Relationship Banking: Concept and Characteristics

Relationship banking is multi-dimensional, and is maintained over cycles of corporate growth and profitability. It is also supported by various institutions that provide the concerned parties with incentives to build and foster the relationships, as is clear for Japanese main banks (MBs) and German universal banks. However, relationship banking involves both promises and perils. The desirability of relationship banking should not be taken for granted, since it ultimately depends on whether the merits can be maximized without being caught in traps.

2.1. What Is Relationship Banking?

Relationship banking may be defined as the provision of financial services by a financial intermediary on the basis of long-term investment in obtaining firm-specific information through multiple interactions with diverse financial services (Boot, 2000). Banks have advantages in gathering/producing information about their clients, thanks mainly to the nature of information production. First, there are economies of scale: the cost of information gathering/production is reduced by learning through repeated transactions. Second, there may also be economies of scope: banks can utilize the information obtained on a type of service for other services (Petersen and Rajan, 1994). Third, financial contracts are typically incomplete: banks and customers can build commitment and reputation through repeated transactions across services, often allowing the low-cost renegotiation of debt contracts (Lehmann and Neuberger, 2001). This characteristic of information production makes it natural for banks to be interested in relationship-based banking (see Box 1 for literature on bank debt and equilibrium financing pattern; see also Yoshitomi and Shirai, 2001 for a more comprehensive discussion of the inherent features of the banking system compared with those of corporate bond markets).

Intensity of relationship banking

Then, what are the operational aspects of relationship banking? What is meant by a strong relationship between a bank and its corporate clients? It is necessary to look at the duration, scope, and extent of multiple banking (Petersen and Rajan, 1994).

• Duration of the bank-borrower relationship. The duration is important because information is accumulated and sharpened through repeated interactions, and is largely non-transferable to those outside of the relationship. Commitment and reputation are also built and verified over time. Risk-sharing and other compensatory pricing practices often take place over the cycles of firm growth and profitability.

  Box 1. Advantages and Disadvantages of Bank Debt, and Equilibrium Financing Pattern

  Some of the advantages and disadvantages of relationship banking derive from the very nature of bank debt compared with public debt. Thus, it would be useful to review the merits and demerits of bank debt before dealing with relationship banking. Compared with public debt (corporate bonds), bank debt is supposed to have several advantages. Bank debt is flexible (Bolton and Freixas, 2000; Rajan, 1992; Diamond, 1993). Loan covenants require the borrower to take or refrain from various actions, giving banks the right to renegotiate or call loans when covenants are violated, enhancing the flexibility and efficiency of financial contracting (Berlin and Mester, 1992; Magee and Sridhar, 1994; Park, 1994). Unlike public debt, the recontracting of existing debt is easy since the bank is a monolithic, readily accessible creditor, which is especially valuable when firms are in financial distress (Berlin and Loeys, 1988). Bank debt likely reduces agency costs. Bank debt is considered to be "inside debt," giving the debtholder access to information from an organization’s decision process that is not otherwise publicly available (Fama, 1985). Banks usually have repeated transactions with their corporate clients over many different banking services. Bank loans are typically short-term, forcing banks to make periodic evaluations of the borrower’s creditworthiness. Also, creditor banks are relatively few (compared with bond holders), giving them stronger incentives to engage in information production and monitoring, mitigating the free-rider problem associated with public debt. Lenders may agree to divide their monitoring tasks among themselves in a way that avoids duplication of monitoring (Rajan, 1992; Diamond, 1984, 1993). Creditor banks may exercise control over the investment decisions of borrowing firms. The control rights specified in loan covenants, together with efficient monitoring and better information, reduce adverse selection or moral hazard associated with external financing (Smith and Warner, 1979; Diamond, 1984; Berlin and Loeys, 1988). The finding that a firm's announcement of a new bank loan or a loan renewal (unlike that of a bond issue) has a significantly positive effect on stock returns indicates the efficiency-enhancing role of banks (James, 1987; Lummer and McConell, 1989). Bank debt, however, does have some disadvantages compared with public debt. Bank debt usually entails higher intermediation costs. These include monitoring costs, bank regulatory taxes, and the agency costs of delegated monitoring (Diamond, 1991; Berlin and Loeys, 1988). This is why more credit-worthy firms rely more heavily on public debt financing (Blackwell and Kidwell, 1988). Private debt has lower agency costs but potentially higher transaction costs given the large economies of scale in issuing public debt. Thus, firms issue public claims when the lower transaction costs of public debt offset the higher agency costs of public debt financing. Control by banks may adversely affect the investment incentives of the client firms. Relying exclusively on short-term private debt can be costly because decisions about the roll-over or calling of the debt and liquidation will be dominated by banks (Diamond, 1993). Short-term bank loans (to better known firms) give little incentive to monitor the borrower, since the bank can liquidate the firm at any early sign of financial distress. Long-term debt with covenants may give banks stronger incentives to monitor (Rajan and Winton, 1995). More seriously, the information acquired by a bank as part of an ongoing relationship can create an "information monopoly" or hold-up problem, in that it is costly for the borrower to switch lenders (Rajan, 1992: and Sharpe, 1990). Borrowing from public markets or multiple bank relationships mitigates the hold-up problem (Rajan, 1992; Hoshi, Kashyap, and Scharfstein, 1993; and Bolton and Scharfstein, 1996). On the basis of these characteristics of bank debt compared with public debt, many theoretical and empirical studies investigate equilibrium corporate financing patterns. More risky firms (with a sufficiently high demand for flexible financing) tend to use bank debt; while safer firms tend to finance from the bond market (avoiding higher cost of financing, Bolton and Freixas, 2000). Firms with large information asymmetries and agency costs for debt depend more on bank debt: these include smaller firms, firms with a higher proportion of intangible assets (where it is more difficult to value assets); and firms with greater growth opportunities. Young firms and older firms with poor performance whose ratings are too low for reputation effects to eliminate moral hazard (but high enough for monitoring to substantially reduce moral hazard) tend to use more debt (Diamond, 1991). Bank debt and public debt are complementary, as bank monitoring creates a public good that reduces the cost of issuing public debt (Gorton and Haubrich, 1987; Fama, 1985). Ariga, Shima, Hutagami, and Kawaguchi (1994), looking at publicly-held Japanese corporations and banks in the 1980s, find that bank lending concentrates on companies whose default risk is high with low average stock return and high volatility. Houston and James (1996), however, find that firms with smaller size and lower leverage borrowed more among publicly-traded American firms. They also find that firms with high growth opportunities and intangible assets or with a single bank lender borrow less, indicating the substantial cost of bank information monopolies (even for larger corporations). For small firms, Petersen and Rajan (1994) find that borrowing from a single lender increases the availability of credit. Hosono (1997) finds that bank loan share to total debt had a negative relationship with R&D expenditure (as well as profitability and size), suggesting that the problem of information asymmetry is not serious for listed machinery companies in Japan. Hoshi, Kashyap, and Scharfstein (1993) find evidence of public financing being a way of insulating firms from bank monitoring: low Q, owner-managed firms are more prone to issue public debt for publicly-traded Japanese manufacturing firms.

