Can we really count on family through all?

The design of social protection policies should take into account the complexity and diversity of family arrangements. Families have different needs and structures as well as various ways of reacting and adapting to changes in their economic environment.

Not only do the poor have lower incomes, but more importantly, these incomes tend to be highly volatile. Thus, poor households are especially vulnerable to unexpected events. In the event of job loss, illness, or a poor harvest, the main source of social protection they might expect to receive is support from their extended family, through the redistribution of resources and money transfers.

As in many developing societies, nuclear families in Senegal are the exception rather than the rule. On average, Senegalese households have eight members of an extended family living and eating under the same roof. Ignoring this diversity in family structures may actually lead to misguided targeting and implementation of policies aimed at promoting inclusion of the poor.

The Poverty and Family Structure (PFS) project provides specific information about household responses to shocks in terms of migration, income sharing, and consumption in Senegal. The main shocks that households encounter are exceptionally good and poor harvests (and related sales conditions), and job gain or loss. Many families may also have a bad year due to death, illness, or loss of livestock. Eager to understand why certain households in Senegal and elsewhere welcome new members and often change their structure, the project researchers examined the various effects that these shocks can have on household composition.

The household adapts its size in hard times or good times through fostering (when other households take responsibility for children) and through migration (in-migration or out-migration of household members). The re-composition of families (i.e. the return of a member who has lived there previously) is more likely after a positive shock, when households are inclined to host new members (see figure). However, urban and rural areas do not attract the same people. When dealing with a positive shock, urban households are more inclined to host adult men, whereas women and girls tend to settle in rural households.

Migration of family members to cities or overseas is a strategy to diversify income sources. However, remittances from these migrants are not shared equally within the households: those who actually benefit from these transfers tend to be those who are the closest relatives to the migrant. Furthermore, the research findings confirm that Senegalese household income is not automatically pooled. Rather, the income is distributed according to who controls the additional monies sent by migrants. These income inequalities within households become even more pronounced when one examines consumption. Intra-household consumption of non-essential products stands out as more unequally distributed than consumption of essential products (such as food). This result probably explains why there may be rich individuals in poor households and poor individuals in rich households.

Policy design should therefore take into account the differences in households’ responses to shocks. Efforts to reduce inequalities are closely linked to the redistribution of resources within extended families. When shocks affect the entire community, the capacity of families to manage risks, cushion shocks, and cope with job losses or poor harvests becomes seriously diminished or nil. In these cases, families can no longer count on the sole support of relatives. In this kind of crisis, public intervention is warranted to prevent long-term negative consequences such as school drop-outs, malnutrition, or the sale of productive assets.