Which Sectors Lead at Each Stage of the Business Cycle?
Sector leadership tends to follow a rough order tied to where growth is accelerating or decelerating — cyclicals and financials early, broad participation mid-cycle, energy and materials late, defensives in a contraction. It is a tendency read from history, not a schedule, and it breaks whenever a single event (a rate shock, a supply crisis, a policy shift) overrides the pattern.
That's the whole idea in one sentence. The rest of this page is what the four stages actually are, why the order tends to hold, and — because this is the slower, easier-to-misuse cousin of the day-to-day rotation read — exactly where it stops being useful.
What is the business cycle, in market terms?
The economy expands and contracts. Not in a straight line and not on a fixed calendar — but growth accelerates for a while, decelerates, sometimes turns negative for a stretch, then accelerates again. Economists give the stretch a name (expansion, contraction) and a rough shape (a wave), and the market spends most of its energy trying to price where in that wave the economy currently sits, well before the official data confirms it.
That's the key distinction this page rests on: the business cycle is measured in growth rate, not price. A recession is a period where growth is negative, not a period where stocks are falling — the two overlap but don't move in lockstep, because the market prices the change in growth months before the change shows up in a GDP report. This is also why the plate above puts a brass dot at "peak growth rate" instead of at the price top: they are not the same point, and conflating them is the single most common mistake in this kind of reading.
What are the four stages, and what tends to lead in each?
Every stage boundary below is fuzzy in practice — nobody rings a bell. Treat the table as a rough map, not a set of dates.
| Stage | What's happening to growth | Groups that have tended to lead |
|---|---|---|
| Early | Growth troughs and starts to accelerate off the bottom | Cyclicals, financials, small caps — the names most beaten down and most sensitive to a turn |
| Mid | Growth is positive and broadly accelerating | Leadership widens; more sectors participate than in any other stage |
| Late | Growth is still positive but decelerating | Energy and materials — the stages where input costs and capacity constraints tend to bite |
| Contraction | Growth turns negative | Defensives — utilities, staples, health care — the groups least tied to discretionary spending |
Two things about that table matter more than the row contents. First, it is a tendency across many historical cycles, assembled after the fact — it is not a forecast, and no cycle has ever matched it exactly. Second, the transitions are usually diagnosed with a lag; by the time "early cycle" is obvious in the data, the market has often already spent weeks pricing it. That lag is the entire reason this read is worth having at all — the alternative is waiting for a confirmation that arrives after the move.
The long version of this is in the book — Become a Cyclitecnical Trader: the cycle ladder, the FLD, and the eight interactions, written out end to end. It's free. Send me a copy. We email it to you. No card, and you can unsubscribe any time.
Why does the order tend to hold at all?
Three mechanical reasons, none of them mysterious:
- Cost of capital changes first. Cyclicals and small caps carry more debt and more operating leverage, so they respond hardest — in both directions — to the same change in growth expectations and financing conditions. That's why they tend to lead off a trough and lag going into a contraction.
- Input costs lag output. Energy and materials names benefit from demand that has already been running for a while, once capacity gets tight — which is why they cluster late rather than early.
- Spending doesn't stop in a contraction — it changes shape. People still buy electricity, groceries, and medicine when they cut back on everything else. That's the entire logic behind defensives outperforming on a relative basis in a downturn — their revenue doesn't grow, it just falls less than everyone else's.
None of these are laws. They're incentive structures that have held often enough, across enough cycles, to be worth knowing — and they get overridden constantly, which is the next section.
How is this different from the day-to-day rotation read?
They are related but answer different questions, on different clocks, and mixing them up is the most common way this concept gets misused.
- The day-to-day rotation read answers: which sector is outperforming right now? It's measured in relative-strength ratios, checked daily or weekly, and it tells you what's happening this week regardless of why.
- This page answers: given where the economy sits in its growth cycle, which groups have tended to lead historically? It's measured in quarters and years, and it's a backdrop — the kind of context you hold loosely, not a signal you check each morning.
