Comparing a stock's P/E to "the S&P 500 average" is one of the most common screening mistakes, because the index average is really a blend of eleven sectors with structurally different valuation norms. A utility trading at 16x earnings isn't cheap relative to the index average of ~20x — it's expensive relative to its own sector, which has historically traded closer to 14–16x for structural reasons that have nothing to do with the broader market's valuation level. Screening sector-relative, rather than index-relative, is the fix.
Last updated: July 2026.
Why sectors trade at structurally different valuations
The S&P 500 is organised into 11 GICS (Global Industry Classification Standard) sectors, and each carries a different typical valuation range for identifiable structural reasons — not market inefficiency:
| Sector | Typical P/E range | Why |
|---|---|---|
| Information Technology | 25–35x | High growth, high margins, asset-light scalability |
| Consumer Discretionary | 20–30x | Growth premium, cyclicality priced in |
| Communication Services | 18–25x | Mix of high-growth platforms and mature telecom |
| Health Care | 16–22x | Defensive demand, patent-driven growth in pharma/biotech |
| Industrials | 16–22x | Cyclical, capital-intensive, moderate growth |
| Consumer Staples | 18–24x | Defensive, low growth, valued for stability |
| Financials | 10–14x | Regulatory capital constraints, cyclical earnings |
| Materials | 14–18x | Commodity-price cyclicality |
| Real Estate | Use P/FFO, not P/E | Depreciation distorts net income (see US REIT screening) |
| Energy | Highly cyclical | Earnings swing with commodity prices; P/E can be misleadingly low at cycle peaks |
| Utilities | 14–18x | Regulated returns, low growth, bond-proxy characteristics |
A blanket "P/E < 15" value screen applied across the full S&P 500 will systematically overweight financials, materials, and utilities while excluding almost the entire technology and consumer discretionary sectors — not because those sectors are overvalued, but because their structural valuation range sits well above 15x by default.
Screening within sectors, not just within the index
The practical fix is running valuation, quality, and growth screens within each sector, then comparing the resulting shortlist against that sector's own historical range — rather than filtering the full S&P 500 universe with one uniform threshold.
Sector-relative value screen (example: Industrials):
- Sector: Industrials
- P/E < 18 (below the sector's typical range, not an absolute market benchmark)
- Operating margin > 10%
- Net Debt/EBITDA < 2.5
- Sort by: P/E ascending
Sector-relative growth screen (example: Technology):
- Sector: Information Technology
- Revenue growth (3yr CAGR) > 15%
- Operating margin > 20%, expanding YoY
- Sort by: revenue growth descending
Running the same structural approach — sector filter first, then valuation and quality thresholds calibrated to that sector's own norms — across all 11 sectors produces a much more comparable set of "cheap relative to peers" or "high quality within its category" candidates than a single index-wide screen ever could.
Cross-sector rotation: using sector dispersion deliberately
Sector valuation gaps aren't static — they widen and narrow with the economic cycle, interest rate environment, and market sentiment. Screening sector-relative dispersion over time (rather than at a single snapshot) surfaces potential rotation opportunities:
When a sector trades unusually cheap relative to its own historical range (not the market average), it can signal either a justified structural re-rating (secular decline, regulatory risk) or a cyclical overreaction worth investigating further with company-specific screens.
When a sector trades unusually expensive relative to its own historical range, it can signal either justified re-rating (genuine structural improvement — margin expansion, new growth drivers) or a crowded, potentially vulnerable position.
The key discipline: always compare a sector to its own history first, and only then consider whether that historical range itself still makes sense given the sector's current fundamentals — comparing directly across sectors (Technology's 30x vs. Financials' 12x) without that structural context leads to superficial and usually wrong conclusions ("Financials are cheap, Technology is expensive").
Building a full sector-dispersion screen
| Filter | Value |
|---|---|
| Sector | (select one of the 11 GICS sectors) |
| P/E | Below the sector's own 5-year median (not an absolute threshold) |
| Operating margin | Above the sector's own median |
| Net Debt/EBITDA | Below 3.0, adjusted for sector norm (higher for Utilities/Real Estate) |
| Sort by | P/E ascending, within the selected sector |
Repeating this same filter structure across each of the 11 sectors in turn — rather than running one blended screen across the full index — is the practical way to build a genuinely comparable, sector-aware shortlist.
Common mistakes when screening across S&P 500 sectors
Using one valuation threshold for the whole index: As shown above, this systematically biases results toward structurally cheaper sectors (Financials, Utilities, Materials) and excludes structurally more expensive ones (Technology, Consumer Discretionary), regardless of company-specific merit within each.
Comparing Energy sector P/E across a full commodity cycle without adjustment: Energy earnings — and therefore P/E — swing dramatically with commodity prices. A low P/E at a cyclical earnings peak can actually signal the market pricing in a coming decline, not cheapness.
Applying P/E to Real Estate at all: As with European REITs and US REITs, depreciation distorts net income for real estate companies. Use P/FFO instead.
Ignoring that sector composition itself changes over time: GICS sector reclassifications (companies moving between Communication Services, Technology, and Consumer Discretionary in particular) have shifted historical sector-level valuation comparisons meaningfully over the past decade — a long-run historical sector P/E comparison should account for these reclassifications rather than assuming perfect continuity.
The same structural-composition logic applies to comparing exchanges rather than just sectors — see Nasdaq vs NYSE stocks for why a raw cross-exchange valuation comparison is largely a disguised sector comparison too.
Bottom line
The S&P 500's headline valuation is a blend of 11 sectors with structurally different, historically persistent valuation norms — comparing an individual stock against that blended average rather than against its own sector produces systematically misleading conclusions about whether it's actually cheap or expensive. Screen within sectors first, calibrate thresholds to each sector's own historical range, and only compare across sectors once that structural context is accounted for.
Frequently asked questions
Why do S&P 500 sectors trade at such different valuations?
Each sector has structurally different growth rates, capital intensity, cyclicality, and regulatory characteristics. Technology commands a growth and asset-light-scalability premium; Financials trade at a discount reflecting regulatory capital constraints and cyclical earnings; Utilities trade like bond proxies given regulated, low-growth returns. These gaps are persistent structural features, not simply market inefficiency.
Is it wrong to compare a stock's P/E to the overall S&P 500 average?
It's a common but misleading shortcut. The index average blends sectors with very different structural valuation norms, so a stock's P/E relative to that blended figure says less about whether it's cheap or expensive than comparing it to its own sector's typical range does.
How should I screen for value within a specific sector?
Filter first by sector, then apply valuation thresholds calibrated to that sector's own historical range rather than an absolute market-wide number — for example, P/E below 18 might be cheap for Industrials but unremarkable for Financials, which typically trades in the 10–14x range.
Does this sector-relative approach apply outside the S&P 500?
Yes — the same structural logic applies to European and Canadian sector screening. Sector-relative valuation, rather than index-relative, is the more reliable comparison framework in any market with meaningful sector composition differences.
Screen S&P 500 and global stocks by sector and valuation → — free, no account required. Filter within sectors across US, Canadian, and European exchanges.