McCormick

McCormick buys pepper, cinnamon, garlic, onion and chillies from farms around the world, mostly abroad, cleans and grinds them, and sells them in two ways. A ten-year-old could follow it. The first way is under its own name: the jars with the red cap in the supermarket, plus brands it has bought over the years, such as French's mustard, Frank's RedHot, Cholula and Lawry's. That is the Consumer segment, about 58% of last year's sales. The second way is to other food companies and restaurant chains, who buy seasonings and flavours to put in their own crisps, sauces and burgers. That is Flavor Solutions, about 42%.

It is not a glamorous business, and that is the point. A jar of oregano costs a few dollars, lasts a year in the cupboard, and the buyer rarely compares brands. In fiscal 2025, to 30 November, McCormick sold $6.84bn, earned an operating margin of about 16%, and turned almost every dollar of profit into cash. Over ten years free cash flow has averaged about 98% of net income. It has raised its dividend for forty years in a row.

Two things changed in 2026. In January McCormick paid $750m to take control of McCormick de Mexico, a joint venture that sells mayonnaise and other products in Mexico, which adds about $800m a year of sales. And on 31 March it agreed to merge with most of Unilever's food business, the home of Knorr and Hellmann's. When that deal closes, expected by the middle of 2027, McCormick will be two and a half times its present size, and its present owners will hold 35% of it. Most of this report is about what that means for a buyer today.

That last cell is the reason for this report. A shareholder who bought five years ago has lost close to half the money, in one of the steadiest businesses in America, and the business did not shrink. Something else happened, and it is worth understanding before deciding whether the fall is an opportunity or a warning.

The lesson of the history is that McCormick has grown in two ways at once. Underneath, it has sold a few percent more each year, with some pricing. On top, roughly every few years, it has borrowed to buy a bigger brand and then spent three or four years paying the debt down. Management says acquisitions have added about two points a year of growth over the last decade, on top of about 4% organic growth. That rhythm worked. French's and Frank's were good purchases.

The Unilever deal follows the same rhythm, but it is a different size. The Reckitt purchase cost $4.2bn when McCormick's own sales were about $4.7bn. Unilever Foods adds about $12bn of sales to a company with $8bn, for $44.8bn. That is not a bolt-on. It is the smaller partner taking over the larger one, paid for partly in borrowed cash and partly by handing over most of the company.

By region, the Americas brought in $4.87bn of fiscal 2025 sales, about 71%; Europe, the Middle East and Africa about $1.27bn; Asia-Pacific $0.70bn. Since January the Americas share has risen with Mexico.

The economics are simple and good, without being extraordinary. Gross margin has averaged about 39% for ten years, with almost no variation: our moat page measures the swing at 4.5%, which is excellent. Capital spending is small, about 3% of sales. The weakness is the return on capital. McCormick has bought most of its growth, so the balance sheet carries $6.3bn of goodwill and a newly recorded $1.47bn "reacquired right" in Mexico. On that capital the ten-year average return has been about 10%, never below about 8%. That is a respectable return for a food company. It is not a Coca-Cola return, and it tells you that past purchases have earned roughly their cost of capital, no more.

Pricing is where the signal lies. In the third quarter organic sales rose 1.9%: price added 2.2% and volume and mix took off 0.3%. In Consumer in the Americas, the most valuable piece and mostly the United States, volume and mix fell 2.5%, after 3.6% in the second quarter. Management talks of a consumer recovery that is "taking longer than expected". It is the same pattern we described at PepsiCo and Hershey this year: the brands can still raise prices, and the customer is buying a little less.

McCormick reported its third quarter, to 31 August, on 1 October. The headline numbers read well, and two of them need correcting before they mean anything.

First correction: the growth was bought. Of the 17.4% rise in sales, 14.6 points came from consolidating Mexico. Organic growth was 1.9%, all of it price. Adjusted operating income rose 22%, also mostly Mexico, plus a cost programme. Adjusted EPS rose by one cent, because the higher profit was eaten by a higher tax rate (24% this year, against 21.5%) and by the interest on the money borrowed for Mexico.

