For example, one of the puzzles in the literature examining the relationship between governance and performance is the lack of evidence that board independence matters. They find that bid premiums increase as target board independence increases and are not affected by target board size. Byrd, Fraser, Lee, and Williams (2001) examine the effect of internal governance arrangements on the likelihood that a thrift will survive the thrift crisis of the 1980s.
The proportion of foreigners in BHCs and the size of the board are large compared to statistics reported from large production samples. It is known that the size of the board can be increased after a merger or acquisition to incorporate some of the target directors. Here we must emphasize that the characteristics of the internal governance we describe are those of the board at the bank holding company level.
For example, the directors of BHC often sit on the boards of subsidiary banks of BHC. This also means that the number of BHC board meetings may underestimate the total interaction between the BHC directors. We discuss our specification of the relationship between performance and board size and composition in Section 3.1.
Since board committee structure and the number of board meetings are reliably related to board size and composition, we cannot be sure that we are not obtaining spurious correlations between board size, composition, and performance if we do not we include these additional governance. characteristics in our performance regressions.
Empirical Results
The most surprising finding from the previous section is the positive relationship between board size and the Tobin-Q. It is also known that board size can increase following a merger or acquisition that includes some of the target directors (e.g. Wulf, 2004). If firms with high Q are more likely to engage in mergers and acquisitions, then our finding of a positive relationship between board size and Q could be due to endogeneity due to omitted variables characterizing BHCs.
As is clear from the table, the coecients on the natural logarithm of board size are very similar to those in Table 2, both in magnitude and significance. Our findings do not appear to be driven by the possibility that firms with a high Q value are more likely to undertake mergers that lead to increases in board size. However, another possibility is that better-performing firms can afford to retain target directors longer, so that mergers and acquisition activities lead to more persistent increases in board size in such firms.
Essentially, we allow for the possibility of board size being temporarily inflated due to mergers and acquisitions. This suggests that the increase in board size due to mergers and acquisitions may explain part of the relationship between Q and board size, but it does not appear to be the main reason why the relationship between Q and board size of the board is non-negative. For the years 1959–1985, we collected information on the size and composition of the boards of directors of our sample companies from Moody's Bank and Finance Manuals.
The trend is consistent with the decline in board size documented by Wu (2000) for the Forbes 500, which she attributes to institutional activism. However, it is fair to say that the decline in bank board size started before the publicized pressure on boards from institutional shareholders. The downward trend in board size does not appear to be accompanied by major changes in board composition.
To assess whether the relationship between board size and composition and performance differs before and after 1986, we regress Tobin's Q on board structure for the years for which stock return data are available on CRSP for our sample banks. As in Table 2, the coefficient on the natural logarithm of board size in all regressions is positive and significant at greater than 10%. In Section 5.1, we identify several specific factors related to the holding company structure that may affect BHC board structure and examine their influence on board size.
BHC organizational structure and board structure
However, the total number of subsidiaries is increasing, primarily as a result of an increase in non-banking companies. In addition, the number of states in which the domestic subsidiaries are located increases from an average of 2.31 in the earlier period to 4.57 in the later period. Different states may also have requirements for board size and composition at the bank level.
Because such regulatory constraints apply to bank-level board structure, the regulatory environment alone does not explain BHC board size and composition. In columns II and III, we also include ROA and two lagged ROAs to control for the fact that performance may affect panel size. The coefficient on the BHC dummy is consistently negative and significant at the 1% level in all specifications.
Consistent with previous research, firm size has a significant and positive effect on board size, but as in Yermack (1996), prior performance does not appear to affect board size.18 Therefore, we let ROA and its lags beyond all further specifications. . The third is the average number of domestic subsidiaries per state and the fourth is the relative size of the lead bank, which we define as the ratio of the consolidated assets of the largest commercial bank subsidiary to the total Tier 1 consolidated assets of the commercial bank. As shown in Column I, the coefficients on the number of subsidiaries and the relative size of the lead bank are negative and significant, while the coefficient on the number of states in which subsidiaries are located is positive and significant and exceeds the 1% level. .
When we break down the subsidiaries by type, we find that the negative correlation between the number of subsidiaries and board size appears to be driven by the number of foreign subsidiaries, both banking and non-banking. In columns II and III, the number of foreign subsidiaries outside the banks is negative and significant at the 1% level. 18This also suggests that our finding that board size is positively related to Tobin's Q is not driven by a positive relationship between past performance and board size.
Although these regressions are exploratory, we believe that the sign pattern is consistent with the impact of an organizational structure induced need for delegation and information sharing on board size. For example, it is intuitive that as the number of states increases, the board size increases to accommodate representatives of subsidiaries from different states. These results and those in Table 8 are also compatible with the downward trend in board size we observe in Figure 2.
BHC organizational structure, board structure and perfor- mance
Even after including controls for both organizational structure and M&A activity, we do not find a strong reversal of our previous finding of a positive and significant relationship between board size and Tobin's Q. The significance of the board size coefficient declines, but the relationship is still non-negative.19. However, it is not clear that unobservable client relationships can cause a spurious non-negative relationship between board size and performance.
In this paper, we investigate the relationship between board size and composition and Tobin's Q in banking firms. The most surprising result is our finding of a non-negative relationship between board size and Tobin's Q, which is contrary to the evidence for non-financial firms. For example, the regional expansion of banking activities may be associated with an increase in BHC board size due to the need to coordinate subsidiary boards located in different states.
We find that board size is significantly related to BHC structure characteristics and that BHC structure can explain part of the relationship between board size and Tobin's Q. However, even after accounting for BHC structure, we do not find a negative relationship between board size and Tobin's Q In line with the negative relationship found between Tobin's Q and board size for non-financial firms, institutional investors often advocate that firms reduce the size of their boards.
Larger Board Size and Decreasing Firm Value in Small Firms.”Journal of Financial Economics pp. A non-insider is defined as any director who is not currently an officer at the banking company's head office. Our measure of Q is the ratio of the company's market value to the book value of its assets.
Ratio of M&A directors to board size is the fraction of directors who joined the board as a result of an M&A transaction. A non-insider is defined as any director who is not currently an officer of the banking firm's headquarters. Data on board size and composition prior to 1986 are from Moody's Bank and Finance Manuals.
OLS regressions of board size on the bank holding company dummy, assets and past return on assets for banking firms 1959-1999. A non-insider is defined as any director who is not currently an officer at the banking company's head office.