The honest way to use them together: the business-cycle read sets a prior — a rough expectation of what should lead if the historical pattern holds. The daily rotation read tells you what's actually leading. When the two agree, the daily read has a tailwind behind it. When they disagree, the daily read is more likely riding something else entirely — a single catalyst, a crowded trade unwinding, a story the macro backdrop doesn't explain — and that disagreement is itself useful information, not a contradiction to paper over.
What breaks this pattern?
Plenty, and it breaks often enough that treating the table above as reliable is the mistake, not the table itself:
- A single dominant catalyst. A rate shock, a war, a pandemic, a policy reversal — any of these can put an entire stage's "usual" leadership on hold or invert it outright. The 2020–2021 cycle broke the pattern almost completely: a policy-driven, compressed cycle where technology led through what should have been a defensive stretch.
- A cycle that never completes. Not every deceleration turns into a contraction. A "growth scare" can look exactly like the early stage of a late-cycle rotation and then simply resolve, with no downturn ever arriving.
- Sector composition changes. What counts as "technology" or "discretionary" shifts over decades. A sector index today holds different kinds of companies than the same sector held twenty years ago, which quietly changes how it behaves relative to the historical pattern.
- Everyone already knows the pattern. If enough capital tries to position ahead of the textbook rotation, the anticipatory buying can front-run the stage itself, compressing or scrambling the sequence.
None of this makes the framework useless. It makes it a prior to hold loosely, not a rule to trade against.
How do you actually use this without overtrading it?
With patience, and by keeping it in exactly one place in the process:
- Check it quarterly, not daily. The stage doesn't change week to week; checking it more often just invites overfitting noise to a slow-moving read.
- Use it to explain a rotation you're already seeing, not to predict one. If the daily rotation read shows a sector improving, ask whether the business-cycle backdrop supports that move or fights it. Support raises confidence a little. A fight is a reason for a smaller position, not a reason to ignore the daily read.
- Never trade the stage label itself. "We're probably mid-cycle" is not a trade. It's one input, sitting well behind price, volume, and the actual structure on the chart, in the order those inputs get weighed.
- Expect to be wrong about the stage sometimes. Stage calls get revised after the fact constantly, by professional economists with better data than a retail read will ever have. Treat any stage call — including this page's — as provisional.
Questions traders ask
How do I know what stage the economy is in right now?
Nobody knows in real time with certainty — stages are usually only obvious in hindsight, once enough data has accumulated to confirm them. The closest a trader gets without a research team is watching the same leadership groups this page describes: if cyclicals and small caps are leading off a period of weak growth, that's consistent with an early stage. If defensives are quietly outperforming while headlines are still bullish, that's consistent with a late or contraction stage. It's a read built from market behavior, not a substitute for one.
Does this apply the same way to every market cycle?
No. It's a tendency drawn from many past cycles, and every individual cycle differs in length, severity, and cause. Some cycles skip stages, compress them into weeks instead of quarters, or get overridden entirely by a single event. Treat the four-stage table as a historical average, not a template any one cycle is obligated to follow.
Is this the same thing as the cycle work on this site?
No, and the distinction matters. This page describes the economic business cycle — a macro concept measured in growth rates over quarters and years, generic across the industry. The Cyclitecnical method works on price cycles — a different, shorter-horizon idea measured directly from price action, with its own vocabulary (FLD, VTL, the nominal ladder). They can coexist as separate inputs, but one is not built from the other, and this page never uses the method's own vocabulary for that reason.
Where can I read the actual data instead of the table above?
The National Bureau of Economic Research (NBER) is the standard reference for U.S. business-cycle dating, published well after the fact and treated as the closest thing to ground truth. Individual sector performance across past cycles is publicly available from any major index provider. Nothing on this page requires taking the table on faith — the underlying history is public.
Keep reading
Get the Cycle Pass — from $297