Bargain or Bonanza


Is discounted cash flow (DCF) still a reliable tool for determining equity cost?

Fortnightly Magazine - August 2013

Capital-hungry electric utilities depend on return on equity (ROE) to appropriately compensate existing equity investors and satisfy new equity investment requirements, and look to the Federal Energy Regulatory Commission (FERC), which has extensive transmission jurisdiction, for protection of transmission ROEs. FERC has used discounted cash flow (DCF) analysis to determine ROEs for many years, but the $54 billion question – EEI’s estimate of near-term transmission investment requirements1 – is whether the DCF method is up to the ROE challenge in the current economic climate. 

FERC has to decide whether rock bottom interest rates, high market prices for utility stocks, and low electric consumption growth prospects are distorting its traditional DCF analysis and causing that analysis to produce unrealistically low common equity costs. In so many words, if FERC were to adopt the low ROEs generated currently by the DCF model, would that result in missed opportunities for investment in needed new transmission facilities, and, as a consequence, produce short-term customer savings – but at the expense of retarding grid development and causing a long-term escalation in customer costs?2

DCF Theory and Practice 

The Supreme Court’s Bluefield and Hope decisions3 form the foundation of FERC’s ROE regulation and require ROEs that are sufficient to maintain a company’s financial integrity and attract capital, and that also correspond to the ROEs of comparable-risk companies. In FERC’s view, the DCF method neatly allows compliance with Bluefield and Hope criteria. Premised on efficient market theory of widely available financial information, the DCF method postulates that common equity cost can be derived by adding the dividend price ratio (D/P) of a company’s common stock to the investors’ expected (constant and infinite) growth rate in the stock’s dividends per share (G).4 According to theory, the DCF method also has the effect of quantifying investor perceptions of relative risk through the differences in DCF results for a particular company’s common stock vis-à-vis other companies’ common stock. Thus, the DCF method in theory produces a result that equals a utility’s equity cost, is consistent with investor expectations, and thus allows the utility to attract capital – and is commensurate with the expected returns realized not only for comparable utilities but, on a relative risk basis, for all U.S. companies with publicly traded common stock.

Although originating in complex mathematical theory and models, DCF analysis is applied in practical circumstances. Currently, such circumstances include massive government intervention in the economy; an emphasis on energy conservation; unacceptably high unemployment; a frail economic recovery; lingering fears of a double-dip recession; and concern over higher taxes, which would hit the utility sector particularly hard if applied to dividends – a critical component of the expected return for investors in utility common stocks. The economic mix also includes a mounting national debt, much of which places us at the mercy of foreign entities and their continued willingness to use the U.S. dollar as the world’s major reserve currency; the upward progression of entitlement spending; and the elusiveness of a long-term budget deal between proponents of higher taxes and advocates of spending cuts. In addition, as non-U.S. economies continue to falter, fears of global economic meltdown weigh heavily on investors in the U.S. stock market.

Within that economic environment, FERC’s DCF model also faces the daunting challenge of producing ROEs that will satisfy large new investment requirements. The EEI $54 billion forecast for new transmission investment is a short-term requirement for only the 2011 through 2015 period. This investment is needed to enhance grid reliability, cut congestion, and connect wind and other environmentally benign resources to the grid, a priority identified by President Obama in his Georgetown University speech on climate change.5 Moreover, the same utility industry, which must raise billions for new transmission, also must raise billions of dollars to expand, modernize and, in many cases, stormproof electric distribution systems and reconfigure them to allow bi-directional flow to accommodate distributed generation. 

If the plate weren’t already full, utilities that have retained the generation function must also spend billions to retrofit existing generation and construct new generation.6 Common equity financing will figure in this capital formation directly through common equity issuances and indirectly through debt financing, which requires a common equity foundation. 

Valid Concerns or Poor-Mouthing?

The DCF method was built for the long haul as reflected in the model’s theoretical assumption that the expected growth rate is constant literally to infinity. The method presumes that the economy will function normally over the long-term. The FERC DCF conundrum is whether the model is equipped to deal with and accurately reflects the cost of capital in the current economic environment, which can hardly be described as normal. 

