88169 FOCUS NOTE Financial Inclusion and Development: Recent Impact Evidence G lobal and national-level policy makers have been embracing financial inclusion as an important development priority. The G20 made the topic one of for economic and social progress on the development agenda. its pillars at the 2009 Pittsburgh Summit (G20 2009). 1. A Vast Majority of Poor By fall 2013, more than 50 national-level policy-making Households Live and Work and regulatory bodies had publicly committed to in the Informal Economy financial inclusion strategies for their countries (World Bank 2013a, AFI 2013). And the World Bank Group in Traditional economic theory distinguishes between October 2013 postulated the global goal of universal the objectives and needs of individual households and access to basic transaction services as an important firms. Individuals are selling their labor power in the milestone toward full financial inclusion—a world market and strive to smooth life-cycle consumption. where everyone has access and can use the financial When people are young, they need to invest; at the services he or she needs to capture opportunities and prime of their earnings power, they save; and in old- reduce vulnerability (World Bank 2013b). age, they dis-save. In the aggregate, the household- sector saves. Firms, on the other hand, compete Policy makers have articulated these objectives in for investable funds to finance their operations the conviction that financial inclusion can help poor and growth. In the aggregate, firms are net users households improve their lives and spur economic of savings. Financial markets are supposed to make activity. But what is the evidence for this type of the match between savers and users and to allocate positive impact? This Focus Note takes impact to capital toward the highest productive usage (e.g., mean those effects that can be traced to a specific Mankiw and Ball 2011). intervention and that would not have occurred otherwise, thus analysis at the micro and local But the poor are typically excluded from the wage- economic levels focuses primarily on the relatively earning employment opportunities that traditional new evidence from randomized control trials (RCTs) economic theory presupposes. They live and work or quasi-randomized impact evaluations. At the in the informal economy—not by choice, but by macroeconomic level it highlights studies using necessity. In economic terms, they are consuming country panel data comparisons. households and self-employed firms at the same time; thus consumption and production decisions are This Focus Note is organized in three sections. The intertwined. As a result, they need a broad range of first section describes the extent to which poor financial services to create and sustain livelihoods, households typically live and work in the informal build assets, manage risks, and smooth consumption. No. 92 economy and explores the implications of this for The traditional distinction between consumer April 2014 how access and use of financial services can benefit financial needs versus the financial needs of firms is them. The second section summarizes recent often blurred. Robert Cull, empirical impact evidence at the microeconomic, Tilman Ehrbeck, and local economy, and macroeconomic levels. The Empirically, financial diaries literature has illustrated Nina Holle third section tees up two areas in which inclusive, this point by showing how poor families in the low-cost financial systems can generate additional, informal economies of developing countries actively indirect benefits for other public-sector and private- manage their financial lives to achieve these multiple sector efforts. objectives (Collins, Murdoch, Rutherford, and Ruthven 2009). They save and borrow constantly In summary, the accumulating body of evidence in informal ways. At any given time, the average supports policy makers’ assessments that developing poor household has a large number of ongoing inclusive financial systems is an important component financial relationships. Financial management is, for 2 the poor, a fundamental and well-understood part A. Microeconomic Level of everyday life. To assess whether any intervention works, the most Estimates of the share of the world population living rigorous method is to ask the counter-factual: what and working in the informal economy vary between would have happened without it. An increasingly 50 percent and 60 percent (World Bank 2012). The influential group of development economists Gallup World Survey 2012 reports that only about argues that the most adequate tool in empirical 40 percent of adults globally have fixed employment microeconomics is the use of randomized evaluations. in excess of 30 hours per week. These are averages across all countries and income groups. The share of This methodology uses an approach similar to clinical informality is considerably higher for poorer countries trials where access to a specific new drug is randomly and poorer income segments and can reach well over assigned, and the impact of a change in access on 80 percent or even 90 