Paper submitted for presentation at the European Financial Management Association Conference, Athens, June, 2000 The authors would appreciate that the copyright of this conference paper be respected and that no part of it is cited without the permission of the authors. The paper was presented by Jose A. Reyes-Gonzalez. Correspondance should be directed to Penny Ciancanelli. Corporate Governance in Banking: A Conceptual framework Abstract In the wake of far reaching financial system reforms, almost three fourths of the member countries of the IMF experienced significant episodes of systemic crisis and associated bank failures.
Notably absent in the ensuing debates on the correlation between financial system reforms and systemic crisis was discussion of corporate governance in the affected banks and the role it may have played in the provoking financial crisis. Consideration of corporate governance in banks is, however, apparently easier said than done. While there is a great deal of empirical research on corporate governance, very little of it concerns the behaviour of owners and managers of banks; all of it assumes that banks conform to the concept of the firm used in Agency Theory.
The aim of this paper is to demonstrate the limitations of that assumption and to propose an alternative conceptual framework more suitable to its analysis. We argue that commercial banks are distinguished by a more complex structure of information asymmetry arising from the presence of regulation. We show how regulation limits the power of markets to discipline the bank, its owners and its managers and argue that regulation must be seen as an external force, which alters the parameters of governance in banks. Key words: Corporate Governance, Banks, Regulation, Agency Theory. 2
Introduction Between 1980 to 1997, over 130 countries, comprising almost three fourths of the member countries of the International Monetary Fund (IMF) have experienced important problems with their banks. (Lindgren, Garcia, Saal, 1996) The fact that these crises occurred after implementation of far reaching reforms of the financial system revived long standing debates in Economics and Finance on role of bank regulation. (Mishkin, 1992; McKinnon, 1993) Notably absent in the debate, however, is consideration of the corporate governance of banks and the role it might play in systemic crisis. Consideration of corporate governance in banks is, however, apparently easier said than done. While there is a great deal of empirical research on corporate governance, very little of it concerns the behaviour of owners and managers of banks. In addition, there is no clear theoretical path between governance as a microeconomic concept and regulation as a macroeconomic concept. There is, therefore, little guidance as to the conceptual framework that is suitable to understanding governance in banks.
This lack of guidance creates a strong theoretical motive for research on these issues. By defining a conceptual framework appropriate to governance in banks, it is possible to contribute to the further development of the microeconomics of banking. Specifically we seek to widen the scope of financial management research so that governance is understood as an integral part of the microeconomic foundations of what is called systemic risk in the banking literature. 2 (OECD, 1995; Davis, 1995; Lindgren, Garcia and Saal, 1996) This paper has two aims.
The first in to demonstrate the limitations of current approaches to the governance in banks. The second aim is to offer a more robust 1 By commercial banks, we mean private sector corporations that intermediate between depositors and borrowers. It is outside the scope of this paper to consider the effect of changes in the financial services industry on the varied ways in which regulators have sought to define a bank. 2 The chief features of systemic risk are well known: Runs (unexpected withdrawal of deposits), nexpected and rapid reversals by securities holders, excessive volatility in the foreign currency market and generalised symptoms of panic amongst financial asset holders. (Sundarajan and Balino,1991) The desire to prevent such episodes is the main rationale for national regulation and the fear of contagion through global systems of intermediation is the main rationale for international efforts at regulation. 3 conceptual framework. We do this by analysing the underlying assumptions of the standard theoretical framework adopted in corporate governance studies and demonstrate two complementary arguments.
Firstly, we show that the assumptions of Agency Theory make it unsuitable for analysing governance in commercial banks because regulations intended to prevent systemic risk (e. g. secure the integrity of the banking system) limit the disciplinary power of market forces. Secondly, we demonstrate that the agency problem in commercial banks is structurally different from that found in other publicly listed firms. Regulation, a transcendental feature of banking, alters the parameters of the agency relationship by introducing a third party–the regulator. This creates additional information asymmetries and associated agency problems.
This implies that any “moral hazard” to which regulation gives rise may be an empirically more robust parameter of poor governance than is the conflict of interest between the owners and managers of commercial banks. Indeed, it can be argued that liberalisation of banking regulation has amplified moral hazard because, on the one hand, the reforms have given banks the freedom to take greater commercial risks and, on the other hand, the regulator shares this risk through his remaining commitment to act as a “lender of last resort”. This paper has three main sections.
In the first section, we critically evaluate the standard framework of corporate governance and its suitability for banks. We argue that the use of Agency Theory implies the adoption of assumptions that obscure the unique governance dilemmas confronting bank owners, managers and regulators. In the second section, we propose a new conceptual framework for analysing corporate governance in banking firms. We are particularly concerned to integrate an analysis of the specific impact of regulation on both the nature of the market and the nature of the principal-agent relationship in bank firms.
