How GDP Growth Feeds Through to Corporate Earnings

Economic growth reaches a company's profit through four links, and each one can break on its own. Growth becomes somebody's spending, spending becomes volume or price, volume and price become revenue, and revenue becomes profit only if the cost base does not rise alongside it. In Sankhya's year 3 one invented business turned a 4.00 per cent real year into profit up 16.76 per cent while another watched profit fall 52.54 per cent.

A published growth rate hides more than it shows. A national growth rate is an aggregateA single number that sums up a whole economy or a whole market, such as total output for a year. An aggregate describes the group and never any one member of it., which means it is an average over millions of producers who did completely different things that year. Some of them made more. Some of them charged more for the same amount. Some of them made less and were carried in the total by their neighbours. When the figure is published as one rate, every one of those differences has already been averaged away, and the rate cannot be un-averaged back into any particular set of accounts.

Gross domestic product (GDP) counts the total output of an economy, and a total measured at this year's prices is not the same as a total measured at a fixed year's prices. Both of those are settled under national output measurement. The narrower question runs like this. Given a visible growth rate, what has to be true for it to arrive as a bigger profit at a named business, and what happens at each of the four points where it might not arrive at all? One growth year can be walked to a profit line twice and end in two opposite answers, and the joint that produced the difference between them can be named.

What are the four links between a growth rate and a profit?

The chain has five stations and four joints, and it is worth saying out loud before it is walked. An aggregate moves. Because the aggregate moved, somebody somewhere spends differently. Because somebody spends differently, a particular business sells a different number of units or gets a different price for them. Because units and prices moved, revenue moves. And because revenue moved, profit moves, but only by whatever is left after the cost base has taken its share.

Each of those four joints can break by itself, and most of the disappointment people feel about growth figures comes from quietly assuming that none of them does. The assumption is easy to make because the chain sounds inevitable said quickly. Growth means more spending means more sales means more profit. Said slowly, with a business in front of the reader, each arrow turns out to be a claim that has to be checked rather than a step that follows automatically.

Think about it at street level first. The economy of Sankhya, an invented republic, grew this year. Did the tea stall outside the bus depot sell more cups? Only if more people passed the depot. Did the depot see more people? Only if the growth happened to be in the kind of activity that puts people on buses. And if the stall did sell more cups, is the owner better off? Only if the price of milk, sugar and gas did not rise by more than the extra cups brought in. Four questions, four places to lose the connection, and none of them is answered by knowing that the economy grew.

If the market grew in rupees, what actually reaches the business?

Start with the joint that catches the most careful readers, the one that hides inside a number rather than inside a business. A market can grow by producing more things, or by charging more for the same things, or by some mixture, and the rupee total looks identical in all three cases. Sankhya's year 3 is the standing example throughout: output rose 4.00 per cent in real terms while the price level rose 5.00 per cent, so the rupee total rose 9.20 per cent. Which of those two numbers reached a business depends entirely on what the business sells.

Take the case where the whole 9.20 per cent was extra things sold and no price anywhere moved. Ambara Foods, an invented staple maker in Sankhya, ran on revenue of Rs 50,00,00,000 in year 2 against a cost base of Rs 45,00,00,000, of which Rs 36,00,00,000 moved with volumeThe number of units actually sold or made, counted in things rather than in rupees, so it does not move when only the price moves. and Rs 9,00,00,000 did not. Sell 9.20 per cent more at unchanged prices and revenue reaches Rs 54,60,00,000, the moving costs reach Rs 39,31,20,000, and the Rs 9,00,00,000 stays exactly where it was because nothing about it was ever connected to units. Profit goes from Rs 5,00,00,000 to Rs 6,28,80,000, a rise of 25.76 per cent on a 9.20 per cent market.

Now take the case where the whole 9.20 per cent was price and Ambara Foods sold not one extra packet. Since 9.20 per cent more rupees is 9.20 per cent more rupees, revenue still reaches Rs 54,60,00,000. But this time everything else is in rupees too. A general price move is general. The moving costs rise 9.20 per cent to Rs 39,31,20,000, and so does the cost base that did not move before, to Rs 9,82,80,000. Profit lands at Rs 5,46,00,000, up exactly 9.20 per cent. In a pure price year every rupee figure in the accounts scales by the same amount, so profit grows at the headline rate and the business is not one packet better off than it was.

