Transforming Indian agriculture: The role of credit policy
14 Dec 2015
Despite various policy attempts at priority sector lending to poor farmers, very little progress has been made on the ground, suggesting problems in the design and implementation of these policies. In this article, Amartya Lahiri and Dilip Mookherjee and explore where the problem really lies and what can be done to address the issues.
Promoting financial inclusion among poor farmers is a high priority for Indian policymakers - for good reason. Most rural poor are excluded from the ambit of the formal financial system, so have to rely on local moneylenders to borrow at high rates. This causes financial distress, besides limiting their ability to raise productivity, grow high-value crops and improve their own living standards. Ultimately, growth in the Indian economy will depend on higher agricultural productivity. Besides raising agricultural output and limiting food price inflation, this is necessary if workers have to transit out of agriculture to higher-productivity occupations in manufacturing and services.
Despite various policy attempts at priority lending to poor farmers, very little progress has been made on the ground. Farm surveys show that access to bank loans is limited to those able to offer collateral (land or other assets) that landless and marginal landowners do not possess. For instance, a survey of such farmers in two districts of West Bengal during 2010 by Maitra et al. (2015)
showed only 5% of agricultural loans were from banks, 24% from cooperatives, 65% from informal lenders, and the remaining 5% from family and friends. Bank and cooperative loans required collateral and charged interest rates of 12% and 16% respectively. Informal loan rates were much higher at 26% but mostly involved no collateral. Those obtaining loans from microfinance institutions (MFI) or self-help groups (SHG) paid interest at rates closer to informal loan rates. In West Bengal these were around 25%. All-India figures for MFI interest rates released by the 2011 Malegam Committee range from 30-50% with an average of 37%.
These facts raise two sets of questions. First, of the few able to access bank or cooperative loans, why are they paying interest rates of 12%, on par with market rates charged to other bank borrowers? Why are MFI loans even more expensive? After all, the government has been making efforts to subsidise lending to poor farmers, so they should be accessing credit at below market rates. Second: why are poor farmers unable to access loans from banks, and thereby forced to pay interest rates that are almost double the bank rate from informal lenders or MFIs?
The inability to translate intent into reality suggests problems in the design and implementation of priority sector lending policies. Understanding this is a complex task. In this article, we try to figure out where the problem really lies, and what could be done to get around these problems.
Direct bank loans
Let us start with direct bank loans. Most of the priority lending initiatives have centred on interest rate subvention schemes that aim to make credit available to farmers at relatively low interest rates. The best known scheme (which has been around since 2006-07) gives banks a 2% ‘subvention’ on their short-term crop loans to farmers up to a loan amount of Rs. 300,000; they are mandated to charge an interest rate of 7% on these loans. There is a further 3% subvention to farmers who repay their loans on or before the due date of the loan in subsequent years, when the banks are supposed to charge only 4%. This scheme was initially only available to public sector banks but has been extended to private sector banks as well since 2013-14.
Another scheme in place aims to prevent distress crop sales by farmers after production. Loans to store post-harvest produce in warehouses by small and medium farmers with Kisan cards1 are eligible for interest rate subventions so that such loans are made available at an interest rate of 7% per year. This scheme is along the same lines as the subvention scheme on crop loans except that it is on loans no longer than six months and there are no additional subventions for timely repayment. This scheme only applies to farmers who took crop loans.
However, these policies are clearly not working. It is evident from the survey evidence cited above that most poor farmers are not getting bank loans, and those that are, are paying interest rates far higher than those mandated by the subvention policy. Moreover, agricultural lending is only about 13% of overall institutional lending in India in 2015.
We speculate that the reason the subvention schemes are not working is that the mandated interest rates are set too low relative to the costs that banks incur in making such loans. A similar view has been expressed by experts such as Nachiket Mor, chair of the Reserve Bank of India (RBI) Committee on Comprehensive Financial Services for Small Businesses and Low Income Households (Mor and George 2014
). If banks can obtain capital at a base rate of 4-5%, they incur transaction costs of the order of at least another 4-5%. These include staff and overhead costs involved in processing loan applications, monitoring borrowers, collecting repayments, besides allowance for bad loans. Hence, a bank needs to charge at least 10-11% in order to break-even. If they were to make loans at the rates mandated by the subvention law, they would lose money. Even public sector banks nowadays are under constant ‘bottom line’ pressure to maintain healthy balance sheets. Since granting loans is ultimately at the discretion of bank officials, they end up making very few loans directly to poor farmers. Of the few that are extended, the interest rate that is charged is large enough (while deviating from the subvention mandates) to enable the bank to break-even.
Lending by non-bank financial intermediaries
How then does 13% of institutional lending to the agricultural sector get channelled? A large share of priority sector lending from the banks goes indirectly through various financial intermediaries: non-bank financial companies (NBFCs) which mainly comprise MFIs, besides regional rural banks and cooperatives. The main attraction of the MFIs is that they typically make non-collateralised loans in contrast to commercial banks which require collateral for all loans above Rs. 100,000. Moreover, given the group liability structure of their loans, MFIs are often able to get around information and other problems that tend to make repayments of non-collateralised loans trickier for standard commercial banks.
