Theme Four of this blog is about household savings and consumption-smoothing. Post #49, in particular, showed that improvements in consumption-smoothing and ability to manage financial shocks are fairly reliable outcomes of microfinance access. This general conclusion is affirmed, in terms of better ability to manage liquidity and risk, by the just-released results of the impact study of Compartamos in the state of Sonora, Mexico, led by Manuela Angelucci, Dean Karlan and Jonathan Zinman. The ability to borrow has a positive impact on consumption-smoothing and shock-coping, but the improvements seem most pronounced when households have expanded access both to credit on reasonable terms and to safe and convenient ways to save. Often the available evidence of such impact is improved household food security. In particular, posts # 48 and 52 relied on measures of food security to verify improvement of consumption-smoothing.
Conversely, the best evidence for improved household food security may be improved consumption-smoothing (and its more extreme version, shock-coping), because food is such a major component of consumption in poor households. But there are other components of consumption to be smoothed, so we could find evidence of consumption-smoothing even in the face of increasing food insecurity. I don’t know of any examples of this; it seems an unlikely scenario, especially for the poor, but it is conceivable. Moreover, consumption-smoothing per se is hard to measure; therefore, little direct evidence is available (see post # 47). We do better to look for direct evidence of food security, which is easier to find, and it is provided in two very different ways in posts # 48 (on Barbara MkNelly’s Credit with Education impact research in Ghana and Bolivia) and # 52 (on Jeffrey A. Flory’s impact research on savings accounts offered in Malawi).
Food-Security Impacts of Credit with Education in Ghana and Bolivia
Briefly, Barbara’s research in Ghana found that the percentage of participant families that experienced a period when they ate less or less well during the previous 12 months was cut almost in half. In contrast, virtually no change was evident for either nonparticipants in program communities or for residents in control communities. For those households that experienced a hungry season, the mean duration of this period was less than one month for participants compared to a mean of almost two months for residents in control communities. Likewise, the coping strategies in the face of food shortages were less disruptive or unpleasant for participants.
The results in Bolivia were different. Following an unusually good harvest, the Credit with Education program seemed to have no significant effects on the incidence and duration of the hungry season; it was less likely or shorter for participants, but the reduction was almost as much for everyone else in the area. However, there was a clearer difference in how the program seemed to affect how participants coped with periods of food shortages. In particular, they were less likely to sell off animals as a coping strategy than residents in control communities. While animals might be considered illiquid savings, they are also typically productive assets, the selling of which undermines the long-term food and livelihood security of the family. Participants were also more likely to use profits from their businesses than were residents in control communities to help them cope. The availability of program loans seemed to increase the incidence of households assuming debt to help their family through the hungry season. But this debt was most commonly to their fellow village bank members through the process of “internal” lending from member savings (similar to borrowing from a savings group).
Food-Security Impacts of Access to Savings in Malawi
In Malawi, Flory found that the total value of inter-household assistance to the most vulnerable households seems to have been increased by the uptake of savings accounts by other, less vulnerable households in rural communities. There were more savings in the community to share with those suffering in the hungry season.
The study provides evidence for improvements in three different welfare indicators: two food-security indicators and one simple health indicator. Highly vulnerable households in treated villages were 11.8 to 16.3 percent more likely than comparable households in control villages to exit the worst food-security category (“severely insecure”) to enter one of the three less-severe categories. They also experience a 1.3 to 1.4 point reduction in a continuous food-insecurity score relative to the highly vulnerable in control villages. This represents a 10 to 12 percent improvement in food-security over baseline values. In addition, highly vulnerable households in treated villages were 12 to 17.4 percent less likely than those in control villages to report any members of the household as recently unwell.
We seldom see such dramatically positive direct, much less indirect, welfare impacts of microfinance or any other type of large-scale intervention, particularly for the poorest, most vulnerable people in the local population. Clearly in this case, there was a fortunate match between a simple information intervention tied to a basic savings-account offer and the needs and functions of a traditional social safety net providing mutual assistance among households of varied wealth and vulnerability status. Portfolios of the Poor and similar financial diary studies show that such informal webs of mutual assistance among households are common in poor communities. The power of microfinance for improving food security is demonstrated when it augments and enhances these informal systems to support the resilience of households in the face of financial setbacks and even severe shocks.