percent in some developing a group is then compared to a second group that countries (ILO 2013). does not have the same access but is otherwise indistinguishable.1 While other methodologies are The share of informal employment is mirrored in the equally important in understanding how financial estimates for financial access. Globally, about half inclusion affects the lives of the poor, this section of all working-age adults are excluded from formal of the Focus Note highlights experimental research financial services. For the lowest income quintile, 77 using RCTs despite their own limitations. percent are excluded (Demirgüç-Kunt and Klapper 2012). In countries such as Cambodia, the Central Despite the still relatively small, albeit growing, African Republic, and Niger only 2–4 percent of all number of this type of randomized evaluation (some adults have an account at a formal financial institution. 25 cited in this overview), the general thrust of Without access to formal financial services, poor this new body of evidence suggests that financial families must rely on age-old informal mechanisms: services do have a positive impact on a variety of family and friends, rotating savings schemes, the microeconomic indicators, including self-employment pawn-broker, the moneylender, money under the business activities, household consumption, and well- mattress. At times, these informal mechanisms being (Bauchet et al. 2011).2 The impact varies across represent important and viable value propositions. individual financial product categories. RCTs to date Often, however, they are insufficient and unreliable, have largely been conducted at individual product and they can be very expensive. Financial exclusion levels, whereas some observers would argue that tends to impose large opportunity costs on those research ought to measure whether access to a broad who most need opportunity. range of services improves household ability to make appropriate choices. 2. Increasingly Robust Evidence of Beneficial Economic Impact Credit. According to the randomized impact evaluations of microcredit to date, two main Across a range of possible impact levels, recent patterns stand out: small businesses do benefit evidence suggests that access to and use of formal from access to credit while the linkage to broader financial services is beneficial. welfare is less clear. 1 The application of this approach to economics is increasingly considered a highly reliable means of assessing micro-level impact. The main strength of this approach is that it corrects for selection bias, a prominent failure of many other approaches. Compared to other methodologies, which are starting from theoretical questions and assumptions, it also has the advantage of not specifically testing one, conceivably narrow underlying economic theory. It simply assesses whether a specific, controlled change has a discernable impact relative to the control group. However, RCTs have their own methodological weaknesses, which are described, e.g., by Ravallion (2009) and Barret and Carter (2010). One main concern is the lack of external validity, which means that inferences for other settings or even scaling up based on the results of an RCT can be difficult. Other caveats include the choice of the proxy variable to measure welfare impact, ethical dilemmas, and cost effectiveness. 2 The experimental literature for financial inclusion is rapidly evolving with new papers being published at a high rate. This part of the Focus Note updates and expands previous work by Bauchet, et al. (2011). 3 Most of the studies to date provide mixed evidence significant effects on household consumption and on the impact of microcredit on important measures expenditures. However, it did find in summary that of household welfare such as an increase in “… the results paint a generally positive picture of consumption or income in poor households over the average impacts of expanded credit access on the typically relatively short time horizon studied well-being: depression falls, trust in others rises, (Banerjee, Duflo, Glennerster, and Kinnan 2010 and and female household decision-making power 2013; Crépon, Devoto, Duflo, and Parienté 2011; increases.” Studies also saw a reduction in the Karlan and Zinman 2011; Angelucci, Karlan, and spending on temptation goods, such as tobacco Zinman 2013). (India, Morocco, and Mongolia). An important point to consider when interpreting results from the An update of the Spandana study in Hyderabad experiments described here is the heterogeneity of (Banerjee, Duflo, Glennerster, and Kinnan 2013), effects across subjects. For subjects that do not own which provides one of the first, longer-term results businesses, microcredit can help their households by going back to borrowers after three years, also manage cash-flow spikes and smooth consumption. did not find later-stage improvements in welfare as Access to microcredit can also lead to a general a result of access to the initial microcredit. There increase in consumption levels as it lowers the need was no evidence of improvements for longer-term