Sectors and cycles

A bank and a soap company do not respond to the same events. That sounds obvious and is routinely ignored by portfolios that hold six companies which all turn out to be the same bet.

Cyclical and defensive

The most useful division is not by industry but by how sensitive profits are to the economic cycle.

Cyclical businesses earn a lot in good times and little — or nothing — in bad ones. Cars, steel, cement, construction, travel, discretionary retail. When incomes are rising and credit is cheap, people buy; when they are not, purchases are postponed, and a postponed car is a sale that simply does not happen.

Defensive businesses sell things people buy regardless. Soap, medicines, electricity, staple food. Demand barely moves with the cycle, so profits are steadier — and rarely spectacular.

Neither is better. They fail in different weather, which is exactly why holding both is different from holding twice as much of one.

The trap in a cyclical's valuation

This is the part that catches careful people, and it inverts the rule they learned.

A cyclical company at the peak of its cycle shows record profits. A price-to-earnings ratio computed on those record earnings looks low — and the stock looks cheap precisely when it is most expensive, because those earnings are about to fall.

At the bottom of the cycle the opposite happens: profits collapse, the P/E on depressed earnings looks enormous or meaningless, and the stock may be at its most attractive.

So for cyclicals, a low P/E is a warning at least as often as an opportunity. The valuation chapters of the company-analysis subject deal with this properly; the point here is that the same number means opposite things depending on what kind of business produced it.

What drives the main sectors

Not a complete list, and enough to read a portfolio:

Banks and lenders — interest rates, credit growth, and how many borrowers stop paying. Their profits depend on other companies' health, which makes them a bet on the whole economy at once.

Technology services — client spending abroad, wage costs, and the currency. A weaker rupee flatters reported earnings for exporters.

Commodities — the price of the commodity, which they do not control. A steel company is, to a first approximation, a leveraged position on the steel price.

Consumer staples — input costs and how much price they can pass on. Steady, and rarely cheap.

Pharmaceuticals — regulation, approvals, patent timing, and the rules of whichever market they sell into.

Property and infrastructure — interest rates and government spending, with long project cycles and heavy debt.

Hidden concentration

The practical failure this chapter is about.

A portfolio of ten stocks feels diversified. If six are lenders, it is one bet expressed six ways — and when credit conditions turn, all six move together. Chapter 9 explains why correlated holdings do not diversify anything.

It happens by accident: you buy what has been doing well, and what has been doing well is usually one sector having a good run. Nobody decides to put 60% in banks. It accumulates.

The check is mechanical and takes five minutes: list your holdings, write the sector beside each, add up the weights. Most people are surprised once.

What this does not license

Knowing sectors respond to different drivers is not a method for predicting which will do well next. Everyone can see the same drivers, and they are already in prices.

What it gives you is a way to see what you are actually exposed to — which is a question about your portfolio, not a forecast about the world.

The point

Cyclicals swing with the economy and look cheapest at the top; defensives grind along and rarely look cheap. Knowing which you hold matters mainly because holding six of the same kind is one bet, however many names are on the list.