Cyclical and Defensive Sectors: How They Behave Through a Cycle

A sector is cyclical when its customers can put the purchase off, and defensive when they cannot. Postponability drives everything else, giving revenue that swings wider than the economy against revenue that barely moves at all. Defensive is not a word about safety, and both labels describe a tendency rather than a rule.

Underneath that sits something simpler than the vocabulary suggests. An economy does not slow down evenly. When confidence drops, some spending stops almost at once and other spending does not stop at all, and the difference between the two is not about how big the purchase is or how important it feels. The difference is about whether the person paying can wait. A workshop that was going to replace a delivery cart this year can run the old one for another eighteen months. The same workshop cannot decide to stop buying soap for the washroom.

Hold that everyday pair in mind. The cart is a purchase with a date attached, and the date can be moved. The soap is a purchase with a rhythm attached, and the rhythm cannot be moved without simply going without. Nothing about the cart is frivolous and nothing about the soap is clever. The only property that separates them is whether the buyer has the option to wait, and that single property is what the two labels name.

Every figure below is worked on the Republic of Sankhya, an invented country used across these notes, and on two sectors invented on top of it. The path Sankhya's output follows across years 4 to 8 is fixed elsewhere and used here unchanged: output up 6.50 per cent, then down 1.00 per cent, then down 2.50 per cent, then up 5.00 per cent. Two sectors sit inside that path, and the arithmetic below ties them back to it.

What these two words do not establish

The vocabulary of cyclical and defensive walks straight into a conclusion nobody has earned. Learning that one sort of revenue holds up in a bad year and another does not is a statement about how businesses behave. A statement about behaviour is not, by itself, a statement about what belongs in anybody's holdings, and the distance between the two is much larger than it looks.

What a sector's revenue does through a cycle, and why the mechanism produces that behaviour, is one question. Which of the two kinds is preferable, and whether a phase of the cycle calls for a position, are different questions. Working out what to do with a view about the economy involves the price already being asked, the alternatives, a time horizon, a tax position and a tolerance for being wrong. Deciding what belongs in a set of holdings is taught under portfolio management and asset allocation, a separate subject with its own reasoning.

What single question decides whether a sector is cyclical or defensive?

The question is put to the customer, not to the product: can this purchase be put off for a year without much cost to the person making it? The question is the whole test, and everything that follows is a consequence of the answer.

Follow what happens when the answer is yes. Delay is free and being wrong about the future is not, so a household or a business that can delay a purchase will delay it as soon as the future looks less certain. The purchase does not disappear. It moves. So the sector selling that purchase sees its order book thin out quickly when confidence turns, and thin out by more than the economy as a whole is shrinking. The economy is not delaying everything at once. This sector's customers very nearly are. Then, when confidence returns, the sector does not simply resume at the old rate. The carts that were not replaced in years 6 and 7 are still not replaced, and the ones due in year 8 are due as well, so two years of buying arrive inside one year.

Now the answer no. A purchase that cannot be put off is one where waiting costs the buyer something immediately: the cooking gas runs out, the medicine is taken daily, the electricity bill arrives whatever the news is doing. The buyer facing a worse year buys a cheaper version, or a smaller quantity, or the same thing from a different seller. The buyer very rarely stops. So the sector selling it sees revenue that barely notices the contractionThe phase of a cycle in which an economy's output is falling rather than rising. Dating the start and the end of a phase is a separate subject with its own definitions. going on around it.

The two words are names for those two answers, and they carry no information the question did not already carry. That matters more than it sounds. A reader who memorises a list of sector names and their labels has learned the output of the test without the test, and the first sector that does not appear on the list, or that appears on it wrongly, will be classified by guesswork. A reader who carries the question can classify anything, including something invented last year.

How does a cyclical sector actually move through the phases?

A cyclical sector moves in the same direction as the economy and goes further, in both directions, every time. Sankhya's coachwork sector, an invented sector inside an invented country, is the one to watch. Coachwork builds carts, trailers, workshop machinery and the bodies that go on delivery vehicles, all of it bought by businesses and households who decide each year whether this is the year.

