Side-by-Side Management of Hedge Funds and Mutual Funds - PowerPoint PPT Presentation

1 / 29
About This Presentation
Title:

Side-by-Side Management of Hedge Funds and Mutual Funds

Description:

Side-by-Side Management of Hedge Funds and Mutual Funds Tom Nohel, Loyola University Z. Jay Wang, University of Illinois Lu Zheng, University of California at Irvine – PowerPoint PPT presentation

Number of Views:278
Avg rating:3.0/5.0
Slides: 30
Provided by: Information284
Category:

less

Transcript and Presenter's Notes

Title: Side-by-Side Management of Hedge Funds and Mutual Funds


1
Side-by-Side Management of Hedge Funds and Mutual
Funds
  • Tom Nohel, Loyola University
  • Z. Jay Wang, University of Illinois
  • Lu Zheng, University of California at Irvine

2
Introduction
  • In recent years the mutual fund scandal has
    been much in the news
  • The focus has been on stale trades, lack of
    independent directors of mutual funds, and
    conflicts of interest in general
  • At the same time, the scrutiny leveled on the
    mutual fund industry is starting to shine a
    spotlight on the hedge fund industry, especially
    given the tremendous growth in hedge fund assets

3
Policy Concerns
  • Among the concerns of the SEC and legislators in
    Washington is the practice of having the same
    individual(s) manage a mutual fund and a hedge
    fund --- side-by-side management
  • Wellington manages the 18 billion Vanguard
    Healthcare fund and offers a healthcare hedge
    fund managed by the same person (Edward P. Owens)
  • Several well-known mutual fund companies have
    such arrangements
  • e.g., Alliance Capital, Invesco, American Express
  • Others forbid the practice due to fears of
    conflicts of interest
  • Fidelity

4
Potential Conflicts in Side-by-Side Arrangement
  • Differing compensation structures between mutual
    funds and hedge funds create potential conflicts
  • Mutual fund managers 1 of assets
  • Hedge fund managers 1-2 of assets a
    performance fee (typically 20 of profits)
  • Thus manager has an incentive to benefit his
    hedge fund at the expense of mutual fund
    investors

5
Manifestation of the Conflicts
  • Front-running of mutual fund trades by hedge
    funds
  • Allocating under-priced IPO shares
    disproportionately to hedge funds
  • Stale trades/Timing of mutual fund shares
  • It is often hedge funds that benefited

6
Manifestation of the Conflicts (cont.)
  • Cherry-picking
  • Allocating trades with favorable subsequent price
    movements to hedge funds
  • Allocation of bunched trades
  • Allocating to hedge funds shares bought (sold) at
    the lowest (highest) price
  • Allocation of trade commissions and soft dollars

7
Policy Debate
  • Due to the potential conflicts of interest, some
    argue to ban side-by-side arrangements
  • Fund companies that allow this practice argue
    that without the lure of the possibility to run a
    hedge fund the best managers will leave
  • Ted Truscott of American Express is quoted as
    saying To attract the best and brightest, we
    have to offer the opportunity for side-by-side
    management (WSJ, 2004)

8
Our Attempt to Inform This Debate
  • Document the status and evolution of side-by-side
    arrangement
  • Analyze the welfare consequences of side-by-side
    arrangement
  • Test for abnormal performance on mutual fund side
  • Test for abnormal performance on hedge fund side
  • Conflicts of interest would suggest that SBS
    mutual funds would under-perform peers, or SBS
    hedge funds should outperform peers (or both)

9
Literature on Money Managers and Incentives
  • There is a sizeable literature showing that
    mutual fund managers respond to incentives
  • Chevalier and Ellison (1997)
  • Brown, Harlow, and Starks (1996)
  • Nanda, Wang, and Zheng (2004)
  • Gaspar, Massa, and Matos (2004)
  • Reuter (2004)
  • Massa, Reuter, and Zitzewitz (2006)
  • Compensation structures encourage strategic
    allocation of returns between mutual and hedge
    funds by side-by-side managers

10
Data Summary Statistics
  • Data sources
  • CRSP Mutual Fund Database
  • TASS Hedge Fund Database from Tremont
  • Construct a unique data set of side-by-side funds
    by matching the managers names from the two
    databases.
  • A total of 112 side-by-side managers who manage
    304 mutual funds and 207 hedge funds
    simultaneously.
  • This covers the period 1990-2005

