Printer Friendly

The relationship between corporate compensation policies and investment opportunities: empirical evidence for large bank holding companies.

The existence and determinants of an optimal corporate policy for compensating top executives are unresolved issues in corporate finance. This paper investigates whether the firm's set of investment opportunities drives the amount and type of executive compensation. The firm's investment-opportunity set potentially affects the nature of agency problems between insiders of the firm and outsiders.(1) Examining the empirical relations between compensation schemes and investment opportunities, therefore, may uncover the degree to which executive compensation packages are able to align the incentives of managers and outside security holders.

We conduct a detailed empirical analysis of the executive compensation plans of large bank holding companies (BHCs).(2) Our approach differs from earlier studies that document systematic differences in corporate-compensation policies across industries. Instead, we examine intra-industry changes in compensation policy over a period of significant change in the investment-opportunity set. We study the banking industry because of changes in the set of investment opportunities that occurred in the 1980s.(3) These changes were responses to financial innovation, to the relaxed bank-regulatory constraints, and to deregulation of the financial services industry in general, which have allowed banks to offer nonbanking financial services.

Smith and Watts (1992) contend that, other things equal, the larger the proportion of a firm's value represented by intangible investment opportunities, the greater the manager's compensation should be. They suggest three reasons for this relationship. First, the investment skills of top executives are scarce resources compared to the supervisory skills of executives. Second, since the risk-reward tradeoff applies to executive compensation, the greater the firm's risk, the greater top management's compensation should be. Third, the higher a firm's proportion of intangible assets, the more likely that top management's remuneration is tied to the firm's value through incentive compensation schemes, which would increase the variability of such compensation.

Formal incentive plans usually take the form of bonuses, stock options, stock-appreciation rights, and performance incentive programs. Typically, such programs make an explicit ex ante linkage between the manager's compensation and the firm's performance as measured by stock price, earnings per share, or some other performance measure. The efficacy of such programs results from a direct connection between the manager's actions and the performance measure. Informal incentive programs, in contrast, take the form of ex post settling up. In these cases, for example, the manager's salary is renegotiated based upon the company's performance over some designated period.

Other things equal, firms with substantial growth opportunities and higher future cash flow volatility are more likely to use formal incentive compensation programs for their top executives. Since BHCs face restricted investment opportunities, they are less likely to use formal incentive plans. Smith and Watts (1982) argue that risk-averse managers, who have formal incentive plans, should be more cognizant of risk-return tradeoffs because their compensation risk relates directly to the firm's risk exposure. BHCs' investment opportunities should expand as financial innovation takes place and as regulatory constraints relax. The resulting increase in growth options should be associated with higher levels of total executive compensation because the chief executive officer takes on greater investment responsibilities and greater risks as investment opportunities expand. Further, we should observe a higher incidence of incentive compensation for the chief executive officer because the firm's growth prospects are less restricted.

We find that from 1979 to 1985 total real compensation and the ratio of incentive compensation-to-total compensation increased substantially at regional BHCs but remained stable at money-center BHCs. These results are consistent with greater growth opportunities at regional banks relative to those of money-center BHCs. As agency theory suggests, the CEOs of these firms were rewarded for exercising these growth options. After controlling for bank size, we find that the ratio of noninterest income to total income appears to be a good proxy for banks' investment opportunities over our test period. Given the decline in the traditional loan-and-deposit business of commercial banking, the success of this measure, which captures, among other things, fees associated with off-balance sheet activities and the movement from implicit pricing (e.g., compensating balances) to explicit pricing (e.g., fees and service charges), has great intuitive appeal to support its statistical significance.

The remainder of the paper is organized as follows. Section I describes the test period, the data, and the sample. Section II presents the empirical model, while Section III discusses the results. Section IV summarizes and concludes.

I. Test Period, Sample Firms, and Compensation Data

Our sample firms are large BHCs consisting of money-center and regional companies. Compensation and financial data for the years 1979 and 1985 are analyzed. This section begins with a brief description of the changing investment opportunities for large BHCs over our test period.

A. Test Period

To test for the effects of changes in the investment-opportunity set on compensation policy, we compare the compensation policies of a sample of large BHCs between 1979 and 1985. We collect only two years of compensation data because of the high cost of collecting complete and detailed information on compensation for any sizable sample. We document our data collection procedures in the following two subsections. We chose 1979 because it represents the beginning of an era of accelerating financial innovation (e.g., securitization and the development of off-balance-sheet activities, such as interest-rate swaps) and deregulation in the financial services industry resulting from the Depository Institution Deregulation and Monetary Control Act in 1980 and the Garn-St Germain Depository Institutions Act in 1982. We chose 1985 because it represents the continuation of the financial innovation trend and the relaxed regulatory constraints in the wake of the deregulation acts of 1980 and 1982, as well as the many reciprocal interstate-banking agreements written by 1985. Our time frame for exploring the effects of deregulation is similar to those used by Collins, Blackwell, and Sinkey (1994) and Crawford, Ezzell, and Miles (1995).