Second correction: the special charges are not all special. GAAP earnings were $0.36 a share. The difference, $0.50, was $95m of costs for the Unilever transaction and integration and a $43m write-off of a pepper-sourcing project in Malaysia, which the company has abandoned. The transaction costs will recur until 2029; management guides about $300m of one-time costs to achieve the savings. An owner pays them, whatever the adjusted figure says.

The guide. McCormick held its 2026 forecast: organic growth of 1–3%, adjusted operating income up 16–20%, and adjusted EPS of $3.05 to $3.13, up 2–5%. Three words moved. In January it promised "sustained volume growth"; now volumes are "expected to be stable". Cost inflation was raised to 6–7%. And on the call management warned of lower margins in the fourth quarter and of further cost pressure in 2027, for which it will guide in January. The shares made a new low that day at $44.81, and over the following week Barclays, JPMorgan, UBS, Bank of America, Stifel and TD Cowen all cut their targets.

This is the picture to hold on to. Adjusted earnings per share were $2.83 in fiscal 2020 and should be about $3.09 this year: 9% more in six years, roughly 1.5% a year. Over the same period the multiple the market pays for them has gone from 33 times to 16. Nothing in the business halved. What halved was the price of a slow-growing, highly predictable consumer company once interest rates rose from almost nothing to about 5%. In 2020 a bond paid you less than 1%, so 3% earnings yield on a spice company looked generous. Today the ten-year Treasury yields about 5%, more than the 3% earnings yield McCormick offered in 2020. That repricing is mostly done. The question for the next five years is the deal.

Buffett's test for a franchise is simple: a product that is needed or desired, thought by customers to have no close substitute, and not subject to price regulation. Let us run McCormick through it, one business at a time.

The spice aisle is the best of it. A jar of ground cumin is cheap, bought rarely, and full of things the buyer cannot check: purity, origin, whether it has been cut with something. A known brand removes the worry. That is why McCormick can sell at a premium and also, quietly, supply many of the store brands next to it. It earns from both sides of the shelf. But a habit protects you less when the household is short of money, and the falling US volume is exactly that: shoppers choosing the cheaper jar, or using the one they have for longer.

Flavor Solutions has the opposite strength and weakness. Once a seasoning is in a customer's crisps or burger sauce, changing it means changing the taste, which customers hate to do. So the business is sticky. But the customer is PepsiCo or a hamburger chain, with buyers who know the cost of every ingredient, and the margin, 14.5%, shows who holds the power. This is the same lesson as Coca-Cola's bottlers or Hershey's cocoa: being essential to a large customer is not the same as being able to charge it.

What is the trend? We would call the moat stable in spices and narrowing slightly at the edges. The new pieces from Unilever are good brands with better margins than McCormick's own, about 24% against 17%, and that is part of the case for the deal. But they are bought, not built, and they come with the competition in mayonnaise and soup that Unilever is now reporting.

On 31 March 2026 McCormick and Unilever agreed a transaction with a long legal name, a Reverse Morris Trust. Stripped of the legal machinery, it works like this. Unilever puts its food business, except in India, Nepal and Portugal and a few drinks and nutrition brands, into a new company. McCormick pays Unilever $15.7bn in cash, which it will borrow. Then that new company merges with McCormick, and McCormick issues new shares for it. When it is over, Unilever's shareholders own 55.1% of the combined McCormick, Unilever itself keeps 9.9%, and today's McCormick shareholders own 35%. The structure exists to make the deal largely tax-free for Unilever's shareholders. Unilever's shareholders do not vote. McCormick's voting shareholders do.

