Stranded Investment: Utility Estimates or Investor Expectations?

Fortnightly Magazine - June 1 1997

Ask this question: Are Investors today earning what they thought they would, back when they last had faith in regulation?


As their customers discover more competitive prices, many utilities remain saddled with the costs of uneconomic plant and power purchase contracts approved under regulation. They seek compensation for these costs, but the amount deserves a close examination.

Some utilities seek remuneration that exceeds the market value of their common stock. Such a settlement seems overly generous for investors, who will continue to own their shares after the payoff. Some utilities with large claims also want a once-and-for-all regulatory determination of the settlement. These utilities want the amount frozen by securitization plans, like those promised in new laws enacted in California and Pennsylvania.

Of course, before regulators can begin to determine how much compensation utilities deserve for stranded costs, they must first determine whether to award any compensation at all. They must also decide which assets deserve payoffs and how utilities ought to dispose of the amounts recovered. The last step, however, is often overlooked: Regulators must design a method of calculating the payoff that promotes efficiency and fairness.

We argue for a reorientation of the stranding debate that emphasizes the fortunes of investors and de-emphasizes the booked costs of utilities. In fact, recovery of expected investor returns should be the basis for a reasonable armistice in the stranded cost wars. If investors are the central actors of the stranding drama, calculations that disregard their fortunes have little claim to primacy in a debate whose stakes are measured in the hundreds of billions.

The calculation, then, should start from forward-looking market data rather than the historical constructs that underlie the lost-revenues and market-to-book methods. Calculations based on investor expectations are conceptually clear and consistent with the realities of finance and power markets. They are already familiar to utilities, regulators and financial analysts, and offer a consistent treatment for the diverse situations of individual utilities.

The Impact of the "Compact"

As the importance of stranded investment became apparent, advocates of full recovery typically faced off against advocates of zero recovery. Those with views between those extremes usually saw a partial payout as a politically expedient strategy rather than a matter of principle. %n1%n The intellectual foundation of full recovery was a metaphorical "Regulatory Compact" that advocates claimed had long governed the industry. Supposedly, this compact guaranteed the returns of utility investors in all but extreme circumstances, because they had forsaken the high risks and high rewards of ordinary stocks for the low risks and low returns of a regulated industry.

On the other side, advocates of zero recovery also noted that the real returns earned by utility investors could hardly be described as "low." They also questioned the veracity of allegations that regulators had at times compelled utilities to make uneconomic investments. If so, as competition arrives, regulators should have no reason to further protect investors from the consequences of decisions by utility management who served at the sufferance of those investors. By contrast, proponents of the Regulatory Compact view incomplete recovery as an uncompensated taking that rises to the status of a constitutional question.

Extrapolating decisions on the regulation of real property to the regulation of utilities is difficult, but the Supreme Court has been generally reluctant to interfere with state-level actions that entail less than a complete taking of a property right or a physical invasion of that property. %n2%n Regarding utilities, the Supreme Court's decision on regulatory disallowances in Duquesne v. Barasch may apply to stranded costs. In that case, the court held that regulators enjoyed wide discretion regarding cost recovery, but suggested that a constitutional issue might arise if uncompensated stranded costs endangered a utility's financial integrity. %n3%n In one recent case, the New York Supreme Court rejected the compact as justification for full recovery, and instead ratified a regulatory decision to determine compensation on a case-by-case basis. %n4%n

If Recovery, How Much?

As a practical matter, regulators and legislators are now coming around to the view that some implied agreement once existed and that some stranded cost recovery is warranted. Given this assumption, one might conclude that ratepayers who left utilities to find their own suppliers breached their obligation to take service. If so, the departing customers could be required to pay "expectation damages" that leave utility investors no worse off than they would have been if the compact had continued to run unabated.

Few participants in the debate have tackled the question of how to compute stranded costs. Some utilities favor charging departing customers amounts that leave revenues intact if the customer had remained in the system. In fact, in its open-access Order 888, the

Federal Energy Regulatory Commission favors this policy for wholesale purchasers who unexpectedly leave utilities. Regulators and legislators have also proposed granting utilities the difference between the booked value of their uneconomic assets and the expected market value of those assets under competition. In California, an important determinant of a utility's compensation will be the amount it receives for plants being divested as part of the transition program.

