A longitudinal study of borrowing by large American corporations.
Richardson, R. J. 1987 "Directorship interlocks and corporate profitability." Administrative Science Quarterly, 32: 367-386.
Roy, William G. 1983 "The unfolding of the interlocking directorate structure of the United States." American Sociological Review, 48: 248-257.
Corporate managers make strategic decisions on a wide range of issues, from mergers and acquisitions to research and development, and anticipated economic costs and benefits of a policy always figure into their decision making. Firm officials are concerned with improving such factors as the organization's growth, profitability, and market share, but the methods by which firms pursue these economic goals are by no means agreed upon. Company officials may disagree on a common measure of firm performance, and particular firms often rely on different measures under different circumstances (Meyer, 1993). Whether a firm engages in a vertical integration strategy or a diversification strategy may be a matter of considerable dispute among a firm's officials, and different firms may employ different strategies because of or regardless of their current financial conditions.
In recent years organizational researchers have devoted increasing attention to the determinants of these corporate strategies. We have seen a flurry of papers on topics such as mergers and acquisitions, adoption of the multidivisional form, make-or-buy decisions, takeover prevention strategies, research and development expenditures, chief executive officer compensation, and political and charitable contributions. Organizational researchers have found that firm strategies in these areas can be accounted for in part by noneconomic phenomena, including a firm's ownership structure, the actions of its peers, and the firm's social ties with other firms. But one crucial strategic decision has received very little attention.
Among the most important decisions that managers make are those involving the firm's financing and capital structure. As Barton and Gordon (1987: 67) noted, "The question of how to finance the firm . . . represents a fundamental functional (financial) decision which should support and be consistent with the long-term strategy of the firm." Yet, with the exception of two studies by Donaldson (1961, 1969), there has been virtually no work on this topic among organizational researchers. This may not be surprising, since financing decisions could be viewed as among the most purely economic decisions that a firm makes. But finance scholars, although they agree on what firms should do, have failed to reach a consensus on how borrowing decisions are actually made (Myers, 1977, 1984; Barton and Gordon, 1987; MacKie-Mason, 1990). The wide range of options open to firms as well as the many alternative ways of evaluating capital projects are two possible reasons for this lack of consensus. We believe that organizational theory may be able to contribute to understanding even this most economic of corporate strategies. In this paper, we draw on several currently prominent organizational models to investigate the role of economic, organizational, and institutional factors in corporate debt financing. Using a time-series model, we then test the ability of these models to account for the borrowing patterns of 22 large U.S. manufacturing corporations over a 28-year period.
Financial Dependence and Managerial Autonomy
The extent to which firms have actually depended on external financing over the years has been debated for decades. Business historians agree that in the early twentieth century, financiers such as J. P. Morgan wielded enormous power within the business community. Investment bankers played an important role in corporate formations and mergers because of their central position in the issuance and sale of new securities (Navin and Sears, 1955; Carosso, 1970; Mizruchi, 1982; Roy, 1983). As the major providers of capital in a period of intense competition and increasing concentration, investment banks appropriated a major share of the promoters' profits for themselves, appointed their own representatives to corporations' boards, and exerted influence over corporate policy (Sweezy, 1956; Chandler, 1977). This period is often referred to as the era of finance capital (Cochran and Miller, 1942).
After 1920, the relationship between investment banks and corporations changed (Sweezy, 1956; Berle and Means, 1968). According to Sweezy (1956), the specter of cutthroat competition ceased in most industrial sectors as large industrial monopolies came to dominate the market. Fewer combinations occurred and in some industries ceased altogether. Investment banks' power decreased correspondingly as issuing new securities, the basis of their power, became less important. Between the 1920s and the 1970s, theories of finance capital virtually disappeared from economic and business writings. Managerialism became the dominant theory of corporate control. According to managerialists, because corporations could successfully and regularly produce enough internal funds (Berle, 1954; Dahrendorf, 1959; Bell, 1961; Baran and Sweezy, 1966; Galbraith, 1967), they no longer had to depend on external capital sources. Oligopolistic market structures ensured adequate profits, and the state, as a major consumer, intervened to prevent economic crises (Baran and Sweezy, 1966; Galbraith, 1967). As a result, firms could finance their own investments and thereby escape dependence on financial institutions. According to Galbraith (1967: 92-93), "Few other developments can have more fundamentally altered the character of capitalism."
Nevertheless, two studies (Lintner, 1966; Stearns, 1986) showed that the managerialists had underestimated corporations' use of external capital sources. Lintner (1966) found that from 1900 through 1953, nonfinancial corporations consistently met between 40 and 45 percent of their total current financial needs with external funds. In addition, although large manufacturing firms relied primarily on internal funds in the 1920s, they increasingly drew on external funding over the next 30 years, a finding directly contrary to the managerialist argument. Stearns (1986) showed that the proportion of total funds obtained from external sources fluctuated after World War II. Although corporations obtained approximately one-third of their total funds from external sources between 1946 and 1965, they obtained almost half of their total funds from these sources between 1965 and 1980. Critics have countered such evidence by maintaining that although corporations do borrow, their borrowing is discretionary (Baran and Sweezy, 1966: 15-16). Instead, they argue, corporations borrow to take advantage of favorable interest rates, tax benefits (Modigliani and Miller, 1963), and investment opportunities (Herman, 1981).
The issues of financial dependence and managerial autonomy continue to influence research on corporations. Pfeffer and Salancik (1978) focused on the limitations of managerial autonomy and organizations' dependence on external resources, including capital. Williamson (1988), in a recent work on financing, emphasized the problematic nature of relations between firms and their capital suppliers. Mintz and Schwartz (1985) argued that corporations are typically dependent on financial institutions for financing. And several theorists within the institutional and network perspectives have suggested that managers are constrained by other actors in their organizational fields, including those on whom they depend for resources (Burt, 1983; DiMaggio and Powell, 1983; Granovetter, 1985). But organizational theorists have dealt with borrowing only tangentially, and almost entirely as an exogenous variable. In this paper we focus directly on the organizational determinants of firm borrowing.
Determinants of Corporate Borrowing
Our examination of corporate borrowing draws on several major perspectives in organizational theory and financial economics. Existing theory leads us to specify four general determinants of a firm's propensity to borrow: the expected return on borrowing, the availability of internal funds, the strategic orientation of the CEO, and the firm's board composition.
Expected return on borrowing. Finance economists have developed a large prescriptive literature detailing the conditions under which corporations should borrow. The basic rule is that a firm should not borrow unless its expected returns exceed the cost of funds. The primary determinants of the amount a firm borrows, according to this model, should be the cost of the funds and the anticipated future earnings on the investments made with them.
