Close
Markets

Kenya’s fruit wins markets on land but loses on proof

(13 min read)

Kenya’s fruit is winning markets on the strength of its land, its altitude and its calendar, and losing ground on proof. That is the plain message of a World Bank Group Country Private Sector Diagnostic published in September 2026, which names avocado and mango as one of three sectors where private money would flow if a handful of public failures were fixed. This article reads the agricultural chapter closely, adds what has happened in the weeks since, and asks what it means for a farmer, a cooperative or an exporter anywhere in Africa who sells fruit to buyers that want to know exactly what is in the box and where it came from.

Read this first. We read the diagnostic’s full text as you supplied it and use only its agricultural chapter and the economy wide context that bears on farming; the tourism and medical consumables chapters are not relevant here. Its job and investment figures are modelled upper bounds, not forecasts. We added recent trade reporting on Kenya’s 2026 avocado season from FreshPlaza and the Nation. Several figures come from the diagnostic’s own secondary sources, and we say where it is internally inconsistent.

Six findings

  • Kenya grows well. Avocado yields average 18 tonnes a hectare against a world average of 11, and the harvest window falls before Peru’s.
  • Much of the crop never earns. About 25% of avocado is lost after harvest, and up to 35% of mango.
  • Quality is a public goods problem. Immature fruit can be exported because inspections are sporadic and the clearance system does not force checks in order.
  • Pests ignore farm boundaries. Area wide control works at about US$33 a hectare, yet the plan funds two new areas for Sh160 million.
  • Mango is mostly a domestic crop. Only about 3% is exported fresh, while processing is the nearer opportunity.
  • The upside is large but conditional. Up to US$167 million of investment and 36,000 better paid jobs over ten years.

The economy around the orchard

Agriculture is about 20% of Kenya’s GDP but grew only 2.5% a year between 2015 and 2024. Private credit fell from 36.7% of GDP in 2015 to 29.1% in 2024 as government borrowing crowded it out, and exporters and processors of fruit borrow at about 18%. Climate shocks cost an estimated 2% to 2.8% of GDP a year. Firms also report more water and licensing trouble than peers, and 75% report frequent outages.

Where Kenyan firms feel the squeeze

Share of firms. World Bank Enterprise Survey 2025, as cited in the diagnostic. Water: firms reporting insufficient supply. Licensing: firms calling it a major or very severe constraint

0%15%30%45%KenyaLower-middle-income averageFirms short ofwaterKenya, Firms short of|water: 37%37%Lower-middle-income average, Firms short of|water: 17.2%17.2%Licensing and permitsa major constraintKenya, Licensing and permits|a major constraint: 25.3%25.3%Lower-middle-income average, Licensing and permits|a major constraint: 15.9%15.9%

A strong orchard, a leaky chain

Kenya is Africa’s largest avocado exporter. Output and export earnings roughly tripled over a decade to 585,000 tonnes and US$162 million in 2024, and the area harvested grew 15% in a year. Yet about a quarter of the crop is lost after harvest through poor handling and thin cold chains, and of the rest most is eaten at home. Only about 22% of what remains is exported fresh, and the processors’ 8% is mostly export rejects.

Where a tonne of avocado goes

Share of output. Diagnostic’s shares applied to total output: 25% lost, then 70%, 22% and 8% of the remaining 75%. The 52.5%, 16.5% and 6% are our arithmetic

Lost after harvest: 25%Used: 75%Lost after harvest: 25%Sold in Kenya: 52.5%Exported fresh: 16.5%Processed (mostly oil): 6%585k t25% lost after harvest52.5% eaten in Kenya16.5% exported fresh6% processed, mostly oil, from export rejects

The farm base is overwhelmingly small. The diagnostic counts more than 770,000 small farmers with under 20 trees each producing about 70% of output, about 145,000 medium farms producing 20%, and the rest from large estates. Small farms have weaker agronomy, little extension or irrigation, and sell through informal brokers, while large farms are often vertically integrated.