B. The Sample: Large BHCs

We required data from the Standard & Poor's Bank COMPUSTAT file, the CRSP daily returns file, and the BHC's annual proxy statement. The sample consists of 66 BHCs for which all types of information are available. These 66 firms are listed on either the New York or American Stock Exchanges, and they control the majority of deposits and assets of commercial banks in the U.S.

We classify each firm as either a money-center BHC or a regional BHC to investigate differences in the investment-opportunity sets of the two groups. Our sample consists of 59 regional BHCs and seven money-center BHCs. We use the BHC classifications from Black, Fields, and Schweitzer (1990) to determine whether a BHC is a money-center or regional firm. McLaughlin and Wolfson (1988) prepared a similar list to gauge commercial bank profitability for 1987. As an alternate example, Citicorp's competitor list shown in Sinkey (1983) includes Bank of America, Bankers Trust, Chase Manhattan, Continental Illinois, First Chicago, Morgan, Chemical, Manufacturers Hanover (and Citicorp). However, since we were unable to obtain sufficient information on compensation for Chemical and Manufacturers Hanover, we restrict our analysis to the other seven money-center banks. We list the sample BHCs by money-center versus regional grouping in the appendix.

Table 1. Total Managerial Compensation: Description and Measurement

This table presents the definitions used to categorize and value the various forms of CEO compensation from the annual proxy statements for 1979 and 1985.

Salary - ash compensation received in recognition of services rendered under the explicit and implicit contracts between an employee and the employer.(a)

Bonus - Cash compensation received in recognition of the achievement of some short-term performance measure under an explicit contract between an employee and the employer. This performance measure is typically the attainment of a specific sales, net income, or earnings per share goal.(a)

Profit-sharing plan - A cash-contribution plan in which the employee typically makes no contribution (although a number of plans either permit or require employee contributions) and where the amount of the company contribution is usually based on company profits. To qualify for tax purposes, the plan must defer payment for at least 2 years. While the amounts may accumulate tax-free during this period, the company is allowed to take a deduction for compensation expense for the year in which the expense was incurred. This short-term plan pays out all of the employer's contribution after a minimum time period, such as 2 years.(a)

Long-term incentive plan - A cash-contribution plan similar in construction to a profit sharing arrangement except that all or a part of the employer's contribution remains in the plan for a period beyond the specified 2-year minimum.(a)

Common-stock award - A common stock purchase with a 100% discount that is given immediately to the employee. Consequently, the employee has full ownership rights attributable to common stock.(b)

Restricted common-stock award - A common-stock award with ownership deferred to a future date. While the individual is enjoined from assigning, transferring, or selling the stock without a tax liability, the executive has the right to vote the stock, receive dividends, and even possess the stock certificate.(b)

Performance incentive shares - Indicates the number of common-stock shares that will be awarded at varying levels of prescribed company performance (individual performance is rarely considered in these plans). A number of companies establish annual awards with payouts after 3 years. Initially this modification requires making three contingent awards: a 1-year, a 2-year, and a 3-year. After the first year, each award is for the future 3-year period.(b)

Common-stock options - An award of common-stock options gives the employee the right to purchase a stated number of shares of common stock at a stated price during a stated period of time. The fair market value of the stock option is represented by the Black and Scholes (1973) Option Pricing Model in its European form:

FMV = O[[S.sub.0] N([[Delta].sub.1]) - [Xe.sup.-[R.sub.f]T]N([[Delta].sub.2])]

where

FMV= Fair market value of the common-stock award.

O = Number of options granted on common shares from the proxy statement.

[S.sub.o]= Closing market price per share of common stock for the fiscal year from the Bank COMPUSTAT tapes.

X = Average exercise price per share for the options awarded in year t from the proxy statement.

[R.sub.f] = Annual risk - free rate of interest from the CRSP bond tape from the Fama-Bliss risk-free rate file for Treasury bills.

T = The time to expiration for the options granted in year t from the stock-option plan descriptions in the proxy statements.

N() = Cumulative probabilities for a unit-normal variable.

[Sigma] = Annual standard deviation of common-stock returns from the CRSP tapes.

[[Delta].sub.1] = {ln([S.sub.0]/X) + T[[R.sub.f] + ([[Sigma].sup.2]/2)]}/[Sigma][square root of T]

and

[[Delta].sub.2] = [[Delta].sub.1] - ([Sigma][square root of T])

Stock appreciation rights (SARs) - Self-standing SARs are simply phantom awards, since there is no accompanying stock option. These rights, however, behave exactly as options with SARs for valuation purposes. The fair market value of stock appreciation rights is represented by the Black-Scholes Option Pricing Model.

