Cross-industry multiple comparisons

Quick definition

Cross-industry multiple comparisons—evaluating a company's valuation by comparing it to companies in different industries—are dangerous without adjustment. A 20× P/E that is expensive for a utility is cheap for a software company. Multiples vary systematically by industry due to growth, profitability, capital intensity, and risk. Comparing unadjusted multiples across industries is a path to overpaying or selling prematurely.

Key takeaways

• Multiples vary systematically by industry: tech 25–35×, utilities 13–15×, energy 12–18×, banks 10–13×, retailers 15–20×
• Growth is the primary driver: high-growth companies justify high multiples; slow-growth command low multiples
• Capital intensity and margins matter greatly: asset-light, high-margin businesses deserve higher multiples than capital-heavy, low-margin ones
• Risk and stability affect multiples: regulated utilities deserve lower multiples than cyclical industrials or startups
• Comparing unadjusted multiples across industries leads to systematic mispricings (tech looks expensive, utilities look cheap, both on misleading signals)
• Adjust multiples for differences using growth rates, margins, ROE, and leverage before cross-industry comparison
• Same multiple can mean different things: 18× P/E is cheap for a high-growth SaaS company, expensive for a mature manufacturer

Why multiples differ so dramatically across industries

Growth as the primary driver

High-growth industries justify higher multiples. A software company growing 20% annually supports a 30–35× P/E; a utility growing 2–3% supports 13–15×.

The PEG ratio—P/E divided by growth rate—illustrates this. A software company at 30× earnings growing 25% has PEG of 1.2 (fair). A utility at 15× earnings growing 3% has PEG of 5.0 (appears expensive). But both are fairly valued relative to growth. The PEG adjustment bridges the cross-industry comparison.

Capital intensity and margins

Software companies convert 30% of revenue to operating profit; they are capital-light. A manufacturer might convert 5% of revenue to profit and require substantial capex for growth. The manufacturer needs more revenue to support the same net income, and that earnings power is less certain. Lower multiples reflect this structural disadvantage.

Example: Software vs. Manufacturing

Metric	Software	Manufacturer
Gross margin	80%	30%
Operating margin	20%	5%
FCF margin	18%	3%
Capex as % revenue	5%	10%
P/E multiple	30×	10×

Both companies earn $100M net income. The software company generated that on $500M revenue; the manufacturer on $2B revenue. The software company's earnings are more durable and less dependent on continued capital expenditure. The 3× higher multiple for software is justified by structural advantages, not overvaluation.

Risk, stability, and regulatory environment

Utilities operate under regulatory frameworks that limit profit but guarantee return on equity. This stability warrants lower multiples. Conversely, a startup with high revenue growth but uncertain path to profitability merits a lower multiple because the downside risk is higher.

Banks face regulatory capital constraints that limit returns and create cyclical earnings volatility. A bank earning 12% ROE but constrained by regulation deserves a lower multiple than a software company earning 35% ROE with no regulatory constraints.

Business model durability

A high-multiple business has durable, predictable cash flows. SaaS companies with sticky customers, high switching costs, and 85%+ net retention justify premium multiples. A distributor of commodities with thin margins and many competitors justifies a discount multiple.

The multiple is compensation for confidence in the durability of earnings. High confidence = high multiple.

Cross-industry multiple framework

The adjusted multiple matrix

Use this framework to compare multiples across industries fairly:

Step 1: Calculate base multiple for each company

• Company A (software): P/E 32×, EV/EBITDA 35×
• Company B (industrial): P/E 12×, EV/EBITDA 10×

Unadjusted, Company A looks expensive; Company B cheap.

Step 2: Calculate growth-adjusted multiple (PEG-like)

• Company A: Growing 25% → PEG = 32 / 25 = 1.28
• Company B: Growing 4% → PEG = 12 / 4 = 3.0

On a growth-adjusted basis, Company A is cheap (PEG 1.28 < 2.0); Company B is expensive (PEG 3.0 > 2.0). The analysis flips.

Step 3: Adjust for profitability (margin multiple)

• Company A: Operating margin 22%, FCF margin 20%
• Company B: Operating margin 5%, FCF margin 2%

Company A's earnings are 4× more profitable on a margin basis. Apply a 20% premium to account for margin durability: 1.28 × 1.20 = 1.54 (adjusted PEG).

