Outperformance Fees Charged by Mutual Funds
What Are the Effects of Charging Outperformance Fees on Fund Managers’
Behaviour Considering the Efficient Market Hypothesis and the Principal-Agent
Theory?
Vincent S. van Leeuwen 0207098
21-07-2013
Abstract. This thesis provides a survey of the existing research on the effects of charging an outperformance fee on fund managers’ behaviour. In all kinds of settings incentive pay causes behavioural change. I will show that charging outperformance fees can have both positive as well as negative effects on fund managers’ behaviour.
1
1. Introduction
Typically mutual funds and other asset managers are remunerated with a base fee, which is a percentage of assets under management. Many asset managers charge an extra fee if their return results beat an a priori chosen index or other kind of benchmark. In most cases they will receive a percentage of the difference. This practice is referred to as charging an ‘outperformance fee’.
One of the major pillars in finance theory is the Efficient Market Hypothesis (EMH). This theory states that returns on stocks are not predictable. The price of a particular stock comes about through a process of supply and demand in which all publicly available
information culminates into a given price. Only unexpected events or new developments can radically change the price of stocks and in doing so will cause an unexpected change in return. In this view it’s mostly luck if fund managers achieve a better return than investors achieved on average in a particular asset market. Persistently achieving a better result than the market is deemed virtually impossible or, in other words, statistically the chance for persistence of superior returns is very low.
As stated earlier, there is a group of asset managers that asks for an extra remuneration for what they call outperformance. They implicitly and sometimes explicitly state that they can achieve a better return for their clients and that they should be remunerated for achieving this excess return. It’s stated that outperformance remuneration aligns the interests of the asset managers and their clients. It seems that the practice of charging outperformance fees is in opposition with the EMH. Furthermore, there are ways in which asset managers on average can achieve a better return than a given benchmark without any skill or superior information. As will be shown in this thesis, the way in which an agent is being remunerated for his
services is bound to affect his behaviour. Maybe, as stated above, in the sense that it will align interests, but perhaps also in other ways which might even be detrimental to the wealth of investors.
The fact that more and more asset managers are remunerated in part on the basis of outperformance and the fact that most people hold securities via asset management companies gives an imputes to research the effects this practice will have on asset managers. Indeed the outperformance fees will be a significant transfer of assets from the investor to the fund
2
manager. In the European Union financial service providers are under increased scrutiny with the intention to protect consumers. One should expect that outperformance fees are going to receive more attention from regulators in the near future.
In this thesis the practice of charging outperformance fees will be put into perspective using the Efficient Market Hypothesis (EMH) and the Principal-Agent theory. The main question is: What are the effects of charging an outperformance fee on the behaviour of fund managers and on the characteristics of the fund? My hypothesis is that charging outperformance fees has a significant influence, both positive as well as negative. I suppose that the negative effect will outweigh the positive in the way that fund managers are more likely to show behaviour that can be classified as moral hazard; they will prioritize their own gain and add more risk to the fund.
The method used for the research that is presented here is a literature survey. This method is chosen to create a summary of existing research on this subject. Scientific papers on the subject of outperformance fees are scarce and they provide information on different kinds of phenomena. My objective is to give a good overview of the state of knowledge to be used as a starting point for future research.
The thesis is presented in the following order: first, I will provide a simple theoretic framework in which I will review the EMH, Arbitrage Pricing Theory and the Principal-Agent theory. Then I will present a more elaborate hypothesis based on these theories. What follows is a review of existing research concerning the effects of outperformance fees on the behaviour of fund managers. The thesis ends with a summarizing conclusion.
3
2. Theoretical Framework: A Review of the Efficient Market Hypothesis,
Arbitrage Pricing Theory, Principal-Agent Theory
In this section I will provide a simple theoretical framework by briefly discussing the Efficient Market Hypothesis (EMH), Arbitrage Pricing Theory (APT) and Principal-Agent Theory. I will finish by elaborating my hypothesis on the effects of charging outperformance fees on the behaviour of fund managers taking these theories into consideration. This will be used to discuss the existing literature on the application of outperformance fees in the next section. As we will see, the research that I’ve found and will be discussed in the next section follow the same rationale.
