Pricing traditional vs. alternative asset management services


Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Pricing traditional vs. alternative asset management services

  1. 1. Pricing traditional vs. alternative asset management services Francois-Serge Lhabitant, Head, Investment Research, Kedge Capital, and Professor of Finance, H.E.C. University of Lausanne, and Professor of Finance, EDHEC Business School The past thirty years have witnessed an increased separation the risk tolerance of the portfolio managers, and their respon- between the ownership and the control of financial wealth. siveness to incentives. The emergence of modern portfolio theory, the increased effi- ciency of markets, and the growing sophistication of financial Traditional vs. alternative incentives instruments have convinced many, if not most, investors to From a theoretical perspective, incentive contracts may com- delegate the management of their portfolios to professional bine three elements, namely, a profit sharing rule (i.e. fee asset managers and their collective investment vehicles. structure) to align incentives in terms of returns; a relative Investment advice is now becoming a commodity. performance component measured against a benchmark to monitor performance, make returns comparable, and audit for Initially, actively managed funds took the lead and intermedi- common uncertainty; and checks on risk-taking, such as max- ated much of the consumers’ investments in financial securi- imum allowable tracking error, reporting requirements, and ties. However, their dismal average performance simply pro- constraints on available investment choices. vided more general evidence of just how difficult it is to beat the market. It also opened the way for passive strategies and How are incentive contracts implemented in practice? indexed funds, which were then perceived as a cost-effective Surprisingly, the empirical evidence seems to suggest that tra- way of buying equity market exposure — a strategy that made ditional and alternative asset managers have taken diametri- sense in an environment of rapidly rising market valuations. cally opposed choices. Most traditional investment managers However, the end of the technology bubble and the subse- are monitored and evaluated against appropriate style bench- quent bear market significantly froze the development of marks, but their compensation is not linked to their relative passive funds and provoked interest in alternative invest- performance. Rather, they charge a management fee that is ments, such as hedge funds and private equity. Since then, generally expressed as a fixed percentage of the assets of the number of highly specialized, non-traditional asset man- their fund. The level of this fee varies depending upon the agement firms has been growing exponentially. Many of them complexity of the strategy and the asset class considered, but are born from the ashes of the failures of mainstream fund is typically between 1 and 3 percent per annum. Over recent managers. years, asset-based fees have been subject to highly competi- tive pressures and declined. This is not surprising, as investors Whatever the investment vehicle and investment strategy have the option of shifting their assets to another asset man- selected, the delegation of portfolio management activities ager or investment vehicle as soon as they identify a better can be seen as a particular case of the principal-agent model opportunity. initially introduced in the seminal work of Jensen and Meckling (1976). Since the costs to wealth owners of monitor- By contrast, alternative asset managers target an absolute ing those who are charged with managing their financial hold- performance, and charge both a management fee (typically ings are rather large, agency theory’s most basic suggestion is 1% of assets under management) and an incentive fee (typi- that principals (investors) should compensate the agents cally 20 percent of profits) based on their fund’s overall per- (portfolio managers) through incentive contracts in order to formance. Anecdotal evidence suggests that for most hedge align their respective interests. The nature and intensity of funds, the management fee is roughly equal to operating these incentives should depend upon a series of parameters, costs1 and the primary compensation is the incentive fee. In such as the incremental profit generated by an additional unit most cases, a hurdle rate of return must be exceeded by some of effort from the manager, the precision with which invest- multiple and any prior losses must be repaid before the fund ment performance and risk can be measured and monitored, manager is eligible to receive any incentive income. Over 1 Liang (1999) calculated the average annual management fee for hedge funds to 12 - The Journal of financial transformation be 1.36%, with a median of 1%. This base fee proved to be much smaller than total management fees surveyed from retail mutual funds.