• Scope of the relationship. The accuracy of information about corporate clients is increased through interactions in other financial services. The fixed cost of producing information about a firm can be spread over multiple products. By handling deposit accounts, the operation of settlement accounts, and foreign exchange transactions, banks acquire good information about their cash flows, liquidity situations, business partners, and the nature of their businesses. The scope is also important because the relationship (and the incentives to produce relationship-specific information) may be continued even when firms no longer rely on bank loans.
• Extent of multiple-bank relationships. Borrowing from a single lender, or loan concentration, is considered to represent a strong relationship banking compared with multiple-banking. This is so because the level of mutual commitment is high (smaller freerider problem) and the scope of the relationship is also likely to be large in a single bank relationship. However, there is a risk that the firm will be informationally captured by the bank. The firm cannot easily turn to other financing sources because other potential lenders and investors have little information about it.

Relationship and stages of monitoring

Relationship financing represents an implicit commitment by banks for additional financing to liquidity-constrained or financially distressed firms contingent on their viability in expectation of various rents to the banks (Aoki and Dinç, 1997).¹ For any corporate client, relationship banking involves information production through stages of monitoring. And there are strong complementarities among these stages. Aoki (1994) argues that relational financiers tend to integrate all the stages of monitoring given the difficulties of information transfer or complementarities among the tasks in stages. Monitoring by stage includes: ex ante monitoring (evaluating the risk characteristics of a borrower’s project before the initial financing); interim monitoring (watching over the borrowing firms after the initial funding to ensure that the borrowers can repay their debts); and ex post monitoring (closely examining the borrowing firms when they show signs of distress and working out a restructuring plan if necessary).

2.2. Incentives and Institutions in Relationship Banking

Relationship banking may be found anywhere around the world as a practice by individual banks. Even in the market-oriented system of the United States, commercial banks tend to practice relationship banking as a form of customer relations, particularly for small and medium-sized firms. Nevertheless, relationship banking is most prevalent in Germany and Japan. Banks typically serve not only as creditors but also as shareholders and are often represented on the client firms' boards of directors. German universal banks provide wide-ranging services including proxy voting and securities businesses, which helps them to maintain relationships with large reputable firms. Japanese MBs have been known for their commitment and reputation in delegated monitoring and extending assistances in times of financial distress.

