Residual Income Model for Cyclical Firms

The residual income model for cyclical firms demands a different anchor than stable-earnings businesses. Instead of next year’s earnings estimate, you anchor the model on normalized mid-cycle earnings—the earnings a cyclical company generates when neither booming nor busting. A residual income model using peak-year earnings will overstate intrinsic value; one using trough-year earnings will understate it. For steel mills, automakers, and construction firms, getting this right is the difference between sound valuation and expensive guessing.

What the Residual Income Model Does

The residual income model values a firm as:

Intrinsic Value = Book Value of Equity + Present Value of Future Residual Income

Residual income in a given year is:

Residual Income = (ROE × Book Value) − (Cost of Equity × Book Value)

Or equivalently:

Residual Income = Net Income − (Cost of Equity × Beginning Book Value)

For a stable-earnings firm—think a utility or staple consumer brand—you forecast Year 1 net income, apply it to current book value to compute Year 1 residual income, then project that forward (assuming steady ROE and cost of equity). For cyclical firms, that approach fails because Year 1 net income is not representative of steady-state earning power.

The Cyclicality Problem

A cyclical firm’s earnings swing sharply with the economy. A steelmaker earning $500 million in an upturn and $50 million in a downturn has average (normalized) earning power of perhaps $275 million—but if you value it using Year 1’s $500 million (the peak), you’ll calculate residual income as:

($500M ÷ Book Value × ROE) − (Cost of Equity × Book Value)

This inflates the value. Conversely, if Year 1 is a trough year at $50 million, the model undervalues the firm.

The problem is that residual income depends on the numerator—net income—directly. Pump in unsustainably high earnings, and the model spits out an unsustainably high value. Cyclical firms force the analyst to step back and ask: what earnings are repeatable and durable?

Estimating Normalized Mid-Cycle Earnings

Identifying the mid-cycle is part art, part rigor. Here are the standard approaches:

Historical Average — Calculate the average net income over a full economic cycle (typically 7–10 years). A steelmaker’s earnings from 2008–2018 span bust, recovery, and peak; the average is a reasonable mid-cycle estimate.

Adjusting for Structural Change — If the firm has grown, shrunk, or changed its business mix, the historical average may not reflect current earning power. A bank that divested a unit in 2015 shouldn’t average in pre-2015 earnings without adjustment.

Peer Comparison — Compare the firm’s return on equity and net profit margin during mid-cycle years to peers. If the firm’s ROE in 2016 (a moderate-cycle year) was 12% and its peers’ was 11%, then 12% is a reasonable forward assumption for ROE at normalized earnings.

Management Guidance and Long-Term Plans — Public companies often guide on long-term earnings power (“we expect 10% ROE over the cycle”). This guidance, while sometimes promotional, can anchor normalized assumptions if industry-checked.

Three-Year Average — A practical shortcut: use a trailing three-year average of earnings per share, which dampens the impact of a single boom or bust year.

Practical Example: An Automotive Supplier

Suppose an auto-parts supplier has Year 1 net income of $100 million (up-cycle year), a book value of equity of $800 million, and your estimate of cost of equity is 9%.

If you value using Year 1’s $100 million:

Residual Income (Year 1) = $100M − (9% × $800M) = $100M − $72M = $28M

Discount this forward assuming it persists, and the value will be inflated.

If you use normalized earnings of $70 million:

Residual Income (Year 1) = $70M − (9% × $800M) = $70M − $72M = −$2M

The firm destroys a small amount of value at normalized earnings—a warning that it needs to either earn more or require less capital. This signals a value closer to book value than the peak-earnings version suggested.

Cycling Through the Model

With normalized earnings anchoring the model, you then:

1. Project normalized earnings forward, adjusting for any structural growth (e.g., 2% real growth plus inflation).
2. Apply normalized ROE and cost of equity to compute residual income year by year.
3. Assume residual income converges to zero in a terminal period (the firm’s competitive advantage erodes), or apply a steady-state assumption (residual income stabilizes at a low rate).
4. Discount all future residual income at the cost of equity.
5. Add book value of equity on the valuation date.

The result is a “normalized intrinsic value” that is independent of the current phase of the business cycle.

Why This Matters for Credit and Equity Analysis

Credit Analysis — A bank or bondholder wants to know if the firm can service debt even in a down-cycle. Using peak-cycle earnings overestimates debt capacity.

Equity Valuation — An investor buying into a cyclical stock should anchor on normalized multiples and normalized returns, not the exciting peak-year numbers that analysts trumpet during upturns.

M&A Pricing — When a buyer acquires a cyclical firm, the purchase price is often structured around normalized EBITDA, not the current year’s figure.

See also