Inflation Pass-Through: From Input Cost to Shelf Price

Pass-through is the share of a cost rise that actually reaches the price a buyer pays. The share is almost never the whole rise and almost never nothing. Three things set the share: how many rivals the seller faces, how quickly its buyers walk away when the price moves, and how long its existing contracts still have to run. The rest comes out of margin.

Underneath that answer sits one fact that does most of the work in this guide. A cost is something that happens to a business. A price is something the business decides. Nothing forces the second to follow the first, and the distance between them is where every interesting thing here lives. A report that a crop failed, or that shipping rates trebled, is information about costs. A cost report is almost no information about prices. It says nothing about what any seller decided to do in response.

The marginWhat is left of the selling price after the cost of making and delivering the thing. Usually quoted as a percentage of sales, sometimes as a rupee amount per unit. is the shock absorber sitting between the two. The margin can take a hit, and when it does, the buyer never sees the cost event at all. The absorbing is not a footnote. Margin is the reason two honest numbers can describe the same year and disagree completely, and it is why anybody reading a price release during a cost shock needs to know what to ask next.

What is pass-through, and why is it a share rather than a rate?

Start with the chain, short enough that every link in it is worth naming. An input price rises. The rise lands on a producer as a higher unit costWhat it costs to make one of the thing, counting materials, labour and the share of overheads attached to it. Rises when any input in it gets dearer.. The producer then decides how much of that to put on its own price. Buyers then decide whether they will keep buying at the new price. And only what survives both of those decisions ever reaches a price indexA weighted average of prices actually charged to buyers, rebuilt each period so that changes in it can be compared. Which goods sit in a price index, and whose prices it collects, are covered separately..

Two of the five links in that chain are decisions rather than mechanisms. Pass-through is a share and not a rate for exactly that reason. A rate is a thing that can be looked up. A share is a thing somebody chose, under pressure, with rivals watching. Asked what the pass-through of edible oil into packed snacks is, the honest answer starts with a question back: for which seller, in which market, on which contracts?

Here is the version at street level. A tea stall outside a bus depot buys milk, sugar and gas. The gas cylinder goes up. The arithmetic is not hard, so the stall owner knows exactly what that costs per glass, to the paisa. The stall owner cannot know whether the customer who buys two glasses a day will still buy two glasses at the higher price, or drop to one, or walk fifty metres to the stall that has not moved its board. So the board sometimes moves and sometimes does not, and when it moves it often moves by less than the cost did. Every large business is doing exactly that calculation at scale, with a pricing meeting instead of a chalk board.

What is Cost-Push Inflation, and why is it the uncomfortable one?

Cost-push inflation is the general price level rising because costs rose and enough of that rise was passed on. Not because buyers turned up with more money to spend. The starting gun is on the production side: a harvest fails, a fuel price jumps, a currency move makes every imported input dearer, a wage settlement lands across an industry. Costs go up first, prices follow to the extent that sellers can make them follow, and the index moves.

Set that beside demand-pull inflationPrices rising because spending across the economy outruns what the economy can produce at current prices. Where it comes from and what sustains it are covered separately. and one difference does all the damage. Cost-push raises prices while squeezing output, so it is the uncomfortable case where both of the things a reader watches move the wrong way at once. When costs rise and sellers pass some of it on, buyers buy less at the higher price, so less gets made and sold. Prices up, output down, in the same episode, from the same cause. Demand-pull does not do that. Under demand-pull the price rise arrives with more being bought and made. Unpleasant in its own way, but at least the two numbers are telling a consistent story.

The difference changes what a rising index is evidence of. A price index moving up alongside output moving up says one thing about an economy. The same index moving up alongside output moving down says something quite different, and the second is the harder situation because there is no comfortable reading of it available. Where each pattern comes from, and what else can produce a general rise in prices, is covered separately under the causes of a general price rise.

What decides how much of a cost rise reaches the shelf price?

