Information asymmetries and agricultural productivity in developing countries

Mar 19, 2020

Introduction

Asymmetric information is quite damaging for credit and insurance markets in developing countries. Farmers, in particular, face unique challenges that make these risk mitigators essential for production to be sustainable. I argue that lenders and insurers may employ a variety of strategies—collateral and group lending, for instance—to reduce moral hazard and adverse selection among borrowers, and policy’s role is, at best, minor and, at worst, indeterminate in resolving these problems. First, I explain why farmers direly need these financial services. Then, I discuss problems posed by information asymmetry in insurance and credit markets. Finally, I discuss potential roles of policy suggested in the literature.

Difficulties Faced in Agriculture

Farmers in poor nations face tremendous risks that promote agricultural inefficiency and precariousness in livelihood. One such challenge is that their farmland is rarely irrigated, which makes production highly dependent on weather. Droughts may cause crop failures, and six months’ of labor may vanish. A powerful illustration of this point is that India’s agricultural wages are twenty-one times more variable than wages in the United States, where insurance and subsidies help to counteract risk. Furthermore, global food prices fluctuate dramatically. They were quite high between 2005 and 2008 but dropped precipitously during the global financial crisis[3]. The confluence of these risks leaves poor farmers vulnerable.

As such, farmers are disposed to risk-mitigation behaviors to stabilize their income. They diversify earnings by splitting time between occupations. Among 27 villages in West Bengal, farming households only reported spending 40% of their time farming. The median family had seven occupations between three laborers. Moreover, families farm multiple plots of land in different locations so that if one is affected by blight, others may survive. Farmers may also manage enterprises conservatively. For instance, they might abstain from investing in new technology because of the high cost and risk of crop failure. Finally, shared land tenancy arrangements are common in developing countries. In these cases, landlords gain a share of the output by absorbing some initial farming costs[3].

Behaviors such as these create tremendous inefficiencies. By splitting time between undertakings, laborers cannot specialize and achieve gains in efficiency that are realizable by farming full-time. Furthermore, technological improvements are needed if yields are to be augmented. And by consolidating efforts to one large plot of land, one can focus on producing one or two crops abundantly rather than many meagerly. Shared tenancy also introduces poor incentives; because landowners may claim half of the output, farmers may be less motivated to work hard. Banerjee and Duflo (2009) reference a study in India that found that farmers invest 20% less effort into shared undertakings than proprietary ones[3]. For these reasons, agricultural productivity demands that risks are managed such that farmers may be free to specialize and invest optimally. In the following sections, I discuss how insurance and lending can serve this purpose and why information asymmetry creates failures in these markets.

Information Asymmetry in Insurance and Market Solutions

Insurance serves an important function in mitigating farming risks. Without it, farmers are averse to the risk of crop failure and are, therefore, inclined to act conservatively and inefficiently. With it, farmers may be assured of their financial security, drought or not, and are free to invest in productive capacity.

Information asymmetry complicates this mission for a few reasons. One issue is moral hazard. Insured farmers may take irresponsible risks in addition to prudent ones because they do not bear the full consequences of their actions. Another problem is adverse selection. Those facing the greatest risks are most likely to want insurance, and insurance-seekers are more aware of their own risk profiles than insurers can be. Insurers may account for this possibility by raising premiums, but to do so would drive away safe customers who might still benefit from being insured. Finally, claimants might defraud insurers. For example, a farmer may issue a claim for the death of insured livestock, and insurers may be unable to assess its veracity. Some insurers have developed innovative strategies to mitigate these possibilities. Weather-related insurance, for instance, may pay a fixed amount based on the premium paid and precipitation levels. Similar strategies may be necessary for other forms of agricultural insurance to be profitable[3].

Information Asymmetry in Lending and Market Solutions

Another type of information asymmetry concerns the relationship between lenders and borrowers. Lending is important for agricultural markets for a few reasons. First, there are significant upfront costs in any profitable farming endeavor. Money is needed to invest in seeds or other agricultural inputs such as fertilizer. Profits are only realized when crops are finally harvested and sold. Lending, therefore, helps to front initial costs when savings are sparse. Second, lending helps to counteract risk aversion. Because farmers’ savings are limited, they may be unwilling to spend funds on technological upgrades, which have a delayed and uncertain payoff. However, lenders may provide the needed injection to intensify farming efforts and improve efficiency. Because the money being spent is not their own, farmers may be more willing to take required risks[7]. Therefore, for poor farmers, lending is an indispensable source of capital.

In deciding whether to extend a loan, asymmetries may arise at three levels. First, prior to granting a loan, the lender may be unable to assess the borrower’s reliability in repaying it. This difficulty in risk assessment may arise either from the inability to acquire such information or to interpret it in context. Second, after a loan has been extended, a lender may be unable to verify that the loan is being put to good use. Because it is not the borrower’s own money that is at stake, he or she might work less hard to ensure its productivity. Third, after the loan has been invested, the lender may be unable to ascertain its profitability and demand repayment. Misleadingly claiming that returns were low, borrowers could withhold the owed amount, and lenders may be unable to prove otherwise[2].

Given information asymmetries, lenders may act in a few ways. One is to demand collateral—for example livestock, land, or housing. Many poor individuals may not have collateral that can be easily seized (untitled land, for instance), and even when it can be, lenders with poverty alleviation objectives may be reluctant to seize it[7, 2]. I return to the topic of collateral when I discuss policy implications.

