Share

BREAD Working Paper No. 281, June. Matthias Doepke and Michèle Tertilt (2010).

This study suggests that though transfer payments to women often increase household spending on children, they may have unintended consequences in some contexts.

Much development policy is based on the finding that money held by women benefits children.  Studies show that transfer payments allotted to married women rather than to their husbands, expenditures on child schooling, clothing, and nutrition increase greatly.  For example, women in Bangladesh use loans to boost household consumption more so than men.  Female-headed households in Kenya, Malawi, and South Africa spend more on food and less on alcohol.  In South African households with a woman receiving an old-age pension, girls have better weight for height and height for age.  No such relationship was found in households with a man with a pension or without pensioners.

Many microfinance institutions have capitalized on this fact and offer loans only to women.  However, Doepke and Tertilt argue that “understanding why women devote more resources to children is extremely important” (1).  Studies have generally not addressed this question, and the answer may affect policy outcomes in different contexts.  This study explores the question of why women spend more on children than men, and in particular develops “a theoretical framework to explore what kind of mechanisms will lead to an asymmetry in child investment by gender” (1). 

The authors explore several economic models of household decision making and bargaining between husbands and wives.  They allow men and women to have the different preferences versus the same preferences with constraints.  The latter model shows that “a gender wage gap can lead women to specialize in home production and therefore act like they have a higher weight on children relative to their husbands” (2).  Additionally, “gender differences in investment opportunities can lead women to act like they value children relatively more” (2). 

This series of models brings several things to light.  Firstly, the authors suggest that some of the existing evidence on the benefits of transfer payments to women should be reinterpreted.  Depending on the constraints and preferences faced by the household, the welfare impact of such payments can be surprisingly uncertain.  For example, “If women spend less on children because they spend less on investment goods, then it is not obvious that giving more money to women is a desirable policy” (3). 

Secondly, development practitioners should be aware that the observed beneficial impacts of lending to women may decrease as gender equality progresses.  Additionally, increased female participation in the labor force makes differences in endogenous preferences between men and women decrease, as well as preference for sons.  Overall, as gender equality progresses and women have less comparative advantage in raising children, their tendency to favour children in their investment decisions will decrease relative to other investment options.  The authors explain that “If women have lower wages and hence their time is less valuable, they will endogenously specialize in providing the public goods that are most time intensive, even though they don’t care about these goods any more than their husbands do.  …If children are relatively time intensive… it may appear as though women care more about child goods, even though in fact they do not.  Once again, we get the result if underlying gender differences were removed (in wages and transfer income) the observed difference in behaviour would also disappear” (31-32).  Insofar as women have a comparative advantage in child rearing due to their lower income, as their income increases, they will specialize in providing child goods less.

Overall, the authors suggest that existing models of family economics (such as the unitary and collective models) do not explain the data well.  They write that “more measurement and empirical work is needed to distinguish between the various theoretical models that can explain the observed empirical pattern.  …Only once we have some confidence in which of these models provides the best guide to reality will we be in a position to provide credible policy recommendations for gender-based development initiatives” (3). 

The authors suggest two policy recommendations beyond additional empirical work on the models proposed above.  First, they caution that even though transfer payments to women increase spending on children, this may come “at the expense of other important public goods” (50).  For example, increased spending may draw from savings and therefore future consumption.  The authors conclude that “In such settings, it is far from obvious whether targeting transfers to women is good policy” (51).  Second, they note that gender spending differences may be a result of comparative advantage that would disappear as gender equality progressed.  Therefore, “while targeting transfers to women may increase spending on children, reducing gender discrimination in goods or labor markets may result in women behaving more like men, which reduces the effect of targeted transfers on public good provision” (51).