In the third and concluding section, we elaborate the idea of a governance gap. We argue that the regulatory structure now in place provides bank owners with incentives to take risks that are greater than those deemed prudent by regulation. (Franke, 2000) The main manifestation of this gap is the correlation between deregulation of banking systems and rapid increase of non-performing loans. We are suggesting that 4 this pattern has its origins in a gap between the expected and actual incentives created by regulation. Section 1.
Banks and the Concept of the Firm Awareness that commercial banks are somehow different from other corporations may explain why there has been very little research–either empirical or theoretical–on their corporate governance. Previous research on corporate governance is limited mainly to empirical research using data on US banks. This research assumes that banks conform to the concept of the firm used in Agency Theory. That is, it assumes that moral hazard in banks is the same as that identified for any company in which control and ownership are separated.
One strand of this literature (Saunders, Strock and Travlos, 1990; Allen and Cebenoyan, 1991; Gorton and Rosen, 1995) is concerned specifically with managerial conduct. (Cf. Diamond, 1984; Glassman and Rhoades, 1980 and Edwards, 1977) The other strand considers the different elements of the corporate governance mechanism in banks. (Prowse, 1995; Houston and James, 1995; and Crawford, Ezzell and Miles, 1995) In both strands, banks are treated as if they were the same as any other firm.
It is possible that the wide publicity given to financial system reforms throughout the world have created the impression that banks have been completely deregulated. This is not the case, however. While most governments have adopted financial deregulation3 as a central element of the reform of national financial systems,4 banks and banking continue to be regulated. Deregulation has been about changes in the rules of the game rather than the abandonment of rules altogether. No country has adopted a complete laissez-faire approach to their financial systems. Vittas, 1992) The important implication for financial management studies is that the regulator and 3 As early as the middle of the 1980’s, some prescient observers pointed out the crisis potential of such rapid and far-reaching financial deregulation. (Diaz-Alejandro, 1985) 4 Three kinds of financial system are identified: stated directed and enterprises financed mainly by bank loans, bank directed and enterprises financed mainly by bank loan, and market directed and enterprises financed by debt and equity securities. (Arbor 1995) 5 egulation are a transcendental feature of bank management everywhere in the world today. From a theoretical perspective, previous research is even more limited. The application of Agency Theory to banks assumes that banks operate in the same type of competitive markets and are structured managerially by the same forces as all other firms. This assumption is completely at odds with that found in the banking literature. Specialists in banking studies (whether economists or political scientists) regard banks as different and distinct from “ordinary firms”, either because that is the empirical state of affairs (e. . they are regulated) or because their specific characteristics require regulation. (Ogus, 1994; Goodhart, Harmann, Llewellyn, RojasSuarez and Weisbrod, 1998) In either case the fact of (or need for) regulation makes it untenable to assume competitive markets. In order to illustrate the lack of fit between standard agency models and the firm known as a bank, it will be useful to review its assumptions and compare those to the characteristics of banks. Agency Theory makes at least hree assumptions: • Normal or competitive markets • The nexus of information asymmetry is the principal-agent relationship between owners and managers • Optimal capital structure requires limited gearing (The Modigliani and Miller Theorems) In contrast, one finds that commercial banks function • In regulated or administered markets • The agency problem is more complex • Capital structure is highly geared reflecting the bank’s function as an intermediary. Owners rarely provide more than 10% of funds loaned; bond holders and depositors provide the rest For governance, the more complex agency problem is of particular importance.
In addition to information asymmetry between owners and managers, there are at least three additional loci of asymmetric information in banks: • Between depositors, the bank and the regulator • Between owner, managers and the regulator • Between borrowers, managers and the regulator 6 The importance of these additional loci of information asymmetry suggest that the nature of the firm called a bank is qualitatively different from the nature of the firm implied by Agency Theory. 5 However, even though we consider Agency Theory to be of limited use in the analysis of bank governance, the agency problem remains an important conceptual tool. As we noted above, the existence of a more complex structure of information asymmetry is a distinctive feature of commercial banks. The question that remains is how might this difference matter. In order to develop our answer to this question, we need to discuss four theoretical issues in some detail. First, what is the nature of the market in which the firm called a bank acts? Second, what is the dynamic relationship between the market and the firm? Third, in theoretical terms, how might regulation affect the market and the firm? Fourth, what is the object of regulation, the firm or the market?
The nature of the market in which the bank operates The foundational work on markets and firms is that of Coase. (Coase 1937) He was the first to go “into the black box” called the firm and draw attention to the fact that the entrepreneur and his employees (factor) were integrated in firms by means of contracts. Jensen and Meckling appear to have supplemented the Coasian framework with the concept of asymmetric information in order to formulate the agency problem in a prescriptive way, e. g. with the idea that contracts could be written in such a way as to overcome the implicit information asymmetry between the factors integrated in the firm. We have identified four sources of moral hazard: Type 1, is the same as that discussed in Agency Theory and is the conventional two person ‘game’ involving the owner (as the principal) and the manager (his agent). Type 2, is that discussed in theories of bank regulation. The depositor is the principal; the bank is the agent. The regulator “insures” the depositor in order to limit systemic risk. Types 3 and 4 are the moral hazards linked to governance. 6 There seems to be some confusion in the literature as to the difference between Agency Theory and the agency problem.