for precautionary savings. welfare indicators, such as education, health, or women’s empowerment. By contrast, for business owners, microcredit can help investments in assets that enable them to start However, some studies suggested nuances and or grow their businesses. In some cases, short-term found some welfare impacts. A study in Mongolia declines in household consumption coincide with (Attanasio et al. 2011) found large impacts of group investment during the set-up and growth phases for loans on food consumption (both more and healthier microbusinesses. Researchers are in fact confirming food). But this same finding did not hold for that access to credit does benefit businesses. individual loans. The authors see better monitoring There is evidence that microcredit both spurred in the group setting and, therefore, larger long- new business creation and benefitted existing term effects as the reason for these results. They microbusinesses in Mongolia and Bosnia (Attanasio hypothesize that “the joint-liability scheme better et al. 2011; Augsburg, de Haas, Harmgart, and ensures discipline in terms of project selection Meghir 2012), although another study in the and execution, so that larger long-run effects Philippines didn’t find such effects. Studies found are achieved.” A South Africa study that looked positive effects on a variety of indicators, including at expanding access to consumer credit found the income of existing businesses (India, the increased borrower well-being: income and food Philippines, and Mongolia), business size (Mexico), consumption went up, measures of decision making and the scale of agricultural activities and the within the household improved, borrower’s status diversification of livestock (Morocco). In addition, in the community improved, as did overall health access to microcredit increased the ability of and outlook on prospects and position. However, microentrepreneurs to cope with risk (the Philippines borrowers were also more subject to stress (Karlan and Mexico). These findings are more remarkable and Zinman 2010). when one considers that most of these studies investigate the effects of credit simply being offered A study of Compartamos borrowers in Mexico to the treatment group, rather than the effects of (Angelucci, Karlan, and Zinman 2013) did not find actual credit uptake and usage.3 In populations with 3 In the parlance of field experiments, they estimate the effects of the “intent-to-treat.” 4 few business owners, credit take-up for investment and Robinson 2013a). However, a parallel study with is likely to be low thus reducing the potential to male rickshaw drivers in the same town did not show identify statistically significant effects. similar welfare impacts. There is also recent experimental evidence suggesting Another Kenya study that looked at the impact of that greater flexibility in product design could result simple informal health savings products found an in improved impacts (Field, Pande, Papp, and Rigol increase in health savings by at least 66 percent forthcoming). When borrowers were given a two- accompanied by very high take-up rates. When month grace period before their first loan payment, using a commitment savings product, investments they diversified their inventory, were more likely in preventative health went up by as much as 138 to purchase durable assets, and had higher profits percent (Dupas and Robinson 2013b). The authors three years later. Although default rates increased found that earmarking for health emergencies somewhat, the patterns indicate that the more increased people’s ability to cope with shocks. The flexible repayment structure encouraged productive study underlines the importance of health savings risk taking. and investments in preventative health in reducing poor people’s vulnerability to health shocks. In their assessment of the microcredit evidence, Banerjee and Duflo (2011, p. 171) concluded that A study on commitment savings in Malawi showed “as economists, we were quite pleased with these positive effects on business investment, increased results: The main objective of microfinance seemed expenditures, and crop outputs (Brune, Giné, to have been achieved. It was not miraculous, but it Goldberg, and Yang 2013). Access to a commitment was working. In our minds, microcredit has earned savings account had positive impacts on female its rightful place as one of the key instruments in the empowerment in the Philippines. Self-reported fight against poverty.” household decision-making increased, particularly for women with little decision-making power at the Savings. The results of studies on the impact of baseline, resulting in a shift toward female-oriented savings are more consistently positive than those durable goods purchased in the household (Ahsraf, for credit, although there are fewer of these Karlan, and Yin 2010). studies. Savings help households manage cash flow spikes and smooth consumption, as well as build Insurance. Another instrument that can help poor working capital. According to researchers, for poor households mitigate risk and manage