FinDev Blog

Detecting Borrower Distress

A new framework for monitoring credit markets for over-indebtedness
Digging a well, India. Photo by Pranab Basak, 2015 CGAP Photo Contest.

In India, signs of borrower distress have been appearing in many states over the past decades. For example, in the north-eastern state of Assam, delinquency increased from 1.5 to 10 percent between September and December 2019, precipitated by the actions of a few lenders who had been allowing customers to borrow beyond their means to repay. Natural disasters and political events at the time further exacerbated the issue, leading to many borrowers being unable or unwilling to repay. The situation is likely to worsen once the regulatory forbearance associated with the COVID-19 pandemic ends.

Over-indebtedness (OI) is a problem in many other countries as well, including the United Kingdom and other parts of Europe. Thus, there is an urgent need globally to design policies to prevent OI and to manage the distress caused by being over-indebted. Before such policies are designed, however, regulators need ways to identify and measure the scope and scale of the problem.

Measuring over-indebtedness

The simplest and most accurate approach to measuring OI is through a quarterly census study of the entire population. However, this is expensive and impractical. Alternatively, the regulator can seek granular borrower-level data from the providers, then validate such data and estimate OI. This exercise is also impractical and could lead to regulatory overreach.

A set of secondary indicators is therefore needed to detect the prevalence of OI. In a report by Dvara Research, titled Detecting Over-Indebtedness while Monitoring Credit Markets in India, we propose such a framework of indicators, and in our Policy Brief, we highlight a pathway to implement it in India. The figure below presents the component indicators of the framework.

Monitoring and Detection - Component Indicators of the Framework

Monitoring and Detection - Component Indicators of the Framework by Dvara Research


Three levels of indicators + algorithms

Market level indicators provide a macro-level overview of the credit markets, including the productivity of credit and the interactions between demand and supply of credit. These indicators help to contextualize those activities of providers that are directly related to the task of estimating the prevalence of OI.

Provider level indicators present the next level of granularity. Features of the providers' asset books shed light on over-lending. For example, providers’ uninhibited growth in saturated areas could inadvertently lead to OI. Instances of a very high number of loans per borrower or rampant renegotiation of credit contracts can also help regulators narrow their focus, and if needed, scrutinize specific geographical areas and product segments.

Borrower level indicators serve as the most direct source of information on OI. Here we propose a mix of ex-ante and ex-post indicators. Ex-ante indicators, such as those related to borrowers’ repayment capacity, help in estimating the likelihood and extent of OI in a specific region. Ex-post indicators, like the number of insolvency and bankruptcy cases or the number of complaints to the regulator, shed light on the prevalence of OI, if any.

The proposed indicators are expected to enhance the market monitoring capabilities of the regulator, but are not sufficient to measure the prevalence of OI when considered alone. Instead, appropriate statistical tools (algorithms) are needed, into which the proposed indicators act as inputs. Periodic sample surveys are also needed to validate such algorithms and fine-tune them. However, the first step would be to capture these indicators comprehensively.

Collecting the necessary data

Internationally, several of these proposed indicators are already captured in some of the studied jurisdictions. For example, the Bank of Zambia’s Credit Market Monitoring Programme (CMMP) requires financial service providers to submit quarterly data disaggregated by borrowers’ income category, on credit demand, disbursement, purpose and geography, among others. The 2018 CMMP report demonstrated that a combined analysis of these data sources can be pivotal in anticipating household financial distress.

Similarly, in Peru, the financial sector supervisory authority requires multiple monthly reports from financial institutions. For example, the Debtor-Creditor Report (RCD) contains information on credit balances at the borrower level and contains a field on the “over-indebtedness condition” of the borrower. Other required reports include sector concentration reports, which indicate lenders’ exposure to various economic sectors, and over-indebtedness reports, which indicate clients’ over-indebtedness levels.

To operationalize the framework in India, it is likely that changes will be needed to supervisory reporting formats, and consequently to credit providers’ IT infrastructure. However, most of the proposed indicators (or their underlying components, such as the borrower’s income profile) are already captured by providers, regulators and credit bureaus. Thus, the task of modifying IT systems will not be prohibitively costly, and in any case, the costs that result are justifiable.

Why this information is important

Once regulators start collecting and analyzing this data regularly, they will be better equipped to monitor credit markets and estimate over-indebtedness. With reliable information on over-indebtedness, regulators can design targeted policies to ensure that no borrower is forced to skip meals or discontinue their children's education to repay debts. Policies to avoid over-indebtedness would also act as a guardrail against mass defaults, which can jeopardize the health of the financial system.

Leave a Comment

Comments on this page are moderated by FinDev Editors. We welcome comments that offer remarks and insights that are relevant to the post. Learn More