Letters to the Editor
April 1, 2003
To the Editor:
An article by Jonathan Lesser ("DCF Utility Valuation: Still the Gold Standard?" Feb. 15, 2003) contains mistakes of both a practical and theoretical nature. Mr. Lesser's conclusion that discounted cash flow (DCF) should not be used by regulators as the primary indicator of equity capital cost is incorrect and is not supported by the statistical or logical evidence he presents.
Mr. Lesser's concerns regarding the DCF are not new, have been rejected by regulators in the past and should be treated similarly now. The DCF continues to be the most accurate method with which the cost of equity capital of regulated industries can be estimated.
Mr. Lesser errs in the first sentence of his article (and shows on which side of the "aisle" his sentiments reside) by stating that the DCF was introduced into regulation in the mid-1960s by "ratepayer advocates."
It was not. The DCF formula with which we are now familiar first appeared in modern finance as a valuation model in 1938, in a text by J. B. Williams, The Theory of Investment Value.1 In attempting to solve the economic conundrum created by the 1929 stock market crash, Professor Williams focused on the dividend rather than earnings as a reliable indicator of investor value. Utilizing a standard mathematical discounting technique, and making the assumption that dividends were expected to grow at a constant rate, the model collapsed into a form which is familiar to us now as the DCF. Williams, however, rejected that DCF model as too simplistic, and it remained forgotten for years.
In the mid-1950s, Professor Myron Gordon began to investigate the model that Williams had abandoned.2 Following a series of other articles on the topic, Gordon formalized his theories in The Investment, Financing, and Valuation of the Corporation, published in 1962.3 Gordon presented the DCF model in a generalized corporate financial context and did not focus on the DCF in relation to public utilities. However, subsequent to the publication of that initial valuation text, in the mid-1960s, Dr. Gordon was asked to present his DCF theories in an AT&T rate proceeding before the Federal Communications Commission. It wasn't until the mid-1970s that Gordon published a text more directly aimed at estimating the cost of capital for utility operations (The Cost of Capital to a Public Utility).4
Therefore, Mr. Lesser's attempt to peg the DCF as the product of "ratepayer advocates," by which, we must assume, he implies a group bent on producing equity cost estimates that are "too low" and are designed to produce an outcome (i.e., lower rates) rather than an accurate cost of equity, is inaccurate. The DCF was first introduced into regulation by academics who were interested in no particular outcome other than the truth-an accurate estimate of the cost of common equity capital.
Second, Mr. Lesser opines that because of the recent stock price volatility of the Dow Jones Utility Index, the DCF is "crumbling at its foundation." This is also untrue, for several reasons:
- The DCF theory makes no assumptions regarding the volatility of market prices, and price volatility has no impact on the "foundations" of the DCF.
- As Lesser himself points out in the third section of his article, there are simple techniques to use to combat day-to-day price volatility while producing an accurate DCF cost of equity estimate.
- The price volatility evidenced in the Dow Jones Utility index over the past year is attributable to illegal activity, loss of investor confidence, and financial meltdown in the unregulated energy trading operations in which some of the utilities contained in the index were involved. Therefore, the recent price volatility of that index in no way indicates that the DCF is no longer applicable to discerning the cost of equity of regulated
utility operations.
Mr. Lesser's claim that the DCF has recently "gone south" and should no longer be a primary indicator of the cost of equity capital in regulation assumes that prior to the recent price volatility in the Dow Jones Utility Index, everything was fine. It is most curious, then, that Mr. Lesser believes the DCF gave reasonable equity cost estimates during the mid-1970s Arab oil embargo, the high inflation of the late 1970s, the 21 percent prime interest rate of the early 1980s, the Three Mile Island nuclear incident, subsequent nuclear plant disallowances, the stock market crash of 1987 (talk about price volatility!), and the onset of deregulation in the electric industry in the mid-1990s-but does not do so now. Such a premise is ultimately illogical and fails to support Mr. Lesser's primary thesis. The current price volatility in the electric utility industry is no more significant an aberration from the norm than the other historical events mentioned above, and it is not reasonable to believe that due to that aberration, we should now, after 30 years of use in regulation, suddenly relegate the DCF to back-burner status.
Third, Mr. Lesser discusses price volatility as the DCF's "ghost in the machine." Here he notes that if the analyst relies on single-day prices, the results of the DCF will be volatile. This is, of course, news to no one. Also, it is precisely why no regulatory body with which I am familiar and very few cost of capital experts I have ever encountered use single-day market prices to estimate the cost of equity capital.