In 2008, the U.S. was on the brink of economic catastrophe with failing financial institutions, decimation of the stock market, a collapse in employment, and the same predicaments facing other economies around the globe. The response – massive government intervention in part to re-liquefy the capital markets by expanding the money supply and pushing interest rates down to artificially low levels – has continued even though the economy has stabilized. The result is that common stock prices, including utility stock prices, have largely recovered and largely exceed 2008 levels. Further, if income investors are viewed as having bid up prices of utility stocks as temporary safe havens until confidence returns in the economy, then the prices of those stocks are higher than justified by economic fundamentals, and wouldn’t survive the yield reductions that would result if ROEs were to be reduced. 

The continuation of these economic conditions and the government intervention in the capital markets to maintain an expanded money supply and low interest rates has a major effect on DCF analysis. The increase in stock prices fueled by that policy has reduced dividend yields, i.e., the ratio of dividends to prices (D/P) in the DCF formula. Hand-in-hand with the D/P reduction, economic anxieties negatively affect expected growth rates (“G” in the DCF formula), which haven’t increased to offset the reduction in D/P. As a consequence, the DCF method is producing comparatively low common equity costs. The dilemma facing regulators, utilities and their customers is whether government intervention and an abnormal economy are having distortive effects on DCF analysis and whether, as a consequence, DCF analysis fails to accurately capture the cost of common equity in today’s capital markets, which could pose greater – not less – risk than in prior years.

For utilities, the answer is easy and obvious. They make the hard-to-refute point that the low DCF results are the product of artificially low interest rates, which are pushing up common stock prices, and of depressed dividend growth rate expectations that arise out of the same economic anxieties that are fueling the government intervention to maintain low interest rates. They regard these government-created low interest rates as proof of the high risk inherent in the economy, since low rates are necessary to reduce apprehension and induce borrowing to stimulate the economy. They contend, with significant plausibility, that it’s counter-intuitive to believe that since 2008 utility risk has receded while the overall level of risk in the economy remains high. They assert that today’s low interest rates, given their provenance, don’t have the force of gravity that would drag down common equity costs. To the contrary, they conclude that interest rate reductions haven’t been matched by corresponding common equity cost reductions. Rather, they see an increase in the spread (the equity risk premium) between the governmentally created low interest rates and common equity costs. They would cite the market volatility following the Federal Reserve’s recent suggestion of the possibility of a prospective tapering of its cheap money policy as showing investor fear that higher interest rates would undermine the stock market and derail the economic recovery and as demonstrating the high level of perceived risk that pervades the capital markets.7 They regard some forms of DCF analysis as oblivious to those realities, and as now sending false signals understating ROE risk and cost just as utility risk and cost and investment requirements are approaching all-time highs. They find the ROEs resulting from some current applications of the DCF model to be inadequate, confiscatory, and below the minimum needed to undertake required new investments. 

FERC-regulated transmitting utilities have a lot at stake since ROE is the sole shareholder compensation. Not surprisingly, these utilities are at loggerheads with their wholesale customers who like the current DCF results and the rate reductions those results would make possible. Their customers reject utility concerns about the DCF model as result-oriented poor mouthing that’s unrelated to any DCF deficiencies. To obtain rate reductions, customers have filed a number of complaints at FERC alleging “significant changes in market conditions” since current ROEs were established,8 and proposing base ROEs that hover in the mid-8 to low-9-percent range. Utility rate filings have met with stiff resistance as protesters aggressively press for lower ROEs, arguing that “the capital markets today are vastly different” from what they were in the past, and that “the cost of capital for electric utilities is far lower than it was … years ago.”9

FERC’s Reaction

FERC has reacted to the ROE controversy in part by placing many ROE decisions on hold. The lack of action on one ROE complaint, which included a request for expedited consideration, prompted the same customers to file a second complaint against the same utility, making essentially the same allegations and seeking essentially the same remedies.10 While FERC has set a few matters for hearing and settlement discussions, ROE issues are still pending in approximately 20 proceedings. The processing of these cases might have been delayed because FERC perhaps is caught between the proverbial devil and the deep blue sea: reluctant to modify its DCF approach, but uneasy that its approach might produce unacceptably low results that will discourage transmission expansion.

