The Allowed Return on Equity

The allowed return on equity is the single most heavily litigated figure in a rate case and the one with the least determinate answer. Every other component of the revenue requirement is measured against something observable: a booked expense, an invoice, a plant account, a tax rate. The cost of equity is not observable. It is an expectation held by investors, estimated indirectly, and the estimate is contested by witnesses whose methods disagree with each other.

The two benchmarks the courts supply

Two decisions frame the question. The earlier holds that a public utility is entitled to rates permitting it to earn a return on the value of the property it employs for the convenience of the public equal to that generally being made at the same time and in the same general part of the country on investments in other business undertakings attended by corresponding risks and uncertainties. The later restates the point and adds that the return to the equity owner should be commensurate with returns on investments in other enterprises having corresponding risks.

The same decision supplies the standard of review. Under the statutory formula of just and reasonable, it is the result reached and not the method employed that is controlling. That sentence has two consequences. A commission is free to weigh methods as it sees fit, and a party challenging an allowed return cannot succeed by showing that a particular model was applied incorrectly if the overall result remains within the zone of reasonableness.

The benchmarks are also comparative rather than absolute. They do not say what number is right. They say the utility should receive what investors could obtain elsewhere for comparable risk, and that the resulting rates should be sufficient to assure confidence in the financial integrity of the enterprise so as to maintain its credit and attract capital.

Assembling the comparable group

Because comparison is the standard, the first analytic step is choosing the companies to compare against. A witness assembles a proxy group of publicly traded companies whose regulated operations dominate their business and whose risk profile resembles the subject utility. Screening criteria typically include the share of revenue from regulated operations, credit rating, dividend history, absence of a pending merger, and coverage by equity analysts sufficient to supply growth estimates.

Group composition is contested because it moves the answer. A group weighted toward companies with substantial unregulated generation produces a higher estimate than one confined to distribution utilities. A group containing companies in jurisdictions with unusual regulatory mechanisms imports the effect of those mechanisms. Each side proposes a group, and the commission either selects one or works from the overlap.

The comparison is never exact, and both sides acknowledge as much. What a commission is actually deciding is whether the subject utility is riskier or safer than the assembled group, and by how much. Evidence on that question includes the regulatory mechanisms in place, the concentration of the customer base, the size and stage of the capital program, and the credit ratings assigned by agencies that assess the same factors independently.

The models and what each assumes

The discounted cash flow model is the traditional workhorse. It infers the return investors require from the current dividend yield plus an expected growth rate, on the assumption that price equals the present value of expected future dividends. Its sensitivity lies entirely in the growth input, which may be drawn from analyst forecasts, from historical growth, or from a sustainable growth calculation, and the choice among those sources can move the result by a full percentage point.

The capital asset pricing model estimates the return as a risk-free rate plus the company's systematic risk multiplied by a market risk premium. Its inputs are more transparent and its assumptions stronger: it presumes that systematic risk alone is priced and that the measured relationship to the market is stable. Utility betas are typically below one, so the model tends to produce lower estimates in periods of low interest rates.

Risk premium approaches estimate the return as the yield on utility bonds plus a premium derived from the historical spread between authorized returns and bond yields. The method is simple and partly circular, since past authorized returns were themselves set by commissions rather than by the market. It nonetheless serves as a reasonableness check on the other two.

Because each model rests on assumptions the others do not share, no witness relies on one alone. The convention is to present two or three, note where they converge, and argue that the convergence identifies the answer. Where they diverge sharply, as they do when interest rates move quickly, the divergence itself becomes the issue, and the commission has to decide which model's assumptions are least strained by current conditions.

Choosing a point within the range

Each model produces a range, and the ranges overlap imperfectly. A commission typically identifies a zone of reasonableness spanning the credible results and then selects a point within it, explaining the selection by reference to factors the models do not capture: the utility's relative risk within the proxy group, the presence of revenue stabilization or cost tracking mechanisms that shift risk to customers, the utility's capital expenditure program, and the effect on customers of the resulting rates.

The capital structure question travels with the return. The overall rate of return is a weighted average, so the equity ratio determines how much of the rate base earns the higher equity return rather than the lower embedded cost of debt. A utility with an unusually high equity ratio can produce a large revenue requirement even with a modest allowed return, which is why intervenors often contest the structure rather than the percentage.

The resulting figure feeds directly into the calculation described in the revenue requirement and how it is built, applied to the balances established under what enters the rate base. Mechanisms that shift specific costs outside the general case, described in riders and trackers outside a rate case, reduce the risk the return is meant to compensate, which is why their existence is itself an argument in the cost of capital phase.