CEO ownership, corporate countrol, and bank performance

Griffith, John M

Abstract

This study examines the relation between CEO ownership and bank performance. In contrast to Pi and Timme (1993), we find that, when economic measures of performance are used, the relation between ownership and the performance of commercial banks is nonlinear. Additionally, in contrast to previous studies, we find the question of whether or not the CEO also holds the title of chairman of the board has an insignificant impact on bank performance. We conclude that, in commercial banks, management entrenchment may offset the effects predicted by Jensen and Meckling’s (1976) convergence-of-interest hypothesis. (JEL G3, G34, N2, N20)

Introduction

Jensen and Meckling (1976) propose, as managerial ownership increases the owner– manager’s interest converges with shareholders. Therefore, there is an increasing incentive for the owner-manager to maximize the value of the firm as managerial ownership increases.

Morck, Shleifer, and Vishny (1988) propose a non-linear model in which increased ownership by the CEO leads to entrenchment, where the manager will indulge in non-value-maximizing behavior. However, management’s self-indulgence should be less than if he or she has control but no claim on the firm’s cash flows. The entrenchment hypothesis predicts the value of the firm will be less when management is free from checks on its control, and suggests firm performance will decline as management ownership increases beyond a certain level. This apparent paradox may be understood if one considers the impact of entrenchment on the marginal effort of the manager to maximize firm value. As the manager’s ownership increases beyond a certain point, the ability of the market to discipline the manager declines, and thus the probability of the market disciplining the manager declines. After this point, the manager is entrenched and may seek to maximize his or her personal utility instead of focusing on shareholder wealth.

The convergence-of-interest and entrenchment hypotheses offer different views of the relation between ownership and value beyond the optimum point suggested by Morck, Shleifer, and Vishny (1988). Each of these hypotheses has merit. Therefore, this study is an attempt to reconcile their implications. This study differs from previous work in this area by using economic measures of bank performance and CEO ownership. The economic measures used are Market Value Added (MVA), Economic Value Added (EVA), and Tobin’s q. Prior studies, examining the relation between bank ownership and performance, have traditionally used accounting measures of performance and management ownership (Pi and Timme 1993; Berger 1992; and Glassman and Rhoades 1980). Because the CEO can be viewed as the classic owner-manager from a responsibility standpoint, he or she bears the ultimate responsibility for the performance of the bank, and the CEO has the most power to affect performance. Hence, we have chosen to examine the relationship between the CEO’s ownership and the bank’s performance (Griffith 1999).

In contrast to prior work, we find the performance of commercial banks is a function of CEO ownership, but it is not always increasing.1 The CEO’s ownership is found to have a dominating influence on the value of the firm and not insiders’ or other managers’ holdings. EVA rises until CEO ownership approaches the 12 percent level and declines until ownership reaches 67 percent. These findings suggest that, in commercial banks, management entrenchment may offset the effects predicted by the convergence-of-interest hypothesis. Furthermore, contrary to Pi and Timme (1993), we find whether or not the CEO also holds the title of chairman of the board has an insignificant impact on bank performance.

The next section discusses the effect management ownership has on firm performance. The third section explains the data and methodology. The fourth section presents the results, and the last section summarizes and concludes the study.

Management Ownership and Performance

The corporate literature extensively addresses the issue of management ownership and performance. However, it has received less attention in the banking literature. In the following paragraphs, we review corporate studies of ownership’s influence on firm performance as well as banking studies of the issue and outline this paper’s contribution to the literature.

Glassman and Rhoades (1980) focus on the relation between control and cost, growth, and profit performance. They conduct an empirical analysis of the relative performance of owner– controlled and manager-controlled banks. In general, the results show owner-controlled banks tend to have higher profit rates than manager-controlled banks. Glassman and Rhoades find sample choice has an important effect on the results; tests on the largest 200 banks showed no relation between profit rate and ownership. This result suggests an explanation for the failure of previous studies to find a statistically significant relation between profit rates and degree of owner control. Tests for nonlinearity indicate the effects of owner control are not evident until a relatively high level of ownership exists.

Demsetz and Lehn (1985) test and find no evidence to support Berle and Means’ (1932) hypothesis that diffuse ownership structure adversely affects corporate performance. Their primary concern is to explore what influences the structure of ownership. They argue the structure of corporate ownership varies in ways that are consistent with maximizing the value of the firm. They find firm size and the instability of profits explain the variation in ownership structure. However, using simple linear regression, they find no correlation between profit rate and ownership by large shareholders.

