Principal Agent In Banks 5d2o1b

1.0 Introduction The principal – agent problem is an idea derived from agency theory, formulated by economists and political scientists, first described by S. Ross in 19731. It usually gets analyzed and disputed in characteristic of those areas of science. In some aspects, agency theory and problems arising from it may be described as having a universal impact on social sciences as a whole, whereby just as in game theory, the incentives attributable to the parties involved may be used to predict people’s behaviour, design organizational systems and draw contracts between parties. Therefore, it is not only possible but desirable to discuss agency theory and the principal – agent problem in of law and legal economics, as the tools derived from analysing the incentives ruling parties’ actions, may help shape legal relationships2. Below I will focus on how the principal – agent models are shaped by law and regulation in companies and their most sophisticated form, banks. I will briefly describe the principle – agent relationships existing within corporate organizations first, in order to focus on describing the specific characteristics and problems of agency based relationships in banks as a notably specific type of company. Finally I try to suggest how bank regulation and governance may ultimately influence the shape that these relationships take, minimising the agency risk.

2.0. The Principal – Agent Theory To briefly describe what is understood as a principal-agent model, I find it necessary to state that the framework occurs in any social, political, or legal situation where two parties align themselves to fit a situation where one party, or the “principal”, with which authority to act originally lies “hires” or diffuses some of this authority onto another party, know as the agent. The agent is the party receiving the authority to act on behalf or in the interest of the originating party. The reason for this alignment from

1

S. Ross, “The Economic Theory of Agency; The Principal’s Problem”, American Economic Review, G3, p 134-139 2 K. Eisenhardt, “Agency Theory: An Assessment and Review”, The Academy of Management Review, vol. 14 no. 1(Jan 1989), p. 57-74

the principal’s side, is that the agent will be able to perform the actions handed down to him in a more efficient way than if the principal would perform them himself. In a perfect world, where information is spread out efficiently throughout the system, both the agent’s and principal’s interests would be ideally in tact, so that the agent would act in perfect accord with the interests of the principal. However, no such system can be identified in reality. First of all, information in any principal-agent model will not be shared symmetrically between the parties. This arises as a consequence of the principal’s delegation of power to the agent. Secondly, however the agent will be working on behalf of the principal, his interests may not be aligned to those of the delegating party. Instead he may be tempted to use his powers in order to satisfy his own interests suring those of the principal. There are three main issues identifiable as risks to achieving an efficient balance in any principal - agent relationship. These main issues are adverse selection, moral hazard and non – verifiability. Adverse selection occurs when the agent may come into possession of information that the principal does not know of. Moral hazard may be described as a situation where the agent may act against the will of the principal unbeknownst to him, possibly to the detriment of that principal’s interests. The last problem is when both parties may own the same information but this remains unknown to any third person3.

1.2. The Principal – Agent Problem The problems mentioned above all may have an influence on how an agency contract will be performed. A “lemons argument” may be used to describe the possible dangers implied in agency relationships. According to this idea, when the agent has an informational advantage over the principal for whom he’s working, he will be tempted to do two things, first of all hide this advantage so that it wouldn’t get discovered by the principal and secondly to maximise the profits he makes out of the relationship. This may be described on the example of a car dealer who trades in proverbial “lemons” and “peaches”, knowing which car is which this dealer, without adequate supervision or disclosure duties will not disclose this information to his customers who are, in this relationship , the principal. Furthermore, this car dealer 3

G. Wood, “Governance or Regulation? Efficiency, Stability and Integrity in the Financial Sector”, Journal of Banking Regulation, vol. 7, no. 1/2 2006, p. 1 - 17

will also try to price all his cars at a relatively similar, but usually artificially high level in order to further obscure his client’s ability to navigate his way through the situation. To an economist, this situation is undesirable due to the fact that it creates inefficiency in the market. To a lawyer, the dealer is committing an act not different to fraud, by choosing to omit significant information about the service he is performing to his principal4. Whatever the approach might be this inefficiency will lead to an unbalanced market. G. Akerlof in his paper suggests that when such an inefficient market with principal – agent problems is identified, the government as a party ruling objectively, as if from behind a veil, may introduce regulation that will effectively “benefit the welfare of both parties”5. The government may introduce uniform regulation that will systemically adjust the incentives of parties, for example increasing the effective costs of withdrawing information by the management of a company from its shareholders, through introducing civil and criminal penalties and prescribing compulsory disclosure of information to a third party auditor6.