German universal banks

German banks play an important role in corporate governance, which is oriented more toward internal than external mechanisms. Large blockholders and universal banks are central to the functioning of internal control mechanisms (Emmons and Schmid, 1998). It is likely that the most distinct features of German banks are their substantial control of corporate equity voting rights largely through proxy voting, and their representation on the supervisory boards of directors.

Proxy voting. German universal banks are rarely blockholders of corporate shares.² Most small shareholders designate a bank or a shareholder association to be their proxy. Universal banks have a competitive advantage in obtaining proxy voting power, as they provide the vast majority of retail brokerage services, and custodial services are needed as most equity shares are in bearer form.

Bank representation on the supervisory board. The supervisory board of German corporations is in charge of supervising management and appointing management board members. It consists of representatives of shareholders and workers in fixed proportions. Having a banker on the supervisory board may help reduce the problem of asymmetric information and lead to better credit support in times of financial distress.

There may be a conflict of interest when a bank exercises votes in its multiple roles as lender, adviser, equity holder, and voting agent (Baums and Randow, 1995). Even though German universal banks don’t seem to actively compete for proxy voting, they may actually be interested in soliciting proxies for various reasons: to protect the interests of the creditor or the value of their own equity investments; or to secure a share of corporate demand for financial services.

The Japanese main bank system

Aoki, Patrick and Sheard (1994) describe the Japanese main bank system as a nexus of relationships consisting of three elements: relationship between a MB and its clients; between a MB and other creditors; and between these parties and the government.

• Relationship between a main bank and its clients. An MB is usually the largest lender for the client firm, providing rescue operations and dispatching directors in times of corporate financial distress. Also, the MB maintains transaction and settlement accounts of the borrowing firms and serves as the trustee of collateral or the guarantor for bond issues. Finally, it is a substantial shareholder of its corporate clients. The close bank-firm ties are depicted as a quasi-internal capital market where the MB internalizes the monitoring and control functions that are undertaken by the capital markets in the Anglo-American system (Sheard, 1989).³ For larger firms, the MB relationship is an essential part of a broader alliance called keiretsu. Common membership in the Presidents' Club and significant cross-shareholding facilitate information exchanges and the coordination of decisionmaking as well as management oversight among affiliated firms. MB management intervention when the firm is performing poorly or in need of restructuring is regarded as an important substitute mechanism for the "missing" takeover market in Japan.
• Relationship between a MB and other creditors. The MB of a firm is implicitly delegated by other financiers to monitor the firm. This ensures that lending banks do not have a freerider problem while avoiding the costly duplication of monitoring for the same firm (as well as risk diversification). However, there is an agency problem between the MB and other creditors, because the monitoring activity is not readily observable by other financiers. Japanese MBs enjoyed a relatively high reputation in this respect until recently, with little evidence of shirking their delegated monitoring.

  Sheard (1994) provides several contributing factors for this high reputation of Japanese MBs in their role as delegated monitors. First, the MB tends to bear a disproportionately large share of any assistance burden or bank losses when a firm falls into financial distress or fails. This may be considered a penalty for the neglect of monitoring or as a device to make the MB’s evaluation of the troubled firm credible. Second, the MB is usually the main provider of financial services for its clients. In this role, it tends to provide insurance to corporate clients by receiving insurance premiums in good times and bearing losses in bad times (Nakatani, 1984). Finally, MBs that neglect their duties may face penalties in various forms. Other MBs may retaliate with similar actions, since they are mutual hostages as members of many loan syndicates. Also, banks with a bad reputation may be excluded from future arrangements of reciprocal delegation arrangements; and the regulatory authorities may exercise various forms of pressure and moral suasion.⁴

• Government regulation in support of the system. The Japanese government has long regulated/protected the banking sector, limiting competition among banks and against the capital markets. In this environment, it was relatively easy for banks to agree on reciprocal arrangements. The rents created by regulation provided an extra incentive for banks to continue behaving as good MBs by rescuing troubled customers and being faithful delegated monitors.

