34967 The World Bank PREMnotes O C T O B E R 2 0 0 5 N U M B E R 1 0 4 G E N D E R Gender and the impact of credit and transfers Ignoring gender in the planning and evaluation of credit and transfer programs can lead to erroneous conclusions about who benefits from them. Access to institutional credit and targeted The added gain from granting transfers can be an important mechanism in transfers and credit to women poverty reduction, social protection, and Improved access to credit and transfers (in- income redistribution programs. These for- kind and cash) has important benefits for mal sources of financing, however, may poor families. Credit programs can increase undermine traditional sources of support, productivity and earnings from self-employ- such as inter-household transfers and infor- ment. By providing additional resources in mal credit from neighbors. The likelihood times of stress, in-kind and cash transfers and the consequences of this happening help poor households overcome short-term depend in part on whether institutional poverty, manage risks, and invest in more pro- transfers and credit target men or women, ductive but riskier assets (Ravallion 2003). whether men and women have access to the Credit and transfers have important same sources of financial support, and additional effects when granted to women. whether using institutional credit or trans- Credit and Program evaluations in countries as diverse fers challenges conformance to traditional transfers in as Bangladesh, South Africa, and the United gender roles. Kingdom have found that credit and trans- the hands of The gender dimensions of public trans- fers in the hands of women result in greater fers and credit discussed in this note have women result spending on children's human capital, im- the following policy implications: proved health and nutrition for both boys in greater · When deciding to whom to target pub- and girls, and increased access to formal lic transfers or credit, placing resources spending on health care for women. Studies from around in the hands of women may benefit children's the world of the relationship between the households more than targeting men. share of resources controlled by women and human · Depending on the current sources of fi- the proportion of household resources go- nancial support for men and women, capital ing to family welfare, particularly to expen- institutional credit and public transfers ditures on children, have come to similar may have to compensate them differ- conclusions. Empirical evidence in Brazil, ently for the potential loss of informal Côte d'Ivoire, Ethiopia, Indonesia, and financial support. South Africa indicates that this effect is in- · Evaluation of public transfer and credit dependent of the source of women's re- programs should assess the differential sources (World Bank 2001). impact that credit and public transfers In addition, female borrowing has a have on men and women via any sub- greater impact than male borrowing on stitution effects on informal sources of household's ability to smooth consumption financing; the impact on household over time (World Bank 2001). In countries welfare via changes in the intra-house- such as Bangladesh, where social norms and hold distribution of resources; and the household work may prevent women from impact on those who send informal looking for paid employment outside the transfers and credit to the household. FROM THE DEVELOPMENT ECONOMICS VICE PRESIDENCY AND POVERTY REDUCTION AND ECONOMIC MANAGEMENT NETWORK household, microcredit allows them to be According to Becker (1974), when infor- self-employed at home (Khandker 1998). mal transfers are based on altruism, chari- Sources of credit and transfers table persons increase the amount they give Individuals can receive credit, monetary when their income increases, and decrease transfers, and in-kind transfers from formal the amount they give when the income of institutions such as banks, microcredit or- their beneficiary increases. When informal ganizations, and governmental programs or transfers are not motivated by altruism but from informal sources such as friends, fam- by exchange (e.g., in return for childcare) or ily members, moneylenders, and shopkeep- by self-interest (e.g., in return for inheritance), ers. How much individuals receive from informal transfers need not necessarily de- each of these sources varies, among other crease with an increase in the beneficiaries' things, according to their sex and the com- income (Cox 1987). munity in which they reside. The literature considers other reasons to In countries like Malawi and Bangladesh, expect a correlation between credit and trans- where microcredit organizations have a large fers from formal and informal sources. How- presence, women are more likely than men to ever, the predicted direction of the correlation receive loans from formal sources. However, is ambiguous. Some authors argue that subsi- loans to women are generally smaller than dized credit programs such as microcredit and loans to men (McKernan, Pitt, and Moskowitz government lending may worsen the terms 2004; Diagne and Zeller 2003). and availability of loans offered by money- Evidence about whether men or women lenders and thus decrease lending from infor- receive more credit and transfers from in- mal sources (Hoff and Stiglitz 1998). Others, formal sources is mixed. In some cases, these like Jain (1999), argue that there may be a posi- transactions are more likely to be directed tive correlation because formal institutions towards women than towards men. In Peru may condition their lending on cofinancing and Mexico, for example, female-headed from informal sources. households have a higher probability than Empirical evidence on the male-headed households of receiving trans- relationship between formal and fers from non-cohabiting children (Cox and informal transfers and credit Jimenez 1992; Teurel