Bank Shareholders
Stakeholders:
Society Depositors
Creditors
Finally, one bank’s reckless risk-taking may lead to international uncertainty about Danish banks in general, making international interbank funding unavailable to the Danish banking sector as a whole, causing a nation-wide funding liquidity crisis.
In this thesis, it is the standpoint that these externalities are not fully accounted for in the corporate governance climate surrounding the banks in the researched period and therefore, it is the view that the bank should compensate for these possible externalities to increase economic efficiency/avoid the impacts outlined in section 1.3.
As bank boards’ composition and structure have been found to be an endogenous factor in determining the risk-profile of the bank in the analysis of this paper, it can be derived that the board’s inner workings can be altered by changing either the components of the board or by changing the incentive structure surrounding it. The former would entail dictating which people the shareholders would be allowed to elect to the board45, which is not deemed a realistic scenario46. Therefore, three incentive-adjusting models are proposed, which all rest upon the theory, the empirical studies and this thesis’ study in their formation47. The purpose of the models is as said to align the externalities of running a bank.
11.1.1 The Actuarial Model
As the bank failures in the latter years of the research period have had visible economic impacts for non-shareholding stakeholders, whom the banks have most probably not compensated these non-shareholding stakeholders for, the insurance view would be that banks have not paid a (high enough) risk premium for the deposit insurance48 and that they should do so. Knowing that the composition and structure of the board matters, the insurer (in effect, the state) could price these two factors into the actuarial evaluation of which risk premium the bank should pay. The underlying idea would then be that banks have insured themselves and that society would not stand to lose should the bank fail.
However, externalities are inherently hard to price, also in this case. Even though measurable differences exist in the risk-taking behavior following the structure and composition of the
45 Not to confuse with dictating which specific people that could sit on the board.
46 Shareholders would have very little incentive to buy the bank’s shares if the board was appointed by e.g. a government agency, because they would lose all and any impact on the bank’s management system.
47 Also, this means that theory already explained would be redundant to re-iterate here.
48 Although the bank does not receive the insurance in the case of default (depositors do), the insurance is valuable to the bank, as it first of all severely limits the possibility of a bank run, leaving the bank in great funding troubles, and second of all it lets the bank take greater risks, because debtholders cease to discipline the bank.
board, the calculation of the possible externalities with one type of board (and governance system in general) versus the other would be extremely hard to price accurately, partly because they are dependent on international business-cycle effects, too (for example it is hard to price the cost of not being able to get funding in the interbank market at a given interest rate). Also, since contagion and uncertainty can be caused by any one bank in the pool of banks, it would be troublesome to assign the cost of the premium fairly to each bank.
Therefore, two other models, which both take the view that bank failures should be avoided altogether are suggested below:
11.1.2 The soft law model – with hard law extensions
As determined in the thesis, no specific soft law for the banking industry exist; instead, banks adhere (to some extent) to the overall corporate governance recommendations. Following the findings of this thesis, a specific set of corporate governance codes could be drafted to include the knowledge put forth here. These codes should include the whole array of stakeholders because of the banks’ special role in society and should give specific recommendations on the knowledge, experience and busyness level of the individual directors to ensure lower levels of risk-taking, as well as recommendations on minimizing incentive programs, as these seem without a doubt to increase risk-taking. Furthermore, a specific set of bank corporate governance codes could recommend that the CEO be elected by the general assembly at specified time intervals, as the CEO’s relative tenure matters in risk-taking and control in the bank, according to this thesis.
The control function of this model would be the managerial labor market and the general populace – but also the DFSA, who would be equipped to replace boards not complying with a certain percentage of the new recommendations49.
11.1.3 The society model
The final model could be a society model in which all banks must have a government-appointed director, who reports to the DFSA, which in this model would be equipped with extra power, too. This would secure representation of the non-shareholding stakeholders and thus mitigate the type-3 agency problem. It is a well-known solution, too, as banks previously had a government-appointed director on the board. This practice could be revived.
49 The DFSA has already had its powers increased, but it is proposed that they should be even stronger in this model.
The relation to the findings of the thesis are that having a responsible government representative on the board would leave the bank with more freedom to choose its board and its risk-level, as long as accordance with the appointed director’s judgment on whether the risk-level was appropriate at any given time. This appointed member should be equipped to order the CEO of the bank to reduce the risk-taking (for example, the loans) at once and before the externalities described above manifest themselves as liabilities to the society.
