Skip to main content

Traps in Valuing Cyclical Stocks

The graveyard of investor losses includes a familiar epitaph: "I bought this stock at 8x earnings—what a bargain." Years later, the investor learned the hard way that those "cheap" valuations masked companies at peak cycle earnings, which collapsed 50-70% when the economic cycle turned. Cyclical stocks—companies whose earnings and cash flows swing dramatically with economic conditions, commodity prices, or other external cycles—are among the most dangerous to value using conventional multiples. The multiples themselves are often reversed: cheapest when the business is best, most expensive when the outlook is worst. This chapter teaches you how cyclical companies actually work and how to avoid the valuation traps that claim otherwise disciplined investors.

Quick Definition​

Cyclical stocks are companies whose earnings and returns swing substantially with economic, commodity, or other external cycles. When the cycle peaks, earnings hit highs and multiples compress. When the cycle troughs, earnings collapse and multiples expand. Successful valuation of cyclicals requires through-the-cycle analysis: normalizing earnings and valuations to a mid-cycle level, not peak or trough. The trap is valuing based on current earnings without anchoring to normalized earnings, leading to massive mispricings.

Key Takeaways​

  • Cyclical multiples are inverted: highest (best) at cycle troughs, lowest (worst) at cycle peaks—opposite of what intuition suggests
  • Trailing P/E is nearly worthless for cyclicals; peak earnings appear cheap, trough earnings appear expensive
  • Through-the-cycle (normalized) earnings are essential; average earnings over full 5-10 year cycles, not current annual earnings
  • EV/EBITDA on normalized EBITDA works better than P/E for cyclicals, but normalization is the real key, not the metric
  • Leverage amplifies cyclicality; a $1 billion swing in EBITDA becomes a $5+ billion swing in equity value if the company is levered 40-50%
  • Cyclical companies are most dangerous to value at peaks (highest confidence often coincides with lowest intrinsic value)
  • Commodity exposure creates second-order cyclicality; oil companies expose you to both oil cycle and credit cycle
  • Valuation multiples that appear "cheap" at cycle peaks are trap signals, not bargain signals

The Inverted Multiple Problem​

The most dangerous trap in cyclical valuation is the inverted multiple illusion. Consider a cyclical business:

Scenario: Homebuilder at Cycle Peak (2021-2022)

  • Earnings: $50 per share (all-time high, driven by housing shortage)
  • Stock price: $300
  • P/E: 6x

That's extraordinarily cheap! Investors pile in. "Only 6x earnings?" "Homebuilders are trading at 1/3 the P/E of the S&P 500!" The stock feels like a bargain. But the market is not mispricing; it's warning. Peak housing starts are coming; supply will normalize; prices will fall.

Three years later, at Cycle Trough (2024-2025)

  • Earnings: $10 per share (housing collapse, oversupply)
  • Stock price: $100
  • P/E: 10x

Now the stock is expensive! Investors avoid. But the market is actually signaling relative value here—earnings will recover; this is the bottom.

The investor who bought at 6x and sold at 10x lost 67% ($300 → $100) despite the multiple expanding. The low multiple was a trap.

Why This Inversion Happens​

At peak cycle, sentiment is best, earnings are highest, and investors are bullish. The market discounts the cycle turning and prices in normalized/normalized recovery. The multiple compresses.

At trough cycle, sentiment is worst, earnings are lowest, and investors are bearish. The market prices the cycle to stay down longer and discounts recovery. The multiple expands.

This is rational. The expected return on a cyclical stock at the trough is higher than at the peak—but only if you hold through the recovery. The multiple must compensate for the high risk and poor near-term outlook.

The Through-the-Cycle Framework​

The solution is through-the-cycle (TTC) analysis: normalize earnings to a mid-cycle level and value based on that, not current earnings.

Step 1: Identify the Cycle​

First, understand what drives your cyclical company's earnings. Is it:

  • Economic cycle: Construction, automotive, retail, capital equipment
  • Commodity cycle: Oil, metals, agriculture
  • Credit cycle: Banks, consumer finance, insurance
  • Real estate cycle: Homebuilders, REITs, commercial real estate

For each, identify historical peaks and troughs.

