How to Structure an Agricultural Business in Kenya Without Holding the Land

Foreign investors setting up an agricultural operating business in Kenya, an avocado orchard, a flower farm, a dairy operation, almost always propose the same structure first: a Kenyan partner or vehicle holds the land, and a separate company the investor controls leases it and runs the farm. It is a reasonable instinct. It is also usually the exact transaction the Land Control Act was built to stop.

We have covered the constitutional and company-registration background for foreign-investor farm setups elsewhere, and the export-licensing mechanics separately. This piece focuses on the part that sits between those two questions: how to actually structure the relationship with the land once a straightforward lease to the operating company will not clear consent.

Why the Direct Lease Usually Fails

Article 65(1) of the Constitution of Kenya, 2010 restricts a non-citizen to leasehold tenure, capped at 99 years. Article 65(3)(a) extends the restriction to companies: a company only counts as a citizen if every share is held by citizens. A foreign investor holding even one share in the operating company is enough to make that company a non-citizen for this purpose.

For agricultural land specifically, the Land Control Act (Cap 302) adds a second, harder obstacle. Section 6 makes a lease, sale, transfer, or comparable dealing with agricultural land inside a declared land control area void unless the Land Control Board consents. Section 9(1)(c) removes the board’s discretion in exactly this scenario: it must refuse consent where the land, or a share in a company owning the land, is to be disposed of to someone who is not a citizen, or to a company that is not wholly citizen owned. Most productive farmland in Kenya sits inside a declared land control area.

So the board is not being unhelpful when it refuses the lease. It is applying the statute as drafted. A presidential exemption exists under section 24, but it is discretionary, granted case by case through a Gazette notice, and not something to build a farming timeline around.

How Land Control Board Consent Actually Works

It helps to understand the fuller test the board applies, not just the non-citizen bar. Section 9(1)(a) and (b) direct the board to weigh the effect on economic development and good husbandry, and to generally refuse consent where the proposed transferee is unlikely to farm the land well, already holds sufficient agricultural land, or where the transaction’s terms are markedly unfair to one party. Section 9(1)(c)’s citizenship bar sits alongside these factors, and unlike them, it is not a matter of the board’s judgment. If a refusal is issued, an applicant can appeal to the provincial land control appeals board within 30 days, and from there to the central land control appeals board within a further 30 days. Neither appeal changes the underlying citizenship bar, which is why appealing a refusal caused by section 9(1)(c) rarely succeeds; the appeal route is more useful where the refusal turned on the board’s discretionary factors than on citizenship itself.

What to Do Instead

The fix is not to find a cleverer way to give the operating company an interest in the land. It is to stop trying to give it one at all.

Keep land ownership with the Kenyan party, and structure the operating company’s return contractually. A supply agreement, a profit share, or a farm-management contract gives the investor’s company its commercial return without any lease, sublease, or other registrable dealing with the land. Nothing about that arrangement engages section 6.

Protect the capital investment through a separate security instrument, not through the land. Trees, irrigation systems, and farm infrastructure are real, recoverable assets whether or not the company holds an interest in the ground beneath them. Kenya’s Movable Property Security Rights Act, 2017 allows a security right to be created over tangible movable assets, including a standing crop, registered on the online collateral registry it established. Combined with contractual step-in rights, compensation-on-termination clauses, and clear contractual title to anything removable, this does the protective work a lease would otherwise have done, without the Land Control Act problem.

Treat it as a governance relationship, not just a land arrangement. The structure depends on an ongoing relationship with whoever holds the title. What happens if that relationship sours. What happens on their death, land passes to their heirs without triggering the Act, since a transmission on death is specifically exempted unless it causes subdivision, but the operating company’s contract with them needs to expressly bind their estate and successors in title. These need to be written down, not assumed.

Get the sequencing right for the next capital raise. Because the land is never inside the operating company’s own asset register, share transfers to future investors should not re-trigger Land Control Board involvement; the Act’s share-transfer trigger under section 6(1)(c) only applies where the company itself owns the agricultural land in question. Build the structure this way from the first round, and a later, larger raise does not reopen the land question.

A Worked Example

Consider a foreign investor developing a modest horticultural pilot on a few hectares, with a local partner who owns or is acquiring the land. Rather than a lease from the partner to the investor’s operating company, the partner retains title throughout. The operating company enters a farm-management and produce off-take agreement with the partner, registers a security right over the standing crop and irrigation equipment under the Movable Property Security Rights Act, and the two parties sign a separate agreement governing what happens to that arrangement on the partner’s death, incapacity, or a breakdown in the relationship. If the pilot succeeds and outside investors join the operating company at a later round, they buy shares in a company that has never held, and does not need to hold, any interest in the land itself.

What This Does Not Solve

This approach trades registered security in the land for contractual and movable-asset security instead. That is a genuine trade-off, not a loophole, and it should be sized to the investment: a modest pilot plot carries this risk differently than a large-scale, multi-decade operation. Whether the trade-off is acceptable for a specific project is a scoping question, not something a general article can answer.