11
Time Trend of Side-by-Side Management (When the
SBS arrangement began)
Year Number of Mutual Funds Number of Fund Managers
1990 3 0
1991 5 1
1992 5 3
1993 12 3
1994 11 6
1995 11 7
1996 20 8
1997 27 10
1998 18 6
1999 21 6
2000 44 14
2001 36 17
2002 47 15
2003 25 7
2004 10 5
12
Side-By-Side Management by Investment Objectives
Mutual Funds
Investment Objectives Number of Mutual Funds Number of Fund Managers
Equity USA Growth (GRO) 58 35
Equity USA Small Companies (SCG) 48 29
Equity USA Midcaps (GMC) 25 20
Equity USA Aggressive Growth (AGG) 21 15
Equity USA Growth Income (GRI) 20 14
13
Side-By-Side Management by Investment Objectives
Hedge Funds
Investment Objectives Number of Hedge Funds Number of Fund Managers
Long/Short 116 71
Event Driven (Merger Arb) 15 9
Equity Market Neutral 13 9
Fund of Funds 11 4
Emerging Market 10 4
14
Summary Stat by Investment Objectives Mutual
Funds
Objective Objective TNA (Mil) Turnover () Expenses () Mgmt Fee ()
Objective Objective Mean Mean Mean Mean
GRO Side-by-Side 1183.24 57.56 1.51 0.93
All 1228.13 83.80 1.33 0.73
SCG Side-by-Side 719.09 80.60 1.63 0.99
All 698.12 98.47 1.43 0.84
GMC Side-by-Side 137.03 159.70 1.70 0.88
All 743.28 121.52 1.52 0.75
AGG Side-by-Side 266.91 208.24 2.44 1.17
All 774.05 267.77 1.91 0.90
GRI Side-by-Side 1293.37 56.30 1.64 0.76
All 2120.51 58.54 1.03 0.54
15
Summary Stat by Investment Objectives Hedge Funds
Objective Objective TNA (Mil) Mgmt Fee () Incentive Fee () Leverage ()
Objective Objective Mean Mean Mean Mean
Long Side-by-Side 154.75 1.09 18.75 42.40
Short All 271.10 1.29 19.12 39.64
Event Side-by-Side 62.61 1.21 18.73 73.96
Driven All 481.20 1.37 18.83 55.24
Market Side-by-Side 122.64 1.12 20.00 53.09
Neutral All 199.10 1.38 19.53 75.06
Fund of Side-by-Side 108.69 1.60 3.64 0.00
Funds All 351.77 1.44 9.15 24.03
Emerg. Side-by-Side 386.38 1.62 13.50 0.00
Markets All 190.09 1.53 17.58 26.51
16
Summarizing
  • Side-by-side management experienced some rapid
    growth in the late 1990s and early 2000s, and
    slowed down after 2002
  • 123 billion under management as of 2004 (in
    mutual funds)
  • Most side-by-side mutual funds are growth
    oriented US equity funds
  • Side-by-side funds have significantly higher
    expense ratios and management fees than their
    peers
  • Side-by-side hedge funds look like their peers,
    though in general smaller

17
Performance Tests Side-by-Side Funds vs. Peer
Funds (Mutual Funds)
  • Side-by-Side Funds
  • Sharpe ratios and 4-factor alphas are estimated
    over the entire side-by-side period based on
    monthly observations (a period of no less than
    two years).
  • 235 Mutual funds have enough data to be included
  • Peer Funds
  • Average Sharpe Ratios and 4-factor as during the
    side-by-side period are estimated for funds with
    the same investment objective but w/o
    side-by-side arrangement.

18
Side-by-Side Funds vs. Peer Funds (Mutual Funds)
Sharpe Ratio Sharpe Ratio 4-Factor Alpha 4-Factor Alpha
Mean Median Mean Median
Side-by-Side Funds 0.139 (0.00) 0.155 (0.00) 0.044 (0.22) 0.033 (0.31)
Funds with Same Investment Objectives 0.108 (0.00) 0.103 (0.00) -0.087 (0.00) -0.085 (0.00)
Performance Difference (Average across funds) 0.032 (0.00) 0.040 (0.00) 0.132 (0.00) 0.080 (0.00)
Performance Difference (average across managers) 0.026 (0.00) 0.033 (0.00) 0.076 (0.04) 0.045 (0.06)
19
Portfolio Approach (Mutual Funds)
Alpha Market SMB HML MOM
Side-by-Side 0.072 (0.32) 0.783 (0.00) 0.197 (0.00) 0.144 (0.00) 0.024 (0.10)
No side-by-side -0.077 (0.08) 0.710 (0.00) 0.084 (0.00) 0.074 (0.00) 0.016 (0.06)
Difference 0.148 (0.00) 0.073 (0.00) 0.113 (0.00) 0.070 (0.00) 0.008 (0.45)
20
Summary and Interpretation
  • Side-by-side funds significantly outperform peer
    funds
  • A bit over 1.5 in 4-factor alpha on an annual
    basis.
  • Applied to the 123 billion under management in
    2004, this translates to 2 billion!
  • The superior performance of side-by-side funds is
    consistent with the existence of a selection bias
  • Side-by-side arrangement is rewarded to better
    skilled managers for superior performance
  • However, this does not preclude the existence of
    conflicts of interest

