Industry Types: How Sectors Behave Differently

Industries differ in ways that decide what any business inside one can earn. Three questions separate them: what happens to demand when the economy turns, what it costs to be in the field at all, and who sets the price. The three answers together give the type. The type does not name a winner; it names the conditions every seller in that field works under.

The method runs four times below. Two of the seven types cannot be settled for either of the two businesses whose figures carry it, and the refusal to settle them comes first, worked through rather than skirted.

What is actually being classified here?

An industry type is a claim about a field, and never a claim about a business. Take that claim apart. Everything downstream rests on it. A field is everybody selling roughly the same thing to roughly the same sort of buyer. A type describes what all of those sellers face and what not one of them chose. A type is a statement about the ground, not about anybody standing on it.

The distinction matters immediately. Anjani Stationers Private Limited, an invented maker of school registers, buys paper in a field where the goods are interchangeable and the rate is the only thing anybody argues about. It sells hard-bound registers into school relationships that run for years. Paper bought and registers sold are two different fields, and this one business stands in both at the same time. Taking a type off the buying side and pinning it to the selling side gets two readings wrong in one sentence.

Any mall shows the same point without a single figure. A chemist, a jeweller, a uniform shop and a phone repair counter sit under one roof, on one floor, paying rent to the same landlord. The four are in four fields, not one. The question is never where a business sits but who buys from it, and what it took to open the shutter in the first place.

Seven types come out of three questions, so the types are answers rather than boxes. The distinction is not decoration. A treatment that hands over seven definitions has hidden the three questions, and the three questions are the only part of any of this that still works next year on a field nobody has described yet. The questions run in a fixed order:

The demand question goes first deliberately, and the reason is uncomfortable. Demand is the question a reader answers fastest and checks least. Somebody reads that a business makes school registers, decides in half a second that school registers sound postponable, and writes down a classification that no evidence anywhere supports. Refusing that move openly, and first, sets the standard for the three questions that follow.

What does demand do when the economy turns?

The demand question is the fastest to answer and the one most often answered on no evidence whatever, so it goes first and it ends in a refusal rather than in a label. Two named types come out of it.

Cyclical Industry

A cyclical industry is one whose demand rises and falls with the wider economy, so what the customers are willing to buy moves with their own conditions rather than with anything the sellers do.

Notice what kind of claim it is. Cyclical is a claim about where the field’s demand comes from, and about that alone. The claim says nothing whatever about whether the sellers inside the field make money, whether they are well run, or whether being in the field is a good idea. A field can be cyclical and full of businesses that earn steadily, and it can be cyclical and full of businesses that lose money in every year of the run.

Think of a lane of shops selling wedding jewellery. Nobody in that lane has cancelled a wedding. Plenty of households instead move one, or shrink one, or buy the smaller set, and the lane feels all of that at once because every shop in it faces the same households. The lane did not choose that. Selling into weddings does that to a lane.

How a field like that behaves once an economy actually turns, including the test that classifies one in the first place and what happens to the field on the way back up, is a full subject on its own, covered under how sectors move with the economy.

Defensive Industry

A defensive industry is the other answer to the same question: the customers keep buying at much the same rate whatever else is happening to them, so the field’s demand holds up when the wider economy turns down. Same kind of claim, opposite answer, and the pair exhausts the question.

A household running on one salary shows the shape of it in a week. When the salary gets tight, the sofa does not get replaced, the holiday does not get booked and the phone lasts another year. The rice, the milk, the soap and the bus fare carry on almost unchanged. Nobody in that household made a decision about sectors. The household made a decision about a sofa, and the sofa's field felt it while the rice's field did not.

One property is the hinge of the whole argument: defensive describes how revenue moves, and it says nothing at all about the cost base. The first question is about the money coming in. Money coming in is only half of what decides what a seller keeps. The other half is what has to be paid whether or not anything comes in at all, and the second question is about exactly that half.

How to Identify an Industry’s Cyclical Drivers

Identifying a driver is a procedure rather than a definition, and the procedure is the most useful thing the demand question yields. Three steps and one refusal.

All three now run on Anjani Stationers, and the third one is where the procedure stops.

Step one. The customers are schools and institutions. Not children, not parents, not households: the entity that writes the cheque is an institution with a purchasing decision and a budget year.

