Limitations of the APV Method in Financial Institution Valuation: Why Banks Rarely Use APV

The HKMA’s 2025 Banking Sector Stability Report, published in March, flagged a 22-basis-point compression in net interest margins (NIMs) across Hong Kong’s retail banks for the 2024 calendar year, driven by the delayed pass-through of US Federal Reserve rate cuts to deposit costs. This margin squeeze, coupled with the HKMA’s parallel consultation on the implementation of Basel III finalisation standards for credit valuation adjustment (CVA) risk, has forced bank CFOs and corporate finance teams to revisit valuation methodologies. The adjusted present value (APV) method, a staple for leveraged buyouts and project finance, is structurally ill-suited to financial institutions. Its core assumption—that debt tax shields can be valued independently of the firm’s operating risk—collapses when applied to banks, where debt is not a financing choice but a regulatory input. This article examines the three structural incompatibilities that make APV a theoretical curiosity rather than a practical tool for bank valuation.

The Regulatory Capital Structure: Why Debt Is Not a Choice

The APV method, as codified by Stewart Myers (1974), separates firm value into two components: the value of the unlevered firm and the present value of the interest tax shield from debt. This decomposition assumes that the firm can choose its debt level arbitrarily, subject to market constraints. For financial institutions, this assumption is invalid.

Capital Adequacy as a Binding Constraint

Under the Banking (Capital) Rules (Cap. 155L), a Hong Kong-incorporated authorised institution must maintain a Common Equity Tier 1 (CET1) ratio of at least 4.5%, a Tier 1 ratio of 6%, and a total capital ratio of 8%. In practice, the HKMA sets institution-specific Pillar 2A and 2B requirements, pushing effective CET1 floors to 9-12% for most systemically important banks. The 2024 HKMA Supervisory Policy Manual (SPM) module CA-G-1 explicitly states that “capital adequacy assessment should form an integral part of the institution’s risk management framework, not a residual decision after asset selection.”

This regulatory overlay means that a bank’s debt-to-equity ratio is not a free variable. The leverage ratio—defined as Tier 1 capital divided by total exposure—is capped at 3% under Basel III. For a bank with HKD 100 billion in risk-weighted assets, the maximum allowable debt (total liabilities) is constrained by the minimum capital requirement, not by an optimisation of the tax shield. The APV method’s central premise—that increasing debt increases firm value through tax savings until the marginal cost of financial distress offsets the benefit—is therefore moot.

The Tax Shield Is Already Priced into the Cost of Capital

Standard corporate finance textbooks teach that the interest tax shield reduces the weighted average cost of capital (WACC). For a non-financial firm, the after-tax cost of debt is r_d(1-T), where r_d is the pre-tax cost of debt and T is the corporate tax rate. For a Hong Kong bank subject to the 16.5% profits tax rate, this adjustment appears straightforward.

However, the HKMA’s 2023 consultation on the Internal Ratings-Based (IRB) approach to credit risk (HKMA Circular, 15 June 2023) clarified that banks must use a “through-the-cycle” (TTC) probability of default (PD) for regulatory capital calculations, not a “point-in-time” (PIT) estimate. The TTC PD embeds a long-run average of macroeconomic conditions, effectively stripping out the cyclicality that the APV tax shield is supposed to capture. When a bank’s cost of debt is already calibrated to a TTC framework, the marginal tax shield from an additional unit of debt is zero—the market prices the debt as if the tax benefit is already reflected in the bank’s regulatory capital structure.

The Inseparability of Operating and Financing Cash Flows

APV’s second foundational assumption is that operating cash flows and financing cash flows can be valued independently. For a manufacturing firm, the cash flows from selling widgets are distinct from the cash flows from servicing debt. For a bank, the two are inseparable.

Net Interest Income as the Core Operating Revenue

A commercial bank’s primary operating revenue is net interest income (NII)—the spread between interest earned on loans and interest paid on deposits. In the 2024 annual reports of the three largest Hong Kong-incorporated banks (HSBC, Bank of China (Hong Kong), and Standard Chartered Hong Kong), NII constituted between 62% and 74% of total operating income. The cost of deposits is not a financing expense; it is a direct input into the bank’s production function.

Consider the following simplified balance sheet for a Hong Kong retail bank:

Item	HKD (billions)
Loans (earning assets)	500
Deposits (liabilities)	450
Equity	50
CET1 ratio	10%

The bank’s cost of deposits—say, 2.5% per annum—is not a financing cost in the corporate finance sense. It is the price the bank pays for its primary raw material (deposits). The APV method would treat the interest paid on deposits as a tax-deductible financing flow, separable from the loan revenue. In reality, the spread between the loan yield (say, 5.5%) and the deposit cost (2.5%) is the bank’s gross margin. If the bank were “unlevered” (i.e., funded entirely by equity), it would have no deposits and thus no loans. The unlevered firm value in APV is a hypothetical construct with no real-world analogue for a deposit-taking institution.

The Fair Value Option and Mark-to-Market Accounting

Under HKFRS 9 (effective 1 January 2018), banks must classify financial assets into three categories: amortised cost (AC), fair value through other comprehensive income (FVOCI), and fair value through profit or loss (FVTPL). For a bank holding a significant portfolio of debt securities—the HKMA reported that Hong Kong banks held HKD 2.3 trillion in debt securities as of 31 December 2024—changes in market interest rates directly affect both the balance sheet (through FVOCI reserves) and the income statement (through FVTPL).

