The economics of cassava are easy to state and hard to win. An acre of cassava can pay well when prices are high, as in 2024, but margins swing widely with the root price, costs are dominated by labour, and the published farm budgets are few and dated, so the figures here are indicative and each names its basis.
The two levers that move the margin are yield and the price you sell into, which is why processing matters more than penny-pinching. Finance is both the enabler and the bottleneck. Treat the farm as a high-variability business that rides on the root price, or move up the value ladder (Pillar 6), where the margin is steadier.
It depends, and the range is wide. A 50-farmer study in Sekyere East, Ashanti's highest-yielding cassava district, found a gross margin of only about USD 23 an acre, a razor-thin return, though the paper also states the figure in cedis, so its currency is uncertain.1A larger study across the Eastern and Brong Ahafo regions put net farm income at about GH¢4,457 per hectare on a 75 percent profit margin, with net income about twelve times the capital invested, with farmers running at 86.4 percent profit efficiency, so the crop is profitable but farmers lose about 13.6 percent of attainable profit to inefficiency, linked in the study to education, experience and membership of a farmer group.2
The sharpest finding is about labour. A value-chain study in Brong Ahafo found that smallholders who pay the market wage for all labour lose money, about GH¢490 an acre on the study's headline measure (GH¢172 on the like-for-like measure used for commercial farms), while commercial farms make about GH¢1,505 an acre, mainly because they harvest far more per acre.3 Those are 2016 figures, so the cedi values are dated; the pattern, not the number, is the point. What this means is that at low root prices whether cassava makes money at all can turn on whether the grower uses family or hired labour and how well the plot yields, while at the high prices of 2024 the same acre paid well.
| Item (per acre) | Indicative GH¢ | Note |
|---|---|---|
| Land rent | 300 | southern-belt, indicative |
| Land preparation (clearing, ploughing) | 300 | |
| Planting material (cuttings) | 200 | |
| Labour and weeding | 1,050 | the dominant cost, over half the total |
| Total variable cost | 1,850 | farmer-reported, Ashanti 2024 |
| Yield (tonnes an acre) | 8 | Kleih / GASIP (2016) basis, about 20 t/ha |
| Farm-gate price (GH¢ a tonne) | about 2,000 | implied by the acre sold standing (2024) |
| Revenue, crop sold standing | 16,000 | farmer-reported, Ashanti 2024 |
| Gross margin | 14,150 | 2024 prices, before valuing family labour |
| Gross margin at the 2016 root price (GH¢260 a tonne) | 230 | stress test: the same costs after a price collapse |
| Break-even farm-gate price (GH¢ a tonne) | about 231 | at 8 tonnes an acre and the 2024 cost |
| Net margin, hired-labour smallholder | (490) | can go negative at market wages (2016 basis) |
| Net margin, commercial farm | 1,505 | scale and lower unit costs (2016 basis) |
Sources: Amos, Agrighanaonline (2024, farmer-reported); Kleih / GASIP (2016); playbook calculation for the margins.
The cost of an acre is dominated by one thing: labour. A 2024 farmer-reported budget in Ashanti put the cost of an acre at about GH¢1,850, of which labour and weeding alone were GH¢1,050, more than half, with land rent, land preparation and planting material making up the rest.4 Figure 12 shows the split.
Ghana is a high-cost place to grow cassava for structural reasons: most operations are still done by hand (Pillar 4), the perishability clock forces a quick and often distress sale (Pillar 3), and inputs were expensive through the 2022 to 2025 inflation surge, with food inflation still 26.5 percent in early 2025.5 The nominal rural wholesale price looks like it is racing up, from about GH¢527 a tonne in 2015 to GH¢5,045 in 2024, but in real, inflation-adjusted terms the 2024 price was essentially flat on 2023, so the 2024 jump was inflation rather than a real gain, although over the decade as a whole the real price still rose about six-fold.6
Costs and prices also move with the wider economy, in both directions. Inflation eased from 23.5 percent in January 2025 to 3.2 percent in March 2026, the lowest reading since the consumer price index was rebased in 2021, then edged back up to 5.0 percent by August 2026, with food inflation at 3.0 percent.7 Cassava needs few bought inputs, but the fertiliser and agrochemicals it does use are imported, so their cost follows the exchange rate: the cedi lost about 9.5 percent against the US dollar in the first nine months of 2026, and the World Bank expects fertiliser prices to rise about 31 percent in 2026, with urea ending the year about 60 percent above 2025.8 Borrowing is still costly, with the Bank of Ghana policy rate held at 14 percent in July 2026.9 On the price side, the planned surplus (Pillar 2) can pull the farm-gate price down in the same season that input costs rise, and a cost build that holds one season can be wrong the next. Re-cost inputs every season, hold a buffer against both a price fall and a cost rise, and prefer offtake contracts that fix the price or tie it to costs.
