امور مالی، کنترل و سودآوری: تاثیر بانک های آلمان
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|12977||2006||20 صفحه PDF||سفارش دهید||9607 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Economic Behavior & Organization, Volume 59, Issue 1, January 2006, Pages 69–88
Bank intermediated finance has been cited frequently as the preferred means for channeling funds from savers to firms. Germany is the prototypical economy where powerful universal banks allegedly exert substantial influence over firms. Despite frequent assertions about the advantages of a bank relation, empirical support is mixed. With a unique dataset and a focus on the fragility/sturdiness of inferences, this paper evaluates German bank influence in terms of three hypotheses: (1) do bank influenced firms enjoy lower finance costs? (No); (2) is bank influence a solution to control problems? (Yes); (3) do bank influenced firms have higher profitability? (No).
Channeling funds from savers to firms is one of the central problems facing an economy. In a frictionless world with widely dispersed and reliable information, financial assets are perfect substitutes and are allocated independently of existing financial markets and intermediaries. In this case, the institutional structure of the financial system is merely a veil. Recent research casts serious doubt on this frictionless model because of implausible assumptions about the availability and reliability of information. In the presence of asymmetric information, financial structure has major impacts on the allocation of funds, the pace of capital formation, and the performance of the economy. When lenders and firms face significant information asymmetries that create possibilities for opportunism by better-informed firms, banks may play an important role in financing and governing firms. Owing to economies of scale and scope, banks are arguably well positioned to finance and monitor firms. Germany is the prototypical economy where universal banks, which offer a wide-range of financial services, allegedly exert substantial influence over firms, and thus Germany is ideal for studying bank-firm relations and bank intermediated finance. The current study extracts testable implications from the German bank influence model (GBIM) and evaluates these hypotheses empirically. The widely held view of German bank influence has three major components: finance, control, and profitability (cf. Edwards and Fischer, 1994, chapter 1). German banks allegedly supply finance relatively cheaply because of technical expertise and superior information. The latter follows from bank representation on firms’ supervisory boards and long-term relations between banks and firms. The second component of German bank influence is that banks reduce managerial agency costs associated with corporate control. The superior information that lowers finance costs also permits banks to monitor management effectively. In addition to representation on supervisory boards, banks have substantial voting, power obtained either directly through ownership or indirectly through proxies, borrowings, or investment companies. Consolidated voting power, supervisory board representation, and long-term relations combine to provide banks with the potential to influence firms substantially. German banks thus would appear to have the power to reduce the agency problem at the core of the corporate control dilemma and, with large ownership stakes, the incentive to exercise control. In turn, lower finance costs for external funds and lower agency costs of corporate control have a favorable effect on firm profitability, the third component of the GBIM. While these arguments are certainly reasonable, and perhaps even persuasive, alternative perspectives exist, and the GBIM needs to be evaluated empirically. Several studies have pointed to positive roles that banks play in the success of the German system of investment finance. This confidence has not gone unchallenged. Wenger (1992) and Wenger and Kaserer (1998) have written extensively on the deleterious effects of German banks. Perlitz and Seger (1994) and Seger (1997) find a negative influence of banks on firm performance. In an important book, Edwards and Fischer (p. 240) conclude that “the commonly-held view of the merits of the German system of finance for investment, in terms of the supply of external finance to firms and corporate control, receives no support from the analysis of available evidence.” However, most existing empirical work on Germany is based on relatively aggregate data or single cross-sections. In her review of Edwards and Fischer's book, Elston (1995) notes that much of the debate on the merits of the bank-based versus market-based finance has stagnated simply due to lack of detailed empirical evidence. Gorton (1995) expresses a similar concern and calls for the use of firm-level cross-sections to assess the influence of German banks. Studies based on cross-tabulations of aggregate financial data, while informative, cannot capture the complexities affecting financing choices. There are many non-bank factors that vary across firms and influence financing decisions. Failing to control for these factors can seriously bias inferences based on aggregate data. Thus, microdata are needed to assess the nature and extent of German bank influence. This paper presents such a microeconomic inquiry by extracting testable implications from the GBIM. Section 2 describes the financial statement and ownership data for German firms. These panel data are transformed into a cross-section, and we discuss the econometric reasons for focusing on cross-sectional variation. The finance, control, and profitability hypotheses are examined in Sections 3, 3.1, 3.2, 4, 4.1, 4.2 and 5 in terms of three questions: (1) do bank influenced firms enjoy lower finance costs? (2) is bank influence a solution to control problems? (3) do bank influenced firms have higher profitability? Apart from a standard regression analysis, we also use Leamer's extreme value analysis (EVA) to assess the fragility/sturdiness of our inferences and to address concerns about multi-collinearity and equation