Marital fertility and exogenous contraints on child quality: a theoretical and empirical approach

This work aims to study the effect of exogenous constraints on child quality on marital completed fertility, both from a theoretical and empirical point of view. The neoclassical theory of fertility, pioneered by Becker et al., considers children in the same way as other “durable goods”: the spouses decide become parents, not only because they get utility from the number of children they bear (quantity), but also from some “desirable” child characteristics (quality). The interaction between these two dimensions gives origin to the well-known quantity/quality trade-off in the demand for children. However, it is obvious that parents do not “own” their children and cannot dispose of their offspring at their will. For instance, legislations on compulsory education and regulating the minimum age to admission to work are almost universally widespread. Furthermore, the expectations of their group of peers may be perceived by the spouses as binding as the law itself. These laws and expectations represent a binding constraint on child quality that parent are compelled to take into account in their childbearing decisions. The first chapter proposes a theoretical approach that generalizes the classic quantity/quality trade-off model by introducing the hypothesis that couples face an exogenously determined quality constraint and taking explicitly into account that remaining childless can be optimal. The childbearing decision consists therefore in a two-stages process: in the first step, the spouses evaluate whether they can be better-off with or without children; in the second one, provided that they decide to bear children, they decide their optimal number and quality. Given the exogenous minimum quality threshold, the couple maximizes then a piecewise utility function under a budget constraint. As long as “minimum quality” (i.e. those characteristics “decided” by an entity external to the couple) and “discretionary quality” (i.e. those child characteristics that parents desire) are no perfect substitutes, the existence of an exogenous minimum quality threshold reduces both the space where becoming parents is the optimal choice and the overall number of children born to the couple. Furthermore, the quantity/quality trade-off is significantly stronger than in the standard case. The second chapter proposes a reduced form estimation of the effect of changes in the marginal expenditures on minimum quality on the total number of children born to the couple. In fact, changes in marginal expenditures take place more often than changes in the threshold only and are easier to locate. A suitable econometric counterpart of the piecewise utility function is, in the case under examination, the complementary log-log – Poisson hurdle model. The cross-sectional dataset used in this work is built from several sweeps of the British National Child Development Study (NCDS) , while the marginal expenditures on minimum quality are proxied by the pupil/teacher ratio in primary schools observed by the partners at the time they got married. There is strong evidence that high marginal expenditures have a strongly negative effect both on the probability of becoming parents and on the overall number of children born to the couple. In turn, the overall effect of the observed income is positive but of small magnitude: in fact, a higher income reduces the predicted probability of becoming parents for nine couples out of ten (i.e. they “surrender to the suggestion” of higher own-consumption). These two results suggests therefore new insights into the interpretation of the low fertility trends experienced in most developed countries.