Here is the sector against the economy that contains it. In year 5 Sankhya's output rose 6.50 per cent and the coachwork sector's output rose 16.00 per cent, from Rs 1,25,000 crore to Rs 1,45,000 crore. In year 6 output fell 1.00 per cent and coachwork fell 8.97 per cent, to Rs 1,32,000 crore. In year 7 output fell 2.50 per cent and coachwork fell 14.02 per cent, to Rs 1,13,500 crore. In year 8 output rose 5.00 per cent and coachwork rose 18.06 per cent, to Rs 1,34,000 crore. Four years, four occasions on which the sector went the same way as the economy and travelled further doing it.

Matching the economy would already be unremarkable, so the going further is the part worth explaining. Two things stack on top of each other. The first is the delay itself. In a year when the economy shrinks 2.50 per cent, the coachwork sector is not losing 2.50 per cent of its buyers. The sector is losing every buyer whose replacement was discretionary this year, and that is most of them. The second is what happens to those postponed purchases. The postponed purchases do not evaporate. They queue. A cart not replaced in year 6 and a cart not replaced in year 7 both still need replacing, so year 8 carries its own demand plus two years of arrears, and the sector rises 18.06 per cent into an economy rising 5.00 per cent.

Arrears are why a cyclical sector's own record looks more dramatic than the economy's record ever does, and why the drama is arithmetic rather than sentiment. Notice also the sector's place in the economy. Coachwork was 7.35 per cent of Sankhya's output in year 5 and 5.96 per cent of it in year 7. Coachwork did not stop existing. It shrank while nearly everything around it shrank less.

How does a defensive sector move through the same phases?

Much less, and that is the entire finding. Sankhya's everyday household goods sector, also invented, sells the soap, the cooking oil, the tooth powder and the cleaning things that leave a household every week whatever the year is doing.

Set it against the same four years. Output rose 6.50 per cent in year 5 and the sector rose 3.00 per cent, from Rs 2,00,000 crore to Rs 2,06,000 crore. Output fell 1.00 per cent in year 6 and the sector rose 1.46 per cent, to Rs 2,09,000 crore. Output fell 2.50 per cent in year 7 and the sector rose 0.48 per cent, to Rs 2,10,000 crore. Output rose 5.00 per cent in year 8 and the sector rose 1.90 per cent, to Rs 2,14,000 crore. Two bad years for Sankhya and the sector did not have a bad year at all.

The good years settle what to make of that. In year 5, the strongest year of the whole path, this sector grew 3.00 per cent against an economy growing 6.50 per cent, so it fell behind by 3.50 percentage points. In year 8 it grew 1.90 per cent against 5.00 per cent, falling behind by 3.10 points. A household that will not stop buying soap in a bad year will not start buying four times as much in a good one, so the property that stops a defensive sector falling in a contraction is precisely the property that stops it surging in an expansion.

The vocabulary invites a reader to hear stability as an advantage that comes free. It does not. Stability is symmetric. A sector whose customers cannot postpone has traded the upside away in exchange for the downside, and the trade was not negotiated by anybody. The trade is simply what selling a weekly necessity does to a revenue line. One more thing is worth noticing. The household goods sector's share of Sankhya's output rose from 10.44 per cent in year 5 to 11.03 per cent in year 7. The sector did not win anything to get there. The denominator fell.

One last way of seeing the pair, and the one most likely to change how the whole comparison reads. Rebased to 100 in year 4, all three series can be followed to where each one ends up. Sankhya's output finishes year 8 at 107.94. The coachwork sector finishes at 107.20. The household goods sector finishes at 107.00. Over the full stretch the three arrive within a point of each other, having taken completely different routes to get there, and the coachwork line went as high as 116.00 and as low as 90.80 on the way while the household goods line never once left the range between 100 and 107.

Why is a defensive sector not the same thing as a safe one?

Because the word describes one property and one property only: how a sector's revenue moves as the economy moves. Defensive is a statement about demand. Defensive is not a statement about the business, about what the business paid for its inputs, about how much it borrowed, about who else is selling the same thing, about who is running it, or about the price somebody paid to own a share of it. Every one of those can go wrong in a year when demand is behaving perfectly.

