How to Value a Conglomerate
Conglomerates—companies that operate multiple unrelated or loosely related businesses—pose unique valuation challenges. A diversified holding company cannot be fairly valued using a single multiple or set of assumptions that works across all divisions. Instead, valuation requires a disciplined, segment-by-segment approach, then aggregation at the enterprise level. This chapter walks through the mechanics of applying sum-of-the-parts thinking to real conglomerates.
Conglomerates succeed or fail based on management's ability to allocate capital across diverse businesses. Valuation must reflect both the economics of each unit and the quality of capital allocation at the parent level.
Key Takeaways
- Conglomerates typically warrant a lower valuation than the sum of their parts would suggest (a discount), reflecting agency costs and capital allocation inefficiencies
- The SOTP approach for conglomerates requires valuing each operating segment independently, then aggregating and adjusting for corporate costs and minority interests
- Some conglomerates trade at a premium when the parent company's capital allocation, synergies, or financial engineering outweigh diversification costs
- Tax considerations, transaction costs, and regulatory constraints affect conglomerate valuation and justify some holding company structures
- Comparing SOTP value to market cap reveals whether the market is discounting the business or pricing in integration risk
- Real-world conglomerate valuations often require scenario analysis, sensitivity testing, and explicit assumptions about management quality
The Conglomerate Valuation Framework
A systematic approach to valuing conglomerates follows this sequence:
Step 1: Segment Definition and Isolation Clearly delineate each operating segment's revenues, operating profit, and capital base. Use segment reporting from the 10-K as described in the previous chapter.
Step 2: Select a Valuation Method for Each Segment For each operating segment, choose the most appropriate method:
- Comparable company multiples (EV/EBITDA, P/E, EV/Revenue, Price/Book)
- Discounted cash flow (DCF) models built from segment-specific assumptions
- Asset-based valuation for real estate, utilities, or insurance companies
- Specialized methods (see below for insurance, real estate, etc.)
Step 3: Identify Comparable Companies for Each Segment For a machinery division, find standalone machinery manufacturers. For a software division, find SaaS comparables. For insurance, find insurers with similar underwriting profiles. Use databases like Bloomberg, FactSet, or S&P Capital IQ to gather multiples.
Step 4: Apply Multiples or Build DCF Models
- Multiples approach: Multiply segment EBITDA (or other metric) by the selected multiple.
- DCF approach: Project segment-specific revenues, margins, CapEx, and discount cash flows at segment-appropriate WACC.
Step 5: Allocate Corporate Overhead and Taxes Corporate general & administrative costs, interest on central financing, and taxes must be allocated fairly to segments. Details are in the Allocating Shared Costs chapter.
Step 6: Aggregate and Adjust Sum segment values, subtract corporate overhead, adjust for minority interests in subsidiaries, and subtract net debt to arrive at equity value.
Step 7: Sensitivity Analysis Vary key assumptions (multiples, growth rates, WACC) to create a range of values, not a point estimate.
Selecting Valuation Methods by Segment Type
Not all segments are valued the same way. Here's a practical guide:
Mature, Stable Cash Generators (e.g., Utilities, Infrastructure, Core Manufacturing) Use EV/EBITDA multiples or low-growth DCF models. These businesses have predictable cash flows and modest growth. Comparable multiples work well because the market prices them consistently.
High-Growth Segments (e.g., SaaS, Emerging Markets) Use EV/Revenue, EV/Sales, or subscription multiples if available. Alternatively, build DCF models with higher growth assumptions. High-growth segments may not have normalized earnings; forward-looking measures work better.
Cyclical Businesses (e.g., Industrial Manufacturing, Chemicals) Use normalized (or through-cycle) EBITDA multiples to smooth cyclical swings. Building a DCF with conservative long-term assumptions (lower growth, normalized margins) is more reliable than using current-year multiples at a cyclical peak or trough.
Financial Services (e.g., Insurance, Asset Management) Insurance companies typically use multiples like price-to-book, price-to-earnings, or a blended approach. Asset managers use EV/AUM (assets under management) multiples or price-to-earnings. Profitability is volatile; use normalized earnings or book value as the base.
Real Estate or Tangible Asset Businesses Use NAV (net asset value) approaches where you estimate the fair value of underlying real estate, equipment, or other assets, then value operating earnings above that base. REIT accounting (depreciation adjustments) complicates earnings multiples; use P/FFO (funds from operations) instead of P/E.
Early-Stage, Loss-Making Segments These are hardest to value. Use venture capital methods (scenario analysis with probability-weighted outcomes) or DCF with explicit assumptions about when/if the business reaches profitability. Multiples don't apply to unprofitable units.