Step two. Their buying is attached to a school-term calendar, and this part is published rather than guessed. The order book refills each spring. The cash credit facilityA borrowing limit set against working capital. The ceiling is agreed once, the business takes down only as much of it as a given month needs, and interest runs on the drawn portion alone. is drawn down across the school-supply season and stands flat again by the year end. Money goes out to build stock, the stock goes out to schools, the money comes back, and the drawing unwinds itself before the books close.

The school term is a driver, it is published, and it is a calendar rather than an economy. So the distinction that matters most is this one: a cycle turns with an economy and a season turns with a calendar. They are not two words for one thing. School terms fall in the same months every year and would carry on falling in the same months whatever else were happening anywhere.

Step three, and here the procedure stops. Anjani Stationers published revenue of Rs 2,70,00,000/- against Rs 2,40,00,000/- the year before, a rise of 12.50 per cent. Reading that as evidence about a cyclical attachment would need a run of the field’s demand beside a run of an economy, and Anjani Stationers publishes neither. Not a short one. None.

Two observations of one business against no economy classify nothing in either direction. A reader arrives wanting to pin the cyclical label on a maker of school registers, on the reasoning that a register is the sort of thing a household could defer. There is no evidence for that and none against it. The honest thing is to say so and stop. A sentence that concedes the temptation before declining it has already picked its answer, and every reader hears the pick rather than the caveat.

What does it cost to be in the field at all?

The first question was about the money coming in. The second is about what has to be in place, and paid for, before any of it can come in. Two businesses selling to the same customers at the same rates can face completely different answers to the second question. So the second question separates fields a reader would otherwise lump together, and its answer decides how much of the money coming in survives the year.

Commodity Industry

A commodity industry is one where what every seller offers is near enough identical that a buyer chooses on rate and delivery alone, so no seller can move the price by itself. The mechanism underneath is simple: near enough identical goods leave no seller with any room to move the price. How that comes about is covered under market structures.

The structural consequence turns up in a buyer’s accounts, published rather than constructed. Anjani Stationers buys paper. Paper is 58.33 per cent of its revenue, the single largest thing the business does with money. The business buys from an incumbent mill, and beside that mill sit nine alternatives, all making the same weight and the same finish. Ten sellers of an identical good are known to exist. A quote comes back in a day. First delivery runs about two weeks.

Now the year. The paper bill was Rs 1,57,50,000/- for 75,000 reamsA standard packaged quantity of paper, counted in sheets. Mills quote and invoice by the ream, so a paper bill divides cleanly into a rate a ream., a weighted average rateAn average where each buy counts in proportion to its size, so a large purchase at one rate moves the average more than a small purchase at another. The weighted average is what the whole year actually cost, spread over the whole year’s quantity. of Rs 210.00/- a ream. The best rate struck anywhere in that year sat Rs 10.00/- below the average, on a fifth of the volume, 15,000 reams. Carry that Rs 10.00/- across all 75,000 reams and the gap is Rs 7,50,000/-. Measured against the year’s operating profit of Rs 41,50,000/-, the gap is 18.07 per cent.

Rs 7,50,000/- is a ceiling, not a saving. Nobody quoted the lower rate on the other 60,000 reams. The lower rate was struck on a fifth of the year and on nothing more. Nothing establishes that the whole year could have gone at that rate, so Rs 7,50,000/- is the largest the gap could conceivably have been worth. Anjani Stationers pays Rs 210.00/- a ream. Printing the figure without the word ceiling quietly invents an offer that was never made.

And now the reading that matters most. Ten sellers of an identical good is why a quote comes back in a day, and it is also why the best rate anybody could strike is only Rs 10.00/- better than the average. The easy quote and the thin gap are not two facts but one structure seen twice. The same interchangeability that makes the field trivially easy to shop is what makes shopping it barely worth the walk.

Four vegetable stalls stand outside one office gate. The same tomatoes, laid out the same way, at rates within a rupee of each other. Walking the row to compare takes a minute, and the minute is the whole point. The walk is easy because the tomatoes are the same, and the walk buys almost nothing for the same reason.

One prohibition, and it is absolute: buying in a commodity field does not make Anjani Stationers a commodity seller. Selling hard-bound registers into a school relationship eleven years old is about as far from an interchangeable good as a stationery business gets. The type is a claim about a field, and this business stands in two fields at once. Which side is under discussion has to be named every time.

