A plant variety right turns a piece of fruit into a licensable property right, which converts a breeder's income from a margin on every box sold into an annuity charged on every hectare planted, and that annuity is only worth owning if the breeder can enforce it in the countries where the fruit eventually grows.
The licence, not the fruit, is the product. The clearest illustration on this page is Zespri. In FY2025/26, the year ended 31 March 2026, Zespri's global kiwifruit net sales were NZ$5,909.4 million against total operating revenue of NZ$6,133.4 million, the difference made up mostly of NZ$224.0 million from plant variety right licences and NZ$74.0 million in royalties. The New cultivars segment delivered profit before tax of NZ$258.7 million, higher than the profit before tax of any of Zespri's fruit segments. Licensing is not a rounding item attached to a fruit business; it is the most profitable part of it, and it earned that by being separable from the weather, the shipping schedule and any single country's harvest. A licence is sold per hectare of permitted planting, so revenue arrives when growers plant rather than when consumers buy.
Enforcement is the running cost of the model. A right that cannot be policed is a marketing claim rather than an asset, and Zespri's disclosures show the machinery. The cooperative filed a formal opposition to a plant variety right application covering the E2 variety, and it continues to protect its sales channels against kiwifruit grown in China from the Gold3 variety. That exposure is not theoretical: New Zealand's own government proposed stronger plant variety right protection in May 2026, which indicates the legal framework itself is treated domestically as a competitive variable. Breeders also extend the same rights north of the equator to shorten the off-season gap: Zespri's northern hemisphere programme amounts to about 3,874 hectares across Italy, France, South Korea and Japan, with the first commercial SunGold production in Greece and Red19 planted commercially in Europe on 170 licensed hectares, every hectare a licence sold in a jurisdiction where the right must be recognised and defended.
Breeding is a long-dated investment, and the only honest way to read it is by segment. T&G Global reports its intellectual property arm, VentureFruit, as a separate line: NZ$9.0 million of revenue and an operating loss of NZ$2.4 million for FY2025. That is a small, loss-making business inside a group with NZ$1.6 billion of revenue, disclosed precisely because the pipeline is early. The pay-off arrives years later: ENVY became the first New Zealand apple brand to pass NZ$1 billion in cumulative global retail sales, is grown in more than 13 countries and sold in more than 55 markets, and lifted its US household penetration from 7.2% in 2023 to 12.4% in 2025. Newer licences follow the same shape, with JOLI licensed across 273 hectares in New Zealand against a target of 1,500 hectares globally by 2035, and TUTTI licensed onto 300 hectares with a Chinese partner in 2025. Driscoll's, by contrast, discloses no revenue and no segment data at all, so the economics of its proprietary berry varieties cannot be checked from outside; any figure circulating for it should be treated as unsourced.
What a serious buyer or investor should ask. Ask which varieties are protected in which jurisdictions and for how long, what proportion of group profit comes from licensing rather than fruit, how much is spent on enforcement, whether licensed hectares are being planted faster than the breeder can sell the resulting fruit, and what happens to royalty income when a protected variety is legally grown at scale somewhere the breeder does not control.
Because a cooperative's advantage is not scale but the elimination of the intermediary margin between grower and shelf, combined with a single export channel that gives a small national industry the pricing power of one seller, and its disadvantage is that it must raise capital from the people it pays, at the pace of a growing season rather than a capital market.
Cooperative ownership is a governance choice, and it changes what the company optimises for. Zespri is the fullest expression of the model on this page. It is owned by its growers, with grower shareholding at about 70% as of July 2026 against a stated target of 80% by 2035, and it represents 2,786 producers and recognised suppliers across 3,404 registered orchards and 16,115 producing hectares, on a payroll of 901 staff. It handles export rights for everything except Australia, sells into more than 50 countries, and routes 77% of its volume through 76 distribution partners rather than through its own overseas subsidiaries. The result for FY2025/26 was NZ$5,909.4 million of global kiwifruit net sales and 248.1 million trays shipped, and Zespri returned a record NZ$3.56 billion to New Zealand growers in fruit and service payments while declaring a net dividend of NZ$1.39 per share. A listed company would have faced pressure to hold some of that back; a cooperative answers to the people who grew the crop.
Sunkist shows the same structure applied to a different commodity and a different mechanism. Sunkist Growers, Inc. is a nonprofit marketing cooperative owned by its grower members, described in a Limoneira Company filing with the US Securities and Exchange Commission as such, and describing itself on its own site as the longest-standing farmer-owned agricultural co-op in the country rather than the oldest outright. Its structure is licensing rather than plantations: packing houses are typically operated by members and licensees under Sunkist packinghouse licences, so the cooperative's asset is a brand and a marketing order rather than property, plant and equipment. The Limoneira agreement is the model in miniature, signed on 6 June 2025 with a term running from 1 November 2025 to 31 October 2028 and auto-extending annually: Limoneira may grade, pack and ship Sunkist grower fruit and use the Sunkist trademark, and in exchange it markets and sells citrus only through Sunkist and joins both the cooperative and Fruit Growers Supply Company. For context on what that shift costs a grower, Limoneira's own consolidated net revenues in the quarter ended 31 January 2026 were US$18.205 million against US$34.305 million a year earlier, with an operating loss of US$10.551 million; those are Limoneira's figures, because Sunkist discloses no revenue at all.
