Capital Budgeting in Utility companies and political risk

capital_budgeting_in_utility_sector.pdf

Home >Business & Finance homework help >Capital Budgeting in Utility companies and political risk

CAPITAL BUDGETING BY UTILITIES

EUGENE F. BRIGHAM and RICHARD H PETTWAY

Dr. Brigham, Professor of Finance and Director of the Public Utilitv Research Center. University of Florida, rs author and coauthor of a number of hooks and many articles in finance. Dr. PeUway. Associate Professor of Finance. University of Florida, has published articles in the Journal of Financial and Quanlitaiivc Analysis, the Financial Analysts Journal, and oiher academic fournals.

he theory of capital budgeting has been studied extensively in recent years, and there is a growing body of literature describing the capital budgeting techniques employed by industrial firms. However, in spite of the importance of public utilities, virtually no studies relating to these firms' capital budgeting practices have appeared in the financial journals. This article is aimed at this gap.

A number of capital investment selection criteria have been identified in the literalurc of finance. The four most frequently mentioned are payback, average rate of return, ARR. internal rate of return. IRR, and net present value, NPV. The NPV method is generally regarded as being the "best" in some the- oretical senses, while the IRR method is a somewhat distant second. Boih payback and ARR, which may be defined in serveral ways, are generally regarded as being distinctly inferior to the two techniques em- ploying discounted cash flow.

Although theory has been extended very elegantly in recent years, the basic techniques were specified reasonably well and widely publicized by the latter 195O's. Once basic theories were accepted academi- cally, various researchers questioned whether or nol business practiced what the academic community

preached. Istvan [4, 5], Pfiomn [7], and Soldofsky [8] studied this question in the early 196O's and re- ported that relatively few firms employed the recom- mended DCF techniques. The studies by Christy [2], the National Association of Accountants [6], and Terborgh [9], all done in the latter half of the l960"s, indicated an increasing use of DCF methods, but they also showed that the payback and ARR were far more widely used. The most recent studies of national firms, the ones by Klammcr [3] and by Abdelsamad [I], showed a continuation of the trend toward DCF; however. 43% of the firms in Klammer's study were still using a non-DCF method in 1970.

Two explanations for the non-use, or at least limited use, of DCF were offered. The first hypoth- esis is that there is simply a learning-and-action lag; the second is that the cost of using a DCF technique may, in some inslances. exceed its benefits. Although neither of these hypotheses has been "proved," our own studies suggest that there is some validity to both. Accordingly, we think thai the use of DCF will increase, but it is most unlikely that any future sttidy will ever find that nil investment decisions are made using a DCF cutoff criterion.

Autumn 1973 11

Capital Budgeting in the Utility Sector

in our work with public utilities it became appa- rent almost immediately that their approach to in- vesting decisions is unlike Ihat of other companies. Regulation itself has led to a modification of tradi- tional approaches to capital budgeting. Consider Exhibit 1. which presents what might be called the "traditional view" of the capital budgeting process. Here, the firm takes on projects so long as their rate of return exceeds the cost of capital, and the capital budget for the period in question is I*. The area under the rate of return schedule, but above the cost of capital schedule, represents what might be called a "producer's surplus." The area labeled "producer's deficit" is rejected.

According to traditional regulatory theory, this conceptual model is not generally applicable to utility companies. In the regulatory process, a target, or allowed rate of return, is specified. This return is, either implicitly or explicitly, recognized as being a point (perhaps the midpoint) within a range of rates of return frequently called the "zone of reasonable- ness." If "good" capital investments cause the actual rate of return to exceed the upper end of this range, then a rate reduction is ordered to drive rates back down to target. Thus, according lo uadiiional re- gulatory theory, the existence of the regulatory pro- cess will eliminate the "producer's surplus" shown in Exhibit 1. If the surplus is eliminated by regulatory

action, this means that Ihe least profitable of the se- lected projects will have a zero NPV, and its IRR will equal the cost of capital. Hence, the rule of choosing projects so as lo maximize NPV does not appear to be operational—at least under the tradi- tional view of regulatory theory.

Public Utility Investment Decisions in Today's Environment

The preceding theoretical discussion must be modi- fied to conform to the reality of the present situation faced by public utilities.

