The more the better? Information sharing and credit risk

Abstract

Estimating correctly the borrower credit risks is the task of particular –and growing- importance for the banks all around the globe. Formal information sharing mechanisms are aimed to reduce information asymmetry in the credit markets and to enhance the precision of those estimates. In the academic literature, however, the answer to the question of whether more detailed borrower information accumulated and shared by credit bureaus and credit registries is always associated with higher quality of bank credit portfolios and lower credit risks is not completely unambiguous. More credit information disclosed by information intermediaries tend to result into the weaker disciplinary effect of credit history, which means higher credit risks. At the same time, the accuracy of assessing the creditworthiness of borrowers grows due to an increase in the predictive power of scoring models, which leads to a reduction in credit risk. In this paper we make a first attempt to examine the nonlinearity of this effect. We study the relationship between the depth of the credit information disclosure and the stability of the banking sector in terms of credit risks. Based on data on more than 90 countries for 2004–2020, we show that the relationship between disclosure and credit risk is non-linear: we observe the lowest levels of credit risk at minimum and maximum levels of disclosure depth. We analyze the influence of the country institutional quality and financial development on the nature of the relationship. We show that credit risks decrease with increasing depth of disclosure by credit bureaus and credit registers in in a high-quality institutional environment and in well-developed financial markets.