Show Summary Details

Page of

date: 15 December 2019

# The Rationale for Interventions to Foster Child Development

## Summary and Keywords

Socioeconomic gradients in health, cognitive, and socioemotional skills start at a very early age. Well-designed policy interventions in the early years can have a great impact in closing these gaps. Advancing this line of research requires a thorough understanding of how households make human capital investment decisions on behalf of their children, what their information set is, and how the market, the environment, and government policies affect them. A framework for this research should describe how children’s skills evolve and how parents make choices about the inputs that model child development, as well as the rationale for government interventions, including both efficiency and equity considerations.

# Scope of the Review

The last several decades have seen much exciting research on the process of human capital formation and, in particular, the important role that in-utero and early childhood experiences play (e.g., Almond & Currie, 2011; Carneiro & Heckman, 2003; Phillips & Shonkoff, 2000). We now know that socioeconomic gradients in health, cognitive, and socioemotional are three different skills start at a very early age, and that well-designed interventions can have a great impact on those skills.

This article provides a framework to describe how children’s skills evolve and how parents make choices about the inputs that model child development. It also includes a discussion on why governments intervene to foster early childhood development (see Berlinski & Vera-Hernández, 2019, for details on the economics behind early childhood interventions). We precede our discussion with a cursory description about the extent and geographical distribution of child development both across and within countries.

# Distribution of Child Development

Differences in child development according to socioeconomic status, both across and within countries, are well documented in the literature.1 Across countries, it is customary to compare stunting rates because anthropometric data is routinely collected and the availability of the WHO growth standards facilitates such comparisons (WHO Multicentre Growth Reference Study Group, 2006; Black et al., 2017). Table 1 reports stunting rates across geographic regions and countries with different income levels. Although stunting rates have been decreasing over time, in 2017 they were still over 30% in South Asia and sub-Saharan Africa. Although the stunting rate in low-income countries is higher than in lower-middle-income countries (35.2% vs. 31.5%), there is a much larger gap between lower-middle- and upper-middle-income countries (31.5% vs. 6.4%).

Comparisons in cognitive and socioemotional development across countries are much harder to make because there is no consensus on a standard, and data on children’s cognitive and socioemotional development is not collected as routinely as anthropometric data. McCoy et al. (2016) compared the Early Childhood Development Index (ECDI) across 35 low- and middle-income countries,2 and found that 14.6% of children aged 3 and 4 have low ECDI scores in the cognitive domain, 26.2% in the socioemotional domain, and 36.8% in either or both of the domains. Extrapolating from the sample, they found that sub-Saharan Africa (44%) and South Asia (37.8%) are the regions with a larger prevalence of children with low ECDI scores, followed by East Asia and the Pacific region (26%; see also Grantham-McGregor et al., 2007, for comparisons for selected LMICs using different instruments).

Table 1. Estimated Percentage of Children Aged 0 to 59 Months Who are Stunted

1995

2005

2015

2017

Global

35.6

29.3

23.2

22.2

Geographic Area

East Asia and Pacific

31.6

18.6

10.2

9.0

Eastern Europe and Central Asia

23.0

15.0

9.4

8.5

Latin America and Caribbean

19.7

14.4

10.3

9.6

Middle East and North Africa

25.7

20.3

15.8

15.0

United States of America

3.3

2.8

2.4

2.3

South Asia

56.3

46.4

36.8

35.0

Sub-Saharan Africa

46.2

40.4

34.9

33.9

World Bank Income Classification

Low income

50.6

43.4

36.5

35.2

Lower-middle income

49.9

41.2

33.0

31.5

Upper-middle income

24.4

13.8

7.3

6.4

High income

3.7

3.1

2.6

2.5

Source: UNICEF, WHO, World Bank Joint child malnutrition dataset, May 2018.

As expected, marked differences in child development are also found within countries. Reynolds et al. (2017) report the socioeconomic gradient in height and receptive language (Peabody Picture Vocabulary Test) for the four countries included in the Young Lives Project: Peru, Vietnam, Ethiopia, and India (for the latter, only the states of Andhara Pradesh and Telangana). The difference in height (measured in percentiles) between the bottom and top wealth quartile ranges from 19 in Vietnam to 22.6 in Peru.3 With respect to vocabulary, differences between the top and bottom quartile at age 5 vary between 21 percentiles for India and 40 for Peru. Hence, it would appear that the socioeconomic gradient in vocabulary is more pronounced than in height (for single-country examples of socioeconomic gradients on child development, see Behrman, 2008; Berlinski & Schady, 2015; Paxson & Schady, 2007; Rubio-Codina, Attanasio, Meghir, Varela, & Grantham-McGregor (2015); for results on five Latin American countries, see Schady et al., 2015).

There is a vast literature reporting a marked socioeconomic gradient in early childhood development within high-income countries. For instance, Carneiro and Heckman (2003) report an average difference of 14 percentiles in the Maths Peabody Individual Achievement Test between the top and bottom income quartile in the children of the National Longitudinal Survey of Youth (1979) in the United States. Using data from the U.K. Millenium Cohort Study, Dearden, Sibieta, and Sylva (2011) document a difference of 22 percentiles in naming vocabulary at age 3 between the top and bottom quintile of a socioeconomic index. Both studies report that the gap widens with age, as Case, Lubotsky and Paxon (2002) and Currie and Stabile (2003) report for the socioeconomic gradient in child health in the United States.

# The Formation of Human Capital

Child development is a dynamic and cumulative process in which children develop a set of skills (e.g., cognitive, motor, socioemotional) over time. Further, the development process is nonlinear (e.g., children experience rapid growth in height and head circumference in the first 6 months of life, faltering thereafter) and far from deterministic (e.g., some children walk as early as 9 months whereas others with healthy development may not walk until 17 months). To study how different choices made by parents and governments affect the formation of human capital, economists have formalized this process using a production function approach (Becker, 1964; Becker & Tomes, 1994; Cunha & Heckman, 2007, 2008; Cunha, Heckman, Lochner, & Masterov, 2006; Heckman, 2007.).