The lost-revenue calculation and the market-to-book comparison both ignore the returns that investors expected to receive under continuing regulation, and neither examines the outcomes investors actually experienced. If fairness requires payments, then the amount should reflect how compensation affects investors' fortunes rather than how it affects the utility's books.

The Financial Expectations Method

To arrive at compensation to keep utility investors whole, begin with the financial expectations of those who bought utility stock when a consensus still prevailed that the old regulatory scheme would persist indefinitely. Call this idea the "Financial Expectations" method. Since stock prices embody these expectations, find the price as of the last date investors expected regulated returns to persist indefinitely. One possible date is the last state rate case in which neither the utility nor its regulators mentioned a threat of retail competition. Assume that this case was decided in 1989, and retail wheeling was set to begin in 1998. To determine stranded cost recovery, estimate returns to investors over a suitably long period (e.g., 20 years) that bridges the transition from regulation to competition. Use this result to estimate the change in the value of equity due to the coming of competition.

This method requires data on the utility's past performance and its estimated future earnings per share, looking forward from 1989. One can then discount these two figures (past earnings from 1980 to the present; future earnings to 2008) by the rate of return on common equity that regulators authorized in the 1989 rate case. This present value indicates calculated (or "realized") returns after 1989. The 1989 stock price shows the "expected" returns in 1989. If the present value of realized returns is less than the price of the stock in 1989, then investors will not have earned the returns they expected.

If future payouts are required to make investors whole, regulators must determine how to build them into rates and allocate them among customer classes. Since only the present value of payments matter for this proposed method, payments may be made in either a lump sum or a steady stream. Adjustments may prove necessary if the utility's realized future performance diverges significantly from the model's predictions, if, for instance, adverse market conditions bring an unexpected revenue shortfall that renders the utility incapable of meeting its commitments to bondholders.

Any adjustments, however, should be determined as the need arises, rather than as fixed amounts computed in the heat of a transition to a market with an uncertain future. Estimates of unrecoverable plant values are sensitive to assumptions about future market conditions and the speed at which retail competition will spread. %n5%n Once-and-for-all determinations are not necessary for access to capital. Uncertainty about future markets and stranding recoveries has adversely affected the financial health of only a few utilities. %n6%n

Identifying the Benefits

Returning to investors an amount consistent with their expectations under the old regulatory regime is fair, efficient and better grounded in reality than giving them a payoff that depends on the vagaries of regulatory accounting. The Financial Expectations method accounts for the returns long-term investors actually realized in the past and for the returns they are likely to earn after competition comes. It reduces their recovery if their past returns exceeded those authorized by regulators, and augments recovery if past performance fell short.

This method is far from a perfect solution, but may emerge as the best without reopening a decade of regulatory dockets. Ideally, investors should be able to keep supernormal returns made by good management that earlier stranded some investments. Investors should also be denied payments that make up for substandard performance by less competent managements. Retrospective redeterminations, however, are unlikely to impose heavy burdens on scarce regulatory resources in return for questionable benefits.

A reasonable compromise on stranded cost begins by presuming the existence of a Regulatory Compact. Under that compact, however, investors accepted a ceiling on their returns. Fairness to ratepayers requires that they pay no more than is required to give the investors the amounts for which they contracted. These amounts may be quite unlike those calculated by the lost-revenues or market-to-book methods.

A firm, bottom-line number based on regulatory accounting mechanisms should appeal to those who still base their thinking on "revenue requirements." Utilities were once dependable sources of long-term returns, recovering the booked costs of their assets and reinvesting the proceeds to produce secure future income. As the industry moves to market, so must this insistence that it is booked costs rather than market returns that matter to investors. Utilities have one further responsibility about which little has been said: However compensation is paid, along with those payments, utilities must put shareholders on notice that they are now are on their own. t

Steven Isser is a consultant with Hagler Bailly Consulting of Arlington, Va. He holds a Ph.D. in economics from the University of Texas and is currently completing law school there. Robert Michaels is Professor of Economics and California State University, Fullerton, and senior advisor to Hagler Bailly Consulting. He holds a Ph.D. from the University of California, Los Angeles. The views expressed in this article are not necessarily those of the authors' affiliations or clients.



Calculating Expectations


On Jan. 1, 1987, an investor buys one share of stock in Utility X for $100. Assume the stock earns a $10 dividend at the end of every year and is expected to do so forever. Regulators have conveniently set the utility's authorized return on equity equal to the rate on similar safe investments (em 10 percent. The market price of the share will thus be $100, equal to the discounted value of its long-lived stream of future dividends.

