Partners (Winter 1999)

THE "BIG PICTURE" ABOUT
RELATIONSHIP-BASED FINANCE

Discussion Comments
Allen N. Berger
Board of Governors of the Federal Reserve System
Wharton Financial Institutions Center

Allen N. Berger is a senior economist at the Board of Governors and a senior fellow at the Wharton Financial Institutions Center. His research covers a variety of topics related to financial institutions such as small business finance, credit rationing and credit crunches, and efficiency and profitability. He is also editor of the Journal of Money, Credit, and Banking and the Journal of Productivity Analysis. Berger has a B.A. from Northwestern University and a Ph.D. from the University of California-Berkeley.

The Three Conditions of Relationship - Based Finance

This session is about the topic of relationship lending by commercial banks, but to look at the "Big Picture" we discuss an even bigger concept, "relationship-based finance," of which relationship lending by banks is but one component. We define relationship-based finance as occurring when the following three conditions are met:

(1) Information is gathered by the provider of funds beyond the relatively transparent data available in the financial statements, observation of any collateral, and other public information;

(2) Information is gathered through continuous contact between the provider and the firm, its owner, the firm's customers, and the local community, etc., often through the provision of multiple financial services;

(3) Information remains confidential to the provider of funds, who uses the information to help make additional decisions over time about future injections of capital, the evolution of contract terms, or monitoring strategies.

Relationship-based finance is one of the major tools used to provide funding to informationally opaque small businesses who may otherwise have much higher capital costs or simply not qualify for sufficient external finance to continue operations. The information is often gathered through multiple contacts, not all of which involve the provision of finance (e.g., checking accounts, investment services, etc.), and not all of which are with the business itself.

For many small businesses, the information gathered about the firm owner is often more important than the information about the firm, since the firm may have little track record or accumulated collateral. Contract terms, covenants, and contract renegotiations are often geared to releasing additional information over time.1

Transaction - Based Finance

At the opposite extreme of relationship-based finance is transactions-based finance, in which funds are provided on the basis of easily available information in financial statements, observation of collateral, and other public information around the time that the funds are provided. The funding is typically a one-time injection of funds, and there is little or no past relationship or information garnered from other accounts, and no setting of contract terms to reveal information to be used in future decisions.

Transactions-based finance is best exemplified by the public stock and bond markets that fund relatively informationally transparent large businesses in which there is typically very little contact between the provider of the funds and the business being funded.

The Difference is Contact

Relationship-based finance more often applies to the funds provided in the private equity and private debt markets that fund relatively informationally opaque small businesses. In these markets, there is typically a considerable amount of contact between the provider of the funds and the business being funded. For example, venture capitalists that provide equity financing maintain an important relationship with the firm and gather information through frequent visits to the firm. They use this information in: (1) helping make managerial decisions; (2) choosing whether to inject more funds; and (3) deciding whether and when the firm should go public.

Trade creditors that provide debt financing may also gain private information about the small business' industry, production process, and financial health from continuous contact, and use this information to decide on credit and price terms.

Role of Commercial Banks

Commercial banks often establish relationships with small businesses through multiple types of contact and reuse the information in their lending decisions. Lines of credit issued by commercial banks often represent continuous, exclusive relationships that provide information over time to the bank, often in conjunction with checking accounts and the handling of accounts receivable.

Typically, small businesses have only one bank providing their lines of credit, and the small businesses may be asked to move their checking accounts and other business to that bank. Commercial banks also specialize in designing contract terms, establishing covenants, and renegotiating contracts—activities that reveal further information over time. However, relationship-based finance does not apply well to all types of private equity and debt.

For example, mortgages, equipment loans/leases, and motor vehicle loans are usually one-time extensions of credit based significantly on the value of pledged collateral. Small businesses often obtain these types of finance from different intermediaries. Thus, even for commercial banks, some of the credit supplied to small business is relationship-based and some is transactions-based.2

The Focus on Relationship Lending by Commercial Banks

Relationship-based finance is more general than the commonly discussed concept of relationship lending by commercial banks because it applies to equity financing as well as debt financing, and it also applies to nonbank intermediaries (e.g., venture capitalists) and other providers of funds (e.g., trade creditors) as well as banks.

