BIS Paper Examines Business Model for Lending by Banks and Bigtechs

In the last decade, large technology companies, also known as big techs, have moved into the provision of financial services, with some becoming substantial players in payments in several advanced and emerging market economies. The big techs have access to massive amounts of data about firms that operate on their online platforms or use their QR-code payment systems. This information can be harnessed to improve the assessment of the credit risk of a firm. Recently, big techs have started competing with banks especially in the market for loans to small and medium-size enterprises (SMEs). The study presented in the paper examines the competition between big techs and banks in the loan market for SMEs where adverse selection and difficulty of enforcing repayment cause frictions. The study focused on loans to firms, in particular SMEs, because big techs mainly provide credit to small vendors in their ecosystem.

The study shows how competition between banks and big techs in attracting borrowers can lead to greater privacy of borrowers, as big techs have an incentive to temper their drive to collect information about firm characteristics. However, the limited capacity of big techs to recognize a firm's type increases the number of inefficient defaults and reduces investment in profitable opportunities. One way to mitigate these inefficiencies is for big techs to share their data with the banks that make loans funded with cheap deposits. The study investigated two information-sharing arrangements. In one, the big tech makes data public for any bank that wants to use them, for example, by conferring the firm type information to a public credit bureau. In the other arrangement, the big tech privately gives the firm type information to the bank, for example, by selling it credit scoring services. These two ways of sharing information lead to different outcomes.

Lack of access to deposits makes funding by big techs more costly than the funding by banks and it is a factor limiting their supply of lending. Conversely, lack of information would result in banks funding insolvent firms. Thus, it seemed natural to investigate whether it is possible to exploit these relative advantages so that the big techs gather the raw data, process them, and share relevant information with the banks that make loans funded with deposits. The study found that it could be more costly for a bank to extract information from the raw data, as big techs are likely to be better placed in terms of the use of credit scoring techniques based on machine learning and big data to better assess the creditworthiness of firms. Thus, a joint venture (bank-big tech) in which the big tech shares the information, such as credit scores, rather than the raw data, would be advisable. However, the analysis of the optimal business arrangement between banks and big techs is an additional relevant area for future research.

Related Links

• Notification
• Report (PDF)

Keywords: International, Banking, Lending, Credit Assessment, Credit Scoring, Financial Disintermediation, P2P Lending, Big Data, Big Tech, Fintech, BIS