Content uploaded by John Mulvey
Author content
All content in this area was uploaded by John Mulvey on May 17, 2016
Content may be subject to copyright.
Risks
The role of
hedge funds for
long-term investors
John M. Mulvey
Professor, Bendheim Center for Finance,
Princeton University
Abstract
Hedge funds have gained popularity for increasing investment
returns. We focus on the role of this asset category for long-
term investors, with attention to rebalancing a portfolio of
diversified assets. An investor must seek out securities with
low correlations to traditional assets to maximize asset
growth. Current hedge fund returns, as measured by average
performance, show dependencies with equity returns. Other
limitations and opportunities for hedge funds are discussed.
23
1For example, the volatility of a target mix such as 70% equity - 30% bonds is
reasonable as a start.
The role of hedge funds for long-term investors
The promise
Hedge funds are defined by the generic description – ‘any
pooled investment vehicle that is privately organized, admin-
istered by professional investment managers, and not widely
available to the public’ [IMF (2000)]. The hedge fund industry
has grown from roughly U.S.$ 100 billion market value in 1990
to approximately U.S.$ 750 billion in early 2004. Almost 6000
funds are now available. There are several underlying causes
for this growth:
■Institutions and individuals are under severe pressure to
generate high returns in order to meet future liabilities and
goals, including under-funded U.S. pension plans, university
endowments, and state retirement accounts.
■Prominent U.S. institutional investors, such as Harvard,
Princeton, and Yale Universities, have achieved superior
returns by committing substantial portions of their capital
to alternative investments, including hedge funds,
in concert with deploying leverage.
■Traditional assets have performed below expectations over
the past few years (notwithstanding the recent rise in
equities).
■Venture capital investments have wilted during the
recent past.
■Hedge funds have held up reasonably well. The average
returns for many hedge fund sectors have been noteworthy
(see below).
■Current academic research supports the supposition that
patterns in returns may be detected by careful analysis
(e.g. market micro-structures).
■Successful hedge fund managers have achieved notoriety
by their proprietary approach to trading and, on occasion,
eye-popping returns.
Should investors put a serious amount of capital into this cate-
gory? Is the game too late? How can the reader make deci-
sions about investing in hedge funds? We will discuss these
issues in this article.
Due to their inflexible structure, short-term investors should
mostly avoid hedge funds. Likewise, conservative investors
should take into account their goals and liabilities within an
asset and liability study [Ziemba and Mulvey (1998]. To this
point, a pension plan will determine its surplus by computing
the market value (assets) – market value (liabilities). Hedge
funds may fit within a carefully crafted asset-liability system,
but this question lies outside the present discussion.
We will focus on long-term investors who possess a moderate-
to-aggressive tolerance for risk. Their goal, and our goal in this
report, is to maximize the growth of assets over a substantial
time period, while maintaining risks within a specified limit1.
Naturally, the investment horizon depends upon circum-
stances - a family trust or pension plan, for example, may set
seven to ten years as a sensible planning horizon.
The reality
Several barriers must be overcome in order to successfully
invest in hedge funds. First, despite recent improvements,
there are severe informational constraints on reported per-
formance. The historical record, largely, began around 1990
when the hedge fund universe was much smaller and less well
established. Not all hedge funds are open regarding their
results due to proprietary and related considerations.
Likewise, it is difficult to estimate the survivor bias, since
hedge funds may stop reporting results when anticipating a
meltdown. Studies have estimated the bias at 1.5 % to 3 % or
more per year. Also, most hedge funds are relatively small -
under U.S.$ 100 million in capital – thereby inaccessible for
many institutional investors.
Second, management costs are higher than traditional mutual
funds. This headwind is partially offset by incentive fees that
pay a bonus only when the fund achieves results above a
designated benchmark. Thus, the goals of the investor and
manager are closely aligned.
A third issue involves the ability of an investor to rebalance his
portfolio on a regular basis. Hedge funds require lock up periods
and other restrictions; it may be difficult to add or subtract
money as conditions warrant. This restriction is particularly
24 - The Journal of financial transformation
The role of hedge funds for long-term investors
constraining on large institutional investors. We will take up
the rebalancing issue below.