Nonetheless, there is a strong research and policy focus on relationship lending by commercial banks for several reasons. Banks are the largest single supplier of external finance to small businesses in the U.S., supplying about 19 percent of total finance or about 37 percent of total debt finance. Although only about 41 percent of small businesses have any bank loans, almost all firms have bank checking accounts, and about 87 percent identify a commercial bank as their primary financial institution.

Small businesses also tend to have long relationships with their banks, almost eight years on average.3 Perhaps the most important research and policy issue regarding relationship lending to small businesses by commercial banks concerns the potential effects of banking industry consolidation on lending to relationship-dependent, informationally opaque small businesses.

Tangible Benefits

The empirical research on relationships between banks and small businesses generally supports the notions that banks use relationships to garner information and that small businesses benefit from these relationships.

The research using U.S. data generally found that small businesses with stronger banking relationships received loans with lower rates and fewer collateral requirements; had less dependence on trade credit; enjoyed greater credit availability; and more protection against the interest rate cycle than other small businesses (Petersen and Rajan, 1994, 1995; Berger and Udell, 1995; Blackwell and Winters, 1997; Berlin and Mester, 1998; Cole, 1998; Hubbard, Kuttner, and Palia, 1998).

The U.S. data also suggest that banks gather valuable private information from depositors, and in some cases use this information in credit decisions (Allen, Saunders, and Udell, 1991; Nakamura, 1993; Frieder and Sherrill, 1997; Mester, Nakamura, and Renault, 1998).

The European and Asian data also usually support the value added of relationships, although some of the European evidence suggests exceptions (Hoshi, Kashyap, and Sharfstein, 1990; Ongena and Smith, 1997; Elsas and Krahnen, 1998; Harhoff and Körting, 1998; Angelini, Di Salvo, and Ferri, 1998).

Bank Consolidation and Small Business Lending

As noted, an important research and policy issue concerns the potential effects of banking industry consolidation on lending to relationship-dependent, informationally opaque small businesses. The main argument behind the issue of whether consolidation reduces the supply of credit to these businesses hinges on the issue of whether there is a significantly different technology used by banks in relationship-based lending versus transactions-based lending.

The larger, more organizationally complex institutions created by consolidation may choose to provide less relationship-based credit to small customers because of Williamson-type (1967, 1988) organizational diseconomies of providing these services along with providing transactions-based wholesale capital market services to large customers. That is, it may be scope inefficient for one bank to produce outputs which require implementation of quite different policies and procedures.

Relationship-based finance requires gaining intimate knowledge of the small business, its owner, and its local market over time through a relationship. Large, organizationally complex institutions may be inefficient at providing these relationship-based services along with transaction-based services. Similarly, there may be scale or organizational diseconomies in making relationship-based loans because of agency costs in monitoring relationship-based information generated by local loan officers in large, organizationally complex financial institutions.

SunTrust Bank
The Bigger the Better?

There has been a substantial amount of recent empirical research on the effects of bank size and organizational complexity on the supply of credit to small businesses. A number of studies have shown that large banking organizations devote lesser proportions of their assets to small business loans than do small organizations (e.g., Berger, Kashyap, and Scalise, 1995; Keeton, 1995; Levonian and Soller, 1995; Berger and Udell, 1996; Peek and Rosengren, 1996; Strahan and Weston, 1996).

As banks get larger, the proportion of assets devoted to small business lending (measured by domestic Commercial & Industrial loans to borrowers with bank credit less than $1 million) declines sharply from about 9 percent of assets for small banks (assets below $100 million) to less than 2 percent for very large banks (assets over $10 billion). The effects of organizational complexity—measured by additional layers of management, operation in multiple states, being in more financial lines of business, etc.—are ambiguous (Keeton, 1995; Whalen, 1995; Berger and Udell, 1996; Berger, Saunders, Scalise, and Udell, 1998).

Effects of Mergers

There has also been a substantial amount of recent empirical research on the effects of bank mergers and acquisitions (M&As) on the supply of credit to small businesses. The effects of bank M&As are not necessarily the same as the static effects of just increasing bank size and complexity-M&As may also involve dynamic changes in organizational focus or managerial behavior.