APV values the tax shield using the contractual interest rate on debt, assuming it is fixed. For a bank, the effective cost of liabilities changes daily as market rates move and as depositors reprice their accounts. The 2024 NIM compression cited earlier was driven precisely by this repricing asymmetry: banks were slow to cut deposit rates despite falling loan yields. An APV model that uses a static cost of debt would systematically overvalue the tax shield during falling rate environments and undervalue it during rising rate environments, producing a valuation that diverges from observable market prices.

The Cost of Financial Distress Is Not a Tail Risk—It Is the Business Model

The APV framework incorporates the cost of financial distress as a convex function of leverage, typically modelled as a percentage decline in firm value when debt exceeds a threshold. For a non-financial firm, financial distress is an event—a bankruptcy filing, a debt restructuring, or a covenant breach. For a bank, financial distress is a continuum that the regulator manages daily.

The HKMA’s Supervisory Intervention Framework

The HKMA’s SPM module IC-1 (Supervisory Review Process) outlines a graduated intervention framework based on a bank’s capital position relative to its “trigger” and “target” ratios. When a bank’s CET1 ratio falls below the trigger level—typically 2 percentage points above the regulatory minimum—the HKMA can impose restrictions on dividend payments, share buybacks, and branch expansion. At the “breach” level (below the regulatory minimum), the HKMA can require the bank to submit a capital restoration plan, restrict asset growth, or, in extremis, apply to the Court of First Instance for a winding-up order under the Banking Ordinance (Cap. 155, Section 122).

This framework means that the cost of financial distress for a bank is not a one-time event cost but a recurring regulatory tax. The bank must hold capital buffers—the HKMA’s 2024 countercyclical capital buffer (CCyB) for Hong Kong was set at 1.0% of risk-weighted assets—that generate no return for equity holders. The opportunity cost of this “dead” capital is a permanent drag on return on equity (ROE), not a probabilistic distress cost.

Deposit Insurance and the Moral Hazard Adjustment

The Hong Kong Deposit Protection Scheme (DPS), established under the Deposit Protection Scheme Ordinance (Cap. 581), covers deposits up to HKD 800,000 per depositor per bank. Banks pay an annual levy to the DPS, calculated as 0.05% of total covered deposits (as of the 2024 levy schedule). This levy is a direct cost of the deposit funding that APV treats as a tax-advantaged debt.

More critically, the DPS creates a government guarantee that reduces the market discipline on deposit costs. Depositors do not demand a risk premium for uninsured deposits because they expect the government to intervene for systemically important institutions. The APV method’s assumption that the cost of debt reflects the marginal risk of default is therefore violated. The true cost of deposit funding is understated by the market, and the tax shield is overstated.

Practical Implications: What APV Misses in Bank Valuation

Given these structural incompatibilities, financial analysts and corporate finance advisors in Hong Kong should adopt alternative valuation frameworks for banks. The most common approach in practice is the dividend discount model (DDM) or the excess returns model, both of which align with the regulatory focus on distributable reserves.

The Dividend Discount Model as the Default

The HKMA requires banks to obtain prior approval for any dividend distribution that would reduce the CET1 ratio below the trigger level (SPM CA-G-5, 2023 revision). This regulatory gate means that dividends are not a discretionary payout but a function of excess capital. The DDM values a bank as the present value of expected dividends, discounted at the cost of equity estimated from the capital asset pricing model (CAPM) or the Fama-French three-factor model.

For the three largest Hong Kong banks, the average dividend payout ratio over 2020-2024 was 62%, with a dividend yield of 4.8% as of 31 December 2024. The DDM captures the regulatory constraint on dividends directly: if capital falls below the trigger, dividends are cut, and the model’s terminal value adjusts accordingly.

The Excess Returns Model for Intrinsic Value

The excess returns model—also called the residual income model (RIM)—values a bank as book value of equity plus the present value of expected excess returns (ROE minus cost of equity) multiplied by beginning book value. This model is particularly suited to banks because book value of equity is a regulatory capital metric that is audited and reported quarterly under HKFRS.

In the 2024 annual results of Bank of East Asia (BEA), the reported ROE was 8.2%, against a cost of equity estimated at 9.5% (using a beta of 0.85 and an equity risk premium of 6.0%). The excess returns model would value BEA at a discount to book value, consistent with its price-to-book ratio of 0.72x as of 31 December 2024. The APV method, by contrast, would require estimating an unlevered cost of equity for a firm that has never been unlevered, introducing unnecessary estimation error.

Actionable Takeaways for Practitioners

1. Do not use APV for any entity whose primary liabilities are deposits or whose leverage is determined by regulatory capital requirements rather than market optimisation.
2. Adopt the dividend discount model for Hong Kong-incorporated banks, and calibrate the dividend growth rate to the HKMA’s trigger capital ratio thresholds (SPM CA-G-5).
3. When valuing a bank in a transaction context (e.g., M&A or IPO), use the excess returns model with a terminal value based on a fade to the cost of equity, not a perpetual growth rate.
4. For cross-border comparisons, adjust the cost of equity for jurisdiction-specific regulatory capital buffers, such as the HKMA’s 1.0% CCyB and the 2.5% capital conservation buffer under Basel III.
5. In any valuation report submitted to the SFC under the Codes on Takeovers and Mergers (2024 edition), explicitly state the model choice and justify the exclusion of APV with reference to the Banking Ordinance (Cap. 155) capital adequacy requirements.