At the end of August, the Argentinean Congress approved a new law with the objective of increasing pension coverage. The new law is an extension of the pension reform known as the “moratorium” or “housewives reform” adopted at the end of 2004 and put into effect in 2007. In a recent paper¹ we analyze some of the consequences of the moratorium for those cohorts of women who were most affected. A summary of the effects of the moratorium on women’s absolute and relative income within the household can be found in the first part of this Post published last June (here). Today, we will describe and discuss the effects of such a large increase in senior women’s income² on the probability of divorce/separation and on two outcomes that proxy for the bargaining power within the household namely, the probability of women being the head of the household, and the distribution of household chores. We find surprising effects; both union dissolution and outcomes related to bargaining power are affected by public transfers. We believe that the Argentinean reform had significant effects on the situation of senior women in the household because not only the transfers were sizable but most importantly because they were permanent.

The moratorium was introduced by two pieces of legislation, Law 25994 art. 6 and Decree 1454/05, approved in December 2004 and December 2005, respectively. When implemented, the moratorium allowed individuals, male and female alike, who had reached the retirement age by 2004—i.e. cohorts 1944 and older if females and cohorts 1939 and older if males— but did not fulfill the 30-year SS contribution requirement, to get a pension and health insurance coverage.³ Our estimates are obtained by comparing the evolution of indicators before and after the reform for cohorts 1941-1944 to the evolution of the same indicators for younger cohorts who were not affected by the reform.⁴ 

The results presented in Part 1 of this post show that women born in 1944 or earlier had very significant effects in absolute and relative income. For example, married women saw their share in the household's and the couple's income increased by 55% and 50 % respectively (i.e., 10 and 11 percentage points, respectively).

Before the reform, only 7% of the married women belonging to the 1941-1944 cohorts were the head of their households, and 44% of these were the only person in charge of the household chores. Domestic service was virtually inexistent, affecting only 3% of the households, although 13% of the husbands were responsible for the housework.

These large asymmetric income effects were accompanied by a significant increase in the divorce/separation probability of 1.8-2.7 percentage points, representing an increase in divorce/separation rates of roughly 18%-19% for women between 60-65 years old. This relatively large impact is likely to reflect a backlog of women who despite wanting a divorce/separation could not afford it. We believe this effect to be plausible given that the minimum pension paid with the moratorium (191 U.S. dollars PPP 2009 monthly during the first 60 months and 374 U.S. dollars PPP 2009 monthly afterward) is not too different from the average personal income of divorcees from these cohorts before treatment (361 USD PPP 2009). Our large estimates may encompass a short-lived effect comparable to the effects of divorce laws adopted in several countries (González and Viitanen 2009).

For those women who remained married or with their partners, we observe a large, although not significant, increase of 1.4 percentage points (or 20%) on the probability that the wife is the head of household, a statistically significant decrease of 5 percentage points (or 11 %) in the probability that wives are the only ones in charge of the housework, and a statistically significant increase in the probability that husbands do most of the housework of 1.6 percentage points (or 13%).

We extend our model to allow different effects of the reform by educational levels. We find an interesting pattern: highly-educated women tended to opt-out of their marriages by increasing their probability of divorce/separation while the low-educated women opted-in and gained more bargaining power within their marriages.

Our results can be generalized insofar as the increase in income is exogenous and unexpected.⁵ More specifically, our results on the sample of married women can be interpreted more broadly as effects of female income shares within the couple on their bargaining power. They imply, for instance, that a 10 percentage points increase in female's income share within the couple leads to a decrease in women's sole participation in household chores of 4.5 percentage points (or roughly 10%) and an increase in the husbands participation in household chores of 1.5 (or roughly 11%).

1. “The Impact of a Permanent Income Shock on the Situation of Women in the Household: the case of a pension reform in Argentina” CEPR DP 10256 (November, 2014), or a previous version Documento de Trabajo #148 (November, 2013), Universidad del Rosario.

2. We adopt the term “senior women” to refer to women above 60.

3. The moratorium also affected younger cohorts who had harsher conditions to access a pension. Our analysis focuses on the female 1941 to 1944 cohorts.

4. We use a simple differences-and-differences methodology where we compare cohorts 1941-1944 with cohorts 1950-1953 during 2004-2009. We treat 2004-2006 as the pre-treatment years and 2007-2009 as the post-treatment years.

5. The idea here is to use the reform as an instrumental variable for the endogenous female income share.
 

References:

González, Libertad and Viitanen, Tarja K. (2009): "The effect of divorce laws on divorce rates in Europe," European Economic Review, 53 (2), pp 127-138.