4.4.3 Studies from 2005 and later

Orr et al. (2005) focused their research on a sample of sixty companies from New Zealand. They formulated the hypothesis that companies with relatively many growth opportunities can benefit more from board independence1 than companies with less growth opportunities2 (Orr et al. 2005: 106-107). In companies with growth opportunities, management may be opportunistic and monitoring by independent directors deters managers from that opportunistic behaviour. The results show that board independence itself has a significant negative impact on company value, measured by the market value of equity (Orr et al. 2005: 111-117). The interaction term of board independence and a dummy variable that takes the value 1 if a company has high growth opportunities shows a significant positive relationship with company value. The positive coefficient of that interaction term can be interpreted as a confirmation of the hypothesis that board independence is more relevant to high-growth companies than to low-growth companies.

Beiner et al. (2006) studied the effects of corporate governance on company value in Switzerland. In their regression analysis they included a variable that captures 38 corporate governance attributes, which were collected through a survey. The attributes were derived from the recommendations in the Swiss Code of Best Practice and cover five categories: ‘corporate governance commitment, shareholders’ rights, transparency, board of directors and executive management, and auditing and reporting’. Besides this corporate governance variable, board independence3 was included in the analysis. In a sample of 109 companies with data from 2002, the authors found, in several models, negative relations between Tobin’s Q and board independence (Beiner et al. 2006: 263-266). These results are not significant. The constructed corporate governance variable does have a positive impact on the performance of the companies in the sample. Finally, a three-stage least squares analysis was carried out. The results between Tobin’s Q and board independence are again negative and not significant (Beiner et al. 2006: 268-269). So corporate governance in general does have a positive influence on performance, but the specific contribution of board independence cannot be measured, according to this study.

Bhagat and Bolton (2008) executed a large-scale analysis on United States companies over the period 1996 until 2003 concerning the relationship between corporate governance and performance. Board independence4 was considered to be an important part of corporate governance and receives attention in this research. 9,317 company year observations are available and show significant negative relationships with board independence in a number of regression and correlation reports (Bhagat and Bolton 2008: 264-267). Contemporaneous and subsequent ROA and Tobin’s Q were the measures of performance in the analysis. The authors remarked that if board independence is intended to improve company performance, this measure should be reconsidered. However, they added that board independence might have merits in poorly performing companies in disciplining management (Bhagat and Bolton 2008: 271-272). In addition the authors questioned whether one corporate governance characteristic can have exploratory power for the performance of a company. Corporate governance measures, which take a number of characteristics into account, might have more power to explain differences in performance. These have a reported positive relationship with performance in this research.

Bruton et al. (2010) investigated the effects of ownership concentration and the type of private equity investor on the post IPO performance of companies in the United Kingdom and France. In a sample of 112 French and 112 British IPO companies over the period 1996 until 2002, the authors found, in four out of five models, a significant negative relationship between board independence5 and performance, measured by per cent price premium6 and ROA (Bruton et al. 2010: 502-505). The fifth model also indicates a negative relationship, but that particular one is not significant.

Carter et al. (2010) focused their research on gender and ethnic diversity in American board rooms and their impact on performance. The number of independent directors on the board7 was one of the control variables in their study. In an unbalanced panel of 641 S&P 500 companies with 2,563 company year observations over the period 1998 until 2002, Carter et al. found solely insignificant negative relationships between the number of independent directors and performance (2010: 405-410). Tobin’s Q and ROA were used as performance measures. Fixed effects least squares and three-stage least squares regressions are the methodologies applied.

Drakos and Bekiris (2010) investigated, in a sample of Greek companies, the relationship between board characteristics – board independence,8 board size and leadership structure – and company performance, measured by Tobin’s Q. In this respect, they took the risk of endogeneity into account by using simultaneous equations. The dataset of Drakos and Bekiris consists of 1,409 company year observations of companies from the Athens Stock Exchange over the period 2000 until 2006. They only found insignificant negative relationships between board independence and Tobin’s Q in their sample (Drakos and Bekiris 2010: 393-394). Therefore, it can be concluded that this article did not find any real evidence for a relationship – either negative or positive – between board independence and performance. It must be noted that board independence regulation in Greece is less stringent than for example in Western Europe or the United States, because only one third of the board is required to be independent. This results in an average percentage of independent NEDs of 31 (Drakos and Bekiris 2010: 392).

Reeb and Upadhyay (2010) focused their research on subordinate board structures, or board committees. They concluded that the existence of board committees is beneficial to financial performance in companies with outsider dominated boards, because larger boards with relatively many independent NEDs suffer more from co-ordination, communication and asymmetric information problems (Reeb and Upadhyay 2010: 485). The dataset consists of 3,335 company year observations of large cap, mid cap and small cap S&P 1500 companies over the period 2000 until 2003. The correlations between board independence9 and performance, measured by both Tobin’s Q and ROA, are negative (Reeb and Upadhyay 2010: 473-475). In addition, they introduced a dichotomous variable that is 1 if a company belongs to 50 per cent of the companies with the highest levels of board independence; and 0 otherwise. In regression analyses, this variable is negative in all models and significant in some of the models (Reeb and Upadhyay 2010: 477-481). This means that belonging to the group of companies with the most independent boards has a negative impact on financial performance.