Skip to Main Navigation

Chapter 4. Lending during the recovery and beyond

When the acute physical hardships of the COVID-19 pandemic begin to recede, a swift return to a healthy economy will depend on lenders providing credit to households and businesses. Although most lenders have not experienced pandemic-related liquidity challenges, they anticipate a rise in nonperforming loans (NPLs) as business activity and household incomes continue to be disrupted. The ongoing economic upheaval produced by the pandemic, its unknown duration, and the lack of clarity on its longer-term impact on borrowers constrain the ability of lenders to extend new loans. Many of the interventions undertaken to support households and businesses experiencing income distress have also made discerning borrowers’ current economic situation and ability to service new debt more difficult. When lenders cannot project the overall path of economic activity, they tend to tighten their credit standards and issue less new credit. A review of quarterly central bank surveys on credit conditions in selected low-, middle-, and high-income economies finds that the majority of countries for which such surveys are available have experienced several quarters of tightening credit standards since the onset of the crisis (figure 4.1).

Figure 4.1 Trends in credit conditions, by country income group, 2018–21 (by quarter)



Source: WDR 2022 team calculations, based on data from survey reports by the central banks of 38 countries published or accessed as of December 15, 2021: Albania, Argentina, Austria, Belgium, Canada, Chile, Cyprus, the Czech Republic, Estonia, France, Germany, Ghana, Greece, Hungary, India, Indonesia, Ireland, Italy, Latvia, Lithuania, Japan, Mexico, the Netherlands, North Macedonia, the Philippines, Poland, Portugal, Romania, the Russian Federation, Serbia, Spain, Thailand, Turkey, Uganda, Ukraine, the United Kingdom, the United States, and Zambia.

Note: The figure shows the net percentage of countries in which banks reported a change in overall credit conditions in quarterly central bank loan officer or credit condition surveys. The net percentage is the difference between the share of countries that report an overall easing in credit conditions and the share of countries that report an overall tightening of credit conditions relative to the previous quarter. A negative net percentage value indicates an overall tightening of credit conditions in the sample of countries covered. For Chile, Japan, Mexico, Poland, Russia, the United States, and Zambia, the overall credit conditions are estimated from an index of reported credit conditions in business and consumer segments.

In a context of tightened credit standards, only the lowest-risk borrowers are able to access credit. This can create a vicious cycle whereby a widespread reduction in credit, including to the micro-, small, and medium enterprises (MSMEs) that make up the majority of businesses in most economies, forces many out of business before they can recover and drive growth. A domino effect may result, as accelerating business failures drive the levels of bad debt higher and cause lenders to further tighten lending standards.

By contrast, if lenders can effectively manage the risk of lending to support businesses and households, they can unleash a virtuous cycle in which credit supports investment and spending that help drive recovery. To achieve this, lenders should adapt underwriting models to the pandemic environment and adopt innovations that leverage new data and digital tools.  

Lenders can start by reassessing and updating their sector and borrower scoring models based on updated information on economic activity by sector or geographical area to capture the structural breaks caused by the pandemic. Retuning credit models is a key first step, which can be supported and encouraged by supervisory authorities. The scale and length of the crisis will likely require lenders to continue to update benchmarking data.

Data and product solutions that raise visibility into the capacity of borrowers’ ability to repay and that improve recourse in the event of a default can help lenders measure and manage risks so that credit can support the virtuous cycle of economic recovery. Such lending innovations are often technology-driven and are enabled by the acceleration in digitalization during the pandemic across all parts of the economy as well as the digital transformation efforts of financial institutions, which often predate COVID-19 but have been reinforced during the pandemic.

Alternative data and product design to improve visibility and recourse

Because of the economic disruptions of the pandemic, the borrowing and repayment histories found in traditional credit reports have become less useful. Meanwhile, alternative data such as transactional records on purchases, income, sales, orders, inventory, bill payments, taxes, court records, and social media profiles are proving to be useful indicators of borrowers’ ability or willingness to repay. Because of accelerated digitalization, such data are becoming more readily available from financial providers, mobile network operators and other utilities, traditional businesses, online platforms, and governments. These data, along with the updated credit models that incorporate them, have proven effective in indicating creditworthiness.1

One lender demonstrating the value of alternative data is Konfío, a digital MSME lender that relies on electronic invoices and other alternative data to supplement traditional credit information. Konfío adapted its credit algorithm in the early months of the pandemic to integrate granular data on the impacts of COVID-19 on small firms in Mexico and doubled its monthly loan disbursements.

