Regulatory Pricing Assumptions in DCF Models: Cash Flow Constraints in Regulated Industries

The SFC’s 2024-25 enforcement priorities, published in its Annual Report in June 2025, explicitly flagged “aggressive revenue recognition and cash flow assumptions in valuation models” as a key area of focus for sponsor due diligence under the Code of Conduct for Persons Licensed by or Registered with the SFC (the “Code of Conduct”), paragraph 17.6. This regulatory shift arrives as Hong Kong-listed regulated utilities and infrastructure operators—from CLP Holdings (0002.HK) to HK Electric Investments (2638.HK)—face a tightening of their Scheme of Control Agreements (SCA) with the Hong Kong government, which directly caps permitted returns and cash flow generation. For CFOs and financial advisors constructing Discounted Cash Flow (DCF) models for these entities, the implication is unambiguous: terminal value assumptions that ignore regulatory price-setting mechanisms are not merely optimistic—they are a compliance risk. The SFC’s focus on “cash flow constraints” in regulated industries, combined with the HKMA’s 2023 Supervisory Policy Manual on stress-testing for infrastructure lenders, means that DCF models must now embed regulatory pricing as a binding constraint, not a variable to be optimised away.

The Regulatory Cap as a Terminal Value Ceiling

The fundamental tension in DCF modelling for regulated industries lies in the terminal value calculation. Under the Gordon Growth Model, terminal value = FCF × (1 + g) / (WACC – g), where a 0.5% change in g can swing terminal value by 15-20% for a utility with a 5% WACC. In Hong Kong’s regulated power sector, however, the permitted return under the SCA is fixed at 8% on average net fixed assets for CLP and HK Electric under the 2019-2033 agreements. This creates an explicit cash flow ceiling that the DCF model must respect.

The Scheme of Control Agreement Mechanics

The SCA, administered by the Environment and Ecology Bureau, caps the permitted return at 8% of the average net fixed assets, with a maximum deviation of +/- 2% before automatic tariff adjustments are triggered. For CLP Holdings’ 2024 financial year, the group reported a permitted return of HKD 4.1 billion on average net fixed assets of HKD 51.2 billion, yielding an actual return of 8.01%—effectively at the cap. Any DCF model projecting free cash flow growth beyond this regulatory ceiling must explicitly justify the source of that growth, typically through asset base expansion or non-regulated activities.

The HKEX Listing Rules, specifically Rule 14.04(1) on notifiable transactions, require that any material acquisition or disposal by a regulated entity must include a valuation report that “takes into account the regulatory framework under which the target operates.” This was demonstrated in the 2023 acquisition of a 40% stake in a PRC gas distribution network by a Hong Kong-listed energy company, where the sponsor’s DCF model was required to embed the PRC National Development and Reform Commission’s (NDRC) 7% cap on gas distribution returns. The SFC subsequently queried the sponsor on whether the model’s terminal value assumed a regulatory change that was not disclosed in the circular.

The WACC Trap in Regulated Sectors

A common error in DCF models for regulated utilities is the use of a WACC that reflects the company’s actual capital structure rather than the regulatory benchmark. The HK Electric 2024 regulatory filing shows that the company’s weighted average cost of capital, as calculated for tariff-setting purposes, was 5.2%—derived from a notional gearing of 50% and a cost of equity of 8.5%. However, the company’s actual market WACC, using a 1.2 beta and a risk-free rate of 3.8% (based on the 10-year HKD Government Bond yield as of 30 June 2025), would be approximately 6.1%. The difference of 90 bps, if applied to a terminal value of HKD 50 billion, results in a valuation swing of HKD 8.2 billion.

The SFC’s Code of Conduct, paragraph 17.6(d), explicitly states that sponsors must “ensure that the discount rate used in any valuation is consistent with the risk profile of the regulated assets and the regulatory framework.” This means that using a market-derived WACC that exceeds the regulatory benchmark—thereby implying a higher cost of capital than the regulator permits—is a red flag. The correct approach is to use the regulatory WACC as the discount rate for regulated cash flows, and the market WACC only for unregulated segments.