The EMH (Fama, 1970) in its semi-strong form, states that security prices reflect all current public information that is known by the investing community. Testing semi-strong market efficiency can be done by looking at the performance of fund managers. If they can’t earn excess returns persistently then indeed stock markets seem to be semi-strong efficient.
Michael C. Jensen (1968) performed such a test. He used the Capital Asset Pricing Model to examine if the performance of mutual funds in the period 1945-1964 was in accordance with the EMH. He found that on average there was statistically no proof that funds were able to beat an indexing strategy with persistence. Even worse, he found that taking only brokerage expenses into account, there still were no funds that significantly outperformed the market (1968, p. 415).
Although Jensen and others after him showed that persistent outperformance in mutual funds is rarely observed, the point has been made that the EMH does allow for minor
abnormal returns (Dimson and Mussavian, 1998, p. 96). The theory states that all public information is reflected in prices, but the investment community can expense effort to uncover information that will alter equilibrium prices of stocks. According to Grossman and Stiglitz (1980, p. 393), when information gathering is costly, a model of market equilibrium should take into account an equilibrium level of disequilibrium. They state that this also means that some amount of arbitrage is still possible, although they warn that competition between investment professionals will insure that reaping rewards from collecting information will be difficult. Berk and Green (2002, p. 3) also state that it’s incorrect to assume away
4
managerial skill and that eighty percent of investment managers are able to at least earn back their fees. Still, they also state that since the research of Jensen there is little evidence of active managers outperforming passive strategies. After performing a detailed analysis of returns and costs, Wermers (2000, p. 1690) also concludes that fund managers have skill, but he states that it’s not enough to earn back their fees.
Malkiel (2003) gives a good summary of the critics of the EMH. He for instance lists the January effect or behavioral biases that lead to over- or under reaction. He also lists research results which actually show that trading strategies on these phenomena are not profitable. He points out that margins are small so transaction costs can also limit traders to take advantage of the anomalies. Indeed, Malkiel (2005) reaffirms the fact that on average fund managers are not beating passive strategies.
So far the mentioned results give no indication of the possibility for fund managers to beat their benchmark with skill or superior information and they also show that the
possibilities for arbitrage are slight at best and that in a lot of cases the extra effort to raise return will lead to higher costs. This in turn will lower return to the point where it is roughly the same as a passive investment strategy. In this light charging outperformance fees seems at odds with reality.
Indeed some scholars suggest that charging an outperformance fee could lead to agents entering the market without skill or superior knowledge and game the fee schedule by adding more risk to the portfolio as opposed to the benchmark and then on average report a higher return (Elton et al., 2003, p. 781).
If we believe that the relation between risk and return indicated by the Capital Asset Pricing Model to hold (Sharpe, 1964), we should expect at least some fund managers that charge an outperformance fee to increase beta. On average this will cause the portfolio return to be higher than the benchmark. The manager will receive extra remuneration for something a smart investor can create himself by for instance leveraging an investment in an index fund. Furthermore, if we believe the Arbitrage Pricing Theory model as derived by Fama and French (1996) to accurately predict the drivers of return we should expect the outperformance fee to cause higher holdings of for instance small cap shares compared to their weight in the chosen benchmark.
5
I will continue discussing the Principal-Agent Theory. Jensen and Meckling (1967, p. 308) define a Principal-Agent relationship as a contract in which one party called the principal hires an agent to perform services on his behalf. They argue that since both parties are mostly interested in their own welfare, that the agent will not always perform to his best abilities or may even conduct activities which are detrimental to the welfare of the principal for his own gain.
The above mentioned behaviour is called moral hazard (Holstrom, 1979, p. 74). The risk of moral hazard is primarily caused by an information asymmetry between the principal and the agent. The principal is not always capable of observing how the agent carries out the contract. Maybe the principal isn’t even knowledgeable enough to appraise the activities performed on his behalf, even if he would be able to observe the actions of the agent perfectly.
