Child Development in a Changing World: Risks and Opportunities Jo Boyden, Stefan Dercon and Abhijeet Singh This review explores current understandings of child development and the consequences for children of risk exposure in low- and middle-income countries by integrating empirical evi- dence from development economics with insights from allied social science disciplines. It provides a holistic perspective that highlights the synergies between children’s developmen- tal domains, drawing particular attention to dimensions such as self-efficacy, self-esteem and aspirations, which have had only limited treatment in the economics literature to date, especially in developing countries. It concludes that there is strong evidence of dynamic re- lationships between risk factors in early childhood and later outcomes across multiple developmental domains, emphasizing the heightened effect of shocks to the care environ- ment and the cumulative effect of multiple shocks. It also concludes that risk is distributed unevenly, with children who are both in poverty and disadvantaged socially according to, for example, their ethnicity bearing the greatest burden; within a household, gender, birth order and other factors mean that some suffer disproportionately from shortfalls and in- complete protection. However, this review finds that low endowments in early childhood can be at least partially compensated for through improved environments and investments in later childhood, emphasizing the resilience of some children. The review goes on to explore the impact on children of dramatic socio-economic changes that have occurred in recent years with rapid growth across most developing countries. It highlights four key forces for change—fall in absolute poverty, increased access to services, changing house- hold incentives for investing in children, and changing social and cultural values—and stresses the ambiguous effects on the welfare of children and their long-term prospects. In so doing, the review aims to consolidate emerging evidence on how risks and opportunities for child development may have changed in these dynamic contexts. JEL codes: O15, I15, I24, I25, I3, J13, J16, Y8 The World Bank Research Observer # The Author 2014. Published by Oxford University Press on behalf of the International Bank for Reconstruction and Development / THE WORLD BANK. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com doi:10.1093/wbro/lku009 Advance Access publication September 24, 2014 30:193–219 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 The last two decades have seen unparalleled global interest in the survival and development of children. This interest reflects commitments made under the terms of the UN Convention on the Rights of the Child and the Millennium Development Goals recognizing improvements to children’s life chances and well-being as a legiti- mate goal of development. Growing awareness that childhood experience is crucial to the adults whom we become has emphasized that enhancing children’s condition is crucial for broader economic and societal development. Thus, investing in chil- dren is not simply the right thing to do for their survival and quality of life; it is also vital for creating and sustaining broad-based economic growth (Commission on Growth 2008). Moreover, there is mounting global consensus that economic growth facilitates, but is not sufficient for, the realization of this human potential. With many children in low- and middle-income countries continuing to experi- ence numerous threats to their well-being (e.g., Engle et al. 2011; Grantham- McGregor et. al. 2007; Walker et al. 2011), this review explores current understandings of child development and the developmental consequences of expo- sure to risk. It integrates empirical evidence from development economics with in- sights from allied social science disciplines, where most evidence is overwhelmingly focused on European and North American contexts (Bornstein et al. 2012). Recognizing the long-term impacts of early deprivation and the potential for signifi- cant change in developmental momentum in middle and late childhood, this review adopts a life-course perspective and reports on research covering early child- hood to youth. It takes a broad view of child development encompassing the entire range from conception to early adulthood. This view is consonant with the UN Convention on the Rights of the Child, which defines a “child” as a person below the age of 18 years, and with research evidence indicating that it is not only the early years of life but also adolescence and youth that can crucially influence functioning and adaptation in adulthood (Durkin 1995; Prentice et al. 2013). Moreover, given that human capital formation in different periods of childhood builds upon investments in previous periods, available evidence can be integrated into a more comprehensive framework. In doing so, this review stresses the impor- tance of understanding the specific risks and protective factors and the developmen- tal milestones and outcomes that are salient in each of the phases of childhood. Integrating this body of evidence from low- and middle-income countries1 into a single discussion informed by decades of research on child development provides fresh insights into the dynamic and complex nature of child development. First, this approach offers a holistic perspective that highlights the synergies between child- ren’s developmental domains, drawing particular attention to dimensions such as self-efficacy, self-esteem and aspirations, which are important components of human agency2 and that have received only limited treatment in the economics lit- erature to date, especially in developing countries. The aim is to provide a coherent framework that draws on a large body of literature on child development to situate 194 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 the many relevant empirical studies in development economics. This review also hopes to broaden the discussion of risks and protective factors as it is usually addressed in the child development literature on high-income countries by integrat- ing empirical evidence from developing countries within the same conceptual framework. The review attempts to be forward-looking and engage directly with the impact on children of dramatic socio-economic changes that have occurred in recent years with rapid growth and the expansion of public services across most developing countries. Specifically, it aims to consolidate emerging evidence on how risks and opportunities for child development may have changed in these very dynamic con- texts. There has been a concerted push in several of the areas of services that most affect children, such as primary education and immunization, and there has been increasing attention in many countries to social safety nets, sometimes with an explicit focus on the young. At the same time, dramatic social change presents the possibility of new risks to children that have as yet been little researched or acknowl- edged. These developments have coincided with a sharp rise in rigorous empirical research on changes in and the impact of interventions on child well-being within economics and other social sciences, encouraging discussion of policy and practice. The emergence of these new opportunities and risks poses important challenges for both policy and scholarship on child well-being. This review consolidates this emerging evidence on selected dimensions, situating it within the context of previ- ous literature across disciplines and signposting areas where knowledge remains patchy. The article is structured as follows. It first discusses the current state of knowl- edge about child development from a variety of disciplines, focusing specifically on the malleability of childhood development and the environmental3 risks confronted by children. The article then discusses changes to the socio-economic and policy contexts that have taken place in developing countries in recent years and the impli- cations for the welfare of children. It concludes by discussing the areas for further research on child well-being and development in low- and middle-income countries. What Do We Know about Children’s Development? The Importance of Environmental Influences Child development, by definition, involves the changes that occur during childhood, the factors that shape these changes, and the outcomes for children in both the short and the longer term. Recognizing the high prevalence of environmental risks and the scarcity of empirical evidence on the outcomes of risk exposure during childhood in low- and middle-income countries, this review is particularly concerned with these Boyden et al. 195 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 external influences that are brought to bear in children’s development. It follows other studies (Aber et al. 1997, 47; quoted in Dawes and Donald 2005, 8) in delin- eating child development as “the acquisition and growth of the physical, cognitive, social and emotional competencies required to engage fully in family and society”. This definition highlights that development entails adaptation to specific social and cultural contexts and comprises multiple functional domains, each of which includes an array of states and competencies. These domains exist in a synergistic relationship; the operation of one affects performance in the other(s), and the influences travel in all directions. For example, neurological shocks can affect social functioning as much as emotional well-being can impact physical growth. Although the diverse developmental domains are both malleable and interacting, they are not all equally liable to change under external influence. Thus, for example, social and emotional competencies are far more responsive to external forces than are sensory and motor functions, which tend to remain relatively stable regardless of experience (Schaffer 1996, 392). Accordingly, children’s feelings about identity, self- worth, and personal well-being are highly dependent on how they understand their relative social position, relative competence, and potential to access opportunities for personal, social, and economic advancement. These are not, for the most part, indi- vidualized processes; they are experienced as part of a family group, peer group, and community (Ridge 2002). In this way, children who experience approval, accep- tance, and opportunities for skill development are far more likely to be resourceful and resilient than are those who are subjected to humiliation, rejection, or failure (Boyden and Mann 2005). Contemporary thinking is noted for the diversity of its conceptualizations of child development. However, there is broad agreement that the process involves a complex transaction between genotypic, biological, and maturational processes that are shaped by children’s experiences, actions, and interactions and by wider environmental influences (Walker et al. 2011; Wachs and Rahman 2013), includ- ing the values of their caregivers, which are embedded in diverse cultural contexts (Rogoff 