We denote the level of a child’s human capital at age t by the vector $θt$, $t∈{1,2,…,T}$.4 We consider three dimensions to it: $θt=(θC,t,θS,t,θH,t)$. The cognitive dimension of human capital is denoted by $θC,t$, the socioemotional dimension is represented by $θS,t$, and the health dimension, $θH,t$. Each dimension of human capital evolves according to its dimension-specific production function, $fk,t$, such that:

$Display mathematics$

where $fk,t$ is strictly increasing in all its arguments, twice differentially continuous, and concave in $It$ that are inputs that affect directly the level of human capital at age t.5

The set of important inputs is vast and ranges from time spent doing cognitive and socioemotionally enriching activities, nutritional intake, exposure to toxins, microbials and pollutants, and availability of toys, books and other material resources. The productivity of certain inputs can depend on the levels of other inputs. For instance, nutrients in food might not be absorbed if exposure to microbials is high (Campbell, Elia, Lunn, 2003; Lin et al., 2013; Lunn, Northrop-Clewes, & Downes, 1991) and the time that parents spend with a child might be more productive if books or other resources are available.

The level of an input can be determined by external shocks that are out of parents’ control (e.g., floods, droughts, epidemics, mass lay-offs) or can be a consequence of parents’ and caregivers’ actions (e.g., time devoted to child’s sleep, food nutritional content, caregiver quality). The precise classification will depend on parents’ information set: an increase to air pollution exposure may be the result of moving to a more polluted area or due to an unexpected increase in manufacturing activity that increases the pollution in the area.6 The level of human capital at birth (determined by genetic factors, inputs received in utero, and the family environment) is denoted by $θ0$.

Previous inputs affect the realization of human capital at t only through their effect on $θt−1$. The model allows for the three dimensions of human capital to play a role in determining the future level of each dimension of human capital (i.e., the entire vector $θt−1=(θC,t−1,θS,t−1,θH,t−1)$ enters in the production of $θk,t$). For instance, social competence, behaviour management, social perception and self-regulatory abilities (i.e., socioemotional skills) can all play an important role in classroom learning and, therefore, on the accumulation of cognitive skills later on. The latter is an example of complementarity between an input and the level of skills. Formally, direct (dynamic) complementarity is defined as:

$Display mathematics$

(Cunha & Heckman, 2007; Cunha et al., 2006; Heckman, 2007; Heckman, Stixrud, & Urzua, 2006). The existence of dynamic complementarities implies that skills produced at a certain age increase the productivity of investments at subsequent stages. Hence, they make a compelling case for early intervention, but also for follow-up investment to make the most of early investment. Their presence reinforces why the socioeconomic gradient on child development starts from an early age: If wealthy parents know that they will be investing more in their children later on, they have incentives to invest earlier. In terms of public policy, dynamic complementarities combined with self-productivity could explain why returns on investments in later childhood and adolescence have a lower rate of return than investments in early childhood (Heckman, 2007).

Recent work has estimated the production function considering the endogeneity of inputs as well as measurement error in outcome variables in the United States (Agostinelli & Wiswall, 2016; Cunha, Heckman, & Schennach, 2010), and in Colombia, India, Ethiopia, and Peru (Attanasio, Cattan, Fitzsimons, Meghir, & Rubio Codina, 2017; Attanasio, Meghir, & Nix, 2017a; Attanasio, Meghir, Nix, & Salvati, 2017b). It is difficult to compare across studies because of the differences in endowments, environments, age ranges, and skills that they consider (cognitive and socioemotional for the United States and Colombia or cognitive and health for Ethiopia, India, and Peru). However, there are some emerging common results that are already very important. One of them is self-productivity, that is, the level of a particular skill increases over the level in the previous period, leading to persistence in the process of human capital accumulation.

Another robust result is that investments are very important for skill accumulation. A key issue is how the effects of these investments vary with age. This would give rise to the existence of “key periods” or “sensitive periods,” in which particular investments are most productive, probably varying by type of skill (Cunha et al., 2006; Knudsen, Heckman, Cameron, & Shonkoff, 2006; Thompson & Nelson, 2001). It seems to be a robust finding that the positive effect of investments on cognitive development and health decrease with age, which highlights the importance of early interventions. Less is known about noncognitive skills, but findings from Cunha et al. (2010) suggest that investments are also very important for noncognitive skills in later stages of childhood.

There is also evidence of cross-productivity, in which the current stock of a particular skill increases the future level of another skill. For instance, health in early years affects future cognitive development in Ethiopia, India, and Peru (Attanasio et al., 2017a; Attanasio, Cattan, et al., 2017). However, there are also results that depend on the setting and age range. Cunha et al. (2010) have found that early socioemotional development enhanced subsequent cognition in a sample of US children and Attanasio, Cattan, et al. (2017) found in a of Colombian children that cognitive development at ages 12 to 24 months enhances socioemotional development at ages 30 to 42 months.

An array of reduced-form evidence from experimental studies is consistent with these results. For instance, intensive parenting and childcare interventions that took place early in childhood have had long-lasting effects on cognitive development (Carneiro & Ginja, 2014; Garces, Thomas, & Currie, 2002; Gertler et al., 2014; Heckman, Moon, Pinto, Savelyev, & Yavitz, 2010b; Walker, Chang, Vera-Hernández, & Grantham-McGregor, 2011), which is consistent with self-productivity in cognitive development, as well as with the malleability of cognitive development at early ages. However, Krishnan and Krutikova (2013) and Akee, Copeland, Costello, and Simeonova (2018) provide evidence of malleability of socioemotional skills later in childhood.

Consistent with cross-productivity, Maluccio et al. (2009) have found that a nutritional intervention in Guatemala that benefited children aged 0 to 36 months also led to improvements in cognitive development and schooling outcomes later in life. Also consistent with cross-productivity, as well as with the malleability of cognitive skills early in life, Barham, Macours, and Maluccio (2013) found that boys exposed to a conditional cash transfer program in utero and up to age 2 had improved nutritional status and cognitive ability up to 7 years after the intervention. However, those who were exposed to the program between 2 and 5 years of age did not experience an improvement in cognitive skills, although they caught up with the younger cohort in terms of nutrition.

