Principle of corporate finance brealey myers

ABSTRACT

The study examines the use of divisional cost of capital by Fortune 1000 companies. Two hundred and ninety eight firms (29.8 percent) responded to the survey. While the concept of weighted average cost of capital is utilized by 85.2 percent of the respondents, less than 50 percent use divisional cost of capital. By using a single firm cut-off criterion for all projects, there is the potential for intrafirm misallocation of capital since projects initiated by high risk divisions are more likely to be accepted because of high returns. Lower return divisions with less risk may be starved for capital when only a single weighted average cost of capital is used. The author also suggests some normative approaches to solve the problem.

INTRODUCTION

In the traditional literature of corporate finance, a key metric is weighted average cost of capital. After it is determined, the WACC is intended to be the cut-off point in capital budgeting decisions. Projects that equal or exceed the hurdle rate are viewed as adding to stockholder wealth maximization, while those that fail the test are viewed as dilutive to value. Perhaps no other principle of finance is further off the mark.

As stated in the Brealey and Myers text on corporate finance, "Company costs of capital are nearly useless for diversified firms" [4]. To the extent that divisions in a corporation have degrees of risk and financial characteristics that are different from the parent corporation, using the overall corporate hurdle rate is certain to lead to incorrect decisions and failure to maximize stockholder wealth [8, 12, 13, 14]. The major consequence of using a single cut-off criterion for all projects is an intrafirm misallocation of capital since projects that are initiated by high risk-divisions are more likely to be accepted because of their potentially higher return. A similar bias works against lower risk divisions in that they may be starved for capital because their relatively low returns do not match up to the corporate cost of capital, which is based on normal risk [8]. In a typical risk-averse environment, these lower-risk projects maybe rejected in spite of the fact that on a risk-adjusted basis they might be quite acceptable. Management may, in fact, have capital budgeting procedures that work against its own objective.

The intent of this study is to determine how many firms actually go to the point of using divisional cost of capital when it is the appropriate measure.

The principles discussed above were illustrated in the Internet boom of the late 1990s. Many large U.S. firms added divisions related to the Internet that were high risk in nature, but were only required to pass the same hurdle rate test as more traditional investments. The same pattern was followed in the telecommunications industry where firms entered new markets for untested means of communication, only to find the market was unaccepting of their new products. Had the full risk associated with these projects been included in their hurdle rates, many new ventures would not have been initiated.

The effect of failing to establish appropriate costs of capital for different projects or divisions is shown in FIGURE 1, which is an adaptation from the Pinches text, Essentials of Financial Management [21]. As demonstrated in FIGURE 1, Project B clearly has a higher internal rate of return (IRR) than Project A. If the deciding factor is the firmwide cost of capital, it will be accepted, while Project A will be rejected. However, if the company establishes project or divisional cost of capital based on risk, Project A (the lower risk project) can be seen as exceeding divisional cost of capital, while Project B (the higher risk project) fails to cover divisional cost of capital. By using divisional cost of capital and project risk considerations, the decision is reversed, and A will be accepted in preference to B and the firm is more likely to maximize stockholder wealth in a risk averse, efficient capital market environment.

THE DATABASE

The Fortune magazine April 2001 listing of the 1,000 largest U.S. corporations served as the database for this study. A carefully pre-tested three-page questionnaire was sent to the top ranking financial officer of each of the firms. Two hundred and ninety-eight useable responses were returned. Key financial attributes of the respondents are presented in TABLES 1, 2, 3, and 4. Participants included such well-known companies as General Motors, Boeing, Intel, Halliburton, and Textron.

A follow-up telephone survey of forty randomly selected non-respondents indicated no statistically significant differences between those that initially answered the questionnaire and those that elected not to participate.

The study first addresses the use of capital budgeting techniques in general and then focuses in on the most important topic of this paper, the use of divisional cost of capital.

CAPITAL BUDGETING PROCEDURES IN GENERAL

In terms of overall capital budgeting procedures, there is clearly a movement toward the normative. Of the 271 survey participants, (90.9 percent) use discounted cash flow as the primary method of evaluation, with a slight preference for the internal rate of return over the net present value method. This is consistent with other survey studies [1, 2, 10, 11, 20, 23, 24, 27] that report similar trends.

In measuring the required rate of return, 254 of the 298 respondents (85.2%) indicated a preference for the weighted average cost of capital, with the other 44 survey participants choosing various other measures. The distribution of answers is presented in TABLE 5. This is once again consistent with a movement toward the normative approach that has been cited in the other research studies mentioned in the paragraph above.

DIVISIONAL COST OF CAPITAL

The same pattern of progressive enlightened responses is less evident on the topic of divisional cost of capital. Only 139 of the 298 respondents (46.6 percent) answered positively to the question: "Do you have different rates of return that are required for different divisions, subsidiaries or projects of the firm?"

The question would appear to be unambiguous and the response indicates that in spite of much progress by corporate management in regard to capital budgeting procedures in general (discussed in prior section), a similar pattern is not evident for the topic of divisional cost of capital.

A series of chi-square tests were run to determine if there was an independence of classification between the use of divisional cost of capital and a number of other variables including revenue, total assets, net profit, or the ratio of fixed assets to total assets. Revenue appeared to be significant at an alpha level of .10 and fixed assets to total assets at a level of significance of .01 (APPENDIX A). The other two variables showed no meaningful relationship.

It is interesting to note that firms with a heavy component of fixed assets in their capital structure are more likely to employ the divisional cost of capital concept. One can only surmise that when firms become increasingly dependent on large, permanent asset acquisitions, the depth of analysis increases.

FIRMS THAT USE DIVISIONAL COST OF CAPITAL

For the 139 firms that use divisional cost of capital in the present study, follow-up questions were presented as to their approaches. As can be seen in TABLE 6, 121 out of the 139 firms that use divisional cost of capital consider risk to be the primary consideration differentiating the required return for the division. Thus, consideration of the topic of risk follows.

DETERMINING RISK AND REQUIRED RATES OF RETURN FOR A DIVISION

For those firms that consider risk to be a key variable in determining divisional cost of capital, the academic literature indicates there are a number of approaches they can take. For example, they can estimate the beta for each division and then incorporate it in the capital asset pricing model for the division to determine the cost of common equity. This value can be combined with an assumed cost of debt and the optimal capital structure for the division to determine the opportunity cost of capital for the division [5, 17, 21].

There are many uncertainties associated with this approach. However, the most important variable to be determined is the divisional beta. The beta might be estimated using an accounting-based approach [6, 15, 19, 22]. A more likely alternative is to use an analogous firm approach. Popularized by Fuller and Kerr [8], and also referred to in the literature as the pure-play approach, a proxy beta is derived from a publicly traded firm whose operations are as similar as possible to the division in question. The proxy beta then becomes the systematic risk that is used in the CAPM. While some have urged that there is great difficulty in finding publicly traded firms that are analogous to a firm's division [14, 18], Fuller and Kerr [8] maintain that this is not the case.