In my last post (# 48), I set out to look for evidence of shock-coping that is improved by having access to microfinance. Shocks are the more extreme (therefore more easily observed) manifestations of the consumption-smoothing capabilities, or resilience, of households.
The last post offered evidence from Freedom from Hunger’s research. A data set of 224 “impact stories” of randomly selected clients in eight countries showed that about one in five clients experienced one or more financial shocks (the majority due to health problems), and 57% of these clients cushioned the shock with the help of microfinance services. Two RCTs, in Bolivia and Ghana in the 1990s, demonstrated the value of village banking participation for reducing the incidence and duration of the “hungry season” in Ghana but not in Bolivia (possibly because the harvest that study year was unusually good). In both countries, those who had shocks to deal with were better able to do so if they were members of village banks.
The statistical power of Freedom from Hunger’s RCTs (among the first for microfinance impacts) was insufficient to allow definitive conclusions. But there have been a good number of statistically powerful RCTs done recently. I have carefully read all those listed by David Roodman in his most recent review (April 2012), plus a couple more, to look for solid evidence relevant to shock-coping. Below is what I found (I mostly quote from the research reports themselves).
It is quite significant that of these 12 RCTs on microfinance impacts, all but one or two offer some evidence of positive impact on shock-coping specifically or consumption-smoothing more generally. In short, the evidence for the emerging microfinance theory of change is much stronger than for the classic theory of change.
Abhijit Banerjee, Esther DuFlos, Rachel Glennerster and Cynthia Kinnan on group microfinance for women, offered by Spandana in Hyderabad, India:
Almost 2/3 of households (64%) experienced a health shock costing Rs. 500 or more in the past year. Property loss costing Rs. 500 or more in the past year was reported by 11% of households. Job loss was experienced by 2.4% of households in the past year (a low number, perhaps because many households do not have a member with a steady job), and 4.8% of households experienced a death in their household the past year. The likelihood that a household borrowed to deal with a shock does not differ across treatment and comparison areas. However, treatment-area households were almost twice as likely to have borrowed from an MFI to deal with a shock (2.2% of treatment-area households vs. 1.2% of comparison-area households), and the amount borrowed from MFIs (including zeros for non-borrowers) is twice as high: Rs. 210 vs. Rs. 91. This is consistent with the idea that microcredit offers a way to deal with shocks that may substitute for holding buffer stocks of assets. In short, we see some evidence that microcredit helps households avoid dealing with shocks using nonpayment of bills and purchasing goods on credit, which may be especially costly responses. However, microcredit does not appear to significantly reduce reliance on moneylenders or relatives and friends. [I interpret their results to mean that Spandana offered clients the opportunity to augment the money they could borrow from more traditional sources rather than replace traditional borrowing.]
Dean Karlan and Jonathan Zinman on individual lending to microentrepreneurs, offered by First Macro Bank in the Philippines:
… increased access to formal-sector credit complements, rather than crowds out, local and family risk-sharing mechanisms. Treated microentrepreneurs have more places to turn for formal and informal credit in a pinch, and consequently rely less on formal-sector insurance. [This echoes and enriches the Banerjee et al. conclusion (above), indicating that shock-coping effects of access to microfinance are additive rather than substitutive for traditional financial assistance.]
Dean Karlan and Jonathan Zinman on individual lending to wage employees, offered by a payday lender in South Africa:
We focus on measuring two simple measures of food consumption. One is whether anyone in the household experienced hunger in the past 30 days (14% of households in the sample reported some hunger). The other is whether the quality of food consumed by the household improved over the last 12 months (26% reported an improvement). Households randomly assigned a loan were an estimated 5.8 percentage points less likely to report hunger (with a p-value of 0.03), and 3.7 percentage points more likely to report a food quality improvement (although this estimate was not statistically significant, with a p-value of only 0.32). The next most common purpose for household borrowing is transportation expenses (19.4%); this and the clothing category are consistent with work-related investments. Indeed we find large treatment effects on employment: treated applicants were 11 percentage points (13%) more likely to be working at the time of the survey. Since everyone in our sample frame had verified employment at the time they entered the experiment, it appears that the treatment effect operates by enabling households to maintain employment by smoothing or avoiding shocks that prevent them from getting to work. Two related results (not shown) are consistent with the story that marginal borrowers use loan proceeds to make investments in maintaining wage employment. First, questions on job history reveal that treated applicants were indeed significantly less likely to report leaving a job since entering the experiment. Second, we find a positive point estimate on the likelihood that treated households repaired their car in recent months. [Access to microcredit seemed to enable prevention of the shock of losing one’s job—in part by dealing with the shock of a car breaking down.]