Another strategy is to charge high interest-rate premiums. This tendency can be modeled using the lender’s risk hypothesis, as Ray (1998) shows. Let p denote an exogenous probability of default for every dollar loaned. L is the total amount loaned, r is the opportunity cost of funds for the lender, and i is the interest rate charged. Assume that competition between lenders causes interest rates to be reduced to where each lender earns zero profit. Profit, then, would be as follows[7]:

If r = 10% and p = 50%, then, irrespective of L, i would be set at 120% per annum—extremely steep. It is clear that lending markets are extremely sensitive to the risk of default. Empirical evidence supports this conclusion. A study by Aleem (1990) in Chambar, Pakistan, found that interest rates among sixty lenders varied between 18 and 200% per year with an average of 79%. Defaults occurred on less than 5% of loans. This tendency reflected lenders’ habit of collecting evidence about individuals and employing “testing loans” before lending substantial amounts[1]. While lenders have had success with high interest rates, there are still many rural workers with insufficient access to credit, in part due to high premiums. There are other problems with the high premium approach. First, they may discourage good borrowers, the target of many micro-financers. As Ray notes, some loans are imprudent irrespective of the premium because “the premium itself affects… chances of repayment”[7]. Second, high premiums do not necessarily solve the problem of bad borrowers. If intending to default, one has no reason to be deterred by high premiums. Even if intending to repay a loan, an unreliable borrower may later default upon failing to achieve sufficient returns[7]. Clearly, high premiums are not enough to produce an efficient allocation of credit.

Finally, a lender might opt for group lending, an approach developed by Grameen Bank in Bangladesh: multiple borrowers together agree that should one group member default, loans are cut off for all. One advantage of this strategy is its efficaciousness in dealing with enforcement, adverse selection, and moral hazard. Loan-reliant group members have an incentive to do what the lender cannot—verify that their peers are working diligently to repay their loans. Second, in forming their groups, borrowers can sometimes accurately discern the reliability of their peers. As such, group lending, in theory, fosters a decentralized selection process that excludes unreliable borrowers. One concern is that groups could collude to defraud lenders. Also, many potential borrowers may lack reliable peers to form groups with, and intra-group interaction may be lacking for a variety of reasons[2]. As such, group lending, though promising, may be insufficient on its own to guarantee widespread access to credit.

Conclusion: The Role of Policy

One key argument against the role of policy in correcting these market failures is that government has no informational or operational advantage over private firms. Like an insurance provider, governments of poor countries have little insight into the fiscal responsibility or fraudulent tendencies of insurance seekers. Neither can governments verify that borrowers will act prudently in the interest of paying back loans. As such, there is little reason for thinking that government may more effectively provide credit or insurance[6]. The remaining viable roles of government are related to regulation or enforcement, and I consider each in turn.

Regarding regulation, Besley (1994) suggests that policy might have a role in reducing informational monopolies. Informal lenders, for example, may have information about borrowers that other lenders do not and are, therefore, positioned to extract maximal surplus from borrowers by charging discriminatory interest rates. However, it is unclear that this outcome is inefficient or that credit would be undersupplied. Good borrowers may be offered more favorable interest rates than they would receive elsewhere, and access may be more extensive when specific lenders have this information. Regulation may only prove useful insofar as other lenders are able to access the same information, which suggests the possibility that government could centralize information so that all lenders know about borrowers’ credit histories[4]. This task, however, is gargantuan, and it is difficult to imagine how the government of a developing country might approach it. Not only would governments need to identify each citizen, which many cannot do, but they would also have to agglomerate specific details about peoples’ credit transactions including collateral, delayed payments, and group borrowing behaviors. Therefore, it seems infeasible that a government could directly involve itself in the realm of information.

However, an indirect role could be feasible. In the realm of enforcement, government could strengthen property rights, which may improve the effectiveness of collateral in mitigating moral hazard and adverse selection. Besley notes that African governments have taken steps to improve titling. But, property rights are not always easily implementable because of culture or ambiguities in ownership[4]. Moreover, Field and Torero (2006) found that titling in Peru had a minimal impact on credit expansion because the program reduced lenders’ confidence in their ability to seize collateral; governments sided with borrowers for political reasons, and property rights provided the requisite legal protection for buyers to retain collateral[5].

Finally, governments could regulate interest rates by mandating reductions or increases. Evidence is lacking that such a mandate would extend access to credit. Instead, lenders would opt not to offer loans at an unprofitable rate that would attract unreliable borrowers[4].

For these reasons, I conclude that policy has a highly contingent and perhaps minimal role to play in correcting these market failures. In most cases, lenders, insurers, and borrowers must develop their own innovative solutions.

Bibliography

[1] Aleem, I. (1990), ““Imperfect Information, Screening, and the Costs of Informal Lending: A Study of a Rural Credit Market in Pakistan”, World Bank Economic Review 4, no. 3, pp 329-349.

[2] Armendáriz, B. & J. Morduch (2010), The Economics of Microfinance, 2nd ed., MIT Press.

[3] Banerjee, A. & E. Duflo (2011), Poor Economics, Penguin Books.

[4] Besley, T. (1994), “How Do Market Failures Justify Interventions in Rural Credit Markets?”, The World Bank Research Observer 9, no. 1, pp 27-47.

[5] Field, A., Field, E. & M. Torero (2006), “Property rights and crop choice in rural Peru, 1994-2004”, IDEAS Working Paper Series from RePEc.

[6] Jain, S. (2019), “Lecture 11: Financing Development Credit Markets in Developing Countries”, Economics of Developing Countries, University of Oxford.

[7] Ray, D. (1998), Development Economics, Princeton UP.

← Back to all articles