Product design can improve visibility, as well as recourse in the event of nonpayment. Products designed to have short loan tenor, for example, limit risk by reducing loan amounts and repayment timelines to the period over which the lender has reasonable insight. Another design option is to accommodate nontraditional forms of collateral, such as movable assets or liens on future cash flows, to serve borrowers who lack traditional collateral or whose collateral value has become uncertain during the pandemic.

Contextual finance is another approach for increasing visibility and recourse. Working capital financing that occurs within a supply chain or is embedded in the workflow of a commercial transaction links credit to an existing commercial relationship, as well as to an underlying economic activity and its associated data. Embedded finance is taking place on e-commerce, logistics, and inventory management platforms. One example is financing for small and medium enterprises selling on e-commerce platforms; data from the platforms on merchant sales patterns, inventories, and customer satisfaction history are used to underwrite the loans. Another example is e-logistics platforms that supply pre-trip working capital to truck drivers based on delivery track records and drivers’ expected revenue. Embedded finance has scope to reduce the cost of lending through process efficiencies and improved recourse such as by automatically setting aside a portion of borrowers’ revenue through the platform to repay loans.

Credit guarantees for managing credit risk

Insuring against loss can help restore credit growth when sufficient visibility and recourse cannot be achieved using the innovations just described. Credit guarantees give lenders recourse to the guarantors in case of default by borrowers. These instruments may be offered by governments, development banks, or donors to promote lending to priority segments where there are market failures in financing, such as small businesses. Guarantees have been a component of the pandemic response in high-income countries and several emerging economies, and partial guarantee schemes may continue to play a role in the recovery. For example, in Burkina Faso the World Bank helped the government set up a credit guarantee scheme focused on restructured and working capital loans for small and medium enterprises struggling from the COVID-19 crisis, but with potential for long-term profitability. Such programs must be carefully designed to be sustainable. As economic conditions improve, guarantors and their partner lenders can progressively narrow eligibility to the sectors or customer segments that continue to show the most need, and guarantee programs can be adjusted to reduce the risk associated with longer-term investments to support priorities such as job creation and financial flows to low-carbon activities.

Policies to encourage safe innovation

Financial innovation has the potential to support responsible financial access, but unsupervised financial innovation may pose risks for consumers as well as for financial stability and integrity. Governments and regulators must modernize policy frameworks to balance the sometimes-conflicting imperatives of encouraging responsible innovation, on the one hand, and protecting customers and the financial sector’s stability and integrity on the other. Updated regulatory and supervisory approaches should seek to recognize and enable the market entry of new providers, the introduction of new products, and innovations in the use of types of data and analytics; enhance financial literacy and consumer protection policies and rules around what finance providers can and cannot offer; and facilitate collaboration between regulators and the government authorities overseeing aspects of digital and embedded finance, as well as competition and market conduct, to prevent regulatory gaps between agencies with overlapping authority. Policies should support the modernization of financial infrastructure to facilitate operational resilience and access, including hard infrastructure related to telecommunications networks, payment networks, data centers, credit bureaus, and collateral registries and soft infrastructure around the policies and procedures that dictate standards, access, and rules of engagement. These government policy responses to support digital transformation can help foster a more robust, innovative, and inclusive financial sector.

Notes

1 Sumit Agarwal, Shashwat Alok, Pulak Ghosh, and Sudip Gupta, “Financial Inclusion and Alternate Credit Scoring for the Millennials: Role of Big Data and Machine Learning in Fintech” (paper presented at Cambridge Centre for Alternative Finance’s “Fifth Annual Conference on Alternative Finance,” Judge Business School, University of Cambridge, Cambridge, UK, June 29–July 1, 2020), https://www.jbs.cam.ac.uk/wp-content/uploads/2020/08/2020-06-conference-paper-agarwal-alok-ghosh-gupta.pdf; Daniel Björkegren and Darrell Grissen, “Behavior Revealed in Mobile Phone Usage Predicts Credit Repayment,” World Bank Economic Review 34, no. 3 (2020): 618–34, https://doi.org/10.1093/wber/lhz006; Leonardo Gambacorta, Yiping Huang, Han Qiu, and Jingyi Wang, “How Do Machine Learning and Non-Traditional Data Affect Credit Scoring? New Evidence from a Chinese Fintech Firm” (BIS Working Paper 834, Monetary and Economic Department, Bank for International Settlements, Basel, Switzerland, 2019), https://www.bis.org/publ/work881.htm.