A key issue in this literature is whether or not there are dynamic complementarities; that is, whether the return (in terms of future skill levels) to an investment increases with the current level of skills. Cunha et al. (2010), Attanasio et al. (2017a, 2017b), and Attanasio, Cattan et al. (2017) all obtained estimates that implied dynamic complementarities. In addition, several studies have attempted to approach the concept of dynamic complementarities from a reduced form perspective, identifying contexts in which the same child is potentially affected by two shocks or policies at different points in the life cycle.7 For example, a rainfall shocks followed by a conditional cash transfer program (Adhvaryu et al., 2018 and Aguilar & Vicarelli, 2015), a visiting nurse program followed by a childcare program (Rossin-Slater & Wüst, 2016), tornado and vitamin supplementation programs (Gunnsteinsson et al., 2016), and facilitating access to abortion services and better schools (Malamud, Pop-Eleches, & Urquiola, 2016). The results using these reduced form approaches are mixed.8

How do parents make choices about the level of inputs to provide? The traditional framework of Becker (1974) and Becker and Tomes (1979) poses that parents are altruistic toward their children (i.e., parents include the utility of their children in their utility function) and parents’ choices to maximize household utility are subject to the production function, the budget constraint (i.e., labor and nonlabor income have be to at least equal the cost of the consumption bundle), and time allocation constraint (i.e., leisure, work, and child-rearing time have to be less or equal than total time available in the day). In other words, they maximize a utility function that includes their own consumption and the welfare of their children subject to a set of constraints. In the simplest model, parents have only one child and maximize their own consumption and the human capital or income of their children. If parents know the production technology and plan to leave bequests to their children, if there are no market imperfections they will make optimal decisions in terms of the welfare function they maximize; that is, to invest in children’s human capital if the rate of return to human capital is larger than the market return for their saving (Becker & Murphy, 1988).

However, the question remains of whether parents are making optimal decisions on behalf of their children. In other words, are they acting as if they are solving the lifetime optimization problem of the child? This might not be the case if parents are not altruistic enough or are unable to borrow against the return of their children’s human capital. In such cases, parents will underinvest with respect to the child’s optimal investment decision.

Further difficulties occur when parents have more than one child and children differ in their innate ability. If parents are expecting to leave a large enough bequest to each of them no further complications arise from multiple children. However, if there are not enough resources to fund the optimal investment plan for each child it is interesting to see how the parents allocate the available resources. For instance, if they are equally altruistic toward their children they should equalize the marginal utility of their future consumption. In such cases, the parents may tend to underinvest in the most able child.

The characterization we have offered so far has a unitary utility model in mind. Outside the unitary models, the relative bargaining power between mothers and fathers is important for the pattern of investment and consumption (Bourguignon, Browning, Chiappori, & Lechene, 1993; Browning, Bourguignon, Chiappori, & Lechene, 1994; Chiappori, 1988, 1992). When determining whether a cash transfer will have a larger effect if received by fathers or mothers, Blundell, Chiappori, and Meghir (2005) emphasize that it is the relative (rather than absolute) differences in the marginal willingness to pay that is important to determine the recipient who will produce the stronger effect.

# Why Intervene in Early Childhood?

Government interventions are usually motivated by equity or efficiency considerations; that is, to “correct” behavior that is not socially optimal due to the existence of market failures.9

## Equity Considerations

Widespread evidence shows marked socioeconomic gradients in child development, and that these gradients start early in life. Initial endowments and parental background are key determinants in the production of human capital. A government motivated to compensate for unequal endowments can focus its interventions in boosting human capital investments for less affluent families during childhood or providing compensation for unequal outcomes later in life through the tax and benefit system. It is not only ethically fair for all children to have equal opportunities right from the cradle, but the public money spent during early childhood may also be more cost-effective. This is the case if there are critical periods for skill formation (i.e., the marginal productivity of inputs are higher at some ages) and if there are dynamic complementarities in the formation of human capital (Heckman, 2008). Further, compensating for outputs later in life through the benefit system might induce moral hazard problems (Currie, 2001).

The rationale for increasing investing in children is less clear when considered from the perspective of equity among generations. Due to economic growth, children’s lifetime income and resources are likely to exceed their parents’, which may suggest if equity is of concern it transfers from children to parents rather than vice versa (Becker & Murphy, 1988). However, because of the high rate of return on investments in childhood, it would be optimal to invest in children and for parents to receive the transfers from their children later on.

A different motivation for childcare interventions and, in particular, subsidizing childcare is one of equity between genders; that is, to ensure that both mother and father can work and that the burden of childcare does not have to fall on the mother.

## Efficiency Considerations

The presence of externalities, informational asymmetries, and liquidity constraints might also justify government interventions to enhance the production of human capital in children. Parents may want to make first best choices on behalf of their children but might be unable to do so because of liquidity constraints (i.e., they are unable to borrow to fund human capital investments in their children). They may also lack the information or knowledge to make the most appropriate choices. Indeed, information failures (moral hazard, adverse selection) may prevent parents from purchasing child-care services of the quality that they wish.10

An important aspect that has gained some attention in the literature is that parents choose inputs based on the perceived production function of human capital, which may be different from the true one. For instance, Cunha, Elo and Culhane (2013) find that disadvantaged U.S. mothers have poor information on the returns to early childhood and Fitzsimons, Malde, Mesnard, and Vera-Hernández (2016) document poor knowledge of child nutrition among Malawian mothers. Related to this, Boneva and Rauh (2018) elicit parental beliefs on the returns to different investments. They find that perceived returns vary substantially across the population, and that on average, parents believe that the returns to late investments are higher than to early investments, which is at odds with the research just summarized.11

There also may be market failure in the provision of good and services for children. For example, childcare services are an experience good, or a good whose quality consumers are likely to ascertain only after consuming it.12 In private markets, prices may provide some signals about quality. Research has shown that where direct information can be obtained (even at a cost), informed consumers can improve the quality of the products offered (Tirole, 1988). Moreover, as the fraction of well-informed parents increases, the likelihood that valuable information will be revealed increases as well. As a result, the public benefit of an informed parent is greater than the private cost the parent is willing to pay. This positive externality provides a rationale for public policy intervention in the childcare market.

Certain benefits from early childhood interventions are not internalized by parents, and therefore parents may invest less compared to a social planner. A classic example is that of contagious diseases. Worms, which are prevalent among school-aged children in developing countries, are transmitted via contact with contaminated soil or feces. In developing countries, interventions such as large-scale deworming with drugs not only reduces the rate of infection in treated beneficiaries but also in those who do not receive treatment (see, e.g., Miguel & Kremer, 2004). There are also benefits for the society that are accrued in the future that parents may not internalize. For example, early interventions such as cognitive stimulation through parenting programs (e.g., Grantham-McGregor, Powell, Walker, & Himes, 1991) and early childhood education (e.g., Havnes & Mogstad, 2011b; Heckman et al., 2010b) have shown reductions in crime and violent behavior (e.g., Heckman et al., 2010a; Walker et al., 2011) and welfare dependency (e.g., Havnes & Mogstad, 2011b).

Not all the interventions that can affect children’s human capital are designed with the sole objective of improving children’s outcomes. There are reasons to intervene, especially in the childcare market, that are linked to parents’ human capital. Childcare interventions may be driven by a motivation to encourage both parents to work, particularly low-income mothers. A childcare subsidy is a less expensive way to increase labor supply than a standard wage subsidy because the latter accrues to everyone working whereas the former only benefits those with children in childcare (Blau, 2003).