3.0. The principal – Agent Framework in Companies What gives the above informational asymmetry problem further significance is its application to relationships within structured organizations and the risks specific to the relationship that eventually have an influence on the governance of these7. Not all corporations may be influenced and suffer from the principal – agent problem in the same way. For example, SMEs8 in general are owner – managed, which makes the principal agent problem irrelevant. The true significance of the problem becomes most clear in companies where ownership structure is less concentrated and the performance of the agent’s duties will regularly mean his privy to specialist and confidential information. This description suits banks perfectly, as most large banks are public companies and the management of a bank will constitute daily with specialist information possibly incomprehensible to the principal, but ultimately desirable by him. What adds significance to the problem is the systemic importance of 4

G. Akerlof, “The Market for “Lemons” Quality Uncertainty and The Market Mechanism”, The Quarterly Journal of Economics, vol 84, no 3 (August 1970), p. 488 - 500 5 G. Akerlof, “The Market…”, p. 488 6 See for example S. 489 of Companies Act 2006, relating to the compulsory appointment of third party auditors in public companies. 7 E. Douglas, “The Simple Analytics of the Principal – Agent Incentive Contract”, 8 Small and Medium Enterprises, mostly being private companies.

banks and their ability to “make or break” a healthy economy. Through their activity of lending money both on a micro an a large scale they are able to facilitate large scale corporate investment and growing consumption by allowing easy access to credit by retail clients9. Information abuse of agents working within these organisations on a systemic scale will upset the whole banking system, creating inefficiencies on an economy – wide scale. If persistent, and unmanaged, it could possibly lead to banks failing abruptly, in a short period of time that would create further disturbance to the system10. Authors agree that market exit in itself is does not constitute a danger to a financial system and in economic may even be identified as healthy as it gives the incentive of introducing market discipline to the banking system. Most states choose to provide their banking sectors with additional regulations, and special regimes that enforce stricter rules on these entities, in order to ensure they are well managed and transparent, not in the aim that market failure as a whole could be avoided. The goal of regulation rather is to limit the inefficiency of the market as a whole as identified by the moral hazard in the “lemons” argument11.

3.1. Shareholder – Manager Relationship The most commonly recognised principal –agent relationship found in enterprises in general is the relationship between the shareholders and the management of the company. This relationship is the most obvious one as it resembles the classic agency agreement under which the principal gives the authority to act to a party that can perform the necessary task more efficiently. Residual evidence of these ancient roots of company law remains to this day in the structural provisions of company law constituting the of a company to be the sole organ responsible for appointment of directors. This basic relationship is clearly visible from an economist’s perspective. However, when we look at it from a legal point of view it is no longer this clear, as company directors, once appointed, legally are bound to the company as a fictional legal person than to the interests of particular shareholders or as a collective, with a stronger bond. This means that directors’ duties in English law 9

K Alexander, R. Dhumale, J. Eatwell, “Global Governance of Financial Systems; The International Regulation of Systemic Risk”, 2006, p. 242 10 G. Wood, ”Governance…” p. 15 – 16 IT should be noted that the failure of a bank that is extended over a long period of time, basically amounting to a slow degradation and eventual disappearance will not create negative effects on the system, in fact may be regarded as strengthening the rest of the economy. Overnight failures, as noted by the author are what creates a negative impact. 11 T. Padoa – Schioppa “Regulating Finance”,2006, p. 1 - 2

