Against the backdrop of the financial sector playing a key role in mobilising resources to achieve economic growth, Zimbabwe has in recent times seen severe macroeconomic constraints and hardship. Over the past two decades, the country’s financial sector has learnt (or not learnt) a number of lessons – lessons that could assist in taking the economy from a recovery path to a growth trajectory.

Let’s take a closer look at the credit market, where non-performing loans remain one of the biggest challenges. The Central Bank has taken measures to address non-performing loans in a bid to stabilise the financial sector. In 2014, the Reserve Bank of Zimbabwe (RBZ) established the Zimbabwe Asset Management Company (ZAMCO) in a bid to resolve non-performing loans. To date, ZAMCO has acquired non-performing loans to the value of one billion USD, but there is increasing evidence that the default rates on loans are on the rise. The big question is: What can be done to sustainably address the problem of borrower default in Zimbabwe?

The insight2impact facility partnered with XDS, the largest credit bureau in Zimbabwe, to understand what drives borrowing and repayment behaviour on consumer loans. Our analysis of demand-side and supply-side data has shown that people see financial services as a means to an end. As a result, people use formal and informal services interchangeably. The use of informal financial products has spill-over effects on how formal financial services are used. Analysis of transactional data in isolation therefore cannot speak to people’s informal financial lives and omits important borrower attributes that are useful in understanding usage behaviour. To overcome this constraint, we merged transactional data and demand-side survey data by matching a sample of clients’ transactional profile from a credit bureau database with survey responses from the same individuals. Our focus on the credit market means that the primary usage behaviour we analysed was frequency of borrower repayments.

Our analysis shows that the current information gathered by financial service providers (FSPs) and available in the credit bureau database provides an inaccurate picture of the credit risk faced by MFIs that wish to extend loans. For example, the survey results show that 36% of the clients in the XDS database also borrowed from informal sources, with most individuals turning to family and friends for credit. On average, borrowers in the XDS database had accessed credit from five sources. Three out of the five credit providers were informal. This shows that informal activity is a significant factor in an individual’s risk profile.

In addition, the cash shortages in the country are resulting in a selection problem in the financial sector. The findings show that people mostly borrow for liquidity needs, which points to high levels of financial distress. The results show that over the period of 12 months, 31% of the borrowers missed at least one payment while 18% defaulted. Our evidence shows that people borrow from the formal sector after exhausting all their options in the informal sector. This means that the formal sector is serving people who are experiencing greater levels of financial distress, which, in turn, implies that a credit risk assessment based only on supply‑side data will omit important information on the client’s risk profile.

Another surprising finding was that higher-income earners were more likely to default on their loans. An increase of one dollar in a borrower’s income was associated with a 23-cent increase in total arrears. A closer look at this finding shows that higher-income individuals tend to borrow from multiple sources. Without proper credit information-sharing, access to credit skewed towards high-income earners poses the risk of over-indebtedness and, thus, higher default rates.

What do these findings mean to policymakers, regulators and FSPs?

The main insight from this study is the importance of understanding the borrower to overcome the problem of non-performing loans. The findings show that inadequate information on the borrower’s risk profile results in FSPs serving already-distressed clients.

There are two ways in which FSPs can address this risk. First, FSPs should consider information-sharing with other FSPs to ensure that they don’t serve overindebted clients. Second, FSPs should broaden their understanding of credit activity in the informal market. This requires utilising stand-alone demand-side surveys in addition to transactional data. This can allow FSPs to more accurately account for informality in credit risk assessments by doing market segmentation based on demand-side behaviour.

Unless ZAMCO has deep pockets to spend on a growing pile of toxic loans, regulators would be well advised to intervene and assist credit providers in incorporating non-transactional data into their risk assessments. A good starting point will be to convene FSPs to start a conversation on the benefits of credit information-sharing and sourcing non-transactional data. Much data is already available, but very few credit providers are using it.

insight2impact (i2ifacility) was funded by Bill and Melinda Gates Foundation in partnership with Mastercard Foundation. The programme was established and driven by Cenfri and Finmark Trust.