Wine Buys EUR 5.1 Billion of Glass a Year From an Industry That Just Started Shrinking

There are two numbers that sit on either side of the wine industry and almost never appear in the same paragraph. The first is 208.0 million hectolitres — the volume our Global Wine report forecasts for 2025. The second is 82.64 per cent — the share of Verallia’s revenue that went on the cost of making its products in the same year.

Convert the first into bottles and you get 27.7 billion of them, 15.25 million tonnes of glass, and — priced at the EUR 0.185 of revenue Verallia booked per container — a glass bill of about EUR 5.1 billion. The second tells you what kind of industry sends that bill.

The chain, in one table

The bottle count, the tonnage and the spend are our arithmetic on published inputs, not a published series — nobody publishes a global wine-bottle bill. What the arithmetic is good for is proportion. At any plausible exchange rate, the container is about one per cent of the value of the wine inside it. It is also 42 per cent of the mass that leaves the winery, the single largest line in wine’s carbon footprint, and the only input in the whole chain that is bought from a listed industrial at a price set by a furnace.

One and a half Verallias

Verallia is the European leader in glass packaging and the third largest in the world: 34 glass plants, 63 furnaces, and close to 18 billion bottles and jars produced in 2025.

The world’s wine alone would need 1.54 times that output. That single ratio explains the structure of the market better than any price chart: no supplier can carry wine, no wine region can be served by one plant, and every producer from a Chianti co-operative to a Champagne house is buying from a regional handful of furnaces whose first question is not what the label says but whether the line can run a lighter mould.

What the supplier’s year actually looked like

A commodity business with a fixed asset base is easy to read when demand is soft, because the arithmetic is unforgiving in one direction only.

Revenue fell 3.6 per cent. Cost of sales, in absolute euros, rose. Gross profit therefore fell 19.31 per cent — a fivefold amplification of a small revenue decline, in the direction that hurts. The EBITDA margin gave up 3.6 points.

O-I Glass, the American giant, produced the same shape of year with different mechanics: net sales of USD 6,426 million against USD 6,531 million, “modestly lower sales volume (in tons)”, and adjusted earnings per share that nearly doubled to USD 1.60 on USD 300 million of cost-out rather than on price. Segment operating profit rose to USD 846 million from USD 748 million while sales fell. Both companies cut their way to a better year.

Volumes went up and revenue went down

This is the part that matters to a winery. Verallia’s 2025 volumes posted “moderate but steady organic growth, with positive growth in each quarter”. Revenue still fell 3.6 per cent, attributed to “a negative price effect and an unfavourable mix”.

Read those two sentences together and 2025 is the year the glass price came back down. It follows four years in which it went the other way — the energy shock of 2022 and 2023 was passed through to fillers, and wineries absorbed it into a cost line they could not renegotiate until the contract expired. The relief arrived in the year demand softened, financed not by a more efficient furnace but by the supplier’s margin: 24.4 per cent of revenue down to 20.8.

The scale of the transfer is not trivial. On a EUR 5.1 billion global glass bill, three to four points of price is close to EUR 200 million moving from the glassmaker back to the filler in a single year — which is a rounding error against the value of the wine and a meaningful number against the value of a wine company.

The tone matters more than the numbers

Verallia is not distressed. It doubled free cash flow, is paying a EUR 1.00 dividend, and is simultaneously proposing to close a site in Germany, shut a furnace in France and shut a furnace in the United Kingdom while restarting a more efficient one nearby in Leeds.

That combination — cash generation, dividend, and capacity withdrawal — describes an industry that intends to meet demand rather than chase it. Which sets the terms for the next few years: the price relief of 2025 was a utilisation event, not a competitive one, and an industry that removes furnaces gets its pricing power back the moment volumes steady.

Three checks

If Verallia’s price effect turns positive in 2026 while volumes are flat, the 2025 relief was a one-off and the filler’s bottle bill returns to its 2023 trajectory.

If a second consecutive year of organic volume growth coincides with falling revenue, glass has become a buyer’s market for the first time since 2019, and procurement teams should be signing multi-year agreements now rather than waiting.

If the closures go ahead while wine volume falls 1.19 per cent a year to 189.0 million hectolitres in 2033, the order book loses 2.53 billion bottles and 1.39 million tonnes of glass over the period. Furnace closures follow volume out of a region, and the bottles then travel further to reach the winery — which is the subject of the companion piece on where the bottle is made.

The bottle is one per cent of the value of a wine and all of the industrial risk in it. Treating it as a marketing decision — a colour, a weight, an emboss — is affordable only because the industry that makes it has, so far, been able to absorb the consequences.