Next, to evaluate a portfolio on an anticipatory basis, we must
be able to estimate the factors that drive returns. In certain
cases, the issue is straightforward. For example, funds in the
statistical arbitrage category generate returns that are rela-
tively uncorrelated with economic factors. This independence
is, in fact, a prime advantage of a hedge fund. But, in other
cases, estimating future returns is complicated by the hedge
fund manager’s freedom to change direction abruptly. A fund
may be 150% long and 50% short in one period, and the
opposite in the next period. It is difficult to construct a reliable
portfolio system under these changing conditions since we
must somehow estimate the manager’s decision processes.
In addition, certain strategies will produce consistent winners
within a rather narrow size limit – e.g. index arbitrage. When a
large number of well-placed investors work in an area, the
large excess returns that initially characterized the area will
fall to a more modest level. Again, it is difficult to reliably esti-
mate this relationship within an anticipatory framework.
A partial solution – fund-of-funds
A funds-of-funds (FOFs) is an entity that invests in a narrow or
more commonly a broad segment of the hedge fund universe.
The FOF conducts the time consuming task of due diligence
for the clients and produces a diversified portfolio of individ-
ual hedge fund managers. Thus, the implosion of any single
fund will cause a minor or temporary distortion to the overall
portfolio.
In several ways, the FOF performs a service that is similar to a
mutual fund by selecting individual fund managers (as com-
pared with stock selection). This service is particularly helpful
for investors who are unable to spend adequate time or
resources on the selection tasks. Of course, the investor must
pay for the help, thus reducing benefits, and the fund-of-fund
must be chosen from a number of providers.
FOFs have achieved good results over the historical period, as
compared with traditional assets, and as measured by the
mean return of reporting funds. We are assuming zero sur-
vivor bias in these discussions; the reader should carefully fac-
tor this issue into future projections, however.
The annual returns and volatilities of asset categories for the
past 13 years (January 1990 to August 2003) are shown below
and plotted in Figure 1.
Table 1
Geo returns Std (returns)
Real Estate Trusts (REIT) 10.7 % 12.0 %
S&P500 10.1 % 14.5 %
Fund of Funds 9.6% 4.4 %
U.S. T-bonds 9.1 % 8.5 %
U.S. T-bills 5.3 % 0.9 %
International Equity (EAFE) 2.0 % 17.4 %
The FOF performance is slightly below the S&P500 index with
a much lower volatility. This performance is consistent with
the mean returns for sub-sectors in the hedge fund universe
achieving the following results [Dow Jones & Co. (2004)]:
Table 2
Geo return Std (returns) Corr.S&P500
Event driven 11.9 % 4.5 % .487
Global emerging 13.0 % 15.7 % .493
Global international 11.4 % 6.9 % .51
Global established 14.5 % 9.0 % .763
Global macro 12.9 % 6.7 % .43
Market neutral 10.6 % 1.5 % .323
25
2Such as lock up periods.
The role of hedge funds for long-term investors
But again, we warn readers that this data does not adjust for
survivor bias or the size effect mentioned earlier. Schneeweis,
Kazemi, and Martin (2001) discuss this hedge fund sub-sector
performance.
Evaluating historical performance within
a portfolio context
Any serious asset allocation study for a long-term investor,
such as a pension plan or family trust, begins with reviewing
historical performance. We all know, of course, that historical
results are no guarantee of future returns. But there are
important lessons to be learned.
Unfortunately, reliable data on hedge funds is unavailable
before 1990. Thus, we must be particularly careful when eval-
uating this data for long-term projections. Also, as mentioned,
several issues complicate the task – including survivor bias
and size limits. But to start, we turn to Figure 1. Here, returns
and risks are shown for the six aforementioned assets over
the period 1990 to 2003.
