Sum-of-the-Parts DCF

A conglomerate is not one business—it’s several operating at different risk levels and growth rates. Sum-of-the-Parts (SOTP) DCF values each segment independently with its own discount rate, then adds them up. This approach often reveals whether the whole is worth more or less than its pieces.

When one WACC isn’t enough

A standard DCF assumes all cash flows are discounted at a single company-wide WACC. This works fine for a focused firm: an airline’s cash flows, whatever their term structure, carry airline-level risk.

But conglomerates—firms with two or more distinct operating segments—pose a problem. A financial-services holding with a stable bank, a volatile insurance unit, and a private-equity arm shouldn’t use the same WACC for all three. The bank’s cash flows justify a 6% discount rate; the PE arm’s, 10%. Using a blended 8% understates the stable bank’s value and overstates the risky fund’s.

SOTP solves this by disaggregating. You build a separate DCF for each segment, assign it a segment-specific cost of capital, discount its cash flows, and sum the resulting equity values. The result is often a more accurate enterprise value than a single monolithic model.

The mechanics: segment-by-segment valuation

The process is straightforward but data-intensive.

First, extract (or estimate) each segment’s free cash flow. This requires reading the company’s 10-K and breaking down operating income, capital expenditures, working capital changes, and taxes by segment. Many large firms disclose segment revenue and operating profit; calculating cash flows requires some inference.

Second, calculate a segment-specific WACC. This step is critical. You can’t use the parent company’s debt-to-equity ratio for each piece; instead, estimate what each segment’s capital structure would be if it were standalone. A utility’s segment WACC might be 5.5% (low leverage, stable cash); a growth software unit’s, 8.5%.

Many analysts use beta to calibrate the difference. A cyclical segment has higher systematic risk and thus a higher cost of equity. A defensible, low-growth segment has lower beta and a lower required return.

Third, apply a terminal value assumption to each segment. Some segments (mature, low-growth) use a 2–3% perpetual growth rate; others (expansion-stage) might use 4–5%. Each segment’s terminal value is then calculated and discounted separately.

Finally, sum the present values:

Total Equity Value = PV(Segment A) + PV(Segment B) + … − Net Debt

An illustrative case

Suppose a conglomerate has two divisions:

• Division A (Insurance): generates $100M free cash flow, $50M invested capital, faces 8% WACC.
• Division B (Asset Management): generates $50M free cash flow, $20M invested capital, faces 10% WACC.

Using a blended company WACC of 9%, you’d discount all cash as if they carry the same risk—wrong.

Under SOTP:

• Division A (8% WACC, 3% terminal growth): FCF discounted at 8%, terminal value calculated at 8%. Result: $1.25B equity value.
• Division B (10% WACC, 4% terminal growth): FCF discounted at 10%, terminal value calculated at 10%. Result: $550M equity value.
• Total: $1.8B.

If you’d used a 9% WACC for both, Division A would be undervalued (too-high discount rate suppresses its stable cash) and Division B might be overvalued (too-low discount rate inflates growth prospects). The SOTP method captures the actual risk profile of each.

Estimating segment WACCs

The challenge in SOTP is assigning credible WACCs to divisions that aren’t standalone entities and don’t trade separately.

One approach: compare pure-play competitors. If the conglomerate’s insurance arm resembles large independent insurers (which have WACC of 6–7%), use that. If the asset-management segment is closest to specialized fund managers (WACC of 8–9%), use that range.

A second approach: model the leverage ratio and cost of debt each segment would carry if independent. A stable utility segment might support 40% leverage; a growth segment, 20%. Apply the parent’s credit spread (or a typical spread for that segment type) to calculate cost of debt. Then solve for cost of equity using the WACC formula.

A third approach: sensitivity-test the valuation across a reasonable WACC range for each segment. If the total value is robust even when you shift segment WACCs by ±50 basis points, your answer is credible. If total value swings wildly, the SOTP answer is fragile and should be used with caution.

When SOTP reveals a discount or premium

Comparing SOTP value to the company’s market price often reveals market mispricing.

If SOTP value is 20% higher than market cap, the market is applying a conglomerate discount—a penalty for holding disparate businesses. This can persist due to agency costs (managers of conglomerates often underinvest in high-return segments and overfund low-return ones) or because the market struggles to analyze multiple segments. A value investor might see a buy opportunity.

Conversely, if SOTP value is 10% lower than market cap, the market is pricing in future synergies, spinoff value, or management improvements you haven’t modeled. A spinoff—separating segments into standalone companies—sometimes creates value by unlocking segment-specific capital structures and investor focus.

Pitfalls and limitations

SOTP’s greatest weakness is that it requires estimates—segment cash flows, appropriate WACCs, terminal growth rates—for multiple divisions. Errors compound. A 0.5% WACC misestimate across four segments can swing total value by 5–10%.

It also ignores genuine synergies if the segments benefit from being together: shared overhead, cross-selling, consolidated bargaining power. If separating divisions would destroy economics, SOTP can overstate standalone values.

Finally, SOTP can be abused. Cherry-picking the highest plausible WACC for undervalued segments and the lowest for priced-in segments biases the total upward. A disciplined analysis uses comparable companies or market data to anchor WACC assumptions, not analyst preferences.

When SOTP is essential

SOTP is most valuable for:

• Holding companies with distinct funds, subsidiaries, or divisions (Berkshire Hathaway, 3M, Alphabet).
• Firms with announced spinoffs or divestitures, where segment values must be isolated for deal modeling.
• Analysts comparing conglomerate value to peer groups, since pure-play comparables don’t exist.
• Private equity or strategic buyers evaluating specific segments of a larger company.

It’s less critical for focused firms or ETFs holding pre-screened, homogeneous assets. But for any complex corporate structure with multiple operating environments, SOTP forces the hard questions: What is each piece worth? At what risk? And is the whole greater than the sum?

See also