Decreasing the cost or improving the quality of childcare should increase the female labor supply, which should induce women to stay longer in school and reduce the negative externalities associated with teenage pregnancy. If childcare is expensive, liquidity constraints may prevent women to gain experience in the labor market and therefore earn higher wages in the future (Adda, Dustmann, & Stevens, 2017; Bertrand, Goldin, & Katz, 2010; Blau & Kahn (2017); Dias, Joyce, & Parodi, 2018; Goldin, 2014).

Another rationale for intervening in the childcare market and, in particular, subsidizing childcare is to improve the efficiency of the tax system. A high-skilled worker who is trying to mimic a low-skilled one will benefit less from a childcare subsidy because she will need to work fewer hours than the low-skilled worker to get the same total pretax earnings. Hence, the childcare subsidy would make it easier to separate high- and low-skilled workers, improving the efficiency of the tax system (Blomquist, Christiansen, & Micheletto, 2010.) This is part of a more general argument from the optimal tax literature to subsidize goods that are complementary to the labor supply (Corlett & Hague, 1953; Bastani, Blomquist, & Micheletto, 2017).

The argument on subsidising goods complementary to labor supply can be applied to the child’s labor supply as an adult. To the extent that interventions in early childhood can improve adult productivity they can also affect the child’s future adult labor supply, depending on the income and substitution effects resulting from higher wages, which will also enhance efficiency (Currie & Gahvari, 2008).

## Interventions in Early Childhood

The type of intervention and its potential impact on early childhood outcomes depends on the underlying factors that are causing poor developmental outcomes in children. For example, if underinvestment in children’s human capital results from lack of information about returns on investments, this might be overcome with the provision of information to parents. Information alone may not be enough and this intervention may need to be supplemented with a combination of encouragement and training (e.g., home visiting programs, midwives who promote breastfeeding).

On the contrary, when liquidity constraints are the source of underinvestment in children’s human capital, policymakers may want to implement cash transfers to generate a pure income effect, which increases consumption as well as inputs that foster child development. Of course, cash transfers will also increase the consumption of other goods and services that are not related to benefits for children.

However, the problem of liquidity constraints is often coupled with parents’ lack of information about the returns on human capital investments at an early age, as well as on what the optimal investments are. Hence, even if liquidity constraints were removed, the allocation that parents will make may be far from optimal. Governments could opt instead to provide cash transfers and information simultaneously. Alternatively, governments could influence parents’ choices through the price system with targeted price subsidies (i.e., childcare subsidies), in-kind transfers (e.g., free school lunches, free milk), or conditional cash transfers that provide families with cash conditional on the consumption of certain good or uptake of certain services.13

The interventions noted above are geared toward affecting the behavior of the demand side of the market. However, governments often introduce policies geared toward improving access and quality of services for young children through the regulation of suppliers. Because of moral hazard and adverse selection, parents may have difficulty ascertaining the quality of childcare services. Regulations such as imposing minimum quality standards and auditing and rating providers are widespread. Another example is the 1981 International Code of Marketing of Breastmilk Substitutes, which promotes breastfeeding through the provision of adequate information on appropriate infant feeding and the regulation of the marketing of breastmilk substitutes. The code prohibits the advertisement or promotion of these products to the public or through the healthcare system. Many countries have adopted this code, which has likely saved many lives in lower- and middle-income countries (Anttila-Hughes, Fernald, Gertler, Krause, & Wydick, 2018).14

Governments can also make investments in infrastructure that indirectly affect the access to goods and services for children but are beyond the realm of individual decision-making. These are more likely to be important in developing countries. For example, the government may construct a road that reduces the costs for attending prenatal services and ensure that maternity healthcare services are of good quality. It may also create (or sponsor through the suitable regulation of providers) water and sanitation infrastructure that are likely to improve the disease environment in which children live and grow (e.g., Galiani, Gertler, & Schargrodsky, 2005).

The companion article (Berlinski & Vera-Hernández, 2019), focuses on concrete example of policies, their rationale and the available empirical evidence on interventions that governments use to promote children’s development.

# Conclusions and Further Research

A child’s early years are a key period in the development of health, cognitive, and social emotional skills in children. Estimates of the production function of human capital help to understand the process of human capital accumulation. In particular, features such as self-productivity, cross-productivity, and dynamic complementarities are essential to understand how skill gaps evolve with age and to determine the optimal timing for investments. Two results are consistent in the literature: (1) the importance of self-productivity and (2) that the effect of investments in cognitive development and health decrease with age. However, there are other issues that are less clear or for which there is very little evidence: (1) how the effect of investments on socioemotional development vary with age and (2) a characterization of the cross-productivity of skills. In regard to the latter, although most studies tend to find evidence on cross-productivity, there is still no consensus on the direction of the cross-productivity effects (i.e., from health to cognition or vice versa) or the age at which they are most important. We expect more work in this area, in particular using richer data (simultaneously including the three skill sets at more ages) combined with methodological advances on estimation.

Dynamic complementarities are also essential for understanding how the developmental gap evolves with age, as well as the optimal timing of investments. Although the structural literature consistently finds dynamic complementarities to be important, the results on the reduced-form literature are more mixed. Although data availability will be a challenge, combining the structural and reduced-form approaches (effectively by using the shocks or natural experiments as sources of variation to estimate the structural model or to validate it) will be a fruitful area of research.

Parents face trade-offs when investing in their children’s development. They make decisions on the basis of their own perceptions of what the production function of human capital looks like, but we now know that some of these perceptions are wrong. However, if as researchers we want to understand how households make choices, we will need a better grasp of parental beliefs about the production function (Attanasio 2015). Although there is some incipient research in this area, we expect it to become more important in the future, which will be combined with estimates of human capital production functions. In parallel, it will be most important to understand how parents form such beliefs so that policies to change them can be designed optimally.

From our own interaction with policymakers, policy decisions on early childhood education tend to be taken more on equity than efficiency grounds. Although market failures (e.g., asymmetric information, externalities, etc.) are usually at the forefront of economists’ rationales for interventions in early childhood, there is little serious research that quantifies such market failures and the welfare loss associated with them. This makes it difficult to guide policy decisions when a menu of policies is offered. To move past cost-effectiveness of policies, we must be able to quantify the losses from market failures and the welfare gains related to fixing them. More research that tackles this issue will be highly valuable.