  2. 2. recent years, these fees have risen, particularly those of strategy that works may continue selling it past the asset established managers who have been able to create a scarcity capacity for which it was designed, just because they are for their fund, which they then use to increase fees and intro- rewarded essentially on the basis of the size of their assets duce a lock-up clause.2 On the contrary, with start-up funds in under management. the course of raising capital, investors often obtain discounts on the fees in exchange for early money. By contrast, performance fees seem to do a better job at align- ing the interests of managers (desire for high fees) and Asset-based fees vs. incentive fees investors (desire for high excess returns). When subject to a One may wonder which of the two models, asset-based or performance fee, a manager will sell his strategy only up to the incentive fees, is preferable to reduce the agency costs of asset capacity for which it was designed. Then, he will close his portfolio management delegation. Fees uniquely based on the fund to additional investment, as he has stronger incentives for size of the assets under management offer a small implicit performance than for asset growth. Adding too much assets incentive to managers. As the assets in the fund grow, due to means being forced to put some money into second-best ideas, capital inflows or the appreciation of the underlying holdings, and these ideas do not often deliver the kind of returns desired, the fee collected will grow in tandem. If on the contrary, assets so asset growth is de-facto limited. At some point, managers decrease, then the fee collected will be reduced proportion- may even have to implement net share repurchases. In this ately. Several empirical academic studies have confirmed the context, an increase in revenues should essentially come from positive relationship that exists between a fund’s relative per- improving the excess returns delivered to investors rather than formance and subsequent inflow of new investments [Sirri and by increasing the assets under management. This partially Tufano (1998)], as well as the fact that some investment funds explain the relatively small size of hedge funds — about 80% of voluntarily waive their stated fees in an attempt to boost net the hedge funds reporting to commercial databases manage performance and, thereby, to attract additional assets (fee less than U.S.$100 million of equity capital. waiving). This suggests that, even though the link between performance and compensation is not direct, it nevertheless However, performance fees also have their drawbacks. The appears to be an important factor in determining fund man- most important ones are linked to their asymmetric nature, agers’ behavior. However, we should also note that academic the manager participates in the upside, but not in the down- research has evidenced the convex nature of the relationship side. This corresponds to a potentially perpetual call option between fund flow and performance. That is, while superior with a path-dependent payoff — the payoff at any time relative performance generates an increase in the growth of depends on the high-water mark, which is related to the max- assets under management and, in turn, managerial compen- imum asset value achieved. This option-like payoff structure sation, there tends to be no symmetric outflow of funds in may lead to possible adverse incentive effects, because the response to poor relative performance, at least over the short- manager simultaneously owns the option and controls its term. The convex flow/performance relationship creates an underlying asset (the portfolio), as well as its volatility. incentive for fund managers to increase risk taking, especially Therefore, near the end of an evaluation period, some man- after poor performance. Therefore, the effective incentive of agers may decide to increase portfolio risk in order to increase an asset-based fee needs to be carefully assessed on a case- the value of their option.3 On the contrary, outperforming by-case basis. However, in the case of skill-based and capacity managers may attempt to lock-in their positive performance constrained strategies, asset-based fees may also create a and dampen portfolio volatility. Alternatively, some fund man- fiduciary conflict because adding new assets can harm the agers may also try to improve the return of their portfolios by interests of existing ones. Managers who have developed a window dressing them, for example by using stale prices 2 As an illustration, Steve Mandel at Lone Pine Capital can charge half of the per- 3 Carpenter (2000) studies the optimal portfolio strategy of a manager compensat- formance fee (i.e. 10%) of any gain the fund makes from its low. This 10% per- ed with a convex option-like payoff and proves this is optimal behavior. 13 formance fee continues until the fund has made up 150% of the drawdown from the previous high, then the standard 20% fee kicks in again.
  3. 3. rather than real market values (or vice-versa) for illiquid stocks business. It establishes that ‘the law or the fund rules must or non-traded assets around the end of an evaluation period. prescribe the remuneration and the expenditure which a man- Between the lack of agreed-upon standards, different views agement company is empowered to charge to a unit trust and about illiquid marks, and moral hazard, valuation can be akin the method of calculation of such remuneration.’ Therefore, to numerical quicksand. legal restrictions to the way companies managing mutual funds can be compensated for their services, if any, are to be It is interesting to note that although mutual funds and hedge found only at the national level. Several countries, such as funds seem to disagree on what is the best choice between Spain, France, or the U.K., have left a large degree of latitude asset-based and performance-based fees for their external when it comes to portfolio managers deciding on the mecha- investors, they both agree on their own internal compensation nism and the value of their compensation. Strikingly, in prac- structures, which involve asset management firms and indi- tice, even though it is legally permissible, most mutual fund vidual fund managers. The compensation of portfolio man- companies are almost never compensated through incentive agers tends to be performance-based, with a fixed base salary contracts. Instead, they are paid a fixed percentage of assets topped by bonuses based, partially or entirely, on relative per- under management, and the incentive intensity is set to zero. formance. This should be kept in mind, as a complete discus- At the other extreme, hedge funds and other lightly regulated sion on the incentives facing mutual funds must consider two private investments companies are primarily charging incen- layers of agency problems: the agency relationship between tive fees. the fund company and the fund investors and the agency rela- tionship between the fund company and fund management The soft dollar arrangements [Chevalier and Ellison (1999)]. Our discussion of fees would not be complete if we did not mention soft dollar brokerage, or simply soft dollars. Soft dol- The regulatory view lar brokerage is a popular arrangement between a fund and its An interesting viewpoint on the question of asset manage- broker. Basically, the fund manager agrees to place a desig- ment fees is that of regulators, which varies from one country nated dollar value of trading commission business with a bro- to another. In the U.S., for example, mutual funds are regis- ker over a given period of time. In exchange for this promise, tered investment companies and they are highly regulated by the broker provides the manager with research credits equal the S.E.C. The latter allows performance incentive fees and to some part, say 50 percent, of the promised commissions. enables a fund to charge higher fees when it beats a bench- Rather than rebating these credits back to investors, the man- mark, so long as it is willing to charge less when it fails to beat ager keeps them and uses them to buy services and any of the it. As one could expect, many fund managers are perfectly large number of broker-approved research products (hard- happy to sell their funds to the public on the grounds that it ware, software, subscriptions, databases, etc.) supplied by can beat the market, but despite the offer, very few of them third-party research vendors. The broker then pays the man- are willing to put their own money where their mouths are and ager’s research bill and simultaneously cancels the appropri- take the other side of the bet. According to the Lipper data- ate number of credits from the manager’s soft dollar account. base, less than 2 percent of the U.S. equity mutual funds apply From a functional perspective, soft dollars are simply one form a performance fee. of bundling research and execution together into a single com- mission payment. They are unique in allowing research and In Europe, a European Council Directive sets the general legal execution to be provided by entirely separate firms, thereby framework within which undertakings for collective invest- promoting vertical disintegration of the research and execu- ment in transferable securities (UCITS) may carry on their tion functions. 14 - The Journal of financial transformation
  4. 4. Do soft dollars reduce or increase agency costs of delegated ity brokerage. The threat of termination dramatically increas- portfolio management? Both views are defendable. On the es the expected losses to brokers who provide low-quality one hand, one may argue that soft dollars allow managers to services, and may therefore perform an effective quality misappropriate investor’s wealth by churning their portfolios assuring function. to subsidize research for which they should pay directly. This, in turn, generates various inefficiencies, such as the choice of What makes a good performance fee? a broker for his willingness to provide research credits rather Coming back to the main topic of our discussion, at this stage, than on expected execution quality. At the end of the day, we may wonder what the necessary characteristics of a good because brokerage commissions are included in the price performance fee should be. Ideally, a performance fee should basis of the underlying security, investors implicitly pay the be structured to achieve five main objectives. It should reward underlying research costs. Soft dollars, therefore, subsidize a proficient manager for excess return earned over the meas- the manager’s use of research inputs, and in some cases the urement period, it should control portfolio risk, it should con- existence or amount of the subsidy is unknown to investors. tain fair but significant consequences for manager underper- Thus, portfolio managers shift expenses that are normally formance, the performance fee agreement should be explicit shouldered by them onto fund shareholders. But on the other in its description of the fee structure to eliminate client mis- hand, one may also argue that soft dollars are aligning the understandings and properly frame client expectations, and it interests of asset managers with those of their investors. Fund should be designed so that there is little economic incentive managers typically own a very small percentage of their port- for the manager to grow the assets under management folio, directly as co-investors or via an annual management beyond the level at which the performance fees max out. The fee. If managers were required to pay for all research and exe- performance fee structure encourages investment firms to cution out of their own pockets, they would bear a dispropor- run their strategies at optimal asset levels that permit the tionate share of the costs of generating portfolio returns in maximization of dollars of excess return. relation to the private benefits based on their portfolio share. Seen in this light, the agency problem faced by portfolio Are hedge fund fees exaggerated? investors is that in the absence of agreement, managers will Throughout the bull market of the 1990s most investors over- do too little research, identify too few profitable trading oppor- looked the fees charged by mutual fund companies because tunities, and execute too few portfolio trades. Thus, soft dollar returns were so impressive. But times have changed. We are arrangements allow investors to subsidize investment now in an era of difficult markets and the level of fees have research and thereby encourage managers to do more of it, come under close scrutiny. Many traditional investors who are which ultimately benefits the portfolio performance. just beginning to venture into alternative investments find their levels of fees overwhelming. If the industry standard Last but not least, soft dollars may also be unique in aligning seems to be 1 percent for the management fees and 20 per- the incentives of brokers and managers. When a broker pro- cent for the performance fee, several funds among the largest vides soft dollar research credits to a manager, it typically and top-performing ones are far above that. For instance, does so in advance of the commission payments it expects Caxton Corporation which oversees more than U.S.