Common-stock options with SARs - A common stock option with SARs permits the optionee to receive the appreciation of the future market value of the award over the option price in stock and/or cash without providing funds for the option price. The fair market value of common-stock options with stock appreciation rights is represented by the Black-Scholes Option Pricing Model.

Dividend units - The recipient is awarded an equivalent for a stated number of shares of company stock. The executive does not receive the stock but will each year receive an amount (cash or stock) equal to the dividend paid. Such grants are often in effect for long periods of time (e.g., until death or age 85 whichever comes last).a

Book-value units - This form of incentive compensation deals with appreciation in a manner similar to that of phantom stock plans related to company stock except, by definition, this uses book value (i.e., assets minus liabilities dividend by the number of shares of common stock outstanding). The idea is that, unlike those plans related to company stock that can bounce up and down, reflecting market assessment of macroeconomic issues rather than company success, book value usually has a nice steady progress. These plans offer an employee the opportunity to receive the company stock at its book value rather than its market value. This is accomplished either by allowing the employee to purchase stock at its current book value or by giving the stock to the employee.(a)

Source: The descriptions in this table are taken from Ellig (1982).

a The stated value in the proxy statement is used as the market value of this form of compensation.

b The stated value in the proxy statement is in units of common stock. The market value of this form of compensation is equal to the number of units of common stock multiplied by the closing price of common stock for the fiscal year.

C. Compensation Data

We obtain information on compensation of the chief executive officer (CEO) from corporate proxy statements filed with the Securities and Exchange Commission. We collect information on salaries, bonuses, and incentive compensation for the years 1979 and 1985 for BHCs appearing on the Standard & Poor's COMPUSTAT file. Incentive compensation includes long-term incentive cash awards, fair market value awards of common stock, restricted common stock, performance incentive shares, common stock options, common stock options with stock appreciation rights, and current value awards of dividend and book-value units. In every case, except for dividend and book-value units, we attempted to assess the current market value of the compensation award. For dividend and book-value units, we use the accounting values for the awards. Dollar amounts of compensation for 1985 are adjusted for inflation using the Implicit Price Deflator from The Economic Report of the President (1987). Deflating the 1985 figures makes comparisons with the 1979 figures more meaningful. Table 1 details the approach taken for valuing each type of compensation.

Table 2 summarizes the compensation data for the entire sample and the two subgroups. We draw three preliminary inferences about the relation between investment opportunities and compensation policy. First, total real compensation (deflated to 1979 dollars) increased for both regional and money-center banks. Second, the amount of compensation tied directly to the value of the firm's stock increased for both categories. This increase in incentive compensation is consistent with Crawford, Ezzell, and Miles (1995), who find a significant increase in the pay-performance relation after deregulation. Finally, both the increase in total compensation and the increased incidence of incentive compensation are consistent with a growing set of investment opportunities for large BHCs in the early 1980s.

Comparing the money-center and regional banks reveals that the CEOs of regional BHCs made the greater strides in compensation. Table 2 reveals that, on average, the 1979 salary plus cash bonus portion of total compensation for money-center CEOs at $495,052 is roughly two and one-half times that of regionals, on average. As shown in this table, however, the regional CEOs' average salary plus cash bonus [TABULAR DATA FOR TABLE 2 OMITTED] ($312,971) gained on the money-center CEOs' average salary in real terms between 1979 and 1985. Additionally, the awards made to money-center CEOs in common stock and stock options ($110,357 and $379,002, respectively) were proportionally much larger than the awards made to regional CEOs in 1979. By 1985, however, the awards for CEOs of regional banks had grown substantially to $13,821 and $79,319 for these same two categories. Finally, total compensation for CEOs at money-center BHCs grew in real terms by 50% from 1979 to 1985, rising from $1,208,609 to $1,828,048. In contrast, total compensation for CEOs at regional BHCs roughly doubled, increasing from $238,509 to $463,917. Because regional BHCs had fewer international subsidiaries and fewer loans to less-developed countries, they were more likely to gain from relaxed domestic regulatory constraints.

Following Froot, Scharfstein, and Stein (1993), the compensation data suggest that, on average, regional BHCs did a better job of risk management than money-center BHCs. Froot et al. regard the sole objective of risk management as ensuring that a company has sufficient internal funds to take advantage of attractive investment opportunities. In banking, insufficient internal funding appears as inadequate capital. Both the marketplace and regulators require that BHCs have adequate capital before they can expand. In this regard, the average money-center BHC's mismanagement of less-developed country (LDC) lending sapped its internal funds, drained precious capital, and did not permit it to tap potential growth opportunities to the extent that the average regional BHC did. As a result, the CEOs at money-center BHCs received proportionately smaller rewards than those at regional BHCs.