Delay in resolving ROE controversies harms the interests of both utilities and their customers. Many utilities face significant prior-period contingent liabilities since their rates are already being collected subject to refund. They have equal concerns over the future. Low DCF outcomes could result in significant future economic losses through prospectively effective ROEs at levels below their perceived cost of common equity capital. Customers want FERC action because they believe existing transmission rates are excessive, unjustly enrich utility investors, and should be reduced as soon as possible. 

The FERC ROE logjam is about to break. An ROE hearing has recently been completed for the ISO-New England Transmission Owners, and an initial decision will be issued after briefing is completed.11 FERC has issued summary judgment orders in two recent cases requiring the rate proponents to conform their ROE evidence to FERC’s prescribed approach regarding the use of proxy groups in DCF analysis.12 And a FERC commissioner has publicly stated that ROE is among the most important issues on FERC’s agenda.13 Thus, the scene is set for FERC to address the backlog of ROE cases and shape a policy that will create a favorable climate for investment in needed transmission infrastructure.

Role of Judgment 

A powerful attraction of DCF methodology is the desire to narrow the scope of controversy in ROE decision-making, provide a more transparent, formulaic approach to determining the allowed ROE, and thereby avoid the pre-DCF painstaking, time-consuming and often highly subjective evaluation of individual utilities’ financial circumstances. FERC’s SoCalEd decision, in which the arithmetic median of the proxy group ROEs is equated with the just and reasonable ROE, epitomizes a mechanistic application of the DCF approach.14 This approach minimizes the exercise of judgment and might be regarded by FERC as the best way or even the only way of avoiding the administrative burden – which detracts from FERC’s ability to address other regulatory responsibilities – of endless, costly and sometimes seemingly pointless litigation about fine points of DCF-ROE methodology and whether that methodology has been applied reasonably and consistently. 

Nonetheless, DCF methodology is an imprecise instrument, and the output of the DCF formula can’t be accepted as the unimpeachable correct result obviating the need for careful ROE evaluation. DCF analysis provides at best an approximation of a utility’s cost of equity, and requires the exercise of judgment to determine whether the estimate needs adjustment in order to coincide with a utility’s actual common equity cost. 

For example, a savvy investor would never dream of predicating an investment decision solely on the output of the DCF formula. Instead, using the DCF method for its intended purpose of common stock valuation, the investor, whether making decisions for himself or an entire mutual fund, would use the DCF result as part of an overall evaluation of qualitative facts and alternative quantitative methods to determine whether a particular security should be acquired. What distinguishes the FERC decision from the private investor decision is that the FERC decision dwarfs the private investment decision in importance. In contrast to the private investor’s pursuit of self-interest, FERC is charged with protecting the public interest, which mandates that it exercise careful judgment for the protection of the legitimate interests of the investor and the long-run best interests of the consumer in a reliable, adequate, and reasonably priced electric supply. 

Judgment must be applied not merely to the DCF output but also to the DCF input. From its early days of DCF utilization, FERC recognized that the objectivity seemingly inherent in DCF methodology was more apparent than real. A 1978 FERC order praised DCF analysis as “forward-looking,” but also cautioned “the result of any DCF analysis depends importantly upon the assumptions used and the choice of input data [which] … are matters of judgment.”15 The following year, a FERC administrative law judge found that “DCF’s appeal as an objective measure” is “illusive.”16

Furthermore, DCF theory doesn’t provide a blueprint for applying DCF methodology to FERC regulation. Subject to arguments by DCF practitioners often seeking to skew results to produce preferred outcomes, FERC has had to adopt various DCF permutations. These entail departures from pure DCF theory. They also entail the risk that strict application of the FERC ROE method simply produces a series of arithmetic calculations, each one of which might appear sound but the aggregate end result of which could lack logical nexus to a utility’s true cost of equity capital.

