The Duvet That Starves Two Economies at Once

Kenya just gave foreign hawkers a week to close shop. South Africa, the United States, Poland and Uganda have already run this experiment — and the results are sitting in the data, not the rhetoric.

The man who washes my car in Nairobi is Burundian, working as a colleague to a Kenyan colleague.

The man who cuts my hair works two chairs down from a Kenyan barber, cutting the same fade, for the same price.

That is the whole story of Kenya's foreign-hawker crackdown in one sentence. Everything below just proves it.

On September 2, 2026, President William Ruto stood before Micro, Small and Medium Enterprise traders at State House and gave foreign hawkers and small-shop owners until the following week to shut down. "We have not improved investor confidence for hawkers to come to Kenya," he told them. By September 7, the deadline had landed. Videos circulated online of Kenyans confronting foreign vendors — in one, goods were taken from a trader without payment; in another, a foreign national was threatened outright. In Nairobi, Burundian nationals began queuing at their own embassy, asking for the travel documents to go home.

Before we decide who should be afraid, it's worth knowing the size of the room we're standing in.

Kenya's informal sector employed 18.1 million people in 2025, according to the KNBS 2026 Economic Survey — 83.8% of everyone recorded as working in the country. Wholesale and retail trade, hotels and restaurants alone absorbed close to 11 million of them. Formal employment, across the whole economy, sits at roughly 3.5 million.

Cut that number in half. In quarters. In tenths.

You have not found a market with a foreigner problem. You have found the entire Kenyan labour market — running on hustle, not headcount.

One failure wearing two faces

That same week, Ruto's government was fighting on a second front entirely. Days earlier, it had ordered Tata Chemicals Magadi — a mining operator operating a century of history in Kajiado County — to cease operations and leave, citing decades of resource extraction with too little local benefit. Days later, it ordered Burundian and Congolese traders selling duvets, groundnuts, and tea to close their stalls.

One target was a century-old industrial concession. The other was a woman selling tea from a kiosk. Both were folded into the same word: foreigner. And oh, has anyone gone to Eastleigh in Nairobi of late and seen the virtual transformation of that part of the city, and no one is ordering Somalis to leave. No. This is because these traders have a political constituency, and Kenyan Somalis are a targeted, voter-rich region politicians cannot touch. Touch the Somali community, and you get into trouble with very vocal politicians saying we are targeted or even discriminated against them because of religion. Who will defend the voiceless Burundians, Congolese, and Ugandans on the farms and in our houses as house managers?

That collapse is the failure. A mining concession and a street kiosk are not the same kind of problem, and treating them as one doctrine — foreignness as the diagnosis, exit as the cure — lets a government look decisive without asking which one, if either, is actually standing between a young Kenyan and a job.

Kenya already has tools that make this distinction unnecessary. A foreign national operating under a Class G business permit must show at least $100,000 in investment capital, tax compliance, and demonstrable benefit to Kenya. Trade Cabinet Secretary Lee Kinyanjui has clarified that visa-free entry does not confer a right to work or trade, and that status can be revoked where it's abused. Foreign Affairs Principal Secretary Korir Sing'oei has said, in the same breath, that lawfully documented foreign traders remain legally protected.

Illegal activity should be stopped because it is illegal — not because the person doing it is foreign. Blur that line, and you haven't solved unemployment. You've just given it a passport to blame.

Here is the part that should stop you cold: the same government telling Burundians to pack up their kiosks is, in the same year, trying to place a million Kenyans annually in jobs abroad, and has signed a labor-mobility pact with Germany explicitly designed so that Kenyan nurses and engineers are judged by their qualifications, not their passports.

Read that twice.

We are demanding of Berlin the exact courtesy we are refusing to extend to Bujumbura, Kinshasa, and Kampala

I did not think about any of this the last ten times I drove into that car wash. I paid, I left, and I never once asked what happens to the man rinsing my windscreen if the directive holds. That's not a small oversight. It's the same blindness this newsletter usually reserves for other people's governments — pointed, for once, at my own Tuesday afternoon.

Why the cycle keeps spinning

Kenya did not invent this instinct, and it will not be the last country tempted by it. The pattern is close to universal, and the data keep telling the same story.

South Africa is the clearest live case. Unemployment there hit 33.6% in the second quarter of 2026 — 8.5 million people, per Stats SA's Quarterly Labor Force Survey — with youth joblessness above 47%. That anger organized itself into movements like Operation Dudula and March and March, which set June 30, 2026, as an unofficial deadline for undocumented migrants to leave. AFP estimates put the number of foreign nationals who have since fled the country above 100,000. Vigilante blocking of migrants from hospitals became common enough that a Johannesburg court had to issue an injunction against the practice in November 2025.

Then came the correction nobody organized for. One sugar farmer on KwaZulu-Natal's North Coast told AFP he lost up to 80% of his cane-cutting crew almost overnight. Reuters visited three Newcastle clothing factories where owners reported losing 12% to 19% of their staff during the protests; the textile workers' union put the townwide loss at around 15% of a 15,000-strong workforce. Yet national agricultural employment did not collapse — it stood 4.3% higher year-on-year even as it dipped quarter-on-quarter. A country can carry brutal unemployment and a real labor shortage in the same season, in the same province. Wages, skills, geography, and networks are not interchangeable, no matter how much a placard wants them to be. I will tell you for free, tea farms in Kiambu and many other parts of this country will suffer. Your car wash price will rise, and the extra money won't go into a Kenyan pocket.