II. The Empirical Model and lOS Proxies

To investigate the relation between BHC compensation and investment opportunities, we specify an empirical model and attempt to capture investment opportunities with traditional proxies and a new one unique to banking. This section specifies the model first followed by a description of the investment-opportunity set (IOS) proxies.

A. The Empirical Model

To explore the relationship between the size of the IOS and compensation policy, we estimate ordinary-least-squares regression equations of CEO compensation on different measures of the IOS, holding the size of the firm constant. We estimate separate equations for 1979 and 1985 and conduct F-tests for differences between money-center and regional subgroups using the following model:

[Y.sub.j] = [[Beta].sub.0] + [b.sub.1] MKT/BOOK + [[Beta].sub.2] NIR/TR (1)

+ [[Beta].sub.3] SIZE + [[Beta].sub.4] MCD + [[Beta].sub.5]([MCD.sup.*]MKT/BOOK)

+ [[Beta].sub.6] ([MCD.sup.*]NIR/TR) + [[Beta.].sub.7] ([MCD.sup.*]SIZE) + [Epsilon]

Two different measures of compensation are employed as the dependent variable:

[Y.sub.1] = IC/TC = the ratio of incentive compensation to total compensation for the chief executive officer. Incentive compensation includes long-term incentive cash awards, fair market value awards of common stock, restricted common stock, performance incentive shares, common stock options, common stock options with stock-appreciation rights, stock-appreciation rights, and current value awards of dividend and book-value units.

[Y.sub.2] = TC = total compensation for the chief executive officer. Total compensation includes the sum of incentive compensation awards plus salary and bonus, as well as any profit sharing awards in cash.

The explanatory variables are:

MKT/BOOK= the ratio of the market value of common equity to the book value of common equity.

NIR/TR = the ratio of noninterest revenue to total revenue. The noninterest component of revenue includes foreign exchange gains and losses, revenues from direct lease financing, trading account net profits or losses less interest, trust department income, rental income, service charges, collection and exchange charges, commissions, and fees.

SIZE = the natural logarithm of firm value. Firm value is the market value of equity at the end of the year plus the book value of debt.

MCD = a binary variable with a value of 1 if the BHC is a money-center BHC and zero otherwise.

B. Measuring the IOS

The literature contains several proxies for the firm's IOS. For instance, Lewellen, Loderer, and Martin (1987), Smith and Watts (1992), and Gaver and Gaver (1993) all use some form of a ratio that captures a firm's market value and its book value. Lambert and Larcker (1987) employ the growth rate in assets as a proxy for the size of the IOS. Arguing that regulated firms have smaller IOSs than larger ones, Smith and Watts (1984) use a binary variable to indicate whether or not the firm operates in a regulated industry. Lehn and Poulsen (1989) use the growth rate in sales to proxy the IOS for firms going private. Gaver and Gaver (1995) examine four separate measures of the IOS. Through factor analysis, they develop an IOS score that effectively captures the intercorrelations among the four separate measures.

We estimate our model with three different measures of the IOS. First, we use the ratio of the market value of common equity to the book value of common equity (MKT/BOOK). The market value of equity reflects the investors' ownership in the market value of assets-in-place plus the market value of growth options while the book value of equity reflects the investors' ownership in the existing assets of the firm. MKT/BOOK should therefore be higher for firms with greater growth opportunities. The level of incentive compensation and the level of total compensation should be positively related to MKT/BOOK.

We propose a new measure of the size of the IOS unique to banking, the ratio of noninterest revenue to total revenue (NIR/TR). Historically, the bulk of banks' sales have come from interest revenues on loans. During the 1980s competition, deregulation, and financial innovation simultaneously eroded this sales base (e.g., the growth of commercial paper as a substitute for bank loans) while presenting growth opportunities in other areas, notably contingent claims or off-balance-sheet activities, which generate fee income. In addition, the relative decline of traditional commercial lending relationships based on prime-rate pricing and compensating-balance requirements led banks to expand and strengthen customer relationships by selling new fee-based products and services. The removal of interest-rate ceilings on deposits and the Congressional mandate that the Federal Reserve price its services on a cost basis also were part of the general shift from implicit to explicit pricing that occurred in the 1980s.

We use NIR/TR to reflect the off-balance-sheet activities of commercial banks and the fee income generated by these activities and to capture the general shift from implicit pricing to explicit pricing. These noninterest sources of revenue include such nontraditional sources as leasing/rental activities, discount brokerage, investment accounts, merchant banking activities, standby letters of credit, revolving credit agreements, and fees for other services not related to the making of loans. Thus, greater noninterest revenue relative to total revenue suggests in part a larger set of investment opportunities. The level of incentive compensation and the level of total compensation should be positively related to NIR/TR.

