in particular may either expose their younger siblings to risky behaviors or even encourage them to engage in these behaviors. Argys, Rees, Averett, and Witoonchart (2006) investigated the association between birth order and adolescent behaviors such as smoking, drinking, marijuana use, sexual activity, and crime. Their estimates show that middle-borns and last-borns are more likely to use substance and be sexually active than their firstborn counterparts. However, Altonji, Cattan, and Ware (2017) question whether the effect of older siblings reflects causal influences or results from shared genes and environment. Their results suggest that only a small fraction of the correlation may be causal.

It is clear from the literature on family structure and child/adolescent health that families matter. The absence of a parent, the event of a divorce, the presence of siblings, and parental communication can all have an effect on the health of the child/adolescent. That said, there is more work to be done here, especially to shore up the case that the estimates are causal and to understand the underlying mechanisms.

Relationships and Health

The strong positive correlation between marriage and health is such an empirical regularity that it has been described as “one of the most robust in the social sciences” (Liu, 2012). Researchers have theorized that this positive correlation could stem from a marriage protection effect: married couples look after each other, making sure that they have regular doctor visits, caring for one another when they fall ill, providing companionship and support in rough times, and keeping each other from engaging in risky behaviors such as smoking and excessive alcohol use. The alternative explanation is that married couples are healthier because of a marriage selection effect: those who are in better health are more desirable as marriage partners than those who are not. However, adverse selection is also possible: those who are less healthy seek out a partner who can care for them.

To determine if marriage is the cause of better health, researchers have often relied on individual fixed-effects models that difference out time invariant unobservable factors. These studies have used many different measures of health and health behaviors, but some generalizations can be made. Married men and women have lower levels of negative health behaviors than those who are not married. They report less problem drinking than divorced or widowed men and women. However, perhaps not surprisingly, men report far more problematic health behaviors than women regardless of marital status (Averett, Argys, & Sorkin, 2013). Married men and women are also more likely to get regular health checkups and engage in preventive health care, such as mammograms and dental screenings (Pylypchuk & Miller, 2014). They also tend to report better health when they respond to a question asking how they would rate their overall health, and they live longer than their unmarried counterparts (Wilson & Oswald, 2005). However, married individuals are more likely to be obese, perhaps because they are no longer searching for a mate and are less inclined to keep themselves in good shape (Averett et al., 2013).

Two studies go beyond fixed effects to separate the selection from the protection effect. One problem with fixed effects is that those who do not change relationship status do not contribute to identification. Using the Blundell and Bond dynamic panel data estimator that uses equations in both differences and levels, Averett and Kohn (2014) find that being in a relationship is good for health, but the benefits are not unique to marriage—cohabitors benefit as well. In another paper, Kohn and Averett (2014) propose a random coefficient mixed logit model to estimate the unobserved heterogeneity associated with both health and relationship choice to identify the selection effect. They find that after more effectively controlling for selection, marriage is not universally better for health. Rather, cohabitation benefits the health of men and women over 45, being never married is no worse for health, and only divorce marginally harms the health of younger men. Using this same method, Sabia, Wooden, and Nguyen (2018) focus on same-sex relationships and health using data from Australia. They find evidence that opposite-sex partnerships are associated with improved health but weaker evidence of such a link for same-sex partnerships.

Thus, the effect of marriage on health appears to be a complex combination of both the selection and protection effects. Moreover, these effects vary with age and relationship status. Furthermore, the effect of relationships on health may change in the future as economic, social, and institutional factors influence the composition and nature of relationships.

Effect of Health on Retirement and Family Finances

At this point the focus changes from the production of health to the effect of health as an input to important life-cycle decisions. Following the theme of this article, rather than the effect of own health, the effect of family members’ health on decisions to retire and on life-cycle financial decisions in retirement are explored. These literatures increasingly include one’s own health and access to health insurance as an important causal variables. However, there are smaller literatures that include the influence of health or health insurance of a family member on an individual’s decision to retire and even fewer that model a family’s joint decision-making. The literature in this area is a mix of structural papers with causal effects identified through exclusion restrictions on model dynamics and functional forms as well as causal effects identified using various econometric methods.

Retirement Decisions

In contemporary developed economies, households often face joint retirement decisions between husbands and wives. Empirically, spouses retire together more frequently than individual models of retirement would suggest, and this joint retirement observation spans different national institutional settings and cultures, though with notable gender differences (Ho & Raymo, 2009; Hospido & Zamarro, 2014; Warren, 2015). Blundell, French, and Tetlow (2016) provide a thorough review of both the theoretical and empirical issues associated with retirement. They include a detailed review of joint models of retirement between husbands and wives and strongly conclude that “ignoring the family context can clearly distort our understanding of retirement behavior and the impact of retirement policies” (Blundell, French, & Tetlow, 2016, p. 111).

Blundell, French, and Tetlow (2016) discuss three mechanisms that link spouses’ retirement decisions: budget constraints, complementary preferences for leisure, and correlated shocks. However, the shocks emphasized in the literature are predominantly wage shocks (e.g., due to involuntary layoffs) rather than health shocks to the non-retiring spouse. Spousal health shocks open a fourth mechanism linking spousal retirement decisions, but the direction of the effect is a priori unclear. On the one hand, a shock to a spouse’s health may cause the working spouse to retire to provide informal care and/or spend what may be perceived to be a shorter life expectancy together. This latter motive is consistent with complementary preferences for leisure addressed in the broader joint retirement literature. On the other hand, a spouse’s health shock may affect the household budget constraint, causing the working spouse to keep working to earn money and/or maintain access to health insurance. Access to employer-provided health insurance (EPHI) and retiree health insurance (RHI) further complicates the channels by which health and finances interact within families to drive retirement decisions.

The limited literature on the effect of spousal health shocks has found conflicting evidence on whether husbands or wives work more or less. That said, an emerging consensus in the literature is that there is a socioeconomic status (SES) gradient in spousal responses: those with lower SES remain in the labor force, while those with higher SES retire. Thus, the differences in the literature can be due in part to different SES distributions in the samples. Another suspected cause of different results is different measures of health shocks, different measures of retirement, and different institutional features in different countries. Garcia-Gomez, van Kippersluis, O’Donnell, and van Doorslaer (2013) summarize the conflicting literature and offer evidence from acute hospitalizations in the Netherlands using a matched difference-in-difference identification strategy. They find that women are more likely to start or remain working in response to a husband’s hospitalization, while husbands are more likely to retire. McGeary (2009) also reviews the literature and offers evidence on the effect of reported health diagnoses on the probability of retirement in the United States using a first-differenced specification to control for time-invariant unobservable heterogeneity. She finds that women are more likely to retire in response to heart disease or arthritis of her spouse but less likely to retire in response to a decline in her spouse’s activities of daily living (ADLs). Men, by contrast, are more likely to retire in response to their wives declining ADLs, but they have no response to other new health diagnoses.