Example: Cyclical Airline Industry

  • Peak: Post-demand surge, before fuel spikes or recession (e.g., 2022)
  • Trough: During recession or fuel crisis (e.g., 2008-2009, 2020)
  • Cycle length: ~7-10 years from trough to trough

Step 2: Normalize Earnings​

Average earnings over a full cycle (or multiple cycles), not just trailing twelve months. A simple approach:

Normalized Earnings = Average earnings over past 5-10 years, excluding extreme years

Or more sophisticated:

Normalized Earnings = Peak Cycle Earnings × (Cycle Position Factor)

Where the cycle position factor adjusts based on where in the cycle you are now.

Example: Oil Company Cycle Normalization

  • Recent peak (2008): $15 EPS at $147/barrel oil
  • Recent trough (2016): $2 EPS at $43/barrel oil
  • Mid-cycle (typical): $8 EPS at $75/barrel oil

If oil today is $80/barrel (slightly above mid-cycle), normalize EPS to roughly $8.50, not current earnings if they're distorted by recent price movements.

Step 3: Calculate Through-Cycle Multiple​

Apply a multiple to normalized earnings, not current earnings.

Cyclical Valuation = Normalized Earnings × Through-Cycle Multiple

What multiple to use? Research the company's historical range. Over 10+ years, a homebuilder might trade at 6-12x normalized earnings. An oil company might trade at 7-11x normalized EBITDA.

Use the median or slightly below (margin of safety). Don't use peak multiple; that's the trap.

Example: Homebuilder at Trough (2024)

  • Trailing EPS: $8 (trough)
  • Normalized EPS: $20 (cycle average)
  • Through-cycle multiple: 10x (conservative relative to historical 8-12x)
  • Fair value: $200

The stock trading at $100 (12.5x trailing, 5x normalized) is undervalued, not overvalued. The low trailing multiple is the trap signal, not the opportunity signal.

Flowchart​

Leverage Amplifies Cyclicality​

Cyclical companies often use significant leverage, which amplifies equity volatility. This is where multiples become truly treacherous.

Example: Leveraged Homebuilder

  • EBITDA at peak: $1 billion
  • Debt: $500 million
  • Equity value: $1.5 billion (at 3x EV/EBITDA)
  • Shares outstanding: 50 million
  • EPS: $20 per share
  • EBITDA at trough: $400 million (60% decline)
  • Debt: $500 million (unchanged; bond covenants prevent paydown)
  • Equity value: $400 million (at 2x EV/EBITDA to reflect distress)
  • Shares outstanding: still 50 million (assuming no dilution)
  • EPS: -$2 per share (negative because interest on debt exceeds EBITDA)

Peak-to-trough EPS declined from $20 to -$2 (110% decline), even though EBITDA declined only 60%. Leverage caused the equity cliff.

This is the equity multiple trap: when leverage is high, trailing P/E at the trough can be worthless (or show negative earnings). Meanwhile, at the peak, a 10x P/E looks reasonable until the next year when EPS collapses 80%.

How to address leverage in cyclicals:

  1. Use EV/EBITDA instead of P/E (removes the leverage distortion from the income statement)
  2. Normalize EBITDA, not earnings
  3. Check debt covenants and refinancing risk—if leverage is unsustainable through the next trough, equity value goes to zero
  4. Stress test equity value at 30-50% EBITDA declines; if leverage causes covenant breaches, the equity is riskier

Commodity-Linked Cycles​

Commodity companies (oil, metals, agriculture) face the complexity that their cycle is partially external—commodity prices—which they cannot control.

Oil Cycle Case Study​

Oil companies face overlapping cycles:

  1. Commodity price cycle: Oil swings $40-120 range; long-term mean ~$70-80
  2. Production cycle: Once you drill a well, it depletes; you need replacement drilling and exploration
  3. Credit cycle: During downturns, banks tighten credit; E&P companies can't finance wells
  4. Geopolitical cycle: Supply disruptions (wars, sanctions) cause price spikes unrelated to supply/demand

Valuing an oil company requires making assumptions about long-term oil price. Two analysts might differ:

Bullish analyst: Long-term oil price = $90/barrel, normalizes FCF at high level, values company at $60/share.