Food-Security Impacts of Saving for Change in Mali
Similar results are reported by the recently completed impact evaluation of the Saving for Change program in Mali by Innovations for Poverty Action (IPA) and the University of Arizona’s Bureau for Applied Research in Anthropology (BARA). It seems clear that Saving for Change builds resilience and helps families cope with food shortages. Households in Saving for Change villages were 10 percent less likely to be chronically food-insecure than households in control villages. This seems a less dramatic impact than in Ghana, but perhaps not. The Ghana data refer just to participant households. The 10 percent data point for Mali refers to all households in the Saving for Change villages, but only about one-third of the eligible women in the average Saving for Change village had joined savings groups at the time of the follow-up research survey. The impact on participant households is diluted by the nonparticipant households.
In addition, livestock holdings increased at the household level, an average for all the participant and nonparticipant households, in Saving for Change villages. These households owned on average US$120 more in livestock than households in control villages, a 13 percent difference. Owning livestock in Mali and so many other developing countries is a risk-mitigation strategy and safety net—to maintain household food security.
Monitoring Program “Impact”
When Freedom from Hunger’s Barbara MkNelly did the impact research in Ghana and Bolivia, we were eager to see evidence of impact, of course. We were also interested to see which of the dependent variables we measured in the research project would be most impacted by program participation, at the same time as being relatively easy to measure, with a view to selecting variables we could use to monitor in all programs as part of a standard process of progress tracking and quality control; i.e., a monitoring system. For both reasons, household food security was the clear favorite over other options such as household income/consumption, women’s self-confidence and nutritional status of children under the age of five years. Food security seems to be quite sensitive to program participation, whether Credit with Education or Saving for Change or variants of these designs. And once we developed the Food Security Survey (FSS), thanks to the work of Hugo Melgar Quiñonez (a nutrition scientist, then at the University of California, Davis) to adapt the similar U.S. Department of Agriculture (USDA) scale for use in developing countries, we had a relatively easy-to-use measurement tool.
After years of delay, Freedom from Hunger finally put a systematic food-security monitoring program in place with its partners, for both credit-led and savings-led programs, starting in 2006. Thanks to the persistence and skill of Bobbi Gray and Megan Gash, we now have the percentage of food-insecure households in a sample of incoming clients/members for 13 microfinance institutions and Saving for Change-facilitating NGOs in nine countries. The average percentage of food- insecure is 61 percent, ranging from 100 percent in Mali to 17 percent in Ecuador. Where we have comparable data for samples of both incoming participants and veteran participants (in the program for 1–3 years), either from the same moment in time or in a before-and-after comparison of the same cohort after three years, we see declines in the percentage of food-insecure households averaging 14 percentage points for 8 programs (ranging from 0 to 38 points). We also have one increase in food insecurity from 53 to 79 percent of sampled households—puzzling over this one deepened our understanding of the sensitivity of this food-security measurement in relation to seasonal changes and the surrounding environment. Coincidentally, the cleanest comparison we have, of a cohort of participants in Ecuador sampled in 2007 and again in 2009, shows a decline of the percentage of food-insecure households from 47 to 34 percent—13 points, which is very close to the average shown above.
To be sure, this is a crude form of program “impact” monitoring. It does not “prove” impact, but it is anchored in solid impact research results obtained by Freedom from Hunger and other researchers in other settings. And it shows movement in the direction predicted by theory tested by field research in several settings, which is reassuring that these programs seem to be on the right track and also that the few field research studies and their results seem to have external validity. Movement counter to the predicted direction stimulates deeper investigation into probable causes. This is what a monitoring system is supposed to do. The overall take-away from both the few formal research studies and the more widespread but shallow monitoring efforts is that increased food security is very likely a reliable and widespread impact of participation in microfinance programs, both credit-led and savings-led.