FERC’s reliance on proxy groups epitomizes its reliance on DCF input judgments. DCF theory doesn’t provide for proxy groups and DCF quantitative analysis isn’t used to select proxy groups. FERC uses the proxy group as a device to offset the imprecision in DCF analysis as applied to any one company. The FERC proxy group theory is that applying DCF analysis to companies that are similar in risk to the rate applicant produces a range of accurate results that delineate the zone of reasonableness within which FERC then sets the rate applicant’s cost of equity. 

Disputes over proxy group selection, analogous to disputes over jury selection in court trials, also show the subjectivity inherent in applying DCF analysis to ROE determinations as parties seek to populate the proxy group with companies that will achieve their ROE objectives. Rate applicants seek proxy groups that include high-ROE utilities and exclude low-ROE utilities as non-comparable. Taking the opposite tack, customers seek proxy groups heavily dominated by low-ROE utilities. FERC ultimately determines the proxy group, and has established guideline selection criteria, but its case-to-case determinations are the product of judgment, not an arithmetic formula.

FERC also exercises judgment in how it uses the proxy group results. FERC’s prior practice was to use judgment based on a rate applicant’s relative circumstances as to where within the proxy group zone of reasonableness (established by the proxy group ROEs) it would establish the just and reasonable ROE. FERC then exercised judgment to equate the mid-point of the proxy group ROEs with the just and reasonable return. In its 2010 SoCalEd decision, FERC again exercised judgment to formally change its then-existing policy by discarding the proxy group midpoint in favor of the proxy group median to determine common equity cost for individual utilities.17 That exercise of judgment can have a material effect on the ROE determination. If the proxy group consists chiefly of low-ROE utilities that cluster below the midpoint, as it would under current low ROE conditions, the ROE applicant will prefer the midpoint, i.e., the average of the high- and low-end results, while customers will prefer the below-average median. 

In an economic blow to utilities, FERC’s decision to use the proxy group median rather than midpoint was recently affirmed by a U.S. Court of Appeals.18 However, the court’s decision had nothing to do with DCF theory and was almost entirely based on FERC’s discretion to exercise judgment and make reasoned choices between two alternatives. Even the underlying FERC decision to prefer the median over the midpoint wasn’t compelled by DCF theory which 1) doesn’t concern proxy groups, and 2) would be equally consistent (or inconsistent) with use of the midpoint or the median. That said, FERC’s adoption of the median does illustrate a potential gap between mechanistic DCF determinations and actual investor decision-making. The choice of the median credits the dubious proposition that a hypothetical investor using DCF theory for valuation purposes and having access to FERC’s proxy group information would deliberately choose to invest in a utility with below-average earnings prospects rather than a comparable-risk utility with better-than-average earnings prospects or with the best earnings prospects. 

Similarly, the expected dividend growth rate (“G” in the “D/P = G” formula) is the product of judgment exercised at each and every stage of its determination. In contrast to the dividend and the price used to calculate D/P, which can be observed from market data, there are no equivalent data to directly determine G. Actually, FERC’s DCF approach considers two growth rates – security analysts’ expected growth in earnings as compiled by IBES, and sustainable internal dividend growth (SIDG) derived from projected Value Line data. With respect to analysts’ forecasts, the availability of alternative sources from multiple publications tempts some parties to cherry-pick among forecasts to support desired results, and requires FERC to resolve disputes and make the final judgment. 

The SIDG calculation entails a different kind of judgment. The SIDG forecast is typically lower than and thus reduces the G that would be derived from analysts’ forecasts. Investors are keenly aware of the analysts’ forecasts, which are the primary focus of the investment community, so according to efficient market theory, investors would include those growth forecasts in their investment decisions. Also, at least in theory, a high ROE could support high dividend growth at the level contemplated by the analysts’ projections. Viewed in that light, FERC is needlessly second-guessing the analysts’ forecasts and finding them unduly optimistic and unrealistic. 