Bruno Crépon, Florencia Devoto, Esther Duflo and William Parienté on solidarity group lending (mainly to men), offered by Al Amana in rural Morocco:
Households that had a pre-existing activity decrease their non-durable consumption and consumption overall, as they save and borrow to expand their activities. Households that had not a pre-existing activity increase food and durable expenditure and no effects on business outcomes are observed. [No mention of shock-coping that I could find, but the positive effect on food expenditure by households without investible businesses indicates the potential for consumption-smoothing by poorer households during difficult periods.]
Orazio Attanasio, Britta Augsburg, Ralph De Haas, Emla Fitzsimons and Heike Harmgart on group vs. individual lending to women, offered by XacBank in Mongolia:
We find robust evidence that access to group loans led to more and healthier food consumption, in particular of fresh items such as fruit, vegetables and dairy products. With the exception of dairy, these effects are not only due to increased home production: we also see treated clients purchasing more. The estimated effect implies that over time food consumption increased by USD 19 more per household in group villages, that is, 6.3 per cent of the loan amount. In contrast to households in group-lending villages, households in individual-lending villages do not experience much change in their consumption as a result of access to credit. We do not find any effects on aggregate consumption and expenditure variables – not even with increased exposure to treatment. A second finding that holds for both treatment programmes is that women with lower education seem to benefit more from the intervention than women with higher education. We interpret the level of education as a proxy poverty measure, more reliable than a wealth indicator given that it is not affected by the programme and is more stable over time. The results therefore suggest that it is the poorer part of the targeted population that benefits more from the microcredit intervention, independent of how it is being delivered. [The effect of group lending on food expenditure, especially in poorer households, indicates the potential for microfinance to enhance survival during a hungry season or livestock die-off.]
Britta Augsburg, Ralph De Haas, Heike Harmgart and Costas Meghir on individual lending to microentrepreneurs, offered by EKI in Bosnia and Herzegovina:
They do not refer to shocks, but they do observe decline in consumption by poorer clients, presumably because they do not have sufficient savings to top up the loans for useful investment in a business and so must reduce consumption to have enough to invest.
Lasse Brune, Xavier Giné, Jessica Goldberg and Dean Yang on savings accounts for tobacco farmers, offered by Opportunity International Bank in Malawi:
In sum, we find that facilitating commitment savings for smallholder cash-crop farmers in Malawi has substantial impacts on savings prior to next planting season, agricultural inputs applied in next season, access to funds during next lean (pre-harvest) period, crop sales at next harvest, and on food and total expenditures after next harvest. By contrast, the impact of facilitating “ordinary” accounts not as large or statistically significant. [Access to funds during the pre-harvest (hungry season) indicates potential for these commitment savings accounts to help smooth consumption during difficult times.]
Pascaline Dupas and Jonathan Robinson on savings accounts for market vendors (mostly women) and bicycle-taxi drivers (men), offered by Bumala Financial Services Association in Kenya:
In the version of the paper I read, Dupas and Robinson made no reference to shock-coping, but Roodman (in the review cited above) reported they “found that in households offered savings accounts, spending fell less when someone in the family got malaria. But the effect wasn’t very significant statistically, which is probably why it has been dropped from the latest draft.” This means there was some indication of shock-coping enhanced by access to savings accounts, but I don’t count this paper as offering solid evidence like the others.
Pascaline Dupas and Jonathan Robinson on five kinds of health savings devices for members of ROSCAs in Kenya:
They looked for the effects of the Lock Box (safe place to save at home with full liquidity of savings in the box) and the HSA (personal health savings account controlled by the ROSCA treasurer for payment of medical expenses only) on the dependent variable “not having enough money to afford full treatment” of any health problem. Both of these “experimental technologies” (devices) were effective. The Lock Box was most valuable for those who are “providers” (likely to be asked for financial assistance by others) and for those who are not “present biased” (have self-discipline to save for a future purpose, in this case, health expenses). However, the HSA was far more effective, despite the commitment for medical expenses only. “This result suggests that people in our sample value earmarking for emergency health savings (a first-order concern to many households) more than for preventative health” [e.g., saving to purchase a malaria-preventing mosquito net]. In other words, given the opportunity, people will save to be able to manage the expense of a health shock.