Real Conglomerate Valuation: Berkshire Hathaway
Berkshire Hathaway is perhaps the most-analyzed conglomerate for SOTP. Here's a simplified framework:
Insurance Operations
- Combined ratio: Measure underwriting profitability (loss ratio + expense ratio)
- Float: Berkshire's insurance subsidiaries float billions of dollars between premium collection and payout; this acts like cheap financing for the parent
- Valuation: Price-to-book (P/B) on book value of float and underwriting reserves, or P/E on underwriting earnings
- Estimated Value: $150–200B (varies with claim experience and market conditions)
Berkshire Hathaway Energy (BHE, ~90% owned)
- Utility with diversified energy assets (renewables, natural gas, transmission)
- Stable, regulated utility with 6–7% growth, 12–15% ROE
- Valuation: 12–15x normalized earnings, or DCF with 3–4% perpetual growth
- Estimated Value: $200–250B
Manufacturing, Service, and Retail (CORT, Markel, GEICO, Precision Castparts, etc.)
- Mix of wholly-owned subsidiaries and partially-owned affiliates
- Each requires individual analysis; Berkshire values them at book value plus an earnings power value calculation
- Estimated Value: $150–200B
Investments and Marketable Securities
- Berkshire holds significant stakes in Apple, Bank of America, Chevron, and others
- Valued at market prices; disclosed in 10-K
- Estimated Value: $300–400B (varies with market)
Cash and Short-Term Investments
- Valued at book value
- Estimated Value: $150–200B
Berkshire's SOTP Range (Illustrative)
\text{Enterprise Value} = 150–200B + 200–250B + 150–200B + 300–400B + 150–200B
\text{= Approx. } 950B–1,250B
Subtract Berkshire's net debt (typically minimal or net cash), and you have an equity value range. Berkshire's market cap at writing is roughly $900B–$1T, suggesting the market prices it near intrinsic value, occasionally at a small premium (due to capital allocation skill) or small discount (due to succession uncertainty).
Valuing General Electric (Case Study)
GE's conglomerate structure historically made it difficult to value. Under Jeffrey Immelt, GE spanned Power Systems, Renewable Energy, Oil & Gas, Aviation, Healthcare, Financial Services, Transportation, and Lighting. Investors struggled to understand consolidated earnings because each division had different growth rates, margins, and risk profiles.
A SOTP analysis circa 2015 might have looked like:
| Segment | Revenue | Op. Margin | EV Multiple | Implied Value |
|---|---|---|---|---|
| Power | $30B | 15% | 10x | $45B |
| Renewable | $8B | 12% | 12x | $12B |
| Aviation | $25B | 18% | 14x | $63B |
| Healthcare | $18B | 22% | 15x | $59B |
| Financial | $45B | 5% | 8x | $18B |
| Industrial (Other) | $25B | 10% | 9x | $23B |
| Sum | $151B | - | - | $220B |
| Less: Corporate Overhead | - | - | - | (10B) |
| Enterprise Value | - | - | - | $210B |
At that time, GE's market cap was roughly $200B, suggesting fair valuation or a small discount. However, subsequent SOTP analyses showed deteriorating margins in Power and Financial Services, higher integration risk, and lower expected synergies—justifying a lower valuation. As GE subsequently divested healthcare, power, and other units, the remaining industrial business re-rated upward on better clarity.
Incorporating Synergies and Dissynergies
A critical and often-ignored adjustment to SOTP is whether combining the segments creates or destroys value.
Synergies (Positive Adjustment to SOTP):
- Shared R&D (e.g., a conglomerate's materials science lab serves multiple divisions)
- Cross-selling (a commercial bank's investment banking team serves clients across divisions)
- Procurement scale (combined purchasing power for commodities, logistics)
- Financial engineering (cheap internal financing from the parent)
Berkshire is famous for creating synergies: its insurance float finances acquisition at below-market rates; its reputation attracts high-quality acquisition targets; its capital allocation skill drives value creation.
Dissynergies (Negative Adjustment to SOTP):
- Distraction (management time split across unrelated businesses)
- Agency costs (conflicts of interest between divisions; parent extracts transfers)
- Capital misallocation (parent invests in declining division to "protect" it)
- Reduced accountability (easy to hide poor performer in consolidated results)
- Higher cost of capital (market discounts diversified firms)
Most diversified conglomerates trade at a discount to SOTP due to dissynergies exceeding synergies. This is the "conglomerate discount," the subject of the next chapter.
To adjust SOTP for synergies/dissynergies:
- Identify quantifiable synergies (e.g., $500M annual cost savings from procurement)
- Estimate realization probability (e.g., 80% likely to achieve the target)
- Calculate net present value (e.g., $500M × 0.80 = $400M annual; $400M / 0.08 = $5B NPV)
- Add or subtract from SOTP value
Valuation Mechanics Illustrated
Special Situations: Insurance Conglomerates
Berkshire Hathaway, Markel, Fairfax Financial, and others are insurance conglomerates. Valuing them requires specialized knowledge:
Float Valuation Insurance float is cash received from policyholders before claims are paid. If premiums exceed claims (underwriting profit), float is cheap or free financing. If claims exceed premiums (underwriting loss), float is expensive. Value float at book value if underwriting is breakeven, or adjust if there's consistent underwriting profit/loss.
Combined Ratio Combined ratio = (Claims + Operating Expenses) / Premiums. Ratio < 100% = underwriting profit. Ratio > 100% = underwriting loss.
