The Core Corporate Governance Puzzle: Contextualizing the Link to Performance
63 Pages Posted: 5 Sep 2019
Date Written: September 4, 2019
There is a puzzle at the core of corporate governance theory. Prior scholarship reports a strong relationship between firms best at creating shareholder value and those rated highly by the established corporate governance indices. Little work explores why, however. We hypothesize that the link between governance and performance depends centrally on context. We illustrate the importance of context by exploring circumstances when a firm's governance structure can operate as a signal of the quality of its management. The idea is that better managers are on average more likely to choose a highly rated governance structure than are bad managers because a structure garnering a high rating increases the risk of job loss more for bad managers than for good ones. Conversely, the choice of a poorly rated governance structure signals negative information about managerial quality because good managers would not wish to make a false negative signal. Signals of managerial quality can take on particular significance under certain circumstances.
This Article tests empirically the hypothesis that a particular context — the existence of an especially high information asymmetry between a firm's insiders and the market concerning the quality of its management — is a situation in which a change in the firm's governance structure will become a stronger signal concerning its management's quality. The test compares ordinary times with 2000-2002, a period of unprecedented corporate accounting scandals that led to greater than usual uncertainty as to which firms had the better managers. We show that an index-score-altering change in governance structure during these accounting scandal years is associated with a much larger change in a measure affirm value creation — Tobin's Q — than a comparable governance change in the years before or after the accounting scandal period. By running both OLS and fixed effect regressions, we are able to show that the market's perception of the effectiveness of a highly-rated governance structure at better incentivizing managers, or at filtering out bad ones, was not significantly different in the scandal years than in the years before or after. Thus, "signaling" — the third possible causal link between good scores and higher Tobin's Q must have been at work. The reasoning is that the clarifying signal arising from a governance change should have a bigger effect in a period of greater uncertainty as to which firms had good managers. This conclusion is further confirmed by empirical evidence that the impact of a governance change on Tobin's Q during the scandal years was especially elevated for firms engaging in substantial amounts of R&D. Such firms have been shown by other studies to be generally more opaque.
These results also teach a larger lesson: the impact of governance is in important respects contextual, depending on the particular circumstances of the time, and the particular characteristics of the firms, involved. This point, largely missed to date, helps illuminate the current debate concerning the corporate governance index studies. It suggests that that there is an empirically verified theory that provides one explanation for the index studies' strong results linking governance structure with firm value creation, but that, rather than a single link between the specified corporate governance provisions and peiformance, a range of linkages are possible whose direction and intensity depend centrally on the particular context in which a firm is operating.
Suggested Citation: Suggested Citation