Perspective
In Defense of the 'Gold Standard'
May 15, 2003
By Jeff D. Makholm
It is hard tyo foresee abandoning the discounted cash flow method relied upon so heavily for the past couple of decades.
In the Feb. 15, 2003, edition of Public Utilities Fortnightly, Jonathan Lesser says that regulators need to rethink the traditional discounted cash flow (DCF) method for finding the cost of capital, or "at the very least, regulators should no longer rely solely on the DCF to set allowed returns."
This is an old issue. Lesser may (or may not) be aware that commissions in the United States have been searching for a way to streamline utilities' rate of return investigations for decades, and yet they still generally rely on the DCF method. For practical purposes, this is unlikely to change in the United States, despite the recommendations of those like Dr. Lesser and those before him who have recommended a different path.
The topic of the fair rate of return continues to be a vexing part of utility regulation. No one, anywhere, has yet devised a way to make the process agreeable. Why has the process been so difficult?
It is not possible to assess the adequacy of particular rate of return techniques without looking more broadly at how those techniques fit into the larger regulatory process.
The Current State of the Ratemaking Process
The current ratemaking process is tortuous and often unpleasant, for commissions, utilities and ratepayers. A Mississippi Supreme Court Judge captured a quintessential aspect of the process when he said:
"Utility rate litigation has become sport, a vent for passions. Each contest satiates for the moment, then fuels the appetite for further fight. We shrink from the thought of the season ending."1
That was true when this quote was written(1989), and it's true today. It is not, however, the direct consequence of the actions of attorneys, consultants, intervenors, commissioners, or staff that creates this problem. It is the regulatory process that makes it almost inevitable that rate case issues are subject to repeated, increasingly detailed, and costly inquiry. This regulatory framework not only provides a questionable incentive for efficient operation for utilities, it also is expensive. Both of these features create an environment for contentiousness over the issue of rate of return.
Incentives for Efficiency
The current regulatory framework sets efficient utility behavior as its goal but always seems to fail to reach it. Why?
First, the definition of efficiency is elusive. It is difficult for regulators, consultants, accountants, and sometimes the company itself, to distinguish efficient behavior from inefficient. While measures of utility efficiency have been developed (e.g., labor productivity, total factor productivity, number of complaints, etc.), there always will be a large component of utility performance that falls outside of what can be objectively analyzed and measured.
This inability to effectively monitor performance means that hands-on regulators are doomed to steer a course between dangerous rocks and a powerful whirlpool. By steering away from the pure cost-plus contract, where ratepayers face runaway costs, regulators risk being drawn into periodic and sometimes large disallowances that threaten utility financial integrity and ratepayer security.
Second, this failure to have objective standards for efficiency is compounded by "information" and/or "agency" problems. It is difficult for outsiders or those without years of experience to evaluate the decisions of utility managers (or to even know what those managers do). Utility managers are likely always to be more informed than regulators or their staffs regarding the company they manage. It is terribly difficult to monitor utility decisions when the information flow is so incomplete or when regulators must rely on utility managers to volunteer information on poor decisions.
Third, and most pertinent to rate of return, objective confirmable standards may never exist to confirm estimates of costs. In the case of rate of return, there is no way of knowing what the true fair rate of return is (or was), even in hindsight. All we ever have is forward-looking rate of return estimates and historical earned returns. This is not so for any other cost category. For example, estimates of depreciable lives can always be updated by experience with actual capital assets. The same is true with estimates of marginal cost-experience will tend to confirm better estimates in the future. But the "true" rate of return is always unverifiable.
Cost of the Process
The second major problem with the current ratemaking process is its cost. Not only does the process serve us poorly, it is very expensive. There have been attempts by the Federal Energy Regulatory Commission (FERC) and the New York Public Service Commission to regularize the rate of return component of rate cases. Both have failed to do so. The Generic Financing Proceeding in New York, conducted in the early 1990s, despite its staggering price tag in professional fees and the loss of productive time for utility and commission employees, failed, as FERC's did before it, to streamline the process.
