Editorial Article
Fourteen Per Cent of the Wine Market Is Growing 1.4 Times as Fast
The premiumisation story has a size, and the size is small
Premium wine is the most repeated growth argument in the drinks industry. It is also, inside our own numbers, a very specific quantity. The premium segment was worth USD 74.22 billion in 2025. The whole wine market was worth USD 528.4 billion. Premium is therefore 14.05 per cent of the category it is credited with transforming.
It is growing faster. Premium publishes 5.34 per cent a year to 2033 and the total market publishes 3.81 — a ratio of 1.402. That is the entire basis of the premiumisation thesis, and it is correct. What almost never gets stated is what it implies for the other 86 per cent.
| 2025, USD bn | 2033, USD bn | CAGR | Share 2025 | Share 2033 | |
|---|---|---|---|---|---|
| Premium wine | 74.22 | 112.35 | 5.34% | 14.05% | 15.77% |
| Rest of the wine market | 454.18 | 600.25 | 3.55% | 85.95% | 84.23% |
| Total wine market | 528.4 | 712.6 | 3.81% | 100% | 100% |
The residual is arithmetic, not a published series: total minus premium, at both ends. Its 3.55 per cent is slower than the market it sits inside, by construction — when one component grows faster than the whole, everything else must grow slower. That is not a finding about everyday wine. It is what the finding about premium wine costs.
Three things that follow
One: premiumisation is redistribution, not new money. The wine market adds USD 184.2 billion of value between 2025 and 2033. Premium wine adds USD 38.13 billion of that — 20.7 per cent of all the growth from 14.05 per cent of the base. Premium punches just under one and a half times its weight. That is a real effect and it is smaller than the rhetoric.
Two: the rate of the shift is 0.215 points of share a year. Premium’s share goes from 14.05 to 15.77 per cent over eight years — 1.72 points, 0.215 a year. At that pace premium’s share reaches 20 per cent in 27.7 years. On the published rates, premiumisation is a slow structural drift that will still be completing itself in the 2050s, not an event that has already happened.
Three: premium may be a price story in a mix costume. Premium grows at 5.34 per cent while the market’s volume falls 1.19 per cent a year. Premium can only be growing on volume if premium volume is rising against a falling market — and neither report carries a premium volume series, so this cannot be tested. If premium volume is flat or falling, then the 5.34 is entirely realisation, and the premium tier is the same price-led mechanism as the category around it, just running hotter.
What our own number cannot tell you
The premium report does not state a price threshold. There is no band — nothing that says a bottle above which price counts as premium — and no statement of whether the boundary is retail price, ex-cellar price, or a tier defined by the producer. The 14.05 per cent is therefore a definition we own, not an industry standard we have adopted.
That matters more than it sounds. Premium thresholds are usually drawn somewhere between USD 8 and USD 15 a bottle at retail in mature markets, and where a line is drawn changes the share materially, because wine’s price distribution is steeply skewed. Move the threshold down and premium’s share rises toward a quarter; move it up and it collapses toward a tithe. Anyone quoting 14.05 should know they are quoting our boundary.
There is a second consequence, and it is the useful one. Because the total market’s value growth is price-driven and its volume is falling, the premium tier’s 1.4-times advantage can be delivered without a single new premium drinker. It can be delivered by the existing premium drinker paying more, and by the everyday drinker being reclassified upward as prices rise across the board. Both are entirely consistent with the numbers we publish, and neither is premiumisation in the sense the word is used commercially.
What the premium thesis needs to survive
Three tests, in order of how cheaply they can be run:
- A volume line for the premium tier. If premium hectolitres are rising while total hectolitres fall, the mix story is real. If they are flat, it is price.
- A stated price threshold, held constant across the forecast. A share series computed on a moving threshold is a series that can be made to say anything.
- A split of the 5.34 into price and mix, which needs the first two and is the only form in which the number is comparable across markets.
Until then, the defensible claim — the one this article makes — is narrow and survives scrutiny: premium wine is one seventh of the world wine market, it grows about 1.4 times as fast as the rest of it, and it will be about one sixth of the market by 2033. Everything beyond that is a story about a number rather than the number itself.
Two companion pieces take the same category apart from other angles: why the wine market grows when its volume falls, and the year spirits passes wine.
More in this series: Spirits passes wine in 2031 · The wine market grows because the bottle got dearer · RTD grows twice as fast as wine on a twelfth of the revenue.