A number of studies examined the effects of bank M&As on small business lending. The most common findings are these M&As in which one or more of the banking organizations is large tend to reduce small business lending, whereas M&As between small organizations tend to increase small business lending, although there are exceptions.

Since M&As involving large organizations dominate M&As in terms of assets, these studies suggest an aggregate net reduction in small business lending by the banks participating in M&As. The total change in the supply of small business credit from M&As also depends on the reactions of other lenders, or the "external effect" of M&As. For example, if banks involved in M&As reduce their supply of relationship lending because of Williamson-type diseconomies, other bank or nonbank lenders that are not burdened by these diseconomies may react by picking up some or all of these credits.

One study measured the external effect of M&As on the lending of other banks in the same local markets and found that changes in the supply of small business credit by these other banks tended to offset much, if not all, of the negative effects of M&A participants (Berger, Saunders, Scalise, and Udell, 1998).

Part of the external effect may be from de novo entry, which has occurred at a torrid pace in recent years. Several studies found that de novo banks tend to lend more to small businesses than do other banks of comparable size (Goldberg and White, 1998; DeYoung, 1998; DeYoung, Goldberg, and White, 1999).

However, the measured effects of M&As on the likelihood of de novo entry are mixed—one study found that M&As increase the probability of entry (Berger, Bonime, Goldberg, and White, 1999) and one study found that M&As decrease the probability of entry (Seelig and Critchfield, 1999).4

More Research Needed

What is needed in this literature is a body of analysis that combines data from banks and small businesses. The relationship-lending literature has used detailed information on the borrowing firms—their risk, industry, size, proxies for informational opacity, etc. However, this research has generally had little information on the banks—their size, organizational complexity, and M&A activity.

The bank consolidation literature has used detailed information on the banks, but generally has had little information on the borrowing firms. Given the research to date, the most informative new research on the effects of consolidation on small business lending would combine the two types of data to try to separate out demand and supply effects, and see if bank size, organizational complexity, and M&As are associated with reduced supply of credit to relationship-dependent, informationally opaque small businesses.

A limited amount of prior research has matched bank and small business information. One study found that large banks tend to charge about 100 basis points less on small business loans, and require collateral about 25 percent less often than small banks, other things equal (Berger and Udell, 1996). This is consistent with the view that large banks tend to issue small business loans to higher-quality transactions-based credits, rather than relationship-based loans that tend to have higher interest rates and collateral requirements.

Another study examined the probability that small business loan applications will be denied by consolidating banks and other banks in their local markets, and found no clear positive or negative effects (Cole and Walraven, 1998).

A third study found that the probability that a small firm obtains a line of credit or pays late on its trade credit does not depend in an important way on the presence of small banks in the market (Jayaratne and Wolken, 1999).

The latter two studies are consistent with a strong external effect of consolidation—that other banks in the market respond when needed.

Current Research Papers

All three of the studies in this session take off on this literature, matching bank data and small business data to disentangle demand and supply effects. Each does it in a different way, each uses the data in a creative way, and each makes a contribution to the literature.

Haynes, Ou, and Berney (1999) have a creative approach to get at the central issue in this literature of whether large and small banks serve relationship-type borrowers equally. They use information from the 1993 National Survey of Small Business Finances (NSSBF) for the small businesses and matching information for the banks that lend to them from the Call Report. The authors do a very good job of controlling for borrower characteristics, and a pretty good job differentiating between types of banks to distinguish demand versus supply factors.

This paper may be improved by use of more bank size classes, and perhaps some organizational complexity variables for the banks. They assess the probability that particular borrowers (large versus small borrowers, risky versus safe borrowers, etc.) with particular types of credit (lines of credit, mortgages, equipment loans, motor vehicle loans, etc.), get credit from large and small banks. This lines up nicely with the theory and earlier empirical results about which types of credit and which types of borrowers are likely to be relationship-driven versus transactions-driven.

The results are intuitive, and consistent with the prior literature. Large banks more often lend to larger, older, more financially secure businesses, consistent with the predicted focus on transactions-driven lending, while the reverse is true for small banks focusing on relationship-driven lending.