The pure price year is the cleanest statement of the trap, and it is worth pausing on. The second year looks like growth in every published line. Revenue up. Profit up. Profit up by exactly the rate everybody is quoting. And nothing happened. A reader who asks only whether the numbers rose has no way to tell the two years apart, and the two years are not remotely the same year.

Why does one growth year reach three businesses months apart?

The second joint is about position rather than composition, and it is the one people treat as a footnote when it is half the answer. Growth is not delivered to every market on the first of April. Growth arrives when somebody makes a particular kind of decision, and different businesses depend on different decisions that get made at very different speeds.

Household scale first. A woman's income goes up this month. Ask what she buys more of tomorrow. Milk, vegetables, an extra bus ride, maybe a second packet of biscuits. The daily purchases take a day to decide. Now ask what she buys more of a year from now. A better cooking gas connection, a scooter, a room added to the house. The larger purchases take a year of watching whether the higher income holds. And the loan for the room is taken only after she has decided on the room. The income moved a year before that decision, and the loan comes a further gap after it.

Three Sankhya businesses sit at exactly those three distances. Ambara Foods sells a staple that gets bought every week, so its share of the growth arrives more or less as the growth does, spread evenly across the four quarters. Korai Finance, an invented lender, is bought when somebody decides to borrow, and that decision follows the decision to spend rather than leading it. Vedhal Machine Works, an invented builder of production equipment, is bought when a customer decides to expand capacity, and that decision is made only after a run of good years has convinced somebody that the demand is real. Vedhal's order bookThe value of orders a business has accepted but not yet delivered, so it shows work already promised rather than work already done. in any year is mostly the residue of decisions taken twelve to eighteen months earlier.

Timing is part of the answer rather than a detail of it. A business whose customers decide slowly reports last year's economy, and everybody reads this year's growth rate at it. The gap is not a lag in some vague sense. The gap is a specific number of quarters, it can be estimated from the length of the decision the customer has to make, and it is the difference between a growth figure being relevant to a set of accounts and being about a period those accounts have not reached yet.

What happened to two Sankhya businesses in the same growth year?

Take Sankhya's year 3 and walk it to a profit line twice. Real output rose 4.00 per cent, the price level rose 5.00 per cent, and the rupee total therefore rose 9.20 per cent. All three of those figures describe one economy and one year. Two invented businesses inside it are about to report results that have almost nothing in common.

Ambara Foods is the easy trace, and its ease is exactly why it needs a warning attached. Assume its volume tracks the real growth of the economy at 4.00 per cent and its realisationThe average price a producer actually gets per unit after discounts and product mix, which can move quite differently from the price on the list. tracks the price level at 5.00 per cent. Revenue then grows 9.20 per cent to Rs 54,60,00,000. The 9.20 per cent revenue figure matching the economy's 9.20 per cent confirms nothing whatsoever. The two are the same relationship written twice. A business whose units grow at the real rate and whose prices grow at the price level has revenue growth of exactly the rupee rate by construction, so arriving at one number from two directions here is arithmetic rearranged rather than a finding. The assumption is the answer. Any trace that agrees with itself this neatly is a description of what was assumed.

Where Ambara Foods becomes informative is below the revenue line. Its moving costs of Rs 36,00,00,000 rise with both units and input prices, reaching Rs 39,31,20,000. Its standing costs of Rs 9,00,00,000 rise only with prices, reaching Rs 9,45,00,000. Profit is Rs 5,83,80,000 against Rs 5,00,00,000, up 16.76 per cent, and the margin edges from 10.00 per cent to 10.69 per cent. The 7.56 points by which profit growth beat revenue growth came from one place and one place only: the Rs 9,00,00,000 that had no reason to rise with units.

Vedhal Machine Works had the same year and a different experience of it. Its customers were deciding whether to add capacity, and Sankhya's pace of real growth had just slipped from 5.00 per cent to 4.00 per cent, so a number of those decisions were postponed. Vedhal's volume fell 3.00 per cent. Worse, two new machine builders had entered its market during the previous year, and the competition for the orders that did exist meant Vedhal's realisation rose only 1.00 per cent while everything it bought rose 5.00 per cent. Revenue therefore fell 2.03 per cent to Rs 19,59,40,000.

Now watch the cost side finish the job. Vedhal's moving costs fell with units but rose with input prices, netting out to a rise from Rs 8,10,00,000 to Rs 8,24,98,500. Its standing costs of Rs 9,90,00,000 rose 5.00 per cent to Rs 10,39,50,000 and did not care that revenue had fallen. Profit collapsed from Rs 2,00,00,000 to Rs 94,91,500, a fall of 52.54 per cent, and the margin went from 10.00 per cent to 4.84 per cent. Same economy, same year, same published growth rate, and two businesses that began the year with identical 10.00 per cent margins ended it 69 points apart on profit growth.

Why does the cost base decide how much of the growth survives?

The fourth joint is where the two traces above actually parted company, and it deserves saying without any of the machinery. A rise in revenue is not a rise in profit. A rise in revenue is a rise in profit minus whatever the costs took on the way past. So the entire question of how much of a growth year a business converts comes down to which of its costs went up alongside revenue and which of them simply carried on.

Household version. Two men each add a Rs 5,000/- month of extra work. The first is a courier who spends Rs 3,000/- of it on extra fuel and keeps Rs 2,000/-. The second is a tailor whose sewing machine, rent and electricity were already being paid whether the work came or not, so he keeps nearly the whole Rs 5,000/-. The tailor converts growth much better. He also has a much worse month when the work stops. The rent arrives regardless, and the good months and the bad months are one fact about the tailor rather than two.

The name for this in a set of accounts is operating leverage. How operating leverage is measured and reported belongs with cost behaviour and margins rather than with anything about aggregates. For reading a growth rate, what matters is the direction operating leverage points. Ambara Foods held only 20.00 per cent of its year 2 cost in a fixed cost baseThe part of a business's costs that stays put across a year whether it sells more or less, such as rent, the salaries of permanent staff, and insurance.. Vedhal Machine Works held 55.00 per cent there. Give both of them Sankhya's year 2 shape, real growth of 5.00 per cent with prices up 5.00 per cent, and Ambara's profit rises 19.70 per cent while Vedhal's rises 36.24 per cent. Vedhal is the better converter, comfortably.

Now give both of them the downturn shape instead, units down 3.00 per cent and realisation up only 1.00 per cent. Ambara's profit falls 42.62 per cent. Vedhal's falls 52.54 per cent. The business that converts growth best is the same business that suffers a slowdown worst. A steep slope is steep in both directions, and nobody gets to choose which half of it applies. A reader who describes a heavily fixed cost base as an advantage in a good year and a problem in a bad year has described the same slope twice and thinks they have found two things.

Why do new sellers take growth that the market really produced?

The third joint is the one that catches the reader who has already learned to check the other three. A market can grow in real units, at a good pace, arriving on schedule at exactly the businesses expected, and a particular business inside it can still finish the year with revenue growth well below the market's. The growth was genuine. Somebody else got it.

A growing market is visible from outside, and the visibility is precisely the problem. Ten shops around a bus stand are doing well, so an eleventh opens. The trade at the bus stand really has grown, and every original shop can still be selling fewer units than last year, or selling the same units at a price they had to cut to keep them. A market growth figure and a company revenue growth figure are therefore two different questions, and the second cannot be inferred from the first without knowing something about how easily an eleventh shop opens.

Vedhal Machine Works is the instance already given. Two builders entered its market during the previous year. Vedhal held on to most of its customers, so its unit fall of 3.00 per cent was much smaller than it might have been. The way it held them was by not raising prices. Realisation rose 1.00 per cent in a year when everything Vedhal bought rose 5.00 per cent, so the difference between the price it charged and the price it paid narrowed by four points on every machine. Vedhal failed to pass throughGetting a rise in a seller's own input costs into the price charged, so the customer carries the increase instead of the seller. its own cost increase, and the reason had nothing to do with the economy and everything to do with who else was quoting.

An analyst who has correctly established that a market grew has established a fact about the market and nothing at all about how it was divided. The division is a separate question with separate evidence behind it, and where the evidence points depends on how hard it is for an eleventh seller to appear, which is a matter of market structure and is settled in its own place rather than here.

What does an analyst actually do with a national growth figure?

Here is the whole procedure, and it is shorter than people expect. Take the growth figure. Ask which market it is supposed to have moved. Ask whether the growth in that market was units or price, and in what proportion. Ask when the customers of the business in question make the decision that turns that market movement into an order. Ask what share of that market the business held last year and whether anything happened to the number of sellers. Then, and only then, write a number next to a named line itemA single named row in a set of accounts, such as revenue, employee cost or finance cost, rather than a summary figure or a ratio built from several rows. in a specific set of accounts.

The test that governs all of it is whether the aggregate lands anywhere. An aggregate that cannot be traced to a specific line in a specific set of accounts is background reading rather than analysis. Treating it as analysis is how a growth rate ends up multiplying an estimate that was never connected to it. If the chain breaks at the first question, because the market that moved cannot be named, then the figure has said something about the country and nothing about the position, and the honest thing is to record that and stop.

Notice what this procedure refuses to do. The procedure never applies a rate to a company. Rates apply to markets; companies have accounts, and the bridge between the two is the four joints. The procedure also produces a written trail, and that matters more than it sounds. Six months later the outcome is in, one of the four assumptions turns out to have been the wrong one, and the trail is what lets somebody check the reasoning.

Who reads this chain for a living, and what do they do with it?

Three people open the same growth release in the same week and none of them is reading it for the rate. An equity analyst is looking for which joint is currently mispriced, a credit officer is looking for whether a borrower's standing costs will still be covered if the joint gives way, and an operator inside a business is looking for how many quarters of warning the chain gives before it reaches the order book.

The equity analyst's job is comparison rather than prediction. Two businesses of similar size in the same market can be expected to report very different years, and most of the difference is already visible in their cost splits and their customers' decision lengths before the year starts. The analyst does not want a multiplier from a growth figure. The growth figure is a reason to ask whether the market has priced in the same joint holding at both, when it holds at one and not the other.

The credit officer runs the chain in reverse and looks only at the bad half of the slope. Vedhal Machine Works entered year 3 with Rs 9,90,00,000 of standing costs against Rs 2,00,00,000 of profit. The ratio between those two figures is the whole story of what happens if orders slip. The standing costs arrive regardless, and the profit is thin cover for them. A lender does not need to forecast Vedhal's year. The lender needs to know how far revenue can fall before the standing costs stop being covered, and that is answerable from the cost split alone, today, without any view about growth at all.

And the person inside the business uses the timing joint as a calendar. Orders at Vedhal Machine Works follow customer expansion decisions by four to six quarters. A finance controller who knows that can read a slowdown in the economy today and say, roughly, which quarter it will show up in the order book. The controller is not forecasting a number. The forecast is of when to look, and a forecast of when to look is a great deal more useful and a great deal more defensible. The cyclicalityHow strongly a business's sales rise and fall along with the wider economy, as against businesses whose sales barely notice what the economy is doing. of the order book is known well before its size is. The same reading tells the controller which of Vedhal's costs must be made to move if the orders do not.

Why can a correctly traced chain still not give an earnings number?

Everything above gives a direction and a rough size. A direction and a rough size do not give a profit figure, and the distance between the two is not a technicality. Walk back through what a complete trace actually rests on: an assumption about how much of the market's growth was units, an assumption about the lag between a customer decision and an order, an assumption about how many sellers are quoting, and an assumption about which costs will move. Four assumptions, each of them a range rather than a point, and they multiply.

Look at what a small error in one of them does. Ambara Foods' profit rose 16.76 per cent on volume growth of 4.00 per cent. Move that volume assumption to 2.00 per cent, one quarter of the change and well inside anybody's error bar, and profit growth would land near half of what was computed. The standing cost stays put and eats a much larger share of a smaller increment. The output moves faster than the input did. A steep slope means exactly that, and it leaves the answer least reliable for exactly the businesses where the chain matters most.

Anyone offering an earnings number derived from a growth number has walked past four assumptions without pricing the uncertainty in any of them, and is guessing with more decimal places than a guess deserves. What the chain honestly supports is a statement of this shape: this growth reaches this business through this joint, the joint currently looks intact for these reasons, so the direction is up and the size is more than the market's rate rather than less. A statement of that shape is a real conclusion. The reasoning goes no further, and pretending otherwise turns a good method into a bad forecast.

Ambara Foods, Vedhal Machine Works, Korai Finance, Palvi Textiles and the Republic of Sankhya are invented.
Educational material. Not advice on any investment, tax, budget or market position.