This is the arithmetic that most commentary skips. On 2025 figures, one McCormick share today stands for about $29.70 of sales, about $5.06 of operating profit and about $17.50 of net debt. After the deal, the same share will stand for about $26 of sales, about $5.57 of operating profit before any savings, or about $6.35 if all $600m arrives, and about $26 of net debt. Each share earns about 10% more before savings and carries about half as much debt again. Interest on the new money will take most of the extra profit, which is why management's own forecast is for earnings per share only mid-to-high single digits above what McCormick would have earned alone in the first year, rising to mid-to-high teens by year three as the savings arrive.

And look at whose profit it will be. McCormick brings about 40% of the combined sales but only about a third of the operating profit. Two-thirds will come from Knorr, Hellmann's and Unilever's food-service business. A shareholder who bought McCormick to own the spice aisle will, from 2027, mainly own stock cubes and mayonnaise, run by McCormick's managers.

Is it a good deal? There are honest arguments on both sides. For: McCormick is buying a better-margin business at the same multiple as its own, with brands at least as famous, a food-service network it could not build, and a credible list of savings, mostly in overheads and buying. It is effectively being paid for its management with two-thirds of a much larger company. Against: Unilever, which knows these brands better than anyone, chose to sell them; its food segment has slowed to almost nothing this year; the price was fixed on 2025 numbers; and the debt is large. Terry Smith's fund sold its whole Unilever stake, saying it did not rate McCormick's management or returns highly. Barclays pointed out in March that Unilever's holders would own the majority. BNP Paribas noted that large consumer-goods mergers have rarely worked, and the nearest example is one Buffett knows well.

The Kraft Heinz shadow. In 2015 Berkshire helped merge Kraft with Heinz, loaded the company with debt and counted on cost savings. The savings came. The brands, starved of investment, did not grow, and Berkshire wrote its stake down. Our X-Ray of Kraft Heinz last month called it "Berkshire's mistake, eleven years later". McCormick is not Kraft Heinz. The leverage is lower, management is promising to reinvest $100m of savings in the brands, and the integration plan has been worked through in detail. Its CEO spent fifteen years at Heinz and knows the story from the inside. But the shape is similar enough that a careful owner should want to see the evidence, not the plan.

Here is the part we find most interesting. We modelled the combined company to fiscal 2030 and asked what earnings per share at $48.40 implies (Part XI has the details). On our central assumptions, with all the savings, McCormick would earn about $4.30 a share in 2030. The price today, at 16 times earnings in 2030, is consistent with about $3.85. That is almost exactly what our model gives with no savings at all. The market, in other words, is paying for the deal as if the $600m never arrives. If even half of it does, today's price leaves some room. If none of it does, and the debt stays high, there is no margin of safety at all.

The most important competitor in the spice aisle is not another brand. It is the store's own label, and in many cases McCormick is happy to make it, because it keeps the factories full. That is a sensible hedge, and it also limits how far the brand can push price. We could not find independent market-share data we trust, so we will not quote one.

Outside the spice aisle, the combined company will meet larger and better-capitalised rivals: Nestlé in soups and bouillon, Kraft Heinz in condiments, Givaudan and Kerry in flavours. None of them is new. What changes is that McCormick will be fighting on more fronts, with more debt, while its managers integrate a business larger than their own.

A curious governance point. McCormick has two classes of common stock that are economically identical. The non-voting shares, ticker MKC, are what almost everyone owns. The voting shares, MKC.V, are about 5.5% of the total and are held heavily by employees, the 401(k) plan and current and former managers. So a transaction that hands 65% of the company to someone else will be approved or rejected by a small group close to management. After closing, a committee with two Unilever members must recommend within a year whether to collapse the two classes into one. Unilever, holding 9.9%, has agreed to a one-year lock-up and to sell down over five years, which means a steady seller in the market from 2028.

Capital allocation. Over the last five years McCormick has returned about 70% of its free cash flow, almost all as dividends; buybacks have been token. It has used its balance sheet for acquisitions and then paid down: debt went from $5.0bn after Reckitt to $4.0bn by 2025, before Mexico pushed it back to about $5.0bn. That is disciplined. What is new is the scale. A management that has done this well at $4bn is now doing it at $45bn, with borrowed cash, a two-year exit from Unilever's systems, and a consumer that is buying less. Integrity is not in question. Judgement on size is.