have a certain duality. Despite the fact that a director’s performance will be assessed from a purely principal – agent relationship perspective when his mandate expires and he will stand for reelection, for the duration of his term, he is bound to follow the interests of the company first. This is the directors’ fiduciary duty to act in the benefit of the company12. As observed by some authors this duty should be observed that in the first order, the company’s profit making needs must be observed. Shareholders may be thought of only in the second order, even though ultimately it is their profit the director is working to achieve as the company is only but a commercial entity. Although directors have a number of other general duties, the one mentioned above is treated as overarching as it sets a direction in which all other duties ought to be executed13. To reassume, this basic principal - agent relationship applicable to all companies will be a simple contract between the directors of a company and its’ shareholders, that once entered into transforms into a relationship where the company serves as a medium for the performance of the task of shareholder wealth maximization.

3.2. Other Principal – Agent Relationships in Companies However, it can’t go unnoticed that a company is much more complicated than just one principal – agent relationship. By any means, in order to get a glimpse at the true nature of any company it must be realized that they are in their purest essence a complex nexus of contracts, with much more than just one agency relationship built into them. This is because any given company although not a real person will form an organism performing more than one function, influenced and influencing other actors in the environment it is set to act14. These obligations will arise out of a number of principal – agent agreements entered into by the management. First, they will carry the responsibility towards their company’s creditors. Even though their responsibility towards the creditors will only become a practical issue, as companies approach insolvency, the directors will in reality be responsible to perform a task for the 12

This rule was expressed for example in Percival v Wright [1902] 2 Ch 421 and incorporated into legislation under Section 172 of Companies Act 2006 as the duty to promote the success of the company. It must be accented that the duty is to promote the company’s success, not that of its’ shareholders. 13 S. Griffin, “Company Law; Fundamental Principles”, 2005, p. 313 14 For theories of the firm see O. Hart, “An Economist’s Perspective on the Theory of the Firm”, Columbia Law Review no. 89, p. 1757-1774

creditors as their principals, this task being to ensure their obligations are performed to the fullest. In Brady v Brady Nourse LJ expressed the opinion that when a company arrives at the threshold of insolvency, the interest of the company becomes that of their creditors15. The management of a company, being in this case the agents of creditors, may be held personally responsible if an abuse of their informational advantage is discovered16. The same bond, but with the opposite balance will exist where a company is itself a debtor. Here it will be the principal in relation to a different company or person being its creditor, and ultimately agent. It may also be said that there is a bond between the management of a company and its employees. Although this view is disputed and employees are generally treated as outsiders under English law, I can understand that in some way the managers may be understood to perform tasks on behalf of employees. In this case it would only be appropriate that the management while performing these tasks avoid abusing their informational advantage. However the ruling approach to this issue will be that a company’s employees are always only auxiliary to the company and therefore should not be treated as principal in a relationship with the management17.

4.0. The Particularity of Banks Banks are essentially intermediaries in the area of finance. They gather capital from depositors who locate it into their s. This capital will constitute for a banks liabilities. Allowing clients to locate capital in their s, banks provide liquidity to them, simultaneously allowing depositors to pay out their money from s usually whenever they want. A bank’s assets will be the investments a bank makes in the form of loans and other contractual agreements that put a bank basically in the position of a provider of capital18.

15

[1988] BCLC 20 at p. 40; I can only find reasons for the application of this principle to banks, as well as other companies. It is always the ultimate issue in a bank failure, to secure the payouts of deposits to depositors who are anything if not a banks creditors (even though the actual of a deposit contract with a bank are different to those of a credit agreement). 16 See for example West Mercia Safetywear Ltd v Dodd [1988] BCLC 250 17 See, S. Griffin “Company Law…”, p. 316 18 S. Heffernan “Modern Banking”, 2008, p. 1-5