Rate of Return Patterns Under Inflation. Exhibit 2{A) shows the rate of return pattern facing a typical utility company when (I) inflation is driving costs up constantly. (2) prices, which are set by regulatory action, are increased at discrete intervals, and (3) no regulatory lag is present. As operating costs rise, profits and. consequently, the realized return on in- vestment decline. When the lower control limit is reached, rates are raised, causing the realized rate of return to rise to ihe target level. However, continued inflation causes the cycle to be repeated, and rates of return are again eroded. The net result is that the rate of return will, on average, fall below the target level.

Exhibit 2(B) shows the effects of regulatory lags. At point A the actual rale of return penetrates the lower control limit, prompting the company to ask

Exhibit 1. Conceptual Model of the Capital Budgeting Process for an Unregulated Firm

IRR, or Marginal Return on Investment Schedule

Percent

"Producer's / Surplus"

/ / J , /

Marginal Cost of Capital Schedule

/"Producer s Deficit"

I* Investment During Period ($)

12 Financial Management

for a rate hearing, which occurs al poinl B. At point C an order is issued permitting the company to raise rates, and the rate increase takes effect at point D.

As we have shown it, the actual rate of return does not return to the target level. The cost figures generally used in the point B rate cases are those of the most recent past year. If inflation continues, by the time the new rates take effect, the cost figures are otitdatcd. that is. they arc too low. Hence, the calculated utility rates are too low lo return the rate of return on investment to the target level.

It would, of course, be possible for regulatory au- thorities to anticipate price increases. In utility par- lance this is called using a forward test year. Alter- natively, the regulatory lag could be shortened by setting the control limits closer to the target rate of return. Such procedures are beginning to be employed by regulatory agencies; the automatic fuel adjustment clause, which permits certain electric utilities to raise prices automatically when fuel costs rise, is an exam- ple. However, the past test year is used more fre- quently than the forward test year, and this has a negative impact on utility profits under inflationary conditions.

A Rising Cost of Capital. Controversy exists o\cr measurement of the cost of capital, but because of an increase in interest rates, no one seriously argties that it has not risen in recent years. However, be- cause of regulatory lags, the target rate of return has generally been set below the actual cost of capital.

Exhibit y illustrates this. From TQ to T ] . the cost of capital is both stable and equal to the al- lowed rate of return. At T| the cost of capital be- gins to rise, and during the interval from T] to TT the rate of return shortfall widens. At T-?, a rate case is held, and the allowed rate of return is ad- justed upward. However, the continuing increase in the cost of capital causes the cycle to be repeated, and over the entire period the actual rate of return averages less than the cost of capital. Note also that the debt cost used in the target rate of return is the "embedded" cost, or the average cost of all out- standing debt. If the cost of new debt is above a company's embedded debt cost-as it has been in re- cent years for virtually all utilities-then the embedded eost will rise over time.

Mandatory and Discretionary Investment Decisions

It is useful to describe now another feature of uti- lity operations. That is. they are legally required to make the investments needed to provide service upon demand. Thus, utility companies' capital investments may be divided into mandatory and discretionary investments. This is illustrated in Exhibit 4, where we show the marginal cost of capital and rate of re- turn schedules for both investment components. As we have drawn it, the mandatory category is sub- stantially larger in dollar terms than the discretionary category; this seems to be In accord with the actual situation.

Exhibit 2. Typical Rate of Return Pattern Under Inflationary Conditions

(a) No Regulatory Lag (b) With Regulatory Lag

Rate of Return

, y.£per Control Limit

Target (or Allowed) Rate of Return

.Actual Rate of Return

Lower Control Limit

Time

Autumn 1973 13

Exhibit 3. Illustration of Rising Cost of Capital Combined with Lagged Changes in the Allowed Rate of Return

Actual Cost of Capital

Percent

Rate of Return x: Target (or Allowed) Rate of Return

Time

Exhibit 4. Capital Budgeting with Mandatory and Discretionary Investments

Percent

Rate of Return on Discretionary Investments

Rates of Return on Mandatory

Cost of Capital

Investment During Period ($)

