The claim.
The 2027 Budget Statement says international cocoa prices "recovered" from an average of US$3,980 a tonne in the second quarter to US$5,830 in the third, "reaching US$5,920 in September", and that the recovery "offers prospects for improved export receipts".
The data. The International Cocoa Organization's daily price was US$4,906.39 a tonne on 10 November (about GHS 57,454 at the Bank of Ghana's 12 November mid-rate of GHS 11.71). That is 17.1% below the September figure the budget cites, and 15.8% below its third-quarter average.
The verdict: outdated. The budget's table rests on World Bank data dated 2 October. Since then the market has turned.
ICCO's price fell US$238.12, or 4.6%, on 10 November alone. It averaged US$5,741.81 in October and US$5,326.22 over the first seven trading days of November. It is 27.6% below its 31 August high of US$6,780.73, and 17.9% below the US$5,977.43 recorded a year earlier.
Why it matters. Cocoa earned Ghana US$3.0 billion in the first nine months of 2026, up 15.9% on a year earlier, according to the same statement. Gold, cocoa and crude together made up 90.4% of export receipts.
Farmers are paid a fixed GH¢42,400 a tonne for 2026/27. COCOBOD says that is 71.18% of the realised free-on-board price, which implies a realised price of about GHS 59,567 a tonne, roughly US$5,087 at today's rate. The world price now sits about 3.5% below that implied level.
That is StatsGH arithmetic; COCOBOD has not published the exchange rate it used.
The counter-view. Spot prices are not what Ghana earns on cocoa already sold forward. And the budget is right that this year has been worse: the statement's own year-to-September average, US$4,543.15, is 8.0% below today's price.
So cocoa is dearer than its 2026 average, but cheaper than the "recovery" the budget banks on. Volumes may not help: StatsGH reported on 10 November that a crop scientist expects output to fall by at least 16% this season as dry weather hits the cocoa belt.
Lower prices and a smaller crop would compound each other.