shocks is households without access to a savings mechanism insurance. Recent randomized evaluations in India it is more difficult to resist immediate spending and Ghana of weather-based index insurance temptations. showed strong positive impact on farmers because the assurance of better returns encouraged farmers When mechanisms for high-frequency, low-balance to shift from subsistence to riskier cash crops (Cole, deposit services are available, they seem to benefit et al. 2013; Karlan, Osei-Akoto, Osei, and Udry the poor. A randomized evaluation in rural western 2014). In Ghana, insured farmers bought more Kenya found that access to a new commitment fertilizers, planted more acreage, hired more labor, savings service enabled female market vendors to and had higher yields and income, which led to mitigate the effect of health shocks, increase food fewer missed meals and fewer missed school days expenditure for the family (private expenditures for the children. were 13 percent higher), and increase investments in their businesses by 38–56 percent over female In Kenya, researchers found index insurance to be vendors without access to a savings account (Dupas a powerful protection against the negative impacts 5 from natural disasters. In the face of a serious drought, also find an increased willingness to send remittances farmers had to sell fewer assets (minus 64 percent), as a result of access to mobile money; however, they missed fewer meals (minus 43 percent), and were less did not examine welfare implications. dependent on food aid (minus 43–51 percent) or any other form of assistance (minus 3–30 percent) (Janzen One randomized evaluation of the impact of a cash and Carter 2013). transfer program delivered via mobile phone (Aker, Boumnijel, McClelland, and Tierney 2011) showed Vulnerability to risk and the lack of instruments to reductions in both the cost of distribution for the cope with external shocks adequately make it difficult implementing agency and the cost of obtaining for poor people to escape poverty. The still limited the cash transfer for the program recipient. The impact evidence to date is focused on relatively few recipients’ cost savings resulted in diversification of insurance products, but suggests that microinsurance expenditures (including food), fewer depleted assets, could be an important mechanism for mitigating risk. and a greater variety of crops grown, especially cash However, demand and uptake—even when offered crops grown by women. for free in the context of these evaluations—is strikingly low (Matul, Dalal, De Bock, and Gelade Due to the relative newness of mobile money and 2013). Key barriers for uptake, including lack of trust product-specific issues in conducting welfare impact and liquidity constraints, have to be addressed to studies such as disentangling channel and product, it realize the full potential of microinsurance to work will take time until we have a robust evidence base for the poor. of how payments and mobile money impact the lives of poor people. Payments and mobile money. To date there have been few randomized evaluations on the impact of B. Local Economic Activity payments and mobile money. Two main patterns stand out so far: Mobile money reduces households’ Financial access improves local economic activity. transaction costs and seems to improve their ability Several settings over the past decades have offered to share risk. an opportunity to assess the impact of financial access compared to a baseline in quasi-experimental Jack and Suri (2014) examine the impact of reduced settings at the local economy level. For example, a transaction costs of mobile money on risk sharing in study using state-level panel data in India provides Kenya. Using nonexperimental panel data, they found evidence that local differences in opening bank that M-PESA users were able to fully absorb large branches in rural unbanked locations (driven by negative income shocks (such as severe illness, job requirements of the Indian regulator between 1977 loss, livestock death, and harvest or business failure) and 1990) were associated with a significant reduction without any reduction in household consumption. By in rural poverty (Burgess and Pande 2005). However, contrast, consumption for households without access the push ultimately proved unsustainable. High bank to M-PESA fell on average 7 percent in response to loan default rates during the 1980s led to the demise a major shock. of the rural branch expansion program after 1990. As the underlying mechanism, researchers identify an In Mexico, research (Bruhn and Love 2013) showed increase in remittances received both in number and that the rapid opening of Banco Azteca branches size and a greater diversity of senders. M-PESA also in more than a thousand Grupo Elektra retail stores facilitates increased risk-sharing among networks of had a significant impact on the region’s economy, friends and family. Two other studies (Blumenstock, leading to a 7 percent increase in overall income Eagle, and Fafchamps 2012; Batista and Vicente 2012) levels relative to similar communities