Cole, Goldberg, and White (1999) also have a creative approach to looking at whether large and small banks tend to concentrate on relationship-based lending (essentially what they call "character lending") versus transactions-based lending (essentially what they call "cookie-cutter lending"). Like Haynes, Ou, and Berney (1999), they use information from the 1993 NSSBF for the small businesses and matching information for the banks that lend to them from the bank Call Report.

An important difference is that instead of a reduced form for whether loans are obtained from large and small banks, they break the decision up into two structural components—whether borrowers apply for loans from large and small banks, and then whether the banks approve or deny. This addresses the approval/denial process more directly.

The authors use good control variables for small businesses and include good relationship variables. Again, it would be nice to see more different bank size classes, and perhaps some organizational complexity variables for the bank.

The results are intuitive, and consistent with the prior literature—large banks more often tend to make transactions-based (cookie-cutter) loans and small banks more often tend to make relationship-based (character) loans.

Scott and Dunkelberg (1999) have a creative approach to looking at supply factors in the banking industry, while controlling for characteristics of the small business. Unlike the other studies, they use information from the 1995 National Federation of Independent Business (NFIB) data set, which asks about whether the small business' principal financial institution was bought out / absorbed.

Their approach has three advantages. First, they use a different data set, which has the benefit of testing the robustness of the findings in the literature, given that most of the recent research using small business data has culled this information from the NSSBF. Second, they have more information about the firm's search efforts, satisfaction of borrowing needs, etc., to measure outcomes. Third, they directly address the issue of consolidation.

As was seen in the prior literature, the dynamic effects of bank M&As are not necessarily the same as the static effects of just increasing bank size/complexity. One disadvantage is that the control variables for the small businesses in the NFIB data set are not quite as detailed as those on the NSSBF.

Small Business
It is a very comprehensive analysis of many dimensions of how well the borrower is treated. The results are somewhat mixed, with consolidation affecting some, but not all of the variables measuring satisfaction of borrowing needs, loan approval/rejection, shopping for lenders, loan rates, etc. in the predicted directions. The mixed results are perhaps not surprising, given the large number of dependent variables employed.

Conclusion

Relationship-based finance is an important tool used by banks, non-bank, intermediaries, and others to finance informationally opaque small businesses who may otherwise have much higher funding costs or be significantly quantity-rationed in capital markets. The issue of relationship lending by commercial banks and the effects of banking industry consolidation on the supply of credit to relationship-dependent, informationally opaque small businesses are also important research and policy topics. All three papers make important contributions to the literature, but more research is needed.

Notes
1 For a more comprehensive review of small business finance, see Berger and Udell, (1998) and other contributions to the August 1998, Journal of Banking and Finance special issue on this topic.
2 See Berger and Udell, (1995) for data on "loyalty ratios" which indicate how often small business borrowers reuse the same bank for the same type of loan.
3 These figures are from Berger and Udell, (1998, Tables 1 and 2), compiled from the 1993 National Survey of Small Business Finances.
4 For a recent review of the research on the consolidation of the financial services industry, see Berger, Demsetz, and Strahan (1999).

References
Allen, L., A. Saunders, and G. F. Udell. "The Pricing of Retail Deposits: Concentration and Information," Journal of Financial Intermediation, 1, 1991, pp. 335-361. Angelini, P., R.D. Salvo, and G. Ferri. "Availability and Cost for Small Businesses: Customer Relationships and Credit Cooperatives," Journal of Banking and Finance, 22, 1998, pp. 929-954.

Berger, A.N., S.D. Bonime, L.G. Goldberg, and L.J. White, "The Dynamics of Market Entry: The Effects of Mergers and Acquisitions on De Novo Entry and Customer Service in Banking," Proceedings of a Conference on Bank Structure and Competition, (Federal Reserve Bank of Chicago, Chicago, Illinois), 1999.

Berger, A.N., R.S. Demsetz, and P.E. Strahan. "The Consolidation of the Financial Services Industry: Causes, Consequences, and Implications for the Future," Journal of Banking and Finance, 23, 1999, pp. 135-194.

 
Return to Index