Sector Drivers: What Moves an Entire Industry at Once

A sector driver is a condition sitting outside any single company that moves the results of all of them together: a shared input price, a demand condition, a change in what is permitted. The defining test is that it acts on competitors at the same moment, and that simultaneity is also how a driver is told apart from something one company did.

Almost every wrong conclusion in company research has the same shape. Something moved in the numbers, the researcher attributed it to the company, and the attribution was never checked against anyone else. The word driver gets used loosely to mean anything that matters. The loose version is useless. Held to the narrow one, the term does real work.

What makes something a driver rather than just an event?

A driver is a condition, not an incident. An incident happens to somebody: a plant catches fire, a chief executive resigns, a large order lands. A condition sits in the environment that everybody in the field operates inside, and it presses on all of them whether they noticed it or not. Nobody has to react for a condition to reach the accounts. A condition arrives on its own.

A sector driver acts on every competitor at the same moment, and an event affecting one company cannot do that. The distinction is not a fussy one. Simultaneity is observable, so the distinction carries the whole of the diagnostic value. Whether the other companies moved too can be looked up, and the answer settles which of the two things is in play.

Take ten shops in one shopping centre. If the centre's management raises the electricity charge for every unit, all ten shops see costs rise in the same month, and none of them did anything. The shared charge is a driver. If one of the ten hires two extra staff and its costs rise, that is a shop doing something. From the outside the two look identical in a single set of accounts, and completely different the moment the shop next door is glanced at. Taking that glance seriously is the whole of the method.

None of this replaces reading industry structureHow an industry is put together: how many players there are, how easy it is to enter, how much power buyers and suppliers hold. The five forces reading belongs to Michael Porter, Competitive Strategy, 1980, and the business analysis material sets it out., which shows how a field is arranged and who is able to keep what. A driver identifies what moved. Structure indicates where the movement is likely to settle. The two answer different questions, and the second belongs to the business analysis material.

How does one input price actually reach every company's accounts?

Traced once, concretely, it never has to be treated as abstract again. A shared input price moves. All of them buy it, so the purchase cost per unit shifts in the same direction for each of them, whoever they are. The purchase cost lands in the cost of materialsThe accounting line that collects what a maker spent on the physical inputs consumed in producing what it sold. Its definition and placement in the statements are settled in the accounting material. line. Subtracting that line from revenue leaves gross marginRevenue less the cost of materials, expressed as a percentage of revenue. The accounting material settles what it does and does not capture., so gross margin moves too. Four steps, no judgement calls, and the same four steps for every company in the field.

There is exactly one company specific part, and it is not skill. It is stock. A maker holding four months of material is still consuming material it bought at the old price long after the new price has arrived in the market. A maker holding six weeks of material is consuming the new price much sooner. The differences in timing between competitors come from how much stock each was holding and not from anything any of them did well. An early margin move is not evidence of a better company.

A household makes this familiar. Two neighbours both cook with the same oil. The shop's price drops on the first of the month. The neighbour who bought a six month tin in February goes on cooking at February's price until the tin is empty. The neighbour who buys weekly is paying the new price by Thursday. Neither of them is a better cook, and neither made a clever decision. The two neighbours had different amounts of oil in the kitchen.

How is a driver told apart from something one company did?

The test is to look at the competitors. Nothing else is in it. If a movement shows up across the field, it is a driver. If it shows up in one company and not in the others, it is that company. There is no third case, and there is no version of this test that requires a model, a call with management or a subscription to anything. The test requires the same line, from the same period, for the companies alongside.

The competitor test is cheap, it takes an afternoon at most, it is skipped almost every time, and running it would prevent a large share of the wrong conclusions in company research. That is the operational heart of the method. Nobody skips it because it is hard. People skip it because they are already researching one company, the movement appeared in that company's numbers, and attributing it to that company requires no extra work at all.

The test has a cost when it returns a positive. If everybody moved, the story about the company under research is lost. Losing it feels like a loss and is actually the finding. The finding is something true about the field instead of something false about a company, and true beats interesting on every day of the week.

Which drivers act on demand, and which act on cost?

Split them. The two kinds arrive at different places, and a researcher who knows which kind they are holding knows where to look. A demand driver changes how much is sold: household formation, a repainting cycle, new construction, industrial capital spending. A demand driver lands at revenue. A cost driver changes what it costs to make what is sold: the input price, freight, energy. A cost driver lands at the cost of materials and at the expense lines below it.