The Evolution of Rate of Return Analysis
The fair rate of return became a hotly contested issue in the early 1970s, when the electric utility business was undergoing the "triple threat" of unprecedented inflation, rapid fuel price increases, and the end of decades of impressive technical advances in lower-cost generating technology. The DCF and capital asset pricing model (CAPM) methods got their start at this time and have survived nearly unchanged as the primary rate of return methods, with the DCF the virtual default method in practically all U.S. regulatory jurisdictions.
Improvements in the theoretical accuracy, objectivity, and reliability of these methods have come at a snail's pace and generally address only minor issues. For example, more than 20 years ago, arguments raged in rate of return proceedings over whether to use forward-looking, rather than historical, information in the financial models used to calculate the rate of return.2
A dozen years ago, the argument had progressed to smaller issues (in terms of the potential effect on rate of return) such as the ex-dividend date adjustments and the inclusion into the sustainable growth model of an allowance for the selling of stock at prices above book value.
Meanwhile, every seeming advance in rate of return analysis is followed by a retreat. Historically based "comparable earnings" analyses, presumed dead after the advent of the well-grounded financial theories like DCF and CAPM, have risen from the ashes of past regulation to be considered a genuine rate of return technique. Furthermore, sound theoretical models are often sacrificed on the altar of ad hoc adjustments, when staff or company analysts scramble to move a model's results down or up for a never-ending variety of reasons that are impossible to verify empirically or theoretically.
It remains true today that most rate case issues, with the exception of major cost items, are capable of being settled in relatively short order except for rate of return, where the old issues are continually battled out. So, what are the options to reduce the scope of the interminable fighting over rate of return?
Previous Attempts to Standardize Rate of Return Analysis
Rate of return techniques abound, but very little time and attention is paid to determining which have practical usefulness. The theories that underlie the empirical determination of the cost of capital have become increasingly arcane and irrelevant to the practical ratemaking world, where common sense, believability, and simplicity determine which techniques an administrative law judge or commissioner will use to set the allowed return. At times it seems that the goals of theoretical accuracy and usefulness are mutually exclusive attributes in rate of return models used in utility rate cases.
Although much time is spent discussing the technical aspects of rate-of-return techniques, we never get around to establishing criteria for determining whether they are any good in real rate cases.
Both FERC and the New York Public Service Commission tried to regularize rate of return investigations by concentrating on the method (or methods) involved. FERC's generic rate-of- return process, begun in 1986, ended in a fog of adjustments for a seemingly endless procession of "special cases." The 1991-1993 Generic Financing Proceeding in New York, which was designed to produce an objective standard for setting the fair rate of return, has not streamlined the process. On the contrary, the outcome of that generic proceeding (which was never adopted by the commission in New York) continues to haunt New York rate cases like a dusty book of alchemy from ancient times-understandable only to those who witnessed its creation.
The methods adopted in New York, for example, were overly complex, ad hoc, and led inevitably to further expensive fights and litigation when the financial winds shifted. Both proceedings consumed a great amount of time, effort, and expense in an attempt to establish generic rules in the first place. In a larger sense, there was so much ground to make up that the proceedings could not be said to have been worthwhile-even if durable rules had come out of them.
Why the Appeal of the DCF?
The DCF method has endured for most of the past two decades for three basic reasons:
- It rests on a solid, straightforward theoretical base;
- It capitalizes on the depth of U.S. capital markets-meaning analysis can use "proxy groups" of publicly traded companies in the same industry to manage the variability of individual company DCF calculations; and
- It makes use of company growth projections from disinterested industry analysts-a key attribute for a method to gauge the opportunity cost of capital in the mind of investors.3
It is difficult to overstate the practical importance of these three attributes of the DCF method. The CAPM, by comparison, is abstruse as a piece of theory. Further, because most of the components of the calculation are common to all companies (i.e., the risk-free rate and the market risk premium), the CAPM cannot make use of the law of large numbers. That is to say, the problems associated with which risk-free rate to pick, or which market risk premium to adopt, hinder the result, no matter how many companies the calculations are performed upon. Finally, the CAPM has no tie to disinterested company analysts that not only reflect, but also shape, the opinions of investors. It is thus no surprise that the CAPM is vastly less popular among U.S. regulatory commissions as a rate of return method.