Who grows Kenya’s avocados

Share of output by farm size. Farm counts as stated in the diagnostic (Ministry of Agriculture, 2026)

70% of output70% of outputOver 770,000 small farms, under 20 trees each20%20%About 145,000 medium farms10%10%About 48,000 large farms

The competitive edge is real. Kenyan yields beat the world average by about 60%, and mid altitude orchards harvest from March to October, before Peru peaks. Kenya’s landed unit values into the EU were about 15% below Peru’s in 2024, which partly reflects quality discounts on immature fruit.

Why the land is the advantage

Tonnes per hectare. Diagnostic citing FAOSTAT and Ministry data; large farm figure is “above 22”

0510152025Avocado|Kenya average: 18 t18 tAvocadoKenya averageAvocado|global average: 11 t11 tAvocadoglobal averageAvocado|large Kenyan farms: 22 t22 tAvocadolarge Kenyan farmsMango|Kenya average: 12 t12 tMangoKenya averageMango|global average: 10 t10 tMangoglobal average

The immature fruit trap

The diagnostic’s sharpest finding is about enforcement. Rejection rates for immature fruit in premium markets reached as high as 30% in 2025, and a 24% dry matter threshold set by the Horticultural Crops Directorate in 2022 is widely breached. Oil processors that buy rejected fruit at dry matter as low as 18% set a floor under early harvesting, and exporters who cut corners can close and reopen under new names. The report finds three failure points: clearance is often granted without inspection, samples are prepared in advance by exporters, and the shared digital platform does not force steps in sequence, so a plant health certificate can be issued before any quality check.

Recent events show how costly this is. Industry reporting says the immature fruit losses exceed Sh7.5 billion (about US$57 million) a year. During 12 closed season weeks in 2026 the directorate approved 33,205 tonnes worth US$38.8 million, according to KenTrade data reported by FreshPlaza. The diagnostic describes that as about a third of normal annual volumes, and our arithmetic against roughly 110,000 tonnes exported in 2024/25 gives about 30%. On 28 October the Agriculture and Food Authority closed the 2026/27 season for sea exports from 30 October after a survey found low volumes, while keeping air exports open subject to inspection and traceability for every consignment. The Nation reports that avocados earned Sh20.8 billion in 2025.

The recommended fix is simple in outline: inspector drawn random samples at every packhouse, with results logged on the National Horticulture Traceability System, which is still under development, and linked so that no consignment moves forward without a logged pass. The report calls packhouses the most feasible first step toward farm to port traceability, and notes Peru already inspects at the packhouse gate as a precondition to certification.

The sequence the report proposes

Recommendations 1a and 1b of the diagnostic; layout is ours

PackhouseInspector draws sample;dry matter loggedPlant healthCertificate only witha logged passCustomsRelease only withsystem documents

Pests do not stop at the fence

The persea mite, a recently identified quarantine pest, is projected to cut avocado harvests in Murang’a County by about 50% in 2026. The county supplies 42% of avocado exports. For mango, fruit fly and the seed weevil cause combined yield and postharvest losses estimated at 35%, and repeated fruit fly interceptions led Kenya to suspend fresh mango exports to the EU in 2014 and resume only in 2021. The diagnostic argues that mandates are split among KEPHIS, county governments and others with no statutory way to fund emergency response, and that county extension varies. Area wide control works: low pest prevalence areas cut fruit fly populations by over 85% in Makueni and lifted mango processing yields from 8% to 14%, for about US$33 a hectare. Yet the government’s agricultural compact targets only two further areas with Sh160 million that is not yet in the budget, while about Sh10 billion went to input subsidies in fiscal 2025, largely for fertiliser.

