Identifying credit supply shocks with bank-firm data: Methods and applications

Abstract

The purpose of these working papers is to promote the circulation of research results (Research Series) and analytical studies (Documents Series) made within the National Bank of Belgium or presented by external economists in seminars, conferences and conventions organised by the Bank. The aim is therefore to provide a platform for discussion. The opinions expressed are strictly those of the authors and do not necessarily reflect the views of the National Bank of Belgium.

Introduction to the Symposium on Contemporary Banking Research: The Use of Fixed Effects to Disentangle Loan Demand From Loan Supply

“…Furthermore, regulators might be particularly interested in the small firms that may be dropped from the analysis when relying upon multiple-bank firms. Degryse et al (2019) introduce and verify a novel method that incorporates single-bank firms into the analysis of the effects of supply shocks on borrowing firms. In their sample of firms incorporated in Belgium, the authors establish that 84% of FT observations belong to firms that do not have multiple-bank relationships, representing 46% of the lending portfolio of banks.…”

“…While Degryse et al () show more formally that firm fixed effects may be substituted with alternative loan demand controls using a clustering procedure, other papers have proceeded with a direct application of such an approach in their empirical analyses. De Jonghe et al () use identical industry‐location‐size clustering to show that less risky firms or firms operating in sectors in which their borrowing bank is more present or specialized are better shielded from banks' funding shocks.…”

Commodity Prices and Bank Lending

“…In this way, under the assumption that all firms in the same district‐industry pair face similar unobserved shocks that can affect their credit demand in any quarter, we can absorb the effect of all macroeconomic shocks on credit demand. In particular, Degryse et al (, p. 34) deal with firms that commonly have bank relationships with only one bank and show that fixed effects for firms clusters (where a cluster comprises firms of similar size in the same industry and location) “perform very well as controls for the firm‐borrowing channel: the bank‐loan supply shocks obtained with such demand controls closely resemble the ‘standard’ bank‐loan shocks (in terms of ordering and magnitude) for the multiple‐bank firm sample.” With this approach, we identify the differential effect of commodity prices on the supply of credit comparing the same borrower applying in the same quarter to banks with different exposures (columns 2, 5, and 8). Finally, we combine these two approaches and we maintain firm fixed effects allowing them to vary across years.…”

The Real Effects of Credit Supply: Review, Synthesis, and Future Directions

“…In addition, it has also been shown that negative effects of credit supply are also reflected in firms' valuation (e.g., Gan, 2007), firms' export growth (e.g., Paravisini et al, 2014) as well as sales growth (e.g., Acharya et al, 2018). These effects are generally stronger for small (e.g., Khwaja and Mian, 2008), young (e.g., Cingano et al, 2016), and single-bank firms (e.g., Degryse et al, 2019). Moreover, cash holdings (e.g., and access to other sources of funding (e.g., Campello et al, 2010) play an important role in the transmission of credit supply into the real economy.…”

“…Despite its advantages and popularity, however, it comes with two important limitations. It is unable to identify aggregate equilibrium outcomes and it is not applicable when firms borrow from only one lender; these limitations are addressed by, respectively, Amiti and Weinstein (2018) and Degryse et al (2019).…”