First, observe the three U.S. assets - equity (S&P500), govern-
ment bonds (T-Bonds), and CASH (T-Bills) display the tradi-
tional lineup - from conservative with low returns to more
aggressive with higher returns; see the drawn efficient fron-
tier. Below this line are two assets - EAFE and GSCI - due to the
abysmal return of Japanese equities and the gradual decline
in interest rates, inflation, and commodity prices since 1990.
Above the line are the superior assets (from the 13-year per-
spective) – real estate investment trusts (REITs) and FOFs.
To calculate the best combination, we could solve an optimal
portfolio problem as is commonly done. But our purposes are
more modest. We focus on two simple portfolios as equal-
weighted combinations of three assets. We call these blended
portfolios. To achieve a blended portfolio, the investor rebal-
ances his assets at the beginning of each month to the desired
ratio. Each asset begins the month at one-third of the
investor’s wealth. Two sets of blended assets are noteworthy.
First, we combine S&P500, REITs, FOFs (point ABD). The FOF
returns are quite good on a standalone basis – showing 9.56%
returns, and 4.4% volatility. The blended portfolio achieves a
solid return equal to 10.5%, and 8.0% volatility.
However, a similar pattern can be achieved by replacing the
FOF with T-Bonds, giving a blend with – S&P500, REITs, and T-
Bonds (point ABC). Even though, FOFs dominate T-Bonds on a
stand-alone basis, the T-Bond volatility works to the advantage
of the long-term investor. Rebalancing gains are higher when
an asset has higher volatility and good expected returns
[Mulvey, Lu, and Sweemer (2001); Mulvey, Pauling, and Madey
(2002)].
Additionally, two practical issues arise when placing FOFs with-
in a blended portfolio. First, institutional constraints with
hedge funds2prevent much of the monthly rebalancing. If fact,
FOF’s low volatility reduces rebalancing gains even if the
transactional constraints are dropped. Second, of course, his-
torical returns may not be accessible for future investors.
What to expect and recommendations
Hedge funds are likely to continue growing in popularity, due
to the increased demand for customized products by institu-
tional investors, wealthy individuals, and family trusts. And the
large potential fee structure will attract entrepreneurial asset
managers. Given this trend, we expect that the area will
receive greater attention from investors and researchers alike.
Surprisingly, perhaps, novel investment opportunities have
become accessible for individuals with modest means. Take
the case of exchange traded funds and single stock futures.
We are seeing a split between inexpensive standardized prod-
ucts, and the high-cost, high potential value-added services
such as hedge funds. This pattern is evident in other domains
such as retail merchandise – Wal-Mart and Sam’s Club on the
one hand, and expensive food-stores such as Wegman’s that
supply superior service and customized prepared foods, on
the other hand (often in the same shopping center). Our rec-
ommendations are as follows. While hedge funds data exists
since 1990, these funds have not been fully battle tested.
26 - The Journal of financial transformation
The role of hedge funds for long-term investors
Thus, many long-term investors should wait for the category
to become more seasoned, especially as the size of the indus-
try grows, before committing a substantial percentage of
assets. As discussed, hurdles must be overcome - lack of trans-
parency, difficulty to pick the better funds (too much survivor
bias and extra costs), and for FOFs the need to generate
returns with lower correlations with market returns. Investors
may put a portion of their assets in hedge funds in order to
gain experience with this asset class. However, a large com-
mitment is risky at present for novice hedge fund investors.
There are several items that would assist in generating
increased interest in hedge funds from long-term investors
such as pension plans. These include:
■Fur ther evidence that hedge fund returns will do well when
traditional assets are performing poorly – to hedge in a
meaningful manner. The correlation of absolute return
hedge funds with the S&P500 should be close to zero.
■Greater flexibility to move money into and out of this
category as conditions warrant (strive for re-balancing gains).
■Some reduction in fees if investor stays with a fund for a
longer time period and the manager is relatively successful.
■Greater reliance on multi-strategy funds that take
advantage of opportunities across hedge fund tactics.
In conclusion, as with all new technologies, the early imple-
menters take on increased risks. But indeed opportunities
exist for superior performance as seen by the results of top
U.S. University endowments.
