Posts
A private-sector framework for food security and competitive agriculture in Trinidad and Tobago.
Before we spend another dollar, can we agree on the numbers?
A plain-language summary of Making It Investable: A private-sector framework for food security and competitive agriculture in Trinidad and Tobago, 2026–2036 — a position paper published by Multicrop Facility Ltd on 8 August 2026.
Disclosure: this is an industry advocacy document, not a government or academic study. It is summarised here on its own terms. Every figure below is the paper's, and where the paper itself flags a number as uncertain, that flag is carried over.
The one thing the paper is actually asking for
Not a new strategy. Not a new fund. One thing: stop committing capital against numbers nobody has reconciled — and publish the reconciliation.
The paper's claim is that four basic facts about agriculture in Trinidad and Tobago are currently unsettled: what share of food is imported, how much corn and feed the country buys, how much state agricultural land is actually being farmed, and whether farming is profitable. All four, it argues, could be settled within months from data the State already holds.
Unusually for an advocacy paper, it tags every claim it makes — verified, reported-but-uncorroborated, contested, or its own judgement — and includes an annex titled What we do not know. Several figures the authors had used in their own earlier submissions have been cut for failing that test.
The commercial case, in short
The food import bill rose from about TT$5.7 billion in 2019 to roughly TT$7.2–7.3 billion across 2022–2024, while population stayed near 1.37 million. The country is paying materially more foreign exchange for the same amount of food, at a moment when foreign exchange is the binding constraint on nearly every business in the country.
Note what the argument is not. It is not about agriculture's share of GDP, which is under one per cent and, the paper accepts, will stay there. It is about foreign-exchange cost and supply fragility — which makes it a manufacturing, retail, logistics and banking problem as much as a farming one.
Government's target — a 25 per cent reduction in the food import bill by 2030, supported by a three-year Priority Commodities Programme — has the paper's support. Its three warnings about measurement deserve the attention of anyone who will be judged against that target:
An import bill falls when world prices fall. Report it nominally and at constant prices, and publish the difference.
An import bill also falls when people eat less. Read it alongside the annual CARICOM–WFP food security survey.
Replacing an import with a domestic product that runs on imported feed, fertiliser, fuel and packaging can cut the bill while barely moving net foreign-exchange use. The paper argues that net foreign exchange saved per commodity — not gross import value displaced — should govern investment, and says no published national analysis appears to calculate it.
The figures it refuses to use
The paper devotes a section to numbers in wide circulation that it declines to cite, including several from its own past submissions:
"80 / 85 / over 96 per cent of food is imported." The spread is too wide to be measurement error; by value, by calorie, by tonne and by retail basket give different answers, and sources rarely say which they mean.
Feed above TT$200m and corn close to TT$256m annually (both from ministerial statements). Corn is a subset of feed, so as stated these cannot both be right — and the entire feed-substitution case rests on the answer. Customs declarations would settle it in weeks.
The agri-export target, which circulates as both TT$1 billion and US$1 billion. Nobody can plan an export business against a target uncertain by a factor of roughly seven.
"About 16 per cent" plot utilisation, from an unpublished Cabinet report referenced in 2021.
The land question
Here the paper's strongest material is also its oldest. Published Caribbean research (Persad, 2004; Persad, Rampersad and Roop, 2009) establishes that Caroni (1975) Ltd controlled about 31,000 hectares — roughly a quarter of the country's best arable land — of which 12,158 hectares were released for diversification; that 67 per cent of the holdings are good-quality Class III/IV land and 33 per cent is marginal; and that 2008/09 field surveys found under 3 per cent of the developed two-acre plots in productive agriculture.
Three conclusions follow. Utilisation moving from under 3 per cent to around 16 per cent is a fivefold improvement from a near-zero base — neither the scandal it is usually presented as, nor good enough. The "production zones" idea now being rediscovered was set out, with a land-capability rationale, in 2009; the obstacle has never been the idea. And any plan treating the full historic acreage as a production base overstates the addressable land by about a thirdbefore drainage, tenure or boundary problems are counted.
Five recent changes not yet in anyone's plan
Customs duty on poultry, cattle and pig feed was removed from 1 January 2026. Good for livestock producers, bad for the domestic feed-crop case — the paper's point is that the second half is not being said.
Guyana is positioning to supply the region in corn, soya, red beans and black-eyed peas. If true, "grow it or import it" becomes a three-way choice, with regional sourcing retaining foreign exchange inside CARICOM. (The paper flags this as single-source and asks readers to press it.)
The 2026 Caroni pilot produced results — six acres, germination above 96 per cent, harvest at 56–60 days, corn harvested 18 July 2026. The paper calls this a promising agronomic result and not an economic one: no yield per acre, no cost of production, no farm-gate price. Its warning to farmers is blunt — if a six-acre trial is scaled nationally and the margins never appear, growers wear the loss.
Roughly US$1.76m in donor money has already been spent on two completed FAO/GCF data and adaptation projects, covering gaps that new strategies propose to commission afresh.
The UN and FAO funding windows are being rewritten now, making the successor country programme the natural vehicle for an agricultural census.
The cheapest win, and it is time-limited
The last full agricultural census was in 2004. Every production, yield and land-use figure in circulation is an extrapolation from a twenty-two-year-old frame. The Central Statistical Office is currently in the field with a Geospatial Data Update Exercise, collecting geo-located data on buildings, households, businesses and agricultural holdings. Capturing agricultural attributes while enumerators are already at the door would be, the paper argues, the cheapest national farm frame the country will ever be offered — and the window closes when fieldwork ends.
What it asks of industry — not just Government
The paper's twelve requests of Government all draw on data or authority the State already holds: publish commodity-level Customs data; publish the fifteen priority commodities; fix the export target in one currency; publish the full pilot economics; publish a drainage responsibility matrix across the Ministry, Works and WASA before more drainage capital is committed; and legislate one annual agricultural performance report laid in Parliament against a fixed indicator set.
But it also puts industry on the hook: stop using the discredited figures; publish contracted volumes, because demand uncertainty is the constraint farmers name most often and it is largely the buyers' to fix; adopt a standard contract-farming instrument covering price formula, rejection, payment timing and dispute resolution; and fund the farm economics survey if the State will not — because the median gross margin per hectare is the single most important missing number in the sector, and nothing else can be sequenced without it.