Confusing stable demand with low risk is the commonest error this vocabulary produces, and it is an error of substitution: a single observed tendency gets used in place of an assessment nobody has actually made. The tendency is real. The tendency is also narrow. Reading it as a verdict on risk is like reading a shop's steady footfall as proof that the shop is solvent, when footfall says nothing about the rent, the loan or the wholesaler who just raised his prices.

Show it rather than assert it. Sthira Household Goods is an invented business inside Sankhya's invented household goods sector, with revenue running at 2.00 per cent of the sector. In year 4 it took Rs 4,000 crore of revenue, paid Rs 2,600 crore for its inputs and Rs 900 crore in other running costs, made Rs 500 crore of operating profit, paid Rs 300 crore of interest on its borrowings, and reported profit before taxWhat is left of revenue after every operating cost and after interest on borrowings, but before any tax is charged on it. How each of those lines is measured and presented is a matter for accounting, and is covered there. of Rs 200 crore.

Then year 5 arrived, the strongest year of the whole cycle. Revenue did exactly what the label says it should: up 3.00 per cent to Rs 4,120 crore, in line with the sector. One imported input Sthira depends on also became dearer, adding Rs 600 crore to an input bill that would otherwise have been Rs 2,678 crore, a rise of 22.40 per cent on that bill. Other running costs stayed at Rs 900 crore and the interest stayed at Rs 300 crore. Operating profit came to Rs 4,120 crore less Rs 3,278 crore less Rs 900 crore, or minus Rs 58 crore, and profit before tax came to minus Rs 358 crore.

Now hold the two facts together. In the best year Sankhya had, a business inside its steadiest sector lost money. And in year 7, the worst year Sankhya had, the input price had normalised and revenue had gone on quietly rising, so the same business reported profit before tax of Rs 270 crore, comfortably positive. The cycle did not cause the loss and the cycle did not cure it. An input price and a borrowing load did both.

A second reading of the same episode explains why the trap catches careful people. A business with heavy fixed costsCosts that stay much the same whether the business sells a lot or a little, such as rent, salaries of permanent staff and the upkeep of a plant. Separating them from the costs that move with volume is a matter for cost accounting. and borrowings has high operating leverageThe degree to which a business's costs stay the same whatever it sells, so that a small change in revenue produces a much larger change in profit. Measuring it is covered under financial analysis., which means a small move in revenue produces a large move in profit. Steady revenue does not protect against that. Sthira's revenue moved 3.00 per cent and its profit before tax moved from plus Rs 200 crore to minus Rs 358 crore, a swing of Rs 558 crore. The move was on the cost side, and almost nothing below it could flex.

Can a sector change which of the two it is?

Yes, and the reason it can is the reason the mechanism matters more than the list. The label follows the customer's behaviour. Change the behaviour and the label changes with it, without a single thing about the product needing to be different.

Sankhya has an invented household connection service that charges a small monthly amount. When it was new it was a convenience, and a convenience is exactly the kind of purchase a household stops when money is tight. Sankhya's household survey, also invented, put it at 22 households in every 100 stopping the service inside a bad year. By year 8, wages arrive through it, school records sit on it and entitlements are claimed with it, and the figure was 3 in every 100. The service is the same service. The monthly charge is roughly the same charge. Stopping the service now costs the household something immediate, so the service is no longer postponable, and a revenue line that used to fall in a downturn now barely notices one.

The label therefore describes a tendency somebody has observed rather than a property the sector holds. A tendency can be true for a decade and then quietly stop being true. The reverse happens as well. A service that everyone treated as unavoidable becomes optional once a cheaper substitute appears or a habit changes, and a revenue line that used to hold through bad years starts sagging in them. Nobody announces either transition. The change shows up in the numbers first and gets renamed afterwards.

Do the two behaviours actually fit inside one economy?