Set against the year three ladder for Sarvani Coatings Limited, each one touches an identifiable row. Revenue Rs 2,415 crore. Cost of materials Rs 1,304 crore, leaving gross profit of Rs 1,111 crore. Then employee cost Rs 205 crore and other expenses Rs 460 crore, giving earnings before interest, tax, depreciation and amortisation (EBITDA) of Rs 446 crore. Depreciation of Rs 92 crore takes it to earnings before interest and tax (EBIT) of Rs 354 crore. A demand driver moves the first row and everything below inherits the movement. A cost driver leaves the first row alone and moves the second.

Because the two kinds land at different rows, knowing which one is in front of the analyst identifies exactly which line should have moved, and a demand story that shows up in a margin line and not in revenue is almost always a mix effectA change in the reported average caused by the blend of what was sold shifting, rather than by any individual price or cost changing. Its measurement is taken up separately. rather than a demand effect. That single sentence catches a great deal of loose writing. When somebody says demand improved and points at gross margin, the questions to put are what happened to revenue, and then what happened to the blend of what was sold.

Two parts of the demand side behave differently, so keep them separate. Household formation and a repainting habit that has been shortening are slow and directional. New construction and industrial capital spending are cyclicalMoving up and down with the broader economic cycle rather than in one steady direction, so a run of good years says little about the next one. The separation from steady growth is worked through separately. and swing about. In the latest year the field grew revenue 11.0 per cent, from Rs 43,500 crore to Rs 48,300 crore. Sarvani Coatings grew 13.9 per cent, 2.9 points more than the field.

What does the test look like on three makers and one input?

Sarvani Coatings Limited makes decorative and industrial coatings, and the dominant cost in that work is a handful of pigments and resins, and what those cost tracks crude oil. The link from crude oil to pigment and resin prices is a mechanism, not a measurement. The shape is what matters: when those inputs move, they move for everyone who buys them, at the same moment, whatever anyone thinks of it.

Take the two published years, year one against year three. Sarvani Coatings' cost of materials fell from 57.0 per cent of revenue to 54.0 per cent, so gross margin went 43.0 per cent, then 46.0 per cent, a gain of 3.0 percentage points. In rupees, revenue went from Rs 1,840 crore to Rs 2,415 crore and gross profit from Rs 792 crore to Rs 1,111 crore. On its own that is a company story, and a flattering one, and it is the point at which most notes stop.

Now run the test. Over exactly the same period Nandivarman Paints Limited gained 2.4 points of gross margin and Kesaria Surface Solutions Limited gained 3.6 points. All three rose. Three risers together are what the competitor test looks like when it returns a positive: a driver moved the field, and the company story about to be written is now a field story instead.

Here is where the honest reading starts, and it is the part that gets left out. The test has narrowed the question and it has not answered it. Sarvani Coatings gained 3.0 points, 0.6 points more than Nandivarman Paints and 0.6 points less than Kesaria Surface Solutions. The three of them are spread across just 1.2 points. Sitting in the middle of the peer group is the least informative place to sit. Nothing in those three numbers separates the part of Sarvani Coatings' gain that came from the shared driver from the part that came from its own pricing or from its blend of decorative and industrial work.

Now do a second reading on a shorter window. A shorter window changes the flavour of the whole thing. Take year two against year three alone. Revenue rose 13.9 per cent, from Rs 2,120 crore to Rs 2,415 crore. Volume rose 6.0 per cent, so realisationWhat a maker actually receives per unit of what it sold, after discounts and after the blend of products it happened to sell. Its build up is taken up separately. per unit rose about 7.5 per cent. Over that same single year the cost of materials per unit of output rose about 3.6 per cent, as a per unit rise rather than a margin point. The cost of materials per unit did not fall. The margin improved because what the company received per unit outran what it paid per unit. The 3.0 point headline above spans two years and cannot be decomposed this way at all, so keep these one year figures well away from it.

The one step up that would genuinely help is not another maker but the supplier. Thottam Chemicals Limited sits one step up the chain, selling resins and additives into this field. Its own margin over the same period would separate two mechanisms that look identical from below: a pricing environment in which everybody's inputs got cheaper, against the makers extracting better terms from their suppliers. The supplier's own margin is the evidence that would settle the question, and it is not available here.