Mandatory Investments

Discretionary Investments

14 Financial Management

An example will illustrate what is involved. As- sume that in certain geographic areas a telephone company may have excess switching capacity, per- mitting it to earn a relatively high rate of return on the small investment needed to serve new customers. Profitable investments of this type give rise to the area designated as A. On the other hand, in some other district where existing capacity is fully utilized, to install a new telephone might require an invest- ment of $2,000, as opposed to an average plant cost of $1,000 for each telephone presently in service. The pricing system used in the regulatory process is, in general, based on average costs, not marginal costs. In the absence of an immediate price increase, growth in the second area necessarily means that the average rate of return on investment will decline. Thus, in- vestment here will correspond to area B in Exhibit 4.

Companies do have a certain amount of discretion in supplying new types of service or in making cost- reducing replacement decisions. For example, electric utilities are sometimes able to negotiate special rates for large industrial customers who seek to purchase interruptabie power, and it is possible for these uti- lities to earn a rate in excess of cost of capital. Sim- ilarly, companies may install new and lower-cost gen- erating equipment to replace obsolete equipment, and the returns on such investments might also exceed cost of capital. Discretionary investments such as these give rise to the "producer's surplus" shown as area C in Exhibit 4.

If area B exceeds the sum of areas A and C, and if regulatory lags are long, then the existence of mandatory investment will cause an erosion of rate of return.

When inflated operating costs, a rising cost of capital, mandatory investments, and regulatory lags are combined, the net result is a substantial diver- gence between the cost of capital and the actual rate of return on total investment. Exhibit 5 illustrates this situation, and the questionnaire results described later suggest strongly that this is indeed the current situation for utilities. Consequently, incremental in- vestment with high IRR's or NPV's would indeed benefit the companies, and their high incremental profits would not be reduced by regulatory actions. Thus, it would seem that the rationale against utili- ties' use of the DCF methods is less valid than under the static conditions assumed in traditional theory.

The Public Utilities' Investment Acceptance Criterion

When choosing among competing projects, the utility industry selects projects whose future costs, when discounted at the cost of capital, are lowest. Future costs, or revenue requirements as they are frequently called, include the following items: (1) labor, fuel, repair parts, and other operating costs; (2) depreciation; (3) property ta.xes; (4) income taxes; and (5) a return on the capital invested in the pro-

Exhibit 5, Combined Effect of Rising Costs, a Rising Cost of Capital, and Regulatory Lag

Cost of Cap i t a l

Percen t

Actual Rate of Return

Time

Autumn 1973 15

ject. The sum of these cost items, all discounted at the current (marginal) cost of new capital, is the present value of revenue requirements.

Utility theory assumes that customers' cash pay- ments will actually equal revenue requirements; hence, the annual revenue requirement is really the expected annual cash flow. Also, note that if revenues arc exactly equal to revenuerequirements,asutility theory assumes they will be, the NPV of any project, or at least the NPV of the total investment required to provide a class of service, will be zero.

The PV of annual cost criterion is applied in two separate but related ways. First, for mandatory in- vestments sales revenues are simply disregarded on the grounds that they will be the same regardless of which mutually exclusive project is chosen. In other words, an electric company may project a requirement to generate an additional 10 million kilowatts to meet service demands, then set about deciding how to provide this added capacity. The theoretically best method—given the assumed level of demand—is the one having the lowest present value of future revenue requirements.

The other way in which the PV of cost criterion is used, and this holds especially when a new type of service not presently offered is being considered, involves (I) calculating the minimum revenue require- ments associated with the new service, then (2) con- ducting some type of demand/regulatory analysis to see if the project will in fact produce revenues equal to its estimated revenue requirements. To illustrate, suppose a telephone company is considering providing data transmission service to a group of business firms. Several switching systems might be used, so they are analyzed to determine the one with the low- est present value of revenue requirements. The com- pany would then attempt to determine whether or not actual revenues, given the proposed price structure, would be sufficient to meet the projected revenue re- quirements. If projected revenues are sufficient, then the project would be undertaken. If they are not, then the project might be deferred, abandoned, or the company might discuss with the regulatory com- mission and the prospective users the possibility of setting higher rates for the service. This type of anal- ysis is really quite similar to the orthodox NPV method. Note, however, that it is used only for dis- cretionary (cost saving or new product) investments. However, mandatory investments are far more impor- tant for most utility companies.