Evidence on the effect of health insurance on the joint retirement decisions of married couples is more limited and mixed than the substantial evidence of own health insurance on own retirement (see Gruber & Madrian, 2004, for a review of the individual literature).3 Blau and Gilleskie (2006) and Kapur and Rogowski (2007) model the joint retirement decision of couples in the presence of EPHI and RHI and find only modest effects; however, these papers include only own insurance, not insurance coverage by a spouse. Congdon-Hohman (2015) explicitly models a husband’s retirement decision when his decision affects his wife’s access to health insurance. Using several probit specifications with the Health and Retirement Survey (HRS), he finds that the husband is 30% less likely to retire if the wife is at risk than if neither husband nor wife were at risk. Notably, Congdon-Hohman (2015) finds too few observations of husbands covered by their wife’s EPHI or RHI to estimate the effect of a husband’s access to insurance on a wife’s retirement decision. Kapur and Rogowski (2011) focus on the effect of EPRI on the retirement of women, including single women, women in dual-earner couples, and women in single-earner couples where either the man or the women is the single earner. Using random-effect probit models (bi-variate probits for joint spousal retirement decisions) and the HRS, they find that the effect of health insurance on retirement for women depends on whether or not their husband is working, offering suggestive evidence (not causal because of their inability to control for the endogeneity of job choice) that there are meaningful cross-effects of insurance among spouses. Boyle and Lahey (2016) estimate the effect of a husband’s gaining government health insurance (an expansion of Veterans Administration [VA] insurance) for himself on his wife’s labor market participation. They use a difference-in-difference strategy measuring the labor market participation of women with and without spouses covered by the VA before and after the VA insurance expansion. Again, the effects can cut both ways: government insurance can prompt the husband to retire, causing the wife to retire also if complementary preferences for leisure dominate; however, since in this case the insurance is only for the husband, health expense risk and/or the joint budget constraint can cause the wife to work. They find that the budget mechanism leads women to work more and that the effect is, not surprisingly, stronger at lower SES levels.

Finally, while the literature has begun to dig deep into the joint retirement decisions of husbands and wives with respect to health, there is scant literature on the effect of the health of family members other than spouses. For example, parents of adult children with disabilities or other health impairments may alter their retirement decisions as a result. In addition, grandparents are increasingly involved in caring for their grandchildren, which like other types of informal care, may have significant effects on both their health and labor market decisions. Anecdotally, drug addiction (e.g., the opioid epidemic in the United States circa 2015) has impaired the health of parents, prompting grandparents to take over parenting their children’s children. There appears to be no work specifically tracing intergenerational health effects on retirement. However, two papers explore the effect of being a grandparent on retirement and suggest that intergenerational family structures may have more impact on retirement decisions than the literature recognizes (Kridahl, 2017; Lumsdaine & Vermeer, 2015).

Finance Decisions

Substantial literatures attempt to explain why retired households decumulate their wealth more slowly and purchase annuities and long-term care insurance (LTCI) less frequently than standard economic life-cycle models predict (see DeNardi, French, & Jones, 2016a, for a review).4 The two predominant explanations for these financial puzzles are both entwined with family and health: bequest motives to leave money primarily to family and precautionary savings for health expenses, particularly long-term care expenses. Both bequests and long-term care expenses are also associated with the provision of informal care addressed in the next section. Despite voluminous work, there is little consensus in the literature regarding either bequests or precautionary savings and thereby no consensus answer to the financial puzzles. Work continues because answers to these puzzles are highly policy relevant as governments search for financially sustainable ways to support aging populations.

Unlike the retirement literature, which has innovated sophisticated collective decision models, the theoretical literature on “household” finances after retirement has almost universally modeled a single retired individual or a household with decisions made by a single representative agent. Hurd (1999) was the first to extend the seminal Yaari (1965) model of life-cycle consumption with uncertain lifetimes to include couples and thereby separate out a true bequest to future generations with a provision for a surviving spouse. Hurd’s theoretical model suggested that couples would decumulate wealth more slowly than singles because of the differences in the couple’s mortality rates—in other words, precautionary savings for the surviving spouse. Hubener, Maurer, and Rogalla (2014) structurally model optimal portfolio choice for retired couples and find that the primary bequest motive is provision for a surviving spouse. Both Hurd (1999) and Hubener et al. (2014) model uncertain lifetimes, but they do not model uncertain medical expenses. Moreover, neither includes children except implicitly in the bequest motive. Thus, there is clearly more theoretical work to be done in this area.

The empirical literature on household financial decisions in retirement has been equally challenging. Separately identifying the effect of precautionary health care savings and the bequest effect has been a critical econometric issue. Advances in identification have come from using additional data on LTCI (Lockwood, 2012), life insurance (Hong & Rios-Rull, 2012; Inkman & Michaelides, 2012) Medicaid (DeNardi, French, and Jones, 2016b), and strategic survey questions (Ameriks, Caplin, Laufer, & Van Nieuwerburg, 2011). Most of these papers find a significant bequest motive. However, Ameriks, Caplin, Laufer, and Van Nieuwerburg (2011) find an equally strong “public care aversion” motive consistent with precautionary savings as well as a bequest motive that extends into the middle class. Still, Ameriks et al. (2011) modeled a “household” of a single individual without allowing for informal care by potential beneficiaries of bequests either strategically or as a substitute to formal spending on long-term care.

The models of life-cycle consumption are complex requiring substantial simplifying assumptions and modeling trade-offs. Pelgrin and St. Amour (2016) summarize in a convenient table many of the modeling trade-offs in the literature, including health production, the retirement decision, endogenous health expenditures, and morbidity and mortality risk. However, all of the papers included in this table model individual, not household decisions. Even in the context of bequests, there is little consensus as to whether to model bequests as a form of altruism, self-satisfaction for the individual leaving the bequest, or exchange for long-term care services from children. Nonetheless, the literature modeling individual life-cycle consumption has concluded that health shocks and health care spending are relevant to explain life-cycle investment and consumption. If, as explored in more detail in the next section, family dynamics are relevant in formal and informal long-term care decisions, then additional work to model family decision-making more extensively may provide additional insights on these pervasive financial puzzles. Indeed, in their review, DeNardi et al. (2016a) note that “parents and children might respond to each other’s needs with transfers of both time and money. An important avenue for future research is better documentation of these interactions and better modeling of the motivations behind them” (p. 200).

