One of the most frequent causes of credit constraints is the presence of asymmetric information between businesses and investors. Asymmetric information is particularly problematic for micro-entrepreneurs where the information about cash flows and investment decisions is not formally recorded. Furthermore, micro-entrepreneurs many times have few assets to pledge as collateral and do not have a guarantor with a solid financial condition, making it even more difficult for them to access the credit market.
Microfinance institutions specialize in lending to these types of borrowers and have lending technologies that do not rely on formal records. Instead, revenues and expenses are estimated based on non-verifiable information collected by loan officers during field visits to the borrowers’ houses and businesses.
During these field visits, loan officers observe the premises of the business, the inventory, and other relevant information the borrowers can demonstrate. They also discuss business matters with the entrepreneurs as well as personal matters that might affect their repayment capacity.
Loan officers’ expertise is crucial to estimate the financial health of a business during these short visits. For example, experienced loan officers are able to estimate businesses’ inventories and revenues by observing key variables, including the products on the shelves or the number of clients that show up at the business during the visit. Although these observations cannot be verified and are considered soft information, the types of skills that lead to such information can be acquired through training, are not loan officer-client specific, and can be applied even if the loan officer is not acquainted with the entrepreneur.
However, an important fraction of the information required to make a microfinance lending decision is private and is collected during the social interactions between the loan officers and the entrepreneurs. The flow of this type of information strongly depends on the interpersonal ties between the borrowers and the loan officers (Uzzi, 1996). For example,it is unlikely that borrowers would disclose their health expenses, alimony expenses, or other expenses they incur to support family members in need to a stranger. Therefore, this type of information is lost when a loan officer leaves the bank, unless the private information can be credibly transferred to a new loan officer, and/or if the departing loan officer can convince borrowers to share “personal information” with the new loan officer.
The social relationship between the loan officers and the borrowers not only helps the bank to make better lending decisions, but also might increase the willingness of the borrowers to get debt. This is particularly important for borrowers that associate a negative connotation with debt, are unfamiliar with financial services, or mistrust financial institutions. On the downside, making lending decisions based on these social interactions makes banks dependent on loan officers and subject to their misinterpretation or misreporting of information.
While it is recognized that the social relationship between the loan officers and the entrepreneurs can have important implications for the lenders and the borrowers, little is known about the costs associated with disruptions to these relationships.
In a recent study, we test the importance of interpersonal relationships in the lending process. In particular, we study whether the banks’ lending decision and the borrowers’ repayment rate, willingness to get debt at the bank, and willingness to get debt at other banks are affected when a loan officer is absent for long periods of time.
We find that the relationship between loan officers and borrowers has first-order implications on entrepreneurs’ credit availability, repayment behavior, and borrowing decisions. When loan officers are absent, we observe a 20% decrease in the probability that clients get a new loan. This reduction is the consequence of both a 5% decrease in the bank’s loan approval rate and a 15% reduction in the number of applications. We do not observe a change in credit terms such as interest rates or maturity; this indicates that the bank adjusts the risk by cutting credit and not by adjusting the price of the loans. We also observe a 22% increase in the probability of missing a payment and an 18% increase in the probability of default for the borrower.
To understand the conditions in which this information can be transferred or generated by the new loan officer, we look at variations in: (1) how well the absence of loan officers can be planned in advance, since it is more difficult to transfer soft information in the case of completely unplanned leaves, and (2) whether the departing loan officers have any incentives to collaborate in conveying information to replacement loan officers. We observe four different types of absences due to sickness, resignation, pregnancy, and dismissal. We use sickness leaves as a baseline, because they are both unexpected and exogenous.
During sickness leaves, we still observe a decrease in lending and an increase in delinquency, but we do not see an increase in outright default. This finding indicates that most payment delays are caused by reduced monitoring and not caused by financial distress. We also observe a decrease in the probability of applying for a new loan at the bank, but an increase in the probability of applying for credit at other banks, suggesting a reduction in the loyalty of clients toward the bank.
We observe similar results during maternity leaves, but we do not observe an increase in the probability of approaching other banks. This is natural since maternity leaves are exogenous but anticipated, and the loan officers can set up their clients with new loans before they leave.
The strongest reduction in credit and the strongest increase in default are observed after loan officers are dismissed. We believe this is the result of poor past performance of the dismissed loan officers as well as a lack of incentive to transfer any soft information.
However, clients of resigning loan officers are less affected by the leave; application probability does not present a significant decrease and the default rate does not increase. This might indicate that when given enough time, loan officers can brief the replacing loan officers about the soft information of the clients. It also suggests that when having the right incentives, the departing loan officers can familiarize the new loan officers with the clients and gain their confidence.
We also study whether the importance of social relationships depends on borrowers’ characteristics. As expected, the probability of the bank approving a loan to firms with larger and better credit risks less affected by loan officer leaves, probably because hard information about the clients is more readily available. However, we also observe that the probability of applying for a new loan at the bank and the probability of applying for a new loan at other banks do not change after the leave. This finding can indicate that the relationship between clients with large and good credit scores and the bank is less dependent on loyalty.
We also find that female clients are more affected by loan officers’ absences, which is probably because female clients have fewer assets and many times operate informal businesses from home. Therefore, cash flows are particularly difficult to verify and personal information is particularly relevant. This finding highlights the importance of micro-credit lending on promoting women’s financial inclusion.
Overall the results in the study support the view that personal relationships are crucial to reduce credit constraints and improve entrepreneurs’ incentives to repay. The results suggest that close social ties between the loan officers and the borrowers can increase the offer of credit to micro entrepreneurs, but also indicate that the demand for credit can depend on social ties. In light of these results, the high turnover observed in the loan officers’ labor market can be very costly for banks and borrowers and can be one of the factors that impede many micro-businesses to grow beyond subsistence level.
Drexler, Alejandro, and Antoinette Schoar. 2012. Do Relationships Matter? Evidence from Loan Officer Turnover. Working Paper.
Uzzi, Brian. 1996. The Sources and Consequences of Embeddedness for the Economic Performance of Organizations: The Network Effect. American Sociological Review pp. 674–698.