Fund of Hedge Funds by sii84458

VIEWS: 63 PAGES: 24

More Info
									Forthcoming: Financial Analysts Journal

                Hedge Fund Performance: 1990-1999




                                Bing Liang

                   Weatherhead School of Management
                    Case Western Reserve University
                         Cleveland, OH 44106

                         Phone: (216) 368-5003
                          Fax: (216) 368-4776
                       E-mail: BXL4@po.cwru.edu




                     Current Version: February 2000
                                       Abstract


   Using a large database, we study hedge fund performance and risk during a ten-year

period from 1990 to 1999. The empirical results show that hedge funds have an annual

return of 14.2% during this time period, comparing with 18.8% of the S&P 500 index.

However, the S&P 500 index is much more volatile than hedge funds as a whole. We

document an annual survivorship bias of 2.43% for hedge funds. We examine the year

1998 in detail since hedge funds are heavily affected by the global financial market

tumble in that year. 1998 had the highest volatility for hedge fund returns. More funds

died and fewer funds were born in 1998 than any other year. There are very few funds

that change their fee structures. We find that fund fee changes are performance related:

poor-performed funds drop incentive fees.




                                            1
   Hedge funds are one of the fastest growing sectors in finance. Recently, hedge funds

make the headline news almost on a daily basis. In particular, after the 1997 Asian

financial crisis and the 1998 near collapse of the Long-Term Capital Management LP,

hedge funds have been catching the eye of investors, money managers, and regulators.

   However, the public has a very limited understanding of the hedge fund industry. The

main reason is that the information on returns, risk, and fee structures of hedge funds are

largely unavailable to the public since hedge funds are either non-regulated U.S.

partnerships or offshore corporations that are not required to disclose their information.

This unregulated feature on hedge funds gives fund managers huge flexibility to operate

in a “black box”, which generates tremendous curiosity from the investment

communities. In addition, a lot of attention has been focused on a few notable hedge

funds, but little effort has been made to consider the industry as a whole.

   Currently, there are a few commercial data vendors/consultants that possess hedge

fund data. Examples are Hedge Fund Research, Inc. (HFR), Managed Account Reports,

Inc. (MAR), and TASS Management Limited (TASS).1 A few academic studies on hedge

funds are usually based on these databases. For example, Fung and Hsieh (1997a) use

combined data from Paradigm LDC and TASS. Ackermann, McEnally, and Ravenscraft

(1999) utilize combined data from HFR and MAR. Liang (1999) also uses data from

HFR. One exception is that Brown, Goetzmann, and Ibboston (1999) employ the hand-

collected data from the U.S. Offshore Funds Directory.

   Liang (2000) indicates that there are differences and errors in some hedge fund

databases. For the same funds tracked by HFR and TASS, returns, assets, fees, and




                                             1
investment style classifications could be different. He suggests that the TASS database

should be used for academic research because of its relative completeness and accuracy.

   In this paper, we use the TASS data containing over 2,000 hedge funds to study their

performance and risk for a ten-year period from January 1990 to July 1999. We evaluate

the industry by using more funds and over a longer time horizon than the previous

studies. Our comprehensive data also allows us to examine how the current financial

crises impact the hedge fund industry. Moreover, we examine not only live funds but also

dead funds, therefore we can study the survivorship bias issue of hedge funds. In

particular, we focus our attention on the year 1998 since that was a special year for the

entire hedge fund industry. In 1998, Russian debt defaulted, the Long-Term Capital

Management LP almost collapsed, and many other hedge funds suffered similar losses as

the Long-Term Capital Management LP.



Data



   We obtain data on fund returns, assets, fees, investment styles, and other fund

characteristics from TASS. As of July 1999, the TASS database covered 2,016 hedge

funds, including 1,407 live funds and 609 dead funds. The total asset under management

is about $175 billion. Among the 2,016 funds, the majority report returns, net of various

fees, on a monthly basis. After deleting the 95 funds with gross returns and quarterly

returns, we have 1,921 funds left for our study.