$10 billon from the manager. But the manager has no legal obligation to charges 3 percent and 30 percent, while Renaissance’s trade and may in particular terminate the executing broker U.S.$6.7 billion Medallion fund charges a 44 percent incentive relationship with the balance of the soft dollar account unpaid. fee, more than twice the industry average. Interestingly, both The broker will then lose a stream of commissions that would funds are closed to new investors and have returned money to have included a premium above the cost of providing low-qual- their existing investors in 2003 in order to be able to maintain 15
  5. 5. positive returns. Of course, only the best performing funds are ture for asset management? One argument often encoun- able to dictate conditions like this. Nevertheless, the list of the tered is that poorly performing managers will be paid less and, top ten earners in the hedge fund industry is impressive. therefore, benefit the plan sponsor. On the other hand, man- According to Institutional Investor, the top 10 managers agers who perform well will also be paid more. But since the earned the following sums in 2003 from a combination of fund earns more, this extra fee will really not cost anything at their share of the fees generated by the funds they managed all. Perhaps, proponents contend, the carrot of higher fees and and the gains on their own capital in the funds: George Soros the stick of lower ones will make the managers work harder. of Soros Fund Management, U.S.$750 million, David Tepper of Appaloosa Management, U.S.$510 million, James Simons of The objectives of performance fees are to reduce them for flat Renaissance Technologies, U.S.$500 million, Edward Lampert and negative performance and to reward managers for posi- of ESL Investments, U.S.$420 million, Steven Cohen of SAC tive absolute performance. Structured properly, this makes a Capital Advisors, U.S.$350 million, Bruce Kovner of Caxton lot of sense for the investor and the manager if added value is Associates, U.S.$350 million, Paul Tudor Jones of Tudor properly identified. Then the client and manager are simply Investment, U.S.$300 million, Kenneth Griffin of Citadel entering a profit-sharing plan, and profit sharing is effective in Investment, U.S.$230 million, Daniel Och of OCH-Ziff Capital aligning incentives. The problem with performance fees starts Management, U.S.$150 million, and Leon Cooperman of when they are not structured properly, that is, if the client is Omega Advisors, U.S.$145 million. giving a manager a fee based on something other than added value (the true alpha). This is not sustainable in the long-run. Not surprisingly, traditional investors’ first reaction may be to Nevertheless, many traditional managers are still reluctant to dismiss the hedge fund industry due to excessive layers of use performance fees. If the entire industry shifted to per- fees. Performance fee structures with 25 and 35 percent carry formance fees, one of the things that might happen is a reduc- can work out to be tremendous fees, and immediately prompt tion in fees in general. For instance, if two-thirds of the man- the question: ‘Does the return justify the fee?’ The answer is agers underperformed, they would draw one-third of their twofold. Firstly, outsiders invest in a hedge fund because they normal fees, while the one-third that outperformed would believe the manager has an expertise that they can not repli- draw four-thirds of their normal fees. The industry-wide fees cate for themselves, or that replication is too costly. This is a would then be cut by a third. Not surprisingly, at least two fact to remember when looking at hedge fund fees — you get thirds of the asset management industry will keep fighting what you pay for. Secondly, if investors achieve their objec- such a trend. tives after expenses, the fees are justified, even if their level is an especially hard pill to swallow.4 But if a fund delivers poor performance, it is not worth a low fee; in fact, it is worth References • Carpenter J. N., 2000, “Does option compensation increase managerial risk no fee at all. Thus, fees should be directly related to provid- appetite?” Journal of Finance, 50, 2311-2331 ing what the investor wants. Consequently, when evaluating • Chevalier J. and G. Ellison, 1997, “Risk taking by mutual funds as a response to or selecting an investment fund, the fee charged should not incentives,” Journal of Political Economy, 105, 1167-1200 • Goetzmann, M., J. Ingersoll, and S. Ross, 1998, “High water marks,” NBER Working be the unique determinant. The investment philosophy and Paper 6413, National Bureau of Economic Research, Cambridge, MA quality and tenure of management are also important con- • Jensen, M. C., and W. H. Meckling, 1976, “Theory of the firm: Managerial behavior, agency costs, and ownership structure,” Journal of Financial Economics, 3, 305-360 siderations, amongst others. • Liang B., 1999, “On the performance of hedge funds,” Financial Analysts Journal, 55, 72-85 • Sirri E. R. and P. Tufano, 1998, “Costly search and mutual fund flows,” Journal of Why so much resistance? Finance, 53, 1589-1622 So, in conclusion, are performance-based fees a desirable fea- 4 As an illustration, Goetzmann et al. (2001) use an option approach to calculate the 16 - The Journal of financial transformation present value of the fees charged by a hedge fund manager and show that the present value of the incentive fees can be quite high (i.e. for a volatility of 15%, the fee can be as high as 13% of the assets under management).