The World Bank M a r c h PREMnotes 2 0 0 0 n u m b e r 3 6 Poverty and Gender Interhousehold transfers: using research to inform policy Interhousehold transfers are an important resource for low-income households, yet have been almost ignored by donors. Though they are no substitute for public programs, these private transfers merit more attention in poverty research, gender analysis, project design, and development strategies. Support from social Anthropologists and sociologists have long garia, the Czech Republic, the Kyrgyz networks helps poor recognized that interhousehold transfers of Republic, Poland, Russia, and Ukraine money, goods, and services through social between 21 and 76 percent of households households survive networks are an important survival mech- give or receive income transfers through anism for poor families in developing coun- private networks. In Russia the transfers tries. In addition to helping cover basic needs received by the poorest 5 percent of house- in times of crisis, social networks mobilize holds are equal to more than one-third of resources for education, housing, commu- formal wage earnings--and in Ukraine, nity infrastructure, small business devel- nearly three-fifths. In these countries house- opment, and international travel in search hold-level adjustments help people who can- of employment. not get good formal sector jobs. Moreover, Yet findings on interhousehold trans- high interhousehold transfers may help fers have had little influence on develop- explain why there has been so little social ment policy. Why? One reason is that most protest in Russia despite a large drop in real studies have focused on specific commu- wages. These transfers do not, however, elim- nities, using a qualitative approach, with inate poverty (Barberia, Johnson, and Kauf- little discussion of how the findings could mann 1998). be generalized. As a result many potentially At the community level, one of the most promising applications to development pol- comprehensive studies of transfers icy have gone unexplored. through social networks was conducted in the Southeast Zone of Cartagena, Colom- The importance of bia, in 1982. Almost half the households interhousehold transfers reported having received a transfer in cash Studies in Colombia, El Salvador, Kenya, or in kind in the month prior to the sur- Peru, and the Philippines have found that vey. Among the poorest quintile, transfers more than one in five households receive accounted for 52 percent of household monetary transfers (table 1). Transfers are income for female-headed households and particularly important for the poorest house- 40 percent for male-headed households holds, in some cases representing nearly half (Bamberger, Kaufmann, and Velez 1999). of household income. The study showed that transfers are impor- Recent studies have also documented the tant both as a source of support by bet- importance of interhousehold transfers in ter-off households for their poorer relatives Eastern Europe and Central Asia. In Bul- and friends, and as an informal social from the development economics vice presidency and poverty reduction and economic management network Table 1 Findings on the magnitude and impact of interhousehold transfers Country, year Findings Brazil (northeast), 1988/89 38% of adult women in sampled communities had helped rear a child who was not their own Colombia (Cartagena), 1998 23% of households in the lowest income quintile received transfers and 13% made transfer remittances, compared with 50% and 3% in the highest quintile El Salvador (Santa Ana), 1976 Transfers went to 58% of households in lowest income decile and 48% in next lowest decile, representing 41% of household income of all households in these deciles India, 1975­83 93% of rural households received remittances representing an average 8% of household income Indonesia, 1982 31% of rural households received transfers representing an average 10% of household income Kenya, 1974 The poorest quintile of urban households received an average of 41% of their income in the form of remittances Malaysia, 1977­78 19­30% of households received transfers representing an average 11% of their household income Peru, 1985 22% of urban households gave and 23% received transfers representing an average 2.9% of income for receiving households Philippines (Tondo), 1978 43% of households in poorest third received transfers representing an average 23% of household income Russia (urban workers), 1996 18% of households received transfers representing an average 1.8% of household income Ukraine (urban workers) 1996 17% of households received transfers representing an average 1.7% of household income Source: Mukras, Oucho, and Bamberger 1985; Barberia, Johnson, and Kaufmann 1998; see also the citations in Bamberger, Kaufmann, and Velez 1999 and Cox and Jimenez 1990. insurance system among low-income Anthropological and sociological households. explanations During follow-up field work in 1998 the For almost 50 years social networks have researchers were able to locate most of