Mr. Lesser provides very clear evidence in his Graph 3 that, even in these "troubled times" of market price volatility, the volatility of a DCF estimate is reduced dramatically through the use of a 30-day market price average-a reasonable and widely used methodology. Moreover, a recent average market price still fulfills the Efficient Market Hypothesis (EMH) requirement that the stock prices used reflect current investor opinion.
Fourth, Mr. Lesser appears to cast aspersions at the DCF by stating that the DCF is based on the EMH. While it is difficult to discern if Mr. Lesser is for or against the EMH theory, the validity of that theory is out of place in his examination of the reliability of the DCF. That is because all market-based cost of equity estimation techniques-capital asset pricing model, risk premium, earning-price ratio, market-to-book ratio, statistical regression analysis-rely on the EMH assumption that securities in today's capital markets are efficiently and accurately priced. Therefore, when Mr. Lesser states, "It might be tempting to conclude the EMH is simply wrong," he is not condemning the DCF alone, but all market-based cost of equity estimation techniques.
Fifth, Mr. Lesser states, "In the traditional DCF form, the cost of equity equals the sum of the current dividend yield and a forecast of long-term earnings growth." Wrong. Perhaps Mr. Lesser simply hasn't done his homework, but the "traditional" DCF formula is based on long-term growth in dividends, not earnings. Of course, there are many surrogates for long-term dividend growth and projected per share earnings growth happens to be one of them-but only one of many. The DCF is predicated on dividend growth, not earnings growth. Moreover, while many analysts mistakenly rely solely on projected earnings growth to determine a DCF equity cost estimate, that methodology can lead to serious errors in the estimation of the cost of equity capital. Mr. Lesser confirms the infirmities of sole reliance on earnings growth by noting that the spread between earnings estimates can be greater than the average estimates themselves (see Table 1 in the Lesser article). Therefore, Mr. Lesser is correct to note that sole reliance on projected earnings growth can produce unreliable DCF estimates. However, he is quite incorrect to state that the DCF is designed to be used with projected earnings growth estimates. It is not. Both historical growth rates and projected growth rates in dividends, book value, sustainable ("b times r"), as well as earnings are widely published, are available to investors, and should be incorporated into any reliable, well-balanced DCF cost of equity estimate.
Sixth, Mr. Lesser poses a rhetorical question that he fails to answer. In discussing "possible solutions" to what he believes to be the "DCF conundrum," he states that using the DCF with other equity cost methods is one solution. On this point he and I agree-I (and most analysts) use other equity cost estimation methods to check and/or temper the results of a DCF analysis; but the DCF analysis remains the primary indicator of the cost of equity capital. Mr. Lesser poses a situation where a DCF produces a 12 percent equity cost estimate, while a capital asset pricing model indicates 8 percent, a risk premium indicates 13 percent and comparable earnings indicates 9 percent, and he concludes only that the arithmetic average of those results is unlikely to be the "correct" cost of equity. By his comments and the tone of his article we can only assume that he believes that all of those equity cost estimates are of equal value. With that premise I certainly disagree. Given the range of equity costs described in Mr. Lesser's example, assuming the methodologies were reasonably applied, I would expect the true cost of equity to be closer to the DCF result than any of the other methodologies.
Finally, the DCF is the method most widely used by regulators for the purpose of setting profit levels in regulated industries for one primary reason-it works, and it works well. Since its inception in the mid-1960s it has been the primary method used in regulation to estimate equity capital costs, and during that time the industry, on average, has been able to raise capital from investors to meet the needs of its customers.
Mr. Lesser's conclusion that "traditional reliance on the DCF needs rethinking by regulators" is the wrong one. Regulators have had the right cost of equity idea for 30 years, and now is no time to change. Indeed, only through continued reliance on the DCF as the primary indicator of the appropriate profit level for regulated firms will both investors and customers benefit in the long run.
Stephen G. Hill
Hill Associates
Winfield, W.V.
Endnotes
- Williams, J. B., The Theory of Investment Value, Cambridge, Mass., Harvard University Press, 1938.
- Gordon, M., and Shapiro, E., "Capital Equipment Analysis: The Required Rate of Profit," Management Science, 3 (October, 1956), pp. 102-110.
- Gordon, M., The Investment, Financing, and Valuation of the Corporation, Richard Irwin, Inc., Homewood, Ill., 1962.
- Gordon, M., The Cost of Capital to a Public Utility, MSU Public Utilities Studies, East Lansing, Mich., 1974.
Articles found on this page are available to Internet subscribers only. For more information about obtaining a username and password, please call our Customer Service Department at 1-800-368-5001.
Public
Utilities Fortnightly & Spark