Bearish analyst: Long-term oil price = $65/barrel, normalizes FCF at lower level, values company at $30/share.

Both might be right about what $90 or $65 oil looks like; they're disagreeing on the probability of each scenario or the long-term equilibrium price.

Solution: Use scenario analysis.

  • Bull case (40% probability): $90 oil → $60/share valuation
  • Base case (50% probability): $75 oil → $40/share valuation
  • Bear case (10% probability): $55 oil → $20/share valuation
  • Probability-weighted fair value = 0.4×$60 + 0.5×$40 + 0.1×$20 = $44/share

This approach acknowledges uncertainty while avoiding the trap of point-estimate pricing.

Real-World Examples​

General Motors and the Auto Cycle (2008-2024)​

In 2007, GM traded at 4x earnings ($30 stock, ~$5 EPS) and looked cheap. The auto industry was at peak cycle: high volumes, high margins, peak production. The multiple compression signaled peak, not value.

In 2009, at the trough, GM earnings went negative and the company filed bankruptcy. The stock crashed 80%. Investors who bought at "cheap" 4x in 2007 lost everything.

By 2015, as the cycle recovered, GM rebuilt and earned $5-6/share again. The stock traded at $35-40, or 6-8x earnings. Now the multiple was appropriate—investors were paying mid-cycle multiples on recovered earnings.

Lesson: A 4x multiple on peak cycle earnings is not cheap; it's a trap.

Banks and the Credit Cycle (2006-2024)​

In 2006-2007, banks traded at 10-12x P/E (moderate multiples). But earnings were at peak cycle: low loan losses, high credit growth, peak origination volumes. The multiple was low despite peak earnings because the market sensed cycle risk.

In 2008-2009, banks traded at 15-20x P/E on collapsing earnings. The multiples expanded because recoveries were expected, but many caught by cycle peak losses never recovered.

By 2010-2015, as credit recovered, banks traded at 8-11x normalized earnings. The multiple reflected mid-cycle expectations.

The trap: Buying banks at 12x earnings in 2007, feeling smart, then losing 80% as credit cycle reversed.

Vale (Mining Cycle), 2008-2024​

Vale, a major iron ore and nickel producer, exemplifies commodity leverage:

2007-2008 Peak:

  • Iron ore price: $170/ton
  • EPS: $8
  • Stock: $80
  • P/E: 10x

2008-2009 Trough:

  • Iron ore price: $40/ton
  • EPS: -$2 (mining costs exceeded price)
  • Stock: $10
  • P/E: negative

Normalized (cycle average):

  • Iron ore price: $80/ton (long-term midpoint)
  • Normalized EPS: $3-4
  • Fair value: $30-40 (at 8-10x multiple)

An investor buying Vale at $80 (thinking 10x is cheap) would suffer 87% loss at the trough. An investor waiting to buy at $10 (when sentiment was worst) could earn 300%+ over the next cycle.

Common Mistakes​

1. Buying Cyclicals at Peak Earnings Based on Low Trailing P/E

The lowest trailing P/E often corresponds to peak earnings, not cheap valuation. A company earning $10/share at peak, trading at 6x ($60), is more dangerous than one earning $5/share at trough, trading at 8x ($40).

2. Assuming the Cycle Will Be "Different This Time"

During booms, executives and investors convince themselves the cycle is broken. Housing will never correct. Oil will stay above $100. Credit growth will be unlimited. It never works. Cycles always turn. Assume the cycle continues.

3. Ignoring Covenant Risk in Levered Cyclicals

A company with $1B EBITDA and $600M debt is fine if EBITDA stays above $700M (ratio of 1.17x). But if a 50% EBITDA decline occurs, EBITDA drops to $500M, debt remains, and covenant breach forces restructuring. Equity is wiped. Check debt covenants and covenant breach scenarios.