The third proxy of the firm's investment opportunities is the natural logarithm of firm size (SIZE). According to Edwards and Heggestad (1973), smaller firms are believed to have greater investment opportunities. This "firm-life-cycle" argument suggests, for example, that regional BHCs, as less mature firms, should have greater growth opportunities than money-center BHCs. The level of incentive compensation and the level of total compensation, therefore, should be negatively related to SIZE. The relation between firm size and compensation could be ambiguous, however. If smaller BHCs are more easily regulated because they have fewer subsidiaries and conduct fewer international transactions, they will possess smaller IOSs than larger firms. Under this alternative explanation, the level of incentive compensation and the level of total compensation should be positively related to SIZE. An alternative view of our tests is that we employ MKT/BOOK and NIR/TR as proxies for banks' investment opportunities with SIZE as a control variable.

III. Empirical Results

We present our empirical findings in three parts. First, we present the results of difference-of-means tests. Then, we present the results of two ordinary least squares regressions using different dependent variables: the ratio of incentive compensation to total compensation (IC/TC) and total compensation (TC).

A. Difference-of-Means Tests

Tables 3 and 4 contain means of the variables used to estimate the regressions. We conduct t-tests for differences in the means between money-center and regional subsamples and for differences in means for each subsample between 1979 and 1985. For both 1979 and 1985, money-center institutions pay higher average total compensation to their CEOs and also pay higher fractions of their total compensation in the form of incentive compensation than do regional institutions. For instance, in 1979 money-center BHCs paid 52% of total compensation in the form of long-term awards while regional BHCs paid only 8%. In 1985, the incentive compensation awards for money-center CEOs were still higher than for regional CEOs relative to total compensation, but the gap between the two narrowed. [TABULAR DATA FOR TABLE 3 OMITTED] Between 1979 and 1985, the regional firms reflect significant increases in total real compensation and in the fraction of total compensation paid in the form of incentive compensation. The money-center firms, however, reveal no significant changes from 1979 to 1985 in either total compensation or long-term awards as a fraction of total compensation.

Tables 3 and 4 also allow us to draw inferences about differences in the IOS between the money-center and regional subsamples and about changes in the IOS between 1979 and 1985. In 1979, our first two measures of the IOS (MKT/BOOK and NIR/TR) are not statistically different between money-center BHCs and regional BHCs, while firm size differs significantly between these two groups. In 1985, on the other hand, the market-to-book ratio is significantly higher for regional banks, indicating a larger IOS for regional banks or an adverse effect of the LDC crisis on money-center banks or both. Since the log of firm value primarily reflects size differences between money-center and regional BHCs, it can also be viewed as a control variable as well as a proxy for investment opportunities. Our results for 1979 and 1985 suggest that log [TABULAR DATA FOR TABLE 4 OMITTED] of firm value is not the best measure of investment opportunities; it simply appears to reflect size differences between money-center and regional banks. Regional banks, however, did experience a significant increase in firm value between 1979 and 1985 whereas money-center banks did not. The differential effects of the international debt crisis across money-center and regional banks no doubt played a major role in explaining these changes in firm values. Furthermore, regional banks experienced greater percentage increases in the market-to-book ratio and the ratio of noninterest revenue to total revenue than did the money-center banks.

The evidence presented so far suggests several conclusions. First, while the IOS increased for all banks between 1979 and 1985, the increase was greater for regional banks than for money-center banks. Second, the relative growth of investment opportunities for regional banks is accompanied by statistically significant increases in both total compensation and the proportion of incentive compensation. Finally, while the incidence of stock-based incentive compensation increased for both money-center banks and regional banks, the increase was greater for regional banks.

B. Regressions on the Ratio of Incentive Compensation to Total Compensation

Table 5 contains estimates of Equation (1) with the ratio of incentive compensation to total compensation as the dependent variable. Although we present separate estimates of Equation (1) for 1979 (Regression 5.1) and 1985 (Regression 5.3), the F-tests indicated no structural differences in the models between the two years. To highlight the relations between investment opportunities and incentive compensation without controlling for differences between money-center and regional banks, we add intercept and slope binaries to regressions 5.1 (1979) and 5.3 (1985). As a result, regressions 5.2 (1979) and 5.4 (1985) specify Equation (1) more completely, allowing us to compare the determinants of CEO incentive compensation for money-center banks with those for regional banks.(4)

The significance of the ratio of noninterest revenue to total revenue (NIR/TR) in both regressions 5.3 and 5.4 suggests a positive relation between the size of the IOS and incentive compensation in 1985. In addition, the coefficient of firm value (SIZE) is positive and statistically significant in both time periods (Regressions 5.1 and 5.3). When we add the intercept and slope binaries for money-center banks to the equation in both time periods (Regressions 5.2 and 5.4), however, the coefficient of SIZE is no longer significant, suggesting that the money-center binaries are capturing differences in firm value between money-center and regional banks, capturing the differential effects of the LDC crisis, or both.