2003; Sameroff 2009). As such, individual characteristics (for instance, personality) and biological forces (including genetics, epigenetics, and neurobiologi- cal factors) work together with family dynamics (for example, family functioning) and broader historical, socio-cultural, economic and other environmental factors to influence children’s growth and adaptation. Bronfenbrenner’s (1979) ecological systems theory has been particularly persua- sive in bringing external contextual influences to the fore in child development re- search and in specifying what the nature of these influences might be.4 One of Bronfenbrenner’s key contributions has been the idea that “what matters for behav- iour and development is the environment as it is perceived rather than as it may exist in ‘objective’ reality” (Bronfenbrenner 1979, 4). In other words, he drew attention not only to material conditions but also to context-specific values and meanings as 196 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 crucial in children’s development. In all contexts, there are different types and amounts of material and psychosocial resources to support children’s development as well as different levels and types of knowledge and beliefs about how best to raise them. These beliefs fundamentally affect children’s everyday lives, including the things that they are taught or encouraged to do and to whom they turn for guidance—all of which produces great variety in the developmental challenges and opportunities that boys and girls experience (Rogoff 2003). Bronfenbrenner envisioned the environment as a series of nested and interdepen- dent structures, of which the most significant for child development are micro- systems such as the family and/or peer group. These constitute the most proximal environments for the young and exercise a direct impact on their development through routine interactions and participation in close relationships and recurring activities. Bronfenbrenner was concerned with highlighting how micro-systems in- fluence children’s development through their interaction, such as when family poverty impairs children’s school performance. He labelled this interaction “meso-systems”. His concept of exo-systems refers to the settings that affect the people who are in proximal relationships with the child while not directly influenc- ing that child, such as when caregivers are members of community associations that moderate the impacts of adversity. Bronfenbrenner’s macro-system refers to the broader political-economic and socio-cultural contexts that shape all other systems. Finally, in his formulation, “chronosystem” refers to the relationship between human-developmental time and the other systems as they evolve over time. In the transactional model outlined by Sameroff and Chandler (1975), changes in the child and in his/her environment are dynamic and mutually constitutive. As such, the child context interface may have very different effects on a child’s develop- ment at different ages (see also Sameroff and Fiese 2000). It has been argued that it is the recurrent, lasting, and reciprocal interactions between an active child and his/her environment that have the most significant effects on the development of characteristics at different points in the life cycle (Rogoff 2003). Consequently, the time spent in poverty or prosperity makes a significant difference. Not only do developmental requirements and outcomes change as children mature, but there are also periods during which children’s responses to external stimuli are heightened (Shonkoff, Boyce, and McEwen 2009; Shonkoff and Levitt 2010). This phenomenon leads to greater vulnerability to long-term harm arising from risk exposure in these periods, but it also offers a window for focusing on policy interventions that will be most potent. Early childhood is generally recog- nized as the most crucial life phase in terms of developmental malleability because this is when change is accelerated and genotypic milestones emerge. Importantly, the time sensitivity of early childhood is also socially structured by influences that include the institutions of education because early cultural learning selects and re- inforces specific cognitive and psychosocial competencies. Unequal participation in Boyden et al. 197 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 early childhood and primary education further determines long-term trajectories in the sense that institutions, teachers, and assessment systems all tend to promote some children over others depending on their perceptions of children’s characteris- tics and potential (Woodhead et al. 2009). Child environment influences operate in both directions. This phenomenon calls attention to the fact that children do not simply absorb and react to external forces but are instrumental in shaping their own environment by “selecting and even cre- ating those settings that are compatible with their individual characteristics” (Schaffer 1996, 394). In other words, the young are seldom the passive victims of overwhelming circumstances but rather are social agents whose aspirations, sense of self and actions affect their own destinies. Even among boys and girls at height- ened risk of poor outcomes due to extreme adversity, there is considerable variation in individual capacity to achieve developmental tasks. Some children thrive despite experiencing multiple developmental threats (Cicchetti and Garmezy 1993; Masten and Obradovic ´ 2006; Obradovic ´ and Boyce 2009). Different children, households, communities, and societies experience different levels of hazard and risk at various points in time. Some contexts, such as those in- volving high levels of exposure to toxic substances, chronically inadequate nutri- tion, or routinely unresponsive social stimulation, are liable to be deleterious to all children and have thus been labelled “chaotic environments” (Weisner 2008). Exposure to multiple risks may have cumulative and/or compound effects, whereby varying causal, moderating, and/or mediating influences at the individual, family, and societal levels interact and intersect to bear down on and overwhelm children (Stevens 2006; Burchinal et al. 2008). The burden of risk varies widely in accor- dance with children’s social attributes, in that different social groups, distinguished, for example, by ethnicity, religion, or caste, are subject to very different levels and forms of structural constraints and challenges. Gender is frequently emphasized as the most significant marker of social difference in childhood, with girls consistently and substantially disadvantaged compared with boys. Gender disparities are gener- ally interpreted as evidence of deep-rooted prejudice and discrimination that is asso- ciated with lower levels of parental and societal investment in females as well as their heightened exposure to harmful practices, such as sexual violence. Given that the effects of adversity on children are contingent upon both suppor- tive and detrimental elements in the wider environment, some researchers have focused on identifying and examining the promotive or protective processes and mechanisms that operate at multiple levels to shield children from various forms of adversity (Rutter 1987). This research has even shown that under certain circum- stances, exposure to stress may be associated with increased resistance to later stress, termed a “steeling effect” (Rutter 2012). Diverse personal and collective re- sources that shape both the strategies children use for overcoming misfortune and their opportunities for mastery are brought to the fore. The characteristics of 198 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 individual children may be especially salient, as may the quality of relationships in the family and school and among peers. The availability of wider formal and infor- mal mechanisms of support, such as the extended family, self-help groups, social protection programs, health services, and education, is also understood as a central part of this complex environment of protection. Empirical Evidence from Development Economics The growing evidence from empirical studies in low- and middle-income countries offers a more detailed understanding of some of the key ways in which external factors influence child development, consistent with the framework in the previous section. In this section, we explore these links directly, making explicit how several streams of research in development economics are of immediate relevance to the findings from allied social sciences. Complementary domains of child development. As discussed, one of the key insights of the developmental sciences literature is that child development is multidimensional and that its many domains interact with each other over time. Intuitively, of course, this has long been recognized in economics. A large body of work has examined the links between nutrition and learning in children.5 For example, an influential study by Glewwe, Jacoby, and King (2001) found that Filipino children who are better nourished in early childhood perform significantly better in school later, both because they enter school earlier and because of greater productivity per year of schooling. Recent research, however, has considerably broadened the empirical evidence on cross-productivity and dynamic complemen- tarities between domains of child development. In particular, in papers based on longitudinal data from the United States, Cunha and co-authors have focused the attention of economists on the importance of “non-cognitive skills”6 and the way in which they interact with “cognitive skills”, as measured conventionally by test scores, throughout childhood as well as the way they continue to shape labor market outcomes and academic achievement in later life (e.g., Cunha and Heckman 2008, 2009; Cunha, Heckman, and Schennach, 2010). Comparative quantitative evidence from developing countries is very scarce. The only peer-re- viewed study of which we are aware is that of Helmers and Patnam (2011), which builds on the methodology in Cunha et al. (2008) and applies a Linear Structural Relations (LISREL) model to data collected in 2002 and 2007 in the state of Andhra Pradesh in India by the Young Lives study.7 For the younger cohort of children born in 2001 – 2, they highlight strong evidence of dynamic relationships between risk factors at the age of one, health in early childhood, and early cognitive achievement (at the age of five). For the older cohort born in 1994 – 5, they find strong evidence of cognitive skills at the age of eight leading to a greater stock of both cognitive skills Boyden et al. 199 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 and psycho-social competencies at age 12. Although they document the cross- productivity of cognitive skills on future psycho-social competencies, in contrast to Cunha and Heckman (2008), they do not find any significant evidence in the oppo- site direction (i.e., higher psycho-social competencies producing higher cognitive skills in the future). Dercon and Sanchez (2013) examine the