2.3. Merits and Demerits of Relationship Banking

Relationship banking can add value through its contractual features that, though mostly implicit, facilitate long-term relations (Ferri, Kang and Kim, 2001; Hoshi and Patrick, 2000).⁵

• Monitoring costs are economized through reciprocal delegated monitoring among credit suppliers, virtually making the loans of a relationship bank subordinate to other banks’ loans and public debt.
• Inefficient closures of distressed but economically solvent firms are prevented, and cases of corporate financial distress are effectively resolved.
• Liquidity constraints are mitigated and business risks shared between a relationship bank and its corporate clients over their cycles of cash flows and profits, since loans are made from a long-term perspective.
• Potential conflicts of interest between the creditor bank and shareholders are controlled through the holding of corporate shares by a relationship bank, easing the problem of asset substitution (investment decisions biased towards projects that enrich stockholders at the expense of debtholders; Prowse, 1990).

However, relationship banking is not without potential perils, which must be minimized by sound business judgment and discipline (Ferri, Kang and Kim, 2001; Hoshi and Patrick, 2000).

• Investment efficiency can be low due to soft-budget constraints. That is, given the good chance of loan renegotiations with their banks, firms with a relationship bank may have weaker ex ante incentives to boost their effort (Bolton and Scharfstein, 1996).
• A relationship bank might extract rents from its clients in the form of higher lending rates and others because they are informationally captured and have difficulties turning to other financing sources.
• Firms with a relationship bank may take too few risks in their businesses, as the bank will discourage investment projects with both high return and high risk.
• The system of relationship banking is often supported by heavy government regulation of the financial markets, which delays capital market (including the market for corporate control) development and results in inefficiency in the banking sector.

Given these merits and demerits, the effects of relationship banking on borrowing cost, credit availability, and corporate performance cannot be predicted ex ante. The average borrowing cost of a firm with a relationship lender will be lower only when the savings on monitoring costs and the positive effect of risk reduction more than compensate for the negative effects of the lender’s monopoly rent extraction. Credit availability is higher only when the positive effects of reduced information asymmetry and reduced risk (as well as soft-budget constraints) outweigh the negative effects of discouraged risktaking (lower investment and slower growth) and the information monopoly by the bank. Likewise, the impact on corporate efficiency and performance will also be determined as a net effect of the various positive and negative factors. Thus, we have to rely on empirical studies to answer these and other related questions on relationship banking.

2.4. Banks' Holding of Equity Shares of Client Firms

Relationship banks in Japan and Germany tend to hold equity shares of their corporate clients as a way of cementing the relationship. As already mentioned, this alleviates potential conflicts of interest between creditors & equity holders and the associated problems of asset substitution, and under- or over-investment (Jensen and Meckling, 1976; Myers, 1977). As residual claimants as well as creditors, the banks have stronger incentives and capacity to monitor client firms, and their incentives for the premature liquidation of troubled firms are reduced.

However, some side effects can also be expected from a creditor bank that is also a shareholder. First, if the client firms of a shareholder-bank face smaller credit constraints, these soft-budget constraints can lead to investment inefficiency. Second, the shareholderbank might use its stronger voice to distort corporate decisions to protect its own interests as a creditor (by discouraging risky but firm-value-increasing projects). This is a result of the fact that banks' equity stakes in client firms are usually much smaller than their stakes as creditors.⁶ Finally, a shareholder-bank’s power over its client firms can lead to the extraction of increased rents (Morck, Nakamura, and Shivdasani, 2000).

Flath (1993) finds that the largest debt holders in Japanese keiretsu firms hold more stock if the firms borrow heavily, have weaker collateral, have greater prospects of growth, or have high levels of spending on R&D or advertising. This finding is consistent with the expectation that creditor banks are more interested in holding the shares of client firms with potentially high agency problems. These firms include those with relatively high information asymmetry and temptations for asset substitution. He also finds that keiretsu firms in which debt holders hold more stock borrow more. Prowse (1995) also finds that banks' shareholdings are significantly correlated with their lending to the firm especially for firms operating in relatively risky environments. This complementarity between equity and debt holding by banks can be interpreted as a result of banks' attempts to protect their position as lenders, or the result of mitigated agency problems.⁷ According to a survey on cross-shareholding with financial institutions conducted on Japanese exchange-listed companies, few firms regard the cross-shareholdings as beneficial in terms of financing (lower borrowing interest rates or increased availability), but most expect support during financial distress and believe the cross-shareholding will be continued in the future (Wakasugi, Omura, and Miyashita, 1994).

The views expressed in this paper are the views of the author/s and do not necessarily reflect the views or policies of the Asian Development Bank Institute nor the Asian Development Bank. Names of countries or economies mentioned are chosen by the author/s, in the exercise of his/her/their academic freedom, and the Institute is in no way responsible for such usage.

Post a Comment

We welcome your feedback on this publication. Post a comment. ADBI is not obliged to acknowledge or publish comments and may abridge or edit them before web posting.

Comment(s)

There are [0] comment(s) for this entry. Post a comment.