21
Does The Side-by-Side Relationship Curtail
Performance?
  • We run a pooled regression with four factor alpha
    as the dependent variable that allows us to
    control for fund size, family size, expenses,
    turnover, and fixed effects for style and time
  • We also include a side-by-side dummy and a pre
    SBS dummy
  • These dummies allow us to compare the performance
    of our side-by-side managers before the SBS
    relationship and while the SBS relationship was
    in place

22
Performance of Side-by-Side Funds Regression
Approach
Four-Factor Adjusted Return (t) Four-Factor Adjusted Return (t)
All Funds Equity Funds
Side-by-side Indicator 0.136 (0.00) 0.154 (0.00)
Pre-Side-by-side Indicator 0.125 (0.00) 0.106 (0.00)
Log Fund TNA (t-1) -0.002 (0.87) -0.010 (0.55)
Log Family TNA (t-1) 0.004 (0.52) 0.006 (0.36)
Log Fund Age (t-12) -0.016 (0.47) -0.018 (0.46)
Expenses (t-12) -8.892 (0.00) -9.382 (0.00)
Time fixed-effect Included Included
Style fixed-effect Included Included
Observations 519,872 385,136
23
Does The Side-by-Side Relationship Curtail
Performance? NO!
  • Our SBS managers appear to be star performers
    prior to the SBS relationship
  • Moreover, if anything, their performance is even
    better once the SBS relationship is in place
  • These results are consistent with the industry
    explanation that SBS privileges are granted to
    the best managers for purposes of retention

24
Hedge Fund Performance
  • We test for abnormal performance on the hedge
    fund side
  • Similar to Mutual Fund tests except that we use
    6-factor alpha to account for left tail risk in
    hedge funds
  • Conflicts of interest should lead to superior
    performance on the hedge fund side
  • Concern over selection bias in hedge fund data
  • Hedge fund data is reported voluntarily, implying
    that managers with poor track records may not
    want to report performance

25
Performance of Hedge Funds with Side-by-Side
Arrangements
Sharpe Ratio Sharpe Ratio 6-Factor Alpha 6-Factor Alpha
Mean Median Mean Median
Side-by-Side Funds 0.182 (0.00) 0.174 (0.00) 0.222 (0.08) 0.167 (0.02)
Funds with Same Investment Objectives 0.236 (0.00) 0.204 (0.00) 0.306 (0.00) 0.330 (0.00)
Performance Difference (Average across funds) -0.054 (0.00) -0.060 (0.00) -0.085 (0.50) -0.106 (0.13)
Performance Difference (Average across managers) -0.053 (0.00) -0.054 (0.00) -0.159 (0.28) -0.059 (0.27)
26
Portfolio Approach (6-factor alpha)
Alpha Market SMB HML MOM
Side-by-Side 0.232 (0.12) 0.486 (0.00) 0.132 (0.00) 0.157 (0.00) 0.106 (0.00)
No side-by-side 0.352 (0.00) 0.295 (0.00) 0.080 (0.00) 0.069 (0.04) 0.076 (0.00)
Difference -0.12 (0.23) 0.191 (0.00) 0.052 (0.02) 0.087 (0.00) 0.030 (0.03)
27
Robustness
  • We eliminate all funds from the peer group that
    CRSP lists as team managed. The results are
    unchanged
  • Fama-MacBeth controlling for past a results are
    qualitatively similar
  • We also run our tests in terms of return gaps,
    and again the results are similar

28
Summary
  • Side-by-side mutual funds significantly
    outperform their peer funds
  • Consistent with self selection Better ability
    managers are managing side-by-side funds.
  • Pooled regression shows evidence consistent with
    high ability managers continuing to outperform
    their peers
  • On the hedge fund side, side-by-side managers are
    at best on par with peers, further weakening the
    case for presence of conflicts of interest
  • Despite potential conflicts of interest in
    side-by-side relationship, no evidence so far
    suggesting a significant loss in investor welfare

29
Limitations and Future Research
  • Incomplete identification of side-by-side funds
  • Only used one hedge fund data base TASS
  • Will include additional data HFR
  • Impact of side-by-side arrangement on the exit
    rate of mutual fund managers
  • Factors affecting the establishment of
    side-by-side arrangement (Retention?)
Write a Comment
User Comments (0)
About PowerShow.com