Another type of problem which is described in the Principal-Agent framework is the risk of adverse selection. To my knowledge only Livio Stracca (2006) has researched this specific problem in the context of the effects of outperformance fees. Stracca (2006, p. 834) states that changing the outperformance fee from a fulcrum fee (where the agent loses money when underperforming) to an asymmetric fee (where the agent doesn’t lose money when underperforming) will lure uninformed agents into the industry.
Of course in the case of mutual funds and other forms of delegated asset management there are Principal-Agent types of contracts between the investor and the firm or party that carries out the actual investing of the money. As in any setting the objective of scholars and practitioners should be to define an optimal contract as to perfectly align the interests of the principal and the agent. Bhattacharya and Fleiderer (1985) for instance state that a quadratic contract in which the agent reveals information about the expected return on a financial asset and in which the agent is penalized for the ex post difference is optimal in the sense of
adverse selection, but also mitigates the hidden action problem. Other scholars focus more on the choice of benchmark and also introduce tracking error into these deliberations.
There are much more scholars that suggest and prove that there are all kinds of possibilities to improve contracts as to limit moral hazard in a delegated portfolio
management setting. In this thesis I focus on the effects of two types of contracts that are still the most used; a symmetric and an asymmetric contract. In the United States portfolio
6
charge a symmetric ‘fulcrum’ fee if they choose to charge an outperformance fee. In many other countries like the Netherlands, Italy etc. regulation is less tight and asymmetric fees are allowed. In this thesis outperformance fee means one or both of these types of contracts, unless when I am quoting research that differentiates between types of contract.
From the theories discussed here, we can deduce that there is some proof of skill in fund management. However, because of the higher costs of active management the gains are virtually non existent. In fact, this corresponds with the EMH. It indicates that some managers are better in finding and interpreting information about stocks, but that competition between the many fund managers ensures that prices of stock move up to their intrinsic value fast. Consistent outperformance with a high margin is not likely. Fund managers know this and should be aware that charging an outperformance fee will not reward them much. The prevalence of outperformance fees especially in countries were they are not limited, must indicate that fund managers offer their services for this type of reward for other reasons.
My hypothesis is that in light of the Principal-Agent theory one should expect that agents show to some extent behaviour that can be classified as moral hazard, because perfect monitoring and contracting is not possible. In the delegated portfolio setting where the agent charges an outperformance fee one should expect the agent to have incentives to game the contract by adding more risk to the portfolio or to change the portfolio to improve the contract payoff in ways that are not value-enhancing for the principal with which I mean not buying or selling assets on the basis of fundamental analysis. If one believes the APT to hold, then one should expect a fund manager to tilt the portfolio weights to certain risk factors to create a difference between the portfolio and benchmark which will on average generate an excess return.
For the same reasons one could expect that when an agent is not remunerated extra on the basis of performance, he will seriously lack proper incentives to expend effort to raise his performance. As discussed, there is some room for skilled fund managers to improve
performance and we should expect this only to be possible after expending considerable effort. Effort expenditure is a disutility for the agent. Granting the fund manager extra remuneration on the basis of performance can possibly align the interest of principal and agent in this respect.
7
In the next paragraph I will discuss the research and outcomes that are reported by scholars that indeed proof that on average outperformance fees change the behaviour of fund managers and that indeed risk characteristics are significantly different between funds that charge an outperformance fee and ones that don’t.
8
3. The Effects of Outperformance Fees on Fund Managers’ Behaviour and on
Fund Characteristics
In this section I will review the research of various scholars on the effects that charging an outperformance fee has on fund managers’ behaviour and on fund characteristics. I will start with the positive effects and continue with the negative effects. Whether an effect is positive or negative will be considered from the principal’s point of view.
Charging an outperformance fee should align the interest of the agent with that of the principal. In the delegated portfolio setting this would mean that fund managers work hard to achieve a better return than the benchmark because of superior stock selection and not because of taking riskier positions. After controlling for differential risk we should expect an excess return if indeed outperformance fees align the interests of principal and agent.
