Information and bank credit allocation

https://doi.org/10.1016/S0304-405X(03)00210-1Get rights and content

Abstract

Private information obtained by lenders leads to borrower capture to the extent that such information cannot be communicated credibly to outsiders. We analyze how this capture affects the loan portfolio allocation of informed lenders. First, we show that banks charge higher interest rates and finance relatively less creditworthy borrowers in market segments with greater information asymmetries. Second, when faced with greater competition from outside lenders, banks reallocate credit toward more captured borrowers (flight to captivity). Third, if borrower quality and captivity are sufficiently correlated, an increase in the competitiveness of uninformed lenders can worsen the informed lender's overall loan portfolio. The model explains observed consequences of financial liberalizations.

Introduction

Over the last two decades, bank activities, which traditionally had been heavily regulated and protected from competition, have been progressively liberalized. Financial sector reforms have eased legal barriers to entry and enlarged the scope of the activities of banks and other financial intermediaries. This process has fostered cross-border lending and foreign entry, reshaping the competitive structure of local credit markets and changing the allocation of bank credit. These changes have important economic consequences as bank lending is an important source of funding not only for individuals and for smaller businesses, but also for large businesses, which make heavy use of bank borrowing as a form of short-term financing. This paper presents a theoretical framework to understand how the allocation of credit responds to shocks to the banking system, such as the entry of a low-cost competitor. It also provides new insights for understanding the observed effects of financial liberalization and cross-border deregulation on bank competition.

We focus on the important role information plays in shaping bank competition. Private information obtained by banks generates a lender-client specificity to the extent that such information cannot be communicated credibly to outsiders. James (1987) is among the first to provide evidence that bank lending is indeed special relative to other sources of finance. He shows that the announcement of a bank loan leads to a positive stock price response for the firm obtaining the loan. Lummer and McConnell (1989) provide similar evidence for the case of loan renewals. We present a model where a lender with an informational advantage competes for borrowers with an outside lender with worse information, but potentially with a cost advantage in extending a loan. We show that the informed lender's informational advantage provides it with some degree of market power and leads to borrower capture, as adverse selection makes it difficult for borrowers to obtain credit from outside lenders. This basic intuition allows us to derive three main results.

First, the degree of borrower capture and spreads on bank loans are higher in markets subject to larger information asymmetries. This allows informed lenders to profitably finance borrowers that are less creditworthy in these markets. An important implication of this result is that the average quality of borrowers obtaining financing from the informed lender is decreasing in its informational advantage. Hence, there are compositional differences in banks’ portfolios across market sectors characterized by different degrees of asymmetric information about borrowers.

The second result is that when faced with greater competition from outside lenders, informed banks shift their credit allocation towards sectors where their competitors face greater adverse selection problems. More generally, when forced to curtail their loan portfolio, informed banks reduce lending to a greater extent in less captive sectors, and retain larger market shares in the more captive but more profitable sectors. We refer to this reallocation as a flight to captivity, since it implies that banks reallocate their portfolio towards more captive borrowers when shocks to their balance sheet, or to their competitive environment, force them to alter their lending patterns. Loans to borrowers in more competitive sectors are also the most liquid in banks’ portfolios, since they could be sold at a price closer to their actual value. In this vein, we show that increases in the competitiveness of outside lenders leads to greater interest-rate reductions for borrowers in low information-asymmetry sectors since these borrowers are most able to take advantage of the competition from these lenders.

Finally, we show that, if a strong negative correlation between borrower quality and degree of information asymmetry exists, an increase in the competitiveness of uninformed lenders can lead to a worsening of the informed lender's overall loan portfolio. Indeed, uninformed lenders better attract borrowers in sectors subject to low information asymmetries, which are more creditworthy. This in turn leaves the informed lender to deal primarily with loans associated with more captured (but also higher risk) borrowers.

The results in this paper derive from the interaction between two distinct, but related, implications of informational specificity. First, banks obtain higher profits from more captured borrowers than from borrowers that have other financing alternatives. Second, given the level of information asymmetry about a borrower, which determines the interest rate offered, loans to higher quality borrowers (higher probability of repayment) are more profitable. Borrower quality by itself has no implications for bank profitability absent some other friction that prevents quality from being perfectly priced. Asymmetric information provides that friction in our model. Consequently, when banks curtail lending we expect to see a relative reallocation of informed credit towards more captured borrowers, as banks attempt to retain their most profitable clients. In equilibrium, we see lower quality but more captured borrowers being retained by an informed bank, even as borrowers of better quality (but less captive) move to alternative sources of financing.