We should note two objections utility executives have raised against the NPV method. First, they point out that no explicit revenue projections are re- quired to use the minimum PV of cost method, but revenues are required to calculate the NPV. We sug-

gest that revenue projections are no more difficult for most utilities than they are for most industrial companies, so this objection to NPV seems of ques- tionable validity. Second, they pointed out that utility revenues are generated by a complex system, yet most investment decisions relate to only one part of the system. We would agree that the PV of annual cost method is quite appropriate whendecidingwhich of two replacement transformers is best and it is known for certain that replacement must occur. How- ever, it seems preferable to us to explicitly consider revenues when analyzing major system additions be- fore the fact, rather than to assume the necessary rate increases.

The Questionnaire Results

At the outset of the project, the plan was to rep- licate the type of survey thai others had done, ex- cept that regulated utilities would be sampled rather than unregulated industrial companies. For the reasons cited above, however, we developed a new question- naire, designed to provide answers to the following set of questions with respect to utilities:

1. What selection techniques are used when choos- ing among alternative investments?

2. How do they account for risk differences among projects?

3. Do they conduct post-audits? 4. Do they experience periods of capital rationing,

and if so, how is this problem handled? 5. What is their most difficult problem encountered

in the capital budgeting process? 6. What is the average embedded (historical)

cost of capital, and how does it compare to the cur- rent (marginal) cost of capital?

7. What capital costs, embedded or current, are used as the hurdle or discount rate?

8. What is the allowed, or target, rate of return, and how does this rate compare to the actual rea- lized rate of return for the current year?

9. Is dividend policy influenced by either capital requirements (investment opportunities) or by condi- tions in the capital markets?

The Sample Companies

During questionnaire development, it became ap- parent that dissimilarities made it impossible to survey electric, gas. telephone and water utilities with the same questions. We concentrated on the 116 electric utilities listed on the Compustat public utility tapes, which account for 99.5% of privately-owned electric

16 Financial Management

company assets. Questionnaires were sent to the chief financial officer of each company. Forty-six percent of the sample completed and returned our question- naire. We compared the responding and nonrespond- ing firms with respect to size and location, and we found no significant differences. The questionnaires were completed in the fall of 1972.

Project Selection Criteria

We asked the following question: "What invest- ment selection technique or techniquesdoes your com- pany use when choosing among alternative projects? If more than one standard is used, please indicate the approximate percentage of the total dollar volume of investment that is evaluated by each method." The responses are given in E.xhibit 6.

Several comments should be made about the results shown. First, most individualcompaniesactuallyindi- cated thai they use only methods I. 2, and 5. Nine- ty-four percent, or 50 out of 53 of the companies, use the DCF method (minimum PV of reventie re- quirements) to analyze at least some of their capital

Exhibit 6. Project Selection Methods Employed by Eleetric Utilities, 1972

Perccnl of loial dollar volume of

capital expcndittiics evaluated by mcUiod

in a typical year*

1. "Urgency": Capital expendi- tures required lo restore service after a system breakdown

2. No formal analysis is made; in- stead, Ihe judgment of the de- cision maker is relied upon

3. Pick project with lowest lolal "first costs" (i.e., the lowest iniiial costs)

4. Pick project with the lowest present value (PV) of initial cosi

5. Pick project with the lowest PV of annual costs

4.1%

17.8

7.4

1.7

69.0

100.0%

•The pcrcel1tage^ given here are unweighted averages of the individual questionnaire responses. •"Companies that use the equated or level annual charge method are included in this group. Generally, revenue requirements equals ihe expected first cost of the project multiplied by an annual cosl percentage which consisls of expected eost of money, property and income taxeb, de- preciation, and maintenance costs.

projects. This contrasts with Klammcr\ finding that only 57% of the Fortune 500 industrial companies used a DCF mehlod.

As indicated earlier, discretionary invcsttiients are generally accepted only if the utility's manager thinks revenue requirements will be realized. If expected reve- nues equal reventie requirements, then e.\pected NPV wili equal zero, while if expected revenues e.xceed revenue requirements, NPV will be positive. Thus, to the extent that discretionary investments are handled in this manner, utilities do, in effect, use the NPV method.