Stulz (1988) analyzes the way in which managerial control of voting rights affects firm value. In his model, the conflict of interest between management and outside shareholders comes solely from the fact that a successful tender offer influences the welfare of managers and outside stockholders differently. Stulz shows the premium presented in a tender offer is an increasing function of the fraction of voting rights of the target controlled by management, and the probability of a hostile takeover is a decreasing function of managerial control of voting rights. As management strengthens its control of voting rights, shareholders’ wealth increases up to a point and then decreases as management’s control of voting rights has a negative effect on the probability of a takeover. Stulz demonstrates, when the probability of a hostile takeover is zero, the value of the firm is at a minimum when management ownership is 50 percent.

Morck, Shleifer, and Vishny (1988) investigate the relation between management ownership and market valuation of the firm, as measured by Tobin’s q, by using a 1980 cross section of 371 Fortune 500 firms.,The study finds a significant relation between Tobin’s q and management ownership. Tobin’s q increases as board ownership increases from 0 to 5 percent, but declines as ownership increases from 5 to 25 percent and increases slowly as ownership rises beyond 25 percent. These breakpoints were arbitrarily selected. They interpret these results as indicating that conditions for entrenchment are significantly correlated with managerial ownership beyond the 5 percent level, but the convergence-of-interest effect exists throughout the entire range of ownership.

Holderness and Sheehan (1988) conduct an examination of publicly held corporations with concentrated ownership by analyzing 114 corporations with majority stockholders listed with the New York or American Stock Exchange. They define a majority stockholder as one person or entity who owns at least 50.1 percent but less than 95 percent of the common stock. They observe that investment policies, accounting returns, and Tobin’s qs are similar for majority-owned and diffusely held firms. These findings are inconsistent with the entrenchment hypothesis.

Using a sample of 1,173 firms for 1976 and 1,093 firms for 1986, McConnell and Servaes (1990) find a curvilinear relation between Tobin’s q and the fraction of common stock owned by corporate insiders. The relationship between q and insider ownership is positive until insider ownership reaches approximately 40 to 50 percent, where it becomes slightly negative. They also find a significant positive relation between q and the fraction of shares owned by institutional investors. The results are consistent with the hypothesis that corporate value is a function of the structure of equity ownership.

The entrenchment hypothesis is supported by the findings of Chen, Hexter, and Hu’s (1993) study, which examines the relation between management ownership structure and corporate value using samples of Fortune 500-sized firms in 1976, 1980, and 1984. It finds corporate value measured by Tobin’s q is a function of management ownership. Specifically, q rises when management ownership is between 0 and 5 percent. It falls as the ownership increases to 12 percent. Beyond 12 percent, the effect varies depending on the sample year.

Griffith (1999) examines the hypothesis that CEO ownership has a dominating effect on the value of the firm. Using a diverse sample of firms, firm value as measured by Tobin’s q is found to be a nonmonotonic function of CEO ownership. Specifically, Tobin’s q rises when the CEO owns between 0 and 15 percent and declines as CEO ownership increases to 50 percent. Beyond 50 percent, the value starts to rise. Firm value is found not to be a function of management ownership when CEO ownership is separated out, indicating that CEO ownership has a dominating effect on firm value.

One study that addresses control and bank performance is Pi and Timme (1993). They investigate the principal-agent conflicts by examining variations in performance and the relations between performance, top management team structure, ownership structure, and the composition of the board of directors. Their sample consists of large, publicly traded U.S. commercial banks. Their results suggest top management team structure affects performance and internal monitoring devices may not be effective. They find when the CEO is also chairman of the board, cost efficiency and return on assets are lower, indicating an increase in the principal-agent conflict because of the consolidation of the decision control processes. Conversely, they find cost efficiency and return on assets are positively related to non-chairman-CEO ownership and are unrelated to institutional and large block-holders’ ownership and the proportion of inside directors.