27
0.00%
0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00% 16.00% 18.00% 20.00%
2.00%
4.00%
T-B ill
D (Fund of Funds)
ABD ABC
ABD - 50% Leverage
ABC - 50% Leverage
C (T-Bond)
B (REIT)
A (S&P 500)
GSCI
EAFE
6.00%
8.00%
10.00%
12.00%
14.00%
Compound Annual Return
Standard Deviation
Risk-Return Profile of Rebalanced Portfolios (monthly returns from January 1990 - August 2003)
Single Assets Portfolios
Rebalanced Portfolios w/o Fund of Funds
Rebalanced Portfolio with Fund of Funds
Figure 1: Historical performance of selected assets (1990 to 2003)
The role of hedge funds for long-term investors
A critical concept for generating superior returns for long-
term investors is to discover novel ways to diversify their port-
folio, thus reducing volatility, and then to leverage the widely-
diversified portfolio to an acceptable risk tolerance. Harvard
University implements this concept as well as anyone [Dow
Jones & Co. (2004)]. Their portfolio consists of a wide range
of assets – timber, inflation linked bonds, hedge funds, venture
capital, etc. In this context, the primary requirement is to
generate good returns with relatively low correlation to the
other assets. To illustrate this point, see Figure 1 for leveraging
the two blended portfolios by 50% – generating mixes with
13% annual returns and 12 % volatility.
Hedge fund managers and potential/actual investors in alter-
native investments should take note of these complementary
goals.
References
•CISDM Hedge Fund Benchmark Series, Center for International Securities and
Derivative Markets, University of Massachusetts, in cooperation with MAR, 2003.
•Dow Jones and Company, Barrons Online, ‘Educating Harvard,
How Two Money Managers + almost $70 Million in Pay = a Bargain,’ Feb 2, 2004.
•International Monetary Fund, Background note on the Hedge Fund Industry,
prepared for the Financial Stability Forum, 2000, http://www.fsforum.org .
•Mulvey, J, N. Lu, and J. Sweemer, 2001, ‘Rebalancing Strategies for Multi-period
Asset Allocation,’ Wealth Magazine, Fall
•Mulvey, W. Pauling, and R. Madey, 2002, ‘Advantages of Multi-Period
Portfolio Models,’ Journal of Portfolio Management, Winter
•Schneeweis, T., H. Kazemi, G. Martin, 2001, ‘Understanding Hedge Fund Performance,’
Lehman Brothers Report, November
•Ziemba, W. and J. Mulvey, (editors), Worldwide Asset and Liability Modeling,
Cambridge University Press, 1998.
Appendix
A description of the sub-sectors of the CISDM/MAR database is listed below.
Event driven: The investment theme is dominated by events that are seen as special sit-
uations or opportunities to capitalize from price fluctuations. They specialize either in
risk arbitrage (merger arbitrage) or distressed securities.
Global emerging: In this type of hedge funds, managers invest in less mature financial
markets. Because shorting is not permitted in many emerging markets, managers must
go to cash or other markets when valuations make being long unattractive. They focus
on specific regions.
Global international: Here, the manager pays attention to economic change around the
world (except U.S.); bottom-up-oriented in that they tend to be stock-pickers in markets
they like. They use index derivatives much less than macro managers.
Global established: They focus on opportunities in established markets. (U.S. opportu-
nity, European opportunity, Japanese opportunity). This type of hedge funds can be sub
classified as Growth, Small-Cap, and Value Global Established.
Global macro: They are the classic opportunistic funds investing anywhere they see
value. They use leverage and derivatives to enhance positions, which will have varying
time-frames from short (less than 1 month) to long (more than 12 months).
Market neutral: They attempt to minimize market risk by using such strategies as con-
vertible arbitrage, stock arbitrage, and fixed-income arbitrage or by taking both short
and long positions in different stocks.
Fund of funds: Th ey are funds, which allocate capital among other investment funds,
including hedge funds. They can either be diversified, which allocate capital to a variety
of fund types or niche, which allocate capital to a specific type of fund.
28 - The Journal of financial transformation