links between nutritional status at seven to eight years of age and three psychosocial traits, i.e., self-efficacy, self-esteem, and aspirations,8 measured at the age of 12 in the four Young Lives countries. They docu- ment that lower height-for-age at eight years of age is strongly predictive of future psychosocial traits. This association remains significant statistically and in magni- tude, even upon the inclusion of controls for current household expenditure, mater- nal psychosocial traits, past cognitive achievement, current body mass index of the child, and community fixed effects. This evidence cannot necessarily be taken as causal because omitted variable bias may persist even after including the control vari- ables. However, the strong pattern of correlations, even when the most common sources of omitted variable bias are controlled for, is suggestive of the type of causal pathways in skill formation that are indicated by the child development literature and the work of Cunha, Heckman and colleagues in economics. Using the same dataset, Dercon and Singh (2013) find that sense of agency and aspirations may be a channel through which inequalities between sexes are reproduced over time. They show that parents in India have significantly higher educational aspirations for their sons than for their daughters when the children are eight years old. This pattern in parental aspirations is transmitted (albeit not perfectly) to children’s aspirations at the age of 12, which, in turn, is associated with a lower sense of self-efficacy, lower test scores, and lower school enrollment among girls at the age of 15. The results in these studies are highly suggestive that there may be important gains in the engagement of economists with the child development literature on skill formation, the role of aspirations in determining investments in children, and the nature of different domains of functioning that develop over the course of childhood.9 Critical periods in child development. We have argued that the child development literature emphasizes the critical periods during childhood when some aspects of children’s development are especially susceptible to influence. In particular, depri- vations and shocks during early childhood are known to have profound and lasting effects across distinct developmental domains. The implication is that recovery is unlikely. Recognition of these critical phases has perhaps been most evident in the field of nutrition. It is now widely accepted that early nutrition deficits have significant impacts on later life outcomes. The literature in this sub-field has increased rapidly in recent years (see Currie and Almond 2011 for an authoritative review of the 200 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 current state of knowledge concerning the long-run impact of shocks in utero and early childhood). This literature is overwhelmingly focused on studies from OECD countries, for which the evidence base is now particularly rich, although there is a growing body of evidence from developing countries. For instance, in an influential study of a sample of women in Indonesia, Maccini and Yang (2009) report that higher rainfall in a respondent’s birth year leads to a lower incidence of (self- reported) poor health, greater terminal height, a greater number of grades complet- ed in school, and a larger accumulation of assets at 25 to 50 years of age.10 A more open question is the extent to which children show resilience through re- covery from early deprivations and shocks, or whether deficits in early childhood can be compensated for by remedial measures in later life. One domain for which re- mediation has been thought to be doubtful is nutritional deficits in early childhood, which lead to stunting (linear growth retardation). There is extensive evidence that early growth deficits persist into adult height, and it is commonly thought that stunting is irremediable after the first two years of life. Nonetheless, a small body of research is now questioning the premise that lifelong height impairment is, in fact, inevitable and is asserting that “catch-up growth” may be possible. The Cebu longi- tudinal study provided some of the first developing-country evidence on this, with Adair (1999) reporting that almost one-third of the children in the sample who were stunted at two years of age experienced catch-up growth by the time they reached 8.5 years. Similarly, using data from Peru, Crookston et al. (2010) found that a large proportion of the children who were stunted at 6 –18 months years of age had caught up by the age of five. In both studies, the magnitude of recovery in children who do catch up is very large (in excess of one standard deviation of the height-for-age z-score). Coly et al. (2006) document a similar magnitude in Senegal, with large changes for those children who were stunted at preschool.11 In a recent contribution using panel data from Brazil, Guatemala, India, Philippines, South Africa, and the Gambia, Prentice et al. (2013) similarly document substan- tial catch-up in height between the age of 24 months and mid-childhood and again between mid-childhood and adulthood. The potential for catch-up has important implications for public policy. Although preventing nutritional deprivation in early childhood remains the key priority and the most efficient form of intervention, it raises the possibility that there may be a role for remedial action for children who have suffered early nutritional deprivation. Currie and Almond (2011) survey a large body of literature on remedial welfare measures, including cash transfers, food stamps, tax credits, health insurance, and home visits by social workers and medical staff. The presence and magnitude of gains differs across interventions and evaluations, but there seems to be clear evidence that low endowments at the time of birth can be at least partially compen- sated for through improved environments and investments in later childhood. As they note in their conclusion, “. . . evidence for long term effects of early insults should Boyden et al. 201 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 not be a cause of pessimism. While children can be permanently damaged at this age, the damage can be remediated. The picture that emerges is one of vulnerability but also of resilience”. 12 Recent evidence from Young Lives is consistent with this possibility. Singh, Park, and Dercon (2014) investigate whether a large, universal, nationally mandated school feeding program in India, the Midday Meal Scheme, which operates in all government primary and upper primary schools, could help ameliorate the impacts of severe droughts in early childhood on children who were approximately five years of age at the time of the intervention. They compare children who attend schools (and have access to the Scheme) with those who have yet to enroll. They use a non- linearity in the effect of age on enrollment as an instrumental variable to correct for self-selection into the Scheme while controlling linearly for age, various socio- economic characteristics, and their nutritional status when the children were aged approximately one year. They find that although drought in early childhood had large and significant negative impacts on children’s height-for-age and weight-for- age z-scores, the school meals entirely compensated for this effect through catch-up growth among the children in the Scheme.13 Crookston et al. (2010) further document that the Peruvian children who were stunted at age one but were no longer stunted at age five demonstrated levels of cog- nitive functioning, measured by testing receptive vocabulary and grasp of quantita- tive concepts, similar to children who were never stunted.14 This finding raises the possibility that not only may catch-up be possible for a longer period than was pre- viously thought likely, but it may even ameliorate the well-documented deleterious impact of nutritional deprivation on cognition. However, the evidence on recovery in cognitive outcomes is mixed. Explicitly testing the existence of a critical window around the first 1000 days, Barham, Macours, and Maluccio (2013) find that there is catch-up in the height of children whose households (experimentally) received conditional cash transfers in later childhood compared with children whose house- holds received these transfers starting in utero and during the first two years of life. Nevertheless, there remains a significant difference in the cognitive outcomes of the two groups, indicating that any catch-up in cognitive functioning is only partial at best. Vulnerability to household shocks. We have noted that a significant proportion of chil- dren in low- and middle-income countries continue to face a multitude of risks of many forms and that exposure to multiple risks can have cumulative effect. In a dis- cussion of risks to child development, it is vital to understand how micro-systems factors, such as family and/or household dynamics, affect children and how chil- dren negotiate these dynamics insofar as they have agency to do so. It is widely ac- cepted that the care environment, generally constituted by a child’s activities and relationships within the home-based family, is the most salient external influence in 202 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 children’s lives, especially during the earliest years of life. Unsurprisingly, shocks to the care regime have been shown to have a significant impact on children’s out- comes across several domains. Bhalotra (2010) uses individual data on infant mortality for approximately 150,000 children born in 1970 – 97 across 15 major Indian states, merged by cohort and state of birth with a state panel containing information on aggregate income, to investigate child mortality and its relation to aggregate income shocks. Specifically, she compares the effects of annual deviations from general trends in income on the mortality risks of children born at different times to the same mother conditional upon a number of state-time-specific covariates, including rainfall shocks and state social expenditure. She finds that rural infant mortality is counter- cyclical in that a negative income shock of a median size (4.4 percent) leads to a rise in infant mortality by 0.136 percentage points, which is almost half of the total annual decline in mortality in India in 1970 – 99. Bhalotra documents that these effects are most likely mediated by changes in health-seeking behavior and mater- nal labor supply in response to the income shock. Thus, in a downturn, rural Indian mothers are significantly less likely to give birth outside the home or to seek antenatal care and are more likely to work outside the home. They are also far less likely to obtain immunization or treatment for their children. In a similar vein, Baird, Friedman, and Schady (2011), in the most geographically comprehensive in- vestigation undertaken on the subject using data on 1.7 million births in 59 devel- oping countries, record a large, negative association between per capita GDP and infant mortality.15 