Where the cooperative model breaks down. Capital has to come from members or debt, so a cooperative cannot fund a breeding programme, a refrigerated fleet and a ripening network at once without charging its own growers for the privilege. Concentration in a single crop leaves the structure exposed to one disease or one bad season, which is why California citrus growers watch huanglongbing so closely. Governance moves slowly, and grower votes on variety mix and market prioritisation can lag by years. The brand can also drift away from the cooperative: the Sunkist name is licensed onto soft drinks and vitamin supplements made and sold by other companies, and the revenue from those products belongs to the licensees, not to the cooperative. A reader comparing Zespri and Sunkist with Dole or Chiquita is comparing a structure that pays growers the residual with one that pays shareholders, and neither is automatically better at moving fruit.
The branded part of a banana, an avocado or a berry is mostly a post-harvest service rather than a botanical fact: the variety determines what the fruit could be, and the cold chain, the ripening room and the pack format determine what the shopper actually receives, which is why companies here invest in containers and ripening centres rather than in advertising.
Ripening is a controlled industrial process, and it is where the consumer experience is manufactured. Del Monte Corporation runs 31 global distribution centres in its network, and those sites include cold storage and banana ripening rather than simple warehousing, alongside 18 fresh-cut processing plants in the United States, the United Kingdom, Japan, South Korea, the United Arab Emirates, Kuwait and Saudi Arabia. The commercial logic is visible in the segment split: for FY2025 the Fresh and Value-Added segment produced US$2,621.9 million, or 61% of net sales, against US$1,490.4 million or 34% from bananas and US$210.0 million from other products. The company sells roughly twice as much value-added fruit as it does plain bananas, and value-added in this context means fruit that has been cut, packed, ripened or otherwise prepared close to the point of sale. Del Monte also holds the physical means to do it: about 11,000 refrigerated containers, four port facilities in the United States and roughly 419 trucks and refrigerated trailers in the United States plus about 241 in the Middle East.
Owning the cold chain is a capital decision, and the balance sheet shows it. Dole plc owns 13 refrigerated vessels, nine of them refrigerated container carriers and four conventional refrigerated ships, with one more on charter, and describes that as the largest dedicated refrigerated containerised fleet in the world. It operates port terminal operations in California, Texas, Mississippi, Delaware and Florida, and of its more than 250 facilities about 75 are pack houses, cold storage and ripening operations, with roughly 160 more acting as marketplace and distribution operations. That is a company choosing to hold the refrigerated corridor itself rather than buy capacity from container lines, and it is only defensible because the volumes are enormous and continuous. The Fresh Fruit segment alone generated US$3,615.127 million of Dole's US$9,172.907 million FY2025 revenue.
Ripening technology can itself become a brand, and Mission Produce shows both the upside and the ceiling. Mission markets its ripening programme under the Mission Control name and holds 49% of Henry Avocado and about 28% of Shanghai Mr. Avocado, which operates four ripening centres in China, giving it a ready-to-eat position in the market that values it most. A frequently repeated claim that Mission operates twelve owned AVO-RIPE ripening centres should not be relied on: no source for that wording could be found, and the company uses no such term. The ceiling on what ripening can protect is equally visible in the numbers. Mission's Q2 FY2026 revenue fell 24% to US$290.9 million because the average selling price per unit of avocado fell 36%, even though volume grew 15% on abundant Mexican supply. A perfect ripening programme does not defend the price of a commodity in an oversupplied season. What it defends is shelf position, repeat purchase and the retailer's willingness to give the brand a second order, which is why the investment continues through a bad price year.
The decision turns on order granularity, delivery frequency, pack formats and who carries the risk of a bad season rather than on headline price, and that is why an integrator with almost no farming of its own can sit on the same shelf as growers many times its size.
An integrator sells availability and assortment, not fruit. Greenyard NV sources about 2.6 million tonnes of fruit and vegetables from more than 80 countries and employs around 8,600 people across 21 countries, but it does not farm them. Its 35 sites split into 12 production sites in the Frozen and Prepared divisions and 23 Fresh service centres that handle ripening, packing, grading and logistics. The Fresh division alone reported EUR 4,356.2 million of segment sales in AY24/25, the year ended 31 March 2025, out of group sales of EUR 5,363.1 million. What a retailer buys from that structure is a supplier that can deliver many categories, in the pack sizes the store planogram requires, every day of the year, on one invoice, absorbing a short crop in Spain and an oversupply in the Netherlands in the same week. The price of that convenience is visible in the margins: Greenyard's AY24/25 EBIT was EUR 61.3 million on EUR 5,363.1 million of sales, a margin of roughly 1.1%, with adjusted EBITDA of EUR 183.0 million or 3.4%, and a net loss for the year of EUR 2.9 million. Integrators survive on volume and working capital, not on unit margin.