SIDG proponents correctly respond that the SIDG calculation is also consistent with efficient market theory, since investors are aware of the SIDG Value Line data. But that response begs three questions. First, it implausibly suggests that the analysts responsible for the growth forecasts, who have access to the Value Line data used for the SIDG calculation, somehow missed the boat, failed to recognize the need for an SIDG calculation, and thus knowingly put out pumped-up growth information that would mislead their subscribers. Second, Value Line’s reports are prepared by an analyst who presumably is aware of other analysts’ growth expectations and who typically develops and publishes his own growth forecast. Nevertheless, FERC’s method derives a growth rate that’s often different from the analyst’s, even though FERC utilizes the analyst’s own data. Finally, the SIDG calculation implies the need for an accuracy check of the analysts’ growth projections (i.e., whether a particular growth rate is sustainable), whereas DCF analysis concerns investor expectations without regard to whether those expectations are accurate and reasonable or inaccurate and arbitrary.

This is not to condemn the SIDG calculation but to illustrate that the exercise of judgment in ROE analysis is inescapable and that FERC’s method for estimating expectations of growth, whether or not the best possible solution to a very difficult task, is subjective and heavily dependent on judgment. This judgment is exercised in the selection of the analysts’ forecasts, in the decision not to place exclusive reliance on those forecasts, and in the decisions made in the development and application of its SIDG calculation. Judgment is even involved in FERC’s decision to blend the two growth measures rather than rely on each of them to facilitate development of a range of growth rates and a range of estimated common equity costs. 

Although the appeal to FERC of DCF analysis is the use of quantitative analysis, there should be no expectation that the just and reasonable FERC ROE springs from a mathematical DCF formula deus ex machina and untouched by human hands. No such formula exists. Critical steps in the ROE determination are infused with the exercise of judgment that have little or nothing to do with DCF analysis. This includes the threshold decision to convert an asset valuation technique into a mechanism for determining ROEs, the decision to use subjectively selected proxy groups, the manner in which proxy group data are employed, the calculation of the dividend growth rate, the decision to use or not use alternative methodologies to corroborate DCF results, and, in earlier days of DCF implementation, the choice of the just and reasonable pinpoint within the zone of the reasonableness based on a utility’s relative financial circumstances. Even the SoCalEd adoption of the arithmetic median of the proxy group ROEs entailed the exercise of judgment that the ROE thereby derived without further adjustment was just and reasonable and in full compliance with Bluefield and Hope requirements. Thus, asking FERC to exercise judgment to determine whether current economic conditions have caused that methodology to malfunction and to produce unrealistically low ROEs is fully consistent with FERC’s exercise of judgment in nearly all aspects of its DCF-ROE methodology.

Are Utilities Facing Tough Times? 

There could be dark clouds on the utility horizon. EEI earlier this year reported nine pending complaints seeking 9.65-percent to 8.7-percent ROEs, and another complaint seeks an even lower return at 8.55 percent.19 A May 13, 2013 customer complaint, claiming to follow FERC’s DCF guidelines, advocates an 8.63-percent to 8.84-percent ROE range.20 More worrisome from the utility perspective, recent FERC staff proposals in pending cases provide ROEs as low as 8.62 percent.21

Even before the DC Circuit affirmed the SoCalEd median decision, FERC on summary judgment had ordered Pacific Gas & Electric and Public Service Co. of New Mexico to calculate their ROEs using the median of the proxy group numbers in the DCF analysis,22 producing ROEs of 8.6 percent and 8.67 percent, respectively.23 These orders appear to etch in stone FERC’s policy of using the proxy group median, and represent a material step toward lower ROEs. Such returns are significantly lower than the returns seen at state commissions for regulated utility operations. They also fall considerably short when measured against the historical spreads between electric returns and the returns approved for FERC jurisdictional natural gas pipelines. 

FERC’s April 15, 2010 order and Oct. 6, 2011 rehearing order in the SoCalEd proceeding also could represent trouble for utilities. In endorsing the proxy group median, FERC heavily relied on the similarity of the proxy group companies to each other and to SoCalEd, on “[t]he laws of statistics,” and the notion that “[t]he median best represents central tendency in a skewed distribution over the mean because the latter is drawn in the direction of the skew more than the median.” A low-ROE proponent might applaud FERC’s acumen in statistics, while a high-ROE proponent might have concern that the emphasis on statistical niceties was oblivious to the turbulence roiling the economy and arguably distorting the results of the DCF analysis.

