I. The concept of dynamics in human capital production tries to derive the proxy variables that affect both cognitive and non-cognitive skills after childhood together with self-productivity & cross-productivity. It is a dynamic factor model made to solve the problems within the endogeneity of inputs over the lifetime, along with a multifactorial approach to determine positive outcomes in later time periods of life (Cunha & Heckman, 2008). This concept has been quantified by economists estimating a linear dynamic factor model to identify general classes of technologies (such as skilled-based formation) recalled on these endogenous inputs. Given that there are numerous inputs that affect outcomes in childhood, economists have had to overcome this issue of multiplicity by creating indices of inputs (to create more manageable dimensions) and by anchoring scales (such as adult earnings or test scores) (Cunha, Heckman & Schennach, 2010). Such studies have found that it is more optimal to invest in children in the beginning of their lives rather than later. These effects have been found to be salient for cognitive skills (such as test scores, reading ability or parental cognitive skills) whereas for non-cognitive skills, investments in human capital are effective throughout longer periods of life.
A study done by Almond (2018) has proposed many policy interactions that provide naturally occurring experiments to substantiate economic hypotheses. A clear example is the introduction of policies made to increase paternity around birth to increase human capital of the child through marriage. Externalities of these policies included negative effects on children’s health due to easier access to the utilization of child-support rather than employer health insurance (Rossin- Slater, 2016). Within human capital production it is important to note that there are major differences when applying policies and programs for different lifetime periods: efficacy cannot be generalized throughout childhood and adolescence. There is in majority a focus on prenatal and early years of life, but policies can affect various periods of life in different ways.
II. In this next section I will be exploring possible interactions between the policy discussed by Rossin-Slater (2016) and welfare reform which was rolled out during the same period. The introduction of in-hospital voluntary paternity establishment (IHVPE) in 1993 for unmarried parents in all of the USA, overruled the long and tedious process of new parents to establish paternity through the courts and DNA testing. Given a third of children are born out of wedlock, this policy was aimed to encourage marriage which has been proven to provide more positive financial outcomes for the family (Rector, 2010). Unfortunately, in this study it was found that while IHVPE does indeed increase paternity establishment rates, it does not affect marriage rates and may even reduce such rates. IHVPE has no benefits on human capital as child household resources and health do not ameliorate or respond to such policy.
In 1996 the Personal Responsibility and Work Opportunity Reconciliation Act was one of the many reforms made in the decade to change welfare programs. This act had the goals of ending dependence on government for needy parents, aiding families caring for children by promoting job preparation, work and marriage as well as preventing and reducing out of wedlock pregnancies and encouraging two-parent families (Bitler et al., 2004). It has been found that this policy gave women greater financial independence, actually decreasing marriage incentives, leading to fewer new marriages as well as divorces. After controlling for economic forces, the 1996 reforms had little impact on work behavior since they were not well-distributed across the less skilled women. Waivers granted in the early 1990s increased income unlike the 1996 reforms. The proximity of these reforms to IHVPE may have created interactions between the two, as well as the externalities such as reducing the amount of two-parent family structures (Bitler et al., 2006).
Rossin-Slater analyses in their paper the possible interaction of IHVPE and the welfare reforms of the 1990s. They disconfirmed the previous studies by Bitler (2004; 2006), showing that welfare reform did not affect marriage rates in their sample. This suggest a higher endogeneity of IHVPE when compared to welfare reform and human capital production.
III. Given that Rossin-Slater and Bitler used two different samples, it would be interesting to compare both samples, both as a cohort and independently to a new random sample. Stratifying such samples across various states and cities according to when we believe both IVHPE and welfare reform were implemented could be helpful in finding new interactions. This has only been done by Rossin-Slater and stratifying Bitler’s sample in this manner as well could be helpful. Creating vectors for various cognitive and non-cognitive outcomes of children could bring us a step further into understanding the role of marriage as an institution as well as the role of non-paternal fathers. This data could be achieved by finding a model sample from the original samples of Rossin-Slater and Bitler, by following up on the families 30 years later. A longitudinal study even if with a sample size leaning on the smaller side, due to the difficulties of reaching out to the participants could still show sizeable effects of the policies aforementioned.
Secondly, Rossin-Slater and Bitler use a linear model. Considering that there are increasing complexities when doing multiple hypothesis testing, using a dynamic factor model as proposed by Cunha (2010), all the data from different states and applications can be left to pick the best combinations of family dynamics in response to these policies. When using a production function approach, the endogeneity of inputs must be accounted for, as well as the wealth of each person. Given that there are some gaps in the data, since not all dates of the beginning of these programs are known for different geographic areas, assuming that the data is missing randomly the sample likelihood can be used instead. This is an endeavor that through online interactions may have very minimal costs to do, linking adults to some of their past exposures to policy through a growing accessibility of “administrative data”.
It would be also interesting to add data regarding child-support establishment and collection. It has been shown that policies that increase child support, such as the ones of the 1980s increase family financial stability and child outcomes (Nepomnyaschy, 2007). Producing policies that create contracts distributing parenting responsibilities, rather than paternal or spousal agreements, may be the key to increasing marriage and therefore family financial capacities.
References
Cunha, F., & Heckman, J. J. (2008). Formulating, identifying and estimating the technology of cognitive and noncognitive skill formation. Journal of human resources, 43(4), 738-782. Cunha, F., Heckman, J. J., & Schennach, S. M. (2010). Estimating the technology of cognitive and noncognitive skill formation. Econometrica, 78(3), 883-931.
Almond, D., Currie, J., & Duque, V. (2018). Childhood circumstances and adult outcomes: Act
II. Journal of Economic Literature, 56(4), 1360-1446.
Rossin-Slater, M. (2016) Signing Up New Fathers: Do Paternity Establishment Initiatives Increase Marriage, Parental Investment, and Child Well-Being? Forthcoming at the American Economic Journal: Applied Economics.
Bitler, Marianne P., Jonah B. Gelbach, Hilary W. Hoynes, and Madeline Zavodny. 2004. “The Impact of Welfare Reform on Marriage and Divorce.” demography 41 (2): 213–36. Bitler, Marianne P., Jonah B. Gelbach, and Hilary W. Hoynes. 2006. “Welfare Reform and Children’sLiving Arrangements.” Journal of Human resources 41 (1): 1–27. Nepomnyaschy, L. (2007). Child support and father-child contact: Testing reciprocal pathways. Demography, 44(1), 93-112.
Rector, Robert. 2010. “Marriage: America’s Greatest Weapon Against Child Poverty.” Heritage Foundation Report 2465.
Sofia Wolfson
Emory University
University of Miami