In our discussion, we distinguish between the two on the basis of the assumptions required when using them, particularly those regarding the market. 7 Their original formulation of the agency problem is quite abstract and concerned with the general problem of control in the Coasian firm. We argue that the while the general notion of agency problem or agency relationship are sufficiently abstract, the particular concepts of the market and the firm, which underpin its use in finance research, are ill suited to the study of banks and therefore of corporate governance in the firm called a bank.
The concept of the market in the Coasian firm assumes perfect competition. However, perfect competition is not a workable concept for explaining the specific features of bank markets. (Campbell, 1994; Hodgson, 1993) We argue that the existence of banking regulation affects the nature of the markets in which the firm called a bank operates. Therefore the specific constraints in bank markets will have to be the building blocks of both prescriptive (normative) models of best practice in corporate governance in banks and empirical analyses (behavioural models) of the origins of control loss (weak corporate governance) in contemporary banks.
The dynamic relationship between the market and the firm called bank In the theory of the firm derived from Coase, the dynamics of the relationship between the market and the firm are not considered. Indeed, there are serious questions as to whether one can say that the concept of the firm in Coasian based theories constitutes a coherent theory of the firm as such. The firm is treated as if it is independent of the market and the market is treated as if it is independent of the firm. However, from a systems perspective, the market and the firm are not independent of each other.
The firm, in order to operate, must contract with other firms for its inputs. Therefore, it is possible to say that the market is the result of a sort of contracting between firms. Contracts with other firms are not the only ones that firms are required to make. They are also required to enter into contracts with regulatory authorities, such as local government, health and safety authorities, officials, and so forth. Thus, if the firm is a nexus of contracts between factors, the market is a nexus of contracts between firms and between firms and regulators.
The treatment of the firm as separate from the market may be an artifact of the partial equilibrium framework used by Coase. (Putterman, 1986) In general, financial management models derived from Coase also treat the firm as separate from the market and the evolutionary models of the modern corporation rarely make an 8 appearance. (Cf. Chandler, 1977) Furthermore, there appears to be a strong assumption that firms are the only logical alternative to markets as a means to allocate the factors of production. This is, of course, not true.
The firm emerges as the single alternative to the market only if one abstracts from the historical facts of their joint evolution. However, if one considers the variety of organisational designs which have developed historically, one is aware of the existence of a third type of firm: The regulated firm operating in a regulated market. Even Coase (1937) himself noted the effect that regulations such as tax would have for his argument, writing: “such a regulation would bring into existence firms which otherwise would have no raison d’etre” (p. 45)
In other words, regulation affects the efficiency calculus between markets and firms. Alternatively, if we are to abstract from history and accept that planning within firms or market price are alternative resource allocation mechanisms, what logic compels us to exclude the government as having powers to regulate allocation? Equally what reasons are there to exclude regulation from any source (including agreements amongst the firms themselves) as a logical third alternative mode of allocation to that of the price mechanism (the market) and the firm?
Of course, for purposes of modelling relations within the firm, it may be sensible to make ceteris paribus assumptions (e. g. treat the market as a parameter). However, a coherent theory of the firm must take into account how the behaviour within firms interacts with that in other firms to constitute market processes through the relays of feedback and spill over effects. Indeed, the lack of such a system perspective on the market makes it possible to argue that financial management has no theory of the firm as such.
On this point, Jensen and Meckling (1976) have written: “While the literature of economics is replete with references to the theory of the firm, the material generally subsumed under that heading is not a theory of the firm but actually a theory of markets in which firms are important actors. ” (p. 211) 9 Regulation and the Agency Problem Coase himself did offer a very brief consideration of regulation in the development of his general theory. He noted that governments and other regulatory bodies often treated the same transactions differently depending on whether they occurred in markets or a firm. Coase, 1937, p. 45) Few working within the Coasian tradition have attended to the potential implications of this observation. Of particular importance is its implication for the agency problem. If we accept that governments and other regulators behave in this way, the possibility exists that regulation creates a specific context for the writing of contracts between the firm and the regulator, and between the principal and the agent within the firm. We argue that bank regulations have quite specific effects on both the market and the firm.
The power of markets to discipline the bank s limited through regulations on entry, mergers, take-overs, administrative rules and so forth. Obviously, these “market” regulations also affect bank management decision-making. In addition, bank firms often face direct regulation, including statutory supervision, limitations on the scope of the business and fiduciary criteria for top managers. The latter implies that the labour market in middle level and top level managers is constrained. In some firms and jurisdictions, the mobility of middle level and top managers is also constrained. The Object of Regulation: The Market or the Firm?
