UNITARY BOARDS AND MUTUAL FUND GOVERNANCE

A unique governance structure for mutual funds is unitary board—one board overseeing all funds in the entire family. We find strong evidence for unitary board as an effective governance mechanism. Funds with unitary boards are associated with lower fees, are more likely to pass the economies of scale benefits to investors, are less likely to be involved in trading scandals, and rank higher on stewardship. In contrast, funds with larger or more independent boards charge higher fees and rank lower on stewardship. Our findings indicate that unitary boards of small size, rather than independent boards, may be more beneficial to fund shareholders.     

SECURITIES INVENTORIES AND EXCESS RETURNS

In this paper, the inventory positions of government security dealers are analyzed for signs of their superior information relative to other participants in the market. Testing the 1966–1980 period on a monthly basis, it does not appear that information on the maturity composition of dealers' positions can be used, either concurrently or when it becomes publicly available, to earn excess returns.     

SIMULATION OF PORTFOLIO RETURNS: VARYING NUMBERS OF SECURITIES AND HOLDING PERIODS

Variability may be measured around the long-run market return by combining the well-developed area of security diversification with the notion of time diversification. This non-traditional concept of diversifiable risk is illustrated with results of a very large scale simulation. The results are very general since the data used in the simulation consist of all firms on the Center for Research in Security Prices (CRSP) monthly return file from January 1926 through December 1977.     

COGNITIVE DISSONANCE AND MUTUAL FUND INVESTORS

We present evidence from questionnaire responses of mutual fund investors about recollections of past fund performance. We find that investor memories exhibit a positive bias, consistent with current psychological models. We find that the degree of bias is conditional upon previous investor choice, a phenomenon related to the well-known theory of cognitive dissonance. Psychological and economic frictions in the mutual fund industry are examined via a cross-sectional study of equity mutual funds. We find an unusually high frequency of poorly performing funds, consistent with investor “inertia.” We also examine the differential responses of investment dollars to past performance, controlling for survivorship. These show that the effect is confined to the top quartile. We find little evidence that the response to poor performance is unusual.     

MUTUAL FUND TRADES: ASYMMETRIC LIQUIDITY PREFERENCES AND FUND PERFORMANCE

We investigate the role of the liquidity of stocks traded by mutual funds on the performance of funds experiencing substantial and sustained redemptions (outflows) or inflows. Accordingly, we identify 770 redeeming fund‐periods and 1,757 inflow fund‐periods and find a statistically significant relation between the liquidity of the stocks they trade and the quantity of the stock traded. Notably, when funds experience redemptions, those with low portfolio liquidity have an elevated preference for selling more‐liquid stocks. In the following period, such funds statistically and economically underperform funds that sell less‐liquid stocks. This is consistent with redemptions detrimentally affecting shareholders that remain in a fund.     

MARKET VOLATILITY AND PERVERSE TIMING PERFORMANCE OF MUTUAL FUND MANAGERS

Prompted by the recent volatility in equity markets, I investigate performance evaluation methods and the mutual fund managers' ability to select undervalued investments and time major market movements during the high-market-volatility period of the 1980s. Specifically, I examine mutual fund managers' stock-selection and market-timing abilities by employing a five-factor risk-adjusted model based on Carhart's four-factor loading model and Bhattacharya and Pfleiderer's quadratic timing model adjusted for perverse timing behavior. Individually, some managers persistently affect fund performance through the selection of undervalued investments, however, at the expense of timing performance. In addition, funds that demonstrate an ability to time major market movements showed persistence in timing performance before and after the October market crash of 1987.     

MARKET TIMING AND MUTUAL FUND PORTFOLIO COMPOSITION

This study uses a methodology that is independent of beta estimates to provide empirical evidence on the success of market timing by mutual fund managers. A fund's success at market timing is evaluated by determining if the percentage of the portfolio invested in stocks increases prior to an upturn in the general level of stock market prices and declines prior to a downturn in the level of stock prices. No evidence is found that managers possess, as a group, any market timing ability.     

A MULTIFACTOR ANALYSIS OF COUNTRY FUND RETURNS

Previous closed‐end country fund research concludes that returns behave more like the U.S. market than like their target markets. We argue this finding may be biased by model misspecification and inappropriate estimation techniques. We propose a single‐equation model containing five hypothesized factors of fund returns. We estimate this model for nineteen pooled seasoned funds using a time‐series cross‐section regression that corrects for two types of autocorrelation. We show that returns are strongly related to target markets. Returns are also related to changes in discounts, exchange rates, and other countries' markets, but are only weakly related to the U.S. market. JEL classification: G10, G12     

THE COST OF MUTUAL FUND DISTRIBUTION FEES

Many individuals purchase shares in mutual funds as investments. With a lack of evidence supporting performance persistence in fund returns, investors should consider expenses as a fund-selection tool since fund expenses have a negative effect on fund returns. One of the largest expenses incurred by fund investors is distribution expenses, which include both load charges and annual fees. Close to two-thirds of all equity funds charge investors for fund distribution. The true cost of these distribution fees to investors is hard to measure because a myriad of distribution arrangements have evolved that vary both the timing and magnitude of distribution charges. We derive a simple methodology that expresses the present value of distribution costs as a percentage of the original investment in fund shares for any expected holding period. This methodology allows direct comparison of the effect on investors of distribution fees for mutual funds with different types of sales arrangements.     

ECONOMIES OF SCALE IN MUTUAL FUND ADMINISTRATION

Since many mutual fund expenses are fixed costs, asset growth should reduce the ratio of fund expenses to average net assets. A translog cost function is estimated for a sample of 2,610 funds to evaluate the existence and extent of economies of scale in mutual fund administration. The elasticity of fund expenses with respect to fund assets is significantly less than one, indicating there are economies of scale in mutual fund administration. Average costs diminish over the full range of fund assets; however, the rapid decrease in average costs is exhausted by about $3.5 billion in fund assets.     

AN EMPIRICAL ANALYSIS OF MUTUAL FUND EXPENSES

Mutual fund investors are subjected to many fees and expenses related to both the management of the fund assets and the sale and distribution of the fund's shares. In recent years these expenses have increased as a percentage of assets. The preoccupation of mutual fund investors with using performance evaluation as a selection criterion is misguided because of the volatility of investment returns. Whether the fund's performance is due to superior management or just good luck is difficult to determine. On the other hand, mutual fund expenses are stable. As such, the mutual fund investor should pursue a policy of choosing funds with low expenses. In this paper we conduct an empirical analysis of these expenses. The results of our analysis of equity funds suggest that expense-conscious investors should look at the fund size, age, turnover ratio, cash ratio, and existence of a 12b-1 fee as key determinants of expenses. Our analysis of bond funds suggests that the key factors are the fund's sales charge, weighted average maturity, size, and existence of a 12b-1 fee.     

INFLUENCES AND TRENDS IN MUTUAL FUND EXPENSE RATIOS

A common question for the mutual fund industry, which has experienced a fivefold increase in assets over the last few years, is whether it shows declining production costs. Prior studies on fund expenses give conflicting results because several factors, from asset size to management style to share class, affect costs. Controlling for these and other factors shows the average stock fund expense ratio declined 1 basis point per year and bond fund ratios decreased 1 to 3 basis points per year, depending on the investment objective, from 1994 to 1998. Funds participating in no‐transaction‐fee supermarkets are more expensive than equivalent nonparticipants. JEL classification: G20, G28