Finance economists have devised several formulas to help managers identify profitable investments. Historically, managers have had considerable discretion in deciding which of these prescriptions to use. Most recently, the preferred formula has been the net present value (NPV) method (Weston and Brigham, 1993). This method determines the profitability of an investment by discounting the anticipated future payoff of an investment by the firm's cost of capital. The discounted cash flow is then summed to compute the project's NPV. NPV is thus defined as
[C.sub.0] + [C.sub.1]/(1 + r) + [C.sub.2]/[(1 + r).sup.2] + . . . + [C.sub.t]/[(1 + r).sup.t],
where [C.sub.0] is the cash outlay at time 0 (that is, the present; this value is negative to indicate the investment on which the return is anticipated), [C.sub.1] through [C.sub.t] are the cash amounts that will be received from the investment in each of the following t years, and r is an anticipated interest rate referred to as the "discount rate." If the NPV is positive, that is, if [C.sub.0] [is less than] [summation of] [C.sub.t]/[(1 + r).sup.t], the project should be accepted. If the NPV is negative, that is, if [C.sub.0] [is greater than] [summation of] [C.sub.t]/[(1 + r).sup.t], the project should be rejected (Brealey and Myers, 1991).
The discount rate varies by company and project based on the risk involved. In an attempt to calculate the risk factor accurately, finance economists devised the capital asset pricing model (CAPM) (Sharpe, 1964). Capital asset pricing theory states that the required rate of return depends on a value referred to as "beta" ([Beta]). In evaluating the relevant rate of return on equity, for example, CAPM risk assessments are computed by the formula
[K.sub.A] = [K.sub.rf] + ([K.sub.p] - [K.sub.rf]) [[Beta].sub.A],
where [K.sub.A] is the required rate of return on firm A's stock, [K.sub.rf] is the rate of return on the most risk-free investment (generally the current Treasury bill rate), [K.sub.p] is the required rate of return on a diversified portfolio of stocks (such as the Standard and Poor's 500 stock index), and [[Beta].sub.A] measures the past volatility of return of firm A's stock relative to the average return on the market during the same period. The beta thus adjusts the rate of return to include the level of risk associated with investing in firm A's stock (see Weston and Brigham, 1993: 182-185 for details on calculating [Beta]).
When the CAPM is used to evaluate the risk and relevant rate of return on debt, adjustments to the firm's beta are required. This can be done by estimating the expected return on each of the securities and then taking a weighted average of these separate returns (Brealey and Myers, 1991). The extent to which a firm borrows funds to finance an investment depends on whether these adjustments decrease the cost of capital. This raises the expected rate of return by increasing the difference between an investment's anticipated future earnings and its cost. Therefore, according to this model, corporations will borrow funds only when their expected returns exceed the cost of capital. This suggests the following:
Hypothesis 1: A firm's level of borrowing will be positively associated with the extent to which the expected returns exceed the cost of funds.
While this hypothesis is a reasonable derivation from the theory, capital asset pricing models have come under attack in recent years (Nichols, 1993). Fama and French (1992) argued that beta is an inappropriate measure of risk, raising the question of whether risk is related to returns in the way financial theorists have suggested. Lowenstein (1991) suggested that the CAPM has contributed to the decline in American competitiveness by limiting managers' investment choices to only safe projects with clear, short-term returns. Such attacks do not alter the basic rule that expected returns must exceed the cost of capital, but they do increase managerial uncertainty. It becomes less clear to managers what analytical measures to use when making investment decisions and how to interpret the results of these measures.
Availability of internal funds. Although economic theory predicts that cost and anticipated future earnings are the primary determinants of firm borrowing, several economists and organizational theorists have suggested alternative factors. In two studies from the 1960s, Donaldson (1961, 1969) found that the firms he studied did not always behave as conventional economic models would expect. Rather, firms tended to rely more on internal funds (and less on external funds) than prevailing theories would have predicted. Two decades later, Stewart Myers (1984), in his presidential address to the American Finance Association, introduced what he called a "modified pecking order" theory of financing, based in part on Donaldson's studies. In this model, retained earnings are seen as a more desirable form of capital than are external funds, primarily because the use of external financing reduces managerial autonomy. This view corresponds to several widely employed models within organizational theory.
Pfeffer and Salancik (1978) argued that an organization's autonomy in relation to its environment is a function of its ability to gain access to resources that are controlled by other organizations. To the extent that capital is a crucial resource and a firm depends on other organizations to provide its capital, the suppliers of capital will exercise power over the firm. Under these circumstances, firms will attempt to minimize their dependence on external financing. The ability to use internally generated capital would thus appear to be an important goal of management. Mintz and Schwartz (1985) have also argued that the availability of capital is a key determinant of managerial autonomy. To the extent that internal funds are abundant, firms would be expected to achieve a certain degree of independence from financial institutions.
In Williamson's (1988) related model, the uses of both debt and equity (stock issues) as financing mechanisms are fraught with peril, because both involve the potential surrender of autonomy. Debt creates transaction costs with financial institutions that include the possibility that a firm will lose autonomy in its decision making. The model thus predicts that unless the costs of capital are low enough to justify external financing, firms will attempt to maximize their use of internally generated funds because external financing involves control costs.
Although a firm with a high level of internal resources may be less likely to borrow, this will not necessarily always be the case. First, a firm's financing needs may be so extensive that even a high level of retained earnings is insufficient. Second, even a firm with a high level of retained earnings might borrow extensively if, as the model presented in the previous section suggests, the expected return makes it economically efficient to do so. Based on both Myers' modified pecking-order theory and the organizational models of Pfeffer and Salancik, Mintz and Schwartz, and Williamson, we believe that, other things being equal, firms will prefer to use retained earnings. Thus, we hypothesize:
Hypothesis 2: A firm's level of retained earnings is inversely related to its level of borrowing.
The strategic orientation of the CEO. Corporate strategies are determined in part by the social definition of what is both possible and appropriate behavior. Fligstein (1990) argued that during the course of the twentieth century, the prescribed form of corporate strategy has shifted from an emphasis on manufacturing to one on sales, to its present focus on finance. In the finance conception, the firm is seen simply as a bundle of assets on which return, usually short-term, is to be maximized, regardless of how or in what sector these profits are realized. This is distinct from the manufacturing conception, for example, in which the focus is on the production of specific goods in which the firm specializes. The prevailing strategic conception, according to this model, will be reflected in the backgrounds of the firm's top managers. Firms in which the finance conception is dominant will be expected to have CEOs with backgrounds in finance or accounting. According to Fligstein, a primary strategy of firms operating under a finance conception of control is growth. Since the quickest route to growth is through acquisition, Fligstein suggested that finance-controlled firms will be more likely than non-finance-controlled firms to engage in acquisitions. Because external financing is usually necessary for acquisitions, firms run according to a finance conception of control will be more likely to use debt than will non-finance-conception firms (Fligstein, 1990: 15). This suggests the following hypothesis:
Hypothesis 3: Firms whose CEOs come from financial backgrounds will have higher levels of borrowing than will firms whose CEOs come from non-finance backgrounds.