4.1. Additional Relationships From this basic outline it becomes visible that the principal – agent framework is more complex than that of other business enterprises. What distinguishes a bank’s principal – agent framework is that in addition to the standard relationships between the managers, shareholders, creditors there will be more element s active. Secondly, because of the nature of a bank’s activity these enterprises are essentially more sensitive to informational disparities between the parties forming principal- agent relationships within their structure. Because of the nature of bank activities, shareholders cannot effectively oversee every decision a bank makes, as this would render the whole business concept functionally impossible. Depositors in a bank have very little say as to how the bank invests their funds and find it hard to monitor these decisions at all. Bank management, does not have the tool s necessary in turn to influence when depositors will be able to withdraw their funds, essentially leaving banks prone to “bank runs” and changing moods based on inadequate information. A depositor will ultimately find himself in a position of a principal in atypical asymmetric information sharing model where he cannot perform effective supervision of his funds because of a number of reasons. This is de to a number of reasons, for example a bank will pool the capital it receives from deposits and diversify it into a large number of loans to limit losses. However losses will always exist, in order to supervise him capital the depositor would have to monitor a very large number of transactions performed by a bank. Also, deposit insurance schemes will further limit the incentive depositors have for examining a bank’s activities19. Insurance schemes are however necessary due to the fact that the incentives guiding shareholders’ actions aren’t exactly in line with their own. Bank shareholders’ incentives to monitor bank activities will be less limited, as they have a legitimate expectation of the bank paying out dividends. Their incentives suffer however from the free – rider problem. They stand a chance to move the risk of making losses onto the shoulders of the depositors, while personally being responsible to for any operational losses with only the capital they chose to invest in the form of 19

S. Heffernann “Modern…”, p. 6 – 8. However, most jurisdictions recognise the need for some sort of deposit insurance. The Unites States was first to develop such legislation in 1933. Since then, in 1991 the US have developed the Federal Deposit Insurance Corporation through the Federal Deposit Insurance Corporation and Institutions Act 1991. In the UK the deposit insurance scheme has been instated by the Financial services and markets Act 2000, under part XV of the act.

share20s. Furthermore, shareholders mostly choose to vote with their feet and focus on the short term percentage increase in net share worth rather than actual dividend. The more shares are held by small individual investors the less effective the supervision and shareholder influence over management. Furthermore, even the powerful investor will tend to favor to put an emphasis on best possible performance by his agent, pushing towards more outcome based contract formulation so that the agent’s remuneration will be mostly aligned with what profits he effectively stands to make. The shareholder will have a better informational relationship with his agent than the depositor, also implementing better reporting measures will increase agency costs to the shareholders, meaning they would stand to make a smaller profit than if no such measures were implemented into the banks internal governance systems. The depositor will only stand to make only a certain predetermined and usually low profit on the money he invested, he stands a much bigger informational disproportion than the shareholder. To this group, introducing more behavior - based measures into the agency contract would be desirable21. In a company that is not a bank this problem of two principal groups, having different incentives would not occur. In a economic system where competition among banks is high would, in my opinion, only increase the shareholder principal group’s incentives to further limit their scrutiny of their agents focusing on limiting agency costs and improving the final profit outcome of the management’s activities. This is a point well – observable when the recent bank crisis in the United States and the UK is observed. The response in the U.S. was to impose a 90% income tax burden on bank manager’s income paid out in bonuses. Although this seems like a measure that could greatly limit the application of result – based agency contracts, I find it to be a comforting half – measure, imposed solely to console those who find bank managers morally responsible for the recent financial crisis22. This is because, performance based manager contracts are purely a form of aligning their interests with those of the shareholders.

20

K. Alexander “Corporate Governance and banks: The role of regulation in reducing the principal – agent problem”, Journal of Banking Regulation, vol. 7 no. 1/2, 2006, p. 17 - 40 21 K. Eisenhardt, “Agency…” p. 57 - 74 22 See, BBC NEWS, available at: http://news.bbc.co.uk/2/hi/business/7953869.stm My opinion is based on the fact that the legislature is clearly based on post facto conditions, being the fact that the tax will be imposed on those managers whose institutions have received public help and the managers already receive more than 250, 000USD remuneration already. Rather than being an attempt at changing parties incentives, it seems like the recently ed bill is more of a legislatively designed financial punishment to those managers.