What is the object of regulation? The market or the firm? Could regulation constrain markets and at the same time offer scope for opportunism by the owners and managers of the firm called a bank? In general, the literature on bank regulation emphasises the stated purpose of regulation as that of maintaining the integrity of the market system. To a great extent this is a legacy of the strong spill over effects of the financial markets crash of 1929. The spectre of debt deflation crises triggered by events in the financial sector continues to preoccupy bank regulators. Vittas, 1992; Hausmann and Rojas-Suarez, 1996; Rojas-Suarez, 1997) Attention is firmly focused on the management of exogenous shocks to the system through “lender of last resort” support from the government. However, lender of last 10 resort operations are inevitably indiscriminate; they bail out all banks, regardless of their previous conduct. In general terms, they shift the cost of risk to all the stakeholders in the system. We argue that crises may derive from poor corporate governance within the sector; in other words, bank governance is an important endogenous source of system weakness.
Excessive risk taking by the firm called bank has important implications for the market system because of economic agents’ common membership in a network of interdependent funds flows and the banks’ monopoly of the payments system. 7 This creates the potential for rapid transmission of governance failures in one bank to the banking sector and thereby to the rest of the economic system. We argue that studies of regulation and the regulators themselves devote less attention to corporate governance as an endogenous source of systemic risk than it devotes to exogenous or market sources of ystemic risk. Section 2. Regulation and Corporate Governance It is clear that developing models of corporate governance in banking requires that we understand how regulation affects the principal’s delegation of decision making authority (Jensen and Smith, 1984) and what effects this has on the behaviour of their delegated agents. (Freixas and Rochet, 1997) We suggest that regulation has at least four effects. Firstly, the existence of regulation implies the existence of an external force, independent of the market, which affects both the owner and the manager.
Secondly, because the market, in which banking firms act is regulated, one can argue that the regulations aimed at the market implicitly create an external governance force on the firm. Thirdly, the existence of both the regulator and regulations implies the market forces will discipline both managers and owners in a different way than that in unregulated firms. Fourth, in order to prevent systemic risk, such as lender of last resort, the current banking regulation means that a second and external party is sharing the banks’ risk. While outside the scope of this paper it may be useful to briefly address the claim that financial innovation have altered this. Financial innovation affects the extent and complexity of the credit linkages in an economic system. They do not, however, replace the fundamental requirement for an end-point. At the end of any chain of credit is the terminal transaction in which the debt is satisfied by the payment of cash and the wealth claim of the creditor is realised.
Banks continue to have the monopoly over the transmission of these essential ‘moments’ of the realisation of asset values and ownership claims in the economic system. 11 Regulation as an additional external governance force In the conventional literature on corporate governance, the market is the only external governance force with the power to discipline the agent. The existence of regulation means there is an additional, external force with the power to discipline the agent. This force is quite different from the market.
It acts in both the macro economic sphere (at the level of the banking sector) and in the microeconomic sphere (at the level of the individual banks). This implies that the power of regulation has different origins and different parameters, and therefore different effects to those produced by markets. Moreover, there is little ambiguity regarding the empirical references for regulation. Banking regulations everywhere are framed in law and have a precise, determinate institutional existence. Specific powers are granted while others are not.
This has two implications. First, the effects of regulation are open to empirical verification; there are definite instruments, means, actions, prescriptions and all of these are subject to tests of any set of behavioural hypotheses. Research can identify those behaviours that have their origins in the fact of regulation and cannot be assigned to generalised processes of market competition. Second, bank regulation reveals the existence of interests separate and distinct from the private interests of the firm.
As a governance force, regulation is intended to serve the public interest, particularly the interest of the consumers of the banking services. Moreover, regulation itself is enforced by an agent of the public interest–the regulator. This agent does not have a contractual relationship with either the principal’s firm or with the banking organisations as an interest distinct from the principal’s. For this reason, the relation between the regulator, the firm and its owners is usually modeled as a political relationship. 8 8
The hypothesis of regulatory capture of industries is widely used in political science and certain strands of industrial relations theory. It may very well be the case that bank regulators are substantively subordinated to the banking industry. However, our argument has a different focus. We are saying that, at a formal institutional level, the regulator is acting as the agent of the public interest and therefore, with respect to any given bank, is in a super-ordinate position. 12 Given the above, the external forces affecting corporate governance in banks nclude not only distinctive market forces but also regulation. The fact of regulation means that governance in banks must be concerned with not only the interests of owners and shareholders but with something called the public interest as well. Additionally, regulation and its agent, the regulator, have a different relationship to the firm than either the market, bank management or bank owners. This relationship is not limited to or defined by financial contracts and is not subordinate to but arguably super ordinate to the firm.