The death of parents or other caregivers is a direct shock to the care environment of children and often to household income. Beegle, De Weerdt, and Dercon (2006) examine a sample of 718 non-orphaned children in AIDS-afflicted areas of Tanzania who were surveyed in 1991–9 and then traced and re-interviewed as adults in 2004. Nearly one-fifth of the sample had lost one or more parents by the age of 15, allowing for an assessment of the permanent health and education impacts of orphanhood during childhood. The authors control for a wide range of child and adult characteristics before orphanhood as well as community fixed effects and find that maternal orphaning has a permanent adverse impact of 2 cm of final height attainment and one year of educational attainment. Expressing welfare in terms of consumption expenditure, they find a gap of 8.5 percent compared with similar children whose mothers had survived until at least their 15th birthday.16 Not everyone is equally vulnerable. Both the development economics literature and the child development literature emphasize that some households are significantly more vulnerable to adversity than others because of their location, social status, or economic factors. Typically, the burden of risk is greatest for poor households, which are also generally the least able to smooth income shocks (Dercon 2002). Boyden et al. 203 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 Moreover, household poverty is a key indicator for multiple developmental risks in children, including malnutrition, environmental toxins, low maternal education level, and family conflict (Wachs and Rahman 2013). We have noted that children’s social characteristics can be a significant indicator of the burden of risk they bear. In poor households, it is the children who are disad- vantaged by gender, birth order or other characteristics who are likely to suffer most from shortfalls and incomplete protection because limited resources are fre- quently distributed unevenly among household members. Bhalotra (2010) shows that not only is the income effect on child mortality significantly larger for children with uneducated mothers (or fathers) and mothers who became pregnant before the age of 18 but that, comparing brothers and sisters, the impact of income shocks is also much larger in girls; indeed, boys are fully protected. Baird, Friedman, and Schady (2011) report that infant mortality in females is more sensitive to negative income shocks than it is in males, and Maccini and Yang (2009) establish that the links between rainfall shocks in the birth year and adverse later life outcomes in Indonesia are apparent only in women. Households are not, in most cases, powerless against shocks, nor do they passive- ly suffer their consequences. Attempts to reduce vulnerability and/or ameliorate impacts include diversification of income sources and crops, informal group-based risk sharing, and increasing labor supply. Coping strategies may result in children moving (temporarily or permanently) into the home of relatives who are better off or in need of more labor, and this can be a source of either developmental risk or learning and social support (Boyden and Howard, 2013). This practice of fostering has been documented quantitatively using data from Burkina Faso by Akresh (2009). He shows that households experiencing negative idiosyncratic income shocks, with gender imbalances among children, located further from primary schools, or with more “good” quality network members (fewer subsistence farmers and unmarried individuals and more educated individuals) are significantly more likely to send a child away. Children commonly share responsibility for household maintenance, very often making significant contributions by undertaking domestic chores, unpaid work in a family enterprise, or paid employment or by caring for younger siblings or inca- pacitated adults (Bourdillon et al. 2010; Heissler and Porter 2013). For many boys and girls, playing an active part in the household economy is vital to their social in- clusion and hence is an important source of identity, pride, and self-efficacy. It also enables them to gradually learn life skills appropriate to their gendered adult roles. Nevertheless, these risk-coping mechanisms can be damaging for the young. For example, Beegle, Dehejia, and Gatti (2006) document a rise in child work in re- sponse to shocks to crop income in rural Tanzania.17 Jacoby and Skoufias (1997) similarly report a reduction in children’s school attendance following shocks in agricultural income in rural India. 204 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 Even when individual and household coping mechanisms provide a shield during crises, they do not always offer complete insurance against the effects of shocks or enduring hardship. Often, this is most evident with adversities that affect entire communities insofar as they commonly limit the ability of households to share risks. Even within households, risk sharing may not be complete. Dercon and Krishnan (2000) test for perfect risk sharing in households confronting adverse shocks and examine whether individuals are able to smooth consumption over time and within households. These authors establish that poorer households are not able to do so. They find that in southern Ethiopian villages, the brunt of adverse shocks and of incomplete risk sharing in the household is borne by women. This discussion of environmental risks is not exhaustive. It is merely intended to illustrate that children, households, communities and population groups may be prey to a multitude of risks, some bearing a far higher risk burden than others, and that such risks may have severe consequences for children that can persist through- out their lives. An understanding of the nature and extent of such risks is therefore crucial for appreciating the potential loss of developmental potential among chil- dren in low- and middle-income countries. Changing Contexts in Developing Countries and Their Effects on Children Many parts of the developing world have undergone rapid socio-economic transfor- mations over the last two decades. In this section, we highlight the impact on chil- dren. We identify four significant trends, as follows: an (uneven) fall in absolute poverty; increased access to services, especially education; changing household incentives for investing in children; and changing social and cultural values. Fall in Absolute Poverty One of the most notable trends in recent decades has been the steady decline in ab- solute poverty, as measured by the ability to meet the cost of “basic needs”, across most parts of the developing world. Chen and Ravallion (2010) analyze 675 repre- sentative household surveys from 115 countries together with internationally com- parable price data to examine trends in global poverty between 1981 and 2005. They report that whereas 52 percent of the population of the developing world was below a poverty line of $1.25 per day in 1981, the comparable figure was 25 percent in 2005. The rate of decline in poverty over this period was 1 percent per year. This decline was far more significant in Asia (especially China) than it was in sub-Saharan Africa, where the gains have been more modest. Importantly, gains have been unequal within countries. For example, whereas India has experienced Boyden et al. 205 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 rapid economic growth and impressive rates of poverty reduction, it remains home to more poor people than any other country in the world. These trends have direct relevance for children’s well-being in low and middle- income countries, with three important implications. First, given that poverty is a significant risk factor for children, any reduction in poverty is certainly good news in and of itself.18 Second, however, there are important grounds for caution. Chen and Ravallion’s analysis (2010) reveals a large “bunching” of people between the poverty line of $1.25 per day and $2 per day. In other words, whereas much of the developing world may now have escaped poverty, many populations remain signifi- cantly vulnerable to income shocks. As shown in the previous section, even tempo- rary income shocks can directly prejudice children’s outcomes, with long-term consequences. Thus, in households that are close to the poverty line, any gains from a reduction in poverty are likely to be significantly lessened by vulnerability to shocks, especially in the absence of supplementary social protection measures. Finally, the trends outlined by Chen and Ravallion (2010) imply that poor house- holds (and, by extension, children in poor households) will increasingly be in middle-income countries, highlighting the importance of measures that address inequalities and cater to children in particularly disadvantaged groups. Access to Services and Inequality in Outcomes Another force for change that crucially affects children in developing countries is the expansion of public infrastructure and services, particularly those services that are aimed at the young, such as education and immunization. This expansion is sig- nificant in at least three respects: it affects the absolute level of child well-being; it can change the level of inequality in the distribution of child outcomes as well as the nature and relative importance of different dimensions in inequality; and it changes the focus of policy debates around public services and of interventions. It is not a priori clear that access to social services will affect equality between children over time inasmuch as increased access does not necessarily have a mono- tonic relationship to societal inequalities. Oster (2009) presents a theoretical model that predicts that when starting at lower initial levels of access, an expansion of ser- vices will lead to higher inequality, whereas when starting at a high level of access, service expansion will reduce inequality. She tests this theory using data on vacci- nations for children in India, with the location of health camps as identification. She rigorously documents that although an increase in the number of health camps initially led to a rise in gendered inequality in vaccinations, further increases led to its decline. This trend directly maps into mortality: in areas that begin with high levels of vaccination, excess female mortality decreases over time. Perhaps the starkest illustration of how increased service access shifts the nature and level of inequalities among children as well as the boundaries of public 206 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 discourse is found in recent advances in primary education in developing countries. Following the push for Education for All and the Millennium Development Goals, primary school enrollment has risen sharply across the developing world, and ever- enrollment into primary schooling is near-universal in most places (Grant and Behrman 2010; World Bank 2012). This expansion has been accompanied by im- portant changes in inequalities in educational investments. In India, for example, data