   According to TASS, there are 15 investment styles. They are top down macro, bottom

up, short selling, long bias, market neutral, opportunities, relative value, arbitrage,




                                             2
discretionary, trend follower, technical, fundamental, systematic, diverse, and other. Note

that these styles may be overlapping.2


Hedge Fund Performance: 1990-1991


    Across Time and Investment Styles. We analyze hedge fund performance and risk

by year and by investment styles. Table 1 shows hedge fund performance by investment

styles from 1990 to 1999. Over the 10-year period, the average monthly return of all

hedge funds in our sample is 1.11% per month, or 14.2% per year. Among them, 1993 is

the best year with an annualized return of 27% and 1994 is the worst year with an annual

return of –0.6%. However, the most volatile year for hedge funds is neither 1993 nor

1994, it is 1998 when many hedge funds run into trouble due to the Russian debt crisis,

soon after the Asian financial crisis in 1997.

    The average standard deviation of returns across 15 styles in 1998 is 2.57%, much

higher than the other years. Note that this ten-year period is roughly coincident with the

longest bull market in the US stock market history. The average annualized return for the

S&P 500 index is 18.8% during the same period.

    Across investment styles, we have winners and losers. The winners are opportunistic

and long bias strategies, and the losers are systematic and technical strategies. Note that

these styles are overlapping since a fund can employ both long bias and short selling

strategies.


    Hedge Funds versus S&P 500 Index. In Figure 1, we plot the cumulative monthly

returns of hedge funds versus the cumulative monthly returns of the S&P 500 index from

January 1990 to July 1999. As we can see, a $1 investment in all hedge funds in January



                                                 3
1990 can grow to $3.39 in July 1999. In contrast, a $1 investment in the S&P 500 index

grows to $4.79 during the same time period. Note that the cumulative monthly return for

all hedge funds is above that for the S&P 500 until the end of 1996. The substantial

growth in the US equity markets from 1997 to 1999 gives the S&P 500 a huge lift, which

contributes to the final result of this horserace between hedge funds and the S&P 500

index. Note that survivorship bias may play an important role here: The TASS database

starts to collect liquidated funds only from 1994, the superior performance of hedge funds

prior to 1994 may be from ignoring the dead funds.

   According to Figure 1, the live fund group significantly outperforms both the dead

fund group and all hedge funds (including both the living the dead funds). For example, a

$1 investment in the dead funds grows to only $1.84, comparing to $3.99 in the live

funds. The performance difference between the live and the dead funds is significant only

after 1994. Again, this may be related to survivorship bias. We will discuss the

survivorship bias issue in the next section.

   Although the S&P 500 index wins the competition during this 10-year period, the

index is more volatile than hedge funds as a group. From January 1990 to July 1999, the

standard deviations of monthly returns for the S&P 500 index, all hedge funds, the live

funds, and the dead funds are 3.89%, 1.67%, 1.70%, and 1.91%, respectively. Therefore,

the S&P 500 index is much more volatile than the overall hedge fund markets.3 As

pointed out by the previous studies, hedge fund strategies are less correlated with each

other and they have low correlations with the traditional asset classes (see Fung and

Hsieh (1997), Ackermann, McEnally, and Ravenscraft (1999), and Liang (1999)). Hedge

funds can effectively reduce their risk by doing dynamic hedging, combining both long




                                               4
and short positions, and diversifying their portfolios across different financial markets.

Note that the dead fund group has a higher volatility than the live fund group. Their

trading strategies are riskier than the live funds, which may attribute to their

disappearance.

   To compare risk-adjusted returns, we calculate the Sharpe ratios. The Sharpe ratios

are 0.27, 0.41, 0.48, and 0.08 for the S&P 500 index, all hedge funds, the live funds, and

the dead funds, respectively. Hence, on a risk-adjusted basis, hedge funds outperform the

S&P 500 index during the period from January 1990 to July 1999. However, we need to

be more cautious when explaining this result since we may underestimate the

survivorship bias in early years when dead funds are not available to data vendors.


