Institutional ownership, CEO incentives, and firm value

Institutional investors claim to monitor and influence executive compensation practices in their portfolio companies. Thus, compensation practices are a reasonable place to look to assess the institutions' effectiveness in achieving their governance objectives. I explore the relationship between institutional ownership and CEO compensation levels and equity performance sensitivity. I address ownership structure endogeneity by using instrumental variables and fixed-effects regression models with S&P 500 index inclusion as an instrument for institutional ownership. Results indicate that institutions fail to decrease compensation levels, but do increase equity performance sensitivities. The results are robust to various model specifications. CRRA simulations suggest that the level premium paid to CEOs is consistent with a utility-neutral shift in compensation structure. At first glance, this seems to support the hypothesis that CEO labor markets are efficient. However, results using an additional measure of the CEO's ability to extract rents are inconsistent with this conclusion. The level premium is negatively related to institutional ownership concentration. This suggests that the IR constraint is slack, but less so the more concentrated the institutional ownership.; The analysis from Chapter 1 is extended to explore the relationship between institutional ownership (with all of its implied monitoring) and firm value. Institutional ownership is shown to be positively related to firm value. Institutional and insider ownership appear to have complementary positive relationships with value and each other-suggesting a tradeoff between the forces implied by the entrenchment, convergence of interests, and external monitoring views of corporate ownership. The positive relationships exist even after controlling for ownership structure endogeneity. On average, a 1% increase in institutional ownership translates to a higher market value (as a percentage of book value) of approximately 0.6%, or {dollar}125M for the mean firm in the cross section, with the effect being slightly smaller for firm random effects models (0.4%). The ownership-value relationship is shown to be the result of improved monitoring and governance rather than price-pressure or institutional stock picking ability.