specification. We find that bank influence is not associated with a reduction of finance costs nor a change in profitability. Our empirical evidence offers little support for the GBIM. Rather a view of bank relations begins to emerge in which banks provide corporate control services and are compensated by charging higher rates for borrowed funds and fees for a variety of other banking services. As discussed in Section 6, bank influence seems to serve as a substitute control mechanism, one of several available for addressing corporate control problems that does not appear to offer any net advantage to firms. This inquiry should be viewed as exploratory for several reasons. The GBIM is a collection of plausible economic relations and behavioral responses, but it is not based on a fully articulated model of optimizing agents constrained by information and agency problems and market forces. Thus, we are reporting some interesting conditional correlations, not identifying deep structural links. Moreover, we have not accounted for the possible endogeneity of bank affiliations. As emphasized by Demsetz and Lehn (1985), control mechanisms may be selected on the basis of particular characteristics of firms and their governance problems. While we control for several of these factors, we cannot entirely discount the possibility of endogeneity bias. Notwithstanding these concerns, explicit testing of the GBIM will be useful in generating a dialogue between empirical results and theoretical models. Indeed, some of the testable propositions require auxiliary assumptions that have not been fully appreciated in the literature. The goal of the current paper is to begin to develop micro-based empirical evidence that will inform views of corporate finance and governance problems and the possibly ameliorative role of close bank relations.
نتیجه گیری انگلیسی
This study has examined the German bank influence model (GBIM) with a unique dataset and a focus on the fragility/sturdiness of inferences. Three implications of the GBIM have been assessed. First, do bank influenced firms enjoy lower finance costs? German banks allegedly supply finance relatively cheaply because of their technical expertise and superior information. In addition to this direct finance channel, we also examine a certification channel that may lower finance costs indirectly. Our results suggest that finance costs are not lower for firms affiliated with banks. Second, is bank influence a solution to control problems? By combining superior information and consolidated power, banks are allegedly well positioned to monitor or discipline management. A sizeable equity stake may create substantial incentives to exercise corporate control. Bank influence and concentrated ownership are negatively related, which suggests that they are substitute means for controlling corporations. Third, do bank influenced firms enjoy higher profitability? The favorable financing and control outcomes associated with the GBIM, implies a favorable shift in the long-run cost curve and an increase in profitability (provided the firms operate in noncompetitive markets). This hypothesis is not supported in the current study. In contrast to Cable (1985), Gorton and Schmidt (2000), and Lehmann and Weigand (2000), we do not find a significant positive relation between bank influence and profitability. Our empirical results do not lend much support to the GBIM, and while deep structural links have not been uncovered, they permit us to draw a preliminary sketch of an alternative model of the German system of corporate control (though additional interpretations are possible). Control dilemmas are omnipresent, and firm owners address this problem by choosing a concentrated ownership structure or a bank affiliation (Table 3).6 Banks must be compensated for the resources employed in creating value for their client firms. Such compensation can take the form of fees for various services or borrowed funds (e.g., compensating balances) or interest charges that are above market levels.7 The empirical results in Table 2 indicate that bank affiliated firms do not hold greater amounts of bank debt, and these weakly negative relations are compatible with the following scenario: the firm participates in a loan market substantially controlled by its primary bank, the bank offers its profit-maximizing price for borrowed funds, and bank affiliated firms respond by borrowing less than independent firms facing lower effective loan rates. Relatively expensive bank lending charges and service fees survive in equilibrium because of the compensating benefits from the resolution of control problems enhancing economic performance. On balance, there is no net effect on profitability relative to independent firms (Table 4) as the benefits of control are counterbalanced by the inflated costs of banking services.8 This equilibrium is sustainable provided banks continue to be effective monitors, a role that has been questioned recently because of several prominent failures of corporate control. The results presented in this study and the above sketch of the German system of finance and control suggests several directions for future research. Demsetz and Lehn (1985) (United States) and Prowse (1992) (Japan) have shown that ownership patterns respond to the economic incentives for control. Our finding on corporate control substitutability raises the further question as to which policy and non-policy factors determine the type of control mechanism used in Germany. An additional issue is the purported role played by banks assisting firms in financial distress. Whether German banks function in a similarly supportive manner remains an important open question, especially since active support of financially distressed firms would impact the interpretation of the bank-profitability relation. Finally, data on service fees and the non-interest terms of lending would be valuable in developing a better understanding of the full cost of a bank relation. Further research on these issues should yield a deeper understanding of the structural characteristics of the German economy and the general nature of control mechanisms.