What does a maker do with a favourable gap, and is it visible?

Here is the part that stops a driver being a straight line to a conclusion. When realisation runs ahead of input cost, a maker holds a gap, and it has a choice about that gap. The maker can keep the gap, in which case margin rises and revenue is whatever the existing price list produced. Or it can hand some of it back in selling prices to win volume, in which case revenue is lower than it would have been and margin barely moves. Same driver, same size, and two completely different sets of published numbers.

How quickly does a driver show up in the reported numbers?

Not immediately, and the delay is not noise. Two things sit physically between a driver and an account. On the cost side it is stock. At the end of year three Sarvani Coatings held inventory of Rs 402 crore against a cost of materials of Rs 1,304 crore. Dividing one by the other gives inventory daysStock on hand expressed as the number of days of consumption it represents. Measured here against the cost of materials rather than revenue. The working capital cycle is taught separately. of 112.5, or a little over one quarter of consumption sitting in warehouses and tanks. An input price that moves today is being consumed at the old price for roughly that long.

On the demand side it is contracts. Industrial coatings are sold to manufacturers on agreed terms, so a change in what customers want does not become a booked sale until the order bookThe work a company has agreed to supply but has not yet delivered and billed. The order book sits between a change in demand and a change in reported revenue. turns over. Both delays mean a driver can be plainly visible in the world and completely absent from the accounts, and that gap is the only place in the whole of company research where a researcher has a genuine informational advantage rather than a story.

Read that advantage carefully. The advantage is narrow and real, and it is not knowing something nobody else knows. The advantage is knowing that a visible condition has not yet arrived in a published number, and knowing roughly how long it will take. Nothing exotic produced that: one balance sheet line, one income statement line, and a division.

What makes a claimed driver testable rather than decorative?

Three things, and a claim missing any of them is not a driver claim at all. A driver claim names the mechanism, so somebody can say how the condition reaches a company. A driver claim predicts where in the profit ladder the movement should show up, so there is a row to go and look at. And it predicts that it should show up across the field, so there is a set of competitors to check it against. Put those three together and the claim can be wrong. Being able to be wrong is the only property that makes a claim worth making.

A driver claim that predicts nothing checkable is a description of the weather, and the reason it survives is that nobody can ever show it to be false. Compare two sentences. Input costs are supportive for the field. Against: the shared pigment and resin input got cheaper, so the cost of materials should fall as a share of revenue for every maker, within about a quarter, and Sarvani Coatings should not be alone in it. The second one can be checked next quarter and can fail. Being checkable is what makes a claim worth writing down.

What does identifying a driver leave unanswered?

Two things, and both of them are the things people most want it to settle. A driver does not identify which company benefits most. The size of the benefit depends on position and on the blend of what each one sells, and the three makers above gained three different amounts from one condition. And it does not show whether any of it is already assumed in what buyers and sellers are paying for the shares.

Everybody looking at the field is looking at the same shared condition, so a driver is the most likely thing in all of research to be already widely known. The input price is public. The competitors' accounts are public. Nothing about spotting a field wide condition is scarce, and scarcity is what an informational advantage is made of.

How does an analyst actually work with this, day to day?

Meghna Iyer covers coatings, and she keeps one habit that costs her almost nothing and saves her repeatedly. Before she writes a sentence attributing any movement to a company, she pulls the same line for the two competitors and the supplier over the same period, and she writes down all four numbers whether or not they help her. The habit is not analytical brilliance. The habit is a rule about the order of operations, and it works because it happens before the story forms rather than after.

The same discipline reads differently depending on who is holding it. A lender looking at Sarvani Coatings cares whether an improving margin is durable enough to service borrowings through a weak year, so a field wide driver that could reverse is worse news than a company specific improvement that will not. A household investor reading a results summary cares about something simpler: whether the good year they are being shown is the company being good or the weather being good. Only one of those is a reason to hold on through a bad one.

And the honest note Meghna Iyer writes at the bottom of the sector review is short. The field gained, the company gained with it, its own share of that gain is not separable from the published statements, and the supplier's margin is the next thing to get. That is a smaller claim than most notes make and it is the largest one the evidence supports.

Which body writes the rules here, and why is none of it quoted?