The firm's board composition. Corporate decision making does not operate in a vacuum. Firms are influenced by the actions of other firms in their environment as well as by interests within the firm's own decision-making apparatus. One important source of influence is the institutions represented on the firm's board of directors. A substantial and rapidly growing literature suggests that a firm's board composition has an important impact on its strategic decision making. In a study of the use of "greenmail," the private repurchase of company stock, Kosnik (1987) found that firms that resisted greenmail had more outside directors and more directors who represented firms with which the focal firm had transactions than did firms that paid greenmail. Studies by Cochran, Wood, and Jones (1985), Singh and Harianto (1989), and Wade, O'Reilly, and Chandratat (1990) found that the proportion of outside directors on a firm's board was positively associated with the existence of "golden parachute" policies for the firm's top executives. Davis (1991), in a study of firms' adoption of takeover defenses (known as "poison pills"), found that firms were more likely to adopt poison pills when they shared directors with firms that had already adopted. Baysinger, Kosnik, and Turk (1991) found a negative association between the proportion of outside board members and firms' R&D expenditures. Clawson and Neustadtl (1989) and Mizruchi (1992) found that director interlocks influenced corporate political strategies. Haunschild (1993), Fligstein and Markowitz (1993), and Palmer et al. (1993) found an association between interlocks and firms' participation in mergers and acquisitions. And Boeker and Goodstein (1993) found that firms with a higher proportion of outsiders on their boards were more likely to appoint CEOs from outside the firm.
Just as the presence of a CEO with a financial background may affect a firm's decision to borrow, so may the presence of representatives of financial institutions on the firm's board. These directors may offer advice and information as well as access to capital. A board member who is also an executive of a financial institution is a source of financial advice and information. Such directors often sit on the firm's finance committee (Bearden, 1986).
A firm may also decide to establish a relation with a financial institution to facilitate access to external funding (Pfeffer and Salancik, 1978; Burt, 1983). In a tight capital market, firms without such interlocks may be less successful in obtaining funds. An interlock between two firms often reflects an already established business relation (Palmer, Friedland, and Singh, 1986; Stearns and Mizruchi, 1986; Baker, 1990). This relation may consist of positive past experiences that would encourage further transactions (Granovetter, 1985) or an unequal distribution of power that could mandate them (Richardson, 1987; Mizruchi and Stearns, 1988). Among both bank hegemony theorists (Mintz and Schwartz, 1985) and agency theorists (Stiglitz, 1985; Eisenhardt, 1989), the presence of a representative of a financial institution on the borrowing firm's board of directors is viewed as an important monitoring mechanism. To the extent that financial institutions are able to monitor their loans, they will be more willing to lend money to a firm. Based on the preceding discussion, firms with representatives of financial institutions on their boards will be more likely to employ external financing than will firms without such representatives:
Hypothesis 4: Firms with financial representation on their boards of directors will borrow more than will firms without financial representation.
Stage of the business cycle. In an earlier paper (Mizruchi and Stearns, 1988), we argued that the strategies of individual firms cannot be understood without knowing the general economic environment within which decision making occurs. One such factor that could affect a firm's level of borrowing is the overall condition of the economy, as measured by the phase of the business cycle. Corporations generally borrow during the expansion phase of the business cycle, when investment prospects are brightest. Short-term funds replenish depleted inventories and finance peak selling cycles, while long-term funds are invested in new capital expenditures (plant enlargement or modernization and new equipment, for example). A study by Stearns (1986), however, showed that the severity of the recessions of 1974-1975 and 1979-1980 prevented firms from leaving the long-term market, as they normally would during contraction periods. Firms continued to borrow long-term funds to meet payoff schedules on their short-term debt and to provide the working capital necessary to remain afloat. Therefore, although we would normally expect to observe a positive association between the expansion stage of the business cycle and borrowing, we do not hypothesize this association in the present study, although we did include stage of the business cycle in our analysis.
The economy-wide level of interest may also be a significant predictor of corporate borrowing. Hypothesis 1 takes into account the firm-specific cost of borrowing and its relationship to a firm's anticipated return. But it may also be that the overall interest rate affects the amount of borrowing in which firms engage. We therefore also accounted for the economy-wide interest rate. Because economy-wide interest rates may be driven up during periods in which the demand for capital is high, it is possible that we will observe a positive association between interest rates and borrowing. To reduce this possibility, we included a variable for the economy-wide level of capital demand.
Corporate borrowing may be affected by additional factors not described in the above hypotheses. Some of these have been theorized by economists and organizational theorists, while others have received little theoretical attention. It is necessary to control for these factors in order to ensure that the effects of our hypothesized variables are not spurious consequences of alternative factors. We do not specify formal hypotheses for these factors, although we do mention possible predicted effects where relevant.
Type of control. Managerialists (Galbraith, 1967; Berle and Means, 1968) have suggested that widespread stock dispersion decreases stockholders' power to press for high dividend payouts. Although shareholders may prefer high dividend payouts, managers prefer to keep dividends low to maintain their resources and, hence, their autonomy. Without adequate internal funds, managers are forced to obtain external capital, thereby subjecting themselves to the monitoring of the capital markets. This suggests that management-controlled firms will be less likely to borrow than will owner-controlled firms. According to agency theorists (Grossman and Hart, 1982; Jensen, 1986), by contrast, management-controlled firms may be more prone to borrow, since debt financing mitigates the conflict between managers and equity holders. By increasing the likelihood of bankruptcy, the use of debt creates incentives for managers to work harder, consume fewer perquisites, and make better investments. This would lead to the prediction that corporations under management control would make more use of external financing than firms under family control. Because we believe that both hypotheses are plausible, we do not offer a specific prediction for the effect of this variable but include it as a control.
Alternative forms of financing. While firms that are unable to use retained earnings for their financing may be forced to borrow, there are alternative sources of financing that enable managers to avoid dependence on financial institutions. One alternative is the public sale of stock. Equity financing, as this is called, is widely seen in the finance literature as an alternative to debt financing (Williamson, 1988). The sale of public stock could be either positively or negatively associated with debt financing. It would be positively associated if both debt and equity were used simultaneously to raise capital. It would be negatively associated if equity were used primarily as a substitute for debt. Regardless of which is the case, it is necessary to control for this variable.(1)
In some cases, firms issue stock specifically for the purpose of financing an acquisition, and the firm does not actually sell its stock on the market. Instead, management promises to provide a certain number of its shares to the stockholders of the target firm as a means of payment. By doing this, the acquiring firm reduces the amount of cash it needs to borrow. The issue of stock to finance an acquisition should not necessarily be negatively associated with the use of debt financing, however, because firms require large sums during an acquisition and are likely to borrow from external sources in addition to any alternatives that might be used. Since the existence of stock issues for the purposes of acquisitions might affect the level of borrowing, we also controlled for this.