In 1985 (Regression 5.4), the money-center intercept binary (MCD) is significant and negative, suggesting that, after holding firm value and investment opportunities constant, CEOs of money-center banks receive a lower proportion of their compensation in the form of incentive compensation than do CEOs of regional banks. The positive and significant coefficient of the MCD*SIZE slope binary in regression 5.4 suggests, however, that the disparity in incentive compensation between money-center and regional CEOs gets smaller, the larger the bank. In other words, the structure of the compensation package for large regional banks is similar to that of the money-center banks. These results suggest that the smaller regional banks offer a higher proportion of incentive compensation than do larger regional banks, a result that appears consistent with the notion that larger, more mature firms have fewer new investment opportunities. These results are also consistent with the interpretation that the CEOs of the money-center banks receive higher proportions of incentive compensation than the CEOs of the largest regional banks.

Finally, following Froot et al. (1993), any firm that mismanages assets-in-place (e.g., LDC loans) will not have sufficient internal funds to expand. Thus, while it is likely (for reasons explained above) that money-center BHCs had fewer growth options, the LDC debt crisis made it difficult for them to tap the opportunities they did have.

C. Total Compensation Regressions

Table 6 contains our estimates of Equation (1) with total compensation as the dependent variable. We present separate estimates of Equation (1) for 1979 and 1985 because F-tests suggest that data for the two years are not poolable. Regressions 6.1 (1979) and 6.3 (1985) do not distinguish between total compensation effects for regional or money-center BHCs, while regressions 6.2 (1979) and 6.4 (1985) add intercept and slope binaries to highlight the differences in total compensation effects for the two groups.

In the basic regression for 1979 (Regression 6.1), SIZE is the only proxy of the IOS with a statistically significant coefficient. In the 1985 equation (Regression 6.3), however, the proportion of noninterest revenue (NIR/TR) is also statistically significant. We conclude that total compensation to bank CEOs is more directly driven by noninterest revenue in 1985 than in 1979. This result appears consistent with our argument that investment opportunities increased for banks in the 1980s because of an expansion of services into nontraditional arenas. It also suggests a positive relation between compensation and the IOS.

Regressions 6.2 and 6.4 allow us to compare how CEO compensation at money-center banks and regional banks differs between 1979 and 1985. The money-center intercept binary (MCD) is not statistically significant in 1979 (Regression 6.2), suggesting no difference in total compensation between money-center and regional CEOs, holding firm size and investment opportunities constant. Of the slope binaries, only (MCD)*(MKT/BOOK) is statistically significant. Its positive sign suggests that in 1979, total compensation of money-center CEOs was more sensitive to changes in the market-to-book ratio than the total compensation of regional bank CEOs. Insofar as MKT/BOOK reflects investment opportunities, it is [TABULAR DATA FOR TABLE 5 OMITTED] conceivable that money-center banks had larger IOSs than regional banks in 1979. Keep in mind that money-center banks had international opportunities not accessible to regional banks and that the new opportunities in domestic markets fostered by the more relaxed regulatory attitudes of the early 1980s had not yet appeared.

In 1985 (Regression 6.4), on the other hand, the money-center intercept binary (MCD) and the slope binaries for firm value (SIZE) and noninterest revenue (NIR/TR) are statistically significant. The negative sign on the coefficient of the intercept binary suggests that CEOs of money-center banks receive less compensation than their regional counterparts, ceteris paribus. While money-center CEOs earn less, ceteris paribus, the difference between their compensation and that of regional CEOs is smaller the larger is the ratio of noninterest revenue to total revenue (NIR/TR) and the larger is the log of firm value (SIZE). This result suggests that differences in compensation between larger regional banks and money-center banks are much less than the differences between smaller regional banks and money-center banks. These results and similar results from Table 5 support the argument of Edwards and Heggestad (1973) that less mature, smaller firms have better investment opportunities.

Finally, because of the LDC debt crisis, money-center BHCs had insufficient internal funds in 1985 to tap the available investment opportunities (Froot et al., 1993). Such was not the case for regional BHCs in 1985, nor was it the case for money-center banks in 1979 prior to the LDC debt crisis.