the been a topic of interest in the ethnographic Cartagenan households on whom case literature on the urban poor. Oscar Lewis's studies had been prepared more than 15 pioneering studies on the "culture of years before. While transfers were still poverty" in Mexico City, Puerto Rico, and important, the portion of households in New York in the 1950s and 1960s found that the poorest quintile receiving monetary poor households regularly provided one transfers had dropped from almost 50 per- another with informal credit services and cent to just 23 percent. The follow-up study shared food, clothes, and furniture. Lack- makes an important point: given the ing access to public services and facing low poverty and vulnerability of most of the and irregular incomes, poor households families comprising support networks, relied on their families, neighbors, and transfers are not a reliable long-term source friends for some assurance that their basic of insurance. needs would be met (Lewis 1961). Many While transfers are often an important later researchers have confirmed the impor- sourceofinsuranceandsupportforthepoor, tance of network support systems in both no research study in any developing coun- developing and industrial nations. try has ever found that more than half of A different form of reciprocity, reported poor households receive them. In fact, the from the Philippines, is utang na loob, or debt share is usually much smaller. Transfers of gratitude. When a household receives are of great importance to the poor, and assistance from another linked to it by kin- we need to understand a lot more about how ship or a close social relationship, it is com- they work. But they must be seen as a com- pelled to return the assistance with interest, plement to--not a substitute for--public to avoid remaining in the donor's perma- transfers. nent debt. The obligation to repay can carry over from one generation to the next. What drives interhousehold transfers? Intergenerational support Researchers have offered a number of expla- Cox and Jimenez (1990), following in the nations for transfers of money, goods, and tradition of an earlier generation of econ- services among households. omists, have shown that intergenerational PREMnote 36 March 2000 transfers act as an informal social security kin and neighbors in a shantytown outside system by which children take care of their Mexico City. Like Oscar Lewis, she found parents in old age or parents provide for that unstable economic conditions led many their children in emergencies. households to develop exchange networks in which they pooled their resources to keep Coping strategies for women the affected households afloat and to max- and female-headed households imize "an important commodity, namely Women, and particularly female-headed security" (Lomnitz 1977, p. 190). The mutual households, increasingly bear the brunt of trust between network households is an economic crises, and it often falls to them important element of these relationships. to avoid the breakup of the household in InspiredbyJohnRawl'sredistributivethe- Women often the face of these pressures. In many coun- ory of social justice and building on the work tries gender relations are changing as a result citedabove,DanielKaufmann(1982)argues bear the brunt of women's increasing labor force partici- thatpoorhouseholds,recognizingthattheir pation and men's inability to fulfill their tra- fluctuatingincomescannotconsistentlycover of economic crises ditional role as principal breadwinner. all their needs, engage in a "contractarian" Because most women have less access interhouseholdexchangeframeworkdriven than men to formal labor and financial mar- by an implicit contract for reciprocal kets and to political networks, they must exchanges over time to help each other in rely more on informal networks to main- times of need. In other words, an informal tain household living standards. Networks insurance mechanism with a network based of poor women and female-headed house- on kin, community relations, or both, whose holds often exchange goods and services purpose is to reduce the risk that any par- that higher-income households can afford ticipating household will fail to have its basic to buy outright. In addition, female entre- needs met over time. The "givers" consider preneurs rely on informal networks for child thetransfersaninvestment,expectingfuture care, protection from the police, and assis- reciprocationiftheirfortunesweretochange. tance in acquiring credit, customers, and raw materials. Policy implications Development policy strives to find cost-effec- Exchange theory models tive and equitable ways to provide social Sahlins (1965) treats transfers between safety nets for vulnerable groups. Although householdsorlargersocialgroupsaspatterns private transfers benefit only a fraction of of exchange, or reciprocity, occurring along poor and vulnerable households, they are a continuum. At one extreme the transfer of sufficient magnitude to merit careful takes the form of a gift with no expectation