The Residual Income Model
The residual income model (RIM) values a company based on its ability to generate returns that exceed the cost of its equity capital. The core insight is straightforward: if a company earns exactly its cost of equity, it creates no value beyond what investors could earn elsewhere at similar risk. Only the earnings above that hurdle rate, the residual income, represent genuine value creation. A company that consistently earns 20% on equity when its cost of equity is 10% is creating substantial value. One that earns 8% on equity with a 10% cost of equity is destroying it.
This framework was formalized by Edwards and Bell in the 1960s and later refined by James Ohlson in his influential 1995 paper. It has gained traction among academics and practitioners because it anchors valuation to the balance sheet rather than to uncertain future cash flow projections. Unlike a DCF model, where most of the value resides in a distant terminal value, the residual income model starts with the book value of equity, a known quantity, and adds only the present value of returns above the cost of capital.
The Core Formula
The residual income model expresses a stock's intrinsic value as:
V0 = B0 + Sum of [RIt / (1 + r)^t] for t = 1 to infinity
Where:
- V0 = intrinsic value of equity today
- B0 = current book value of equity per share
- RIt = residual income in period t
- r = cost of equity
Residual income in any period is:
RIt = Earningst - (r x Bt-1)
Or equivalently:
RIt = (ROEt - r) x Bt-1
Where ROEt is return on equity in period t and Bt-1 is book value of equity at the start of the period.
This formula makes the value drivers explicit. A company creates value (positive residual income) when ROE exceeds the cost of equity. It destroys value when ROE falls below that threshold. And the amount of value created depends on the spread between ROE and the cost of equity, multiplied by the equity base over which that spread is earned.
Why Residual Income Differs From DCF
At a theoretical level, the residual income model and the DCF model should produce identical valuations, because they are mathematically equivalent under the clean surplus accounting assumption (all changes in book value flow through the income statement). In practice, the two models often produce different results because of the way analysts implement them.
The key practical differences:
Terminal value sensitivity. In a DCF, the terminal value typically accounts for 60-80% of total enterprise value. In a residual income model, the book value of equity is already known, and only the excess returns need to be projected. If a company's ROE is expected to converge to its cost of equity over time (as economic theory suggests competitive forces will cause), the terminal value of residual income approaches zero. This means more of the valuation is anchored in observable, current data.
Accounting-based inputs. The residual income model uses earnings and book value, which come directly from financial statements, rather than free cash flow, which requires adjustments for capital expenditures, working capital, and depreciation. For companies with volatile capital expenditure cycles, this can produce more stable valuation estimates.
Book value as anchor. Starting with book value provides a tangible baseline. The question becomes: how much is the company worth above (or below) its book value? This framing is particularly useful for financial companies, where book value represents tangible assets.
A Worked Example
Consider a hypothetical bank with the following characteristics:
- Current book value per share: $40
- Cost of equity: 10%
- Expected ROE for years 1-5: 14%
- Expected ROE from year 6 onward: 10% (converges to cost of equity)
Residual income in year 1: RI1 = (14% - 10%) x $40 = $1.60
Assuming the bank retains all earnings (no dividends for simplicity), book value grows to $40 x 1.14 = $45.60 at the end of year 1.
Residual income in year 2: RI2 = (14% - 10%) x $45.60 = $1.82
Continue this projection through year 5. From year 6 onward, ROE equals the cost of equity, so residual income is zero and the terminal value of the residual income stream is zero.
Discounting the five years of residual income back at 10% and adding them to the current book value of $40 produces an intrinsic value above book value. The premium to book reflects the bank's ability to earn above-cost returns during the five-year period.
If the same bank's ROE were only 8%, below the cost of equity, residual income would be negative, and the model would value the stock below book value. This is consistent with what the market does in practice: banks with below-cost returns on equity, such as many European banks in the 2010s, trade at persistent discounts to book value.
Residual Income and the Price-to-Book Ratio
The residual income model provides a theoretical foundation for the price-to-book ratio. The formula can be rewritten as:
P/B = 1 + Present Value of Future (ROE - r) / r
This means:
- If ROE = cost of equity, the stock should trade at 1.0x book. No excess value is being created.