Health, Wealth, and Family at the End of Life

There has been an explosion of economic work on end-of-life (EOL) care, correlated with globally aging societies of which the economists writing these papers are not immune. Phillipson, Becker, Goldman, and Murphy (2010) note that “it seems generally agreed upon that medical resources are being wasted on excessive end-of-life treatment” (p. 3). and that end-of-life care accounts for a substantial portion of health care spending and rates of spending growth. Anecdotal stories of distraught family members demanding futile care abound in the clinical literature, yet no economic literature appears to quantify the effect of family demands on end-of-life medical care spending. Moreover, multi-national estimates have found that spending in the last year of life accounts for a lower percentage of aggregate health care spending (though still a considerable proportion of an individual’s lifetime health care spending) than previously reported (see French et al., 2017). Still, looking at family rather than national finances, Dalton and LaFave (2017) present empirical work using fixed effects and sensitivity analysis on subgroups to address the endogeneity of health and wealth, and they find significant impacts of health shocks on the financial well-being of family members. Interestingly, they find differential gender effects, with male children reducing home equity and savings while female children reduce their own consumption and health expenditures in response to their parents’ declining health.

This section will examine the literature both from the elder’s perspective in terms of family inputs to improving the elder’s quality of life at the end and from the family members’ perspective in terms of the decision-making and consequences of providing informal care to aging family members. Taking a family rather than individual perspective on end-of-life care opens up new avenues to explain observed behavior and design public policy to improve outcomes at the national, family and individual levels.

End-of-Life Planning and Quality of End-of-Life Care

EOL planning encompasses family activities that are both formal (e.g., writing a living will, assigning power of attorney, purchasing LTCI) and informal (talking with loved ones about preferences). The clinical literature has found some evidence that EOL planning is associated with improved quality of care defined as increased hospice use, less death in the hospital, and more concordance with preferences stated in EOL planning (Bischoff, Sudore, Miao, Boscardin, & Smith, 2013; Brinkman-Stoppelenburg, Reitjens, & van der Heide, 2014). Thus, if EOL planning can improve outcomes, the question is: What are the family dynamics associated with whether or not a family engages in EOL planning? There is a limited literature, primarily clinical rather than economic, that addresses this question. Nonetheless, this literature suggests positive spillover effects on children from observing their parents’ EOL. Specifically, both Woosley, Danes, and Stum (2017) and Carr (2012), using data from Wisconsin and New Jersey, respectively, find that adult children are more likely to engage in EOL planning if their parents did. Interestingly, Carr (2012) finds positive role model effects but not negative ones: observing a “bad death” did not appear to prompt EOL planning. Another study by Boerner, Carr, and Moorman (2013) finds that individual advanced care strategies are associated significantly with general family functioning, criticism from children, gender, ethnicity, age, and educational attainment but not marital relations.

Not surprisingly, there is more economic literature about family decision-making to purchase LTCI. LTCI is a substitute to informal long-term care addressed in the next section. The literature universally characterizes the low rate of LTCI purchase as an economic puzzle to be answered, and family factors remain a promising explanation. However, the econometrics of estimating causal effects of family factors on LTCI purchase are particularly challenging because LTCI underwriting (purchase of insurance) must occur many years before need, and complex family factors require rich data over long panels. Moreover, the theory associated with family influence on LTCI purchase offers few clear testable hypotheses. Specifically, the presence of children can reduce demand for LTCI by substituting informal care or increase demand by providing a greater incentive to preserve wealth for bequests. Van Houtven, Coe, and Konetzka (2015) clearly summarize the theoretical and empirical issues in this literature and estimate a model that addresses some of the empirical challenges by using a long panel of the HRS with extensive family-level variables as well as state fixed effects. They find, consistent with prior literature, that individual characteristics (age, wealth, education) remain primary predictors of LTCI and that spousal characteristics are significant, but other than having a co-resident child, child characteristics are not significant. However, given the theoretical and econometric issues along with the changing family and institutional circumstances in multiple countries, there is still much more work to be done to explain LTCI purchase decisions.

Provision of Informal Care

Informal care—a synonym for unpaid care the substitute for which is paid nursing home and home care—is often noted in the literature as the most common form of long-term care. Quantifying informal care is challenging, and there is no consensus methodology (van den Berg & Spauwen, 2006). Challenges include accurately measuring the time spent, avoiding double-counting housekeeping and other activities that would be done anyway, and associating a money rate with time since informal care is a non-market good. Economists have ample incentive to grapple with these challenges as informal care sits at the nexus of health, labor market, insurance market, government policy, cultural norms, and, for our purposes here, family dynamics not only between spouses but also between parents and their children and among siblings.

The vast majority of literature regarding informal caregivers models a single altruistic decision-maker (typically an adult child) by including the elder’s (typically the parent’s) health in the decision-maker’s utility function (see Norton, 2016). Papers that estimate a contemporaneous cross-section tend to find a trade-off between providing informal care and labor market outcomes on the extensive margin (probability of any work) but less consensus on the intensive margin (hours worked). Notably, estimates of both working and hours worked vary considerably due to different samples, institutional settings, and econometric methodologies. More importantly, the static papers underestimate labor market effects of informal caregiving because they do not account for the long-term difficulties of re-entering the workforce after leaving (or reducing hours) to provide informal care. Skira (2015) poses a dynamic discrete choice model that assumes an altruistic daughter who gets utility from providing care that varies with the health of her parent. Skira (2015) estimates her model with the HRS and finds that the cost of care provision for women in their fifties is comparable to two years in a private nursing home—substantially higher than costs estimated in the static papers. Schmitz and Westphal (2017) reach similar conclusions, estimating reduced-form models with German longitudinal data. They conclude not only that caregivers are less likely to work full time but also that while wages are unaffected contemporaneously, they nonetheless decline over eight years after initiating informal care.

