Capital budgeting models: theory vs. practice.

Capital budgeting decisions are crucial to a firm's success for several reasons. First, capital expenditures typically require large outlays of funds. Second, firms must ascertain the best way to raise and repay these funds. Third, most capital budgeting decisions require a long-term commitment. Finally, the timing of capital budgeting decisions is important. When large amounts of funds are raised, firms must pay close attention to the financial markets because the cost of capital is directly related to the current interest rate.

The need for relevant information and analysis of capital budgeting alternatives has inspired the evolution of a series of models to assist firms in making the "best" allocation of resources. Among the earliest methods available were the payback model, which simply determines the length of time required for the firm to recover its cash outlay, and the return on investment model, which evaluates the project based on standard historical cost accounting estimates. The next group of models employs the concept of the time value of money to obtain a superior measure of the cost/benefit trade-off of potential projects. More current models attempt to include in the analysis non-quantifiable factors that may be highly significant in the project decision but could not be captured in the earlier models.

This article explains budgeting models currently being used by large companies, the division responsible for evaluating capital budgeting projects, the most important and most difficult stages in the capital budgeting process, the cost of capital cutoff rate, and the methods used to adjust for risk. A possible rationale is provided for the choices that firms are making among the available models. The discussion identifies difficulties inherent in the traditional discounted cash flow models and suggests that these problems may have led some firms to choose the simpler models.

Previous Research Studies

Capital budgeting decisions are extremely important and complex and have inspired many research studies. In an in-depth study of the capital budgeting projects of 12 large manufacturing firms, Marc Ross found in 1972, that although techniques that incorporated discounted cash flow were used to some extent, firms relied rather heavily on the simplistic payback model, especially for smaller projects. In addition, when discounted cash flow techniques were used, they were often simplified. For example, some firms' simplifying assumptions include the use of the same economic life for all projects even though the actual lives might be different. Further, firms often did not adjust their analysis for risk (Ross, 1986).

In 1972 Thomas P. Klammer surveyed a sample of 369 firms from the 1969 Compustat listing of manufacturing firms that appeared in significant industry groups and made at least $1 million of capital expenditures in each of the five years 1963-1967. Respondents were asked to identify the capital budgeting techniques in use in 1959, 1964, and 1970. The results indicated an increased use of techniques that incorporated the present value (Klammer, 1984).

James Fremgen surveyed a random sample of 250 business firms in 1973 that were in the 1969 edition of Dun and Bradstreet's Reference Book of Corporate Management. Questionnaires were sent to companies engaged in manufacturing, retailing, mining, transportation, land development, entertainment, public utilities and conglomerates to study the capital budgeting models used, stages of the capital budgeting process, and the methods used to adjust for risk. He found that firms considered the internal rate of return model to be the most important model for decision-making. He also found that the majority of firms increased their profitability requirements to adjust for risk and considered defining a project and determining the cash flow projections as the most important and most difficult stage of the capital budgeting process (Fremgen, 1973).

In 1965, J William Petty, David P. Scott, and Monroe M. Bird examined responses from 109 controllers of 1971 Fortune 500 (by sales dollars) firms concerning the techniques their companies used to evaluate new and existing products lines. They found that internal rate of return was the method preferred for evaluating all projects. Moreover, they found that present value techniques were used more frequently to evaluate new product lines than existing product lines (Petty, 1975)

Laurence G. Gitman and John R. Forrester Jr. analyzed the responses from 110 firms who replied to their 1977 survey of the 600 companies that Forbes reported as having the greatest stock price growth over the 1971-1979 period. The survey containing questions concerning capital budgeting techniques, the division of responsibility for capital budgeting decisions, the most important and most difficult stages of capital budgeting, the cutoff rate and the methods used to assess risk. They found that the discounted cash flow techniques were the most popular methods for evaluating projects, especially the internal rate of return. However, many firms still used the payback method as a backup or secondary approach. The majority of the companies that responded to the survey indicated that the Finance Department was responsible for analyzing capital budgeting projects. Respondents also indicted that project definition and cash flow estimation was the most difficult and most critical stage of the capital budgeting process. The majority of firms had a cost of capital or cutoff rate between 10 and 15 percent, and they most often adjusted for risk by increasing the minimum acceptable rate of return on capital projects (Gitman, 1977).

In 1981, Suk H. Kim and Edward J. Farragher surveyed the 1979 Fortune 100 Chief Financial officers about their 1975 and 1979 usage of techniques for evaluating capital budgeting projects. They found that in both years, the majority of the firms relied on a discounted cash flow method (either the internal rate of return or the net present value) as the primary method and the payback as the secondary method (Suk, 1981).

Capital Budgeting Techniques

Several models are commonly used to evaluate capital budgeting projects: the payback, accounting rate of return, present value, internal rate of return, profitability index models and others.

The payback model measures the amount of time required for cash income from a project to exactly equal the initial investment. The accounting rate of return is the ratio of the project's average after-tax income to its average book value.

Academicians criticize both the payback and the accounting rate of return models because they ignore the time value of money and the size of the investment.

When the net present value model is used, the firm discounts the projected income from the project at the firm's minimum acceptable rate of return (hurdle rate). The net present value is the difference between the present value of the income and the cost of the project. If the net present value of the project is positive, the project is accepted; conversely, if the net present value is negative, the project is rejected. The internal rate of the return model equates the cost of the project to the present value of the project. The net present value and the internal rate of return models overcome the time value of money deficiency; however, they fail to consider the size of a project.

Furthermore, the payback model does not consider returns from the project after the initial investment is recovered. The profitability index is a ratio of the project's value to its initial investment. The firm then selects the project with the highest profitability index and continues to select until the investment budget is exhausted. The profitability index overcomes both the time value of money and the size deficiencies.

Some decision makers have criticized the net cash flow method because they simply do not agree with the decisions indicated by the results from the models. In some cases, managers are reluctant to make important decisions based on uncertain estimates of cash flows far in the future. Thus, they consider only near-term cash flows or are distrustful of the output of the models. In others, managers may have predetermined notions about which projects to adopt and may, therefore, "massage" the numbers to achieve the result they desire. Thus, in many cases, the negative results occurred because of inappropriate input into the models, rather than from the models themselves. One area of particular concern is the choice of discount rate. For example, Robert S. Raplan and Anthony A. Atkinson suggested, in 1985, that users often employ too high a discount rate, either by choosing too high a cost of capital or by using a higher rate as an adjustment for risk. An inappropriately high discount rate yields too high a hurdle rate or too low a net present value and thus a negative signal about the project. They recommend using a discount rate that reflects the firm's true cost of capital according to sound theory of finance. Moreover, they say that risk should be analyzed by modeling multiple scenarios (best to worst cases) in a manner similar to flexible budgeting. Finally, when the discount rate incorporates inflation, the user must be careful to adjust future cash flows for inflation as well (Kaplan, 1985).