A second SoCalEd ruling also might concern high-ROE proponents. Relying on its long-standing policy, FERC adopted a downward ROE adjustment based on bond yield changes that post-dated the close of the record and rejected SoCalEd’s proffer of evidence that the bond yield reductions shouldn’t be considered since the declining interest rates were due to the unique conditions of the 2008 market collapse. While the appeals court remanded this ruling on evidentiary and procedural grounds, the FERC ruling could signal FERC’s view that common equity costs change in lock-step with interest rate changes – regardless of prevailing economic conditions. 

Finally, FERC’s 2005 rulemaking to comply with the incentive provisions of FPA Section 21924 elicited comments on alternatives to its DCF analysis for establishing the base ROE. FERC elected to continue using its conventional DCF analysis for ROE, coupled with Section 219 incentive adders enhancing the ROE for projects with increased risk. The risk for utilities is that FERC will use incentive ROE adders as a backstop and justification for the low ROEs produced by its conventional DCF approach. Utilities would contend that Section 219 was never intended for that purpose and that Section 219 might never have been necessary if FERC’s historic DCF methodology had provided transmission-owning utilities adequate ROEs.

What Should FERC Do? 

In our opinion, as its first step, FERC should take a hard look at its 2010 SoCalEd decision. That decision’s problematic aspect isn’t the replacement of the proxy group midpoint with the proxy group median, which can be a useful data point for determining a just and reasonable ROE. Rather, the fundamental SoCalEd concern is whether the adoption of the proxy group median as the just and reasonable ROE elevates arithmetic over judgment, and, contrary to a fundamental Hope precept, elevates the “method used” over the reasonableness of the “result reached”25 as the dominating factor in the ROE calculus.

As part of a needed ROE reconsideration, FERC should determine whether currently low interest rates reflect a corresponding decline in common equity costs or alternatively, an increase in the equity risk premium. Utilities and their customers will be eager to share their different views with FERC. FERC then must make the ultimate decision based on a reasoned analysis of the available economic evidence, but can’t simply assume that common equity costs and debt costs continuously march in the same direction in lock-step with each other. FERC must determine whether prevailing economic conditions are having a distortive effect on DCF calculations and related aspects of its ROE methodology, and whether adaptations and modifications are needed to ensure that its ROE rulings correspond to actual utility common equity costs.

FERC also should consider a more flexible ROE approach. This wouldn’t require abandonment of DCF methodology. To the contrary, abandoning a useful source of ROE information such as DCF analysis would be inconsistent with a reasoned approach that would take account of facts and figures from a variety of sources, and developed through a variety of methods. As a first step in a new approach, FERC might consider modest adaptations to its DCF approach to reflect current economic conditions. These could include enlarging proxy groups, more broadly defining the zone of reasonableness, and approving DCF results at the upper end of the zone to provide ROE headroom and reflect the fact that common equity risk premiums might be significantly greater now than in the past due to the unique financial risks in the current economic environment. 

In this new regime, the proxy group median would continue to be relevant but only as one among the relevant benchmarks for setting the ROE within the zone of reasonableness. The actual ROE would be selected based on judgment that would factor in the proxy group results, current economic conditions, and the particular utility’s financial and economic data presented in each case. FERC also could consider adjusting DCF results based on consideration of historic interest rates. For example, FERC could consider an adder to the DCF median to reflect the government’s artificial lowering of interest rates. With Treasury bill rates effectively zero and long-term rates at about 2 percent, ample evidence might suggest an adder of 2 percent to the median results based upon historical bond yields during normal times. 

FERC should use DCF results as a starting point rather than the ending point in ROE analysis. Following its early precedent, FERC should recognize that no single approach can reliably achieve an appropriate ROE level, and consider alternative methods, in addition to the DCF method, such as the capital asset pricing model (CAPM) and risk premium analysis applied to forward-looking data. FERC used such methods to “corroborate” the SoCalEd DCF results without needing to address the remedy if the alternative methods and DCF analysis produced inconsistent results.26 Consulting more than one method of estimating cost of equity capital would enable FERC to weigh their respective strengths and weaknesses in an effort to develop the most reliable and accurate estimates of common equity cost. 