Elton et al. (2003, p. 789) use an APT type pricing model to regress fund results of funds that do and funds that don’t charge an outperformance fee. With their APT model they control for the influence of small stock, international stock, bonds, etc. They find a difference of 0.084% per month return between these two fund groups. Even after they control for difference in expenses, the difference between return is significant at the 5% level implying that indeed fund managers have stock selection skills and that indeed the outperformance fee raised the return implying that an outperformance fee aligns the interest of the principal and the agent. The only drawback in this analysis is that on the basis of these data it is impossible to distinguish between adverse selection and moral hazard effects. One could state that even if the selection effect is present the outperformance fee still incentivized the agents to perform and achieve a higher return.
Edwards and Caglayan (2001) show that skill can be proven for hedge fund managers. By using a multifactor APT, they control for risk factors that could ameliorate results with respect to the benchmark without proof of skill. They actually find an excess return of 3% to 6% per annum when managers receive a 20% or more outperformance fee as compared to managers that received a lower outperformance fee. Furthermore they prove that the results of the top performers show consistency as do the low performers as extra proof of the existence of skill.
9
As stated, Elton et al. (2003) show that the results of outperformance fee charging mutual funds are statistically better than the funds that do not charge an outperformance fee. Half of the difference is due to an overall lower fee, but after this fact there is still significant
outperformance. They do show as well that the average outperformance fee is slightly less than zero and that after costs on average a passive strategy would amount to almost the same return. This in turn implies that non outperformance fee funds (or base fee fund) were actually underperforming. Kahn et al. (2006, p.62) explain that a hedge fund manager has less
constraints and can for this reason accomplish more with his skill. This would indeed account for the dramatic difference in excess return between the two studies.
It should be taken into account that the positive effect of slightly better performance when applying outperformance fees that Elton et al. (2003) show, is based on data on funds in the United States. We should be careful applying their results to other countries. The stock markets in the United States are presumed to be more efficient than stock markets in third world countries and developing countries, because of laws that promote more information disclosure and sophisticated regulators. Bialkowski and Otten (2011) show that Polish fund managers specializing in Polish stocks also outperform the proper benchmark before costs, but not after. They actually claim that this is proof that they overcharge for their creation of alpha. So if it’s true that in some countries stock markets are less efficient one should expect that the practice of charging an outperformance fee would lead to sharper differences than in the Elton et al. sample. To my knowledge, no research on this subject has been done.
Another positive effect of outperformance fees is that it possibly shifts the focus of the agent from adding more assets to the total assets under management to achieving a higher return on the assets under management. Indeed the only way a fund manager could improve total profits if he or the fund only receives base pay is to enlarge total assets under
management.
Focusing on enlarging assets can hurt performance. First, because opportunities to create excess return are scarce. As discussed in the section on EMH, there is much proof that markets are efficient enough not to offer a lot of arbitrage or mispricing opportunities. Skilled fund managers are in this view aware of some opportunities to create excess return, but after these have been had, they will have to invest in lower return investments and so naturally erode excess return on average. Chen et al. (2002) proof that this effect is significant. Another
10
reason could be more direct in the sense that the time spent on getting more funds cannot be used to acquire information on investment opportunities.
Now I turn to the possible negative effects of charging an outperformance fee. The research on EMH that has been done shows that consistently and with a high margin beating the
overall market will be difficult. To earn more, a fund manager could always raise the riskiness of the portfolio, for instance along the lines of the APT model. In this section I will review research that indicate not only overall riskiness is increased by charging outperformance fees, but also that fund managers change risk characteristics midyear to reap higher rewards. Although not scientifically proven also rogue trading and window dressing can be concerns. First, it is important to mention that even without outperformance fees there is an incentive to increase the riskiness of a portfolio in order to achieve higher returns (Brown et al., 1996).