   Survivorship Bias. Survivorship bias results from the fact that poor performed funds

disappear over time and calculation of fund returns based on survived funds only can

generate an upward bias in fund returns. So far, different studies about hedge fund

survivorship bias report different results. For example, Fung and Hsieh (1998) document

an annual survivorship bias of 1.5%. Brown, Goetzmann, and Ibboston (1999) report an

annual survivorship bias of 3% for offshore funds.4 However, Ackermann, McEnally, and

Ravenscraft (1999) indicate that the survivorship bias is small at an average magnitude of

0.16% per year. Liang (2000) reconciles the conflicting results about survivorship bias in

the above studies by showing that two major hedge fund databases contain different

amounts of dissolved funds, hence different estimates on survivorship bias.

   Table 2 shows that hedge funds have an average attrition rate of 8.54% per year. On

average, there are about 8.54% funds that disappeared each year. If the main reason for a

fund’s disappearance is poor performance, then an upward survivorship bias will exist in



                                            5
fund returns.5 In Table 3, we show that the average monthly returns for the S&P 500

index, all hedge funds, the live funds, and the dead funds are 1.45%, 1.08%, 1.22%, and

0.55%, respectively.6 There is a substantial return difference between the live group and

the dead group. Following previous literature, we calculate survivorship bias as the return

difference between the survived funds and all funds. The result indicates that the bias is

0.14% per month or 1.69% per year during the ten-year period. Note that TASS collected

the dead fund data starting in 1994, therefore a meaningful calculation of survivorship

bias should be based on data from 1994 and on. As a result, the return difference

between the survived funds and all funds during the period from 1994 to 1999 is as high

as 0.2% per month or 2.43% per year. This is higher than the 1.69% from the whole ten-

year period. Hence, the survivorship bias of hedge fund returns should be 2.43% per year,

which compares with the 1.5% bias in Fung and Hsieh (1998), the 3% bias in Brown,

Goetzmann, and Ibboston (1999) for offshore funds, and the 2.24% bias in Liang (2000).

The positive survivorship bias confirms that the main reason for a fund’s disappearance is

poor performance.



Hedge Funds in 1998


   1998 was a disaster year for the hedge fund industry. On August 17, Russia defaulted

its ruble debt and domestic dollar debt. Trading in Russian debt was halted, the stock

market tumbled, and the ruble was depreciated. The crisis in Russia soon spread to the

financial markets in other countries and caused a panic among investors. In response,

investors’ money flowed to high quality debt instruments such as the US Treasury

securities and other government debt. Credit spread widens between the high quality and



                                            6
risky debts. This credit spread widening reverses a multiyear trend toward spread

tightening. In addition, due to massive sell-offs, liquidity risk premium also increases for

corporate bonds, mortgage-backed securities, and other illiquid securities. As a result,

many hedge funds like the Long-Term Capital Management that is betting on

convergence in yield spreads and invest heavily in the fixed-income security markets,

suffer tremendous losses and face margin calls from their lenders. Under pressure, hedge

funds are forced to liquidate their portfolios, to deleverage their positions, or to go out of

business.

   Consequently, hedge fund lenders such as investment banks, commercial banks,

brokerage houses, and other counterparties tighten their credit to hedge funds although

some of them seldom requested a collateral or a “haircut” before.7 This kind of credit

squeeze, together with investor withdrawal, creates more pressure and losses for hedge

funds because hedge funds such as fixed income arbitrage funds and global macro funds

rely heavily on leverage to achieve large positions and to boost returns.



   Monthly Returns. Here we examine hedge fund returns in each month of 1998 and

also look at the first half and the second half of 1998. Figure 2 shows the 12 monthly

returns for hedge funds in 1998. We expect that August is the worst month for many

hedge funds since the default on Russian debt triggered a ripple effect to the global

economy. As a matter of fact, the loss in August is as high as 4.88% for all the funds we

studied, far beyond the gains and losses in other months. However, there was some

recovery in September and October after the Fed coordinated 14 of the largest financial

institutions to bailout the Long-Term Capital Management and further lowered the key




                                              7
interest rates to ease the crisis. By November, the Long-Term Capital Management had

already made profit, and returned $2.6 billion of the $3.63 billion the bailout group

injected into the fund. The fund’s founder John Meriwether had been approved by the 14

institutions to launch a new hedge fund.