Three things, and they are worth holding as three because a seller can be strong on one and helpless on another. The first is how many rivals the seller faces. If eleven other producers make something near enough identical and none of them has moved, the seller who moves first loses volume to the ten who did not. A seller with many close rivals passes on less, not because it wants to, but because the first mover pays for the move. Where that competitive position comes from, and how a market with three sellers differs from a market with three hundred, is market structureHow many sellers a market has, how alike their products are, and how hard it is for a new seller to get in. Market structure is set out in full at the start of the market work., which is set out separately and used here rather than rebuilt.

The second is how sharply buyers cut back when the price moves. Buyer sensitivity to price is the elasticity of demandHow much the quantity people buy responds to a change in price. High elasticity means buyers cut back sharply; low elasticity means they mostly keep buying. Elasticity is explained in full elsewhere.. Elasticity is covered separately in full. The direction is what matters here. A seller whose buyers barely change what they buy when the price moves can put most of a cost rise on the price and keep its volumes. A seller whose buyers walk away over a five rupee difference cannot, and knows it before the meeting starts.

The third gets forgotten more often than the other two, and it is the most mechanical of them. A seller who signed a twelve month supply contract at a fixed price in March cannot pass on anything at all in June. The price is written down. No amount of seller strength or buyer loyalty changes that. The contract runs until it runs out. Contract length is why pass-through is spread over time rather than arriving all at once, and it is the reason a cost shock and its price consequence never share a calendar.

Why does pass-through take time to arrive?

Because contracts do not all end on the same day. The renewal calendar sounds too small to explain anything, and it explains almost all of it. Sarani Foods, an invented Sankhya producer that packs edible oil into one litre pouches, supplies four large buyers. Each is on a twelve month contract, and the four contracts were signed in different quarters, so exactly one of them comes up for renewal each quarter. When the oil price jumps, Sarani can reprice one buyer immediately and cannot touch the other three.

Work the average through. Suppose Sarani has settled on passing 7.20 per cent onto its price whenever a contract turns over. In the first quarter one buyer in four is on the new price, so the average price Sarani actually charges has risen 1.80 per cent. In the second quarter it is two in four, so 3.60 per cent. In the third, 5.40 per cent. Only in the fourth quarter, a full year after the cost event, is every buyer paying the new price and the average finally reaches 7.20 per cent.

A slow looking price rise can be a fast cost rise arriving through slow contracts, and the price index cannot tell the two apart. A gentle four quarter climb of 1.80, 3.60, 5.40 and 7.20 per cent looks in the data exactly like a mild and gradually worsening cost problem. The cost actually moved in a single step in one week, filtered through a renewal calendar. Telling those two apart is beyond the index. The contract terms do it.

What happens to the part that is not passed on?

The absorbed part does not disappear, and it does not stay with the supplier who raised the price. The seller's margin pays for it. The Sarani Foods case puts numbers on it. Oil, the dominant input in a pouch of packed edible oil, rises 12.0 per cent, the food component of the Sankhya price spine. Full pass-through would mean lifting the shelf price of the pouch by the same 12.0 per cent. Sarani lifts it by 7.20 per cent instead, or 60 per cent of the cost rise. The remaining 4.80 percentage pointsThe plain difference between two percentages. Going from 7.20 per cent to 12.0 per cent is a rise of 4.80 percentage points. A rise of 4.80 per cent would be a different thing. are the price rise Sarani chose not to take.

In rupees, on the Sankhya pouch that sold for Rs 250/- before any of this happened. Full pass-through would have put the pouch at Rs 280/-. Sarani puts it at Rs 268/-. The gap is Rs 12/- a pouch, every pouch, and Rs 12/- on Rs 250/- is exactly the 4.80 per cent that never reached the shelf. Sarani sells 4,00,000 pouches over the year, so the absorbed part comes to Rs 48,00,000/- of margin given up against what full pass-through would have kept.

The absorbed part is invisible in every price index and shows up in the profit account instead, so an index that barely moves can still be describing a very large cost event. The Rs 48,00,000/- is real money that Sarani no longer has. The sum is simply recorded in a different document, published by a different party, on a different date, and read by a different set of people.