The two behaviours have to fit, and making them fit is what turns the adjectives into measured numbers. Sankhya's output is a published path. The coachwork sector and the household goods sector both sit inside it. So at every year the two of them, plus everything else Sankhya produces, must add up to the figure Sankhya published, and if they do not, one of the three is wrong.

Where these numbers come from. Sankhya's output path is the invented path used throughout these notes, and the two sectors below were invented on top of it. The two were chosen so that the pair of them, added to everything else Sankhya produces, rebuild Sankhya's published output in every single year. Sector levels are exact whole Rs crore. Every growth rate, share and ratio here was divided out of those levels and then rounded to two decimals, so a rate read back off the table will not always rebuild a level to the last rupee. Producer figures are stated to the nearest crore.

The last two rows of that table are where the adjectives stop and the measurement starts. Take each series and find the distance between its highest growth rate and its lowest one across the four years. Sankhya's output ran from plus 6.50 per cent down to minus 2.50 per cent, a range of 9.00 percentage points. The household goods sector ran from plus 3.00 to plus 0.48, a range of 2.52 points. The coachwork sector ran from plus 18.06 down to minus 14.02, a range of 32.08 points. Divided by the economy's own 9.00, that is 3.56 times for the coachwork sector and 0.28 times for the household goods sector. Wider and narrower mean exactly that, once somebody insists on a number.

Now take one year apart rather than the whole stretch. One year shows something the ratios hide. Between year 6 and year 7 Sankhya's output fell from Rs 19,52,690 crore to Rs 19,03,873 crore, a fall of Rs 48,817 crore. The coachwork sector fell Rs 18,500 crore of that. The household goods sector actually rose Rs 1,000 crore, pushing the other way. Everything else fell Rs 31,317 crore. The three changes add to minus Rs 48,817 crore and close the year exactly.

Turn those same three numbers into shares and the point lands harder. In year 6 the coachwork sector was 6.76 per cent of Sankhya's output. Of the fall that happened next, it supplied 37.90 per cent. A 37.90 per cent share of the fall is 5.61 times its share of the economy, and that is what it actually means to say a sector amplifies a downturn: a small part of the economy delivering a large part of the movement. The household goods sector was 10.70 per cent of output and supplied minus 2.05 per cent of the fall, because it grew while the economy shrank, and everything else supplied the remaining 64.15 per cent.

What does a lender actually do with this distinction?

Something more specific than applying the label, and watching a lender do it is the quickest way to see why the mechanism beats the vocabulary. A credit officer assessing a Sankhya business does not ask which sector it is filed under. The question is put to the borrower's own revenue: how much of last year's sales could the borrower's customers have chosen not to buy, or bought a year later, at little cost to themselves?

Suppose the answer is half. Half is not a label. Half is a number the officer can work with, and it says that in a year when confidence drops, roughly half this borrower's revenue is exposed to a decision the customer gets to make, and the other half arrives more or less regardless. From there the officer can size the repayment against the half that arrives regardless rather than against the total, look at whether the borrower's working capitalThe money tied up in running the business day to day, mostly in stock held and in amounts customers have not yet paid, less the amounts the business has not yet paid its own suppliers. Sizing and managing it is a separate subject. cycle can survive half a year of thin orders, and set a covenantA condition written into a loan agreement that the borrower agrees to keep meeting, such as holding a ratio above an agreed level. Breaching it usually gives the lender rights it did not otherwise have. against a measure that will actually move if the postponable half disappears.

Two businesses filed under the same sector can have completely different answers to that question, so the number, not the sector label, is what tells the lender how the borrower will behave in a contraction. A workshop inside the coachwork sector that also holds a long maintenance contract with a large buyer has a floor under its revenue that its neighbour does not. The label would treat them identically. The question would not.

The same habit works in the other direction. A lender looking at a business whose sector is famously steady still asks what would happen if one input became 22 per cent dearer for a year. Steadiness does not cover an input price, and Sthira Household Goods is what the answer looks like when nobody asked.

The Republic of Sankhya, its coachwork sector, its everyday household goods sector, its household connection service, its household survey, Vahini Coachworks and Sthira Household Goods are invented.
Educational material. Not advice on any investment, tax, budget or market position.