Regulated markets. Regulation creates constraints on market processes, limiting their nature and scope and subjects all firms to the threat of administrative action by the regulator. In standard corporate governance models it is claimed that the market has super ordinate powers over the firm because it is a staging ground for significant threats to managerial control (e. g. new entrants to the market, mergers, acquisitions, and take overs). The market in managerial labour is another staging ground for threats against manager’s misbehaviour.
The existence of specific controls on mergers, risk taking and on employees as fiduciaries all limit the disciplinary power of markets. Of course, the disposition of regulatory power will rarely occur without reference to and consideration of issues, structures, processes and concerns articulated by both the market and by individual firms. Nevertheless, it is accepted that the public interest is supposed to be the overriding consideration in the deliberations of the regulator. (Visentini, 1997).
At the most general level, regulation is associated with the resolution of market failure in provision of the public good of financial stability. Whether specific regulations fulfil this function or not, their existence necessarily alters the parameters of competition between regulated firms. In other words, the characteristic limitations (restrictions) imposed are not concerned with market structure per se (e. g. barriers to entry or market monopoly power) as a means to ensure competition. Instead, the constraints imposed by bank regulators in many countries attempt the opposite.
Regulators seek to constrain price and other forms of competition. They seek to restrain new entries, prevent mergers, acquisitions, take overs, and, in general, 13 promote only those mergers and acquisitions that assist in reducing system risk. In addition, regulations often establish minimum qualifications, require character references and other evidence of probity for persons considered for management positions within a bank. Hence, following up on the claim of Visentini (1997) that “the banking and financial markets take on the characteristics of administered markets” (p. 75), one could argue that regulations are intended to restore a welfare rather than a competitive equilibrium. Regulation as representing the public interest. In the conventional view of the firm found in financial management theory, attention is mainly focused on identifying what we will characterise as those interests internal to the organisation. These include such things as the maximisation of wealth as the principal’s primary (if only) interest. To fulfil that interest the agent has a well-specified objective function: the maximisation of shareholder wealth.
The manager is expected to act and take decisions on behalf of the owner’s interest. The implication is that any system or method of corporate governance ought to take as their objective safeguarding the interests of the principal, which means maximising shareholder wealth. However, as we have seen, in the banking firm, there exists another interest–that of the regulator acting as an agent for the public interest. This interest exists outside of the organisation and is not necessarily associated, in an immediate and direct way, to maximisation of bank profits.
The mere existence of this outside interest will have a profound effect on the construction of interests internal to the firm. Thus, because the public interest plays a crucial role in banking, pursuit of interests internal to the firm requires individual banks to attend to interests external to the firm. This implies a wider range of potential conflict of interests than is found in a non-bank corporation In bank corporations, the agent should respond not only to the owner’s interest, but also to the public interest expressed by regulation through administrative rules, codes, ordinances, and even financial prescriptions.
In order to assure and to protect the public interest, regulation imposes a form of external governance on the agent. Thus, 14 the agent is monitored by the regulator, to prevent misconduct. If the manager does not act in conformance with the regulations, he or she can be disciplined through extra market administrative action including the possibility of being excluded from employment in the sector altogether. The above implies that the bank’s manager must function in light of two distinct sets of interests. One is the private interests internal to the firm; the other is the interests external to the firm.
From the perspective of governance, the agent will seek to ensure that behaviour beneficial to the firm’s interest does not compromise the public interest. Sharing the bank’s risk. A centrepiece in analyses of the agency problem is the proposition that the owners’ interest may be affected by the self-regarding actions of the agent. In banks, the main action the principal delegates to the agent are lending decisions. The owner or owners is or are thought to bear the risks taken. This fact, together with the delegation of the firm’s management to a second party, creates the rationale of the so-called agency problem.
Thus, as the contingent claimant on the organisation resources, shareholders bear any business risk that the firm faces in its everyday of operations. It is often assumed that on average owners/investors are risk averse. (Fabozzi, Modigliani and Ferri, 1994) Thus, investors seek to minimise risk for a given level of return. Therefore, one of the main objectives of corporate governance is taken to be the creation of decision structures which prevent the agent from engaging in activities which expose the investor to a higher level of risk than that desired by the shareholders.
Therefore, proper governance is deemed to require systems that prevent this problem, such that the agent finds it difficult to take higher risks than desired by owners. In banks, the framework of action, motivation and behaviour is quite different. Because current banking regulation is concerned first and foremost with the existence of systemic risk, regulation applies those policy instruments deemed effective in limiting systemic risk. Of those instruments, the lender of last resort and systems of 15 deposit insurance are the ones deemed to be the best means to prevent contagion, bank runs and other threats to system integrity.
