Sector rotation — descriptive, not predictive

Watch enough cycles and a pattern seems to appear. Certain kinds of company tend to do well while the economy is climbing; different kinds tend to hold up better when it stumbles; and the leadership seems to hand off from one group to the next as the cycle turns. Practitioners gave this observed hand-off a name: sector rotation — the tendency for different sectors of the market to lead and lag at different phases of the economic cycle.

The temptation is immediate and powerful. If money rotates through sectors in a knowable order, surely you can get ahead of it — buy the sector about to lead, sell the one about to lag, and ride the cycle around the dial. This module exists to make one distinction with great care, because a lot of money has been lost blurring it: sector rotation is a description of the past, useful for understanding why sectors moved as they did. It is not a prediction engine you can trade, because it cannot tell you which phase you are in now, cannot promise this cycle will rhyme with the last, and describes a pattern so public that prices already anticipate it. A map of where the money has been is not a timetable for where it will go.

Start by granting what is true, because the pattern is not nonsense. There are honest reasons different sectors tend to lead at different points, and they follow straight from the last two modules.

Early in a recovery, when the cost of money is easing and confidence is returning, the sectors most sensitive to borrowing and to fresh demand — lenders, property, cars, capital goods — have the most to gain, because the thing holding them back (tight money, nervous buyers) is exactly what is lifting. Through a strong expansion, the makers of raw materials and heavy goods often shine as everyone builds and buys at once and their operating leverage does its amplifying work. As growth tires and prices of everything have run up, the steadier businesses — staples, everyday medicines, utilities — start to look relatively attractive because their earnings do not depend on the boom continuing. And in an outright downturn, those same steady-demand sectors tend to fall least, because people keep eating, washing and taking their medicines whatever the economy does.

Every one of those tendencies has a real mechanism behind it. That is precisely what makes rotation seductive — it is not a superstition, it is a pattern with plausible causes. But a real pattern in the past is a long way from a usable signal for the future, and the gap between them is where the danger lives. Three things break the bridge: you cannot reliably name the current phase, the next cycle need not repeat the last, and the pattern is so widely known that prices move ahead of it.

Here is the historical tendency, laid out plainly — and read the whole of this section as a museum exhibit, a record of what has often happened, never a forecast of what will.

Read the dial and the logic is familiar from the previous module: the rate-sensitive and high-leverage sectors tend to shine when the economy is climbing, and the steady-demand sectors hold up when it stumbles. That is not a coincidence — it is the same operating-leverage-and-cyclicality story, viewed from the sector shelf instead of the single company.

But now look at what the map does not contain. It does not contain a "you are here" marker, because no one can place one honestly in real time. It does not contain a guarantee that the arrows will point the same way next time, because each cycle has its own cause — a rate shock, a pandemic, an oil spike, a credit freeze — and its own starting valuations. And it does not contain a price, so it cannot tell you whether the "next leader" is cheap or already bid up by everyone else reading the same dial. The map is genuine. Everything you would need to trade it is missing from the map.

Take the same rotation map and use it two ways, so the difference is concrete. illustrative

The sound use — describing. You look back over a stretch where, it later became clear, the economy had slid into a slowdown. You notice that a basket of steady-demand names — a staples maker, a pharma company — held up while a basket of metals and capital-goods names fell hard. You reach for the rotation idea and it explains what you are seeing: defensive earnings held their value while cyclical earnings deflated, exactly as the framework would suggest. You have understood the past better. Nothing about this required you to have known the phase in advance; you are reading history with a helpful lens.

The unsound use — predicting. Now you take the same map and say: "the dial shows staples lead into a slowdown, I think we're heading into one, so I'll rotate into staples to get ahead of the money." Every word after "so" is a leap. You do not actually know a slowdown is coming — the phase is only nameable afterwards. You are assuming this cycle repeats the pattern. And the staples names may already be bid up by everyone else who read the dial and reached the same conclusion. You have converted a description into a bet, and stapled false confidence to it.

The tell is the direction of reading. Backward — "does the cycle help explain what already moved?" — is legitimate. Forward — "the cycle tells me what will move next" — is where the map stops being a map and starts being a fortune-teller. Dressed up as a rotation strategy, phase-timing is still forecasting, and forecasting the turn is closer to astrology than analysis.

Everything load-bearing about this module lives here.

It cannot locate the present. The rotation dial has phases but no "you are here." Since the phase is a hindsight label, any rotation trade rests on a guess about the current phase dressed as a fact. You can read the temperature around you, but you cannot pin your exact spot on the dial in real time.

It cannot promise the pattern repeats. Each cycle is driven by a different shock and begins from different valuations. The sector that led the last recovery may lag the next because it entered expensive, over-owned, or freshly disrupted. And the opposite reflex — assuming this cycle must rhyme exactly with history — is its own trap; every cycle is a little bit genuinely different, which is why neither "it always works" nor "it never works" is the right posture.

It cannot beat the crowd that shares it. A pattern in every textbook is priced by everyone reading those textbooks. If rotation were a reliable, easy edge, the act of chasing it would move prices ahead of the phase and erase the edge. Fame is a reason to expect anticipation, not advantage. And markets can wander from any tidy pattern for longer than you can hold a position built on it.

It cannot supply a price. The dial says which sector has tended to lead, never at what valuation. A "leader" bought expensive can lose you money even if the rotation call is dead right.

1. Reading the map forward. The core error, from which the rest flow: treating a backward description as a forward signal. A tendency in history is not an instruction for tomorrow.

2. Assuming you know the phase. Rotation trades smuggle in a confident phase call that the data cannot support in real time. If you cannot honestly name where you are, you cannot honestly rotate.

3. Ignoring valuation. The dial names sectors, not prices. Rotating into a "due to lead" sector that is already richly valued can lose money even when the macro read is correct.

4. Forgetting everyone has the same map. A famous pattern is a priced pattern. The edge you imagine you are seizing has usually been competed away by the crowd reading the identical dial.

5. Over-trading a slow idea. Even where rotation describes something real, it plays out over long, uncertain stretches. Churning your portfolio around a foggy phase call mostly generates costs and mistakes, not returns.