When to Use P/E and EV/EBITDA

Despite their flaws, P/E and EV/EBITDA multiples remain the most widely used valuation tools. They're fast, intuitive, and marketable to non-technical audiences. The mistake isn't using them; it's using them without understanding their limitations. A P/E ratio is useful when applied carefully with proper adjustments. The key is knowing exactly what you're measuring and what assumptions are baked into that number.

Quick definition: P/E ratio (Price / Earnings) values a company by its earnings multiple; EV/EBITDA (Enterprise Value / EBITDA) values it by operating earnings independent of capital structure and tax rate. Both are shortcuts that assume comparability across companies.

Key Takeaways

• P/E works best for stable, mature companies with predictable earnings
• EV/EBITDA works best for comparing companies with different capital structures
• Always adjust multiples for growth rate, ROIC, and capital intensity
• PEG ratio (P/E / growth rate) is a useful adjustment, though imperfect
• Trailing multiples use past earnings; forward multiples use forecasts (both have risks)
• True comparability requires normalizing for accounting, leverage, and business model

The P/E Ratio: When It Works

Best Use Case: Stable Mature Companies

A 20-year-old consumer staples company with consistent 5% earnings growth, minimal capex requirements, and stable margins. Here, P/E is relatively meaningful because earnings are predictable and stable.

Example: Procter & Gamble with $6B annual earnings and a market cap of $320B would have a P/E of 53x. Too high? Only if you assume no growth. If P&G can grow earnings at 4–5% indefinitely (a reasonable assumption given its moat), then 53x isn't absurd. Comparing to competitors at 45x gives you directional guidance.

Why It Works Here:

• Earnings are stable and predictable
• The business won't change materially
• Leverage and accounting are consistent
• Growth rate is modest and built into the multiple

Worst Use Cases: High-Growth and Cyclical Companies

A high-growth company with volatile earnings makes P/E meaningless. Amazon reported losses or minimal earnings for two decades; its P/E was infinity or undefined. Yet it was obviously valuable. Using P/E to value Amazon vs. Walmart tells you nothing.

A cyclical company at peak earnings might show a P/E of 8x, appearing "cheap." But if earnings are about to collapse, the true earnings power is lower, and 8x is expensive. Normalizing for the cycle changes the calculation entirely.

The EV/EBITDA Ratio: When It Works

Best Use Case: Comparing Companies with Different Leverage

Company A has $100M EBITDA, $10M debt, and a market cap of $500M. Company B has $100M EBITDA, $50M debt, and a market cap of $450M.

P/E comparison is difficult because different leverage affects net earnings. But EV/EBITDA (Enterprise Value / EBITDA):

• Company A: EV = $510M; EV/EBITDA = 5.1x
• Company B: EV = $500M; EV/EBITDA = 5.0x

EV/EBITDA shows they're valued identically despite different capital structures. This clarity is why bankers and DCF modelers prefer EV-based metrics.

Why It Works Here:

• Removes the distortion of capital structure
• Removes tax effects (EBITDA is pre-tax)
• Allows comparison of companies with different depreciation policies
• Works for unprofitable companies (if EBITDA is positive)

Worst Use Cases: Capital-Intensive vs. Asset-Light

A capital-intensive manufacturer with $100M EBITDA but $50M annual capex requirements has real free cash flow of $50M. A SaaS company with $100M EBITDA and $5M capex has $95M free cash flow. Same EV/EBITDA, wildly different cash generation.

A company with large non-cash charges (depreciation) has inflated EBITDA relative to cash earnings. A company with off-balance-sheet capex (operating leases) has understated capex in the multiple.

Adjustments That Make Multiples Meaningful

1. PEG Ratio: Adjust for Growth

PEG = P/E / Expected Annual Growth Rate

A stock at 20x earnings with 20% expected growth has a PEG of 1.0. One at 20x with 10% growth has a PEG of 2.0.

Limitations:

• Assumes the growth rate is accurate (analyst estimates are often wrong)
• Doesn't account for ROIC (growth at 5% ROIC is worthless)
• Penalizes high-growth companies if growth disappoints slightly

Better Use: PEG is a screening tool, not a final arbiter. Use it to rank stocks, then investigate the most attractive ones.

2. Adjusted EV/EBITDA: Account for Capex

Free Cash Flow / Market Cap is more accurate than EV/EBITDA.

Or adjust EV/EBITDA for typical capex:

EV / (EBITDA - Normalized Capex)

A company with $100M EBITDA, $30M normalized capex, and $500M EV would have:

• Raw EV/EBITDA: 5.0x
• Adjusted EV/(EBITDA - Capex): 7.1x

The adjusted multiple is more honest about returns to shareholders.

3. Earnings Quality Adjustments

Calculate earnings before suspect items:

Remove one-time gains, aggressive accounting, and stock-based comp.

Example: A company reports $100M net earnings. Adjustments:

• Add back: $10M one-time gain from asset sale
• Add back: $8M stock-based compensation (dilution)
• Remove: $5M tax benefit from restructuring (non-recurring)
• Adjusted earnings: $87M

If the stock trades at $1,000M market cap, the raw P/E is 10x; the adjusted P/E is 11.5x. The adjusted multiple is fairer.

4. Normalized Earnings for Cycles

Cyclical Business Adjustment:

For a cyclical company, calculate earnings at mid-cycle rather than peak or trough.