# Acknowledgments

We are thankful to Raquel Bernal, Karen Macours, and Francesca Salvati for their comments on selected parts of this review. Any errors are the sole responsibility of the authors. The views expressed herein are those of the authors and should not be attributed to the Inter-American Development Bank, its Executive Directors, or the governments they represent.

Agostinelli, F., & Wiswall, M. (2016). Estimating the technology of children’s skill formation (NBER Working Paper w22442). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Almond, D., Currie, J., & Duque, V. (2018). Childhood circumstances and adult outcomes: Act II. Journal of Economic Literature, 56(4), 1360–1446.Find this resource:

Attanasio, O. P. (2015). The determinants of human capital formation during the early years of life: Theory, measurement, and policies. Journal of the European Economic Association, 13(6), 949–997.Find this resource:

Attanasio, O., Cattan, S., Fitzsimons, E., Meghir, C., & Rubio Codina, M. (2017). Estimating the production function for human capital: Results from a randomized controlled trial in Colombia (IFS Working Paper W17/06). London, U.K.: Institute for Fiscal Studies.Find this resource:

Attanasio, O., Meghir, C., & Nix, E. (2017). Human capital development and parental investment in India. (NBER Working Paper 21740). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Attanasio, O., Meghir, C., Nix, E., & Salvati, F. (2017). Human capital growth and poverty: Evidence from Ethiopia and Peru. Review of Economic Dynamics, 25, 234–259.Find this resource:

Becker, G. S. (1964). Human capital theory. New York: Columbia University Press.Find this resource:

Becker, G. S., & Tomes, N. (1979). An equilibrium theory of the distribution of income and intergenerational mobility. Journal of Political Economy, 87(6), 1153–1189.Find this resource:

Behrman, J. R. (2008). Early life nutrition and subsequent education, health, wage, and intergenerational effects (Working Paper 57732). Washington, DC: The World Bank.Find this resource:

Blau, D., & Currie, J. (2006). Pre-school, day care, and after-school care: Who’s minding the kids? In E. A. Hanushek & F. Welch (Eds.), Handbook of the Economics of Education, Vol. 2 (pp. 1163–1278). Amsterdam, The Netherlands: Elsevier.Find this resource:

Blomquist, S., Christiansen, V., & Micheletto, L. (2010). Public provision of private goods and nondistortionary marginal tax rates. American Economic Journal: Economic Policy, 2(2), 1–27.Find this resource:

Boneva, T., & Rauh, C. (2018). Parental beliefs about returns to educational investments: The later the better? Journal of the European Economic Association, 16(6), 1669–1711.Find this resource:

Carneiro, P., & Heckman J. (2003). Human capital policy. In J. Heckman & A. Krueger, Inequality in America: What role for human capital policies? (pp. 77–240). Cambridge, MA: MIT Press.Find this resource:

Case, A., Lubotsky, D., & Paxson, C. (2002). Economic status and health in childhood: The origins of the gradient. American Economic Review, 92(5), 1308–1334.Find this resource:

Cunha, F. (2011). Recent developments in the identification and estimation of production functions of skills. Fiscal Studies, 32(2), 297–316.Find this resource:

Cunha, F., & Heckman, J. (2007). The technology of skill formation. American Economic Review, 97(2), 31–47.Find this resource:

Cunha, F. & Heckman J. J. (2008) Formulating, identifying, and estimating the technology for the formation of skills. Journal of Human Resources, 43(4), 738–782.Find this resource:

Cunha F., Heckman J. J., Lochner, L., & Masterov, D.V. (2006). Interpreting the evidence on life cycle skill formation. In E. Hanushek & F. Welch (Eds.), The Handbook of Economics of Education (pp. 697–812). Amsterdam, The Netherlands: North Holland.Find this resource:

Cunha, F., Heckman, J. J., & Schennach, S. M. (2010). Estimating the technology of cognitive and noncognitive skill formation. Econometrica, 78(3), 883–931.Find this resource:

Currie, J. (2009). Healthy, wealthy, and wise: Socioeconomic status, poor health in childhood, and human capital development. Journal of Economic Literature, 47(1), 87–122.Find this resource:

Currie, J., & Almond, D. (2011). Human capital development before age five. In D. Card & O. Ashenfelter (Eds.), Handbook of labor economics, Vol. 4 (pp. 1315–1486). Amsterdam, The Netherlands: Elsevier.Find this resource:

Currie, J., & Stabile, M. (2003). Socioeconomic status and child health: Why is the relationship stronger for older children? American Economic Review, 93(5), 1813–1823.Find this resource:

Gertler, P., Heckman, J., Pinto, R., Zanolini, A., Vermeersch, C., Walker, S., . . . Grantham-McGregor, S. (2014). Labor market returns to an early childhood stimulation intervention in Jamaica. Science, 344(6187), 998–1001.Find this resource:

Heckman, J. J. (2007). The economics, technology, and neuroscience of human capability formation. Proceedings of the National Academy of Sciences, 104(33), 13250–13255.Find this resource:

Heckman, J. J., & Mosso, S. (2014). The economics of human development and social mobility. Annual Review of Economics, 6(1), 689–733.Find this resource:

Phillips, D. A., & Shonkoff, J. P. (Eds.). (2000). From neurons to neighborhoods: The science of early childhood development. Washington, DC: National Academies Press.Find this resource:

Thompson, R. A., & Nelson, C. A. (2001). Developmental science and the media: Early brain development. American Psychologist, 56(1), 5–15.Find this resource:

## References

Adda, J., Dustmann, C., & Stevens, K. (2017). The career costs of children. Journal of Political Economy, 125(2), 293–337.Find this resource:

Adhvaryu, A., Nyshadham, A., Molina, T., & Tamayo, J. (2018). Helping children catch up: Early life shocks and the PROGRESA experiment (Working Paper No. 24848). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Agostinelli, F., & Wiswall, M. (2016). Estimating the technology of children’s skill formation (Working Paper w22442). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Aguilar, A. & Vicarelli, M. (2015). El Nino and Mexican children: Medium-term effects of early-life weather shocks on cognitive and health outcomes. Working paper.Find this resource:

Aizer, A., & Cunha, F. (2012). The production of human capital: Endowments, investments and fertility (NBER Working Paper No. 18429). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Akee, R., Copeland, W., Costello, E. J., & Simeonova, E. (2018). How does household income affect child personality traits and behaviors? American Economic Review, 108(3), 775–827.Find this resource:

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

Anttila-Hughes, J. K., Fernald, L. C., Gertler, P. J., Krause, P., & Wydick, B. (2018). Mortality from Nestlé’s marketing of infant formula in low and middle-income countries (NBER Working Paper No. w24452). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Attanasio, O. P. (2015). The determinants of human capital formation during the early years of life: Theory, measurement, and policies. Journal of the European Economic Association, 13(6), 949–997.Find this resource:

Attanasio, O., Cattan, S., Fitzsimons, E., Meghir, C., & Rubio Codina, M. (2017). Estimating the production function for human capital: Results from a randomized controlled trial in Colombia (IFS Working Paper W17/06). London, U.K.: Institute for Fiscal Studies.Find this resource:

Attanasio, O., Meghir, C., & Nix, E. (2017a). Human capital development and parental investment in India (NBER Working Paper 21740). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Attanasio, O., Meghir, C., Nix, E., & Salvati, F. (2017b). Human capital growth and poverty: Evidence from Ethiopia and Peru. Review of Economic Dynamics, 25, 234–259.Find this resource:

Barham, T., Macours, K., & Maluccio, J. A. (2013). Boys’ cognitive skill formation and physical growth: Long-term experimental evidence on critical ages for early childhood interventions. American Economic Review, 103(3), 467–471.Find this resource:

Bastani, S., Blomquist, S., & Micheletto, L. (2017). Child care subsidies, quality, and optimal income taxation (CESifo Working Paper No. 6533).Find this resource:

Becker, G. S. (1974). A theory of social interactions. Journal of Political Economy, 82(6), 1063–1093.Find this resource:

Becker, G. S., & Murphy, K. M. (1988). The family and the state. Journal of Law and Economics, 31(1), 1–18.Find this resource:

Becker, G. S., & Tomes, N. (1979). An equilibrium theory of the distribution of income and intergenerational mobility. Journal of Political Economy, 87(6), 1153–1189.Find this resource:

Becker, G. S., & Tomes, N. (1994). Human capital and the rise and fall of families. In G. S. Becker (Ed.), Human capital: A theoretical and empirical analysis with special reference to education (3d ed.; pp. 257–298). Chicago, IL: University of Chicago Press.Find this resource:

Behrman, J. R. (2008). Early life nutrition and subsequent education, health, wage, and intergenerational effects (Working Paper No. 57732). Washington, DC: The World Bank.Find this resource:

Berlinski, S., & Schady, N. (Eds.). (2015). The early years: Child wellbeing and the role of public policy. New York, NY: Palgrave Macmillan.Find this resource:

Berlinski, S., & Vera-Hernández, M. (2019). The economics of early interventions aimed at child development. Oxford Research Encyclopedia of Economics and Finance. New York, NY: Oxford University Press.Find this resource:

Bernal, R. (2008). The effect of maternal employment and child care on children’s cognitive development. International Economic Review, 49(4), 1173–1209.Find this resource:

Bertrand, M., Goldin, C., & Katz, L. F. (2010). Dynamics of the gender gap for young professionals in the financial and corporate sectors. American Economic Journal: Applied Economics, 2(3), 228–255.Find this resource:

Black, M. M., Walker, S. P., Fernald, L. C. H., Andersen, C. T., DiGirolamo, A. M., Lu, C., . . . Grantham-McGregor, S. (2017). Early childhood development coming of age: Science through the life course. The Lancet, 389(10064), 77–90.Find this resource:

Blau, D. M. (2003). Child care subsidy programs. In R. Moffitt (Ed.), Means tested social programs (pp. 443–516). Chicago, IL: University of Chicago Press.Find this resource:

Blau, D., & Currie, J. (2006). Pre-school, day care, and after-school care: Who’s minding the kids? In E. A. Hanushek & F. Welch (Eds.), Handbook of the Economics of Education, Vol. 2 (pp. 1163–1278). Amsterdam, The Netherlands: Elsevier.Find this resource:

Blau, F. D., & Kahn, L. M. (2017). The gender wage gap: Extent, trends, and explanations. Journal of Economic Literature, 55(3), 789–865.Find this resource:

Blomquist, S., Christiansen, V., & Micheletto, L. (2010). Public provision of private goods and nondistortionary marginal tax rates. American Economic Journal: Economic Policy, 2(2), 1–27.Find this resource:

Blundell, R., Chiappori, P.-A., & Meghir, C. (2005). Collective labor supply with children. Journal of Political Economy, 113(6), 1277–1306.Find this resource:

Boneva, T., & Rauh, C. (2018). Parental beliefs about returns to educational investments: The later the better? Journal of the European Economic Association, 16(6), 1669–1711.Find this resource:

Bourguignon, F., Browning, M., Chiappori, P.-A., & Lechene, V. (1993). Intra household allocation of consumption: A model and some evidence from French data. Annales d’Economie et de Statistique, 29(29), 137–156.Find this resource:

Browning, M., Bourguignon, F., Chiappori, P.-A., & Lechene, V. (1994). Income and outcomes: A structural model of intrahousehold allocation. Journal of Political Economy, 102(6), 1067–1096.Find this resource:

Campbell, D. I., Elia, M., & Lunn, P. G. (2003). Growth faltering in rural Gambian infants is associated with impaired small intestinal barrier function, leading to endotoxemia and systemic inflammation. The Journal of Nutrition, 133(5), 1332–1338.Find this resource:

Carneiro, P., & Ginja, R. (2014). Long-term impacts of compensatory preschool on health and behavior: Evidence from Head Start. American Economic Journal: Economic Policy, 6(4), 135–173.Find this resource:

Carneiro, P., & Heckman J. (2003). Human capital policy. In J. Heckman & A. Krueger, Inequality in America: What role for human capital policies? (pp. 77–240). Cambridge, MA: MIT Press.Find this resource:

Case, A., Lubotsky, D., & Paxson, C. (2002). Economic status and health in childhood: The origins of the gradient. American Economic Review, 92(5), 1308–1334.Find this resource:

Chiappori, P.-A. (1988). Rational household labor supply. Econometrica, 56(1), 63–90.Find this resource:

Chiappori, P.-A. (1992). Collective labor supply and welfare. Journal of Political Economy, 100(3), 437–467.Find this resource:

Corlett, W. J., & Hague, D. C. (1953). Complementarity and the excess burden of taxation. Review of Economic Studies, 21(1), 21–30.Find this resource:

Cunha, F. (2011). Recent developments in the identification and estimation of production functions of skills. Fiscal Studies, 32(2), 297–316.Find this resource:

Cunha, F., Elo, I., & Culhane, J. (2013). Eliciting maternal expectations about the technology of cognitive skill formation (Working Paper No. 19144). Washington, DC: National Bureau of Economic Research.Find this resource:

Cunha, F., & Heckman, J. (2007). The technology of skill formation. American Economic Review, 97(2), 31–47.Find this resource:

Cunha, F., & Heckman, J. J. (2008). Formulating, identifying, and estimating the technology for the formation of skills. Journal of Human Resources, 43(4), 738–782.Find this resource:

Cunha F., Heckman J. J., Lochner, L., & Masterov, D. V. (2006). Interpreting the evidence on life cycle skill formation. In E. Hanushek & F. Welch (Eds.), The Handbook of Economics of Education (pp. 697–812). Amsterdam, The Netherlands: North Holland.Find this resource:

Cunha, F., Heckman, J. J., & Schennach, S. M. (2010). Estimating the technology of cognitive and noncognitive skill formation. Econometrica, 78(3), 883–931.Find this resource:

Currie, J. (2001). Early childhood education programs. Journal of Economic Perspectives, 15(2), 213–238.Find this resource:

Currie, J., & Gahvari, F. (2008). Transfers in cash and in-kind: Theory meets the data. Journal of Economic Literature, 46(2), 333–383.Find this resource:

Currie, J., & Stabile, M. (2003). Socioeconomic status and child health: Why is the relationship stronger for older children? American Economic Review, 93(5), 1813–1823.Find this resource:

Dearden, L., Sibieta, L., & Sylva, K. (2011). The socio-economic gradient in early child outcomes: Evidence from the Millennium Cohort Study. Longitudinal and Life Course Studies, 2(1), 19–40.Find this resource:

Del Boca, D., Flinn, C., & Wiswall, M. (2016). Transfers to households with children and child development. Economic Journal, 126(596), F136–F183.Find this resource:

Dias, M. C., Joyce, R., & Parodi, F. (2018). The gender pay gap in the UK: Children and experience in work (IFS Working Paper 18/02). London, U.K.: Institute for Fiscal Studies.Find this resource:

Fitzsimons, E., Malde, B., Mesnard, A., & Vera-Hernández, M. (2016). Nutrition, information and household behavior: Experimental evidence from Malawi. Journal of Development Economics, 122, 113–126. 2Find this resource:

Galiani, S., Gertler, P., & Schargrodsky, E. (2005). Water for life: The impact of the privatization of water services on child mortality. Journal of Political Economy, 113(1), 83–120.Find this resource:

Garces, E., Thomas, D., & Currie, J. (2002). Longer-term effects of Head Start. American Economic Review, 92(4), 999–1012.Find this resource:

Gertler, P., Heckman, J., Pinto, R., Zanolini, A., Vermeersch, C., Walker, S., . . . Grantham-McGregor, S. (2014). Labor market returns to an early childhood stimulation intervention in Jamaica. Science, 344(6187), 998–1001.Find this resource:

Goldin, C. (2014). A grand gender convergence: Its last chapter. American Economic Review, 104(4), 1091–1119.Find this resource:

Grantham-McGregor, S., Cheung, Y. B., Cueto, S., Glewwe, P., Richter, L., & Strupp, B. (2007). Developmental potential in the first 5 years for children in developing countries. The Lancet, 369(9555), 60–70.Find this resource:

Grantham-McGregor, S., Powell, C., Walker, S., & Himes, J. (1991). Nutritional supplementation, psychosocial stimulation and mental development of stunted children: The Jamaican study. The Lancet, 338, 1–5.Find this resource:

Gunnsteinsson, S., Adhvaryu, A., Christian, P., Labrique, A., Sugimoto, J., Shamim, A. A., & West, K. P., Jr. (2016). Resilience to early life shocks: Evidence From the interaction of a natural experiment and a randomized control trial. Unpublished manuscript.Find this resource:

Havnes, T., & Mogstad, M. (2011a). Money for nothing? Universal child care and maternal employment. Journal of Public Economics, 95(11), 1455–1465.Find this resource:

Havnes, T., & Mogstad, M. (2011b). No child left behind: Subsidized child care and children’s long-run outcomes. American Economic Journal: Economic Policy, 3(2), 97–129.Find this resource:

Heckman, J. J. (2007). The economics, technology, and neuroscience of human capability formation. Proceedings of the National Academy of Sciences, 104(33), 13250–13255.Find this resource:

Heckman, J. J. (2008). Schools, skills, and synapses. Economic Inquiry, 46(3), 289–324.Find this resource:

Heckman, J., Moon, S. H., Pinto, R., Savelyev, P., & Yavitz, A. (2010a). Analyzing social experiments as implemented: A reexamination of the evidence from the HighScope Perry Preschool Program. Quantitative Economics, 1(1), 1–46.Find this resource:

Heckman, J. J., Moon, S. H., Pinto, R., Savelyev, P. A., & Yavitz, A. (2010b). The rate of return to the High Scope Perry Preschool Program. Journal of Public Economics, 94(1–2), 114–128.Find this resource:

Heckman, J. J., Stixrud, J., & Urzua, S. (2006). The effects of cognitive and noncognitive abilities on labor market outcomes and social behavior. Journal of Labor Economics, 24(3), 411–482.Find this resource:

Knudsen, E. I., Heckman, J. J., Cameron, J. L., & Shonkoff, J. P. (2006). Economic, neurobiological, and behavioral perspectives on building America’s future workforce. Proceedings of the National Academy of Sciences, 103(27), 10155–10162.Find this resource:

Krishnan, P., & Krutikova, S. (2013). Non-cognitive skill formation in poor neighbourhoods of urban India. Labour Economics, 24, 68–85.Find this resource:

Lin, A., Arnold, B., Afreen, S., Goto, R., Nurul Huda, T., Haque, R., . . . Luby, S. (2013). Household environmental conditions are associated with enteropathy and impaired growth in rural Bangladesh. American Journal of Tropical Medicine and Hygiene, 89(1), 130–137.Find this resource:

Lunn, P., Northrop-Clewes, C., & Downes, R. (1991). Intestinal permeability, mucosal injury, and growth faltering in Gambian infants. The Lancet, 338(8772), 907–910.Find this resource:

Malamud, O., Pop-Eleches, C., & Urquiola, M. (2016). Interactions between family and school environments: Evidence on dynamic complementarities? (Working Paper No. 22112). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Maluccio, J. A., Hoddinott, J., Behrman, J. R., Martorell, R., Quisumbing, A. R., & Stein, A. D. (2009). The impact of improving nutrition during early childhood on education among guatemalan adults. Economic Journal, 119(537), 734–763.Find this resource:

McCoy, D. C., Peet, E. D., Ezzati, M., Danaei, G., Black, M. M., Sudfeld, C. R., . . . Fink, G. (2016). Early childhood developmental status in low- and middle-income countries: national, regional, and global prevalence estimates using predictive modeling. PLOS Medicine, 13(6), e1002034.Find this resource:

Miguel, E., & Kremer, M. (2004). Worms: identifying impacts on education and health in the presence of treatment externalities. Econometrica, 72(1), 159–217.Find this resource:

OECD. (2009). Doing better for children. Paris, France: OECD Publishing.Find this resource:

Paxson, C., & Schady, N. (2007). Cognitive development among young children in Ecuador: The roles of wealth, health, and parenting. Journal of Human Resources, 42(1), 49–84.Find this resource:

Phillips, D. A., & Shonkoff, J. P. (Eds.). (2000). From neurons to neighborhoods: The science of early childhood development. Washington, DC: National Academies Press.Find this resource:

Reynolds, S. A., Andersen, C., Behrman, J., Singh, A., Stein, A. D., Benny, L., . . . Fernald, L. C. H. (2017). Disparities in children’s vocabulary and height in relation to household wealth and parental schooling: A longitudinal study in four low- and middle-income countries. Population Health, 3, 767–786.Find this resource:

Rossin-Slater, M., & Wüst, M. (2016). What is the added value of preschool? Long-term impacts and interactions with a health intervention (Working Paper No. 22700). Cambridge, MA: National Bureau of Economic Research.Find this resource:

Rubio-Codina, M., Attanasio, O., Meghir, C., Varela, N., & Grantham-McGregor, S. (2015). The socioeconomic gradient of child development: Cross-sectional evidence from children 6–42 months in Bogota. Journal of Human Resources, 50(2), 464–483.Find this resource:

Schady, N., Behrman, J., Araujo, M. C., Azuero, R., Bernal, R., Bravo, D., . . . Vakis, R. (2015). Wealth gradients in early childhood cognitive development in five Latin American countries. Journal of Human Resources, 50(2), 446–463.Find this resource:

Thompson, R. A., & Nelson, C. A. (2001). Developmental science and the media. Early brain development. American Psychologist, 56(1), 5–15.Find this resource:

Tirole, J. (1988). The theory of industrial organization. Cambridge, MA: MIT Press.Find this resource:

Todd, P. E., & Wolpin, K. I. (2003). On the specification and estimation of the production function for cognitive achievement. Economic Journal, 113(485), 3–33.Find this resource:

Walker, S. P., Chang, S. M., Vera-Hernández, M., & Grantham-McGregor, S. (2011). Early childhood stimulation benefits adult competence and reduces violent behavior. Pediatrics, 127(5), 849–857.Find this resource:

WHO Multicentre Growth Reference Study Group. (2006). WHO child growth standards: Length/height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index-for-age: Methods and development. Geneva, Switzerland: World Health Organization.Find this resource:

## Notes:

(1.) See for instance Behrman (2008); Berlinski and Schady (2015); Black et al (2017); Carneiro and Heckman (2003); Case, Lubotsky, and Paxson (2002); Cunha et al. (2006); Currie and Stabile (2003); Grantham-McGregor et al. (2007); Heckman (2008); Paxson and Schady (2007); Phillips and Shonkoff (2000); Rubio-Codina et al. (2015).

(2.) The ECDI comprises 10 yes/no questions asked to the caregiver of children aged 36 to 59 months to assess their literacy-numeracy, learning/cognition, physical development, and socioemotional development. The data to compute it is included in recent waves of the Demographic and Health Surveys, and the Multi Indicator Cluster Surveys.

(3.) The results on percentiles correspond to children aged 1, however, they cannot reject that the socio-economic gradient on height is constant across ages except for India. The results on vocabulary are for age 5.

(4.) We heavily rely on Cunha, Heckman, and Schennach (2010) for the formal description of the evolution of human capital.

(5.) For simplicity we do not include an idiosyncratic error term in the production function. However, this is not to deny that there are inputs that influence the accumulation of human capital that neither parents nor scientists know about. Moreover, if we were measuring skills the measures would be imperfect for the latent construct and would also justify the inclusion of an idiosyncratic error term.

(6.) We will not make a distinction between inputs that are observable or unobservable to the econometrician because methods to estimate the production function are outside of the scope of this review, see Agostinelli and Wiswall (2016); Attanasio et al. (2017a); Bernal (2008): Cunha (2011); Cunha, Heckman, and Schennach (2010); Del Boca, Flinn, and Wiswall (2016); Todd and Wolpin (2003).

(7.) Aizer and Cunha (2012) also attempted to identify dynamic complementarities but with just one shock: the introduction of Head Start. They found that the intervention provided the most benefits to those with lower baseline skill levels.

(8.) Malamud et al. (2016) point out that the reduced-form approach might underestimate the presence of complementarities in the production function, because households might undo the effect of the first shock.

(9.) Most of the ideas that we present in this section are based on Blau and Currie (2006) and Blomquist et al. (2010).

(10.) Less educated children are also more likely to participate in criminal activities when they grow up. Therefore, there is a social benefit to improving human capital levels that may not be internalized when parents make decisions about how to invest in child development.

(11.) They also document that parents’ beliefs about the returns on early investments are positively correlated with household income, which may contribute to the socioeconomic gradient on child development.

(12.) As an incomplete contract in the presence of asymmetric information, childcare services are defined as an experience good. Although the economic transaction between parents and childcare providers is very straightforward (paying for childcare services), the resulting relationship between buyers (parents) and sellers (childcare providers) is rather complex. First, the contract between parents and the provider is a highly incomplete contract because it is impossible to specify how the provider should act in every possible circumstance. Second, parents cannot observe what happens in the center while they are away and children can communicate only partial information about what goes on there (thus, there is asymmetric information). Third, providers may over-invest in aspects of quality that are easily seen by parents, such as infrastructure, and under-invest in process quality, which parents do not see and may not understand the significance of.

(13.) Compulsory laws combined with public provision have also been used to compel parents to use certain services. For instance, the compulsory attendance act of 1852 enacted by the state of Massachusetts made school mandatory for children between the ages of 8 and 14 for at least three months out of each year. The penalty for not sending a child to school was a fine and the violators were to be prosecuted by the city. These laws are usually hard to enforce. However, in this case they contributed to an increase in the salience of public schooling ensured that the state would fund the operation of free public schools.

(14.) Anttila-Hughes et al. (2018) estimate that the lack of availability of formula in lower- and middle-income countries resulted in approximately 66,000 infant deaths in 1981 at the peak of the Nestlé infant formula controversy.