Providing informal care can also have negative contemporaneous and long-term effects on the caregiver’s own health. However, estimating a causal effect of informal caregiving on caregiver health must address a significant selection effect: those in worse health are less likely to be in the labor market and thereby more available to provide informal care in the first place. In addition, other observable and unobservable individual characteristics (wealth, preference for leisure) will be correlated with both the decision to provide informal care and caregiver health. Papers that use different data, populations, and econometric methods from fixed effects to instrumental variables and propensity score matching to address selection all find a negative effect of caregiving of a parent on the health of an adult child. DeZwart, Bakx, and van Doorslaer (2017) briefly review this literature, clearly noting the sample ages, countries, and econometric strategies of numerous papers before contributing an analysis on the effect of spousal health rather than child health. Using the European Survey on Health, Ageing and Retirement in Europe (SHARE) data and statistical matching to control for selection, they also find a significant negative effect on caregiver health when the caregiver is a spouse rather than a child.

Rather than altruism, another branch of the literature models strategic financial exchanges from parents to compensate children for informal care.5 Lopez-Anuarbe (2013) allows for both altruism and monetary transfers in a game-theoretic model of parental gifts and informal assistance from children that results in best response functions for parents (children) that are linear and positive with respect to the child’s assistance (parent’s gifts). However, Lopez-Anuarbe (2013) uses U.S. HRS data and estimates the best response functions independently using OLS and logit models that do not account for unobservable heterogeneity or any joint distribution of the error structure. It is not difficult to hypothesize sources of joint unobservable heterogeneity that may bias OLS results, in particular the quality of the relationship between parent and child. Groneck (2017) uses family fixed effects and instrumental variables as well as specific data on bequests from the HRS in an empirical model with reference to both altruism and exchange theories and finds strong evidence for the exchange of bequests for informal care from children. Notably, Groneck (2017) concludes that a written will, which is like a contract, is an important determinant for the exchange.

Two contributions that use a game-theoretic framework focus on the interactions among siblings to determine who cares for the parents. Maruyama and Meliyanni (2017) models the children’s decisions of where to live (far or near parents) assuming that those who live closer to their parents will provide more care. Maruyama and Meliyanni (2017) both posits a theoretical model and uses a cross-section of the HRS to estimate altruism, private cost, and cooperation. Though their model requires significant restrictions for tractability and identification (including exogenous income and one-time location decisions), they are the first to quantify the free-rider problem associated with providing the public good of informal care. They find moderate altruism and cooperation and limited strategic behavior (little first-move advantage) but significant free-riding resulting in predictions of 18.3% more parents receiving care from children than actually observed in the data (Maruyama & Meliyanni, 2017). Komura & Ogawa (2017) focus on the effect of income on male siblings’ location choice and illustrates theoretically that when incomes are similar, the elder child has an incentive to exercise his or her first mover advantage and move farther away. However, in contrast to other literature, Komura and Ogawa (2017) find that the child earning more (which due to greater education is often the elder) provides informal care no matter where the younger child locates (Komura & Ogawa, 2017). While Komura and Ogawa (2017) do not empirically estimate the model, they do suggest that the findings can explain the differences in care arrangements in Western and Eastern countries. Rather than children moving to help elderly parents, a related literature considers parents’ moving closer to children to receive assistance (see the seminal work by Litwak & Longino, 1987).

Thus, our chapter has come full circle. Just as parents’ decisions are critical to the health of their young children, adult children’s decisions are critical to the health of their aging parents. These decisions at any stage in the life cycle cannot be disassociated from related economic decisions about labor market activity and household finances. Moreover, these decisions must be considered in the context of diverse institutional and cultural circumstances.

Conclusion

Health is intertwined with every economic decision starting with whether to have children and then how to invest in them, whether and how much to work inside and outside the home, who and when to marry, when to retire, and how to care for ill family members at all stages of life. Our analysis of the complex interrelationships between health and family over the life cycle has exposed a potential mechanism driving economic inequality: the intergenerational transfer of disadvantage associated with investment in and spillover effects from health within the family. The literature is clear that decisions of family members affect an individual’s health, which in turn affects nearly every other economic outcome, including education and labor market success. The literature is also clear that there is a spillover effect of an individual’s health on both the physical and financial health of family members. As a result, it is conceivable that a health shock can send shockwaves through a family in multiple directions over several generations.

For example, in our aging societies adult children provide substantial informal care, which can have long-term effects on both the caregiver’s own health and wealth, which can be significant and persistent. It is perhaps not surprising that elders with greater wealth rely less on family for long-term care, putting a greater burden on families with less wealth. More specifically, children with poor labor market prospects are more likely to leave the labor market to provide informal care, thereby further reducing their labor market prospects going forward. If these caregivers then have less time and economic resources to invest in their children’s health, then these children may in turn have more health issues, resulting in worse economic outcomes, which may place greater burdens on their children, and so on.

While intriguing, establishing any such intergenerational transfer of health and wealth disadvantage requires substantially more theoretical and empirical work. A substantial theoretical challenge is to model game-theoretic strategic behavior beyond two-person households—either spouses or parents and children—as adding decision-makers increases the model’s complexity exponentially. Such complexity increases even more when strategic decisions play out dynamically over time. Nonetheless, to establish any intergenerational transfer of health disadvantage requires a coherent model of household decision-making over the life cycle that can capture the trade-offs between individual and family decisions about not only health but also other human and financial capital investments.

Dynamic bi-directional causality between health and wealth plus myriad sources of unobservable heterogeneity make the empirical challenges of establishing the causal effects associated with health and family equally daunting. However, as social survey panel data sets get longer and include richer health data, opportunities to empirically test theory and establish causation improve. Unfortunately, many data sets still lack detailed information on time inputs and individual consumption of household public goods (e.g., child health) by family members, making time contributions and sharing rules in the presence of household production of health difficult to assess empirically. Nonetheless, advances in econometric methods to control for selection in dynamic panel data models plus new natural experiments and creative data sources offer future researchers exciting empirical opportunities.

Empirical economists should also consider that changing family dynamics, health technologies, and social policies may require reconsidering apparently settled research questions. For example, communications technology is increasing the frequency with which children can communicate with their parents. Rather than rationing expensive long-distance phone calls, families can now text, tweet, post, and chat almost continuously. In addition, technology has lowered the cost of monitoring family members. Whether these technological advances and those to come meaningfully change decisions about family dynamics with respect to health is an interesting avenue for research. Similarly, the diffusion of evidence on the impact of early childhood investments may change family decision-making about such investments in the future. Finally, the institutional settings within which families make decisions about health continue to evolve. Such institutional considerations include not only health insurance and health care systems but also changing social mores associated with women in the labor market and men doing more household production.