Most respondents indicated that at least some pro- jects are accepted on the basis of urgency, and our discussions with utility e.xecutives lead us to conclude that the urgency criterion is eminently reasonable. Similarly, almost all the companies indicated that some projects are accepted without formal analysis, relying instead upon judgment. A typical example is the worn out transformer, which the engineer decides to replace with whatever new transformer he believes to be the best. As with the urgency criterion, our dis- cussions with utility e.xeeutives convinced us that the nonuse of formal capital budgeting procedures for this set of projects does not necessarily imply ineffi- cient or unsophisticated management. Rather, it sug- gests a conscious comparison of the costs of follow- ing formal procedures versus the benefits gained by using informal procedures.

Adjustments for Risk

If all projects under consideration are not equally risky, then this fact should be taken into account. The two procedures most commonly recommended in the finance literature are (I) the use of risk-adjusted discount rates and (2) the use of certainty equiva- lents. Exhibit 7 shows what electric utility companies actually do. First, no respondent indicated that his company used certainty equivalents, and only about 15% of the companies use the risk-adjusted discount rate technique.

This is not to say, however, that most electric uti- lity companies indicated no formal recognition of risk differentials; 58% of the companies did acknowledge risk in some manner. The two most commonly used procedures are (1) sensitivity analysis of cost and revenues under alternative conditionsabout investment alternatives; and {2)an arbitrary downward adjustment in the expected life of an abnormally risky project.

It is interesting that utilities do formally analyze risk to a greater extent than the Fortune 500 indus- trial companies. Klammer found that only 40% of the industrial firms surveyed explicitly analyze risk versus 58% of the utility companies.

Autumn 1973 17

Exhibit 7. Procedures Used to Account for Differing Degrees of Project Risk

Primary method

used**

Secondary method used (if

an\ indicated)*'

1. Raise the cost of capital used in cal- culating revenue requirements for riskier projects

2. Adjust downward the e.xpccled life if the project is more risky than normal

3. No formal differ- entiation is recog- nized

4. Use ''sensitivity analysis" (i.e.. formally consider what will happen lo eosts and reve- nues under alterna- tive conditions, and use this infor- mation in a judg- mental manrtcT to reach a decision as to the best alierna- tive)

4.4%

1.0

42.3

10.4%

8.4

42.3

100.0%

*Only 32.5% of the responding ciimpanics indicated thnl they used two methods to account for riiik differentials. •*The percentages given here are unweighted averages of the indi\iduai questionnaire responses.

Post-Audits of Investment Projects

Post-audits supposedly lead to better capital bud- geting by (I) uncovering serious weaknesses or sys- tematic biases and (2) stimulating decision makers to be more careful.

Exhibit 8 shows tlic percentage of the electric companies that conduct post-audits. The table is di- vided into two sections, one for residential and com- mercial investments, the other for industrial invest- ments. The primary reason for using this breakdown is that industrial service is frequently discretionary, and some utility executives feel that p9st-audit$ are more applicable for investments of this type. The table also recognizes that post-audits can be made separately for construction costs, operating costs, and operating revenues.

Exhibit 8. Post-Audits of Investment Projects

Percentage of respondents thai conduct post-audit*.

Post-audit of initial outlay costs

Post-audit of operating costs

Posl-audii of operating revenues

Rcsiitcniial and conimi-rciai rnM-'stmtrnts

Industrial service

investments

60.9% 63.0%

30.2% 38.6%

25.6% 35.7%

Only a little over 60% of the titilities conduct post-audits. This compares with Klammcr's finding that 88% of the largest industrial firms employed post-audit^i of construction costs. One explanation given by a utility company executive for his own company's lack of interest in construction cost post- audits for all projects relates to the very long con- struction periods sometimes involved. Today it takes an average of 14 years to plan and build a nuclear plant. With such a long time frame, the initial cost estimates are simply not relevant. Early estimates are avaiiabie and could be looked up and analyzed, but why bother? This executive also suggested that a considerable amount of utility investment is done under fixed cost contracts, and post-audits are obviously not useful in these instances.