   As a group, hedge funds are able to generate returns of 2.36% and 1.95% in

November and December of 1998, respectively. During the first seven months of 1999,

hedge funds were able to produce an average monthly return of 1.44% from January 1999

to July 1999. When we further break down 1998 into two-halves, we find that the average

monthly return in the first half is 0.61%, comparing with –0.12% in the second half. The

difference is statistically significant at the 1% level (t=5.99). Hence, the poor

performance if hedge funds in 1998 was mainly driven by the second half when the

global financial markets experienced profound turbulence.



   Death versus Birth. We may expect that more funds died in 1998 than in other

years. As a matter of fact, Table 4 shows that in 1998, 179 funds died (29.4% of the 609

dead funds) and 202 funds were born. The total number of dead funds in 1998 was the

highest during the period of 1994 to 1998 while the total number of newborn funds was

the lowest during the same time period. In fact, the fund attrition rate in 1998 was 13%,

much higher than the 8.5% average from 1994 to 1999 (see Table 2). In addition, Table 4

shows that, among the 179 dead funds, 69 died in the first half and 111 died in the second

half. In contrast, among the 202 newborn funds, 117 were born in the first half versus 85

in the second half of 1998.




                                            8
   Fee Changes and Fund Performance.             Once determined, hedge funds seldom

change their fee structures. The median management fee is 1% of fund assets and the

median incentive fee is 20% of profits (see Ackermann, McEnally, and Ravenscraft

(1999) and Liang (1999)). In general, the management fees and incentive fees are very

stable over time for hedge funds. For example, for the 2,016 funds we studied, there are

only 12 funds that changed management and/or incentive fees from 1997 to 1998.8 These

fee changes and the corresponding fund performance in 1998 are reported in Table 5.

Among the 12 funds, there are 8 funds that dropped their management fees and 4 funds

that increased their management fees. In contrast, there are 7 funds that dropped incentive

fees while 5 increased incentive fees. The average management fee change for these 12

funds in 1998 was –0.67%, compared to the average incentive fee change of –5.13%.

Therefore, both the magnitude of fee reduction and the number of funds that reduce their

fees are higher than those funds with increased fees.

   For the 7 funds with reduced incentive fees, the average monthly return in 1998 is

–1.67%, compared with 0.90% for the 5 funds with increased incentive fees. More

interestingly, the 7 funds with negative fee changes in 1998 had an average monthly

return of 2.14% in 1997 while the 5 funds with positive fee changes in 1998 had an

average return of 1.74% in the previous year. The funds with reducing fees performed

very poorly in 1998 and were substantially below their performance in 1997. It seems

that poor performance is a major factor for a fund to reduce its incentive fee. This is

consistent with the design of incentive fees. The result also suggests that a fund changes

its incentive fee not only based on the current year’s performance but also based on last

year’s performance.




                                             9
Conclusion

   Using a large sample, we investigate the hedge fund performance and risk over a 10-

year period from 1990 to 1999. In particular, we examine the year 1998 to see the impact

of the global financial crisis on the hedge fund industry. We also study the survivorship

bias issue for hedge funds.

   Hedge funds enjoy sizable returns during this 10-year bull market period. The

average annualized return is 14.2% for all hedge funds, compared to 18.8% of the S&P

500 index. Although the total return for the index is higher than that of the hedge funds,

hedge funds as a group are much less volatile than the index due to cross-style

diversification, dynamic hedging, cross-border investing, and varieties of non-traditional

financial instruments used. During this time period, hedge funds have a Sharpe ratio of

0.41, much higher than 0.27 for the index. It seems that hedge funds offer better risk-

return trade-off than pure equity trading strategies. Empirical results show that the

average survivorship bias for hedge fund returns is 2.4% per year.

   Hedge funds as a whole are severely affected by the economic crisis in 1998. There

are more funds that died than in any other year, especially in the second half of 1998

when Russia defaulted its debt. In 1998, the number of dead funds is the highest and the

number of newborn funds is the lowest since 1993. 1998 also has the highest volatility in

hedge fund returns. In general, there are very few funds that change incentive fees and

management fees. Changes in fees are performance-related: poorly performed funds in

1998 dropped their incentive fees. This is consistent with the design of incentive fees.