A second study that considers the impact of ownership structure on bank performance is DeYoung, Spong, and Sullivan (2001). They examine the effectiveness of managerial shareholdings as a tool to mitigate agency costs associated with hired managers in small, closely held commercial banks. Their findings indicate entrenchment can occur if managerial shareholdings are over-utilized and that under-utilization may result in reduced profits. Hence, their results indicate the existence of an optimal level of managerial shareholdings. Additionally, these findings point to the need for further research in this area to determine if the authors’ conclusions can be extended to large bank holding companies, the group examined in this paper.

These studies indicate there is disagreement over the impact of ownership structure on the value of the firm. In light of the findings of Griffith (1999) and DeYoung, Spong, and Sullivan (2001), we seek to resolve some of that disagreement by focusing specifically on how CEO ownership influences bank performance in bank holding companies as measured by MVA, EVA, and Tobin’s q.

Data and Methodology

This study uses a panel data set and a random effects model to examine the impact of CEO ownership on firm performance. The sample consists of 100 bank holding companies obtained from Stern Stewart & Co.’s 2000 database for years ending 1995-1999. COMPUSTAT data is used to calculate selected measures, and supplemental information is drawn from the Standard and Poor’s Stock Guide.

EVA

Prior studies traditionally used net income (NI), earnings per share (EPS), return on equity (ROE), or return on assets (ROA), all accounting measures of performance. Unfortunately, none of these performance measures tells us how much management has increased shareholders’ wealth. EVA can lead to management decisions that are different from those based on traditional measures. Traditional measures do not reflect risk and may promote behavior that aims to maximize earnings or prevent the dilution of returns. Return or ratio measures provide an indication of average profitability alone, without any assessment of shareholder value creation. The advantage of EVA is that it is dollar-based, and thus EVA maximization correlates with wealth maximization. Uyemura, Kantor, and Pettit (1996), in their study of EVA for banks, find that EVA has the strongest correlation with market value added (MVA). They show that no “return” or “ratio” measure can accurately assess shareholder value creation as well as EVA. They conclude that EVA provides the best operational performance measure. They also provide an excellent example of why EVA should be used to measure performance versus traditional measures of performance.

In contrast to Uyemura, Kantor, and Pettit’s (1996) findings for banks, Kramer and Peters (2001) find that in most industrial sectors (53) EVA is not an effective proxy for MVA. However, they did find that EVA does not suffer from any industry-specific bias when used as a proxy for MVA. They found that NOPAT, which is a component of EVA, consistently outperformed EVA as a proxy for MVA across industry sectors.

Ownership

For this study, we obtained ownership data from proxy statements for the year-end 1995– 1999. Anderson and Lee (1997) examine ownership data provided by four databases and proxy statements. They suggest that data collected from proxy statements, while expensive and labor intensive, is superior to other data sources.

Summary Statistics

Table 1 presents descriptive statistics for measures of bank performance and ownership characteristics. It is interesting to note that bank Tobin’s qs range from 0.0375 to 1.6736 with a mean q ratio of 1.2045, which is significantly greater than 1.0 at the 1 percent probability level for the period of 1995-1999, A value greater than one indicates the going-concern value of a firm exceeds the current cost of the assets necessary to generate the cash flow. If the value of a firm is separated into assets-in-place and growth opportunities, then a q greater than one indicates the firm has positive NPV projects.

Table 1 also indicates the mean EVA for our sample is $66.39 million and the median EVA is $21.82 million. The results in Table 1 show the mean bank market value increased by $2.66 billion and the median MVA is $816.59 million for the sample period. The large difference between mean and median for both EVA and MVA indicates the distributions are skewed toward the higher measures.

The results in Table 1 show the average bank has net operating profit after tax (NOPAT) of $432.75 million, a return on capital (ROC) of 13.28 percent, and a cost of capital (COC) of 10.57 percent for the period. The mean statistics for total assets and capital show the average bank has approximately $34.55 billion in total assets and $3.82 billion in capital for the period.

The ownership structure for our sample presented in Table 1 shows the mean ownership position for CEOs is relatively small. The mean and median holdings of the CEO are 2.88 percent and 0.72 percent of total ownership, respectively, thus indicating the distribution of CEO ownership is skewed by large holdings of some CEOs. The average holdings of the two classes of CEO, chairman-CEOs and non-chairman-CEOs, are not significantly different. The mean and median statistics for inside ownership show insiders, employees, or board members excluding the CEO own 6.84 percent and 3.27 percent, respectively, for the period. This is, on average, significantly smaller than the insider holdings of non-bank corporations, 13.06 percent and 7.35 percent (Griffith 1999).