Our analysis sheds new light on some unexplained consequences of regulatory reforms. Financial liberalizations, such as those that have occurred recently in emerging markets and transition economies, have lead to the entry of foreign banks that, relative to local banks, have concentrated their activities on wholesale banking and on lending to large borrowers (see Berger et al., 2001; Clarke et al., 2001). This pattern of entry has obviously benefitted borrowers in these sectors of the economy. However, small borrowers appear also to have benefitted from these reforms through both increased availability of credit and reduced intermediation spreads, in spite of their limited access to lending from foreign banks (see Martinez Peria and Mody, 2002; Clarke et al., 2002; Laeven, 2003). This pattern of entry, as well as the positive spillover for small borrowers, is not easily understood within the context of existing models that ignore how information asymmetries affect the allocation of domestic bank credit. Most of the agency cost literature focuses on cream-skimming effects associated with observable borrower quality. However, this framework explains the observed pattern of entry and credit reallocation only to the extent that quality and the availability of borrower-specific information are highly correlated. Our model predicts that foreign banks will be more effective at competing away from local banks the borrowers for whom informational disadvantages are smaller. In turn, local banks will reallocate their portfolios towards borrowers whose quality is less discernible by external lenders. In other words, we show that the informational structure of the industry determines the height of the barriers to entry across different groups of borrowers and creates a segmentation of lending markets between foreign and domestic banks. To the best of our knowledge this implication is unique to our model and is in accordance with existing evidence.

The issue of how changes in the competitive structure of credit markets affect both the portfolios of the informed banks as well as borrower welfare has been little studied. Previous theoretical literature by Sharpe (1990), Rajan (1992), and von Thadden (2002) demonstrates that informed lenders can reap benefits from their information even when faced with competitive threats. Dell'Ariccia et al. (1999) show that information asymmetries between lenders can hinder entry into banking markets. However, this literature does not address how information affects the way credit is distributed across borrowers if lenders face constraints, such as an upward-sloping industry supply curve of funds, or how that allocation changes when these constraints tighten. From an empirical perspective, Berger and Udell (1995) find that firms with longer banking relationship are less likely to have collateral requirements. Berger et al. (1998) discuss similar issues in the context of bank mergers. These results provide evidence that banks solve information problems over the course of the lending relationship. However, by focusing on the firms’ side of the issue, this literature does not analyze the implications for bank credit allocation. Petersen and Rajan 1994, Petersen and Rajan 1995 find that the availability of credit to small businesses increases as banking relationships lengthen, or as competition in credit markets decreases. This latter result is consistent with our finding that borrower quality is lower in markets with greater information asymmetries, where banks can profit from lending to marginal quality borrowers. Our model can therefore be seen as providing both a foundation for their analysis of exogenous market power and credit market competition, as well as extending the model to analyze changes in banks’ portfolios following a shock to their liabilities.

The choice of bank versus arm's-length financing is examined in the agency cost literature (e.g., Besanko and Kanatas, 1993; Diamond, 1991; Boot and Thakor, 1997; or Chemmanur and Fulghieri, 1994), which predicts that only borrowers of higher observable quality will obtain financing from nonmonitoring lenders. Our model focuses instead on competition among banks and on their sources of market power and suggests that entrants should concentrate their activities where information asymmetries are lower as a way of avoiding adverse selection problems. This captivity dimension coexists with the quality dimension found in these models, and helps explain the patterns of foreign bank entry described above.

Finally, our analysis also has implications for other shocks to banks’ balance sheets, such as changes in monetary policy. In this context, the bank lending channel view holds that monetary contractions reduce the supply of bank loans relative to other sources of finance (see Kashyap and Stein, 2000). In our framework, a relative increase in the informed bank's cost of funding induced by a tightening of monetary policy reduces its lending to lower quality borrowers. This effect is similar to the well-known flight to quality, which argues that creditors will optimally shift funds away from high agency cost borrowers during periods of recessions or when borrowers’ balance sheets deteriorate. Lang and Nakamura (1995) and the references therein provide a discussion of this effect. This reduction in lending, however, is more pronounced for market sectors where borrowers are less captive. This latter effect is exactly our flight to captivity. We also argue that this contraction not only affects banks’ portfolios, but also affects borrowers through the interest rates they pay on loans.