Priorities in shillings

Diagnostic figures: input subsidies fiscal 2025 and the budget for two new low pest areas. The 1.6% share is our arithmetic

Input subsidies, fiscal 2025Sh10bnTwo new low-pest areas, plannedSh160m (1.6%; our arithmetic)

Mango: a big crop that mostly stays home

Mango output reached 783,000 tonnes in 2024 from nearly 800,000 farmers. Most is Apple and Ngowe, varieties for the home market, while export varieties are under 15% of output. About 80% passes through informal brokers, postharvest losses reach 35%, and only 3% is exported fresh, worth US$14 million. Red Sea disruption has stretched Mombasa, Rotterdam transit from about 25 to 45 days, longer than fresh mango lasts. The diagnostic therefore favours pulp for domestic and regional juice makers: six processors have about 166,000 tonnes of capacity, running at 40% because the season lasts about three months, and regional pulp and juice exports doubled from US$8.3 million in 2015 to US$16.4 million in 2024.

Where a tonne of mango goes

Units per 100 of output. Diagnostic: losses up to 35%; of survivors 87% sold fresh at home, 10% processed, 3% exported. Resulting values are our arithmetic and treat losses as the upper bound

0255075100Output (2024): 100100Output(2024)Lost on the way: −35−35Lost onthe waySold fresh in Kenya: −56.6−56.6Sold freshin KenyaProcessed (pulp, dried): −6.5−6.5Processed(pulp, dried)Exported fresh: 22Exportedfresh

The diagnostic’s executive summary states the 10% and 3% shares against total production, while the chapter applies them to what survives; we use the chapter’s version. Margins explain the pull of each channel. Air freighted fresh mango to Europe earns about 63% gross margin, against 25% to 45% for exported dried mango and 25% to 35% for pulp sold in Kenya, but volumes by air are too small to attract investment. A December 2025 pilot shipped about one tonne of Apple mango to the UK by air, which the report says was donor funded and not easily replicated.

Margins by channel

Gross margins from stakeholder interviews cited in the diagnostic. Fresh air freight is a single point estimate

Fresh mango, air freight to EU63%Dried mango, exported25–45%25 to 45%Mango pulp, sold in Kenya25–35%25 to 35%

What the buyers pay for

Kenyan avocados sell for roughly Sh40 to 60 a kilogram at the farm gate and in informal markets, against Sh178 to 320 for export grade fruit in Dubai and Europe, which is why the export channel pulls so hard. Europe takes 70% of avocado export proceeds, with the Netherlands alone at 31% of exports, and the Middle East 29%, where prices are lower and Kenya is the leading supplier.

Farm gate against export price

Shillings per kilogram. Diagnostic footnote 31; ranges as stated

Farm-gate and informal market, Kenya40–60 Sh/kg40 to 60 Sh/kgExport-grade, Dubai and Europe178–320 Sh/kg178 to 320 Sh/kg
Who buys Kenya’s avocados

Share of export proceeds by region, with main countries as shares of total exports. The diagnostic gives country shares of all exports

Europe: 70%Middle East: 29%Other: 1%70%Europe share of export proceedsNetherlandsNetherlands: 31%31%SpainSpain: 12%12%FranceFrance: 11%11%GermanyGermany: 10%10%UAEUAE: 17%17%Saudi ArabiaSaudi Arabia: 3%3%TürkiyeTürkiye: 3%3%

The size of the prize

The modellers assume reforms would lift avocado area growth to 9% a year, 2.6 points above the 2021 to 24 baseline and equal to Peru’s growth after it formalised inspection. They assume export intensity rises 20 points from 22.2% to 42%, and apply a 39% export wage premium found in earlier research on Kenyan avocado farmers. That yields up to US$160 million and 30,000 better paid jobs for avocado, and US$7 million and 6,000 for mango. The report is clear these are relative to a business as usual path and depend on wider reforms.