Investors and people deciding whether to invest can easily find information about mutual fund performance; newspapers publish return information and the internet brings a wealth of information about fund characteristics. Of course the objective of these investors is finding mutual funds that will maximize risk-adjusted returns. Indeed, Capon et al. (1996, p. 68) show that return characteristics are the most important in the choice of mutual fund for households. This investment decision process has led several authors to believe that viewing the fund managers’ behaviour in terms of a tournament to reach the top of the return charts will lead to a better understanding of their investment decisions (Brown et al., 1996, p. 85). They show proof that fund managers indeed alter the riskiness of their portfolio to the end of increasing the amount of assets under their management. Furthermore, it has been stated that this incentive is aggravated by the fact that fund inflows and outflows are not symmetrical. The winners in the tournament experience relatively more inflows than the losers experience outflows of assets (Sirri and Tufano, 1998, p. 1595).
To further research the effect an outperformance fee will have on overall year-to-year riskiness, it is important to keep in mind that there needs to be a clear difference between risk characteristics between funds that do and funds that don’t charge an outperformance fee. Elton et al. (2003, p. 802) indeed show that funds that charge outperformance fees have a higher beta than funds that do not. They state that the most common strategy of funds with performance fees in their sample was an overexposure to small cap stocks compared to the benchmark they use. Surprisingly, the average beta was still lower than one.
11
Golec and Starks (2004) analyse the effect of the change in regulations in the US in 1971. Congress and the Securities and Exchange Commission feared that asymmetric fee schedules would entice fund managers to gamble and get rewarded for gambling so they decided to implement law stating that firms that charge an outperformance fee are obliged to make the fee symmetric. This should mean ceteris paribus that, on average, firms that charged
asymmetric fees before should show a reduction in overall risk and indeed Golec and Starks show this to be true. They state that the larger the penalizing effect of the fulcrum fee the lower the average risk. Still, risk is higher in the fulcrum fee funds than in base fee funds. Already Kritzman (1987, p. 23) pointed out that fulcrum fees are still asymmetric when the fund management is organized in a firm with limited liability. Indeed, after a very bad beat the firm can declare bankruptcy so downside risk is mitigated for the agent.
The penalizing effect is also proven in a research among hedge funds. Kouwenberg and Ziemba (2005) show that hedge funds where the manager has a substantial stake in the fund (30%) are considerably less risky than hedge funds where the manager has no or a much smaller stake in the ownership. This relationship is very much similar to the Golec and Starks research because the fund manager is penalized when negative returns occur because of the size of his stake in the results. Furthermore Kouwenberg and Ziemba (2005) show a positive risk and incentive fee relationship in fund of funds.
Indeed we should expect fund managers to take higher risks, because on average we would expect positive rates of return on the benchmark, so on average raising beta or other risk factors above the benchmark should earn a positive outperformance fee. The payoffs are lower ceteris paribus when forced to make the fee schedule symmetric, because payoffs from raising risk are lower and because the agent is risk averse.
The second negative effect of charging outperformance fees on the behaviour of fund managers is that they might change the risk characteristics of a fund midyear. Kritzman (1987, p. 21) analyses the outperformance fee as giving the fund manager a call option where the value of the option is dependent on the spread between the standard deviations of the fund and the benchmark and the level of correlation between the two. His and other analysis in this way is interesting, because in a sense these options can be exercised. Indeed Kritzman (1987, p. 23) points to this possibility which is characterized by many researchers as ‘lock-in’.
Orphanides (1996, p. 3) states that the most common performance evaluation period is one year. In his analysis he found that early in his sample from 1976 to 1993 there is no
12
evidence of fund managers altering the risk midyear, but in later years he did find enough evidence of this behaviour. In his conclusions he writes that this is probably the result of increasing proliferation of outperformance fee schedules. He doesn’t test whether there is a significant difference between firms with or without a performance fee schedule. This is important because Chevalier and Ellison (1997) proved that the performance flow relationship was enough incentive to change risk taking behaviour within performance evaluation periods.