Asset-Light Industry

An asset-light industry is one where a seller can operate at scale without holding the things being sold or the plant that makes them, so what it carries is an arrangement rather than a stock. Setu Bazaar, an invented online marketplace, is the worked case. Rs 5,00,00,00,000/- of goods crosses it in a year and it holds not one rupee of that at any point, and Rs 4,80,00,00,000/- of the total reaches the sellers who listed the goods.

Now the correction, and it is the one that matters most. Asset-light is a claim about what a business holds, and not a claim about what it costs that business to stand still. Holding and standing still are different columns, and they get read as one column constantly.

Set the two published standing costsThe part of a cost base that has to be paid across a year whether or not anything is produced or sold: the rent, the salaries, the software, the maintenance. How it is separated from the variable part is settled elsewhere. side by side as shares of revenue. The two businesses are nothing like the same size, and shares of revenue are the only fair comparison between them.

The business holding almost nothing carries by far the heavier standing base as a share of what it takes, and both figures were published beforehand. Note carefully which 62.50 per cent that is: it is Setu Bazaar's standing base as a share of its revenue, and a different 62.50 per cent turns up below, under capital-intensive industry, meaning something else entirely.

Why does it come out that way? An arrangement has to be kept running whether or not anything crosses it. The listings have to stay up, the payments have to clear, the disputes have to be answered and the people who do all of that have to be paid in a quiet month exactly as in a busy one. None of that cost shrinks because nothing was held.

A courier firm with no vans shows the same shape at a smaller size. The firm hires every vehicle it uses, so on any given morning it holds nothing that moves. The firm still pays the office, the software, the phone lines and the people who answer them, on the first of every month, whether the parcels came in that month or not.

Two structures, set beside each other, and the comparison stops there. Which of the two is the better business to be is a different question with a different kind of evidence behind it.

Capital-Intensive Industry

A capital-intensive industry is one where a large amount has to be spent, and in place, before the first unit can be sold, and where that spend then sits there whatever the output turns out to be. The spending comes first and the selling comes after, and the order is the whole difference.

Anjani Stationers works it. Its rated capacityWhat a works can produce in a period when everything runs as designed. Rated capacity is an engineering statement about the plant, not a forecast of what anybody expects to sell. is 4,00,000 registers, built out of 4,000 line-hoursOne hour of one production line running. Capacity is often stated as a rate an hour multiplied by the hours available, so line-hours are the unit that connects the two.. The works made 2,50,000. Divide the second by the first and utilisationOutput measured against what the plant was rated to produce over the same period, expressed as a share. Utilisation compares a works with itself and with nobody else. is 62.50 per cent, and that single division is all the arithmetic the type needs. The standing base was Rs 74,00,000/- for the year, paid at that output and much the same at any output the works could have reached.

So what does the capital actually buy? Two things, and a third that gets assumed and is not there.

The capital buys a ceiling, and the ceiling is real. 4,00,000 is a wall. No order beyond it can be taken this year, whatever anybody offers and however good the terms are. The wall is a genuine limit on what the business could ever do inside twelve months, and it is knowable in advance.

The capital buys a floor under the cost of standing still, and the floor is real too. Rs 74,00,000/- was going out whether the works ran at 2,50,000 or somewhere else within reach. The floor is set by what was built, not by what got sold.

The capital does not buy demand. Nothing about having built to 4,00,000 causes anybody to order 4,00,000. A field where every seller has built capacity and none of them can fill it is the ordinary shape of a capital-intensive field, not a strange one, and it is worth expecting rather than being surprised by.

One thing the works' own figures settle: the 62.50 per cent utilisation is a rate shortfall and not an hours shortfall. Every one of the 4,000 line-hours ran. The gap sat in what each hour produced, a different problem entirely from a works standing idle.

A wedding hall makes the same point at a size anybody can picture. A hall costs the same to keep whether it is booked forty times a year or ninety. Building a bigger hall raises what can be booked, and raises what has to be paid, and does absolutely nothing to the number of weddings in the neighbourhood.

Scarcity, and what cannot be added at will

Something is scarce in a field when more of it cannot be added at will, and for exactly that reason scarcity belongs with what it costs to be in the field rather than with anything a seller decided. The placement is the whole point. Scarcity is not a policy. Scarcity is a fact about the ground that every seller standing on it inherits.

A definition that is never run is an assertion, so the test now runs across the three types already named.

Two hundred seats in a hall on the one evening in December everybody wants is scarce. The identical two hundred seats in the identical hall on a Tuesday in June are not. Nothing about the hall changed. The change is in whether more seats could be produced at the moment they were wanted, and that was never the hall’s decision to make.