Labour and weeding is GH¢1,050 of a GH¢1,850 acre, more than half, with land rent, land preparation and planting material making up the rest, from a 2024 farmer-reported budget in Ashanti.
The margin rides on the two revenue drivers. On an indicative model, yield and farmgate price move the margin far more than any cost line. At the 2024 farmer-reported basis (an acre sold standing for about GH¢16,000 against about GH¢1,850 of cost), a 20 percent fall in price or yield cuts the margin by about a quarter, while a 20 percent rise in labour cost moves it by under 2 percent (Figure 13). At the much lower root prices of 2016, the same 20 percent swing in price could erase the margin entirely, which is why price risk is the farmer’s main exposure, well ahead of cost.
Yield and farmgate price rank equal first, swinging the margin by about 23 percent for a 20 percent move at the 2024 basis; labour and weeding is third at about 1.5 percent, land preparation fourth at about 0.4 percent, and planting material fifth at about 0.3 percent.
Effort and money are best spent on the things that move the margin most: raising yield through clean planting material (Pillar 1) and agronomy (Pillar 4), and lifting the price realised through processing and branding (Pillars 2 and 6). Shaving the smaller cost lines barely moves the result. It also means the farmer is dangerously exposed to a price fall, which is exactly the glut risk the planned surplus creates (Pillar 2).
Numbers on a page are one thing. Move the levers yourself. Set the yield, the farmgate price and the labour bill to your own plot’s figures, then swing them one at a time. Push the yield or the price slider and watch the margin move fast, then push the labour slider and watch it barely stir. Then drag the price down toward the 2016 level of GH¢260 a tonne and see how quickly the margin disappears: price risk is the grower’s main exposure.
Slide the three levers a grower can hold: the yield in tonnes an acre, the farmgate price you sell into, and the labour and weeding bill, the largest cost. The model starts at the 2024 farmer-reported basis, an acre sold standing for about GH¢16,000 against about GH¢1,850 of cost. Drag the price down toward the 2016 level and watch how fast the margin goes.
At the 2016 root price (GH¢260 a tonne) the same costs leave about GH¢230; break-even is about GH¢231 a tonne at 8 t/acre.
Base case, per acre: yield about 8 t/acre (Kleih / GASIP 2016) at about GH¢2,000 a tonne gives revenue of about GH¢16,000, the farmer-reported 2024 standing-crop sale; taking off the variable cost of about GH¢1,850 (land rent 300, land preparation 300, cuttings 200, labour and weeding 1,050) leaves a gross margin of about GH¢14,150 an acre. Sources: Amos, Agrighanaonline (2024, farmer-reported); Kleih / GASIP (2016); playbook calculation for the margins.
The same model, re-run as a sensitivity read. Each bar is how far the margin moves when that one driver changes by 20 percent, holding the others where you left them. Two drivers lead, and neither of them is a cost line.
How many tonnes the acre actually brings in, driven by clean planting material (Pillar 1) and agronomy (Pillar 4). It is the joint-widest lever and the one the grower can build, so it moves the margin far more than shaving any cost line.
The market sets this price and it is volatile: about GH¢2,000 a tonne is implied by the 2024 standing-crop sale, against GH¢260 in 2016. It moves the margin as much as yield does, and lifting it runs through processing and branding (Pillars 2 and 6) rather than through selling a raw root at the gate. It is also the exposure the glut risk of Pillar 2 creates.
The largest cost line, over half the bill, and the one the operator can actually cut through mechanised weeding, canopy-closing spacing and shared equipment (Pillar 4). At the 2024 basis a 20 percent change moves the margin by under 2 percent, so it is worth cutting but it is not where the margin is won.
Clearing and ploughing. A shared tractor or ploughing service can trim it, but it is a small line and it barely moves the margin.
The cuttings. Clean, certified material earns its keep by avoiding replant and disease loss rather than by being cheap; as a cost line it is the smallest swing of all.
Sensitivity holds two levers where you left them and moves the third by 20 percent: yield and price both act on revenue, so they lead and tie; labour and weeding is the largest but controllable cost and swings the margin far less; land preparation and planting material are small lines that barely move it. Ranks and swings match the Sensitivity sheet of the Cassava Cost and Returns Model v2 (2024 basis).