11.1.4 Final recommendation
In summary, the thesis recommends a mix of all three models. It is the viewpoint that the bank and its supreme governing body, the board, should pay a risk premium for having deposit insurance and that this premium should reflect the risk-level of the bank. It is also recommended that financial institutions, which exist for more reasons than to create shareholder wealth, be given specific corporate governance codes, based in actual research.
Finally, the re-introduction of a government-appointed director is recommended by this paper, to ensure representation of all groups that could be economically affected by the bank’s actions.
11.5 Partial conclusion
The final part of the thesis was found to serve as an extension, providing ground for further research. The recommendations of the thesis were found to be that the mismatch between those who benefit from the bank’s actions and those who stand liable, should the bank fail, should be solved by applying a combination of three models: the actuarial model, which prescribes charging banks with a fair risk premium for the insurance they receive, the soft law-model, which prescribes altering the corporate governance recommendations in specific bank version and finally, the society model in which the non-shareholding stakeholders are represented on the board of the banks.
12. CONCLUSIONS
The thesis was motivated by the criticism directed at the Danish banks and their boards in particular in the wake of the financial crisis. This led to the research question:
“Have the structure and composition of Danish bank boards affected risk-taking in the period from 2003-2008?”
It was concluded that some aspects of the structure and composition have affected the risk-taking to a significant degree.
The thesis investigated the research question deductively through the construction of ten hypotheses; six hypotheses regarding the composition of Danish bank boards, four regarding the structure. The hypotheses emerged from a thorough theory review of the corporate governance field with a section dedicated to agency theory, as well as from detailed literature reviews on the theory and empirical studies within the fields. The sample was constructed from the 67 largest Danish banks, comprising more than 95 % of the total working capital in the Danish banking industry.
The dependent variable best suited to reflect the research question’s “risk-taking” was determined as the loan/deposit ratio of the Danish banks in 2007.
The answer to the research question was found through the subsequent data analysis, building on 4.829 manually collected, unique data points. The thesis found that the structure and composition of Danish bank boards in the period 2003-2008 have indeed affected risk-taking in Danish banks. The basis for this result can be summarized as follows:
• Independent directors were found to allow more risk-taking than dependent directors*.
• Experienced board members were found to allow less risk-taking than inexperienced ones (less than five years of board experience)
• Directors with more than three simultaneous directorships were found to allow more risk-taking than directors with three or less directorships
• Directors were found to allow increasingly more risk as their years on the board increases
• Directors with a master’s degree in finance, economics, business economics or accounting were found to allow more risk-taking than those without such an educational background
• The existence of incentive programs (bonuses and stock (option) plans) was found to be positively related to risk-taking
• It was found that if the CEO had been in place longer than the board on average, risk-taking had been increased relative to a situation in which the board had served longer than the CEO*.
*: Only in the models without employee-elected representatives
The parameters ”gender” and ”board size” were both found to be random in relation to the risk-taking of the bank.
It was discussed that independent board members do not seem to be impaired by higher monitoring costs as they do increase risk-taking, but also that the sample was flawed by its homogeneity. It was also discussed whether experienced board members might be better at recognizing a non-linear relation in risk-taking and therefore, could be risk-reducing as well.
On the other hand, the experienced members might be diversifying risk unnecessarily.
Furthermore, the thesis suggested that women, who make up under ten percent of outside directors, might not have a say at board meetings and therefore do not influence risk-taking.
On the other hand, female directors might just have risk preferences equal to those of males.
Directors holding more than three simultaneous board seats were argued to be conducive to risk-taking because of their attention to the managerial labor market and thus, to shareholders.
However, it also has to be considered that directors with many simultaneous board positions might just be rubberstamping a risk-seeking CEO’s proposals. A board director with longer time on the board was argued to be good at understanding the shareholders’ needs and incentives. However, that must be weighed against the possibility that a board member with long tenure could be manager-friendly to a seeking CEO. Financial education as a risk-taking trait in a director was suggested to stem from these directors’ perceived better understanding of risk. Nevertheless, there was no evidence that financially educated board members prevented hazardous levels of risk.
Within the confines of the research at hand, board size was seen to be a random variable either because of smoothly functioning market forces or because board size is determined by other things than market economics, such as politics.
Incentive programs emerged as the most powerful explanatory variable in risk-taking in Danish banks, which was discussed from the perspective that the more the CEOs salary that stems from performance-variable pay, the less value he places on keeping the bank going as a concern. Finally, it was discussed that CEOs with longer tenure than the average board may recognize that their salary’s NPV from keeping the bank a going concern gradually becomes smaller and therefore, they might be motivated to take more risk for reasons of prestige, fame etc.