The paper proceeds as follows. 2 Model, 3 Cross-segment analysis present the model along with a characterization of the equilibrium. Section 4 examines the cross-sectional implications of shocks to the competitive environment. Section 5 discusses how these shocks affect banks’ portfolios when creditworthiness and borrower capture are correlated. Section 6 provides further implications and discusses the findings in light of existing evidence. Section 7 concludes.

Section snippets

Model

Consider an economy with a continuum of entrepreneurs, where each entrepreneur is endowed with an investment project that requires a capital inflow of $1, but has no private resources, so that the entrepreneur must look to a lender to obtain this financing. Projects pay off an amount R with probability θ, and 0 with probability 1−θ, and we assume that this outcome is perfectly observable and contractible by both parties, but that the parameter describing the probability of success, θ, is

Cross-segment analysis

In this section, we characterize the equilibrium of the model, and analyze how credit allocation and interest rates vary across groups of borrowers characterized by different informational structures. Since the market segments are perfectly distinguishable on the basis of λ, and both lenders’ costs are linear, we can treat all markets separately. Therefore, for a fixed informational structure (for a fixed λ) we can write the profits for each lender as a function of their interest rate offers in

Credit reallocation

As emphasized earlier, one issue of interest is how informed lenders adjust their credit allocation when there is a change to the competitive environment in which they operate. One example of such change can be modelled as a shock to δ which alters the relationship between lender 1's information advantage and the uninformed lender's cost advantage.

First, consider how a change in the informed lender's cost advantage changes the equilibrium expected credit allocation and interest rates in any

Correlation between quality and information asymmetries

For the sake of generality, we have so far not assumed any correlation between θ and λ. However, in practice there likely exists a negative correlation between borrower quality and the degree of adverse selection in the market, as better quality borrowers tend to also be those that are older or larger, and therefore less likely to be captured by the informed lender. Incorporating this correlation into the model has an important implication concerning the quality of the portfolio of the informed

Implications and empirical evidence

In this section, we compare the implications of our model with the empirical findings concerning financial liberalization. The existing evidence is consistent with the predictions of our model. Furthermore, the model provides an explanation for some observed consequences of liberalization which have been difficult to reconcile with existing theoretical frameworks. We also briefly discuss the implications of our model in the context of other shocks to banks’ balance sheets.

Conclusions

This paper presents a framework for analyzing how competition among financial intermediaries affects credit allocation under asymmetric information. It shows that, when there are informational asymmetries among lenders, the profitability to a bank of granting a loan is determined along two different dimensions. First, the creditworthiness of the borrower affects the bank's expectation of recovering the invested funds. Second, the ability of the borrower to credibly communicate its quality to

References (38)

  • A. Berger et al.

    Relationship lending and lines of credit in small firm finance

    Journal of Business

    (1995)
  • Berger, A., Miller, N., Petersen, M., Rajan, R., Stein, J., 2002. Does function follow organizational form? Evidence...
  • M. Berlin et al.

    Deposits and relationship lending

    Review of Financial Studies

    (1999)
  • D. Besanko et al.

    Credit market equilibrium with bank monitoring and moral hazard

    Review of Financial Studies

    (1993)
  • Bonaccorsi di Patti, E., Dell'Ariccia, G., 2003. Bank competition and firm creation. Journal of Money, Credit, and...
  • Bonin, J., Mizsei, K., Szekely, I., Wachtel, P., 1998. Banking in Transition Economies: Developing Market Oriented...
  • A. Boot et al.

    Financial system architecture

    Review of Financial Studies

    (1997)
  • A. Boot et al.

    Can relationship banking survive competition?

    Journal of Finance

    (2000)
  • T. Broecker

    Credit-worthiness tests and interbank competition

    Econometrica

    (1990)
  • Cited by (317)

    • Bank market power and financial reporting quality

      2024, Journal of Corporate Finance
    View all citing articles on Scopus

    We benefitted from the comments of Tito Cordella, Douglas Diamond, Vincent Hogan, Anil Kashyap, Ashoka Mody, Gordon Phillips, Nagpurnanand Prabhala, Brian Sack, Lemma Senbet, Jeremy Stein, and participants in seminars at McGill University, University of Maryland, Washington University, CEMFI, Birkbeck College, the 2002 European Meetings of the Econometric Society, and the Southeast Theory and International Economics Conference. The views expressed in this paper are those of the authors and do not necessarily represent those of the IMF.

    View full text