- If ROE > cost of equity, the stock should trade above book. The premium reflects value creation.
- If ROE < cost of equity, the stock should trade below book. The discount reflects value destruction.
This relationship explains why high-ROE companies like Apple (ROE consistently above 100% due to negative tangible equity from buybacks) trade at massive premiums to book, while low-ROE companies like certain regional banks trade near or below book value.
It also explains the "value trap" phenomenon. A stock trading at 0.7x book value is not necessarily cheap. If the company's ROE is persistently below its cost of equity, the discount is justified, and the stock may deserve an even larger discount.
The Fade Factor
In reality, few companies sustain above-normal returns on equity indefinitely. Competition, technological disruption, regulatory changes, and management missteps all erode competitive advantages over time. The residual income model handles this by incorporating a "fade" factor that gradually reduces ROE toward the cost of equity.
The speed of the fade depends on the durability of the company's competitive advantage:
Slow fade (10-20 years). Companies with strong moats: Visa's payment network, ASML's monopoly in EUV lithography, or Moody's dominance in credit ratings. These businesses have structural advantages that competitors cannot easily replicate.
Medium fade (5-10 years). Companies with strong but vulnerable positions: a leading pharmaceutical company with patent cliffs, a consumer brand facing private-label competition, or a technology company in a rapidly evolving market.
Fast fade (3-5 years). Companies in highly competitive industries with low barriers to entry: commodity producers, most retailers, and service businesses without significant switching costs.
The fade assumption has a large impact on valuation. A company earning 25% ROE with a 10% cost of equity is worth much more if that 25% ROE persists for 15 years than if it fades to 10% within five years.
Advantages Over Other Methods
Less reliance on terminal value. Because the model starts with book value and projects only the excess, the terminal value component is smaller and often approaches zero as ROE converges to the cost of equity. This reduces the single biggest source of error in DCF models.
Works well for financials. Banks, insurance companies, and other financial institutions are notoriously difficult to value with DCF because their cash flows are intertwined with their capital structure. The residual income model, which uses earnings and book value, is better suited to these businesses.
Integrates profitability and growth. The model explicitly captures the interaction between ROE, growth, and the cost of equity. It rewards growth only when ROE exceeds the hurdle rate. A company that grows rapidly but earns below-cost returns is worth less in this model, not more, reflecting the economic reality that growth at inadequate returns destroys value.
Clean surplus validation. The model can be cross-checked against the clean surplus relation: ending book value equals beginning book value plus earnings minus dividends. Any deviations highlight accounting distortions (comprehensive income items, share buybacks at above-book prices) that the analyst should investigate.
Limitations
Accounting quality dependency. The model relies on earnings and book value as reported under accounting standards. If accounting standards do not reflect economic reality (as with R&D expensing, operating lease treatment prior to ASC 842, or pension accounting), the model's outputs will be distorted. Adjustments to restate accounting numbers on a more economic basis are often necessary.
Negative book value. Companies that have repurchased massive amounts of stock may have negative book value (McDonald's and Philip Morris International are examples). The residual income model becomes less intuitive and mechanically problematic when the equity base is negative.
Clean surplus violations. The model assumes all changes in equity flow through the income statement. In practice, items like foreign currency translation adjustments, unrealized gains and losses on certain investments, and pension remeasurements bypass the income statement and go directly to other comprehensive income. These violations can distort the model's output.
Less intuitive for non-financial companies. While the model works theoretically for any company, analysts and investors are more accustomed to thinking in terms of cash flows and earnings multiples. The residual income framework requires a different mental model that some find less natural to apply.
Despite these limitations, the residual income model remains a valuable addition to an analyst's toolkit, particularly for financial companies and for any situation where anchoring the valuation to the balance sheet reduces uncertainty relative to projecting distant cash flows.
Put these principles into practice. Track fundamentals, build portfolios, and analyze stocks with AI-powered insights.
Start Free on GridOasis →