Example: Steel company earnings:

• Peak (2011): $20 per share
• Trough (2016): -$5 per share (loss)
• Mid-cycle estimate: $8 per share

If it trades at $50 per share, the apparent P/E using peak earnings (2.5x) is misleading. Using normalized earnings (6.3x), the valuation becomes clearer.

Forward vs. Trailing Multiples

Trailing P/E: Based on the last 12 months of actual earnings. Trustworthy but backward-looking.

Forward P/E: Based on analyst estimates for the coming 12 months. Forward-looking but optimistic.

When earnings are rising, forward P/E is lower and looks more attractive (if the forecast is right). When earnings are declining, forward P/E is higher and looks less attractive.

In a recession, trailing P/E spikes because earnings have collapsed, while forward P/E might be reasonable (assuming earnings recovery). Use both, and understand which story they're telling.

Practical Framework: Multi-Multiple Valuation

Use all three together:

1. Calculate P/E — What's the current earnings multiple?
2. Calculate EV/EBITDA — What's the operating earnings multiple?
3. Adjust both — Account for growth, leverage, capex
4. Compare to peers — Do the adjusted multiples align with comparable companies?
5. Reverse to intrinsic value — If fair multiple is 12x earnings, what's intrinsic value?

This sequence uses multiples as a tool, not an answer.

When Multiples Fail: Red Flags

Spread from Fundamentals

If your multiple analysis suggests fair value is $80, but intrinsic value (via DCF) is $120, something's wrong. Either your multiple adjustments are off, your DCF assumptions are aggressive, or the market is mispricing the stock.

Rapid Changes

If a stock's P/E has jumped from 12x to 25x without earnings growth, investor sentiment has shifted. The multiple expansion might be justified (if future growth is real) or might be a bubble (if growth is unlikely). Investigate the cause.

Peer Divergence

If one peer trades at 10x and another at 20x despite similar earnings and growth, one is likely mispriced. But verify that they're truly comparable before assuming the cheaper one is a bargain.

Real-World Applications

Screening: Run a stock screener for lowest P/E, lowest EV/EBITDA, etc. Filter for companies with decent growth (avoid pure value traps). You've found candidates to analyze deeper—not final investment decisions.

Quick Due Diligence: At an earnings call, the CFO mentions 20% expected earnings growth. A P/E of 16x with PEG of 0.8 looks reasonable. This quick check passes the bar for further analysis.

Exiting Positions: Your investment has tripled; stock trades at 30x earnings. Peers are at 20x. Is it fairly valued or a sell? Adjust for growth; if growth rate justifies 30x, hold. If not, consider selling.

Evaluating Analyst Reports: An analyst recommends a stock at "undervalued at 12x," but doesn't adjust for that the company is growing 3%. Raw 12x isn't cheap for 3% growth; it's fair at best. Question the analysis.

Common Mistakes

Using raw P/E without normalization. A peak-cycle company at 8x earnings isn't cheap; it's at risk of compression.

Confusing multiple expansion with value creation. A stock can rise due to P/E expansion (sentiment), not earnings growth. This is wealth transfer, not wealth creation.

Comparing forward P/E to trailing EV/EBITDA. Mix apples with apples: forward P/E to forward EV/EBITDA, or trailing to trailing.

Ignoring the interest rate environment. In a 1% rate environment, 25x earnings is reasonable for a stable business. In a 6% rate environment, it's expensive. Always adjust for rates.

Using one multiple for a complex business. A conglomerate with different divisions requires breaking it down by segment and comparing multiples separately.

FAQ

Q: Should I ever use only P/E without adjustments? A: Only for mature utilities and consumer staples where earnings are truly predictable. For most companies, raw P/E is misleading.

Q: Is EV/EBITDA better than P/E? A: For comparing companies with different leverage, yes. For comparing growth rates, no. Use both.

Q: How do I handle negative EBITDA or losses? A: You can't. If EBITDA is negative, the company is burning cash, and multiples are meaningless. Move to a different valuation method (DCF, sum-of-the-parts).

Q: What's a "fair" P/E multiple? A: It depends on growth, ROIC, leverage, and interest rates. A mature company: 12–18x. A growing company: 18–30x. A high-growth company: 25–50x. These are ranges, not rules.

Q: How do I compare P/E across different countries? A: Adjust for expected growth rates and cost of capital differences. A 15x P/E in Japan on 1% growth is expensive vs. 15x in India on 10% growth. Compare real returns, not nominal multiples.

Related Concepts

• Trailing vs. Forward Earnings: The timing and reliability of earnings estimates
• Earnings Quality: How trustworthy reported earnings are
• Comparable Companies (Comps): The peer group used in multiple analysis
• Margin of Safety: Using a discount to fair multiple-based value
• Multiple Expansion/Compression: How P/E ratios change independent of earnings growth

Summary

P/E and EV/EBITDA multiples are valid tools when used with discipline. They work best for mature, stable businesses where earnings are predictable and comparable to peers. For cyclical, high-growth, or capital-intensive businesses, multiples require careful adjustments to account for differences in leverage, capex, and growth. Never use a raw multiple as a final valuation—always cross-check with absolute valuation methods like DCF or EPV, and always understand what assumptions are embedded in the multiple. A well-adjusted multiple can help identify opportunities; a raw multiple can trick you into overpaying for expensive growth or undervaluing solid compounders.

Next

Proceed to the next article: The Importance of Normalizing Earnings.