FERC itself has stated that DCF methodology is “a reasonable choice from among a number of possibilities,”27 that it shouldn’t accept “blindly” or “uncritically” the results derived from market-oriented formulas such as DCF, and that it wouldn’t “abandon independent judgment of the reasonableness of the final [ROE] result of any analytic technique used.”28 This arsenal of decisions establishes that under current conditions FERC can’t imprison itself in a single ROE methodology, must always engage in reasoned decision-making, and must recognize the distinct possibility that DCF analysis in a challenged, unstable economic climate might produce ROEs that are on the border of – or even in outright violation of Bluefield and Hope requirements. 

We close with these considerations. First, the cost-benefit arithmetic favors resolving any doubt about the fair ROE, which could be inherently unknowable and subject to a wide band of error on both its upward and downward sides, in favor of higher ROEs. This is because the customer cost effect of higher ROEs, which will encourage transmission construction, is small in comparison to the much greater financial savings and environmental benefits new transmission facilities will provide.29 Second, even if FERC-allowed ROEs are inadequate, many or most utilities might attempt to engage in needed new construction in fulfillment of their duty to provide their customers economic and reliable service. However, scrimping on the ROE allowance provides no long-term benefits because inadequate ROEs ultimately will increase the utility’s cost of capital and wipe away any temporary customer cost savings. 

Finally, the regulatory compact involves reciprocity of commitments. A utility’s willingness to fulfill its part of the regulatory bargain by engaging in required new construction regardless of the ROE allowance imposes a corresponding FERC obligation. There must be assurance that allowed ROEs aren’t simply the artifact of a disembodied arithmetic or statistical formula, but must be the product of judgment that takes account of the financial and economic conditions that affect the capital markets, provide definition to investor risk perceptions, and determine a utility’s true common equity cost.


1. Edison Electric Institute, Transmission Investment: Adequate Returns and Regulatory Certainty Are Key, at 5 and 6 (June 2013). 

2. Editor’s Note: Recent Fortnightly articles also have addressed these topics. See “Investor Sequester,” by Bruce W. Radford, June 2013; and “Transmission Policy in Flux,” by David Raskin, May 2013. 

3. Bluefield Waterworks & Improvement v. Pub. Serv. Comm’n of W. Va. (“Bluefield”), 262 U.S. 679 (1923); FPC v. Hope Natural Gas (“Hope”), 320 U.S. 591 (1944). See also, Duquesne Light v. Barasch, 488 U.S. 299 (1989).

4. Charles F. Phillips, Jr., The Regulation of Public Utilities: Theory and Practice, ch. 9 (2005).

5. Remarks by the President on Climate Change, June 25, 2013.

6. The Brattle Group on behalf of the Edison Foundation, Transforming America’s Power Industry: The Investment Challenge 2010-2030, at 13 & 43 (Nov. 2008): “[C]apacity expansion would entail spending $697 billion over the 2010 to 2030 period … assuming a 1.9-percent annual inflation rate,” and “[t]he total distribution costs for the 2010 to 2030 period … are $582 billion in nominal terms.” WIRES, a non-profit association concerned with transmission development, in a petition to FERC regarding ROE policy reported a $300 billion transmission investment need for the 20-year period ending 2030.  See Petition for Statement of Policy, Docket No. RM13-18-000 at p. 7 (filed June 26, 2013).

7. See Federal Reserve Press Release dated June 19, 2013 describing the policy action of the Federal Open Market Committee.

8. See, e.g., Complaint Requesting Fast Track Processing of Seminole Electric Cooperative and Florida Municipal Power Agency, Docket No. EL13-63-000 (filed May 13, 2013), renewing complaint pending in Docket No. EL12-39-000 (filed Feb. 29, 2012).

9. See, e.g., Protest and Motion to Consolidate of Arkansas Electric Cooperative, et al., Docket Nos. EC12-145-000, et al., at pp. 63-64 (filed Jan. 22, 2013).