How far export intensity would have to move

Share of avocado output exported. Kenya 2024 and the reform scenario: diagnostic modelling. Peru: “over 60%” per the diagnostic

0%10%20%30%40%50%60%70%Kenya|2024: 22.2%22.2%Kenya2024Kenya with|reforms (model): 42%42%Kenya withreforms (model)Peru|(above): >60%>60%Peru(above)

Where verified field data fits

The diagnostic ends its fruit chapter on a point that matters beyond Kenya: fragmented smallholders and heavy intermediation make it hard for buyers to invest in farmers in exchange for traceable, quality supply, and cooperative capacity building is one remedy. Packhouse inspection fixes the quality gate, but the plot where fruit grew sits upstream. AgriKonnekt is built to check farm plots with sensors and satellites so cooperatives can show results to funders, as described on the AgriKonnekt page. We have not seen it applied to avocado traceability or to Kenya’s traceability system, so that role is our proposal.

What to do next

  1. Cooperatives and aggregators should record where every consignment was grown and when it was picked, so mature, traceable fruit can be shown to a packhouse.
  2. Farmers should wait for maturity checks before harvest, since early sales risk licences and market access for everyone.
  3. Exporters should test dry matter at their own packhouse gates, as the report recommends doing at national scale.
  4. Counties should organise neighbouring farms into area wide pest control, since individual spraying cannot stop pests that spread across fences.
  5. Policymakers should link inspection results across the three agencies, pass the Plant Protection Bill and defer import VAT on processing equipment, as recommended.
  6. Processors and funders should look first at mango pulp for regional markets, where the diagnostic finds the nearer opportunity.

What we can and cannot claim

StatementEvidence status
Avocado output of 585,000 tonnes and US$162 million in 2024World Bank diagnostic citing FAOSTAT and Harvard Atlas; other reports give different 2025 totals
25% of avocado lost after harvest; mango losses up to 35%Diagnostic citing ministry and academic sources
Rejection rates as high as 30% in 2025Authority figures cited in the diagnostic and trade press
33,205 tonnes approved during the 2026 closed seasonKenTrade data as reported by FreshPlaza
Mango shares of 10% processed and 3% exportedApplied to survivors in the chapter and to total output in the summary
US$167 million investment and 36,000 jobsModelled upper bounds; assumptions stated in the appendix
Plot level data could support packhouse traceabilityOur argument; not tested

Kenya’s fruit story is a reminder that in modern food trade a crop’s value is set as much by trust as by yield. The land and the calendar are an advantage that rivals cannot copy, yet quality leaks, pests and slow processing keep too much of that value from reaching farmers. The remedies the diagnostic proposes are modest in cost compared with the gains: inspect where the fruit is packed, record what is found, tie agencies to one system and control pests across whole areas. Cooperatives and exporters that can prove where and when their fruit was grown will be best placed when buyers, regulators and funders start paying for proof.

Sources & notes

SOURCES: World Bank Group (IFC and partners), Kenya: Country Private Sector Diagnostic (September 2026); FreshPlaza, Kenya enforces stricter avocado export rules as season opens; FreshPlaza, Kenyan avocado exports during ban trigger shortages; FreshPlaza, sea export suspension from 30 October; Nation, Kenya shuts avocado exports by sea; Kenya Times on the Agriculture and Food Authority notice; FreshPlaza on the 2026 season rules; AgriKonnekt.

Differences between sources: the diagnostic’s summary and chapter apply mango processing and export shares to different bases; closed season avocado exports of 33,205 tonnes are called about a third of annual production in the diagnostic, while they equal about 30% of 2024/25 exports on our arithmetic; avocado output is 585,000 tonnes in 2024 in the diagnostic and 694,000 tonnes for 2025 in other trade reporting.

Our own arithmetic: the 52.5%, 16.5% and 6% avocado shares; the mango waterfall values; the 1.6% budget ratio.

Not read: the diagnostic’s forthcoming companion reports, the Plant Protection Bill, the underlying Ministry strategies and KenTrade data. Modelled job and investment figures are the World Bank’s.