Another alternative source of financing is an employee stock ownership plan (ESOP). Establishing ESOPs can provide additional funds to the firm, directly or indirectly. By issuing stock to its employees, the firm's management can raise capital while maintaining control over the firm's operations. In addition, many large corporations establish ESOPs to take advantage of the corporate tax credits accompanying them. These credits lower firms' income taxes, thus increasing their internal funds. Theoretically, employees who have amassed enough stock could take over a firm's operations, but in actuality this has rarely occurred. Few ESOPs involve majority ownership of firms, and the few that do have often retained existing management. We expect that firms with ESOPs will be less likely to use external financing than will firms without ESOPs.
Debt ratio. Our dependent variable is the amount of debt financing initiated by a firm in a given year. We are concerned not with the firm's level of debt financing per se but, rather, with a decision to borrow at a particular time. Whether a firm will attempt to borrow will depend partly on its current level of indebtedness. A firm that recently secured a large long-term loan may be less likely to borrow in the subsequent year. At the same time, firms that have used external financing in the past may be more likely to borrow again in the future. Agency theorists argue that as a firm's debt ratio increases, it becomes less risk averse. In this view, highly leveraged firms have an economic incentive to borrow. If a risky project is successful, the firm captures most of the gains; if it turns out poorly, the creditors bear most of the costs (Jensen and Meckling, 1976). Other observers suggest that firms with low debt ratios take on additional debt to render themselves less attractive as takeover targets (Davis, 1991). But regardless of whether a firm's level of indebtedness is positively or negatively related to the amount of external financing its management uses, it is likely that the firm's debt ratio is taken into account by both the firm's management and financial institutions when such a decision is made (Barton and Gordon, 1987). We therefore controlled for the firm's debt ratio.
Industry. It is known that members of different industries vary in their use of external financing (Gupta, 1969; MacKie-Mason, 1990). Capital-intensive industries may have larger financing requirements than will labor-intensive ones. tn order to ensure that the effects of the theoretically specified variables are not spurious consequences of industry-specific characteristics, we included dummy variables for the industries represented by the firms in our sample.
Our data consist of information about 22 major U.S. industrial corporations between 1956 and 1983. This period covers the peak of corporate and managerial independence in the 1950s and early 1960s through the trend toward greater dependence on financial institutions that began in the late 1960s. Our data end just before the explosion of alternative financing schemes in the 1980s, which corresponded with the takeover binge of the mid-1980s and the resurgence of investment banks.
The 22 firms consist of two firms each from 11 two-digit Standard Industrial Classification manufacturing industries. For each of the 11 industries, we selected for study one of the largest and one of the smallest firms within the Fortune 500 whose primary activities took place in that industry. The 22 firms were (based on their names as of 1983) American Motors, Beatrice, Colgate-Palmolive, Crown Central Petroleum, Dayco, DuPont, Exxon, General Motors, General Tire, Grumman, Interlake Iron, IBM, ITT, Johnson & Johnson, Koppers, NCR, Procter & Gamble, Sterling Drug, U.S. Steel, United Technologies, Westinghouse, and W.K. Wrigley.
We sampled on the basis of primary industry in order to observe a wide range of large manufacturing firms (our target population). Although the number of firms in our sample makes it difficult to generalize our findings to all corporations, our previous studies with this data set (Stearns and Mizruchi, 1986; Mizruchi and Stearns, 1988) have consistently replicated findings from the literature based on a wide range of samples. We believe, based on our findings in these studies, that our results are representative of the behavior of large American manufacturing corporations in general.
For each company-year, we collected general financial data, such as sales and assets, net income, short- and long-term debt, use of the stock and bond markets, mergers and acquisitions, and any restrictive covenants attached to loans. We also collected information on the firm's top management and board of directors, including the number of inside and outside directors and representatives of financial institutions.
In addition, we coded data on general economic conditions, such as the prime rate of interest among the largest New York banks, the cost of AAA bonds, the aggregate demand for capital (in the U.S. financial market) of all nonfinancial corporations in the economy, whether the Federal Reserve was following a "tight" or "easy" monetary policy, and whether the economy was in an expansion or contraction phase of the business cycle. Our primary data sources were Moody's, for director and financial data, Standard and Poor's Directory of Executives, for information about the principal affiliations of outside directors, and the Federal Reserve Board Bulletins, the Federal Reserve Flow of Funds Accounts, and the National Bureau of Economic Research for data on economic conditions.
Our dependent variable, total debt financing, was operationalized as the total amount of money borrowed, both short-term and long-term, including public bond issues. We controlled for firm size by dividing total debt financing by total assets. The dependent variable and all independent variables affected by inflation were adjusted to constant dollars (1967 = 100), although this was unnecessary given the nature of the data.
Available internal funds was operationalized in terms of retained earnings, the amount of internal funds left after a corporation has paid its stock dividends for the year. We controlled for firm size by dividing retained earnings by total assets.
Our test of hypothesis 1, involving the cost of capital and the firm's anticipated future earnings, required that we operationalize both variables and determine whether the anticipated earnings exceeded the cost of capital. An individual firm's specific cost of capital is difficult to operationalize. Data on actual interest rates paid by firms at the initiation of their loans are unavailable. Investor services, the most prominent of which are Moody's and Standard and Poor's, however, provide ratings for the bonds of major corporations at the time of issue. These ratings, which in Moody's scheme range from a high of AAA to a low of D, indicate the degree of risk that an investment entails. For investors, greater risks require a higher rate of return. This means that bonds of companies with poorer ratings pay higher interest. Moody's bond rating is an imperfect measure of the firm's cost of capital, however, for two reasons. First, it applies only to long-term public debt (bonds). Second, it describes the firm's financial condition only at the time of the bond's issue. The first issue is not a problem because firms that pay relatively more or less in bond issues are likely to pay relatively more or less for other types of debt. The second issue is more problematic, since not all firms issued bonds every year. Fortunately, most firms issued either several new bonds during the period of our study or had their bonds reevaluated during this period. Only four of our 22 firms issued no bonds during the period of our study. Firms' bond ratings were strongly associated with their level of indebtedness, in that firms with high debt ratios tended to have low bond ratings (the correlation between the two was -.7). For years in which firms had missing data on the bond rating, we inserted a predicted value based on a linear function of the debt ratio and dividends per share. We experimented with several predictors of a firm's bond rating and found that these two variables accounted for 52 percent of the variation in the rating. The use of this predicted value for the cases with missing data created some overlap with the variable debt ratio, but this is not a problem because the debt ratio was used only as a control.
Anticipated future earnings was operationalized as the product of the firm's profitability and growth in the period leading up to the year in question. Because of occasional large year-to-year fluctuations in profitability and growth, we averaged each firm's values on these variables in the three-year period prior to the year for which the variable was coded. We then computed the firm's anticipated earnings by multiplying the firm's profitability in the previous three-year period by its growth rate in the previous three-year period.
