Knowledge that Transforms

This research investigates the influence of market microstructure on liquidity premiums. Premiums of a competitive, multiple-dealership market (NASDAQ) are contrasted with those of a monopolistic, specialist system (NYSE). Differences in liquidity premiums are estimated from monthly stock returns. For small firms, the average returns of NYSE securities exceed the average returns of NASDAQ securities. This return differential persists after controlling for size, risk, and liquidity-related variables. Neither the NYSE nor NASDAQ dominates the other in providing liquidity. The NASDAQ appears to have a liquidity advantage over the NYSE for small firms but not for large companies.

This paper presents a detailed case history of voluntary restructuring by General Mills. During the 1980s the company reversed the extensive diversification of the two preceding decades by returning to its traditional core of packaged foods and food-related services. Drawing extensively on field interviews with the parties directly involved in the decision process, the paper explains the rationale behind both the structuring and the restructuring process. It highlights the role of the internal corporate governance process, the internal and external forces for change, and the consequences for financial performance and shareholder value.

A portfolio formed from a given list of assets is defined as a numeraire portfolio for the list if (a) it is self-financing, (b) its value is always positive, and (c) zero is always the best conditional forecast of the numeraire-dominated rate of return of every asset on the list. The numeraire portfolio exists if and only if there are no profit opportunities from trading assets on the list. For a sample list of heterogeneous assets (NYSE size-quintile portfolios, corporate bonds, and short-term bills), numeraire-dominated returns are similar to market-model forecast errors and, as abnormal return measures, clearly dominate market-adjusted returns.

In 1972 a group of electric utilities announced plans to construct Seabrook Station, a nuclear generating facility. In 1988, the lead partner in the venture, Public Service Company of New Hampshire (PSNH), filed for bankruptcy. Examination of the stock price effects of a variety of financial events preceding the bankruptcy filing shows that information about cash flows paid to PSNH security holders affected the common stock prices of PSNH and its Seabrook partners. whereas information about investment and operating cash flows had little or no effect.

We examine an exhaustive set of initial public offerings (IPOs) by venture-capital-backed companies between 1978 and 1987. We find that venture capitalists specialize their investments in firms to provide intensive monitoring services. Consistent with their monitoring role, the venture capitalists take concentrated equity positions, maintain their investment beyond the IPO, and serve on the boards of their portfolio firms. The quality of their monitoring services appears to be recognized by capital markets through lower underpricing for IPOs with better monitors.

This paper investigates whether union wealth concessions caused by changes in real wage growth following takeovers explain target-firm share-price premiums. Point estimates imply that union wealth changes in the six years following the acquisition account for 1% to 2% of shareholder premiums. This figure rises to 5% over 18 years. For hostile takeovers, union wealth increases by roughly 3% and 10% of shareholder gains over 6- and 18-year periods. The 95% confidence interval for wage changes implies that unions may lose up to a quarter, or gain up to a fifth, of what target shareholders gain.

I analyze financing policies in a firm owned by atomistic shareholders who observe neither cash flows nor management's investment decisions. Management derives perquisites from investment and invests as much as possible. Since it always claims that cash flow is too low fund all positive net present value projects, its claim is not credible when cash flow is truly low. Consequently, management is forced to invest too little when cash flow is low and chooses to invest too much when it is high. Financing policies, by influencing the resources under management's control, can reduce the costs of over- and underinvestment.

We examine the importance of dividend-capture trading in NASDAQ stocks by testing for cross-sectional relations between ex-day abnormal returns and bid-ask spreads. Throughout, we find that ex-day returns and spreads are positively related. The relation increases across dividend-yield quintiles and is strongest in high-yield stocks. The relation does not appear in a sample of non-ex-dividend days. These findings indicate that dividend-capture trading affects the ex-day returns of at least some, particularly high-yield, NASDAQ stocks, and that dividend-capture trading is important for understanding ex-dividend-day returns.

We investigate the relation between Tobin's Q and the structure of equity ownership for a sample of 1,173 firms for 1976 and 1,093 firms for 1986. We find a significant curvilinear relation between Q and the fraction of common stock owned by corporate insiders. The curve slopes upward until insider ownership reaches approximately 40% to 50% and then slopes slightly downward. We also find a significant positive relation between Q and the fraction of shares owned by institutional investors. The results are consistent with the hypothesis that corporate value is a function of the structure of equity ownership.

Previous research examining the relation between dividend yield and equity returns documents a U-shaped pattern arising from the positive CAPM-adjusted average excess return of zero-dividend firms. In contrast, this paper reports that zero-dividend firms earn negative average excess returns relative to firms of similar size. Despite the apparent conformity of these results to the predictions of after-tax asset pricing models, the negative size-adjusted excess returns cannot be drive solely by tax effects. These excess returns, which are concentrated in the initial zero-dividend years and approach - 1% per month, are attributed to possible dividend-expectation effects rather than taxes.