Hidden Gems and Borrowers with Dirty Little Secrets: Investment in Soft Information, Borrower Self-Selection and Competition

Abstract

Hidden gems and borrowers with dirty little secrets: investment in soft information, borrower self-selection and competition ECB Working Paper, No. 1555 Provided in Cooperation with: European Central Bank (ECB)Suggested Citation: Gropp, Reint; Gruendl, Christian; Guettler, Andre (2013) : Hidden gems and borrowers with dirty little secrets: investment in soft information, borrower self-selection and competition, ECB Working Paper, No. 1555, European Central Bank (ECB) We focus on the difference between firms with positive soft information and firms with negative soft information. Recent theoretical models (Hauswald and Marquez, 2006;Inderst and Mueller, 2007) suggest that firms with positive soft information would tend to self-select to relationship banks, because relationship banks can take the positive private signal into account in the lending decision. This creates an Akerlof-type adverse selection problem, in which transaction banks tend to receive applications from borrowers with on average negative soft information. In response, transaction banks may apply a negative adjustment to the rating of all their loan applicants. However, if they do, even borrowers with slightly negative private information may be better off obtaining a loan from a relationship bank, resulting in an even worse pool of loan applicants with respect to the private signal and so forth. Ultimately, in the absence of any offsetting factor, transactions banks would no longer participate in the market for small business loans and some positive NPV firms may no longer receive credit. The selection effect may explain why banks ultimately specialize in either relationship or transaction banking.Furthermore, theory would predict that there are interaction effects with the degree of interbank competition. In particular, Boot and Thakor (2000) argue that there are two effects. When competition is introduced, banks' marginal rents from relationship lending are smaller and each bank thus reduces its investments in soft information. However, competition affects the bank's profits from both relationship and transaction lending asymmetrically. A relationship orientation helps to partially insulate the bank from pure price competition, so that an increase in competition from other banks hurts the bank's profits from transaction lending more than its profits from relationship lending. Thus, increased competition between banks may encourage banks to shift from transaction to relationship lending.We use a matched bank-borrower dataset of German savings banks to test these predictions.These banks provide an ideal laboratory, as they compete with pure transaction banks, such as 2 Deutsche Bank and with pure relationship banks, such as the large number of extremely small cooperative banks in Germany. At the same time, there is sufficient variation within the savings bank sector in the degree to which banks incorporate soft information in their lending decisions.Most importantly, the dataset allows us to construct a proxy for the case when the private no...

Hardening soft information: does organizational distance matter?

“…Our findings contribute to a fast growing literature on loan officer discretion, information production and manipulation in credit-scored lending (Puri et al, 2011;Bouwens and Kroos, 2012;Brown et al, 2012;Campbell, 2012;Berg et al, 2013;Berg, 2014;Degryse et al, 2014;Mosk, 2014;Nakamura and Roszbach, 2018;Campbell et al, 2019). Perhaps in a paper most closely related to our research, Gropp and Guettler (2018) measure the discretionary use of soft information as the difference between the final and the hard-information-based ratings (this difference we label as "discretion_1") and show that loan officers at small savings banks in Germany are more likely to upgrade the financial rating of applicants than their colleagues at large savings banks, while the downgrades of financial ratings are not significantly different between the two groups of banks. Our paper differs from Gropp and Guettler (2018) in two major ways.…”

“…The second type of hardening allows a loan officer to upgrade or downgrade a final rating. We refer to this as "uncodified discretion" -a type of hardening also documented by supervisors (BCBS, 2005;Fed, 2011) and in the academic literature (Brown et al, 2012;Brown et al, 2015;Gropp and Guettler, 2018). It is not uncommon to find both codified and uncodified discretion in the same rating scheme as was the case in the six banks studied in Brown et al (2015).…”

“…Perhaps in a paper most closely related to our research, Gropp and Guettler (2018) measure the discretionary use of soft information as the difference between the final and the hard-information-based ratings (this difference we label as "discretion_1") and show that loan officers at small savings banks in Germany are more likely to upgrade the financial rating of applicants than their colleagues at large savings banks, while the downgrades of financial ratings are not significantly different between the two groups of banks. Our paper differs from Gropp and Guettler (2018) in two major ways. First, we identify the location of our loan officers and assess the impact of communication frictions in information transmission related to the exact within-bank geographical distance between bank officers and bank decisionmakers rather than just proxying for communication frictions with the size of the bank.…”

“…To this end we use two indicator variables. In line with Brown et al (2012) and Gropp and Guettler (2018), discretion_1 assumes the value of 1 if the final rating is different from the modified statistical rating produced by the scoring algorithm on the basis of pure hard information and 0 otherwise. However, discretion_1 does not distinguish between the case of non-use of the discretion by the loan officer, in which final rating = integrated rating = modified statistical rating, from that of a double use of both codified and uncodified discretion which counterbalance each other, in which (final rating -integrated rating) = -(integrated rating -modified statistical rating) leaving the final rating equal to the statistical rating.…”

The Effects of Corporate and Country Sustainability Characteristics on The Cost of Debt: An International Investigation

“…When assessing the credit quality of an institution seeking to borrow funds, the traditional approach of banks has been to base this assessment on reasonably objective data that are quantitative, easily verifiable by third parties, and which are usually financial in nature (e.g., profitability, current operational leverage, liquidity, market value and credit ratings assigned by rating agencies). This emphasis on “hard information” is the defining characteristic of “transaction‐based banking” (Gropp et al., ). In contrast, the lending decisions for “relationship banking” are based not only on the above, but also on “information which is harder to accurately summarize in a numeric score” (Petersen, ), and which is often collected in person, difficult to verify by third parties, and more subjective in nature (e.g., managerial competence, trustworthiness and innovative thinking), or “soft information.” As Goetzmann et al.…”