However,as noted above, MFI loans do not come cheap: most MFIs charge interest rates ranging between 30-50%. Such high interest rates and allegations of coercive loan collections by MFI officers gave rise to the microfinance crisis in Andhra Pradesh (AP) a few years ago. In the wake of that crisis, the Malegam Committee set up by the RBI investigated the structure of costs incurred by MFIs. For loan capital they pay the banks around 12%. Staff costs and overheads amount to 14%, and provisioning for bad loans costs 2%, amounting to a total cost of 28%. No wonder, then, that MFI interest rates are upwards of 30%, since they have to break-even. On average they earn a profit margin of 9-10%, so the average interest rate charged is 37%.
The other problem with credit disbursed by MFIs is that they do not seem to make any dent in raising agricultural productivity, as shown by a large and growing body of evidence (Banerjee et al. 2015
). Part of the reason for this is that MFI loans typically require regular (monthly) repayments and are usually very short term. In addition, the loan terms entail regular group meetings and oversight which tend to raise the administrative cost of these loans. These conditions tend to discourage long-term investments as well as relatively risky ventures which might otherwise have high expected returns. Besides there is no mechanism to ensure that only high-productivity borrowers apply for these loans.
In light of the AP microfinance crisis, the Malegam Committee recommended a cap on MFI interest rates of 24%, and a 10% cap on margins above the cost of capital. Clearly, most MFIs will not be able to break-even if these caps were to be enforced. Whatever credit has been flowing to the priority sector will then dry up. And even if somehow the MFIs managed to tighten their belts and lend viably at 24%, borrowers will still be paying interest rates close to the high rates they are already paying in the informal sector. Perhaps being aware of this, the Malegam Committee recommended a return to direct lending by banks, through SHGs, with the loans being mediated by banking correspondents and banking facilitators (BC/BFs). Yet it did not issue any explicit guidelines how such BC/BFs were to be appointed, or incentivised by suitable commissions. In practice we believe such lending constitutes only a small trickle. We suspect a key reason is that direct bank lending to poor farmers remains constrained by the subvention laws: bank officers have no incentives to lend directly at the mandated interest rates.
Herein lies the Catch-22 characterising rural sector lending policies. Direct lending by banks is hamstrung by the subvention laws which mandate interest rates that are so low that the banks would incur losses. And lending through NBFCs such as MFIs is constrained by the high costs the MFIs incur in administering the loans, and are likely to be even more constrained if the RBI were to implement the Malegam Committee recommendations. Hence, policies though well-intentioned have failed, and will continue to fail to deliver.
What are the likely solutions? One route is to do away with the subvention restrictions on interest rates that banks must charge poor farmers. Then banks would be free to set interest at rates that cover their costs. Competition among banks, if sufficient, would eliminate efforts at rate-gouging. Otherwise, to control interest rates, regulators could set higher ceilings on margins above costs of capital, which would help banks recover their lending costs. Our discussions with some senior bank officials with experience in rural lending suggest these costs would be at least of the order of 7-8%. If banks could access priority sector funds at 5%, they ought to be able to lend profitably if allowed to charge interest rates up to 15%.
The other alternative is to channel rural credit through non-bank intermediaries such as MFIs. However, the most cost-efficient of these institutions incur transaction costs of the order of 7-8%. So if they could also access priority sector funds from banks at roughly 5-6%, the rate at which the banks access these funds, MFIs would also be able to break-even if they charged borrowers 15%. We do not see any reason why the banks should charge market rates of 12% to MFIs, while accessing priority sector funds at 5-6%, and pocket the difference.
Either way, policymakers have to acquire a realistic sense of the costs of delivering institutional credit without collateral to poor borrowers, and permit interest rates of the order of 15%. This has the potential to raise the volume of credit substantially to the rural sector, at rates still considerably below what they are currently paying to informal lenders.
The discussion above highlights a more general issue with imposing artificial ceilings on interest rates. These policies often lead to financial repression due to credit being rationed by lending agencies in response. A more fruitful approach might be to give subsidies to banks to lower their cost of funds and then cap the margins that they can charge over their subsidised cost of funds. Farmers are willing to pay higher interest rates than the subvented rates mandated by policies. Incentivising lenders to provide farmers with credit is key. The subventions have just tended to squeeze their access to formal credit.
The other part of the problem is to find ways to target credit delivery to truly productive farmers, and lend in a way that does not stifle their ability to diversify into high-value crops. BC/BFs need to be appointed from those with experience in lending within local communities, and adequately incentivised with commissions linked to loan repayments. Individual liability loans with loan durations matching high-value crop cycles need to be offered. That such schemes can be successful in increasing agricultural productivity have been shown in the West Bengal context in a field experiment by Maitra et al. (2015)
. There is an urgent need for such new initiatives to ultimately realise the objective of raising agricultural productivity and living standards in rural India.
A shorter version of this article was earlier published in Times of India.
- Banerjee, Abhijit, Dean Karlan and Jonathan Zinman (2015), “Six Randomized Evaluation of Microcredit: Introduction and Further Steps”, American Economic Journal: Applied Economics 7(1):1-21.
- Maitra, P, S Mitra, D Mookherjee, M Torero, and S Visaria (2015), ‘Asymmetric Information and Middleman Margins: An Experiment with West Bengal Potato Farmers,’ Working Paper No. 2015-23, Hong Kong University of Science and Technology.
- Mor, N and D George (2014), ‘Strengthening rural lending’, Special article, Yojana – A development monthly, March 2014.