Administrative Choice: Mutual Funds and the Performance of 401(k) Plans. Martin J. Gruber June Maier. Plan

Transcription

1 Administrative Choice: Mutual Funds and the Performance of 401(k) Plans Martin J. Gruber June 2012 Maier Plan We will first examine some facts about the performance of mutual funds We will examine how well plan administrators do in selecting mutual funds for their participants and how well participants do in allocating their pension assets among funds 1

2 Academic Credentials Professor of Finance at NYU since 1965 (Scholar in Residence since 2010) Past President of the American Finance Association Co-author of Modern Portfolio Theory and Investment Analysis (8 th edition 2009) Co-author of over 100 articles on Portfolio Management, Mutual Funds, and Pension Funds Professional Credentials Consultant on Mutual Funds Member of Board of Directors National Bureau of Economics Research (NBER) Daiwa Closed End Funds Member of the executive committee and the Investment Committee of the NBER Formerly Member of Board of TIAA Chairman of Board of CREF DWS (Chairman of Equity Committee) Member of Board of SG Cowen Funds 2

3 U.S. Retirement Market (2011) 17.5 Trillion Dollars 4.7 Trillion a 25% in Employee Sponsored Defined Contribution Plans DC plans had 2.5 trillion in mutual funds, 53% of their assets IRAs 48% in mutual funds 3

4 Growth of U.S. Mutual Funds by Type (Market Value) Year Money Market Bond & Income Equity Total Number of Mutual Funds Year 8 4

5 Number of Mutual Fund Complexes Over 650 in 2010 Top Ten Companies Hold 56% of Assets Top 25 Complexes 76% of Assets Percent of U.S. Households Owning Mutual Funds 60% 50% 40% 30% 20% 10% 6% 11% 12% 37% 20% 24% 25% 27% 31% 44% 49% 50% 48% 49% 46% 43% 0% % only invest in employee sponsored plan 37% both inward and outward 31% only outward employee invested plan 5

6 Channels Used for Mutual Fund Investments (Percent of Respondants) 40% 35% 30% 25% 20% 15% 10% 5% 0% Full Service Broker Discount Broker Insurance Agent Bank Representative Independent Direct Marketer Financial Planner 25% 13% 7% 15% 22% 19% *1/3 Online 2/3 Office # Funds Expense Ratio Total Load 12B-1 Fees All Equities No Load Front Load Deferred Load All Bonds No Load Front Load Deferred Load

7 Performance of Mutual Funds Return Per Year Years Manager A Manager B Manager C % 0.6% 0.8% % 37.4% 36.1% % 18.8% 19.2% % 0.6% 1.0% % 24.5% 24.4% % 10.9% 11.1% % 16.4% 16.2% % 3.6% 3.6% S&P 500 Index Small Stocks Index Small Stocks Manager 7

8 Morningstar Categories Investment Style Large Growth Medium Growth Small Growth Large Blend Medium Blend Small Blend Large Value Medium Value Small Value Morningstar Categories Investment Style Growth: (relative P/E + relative P/B) greater than 2.25 Blend: (relative P/E + relative P/B) greater than or equal to 1.75 but less than or equal to 2.25 Value: (relative P/E + relative P/B) less than 1.75 Large: geometric mean market capitalization of greater than $9.1 bln Medium: geometric mean market capitalization of greater than or equal to $1.6 bln but less than or equal to $9.1 bln Small: geometric mean market capitalization of less than $1.6 bln 8

9 α = average annual return above that of index funds with the same risk (of the same type) Average annual α = -.65% Average of expenses = 1.35% Average α for non-surviving funds= % Future Performance Ranked By Previous Performance Alpha Decile Alpha Next Year Worst 10% Best 10% Spearman Rank * * Top Bottom * Alpha = average return above that of index funds with the same risk * Significant at 1% level ** Significant at 5% level 9

10 Ranked By Previous Expense Ratio Decile Alpha Next Year Highest 10% Lowest 10% Spearman Rank Low High * * Significant at 1% level Future Three Year Performance Predicted By Previous Three Years Decile Alpha Next 3 Years Worst 10% Best 10% Spearman Rank * Top Bottom * * Significant at 1% level 10

11 Annual Cash Flows for Deciles Formed on the Basis of Previous Years Performance in the Year Following Formation % Change in Cash Decile Flow next year Worst 10% Best 10% 29.0 Spearman Rank 1.00 * Top Bottom * * Significant at 1% level Annual Realized Cash Flow Weighted Alpha 1 Year Positive Negative Weighted Average

12 1. Index funds outperform actively managed funds of the same risk Active funds 65b.p. below indexes. Charge fee of 130 b.p. Bring information worth 65b.p. but charge 130b.p. 2. Funds that charge higher fees tend to do worse that funds that charge lower fees 3. Funds that perform well have a slight tendency to continue to perform well Funds that perform really badly tend to continue to perform really badly 4. Investors seem to be aware of this. Funds that perform well have higher subsequent cash flows. Funds that perform badly have high cash out flows. 5. The cash flow in and out of funds has positive alpha greater than 20 b.p. 12

13 Pension Plans There has been a lot of research on how participants behave. Surprisingly, there has been almost no research on the actions of plan administrators on the choices given to plan participants. The action of a participant is a result of two decisions: the choice the participant is offered and how he or she allocates among these choices. This is the first set of research to examine the appropriateness of the choices given to the participant: an examination of the decisions of the plan administrators rather than the participant. Do administrators choose the right funds? 13

14 Return Based Style Analysis α = how much better did a manager due than holding a combination of indexes with the same risk 27 Methodology A. Alpha R it R rt = α i + β ij I jt + e it Stock Funds: S&P 500, Fama French Small-Large and high minus low, Lehman Gov/Credit, and MSCI Europe Bond Funds: Lehman Gov/Credit, Lehman Mortgage- Backed, Credit Suisse High-Yield Index, Salomon non-dollar World Gov. Bond Index International: S&P 500 and the three MSCI Indexes (Europe, Pacific, and Emerging Markets 28 14

15 B. Differential Alpha Mutual funds, in general, have negative alpha. We took the alpha for each mutual fund minus the average alpha for funds of the same general size from the same ICDI category. 29 Performance of Administrators α α = average annual return above that of index funds with the same risk Differential α = annual return above that of the average active manager Equal Wts. Alpha Diff. α Average P-Value Fee difference

16 Given the type of fund offered, administrators tend to hold better than random funds, but much of the difference is due to lower expense ratios. They do not do better than holding index funds Do participants allocate savings wisely given the choices they are offered? 16

17 Performance of Administrator and Participant α Equal Wts. Participant Wts. Alpha Diff. α Alpha Diff. α Average P-Value The choices of plan administrator is very important because participants destroy value rather than increases value by their decisions

18 Does the past performance of administrators tell you anything about their future performance? Past Performance Quartiles Average Future Differential Alpha 1 (lowest) (highest)

19 Plan administrators who outperform in the past have a tendency to outperform in the future. Plan administrators who underperform in the past have a strong tendency to underperform in the future. 37 Conclusions 1. Actions of plan administrators vitally important 2. Offer portfolios of better active funds 3. Nevertheless, don t outperform index funds 4. Additions and deletions not as skillful 5. Some do better than others over time 6. Participants don t add value to the decisions of the plan administrators 19

20 Performance of Added minus dropped funds - annual α Alpha Differential Alpha Before P-Value After *not significant at 1% level 20