10. See Seminole complaints, note 7, supra.

11. Complaint of Martha Coakley, Attorney General of the Commonwealth of Massachusetts, et al. v. Bangor Hydro-Electric, et al., Docket No. EL11-66-000 (filed Sept. 30, 2011).

12. Pac. Gas & Elec., 141 FERC ¶ 61,168, at P 23 (2012) (citing Pub. Serv. Co. of N.M., 137 FERC ¶ 61,119, at P 13 (2011)); Pub. Serv. Co. of N.M., 142 FERC ¶ 61,168, at p. 28 (2013).

13. Commissioner John R. Norris, Remarks at the 992nd Commission Meeting (March 21, 2013) (grouping ROEs with Order No. 1000 proceedings) (transcript available at:

14. S. Cal. Edison, 131 FERC ¶ 61,020 (2010) (“Paper Hearing Order”), reh’g denied, 137 FERC ¶ 61,016 (2011) (“Rehearing Order”) aff’d in part and remanded in part, S. Cal. Edison v. FERC, No. 11-1471, 2013 WL 1920937 (D.C. Cir. May 10, 2013) (“SoCalEd v. FERC”).

15. Minn. Power & Light, 3 FERC ¶ 61,045, at 61,133 (1978).

16. Cent. Ill. Light, Initial Decision, 6 FERC ¶ 63,016, at 65,137-138 (1979), aff’d, 10 FERC ¶ 61,248 (1980), aff’d and remanded on other grounds sub nom. Chatham v. FERC, 662 F.2d 23 (1981).

17. See SoCalEd, note 13, supra, Paper Hearing Order at P 92. FERC previously used the midpoint in Golden Spread Elec. Coop. v. Sw. Pub Serv., 123 FERC ¶ 61,047 (2008).

18. See SoCalEd v. FERC, note 13, supra.

19. Edison Electric Institute, Electric Power Industry Outlook: 2013 Wall Street Briefing, at 8 (Feb. 6, 2013).

20. Complaint of Seminole Electric Cooperative and Florida Municipal Power Agency Requesting Fast Track Processing, Docket No. EL13-63-000 (filed May 13, 2013).

21. Midwest Independent System Operator and Ameren Illinois, Prepared Direct and Answering Testimony of Trial Staff Witness Robert J. Keyton, Docket No. ER11-2777-000 (filed Feb. 17, 2012). See also Martha Coakley, et al. v. Bangor Hydro-Electric, et al., Direct and Answering Testimony of Trial Staff Witness Sabina U. Joe, Docket No. EL11-66-001 (filed Jan. 18, 2013) (proposing base ROE of 9.66 percent).

22. Pac. Gas & Elec. Co., 141 FERC at p. 23 (citing Pub. Serv. Co. of N.M., 137 FERC at p. 13: “The Commission has previously determined that, with regard to a single utility’s proposed base ROE, the use of the median – not midpoint – DCF value from the proxy group selected is the most accurate measure of central tendency for a single utility of average risk, such as PG&E.”)

23. Pub. Serv. Co. of N.M., 142 FERC at p. 28.

24. Promoting Transmission Investment through Pricing Reform, Notice of Proposed Rulemaking, 113 FERC ¶ 61,182, at p. 20 (2005). FPA § 219(b)(2) which requires FERC to “provide a return on equity that attracts new investment in transmission facilities.” 16 U.S.C. § 824s(b)(2).

25. Hope, note 2, supra, at 602 (“Under the statutory standard of ‘just and reasonable’ it is the result reached not the method employed which is controlling”).

26. See SoCalEd, note 13, supra, Paper Hearing Order at PP 114-116.

27. Transcon. Gas Pipe Line Corp., 85 FERC ¶ 61,323, at 62,269 (1998).

28. Consol. Gas Supply, Initial Decision, 8 FERC ¶ 63,012, at 65,170 (1979) (citing Minn. Power & Light Co., 3 FERC at 61,132-33); Consol. Gas Supply, 10 FERC ¶ 61,029, at 61,053 (1980); Consol. Gas Supply v. FERC, 653 F.2d 129, 134 (4th Cir. 1979).

29. See EEI report, note 1, supra, at 4-5.