Sebstad and Cohen identified three main pathways through which microfinance services can reduce vulnerability: income smoothing, building assets (including financial, physical, human and social assets), and empowering women.
In my last post (# 37), I concluded that microfinance increases opportunities to diversify sources of income, but we lack evidence-based confirmation that this leads to better income smoothing through the year for microfinance clients. Remembering that lack of evidence of impact is not evidence of lack of impact (echoing David Roodman), I believe this lack of confirmation of income smoothing comes from the difficulty of measuring income variability (requires a time series of data), which therefore is seldom measured.
What about building assets? Certainly physical assets are much easier to observe and quantify. Financial assets are harder to get at, but they too can be inventoried at one point in time and compared to the inventory at a later point in time. Human and social assets are often ill-defined, but if defined clearly, they can be observed with some reliability. In this post, I start with the easiest—physical assets.
Loans are often used to build an enterprise’s or a household’s inventory of physical assets. Sebstad and Cohen identify two types of physical assets—“productive assets,” which can be used to generate income without actually selling the asset itself, and “other household physical assets,” which only generate cash when sold off.
Regarding “productive assets,” Sebstad and Cohen cite a review of ten microcredit impact studies that looked for changes in enterprise assets. In seven of ten countries, there was positive change in the value of fixed assets among borrower enterprises. In another program studied in Malawi, one quarter of client enterprises increased their assets, one half had no change, and one quarter had no assets to begin with, which Sebstad and Cohen assert is a general pattern of variation in asset change. Yet another study of a program in the Philippines found that the value of livestock and accumulation of capital in machinery, tools, and equipment went up substantially with the number of loans. Research on BRAC members in Bangladesh found that successive loans lead to a buildup of assets over time and that the structure of assets shifts in favor of more productive assets. In general, asset accumulation seems to increase with the cumulative value of loans taken, generally associated with duration of participation in the microfinance program.
Freedom from Hunger’s research in Mali (reported by MkNelly and McCord, p. 14) similarly found two-year clients were significantly more likely then incoming clients to have acquired business assets in the prior 12 months. Yet, when the responses of the one-year clients are combined with those of the two-year clients, there were no significant differences compared to incoming clients, indicating that a certain duration of program participation is required before impact on enterprise assets will occur (but see David Roodman’s caution regarding this kind of reasoning from non-randomized assignment of duration of access to the financial service).
In his review of seven recent RCTs, David Roodman found evidence of increased investment in enterprise (including business inventory and durable assets) in four of five microcredit programs and two of three microsavings programs. Specifically, the RCT study of the MFI Spandana’s clients in Hyderabad, India found that average spending on durables used in a business was 140% higher in treatment areas compared to control areas.
MkNelly and McCord cite Michael Kevane’s in-depth qualitative research in Burkina Faso which found that many women clients of a credit-union-managed Credit with Education program made significant investments in items such as cooking pots, storage containers and their marketing sites, despite the small scale of their enterprises. They also point out that two-year clients of a Credit with Education program in Mali were significantly more likely than incoming clients to report owning a pasta-making machine (field agents explained that women value these machines (with an estimated cost of approximately $40) both for their enterprises and for home use). All of the two-year clients owning a pasta machine had acquired it since joining the program.
The RCT study of the MFI Al Amana in Morocco (linked in Roodman’s review) found that greater access to microcredit led to expanded livestock activities but mostly from an increase in the stock of livestock. Similarly, in the study of CRECER in rural Bolivia (see MkNelly and McCord review), Credit with Education participants were more inclined than non-participants to invest in animals, either for their families or to fatten and sell. One-third reported using loan capital to buy animals such as sheep, pigs, cows and bulls, which are perhaps the most important stock of wealth on the Altiplano.
Regarding “other household physical assets,” the Bolivia study illustrates the blurring of the distinction from “productive assets”—animals can be consumed by the household or used to generate regular income for the household or held as a store of wealth for sale only when lump sums of cash are needed. The research literature tends to focus on enterprise assets even when these may also be used for household needs. The Spandana study makes a distinction between all “durables” and “durables used in a business,” finding that average spending on all durables was 19% higher in treatment areas than in non-treatment areas, but durables for business averaged 140% higher, as stated above. In other words, microcredit access was shown to have a statistically significant effect on asset accumulation (within only about one year to 18 months) but much more so when the household is engaged in business.
An RCT study of the MFI XacBank in Mongolia (linked in Roodman’s review) compared the impacts of group lending vs. individual lending offered to women. Women who obtained access to microcredit often used the loans to purchase household assets, in particular large domestic appliances. However; only women offered access to group loans were more likely to own enterprises (and therefore enterprise assets) than the control women; not the women offered access to individual loans. I will return to this study in later posts, because of its implications for Freedom from Hunger’s Credit with Education strategy.
Regarding saving groups independent of external lenders, I introduced in post #31 the excellent survey of emerging impact research results by my Freedom from Hunger colleague, Megan Gash. She has written a chapter (“Pathways to Change”) for the forthcoming book Savings Groups at the Frontier that comes out of the Arusha Savings Group Summit (October 4–6, 2011 in Arusha, Tanzania), expected to be published next month (November 2012). Here is Megan’s summary of results regarding asset accumulation:
Asset accumulation is a very common finding among the studies reviewed, although type varies and seems related to length of membership. In Kenya, researchers found that newer members buy inexpensive items such as wrappers (lessos), cooking utensils, chicken and goats, and few members were able to report a solid growth in physical assets, likely due to being in the program for only a year. A study in Burkina Faso looked at two cohorts of members, one in the program for one to two years and another for two to three years, and saw some differences in their asset growth. Many purchased both productive and non-productive assets, such as a goat, sheep, poultry, agricultural inputs, utensils, storage trunks and festival attire (dresses), but those in the older group showed more purchases of expensive items such as bicycles and jewelry. Reasons behind asset accumulation can vary, with some members buying them for immediate use and others as a long-term savings strategy (something to be sold at a later date), or even to assure the ability to repay loans to the group in the future. It would be logical to see that as members accumulated larger savings over time, or took larger loans from a greater pool of money, they could purchase more expensive assets. Since evidence of asset accumulation was noted in several of the studies, there is a high likelihood that savings group participation will result in asset accumulation.
From the evidence presented above, it seems very likely that Megan’s conclusions about asset accumulation by members of savings groups apply equally to members of credit groups. The evidence is quite solid that microfinance participation increases ownership of physical assets. Whether they be for income-generating purposes, household use or storing wealth for future liquidation, these physical assets provide insurance against emergency demand for cash and a means to save for expected lump sum expenses in the future.