analysis of whether they could complement of reciprocity. Moving away from altruism, limited public resources. This raises a num- resources are transferred with the under- ber of questions for policy: standing that they will be repaid if and when · What net redistributive effect do public it becomes possible to do so, but without transfers have on health, old age security, anyspecificscheduleorformofrepayment-- and the like after their effects on net pri- as with the resources parents give children. vate flows have been taken into account? Furtheralongthecontinuumaremarket-ori- · Can some services--health, education, ented exchanges in which each party aims to nutritional supplements--be provided maximize its utility from the transaction. At more efficiently through private rather the other extreme is "negative reciprocity," than public transfers? where parties try to extract goods or ser- · Can some groups--ethnic minorities, vices from each other by force. female-household heads--be reached Sahlins influenced a generation of social moreeffectivelythroughprivatetransfers? scientists. Larissa Lomnitz studied patterns · Can informal private networks make pub- of long-term reciprocal exchange among lic sector targeting more effective? PREMnote 36 March 2000 Different issues arise when designing pro- Transfers and Survival Strategies of Low- jectssuchashousing,transport,watersupply, income Households." World Bank, Wash- and urban development to leverage private ington, D.C. transfers.StudiesoftheWorldBank'slow-cost Barberia, Lorena, Simon Johnson, and housing programs have found that people Daniel Kaufmann. 1998. "Social Networks areoftenwillingtoprovidemoney,materials, in Transition" William Davidson Institute and labor to help parents, children, or sib- Working Paper 102. University of Michi- lings capitalize on the investment opportu- gan. nitiesaffordedbyaccesstoservicedplots.The Cox, Donald, and Emmanuel Jimenez. 1990. sameincentivesdevelopwhentransportpro- "Achieving Social Objectives through Pri- jects create opportunities for small entre- vate Transfers: A Review." World Bank We need a better preneurs to enter the passenger or goods Research Observer 5(2): 205­18. transportbusiness.Questiontobeconsidered Kaufmann, Daniel. 1982. "Social Interac- understanding of include: tions As a Strategy of Survival among the · How can the volume and distribution Urban Poor: A Theory and Some Evi- the determinants of the private transfers generated by dif- dence." Ph.D. diss. Harvard University, ferent kinds of projects be estimated? Department of Economics, Cambridge, and uses of · Can transfers be measured with enough Mass. accuracy and reliability to help estimate Lewis, Oscar. 1961. The Children of Sanchez. interhousehold the capacity and willingness of different New York: Random House. groups to pay for goods (housing, roads) Lomnitz, Larissa. 1977. Networks and Mar- transfers and services (water, transport, health, ginality: Life in a Mexican Shantytown. New education)? York: Academic Press. · How can private transfers be leveraged Mukras, Mohammad, John Oucho, and for social or physical infrastructure pro- Michael Bamberger. 1985. "Resource jects such as housing, construction, and Mobilization and the Household Econ- minor irrigation projects? omy in Kenya." Canadian Journal of African Finally, interhousehold networks repre- Studies 19(2). sent a crucial but largely unexplored aspect Sahlins, Marshall. 1965. "On the Sociology of social capital. Willingness to engage in of Primitive Exchange." In M. Banton, exchange networks is based on trust and a ed., The Relevance of Models for Social Anthro- number of dimensions of social solidarity pology. London: Tavistock. that we do not fully understand. A better understanding of the determinants and uses This note was written by Michael Bamberger (Senior of interhousehold transfers, and the ways in Sociologist, Gender Division, PREM Network), which public policy could stimulate these Daniel Kaufmann (Senior Manager, Governance, transfers, could help the Bank strengthen Regulation, and Finance, World Bank Institute), social capital as a building block in the Com- and Eduardo Velez (Sector Leader, Human Devel- prehensive Development Framework. opment, Mexico Management Unit). If you are interested in similar topics, consider Further reading joining the Gender and Development Thematic Bamberger, Michael, Daniel Kaufmann, and Group. Contact Wendy Wakeman (x 33994) or Eduardo Velez. 1999. "Interhousehold click on Thematic Groups on PREMnet. This note series is intended to summarize good practice and key policy find- ings on PREM-related topics. The views expressed in these notes are those of the authors and do not necessarily reflect the views of the World Bank. PREM- notes are distributed widely to Bank staff and are also available on the PREM website (http://prem). If you are interested in writing a PREMnote, email your idea to Sarah Nedolast. For additional copies of this PREMnote please con- tact the PREM Advisory Service at x87736. Prepared for World Bank staff