Association for Public Policy Analysis & Management

The purpose of this paper is to study the association between noncustodial father’s alimonies and economic well-being in households composed by mothers and children after divorce in Uruguay. We consider a wide set of household and child outcomes such as  income, crowding, access to durable goods, school attendance, repetition and leisure activities.

The aim of the present study is to analyze whether results differ when fathers fail to provide a maintenance allowance. There is scarce literature analyzing this effect in the context of developing countries.

The divorce law was passed early in Uruguay in the Latin American context (1907), though the number of divorces increased only after 1990s. When divorced or separated, couples must agree upon a child support and visits regime. If an agreement is not reached, the court fixes a schedule. Non-compliance with these conditions constitutes grounds for legal action. However, the legal system has little power to enforce agreements and court decisions

The data used in this study were gathered in Estudio Longitudinal del Bienestar in Uruguay. The sample is representative of children attending the first grade of primary public schools in 2004 and their families (85% of the cohort) in urban areas, which account for 87% of the Uruguayan population. The whole data set comprises four waves, and this paper is based on two of them carried out on 2004 and 2011/12.

Panel attrition is around 30%, but an analysis of the distribution of relevant variables indicates that there are not significant biases. We use a sub-sample of around 1300 women who report to live with their children in 2004 and 2011/12.

To control for selection bias, we develop two different identification strategies. We first restrict our analysis to cohabiting couples in 2004 (baseline) and use propensity score matching methods to obtain a comparable treatment group (those couples that were separated in the 2011/12 wave) and a control group (couples that cohabited in 2011/12). We carry out the subsequent analysis using those cases that belong to the common support. On this basis, we estimate a combined PSM-difference in difference model (Heckman et al, 1997; Aasve et al., 20017 and Ongaro et al., 2009) and a separate difference in difference analysis.  

Our preliminary results show that separation does not affect household well-being measured by an assets index and household income, but it decreases when subtracting public transfers. However, mothers’ income and their probability of being employed increases. Alimonies do not have a separate effect in these outcomes.

With regards to children outcomes there is a negative effect of separation on school attendance and repetition. However, alimonies outweigh these effects. We also study non-classroom schooling activities, finding no effect of separation on time spent on reading, computer activities and TV watching. Meanwhile, we find a negative effect on time devoted to sports practicing and study. Again, alimonies have a separate effect that counteracts the reduction on time spent on study.