Hypothesis 1 involves a comparison between anticipated cost and anticipated earnings. Our operationalizations for these variables are calculated in very different units, however, which makes it impossible simply to subtract anticipated cost from anticipated earnings. To allow for a direct comparison of these two measures, we transformed them to standardized scores, in which each value is expressed in terms of its standard deviation units from its mean. We then subtracted the standardized score for cost of capital from the standardized score for anticipated earnings on investment. This created a new variable, which we refer to as the "expected return on borrowing" (ERB). Hypothesis 1 suggests that the greater the difference between a firm's anticipated earnings and the cost of borrowing, the more borrowing in which the firm should engage. We should thus anticipate a positive association between a firm's ERB and its amount of borrowing.
The presence of a finance conception of control was operationalized, following Fligstein (1987), as the employment background of the person identified as being in control of the firm. CEOs with previous positions exclusively in finance and/or accounting were coded as finance CEOs. Those with backgrounds in finance and other areas (such as sales, marketing, or production) were coded, following Fligstein, as having backgrounds in general management. The presence of a finance CEO was operationalized as a dummy variable coded 1 for yes.
Financial representation on a firm's board was initially computed as the number of directors whose principal affiliation was with a commercial bank, an insurance company, or an investment bank or diversified financial corporation. Principal affiliation was determined by identifying the individual's full-time position (usually an officership). But the number of financial representatives on the firm's board may be affected by the size of the board. It is possible to argue that regardless of the size of the board, the number of financial representatives is the relevant variable, but this number may also be confounded by the firm's size. To be conservative, some type of transformation appears to be necessary. One way to handle this problem is to compute the proportion of board members whose primary affiliations are with financial institutions. Another approach is to collapse financial representation into a dummy variable, coded 1 for the presence of one or more financial representatives on the firm's board. Because we view the key issue as the existence of financial representation on the firm's board and because the presence of additional financial directors might be redundant, we have chosen to use the dummy variable approach. We did compute the equations with the proportional variable, and the results were virtually identical.
Total demand for capital was operationalized as the aggregate purchase of credit (in constant 1967 dollars) by U.S. nonfinancial corporations in a given year. Phase of the business cycle (a dummy variable coded 0 for contraction and 1 for expansion) was coded based on assessments by the National Bureau of Economic Research (U.S. Bureau of Economic Analysis, 1986). The New York prime interest rate, adjusted for inflation, was used as the measure of the economy-wide cost of capital.
Type of control (a dummy variable coded 0 for family-controlled firms and 1 for management-controlled firms) was coded based on an examination of the classification schemes of Larner (1970), Burch (1972), Kotz (1978), and Herman (1981). Larner treated a 10-percent ownership block as necessary for control. The other authors used a 5-percent criterion as well as additional information, including, in Burch's case, reports from the business press. In the few cases in which different authors presented competing classifications, we used our best judgment as to whose classification was more compelling. Four of our firms were classified by Kotz as under "financial control." Because all four firms were classified as management controlled by both Burch and Larner, and because a previous study (Stearns and Mizruchi, 1986) revealed no differences between these four firms and the remaining management-controlled firms on a series of variables, we treated them as management controlled in the current study.
Whether a firm had an employee stock ownership plan (ESOP) was a dummy variable coded 1 for yes. The value of the firm's stock issue was calculated as the number of shares times the firm's average stock price for the year. The value of the firm's stock used to finance the acquisition of another firm was calculated in the same way.
The firm's debt ratio was operationalized as the firm's long-term debt plus current liabilities, divided by total assets. Finally, dummy variables were created for ten of the eleven industries represented in our data set. The industries (based on Fortune's classification of the firms' area of primary production) were food, chemicals, petroleum refining, rubber, metal manufacturing, electronic appliances, motor vehicles, aerospace, pharmaceuticals, soaps and cosmetics, and office equipment.
Our design is a pooled cross-sectional time series commonly found in econometric analyses. In these models, the unit of analysis is the company-year. With 22 firms and 28 time points, our data set contains 616 observations. Because the same units are observed at several points in time, however, the observations are not statistically independent. Thus our true number of observations is somewhere between 22 and 616. There are three potential sources of bias in analyses of this type: serial and contemporaneous correlation of the residuals and heteroskedasticity. These biases can be minimized with the use of a generalized least squares (GLS) model (Hannan and Young, 1977; Maddala, 1977). An approach well suited to our data is a modified GLS estimation technique discussed by Parks (1967). This model assumes the existence of a first-order autoregressive process, in which the correlations of the within-company error terms are likely to be highest in one-year intervals and smaller as the time between the observations increases. This assumption is particularly useful when relatively long time periods are examined. Parks' model also corrects for heteroskedasticity and contemporaneous correlation of the residuals. The SAS routine used to compute our equations provided two additional estimation procedures. We tried these as a check on the robustness of our results. The results produced a few minor differences but none large enough to alter our substantive conclusions.
Several of our variables were measured with one-year or partial lags. The debt ratio and the cost of capital variables must be lagged a full year because their numerators for the current year are components of the dependent variable (amount borrowed). Retained earnings and capital demand were lagged by an interpolated half-year, by averaging the previous and current years' values. This "straddled lag" was used for the following reason: On the one hand, corporate financing decisions are generally made based on up-to-the-minute information and availability of resources. Last year's retained earnings, for example, may provide a poor indicator of the firm's available cash at the point of decision. On the other hand, not to lag the variable at all creates the possibility that the firm's retained earnings include income earned subsequent to the firm's decision to borrow. Although the straddled lag does not entirely eliminate the latter possibility, it significantly decreases its likelihood. The same logic applies to the effect of capital demand. Because of the often rapid fluctuations in business conditions, the conditions that led firms to borrow in the previous year may have changed. If we do not lag the variable at all, however, we may fail to allow sufficient time for the effect to occur and we increase the probability that the aggregate capital demand changes subsequent to the firm's borrowing decision. In this case, too, the straddled lag appears to best address the rapidity of response while maintaining the appropriate temporal ordering. Two additional variables, the presence of a finance CEO and the presence of a representative of a financial institution on the firm's board, were also measured with straddled lags. Because of these variables' high level of stability, the specific way in which they were lagged made little difference in the data.
Table 1 presents the means, standard deviations, and correlations among the variables in the analysis. The industry dummy variables are excluded to conserve space. Few correlations involving these variables reached even the .3 level, and most were much lower. Only two stand out. The motor-vehicle firm dummy was correlated .57 with the issue of stock for the purpose of acquisitions; and the office equipment (computer) industry dummy was correlated .53 with the presence of an ESOP. The equations were computed with and without the industry dummy variables. The effects of the stock-for-acquisition and ESOP variables were unaffected by the presence of the industry dummies.
TABULAR DATA OMITTED
On the basis of the correlation matrix, two variables stand out as potential sources of multicollinearity. The largest potential problem is with the control for debt ratio. A firm's debt ratio has a simple correlation of .67 with the amount of borrowing in a particular year. This indicates that firms that are already heavily in debt are more likely to borrow than are firms that are relatively solvent. This finding is consistent with the agency theory prediction described above. A firm's debt ratio from the previous year is strongly inversely correlated with both its retained earnings (r = -.78) and its expected return on borrowing (r = -.66). The simultaneous presence of these variables in the equation raises the potential for distorted estimates. Similarly, the correlation between interest rates and stage of the business cycle is also negative (r = -.52; interest rates tend to be higher during recessions). And the correlation between retained earnings and the ERB was a positive .62. To ensure that the simultaneous inclusion of these highly correlated variables does not distort our findings, we present equations with and without the variables for debt ratio and stage of the business cycle. We also computed the equations with retained earnings removed. Because the effects of the remaining variables remained unchanged in these equations, we do not report them here. They are available on request.
Table 2 presents five MGLS equations for the determinants of corporate borrowing. Equation 1 contains the full model with all variables included. Equations 2 through 5 omit particular variables. The hypotheses identified four primary determinants of borrowing: the expected return on borrowing (operationalized as the difference between the standardized values of anticipated earnings and cost of capital); the availability of funds (retained earnings); the strategic orientation of the CEO (the existence of a CEO from a financial background); and the firm's board composition (the presence of a representative of a financial institution on the firm's board). Equation 1 indicates that the higher the retained earnings, the less likely are firms to use debt financing. Since this effect holds even when market conditions are controlled, it suggests that firms do prefer to use retained earnings for financing if they are available. This is consistent with the organizational autonomy hypothesis (H2).
Hypothesis 1, on the expected return on borrowing, yielded some unexpected but intriguing findings. A firm's ERB, as represented by the difference between its anticipated earnings and anticipated cost, was negatively associated with its level of borrowing, so that the greater the expected return, the less the firm borrowed. The reasons for this, we believe, are twofold. First, there is a high association between firms' anticipated costs (that is, their bond ratings) and their previous indebtedness.(2) Firms that are heavily in debt tend to be viewed as greater risks and thus have poorer bond ratings. Yet firms that have borrowed significantly in the past tend to borrow heavily in the present as well, evidenced by the positive effect of previous debt ratio on current borrowing. This creates the paradoxical result that firms that pay more for capital nevertheless borrow more.
A second reason that firms with low ERB actually borrow more is the negative association between anticipated earnings and amount of borrowing. Firms that have experienced high levels of profits and growth in recent years borrow less than do firms that have experienced low levels of profits and growth. This finding makes sense in terms of hypothesis 2: Firms that do well have more cash, which they use to avoid external financing. Thus, both components of our measure of the extent to which it "pays" for firms to borrow, anticipated earnings on new investment and the firm-specific cost of capital, are associated with borrowing, but in counterintuitive ways. Firms whose bond ratings are poor borrow more. And firms with high levels of profits and growth borrow less. As a result, firms whose relative anticipated earnings outstrip their relative costs are the least TABULAR DATA OMITTED likely among those in our sample to borrow large sums of cash.
To further examine our finding on the expected return on borrowing, we modified our measure of anticipated earnings by removing growth and focusing on profitability. We then recomputed our regression equations with this modified measure of ERB. The effect of ERB remained negative but was no longer statistically significant (the effects of the other variables were unchanged). The modified measure of ERB was correlated .89 with retained earnings, however, which may explain the weakened result. To check this, we recomputed the five equations with retained earnings omitted. In all five equations the effect of ERB was significantly negative. This demonstrates that regardless of whether or not we include growth in our measure of anticipated earnings, the effect of ERB on borrowing remains negative.
To examine whether the simultaneous inclusion of interest rates and phase of the business cycle was a possible source of multicollinearity, we removed phase of the business cycle from the equation. With expansion removed from the equation, the coefficient for interest rate becomes negative, as predicted, although the effect does not reach statistical significance. The sizes of the coefficients of the other variables in the model are virtually identical in equation 2 to those in equation 1. Retained earnings remains a strong negative predictor of the use of debt financing. Two of the three alternative financing variables, the presence of an ESOP and the issue of stock for the purpose of an acquisition, are strong predictors of debt financing in both equations. As expected, the existence of an ESOP is negatively related to debt financing, indicating that ESOPs offer firms an opportunity to increase their capital supply through selling shares to their employees or benefitting from tax deductions. The issue of stock independent of its use in acquisitions is also positively associated with borrowing, although the coefficient is only marginally statistically significant. This finding does suggest that stock is more likely to be used as a complement to rather than a substitute for debt financing.
As predicted by hypothesis 3, firms with CEOs from financial backgrounds were more likely to use debt financing than were firms with nonfinancial CEOs, suggesting that firms with a finance conception of control are more likely than other firms to borrow. A firm is more likely to use debt financing when other firms are behaving similarly. This effect holds regardless of whether or not interest rates or the stage of the business cycle, both of which are associated with capital demand, are inserted into the equation.
One of the strongest predictors of borrowing was the presence of a representative of a financial institution on the firm's board. This finding supports hypothesis 4. Management-controlled firms are less likely to use debt financing than are family-controlled firms.
Equation 3 of Table 2 is identical to equation 2 except that stage of the business cycle has been substituted for interest rate. Despite the unusual character, noted above, of the recessions of 1974-1975 and 1979-1980, firms were more likely to use debt financing during expansion periods than during contraction periods. The effects of the other variables in the model are generally unchanged from equations 1 and 2, although the coefficient for retained earnings becomes much stronger and that for management-controlled firms becomes somewhat weaker. Equations 4 and 5 are identical to equations 2 and 3 except that the debt ratio has been removed. Even with debt ratio omitted, the effects of most of the variables remain quite consistent. Overall, the substantive conclusions based on these equations remain the same as those based on equations 1-3.
Finally, one further issue requires attention. One of the assumptions in longitudinal models such as ours is that the effects of the exogenous variables are constant over time. The period spanned by our study (1955-1983), however, encompasses two different eras in the relation between nonfinancial and financial institutions. As we noted earlier, the use of external financing among large U.S. firms increased sharply in the late 1960s and remained high in subsequent years (Stearns, 1986). Although there is no theoretical basis for predicting differences in the individual effects, it is possible that the factors that determined corporate borrowing in the post-1966 period differed from those in the pre-1966 period. To examine this possibility, we divided our dataset into two periods, 1956-1966 and 1967-1983, and recomputed the equations for each period. Because of the smaller number of observations in each period, the results of the MGLS equations were of questionable reliability and are not reported here. To the extent that the equations are accurate, however, most of our findings held during both periods.(3)
DISCUSSION AND CONCLUSION
Our findings are consistent with hypotheses drawn from several branches of economic and organizational theory. Economic considerations play an important role in financing decisions, although not in exactly the way that our first hypothesis predicted. On the one hand, borrowing increases during periods of economic expansion and is negatively (although only slightly) associated with economy-wide interest rates. Also, firms that have borrowed heavily in the past are more likely to borrow heavily in the future, perhaps because highly leveraged firms have an economic incentive to borrow (Jensen and Meckling, 1976). If a risky project is successful, the firm captures most of the gains, while if the project turns out poorly, the firm's creditors bear most of the costs. On the other hand, firms with high expected returns on borrowing borrow less than do firms with lower expected returns. This latter finding is inconsistent with the predictions of finance theory. Furthermore, the strong effects of several alternative variables indicate that the cost of capital is only one of several factors that determine a firm's level of borrowing.