from the mid-1990s reflect that the primary channel for gender bias in educa- tion was through enrollment, with boys significantly more likely to be enrolled than girls, but that, conditional on enrollment, there were no significant biases in spend- ing on education (Kingdon 2005). However, this pattern has almost entirely reversed since then. Using data from 2005 and replicating the methodology of Kingdon (2005), Azam and Kingdon (2012) report that fewer states present evidence of gender bias in 2005 compared with 1993 and that there is little evi- dence of any gender bias in enrollment in the five to nine year age group (although there is some evidence for later years). Maitra, Pal, and Sharma (2011) use the same data to document that the gender gap in private school enrollment is double that in enrollment overall and has increased over time in rural areas.19 Given the above, it should not be surprising to find that the relative importance of diverse axes of inequality is very different from the received wisdom, with gender differences as the most obvious example. Dercon and Singh (2013) systematically investigate gender bias at the ages of eight, 12 and 15 years for children in the four Young Lives countries.20 They show that gender preferences are often specific to age, contexts, and indicators and are not always as expected. Not only are the gaps associated with other socio-economic factors often larger, but gender gaps may also be absent across most outcomes (in Peru) and may even reveal a pro-girl bias (in Vietnam). Using DHS data from 38 countries, Grant and Behrman (2010) also document considerable heterogeneity in gender bias across regions of the develop- ing world and phases of childhood. Interestingly, they find that, conditional on ever- enrollment, girls seem to have a consistent advantage in school progression, leading these authors to conclude that “developing countries are becoming more like devel- oped countries with gender gaps that increasingly favor, rather than discriminate against, females” (2010, 71). The rise in school availability has also changed the academic and policy discours- es about education. The focus is shifting from access, which dominated policy dis- cussions until the 1990s (embodied, for example, in several enrollment-related targets in the MDGs), to quality and relevance. As much work has documented, ed- ucation quality remains poor, and problems of school accountability are endemic in many low- and middle-income countries. In an influential study, Chaudhury et al. (2006) found that in six developing countries, up to one-quarter of teachers were absent at the time of unscheduled visits to schools. In India, half of the teachers who were present were not teaching. Moreover, many children complete primary Boyden et al. 207 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 school without the ability to perform basic tasks, such as reading a simple sentence or resolving a simple division problem (Pratham 2012). Recently, there has been a marked rise in private schooling in many low- and middle-income countries, reflecting, at least in part, the poor quality of education in government schools.21 Although causality is debated, several studies have docu- mented the higher median achievement of children in private schools, though the average expenditure per child tends to be far lower than that in government facili- ties. For example, using yearly child-level panel data on private and state school pupils, this pattern has been documented for rural areas of Pakistan by Andrabi et al. (2011). Bold et al. (2011) report similar findings from Kenya using several years of data from an exam administered to primary school students throughout the country. Muralidharan and Sundararaman (2013) and Singh (2013) both report much greater productivity of private schools and causally greater absolute effects on some aspects of learning (notably English) in India. This remarkable transformation in the educational landscape, with high enroll- ment (at least at the primary level), shifting inequalities, and a burgeoning private sector, presents both opportunities and potential risks for children’s development, some of which are only now being explored in depth. The important consideration here is not to definitively pronounce the aggregate implications of such changes, which may not be possible given the significant gaps in our knowledge. Rather, it is important to note that an understanding of child development in low- and middle- income countries must necessarily consider recent policy developments. Changing Household Incentives for Investing in Children Inequalities and differential investments in and opportunities presented to different children are also affected by broader economic trends in developing countries. To emphasize these links, this sub-section highlights two studies that show how even the most entrenched inequalities may be affected. We focus on gender discrimina- tion in India, which has been widely documented in the literature and builds on our previous discussion. Concentrating on educational access, Munshi and Rosenzweig (2006) document how patterns of enrollment across public and private schools changed in a part of urban Bombay (Mumbai) between 1980 and 2000 in response to a transformation in the labor market returns to English-medium education. In stark contrast with results from the rest of the country, these authors find that although the gaps in the proportion of children studying in English-medium schools have declined con- siderably between different caste groups in this period, the decline has mostly been caused by a rapid increase in the proportion of enrollment among girls from lower castes. They explain this finding using caste networks in employment, which provide lower-caste boys with blue-collar jobs in manufacturing that have not 208 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 traditionally required familiarity with English.22 Clearly, an understanding of trends in inequality needs to account for the changing incentives experienced by households in the context of economic growth. Jensen (2012a) presents striking evidence of how transformations in economic opportunities in developing countries can affect investments in children and youth at different ages. He uses an innovative experimental research design to examine the impact of expanding labor market opportunities for women in India. Women in randomly selected villages were connected to recruiters for the Business Process Outsourcing (BPO) industry. Jensen evaluated the effect of this (by design) exoge- nous change in labor market outcomes on outcomes for girls and young women in these villages. Using panel data spanning a three-year period, he found that women aged 15 to 21 at baseline from villages exposed to the intervention were 4.6 per- centage points more likely to work in a BPO job than were women in control villages and 2.4 percentage points more likely to work at all for pay outside the home. Additionally, these young women expressed a greater interest in working through- out their lives, even after marriage and childbirth, indicating shifting aspirations toward paid work as a career. The higher educational requirements and greater returns to human capital in the BPO sector also led to increased investments for women. The cohort of 15- to 21-year-old women from treatment villages was sig- nificantly more likely to enroll in computer or English-language courses at private, for-fee training institutes, indicating a willingness to invest in getting a job or build- ing a career when suitable opportunities are available. Significantly, even younger, school-aged girls had increased school enrollment and a greater body mass index (BMI), reflecting better nutrition and/or health investments. Changing Social and Cultural Values In many developing countries, economic growth and the expansion of infrastruc- ture and social policy has occurred alongside the spread of information and com- munications technology and the media, all of which can have dramatic and lasting effects on material conditions, norms, values and practices. In this section, we high- light recent evidence of social changes that have led to important differences in indi- vidual and community outcomes. Although the link has not been made explicitly, the implications for children’s development are likely to be significant. La Ferrara et al. (2012) study how fertility choices in Brazil shifted significantly between 1970 and 1991 in response to the introduction of television, specifically soap operas. They identify the effect through variation in the timing of the introduc- tion of the main television network in different areas and show that their results are robust to possible concerns of selection bias. The decline in fertility occurs not through delayed commencement of childbearing but by stopping childbearing earlier.23 Similarly, Jensen and Oster (2009) study the impact of the introduction of Boyden et al. 209 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 cable television on women’s status in India using a three-year individual-level panel dataset and find that it led to a decline in female fertility. Equally importantly, they also report evidence of a decline in the acceptability of violence against women and an increase in school enrollment among younger children, which they hypothesize as arising from an increase in women’s participation in decision making.24 Not all of the outcomes of new information and communications technology and entertainment sources are beneficial. Olken (2009) uses the variation in television and radio reception across 600 different Indonesian villages to investigate the impact on social capital. Using arguably exogenous variation induced by differences in terrain and in the timing of the introduction of private television and radio,25 he demonstrates that increased signal reception leads to less participation in social or- ganizations and lower self-reported trust. Although his analysis does not include child outcomes, it is likely that these types of changes have had indirect effects on children growing up in these communities. The rapid recent expansion of television availability is only one of many changes that may affect the social values that shape children’s lives; this example was select- ed here mostly because of the availability of peer-reviewed evidence. Our key point is that such changes need to be studied explicitly if we want to understand trends in child well-being in the rapidly changing contexts of many developing countries. Conclusions and Future Directions The broad spread of our discussion precludes us from offering specific directions for either research or policy. However, it does allow us to draw out commonalities between the strands of research that have proven most fruitful and to frame how the effect of changing environments on children may best be understood. The first lesson we draw from reviewing current evidence is the importance of cohort studies that encompass diverse dimensions of child well-being. We argue that data of this type are extremely important for at least two crucial reasons. First, child development is a sequential process involving dynamic complementarities; thus, the long-term consequences of external influences at different stages of childhood and adolescence can only be convincingly studied if data are available for the same indi- viduals over time. Although some questions, such as the impact of shocks in early childhood, may be examined by matching previous external data sources to existing cross-sectional studies such as the DHS and LSMS surveys, others (such as the possi- bility of catch-up growth in later childhood) cannot be persuasively researched without access to high-quality panel data. It is precisely in the latter type of ques- tions that our empirical evidence may be most effectively augmented. The second reason is more specific to developing countries. Rapid change in socio-economic con- texts has ambiguous effects on the welfare of children and their long-term prospects. 