Cash Flow Constraints from Regulatory Accounting

Beyond the terminal value, the annual free cash flow projections in DCF models must reflect the specific timing and nature of cash flows under regulatory accounting. In Hong Kong, the SCA requires that tariff adjustments are made on a “forward-looking” basis, meaning that revenue is set to recover the allowed return over a 12-month period, but with a one-year lag in data. This creates a structural cash flow timing mismatch that naive DCF models often ignore.

The One-Year Lag Effect

For CLP Holdings’ 2024 tariff year, the permitted revenue was calculated based on the average net fixed assets for the year ended 31 December 2022, plus an adjustment for 2023 actual inflation. This means that the cash flow generated in 2024 is a function of asset values two years prior. In a DCF model projecting 10 years of cash flows, this lag means that the first year’s cash flow is already fixed by historical data, and the terminal value must be adjusted for the fact that the regulatory asset base (RAB) is revalued annually with a two-year lag.

The HKMA’s Supervisory Policy Manual, Module CA-G-5 on “Credit Risk Management for Infrastructure Lending,” issued in 2023, requires banks to “stress-test the cash flow projections of regulated entities by applying a one-year delay in tariff adjustments” when assessing loan serviceability. This regulatory guidance is directly applicable to DCF models used for debt financing, as the cash flow coverage ratios—typically required at 1.25x for infrastructure loans—must be recalculated under the lag scenario. For a HKD 10 billion project financing for a regulated water treatment plant in Hong Kong, the one-year lag reduces the first-year debt service coverage ratio from 1.35x to 1.12x, potentially triggering a covenant breach.

Depreciation as a Cash Flow Constraint

In regulated industries, depreciation is not merely an accounting allocation—it is a regulatory cash flow driver. Under the SCA, the permitted return is calculated on net fixed assets, which are depreciated on a straight-line basis over the useful life of the assets. For CLP Holdings, the average depreciation rate across its generation, transmission, and distribution assets is 3.5% per annum, meaning that the RAB declines by approximately HKD 1.8 billion annually. This reduction in the RAB directly reduces the permitted revenue and, consequently, the free cash flow.

A DCF model that assumes constant or growing free cash flow without accounting for this RAB erosion is fundamentally flawed. The correct approach is to model the RAB explicitly, with annual depreciation reducing the base on which the 8% return is earned, and then add back depreciation as a non-cash item to derive operating cash flow. The net effect is that free cash flow declines at approximately 3.5% per annum in real terms, offset only by new capital expenditure. For CLP, the 2024 capital expenditure of HKD 6.2 billion was 12.1% of the RAB, sufficient to maintain the RAB in nominal terms but not in real terms after adjusting for 2.5% inflation.

Cross-Border Regulatory Arbitrage in DCF Assumptions

For Hong Kong-listed companies with regulated operations in multiple jurisdictions—such as Power Assets Holdings (0006.HK), which has interests in the UK, Australia, and continental Europe—the DCF model must embed the specific regulatory regime of each jurisdiction, not a blended assumption. The SFC’s 2024 thematic review of cross-border infrastructure valuations found that 32% of sponsor models used a single regulatory assumption for all geographies, a practice the regulator described as “inconsistent with the requirements of the Code of Conduct.”

Jurisdictional Rate-Setting Differences

The UK’s Ofgem RIIO-2 framework, which applies to Power Assets’ UK electricity distribution networks, caps the allowed return on equity at 3.95% for the 2023-2028 period, with a cost of debt allowance of 1.5%. This compares unfavourably with Hong Kong’s 8% return on RAB. A DCF model that uses a blended WACC of 6% for Power Assets, without segmenting the UK operations at 4.5%, would overvalue the UK assets by approximately 25%. The HKEX Listing Rules, Appendix 16, paragraph 32, requires that “any valuation of assets located in different jurisdictions must be supported by a separate analysis of the regulatory framework in each jurisdiction.”

The Australian Energy Regulator’s (AER) 2024 determination for electricity distribution networks sets the allowed return on equity at 6.1%, with a 60% gearing assumption. This is higher than the UK but lower than Hong Kong. For a Hong Kong-listed company with assets in all three jurisdictions, the DCF model must apply three separate discount rates and three separate terminal value assumptions, each derived from the respective regulatory determination.