[TABULAR DATA FOR TABLE 6 OMITTED]

IV. Summary and Conclusions

Agency theory suggests a positive relationship between the level of executive compensation, the proportion of incentive compensation, and the size of the firm's IOS. Potential proxies for banks' investment opportunities include firm size, the market-to-book ratio, and a proxy unique to our sample of large bank holding companies, the proportion of noninterest revenue-to-total revenue.

Historically, the bulk of banks' revenue has come from interest on loans and investment securities. During the 1980s, competition, deregulation, and financial innovation simultaneously eroded this base of interest revenue (e.g., the growth of commercial paper as a substitute for bank loans) while presenting growth opportunities in other areas, notably contingent claims or off-balance-sheet activities where revenues take the form of fee income. In addition, the decline of the prime-rate and compensating-balance conventions led banks to expand and strengthen customer relationships by selling new fee-based products and services. The removal of interest-rate ceilings on deposits and the Congressional mandate that the Federal Reserve price its services to member banks on a cost basis also were part of the general shift from implicit to explicit pricing that occurred in the 1980s.

Our empirical findings show that from 1979 to 1985 total real compensation and the ratio of incentive compensation-to-total compensation increased substantially at regional BHCs but remained relatively stable at money-center BHCs. These results are consistent with greater growth opportunities at regional banks relative to money-center BHCs. Since the international debt crisis primarily affected money-center banks and seriously depleted their real net-worth positions, these companies seem to have had insufficient internal funds with which to fully tap growth opportunities. In contrast, regional banks had only limited LDC exposure, and through regional interstate banking compacts, they were presented with growth options not available to money-center firms. Using their stronger capital positions, many regional BHCs grew to become "superregional banks." Agency theory suggests that CEOs should be rewarded for exercising these growth options. Our findings indicate that the amount of total compensation and the incidence of long-term incentive compensation of large BHCs were positively related to the expanded investment opportunities available to banks during the 1980s. Thus, we show that the observed level and structure of executive compensation are consistent with agency theory.

The authors are grateful for comments on earlier drafts of this manuscript from Jennifer Gaver, Ed Kane, Ken Wiles, and the Editors. We are also thankful for comments received from seminar participants at Arizona State University, the 1990 Financial Management Association Annual Meeting, The University of Arizona, and The University of Tennessee. The authors are responsible for any remaining errors.

1 See Smith and Watts (1992) and Gaver and Gaver (1993).

2 We use the terms bank and bank holding company interchangeably.

3 Collins, Blackwell, and Sinkey (1994) document the effects of financial innovation and deregulation on bank investment opportunities.

4 Market-to-book and noninterest revenue-to-total revenue are significantly correlated (p = 0.56), but regressions using each measure separately produced results similar to those reported in Tables 5 and 6.

References

Black, F. and M. Scholes, 1973, "The Pricing of Options and Corporate Liabilities," Journal of Political Economy (May-June), 399-418.

Black, H.A., M.A. Fields, and R.L. Schweitzer, 1990, "Changes in Interstate Banking Laws: The Impact on Shareholder Wealth," Journal of Finance (December), 1663-1671.

Chow, G.C., 1960, "Tests of Equality Between Sets of Coefficients in Two Linear Regressions," Econometrica (July), 591-605.

Collins, M.C., D.W. Blackwell, and J.F. Sinkey, 1994, "Financial Innovation, Investment Opportunities, and Corporate Policy Choices for Large Bank Holding Companies," Financial Review (May), 223-247.

Crawford, A.J., J.R. Ezzell, and J.A. Miles, 1995, "Bank CEO Pay-Performance Relations and the Effects of Deregulation," Journal of Business (April), 231-256.

Economic Report of the President, 1987, Washington, D.C., United States Government Printing Office, 251.

Edwards, F.R. and A.A. Heggestad, 1973, "Uncertainty, Market Structure and Performance: The Galbraith-Caves Hypothesis and Managerial Motives in Banking," Quarterly Journal of Economics (August), 455-473.

Ellig, B.R., 1982, Executive Compensation - A Total Pay Perspective, New York, NY, McGraw-Hill.

Froot, K.A., D.S. Scharfstein, and J.C. Stein, 1993, "Risk Management: Coordinating Corporate Investment and Financing Policies," Journal of Finance (December), 1629-1658.

Gaver, J.J. and K.M. Gaver, 1993, "Additional Evidence on the Association Between the Investment Opportunity Set and Corporate Financing, Dividend, and Compensation Policies," Journal of Accounting and Economics (January/April/June), 125-160.

Gaver, J.J. and K.M. Gaver, 1995, "Compensation Policy and the Investment Opportunity Set," Financial Management (Spring), 19-32.

Lambert, R.A. and D.F. Larcker, 1987, "An Analysis of the Use of Accounting and Market Measures of Performance in Executive Compensation Contracts," Journal of Accounting Research (Supplement), 85-125.