Then assume deregulation comes without warning at the end of 10 years of dividend payments, immediately cutting the stock price to $50. How much stranding compensation, if any, is due for stranded costs under each of three different scenarios: 1) constant (expected) end-of-year dividends of $10; 2) a higher (unexpected) annual dividend of $12; or 3) a lower (unexpected) dividend of $8?

1. Constant Dividend.

For 10 years, the stock fulfills the investor's expectations, with $10 dividends paid annually. The price remains steady at $100. (Capital gains are omitted here but can be included at the cost of algebraic complexity.) Just after the 1997 end-of-year dividend is paid, however, instant deregulation occurs and the $100 price falls to $50 on Jan. 1, 1998. If, on that date, the investor sells the share at $50 and receives $50 for stranded costs, he will still have $100 in assets, as expected on the purchase date of the stock:

[$10 ( 1.10] + [$10 ( (1.10)²] + ... [$10 ((1.10)10 ] + [($50) ( (1.10)10] + [($50) ( (1.10)10] = $100

The final three terms are the Dec. 31 dividend payment, the Jan. 1 sale price of the stock and the Jan. 1 receipt of stranding compensation.


2. Rising Dividend.


Next, assume that just after the investor bought the stock for $100 on Jan. 1, 1987, the utility raised its dividend to $12 per year while its cost of capital remained the same. (This situation is realistically more akin to capital gains.) As before, after 10 years of $12 dividends, sudden deregulation depresses the stock price to $50. An investor who also receives an additional $50 for strandings is overcompensated relative to expectations at the time of purchase. Stranding compensation of $18.15 is all that is necessary to fulfill his 1987 expectations:

[$12 ( 1.10] + [$12 ( (1.10)²] + ... [$12 ((1.10)10 ] + [($50) ( (1.10)10]+ [($18.15) ( (1.10)10] = $100

3. Falling Dividend.


This method can also calculate compensation for investors whose expectations were frustrated (presumably not because of utility imprudence). Just after the investor buys the stock, dividends fall to $8 and continue at that level over the first 10 years. Just after the tenth dividend payment, the stock price falls to $50 with deregulation. Now a Jan. 1, 1998 stranding payment of $81.88 leaves the investor with an income stream that has a present value of $100, looking forward from 1987:

[$8 ( 1.10] + [$8 ( (1.10)²] + ... [$8 ((1.10)10 ] + [($50) ( (1.10)10] + [($81.88) ( (1.10)10] = $100


1For polar views, see William J. Baumol & J. Gregory Sidak, Transmission Pricing and Stranded Costs in the Electric Power Industry (Wash. D.C.: AEI Press, 1995), and Robert J. Michaels, "Unused and Useless: The Strange Economics of Stranded Investment," 7 Electricity Journal 12 (Oct. 1994), 12-22. On what utilities should do with the funds, see Robert Michaels, "After Stranding Recovery, What?," PUBLIC UTILITIES FORTNIGHTLY, June 1, 1996, 14-16.

2See J. Gregory Sidak and Daniel F. Spulber, Deregulatory Takings and Breach of the Regulatory Contract, 71 N.Y.U. Law Review 851 (1996); Lucas v. South Carolina Coastal Council, 112 S. Ct. 2886 (1992)[deprivation of all economically beneficial use required]; Yee v. City of Escondido, 112 S.Ct. 1522 (1992)[upholding rent control]; but see Dolan v. City of Tigard, 114 S.Ct. 2309 (1994).

3Duquesne Light Co. v. Barasch, 488 U.S. 299, 312-314 (1988).

4Energy Asso. of New York State et al., v. New York PSC, 653 N.Y.S.2d 502, 174 PUR4th 406 (Sup.Ct.1996).

5Moody's recent update of a 1996 report showed little change in aggregate stranded costs, but estimates for over 75 percent of utilities changed by more than 10 percent over the year. "New Moody's Survey Shows Many Changes in Estimated Stranded Costs and Prices," Electric Utility Week, Jan. 27, 1997, 11.

6Donaldson, Lufkin, and Jenrette, Electric Utility Company Outlook, Dec. 1996. Some utilities with large exposures have claimed that less than full recovery will lead them into insolvency, or, at worst, bankruptcy.



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.