and Davis 2000). In other cases, for example Malawi, loans from The empirical evidence to date suggests that: family members, friends, and moneylend- 1) few households borrow or get transfers ers are equally distributed between men and from both formal and informal sources; 2) women (Diagne and Zeller 2003). In still formal credit and transfers diminish, but do other cases, men are more likely than not completely eliminate, the informal trans- women to receive credit or transfers from fers and credit that flow into a household; informal sources. In rural Bangladesh, for and 3) the effect that formal credit and trans- instance, men receive 70 percent of all credit fers have on informal credit and transfers and transfers from family members, friends, received and sent can depend on the gen- neighbors, and moneylenders and send 95 der of the borrower. percent of all interhousehold transfers Evidence from India, Nepal, and (McKernan, Pitt, and Moskowitz 2004). Bangladesh suggests that few households borrow or get transfers from both formal Theoretical relationship between and informal sources (Kochar 1997; Besley, formal and informal transfers and Jain, and Tsangarides 2001; McKernan, Pitt, credit and Moskowitz 2004). Because men and Ideally, transfers from formal institutions women do not have the same access to for- should increase the income of a household mal and informal credit and transfers, the or individual by the same amount that is be- simultaneous use of formal and informal ing transferred. However, the effect of for- sources is even less frequent when one looks mal transfers may be less than intended be- at transactions separately by gender. In rural cause these transactions can result in a de- Bangladesh, 49 percent of households received crease in the amount that households or in- some type of formal or informal credit or trans- dividuals receive from informal sources. fer during 1998, but only 6 percent received This hypothesis is known as "crowding out" credit or transfers from both sources. Women (Becker 1974). received some kind of formal or informal PREMNOTE OCTOBER 2005 credit or transfers in 30 percent of the house- extent for women and men and for differ- holds, but in only 2 percent of the households ent reasons. A 100 Taka increase in did women receive transfers or credit from microcredit granted to men causes a 31 Taka both sources. Similarly, men received some decrease in the household's net flow of in- kind of formal or informal credit or transfers formal transfers and credit. This change is in 24 percent of the households, but in only caused by an increase in the amount that the 2 percent of the households did men receive men in the household send to relatives as transfers or credit from both sources gifts and loans. A 100 Taka increase in (McKernan, Pitt, and Moskowitz 2004). microcredit granted to women causes a Why households rarely rely on both for- somewhat smaller decrease in the mal and informal sources for credit or trans- household's net flow of informal transfers fers is unclear. One possibility is that formal and credit (25 Taka), but the change is caused sources reach individuals who are unable to by a reduction in the gifts and loans that the access money from informal sources. An- household receives from relatives rather other possibility is that microcredit organi- than by an increase in the gifts that the zations and other formal sources may be household sends to other households. The crowding out the informal sector, replacing authors suggest, based on conversations informal with formal credit and transfers. with Grameen Bank officials and the nature The few causal analyses that examine of traditional gender roles in Bangladesh, whether formal sources crowd out informal that the loss of gifts to households in which sources do find some evidence of crowding women receive microcredit results from a out. However, in none of these studies does perceived decrease in the economic need of an increase in credit and transfers from for- the household and a backlash against mal institutions completely substitute for women who gain economic independence. Gender plays a credit and transfers from informal sources. The extension of the Old Age Pension role in whether For instance, Rosenzweig and Wolpin (1994) Program (OAP) in South Africa is a case found that in the United States, increases in where the gender of the beneficiary does not formal transfers welfare benefits only marginally reduced the make any difference in how much and credit proportion of young women who receive interhousehold transfers change in response parental support in the form of shared hous- to institutional transfers. Jensen (2003) finds crowd out infor- ing and cash transfers. Cox and Jimenez that a one-rand increase in both male and fe- mal transfers (1995) ran simulations using data from the male income from old age public pensions and credit Philippines and found that an increase in leads to a 0.25 to 0.30 rand reduction in trans- unemployment insurance would reduce, but fers from children living away from home. not eliminate, the amount that households That the OAP is a society-wide program that receive through interhousehold transfers. covers everyone who is not covered under Gender plays a role in whether formal occupational pensions may contribute to the transfers and credit crowd out informal program having the same effect on men and transfers and credit. However, the literature women. So, too, may less differentiated gen- finds inconsistent conditioning of the cor- der roles than in countries like Bangladesh. relation between formal and informal trans- Policy implications fers by sex. In some instances, the crowding Evidence suggests that credit and transfers out is greater for females than for males, but in the hands of women have additional house- in others, the crowding out is similar for hold benefits--such as improved health and both sexes. The