From a governance perspective, however, the presence of these policy instruments dramatically changes the relationship between the agent and the principal in banks and the conceptual framework required to understand it. First of all, these standard policies of system risk limitation imply that bank owners are in a risk sharing relationship with an external authority. The business risk, that in ordinary firms would be borne totally by the shareholders, is now only partially assumed by them. Of course, in ordinary firms, creditors and other commercial entities take some risk with any firm they do business with.
However, because the firm called a bank is in a risk sharing relationship, it is possible that it is willing to assume much higher levels of risk than unregulated firms. Indeed and perversely, excessive risk taking in lending is the most rational course of action by bank firms precisely because it is, in a sense, a one way bet. If the risk taken leads to a very high return, the bank pockets excess profits. If the risks taken result in a bankruptcy that is perceived to be threat to the system, the owners are bailed out.
It has been documented in the bank literature of various countries that some banks are “too big too fail” and regardless of the risky lending behaviour engaged in, they are inevitably bailed out because not to do so would threaten the banking system. (Miskin, 1992; Vittas, 1992; Goodhart, Harmann, Llewellyn, Rojas-Suarez and Weisbrod, 1998) In principle, there appears to be no upper limit to the risk the regulator is willing to bear, the question that arises from a governance perspective is would investors (principals) in banks have an incentive to encourage excessive risk taking by their agents?
Banking regulations normally include some attention to risk taking by bank managers. However, they usually do not impose such limits on owners. Therefore, if the shareholders are able to engage in riskier behaviour than considered desirable by the regulator, governance in banking might need to focus on the owner rather than the manager. 16 The view that bank owners may be more problematic for the regulator than the managers may seem arguable.
However, we argue that regulation, in addition to establishing risk sharing between the government and the owners, also and at the same time gives them additional freedom of action. This is because regulation limits competition and therefore owners are protected against the discipline of the market. In contrast, the agent may actually have less freedom than managers of other types of firms since they are often directly scrutinised by the regulator.
To summarise our argument thus far, we are saying that a theory of corporate governance in banking requires consideration of the following issues: • Regulation as an external governance force separate and distinct from the market • Regulation of the market itself as a distinct and separate dimension of decision making within banks • Regulation as constituting the presence of an additional interest external to and separate from the firm’s interest • Regulation as constituting an external party that is in a risk sharing relationship with the individual bank firm Theories of corporate governance in banking which ignores regulation will misunderstand the agency problems specific to banks. This may lead to prescriptions that amplify rather than reduce risk. In other words, if one accepts that regulation affects the banking sector in an important way, one must also accept the fact that this has important implications for the structure and dynamics of the principal agent relationship in banks.
Moreover, if the structure and dynamics of the principalagent relationship are different, we cannot assume that the lessons derived by Agency Theory research on non-regulated firms are relevant or useful. Certainly, they cannot be applied in a mechanical way. Section 3. The Governance Gap in Bank Regulation In this section we explore in greater detail how governance at the firm level is in a dynamic feedback relation with regulation at the system level. In our view, a governance gap has been created by the deregulation of financial systems without considering their implications for the management of banks. Episodes of difficulty in 17 the banking sector correlate strongly with the reforms and evidence an increase in systemic risk.
The reason for increased systematic risk is that liberalisation lifts important constraints on the allocation decisions of bank managers and leaves to the collective decisions of the market the “price” or level of interest rate to be charged. Absent robust structures of corporate governance in bank firms, existing regulations make it difficult to prevent rational owners from taking excessive risks with depositors’ money. The extent of the banking problems emerging in the wake of the financial reforms has differed from place to place: In some cases it has affected the entire national banking system, in others it has been restricted to the major banks only. In Japan, the world’s second largest economy, serious trouble in the banking sector continues without much respite after nearly six years of concerted government efforts. More recent manifestations of structural weaknesses are found in Asia and Southeast Asia. South Korea and Indonesia have experienced profound and destabilising shocks to the entire array of economic and political institutions. 10 Because of the pervasive importance of the banking system to the operations of national economies, these failures have attracted a great deal of attention from policy makers and public policy researchers. Of specific concern to them are the fiscal consequences of lender of last resort interventions.
Thus, the fiscal cost to the Mexican government for rescuing commercial banks is estimated to have now reached $100 bns or 20% of GDP; in Japan’s case, it is thought that the cost to the government has now reached $300 bns. 11 The OECD (1995) has classified the purposes of regulatory policies into three main categories: 9 There is a degree of judgement in these classifications. However, following Sundarajan and Balino (1991), crisis refers to cases of runs, other substantial portfolio shifts, collapses of financial firms requiring general government intervention. They classify evidence of generalised weakness of the banking system as a significant problem. 10 Rojas-Suarez (997) has pointed out that the Latin American region offers numerous examples of eficiencies in regulatory and supervisory procedures that, she argues, causes the newly liberalised financial institutions to operate under a system of “distorted incentives”.
