3. The Monitoring and Advisory Functions of Corporate Boards

Specifically, these studies suggest, for example, that financial expert directors (Xie et al. 2003), political directors (Goldman et al. 2009), female directors (Adams and Ferreira 2009), CEO directors (Fahlenbrach et al. 2010), and foreign directors (Masulis et al. 2012a) may significantly affect the effectiveness of the board in performing its functions.

A director is regarded as an executive director if he/she is the CEO, Chairman of the board, president or holds another senior executive position in a firm.

Following Cheng (2008), we define a receiver firm as a company “who hires executives from other firms as independent directors. The executives are confined to those who also sit on boards as inside directors. (…) A sender firm is a firm who has at least one executive board member serving as independent directors in other firms”.

We note that, given the relatively small number of such independent executive directors, our results remain unchanged whether or not we discard these observations.

For example, if the benefits of intense board oversight outweigh the negative effects on board advising, we should observe a positive relation between the intensity of board monitoring and firm value, and a negative relation otherwise. If the value added by advisory directors’ presence on the board is eclipsed by their negative impact on board monitoring, we should find a negative relation between advisory directors’ presence on the board and firm value, and a positive relation otherwise.

The RiskMetrics Database contains data only on the audit, compensation, nominating and corporate governance committees.

Appendix A provides detailed definitions and sources of all variables.

This is below the threshold of 0.8 beyond which multicollinearity problems arise. To double check for any multicollinearity issue, we also perform the Variance-Inflation-Factor (VIF) analyses for all tests. We find that the highest VIF is around 3, which is considerably less than the 10 threshold above which multicollinearity could be an issue (see Gujarati 2003).

The mean and median values were obtained by excluding firm-years with missing R&D data.

The weighting index is constructed using a quasi-structural model, where the shape of the distribution of citation-lags (defined as the time difference between the application year of the patent and the year the citation is received) is econometrically estimated.

These statistics were calculated by dropping observations with missing patent data.

Likewise, other studies argue that firms characterized with weaker monitoring, such as those in which the CEO and chairman positions are held by the same person (Goyal and Park 2002), those with busier boards (Fich and Shivdasani 2006) or those with foreign independent directors (Masulis et al. 2012b) are less likely to terminate underperforming CEOs.

Following Chhaochharia and Grinstein (2009), we measure firm performance using the prior year annual stock return.

In line with the theoretical issues outlined here, a considerable body of research has attempted to assess the link between CEO compensation and the strength of corporate boards. Specifically, Faleye (2007) finds that firms with classified boards reward their CEOs with higher levels of compensation and lower compensation incentives. Similarly, Core et al. (1999) argue that CEO compensation increases with board size, the proportion of busy directors on the board, the CEO–Chairman duality and director age. Overall, these studies converge on the conclusion that CEO pay incentives improve with the quality of corporate boards.

To measure firm performance, we use the annual stock return instead of industry- or market-adjusted stock returns, because according to Jenter and Kanaan (forthcoming), firms compensate their CEOs based on absolute performance rather than relative performance.

According to the classification criteria proposed by Hennes et al. (2008), a restatement is classified as an intentional misreporting if one of the following conditions is satisfied: (a) the company uses words like “fraud” or “irregularity” in the US Securities and Exchange Commission (SEC) filings or restatement announcements; (b) the company is under investigation by the SEC or the Department of Justice; and (c) the company conducts independent internal investigations. The authors also highlight the relevance and effectiveness of their partitioning procedure by conducting several tests. They find, inter alia, that restatements attributed to accounting irregularities generate significantly more negative market reactions around the announcement, compared to accounting error restatements.

We recognize that we cannot completely rule out the possibility that our results are driven by omitted variables.

We also note that, while the firm value equation is analogous to that of the first column of Table 3.8, the monitoring proxy equation is a function of Tobin’s Q, board size, board independence, directors’ external business, CEO–Chairman duality, firm size, leverage, corporate diversification, R&D expenditures, operating performance, CEO ownership, independent directors’ ownership, institutional ownership, blockholders ownership and the combined size of the audit, compensation and nominating committees, and the advisory proxy equation is a function of Tobin’s Q, board size, board independence, CEO–Chairman duality, firm size, leverage, corporate diversification, operating performance, CEO ownership, independent directors’ ownership, the number of companies in each 2-digit SIC code industry, and the average number of board seats held by independent directors in the industry.