extent of crowding out ap- nutrition for both boys and girls--over credit pears to depend on, among other things, the and transfers in the hands of men. Conse- social norms governing the economic re- quently, it is important for programs to tar- sponsibilities that adult children have for get women, especially in countries such as their parents and whether engaging in in- Bangladesh, where few women receive gifts stitutional credit or transfers challenges con- and loans from informal sources. formance to society's gender norms. Evidence also suggests that formal credit In their study of rural Bangladesh, and transfer programs reduce net credit and McKernan, Pitt, and Moskowitz (2004) transfers from informal sources, and that this found that formal microcredit provided to effect may differ for men and women. It is men and women reduces net informal trans- important to keep these differences in mind fers into the household, but to a different when deciding which household member to PREMNOTE OCTOBER 2005 target as well as the amount to provide as a Cox, D. 1987. "Motives for Private Income Trans- credit or transfer. For example, the amount fers." Journal of Political Economy 95(3): 508­43. lent or transferred to women from institutional Cox, D., and E. Jimenez. 1992. "Social Security sources may have to be larger in countries with and Private Transfers in Developing Coun- traditional gender roles where women receive tries: The Case of Peru." The World Bank Eco- nomic Review 6(1): 155­69. less informal support and face greater degrees ______. 1995. "Private Transfers and the Effective- of crowding out than do men. ness of Public Income Redistribution in the Phil- Finally, this note points to three impor- ippines." In Public Spending and the Poor: Theory tant implications for program evaluation. and Evidence, ed. Dominique Van De Walle and First, it is critical for evaluations to assess Kimberly Nead, 321­46. Baltimore: Johns the differential impact of credit and trans- Hopkins University Press. fer programs on men and women's access Diagne, A., and M. Zeller. 2003. "Access to Credit to informal credit and transfers. Second, and Its Impact on Welfare in Malawi." IFPRI credit and transfer programs are likely to Research Report 116. Hoff, K., and J. Stiglitz. 1998. "Moneylenders and change the intra-household distribution of Bankers: Price-Increasing Subsidies in a Mo- resources both directly and indirectly nopolistically Competitive Market." Journal of (through changes in informal sources of fi- Development Economics 55: 485­518. nancing.) Program evaluations need to ex- Jain, S. 1999. "Symbiosis vs. Crowding Out: The plore the impact of these changes fully since Interaction of Formal and Informal Credit Markets resources in the hands of women tend to ben- in Developing Countries." Journal of Development efit households more than in the hands of men. Economics.59(2):419­44. Third, some of the benefits of credit and trans- Jensen, R. T. 2003. "Do Private Transfers `Dis- fer programs will be shared with third par- place' the Benefits of Public Transfers? Evi- dence from South Africa." Journal of Public ties--those who send and receive informal Economics, 88: 89­112. transfers and credit to the household. It is likely Khandker, S. R. 1998. Fighting Poverty with that many of these third party impacts are Microcredit: Experience in Bangladesh. New York: positive. Formal programs will improve the Oxford University Press for the World Bank. welfare not just of program participants, but Kochar, A. 1997. "An Empirical Investigation of of related nonparticipants. However, men and Rationing Constraints in Rural Credit Markets women often have different probabilities of in India." Journal of Development Economics sending informal credit and transfers, and 53(2):339­71. thus of benefiting through third party ef- McKernan, S., M. Pitt, and D. Moskowitz. 2004. "Use of Formal and Informal Financial Sec- fects. Sex-disaggregated measures of these tors in Bangladesh: Does Gender Matter?" indirect benefits are needed to assess the full World Bank Working Paper Series 0-3045. impacts of credit and transfer programs. Ravallion, M. 2003. "Targeted Transfers in Poor This note was written by M. Estela Rivero- Countries: Revisiting the Trade-offs and Policy Fuentes (Extended Term Consultant, PRMGE) Options." World Bank Policy Research Work- and Signe-Mary McKernan (The Urban Insti- ing Paper Series 3048. tute) under the supervision of Karen Mason and Rosenzweig, M. and K. Wolpin. "Parental and Nayantara Mukerji. Public Transfers to Young Women and Their Children." American Economic Review. 84 (5): Further reading 1195­1212. Becker, G. S. 1974. "A Theory of Social Interac- Teruel, G. and B. Davis. 2000. "An Evaluation of tions." Journal of Political Economy 82: 1063­93. the Impact of PROGRESA Cash Payments on Besley, T. J., S. Jain, and C. Tsangarides. 2001. Private Inter-Household Transfers." IFPRI "Household Participation in Formal and In- Research Report. formal Institutions in Rural Credit Markets in World Bank. 2001. Engendering Development Developing Countries: Evidence from Nepal." Through Gender Equality in Rights, Resources and Background paper prepared for World Devel- Voice. Washington DC: The World Bank and opment Report 2001/2002: Institutions for Mar- Oxford University Press. kets, Washington, DC. This note series is intended to summarize good practices and key policy findings on PREM- related topics. The views expressed in the notes are those of the authors and do not neces- sarily reflect those of the World Bank. PREMnotes are widely distributed to Bank staff and are also available on the PREM Web site (http://prem). If you are interested in writing a PREMnote, email your idea to Madjiguene Seck. For additional copies of this PREMnote please contact the PREM Advisory Service at x87736. PREMnotes are edited and laid out by Grammarians, Inc. Prepared for World Bank staff