Exhibit ii also shows that post-audits of operating costs and operating revenues are not conducted gen- erally. A noticeably larger percentage, however, of industrial as opposed to commercial residential pro- jects are subjected to post-audits. The principal reason for the companies" infrequent use of operating cost- revenue post-audits is. apparently, that since most of their investments are mandatory, they simply must be made regardless of either the operating cost of the project or its revenues.

Capital Rationing

Exhibit 9 indicates that 40% of the companies surveyed have been subject to capital rationing. Of the firms, 89% indicated that in response to funds shortage they would apply for a rate increase. If a rate increase were granted, then their higher earning power would presumably enable them to obtain the capital necessary for making alt "identified and justi- fiable" investments.

If rate increases were not granted. 75% of the companies indicated that they would eliminate or postpone those projects that would be least likely to

18 Financial Management

Exhibit 9. Capital Rationing in the Electric Utility industry

I. Percentage of respondents that have experienced capital raiioning during the past 5 years*

II. Procedures for dealing with Capital raiioning

1. Apply for a rate increase

2. Eliminate or postpone thoie projects that are least likely lo meet revenue requirements

3. Lca!>e fixed assets

4. Make less capital intc^^ive in\cstments (i.e.. accept Ihe al(erna[i\c with ihc lower first cost or initial outlay)

Have had Capital

Raiioning

40%

Percentage of rcspondcnis thai indicated their firm

would l.-ike the action noted

89%

75%

55%

' . \ periixl of capital rationing is defined as a period when the firm could not obtain sufficient funds at or below its allowed rate of return to make all its identified and justifiable investments.

meet revenue requirements, and over half the com- panies indicated that they would lease rather than purchase fixed assets. The willingness to lease was somewhat surprising, but apparently utilitycompanies that are strapped for capital are increasingly resorting to leasing arrangements. The fourth alternative men- tioned was to make less capital intensive investments.

Perceived Problem Areas in Capital Budgeting

Far and away their most serious problem in the eyes of utility executives is obtaining permission from environmental protection agencies and. or the Atomic Energy Commission to build new generating plants. No other factor was considered to be a serious pro- blem by even half as many respondents.

The remainder of Exhibit 10 was somewhat sur- prising. We expected the companies to have trouble estimating annual costs and revenues and cost of capital, but obviously they do not consider these esti- mates 10 be serious problems. In retrospect, it is easy to see why this is so. The cost of capita! for utility companies is, rightly or wrongly, determined in rate cases. Also, capital budgeting techniques used tend to suppress revenue estimates; re\enue shortfalls are supposed to be made up by rate increases. Further, the companies frequently assume that, once a project is in operation, the regulatory process will provide sufficient revenues to cover oper- ating costs.

It is also interesting to examine the second col- umn in the table headed **A Fairly Serious Pro- blem." Many items not consideied to be "very serious" are considered, nevertheless, to be "fairly serious". For example, estimating the annual oper- ating costs of a project, response 9 in E.\hibit 10, is not generally considered to be a very serious pro- blem, but it is considered to be a fairlv serious one.

Cost of Capital, Allowed Rates of Return, and Realized Rates of Return

The average after-tax current cost of capital. 9.3%, indicated in Exhibit 11, is well above the indicated embedded cost of capital. 8.0%. This differential is, presumably, caused by the fact that the embedded cost of debt for most companies is well below the current rate of interest on long-term bonds. It is also interesting to note that the average allowed rate of return as prescribed by regulatory authorities, 7.6%, is below the indicated 8.0% a\erage embedded cost of capital. There are a large number of rate cases in process across the country today, and allowed rates of return will presumably be increased somewhat.

The last item shown in Exhibit II. the current rate of return on in\estment. is substantially lower than either the allowed rate of return or the cost of capital. Thus, the situations shown in both Exhibits 2 and 5 seem to exist today.