Acting as agents in banks, managers invest both considerable amounts of time capital and reputation, to produce results for shareholders. Although there may exist problems in the relationship between shareholders and managers, arising from informational asymmetry and interest issues such as enjoying job – perks or empire building by managers, there is a basic foothold aligning the interests of both the parties. This is because in essence managers stand to lose not only the considerable amounts of time invested in the success of the operations of the bank, but also their reputation and any future incomes that would come their way if they didn’t take excessive risks. So there is a reasonable ground to believe that agents in essence may be more risk averse than their principals in this framework and shareholders may find it necessary to bring forward incentives for additional risk taking23.

5.0. Governments and the role of regulators As I mentioned before, there is a legitimate and empirically proven need for government involvement in minimizing the impact of agency problems in some markets that would otherwise be inefficient and create social costs. In the case of banks this need for involvement through regulation is evident on due to the role that a crisis in one bank may spread to other institutions through contagion and onto the rest of the economy. Bank runs may cause a dramatic loss of confidence in the market by investors and consumers, causing them to limit their spending, changing habits etc. leading to a slow down in growth or even an economic recession24. It may therefore be said that to some extent, the weight of responsibility for the well being of the whole economy is placed on banks and their management sharing a balanced principal – agent relationship allowing regulators to spot red flags before an meltdown. In the case of privately owned banks, it remains fairly clear that however the role banks play may be important to the system and therefore to the general public’s well being and economic success, the general public’s interest in the performance of banks is identifiable solely as those between institutions and stakeholders25. However, there are undisputable grounds to assume that there may be informational disparity leading to moral hazard between regulators and bank owners. 23

J. Macey, M. O’Hara, “The Corporate Governance of Banks”, FRBNY Economic Policy Review, April 2003, p. 91 - 107 24 S. Heffernan “Modern…”, p. 351- 358 25 R. Levine, “Bank Regulation and Supervision”, paper for the National Bureau of Economic Research,2005, available at: http://www.nber.org/reporter/fall05/levine.html

While the regulator will assume a position to instate best practice procedures and governance rules to limit the possible social costs of risks in bank activity, bank owners will be prone to deg their principal – agent contracts in ways prompting bank managers to take excessive risks. This issue is even more underlined by the existence of bail – out program procedures and lender of last resort schemes, which effectively move the most of the risk onto the regulator and effectively the general public. Systemically important banks will be most prone to this problem26.

5.1. Deposit Insurance and Lender of Last Resort Having generally outlined the framework of principal – agent relationships that function in the system of bank corporate governance, I find it necessary to point out to a some issues. Steps undertaken by governments such as lender of last resort programs, deposit insurance schemes, seem to perform the function of protecting the party most at risk from suffering from principal – agent problems. However, these steps may be criticized as alleviating from suffering any of the consequences of undertaking excessive risk altogether. Although capital adequacy schemes have multiple requirements, for example the requirement for the bank’s senior management to have adequate knowledge of a banks procedures, or the Pillar One requirement for banks senior management to oversee and approving the risk assessment processes, they seem to be purely check – boxes for regulators to tick instead of having any actual binding power27. There seems to be no empirical measure of what constitutes adequate information of a bank’s operations on behalf of senior management. There is also no test of what constitutes adequate involvement and understanding of a bank’s capital rating system. Based on the above, I do not find it out of line to state that where bank regulation and governance fail, is where it is made to rely too firmly to rely on its self – fulfilling power.