Sensitivity says what moves the margin; cost-cutting says what the operator can actually control. Table 10 sets out the controllable cost lines, the lever on each, an indicative saving, and how to check it on your own plot.
| Cost line | Indicative base | The lever | Indicative saving | How to check on your plot |
|---|---|---|---|---|
| Labour and weeding | GH¢1,050/acre | Mechanised weeding and herbicide, plus spacing that closes canopy sooner (Pillar 4); shared equipment as a service (Pillar 4) | ~30% | Time the labour before and after |
| Land preparation | GH¢300/acre | A shared tractor or ploughing service in place of hired clearing gangs | ~20% | Compare a hire quote against manual |
| Planting material | GH¢200/acre | Clean, certified cuttings that establish better and resist disease (Pillar 1) | avoids replant and disease loss | Track establishment and survival rate |
| Harvest and haulage loss | value lost | Process near the farm to beat the perishability clock (Pillar 3) | ~15% of value | Weigh what spoils or is discounted |
| Working-capital cost | 12 to 30% interest (2019) | A GIRSAL-guaranteed or group loan in place of an informal lender | a lower rate | Compare your borrowing terms |
Source: playbook synthesis, anchored to the cost breakdown in Table 9 and the levers in Pillars 1, 3 and 4.
Finance is where cassava economics is won or lost, and it is the hardest piece. In 2019 commercial agricultural lending in Ghana ran at 12 to 30 percent interest, and borrowing costs have swung widely since, with the Bank of Ghana policy rate at 14 percent in July 2026.10 Agriculture takes only a small share of bank credit, 3.5 to 4.7 percent on the Feed Ghana Programme's own two baselines, which the government plans to lift to 10 percent by 2028, with loans disbursed under GIRSAL guarantees targeted to rise from about GH¢289 million a year in 2024 to GH¢824 million a year by 2028.11
The main instrument to close the gap is GIRSAL, the state risk-sharing scheme, which guarantees up to 70 percent of a loan's principal and by mid-2026 had guaranteed loans worth about GH¢1.85 billion over six years, supporting 6,695 agribusinesses through 40 partner financial institutions in 88 districts.12 It lists roots and tubers, cassava included, among its eligible sectors.13 The structure exists and is growing, but GIRSAL publishes no breakdown by crop or farm size, so how much of it reaches cassava smallholders is unknown. Digital-credit trials in southern Ghana raised farm investment but did not automatically lift yields or incomes, a reminder that credit alone is not the answer.14 The opening is to package supply: an aggregator or processor that bundles smallholders under an offtake contract and a GIRSAL guarantee turns a scatter of unbankable farmers into a lendable business, the same clean-supply-plus-buyer model that runs through Pillars 2 and 6.
The margin rides on revenue, so a fall in yield or farmgate price, the glut risk of Pillar 2, shrinks it fast, and at the low root prices of 2016 it could erase it entirely.
Every grower selling at the spot price, and every lender or investor whose case does not stress-test a 20 percent fall in yield or price.
contract a price or offtake before planting; lift yield; move volume up the value ladder into a processed product.
Paying the market wage for weeding can turn a subsistence plot’s margin negative, about GH¢490 an acre in one study.
Smallholders who pay the market wage for weeding, whose plot can run at a loss, and anyone scaling hired labour without a yield and price to carry it.
use family or exchange labour, or mechanise; do not scale hired labour without a yield and price to carry it.
Agriculture gets a small share of bank credit at high interest (12 to 30 percent in 2019), and guarantee schemes publish no data on how far they reach the individual smallholder.
Individual smallholders borrowing alone at high rates, and every plan that assumes a guarantee scheme will reach a lone farmer.
borrow as an aggregated, offtake-backed group under a GIRSAL guarantee rather than as a lone farmer.
Cassava is a high-variability crop at the farm: from a loss for hired-labour smallholders to about GH¢1,500 an acre for commercial farms at 2016 prices, and far more when prices spiked in 2024.
Labour and weeding are more than half the cost of an acre (about GH¢1,050 of GH¢1,850); Ghana is high-cost because it is hand-powered.
The margin is made off the farm: yield and price swing it far more than any cost line, so raise yield (Pillar 1) and lift price through processing (Pillars 2 and 6).
The cost model here is indicative and each figure names its basis and year; the returns depend above all on the root price and yield in the season you sell.
Finance exists in structure (GIRSAL, roots and tubers eligible), but its reach to smallholders is unmeasured; the route to bankability is an offtake-and-guarantee package around aggregated smallholders.
Where this connects.The yield that drives the margin is Pillar 1’s; the price that drives it is Pillars 2 and 6’s, since processing is where the real margin sits; and the market intelligence and the policy that could unlock the sector at scale are the subject of Pillars 7 and 8.