Autumn 1973 19

Exhibit 10. Percei>ed Problem Areas in Capital Budgeting

Obtaining rcgulalory approval for new plants from environmental protcelion agencies and/or AEC

Specification of first cost or capital requirements of a new investment

Estimation of the cosl and availability of llie input factors (i.e., fuel, labor)

Estimation of ihc project's economic life giving regard lo bolh demand factors and obsolescence of ihe invcsiment due to new technology

Estimation of when the plant will be placed in service

Making in\cstmcnis that should be profitable, given demand and technology factors, but thai are nol allowed lo earn their expected return by regulatory authorities

Making sure all reasonable alternatives have been considered

Specification of the effects of inflation on annual costs in general

Estimation of annual operating cost of the project

Predicting the needs of the franchise area in advance

Esiimaiion of annual revenue attributed to the project

Specification of a "cost of money" or cost of capital

Estimation of project life from a wear/iear standpoint

Perccni of Respondents Stating that the Indicated F-'actor is:

A Very A Fairly Not at Scriouii Serious all

Problem Problem Serious

The Cost of Capital Used as the "Hurdle Rate"

We asked the companies to indicate which cost of capital, the embedded cost or the current (or mar- ginal) cost, was used in llie capital budgeting pro- cess. The o\orwhclming majority of tlie companies used cither the current cost of capital or a figure very close to the current cost; no company used the embedded cost of capital when analyzing new invest- ments.

Dividend Policy

At least some of ihe writinjjs in finance suggest that companies should alter their dividend payout policies as changes occur in either investment oppor- tunities or in capital market conditions. Todetermine whether or nol utility companies do adjust their di- vidend policies, we asked the following: It has been suggested that utility companies'dividend policies may

be affected hy capital investment opportunities or requirements and by capital market conditions (i.e., ihe slate of stock and bond markets). For example, in a period of high investment demand and light money, companies might not increase dividends if earnings increased, thus reducing the payout ratio, or they might even cut dividends in order to con- serve capital. Recognizing that it might take several years to effect such u change, do you think that your own company's dividend policy would be affected by:

Percent responding:

Yes No a. Changes in capital expenditure

opportunities or requirements?

b. Capital market conditions?

34%

40'̂ ,, 60%

According to the respondents, only about one-third of ihc utility companies' dividend policies are ad- justed in response lo changing investment oppor- tunities or capital market conditions.

20 Financial Management

Exhibit U . Cost of Capital, Allowed Rates of Return, and Realized Rates of Return, Electric Companies, 1972*

!, Average After-Tax Current (or Marginal) Cost of Capital 9..V'',

2. Average After-Tax Embedded Cost of Capiial 8.0%

3. Allowed, or Target, Rate of Return as Prescribed by Regulatory Agencies 7.6%

4. Current Actual Rate of Return on Investment 1 .IX

•The cost of equity capital is defined as the rate of return on book equity thai was authorized if a rate caî e was recently concluded, or the rate of return most likely to be allowed if a rate case were lo be decided now. The problems encountered when attempting to measure the cost of equity are well known, and il is possible iha[ Commission-determined costs of capital arc seriously over- or undersuued. We have simply avoided this issue by accepting ihe Commission's estimates.

It should be noled that iho figures given are returns on hook equity, which may be different from investors' re- quired rates of return on market values. For a discussion of this point, sec the discussion of A.A. Robichek in the 1971 AT&T rate ease (FCC Doeket No. 19129) or E.F. Brigham in the 1972 Conisal rate case (FCC Doeket No. 16070).

Also, it should be noted thai different companies cm- ploy different rate base valuation methods (i.e.. original cost vs. "fair value"), and different rates of relurn on these different rate bases are appropriate. Such differences were considered in ihe study upon which Exhibit II is based.

Source: Eugene F. Brigham and Richard H. Pettway. "Capital Budgeting in ihe Public Utility Sector." University of Florida, Public Uliiily Research Center. Working Paper No. 3-73. October 1973.

One thing was very clear from comments attached to the questionnaire—the utility company executives very definitely think thai the market price of their stock is influenced by dividend policy. Quite a few respondents made note of the fact that Potomac Electric Power Company, in a well-known case, took exactly the action suggested inourquestionnaire, and, apparently as a resull of this action, the price of the stock dropped precipitously. Academicians mighl ar- gue that the stock price declined because of other factors, but it would be hard to convince a number of utility company executives that ihis was ihc case.

