Cyclicals
Take a cyclical business and locate it in its cycle, normalize its earnings, run the P/E inversion, read the top/bottom tells, and end in a disposition framed as thinking. Cyclicals break naive P/E thinking - the job is to value them on mid-cycle (normalized) earnings and judge where in the cycle they sit, never to trust spot earnings or a spot multiple. This is a thinking tool, not financial advice; never tell the user to buy or sell, surface the reasoning and a disposition.
The playbook (what counts as cyclical, the tells, the normalized-earnings method, Lynch's framing) lives in
next to this file. Read it at the start of every run; it is a living list the user grows.
When to use this: ,
, and
are independent tools. Reach for this one when the business is cyclical and the cycle itself is the crux.
Workflow
1. Confirm it is actually a cyclical
Apply the earnings-line test: a growth business's earnings slope steadily up; a cyclical's wobble up and down. Does it sell expensive or deferrable goods (postponable when cash is short), as opposed to staples? If it is not cyclical, say so plainly and point the user to
or
instead.
2. Locate the cycle position - with evidence
Where in the cycle is it, and what is the evidence: demand trend, capacity utilization, order book/backlog, inventories, pricing power, margins vs the cycle's own history, capex, balance sheet/dividend, and sentiment. Lynch's buy zone: economy weakest, earnings lowest, sentiment bleakest.
3. Normalize the earnings
Estimate mid-cycle EPS across a full peak-to-trough cycle (or a normal margin on normal volume). State whether current earnings sit near peak or trough. The airline lesson: "$15 a share one year, minus $15 the next - a $15-a-share operation no matter what." Never extrapolate peak or trough earnings as permanent.
4. Run the P/E inversion (mandatory - the central trap)
A low P/E on peak earnings is expensive and late (the multiple is low because the market knows earnings will fall) - the classic value trap. A high or N/A P/E on trough earnings can be cheap and early. Value on the normalized number from step 3, not on the reported one.
5. Read the tells
From
, name which
top tells and which
bottom tells are firing, each tied to a concrete fact about this business.
6. Disposition
Synthesize: where in the cycle, the normalized earnings power, the dominant tells, what would confirm or break the read, and the asymmetry. Frame it as thinking - e.g. "early-cycle, sentiment bleak, normalized EPS ~$X, capacity being shut" - never a buy/sell directive.
Rules
- Not financial advice. Reasoning and a disposition, never a directive to trade.
- Normalize, never trust spot. Peak earnings at a low multiple is the trap; trough losses are not by themselves a reason to avoid.
- The P/E inversion (step 4) is mandatory.
- Specific, not generic. Tie every tell to a concrete fact and quote it; no textbook recitation.
- Sentiment is a signal. "This one is different" at the bottom and universal enthusiasm at the top both count - but distinguish a normal cycle from a true structural break.
- is the source of truth for the tells and sectors - keep it growing there, not here.