Lehn, K. and A. Poulsen, 1989, "Free Cash Flow and Stockholder Gains in Going Private Transactions," Journal of Finance (July), 771-787.

Lewellen, W., C. Loderer, and K. Martin, 1987, "Executive Compensation Contracts and Executive Incentive Problems: An Empirical Analysis," Journal of Accounting and Economics (December), 287-310.

McLaughlin, M.M. and M.H. Wolfson, 1988, "The Profitability of Insured Commercial Banks in 1987," Federal Reserve Bulletin (January/April), 403-418.

Satterthwaite, F.W., 1946, "An Approximate Distribution of Estimates of Variance Components," Biometrics Bulletin (December), 110-114.

Sinkey, J.F., Jr., 1983, Commercial Bank Financial Management, New York, NY, Macmillan.

Smith, C.W., Jr. and R. Watts, 1992, "The Investment Opportunity Set and Corporate Financing, Dividend, and Compensation Policies," Journal of Financial Economics (December), 263-292.

Smith, C.W., Jr. and R. Watts, 1982, "Incentive and Tax Effects of U.S. Executive Compensation Plans," Australian Journal of Management (December), 139-157.

Smith, C.W., Jr. and R. Watts, 1984, "The Structure of Executive Compensation Contracts and the Control of Management," University of Rochester Working Paper.

Appendix: Sample BHCs Listed by Money-Center Versus Regional Designation

Panel A: Money-center BHCs(a)

Bank America Corp. Bankers Trust NY Corp. Chase Manhattan Corp. Citicorp Continental Illinois Corp. First Chicago Corp. Morgan (J.P.) & Co.

Panel B. Regional BHCs

Affiliated Bankshares, CO Allied Bankcshares Inc. - TX Ameritrust Corp. Bank One Corp. Bancoklohoma Corp. Bancorp Hawaii Inc. Bank of New York Co. Inc. Barnett Banks Inc. Baybanks Inc. Boatmen's Bancshares Inc. Central Fidelity Banks Inc. Colorado National Bankshares Commerce Bancshares Inc. Dauphin Deposit Corp. Deposit Guaranty Corp. Dominion Bankshares Corp. Fidelcor First Bank System Inc. First Florida Banks Inc. First Hawaiian Inc. First Jersey National Corp. First Kentucky National First Maryland Bancorp First National Cincinnati Corp. First Pennsylvania Corp. First Security Corp. - Del First Tennessee National Corp. First Union Corp. (N.C.) First Virginia Banks Inc. First Wachovia Corp. First Wisconsin Corp. Florida National Banks of Florida Hartford National Corp. Indiana National Corp. Irving Bank Corp. Manufactures National Marine Corp. - WISC Marshall & Ilsley Corp. Mercantile Bancorporation Mercantile Bankshares Michigan National Corp. NCNB Corp. Old Kent Financial Corp. PNC Financial Corp. Rainier Bancorporation Republic New York Corp. Security Pacific Corp. Society Corp. South Carolina National Corp. Southeast Banking Corp. State Street Boston Corp. Texas American Bancshares Texas Commerce Bancshares U.S. Bancorp U.S. Trust Corp. Union Planters Corp. United Banks of Colorado United Jersey Banks Wells Fargo & Co.

a Since proxy statements for two traditional members of the money-center group, Chemical and Manufacturers Hanover, were unavailable, we omitted them from the analysis.

M. Cary Collins is Associate Professor of Finance at The University of Tennessee at Knoxville, TN. David W. Blackwell is Associate Professor of Finance at the Goizueta Business School at Emory University, Atlanta, GA. Joseph F. Sinkey, Jr. is Edward W. Hiles Professor of Financial Institutions at The University of Georgia, Athens, GA.
COPYRIGHT 1995 Financial Management Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1995 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:includes appendix
Author:Collins, M. Cary; Blackwell, David W.; Sinkey, Joseph F., Jr.
Publication:Financial Management
Date:Sep 22, 1995
Words:6637
Previous Article:Who opts out of state antitakeover protection?: the case of Pennsylvania's SB 1310.
Next Article:Monitoring versus bonding: shareholder rights and management compensation.
Topics:


Related Articles
Management ownership and firm compensation policy: evidence from converting savings and loan associations.
Compensation policy and the investment opportunity set.
Monitoring versus bonding: shareholder rights and management compensation.
The director as employee of management.
Tax policy and investment.
Uncertainty in Executive Compensation and Capital Investment: A Panel Study.
Firm performance and executive compensation in Australia and Canada.
Venture capitalists as economic principals.
Choosing the best compensation program: fee or non-fee? It's a quandary many corporate decision-makers encounter when choosing cash managers. A noted...

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |