The Kshetriya Gramin Financial Services (KGFS) model since its inception has focused on providing a range of high quality financial services through its geographically-focused community financial institutions using a customised wealth management approach.
Utilising an experimental approach based on randomised rollout of KGFS branches in Pudhuaaru, Vellaru and Thenaaru in rural Tamil Nadu, an impact evaluation is presently being conducted by researchers from Harvard and Duke University in partnership with the Centre for Microfinance and IFMR Rural Channels to rigorously evaluate the impact of rural bank branch expansion at both the household and village levels. Specifically, this unique evaluation will identify how household financial behavior (including household borrowing, savings, purchase of various financial instruments etc.), livelihood choices, agricultural decision-making, social networks and stress related health outcome are impacted by the provision of comprehensive financial services through a localized financial service provider like KGFS.
Out of 160 proposed study sites, KGFS branches are being rolled out in one half (i.e. the treatment group) while the other half (i.e. the control group) will receive KGFS branches after the completion of the study. The assignment to treatment has been done such that each treatment branch has been matched with a control branch based on a variety of geographic traits mitigating the effects that may result from seasonality or geography rather than the introduction of banking services.
Before the rollout of each KGFS branch, a baseline survey is conducted; in total, the study which will cover approximately 6,800 households in the experimental area including 3,400 households in the treatment group and the remainder 3,400 households from the control group. To date, approximately 1300 households have been surveyed.
Among the baseline households which have already been surveyed, a follow-up survey of 673 households was conducted between October 2011 and January 2012 covering eight treatment branches where Pudhuaaru KGFS (PKGFS) offered financial service starting 2010 and corresponding eight control areas. The data from 673 household in the follow up survey was used to compare the borrowing patterns of the areas served by Pudhuaaru KGFS with the respective control areas without Pudhuaaru KGFS. The findings from this follow up survey as published in a recent policy memo present interesting results on the impact of Pudhuaaru KGFS on the borrowing patterns in the treatment areas as compared to the control areas.
This post presents the main findings that researchers identified in the policy memo.
Impact on the Borrowing from Formal Sources1
Comparison of formal borrowing (from scheduled commercial banks, MFIs, Self Help Groups and credit cooperatives) between treatment and control groups (Figure 1) indicates that the average outstanding amount of formal borrowing per household in areas served by a Pudhuaaru KGFS branch is 38% higher than the areas without a Pudhuaaru KGFS branch. Also, the average number of formal loans outstanding per household in the treatment branches is approximately 20% more than in the control branches (Figure 2). Comparing the average amount of borrowing from formal sources including repaid and outstanding, it seems that the average amount of formal loans repaid by the treatment households was almost twice more than the control household (Rs. 7264 by the treatment households as compared to Rs. 3628 by the control). Data on average number of formal loans repaid in the past year indicates that 28% of households in the treatment area have repaid formal loan as compared to 20% in the control areas. These findings clearly demonstrate that the extent of formal borrowing increased in the branch service areas as compared to the similar areas that do not have a KGFS branch. Thus opening of Pudhuaaru KGFS branches positively impacted households’ borrowing from the formal sources.
Impact on the Borrowing from Informal Sources
Comparison of borrowing from informal sources (Figure 3) which include friends, relative, neighbours, shopkeepers, moneylenders, pawnbrokers, landlords and employers shows that average informal borrowing per household in the treatment branches is significantly lower than the average informal borrowing per household in the control branches. As shown in Figure 3 the average amount of informal outstanding per household in treatment areas is 19% lower as compared to the control areas, while the average amount of informal loans per household including outstanding and repaid is 35% lower in treatment areas as compared to the control areas.
The comparison of total loan outstanding in the treatment and control samples (Figure 4)2 indicates that total amount of borrowing from moneylenders in the treatment areas is 11% lower than in the control areas, while the total amount of borrowing from friends, relatives and neighbours in the treatment areas is 22% lower than in the control areas. The results also show that not only the amount of indebtedness, but also the proportion of households reporting any informal loan outstanding is significantly lower in treatment areas (63% with any informal outstanding) as compared to the control areas (71% with any informal outstanding). So, these findings suggest that the opening of Pudhuaaru KGFS branches has significantly reduced the proportion and amount of households’ indebtedness to moneylenders, friends, relatives and neighbours.
Combining the findings related to formal and informal borrowing, the results provide strong evidence that arrival of Pudhuaaru KGFS branches has significantly shifted households’ dependence from informal sources of loan to formal sources of loan. Thus, the interim results from this impact assessment study clearly demonstrate that geographically-focused community financial institutions like Pudhuaaru KGFS can play very effective role in reducing households’ dependence on informal sources of borrowing by providing effective access to formal sources of finance.
- Households were asked to only report loans greater than Rs.2,000 so these estimates are likely to underestimate the total effect on borrowing.
- The formal loan in figure 4 excludes SHGs as it is separately categorized in the figure.