210 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 Significant changes may occur even in a very short period of time and may not be foreseen, and the outcomes for children may be unknown. This understanding will be critical for designing timely and effective policy responses to changing circumstances to safeguard children’s well-being. An existing system of longitudinal surveys is perhaps the best means for evaluating change as and when it occurs. More specifically, we would like to highlight two particular strands of research in child development that could benefit from much greater attention. The first seeks to understand which periods are critical for different domains of child development, whether deprivations or abuses to children in these critical periods are irreversible, the relative costs of later remediation, and, finally, how remediation may best be achieved. The answers to these questions are of first-order importance for policy. Even if preven- tion remains the most cost-effective form of intervention, it is vital to be able to offer effective support to recovery, if possible. Our knowledge in this area remains patchy. The second area of research involves seeking better understanding of the inter- dependence and complementarities across domains of children’s development. The recent focus in the economics literature on outcomes beyond nutrition and cogni- tion is welcomed. However, as we have highlighted, evidence on developing coun- tries remains sparse, and it cannot be presumed that the relationships documented in OECD countries, such as in the interaction between cognitive and psycho-social skills, would necessarily be generalizable to developing countries. Children’s devel- opment is not merely driven by biological imperatives but is also shaped by social and other environmental factors that differ greatly both across and within develop- ing countries. It is vital that this complex process be studied in specific socio- economic contexts. Furthermore, the synergies between developmental domains call for the need to employ an integrated approach to policy across different sectors and levels of government and in international agencies. We end by highlighting that policies aimed at supporting households can have major impacts on children. Although this point may seem rather obvious (and has been recognized frequently in previous work), very often, such policies work from assumptions rather than evidence of the challenges children face, their needs and the impacts of interventions on them. Supporting households without attending to the specific conditions and circumstances of children may be inefficient at best and detrimental to the young at worst. Children merit consideration not merely on ethical grounds but also because even temporary setbacks can have significant long-term impacts on their lives. Notes Corresponding author Jo Boyden, Department of International Development, University of Oxford , jo.boyden@qeh.ox.ac.uk . , Stefan Dercon, Department of International Development, University of Boyden et al. 211 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 Oxford, (at the time of writing, Stefan Dercon was also Chief Economist of the UK Department for International Development, DFID), Abhijeet Singh, Department of International Development, University of Oxford. Young Lives is funded by UK aid from the Department for International Development (DFID) and co-funded by the Netherlands Ministry of Foreign Affairs from 2010 to 2014, and by Irish Aid from 2014 to 2015. The views expressed here are those of the author(s). They are not necessarily those of Young Lives, the University of Oxford, DFID or other funders. 1. The challenges of delivering basic services, such as health and education, and the volume and severity of threats to children’s well-being in developing countries have recently provided growing mo- mentum to research on selected aspects of child development within development economics (e.g., Edmonds 2007; Glewwe and Miguel 2007; Glewwe and Kremer 2006; Ferreira and Schady 2009; Almond and Currie 2011). 2. The term ‘agency’ is used here to refer to the capacity of an individual to take action in a partic- ular situation. 3. In this paper, “environment” includes the physical, institutional, social, and cultural factors that are external to the individual. 4. Bronfenbrenner critiqued theories prevalent in the 1970s in which environmental factors were “limited to a few crude and undifferentiated categories that do little more than locate people in terms of their social address” (1979, 17), listing family size, single- versus two-parent households as well as variation by social class or ethnic background as examples of factors typically used in studies to describe contexts. He further criticized models that limit environmental influences on behavior and de- velopment to interpersonal processes, such as modelling, identification, and social learning. He argued that these do not sufficiently account for non-social aspects of the environment, including the substantive nature of the activities engaged in, and he delimited the concept of environment to a single immediate setting containing the subject. His concern with the latter was that it could distort the attribution of the determinants, processes, and potential of human development because it does not consider a larger scope of potential effects. 5. See Glewwe and Miguel (2007) for an authoritative review of several studies documenting a link between the health and educational outcomes of children. 6. Referred to in this paper as “psycho-social competencies” to emphasize that this is an important area of child development in its own right that contains several distinct domains with associated skills and states and is not merely an umbrella for everything that is not cognition. These are also referred to in some disciplines as socio-emotional skills. 7. Young Lives , www.younglives.org.uk . is a longitudinal study of childhood poverty in Ethiopia, India (Andhra Pradesh state), Peru and Vietnam. It follows nearly 12,000 children in two age cohorts. Currently, three rounds of household-based data have been collected and are available for public use. 8. The authors use self-esteem to refer to an individual’s feelings of his or her own merit and com- petence and follow Albert Bandura’s definition of self-efficacy as “belief in one’s capabilities to organize and execute the courses of action required to produce given attainments” (1997, 31). Aspirations refer to an individual’s desired goals or objectives. The hypothesis is that individuals who perceive them- selves to be more worthy and efficacious and with higher aspirations are more motivated and likely to achieve desired goals. 9. This interaction between allied social science disciplines can be an advantage for both theoreti- cal and empirical work in economics. Although this review is overwhelmingly focused on recent em- pirical evidence, Mookherjee, Napel, and Ray (2010) provide an interesting theoretical contribution in which they model how parental aspirations influence investments in children and how these aspira- tions are themselves affected by social interactions with peers. 10. Other studies from developing countries that examine the long-term impacts of early childhood shocks include Alderman et al. (2001), Glewwe and King (2001), Glewwe, Jacoby, and King (2001) and Maluccio et al. (2009). 212 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 11. The phenomenon of catch-up growth and older available evidence is reviewed in Golden (1994) and Boersma and Wit (1997). 12. There are two important points to note about the Currie and Almond (2011) review: it relies mostly on evidence from OECD countries, and the primary focus is on remediation in cognitive skills and behavioral outcomes. Their insight on the possibility of remediation in these outcomes is shared by the child development literature (Rutter 2012). 13. See also Schott et al. (2013), who present factors at the household and community levels that are associated with catch-up in the four Young Lives countries. Although this evidence is correlation- al, it may hold valuable information for designing possible interventions to aid recovery from stunting. 14. See also several related contributions based on Young Lives data, particularly Lundeen et al. (2013), who document that the incidence of recovery from stunting between 5 and 8 years ranged from 27 percent in the Vietnamese sample to 53 percent in the Ethiopian sample; Crookston et al. (2013), who document that changes in growth after infancy are significantly associated with test scores and being in the right grade-for-age at the age of 8 years; and Fink and Rockers (2014), who, similar to Crookston et al. (2010), document that children who recover from stunting show smaller cognitive deficits compared to persistently stunted children. 15. These recent explorations of the effects of aggregate economic shocks on infant mortality high- light the importance of not uncritically generalizing findings from research on childhood in OECD countries. The relationship between economic cycles in OECD countries has been shown to be the op- posite of the relationship documented in these papers. Recent longitudinal research on the US, for example, finds that child mortality is lower in recessions and higher during periods of growth. This may reflect the substitution of time away from the labor market into health-preserving activities by mothers during recession (Dehejia and Lleras-Muney 2004). Note, however, that this does not negate the possi- bly important role of heterogeneity between developing countries. Miller and Urdinola (2011) find that the relationship between global coffee prices and infant mortality in Colombia (a major coffee ex- porter) is similar to the results of economic shocks in the US: infant mortality is lower when coffee prices are low and higher when coffee prices are high, a result that is in stark contrast to those of Bhalotra (2010) and Baird, Friedman, and Schady (2011). 