Skip to main content

Regulated Returns & Rate Base

Updated Jun 24, 2026 at 8:22pm

Research Draft Medium 1,254 words

A regulated utility is not a free-market business — it is a government-sanctioned monopoly whose profits are set by a regulator rather than by competition. The central machinery is the rate base (the depreciated value of capital the utility has prudently invested in "used and useful" infrastructure) and the allowed return the regulator lets it earn on that capital. Because earnings flow from invested capital rather than from sales volume, the utility's growth thesis is fundamentally how fast it can grow rate base while continuing to earn close to its allowed return. The core tension is the regulatory bargain: the utility gets a near-guaranteed return in exchange for capped profits and an obligation to serve — which simultaneously makes the stock bond-like and creates a structural incentive to over-invest.

How it's calculated / formed

The regulator sets a utility's total allowed revenue through the cost-of-service / revenue-requirement formula (verified across Energy IB and Enerdynamics):

Revenue Requirement = O&M Expenses + Depreciation + Taxes + (Rate Base × Allowed Rate of Return)

  • Rate base = original cost of plant − accumulated depreciation + approved adjustments (working capital, certain regulatory assets, and — where allowed — Construction Work in Progress, "CWIP"). It is the equity-plus-debt capital the utility has sunk into poles, wires, pipes, substations, generation, and meters that are deemed used and useful.
  • Allowed rate of return is the weighted average cost of capital (WACC): WACC = (Equity % × Allowed ROE) + (Debt % × Cost of Debt) A typical capital structure is roughly 50% equity / 50% debt. Because the allowed ROE on the equity slice (commonly ~9.5–10%) exceeds the cost of debt (often ~4–5%), the headline allowed ROE is always higher than the blended WACC.

Regulators set the allowed ROE using a blend of financial models — primarily Discounted Cash Flow (dividend yield + growth), the Capital Asset Pricing Model, comparable-earnings benchmarking against peers, and a bond-yield-plus-risk-premium approach (Enerdynamics; Energy IB). The chosen ROE is a contested, testimony-driven number, not a market output.

The number is reset in a rate case: the utility files testimony built around a test year (a representative 12-month cost picture), intervenors and commission staff challenge it through discovery and hearings, an administrative law judge issues a recommended decision, and the commission votes a final order. The process commonly runs 6–18 months (Energy IB).

How it's used in practice

For an equity analyst, the regulated-utility model collapses to a few drivers:

1. Rate base × allowed ROE ≈ the equity earnings power. The single most-watched variable for forward EPS is the rate-base growth rate, because each dollar of net new capital lifts allowed earnings proportionally. Utility EPS growth (commonly cited at ~5–8% for large IOUs) is engineered from rate-base CAGRs of roughly 7–10% partially offset by dilution and lag. 2. Capex is the fuel, not a cost to minimize. Spending that gets approved into rate base creates future earnings. U.S. investor-owned energy-utility capex has surged — S&P Global Market Intelligence (RRA) reported ~$173B in 2024 and projected ~$215B for 2025 (rising toward ~$228B in 2026), driven by grid hardening, electrification, and data-center load growth. 3. Jurisdiction quality is decisive. "Constructive" states (commonly cited examples: Florida, Texas, Virginia, Indiana, Wisconsin) offer forward test years, rider mechanisms, CWIP in rate base, and timely orders; "challenging" ones (parts of the Northeast, California) use historical test years and tougher consumer advocacy. The same balance sheet earns more in a constructive state. 4. Earned vs. allowed ROE is the report card. A utility rarely earns exactly its allowed ROE; the gap reveals execution and regulatory lag.

Adoption, debate & evidence

Cost-of-service rate-of-return regulation is the dominant framework for U.S. investor-owned electric, gas, and water utilities — it is not fringe. The empirically grounded facts are the rate-case numbers, which are tracked authoritatively by S&P Global Market Intelligence's Regulatory Research Associates (RRA). Per RRA-sourced data, the median authorized electric ROE was ~9.70% in full-year 2024 and again in 1H 2025 (the 2023 full-year median was ~9.50%, average ~9.60%; gas runs slightly higher). These figures are well-documented, not folklore — though note that median, average, and average-excluding-rider-cases differ by ~10–20 bps, so headline "~9.7%" comparisons should specify the measure.

What is genuinely contested:

  • Are allowed ROEs too high? Consumer advocates and groups like RMI argue authorized ROEs have stayed sticky near ~9.5–10% even as interest rates fell for much of the past decade, transferring excess profit to shareholders and inflating bills. The Energy & Policy Institute estimates investor-owned electric utility profit averaged ~12.8% of customer bills nationally (2021–2024), rising to ~14.6% in 2025, with some utilities (e.g. PPL, PSE&G) near ~18–20%. This is a values-and-evidence dispute, not a settled fact.
  • The Averch–Johnson effect. A classic 1962 economic result predicts that when the allowed return exceeds the true cost of capital, firms have an incentive to over-invest in capital ("gold-plating") to inflate the base on which they earn. The theory is widely taught; the empirical support is mixed — some studies find evidence of input inefficiency or capital bias, others fail to confirm systematic overcapitalization (per surveys of the AJ literature, e.g. Wikipedia's review and academic working papers). Treat AJ as a credible incentive concern, not a measured law.
  • Performance-based / multi-year alternatives are gaining ground precisely to blunt both regulatory lag and the AJ incentive, but cost-of-service remains the backbone.

Strengths & limitations

When it works: the model delivers low-volatility, bond-proxy earnings with high visibility, which is why utilities are classic defensive, income holdings. In a heavy-capex super-cycle with constructive regulators, rate-base growth compounds into reliable mid-single-digit EPS and dividend growth.

When it fails:

  • Regulatory lag. With historical test years, the utility spends capital today but only earns on it after the next case clears — so it perpetually under-earns its allowed ROE while "playing catch-up" (S&P Global; Energy IB). Rising rates and heavy capex widen the gap.
  • Disallowances. Capital deemed imprudent or not used and useful is excluded from rate base — a real, sometimes large, write-off risk.
  • Affordability backlash. Big bills can flip a constructive commission hostile.

The #1 misuse: treating the allowed ROE as the earned return. A utility advertising a 10% allowed ROE may chronically earn 8–9% after lag, financing, and disallowances. Always separate authorized from achieved.

Sources

Disputes flagged: (1) Whether authorized ROEs are excessive is a live, ideologically charged debate — RRA reports the numbers; advocacy groups contest their fairness. (2) The Averch–Johnson over-investment effect is theoretically standard but empirically mixed (Courville/Spann found input bias; Boyes/Baron-Taggart found none) — treat as an incentive concern, not a measured law. (3) Specific authorized-ROE figures (~9.70% median, 2024–1H 2025) and capex figures are RRA/S&P-sourced, but median vs. average vs. average-ex-rider differ by ~10–20 bps. The profit-share-of-bill range is the Energy & Policy Institute's estimate (~13–15% national average, ~20% at the high end), not a regulatory constant. Rate-base-multiple and EPS-growth ranges are commonly-cited industry rules of thumb, not precise measured constants.