Pay and incentives. The CEO earned $10.8m in 2025, about 79% of it in stock awards. Long-term awards are tied to sales, earnings per share and return on capital. We would watch the last one most closely after the deal, because a $45bn purchase will drive return on invested capital down for years even if it works, and we would want the board to say how it will measure success.

* 2018 includes a one-off benefit from the US tax reform. † Includes the $866.8m Mexico gain; without it, about $2.31. Per-share figures are adjusted for the 2020 two-for-one split.

Ten years of figures tell a consistent story. Sales grew about 5% a year, a good part of it from acquisitions. Operating margin has moved between about 14% and 18% and is lower today than in 2020. Earnings per share rose about 5% a year to 2025, almost all of it before 2020. Free cash flow has tracked net income closely, with one bad year: in 2022 it fell to $390m, below the dividend, as working capital swelled in the supply-chain crisis. The share count barely moved. This is a business that grows with the economy, prices a little above inflation, and spends its spare cash on dividends and the occasional large purchase.

Our reading: the dividend is safe and the record will very probably survive, because management has said so publicly and has the cash flow to do it. What will not survive, for a few years, is growth of 7% a year. A company carrying four times EBITDA in debt and promising to halve it will raise by a token amount. This is the PepsiCo lesson from July, in sharper form: the dividend is safe; the dividend growth is not.

The three lenses agree. Unlike Microsoft or Alphabet, where heavy investment drives a wedge between earnings and cash, McCormick is a Coca-Cola-type case: capital spending is small and roughly equal to depreciation, so earnings, owner earnings and free cash flow land within about 15% of each other. As a standalone business at $48.40, you would be paying about 16 times owner earnings for something growing perhaps 3–4% a year, with a 4% dividend. On its own, that would be a fair price for a good, slow business.

But a standalone McCormick is not on offer. We therefore value the combined company per McCormick share. Our model starts from 2025 combined sales of about $20bn and an operating margin of about 21% before savings, adds the savings, takes off interest on about $17–18.5bn of net debt by 2030 and tax at 25%, and divides by about 769m shares. We put a multiple on fiscal 2030 earnings, discount four years at 10%, and add the dividends received on the way.

The central value moves to about $48 at 14 times and $59 at 18 times; to about $55 at a 9% discount rate and $52 at 11%. Take away the savings entirely and it is about $47; with half of them, about $50. Our central 2030 EPS of about $4.30 is close to the analysts' $4.43, which suggests the consensus already assumes the deal works.

What is priced in? At $48.40 and 16 times, the market needs about $3.85 of earnings per share in 2030, 89% of our central figure, which is roughly what the combined company earns with no savings. So the price already assumes a disappointing deal. That is the case for the stock. The case against it is that "disappointing" can be worse than our low case: Unilever Foods may keep growing at 0.1%, the savings may be eaten by reinvestment and lost sales, and a company at four times EBITDA has little room if something else goes wrong.

A comparison. Our X-Ray of Hershey last month put that company at 17–19 times normal owner earnings, with a buy-zone of $150; we liked the brand more than the price. McCormick is close to the opposite case: we are less sure of the business it is becoming, and the price is low enough to reflect that. Kraft Heinz, at about 11 times owner earnings, was cheaper still and we passed, because its brands were shrinking. McCormick's brands are not shrinking, but they are not growing in volume either.

Why $42? It is about 21% below our central value, about 13.6 times this year's guided earnings and a dividend yield of about 4.6%. For a business we rate 3 out of 5 on what can be known, with a large transaction ahead, the usual rule would ask for 30% or more, and we would not argue with a buyer who waited for less. But at $42 even the low case, with half the savings and four more years of flat volumes, returns most of the money, and the dividend pays you about 4.6% while you wait. The 52-week low is $43.25, so the market has come close.