where no Banco 6 Azteca branches had been opened. Households were development (measured by private credit and bank better able to smooth consumption and accumulated branch growth), but it declines sharply for countries more durable goods in communities with Banco at intermediate and advanced stages of financial Azteca branches (Ruiz 2013). At the same time, the development (Jahan and McDonald 2011). One proportion of households that saved declined by interpretation is that higher income segments initially 6.6 percent in those communities, suggesting that benefit more from deeper financial intermediation, households were able to rely less on savings as a but as it progresses, poorer segments benefit, too. buffer against income fluctuation when formal credit Regressions that account for country characteristics became available. and address potential reverse causality show a robust negative relationship between financial C. Macroeconomic Level depth and the Gini coefficient (Clarke, Xu, and Zhou 2006). Moreover, financial depth was associated At the macroeconomic level, the evidence has to rely with increases in the income share of the lowest on cross-country comparisons. The well-established income quintile across countries from 1960 to literature (summarized, for example, in Levine 2005, and countries with higher levels of financial 2005 and Pasali 2013) suggests that under normal development also experienced larger reductions in circumstances, the degree of financial intermediation the share of the population living on less than $1 per is not only positively correlated with growth and day in the 1980s and 1990s. Controlling for other employment, but it is generally believed to causally relevant variables, almost 30 percent of the variation impact growth. The main mechanisms for doing so across countries in rates of poverty reduction can are generally lower transaction costs and better be attributed to cross-country variation in financial distribution of capital and risk across the economy. development (Beck, Demirgüç-Kunt, and Levine Broader access to bank deposits can also have a 2007). Financial inclusion seems to reduce inequality positive effect on financial stability. by disproportionally relaxing the credit constraints on poor people, who lack collateral, credit history, and However, there are some caveats. Some research connections. Research by the World Bank (Han and indicates that the positive growth impact from financial Melecky 2013) also suggests that broader financial intermediation does not hold in economies with weak inclusion can coincide with greater financial stability, institutional frameworks (Demetriades and Law 2006), though sorting out the lines of causation between such as poor or nonexistent financial regulation, or those two sets of variables remains a challenge. It in extremely high-inflation environments (Rousseau seems plausible, however, that greater access to bank and Wachtel 2002). Evidence also indicates that deposits can make the funding base of banks more the positive long-run relationship between financial resilient in times of financial stress. The authors stress intermediation and output growth co-exists with a that policy efforts to enhance financial stability should mostly negative short-run relationship (Loayza and thus not only focus on macroprudential regulation, Ranciere 2006). More recent work following the global but also recognize the positive effect of broader financial crisis also suggests that the relationship access to bank deposits. between financial depth and growth might not be linear, but shaped like an inverted “U”—i.e., at very 3. Additional Indirect Benefits low levels of financial intermediation and at very high of Financial Inclusion levels, the positive relationship disappears (Cecchetti and Kharroubi 2012). In addition to the direct economic benefits, two recent developments suggest benefits for other Bivariate relationships indicate that inequality government and private-sector efforts that might as measured by the Gini coefficient increases as arise from inclusive low-cost financial systems that countries progress through early stages of financial reach larger numbers of citizens. 7 First, policy makers increasingly recognize that cost electronic retail payment systems have reached a financial market that reaches all citizens allows critical scale and no studies have been conducted for more effective and efficient execution of other as to the possible household welfare impact due to social policies. For example, financial inclusion access to these types of novel services. improves the payment of conditional transfers such as when parents are rewarded for ensuring 4. Conclusion their children get recommended vaccinations or for sending their daughters to school. Because of the Global and national policy makers are committing potential cost savings, a number of countries are to advance financial inclusion. Financial services are switching their government payments to electronic a means to an end, and financial development must means to improve targeting of beneficiaries and take