Elton et al. (2003, p. 798) do test if there is a difference between funds that do or don’t charge an outperformance fee. They examine funds from 1990 till 2000 that use a three years evaluation period and look at the changes in variance in the last year. They find very strong evidence that indeed top performing funds have lower variance than the bottom funds. Also significant at the 10% level they show that variance differs from no outperformance fee funds.
Whereas higher than benchmark performing funds can be shown to lower their risk to lock-in gains, underperforming funds have the incentive to raise risk along the same lines of reasoning. Experiments carried out by Brown et al. (2005, p. 1) have shown that people are willing to sell out on gains, but increase their position on a loss. In the worst case this could mean that a fund which is underperforming takes on so much risk to reverse that a large part or all of the investment in the fund is lost. A prime example is the Barings Bank fiasco where the trader Nick Leeson brought the entire bank down.
Orphanides (1996, p. 8) states that monitoring of fund managers is biased towards excessive risk taking. Further, he states that early in his sample underperforming funds would take on more risk, but later in his sample this tendency disappeared. He figures that this could be due to better monitoring. Of course mutual funds will have monitoring systems to keep a fund manager in check. However, a financial adviser who caters to private investors directly and manages their wealth may not be scrutinized as much. Therefore, in that situation this form of excessive risk taking can cause a lot of damage.
The last form of perverse incentive that will be discussed is ‘window dressing’. Window dressing is to dump losing stocks on the last possibility of trading or to buy extra stock of stock that is already owned to inflate the price of this stock so the total holdings of the stock are worth more. These trades take place on the last possible moment of trading before a moment of disclosure or a cut-off moment for performance measurement. Because the trades are at the last possible moment, the price goes up and can’t be brought down again by another party trading in that specific stock. Stocks that are most suited for window
13
dressing are small cap stocks, because they are more illiquid and because supply is smaller so the price of the stocks rises more rapidly. Window dressing is costly and the gains are quickly reversed the next day.
Carhart et al. (2002, p. 662) proves that open-end equity mutual funds exhibit abnormally large returns on quarter ends and this effect is largest at the end of the fourth quarter. They test this with data containing returns of 2829 funds. Of these funds 80% report a better return than the S&P 500 on the last trading day in the fourth quarter. So there is no doubt that window dressing actually occurs. However, it must be seen as a form of market manipulation and therefore illegal in the United States.
After testing for the occurrence of window dressing Carhart et al. (2002) continue testing two distinct models. The first model suggests that fund managers practice window dressing to beat the benchmark and rake in a higher fee. They test this model by examining if the distribution of fund returns around the benchmark return leans to a slightly higher return. They conclude that there is no significant proof for window dressing behaviour stemming from this model’s incentive.
The second model they test is the occurrence of window dressing to gain more capital to invest. They test this model by looking at the standing of a fund a day before the last day, figuring that top performing funds have the most to gain by window dressing. Indeed, it is clear that top 20 percentile funds have considerably higher returns on the last day compared to the S&P 500.
Carhart et al. (p. 690) further state that there is other research which concludes that outperformance fees have a positive relationship with window dressing. However, they don’t mention a specific research and, as stated, their own research indicates that there is no
evidence for this type of incentive. Interestingly, anecdotal evidence does as well point to the occurrence of window dressing and people explicitly state that this practice is done to
ameliorate stock rises and to reap higher fees as well as gain more in investments.1 I suggest that Carhart et al.’s specification of the testable hypothesis is flawed.
I propose this for two reasons. First, outperformance fee funds have more to gain by taking more risk during the year than the benchmark. Theoretically this means that their return
1 http://online.wsj.com/article/SB10001424052970203686204577116352886527644.html &
14
is significantly lower or higher than the benchmark but not on or close to the return of the benchmark. Actually most funds in that region of the return distribution are index trackers that have no incentives to window dress. Second, it doesn’t really matter if the funds have a return close to the benchmark or higher. In both cases they gain the same performance fee by
window dressing. Further complicating the analysis is the fact that it’s not known how much of the more than 2800 funds in their test use an outperformance fee schedule.