Ronald Abraham, Felipe Kast and Dina D. Pomeranz on savings accounts for microentrepreneurs in Chile:
I am unable to find the full research report in a format that is readable, but here is the abstract:
Poverty is often characterized not only by low average income, but also by highly variable income and expenditures, and a lack of access to insurance services that can help smooth consumption. We investigate whether access to a formal savings account can provide a vehicle for self-insurance, by allowing participants to build a buffer stock of precautionary savings. In a randomized field experiment in Chile, about 3000 low-income micro-entrepreneurs are provided access to a formal savings account with no minimum balance or maintenance fees. Evaluating the impact after one year, we find that access to such accounts helps participants alleviate the burden of economic shocks, both objectively and subjectively. Participants with access to a savings account have less informal debt, fewer outstanding payments, and less often need to reduce consumption due to economic difficulties. Subjectively, they report being less worried about their financial future, and evaluate their recent economic situation as less severe. We therefore conclude that formal savings accounts can be an effective vehicle to provide a means for consumption-smoothing in contexts where many other forms of insurance are lacking. [Clear impact of savings accounts on consumption-smoothing and shock-coping.]
Jeffrey A. Flory on savings accounts offered by mobile units of the Opportunity International Bank in rural Malawi:
This study is primarily concerned with the indirect effects of formal savings on consumption-smoothing of non-users of the savings accounts, specifically the impact on assistance receipts by the most vulnerable households in particular. “In qualitative interviews in rural areas of central Malawi, formal-savers report the top reasons people ask them for cash help are for medical expenses and sickness-related issues, to buy food, or to pay for funeral expenses.” I plan to do a separate post to more deeply examine this unprecedented study, but for the moment, here is the results part of the abstract of the research report.
Using a panel dataset of over 2,000 households collected during a rapid expansion of formal savings services in Central Malawi, this paper shows that experimentally boosting use of formal savings in rural areas sharply increases inter-household transfers during peak periods of hunger. The impact on transfer receipts is strongest among the poorest households, a de facto financial services-ineligible group, among whom the effects are also linked to significant changes in welfare. The strong impacts of formal savings expansion on non service-users suggests that formal finance can have much greater immediate-term effects than would be suggested by focusing exclusively on impacts experienced by service-users. The findings also highlight the sensitivity of traditional safety nets and welfare outcomes among the highly vulnerable in villages to expansion of formal financial markets. [Remarkable evidence of the indirect effects of savings accounts on community-wide shock-coping and consumption-smoothing—not just including the poorest households but most impactful for the poorest households! More on this in an imminent post.]
William Jack and Tavneet Suri on “The Risk Sharing Benefits of Mobile Money” (Working Paper, January 2011), part of a larger study of the use of M-Pesa in Kenya:
I heard Billy Jack (Georgetown University economist) present on this (and Jake Kendall of the Gates Foundation also cited this paper) at the USAID Evidence Summit in December 2012 (see my posts on that meeting), but cannot find this paper on the Web. However, Amolo Ng’weno of the Gates Foundation wrote the following about this study in a brief for the Global Savings Forum, November 2010:
The ability of friends and family to make immediate, real-time transactions of as little as $2 helps prevent unnecessary periods of going without food. Early research shows that households with M-PESA are more successful at weathering negative events and, specifically, do not reduce their food consumption when faced with a shock. Additionally, being able to get money quickly allows a family to take a relative to see the doctor on the first day of their illness, and children are able to attend school when they would have otherwise been forced to skip or drop out for lack of fees.
These results in Kenya tie in nicely with the indirect effects of savings accounts in Malawi. Both enhance traditional mutual assistance networks: the savings accounts accumulate money for use in inter-household transfers of financial assistance, and M-Pesa makes those transfers more timely and from a geographically broader network of mutual assistance.
In conclusion, all these RCTs together provide an impressive body of evidence that microfinance does indeed improve the effectiveness of household efforts to smooth consumption and cope with financial shocks—often substantially. However, it is notable that only two of these studies were focused on this question. There needs to be more and better research on this question. In particular, we need studies that actually measure variability of consumption over periods long enough to include the inevitable shocks that impact the poor. And we need these studies to disaggregate the poor into various levels of vulnerability to shocks, like the excellent Flory study in Malawi.