One sentence carries the whole of scarcity. A charge can only hold where the thing being charged for cannot simply be added, and that is a property of the field rather than of the seller. In reverse it is even clearer: where more can be added at will, somebody adds it, and the room to charge closes behind them. How much anybody should charge is a separate subject.

Who sets the rules?

The first two questions were answered out of figures a business publishes. The third cannot be. There is no line in anybody's accounts that reports who decides the price, and no set of statements anywhere that reveals whether a newcomer would be allowed to start. Saying so plainly is part of the method, not an apology for it. The two types still get named and still get a test that can be run tomorrow, and neither Anjani Stationers nor Setu Bazaar sells in either of them.

Regulated Industry

A regulated industry is one where the price, or who may sell, or both, is set by somebody who is not a party to the sale. The definition is a test rather than a list, and a test is the right form: a list of sectors goes out of date, and a test does not.

So the question is simply this. Is the price set by the seller, or by somebody who is not a party to the sale? Two people are in a transaction. If a third decides what it costs, the field answers yes.

Three things follow structurally, and none of them needs a figure to see.

An auto rickshaw shows the first one in ten seconds. The fare comes off a card the driver did not write, in a currency the driver does not set, for a distance a meter the driver did not choose measures. The driver runs the vehicle, keeps the takings and carries the costs, and holds no view at all on what the ride costs that anybody has to listen to.

And now the declaration, in plain words and without embarrassment. Neither Anjani Stationers nor Setu Bazaar sells in a regulated field, so the type gets its test and no worked business. Inventing a tariff orderA written decision by a sectoral regulator setting what may be charged for a regulated service. The contents of any particular one are read at the regulator’s own publication, never from memory. or a licence to illustrate the point would be manufacturing exactly the kind of outside fact that nobody can get from a set of accounts.

Deregulated Industry

A deregulated industry is a field where the rules that used to decide the price or the entry have been lifted, so both now sit with the sellers and the buyers instead. Deregulated is the same two tests, answered the other way, after something changed.

The second test is the one that matters for deregulation: can a new seller start by deciding to? A head count can never reveal the answer, so a list of who currently sells says nothing about it, and that is exactly why the question is the right one. Ten sellers in a field where anybody may join and ten sellers in a field where a licence regimeA rule under which selling in a field requires prior permission from an authority, rather than only capital and a willingness to trade. The requirements of any particular regime are read at its own source. stands between a newcomer and the first sale are two entirely different fields that happen to produce the same count.

Three consequences of lifting a rule, reasoned rather than asserted:

And the sentence that keeps the word honest: deregulated is a statement about what changed and when, so it always carries a date. A field described as deregulated with no date attached is a field being described from somebody's memory, and the memory could be a decade stale without anybody in the room noticing.

Consider a road where only two bus operators were once permitted to run, and where anybody with a bus may now run. Everything about how that road works turns on when the change happened. A year in, the field is still sorting itself out. Fifteen years in, it has long since settled into whatever it settled into, and calling it deregulated describes its history rather than its present.

Regulated vs Deregulated Industry

Set the pair against each other on the two tests already stated, and refuse to add a third. The first test is who sets the price. The second is whether a new seller can start by deciding to. Nothing else.

And now count the cells honestly, the part almost every treatment of this skips. Two tests with two answers each produce four combinations, not two.

Two of those four carry the familiar words. The other two are just as real, turn up just as often, and have no short name at all. Two unnamed cells are precisely why running two tests beats reaching for a label: the label has vocabulary for half the possibilities and silence for the other half, and silence tends to get rounded to the nearest available word.

Deregulation is a movement between cells rather than a cell. So the useful question is never which of two boxes a field sits in. The useful question is which of the two tests changed, in which direction, and when. Asked that way it yields an answer with a date in it, and a date is checkable. Asked the other way it yields a word, and a word is not.

What does the answer buy, and how is it found out?

A type is worth having only if it changes what gets looked at next. The fourth question produces no new type. Instead it asks what the three answers do once they are in hand, and where each of the three physically comes from, and the answer decides whether the first three were a classification exercise or real work.

How Industry Structure Affects Profitability

One claim carries how structure affects profitability. The type sets the conditions every seller in the field works under, and it sets them for all of them at once. Not for the good ones, not for the large ones. For all of them, including the ones who have not started yet.

Take the three worked types and say what each does to what a seller keeps. Each of these follows from something already shown above rather than from a new assertion.