Company B's thin margins indicate capital intensity and earnings fragility. Apply a 20% discount: 3.0 × 0.80 = 2.4 (adjusted PEG).

Even after adjustment, Company A at 1.54 looks fair to attractive; Company B at 2.4 looks expensive.

Step 4: Adjust for capital intensity and FCF

EV/EBITDA is less distorted by capital intensity than P/E, but still matters.

• Company A: EV/EBITDA 35×, FCF/EBITDA 85%
• Company B: EV/EBITDA 10×, FCF/EBITDA 25%

On a free cash flow basis, Company A's multiple compresses: 35× × 0.85 = ~30× FCF multiples. Company B's compresses more: 10× × 0.25 = ~4× FCF multiple. Company A is slightly expensive on FCF; Company B is reasonable.

Step 5: Risk and leverage adjustment

• Company A: Net debt/EBITDA 0.5×, ROE 28%, growth stable
• Company B: Net debt/EBITDA 2.5×, ROE 8%, growth cyclical

Company A has lower leverage, higher returns, more stable growth. No discount needed; perhaps a 5% premium for quality.

Company B has higher leverage, lower returns, cyclical exposure. Apply 15–20% discount for risk.

Final adjusted comparison

Factor	Company A	Company B
Base multiple	32×	12×
Growth adjustment	÷ 25 = 1.28×	÷ 4 = 3.0×
Margin adjustment	× 1.20 = 1.54×	× 0.80 = 2.4×
Risk adjustment	+ 5% = 1.62×	- 20% = 1.92×

On an adjusted basis, Company A at 1.62× adjusted multiple is fair to attractive; Company B at 1.92× is expensive despite trading at 12× P/E unadjusted. The unadjusted multiple comparison was misleading.

Industry multiples: baseline expectations

Technology sector

SaaS and enterprise software: 25–35× P/E, 8–12× EV/Sales

• High growth (15–40% annually)
• High margins (20–30% operating)
• Low capital intensity (5% of revenue)
• Strong ROE (25–40%)

Cloud infrastructure (IaaS): 20–30× P/E, 6–10× EV/Sales

• Very high growth (20–40%)
• Lower margins (10–20% operating) due to capital for data centers
• Moderate capital intensity (15–20% of revenue)
• Strong ROE (15–30%)

Semiconductors: 15–25× P/E, 4–8× EV/Sales

• High growth (10–25%)
• Moderate-to-high margins (20–35% operating)
• High capex intensity (20–30% of revenue) driven by fab investment
• Good ROE (15–25%)

Healthcare

Pharmaceuticals: 12–18× P/E, 3–6× EV/Sales

• Moderate growth (4–8%)
• Very high margins (25–40% operating)
• Moderate capex (8–12% of revenue for R&D and manufacturing)
• Excellent ROE (15–25%)
• Patent-driven multiples (cliff risk when patents expire)

Medical devices: 18–28× P/E, 5–8× EV/Sales

• Moderate-to-good growth (8–15%)
• Good margins (20–30% operating)
• Moderate capex (8–10% of revenue)
• Good ROE (15–20%)

Financials

Banks: 8–13× P/E, 0.8–1.2× Price-to-Book

• Slow growth (3–6%)
• Moderate margins (20–35% net margin on revenue, but earnings compressed by leverage)
• Low capex (not capital-intensive in traditional sense; deposits are "capital")
• ROE typically 10–15%, constrained by regulation

Insurance: 10–15× P/E, 0.8–1.3× Price-to-Book

• Low-to-moderate growth (3–8%)
• Moderate underwriting margins (2–5% operating, but investment income adds)
• Very low capex
• ROE typically 10–15%

Consumer and Retail

Consumer discretionary (branded): 15–25× P/E, 1–2.5× EV/Sales

• Moderate growth (5–12%)
• Moderate margins (8–15% operating)
• Moderate capex (3–5% of revenue for stores/distribution)
• Good ROE (10–20%)

Consumer discretionary (discount retail): 12–18× P/E, 0.5–1.5× EV/Sales

• Low growth (2–5%)
• Low margins (2–5% operating)
• High capex intensity (4–6% of revenue for stores)
• Lower ROE (6–12%)

Consumer staples: 16–22× P/E, 1.5–2.5× EV/Sales

• Low growth (2–5%)
• Moderate margins (10–15% operating)
• Low capex (2–4% of revenue)
• Good ROE (12–18%)