11 The Economist, November 6th-12th 1999. 18 • To meet government objectives on resources allocation • To provide instruments of monetary control • To correct market failures12 The past decades’ reform of financial systems involved the elimination of most regulations in the first two categories. Having liberalised credit allocation and eliminated ceilings on the rate of interest, regulators turned their attention to the issue of market failure. (Caprio, Atiyas, and Hanson, 1994)13 The issue of market failure is linked to systemic risk and the monitoring of the behaviour of financial intermediaries, especially banks.
According to the OECD, systemic risk “refers to potential threats to stability of the financial system as a whole arising from risk taking by individual financial actors”. (1995, p. 11) The discussions of these issues are mainly concerned with the protection of the consumers of banking services. However, from a governance perspective, management of systemic risk involves three main types of policy instruments including: • Prudential supervision such as minimum capital standards14 • Lender of last resort and deposit insurance • Policies that limit competition We discuss each of these in turn, focusing on issues related to bank governance issues.
Prudential Supervision The deregulation of credit allocation implied the need for regulatory expertise in assessing the overall risk exposure of individual banks. However, as the crises revealed, the regulators often lacked the expertise required for prudential supervision of risky lending practices. The near universal adoption of the Basle capital adequacy framework reflected an official consensus as to the minimum of owners’ capital that ought to be at risk in lending. Losses from risky lending practices were to have been 12 This refers to market failure seen to arise from very rapid internationalisation and increased competition between financial institutions. Swary and Topf 1992) 13 In some developing countries, privatisation was a matter of transferring the ownership of public assets rather than to improve banking performance through market forces. (Reyes 1994) 14 The most relevant international event in this regard was the so-called Basle Agreement on the measurement of bank capital and agreement to minimum thresholds. (1988) 19 absorbed by additional owners’ capital. However, because of the agency problems discussed earlier (strong information asymmetries between the regulator, the bank and the weaker commitment of investors) owners were able to shift costs onto regulators and few banks increased the capital asset ratio in line with the riskiness of their lending decisions.
Lender of Last Resort and Deposit Insurance The lender of last resort is an important regulatory power normally assigned to the Central Bank. Deposit insurance is often compulsory and is under the control of either the Central Bank or a non-governmental regulatory institution. Both the lender of last resort and deposit insurance are intended to function as the joint and several means available to the regulator to prevent runs and to manage severe liquidity problems. It is widely recognised that however effective these instruments may be in fulfilling their stated purpose, they present certain characteristic problems. We are arguing that these characteristic problems are intimately connected to the quality of corporate governance in banking.
In the case of the lender of last resort function, the danger is the possibility regulators may confuse generalised liquidity problems with an insolvency problem provoked by weak financial controls at the firm level or deficient overall corporate governance in the banks that make up the banking industry in the country. This seems to have been the case in Finland, (1991-1992), Norway (1988-1992), Sweden (1991-1993). The failures to prevent and then carefully identify the origins of the crisis have meant that the costs of the crisis have been shifted arbitrarily to the public purse. In the case of Sweden, for example, the fiscal cost was over 3 percent of the GDP (OECD, 1995).
In Venezuela (1994) the cost of restructuring the banking system was 13 percent of the GDP (Hausman and Rojas-Suarez, 1996) and in Mexico is expected that the fiscal cost of rescuing the commercial banks could be around 20 percent of the GDP (Lopez, 1999). It is also well understood that there are hazards associated with deposit insurance. Its existence reduces the incentive for both bank management and depositors to attend 20 to the overall risk exposure created by lending decisions. The perverse results of deposit insurance schemes are evidenced by, for example, the costs of public rescue operations in the US; these are estimated at 3. 2% of GDP for the period 1980-1992.
Restrictions on competition Restriction on competition is accomplished through regulations directed at new entrants to the sector and regulations which limit the behaviour of existing members of the sector, including constraints on mergers, acquisitions and take overs. Regulation of ownership has several purposes: To prevent the creation of structural sources of hazard, such as, lenders and borrowers being controlled by a common owner or corporate group; To prevent excessive market concentration and other sources of conflict of interest between the financial sector and the rest of the economy. Restriction on ownership may or may not be associated with economy wide antitrust regulation. Whatever its form or source, it is must be accepted that limitations on ownership constitute intervention in the operations of markets that Agency Theory assumes are available to discipline managers.
Policy instruments devoted mainly to systemic risk impose a certain degree of restrictiveness on market forces. Therefore, one of the most important aspects of the effectiveness of the standard framework of regulation is the creation of barriers to competition in the banking sector. This has far reaching consequences for governance in banks, especially for the agency problem within banks. However, up to now, financial deregulation has concentrated more on the ability of banks to conduct business at market clearing prices than on the nature of the market itself. We argue that from a governance perspective, this constitutes a strong gap in control.