A grower sells provenance, consistency and a story the retailer can print on the pack. Buying direct makes sense where the product is differentiated, the season is known and the retailer will commit volumes in advance. Dole plc distributes in more than 85 countries and operates in 30, and its FY2025 sales were spread across the United States at 34%, the United Kingdom at 11%, Spain at 9%, Sweden at 7%, Ireland at 5% and 34% elsewhere, so a direct relationship is really a relationship with a global sourcing machine, not a single farm. T&G Global's decision to exit domestic fresh produce distribution is instructive: on 31 July 2026 it signed and largely completed three transactions disposing of the T&G Fresh businesses, keeping the premium branded apple and intellectual property operations. Distribution is a low-margin, high-touch, local business and a brand is a global one, and T&G chose the brand. Retailers also increasingly prefer one counterparty for a whole category, which is the logic behind the combination of Mission Produce and Calavo Growers completed on 28 May 2026, a platform with a stated annualised synergy target above US$30 million that reported record quarterly revenue of US$450.0 million in Q3 FY2026.
Some of the most-cited integrator relationships are not contracts at all. Greenyard's relationships with major European retailers include an agreement to strengthen a commercial relationship in Belgium and a signed letter of intent for a strategic partnership in Germany covering value-added services such as ripening; neither retailer appears in Greenyard's annual report as a disclosed material customer. Greenyard's prospectus also disclosed that sales outside Europe, including the United Kingdom, were less than 5% of sales in FY23/24, which is inconsistent with the widely circulated suggestion of a sizeable China or Asia revenue line for that company.
The questions to put in a tender. Ask who bears the cost of a rejected pallet, who holds inventory risk across a price collapse, how much stock the supplier can hold in its own cold chain, what share of the category it sources directly rather than through other traders, and which party is contractually obliged to keep the shelf full during a shortage.
Real integration is proven by three separate balance-sheet facts, namely owned acreage, owned pack and processing capacity and owned cold chain, and a trading business can imitate all three in its marketing material while owning none of them, so the test is to read the disclosures by category and check where the revenue actually comes from.
Acreage is the hardest asset to fake, and the disclosures are specific where they exist. Dole plc farms about 110,000 acres of its own production, roughly 100,000 acres of it Fresh Fruit land in Central and South America, a figure specific enough to be checkable. Mission Produce holds 4,000-plus hectares across California, Peru, Colombia, Guatemala and South Africa, growing avocados and mangoes, and owns pack houses in Mexico, Peru and California; the widely repeated claim that these are Peruvian blueberry farms in a single country is wrong on both geography and crop. A buyer should look for a named country, a hectare count and a crop, and treat a claim lacking all three as unverified.
Facility counts must be read by type rather than by total. Del Monte Corporation's FY2025 filing describes 18 fresh-cut processing plants against 31 global distribution centres, four US port facilities and about 11,000 refrigerated containers. Greenyard NV reports 35 sites, but only 12 are production sites in Frozen and Prepared; the other 23 are Fresh service centres for distribution, ripening, packing and logistics, so counting all 35 as processing plants overstates the industrial footprint threefold. Zespri's 39 pack houses and 64 cool stores in the 2025/26 season are partner or contracted facilities rather than wholly owned plants, and Sunkist's packing houses are typically operated by member licensees under its trademark rather than owned by the cooperative. The same care applies to volume: Chiquita owns around 70 banana plantations across five Central and South American countries, but about 50 third-party suppliers account for roughly 60% of its volume, so its fruit is mostly other people's even though its brand and ripening discipline are its own.
Owning the shipping lane in a perishable trade is the single strongest test, and the one almost nobody passes. Dole owns 13 refrigerated vessels plus one on charter and runs port terminals in five US states; Del Monte owns six and charters one. Between them they carry their own fruit across the Atlantic and Pacific on their own schedules, a level of control that cannot be rented on a spot basis when the market is tight. Almost every other company here depends on third-party container lines, chartered vessels or trucking, and says so in its disclosures, which a buyer should read as a declared limit rather than a hidden weakness.
Buying on the open market is a legitimate business, provided it is described honestly. Joy Wing Mau is the largest pure example here. It owns essentially no production, yet it moves more than 2,000 tonnes of fruit a day through 30-plus cold-chain logistics centres into more than 10,000 supermarket stores and 25,000 fruit specialty stores in more than 300 Chinese cities, operates in more than 40 countries, and was licensed 300 hectares of T&G Global's TUTTI apple in 2025. Its scale should be quoted with care: revenue above RMB 20 billion for 2024, roughly US$2.8 billion, is reported through district industry and commerce data and Chinese media rather than audited filings, no verified 2025 figure exists, its mainland listing plan failed and it now targets Hong Kong by 31 December 2027. The checklist is short: named acreage, named pack houses, owned or contracted cold chain stated as such, and a revenue breakdown separating what the company grew from what it resold.