Industrials and Manufacturing

Industrial equipment: 12–18× P/E, 1–2× EV/Sales

• Low-to-moderate growth (4–10%)
• Moderate margins (8–15% operating)
• Moderate capex (4–6% of revenue)
• Good ROE (12–18%)

Machinery and cyclicals: 8–14× P/E, 0.8–1.5× EV/Sales

• Variable growth (highly cyclical)
• Lower margins (5–12% operating, cyclical)
• Moderate-to-high capex (5–10% of revenue)
• Cyclical ROE (8–20%)

Utilities and Energy

Utilities (regulated): 13–16× P/E, 1.5–2.5× EV/Sales

• Very low growth (1–3%)
• Moderate margins (25–35% operating)
• High capex (8–12% of revenue for infrastructure)
• Steady, regulated ROE (8–10%)

Integrated oil & gas: 8–15× P/E, 0.5–1.5× EV/Sales

• Variable growth (cyclical, depending on commodity prices)
• Highly variable margins (10–40% operating, cyclical)
• High capex (10–20% of revenue for exploration and development)
• Cyclical ROE (5–25%)

Renewables: 12–20× P/E, 2–4× EV/Sales

• Moderate growth (5–15%)
• High margins (50–70% EBITDA margins)
• High capex (30–50% of revenue during construction, lower post-construction)
• Stable, regulated-like ROE (7–12%)

Real-world examples

Microsoft vs. Duke Energy

Microsoft: P/E 30×, growth 12%, operating margin 42%, capex 5% of revenue, ROE 35%

Duke Energy: P/E 14×, growth 2%, operating margin 32%, capex 12% of revenue, ROE 8%

Unadjusted: Microsoft at 30× looks expensive; Duke at 14× looks cheap.

Adjusted:

• Growth: Microsoft justified at 30/12 = 2.5× earnings multiple per percentage point of growth. Duke at 14/2 = 7× earnings per point. Microsoft looks cheap on a growth-adjusted basis.
• Margins: Microsoft's higher margins and lower capex mean its earnings are more durable and capital-efficient. Worth 20% premium.
• ROE: Microsoft's 35% ROE is exceptional; Duke's 8% is constrained by regulation. This reinforces the multiple premium for Microsoft.

Conclusion: Both fairly valued on an adjusted basis. Microsoft's 30× is not expensive for its fundamentals; Duke's 14× is not cheap relative to its constraints.

Nvidia vs. Walmart

Nvidia: P/E 45×, growth 60%, operating margin 35%, capex 8%, ROE 70%

Walmart: P/E 25×, growth 3%, operating margin 6%, capex 3%, ROE 10%

Unadjusted: Nvidia expensive at 45×; Walmart reasonable at 25×.

Adjusted:

• Growth: Nvidia's 45/60 = 0.75× per-percentage-point multiple is very attractive. Walmart's 25/3 = 8.3× is expensive relative to growth.
• Margins: Nvidia's 35% operating margin vs. Walmart's 6% is a massive structural advantage. Nvidia's earnings are 6× more durable per revenue dollar.
• Capex: Nvidia's earnings convert to robust FCF; Walmart's lower capex but thin margins means limited FCF.

Conclusion: Nvidia's 45× is attractive for the growth and margin profile; Walmart's 25× is expensive relative to 3% growth and thin margins. The unadjusted comparison was reversed.

JPMorgan vs. Blackrock

JPMorgan: P/E 12×, ROE 13%, net interest margin 1.9%, growth 3%

Blackrock: P/E 18×, ROE 12%, asset growth 5%, growth 4%

Unadjusted: Blackrock at 18× is more expensive than JPMorgan at 12×.

Adjusted:

• ROE: JPMorgan's ROE is actually higher (13% vs. 12%), but both are constrained by regulation (banks) and competitive dynamics (asset managers).
• Growth: Blackrock grows slightly faster (4% vs. 3%), but both are mature financial services.
• Business model: JPMorgan is capital-constrained by regulation; Blackrock is less constrained. Blackrock's multiple should be higher to reflect greater capital flexibility.

Conclusion: Blackrock's 18× is slightly expensive for 4% growth, but justified by higher operational leverage. JPMorgan at 12× is fair for a constrained bank. Not a screaming buy or sell either direction; fair valuations given constraints.