16. These studies highlight that shocks, even in middle childhood, can have large negative impacts, and vulnerability in many key domains remains even beyond early childhood. 17. Child work is not inherently damaging to children, but it can be harmful under certain condi- tions and may detract from children’s schooling. Beegle, Dehejia, and Gatti (2009) show that child work at around 10 years of age in Vietnam, arising in response to changes in rice prices, resulted in significantly lower educational attainment, a lower probability of being enrolled in school, and a much greater probability of taking up wage labor five years later. 18. Whether income growth translates into improvement of outcomes for children depends cru- cially on questions of intra-household allocation of resources. As our previous discussion highlights, it is not obvious that all children will benefit from these gains (for example, there may be important intra-household differences based on gender or birth order), but, on aggregate, the rise in incomes is a cause for optimism. 19. A similar pattern has been documented by Woodhead, Frost and James (2013) using Young Lives data from Andhra Pradesh. 20. Dercon and Singh (2013) investigate gender biases across 13 different indicators relating to nutrition (height-for-age, weight-for-age and BMI-for-age z-scores); education and achievement (enrollment as well as test scores in arithmetic and the Peabody Picture Vocabulary Test); educational aspirations (reported by children and parents); subjective well-being (reported by children using the “ladder of life”, a measure of life satisfaction); and four psychosocial competencies (agency and efficacy, trust, pride and self-esteem, and sense of inclusion). 21. In India, for example, even in rural areas where private schooling accounted for only 10 percent of the enrollment in 1993 (Kingdon 2007), its share had risen to approximately 25 percent of total enrollment by 2011 (Pratham 2012). The availability of private schools is considerably higher in urban areas. As the data from the India Human Development Survey (2005) document, in many Boyden et al. 213 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 populous states such as Andhra Pradesh, Uttar Pradesh, Punjab, Haryana and Madhya Pradesh, the share of private schooling is considerably higher than the share of state schools in the enrollment of 6- to 14-year-old children in urban areas. 22. Our discussion above should not be taken to imply that rapid changes in the incentives to invest in education have not been documented previously. For example, Foster and Rosenzweig (1996), using panel data on Indian households in 17 states, document that the Green Revolution raised returns to education, which increased private investments in education. The change from the pattern demonstrated there and the pattern documented in Munshi and Rosenzweig (2006) is that in the latter, returns to a particular type of education (English-medium education) have grown much faster post-liberalization than returns to vernacular education. Furthermore, there are documented hetero- geneities in the rates of return for different types of children (boys and girls). These results, and the pre- vious results of Foster and Rosenzweig (1996), highlight that to understand the impact of aggregate economy-wide changes on educational investments (or investments into childhood more broadly), we need to focus on heterogeneous impacts across children in different categories. 23. Although they do not explicitly evaluate the effect of television availability on children’s welfare, it is quite plausible that the rapid decline in fertility documented in the paper has direct impli- cations for the welfare of children in these households. The general literature on the relationship between family size and investments in children (the “quality-quantity trade-off ”) is vast and has used several strategies to examine this effect, including unanticipated shocks to fertility from twin births (Rosenzweig and Wolpin 1980) and the sex composition of the first two children (Angrist and Evans 1998). For example, Jensen (2012b) demonstrates a direct link between higher fertility and child mal- nutrition, using secondary infertility as an instrument to correct for endogenous fertility choices. 24. A central concern in their identification strategy is whether the results merely reflect differen- tial trends across communities that received cable TV and those that did not. They argue convincingly that there is no evidence of pre-existing trend differences between communities that received access to cable TV during the period of their study and other communities and that the results are causal. 25. Olken does not differentiate between the effect of radio and television reception in this study but rather reports the combined effect. In his data, television and radio reception are very strongly cor- related even within districts. References Aber, J. L., M. Gephart, J. Brooks-Gunn, and J. J. Connell. 1997. Development in context: Implications for studying neighborhood effects. In: J. Brooks-Gunn, G. J. Duncan, and J. L. Aber, eds., Neighborhood poverty. Volume 1: Context and consequences for children. New York: Russell Sage Foundation. Adair, L. S. 1999. “Filipino Children Exhibit Catch-Up Growth from Age 2 to 12 Years.” The Journal of Nutrition 129 (6) : 1140–8. Akresh, R. 2009. “Flexibility of Household Structure Child Fostering Decisions in Burkina Faso.” Journal of Human Resources 44 (4) : 976– 97. Alderman, H., J. R. Behrman, V . Lavy, and R. Menon. 2001. “Child Health and School Enrollment: A Longitudinal Analysis.” Journal of Human Resources 36 (1) : 185–205. Andrabi, T., J. Das, A. I. Khwaja, and T. Zajonc. 2011. “Do Value-added Estimates Add Value? Accounting for Learning Dynamics. “American Economic Journal: Applied Economics 3 (3) : 29 –54. Angrist, J. D., and W. N. Evans. 1998. “Children and Their Parents’ Labor Supply: Evidence from Exogenous Variation in Family Size.” American Economic Review 88 (3) : 450–77. Azam, M., and G. G. Kingdon. 2012. “ Are Girls the Fairer Sex in India? Revisiting Intra-Household Allocation of Education Expenditure.” World Development 42 : 143 –64. 214 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 Baird, S., J. Friedman, and N. Schady. 2011. “Aggregate Income Shocks and Infant Mortality in the Developing World.” Review of Economics and Statistics 93 (3) : 847–56. Bandura, A. 1997. Self-efficacy: The exercise of control. New York: Freeman Barham, T., K. Macours, and J. A. Maluccio. 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– 71. Beegle, K., R. J. Dehejia, and R. Gatti. 2006. “Child Labor and Agricultural Shocks.” Journal of Development Economics 81 (1) : 80 –96. . 2009. “Why Should We Care about Child Labor? The Education, Labor Market, and Health Consequences of Child Labor.” Journal of Human Resources 44 (4) : 871 –89. Beegle, K., J. De Weerdt, and S. Dercon. 2006. “Orphanhood and the Long-run Impact on Children.” American Journal of Agricultural Economics 88 (5) : 1266–72. Bhalotra, S. 2010. “Fatal Fluctuations? Cyclicality in Infant Mortality in India.” Journal of Development Economics 93 (1) : 7–19. Boersma, B., and J. M. Wit. 1997. “Catch-up Growth.” Endocrine Reviews 18 (5) : 646–61. Bold, T., M. Kimenyi, G. Mwabu, and J. Sandefur. 2011. “Why Did Abolishing Fees Not Increase Public School Enrollment in Kenya?” Working Paper 271, Washington DC: Center for Global Development. Bornstein, M., P. Rebello Britto, Y. Nonoyama-Tarumi, Y. Ota, O. Petrovic, and D. L. Putnick. 2012. “Child Development in Developing Countries.” Child Development 83 (1) : 16 –31. Bourdillon, M., D. Levison, W. Myers, and B. White. 2010. Rights and Wrongs of Children’s Work, New Brunswick, NJ: Rutgers University Press. Boyden, J., and N. Howard. 2013. “Why Does Child Trafficking Policy Need to be Reformed? The Moral Economy of Children’s Movement in Benin and Ethiopia.” Children’s Geographies 11 (3) : 354 –68. Boyden, J., and G. Mann. 2005. “Children’s Risk, Resilience and Coping in Extreme Situations.” In M. Ungar, ed., Handbook for Working with Children and Youth: Pathways to Resilience across Cultures and Contexts. Thousand Oaks, CA: Sage Publications. Bronfenbrenner, U. 1979. The Ecology of Human Development: Experiments by Nature and Design. Cambridge, MA: Harvard University Press. Burchinal, M. R., J. E. Roberts, S. A. Zeisel, and S. J. Rowley. 2008. “Social Risk and Protective Factors for African American Children’s Academic Achievement and Adjustment during the Transition to Middle School.” Developmental Psychology 44 (1) : 286 –92. Chaudhury, N, J. Hammer, M. Kremer, K. Muralidharan, and F. H. Rogers. 2006.” Missing in Action: Teacher and Health Worker Absence in Developing Countries.” Journal of Economic Perspectives 20 (1) : 91 –116. Chen, S., and M. Ravallion. 2010. “The Developing World is Poorer than We Thought, but No Less Successful in the Fight against Poverty.” The Quarterly Journal of Economics 125 (4): 1577–625. Cicchetti, D., and N. Garmezy. 1993. “Milestones in the Development of Resilience.” Development and Psychopathology 5 (4) : 497 –502. ´ fice, F. Simondon, S. Wade, and K. B. Simondon. 2006. ´ ne Coly, A. N., J. Milet, A. Diallo, T. Ndiaye, E. Be “Preschool Stunting, Adolescent Migration, Catch-up Growth, and Adult Height in Young Senegalese Men and Women of Rural Origin. The Journal of Nutrition 136 (9) : 2412–20. Commission on Growth and Development. 2008. The Growth Report: Strategies for Sustained and Inclusive Development. Washington DC: World Bank. Boyden et al. 215 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 Crookston, B. T., M. E. Penny, S. C. Alder, T. T. Dickerson, R. M. Merrill, J. B. Stanford, C. A. Porucznik, and K. A. Dearden. 2010. “Children Who Recover from Early Stunting and Children Who Are Not Stunted Demonstrate Similar Levels of Cognition.” Journal of Nutrition 140 (11) : 1996–2011. Crookston, B. T., W . Schott, S. Cueto, K. A. Dearden, P. Engle, A. Georgiadis, and E., Lundeen et al. 2013. “Postinfancy Growth, Schooling, and Cognitive Achievement: Young Lives.” American Journal of Clinical Nutrition 98 (6) : 1555–63. Cunha, F., and J. Heckman. 2008. “Formulating, Identifying and Estimating the Technology of Cognitive and Non-cognitive Skill Formation.” Journal of Human Resources 43 (4): 738–82. . 2009. “The Economics and Psychology of Inequality and Human Development.” Journal of the European Economic Association 7 (2– 3) : 320–4. Cunha, F., J. Heckman, and S. Schennach. 2010. “Estimating the Technology of Cognitive and Non- cognitive Skill Formation.” Econometrica 78 (3): 883–93. Currie, J., and D. Almond. 2011. “Human Capital Development before Age Five.” Handbook of Labor Economics 4 : 1315–486. Dawes, A., and D. Donald. 2005. Improving Children’s Chances: Linking Developmental Theory and Practice. Richmond, VA: Christian Childrens Fund. Dehejia, R., and A. Lleras-Muney. 2004. “Booms, Busts, and Babies’ Health.” The Quarterly Journal of Economics 119 (3) : 1091–130. Dercon, S. 2002. “Income Risk, Coping Strategies, and Safety Nets.” World Bank Research Observer 17 (2) : 141– 66. Dercon, S., and P. Krishnan 2000. “In Sickness and in Health: Risk Sharing within Households in Rural Ethiopia.” Journal of Political Economy 108 (4) : 688 –727. ´ nchez. 