tl;dr
In Kenya, neighbors issue their own money as promises to each other. It works — and it's the closest real thing to money flowing like love.
Each voucher is a promise of real goods from a real neighbor.
The money works because the people know each other.
Proof that small circles can run their own flows of care.
The story
In Mombasa, small businesses printed vouchers backed by their own goods — and trade grew.
more
Will Ruddick — an American physicist turned Peace Corps volunteer — launched Bangla-Pesa in 2013 in Bangladesh, an informal settlement in Mombasa: paper vouchers issued by ~200 small businesses to each other, denominated in their own goods and services. Within weeks the Kenyan government arrested him and five community members for forgery; the charges collapsed, the network reopened, and trade in the settlement measurably increased — the vouchers were adding liquidity where shillings were scarce but capacity was not. Grassroots Economics Foundation grew from there: more Kenyan communities, the digital Sarafu network, a COVID-era partnership with the Red Cross that pushed vouchers through tens of thousands of households as humanitarian aid, and, most recently, commitment pooling — a protocol generalizing what the currencies were always doing: pooling promises of future service and letting them circulate.
The underlying diagnosis is Graeber's, made operational (gift-economies): communities are poor in national currency, not in capacity or care. People stand ready to teach, farm, cook, build, and watch each other's children; what is missing is the token that lets those offers find each other. So issue the token locally, back it with the offers themselves, and let the scarce shilling stop being the bottleneck on neighbors serving neighbors.
The mechanism, reread through love
Each voucher is a neighbor's promise, so trust is built into the money.
more
A voucher in a Sarafu-style network is a named promise: redeemable against a particular person's bread, tutoring, water delivery. This is money with almost none of money's usual physics (metta):
- It is particular, not fungible. Its value depends on who issued it and whether you trust their promise — Eisenstein's "sacred = unique and related" as a working currency, not a proposal (sacred-economics).
- It cannot meaningfully pool toward power. Hoarding a neighbor's promises is pointless; they are worth something only in circulation, Hyde's gift-that-must-move enforced by the token design itself.
- It is naturally Goodhart-resistant. There is no proxy to farm: a sybil cannot fake a reputation for delivering bread to people who live next door (goodharts-law, localism). Legibility is replaced by acquaintance.
- Issuance is commoning. Deciding who may issue vouchers, how many, and what backs them is exactly Ostrom's governance-by-members — boundaries, monitoring, graduated sanctions — applied to money creation itself (commons).
What this teaches
Small circles can run their own flows of care — no bank required.
more
- The strongest existence proof on the shelf. Every other lens here is literature or a partial experiment; Grassroots Economics is a running system where money behaves relationally at village scale. When the experiments begin, "a small circle issues promises to each other" is a tested mechanism, not a thought experiment.
- Elicit offers, not just loves. the founding question here is what do you love? (relationality). Ruddick's networks ask the complementary question — what can you offer? — and the pairing matters: a commons of loves plus a commons of offers is a flow waiting to happen; either alone is a list.
- Digitize the existing practice, don't install a new one. Sarafu worked where it wrapped chamas and mweria that already existed — the pre-existing web of obligation was the asset, the currency just made it liquid. This is the answer to the gift literature's hardest question — what minimum viable relationship must exist before a gift behaves like one (gift-economies): inherit the relationship from practices already alive, as dharma-gift inherits from dharma communities.
- Denominate in service, not in currency. A flow like this could be a commitment pool — promises of attention, teaching, care — with money entering only at the edges. The love stays the unit of account (attention).
The caution
It worked because it stayed small and known — and regulators still pushed back.
more
Three, all earned in the field:
- The state may read neighbor-money as counterfeit. Bangla-Pesa's founders were charged with forgery for printing promises. Monetary relation-building at any scale touches the state's oldest monopoly; experiments here that look like currency issuance should expect the same immune response and design for it (small, legible-to-members, framed as vouchers and gratitude, not "money").
- The convener is a hidden subsidy. Multiple currencies faded when Grassroots Economics' staff and donor funding moved on. If a flow needs a paid outsider to keep circulating, the relation hasn't taken — the same dependency test this inquiry should apply to itself from day one.
- Rails carry values. Moving Sarafu onto blockchain brought interoperability and airdrop-shaped, speculation-shaped expectations along with it — the fief-logic warned of in technofeudalism arrives bundled with the infrastructure. Choose rails for the relation they afford, not the scale they promise.