Common mistakes in cross-industry analysis

Mistake 1: Using P/E alone without context

A 15× P/E is cheap for a software company and expensive for a utility. Using absolute P/E levels to screen across industries leads to systematic mispricings.

Mistake 2: Forgetting about capex in capital-intensive industries

A capital-intensive business with 12× P/E but high capex might have lower FCF yield than a 20× software company with minimal capex. Comparing P/E misses the cash return to shareholders.

Mistake 3: Comparing stable to cyclical without adjusting

A cyclical industrial at 10× earnings might be trading at peak cycle earnings. A utility at 15× has more stable earnings. The industrial looks cheaper on P/E, but on normalized earnings might be more expensive.

Mistake 4: Ignoring regulatory constraints

A utility at 15× with 2% growth looks expensive relative to peers. But regulation guarantees ROE. A 15× multiple is fair for regulated, stable cash flows. Comparing a utility P/E to an unregulated industrial P/E is inappropriate without adjustment.

Mistake 5: Confusing margin expansion with multiple expansion

A company improving from 10% to 15% operating margin warrants a multiple expansion. But comparing two companies at different margins without recognizing they are at different efficiency levels (not different growth or risk) is naive.

Mistake 6: Forgetting leverage in EV multiples

EV/EBITDA is supposed to be leverage-neutral, but it is not entirely. A leveraged company with 3× debt-to-EBITDA and a conservative company with 1× debt-to-EBITDA might have the same EV/EBITDA but different equity valuations. Adjust for leverage differences.

FAQ

Is a software company at 25× P/E cheap compared to a utility at 15×?

Not necessarily. On a growth-adjusted basis (PEG), the software company's multiple might be attractive. But you must adjust for capital intensity, margins, and risk. Without adjustment, the comparison is meaningless.

How do I adjust for cyclicality when comparing across industries?

Use normalized or trough earnings, not current earnings. A cyclical industrial at peak cycle is more expensive than it appears; one at trough is cheaper. Calculate multiples using 3–5 year average earnings for cyclical businesses.

Should I use P/E or EV/EBITDA for cross-industry comparisons?

EV/EBITDA is slightly better because it neutralizes leverage and capital structure. P/E is distorted by different tax rates and financing mix. But neither is perfect. Use both; if they diverge, investigate.

Can I compare tech to financial services?

Only with heavy adjustments. Different growth rates, margins, capital constraints, and risk profiles. A software company and a bank at the same P/E are not at the same valuation on an adjusted basis. The software company is almost certainly cheaper.

How do I know if my adjustments are too aggressive?

Check your adjustments against historical norms and other analysts' work. If your adjusted multiples for two industries converge significantly, you may be over-adjusting. Some multiple premium for higher-quality businesses is structural and appropriate.

Related concepts

• Price-to-earnings ratio — baseline metric; must be adjusted for cross-industry use
• Enterprise value and multiples — EV/EBITDA and EV/Sales as alternatives, slightly better for leverage-neutral comparison
• PEG ratio — growth-adjusted P/E, useful for cross-industry comparisons
• Return on equity and invested capital — measures of business quality that justify multiple premiums
• Capital intensity and cash flow — understanding how capex affects earnings quality
• Regulatory environment and returns — how regulation constrains multiples for utilities and financials

Summary

Multiples differ dramatically across industries due to growth, profitability, capital intensity, and risk. Comparing unadjusted multiples across industries is misleading and leads to systematic mispricings—tech companies appear expensive, utilities appear cheap, but both can be fairly valued when adjusted. Use a structured framework: adjust for growth (PEG-like), margins (margin efficiency), capital intensity (capex as percentage of revenue), and risk (leverage, stability, regulatory constraints). Software companies rightfully trade at 25–35× earnings; utilities rightfully trade at 13–15×; the difference is structural, not a valuation opportunity. The most dangerous mistake is comparing P/E multiples across industries without recognizing these differences. Always adjust before concluding a cross-industry comparison. Combined with fundamental analysis of business quality, adjusted relative valuation is a powerful tool; without adjustment, it is a path to value destruction.

Next

Read Valuing cyclical stocks with multiples to understand how to apply multiples to cyclical businesses, where earnings swing dramatically and multiples must be normalized.