Table 1 shows the average bank has three inside and 13 outside directors for the period 1995– 1999. Directors are classified as insiders if they are officers, ex-employees, or related to an officer of the bank. These results indicate these large bank holding companies do not pack their boards with insiders. However, the average tenure of insiders on the board was 12.70 years, and the average tenure of outsiders on the board was 9.83 years.

Methodology

From the data described above, we sought to extract certain measures of performance, ownership, and control and to determine which, if any, of the ownership or control measures were statistically significant as predictors of performance. Because dollar values of EVA and MVA are a function of bank size, we standardized both using year-end capital to eliminate any size effect.

We contend the percent ownership of the CEO captures the differences in behavior between the cases where the CEO is the primary owner and the CEO is merely hired to run the bank. Therefore, we do not subdivide the data into sub-samples based upon whether or not the CEO is an owner manager or simply hired to run the bank.

For measures of ownership and control, we constructed the following variables. CHM is a dummy variable that takes on a value of one if the CEO is also the Chairman of the Board, and takes on a value of zero otherwise. TEN is the natural log of CEO tenure, expressed in years. AGE is the natural log of the age of the CEO. INS-BD is the natural log of the ratio of inside directors to total board members. INSIDE is the natural log of the ratio of insider ownership to total ownership. Finally, CEO is the log of the ratio of CEO ownership to total ownership. In order to test for possible nonlinearity in the relation between ownership and performance, squared and cubed values of the ownership variables were constructed and used in the model.

After constructing these variables for ownership and control, correlation coefficients were estimated. Table 2 is the correlation matrix. Using the Pearson product moment correlation coefficient to measure the degree of linear relationship between variables and finding a high degree of correlation between most variables, it was concluded that only CHM and CEO were sufficiently uncorrelated with the other variables to be included in our model. Therefore, all variables other than CHM and the various ownership variables were dropped. Squared and cubed values of ownership were included in order to test for possible nonlinearity.

Testable Hypotheses

The first testable hypothesis is that the identity of the CEO as chairman of the board (CHM) is significant as a predictor of firm performance. This hypothesis is built on the belief that, if an individual is both CEO and the chairman of the board, he or she has the power to have a greater impact on firm value, either positively and negatively, than if he or she is CEO only. The second testable hypothesis, which is based on Jensen and Meckling’s (1976) convergence-of-interest hypothesis, is that a significant positive relation exists between the percentage of stock owned by the CEO and performance. The third testable hypothesis is that the relation between CEO ownership and performance is linear. We examine this hypothesis by including the squared and cubed terms of CEO in our regression model. If the relation between CEO ownership and performance is significant, and if either the squared or the cubed values of CEO are significant, we conclude the relation is nonlinear. The last and fourth testable hypothesis is that the relation between CEO ownership and firm performance is an increasing monotonic relation. We base this on the belief that the effects of entrenchment, if any, are not great enough to offset the effects of the convergence of interest that occurs as the CEO’s ownership increases. It is possible to have a nonlinear but monotonic relation between CEO ownership and firm performance. To test this hypothesis, we examine the rank correlation coefficient between CEO ownership and performance. Hence, the testable hypotheses are as follows:

H1: The relation between CHM and firm performance is significant.

H2: The relation between CEO and firm performance is positive and significant.

H3: The relation between CEO and firm performance is linear.

H4: The relation between CEO and firm performance is monotonically increasing.

Results

Table 3 presents the results of the cross-sectional regressions of EVA on management ownership, CEO ownership, and chairman status. We plot the results for the cubic model of management, CEO, and management less CEO ownership in Figure 1. The first column of Table 3 contains the results for management. The estimated coefficients are statistically significant. The model indicates EVA rises when management ownership is between 0 and 15 percent, falls in the 15 to 70 percent range, and rises when management ownership is greater than 70 percent. These findings support both the convergence-of-interest and entrenchment hypotheses. Firm value initially increases as management’s ownership rises or interests converge with shareholders; firm value subsequently falls as management’s control increases and management becomes entrenched, and then firm value rises again when management owns considerably more than a majority of the firm. Hence, the convergence-of-interest influences management’s decisions throughout all ranges of ownership but entrenchment has a greater marginal impact in the 15 to 70 percent range.