Because the use of debt financing is likely to be accompanied by a loss of managerial autonomy, firms may attempt to finance their investments with internally generated funds. In line with this argument, borrowing increases when a firm's retained earnings are low. This finding is consistent with the interview data presented by Donaldson (1961, 1969). The firms that Donaldson studied borrowed considerably less than what finance theory would have recommended. Donaldson attributed this to the conservative nature of firm financial officers, who preferred to avoid debt in order to preserve management's autonomy and security.
The amount that a firm borrows is related to the extent to which other firms borrow. This finding may be due to variables that were not included in our model, but it is also possible that it reflects a mimetic process, in which firms observe the actions of their peers and respond accordingly (DiMaggio and Powell, 1983). Whether this tendency represents a mimetic process or simply a similar response to a common set of conditions is an issue worth examining.
The existence of alternatives, whether retained earnings or ESOPs, lowers the use of debt financing. When these alternatives are available, corporate managers appear to use them. There are, however, conditions under which firms' borrowing increases, including cases in which firms issue stock for the purpose of an acquisition.
Finally, among our findings are two organizational variables, both of which may operate independent of market conditions: the employment background of the firm's CEO and the presence of a representative of a financial institution on the firm's board. Consistent with an argument by Fligstein, firms whose CEOs have backgrounds in finance or accounting borrow more than do other firms. This finding suggests that a CEO from a finance background is likely to have a different perception of the firm's needs than would a CEO from a different functional background.
It is possible that the original decision to appoint a CEO with a finance background was a consequence rather than a cause of borrowing. It should be noted, however, that CEO background was used as an indicator of a larger concept: the firm's conception of control, a general orientation that is viewed as affecting managers' decision making. The firm's choice of a CEO from a finance background may itself reflect a perception by top management, the board, or even the stockholders that a financial specialist represents the most appropriate orientation for the firm. We assume, therefore, that it is the finance conception of control rather than CEO background per se that leads firms to engage in greater debt financing.
Firms with representatives of financial institutions on their boards of directors are more likely to borrow than are firms without financial directors. Consistent with the resource dependence model (Pfeffer and Salancik, 1978), firms may decide to borrow when representatives of financial institutions on their boards provide information, advice, and access to capital. Consistent with both agency theory (Stiglitz, 1985) and the finance hegemony model (Mintz and Schwartz, 1985), financial institutions may be more willing to lend when they are confident that their loans can be monitored. This finding holds even when the firms' individual financial situations and aggregate market factors are controlled. It is consistent with the argument that a firm's strategic decision makers are embedded in networks of social relations that transcend the boundaries of their firm (Granovetter, 1985).
As with our variable for the background of the CEO, it is possible that the causal ordering of this effect is reversed, that a firm's decision to borrow leads it to appoint representatives of financial institutions to its board. This is a plausible interpretation that is consistent with earlier evidence, including a study of ours (Mizruchi and Stearns, 1988), and it probably affects some of our cases. We think it is unlikely to apply widely here, however, for two reasons.
First, the appointment of new representatives of financial institutions was a relatively rare event in our data (only 51 occurrences), but financing decisions occurred almost yearly in almost all of our companies. Second, the average financial representative in our dataset had, at a given moment, been on the firm's board between twelve and thirteen years. These facts suggest that the appointment of a financial representative to a firm's board almost certainly preceded by several years a given decision to borrow. The length of the board relation between the financial and nonfinancial firms facilitates the conditions for mutual cooptation (Granovetter, 1985; Mizruchi and Stearns, 1988; Stearns and Mizruchi, 1993): The financial institution can use its position on the board to seek out potential business or to advise and monitor the firm once a loan is extended; the nonfinancial firm can use the board appointment to encourage the financial institution to provide funds should the firm need them. Even if the initial board appointment was a consequence of an earlier financing strategy, once the banker is on the board, his or her presence may facilitate access to additional financing. Although we therefore believe that financial representation on a firm's board exerts an independent effect on the firm's use of debt, in both this case and that of CEO background, a more refined analysis, including case studies, may be necessary to resolve this issue.
Many of a firm's most critical decisions, including plant expansion, development of new technologies, entering new product lines, and acquisitions of other firms, require the use of external financing. Although this study is a preliminary effort based on a limited number of firms, we hope that this work will encourage organizational researchers to examine corporate financing in the context of the firm's larger strategic planning. All corporations borrow money. Further study of the ways in which corporate executives manage their capital dependence will provide greater insight into how firms manage environmental uncertainty to reach their economic goals.
In an analysis not reported here, we considered equity as a component of external financing, in addition to debt. The results were virtually identical to those reported below, in which debt only is used as the dependent variable. One reason for the similarity is that new stock issues play a relatively small role in total external financing, under 5 percent in our data. This was characteristic of the entire population of U.S. firms during this period (Stearns, 1986).
2 The high negative correlation between the firm's debt ratio and its expected return on borrowing might indicate that the strong effect of ERB is a result of collinearity. The ERB effect remains almost exactly the same, however, even when debt ratio is removed from the equation (equations 3-5).
3 The transformed matrices in these regressions did not have full rank. rendering matrix inversion problematic. There was some fluctuation in the reported coefficients across equations, although most of the differences occurred between equations 1-2, which included the debt ratio, and 3-5, which did not, but the substantive coefficients were generally consistent across all equations. Standard errors also appeared dubiously low in several instances. The results are available on request.
Baker, Wayne E. 1990 "Market networks and corporate behavior." American Journal of Sociology, 96: 589-625.
Baran, Paul A., and Paul M. Sweezy 1966 Monopoly Capital. New York: Monthly Review Press.
Barton, Sidney L., and Paul J. Gordon 1987 "Corporate strategy: Useful perspective for the study of capital structure?" Academy of Management Review, 12: 67-75.
Baysinger, Barry D., Rita D. Kosnik, and Thomas A. Turk 1991 "Effects of board and ownership structure on corporate R&D strategy." Academy of Management Journal, 34: 205-214.
Bearden, James 1986 "Social capital and corporate control: The Singer Company." Journal of Political and Military Sociology, 14: 127-148.
Bell, Daniel 1961 The End of Ideology. New York: Collier.
Berle, Adolph A. 1954 The 20th Century Capitalist Revolution. New York: Harcourt, Brace and World.
Berle, Adolph A., and Gardiner C. Means 1968 The Modern Corporation and Private Property. (Originally published in 1932.) New York: Harcourt, Brace and World.
Boeker, Warren, and Jerry Goodstein 1993 "Performance and successor choice: The moderating effects of governance and ownership." Academy of Management Journal, 36: 172-186.
Brealey, Richard A., and Stewart C. Myers 1991 Principles of Corporate Finance, 4th ed. New York: McGraw-Hill.