                                             10
   I would like to thank H. Gifford Fong (the editor) for his comments. The paper is

supported by a research grant from the Weatherhead School of Management at Case

Western Reserve University. I am grateful to TASS Management Limited for providing

the data.




                                        11
References

Ackermann, C., R. McEnally, and D. Ravenscraft, 1999. “The Performance of Hedge

   Funds: Risk, Return and Incentives.” Journal of Finance, vol. 54, no. 3 (June):833-

   874.

Brown, S. J., W. N. Goetzmann, and R. G. Ibbotson, 1999. “Offshore Hedge Funds:

   Survival & Performance 1989-95.” Journal of Business, vol. 72, no. 1 (January):91-

   117.

Brown, S. J., W. N. Goetzmann, R. G. Ibbotson, and S. A. Ross, 1992. “Survivorship

   Bias in Performance Studies.” Review of Financial Studies, vol. 5, no. 4

   (Winter):553-580.

Brown, S. J., W. N. Goetzmann, and J. Park, 1999. “Conditions for Survival: Changing

   Risk and the Performance of Hedge Fund Managers and CTAs.” Forthcoming,

   Journal of Finance.

Fung, W., and D. A. Hsieh, 1998. “Performance Characteristics of Hedge Funds and

   CTA Funds: Natural Versus Spurious Biases.” Forthcoming, Journal of Financial

   and Quantitative Analysis.

Fung, W., and D. A. Hsieh, 1997a. “Empirical Characteristics of Dynamic Trading

   Strategies: The Case of Hedge Funds.” The Review of Financial Studies, vol. 10, no.

   2 (Summer): 275-302.

Fung, W., and D. A. Hsieh, 1997b. “Survivorship Bias and Investment Style in the

   Returns of CTAs.” The Journal of Portfolio Management, vol. 24, no. 1 (Fall):30-41.

Liang, B., 1999. “On the Performance of Hedge Funds.” Financial Analysts Journal, vol.

   55, no. 4 (July/August):72-85.




                                          12
Liang, B., 2000. “Hedge Funds: The Living and the Dead.” Forthcoming, Journal of

   Financial and Quantitative Analysis.

Malkiel, B. G., 1995. “Returns from Investing in Equity Mutual Funds, 1971 to 1991.”

   Journal of Finance, vol. 50, no. 2 (June):549-572.




                                           13
Note:

1. The other big name is Van Hedge Fund Advisors, but their data is not available to

   academics. In general, hedge funds report to data vendors voluntarily. Because hedge

   funds are not allowed to advertise to the public, hedge funds view this voluntary

   reporting as a way to distribute their fund information and attract investors for more

   assets.

2. Although the styles are overlapping, all live funds and dead funds are classified under

   these style definitions. Following these definitions, we can minimize the survivorship

   bias problem.

3. Hedge funds as a group have low volatility since different investment strategies are

   less correlated, but individual styles or individual hedge funds may still be very

   volatile, depending on what investment strategy and what financial instruments are

   used.

4. Liang (1999, 2000) indicates that offshore funds are riskier than onshore (US) funds.

5. One may argue that hedge funds voluntarily stop reporting to data vendors because

   they are doing so well and do not need any more investors. This explanation is

   unlikely the major reason for a fund’s disappearance because disappeared funds

   significantly underperform the survived funds.

6. The 1.08% return for all funds is slightly different from 1.11% in Table 1 because the

   investment styles in Table 1 are overlapping.




                                           14
7. A haircut is the difference between the market value of an asset posted as collateral

   and the value attributed to such an asset by a lender in determining whether the

   collateral has been met.

8. Funds that died before 1997 may have also changed fees but we don’t have the fee

   information to evaluate them.




                                          15
                                         Table 1. Hedge Fund Returns by Investment Strategies: 1990-1999
Strategy       1990              1991            1992           1993           1994            1995          1996           1997           1998         1999*           Avg.
Top down macro 0.85       1.21   1.99     2.12    1.22   1.67   2.71    1.68   -0.21    1.87   1.05   1.43   1.43    1.98    1.45   2.68   -0.38   3.62 1.77     2.35    1.19
Bottom up       0.58      2.03       2    1.75    1.21   1.63   2.35    1.23   -0.02    1.68   1.32   1.25   1.57    1.71    1.48   2.56    0.28   3.89 2.17     2.42    1.29
Short selling   1.01      0.99   1.58     1.12    1.26   1.09   1.92    1.06     0.1    0.94   1.32   0.96   1.31    1.37    1.39   2.04    0.54   2.81 1.74     1.96    1.22
Long bias       0.46      2.28     2.2    2.01    1.32   1.92   2.34    1.39   -0.12    1.93   1.25    1.4   1.48    1.86    1.55   2.78     0.1   4.49 2.34     2.81    1.29
Market neutral  0.73      1.05     1.4    0.79    1.07   0.75   1.71    0.76    0.14    0.71   1.08   0.53   1.08    0.72     1.1   0.82    0.37   1.46 1.17     0.64    0.99
Opportunities   0.93      1.53   2.01     1.73     1.3   1.53     2.3   1.26    0.09    1.38   1.37   1.14   1.58    1.74    1.51   2.31    0.11   3.36 2.00     2.28    1.32
Relative value  0.63      1.23   1.58     1.88    1.11    0.9   2.31    1.21   -0.04    1.33   1.09   0.93     1.3   1.07    1.23   1.58   -0.01   2.72 1.64     1.69    1.08
Arbitrage       0.97       0.8   1.24     1.09    1.16   0.98   2.17    0.95    0.05    0.92   1.23   0.75   1.21    0.88    1.27   1.13     0.1   2.03 1.47     1.02    1.09
Discretionary   1.38      1.45   1.27     3.35    0.84   1.62   1.86    1.61    0.11    0.91   1.07    1.2   1.05    1.74    1.09   2.21    0.02    1.8 1.26     1.67    1.00
Trend Follower  2.63      3.62   1.35     6.02    0.53   4.27   1.55    2.42   -0.18    2.06   1.32   2.68   1.03    3.22     0.9   2.95    0.84   2.28 0.34     2.11    1.03
Technical          2      2.45   1.16     4.28    0.52   2.52   1.59    1.81   -0.11    0.29   1.11   1.77   1.01    2.38       1   2.37    0.65   1.31 0.64     1.76    0.96
Fundamental     1.13      0.91   1.65     1.99    1.03   0.95   1.99    1.16   -0.02    1.18   1.13   0.97   1.31    1.58    1.34    2.4    0.11   3.00    1.7   2.08    1.14
Systematic      2.28      2.91   1.32     4.88     0.6   3.09   1.56    1.89   -0.38    1.53    1.1   1.96   1.07    2.66    1.05   2.63    0.57    1.4 0.46     1.94    0.96
Diverse          1.6      1.13   1.32     2.46    0.91   1.02     2.2   1.46   -0.16     1.1   1.11   1.38   1.33    2.14    1.18   2.51    0.15   1.86    1.1    1.8    1.07
Other           1.04      0.77   0.92      0.8    1.14    0.9   1.65    0.87       0    0.79   1.13    1.2   1.18    1.01    1.13   1.86     0.3   2.46 1.67     1.96    1.02

Average            1.21 1.62 1.53 2.42 1.01 1.66 2.01 1.38 -0.05 1.24 1.18 1.30 1.26 1.74 1.24 2.19 0.25 2.57 1.43 1.90 1.11
Note: The TASS data contains 2,016 funds, including 1,407 live funds and 609 dead funds. Calculations are based on 1,921 funds with returns net of all fee s and
with monthly returns only.
*As of July 1999.




                                                                                   14
                       Table 2. Hedge Fund Attrition Rate: 1993-1999
          Year              Start       Funds born        Funds died          Year end    Attrition Rate
                                                                                               (%)
      Pre-1993                                                                     540
          1993               540               234                                 774
          1994               774               242                32               984              4.13
          1995               984               229                77             1,136              7.83
          1996             1,136               245               142             1,239             12.50
          1997             1,239               262               124             1,377             10.01
          1998             1,377               202               179             1,400             13.00
         1999*             1,400                60                53             1,407              3.79

           Total                             2,014**            607**                                 8.54
Note: The attrition rate the defined as the ratio of the number of funds died during a year to the number of
funds at the beginning of the year.
*Through July 1999
**Two funds with unknown birth date and death date




                                                     15
                       Table 3. Hedge Fund Survivorship Bias: 1990-1999
            Year    S&P500 (%)       All funds (%)    Live funds (%)    Dead funds (%)        Bias (%)
            1990      -0.14                1.47            1.32              1.80              -0.15
                      (5.31)             (1.13)           (1.03)            (1.96)
            1991       2.34                1.53            1.70              1.15               0.18
                      (4.56)             (2.50)           (2.45)            (2.70)
            1992       0.64                0.91            0.94              0.87               0.03
                      (2.13)             (1.04)           (1.01)            (1.30)
            1993       0.81                1.81            2.05              1.45               0.24
                      (1.77)             (1.25)           (1.40)            (1.19)
            1994       0.15               -0.09            0.02             -0.25               0.11
                      (3.04)             (0.92)           (1.03)            (0.83)
            1995       2.70                1.14            1.49              0.53               0.36
                      (1.50)             (1.10)           (1.23)            (1.04)
            1996       1.79                1.26            1.52              0.65               0.26
                      (3.15)             (1.60)           (1.71)            (1.40)
            1997       2.52                1.26            1.41              0.62               0.16
                      (4.60)             (2.12)           (2.12)            (2.22)
            1998       2.30                0.25            0.44             -1.31               0.19
                      (6.21)             (2.13)           (2.13)            (2.16)
          1999*        1.29                1.44            1.47             -0.43               0.03
                      (3.93)             (1.80)           (1.80)            (1.88)

         Average         1.45             1.08**             1.22              0.55             0.14
Note: The survivorship bias is calculated as the monthly return difference between survived funds and all
funds.
*Till July 1999
**Slightly different from 1.11% in Table 1 because the styles are overlapping in Table 1.




                                                     16
Table 4. Numbers of Funds that Were
Born and Died in 1998
   Month        Birth        Death
   9801          36            6
   9802          15            8
   9803          16           13
   9804          19           13
   9805          16            8
   9806          15           20
   9807          32           29
   9808          20           18
   9809           9           17
   9810           2           14
   9811           8           11
   9812          14           22

   Total         202          179




                  17
   Table 5. Fund Fee Changes and Performance in 1998
       Fund         1998 return (%)        Incentive fee Management fee
                                            change (%)     change (%)
        1                      -1.91                 -20.0          0.50
        2                      -0.57                 -20.0         -2.75
        3                       2.17                   2.5          1.00
        4                      -8.95                  -5.0         -0.50
        5                       0.18                  20.0          1.00
        6                      -0.53                 -20.0         -1.50
        7                       0.62                  -5.0         -0.50
        8                       1.01                   5.0         -0.10
        9                       0.16                  20.0          1.00
        10                      0.96                   5.0         -1.00
        11                      0.05                 -24.0         -3.50
        12                     -0.39                 -20.0         -1.70

      Average                   -0.60              -5.13             -0.67
    Note: If a fund changes fees, it usually occurs at the year-end. Therefore
the fee change here is from the end of 1997 to the end of 1998. The
difference is the fee in 1998 minus the fee in 1997.




                                      18
                               Cumulative return




                   0
                       1
                           2
                                          3
                                                            4
                                                                5
                                                                    6
            date
            9004
            9008
            9012
            9104
            9108
            9112
            9204
            9208
            9212
            9304
            9308
            9312
            9404
            9408




     Date
            9412




19
            9504
            9508
            9512
            9604
            9608
            9612
                                                                        Figure 1. Hedge Fund Returns versus S&P 500




            9704
            9708
            9712
            9804
            9808
            9812
            9904
                                              All
                                       Live
                                Dead
                                                    sp500
                                      Figure 2. Hedge Fund Returns in 1998

        4


        3


        2


        1


        0
             Jan   Feb   M ar   Apr     M ay     Jun      Jul   Aug     Sep   Oct   Nov   Dec
Month




        -1


        -2


        -3


        -4


        -5


        -6
                                                  Return (%)




                                                          20

								
To top