The economic cycle wheel: how sector leadership rotates

What sector rotation actually means

Sector rotation is the observation that different parts of the stock market lead at different points in the economic cycle, in a repeating order. When growth is accelerating from a low base, money tends to favour the sectors whose earnings are most sensitive to that acceleration — banks, industrials, transport. When growth is contracting, it favours the sectors whose earnings barely notice — utilities, staples, health care — along with government bonds.

The business-cycle framing is not new or proprietary. The version most people have seen traces back to Sam Stovall’s sector-rotation work at S&P in the 1990s, and Fidelity has published a business-cycle sector framework along the same lines for years. What follows is that standard map, with the tickers checked and the caveats left in.

The five phases

Each phase is defined by what growth and inflation are doing, not by what the index is doing. Leadership is the symptom; the cycle is the cause.

1 · Recovery — early cycle

Growth is negative but improving. Central banks are still easy, the yield curve is steepening, and the sectors that make money on credit volume start to work first: banks (XLF, KRE, IYF) and the industrial and transport complex (XLI, XTN, VIS). Regional banks are the sharpest expression of the credit leg and the most volatile.

2 · Expansion — mid cycle

The longest phase, and the narrowest leadership in this framework: technology and high-beta growth (XLK, VGT, QQQ). Earnings are compounding, credit is available, and the market pays up for duration in equities the same way it does in bonds.

3 · Peak — late cycle

Growth is still positive but decelerating, and inflation is at its highest. Real assets lead: gold (GLD, IAU, GDX), broad commodities (DBC) and energy (XLE, VDE). This is the phase most often misread in real time, because the index can keep making highs while the internals have already turned.

4 · Slowdown — defensive

Earnings estimates start coming down. Energy often keeps working into the early part of this phase (XLE, VDE, IEO) while the defensive bid builds in utilities (XLU, VPU, IDU). The overlap is real and it is why phases 3 and 4 are the hardest pair to separate.

5 · Recession — contraction

Growth is negative. Rates fall, so duration works: Treasuries and aggregate bonds (IEF, TLT, AGG). In equities the bid is for earnings that do not depend on the cycle — health care (XLV, VHT, IXJ).

The distinction that actually matters

Five phases is more precision than the data usually supports. The split that survives contact with a live tape is the binary one underneath it: cyclical versus defensive.

• Cyclical — technology, financials, industrials, consumer discretionary, energy, materials, communication services. These lead when the market is willing to pay for growth.
• Defensive — consumer staples, utilities, health care, real estate. These lead when it is not.

You do not need to name the phase to use that. If defensives are top of the one-month relative-strength ranking, the market is expressing a late-cycle or recessionary view, whatever the GDP print eventually says.

How to tell where you are — without waiting for the data

The trap in this framework is treating it as a forecast. Economic data is revised and arrives late; the NBER dates recessions long after they end. Relative strength does not have that problem, because leadership is observable today.

The practical method is to rank the eleven sectors by their relative performance against the S&P over a rolling window — a month is the common choice — and read which side of the cyclical/defensive split is on top. That is a measurement, not a prediction, and it updates every session.

Where the wheel stops working

The order is the reliable part. Almost everything else about it is softer than the diagram suggests, and a page that sells you the diagram without saying so is doing you a disservice.

• Phases have no fixed length. The NBER dated the 2020 recession from February to April — two months, the shortest in the record. An expansion can run for a decade. Nothing on the wheel tells you which you are in.
• Phases overlap and occasionally skip. A policy shock can compress three phases into one quarter. In 2022 bonds and equities fell together, which the recession leg of this map does not anticipate at all.
• Index concentration distorts it. When a handful of mega-cap names drive most of the index return, “technology is leading” can describe five companies rather than a sector, and the cycle read is much weaker than it looks.
• It says nothing about entries. This is a context tool over weeks and months. It has no opinion about where price is right now, and using it as one is the most common way people lose money with it.

Using it without fooling yourself

Read leadership first and label the phase second — never the reverse. If you start from “we are late cycle” you will find the sectors that confirm it. If you start from the relative-strength ranking, you get a measurement that can disagree with you, which is the only kind worth having.

Then treat the result as one input among several. Sector leadership tells you the regime the market is currently pricing. It does not tell you what happens on Tuesday.