into account vulnerabilities and ward off possible reduce transactions costs. In Brazil, the bolsa familia unintended negative consequences. However, recent program (a conditional cash transfer program that evidence using rigorous research methodologies serves 12 million families) reduced its transaction appears to generally confirm the policy makers’ costs from 14.7 percent of total payments to 2.6 convictions that inclusive and efficient financial percent when it bundled several benefits onto one markets have the potential to improve the lives of electronic payment card (Lindert, Linder, Hobbs, citizens, reduce transaction costs, spur economic and de la Briére 2007). A low-cost financial system activity, and improve delivery of other social benefits helps governments better execute other social and innovative private-sector solutions. policies. Whether those payments can in turn lead to a virtuous cycle of including more citizens in the This Focus Note summarizes recent evidence financial system, and keeping them there, is not on three different economic levels. At the yet clear. microeconomic level, it synthesizes the evidence of how the use of different financial products affects the Second, financial innovation that dramatically lives of the poor. Studies show that small businesses lowers transaction costs and increases reach is benefit from access to credit, while the impact on enabling new private-sector business models that the borrower’s household’s broader welfare might help address other development priorities. In Kenya, be more limited. Savings help households manage where mobile money services such as M-PESA reach cash flow spikes, smooth consumption, as well as more than 80 percent of the population, a wave build working capital. Access to formal savings of second-generation innovative businesses and options can boost household welfare. Insurance uses is emerging on the M-PESA infrastructure. The can help poor households mitigate risk and manage presence of a ubiquitous, low-cost electronic retail shocks. New types of payment services can reduce payment platform increases the viability of new transaction costs and seem to improve households’ business models that need to collect large numbers ability to manage shocks by sharing risks. Research of small amounts. This may also help address other also suggests that financial access improves local development priorities. For instance, M-Kopa economic activity. in Kenya or Mobisol in Tanzania have created microleasing for off-grid, community-based solar At the macroeconomic level, the empirical evidence power—an example of innovation in the context shows that financial inclusion is positively correlated of climate-change adaptation. Similar advances are with growth and employment. The researchers being made with respect to water services to low- generally believe in underlying causal impact. The income households and communities. So far, this main mechanisms they cite for doing so are generally type of leverage has by definition occurred only in lower transaction costs and better distribution of geographies such as Kenya or Tanzania where low- capital and risk across the economy. Evidence of a 8 more preliminary nature suggests that broader access Augsburg, Britta, Ralph de Haas, Heike Harmgart, to bank deposits can also have a positive effect on and Costas Meghir. 2012. “Microfinance at financial stability that benefits the poor indirectly. the Margin: Experimental Evidence from Bosnia and Herzegovina.” Working Paper 146. In addition to the direct economic benefits, two London: European Bank for Reconstruction and recent developments suggest benefits for other Development. government and private-sector efforts that might arise from inclusive low-cost, financial systems that Banerjee, Abhijit V. 2013. “Microcredit under the reach a larger number of citizens. First, financial Microscope: What Have We Learned in the Past Two inclusion can improve the effectiveness and efficient Decades, and What Do We Need to Know?” Annual execution of government payment of social safety net Review of Economics, 5: 487–519. transfers (government-to-person payments), which play an important role in the welfare of many poor Banerjee, Abhijit V., and Esther Duflo. 2011. Poor people. Second, financial innovation can significantly Economics. New York: Perseus Books. lower transaction costs and increase reach, which is enabling new private-sector business models that Banerjee, Abhijit V., Esther Duflo, Rachel help address other development priorities. Glennerster, and Cynthia Kinnan. 2010. “The Miracle of Microfinance? Evidence from a Randomized References Evaluation.” Cambridge, Mass.: Abdul Latif Jameel Poverty Action Lab and Massachusetts Institute of AFI (Alliance for Financial Inclusion). 2013. “Putting Technology, June. Financial Inclusion on the Global Map. The 2013 Maya Declaration Progress Report.” Bangkok: AFI. ———. 2013 (Update). “The Miracle of Microfinance? Evidence from a Randomized Aker, Jenny, Rachid Boumnijel, Amanda McClelland, Evaluation.” NBER Working Paper No. 18950. and Niall Tierney. 