Burch, Philip H., Jr. 1972 The Managerial Revolution Reassessed. Lexington, MA: D.C. Heath.
Burt, Ronald S. 1983 Corporate Profits and Cooptation. New York: Academic Press.
Carosso, Vincent P. 1970 Investment Banking in America. Cambridge, MA: Harvard University Press.
Chandler, Alfred D., Jr. 1977 The Visible Hand. Cambridge, MA: Harvard University Press.
Clawson, Dan, and Alan Neustadtl 1989 "Interlocks, PACs, and corporate conservatism." American Journal of Sociology, 94: 749-773.
Cochran, Philip L., Robert A. Wood, and Thomas B. Jones 1985 "The composition of boards of directors and incidence of golden parachutes." Academy of Management Journal, 28: 664-671.
Cochran, Thomas C., and William Miller 1942 The Age of Enterprise. New York: Harper & Row.
Dahrendorf, Ralf 1959 Class and Class Conflict in Industrial Society. Stanford, CA: Stanford University Press.
Davis, Gerald F. 1991 "Agents without principles? The spread of the poison pill through the intercorporate network." Administrative Science Quarterly, 36: 583-613.
DiMaggio, Paul J., and Walter W. Powell 1983 "The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields." American Sociological Review, 82: 147-160.
Donaldson, Gordon 1961 Corporate Debt Capacity: A Study of Corporate Debt Policy and the Determination of Corporate Debt Capacity. Boston: Harvard University, Graduate School of Business, Division of Research.
1969 Strategy for Financial Mobility. Boston: Harvard University, Graduate School of Business, Division of Research.
Eisenhardt, Kathleen M. 1989 "Agency theory: An assessment and review." Academy of Management Review, 14: 57-74.
Fame, Eugene F., and Kenneth R. French 1992 "The cross-section of expected stock returns." Journal of Finance, 47: 427-465.
Fligstein, Neil 1987 "The intraorganizational power struggle: Rise of finance personnel to top leadership in large corporations, 1919-1979." American Sociological Review, 52: 44-58.
1990 The Transformation of Corporate Control. Cambridge, MA: Harvard University Press.
Fligstein, Nell, and Linda Markowitz 1993 "Financial reorganization of American corporations in the 1980s." In William Julius Wilson (ed.), Sociology and the Public Agenda: 185-206. Newbury Park, CA: Sage.
Galbraith, John Kenneth 1967 The New Industrial State. New York: New American Library.
Granovetter, Mark 1985 "Economic action and social structure: The problem of embeddedness." American Journal of Sociology, 91: 481-510.
Grossman, Sanford J., and Oliver Hart 1982 "Corporate financial structure and managerial incentives." In John J. McCall (ed.). The Economics of Information and Uncertainty: 107-137. Chicago: University of Chicago Press.
Gupta, Manak C. 1969 "The effect of size, growth, and industry on the financial structure of manufacturing companies." Journal of Finance, 24: 517-529.
Hannan, Michael T., and Alice A. Young 1977 "Estimation in panel models: Results on pooling cross-sections and time-series." In David R. Heise (ed.), Sociological Methodology, 7: 52-83. San Francisco: Jossey-Bass.
Haunschild, Pamela R. 1993 "Interorganizational imitation: The impact of interlocks on Corporate acquisition activity." Administrative Science Quarterly, 38: 564-592.
Herman, Edward S. 1981 Corporate Control, Corporate Power. New York: Cambridge University Press.
Jensen, Michael C. 1986 "Agency costs of free cash flow, corporate finance, and takeovers." American Economic Review, 76: 323-329.
Jensen, Michael C., and William H. Meckling 1976 "Theory of the firm: Managerial behavior, agency costs, and ownership structure." Journal of Financial Economics, 3: 305-360.
Kosnik, Rita D. 1987 "Greenmail: A study of board performance in corporate governance." Administrative Science Quarterly, 32: 163-185.
Kotz, David M. 1978 Bank Control of Large Corporations in the United States. Berkeley: University of California Press.
Larner, Robert J. 1970 Management Control and the Large Corporation. New York: Dunellen.
Lintner, John 1966 "The financing of corporations." In Edward S. Mason (ed.), The Corporation in Modern Society: 166-201. Cambridge, MA: Harvard University Press.
Lowenstein, Louis 1991 Sense and Nonsense in Corporate Finance. Reading, MA: Addison-Wesley.
MacKie-Mason, Jeffrey K. 1990 "Do firms care who provides their financing?" In R. Glenn Hubbard (ed.), Asymmetric Information, Corporate Finance, and Investment: 63-103. Chicago: University of Chicago Press.
Maddala, G. S. 1977 Econometrics. New York: McGraw-Hill.
Meyer, Marshall W. 1993 "Organizational design and the performance paradox." In Richard Swedberg (ed.), Explorations in Economic Sociology: 249-278. New York: Russell Sage Foundation.
Mintz, Beth, and Michael Schwartz 1985 The Power Structure of American Business. Chicago: University of Chicago Press.
Mizruchi, Mark S. 1982 The American Corporate Network, 1904-1974. Beverly Hills, CA: Sage.
1992 The Structure of Corporate Political Action. Cambridge, MA: Harvard University Press.
Mizruchi, Mark S., and Linda Brewster Stearns 1988 "A longitudinal study of the formation of interlocking directorates." Administrative Science Quarterly, 33: 194-210.
Modigliani, F., and M. H. Miller 1963 "Taxes and the cost of capital: A correction." American Economic Review, 53: 433-443.
Myers, Stewart C. 1977 "Determinants of corporate borrowing." Journal of Financial Economics, 5: 147-175.
1984 "The capital structure puzzle." Journal of Finance, 39: 575-591.
Navin, Thomas R., and Marian V. Sears 1955 "The rise of a market for industrial securities, 1888-1902." Business History Review, 29: 105-138.
Nichols, Nancy A. 1993 "Efficient? Chaotic? What's the new finance?" Harvard Business Review, 71 (March-April): 50-60.
Palmer, Donald, Brad M. Barber, Xueguang Zhou, and Yasemin Soysal 1993 "The other contested terrain: The friendly and predatory acquisition of large corporations in the 1960s." Paper presented at the Annual Meeting of the American Sociological Association, Miami Beach.
Palmer, Donald, Roger Friedland, and Jitendra V. Singh 1986 "The ties that bind: Organizational and class bases of stability in a corporate interlock network." American Sociological Review, 51: 781-796.
Parks, Richard W. 1967 "Efficient estimation of a system of regression equations when disturbances are both serially and contemporaneously correlated." Journal of the American Statistical Association, 62: 500-509.
|Printer friendly Cite/link Email Feedback|
|Author:||Mizruchi, Mark S.; Stearns, Linda Brewster|
|Publication:||Administrative Science Quarterly|
|Date:||Mar 1, 1994|
|Previous Article:||Self or group? Cultural effects of training on self-efficacy and performance.|
|Next Article:||A social movement perspective on corporate control.|