Regression 2 examines the dominance of the CEO’s ownership. The results show, if the CEO’s ownership is not included in management’s holdings (Mgmt-CEO), the relation between management’s ownership and EVA is insignificant. The adjusted RZ for the model is less than 1 percent, and the coefficients for all variables except the first order variable are insignificant. The first order variable is significant at the 10 percent level. This result confirms the belief that it is primarily the CEO’s ownership position in the firm that influences firm value, and what appears in previous studies’ examination of management’s ownership is actually the dominating effect of the CEO.

In regression 3, CEO ownership is introduced to examine how much of management’s or insiders’ influence is attributable to the CEO. The results show a cubic relation also exists between EVA and CEO ownership, and all coefficients are significant at better than the 0.01 level. EVA rises until CEO ownership approaches the 12 percent level and declines until ownership reaches 67 percent. These results support the convergence-of-interest hypothesis. This increase in the value of the firm at lower levels of CEO ownership also supports the hypothesis that the market disciplines the CEO and forces him or her to maximize the wealth of shareholders (Fama 1980; Jensen and Ruback 1983). At higher levels of CEO ownership (greater than 15 percent), the decline in value can be attributed to the inability of the market to discipline the CEO once he or she has obtained enough control to prevent ouster either through the managerial labor market (Fama) or the takeover market (Jensen and Ruback). Weston (1979) finds when insiders own more than 30 percent of the firm, there have been no firms acquired through hostile takeover. Once the CEO’s ownership reaches the 12 percent level, it appears entrenchment begins to dominate the effect of convergence of interest. The rise in value, once majority ownership is obtained, shows that at high levels of ownership the marginal impact of convergence of interest is greater than that of entrenchment.

We introduce the impact of CHM, the log of total assets and loans to total assets into the model in regressions 4, 5, and 6. All three have been used before as control variables. If the CEO is both CEO and the chairman of the board, he or she may have a greater impact on firm value, either positively or negatively, than if he or she is just the CEO. CHM is a dummy variable, which takes on a value of 1 when the CEO is also chairman of the board. We find, when the CEO also holds the title of chairman of the board, firm value is not significantly affected. Hence, our results do not support the findings of Pi and Timme (1993). There are several possible explanations. First, holding both positions has no significant impact on performance because the added title and added responsibilities do not significantly add to his or her ability to affect performance. Second, the CEO’s ownership drives performance, not his or her title. Finally, when he or she is only CEO, the chairman’s monitoring has little impact on his or her performance.

The coefficients for all previous significant variables remain statistically significant with the exception of the first order variable in regression 5. The adjusted R’s for each model are lower, indicating that these variables add little if any explanatory value to the model. The results show that a strong nonlinear relation between CEO ownership and EVA still exists after other explanatory variables have been included in the model.

Table 4 contains our cross-sectional regression results for CEO ownership. Contrary to the results predicted by the convergence of interest hypothesis, when MVA is used as a measure of performance, only CEO ownership is statistically significant. The relation between CEO ownership and performance is significant and negative, but not linear, thus suggesting the marginal positive impact of CEO ownership is less than the marginal negative impact of CEO entrenchment. CHM has a t-value of -0.27 and is not significant at the 5 percent level. CEO has a t-value of -2.24 and is significant at the 5 percent level.

The extra sum of squares method is used to test the hypothesis of linearity. An F-statistic of 0.574 leads us to reject our hypothesis of linearity. In testing for monotonicity, we report a rank correlation coefficient of -0.008, which is not significant at the 5 percent level.

Therefore, when MVA is used as a measure of performance, we reject the following hypotheses: (1) when the CEO also holds the office of chairman, performance is significantly influenced; (2) CEO ownership is a positive predictor of firm performance; (3) the relation between CEO and firm performance is linear. However, we cannot reject the hypothesis that the relation between CEO ownership and performance is monotonic. We find CEO ownership is a significant negative predictor of MVA. We conclude the entrenchment hypothesis and the inability of the market to discipline the CEO once his or her ownership reaches certain levels explain the relation between MVA and CEO ownership.

Next, we find a similar relation between CEO ownership and Tobin’s q. Given the construction of Tobin’s q, market value divided by book value, this result is not surprising. Only CEO is statistically significant, and the relation between CEO and performance is significant and negative, but not monotonic. CHM has a t-value of -0.47 and is not significant at the 5 percent level. CEO has a t-value of -2.62 and is significant at the 1 percent level.

Again, the hypothesis of linearity is tested, using the extra sum of squares method. An Fstatistic of 0.458 leads us to reject our hypothesis of linearity. To test for monotonicity, we report a rank correlation coefficient of -0.011, which is not significant at the 5 percent level, and conclude that we cannot reject the hypothesis that this relation is monotonic.

Next, we use a cubic model in column 3 to examine the relation between standardized EVA and CEO ownership. The results for the cubic model show that whether the CEO is also the chairman of the board is not a significant predictor of firm performance, but his or her ownership is a significant predictor of bank performance. The results indicate as CEO ownership increases, performance first increases, then decreases, and then increases again (Figure 1). Note that CHM has a t-value of -0.12 and is not significant at the 5 percent level. CEO has a t-value of 2.49 and is significant at the 5 percent level. The squared term of CEO has a t-value of -2.48 and is significant at the 5 percent level. Finally, the cubed term of CEO has a t-value of 2.11 and is significant at the 5 percent level.

These findings support both the convergence-of-interest and entrenchment hypotheses. Firm value increases as the CEO’s ownership rises or interests converge with shareholders; firm value falls as the CEO’s control continues to increase and then rises once again when the CEO owns considerably more than a majority of the firm, thus indicating that the convergence-of-interest influences management’s decisions throughout.

Further tests indicate, in contrast to prior work (Pi and Timme 1993; Berger 1992; and Glassman and Rhoades 1980), the relation between CEO ownership and EVA is significant but not necessarily monotonic. To test for monotonicity, we report a rank correlation coefficient of 0.075, which is not significant at the 5 percent level. Thus, when EVA is used as a measure of bank performance, H3 is rejected. However, due to the reduced capacity of the market to discipline the CEO, H2 is only rejected for ownership levels between 12 and 67 percent.

We also examined the impact of board composition on firm performance. Our results indicate performance is not significantly influenced by board composition. These findings are consistent with Pi and Timme (1993) but in contrast to those of Griffith (1999).

Conclusions

This study employs a sample of the 100 largest U.S. bank holding companies from 1995 through 1999 and utilizes three different measures of bank performance to examine the relation between bank performance, CEO ownership, and CEOs who are also chairman of the board. The three measures of bank performance used are Tobin’s q, EVA, and MVA. All are economic measures of performance indicating the contribution of management to shareholders’ wealth.

We find no relation between the CEO serving as Chairman of the board and bank performance using Tobin’s q, EVA, or MVA. These results are contrary to those of Pi and Timme (1993), who find cost efficiency and return on assets are negatively related to CEO-chairmanship.

When we use market-value-added (MVA) and Tobin’s q to measure bank performance, we find a significant and negative relation between CEO ownership and performance. This differs from the results of Pi and Timme (1993) and suggests management entrenchment, or some other mitigating factor, may be influencing results. This difference in results may be due to the use of a market-based measure of bank performance. Furthermore, when we use economic value added (EVA) as a measure of performance, we show a significantly positive nonlinear relation between CEO ownership and performance. Performance rises until the CEO’s holdings reach approximately 12 percent and then declines until his or her ownership reaches 67 percent of the firm. It seems counterintuitive for a CEO with majority interest in the firm not to seek to maximize shareholder wealth. However, the CEO may accept sub-optimal returns in order to reduce firm specific risk due to inadequate diversification within his or her own portfolio. Therefore, both convergence of interest and entrenchment impact performance, but the marginal impact of these factors varies with the level of CEO ownership. Using EVA as a measure of performance, these results confirm the findings of McConnell and Servaes (1990), Morck, Shleifer, and Vishny (1988), and Chen, Hexter, and Hu (1993).

Whether we use Tobin’s q, EVA, or MVA as a measure of performance, it appears our results differ from the predictions of the “convergence of interest” model (Jensen and Meckling 1976). Our results suggest “entrenchment” or some other mitigating factor, such as the inability of the market to discipline the CEO, may be the dominant factor influencing bank performance for mid– range values of CEO ownership. Thus, the “convergence of interest” model may not provide a meaningful explanatory model of bank performance.

1Similar results for industrials were found by Morck, Shleifer, and Vishny (1988) and Griffith (1999).

References

Akhavein, Jalal D., Allen N. Berger, and David B. Humphrey. 1997. “The Effects of Megamergers on Efficiency and Prices: Evidence from a Bank Profit Function.” Review of Industrial Organization 12: 95-139.

Berger, Allen N., Rebecca S. Demsetz, and Philip Strahan. 1999. “The Consolidation of the Financial Services Industry: Causes, Consequences, and Implications for the Future.” Journal of Banking and Finance 23: 135-194.

Berger, Allen N., and David B. Humphrey. 1992. “Megamergers in Banking and the Use of Cost Efficiency as an Antitrust Defense.” The Antitrust Bulletin 37: 541-600.

Berger, Allen N., Anthony Saunders, Joseph M. Scalise, and Gregory F. Udell. 1998. “The Effects

of Bank Acquisitions on Small Business Lending.” Journal of Financial Economics 50: 187229.

Canada NewsWire. www.newswire.canot.com

Chang, Saeyoung, and David Y. Suk. 1998. “Failed Takeovers, Methods of Payment and Bidder Returns.” The Financial Review 33: 19-34.

Cornett, Marcia M., and Hassan Tehranian. 1992. “Changes in Corporate Performance Associated with Bank Acquisitions.” Journal of Financial Economics 31: 211-234.

Demsetz, H. 1968. “Why Regulate Utilities?” Journal of Law and Economics April: 55-65.

Hannan, Timoth H., and Robin A. Prager. 1998. “Do Substantial Horizontal Mergers Generate Significant Price? Effects Evidence from the Banking Industry.” Journal of Industrial Economics 46: 433-452.

Healy, Paul M., Krishna G. Palepu, and Richard Ruback. 1992. “Does Corporate Performance Improve After Mergers?” Journal of Financial Economics 31: 135-175.

Holthausen, Robert W., and Richard W. Leftwich. 1986. “The Effect of Bond Rating Changes on Common Stock Prices.” Journal of Financial Economics 17: 57-89.

Hotelling, H. 1929. “Stability in Competition.” The Economic Journal 39:41-57.

Kaen, Fred R., and Hassan Tehranian. 1990. “Information Effects in Financial Distress: The Case of Seabrook Station.” Journal of Financial Economics 26: 143-171.

Nathan, Ali, and Edwin H. Neave. 1992. “Operating Efficiency of Canadian Banks.” Journal of Financial Services Research 6: 265-276.

Pilloff, Steven J. 1996. “Performance Changes and Shareholder Wealth Creation Associated with Mergers of Publicly Traded Banking Institutions.” Journal of Money Credit and Banking 28: 294-310.

Prager, Robin A., and Timothy H. Hannan. 1998. “Do Substantial Horizontal Mergers Generate Significant Price Effects? Evidence from Banking Industry.” Journal of Industrial Economics 46: 433-452.

Resti, Andrea. 1998. “Regulation Can Foster Mergers; Can Mergers Foster Efficiency? The Italian Case.” Journal of Economics and Business 50: 157-169.

Rose, Peter S. 1987. “The Impact of Mergers in Banking: Evidence from a Nationwide Sample of Federally Chartered Banks.” Journal of Economics and Business 39: 289-312.

Saunders, Anthony, and Berry Wilson. 1999. “The Impact of Consolidation and Safety Net Support on Canadian U.S. and U.K. Banks: 1893-1992.” Journal of Banking and Finance 23: 537-571.

Task Force on the Future of the Canadian Financial Services Sector. 1998. Report of the Task Force Change Challenge Opportunity. Ottawa, Canada.

Zhang, Hao. 1998. “U.S. Evidence on Bank Takeover Motives: A Note.” Journal of Business Finance and Accounting 25: 1025-1032.

John M. Griffith, Lawrence Fogelberg, and H. Shelton Weeks*

John M. Griffith, College of Business and Public Administration, Old Dominion University, Norfolk, VA 23529, jgriffit@odu.ed; Lawrence Fogelberg, Sorrel College of Business, Troy State University; H. Shelton Weeks, College of Business, Florida Gulf Coast University.

Lithium-ion battery chemistry offers a range of advantages for cordless power tools

Hit the mark