The Principal-Agent theory would suggest that the fund manager will window dress if he gains in outperformance fee reimbursement. Although it’s true that the gains reverse the next day which makes it more difficult the next year to beat the benchmark, it’s still true that the fund manager gets an extra fee now for a lower one a year later, so time value of money considerations would dictate that there is a gain. Furthermore there is always a chance next year the result will be worse than the benchmark anyway, meaning more than the negative return on the first day of the new evaluation period.
Although, to my knowledge, there is no other research on the subject of the relation between outperformance fees and window dressing, the anecdotal evidence suggests that outperformance fee incentives do cause window dressing. Parallel to the other forms of behaviour it is reasonable to assume that this is in fact the case.
As shown outperformance fee alter behaviour and change fund characteristics. The positive effects are an on average higher return and preventing excessive assets gathering. The proposed negative effects are also present. On average portfolio risk is higher and managers lock-in gains. Furthermore, there is some evidence of window dressing. Rogue trading could be an issue as well. None of the research indicates if the negative or positive effects outweigh each other.
15
4. Conclusion
Based on the existing research that has been examined in this thesis, it can be concluded that various significant effects of charging an outperformance fee on the behaviour of fund managers and on the fund characteristics occur. These are both positive and negative effects.
The EMH and the well-known tests of this theory must make fund managers and investing public weary of claims of superior performance and the claim that an
outperformance fee strictly aligns the interest of the investor with the interest of the fund manager. CAPM and APT models dictate that expected return is positively related to sources of risk. If believed to be true it is not hard to see that outperformance fees can be gamed in this sense in a way that requires little skill.
Proven negative effects of charging outperformance fees are increasing the overall riskiness and changing the risk characteristics of the fund midyear to lock-in gains or to reverse previous bad fortune. Further, indications for rogue trading and window dressing exist. In all of these, the interests of the principal and agent aren’t in alignment. The results are in the interest of the agent and might even be detrimental to the wealth of investors. Therefore they are considered negative from the principal’s point of view.
On the other hand, the in this thesis presented research has shown that outperformance fees do raise returns, although it’s not entirely clear if for which part this is due to less adverse selection. In any case it must also be due to overcoming moral hazard problems. Furthermore encouraging better stock selection through an outperformance fee makes expending effort on asset gathering less attractive ceteris paribus. So in this sense outperformance fees do align the interest of the principal and the agent.
Based on the reviewed literature it doesn’t become clear whether the negative effects outweigh the positive effects as has been supposed in the hypothesis. Although the evidence supports the hypothesis, it is also true that the lack of incentives might have negative effects for the principal, due to the likely underinvestment of effort on the agent’s behalf. None of the research suggest which effect is the strongest.
As I have shown there is some research to be found on the topic of outperformance fees in delegated portfolio management settings, although in my opinion more research needs
16
to be done on this topic. Since more and more fund managers are charging outperformance fees, to me it seems imperative that more research on this topic will be done. I would like to suggest for more qualitative research, both amongst fund managers and principals. Both symmetrical and asymmetrical contracts should hereby taken into account. I would also like to propose doing more of the same tests using different data from different countries to for instance check if the perceived efficiency of the market makes a difference or if legislation has an impact on characteristics by curtailing the options an agent has.
More knowledge on this subject could make investment advisers and consumers better informed on the setting in which outperformance fees make the most difference. This could help them with selecting between funds and fund managers.
17
6. Bibliography
Bhattacharya, S., Pfleiderer, P. (1985). Delegated Portfolio Management. Journal of Economic Theory, 36, (1), 1-25
Berk, J.B., Green, R.C. (2002). Mutual Fund Flows and Performance in Rational Markets, Working Paper.
Bialkowski, J., Otten, R. (2011). Emerging Market Mutual Fund Performance: Evidence for Poland. North-American Journal of Economics and Finance, 22, (2), 118-130
Brown, S.J., Gallagher, D.R., Steenbeek, O.W., Swan, P.L. (2005) Double or Nothing: Patterns of Equity Fund Holdings and Transactions, Working Paper.
