Pension Plans vs. Mutual Funds: Is the Client Victim or Culprit?

See Bogle John C. Twardowski Jan M. , “Institutional Investment and Performance Compared,” Financial Analysts Journal (January/February 1980), pp. 33–41; Schotland Roy A. , “Why Mutual Funds are Top Performers,” Pension and Investment Age, July 20, 1981, p. 13; Munnell Alicia H. , “Who Should Manage the Assets of Collectively Bargained Pension Plans?,” New England Economic Review (July/August 1983), pp. 18–30.

Monks Robert A.G. , “How to Earn More on $1 Trillion,” Fortune, September 2, 1985, pp. 98–99.

Op. cit., footnote 1; LeBaron Dean , “Reflections on Market Inefficiencies,” speech, Institute for Quantitative Research in Finance, Scottsdale, Arizona, October 10, 1982; and Heard James E. , “Pension Funds and Contests for Corporate Control,” California Management Review (Winter 1987), pp. 89–100.

See, for example, Ambachtsheer Keith , Pension Funds and the Bottom Line (Homewood: IL: Dow-Jones Irwin, 1986); Copeland Thomas E. , “An Economic Approach to Pension Fund Management,” in Stern J. M. Chew D. H. Jr. , eds., The Revolution in Corporate Finance (New York, NY: Basil Blackwell, Inc. 1986), pp. 282–296.

The data were purchased under DOL Contract Number J-9-P-4-0109.

For a definition, see Brealy Richard Myers Stewart , Principles of Corporate Finance, 2nd ed. (New York, NY: McGraw Hill, 1984), pp. 770–771.

These data were also purchased under DOL Contract Number J-9-P-4-0109.

Sharpe William F. , “Mutual Fund Performance,” Journal of Business (January 1966), pp. 119–138.

Recent work has called into question the ability to make appropriate adjustments for risk without collecting far more information than current data bases contain. For instance, Jobson J.D. Korkie Bob , “Potential Performance and Tests of Portfolio Efficiency,” Journal of Financial Economics (December 1982), pp. 433–466. Jobson and Korkie suggest an approach to overcome problems with traditional measures that requires knowledge of historical return experience for a portfolio's component securities. Given most managed portfolios have 100 to 200 securities in them and portfolio turnover of nearly 100 percent per year, this means a huge number of stocks must be tracked. Further, others argue that appropriate risk adjustment for performance measurement can only be done using portfolio managers' own forecasts. See Admati Amat R. Ross Stephen A. , “Measuring Investment Performance in a Rational Expectations Equilibrium Model,” Journal of Business (January 1985), pp. 1–26.

Because Sharpe performance measures are correlated across portfolios, we had to make adjustments to the standard statistical tests, namely t-tests. The adjustments made followed those developed by Jobson J.D. Korkie Bob M. , “Performance Hypothesis Testing with Sharpe and Treynor Measures,” Journal of Finance (September 1981), pp. 889–908. We assumed a correlation of .75 between the groups. This more sophisticated analysis strengthens the conclusion.

See Jensen Michael C. , “The Performance of Mutual Funds in the Period of 1945–1964,” Journal of Finance (May 1968), pp. 389–416; “Risk, the Pricing of Capital Assets, and the Evaluation of Investment Portfolios,” Journal of Business (April 1969), pp. 167–247.

See Lehman Bruce N. Modest David M. , “Mutual Fund Performance Evaluation: A Comparison of Benchmarks and Benchmark Comparisons,” Journal of Finance (June 1987), pp. 233–265. They show in a very rigorous way the choice of market proxy and other benchmarks seriously affects the interpretation of performance results and may hamper inference regarding superior or inferior performance.

Shanken Jay , “Multivariate Proxies and Asset Pricing Relations,” Journal of Financial Economics (March 1987), pp. 91–110.