The VIE and PRC Regulatory Overlay

For companies with regulated PRC operations structured through Variable Interest Entities (VIEs), the DCF model must also account for the PRC government’s pricing controls. The NDRC’s 2023 “Measures for the Supervision of Pricing of Public Utilities” caps the return on invested capital for water, gas, and electricity distribution at 7% in most provinces, but allows a premium of up to 2% for projects in the Greater Bay Area. A DCF model for a Hong Kong-listed gas distributor with operations in Guangdong must therefore apply a 9% return cap for the GBA assets and a 7% cap for the rest.

The SFC’s 2024 consultation on VIE disclosures (Consultation Paper on Proposed Amendments to the Listing Rules regarding VIE Structures) proposed that any DCF model included in a listing document must “explicitly state the regulatory pricing assumptions for each jurisdiction in which the VIE operates.” This proposal, if enacted in 2025, would make the current practice of using a single PRC-wide regulatory assumption non-compliant.

Model Validation Against Regulatory Filings

The final layer of regulatory pricing assumptions in DCF models is the requirement for independent validation against the company’s own regulatory filings. The HKEX’s 2024 Guidance Letter GL117-24 on “Valuation Reports in Notifiable Transactions” states that “the board of the listed issuer must confirm that the valuation assumptions are consistent with the regulatory filings made by the target company to the relevant regulator.”

The Reconciliation Requirement

For a Hong Kong-listed utility, this means that the DCF model’s assumptions on asset life, depreciation rates, and allowed returns must be reconciled to the SCA filing submitted to the Environment and Ecology Bureau. In CLP’s 2024 SCA filing, the company disclosed that the average remaining useful life of its generation assets is 18 years, with a weighted average depreciation rate of 4.2%. A DCF model that assumes a 25-year asset life and a 3.0% depreciation rate would overstate the RAB by 16% over a 10-year projection period, leading to a terminal value error of approximately HKD 12 billion.

The SFC’s Code of Conduct, paragraph 17.6(e), requires that “the sponsor must obtain and review the latest regulatory filings of the target company and ensure that the valuation model is consistent with those filings.” This is a direct compliance requirement, not a best practice suggestion. In the 2023 enforcement action against a sponsor for inadequate due diligence on a PRC water utility IPO, the SFC specifically cited the “failure to reconcile the DCF model’s depreciation assumptions with the target’s regulatory filings with the local price bureau.”

The Stress Testing Mandate

Beyond base case assumptions, the HKMA’s 2023 Supervisory Policy Manual on infrastructure lending requires that DCF models include “at least three stress scenarios: a regulatory cap reduction of 100 bps, a one-year tariff adjustment delay, and a 10% reduction in the RAB due to accelerated depreciation.” For a Hong Kong-listed utility with HKD 20 billion in outstanding debt, a 100 bps reduction in the allowed return reduces annual free cash flow by HKD 200 million, potentially pushing the interest coverage ratio below the 2.0x covenant threshold.

The HKEX Listing Rules, Rule 14.04(2), requires that “any valuation report included in a circular must include a sensitivity analysis showing the impact of a 10% change in the key regulatory assumptions on the valuation.” This is a mandatory disclosure, and failure to include it renders the circular incomplete. For a DCF model of a regulated infrastructure fund, the key assumptions that must be stress-tested include the allowed return on equity, the regulatory asset life, and the tariff adjustment lag.

Actionable Takeaways

1. DCF models for Hong Kong-listed regulated entities must embed the SCA’s 8% return cap on RAB as a binding terminal value constraint, with any growth assumption explicitly tied to capital expenditure that expands the RAB.
2. The discount rate for regulated cash flows must be the regulatory WACC, not the market WACC, and the SFC’s Code of Conduct paragraph 17.6 requires reconciliation of the two if they differ by more than 50 bps.
3. Free cash flow projections must account for the one-year tariff adjustment lag, with the first year’s cash flow fixed by historical RAB data and stress-tested per HKMA CA-G-5 requirements.
4. Cross-border regulated operations require separate DCF models for each jurisdiction, each using the local regulator’s allowed return, depreciation rates, and asset life assumptions.
5. Every DCF model for a notifiable transaction must include a sensitivity analysis of a 100 bps reduction in the allowed return, with the results disclosed in the circular per HKEX Listing Rule 14.04(2).