A SOTP for an insurance conglomerate:
| Component | Value Method | Estimated Value |
|---|---|---|
| Underwriting Earnings | 12–15x P/E | $5–10B |
| Float Value | Book Value + Underwriting Edge | $20–30B |
| Operating Subsidiary Earnings | Segment-specific multiples | $2–5B |
| Investments (at Market) | Market price | $50–100B |
| Total Enterprise Value | - | $77–145B |
Special Situations: Real Estate Companies
Real estate-heavy conglomerates (REITs, real estate operating companies) use different methods:
Net Asset Value (NAV) Estimate the fair value of all real estate properties (sometimes using appraisals, sometimes market sales comps), subtract liabilities, divide by shares outstanding. This is the fundamental value.
Price-to-FFO (Funds from Operations) FFO adjusts accounting earnings for non-recurring or non-cash items, focusing on cash flow. FFO-based multiples (e.g., 12x FFO) are commonly used.
A SOTP for a mixed real estate/industrial company:
| Property Type | Estimated Fair Value | FFO Multiple | Operating Value |
|---|---|---|---|
| Residential | $200M (appraisal) | - | $200M |
| Commercial | $300M (appraisal) | - | $300M |
| Industrial (operating) | - | 12x | $50M FFO × 12 = $600M |
| Total | - | - | $1,100M |
Common Mistakes
Using a Single Multiple for Disparate Businesses Applying a 12x EBITDA multiple to a conglomerate's consolidated EBITDA ignores that different segments deserve different multiples. Always segment first.
Forgetting Corporate Overhead Segment operating income excludes corporate G&A. If you sum segment values without subtracting corporate overhead, you've overstated enterprise value.
Overestimating Synergies Synergies are often oversold in M&A contexts. Be conservative; use probability-weighted values and require multiple sources of evidence.
Ignoring Minority Interests If a conglomerate owns 90% of a subsidiary and outside investors own 10%, the SOTP value of that subsidiary must be adjusted. The conglomerate's claim is 90% of the value, not 100%.
Applying Wrong Discount Rates Each segment should have a WACC reflecting its risk profile. A utility subsidiary (low beta, low WACC) requires a different discount rate than a high-growth tech subsidiary. Don't use the parent's WACC for all segments.
Failing to Update Assumptions Conglomerates' segment mix and relative performance evolve. An analysis from three years ago may not reflect today's reality. Always refresh SOTP annually.
Frequently Asked Questions
Q: Should I value a holding company like Berkshire the same way as an operating conglomerate like 3M? A: Partially. Both use SOTP; the difference is that holding companies (like Berkshire) often own large stakes in external companies (Apple, Bank of America) valued at market prices, while operating conglomerates (like 3M) value segments using multiples of internal earnings. Both require careful segment-by-segment analysis, but holding companies have simpler mechanics because some stakes are directly observable.
Q: What if a segment is unprofitable or in turnaround? A: Use DCF or scenario analysis. Project when the segment returns to profitability, model recovery cash flows, and discount. Multiples don't work for negative earnings. Be conservative; many turnarounds fail.
Q: How do I handle tax efficiency or tax shield benefits in a SOTP? A: Corporate tax planning, loss carryforwards, and transfer pricing all affect consolidated taxes. At the segment level, use the company's effective tax rate (or a normalized rate) and let corporate tax impacts flow through the reconciliation. Don't try to value tax shields segment-by-segment; it's too complex.
Q: Should I assume the segments stay together or get spun off? A: Do both. One scenario assumes segments stay combined; calculate the conglomerate value (with overhead allocation, synergies, etc.). Another scenario assumes each segment is spun off and operated independently; calculate standalone valuations and sum them. The difference is the conglomerate discount or premium.
Q: How sensitive is my SOTP to the choice of multiples? A: Very. A 1–2x change in EBITDA multiples (e.g., 11x vs. 13x) can shift total valuation by 5–15%. Always stress-test by varying multiples ±1–2x for each segment and showing the range.
Related Concepts
- What is Sum-of-the-Parts Valuation?: Foundational framework for this chapter.
- Identifying Business Segments: Extract data to feed into conglomerate valuations.
- The Conglomerate Discount Trap: Understand why conglomerates often trade at discounts and when those discounts are justified.
- Valuing Spin-offs and Divestitures: Apply conglomerate valuation logic to pre/post-transaction scenarios.
- Comparable Company Analysis: Build segment-level comparables.
- Discounted Cash Flow Valuation: Build DCF models for complex or high-growth segments.
Summary
Valuing a conglomerate is as much an art as a science. It requires segmenting the business cleanly, choosing appropriate valuation methods for each segment's economics, and then carefully aggregating while accounting for corporate costs, tax effects, and potential synergies or dissynergies. The best approach combines multiple methods—multiples, DCF, NAV—and triangulates to a value range rather than a point estimate. Sensitivity analysis is essential because small changes in multiples or growth assumptions can meaningfully shift the valuation. The payoff for disciplined conglomerate valuation is the ability to identify mispricings, spin-off opportunities, and value traps that investors using single multiples or generalized DCF models miss.
Next
Continue to The Conglomerate Discount Trap to understand why conglomerates often trade at valuations below their SOTP and when that discount reflects genuine problems vs. opportunity.