As Heckman (2012) notes, health policy is family policy. This article has dealt with the theoretical and empirical underpinnings regarding families and health. Efforts to theoretically and empirically model the complex and dynamic interactions between families and health are ongoing, leaving many interesting and policy-relevant avenues for research. Such efforts are important not only to inform health and family policy but also to possibly uncover new mechanisms driving economic inequality that can inform economic policies more broadly.

Acknowledgment

We gratefully acknowledge the helpful research assistance of Jean Donovan Rasamoelison (Lafayette College), Leah Nadel, and Zelyn Kwok (Drew University). We also thank our many colleagues who specialize in the areas covered in this chapter who have talked with us about these issues over several years.

Further Reading

Household Decision Theory

Chiappori, P.-A., & Mazzocco, M. (2017). Static and intertemporal household decisions. Journal of Economic Literature, 55(3), 985–1045.Find this resource:

Munro, A. (2018). Intra‐household experiments: A survey. Journal of Economic Surveys, 32(1), 134–175.Find this resource:

Maternal Investments in Infant Health

Almond, D., Edlund, L., Joffe, M., & Palme, M. (2016). An adaptive significance of morning sickness? Trivers–Willard and hyperemesis gravidarum. Economics & Human Biology, 21, 167–171.Find this resource:

Barcellos, S. H., Carvalho, L. S., & Lleras-Muney, A. (2014). Child gender and parental investments in India: Are boys and girls treated differently? American Economic Journal: Applied Economics, 6(1), 157–189.Find this resource:

Currie, J. (2011). Inequality at birth: Some causes and consequences. American Economic Review, 101(3), 1–22.Find this resource:

Childhood and Adolescence

Cenegy, L. F., Denney, J. T., & Kimbro, R. T. (2018). Family diversity and child health: Where do same‐sex couple families fit? Journal of Marriage and Family, 80(1), 198–218.Find this resource:

Courtemanche, C., Tchernis, R., & Zhou, X. (2017). Parental work hours and childhood obesity: Evidence using instrumental variables related to sibling school eligibility (No. w23376.). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Rossin-Slater, M. (2015). Promoting health in early childhood. The Future of Children, 25(1), 35–64.Find this resource:

Tracey, M. R., & Polachek, S. W. (2018). If looks could heal: Child health and paternal investment. Journal of Health Economics, 57, 179–190.Find this resource:

Yi, J., Heckman, J. J., Zhang, J., & Conti, G. (2015). Early health shocks, intra‐household resource allocation and child outcomes. The Economic Journal, 125(588), F347–F371.Find this resource:

Relationships and Health

Mata, J., Frank, R., & Hertwig, R. (2015). Higher body mass index, less exercise, but healthier eating in married adults: Nine representative surveys across Europe. Social Science & Medicine, 138, 119–127.Find this resource:

Pylypchuk, Y., & Kirby, J. B. (2017). The role of marriage in explaining racial and ethnic disparities in access to health care for men in the US. Review of Economics of the Household, 15(3), 807–832.Find this resource:

Spillover Effects and Economic Evaluation of Illness on Families

Al-Janabi, H., Van Exel, J., Brouwer, W., & Coast, J. (2016). A framework for including family health spillovers in economic evaluation. Medical Decision Making, 36(2), 176–186.Find this resource:

Fletcher, J., & Marksteiner, R. (2017). Causal spousal health spillover effects and implications for program evaluation. American Economic Journal: Economic Policy, 9(4), 144–166.Find this resource:

Ride, J. (2018). Setting the boundaries for economic evaluation: Investigating time horizon and family effects in the case of postnatal depression. Value in Health, 21(5), 573–580.Find this resource:

Health and Retirement

French, E., & Jones, J. B. (2017). Health, health insurance, and retirement: A survey. Annual Review of Economics, 9, 383–409.Find this resource:

Long-Term Care

Barczyk, D., & Kredler, M. (2018). Evaluating long-term-care policy options, taking the family seriously. The Review of Economic Studies, 85(2), 766–809.Find this resource:

Caplin, A., Luo, M., & McGarry, K. (2018). Measuring and modeling intergenerational links in relation to long‐term care. Economic Inquiry, 56(1), 100–113.Find this resource:

Gentili, E., Masiero, G., & Mazzonna, F. (2017). The role of culture in long-term care arrangement decisions. Journal of Economic Behavior & Organization, 143, 186–200.Find this resource:

References

Abrevaya, J., & Dahl, C. M. (2008). The effects of birth inputs on birthweight: Evidence from quantile estimation on panel data. Journal of Business & Economic Statistics, 26(4), 379–397.Find this resource:

Aizer, A., & Currie, J. (2014). The intergenerational transmission of inequality: Maternal disadvantage and health at birth. Science, 344(6186), 856–861.Find this resource:

Alessiea, R., Lusardib, A., & Kapteync, A. (1999). Saving after retirement: Evidence from three different surveys. Labour Economics, 6(2), 277–310.Find this resource:

Almond, D., & Currie, J. (2011). Killing me softly: The fetal origins hypothesis. The Journal of Economic Perspectives, 25(3), 153–172.Find this resource:

Altonji, J. G., Cattan, S., & Ware, I. (2017). Identifying sibling influence on teenage substance use. Journal of Human Resources, 52(1),Find this resource:

Ameriks, J., Caplin, A., Laufer, S., & Van Nieuwerburg, S. (2011). The job of giving or assisted living? Using strategic surveys to separate public care aversion from bequest motives. The Journal of Finance, 66, 519–561.Find this resource:

Anderson, P. M. (2012). Parental employment, family routines and childhood obesity. Economics & Human Biology, 10(4), 340–351.Find this resource:

Anderson, P. M., Butcher, K. F., & Levine, P. B. (2003). Maternal employment and overweight children. Journal of Health Economics, 22(3), 477–504.Find this resource:

Antecol, H., & Bedard, K. (2007). Does single parenthood increase the probability of teenage promiscuity, substance use and crime? Journal of Population Economics, 20(1), 55–71.Find this resource:

Apps, P., & Rees, R. (1996). Labour supply, household production and intra-family welfare distribution. Journal of Public Economics, 60, 199–219.Find this resource:

Apps, P., & Rees, R. (1997). Collective labour supply and household production. Journal of Political Economy, 105, 178–190.Find this resource:

Argys, L. M., Rees, D. I., Averett, S. L., & Witoonchart, B. (2006). Birth order and risky adolescent behavior. Economic Inquiry, 44(2), 215–233.Find this resource:

Averett, S., & Estelle, S. (2014). Will daughters walk mom’s talk? The effects of maternal communication about sex on the sexual behavior of female adolescents. Review of Economics of the Household, 12(4), 613–639.Find this resource:

Averett, S. L., & Fletcher, E. K. (2016). Prepregnancy obesity and birth outcomes. Maternal and Child Health Journal, 20(3), 655–664.Find this resource:

Averett, S., & Kohn, J. (2014). Can’t we just live together? New evidence of the effect of relationship status on health. Journal of Family and Economic Issues, 35, 295–312.Find this resource:

Averett, S., Argys, L., & Sorkin, J. (2013). In sickness and in health: An examination of relationship status and health using data from the Canadian National Public Health Survey. Review of Economics of the Household, 11(4), 599–633.Find this resource:

Averett, S., Argys, L. M., & Rees, D. I. (2011). Older siblings and adolescent risky behavior: Does parenting play a role? Journal of Population Economics, 24(3), 957–978.Find this resource:

Barker, D. (1998). Mothers, babies and health in later life (2nd ed.). Edinburgh: Churchill Livingstone.Find this resource:

Barker, D. (1990). The fetal and infant origins of adult disease. BMJ: British Medical Journal, 301(6761), 1111.Find this resource:

Basu, A., Elstein, D. W., & Meltzer, D. (2010). A time tradeoff method for eliciting partner’s quality of life due to patient’s health states in prostate cancer. Medical Decision Making, 30, 355–365.Find this resource:

Basu, A., & Meltzer, D. (2005). Implications of spillover effects within the family for medical cost-effectiveness analysis. Journal of Health Economics, 24(4), 751–773.Find this resource:

Bischoff, K. E., Sudore, R., Miao, Y., Boscardin, W. J., & Smith, A. K. (2013). Advance care planning and the quality of end-of-life care among older adults. Journal of the American Geriatric Society, 61(2), 1–12.Find this resource:

Black, S. E., Devereux, P. J., & Salvanes, K. G. (2016). Healthy (?), wealthy, and wise: Birth order and adult health. Economics & Human Biology, 23, 27–45.Find this resource:

Blau, D., & Gilleskie, D. (2006). Health insurance and retirement of married couples. Journal of Applied Econometrics, 21(7), 935–953.Find this resource:

Blundell, R., Crawford, R., French, E., & Tetlow, G. (2016). Comparing retirement wealth trajectories on both sides of the pond. Fiscal Studies, 37, 105–130.Find this resource:

Blundell, R., French, E., & Tetlow, G. (2016). Retirement incentives and labor supply. Handbook of the Economics of Population Aging, 1, 457–566.Find this resource:

Bobinac, A., van Excel, N. J., Rutten, F., & Brouwer, W. (2010). Caring for and caring about: Disentangling the caregiver effect and the family effect. Journal of Health Economics, 29, 549–556.Find this resource:

Boerner, K., Carr, D., & Moorman, S. (2013). Family relationships and advance care planning: Do supportive and critical relations encourage or hinder planning? Journals of Gerontology, 68(2), 246–256.Find this resource:

Bolin, K., Jacobson, L., & Lindgren, B. (2001). The family as the health producer—when spouses are Nash-bargainers. Journal of Health Economics, 20, 349–362.Find this resource:

Bolin, K., Jacobson, L., & Lindgren, B. (2002). The family as the health producer—when spouses act strategically. Journal of Health Economics, 21, 475–495.Find this resource:

Boyle, M. A., & Lahey, J. N. (2016). Spousal labor market effects from government health insurance: Evidence from a veterans affairs expansion. Journal of Health Economics, 45, 63–76.Find this resource:

Brenøe, A. A., & Molitor, R. (2018). Birth order and health of newborns. Journal of Population Economics, 31(2), 363–395.Find this resource:

Brinkman-Stoppelenburg, A., Reitjens, J. A. C., & van der Heide, A. (2014). The effects of advance care planning on end-of-life care: A systematic review. Palliative Medicine, 28(8), 1000–1025.Find this resource:

Browning, M., Chiappori, P. A., & Weiss, Y. (2014). Economics of the family. New York: Cambridge University Press.Find this resource:

Buckles, K., & Kolka, S. (2014). Prenatal investments, breastfeeding, and birth order. Social Science & Medicine, 118, 66–70.Find this resource:

Carr, D. (2012). “I don’t want to die like that . . .”: The impact of significant others’ death quality on advance care planning. The Gerontologist, 52(6), 770–781.Find this resource:

Case, A., & Paxson, C. (2001). Mothers and others: who invests in children’s health? Journal of Health Economics, 20(3), 301–328.Find this resource:

Cawley, J., & Liu, F. (2012). Maternal employment and childhood obesity: A search for mechanisms in time use data. Economics & Human Biology, 10(4), 352–364.Find this resource:

Cherchye, L., De Rock, B., Lewbel, A., & Vermuelen, F. (2015). Sharing rule identification for general collective consumption. Econometrica, 83(5), 2001–2041.Find this resource:

Chiappori, P. A. (1997). Introducing household production in collective models of labor supply. Journal of Political Economy, 105, 191–209.Find this resource:

Congdon-Hohman, J. (2015). Love, toil, and health insurance: Why American husbands retire when they do. Contemporary Economic Policy, 33(1), 118–140.Find this resource:

Conway, K., & Li, M. (2012). Family structure and child outcomes—A high definition, wide angle “snapshot.” Review of Economics of the Household, 10(3), 345–374.Find this resource:

Corinne, R., Russell, S., Hui, L., & Robert, C. (2016). Family structure and child health: Does the sex composition of parents matter? Demography, 53(5), 1605–1630.Find this resource:

Currie, J., Neidell, M., & Schmieder, J. F. (2009). Air pollution and infant health: Lessons from New Jersey. Journal of Health Economics, 28(3), 688–703.Find this resource:

Currie, J., & Rossin-Slater, M. (2015). Early‐life origins of life‐cycle well‐being: Research and policy implications. Journal of Policy Analysis and Management, 34(1), 208–242.Find this resource:

Dalton, M., & LaFave, D. (2017). Mitigating the consequences of a health condition: The role of intra and interhousehold assistance. Journal of Health Economics, 53, 38–52.Find this resource:

Dalton, M., & LaFave, D. (2017). Mitigating the consequences of a health condition: the role of intra and interhousehold assistance. Journal of Health Economics, 53, 38–52.Find this resource:

DeNardi, M., French, E., & Jones, J. B. (2016a). Savings after retirement: a survey. Annual Review of Economics, 8, 177–204.Find this resource:

DeNardi, M., French, E., & Jones, J. B. (2016b). Medicaid insurance in old age. The American Economic Review, 106(11), 3480–3520.Find this resource:

deZwart, P. L., Bakx, P., & van Doorslaer, E. K. A. (2017). Will you still need me, will you still feed me when I’m 64? The health impact of caregiving to one’s spouse. Health Economics, 26(S2), 127–138.Find this resource:

Fertig, A. (2010). Selection and the effect of prenatal smoking. Health Economics, 19(2), 209–226.Find this resource:

Fitzsimons, E., & Vera-Hernández, M. (2013). Food for thought? Breastfeeding and child development. (IFS Working Papers No. W13/31.) London: Institute for Fiscal Studies.Find this resource:

Fletcher, J. M., & Sindelar, J. L. (2012). The effects of family stressors on substance use initiation in adolescence. Review of Economics of the Household, 10(1), 99–114.Find this resource:

French, E. B., McCauley, J., Aragon, M., Bakx, P., Chalkley, M., Michaud, P. C., . . . Kelly, E. (2017). End-of-life medical spending in last twelve months of life is lower than previously reported. Health Affairs, 36(7), 1211–1217.Find this resource:

Gao, J., & Baughman, R. A. (2017). Do smoking bans improve infant health? Evidence from US births: 1995–2009. Eastern Economic Journal, 43(3), 1–24.Find this resource:

Garcia-Gomez, P., van Kippersluis, H., O’Donnell, O., & van Doorslaer, E. (2013). Long term and spillover effects of health shocks on employment and income. Journal of Human Resources, 48(4), 873–909.Find this resource:

Green, T. (2011). Infant feeding and asthma: Is breast milk best? Review of Economics of the Household, 9(4), 487–504.Find this resource:

Groneck, M. (2017). Bequests and informal long-term care: Evidence from HRS exit interviews. Journal of Human Resources, 52(2), 531–572.Find this resource:

Grossman, M. (1972). On the concept of health capital and the demand for health. Journal of Political Economy, 80(2), 223–255.Find this resource:

Gruber, J., & Madrian, B. C. (2004). Health insurance, labor supply, and job mobility: a critical review of the literature. In C. G. McLaughlin (Ed.), Health policy and the uninsured (pp. 97–178). Washington, DC: The Urban Institute.Find this resource:

Heckman, J. J. (2012). The developmental origins of health. Health Economics, 21(1), 24–29.Find this resource:

Himmelweit, S., Santos, C., Sevilla, A., & Sofer, C. (2013). Sharing of resources within the family and the economics of household decision making. Journal of Marriage and Family, 75, 625–639.Find this resource:

Ho, J. H., & Raymo, J. M. (2009). Expectations and realization of joint retirement among dual-worker couples. Research on Aging, 31(2), 153–197.Find this resource:

Hong, J., & Rios-Rull, J. V. (2012). Life insurance and household consumption. American Economic Review, 1023, 701–3730.Find this resource:

Hospido, L., & Zamarro, G. (2014). Retirement patterns of couples in Europe. IZA Journal of European Labor Studies, 3(12), 1–18.Find this resource:

Hubener, A., Maurer, R., & Rogalla, R. (2014). Optimal portfolio choice with annuities and life insurance for retired couples. Review of Finance, 18, 147–188.Find this resource:

Hurd, M. D. (1999). Mortality risk and consumption by couples. Cambridge, MA: National Bureau of Economic Research.Find this resource:

Inkman, J., & Michaelides, A. (2012). Can the life insurance market provide evidence for a bequest motive? Journal of Risk and Insurance, 79, 671–695.Find this resource:

Jacobson, L. (2000). The family as producer of health—an extended Grossman model. Journal of Health Economics, 19, 611–637.Find this resource:

Kapur, K., & Rogowski, J. (2007). The role of health insurance in joint retirement among married couples. Industrial and Labor Relations Review, 60(3), 397–407.Find this resource:

Kapur, K., & Rogowski, J. (2011). How does health insurance affect retirement behavior of women. Inquiry, 48, 51–67.Find this resource:

Kohn, J., & Averett, S. (2014). The effect of relationship status on health with dynamic health and persistent relationships. Journal of Health Economics, 36, 69–83.Find this resource:

Komura, M., & Ogawa, H. (2017). The prodigal son: Does the younger brother always care for his parents in old age? Applied Economics, 49(22), 1466–4283.Find this resource:

Kridahl, L. (2017). Retirement timing and grandparenthood in Sweden: evidence from population-based register data. Demographic Research, 37, 957–994.Find this resource:

Krol, M., Papenberg, J., & van Excel, J. (2015). Does including informal care in economic evaluations matter? A systematic review of inclusion and impact of informal care in cost-effectiveness studies. PharmacoEconomics, 33(2), 123–135.Find this resource:

Lavelle, T. A., Wittenberg, E., Lamarand, K., & Prosser, L. A. (2014). Variation in the spillover effects of illness on parents, spouses and children of the chronically ill. Applied Health Economics and Health Policy, 12(2), 117–124.Find this resource:

Litwak, E., & Longino, C. (1987). Migration patterns among the elderly: A developmental perspective. Gerontologist, 27(3), 266–272.Find this resource:

Liu, H. (2012). Marital dissolution and self-rated health: Age trajectories and birth cohort variations. Social Science and Medicine, 74, 1107–1116.Find this resource:

Lockwood, L. M. (2012). Bequest motives and the annuity puzzle. Review of Economic Dynamics, 15, 226–243.Find this resource:

Lopez-Anuarbe, M. (2013). Intergenerational transfers in long term care. Review of Economics of the Household, 11, 235–258.Find this resource:

Ludwig, D. S., & Currie, J. (2010). The association between pregnancy weight gain and birthweight: A within-family comparison. The Lancet, 376(9745), 984–990.Find this resource:

Lumsdaine, R. L., & Vermeer, S. J. C. (2015). Retirement timing of women and the role of care responsibilities for grandchildren. Demography, 52(2), 433–454.Find this resource:

Maruyama, S., & Meliyanni, J. (2017). Do siblings free-ride in “being there” for parents? Quantitative Economics, 8, 277–316.Find this resource:

McGeary, K. A. (2009). How do health shocks influence retirement decisions? Review of Economics of the Household, 7, 307–321.Find this resource:

Modrek, S., Basu, S., Harding, M., White, J. S., Bartick, M. C., Rodriguez, E., & Rosenberg, K. D. (2017). Does breastfeeding duration decrease child obesity? An instrumental variables analysis. Pediatric Obesity, 12(4), 304–311.Find this resource:

Norton, E. C. (2016). Health and long-term care. Handbook of the Economics of Population Aging, 1B, 951–989.Find this resource:

Pelgrin, F., & St-Amour, P. (2016). Life cycle responses to health insurance status. Journal of Health Economics, 49, 76–96.Find this resource:

Phillipson, T., Becker, G., Goldman, D., & Murphy, K. (2010). Terminal care and the value of life near its end. National Bureau of Economic Research No. w15649.Find this resource:

Price, J. (2008). Parent-child quality time: Does birth order matter? Journal of Human Resources, 43(1), 240–265.Find this resource:

Prosser, L. A., & Wittenberg, E. (2013). Disutility of illness for caregivers and families: A systematic review of the literature. PharmacoEconomics, 31, 489–500.Find this resource:

Pylypchuk, Y., & Miller, E. (2014). Marital status, spousal characteristics, and the use of preventive care. Journal of Family and Economic Issues, 35(3), 323–338.Find this resource:

Reichman, N., Corman, H., & Noonan, K. (2004). Effects of child health on parents’ relationship status. Demography, 41(3), 569–584.Find this resource:

Reichman, N., Corman, H., Noonan, K., & Dave, D. (2009). Infant health production functions: What a difference the data make. Health Economics, 18(7), 761–782.Find this resource:

Reichman, N. E., Corman, H., Noonan, K., & Schwartz-Soicher, O. (2010). Effects of prenatal care on maternal postpartum behaviors. Review of Economics of the Household, 8(2), 171–197.Find this resource:

Sabia, J., Wooden, M., & Nguyen, T. T. (2018). Sexual identity, same-sex relationships, and health dynamics: New evidence from longitudinal data in Australia. Economics and Human Biology.Find this resource:

Schmitz, H., & Westphal, M. (2017). Informal care and long-term labor market outcomes. Journal of Health Economics, 56, 1–18.Find this resource:

See, S. G. (2016). Parental supervision and adolescent risky behaviors. Review of Economics of the Household, 14(1), 185–296.Find this resource:

Simon, D. (2016). Does early life exposure to cigarette smoke permanently harm childhood welfare? Evidence from cigarette tax hikes. American Economic Journal: Applied Economics, 8(4), 128–159.Find this resource:

Skira, M. (2015). Dynamic wage and employment effects of elder parent care. International Economic Review, 56(1), 63–93.Find this resource:

Slade, A. N., Beller, A. H., & Powers, E. T. (2017). Family structure and young adult health outcomes. Review of Economics of the Household, 15(1), 175–197.Find this resource:

Sonchak, L. (2015). Medicaid reimbursement, prenatal care and infant health. Journal of Health Economics, 44, 10–24.Find this resource:

Spicer, A., Stavrunova, O., & Thorp, S. (2016). How portfolios evolve after retirement: Evidence from Australia. Economic Record, 92, 241–267.Find this resource:

van den Berg, B., & Spauwen, P. (2006). Measurement of informal care: An empirical study into valid measurement of time spent on informal caregiving. Health Economics, 15, 447–460.Find this resource:

Van Houtven, C. H., Coe, N. B., & Konetzka, R. T. (2015). Family structure and long-term care insurance purchase. Health Economics, 24(1), 58–73.Find this resource:

Vermeulen, F. M. P. (2002). Collective household models: Principles and main results. Journal of Economic Surveys, 16(4), 533–564.Find this resource:

Villanueva, E. (2005). Inter vivos transfers and bequests in three OECD countries. Economic Policy, 20(43), 506–565.Find this resource:

Warren, D. A. (2015). Retirement decisions of couples in Australia: The impact of spousal characteristics and preferences. The Journal of the Economics of Ageing, 6, 149–162.Find this resource:

Wax, A. (2017). Family and household economics. In F. Parisi (Ed.), Oxford handbook of law and economics, Vol. 2: Private and commercial law. New York: Oxford University Press.Find this resource:

Wehby, G. L. (2014). Breastfeeding and child disability: A comparison of siblings from the United States. Economics and Human Biology, 15, 13–22.Find this resource:

Wilson, C., & Oswald, A. (2005). How does marriage affect physical and psychological health? A survey of the longitudinal evidence. (IZA Discussion Paper No. 1619.)Find this resource:

Wittenberg, E., & Prosser, L. A. (2013). Disutility of illness for caregivers and families: A Systematic review of the literature. PharmacoEconomics, 31, 489–500.Find this resource:

Woosley, A., Danes, S. M., & Stum, M. (2017). Utilizing a family decision-making lens to examine adults’ end-of-life planning actions. Journal of Family and Economic Issues, 38, 33–44.Find this resource:

Yaari, M. E. (1965). Uncertain lifetime, life insurance and the theory of the consumer. Review of Economic Studies, 32(2), 137–150.Find this resource:

Yan, J. (2017). The effects of prenatal care utilization on maternal health and health behaviors. Health Economics, 26(8), 1001–1018.Find this resource:

Yan, J. (2013). Prenatal smoking cessation and infant health: Evidence from sibling birth. Southern Economic Journal, 80(2), 299–323.Find this resource:

Yan, J. (2014). The effects of a minimum cigarette purchase age of 21 on prenatal smoking and infant health. Eastern Economic Journal, 40(3), 289–308.Find this resource:

Yan, J. (2015). Maternal pre-pregnancy BMI, gestational weight gain, and infant birth weight: A within-family analysis in the United States. Economics and Human Biology, 18(2), 1–12.Find this resource:

Notes:

(1.) In this chapter we use the terms family and household interchangeably.

(2.) Black, Devereux, and Salvanes (2016) explores the potential effects of birth order on health in adulthood and finds that firstborns are more likely to be overweight, to be obese, and to have high blood pressure and high triglycerides.

(3.) This literature is predominantly from the United States, where the focus is on retirement prior to age 65 Medicare eligibility.

(4.) Most work has been done on U.S. data. However, several studies document similar household financial profiles in different countries (Alessiea, Lusardib, & Kapteync, 1999; Blundell, Crawford, French, & Tetlow, 2016; Spicer, Stavrunova, & Thorp, 2016; Villanueva, 2005).

(5.) Some of the strategic intra-family transaction literature models payments from adult children to parents to compensate for informal child care. Our focus in this section is on long-term care rather than child care.