26

K. Alexander, “Corporate…”, p. 21 Basel Committee on Banking Supervision, Pillar One, part 3, H, par. 5, sub. Par. 438 .Quoted in K. Alexander, “Corporate …”, supra note 47 27

6.0. Developing Regulation and Reform If I were to suggest a way of setting up a principal – agent framework for banks I would focus reform efforts on three issues. These issues are shareholder monitoring and influence improvement, depositor disclosure improvement and greater regulator involvement in the daily operations of banks. To minimize the chance of moral hazard occuring along the main principal – agent relationship between shareholder and management, a greater degree of shareholder activism is necessary. Institutional shareholders should be given greater powers to monitor the actions of the management. Shareholders owning more than a specified percentage of shares should be allowed to appoint one director to the board. A manager appointed this way will be able to know to whom he owes his seat of power, at the same time being limited by the fiduciary duty of owing loyalty to the company28. This solution would limit a manager’s incentive for empire building and the proverbial “easy life”, also aligning each particular manager’s remuneration with his actual performance and giving the principal a greater incentive to scrutinize it. Some authors believe that in relation to banks there should a regime limiting the extent of limited liability of shareholders, so that to limit the extent to which they might urge their agents to take excessive risks29. Sufficient evidence exists that banks would perform better under a stricter shareholder liability regime, based on historical data, but the idea itself seems too outrageous to be accepted into modern into banking regulation. Once shareholders’ interests have been given greater attention, the need arises to counter possible problems that may be caused by the shareholder’s abusing their comfortable situation and transferring risks towards the depositors. A government could increase the trust in the system as a whole if greater measures to increase disclosure towards depositors were to be introduced. Thaler and Sustein have developed a theory dubbed “libertarian paternalism”. This idea, basing on empirical 28

L. Zingales, „The Future of Securities Regulation”, Center for Economic Policy Research Discussion Papare Series, available at: www.cepr.org/pubs/dps/DP7110.asp 29 L. Evans N. Quigley “Shareholder Liability Regimes, Principal Agent Relationships and Banking Industry Performance”,Journal of Law and Economics no.38,1995, p. 498 - 520

evidence, suggests that some rules ought to be designed so that the default option chosen by most people will maximize social welfare30. An example of such a rule is that whenever someone would choose to invest money in a bank through opening a money management or making a deposit, they should receive a pre – established set of disclosures about the investment they are making, containing a set of financial data, historical information, a complete set of information regarding a bank’s operations in relation to money paid in by any depositor. Finally, in order to ensure that regulators and regulation is even able to have an influence on the banking sector I believe that a separation of retail banking from investment banking from the investment banking is a necessity. Furthermore, I believe that banking institutions should be separated from other financial institutions and their activities should be strictly numerated, so that financial innovation could take place in the system through financial institutions, but separately from banks, effectively cutting these innovations from the scale effect that banks produce in their activities. A Glass – Steagall Act type legislation would be a significant step towards ensuring soundness within the system of bank corporate governance31. The basic critique of such an arrangement is that this would mean that this would limit the banking sector’s efficiency and profitability. Although this might be true, such an arrangement would be systemically desirable, as part of a possible macro – regulation framework that would perhaps limit some profits during “high - times” in the banking sector, but would also prevent bubbles form occurring or regulators losing touch with innovation.

7.0. Conclusion In the area of corporate governance, banks seem to stand out as a subject of research. Organizationally more complex and more important to the system as they are, they are also more vulnerable to the imperfections that arise in relations between humans. This particularity of the principal – agent framework existent in banks leads to a necessity to ask some specific questions. First, what can be done to banks specifically from a 30

T. Thaler and C. Sustein, „Libertarian Paternalism”, The American Economic Review, Vol. 93:2, p. 175-179 31 Some authors suggest the return of similar legislation in different forms. See : L. Zingales ”The Future…”p. 3

regulatory perspective to upgrade the efficiency of their operations? Secondly, do these improvements require a new banking law with a set of hard legislative rules, or does secondary regulation released by legislative empowered regulators suffice? Having answered the first question in the preceding paragraphs I wish to focus on the second issue. The extent to which banks are different from typical companies and their importance to national and recently also to the global financial systems suggests that banks deserve to be managed by specific legislation, that would address them specifically. Most legal systems in the world already have such legislation, addressing banks as a particular type of company. These legislations should evolve to address more of the specific issues relating to corporate governance and existing principal – agent problems experienced by banks.