The Only Wild-Harvested Thing in the Bottle

Gin’s cost structure is unusually well known for a spirits category, and that is the first strange thing about it. Duty is a published rate. The still is a capital cost that has been measured. The neutral spirit is a commodity. Everything except the botanical basket can be looked up, and the botanical basket is the part the industry will not discuss.

The category moved 109.0 million nine-litre cases — 981.0 million litres — and USD 14.0 billion at retail in 2025, up 1 per cent in volume and 2 per cent in value with pricing adding 3. That is USD 14.27 a litre, the cheapest liquid in premium-plus spirits. The cheapness is not a marketing position. It is an arithmetic result, and the arithmetic leaves almost nothing for the ingredients.

Every other line in the bottle is fixed

UK spirit duty is charged per litre of pure alcohol, so it falls identically on a London dry and on a supermarket own-label: GBP 30.62 per litre of pure alcohol in the 8.5 to 22 per cent band from 1 February 2026. A 70cl bottle at 37.5 per cent contains 0.2625 litres of pure alcohol, which is GBP 8.04 of duty.

Set that against a shelf price and the space available for ingredients disappears.

GBP 1.96 has to cover the liquid, the bottle, the cap, the label, the freight and two margins. Juniper’s share of that is pence. The most important agricultural input in the product is the smallest number in its price, which is precisely why nobody prices it properly and why a doubling in berry cost is invisible to the drinker and material to the producer.

The capital side is just as subdued. The best-documented capacity increment of the year is The Boatyard Distillery spending GBP 2 million to go from roughly 300,000 bottles a year to more than 1.6 million, with GBP 510,000 from Invest Northern Ireland. That is GBP 1.54 of capital per bottle of annual capacity — against GBP 8.04 of duty on the one bottle that comes out of it. Capacity is cheap. Tax is not. Botanicals are neither, and that is the problem.

Juniper is not a crop, it is a harvest

Coriander seed, the second botanical in most recipes, behaves like a crop and can be treated like one. When the price spiked, farmers planted it, and the market flooded. War in Ukraine forced buyers off Ukrainian and Russian seed and onto Bulgarian supply, and one UK supplier now has 36 hectares of coriander growing domestically.

Juniper does not do that. There is no meaningful farmed supply at the volumes the category needs, and the producers say so plainly. Four Pillars’ head of production, quoted in The Spirits Business, is direct about it: juniper is something the Australian distillery “will probably never be able to buy locally… we’re never going to be able to grow the volumes that we need.” Four Pillars bought 50 tonnes of juniper in a year, sourced from Macedonia and Serbia through a UK commodity supplier.

That is a wild harvest with a hand-picking labour line, a short season and a weather exposure — sourced from the Balkans, sold as a dried good, and bought before the orders arrive. The buyer’s own description of it is the most useful evidence in the file. Beacon Commodities’ Tommy Haughton calls juniper “always universally difficult, probably because it’s so important,” adds that “you don’t want to overbuy… there’s a lot of guesswork,” and notes that it is now “two or three or four” botanicals that are a problem every year, rotating.

At the same industry conference, the supplier side described the buying pattern without embarrassment: “I’ll buy a huge quantity and maybe in two years I won’t buy the product, and buy very low quantities.”

That is a market clearing on guesswork rather than on a curve. There is no juniper futures contract, no published index, no origin disclosure requirement on a bottle, and no substitutable input that performs the same function — juniper is what makes gin gin. Every other category with a single irreplaceable agricultural input has built a pricing mechanism around it. Gin has built a silence.

The standard lets a producer opt out of the risk entirely

The regulatory definition compounds it. A distilled gin is made by re-distilling ethyl alcohol of agricultural origin in the presence of juniper and other botanicals. A cold-compounded gin skips the still: essences and flavourings are added to spirit. Both are legally gin. Only the first can be a London gin, and only if the distillate reaches 70 per cent alcohol by volume with all the natural plant materials present.

So the product standard does not require a berry. A compounding operation needs almost no plant, no seasonal buying programme and no agricultural exposure — it buys flavour, not harvests. In a category where 45.6 per cent of US volume sits in the value tier, that is not a technicality. It is the cheapest route to market, and it carries none of the supply risk that the distilled-gin producer is quietly holding.

The consequence is a split the price does not show. Part of the category is exposed to a wild harvest in the Balkans; part of it is buying an essence; and both arrive on shelf at the same duty rate, because duty is blind to quality. The producer who takes the agricultural risk is not compensated for it in the tax treatment, in the retail price, or in the disclosure on the label.

What would actually tell you the harvest has turned

Price is the wrong signal, because nobody publishes one and because the cost is too small a share of the bottle to see. The usable signal is buying behaviour. Botanicals are dried goods that store for 12 to 24 months, so a single bad harvest is absorbed without a visible price event and two in a row are not. A producer buying two years forward, then buying nothing, is telling you what it thinks the harvest will do — and the supplier quoted above described exactly that pattern as normal.

There is a second signal, in the volume mix. World gin volume is 26 per cent above 2019 while the United Kingdom — where the modern category was made — is 40 per cent below its peak. The growth is concentrated in markets where the liquid is sold at USD 2.53 a litre, and at that price the botanical bill is a larger share of the producer’s revenue than it is anywhere in the premium tier, on a smaller revenue base. The category is growing where it can least afford a bad harvest.

What this report does not hold

We do not hold a verified cost of botanicals per litre, of juniper per litre, or of neutral grain spirit per litre, and we have not invented one. Those numbers are commercially confidential at the scale that would matter, and any cost build-up published here would be a construction rather than a measurement. What can be stated without construction is the structure: a category that sells at USD 14.27 a litre, whose largest single price component is a tax that ignores quality, whose plant costs GBP 1.54 per bottle of annual capacity, and whose one irreplaceable input is a wild crop bought on guesswork by some producers and avoided altogether by others.

That is a supply risk sitting inside the cheapest liquid in premium spirits, and it is unpriced because the product is too cheap for the risk to show up in the bottle.