Fertility and household savings: the case of Chile
Fernando Brito
There is consensus among economists and policy makers that higher saving rates foster domestic investment and economic growth. Therefore, understanding the determinants of savings is a fundamental concern for economic development. Consistent with the life-cycle hypothesis, one of the key determinants of savings is demographics. The decline in fertility and the forthcoming population aging, particularly in emerging economies, have the potential to reshape age structure and affect domestic savings. Thus, exploring the dynamics of demographics plays a central role in the study of the main determinants of household savings.
This study estimates the impact of the quantity of children on household savings. To explore this issue, this paper employs Chilean cross-sectional micro-data from the Household Expenditure Survey (waves 1987, 1997, 2007 and 2012). A main problem in the study of the relationship between the number of children and household savings is endogeneity. The number of children is likely to affect household savings, but the reverse causal effect may also be true. The household data set used in this study not only allows us to control for attributes at households level, but also to deal with potential endogeneity.
Specifically, this paper contributes to the literature on the demographic-savings nexus using an Instrumental Variable approach to avoid potential endogeneity biases. Specifically, this study exploits the fact that sex sibling composition generates an exogenous variation of the household's quantity of children. Consistent with the idea that Chilean parents prefer balanced sex ratios in their family composition, this study cannot reject the null hypothesis that the sex sibling composition of the first two children significantly affects the probability of having a third child.
Another advantage of this instrument is that the sex of a child is randomly determined. Thus, an instrumental variable constructed from the sex sibling composition proves to satisfy both the relevance and exclusion conditions. Studies that do not take into account potential endogeneity problems found little effect (see, e.g., [Harris et al., 2012] or [Gallego and Butelmann, 2001]).
However, once we deal with potential endogeneity, this paper finds that the effect of the number of children on household savings is statistically significant and economically meaningful. This study finds an average effect of -13.98%. This effect is progressive in the sense that it is small for the poor (-8.05%) and large for the rich (-18.29%).
The main conclusion of this paper is that the demographic transition increased average savings rates in Chile. Specifically, while parents (of all socioeconomic segments of population) begun to have less children and postponing parenting they automatically decreased household's consumption and via labor offer they also increased household's income.
Additionally, they had an important precautionary motive for increasing savings rate. Increased savings rates due to fertility trends, however, was mostly observed among the rich. This show us how much can be gained from well run public policies targeted to the poor. Governmental aid that not only provides basic goods and services to reduce marginal propensity to consume, but also that promotes basic financial education to make young people conscious on the consequences of their consumption lifestyles, can have a dramatic positive effect and can help to match up opportunities.

República 701, Santiago, Chile. Teléfono: (+562) 2978 4054 / (+562) 2978 4914

E-Mail: cf@dii.uchile.cl

El Centro de Finanzas agradece el significativo aporte del Banco de Crédito e Inversiones, BCI, a esta iniciativa