4. Using Single Point-Estimate Oil/Commodity Price

Oil prices are volatile and unpredictable. Don't value oil companies at a specific price ($75 oil); use scenarios. If the company earns well at $60 oil, it's safer than one that earns well only at $100 oil.

5. Confusing Recovery to Prior Peak with Indefinite Growth

When a cyclical recovers from trough, emotions run high. But recovery to prior peak is not growth beyond that. A steel company earning $8 at prior peak, then recovering to $8 from a $2 trough, is not growing. It's normalizing. Don't use growth multiples for normalization stories.

6. Forgetting That Multiples Normalize More Slowly Than Earnings

During recovery, earnings can double in a year. But the multiple typically compresses from trough (expansion phase) back to normalized levels. If a company at trough trades at 8x (unusual multiple for normal earnings) and earnings double, the stock may trade at 8x earnings instead of 12x. Investors get earnings recovery but multiple compression, reducing returns.

FAQ​

Q: How do I know what the "normalized" earnings level should be?

Average earnings over a full historical cycle (5-10+ years). If that's not available, estimate mid-cycle by looking at years near the middle. For commodity companies, use historical commodity prices adjusted to long-term equilibrium. Financial companies: use years when loan losses were normal (not artificially low or high).

Q: Should I ever buy a cyclical stock at peak cycle?

Rarely, and only with high margin of safety (>30% discount to normalized value). Even then, you're betting on the cycle not turning, which is a poor bet. Better to wait for cycle weakness to build a position.

Q: How do I determine where in the cycle we are now?

Compare current metrics to historical range: earnings, margins, capacity utilization, growth rates. If a company's growth is decelerating and margins are at highs, you're likely late in the cycle. If earnings are depressed and sentiment is worst, you're likely at the trough.

Q: Can I use DCF for cyclicals?

Yes, but be careful. Project through-cycle cash flows (not peak-cycle or trough-cycle assumptions). Use a mid-cycle ROIC and growth. If you use peak-cycle assumptions, your DCF will overvalue; trough-cycle assumptions will undervalue. Many DCF mistakes come from embedding current cycle position into perpetual forecasts.

Q: Is leverage always bad for cyclicals?

Not always. Some leverage is manageable if debt is taken during booms and paid down during booms (before the trough hits). But many cyclical companies increase debt during booms to fund buybacks or acquisitions, leaving them vulnerable. Check if leverage is rising or falling through the cycle.

Q: How do I value a cyclical in the middle of the cycle?

Use through-cycle multiples on normalized earnings. If the company's current earnings are $6 and normalized earnings are $8, value at $8 × median multiple, not $6 × elevated peak multiple.

  • Sector-Specific Multiples — Learn how to apply appropriate multiples to energy, materials, and other cyclical sectors.
  • DCF Valuation Deep Dive — Master discounted cash flow analysis for cyclicals, including through-cycle cash flow assumptions.
  • The Problem with Multiples — Understand when multiple-based valuation fails most catastrophically and why supplementary DCF is essential.
  • Managing Cyclical Risk in Portfolios — Learn portfolio-level strategies for managing cyclical stock positions and reducing drawdown risk.

Summary​

Cyclical stocks are among the most dangerous to value because the signals that appear most attractive are often traps. The lowest trailing P/E often appears at peak cycle earnings; the highest appears at trough earnings. Investors who mechanically buy cheap multiples get caught in the cycle turn and suffer catastrophic losses. The solution is through-the-cycle analysis: normalizing earnings to a mid-cycle level and valuing based on normalized metrics, not current performance. For leverage-laden cyclicals, EV/EBITDA on normalized EBITDA works better than P/E. For commodity companies, scenario-weighted valuations acknowledge that long-term prices are unknowable. The key discipline is resisting the emotional pull of peak-cycle optimism (when multiples are lowest and risk is highest) and building positions at cycle troughs (when multiples are highest and risk is lowest). This contrarian positioning is uncomfortable but profitable—the opposite of what unprepared investors do.

Next​

Continue to The Problem with Multiples to learn when multiple-based valuation breaks down entirely and why pairing relative valuation with DCF analysis is essential for avoiding costly errors.