2011. “Zap It to Me: The Short- Cambridge, Mass.: National Bureau of Economic Term Effects of a Mobile Cash Transfer Program.” Research, May. Working Paper No. 263. Washington, D.C.: Center for Global Development. Barrett, Christopher, and Michael Carter. 2010. “The Power and Pitfalls of Experiments in Development Angelucci, Manuela, Dean Karlan, and Jonathan Economics. Some Non-random Reflections.” Applied Zinman. 2013. “Win Some Lose Some? Evidence Economic Perspectives and Policy, Vol. 32, No. 4: from a Randomized Microcredit Program Placement 515–48. Experiment by Compartamos Banco.” NBER Working Papers 19119. Cambridge, Mass.: National Bureau of Economic Research, May. Batista, Cátia, and Pedro C. Vicente. 2012. “Introducing Mobile Money in Rural Mozambique: Ashraf, Nava, Dean Karlan, and Wesley Yin. 2010. Evidence from a Field Experiment.” Universidade “Female Empowerment: Impact of a Commitment Nova de Lisboa, mimeo. Preliminary, October. Savings Product in the Philippines.” World Development 38 (3): 333–44. Bauchet, Jonathan, Cristobal Marshall, Laura Starita, Jeanette Thomas, and Anna Yalouris. 2011. Attanasio, Orazio, Britta Augsburg, Ralph de Haas, “Latest Findings from Randomized Evaluations of Emla Fitzsimons, and Heike Harmgart. 2011. “Group Microfinance.” Forum 2. Washington, D.C.: CGAP, Lending or Individual Lending? Evidence from a Financial Access Initiative, Innovations for Poverty Randomised Field Experiment in Mongolia.” Working Action, and Abdul Latif Jameel Poverty Action Paper W11/20. London: Institute for Fiscal Studies. Lab. 9 BBeck, Thorsten, Aslı Demirgüç-Kunt, and Ross Cole, Shawn, Xavier Giné, Jeremy Tobacman, Petia Levine. 2007. “Finance, Inequality, and the Poor.” Topalova, Robert Townsend, and James Vickery. Journal of Economic Growth 12(1): 27–49. 2013. “Barriers to Household Risk Management: Evidence from India.” American Economic Journal— Blumenstock, Joshua, Nathan Eagle, and Marcel Applied Economics, 5 (1): 104–35. Fafchamps. 2012. “Charity and Reciprocity in Mobile Phone Based Giving: Evidence in the Aftermath of Collins, Daryl, Jonathan Morduch, Stuart Rutherford, Earthquakes and Natural Disasters.” Working Paper. and Orlanda Ruthven. 2009. Portfolios of the Poor: Berkeley, Calif.: University of California, Berkeley, How the World’s Poor Live on $2 a Day. Princeton, October. N.J.: Princeton University Press. Bruhn, Miriam, and Inessa Love. 2013. “The Economic Crépon, Bruno, Florencia Devoto, Esther Duflo, and Impact of Expanding Access to Finance in Mexico.” William Parienté. 2011. “Impact of Microcredit in In Robert Cull, Aslı Demirgüç-Kunt, and Jonathan Rural Areas of Morocco: Evidence from a Randomized Morduch, eds., Banking the World: Empirical Evaluation.” Working Paper. Cambridge, Mass.: Foundations of Financial Inclusion. Cambridge, Massachusetts Institute of Technology, March. Mass.: Massachusetts Institute of Technology Press, pp. 137–56. Demetriades, P. O., and S. H. Law. 2006. “Finance, Institutions and Economic Development.” ———. 2009. “The Economic Impact of Banking the International Journal of Finance and Economics, Unbanked: Evidence from Mexico.” Policy Research 11(3): 245–60. Working Paper No. 4981. Washington, D.C.: World Bank. Demirgüç-Kunt, Asli, and Leora Klapper. 2012. “Measuring Financial Inclusion: The Global Findex Brune, Lasse, Xavier Giné, Jessica Goldberg, and Database.” World Bank Policy Research Paper 6025. Dean Yang. 2013. “Commitments to Save: A Field Washington, D.C.: World Bank, April. Experiment in Rural Malawi.” http://econweb.umd. edu/~goldberg/docs/bggy_mwisavings.pdf. Dupas, Pascaline, and Jonathan Robinson. 2013a. “Savings Constraints and Microenterprise Burgess, Robin, and Rohini Pande. 2005. “Do Rural Development: Evidence from a Field Experiment Banks Matter? Evidence from the Indian Social in Kenya.” American Economic Journal—Applied Banking Experiment.” American Economic Review, Economics, 5 (1): 163–92. Vol. 95, No. 3: 780–95. ———. 2013b. “Why Don’t the Poor Save More? Cecchetti, Stephen G., and Enisse Kharroubi. 2012. Evidence from Health Savings Experiments.” “Reassessing the Impact of Finance on Growth.” The American Economic Review, 103 (4): 1138– Working Paper No. 381. Basel: Bank of International 71. Settlements, July. Field, Erica, Rohini Pande, John Papp, and Natalia Clarke, George R. G., L. Colin Xu, and Heng-fu Rigol. Forthcoming. “Does the Classic Microfinance Zou. 2006. “Finance and Inequality: What Do the Model Discourage Entrepreneuership Among the Data Tell Us?” Southern Economic Journal 72 (3): Poor? Evidence from India.” American Economic 578–96. Review. 10 G20 Information Centre. 2009. “G20 Leaders Karlan, Dean, Isaac Osei-Akoto, Robert Osei, and Chris Statement: The Pittsburgh Summit.” http://www.g20. Udry. 2014. “Agricultural Decisions after Relaxing utoronto.ca/2009/2009communique0925.html Credit and Risk Constraints.” Quarterly Journal of Economics. Gallup. 2012. “Employed Full Time for an Employer Index.” http://www.gallup.com/poll/145595/ Levine, Ross. 2005. “Finance and Growth: Theory and worldwide-employed-full-time-employer.aspx Evidence.” In Philippe Aghion and Steven Durlauf, eds, Handbook of Economic Growth, edition 1, volume 1. Han, Rui, and Martin Melecky. 