Tinic Seha M. West Richard R. , “Risk, Return, and Equilibrium: A Revisit” Journal of Political Economy (February 1986), pp. 126–147.

We chose to use rate of return as the dependent variable, and standard deviation as an explanatory variable rather than place standard deviation on the dependent variable side of the equation, as would have been the case if we used the Sharpe Measure as the dependent variable. Using the Sharpe Measure as a dependent variable implicitly assumes the coefficient on the risk variable is unity in every year. The way we approach the issue allows the coefficient to take on any value the data suggest. Since there is no theory regarding what value this variable should take, we chose not to impose any restrictions on it.

Logue Dennis E. Rogalski Richard J. , Managing Corporate Pension Plans: The Impacts of Inflation (Washington, D.C.: American Enterprise Institute, 1984).

Henriksson Roy D. “Market Timing and Mutual Fund Performance: An Empirical Investigation,” Journal of Business (January 1984), pp. 73–96; Lehmann Bruce N. Modest David M. , “Mutual Fund Evaluation: A Comparison of Benchmarks and Benchmark Comparisons,” Journal of Finance (June 1987), pp. 233–265.

In a world with a fixed supply of securities and no new infusions of cash, market clearing requires that no aggregate rebalancing take place. In such a world, management commitment change, the first measure noted, is the more appropriate measure. In a world where the supply of new securities changes, and there are new infusions of cash, market clearing requires some rebalancing. For example, corporate debt-to-equity ratios in market value terms have remained fairly constant since the mid-1970s though rates of return on stocks have exceeded those on debt. (See Jensen M.C. , “The Takeover Controversy: Analysis and Evidence”, Midland Corporate Finance Journal, (Summer 1986), pp. 6–32, especially Figure 2 [p. 22].) To have achieved this, companies must have issued more debt than equity in the aggregate. Investors must—since markets cleared—have gone through a continuous process of rebalancing: That is, they adjusted the position of their total portfolios allocated to equities, probably through the purchase of new debt with fresh cash generated by investment income and new investment. To illustrate, suppose pension funds hold $100 worth of equity and $100 worth of debt at the start of a period. Within the period the value of stocks rises to $110 while debt remains at $100. If public companies maintain constant debt-equity ratios, $10 more debt will be sold. This must be bought. If pension funds buy with fresh cash, they move towards their original 50%–50% allocation. If they let their chips ride, they use 55% of their new funds to buy stock. This would not allow the debt markets to clear. In any event, the distinctions between the two measures are empirically trivial.

Unfortunately, we could not obtain as high quality quarterly mutual fund data on asset allocation and the like as we had for pension plans. Thus, our tests simply seek to distinguish better from worse pension plans. Moreover, in some respects it would be unfair to incorporate mutual funds into this part of the analysis. We have been evaluating pension and endowment plans, not individual funds. One may think of plans as comprising many funds, just as one may think of investors holding shares in several different mutual funds. The mutual funds themselves may pursue constant policies, but some investors may do a lot of switching among mutual funds. Many may harm themselves. The same may be true for total plans. Each component fund might follow constant policies, but the fiduciary trustee may switch much money among funds. Accordingly, a plan sponsor or trustee is analogous to the mutual fund investor not the mutual fund. We obviously could not get data on how investors in multiple mutual funds may have helped or hurt their wealth through their own switching behavior.

Arnott Robert D. , “The Pension Sponsor's View of Asset Allocation,” Financial Analysts Journal (September/October 1985), pp. 17–23.

See, for example, Black Fischer , “The Tax Consequences of Long Run Pension Policy,” Financial Analysts Journal (July/August 1980), pp. 25–31; Sharpe William F. , “Corporate Pension Funding Policy,” Journal of Financial Economics (June 1976), pp. 183–193; Ambachtsheer Keith P. , Pension Funds and the Bottom Line (Homewood, IL: Dow-Jones Irwin, 1986). Each has a different prescription, but none urge market timing be attempted.