2013. “Height in Mid Childhood and Psychosocial Competencies in Late Dercon, S., and A. Sa Childhood: Evidence from Four Developing Countries.” Economics & Human Biology 11 (4) : 426 –32. Dercon, S., and A. Singh. 2013. “From Nutrition to Aspirations and Self-Efficacy: Gender Bias over Time among Children in Four Countries.” World Development 45 : 31 –50. Durkin, K. 1995. Developmental Social Psychology: From Infancy to Old Age. Malden: Blackwell Publishing. Edmonds, E. V. 2007. “Child Labor.” Handbook of Development Economics 4 (5) : 3607–709. Engle, P. L., L. C. H. Fernald, H. Alderman, J. Behrman, C. O’Gara, A. Yousafzai, and M. Cabral de Mello, et al. 2011. “Strategies for Reducing Inequalities and Improving Developmental Outcomes for Young Children in Low-income and Middle-income Countries.” The Lancet 378 (9799) : 1339–53. Ferreira, F. H., and N. Schady. 2009. “Aggregate Economic Shocks, Child Schooling, and Child Health.” The World Bank Research Observer 24 (2) : 147– 81. Fink, G., and P. C. Rockers. 2014. “Childhood Growth, Schooling, and Cognitive Development: Further Evidence from the Young Lives Study.” The American Journal of Clinical Nutrition 100 (1) : 182 –8. Foster, A., and M. Rosenzweig. 1996. “Technical Change and Human-Capital Returns and Investments: Evidence from the Green Revolution.” American Economic Review 86 (4) : 931–53. Glewwe, P ., H. G. Jacoby, and E. M. King. 2001. “Early Childhood Nutrition and Academic Achievement: A Longitudinal Analysis.” Journal of Public Economics 81 (3) : 345 –68. Glewwe, P., and E. M. King. 2001. “The Impact of Early Childhood Nutritional Status on Cognitive Development: Does the Timing of Malnutrition Matter?” The World Bank Economic Review 15 (1) : 81 –113. Glewwe, P., and M. Kremer. 2006. “Schools, Teachers, and Education Outcomes in Developing Countries.” Handbook of the Economics of Education 2 (2) : 945–1017. 216 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 Glewwe, P., and E. A. Miguel. 2007. “The Impact of Child Health and Nutrition on Education in Less Developed Countries.” Handbook of Development Economics 4 (5) : 3561–606. Golden, M. H. 1994. “Is Complete Catch-Up Possible for Stunted Malnourished Children?” European Journal of Clinical Nutrition 48 : S58 –71. Grant, M. J., and J. R. Behrman. 2010. “Gender Gaps in Educational Attainment in Less Developed Countries.” Population and Development Review 36 (1) : 71 –89. Grantham-McGregor, S., Y. Cheung, S. Cueto, P . Glewwe, L. Richter, and B. Strupp, and the International Child Development Steering Group. 2007. “Developmental Potential in the First 5 Years for Children in Developing Countries.” The Lancet 369 (9555) : 60 –70. Heissler, K., and C. Porter. 2013. “Know Your Place: Ethiopian Children’s Contributions to the Household Economy.” European Journal of Development Research 25 (4) : 600 –20. Helmers, C., and M. Patnam. 2011. “The Formation and Evolution of Childhood Skill Acquisition: Evidence from India.” Journal of Development Economics 95 (2): 252 –66. Jacoby, H. G., and E. Skoufias. 1997. “Risk, Financial Markets, and Human Capital in a Developing Country.” The Review of Economic Studies 64 (3) : 311–35. Jensen, R. 2012a. “Do Labor Market Opportunities Affect Young Women’s Work and Family Decisions? Experimental Evidence from India.” The Quarterly Journal of Economics 127 (2) : 753–92. . 2012b. “Another Mouth to Feed? The Effects of (in) Fertility on Malnutrition.” CESifo Economic Studies 58 (2) : 322– 47. Jensen, R., and E. Oster. 2009. “The Power of TV: Cable Television and Women’s Status in India.” The Quarterly Journal of Economics 124 (3) : 1057–94. Kingdon, G. 2005. “Where Has All the Bias Gone? Detecting Gender Bias in the Intrahousehold Allocation of Educational Expenditure.” Economic Development and Cultural Change 53 (2) : 409–51. . 2007. “The Progress of School Education in India.” Oxford Review of Economic Policy 23 (2) : 168–95. La Ferrara, E., A. Chong, and S. Duryea. 2012. “Soap Operas and Fertility: Evidence from Brazil.” American Economic Journal: Applied Economics 4 (4) : 1–31. Lundeen, E. A., J. R. Behrman, B. T. Crookston, K. A. Dearden, P . Engle, A. Georgiadis, and M. E., Penny, et al. 2013. “Growth Faltering and Recovery in Children Aged 1–8 Years in Four Low- and Middle-Income Countries: Young Lives.” Public Health Nutrition Nov. : 1–7. Maccini, S., and D. Yang. 2009. “Under the Weather: Health, Schooling, and Economic Consequences of Early-Life Rainfall.” American Economic Review 99 (3) : 1006–26. Maitra, P., S. Pal, and A. Sharma. 2011. “Reforms, Growth and Persistence of Gender Gap: Recent Evidence from Private School Enrolment in India.” IZA Discussion Paper 6135, Instutute for the Study of Labor, Bonn, Germany. Maluccio, J., J. Hoddinott, J. Behrman, R. Martorell, A. Quisumbing, and A. Stein. 2009. “The Impact of Improving Nutrition During Early Childhood on Education among Guatemalan Adults.” Economic Journal 119 (537) : 734–63. ´ . 2006. “Competence and Resilience in Development.” Annals of the Masten, A. S., and J. Obradovic New York Academy of Sciences 1094 : 13 –27. Miller, G., and B. P. Urdinola. 2010. “Cyclicality, Mortality, and the Value of Time: The Case of Coffee Price Fluctuations and Child Survival in Colombia.” Journal of Political Economy 118 (1) : 113–55. Mookherjee, D., D. Ray, and S. Napel. 2010. “Aspirations, Segregation, and Occupational Choice.” Journal of the European Economic Association 8 (1) : 139– 68. Boyden et al. 217 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 Munshi, K., and M. Rosenzweig. 2006. “Traditional Institutions Meet the Modern World: Caste, Gender, and Schooling Choice in a Globalising Economy.” American Economic Review 96 (4) : 1225–52. Muralidharan, K., and V . Sundararaman. 2011. “Teacher Performance Pay: Experimental Evidence from India.” Journal of Political Economy 119 (1) : 39 – 77. . 2013. “The Aggregate Effect of School Choice: Evidence from a two-stage experiment in India.” NBER Working Paper 19441, National Bureau of Economic Research, Cambridge, MA. ´ , J., and T. W Obradovic . Boyce. 2009. “Individual Difference in Behavioural, Physiological, and Genetic Sensitivities to Contexts: Implications for Development and Adaptation.” Developmental Neuroscience 31 (4) : 300– 8. Olken, B. A. 2009. “Do Television and Radio Destroy Social Capital? Evidence from Indonesian Villages.” American Economic Journal: Applied Economics 1 (4) : 1 –33. Oster, E. 2009. “Does Increased Access Increase Equality? Gender and Child Health Investments in India.” Journal of Development Economics 89 (1) : 62 –76. Pratham. 2012. Annual Survey of Education Report 2012. New Delhi: Pratham. Prentice, A. M., K. A. Ward, G. R. Goldberg, L. M. Jarjou, S. E. Moore, A. J. Fulford, and A. Prentice. 2013. “Critical Windows for Nutritional Interventions against Stunting.” The American Journal of Clinical Nutrition 97 (5) : 911 –8. Ridge, T. 2002. Childhood Poverty and Social Exclusion: From a Child’s Perspective. Bristol: Policy Press. Rogoff, B. 2003. The Cultural Nature of Human Development. Oxford: Oxford University Press. Rosenzweig, M. R., and K. I. Wolpin. 1980. “Testing the Quantity-Quality Fertility Model: the Use of Twins as a Natural Experiment.” Econometrica 48 (1) : 227 –40. Rutter, M. 1987. “Psychosocial Resilience and Protective Mechanisms.” American Journal of Orthopsychiatry 57 (3) : 316 –31. . 2012. “Resilience as a Dynamic Concept.” Development and Psychopathology 24 (2) : 335–44. Sameroff, A. 2009. The Transactional Model of Development: How Children and Contexts Shape Each Other. Washington DC: American Psychological Association. Sameroff, A., and M. J. Chandler. 1975. “Reproductive Risk and the Continuum of Caretaking Casualty.” In F. D. Horowitz, E. M. Hetherington, S. Scarr-Salapatek, and G. M. Siegel, eds., Review of Child Development Research. Chicago: University of Chicago Press. Sameroff, A. J., and B. H. Fiese. 2000. “Transactional Regulation: the Developmental Ecology of Early Intervention.” In J. P . Shonkoff, and S. J. Meisels, eds., Handbook of Early Childhood Intervention 2nd ed. New York: Cambridge University Press. Schaffer, R. 1996. Social Development. Oxford: Blackwell. Schott, W. B., B. T. Crookston, E. A. Lundeen, A. D. Stein, and J. R. Behrman, and the Young Lives Determinants and Consequences of Child Growth Project Team. 2013. “Periods of Child Growth up to Age 8 Years in Ethiopia, India, Peru and Vietnam: Key Distal Household and Community Factors.: Social Science & Medicine 97 : 278–87. Shonkoff, J, W. Boyce, and B. McEwen. 2009. “Neuroscience, Molecular Biology and the Childhood Roots of Health Disparities: Building a New Framework for Health Promotion and Disease Prevention.” Journal of the American Medical Association 301 (21) : 2252–9. Shonkoff, J., and P. Levitt. 2010. “Neuroscience and the Future of Early Childhood Policy: Moving from Why to What and How.” Neuron 67 (5) : 689 –91. Singh, A., A. Park, and S. Dercon. 2014. “School Meals as a Safety Net: An evaluation of the Midday Meal Scheme in India.” Economic Development and Cultural Change 62 (2) : 275– 306. 218 The World Bank Research Observer, vol. 30, no. 2 (August 2015) Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017 Singh, A. 2013. “Size and Sources of the Private School Premium in Test Scores in India.” Young Lives Working Paper 98, University of Oxford. Stevens, G. D. 2006. “Gradients in the Health Status and Developmental Risks of Young Children: The Combined Influences of Multiple Social Risk Factors.” Maternal and Child Health Journal 10 (2) : 187–99. Wachs, T. D., and A. Rahman. 2013. “The Nature and Impact of Risk and Protective Influences on Children’s Development in Low-Income Countries.” In P . R. Britto, P. L. Engle, and C. M. Super, eds., Handbook of Early Childhood Development Research and Its Impact on Global Policy. New York: Oxford University Press. Walker, S. P., T. D. Wachs, S. Grantham-McGregor, M. M. Black, C. A. Nelson, S. L. Huffman, and H. Baker-Henningham, et al. 2011. “Inequality in Early Childhood: Risk and Protective Factors for Early Child Development.” The Lancet 378 (9799) : 1325–38. Weisner, T. 2008. “Well-being, Chaos, and Culture: Sustaining a Meaningful Daily Routine.” In G. W. EvansT. D. Wachs, eds., Chaos and Children’s Development: Levels of Analysis and Mechanisms. Washington, DC: American Psychological Association. Woodhead, M., P. Ames, U. Vennan, W. Abebe, and N. Streuli. 2009. “Equity and Quality? Challenges for Early Childhood and Primary Education in Ethiopia, India and Peru.” Studies in Early Childhood Transitions Working Paper 55, The Hague: Bernard Van Leer Foundation. World, Bank. 2012. World Development Indicators 2012. Washington, DC: World Bank Publications. Boyden et al. 219 Downloaded from https://academic.oup.com/wbro/article-abstract/30/2/193/1652371/Child-Development-in-a-Changing-World-Risks-and by World Bank Publications user on 06 October 2017