The perception gap is the real story underneath the anger. Survey research cited by The Conversation found up to 70% of South Africans believe migrants take local jobs, while administrative tax data shows foreign nationals hold less than 4% of formal employment — rising to roughly 20% of informal-sector participants, concentrated in exactly the businesses South Africans see every day. In KwaZulu-Natal, the provincial government's own 2026 research found 57.8% of spaza shops are foreign-owned — largely, officials say, because of bulk purchasing, wholesaler relationships, and cooperative buying that any Kenyan trader could copy tomorrow, not because of a passport. A World Bank study covering 1996 to 2011 found immigrant workers generated roughly two jobs for South African citizens each, through spending, hiring, and businesses started — not jobs simply taken.

None of this is theoretical for Kenya. The EAC Common Market Protocol, in force since 2010, guarantees the right of establishment to nationals of partner states and bars new nationality-based restrictions except on narrow public-policy grounds; protecting local traders from competition is not one of the grounds it lists. Tanzania already tested that boundary in 2025 with a licensing order barring non-citizens, including fellow East Africans, from fifteen categories of business — retail among them. Kenya's own directive risks the same collision with a treaty it signed.

Kenya has, in fact, already run a version of this experiment on itself. In 2014, Operation Usalama Watch sent thousands of security personnel into Nairobi's Eastleigh estate, ostensibly hunting terror suspects, and ended up detaining and expelling large numbers of Somalis — Kenyan and foreign alike. Traders in Eastleigh, a district then estimated to generate around 30% of Nairobi's council revenue, reported business down as much as 75% during the operation. A bomb attack near the same market killed a dozen people while the crackdown was still underway. The economic cost was immediate and real. The security gain was not obviously there.

Washington offers the newest warning, and it's an inconvenient one for anyone hoping enforcement is a free lunch. A Brookings Institution study, updated as recently as this month, found that 2025's ICE enforcement surges cost at least 668,000 jobs across 86 U.S. metro areas — with as many as 297,000 of them held by workers born in the United States. Roughly thirteen jobs vanished for every additional arrest. Remove enough people from one link in a local economy, and the ripple reaches people who were never the target.

Europe shows the mirror image. Poland and Hungary have tightened rules on non-EU workers under public pressure, even though the Polish Economic Institute projects Poland could lose 2.1 million workers by 2035, and a 2026 industry report found non-EU labor already contributes over 10% of Polish economic output. Demographics do not negotiate with nationalism.

And East Africa has its own extreme precedent. In 1972, Idi Amin expelled roughly 80,000 Ugandan Asians and handed their businesses to citizens with no time to learn how to run them. GDP fell 5% over the following three years; manufacturing output collapsed from 740 million to 254 million Ugandan shillings by 1979; real wages fell 90% within a decade; by the time Amin was overthrown, the economy had shrunk more than 40% from its 1971 size. Ownership can be reassigned overnight. The relationships, credit histories, and know-how sitting inside a business cannot.

What already works — and is being ignored at scale

None of this means Kenya should abandon its own traders. It means the fight is aimed at the wrong target.

The Local Content Bill, 2025, already before Parliament, points to a better instinct: it proposes that covered foreign companies source at least 60% of goods and services locally and keep at least 80% Kenyan staff. Whatever the right thresholds turn out to be, the question underneath it is the right one — how much Kenyan capability is this investment building? — and it applies as well to a factory as to a market stall. Over the weekend I was talking to someone working in the car assembly industry. The big firms import almost everything from Thailand to put together these cars you call Kenyan-assembled. However, that was not a secret, but the shocking part is the second part of the story. There are local content rules they are trying to follow. Who owns these factories they have to buy from? Your guess is as good as mine. They are owned by politicians and political elites and their friends. And that is not bad enough. The trouble is these assembly plants receive low-quality local content parts, and their cars are suffering, and the customers are suffering.

Where the advantage is illegal — counterfeit goods, unpaid tax, undocumented employment — the answer is enforcement, and Kenya already has the law for it. Where the advantage is organizational — the bulk buying and wholesaler ties that let KwaZulu-Natal's foreign-owned spazas out-price local ones — the answer is helping Kenyan traders reach the same scale through cooperative purchasing and credit, not removing the benchmark that shows what's possible.

And reciprocity is not sentiment. It is leverage. Actually, have you gone to Burundi, Rwanda, and Uganda and counted the number of Kenyans operating all manner of businesses in these countries, from M-Pesa to hotels and corporate positions? Do you want these countries' total Kenyans to leave too? Every Kenyan nurse on a ward in Germany, every engineer on a site in the Gulf, is protected by exactly the principle we would be discarding at Gikomba: that a worker should be judged on legality and skill, not on nationality. We cannot spend that argument twice — once when we're exporting labor, and again when we're importing it.

This is not a technical gap. It is a choice about which economics we're willing to look at honestly — and which anger is easier to point somewhere else.

Ubuntu, or the economics we keep forgetting

There's a reason that barbershop works. My Burundian barber and his Kenyan colleague aren't really competing for the same customer; they're building the reputation that brings customers to both of their chairs. Ubuntu — umuntu ngumuntu ngabantu, a person is a person through other persons — was never only a phrase for funerals and campaign speeches. It's also, quietly, a description of how a two-chair barbershop, or an 18-million-person informal economy, actually survives: through networks, not through solo operators sorted by passport.

Kenya can keep that insight or discard it. But it should at least notice that it already lives it, every time it sits down in that second chair.

What would it take to make judging a trader in Gikomba by his tax compliance as automatic as we expect a Kenyan nurse in Berlin to be judged by her certificate?