Bank Board Structure and Performance

Banks clearly appear to have different governance structures than non-financial firms. The question is whether these governance structures are ineffective and whether implementing independence standards imposed by Dodd-Frank, SOX and the major stock exchanges will improve bank governance. In our paper, Bank Board Structure and Performance: Evidence for Large Bank Holding Companies, which was recently made publicly available on SSRN, we try to provide an answer to this question by examining the relationship between board composition and size and bank performance. We focus on large, publicly traded bank holding companies (BHCs) in the U.S., which are the banks that are mentioned most often in the context of the crisis.

We first examine the relationship between board composition and size and performance in a sample of data on 35 BHCs from 1986-1999. We deliberately focused on a relatively small number of BHCs over a longer period of time to ensure that there would be sufficient variation in governance variables which typically do not change much over time. In addition, we collected detailed data on variables that have received attention in the law, economics, and organization literature and which are recognized to be correlated with sound corporate governance, but that are generally not studied as a group due to high data collection costs.

Since internal governance mechanisms are ultimately simultaneously chosen, the richness of this data enables us to limit omitted variable bias in performance regressions both by using firm fixed-effects and by controlling for possible interdependencies among governance mechanisms. To investigate alternative explanations for our findings, we extend this sample by collecting data on board size, board composition and performance from 1964-1985. Our data provides information on bank governance over a 34 year time period prior to the governance reform movement embodied in SOX and the revised listing requirements at the major exchanges. Thus, it helps us document persistent governance choices banks have made in the absence of strong external governance pressure. As such, it serves as a useful baseline against which to analyze any proposed governance changes.

Consistent with previous studies in governance, we examine the relationship between banking firm board structure and performance as proxied by a measure of Tobin’s Q. As in many studies of non-financial firms, we find that the proportion of independent outsiders on the board is not significantly related to performance. However, in contrast to findings for non-financial firms in Yermack (1996), Eisenberg, Sundgren and Wells (1998) and Coles, Daniel and Naveen (2008), the natural logarithm of board size is, on average, positively related to Tobin’s Q in our sample.

Since we use firm fixed-effects in our regressions, this finding is consistent with the idea that increases in board size add value because banks are growing in complexity over time. However, the time period of this sample was also a period of active consolidation in the banking industry as a consequence of the deregulation of interstate banking restrictions. Thus, a concern is that our results could be driven by endogeneity due to omitted variables related to M&A activity. We investigate this possibility but find little evidence that M&A activity is driving our results. For example, we find that board size is positively correlated with performance during the period 1965-1985, a period in which there was relatively little M&A activity because of regulatory restrictions on interstate banking.

Our results are robust to addressing endogeneity concerns that arise because of variables that are plausibly omitted from our performance regressions. However, an obvious concern is that our results could be driven by reverse causality. In order to address this endogeneity problem, we need an instrumental variable that is correlated with board size but uncorrelated with performance except through variables already included in our regression. Unfortunately, it is difficult to come up with valid instruments in the context of governance regressions. The factors that are arguably most correlated with the endogenous variable are other governance or firm characteristics that are already (or should be) included in performance regressions. Thus, rather than identifying the causal relationship between board size and performance econometrically, we try to increase confidence that our results are not spurious by examining whether there is a plausible mechanism that might drive it. In particular, we examine whether the value of large boards appears to be driven by BHC complexity.

We examine measures of operational, geographic and financial complexity, but do not find that large boards add more value as BHC complexity grows. We argue that one reason for this may be that some directors are more suited than others to help the BHC’s management deal with complexity. In particular, BHC directors often have subsidiary directorships. It is plausible that these directors play a particular role in coordinating activities in the holding company and thus dealing with complexity. Consistent with this argument, we find that when complexity increases, firm performance improves when BHCs have more of their directors sitting on subsidiary boards. There can be several reasons why large boards may add value in BHCs. But, one reason may be that larger boards contain a larger number of directors who also sit on subsidiary boards. Although we cannot claim to have identified the causal effect of board size on performance, we believe our results are at least suggestive that for banking firms the advantages of larger boards outweigh their costs.

Our paper contributes to the literature in several ways. First, to our knowledge we analyze the most extensive time series of data on bank board governance in the literature. Second, our paper complements a stream of research documenting that shareholder governance mechanisms, such as ownership, are important for bank behavior. Our paper also complements the literature on non-financial firms arguing that some firms may benefit from large boards. However, in our sample, increases in board size are not generally value-enhancing as firm complexity increases. Instead, increases in board size due to additions of directors who also sit on subsidiary boards appear to be important. As far as we know, there is no literature documenting that subsidiary directorships are common in non-financial firms. We appear to be the first to document that such subsidiary directorships may be important for banks. Overall our findings highlight the need to exercise caution in reforming bank governance following the subprime mortgage crisis. Simply adopting proposals that are largely motivated by research on non-financial firms are unlikely to be effective.

The full paper is available for download here.