And now the limit, and it matters more than anything above it. A type explains what every seller faces and explains nothing whatever about why one of them did better than another. Anjani Stationers earned Rs 41,50,000/- of operating profit in its year. Setu Bazaar published a loss of Rs 2,50,00,000/- in its year. Neither of those two results follows from either business's type, and neither could have been predicted from the answers to the three questions.

The type is the field and the result is the business. Two uniform shops in the same market, same cloth, same rates, same landlord, same customers: one of them opens an hour earlier and one of them does not. Nothing in the field explains that difference. The field is identical for both of them, and that sameness is what makes it a field.

Which leaves one last question worth asking out loud. Why did the type have to be established from outside the accounts in the first place? Because a set of accounts can acquit a rival and can never convict one. A margin that held steady cannot have been forced down by anybody, so the accounts can rule a rival out. Ruling one in needs three facts that no profit and loss statement contains: the number of sellers in the field, how easily one more could set up, and what the buyer could switch to instead.

How to Research an Industry

Finding out a field’s type is a narrow job, and it is not a general method for analysing the field. The general method is a separate subject with its own procedure, and stretching the narrow job to cover it would produce a worse version of both.

Three questions, in the order they should be asked, and where each answer physically comes from.

The ordering rule is the one thing about the three questions that is not obvious in advance. The type shows which evidence is worth chasing, so the type is settled first. Chasing a rival's cost sheet in a commodity field can consume a year establishing that everybody buys at much the same rate, which the structure of the field would have shown in a week. The type does not answer the question. The type shows which question is worth the year.

One closing note. None of those three answers sits in anybody’s accounts, and that is not an inconvenience to be worked around. Sitting outside the accounts is the defining property of this kind of work, and the reason it takes longer than reading a statement and cannot be automated away.

What goes wrong, and what does the mistake actually cost?

A third one is shorter and turns up constantly: reading asset-light as cheap to stand still. Setu Bazaar holds none of the Rs 5,00,00,00,000/- of goods that crosses it and carries a standing base of 62.50 per cent of its revenue, a share of revenue rather than any utilisation figure. Anjani Stationers runs a works and carries 27.41 per cent. A reader who treats asset-light as a statement about cost has read the wrong column entirely. Asset-light says what a business holds and says nothing about what has to be paid whether or not anything moves.

A fourth failure is one any analyst can commit in their own draft. A refusal that leans is not a refusal. A sentence such as it would be tempting to call this cyclical, and one can see why, has picked its answer while appearing not to, and every reader hears the pick rather than the caveat. Write the refusal flat: there is no evidence either way, and here is what is published instead.

How would somebody use a type on an ordinary working day?

A classification like this earns its place only if the reading changes something on a Monday. Three people use the same three answers for three different purposes, and none of them uses it to decide what anything is worth.

An analyst uses the type to choose which line of the accounts to interrogate first, before opening the accounts at all. Every difference between two sellers in a field where nobody can move the price has to live on the cost side, so the field sends them straight there. Whether a heavy standing base gets carried is a question about counts, so a field where one sits under everybody sends them to the volume line and to capacity. Same statements, different reading order, and reading order is most of what separates a fast week from a slow one.

A lender uses the type to work out what has to be paid in a month when nothing much happens. Standing still is a structural question and it has a structural answer: what is the standing base, and how much of the year's income has to arrive before it is covered? A lender looking at Setu Bazaar sees a standing base at 62.50 per cent of revenue, a share of revenue and not a utilisation, sitting under a business that holds none of the goods it moves. A lender looking at Anjani Stationers sees 27.41 per cent, sitting under a works. The two are genuinely different lending propositions, and the labels asset-light and capital-intensive would have suggested the reverse.

A buyer uses the type to decide how much effort to spend shopping. The buyer’s use is the cheapest and the most immediately useful. If three sellers quote inside a day on a near enough identical good, the field has shown that shopping harder buys very little, and Anjani Stationers' own year says the same thing in rupees: the best rate anybody struck was Rs 10.00/- under the weighted average, and the Rs 7,50,000/- that would represent across the year is a ceiling and not a saving. If instead one seller answers in a fortnight and the others do not make the same thing at all, the field has shown the opposite, and the effort belongs there.

Notice what none of the three did. None of them put a value on anything, none of them said which field is better to be in, and none of them predicted a result. All three used the type to decide where to point their attention, and pointing attention is the entire job a type does.

What is not covered here?