Getting the Structure Right Isn’t Enough: A Land Lesson From a Kenya Tourism Deal

A foreign hospitality investor came to us this year wanting to build a small lodge on private land near one of Kenya’s wildlife areas. The corporate side was straightforward. The land side taught us something worth writing down, because the trap here isn’t obvious even to a well-advised buyer.

The first idea, and why it doesn’t work

The initial plan was simple enough on paper: have a trusted Kenyan individual buy the land, then grant the foreign company a long lease. It’s a structure we hear proposed often, usually with good intentions and no idea it’s legally hollow.

Article 65 of the Constitution limits non-citizens to leasehold land, capped at 99 years. That much is well known. Less well known is Article 65(3)(b), which addresses exactly this workaround: property held in trust counts as held by a citizen only if all of the beneficial interest belongs to citizens. If the real economic benefit of the land, the rent, the right to develop it, the value if it’s ever sold, actually flows to the foreign company, the arrangement is treated in law as non-citizen ownership regardless of whose name sits on the title. The nominee structure doesn’t evade Article 65. It just adds a person between the investor and the land who can mortgage it, die and pass it into an estate dispute, or have creditors attach it, for no legal benefit at all.

The better idea

Our client’s next proposal was more sophisticated, and it’s a structure we generally recommend: a Kenyan-owned company holds the freehold, and leases it to the foreign-owned operating company for up to 99 years. Two separate entities, one holding the asset, one running the business.

This works, provided it’s genuine. Article 65(3)(a) makes a company a citizen only if it’s wholly owned by citizens, so a properly constituted, wholly Kenyan-owned land company can hold freehold outright. The lease to the operating company sits exactly at the constitutional ceiling. Done properly, with the lease registered, this is a clean, standard way to separate land risk from operating risk, and it has a useful side benefit: a claim against the hotel business can’t reach the land underneath it. It’s the same logic behind the land-holding company structures we cover in our guide to family land, succession, and land-holding companies in Kenya.

The catch that neither structure escapes

Here’s the part that took us back to the statute books. Kenya’s Land Control Act, a much older piece of legislation than the current Constitution, imposes its own separate regime on top of Article 65, and it applies specifically to agricultural land in a declared land control area. We’ve written before about how to tell the two categories apart in our guide to controlled versus uncontrolled land in Kenya; this case study is about what happens when a foreign investor lands on the wrong side of that line.

Section 6 of the Act makes the sale, transfer, or lease of such land void unless a Land Control Board consents. Section 9 tells the board how to decide, and one part of it isn’t discretionary: the board must refuse consent where the land, or a lease of it, is going to anyone who isn’t a citizen, a wholly citizen-owned company, group representatives incorporated under the Land (Group Representatives) Act, or a state corporation. A foreign-owned operating company doesn’t qualify. If the parcel in question turns out to be agricultural land within a land control area, which describes most rural land outside Kenya’s towns, that mandatory refusal applies to the lease from the Kenyan land company to the foreign operating company just as much as it would to a straight sale to a foreigner. The improved structure doesn’t sidestep this. Article 65 and the Land Control Act are simply answering different questions, and a deal can satisfy one while failing the other.

Ordinary land registration law says the same thing from a different angle. Section 107(3) of the Land Registration Act, 2012 states plainly that, for the avoidance of doubt, any lease granted to a non-citizen shall not exceed ninety-nine years. Three separate statutes, the Constitution, the Land Control Act, and the Land Registration Act, all converge on the same non-citizen leasehold limit, which is exactly why a structure that satisfies one but not another still fails as a whole.

There’s a real argument that this sits awkwardly against a Constitution that explicitly permits 99-year leases to non-citizens. That argument hasn’t, to our knowledge, been tested and settled in court, so it isn’t something to build a deal on. What can be built on is the Act’s own definition of agricultural land, which excludes land where the title itself carries a covenant restricting it to non-agricultural use, tourism development being the obvious example. Land already carrying that kind of restriction sits outside the Land Control Act altogether, and the lease proceeds on ordinary land registration rules. Land that doesn’t carry it needs either a different site or a Presidential exemption under section 24 of the Act, which is discretionary and shouldn’t be assumed.

One more wrinkle specific to this kind of deal

Investors looking at land near a conservancy hit a second, more practical trap: a lot of what tourism marketing calls a “private conservancy” is Maasai group-ranch or community-trust land leased to an operator, not privately titled land at all. Community land carries its own disposal rules under the Community Land Act, 2016, entirely separate from everything above; we go through those rules in detail in Community Land in Kenya: The Community Land Act 2016 Explained. The two issues compound each other. A parcel can look private because a lodge has operated there for years under a long-term community lease, while the underlying land is neither privately titled nor available for the kind of ownership structure a foreign investor actually needs.

What this means in practice

The lesson isn’t that foreign investment in Kenyan land is impossible, plenty of it happens successfully. It’s that the corporate structure and the land’s own legal classification are two separate questions, and getting the first one right tells you nothing about the second. Before negotiating price, before commissioning designs, the two things worth confirming on any candidate parcel are whether the title is individually registered private land (not an undivided community or group-ranch interest), and whether that land counts as agricultural land in a land control area, and if so, whether its title already carries a non-agricultural use restriction. Both are checkable at the land registry before any money changes hands. Neither is expensive to check. Both are expensive to discover late.

There’s a sequencing lesson in here too. It’s tempting, understandably, to fall in love with a view, a river frontage, a wildlife corridor, and treat the legal work as a formality to tidy up once the deal feels agreed. Land control classification runs the other way. It’s the cheapest, fastest check available, and it’s the one that decides whether the rest of the plan is even worth pursuing on that particular parcel. Running it first, on a shortlist rather than a single favourite, costs a few days. Running it last, after site visits, architectural concepts, and a verbal agreement with a seller, costs a lot more than money if the answer turns out to be no.

None of this is unique to hospitality projects. Any foreign-backed venture that needs Kenyan land, agriculture, manufacturing, renewable energy, warehousing, runs into the same two-layer test: does the structure satisfy Article 65, and separately, does the specific parcel’s classification let a non-citizen entity hold an interest in it at all. Advisors who only check the first layer, because it’s the one everyone already knows about, can hand a client a structure that is entirely correct and still doesn’t work on the ground.