The inheritance of human capital
Rethinking how much parents’ investments and endowments influence children’s human capital in low- and middle-income countries
The measure of human capital —the economic value of one’s skills and experience— acknowledges that investments in people’s cognitive and emotional skills, and health and nutrition, increase their productivity. Beyond economic gains, human capital is widely considered critical for many dimensions of individual fulfillment, and for successful societal development in low- and middle-income countries (LMICs). Indeed, the extent and distribution of human capital is considered by many to be at the heart of the development process.
Children’s human capital is of considerable interest, too, because of the positive impacts it has on a child’s future productivity and welfare, and its relevance for overall levels of equality and social mobility. Parental human capital levels and investments in their children are widely understood to play a significant role in children’s human capital, with many studies focusing especially on schooling attainment. But, a key insight from my forthcoming review of current knowledge on human capital, and its transfer to the next generation, is that there are multiple channels through which children develop their human capital that deserve more research.
Knowing what we don’t know
What we know and don’t know about how parental investments in children’s human capital impact inequality, poverty, and social mobility in LMICs is the central question addressed in my new Element. In this book, I argue that human capital has multiple important dimensions, particularly cognitive and socioemotional skills and physical and mental health. But, to understand how parental human capital and parental endowments determine children’s human capital in LMICs (as many existing studies in the field attempt to do) challenges appear.
For example, simple empirical associations between parental and children’s characteristics do not reflect causal relations because of unobserved factors, such as genetics and cultural norms, that can confound estimates. For this reason, I focus on studies that attempt to deal with these estimation issues through unusual data, such as studies of twins, randomized controlled trials, natural experiments, or appropriate estimation techniques, such as the method of instrumental variables.
These studies provide an important counterweight to studies that measure the impact of parents by using simple associations. In some cases, the estimated effects are considerably larger, once measurement error and nonlinearities are controlled for. In a much larger number of cases, the estimated effects are much smaller.
For example, the simple associations between parental schooling and child schooling are not likely to capture causality, because of confounding factors such as genetics, family resources (reputation, social networks, or the contributions of grandparents), and family culture. Therefore, although parents with high schooling attainment are likely to have children with high schooling attainment, it would be misleading to infer that this is entirely the causal effect of parental schooling attainment. Moreover, unobserved parental endowments are often found to have substantially large effects in the limited number of studies that explore their impacts.
When simple data analysis misleads
One implication is that the literature to date is likely to overstate the influence of characteristics such as the schooling attainment of parents. Therefore, policies directed towards increasing parental human capital are not likely to be as effective as many studies suggest, since unobserved endowments are unlikely to be affected by these policy interventions.
the key thing to understand is that efforts to improve the transfer of human capital from parents to children... may be less successful at achieving their aims than the literature, overall, implies
For policymakers seeking to enhance future development, equality, or social mobility, the key thing to understand is that efforts to improve the transfer of human capital from parents to children (such as targeted parental training programmes, social transfers or tax credits for households with dependents, access to social workers or counselling services for parents, etc.) may be less successful at achieving their aims than the literature, overall, implies. By comparison, policy efforts that aim universally to raise the human capital of society’s children —regardless of the a priori parental attainment of human capital— (such as universal early childhood education, nutritional programs, broad provision of pre- and post-natal care, etc.) may be more successful.
The related point is that in societies with highly unequal levels of individual human capital attainment, when studies find a weaker link between the current and the future generation, this will be better for future mobility, as this means that children from disadvantaged backgrounds are more likely to achieve higher human capital levels than their parent’s characteristics imply.
What we still need to find out
The rigorous examination of the current literature and the several policy case studies in my new book leads us towards a much more cautious assessment of the effects of parental human capital on children’s human capital than is conventional. There are also other important gaps in the field that need to be addressed.
Firstly, there is too much heterogeneity in contexts —market development, policies, culture, demography, and resources— to generalize to other contexts from the relatively few available studies that plausibly estimate causal relations. Further research is needed to gain additional insight in a broad range of contexts.
Secondly, data are very limited, particularly on important cognitive and socioemotional skills and health outcomes, including mental health, and over long enough spans of the children’s lives to be able to assess impacts when the children become adults. Therefore, new data are essential for solidifying knowledge on the impacts of parental human capital on their children’s human capital in LMICs and to inform related policies.
In the meantime, this book does more than provide an analysis of the current research, it encourages policymakers and others to be more critical and to question the reported effects of parental investments on children’s human capital. While well intentioned, the impacts of policies designed to improve children’s human capital, and intergenerational mobility, are often not being properly assessed.
It is important that researchers and policymakers go forward with a broader idea of what human capital is and with greater caution in interpreting associations as due entirely to causal relations. This will help to drive us towards filling the critical gaps in research and data, and put us on a path towards more effectively building the human capital of future generations.
Jere R. Behrman is Professor of Economics at University of Pennsylvania and is a leading international researcher in empirical microeconomics, with emphasis on developing economies.
The views expressed in this piece are those of the author(s), and do not necessarily reflect the views of the Institute or the United Nations University, nor the programme/project donors.