Brown, K.C., Harlow, W.V., Starks, L.T. (1996). Of Tournaments and Temptations: An Analysis of Managerial Incentives in the Mutual Fund Industry. The Journal of Finance, 51, (1), 85-110
Capon, N., Fitzsimons, G.J., Prince, R.A. (1996). An Individual Level Analysis of the Mutual Fund Investment Decision. Journal of Financial Services Research, 10, (1), 59-82
Carhart, M.M., Kaniel, R., Musto, D.K., Reed, A.V. (2002). Leaning for the Tape: Evidence of Gaming Behavior in Equity Mutual Funds. The Journal of Finance, 57, (2), 661-693
Chen, J., Hong, H., Huang, M., Kubik, J.D. (2002). Does Fund Size Erode Performance? Liquidity, Organizational Diseconomies and Active Money Management, Working Paper
Chevalier, J.A., Ellison, G.D. (1997). Risk Taking by Mutual Funds as a Response to Incentives, Working Paper.
Dimsom, E., Mussavian, M. (1998). A Brief History of Market Efficiency. European Financial Management, 4, (1), 91-103
Edwards, F.R., Caglayan, M.O. (2001). Hedge Fund Performance and Manager Skill, Working Paper
Elton, E.J., Gruber, M.J., Blake, C.R. (2003). Incentive Fees and Mutual Funds. The Journal of Finance, 58, (2), 779-804
Fama, E.F. (1970). Efficient Capital Markets: A Review of Theory and Empirical Work. The Journal of Finance, 25, (2), 383-417
Fama, E.F., French, K.R. (1996). Multifactor Explanations of Asset Pricing Anomalies. The Journal of Finance, 11, (1), 55-84
Golec, J., Starks, L. (2004). Performance Fee Contract Change and Mutual Fund Risk. Journal of Financial Economics, 73, (1), 93-118
Grossman, S.J., Stiglitz, J.E. (1980). On the Impossibility of Informationally Efficient Markets. The American Economic Review, 70, (3), 393-408
Holstrom, B. (1979). Moral Hazard and Observability. The Bell Journal of Economics, 10, (1), 74-91
18
Jensen, M.C. (1968). The Performance of Mutual Funds in the Period 1945-1964. The Journal of Finance, 23, (2), 389-416
Jensen, M.C., Meckling, W.H. (1967). Theory of the Firm: Agency Costs and Ownership Structure. Journal of Financial Economics, 3, (4), 305-360
Kahn, R.N., Scanlan, M.H., Siegel, L.B. (2006). Five Myths About Fees: The Truth Behind Analyzing Fees, in the Context of Investment Goals. Journal of Portfolio Management, Spring 2006, 56-63
Kouwenberg, R., Ziemba, W.T. (2005). Incentives and Risk Taking in Hedge Funds. Journal of Banking and Finance, 31, (11), 3291-3310
Kritzman, M.P. (1987). Incentive Fees: Some Problems and Some Solutions. Financial Analysts Journal, 43, (1), 21-26
Malkiel, B.G. (2003). The Efficient Market Hypothesis and its Critics. Journal of Economic Perspectives, 17, (1), 59-82
Malkiel, B.G. (2005). Reflections on the Efficient Market Hypothesis: 30 Years Later. The Financial Review, 40, (1), 1-9
Orphanides, A. (1996). Compensation Incentives and Risk Taking Behavior: Evidence from Mutual Funds. Working Paper.
Sharpe, W.F. (1964). Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk. The Journal of Finance, 19, (3), 425-442
Sirri, E.R., Tufano, P. (1998). Costly Search and Mutual Fund Flows. The Journal of Finance, 53, (5), 1589-1622
Stracca, L. (2006). Delegated Portfolio Management: A Survey of the Theoretical Literature. The Journal of Economic Surveys, 20, (5), 823-848
Wermers, R. (2000). Mutual Fund Performance: An Empirical Decomposition into Stock-Picking Talent, Style, Transaction Costs, and Expenses. The Journal of Finance, 55, (4), 1655-1695