Conclusions

Under inflation the established pattern of rate regulation has nol worked oui as utility Ihcory as-

sumes, and, as a resull, the utility companies have been placed in a difficult position. On the one hand, they must make whatever invcsttneni is necessary to meet service demands, yet rising costs, coupled wiih prices of their products ihat rise only with a lag, have caused rates of return to erode. Thus, many utilities are placed in a position where they must ac- cept projects whose internal rates of return are less than their cost of capital.

Frotii a survey we conclude the following abou( capital budgeting by electric utilities.

1. Utility companies use a DCF selection criterion (minimum PV of revenue requirements) to a greater extent than do the Fortune 500 industrials. This dif- fercniial usage probably results from the fact that the utilities arc large and capital intensive, make very long-term investments, and can estimate cash flows better than firms more subject to competitive pres- sures.

2. Utilities seem to recognize risk differentials among projects to at least as great an extent as do industrial companies, but since these differences can- not generally be quantified, they influence project se- lection in a judgmental manner, not through a for- mal technique such as certainty equivalents or risk- adjusted discount rales.

3. Utility companies do not employ post-audits of investment projects to as large an extent as do in- dustrial firms.

4. Capital rationing is becoming a problem for utililies. Their first reaction is lo seek rate increases which will enable them to raise additional funds, but if rate increases are nol forthcoming, then projects will be eliminated or postponed, assets will be leased, or less capital intensive alternatives will be accepted.

5. Utility companies do not generally consider in- put estimates to be a very serious problem. Inter- estingly, they overwhelmingly consider obtaining ap- proval for new generating plants from environmental protection agencies or the AEC lo be the single most difficult aspect of capital budgeting.

6. The current cost of capital exceeds ihe embedded cost, and this cost exceeds both Ihe allowed and rea- lized rates of return. This situation has given rise to a large number of pending rate cases.

7. When utiiiiies use the discounted cash flow tech- niques, they use the marginal cost of capital as a hurdle rale.

8. The majority of the companies Indicated thai their dividend policy is nol influenced by capital needs or by capita! market conditions, al least not in the short run.

Overall, the electric companies seem to be oper- ating largely in a manner that, while different be- cause of their regulatory environment, is generally

Autumn 1973 21

consistent wiih the types of capital budgeting tech- niques recommended in the aeadeniic literature. How- ever, we do feel that public utilities should at least consider employing the NPV method ralhcr than ihc PV of annual cost method for both discretionary and

mandatory system expansion investments. While dif- ficulties would certainly be encountered in making these calculations, the NPV method would provide valuable data on the explicit impact of expansion on both profitability and revenue requirements.

REFERENCES

1. Mouslafa Abdclsamad. A Guide to CapitalE.xpemtiture Analysis, New York, AMACOM. American Management Association, 1973.

2. George A. Christy, Cupiial Budgeting—Curreni Pruc- lices and Their Efficiency, Eugene, Oregon, Bureau of Business & Economic Research, University of Oregon. 1966.

3. Gordon R. Corey. "The Avcrch and Johnson Pro- position: A Critical Analysis," Bell Journai (Spring 1971). pp. 358-373.

4. Donald F. Istvan. Capital Expenditure Decisions: Haw They arc Made in Large Corporations. Bloomington, Indiana, Bureau of Business Research, Indiana University, 1961.

5. Donald F. Istvan, "The Hconomie Evaluation ofCap- it;il Expenditures," ' The Journal of Business {\9(}\).

6. Nat ional Assoeiation of Accountants , Financial Anal- ysis to Guide Capiial Expenditure Decisions. Research Report 43. New York. Niiiional .Association of Accoun- tants. 1967.

7. Norman P, Pflonin, "Managing Capital Expendi- tures," Studies in Business Policy, 107, New York. The National Industrial Conference Board, 1963.

8. Robert M. Soldofsky, "Capital Budgeting Praetiees in Small Manufacturing Companies," .Siudie.s in the Factor Markets for Small Business Firms, Ames, Iowa, Iowa State University, 1963.

9. George Tcrborgh, Business Investment Management. a MAPI Study and Manual. Washington. D.C.. Ma- chinery and Allied Products Institute and Council for Teehnologica! Advancement. 1967.

22 Financial Management