2013. “Financial Amsterdam: Elsevier. Inclusion for Financial Stability: Access to Bank Deposits and the Growth of Deposits in the Lindert, Kathy, Anja Linder, Jason Hobbs, and Global Financial Crisis.” Washington, D.C.: World Bénédicte de la Briére. 2007. “The Nuts and Bolts Bank. https://openknowledge.worldbank.org/ of Brazil’s Bolsa Família Program: Implementing handle/10986/16010 Conditional Cash Transfers in a Decentralized Context.” Social Protection Working Paper No. 0709. ILO (International Labour Organisation). 2013. Women Washington, D.C.: World Bank, May. and Men in the Informal Economy: A Statistical Picture, Second Edition. Geneva: ILO. http://www. Loayza, Norman V., and Romain Ranciere. 2006. ilo.org/wcmsp5/groups/public/---dgreports/---stat/ “Financial Development, Financial Fragility, and documents/publication/wcms_234413.pdf Growth.” Journal of Money, Credit, and Banking, 38(4): 1051–76. Jack, William, and Tavneet Suri. 2014. “Risk Sharing and Transactions Costs: Evidence from Kenya’s Mobile Mankiw, Gregory, and Laurence M. Ball. 2011. Money Revolution.” American Economic Review, Macroeconomics and the Financial System. New York: 104(1): 183–223. Worth Publishers. Jahan, Sarwat, and Brad McDonald. 2011. “A Bigger Matul, Michal, Aparna Dalal, Ombeline De Bock, Slice of a Growing Pie.” Finance and Development, 66, and Wouter Gelade. 2013. “Why People Do Not September. Washington, D.C.: International Monetary Buy Insurance and What Can We Do about It.” Fund. Microinsurance Paper No. 20. Geneva: ILO, February. Janzen, Sarah A., and Michael R. Carter. 2013. Merton, Robert C., and Zvi Bodie. 1995. “A Conceptual “After the Drought: The Impact of Microinsurance Framework for Analyzing the Financial Environment.” on Consumption Smoothing and Asset Protection.” In D. B. Crane, et al., eds, The Global Financial System: NBER Working Paper No. 19702. Cambridge, Mass.: A Functional Perspective. Boston, Mass.: Harvard National Bureau of Economic Research, December. Business School Press, 3–31. Karlan, Dean, and Jonathan Zinman. 2010. “Expanding Pasali, Selahattin Selsah. 2013. “Where Is the Credit Access: Using Randomized Supply Decisions to Cheese? Synthesizing a Giant Literature on Causes Estimate the Impacts.” Review of Financial Studies, and Consequences of Financial Sector Development.” 23: 433–64. World Bank Policy Research Working Paper 6655. Washington D.C.: World Bank, October. ———. 2011. “Microcredit in Theory and Practice: Using Randomized Credit Scoring for Impact Ravallion, Martin. 2009. “Should the Randomistas Evaluation.” Science, 332:1278–84. Rule.” The Economist’s Voice, vol. 6, no. 2: 1–5. 11 Rousseau, P., and P. Wachtel. 2002. “Inflation World Bank. 2012b. World Development Report 2013. Thresholds and the Finance-Growth Nexus.” Jobs. Washington, D.C.: World Bank. Journal of International Money and Finance, 21(6): 777–93. World Bank. 2013a. Financial Inclusion Strategies Database. http://econ.worldbank.org/WBSITE/ Ruiz, Claudia. 2013. “From Pawn Shops to Banks. The EXTERNAL/EXTDEC/EXTGLOBALFINREPORT/0,,con Impact of Formal Credit on Informal Households.” tentMDK:23491959~pagePK:64168182~piPK:64168 World Bank policy research working paper 6634. 060~theSitePK:8816097,00.html# Washington, D.C.: World Bank. World Bank. 2013b. “Universal Financial Access Is Vital World Bank. 2012a. World Bank Findex Database. to Reducing Poverty, Innovation Key to Overcoming http://econ.worldbank.org/WBSITE/EXTERNAL/ the Enormous Challenge, Says President Jim Yong EXTDEC/EXTRESEARCH/EXTPROGRAMS/ Kim.” Press release, 11 October. http://www. EXTFINRES/EXTGLOBALFIN/0,,menuPK:85196 worldbank.org/en/news/press-release/2013/10/11/ 98~pagePK:64168176~piPK:64168140~theSit universal-financial-access-vital-reducing-poverty- ePK:8519639,00.html innovation-jim-yong-kim No. 92 April 2014 Please share this Focus Note with your colleagues or request extra copies of this paper or others in this series. CGAP welcomes your comments on this paper. All CGAP publications are available on the CGAP Web site at www.cgap.org. CGAP 1818 H Street, NW MSN P3-300 Washington, DC 20433 USA Tel: 202-473-9594 Fax: 202-522-3744 Email: cgap@worldbank.org © CGAP, 2014 The authors of this Focus Note are Robert Cull, lead economist at the World Bank; Tilman Ehrbeck, CEO of CGAP; and Nina Holle, financial sector development analyst at CGAP. The authors are thankful for review and valuable comments from Katharine McKee, Mayada El-Zoghbi, Gerhard Coetzee, and Greg Chen from CGAP. We also want to thank Xavier Giné from the World Bank Development Research Group for his valuable input on sections of this paper. The suggested citation for this Focus Note is as follows: Cull, Robert, Tilman Ehrbeck, and Nina Holle. 2014. “Financial Inclusion and Development: Recent Impact Evidence.” Focus Note 92. Washington, D.C.: CGAP. Print: ISBN 978-1-62696-036-7 epub: ISBN 978-1-62696-038-1 pdf: ISBN 978-1-62696-037-4 mobi: ISBN 978-1-62696-039-8 UKa from the British people