The other methods mentioned by Lesser have more debilitating attributes still. The comparable earnings method is generally irrelevant to investor expectations, to the extent that is uses historical earnings data. Risk premium analyses take the cost of debt as given (easy to do, as debt costs are observable), but struggle perpetually with how to calculate the equity risk premium. To the extent that the equity risk premium uses historical data, it is again generally irrelevant to investor expectations. If it derives the premium by reference to a prospective cost of capital method (like the DCF or CAPM), then it is simply not an independent method at all.
In the context of U.S. rate cases, the DCF method's attributes are magnified, as are the drawbacks of the other methods. The nature of the methods to resolve disputes in U.S. utility rate cases is at least as important as the theoretical attributes of the particular methods employed. "Informationally demanding" methods, like the CAPM, do not stand a chance as a method for resolving conflict between contending parties compared to the "informationally simple" methods, with tangible parameters, like the DCF.
As a result, Lesser's "more methods" recommendation is a dead end. He is correct to point out the greater variability in estimates now than in years past. Mergers, the advent of holding companies, and deregulation all have served to shrink the number of companies to which the DCF analysis can be applied. Nevertheless, the marginal reduction in proxy group size or stability in the past decade does not countervail the three underlying reasons why the DCF is so popular.
Perhaps the best way to deal with the perpetual contention surrounding the rate of return is not to repudiate the overwhelmingly preferred DCF method or pretend that more methods and more investigation might work, but rather to shrink the scope for contention surrounding the issue.
Reducing Rate of Return Conflicts
One tried and tested method to reduce rate of return contention is to turn to alternative regulatory frameworks that either eliminate the need to set the fair rate of return or that lengthen the time between rate cases. There are at least three potential ways to reduce rate of return contention.
Unbundling and deregulation. The airline industry, trucking industry, gas production and transport, and electricity generation capacity once fell under comprehensive rate of return regulation but were subsequently deregulated either partially or fully.
Unbundling and deregulation reduces rate of return battles because they reduce the size of the asset base subject to rate regulation. In other words, with a smaller pie, there should be less incentive to fight. For example, in what I called the "contractualization" of the U.S. interstate gas transport industry, the determination of the fair rate of return has become increasingly less important as contractual obligations between gas transporters and distributors replace traditionally regulated rates. And if rate regulation ends completely, then the reason for the fight over rate of return vanishes.
Reduce the number of contested issues. Permitting cost pass-throughs like fuel adjustment clauses, weather adjustments, revenue decoupling mechanisms, and other techniques that remove attrition,4 reduces the need for filing frequent rate cases because it eliminates factors outside of management's control.
Institutionalized price cap regulation. Price cap regulation, or performance-based regulation (PBR), allows prices to be indexed to both the general price level and durable industrywide productivity trends. As such, it has reduced the frequency of contested price-setting cases where rate of return is an issue. However, PBR regulation does not prevent rate of return from arising as an issue when it does appear.
In conclusion, contention over the fair rate of return is an unavoidable component of utility regulatory oversight, even under alternative frameworks. Efforts to make the process objective and mechanical are futile as an administrative and practical matter. The only realistic way to reduce rate of return contention over the long term is to unbundle and deregulate utility functions (like gas transmission), and lengthen the time between rate cases by instituting PBR or other progressive regulatory programs.
For the rate of return contention that does exist, it is hard to foresee abandoning the DCF as the gold standard relied upon so heavily by U.S. commissions and utilities for the past couple of decades. From the perspective of dispute-resolution techniques and simple efficiency, the DCF-despite its difficulties-retains attributes that other rate of return methods just cannot match.
Endnotes
- Justice Robertson, Mississippi Supreme Court, State of Mississippi et al, v. Mississippi Public Service Commission and Mississippi Power Company, Jan. 4, 1989.
- In reality, this issue has, alas, never gone away entirely.
- Regulatory commissions outside the United States do not have the luxury of either such deep capital markets (with many publicly traded companies in the same industry) or the associated vast array of stock analysts. As a result, they use other methods, but with less robust results and often more extensive contention.
- Attrition occurs when earnings are depressed over time because the marginal cost of new plant and equipment exceeds average costs and average prices.
Jeff Makholm is a senior vice president at National Economic Research Associates Inc. (NERA) and co-chair of NERA's Energy Practice.
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