Why does the same cost rise reach two businesses differently?

Because pass-through is a property of where a seller stands, not of the cost that hit it. Put Sarani Foods next to Devkot Snacks, a second invented Sankhya producer making a near identical pouch. The same oil rise of 12.0 per cent lands on both in the same week and in the same proportion. Sarani has a pouch that shops stock because customers ask for it by name, and only two producers make anything comparable in its region. Devkot has nine near identical rivals within twenty kilometres and shelf space that goes to whoever is cheapest that month.

Sarani passes on 60 per cent, so its shelf price moves 7.20 per cent, and the pouch goes from Rs 250/- to Rs 268/-. Devkot passes on 25 per cent, so its shelf price moves 3.00 per cent, and its cheaper pouch goes from Rs 200/- to Rs 206/-. Against full pass-through Sarani gives up Rs 12/- a pouch and Devkot gives up Rs 18/-. On the same 4,00,000 pouches a year, that is Rs 48,00,000/- for Sarani and Rs 72,00,000/- for Devkot.

The price index records 7.20 per cent for one and 3.00 per cent for the other from a cost event that was identical for both. The index is measuring the pass-through decision at least as much as it is measuring the cost. If both pouches sat in a consumer basket with the same basket weightThe share of household spending a category is given inside a price index, so that a big spending item moves the index more than a small one. How the weights are set is covered separately., the index would show the pair contributing an average of 5.10 per cent, against an input cost event of 12.0 per cent for both of them. Nothing about that reading is wrong. The reading is simply an answer to a different question from the one most people think they are asking.

Where these numbers come from. Every percentage above belongs to the Republic of Sankhya. The 12.0 per cent input rise is the food component of the Sankhya price spine. The shares of 60 per cent and 25 per cent are settings chosen to make the arithmetic visible. Measured Indian figures come from the issuers named below.

What do the two limit cases look like?

Both ends of the range are worth naming, not because either is common, but because a reader who can place them has somewhere to put every real case in between. At one end sits full pass-through. The shelf price rises by the same percentage the input did, so the seller keeps what it kept per rupee of cost, and the index records the whole cost event. At the other end sits nothing passed on. The shelf price does not move at all, the index records no price change whatsoever, and every rupee of the cost rise lands on the profit account.

The two limit cases produce identical cost events and completely opposite public records. A price index measures the decision rather than the shock, and nothing shows it more cleanly. At full pass-through on the Sankhya pouch, the index shows 12.00 per cent and Sarani gives up nothing. At zero, the index shows 0.00 per cent and Sarani gives up the full 12.0 per cent. On a Rs 250/- pouch that is Rs 30/- each and Rs 1,20,00,000/- across 4,00,000 pouches in the year. Same oil, same week, same country, and one of those two years leaves no trace in the price statistics at all.

What does an analyst ask first when input costs jump?

Not how much costs rose. Commodity prices and freight rates are visible to everybody at once, and nobody has to be asked, so the size of a cost rise is usually the easiest thing to establish. An analyst who stops there has learned the part of the story that was already public.

The two questions that actually change a forecast are how much of the rise this particular business can put on its price, and how long its existing contracts have left to run. Pricing power and contract length together decide whether a cost shock is a margin event or a price event, and the same rise can be either depending only on the answers. A lender asks the same pair for a different reason: a borrower that can pass costs on will service its loan through a cost shock, and a borrower that cannot will do it out of a margin that was already thin.

A household runs the identical test without calling it that. When the cooking gas price moves, the size of the move does not decide whether the household notices. The restaurant down the road, the auto driver and the tuition centre decide it, by whether each can put the gas rise on its own price, and how soon. If they can, the household meets the cost rise several times over in a month. If they cannot, the household never sees it, and somebody else absorbs it quietly.

The Republic of Sankhya, Sarani Foods and Devkot Snacks are invented.
Educational material. Not advice on any investment, tax, budget or market position.