On the one hand, banks can conduct business at market clearing prices, and, on the other hand, they do not face the discipline of a competitive market. Additionally, regulators do not appear to realise that corporate governance in banks as an endogenous source of systemic risk. The most important evidence for this argument lies in two often overlooked features of many of the commercial banks implicated in recent crises–the owners of the banks 21 were often its top managers and accounting information on non performing loans was limited or non-existent. This may help to explain why systemic crises were triggered in countries as different as Sweden and Mexico.
The common factor appears to have been a “governance gap” created when the banks were deregulated without attention to the need to craft new governance structures suitable to the new conditions. Concluding Comments We have argued that financial regulations have structural consequences at two levels: on the sector as a whole and on the individual firms that compose the sector. In other words, regulation creates a unique type of firm whose specific characteristics have only recently begun to attract the attention of financial management scholars. (Freixas and Rochet, 1997) Because regulation imposes constraints on market forces and because the regulator acts on behalf of the public to share risk with the owners, corporate governance of banks must be seen to be different from that of the average or typical firm.
Thus, in banks, it is to be expected that (I) the problem of governance will be more complex (ii) the relationship between the agent and the principal is unique in being mediated by an external force and (iii) the owners may be considered as the single most important source of moral hazard. Therefore, in general, we argue that a conceptual framework for corporate governance needs to include (as a management control issue) the extent to which operating practices in banks amplify systemic risk and generate crises and to analyse the ability of the new rules of the regulation game to contain them. The views of Fama (1980) on differences in the “commitment” of owners and managers to the firm as an organised entity are relevant at this point.
Fama argues that the corporate form and developed financial markets makes owners less committed to a specific firm than its managers will be. Fama writes: “… the risk bearers in the modern corporation also have markets for their services–capital markets–which allow them to shift among teams with relatively low transaction costs and to hedge against the failing of any given team by diversifying their holdings across teams. ” (p. 137) 22 Thus, investment capital is likely to be more mobile than the managers of a firm. Fama (1980) writes: “If there is a part of the team that has a special interest in its (the firm’s) viability, it is not obviously the risk bearers”. (p. 37) If one accepts this as valid in general terms, the implications for banks is that the commitment of investors in banks may be even weaker than that found in other firms. This is because the owners of banks share risk with the regulator. Thus, bank owners have less incentive to monitor excessive risk; indeed, they have an incentive to take risks greater than those deemed prudent by the regulator. The governance gap created by deregulation places them in the unique position of sharing the costs of risk but not the profits earned in taking it. Indeed, if any owner has a directive or executive power in the bank, such a bank might be even riskier than other banks since they would be in the position to override the stronger commitment of the managers to the continued existence of the organisation as such.
Indeed, the fact of the regulation of the banking labour market may be expected to reinforce the commitment of the manager making it higher in banks than in other firms. However, if corporate governance proposals to “align” the interests of the principal and the agent result in actions (e. g. profit related bonuses, stock options and so forth) which make the bank manager a bearer of risk as well, her or his superior commitment to the firm called a bank could be reduced thereby increasing the moral hazard problems. Our analysis implies that owner managed banks are potentially the greatest endogenous source of systemic risk. Such banks are in a position to engage in the riskiest of actions knowing full well that they may keep the rewards without bearing the full cost of the risks taken.
This may explain why owner managed banks were key actors in the recent crises in Eastern Europe, Latin America, Southeast Asia and Asia. The reforms undertaken in many countries altered the parameters of control in banking firms in ways that appears to have amplified moral hazard and in certain cases to have created fertile soil for opportunistic behaviour by both the principal and the agent. Governance structures that may have been suitable in the prior regulation 23 regimes were evidently an unexpected source of systemic risk in the reformed regulation regimes. It is accepted that each country has a specific sort of corporate governance and diverse methods of corporate control as result of its history, economic and business culture. Ciancanelli and Reyes, 1999; Ciancanelli and Scher, 1999; Prowse, 1994) 15 Therefore, it follows that there is not one “best” type of corporate governance. At the same time, one can say that all corporations, whatever the particular forces that constrain their behaviour, act in light of both internal and external rules. The latter emanate mainly from the legal system, regulation and codes of conduct. (Cadbury, 1998) Of these the government and regulatory bodies are deemed to play a transcendental role in corporate governance. (Charkham, 1995)
The observed forms of corporate governance of banks emerge in the course of their operations as entities having to respect the private interest of owners, on the one hand, and the public interest in the overall stability of the system, on the other hand. Visentini, 1997) Thus, if one accepts that the banking regulatory framework is one of the most important external forces in shaping the behaviour of banks, a theory of corporate governance needs to address the integration of both internal and external forces in order to attain an optimal balance of public and private interests. 16 15 In U. K. , as result of the Cadbury Committee (1992), corporate governance entered the national debate regarding the relationship between the public and private interests. 16 In the case of U. K. checks and balances upon the power of the CEO have been legislated following up on the recommendations of the Cadbury Committee, an independent investigator into the issues set up by the national government.