The Informal Stack

Hernando de Soto estimated that the total value of real estate held but not legally titled by the poor in developing nations exceeded $9.3 trillion.¹ That was in 2000. In Egypt alone, the extralegal wealth of the poor was 55 times all foreign direct investment ever recorded in the country, including the Suez Canal and the Aswan Dam. In Haiti, 150 times all foreign investment since independence in 1804.

The informal economy is the economy. The formal layer — registered businesses, titled property, banked transactions — sits on top of an informal base that most investors, most policy makers, and most tech companies treat as a problem to solve rather than a system to understand.

Three lenses on the same blindness

De Soto diagnosed the problem as legal exclusion. The poor have assets — homes, businesses, vehicles — but no legal title. Without title, assets cannot be used as collateral, cannot be split into shares, cannot be insured or leveraged. He called this "dead capital" — wealth that exists physically but sits economically inert because the legal system does not see it. His research team documented what formalization actually costs: in Peru, legally registering a small garment workshop took 289 days.² In Egypt, obtaining legal rights to a plot of desert land required 77 procedures across 31 agencies over 5 to 14 years. The system was not broken. It was designed to exclude.

James Scott saw something different. In Seeing Like a State, he argued that the modern state's project of making society legible — readable, controllable — systematically destroys practical knowledge embedded in local systems. He called this knowledge metis: the farmer who reads soil by touch, the street vendor who knows foot traffic by season, the informal builder whose structures survive earthquakes without engineering degrees. Tanzania's villagization program in the 1970s forcibly relocated 5 million people into planned grid villages³, severing them from the fields, water sources, and grazing patterns they had adapted to over generations. Agricultural production collapsed. The plan was legible to Dar es Salaam. Illegible to the actual ecology.

Daron Acemoglu and James Robinson added the political layer. Informality persists not because governments cannot formalize but because elites benefit from the exclusion. Extractive institutions channel resources to a narrow group. Formalization would empower a broader population, threatening the monopoly on rents. The system stays informal because the people with power to change it prefer it that way.

Legal systems exclude the majority. Formalization programs destroy the local knowledge that makes the informal economy functional. And powerful people have an interest in keeping it that way. Miss any of the three layers and your intervention fails.

What the informal stack actually looks like

I have spent a decade working in cities where the informal economy is the dominant economy. What you see on the ground does not match any single framework perfectly. It confirms all three partially.

In most African cities, commerce is relational before it is transactional. A woman selling fabric in a market in Abidjan has a customer base built on years of trust relationships. She extends credit informally — no contract, no interest rate, no collateral. Her customers pay when they can. The default rate is low because the social cost of default in a tight community is higher than any legal penalty a court could impose. This is metis. It works. And it is invisible to every formal metric.

The largest e-commerce platform in Africa is arguably WhatsApp. Not because anyone designed it that way, but because commerce in these cities is social first and transactional second. A buyer sends a photo of what they want. The seller responds with a price. Negotiation happens in the chat. Payment is M-Pesa or cash on delivery. No shopping cart. No checkout flow. No formal record. Billions of dollars in transactions flowing through a messaging app because the app matched the existing social pattern of how commerce already worked.

M-Pesa itself followed the same logic. People in Kenya were already sending mobile airtime as a proxy for money transfers. Safaricom did not invent mobile payments. They built a product on top of a behavior that already existed informally. The infrastructure met the pattern. That is why it scaled — and why dozens of formal fintech products that tried to impose a checkout-button model on these same markets did not.

The formalization trap

De Soto's prescription was to formalize — create legal bridges so informal assets become formal capital. The logic is sound. A titled house can be mortgaged. A registered business can get a bank loan. Dead capital comes alive.

But Scott's warning applies. Every act of formalization is also an act of simplification. When you register the fabric seller's business, you impose fixed costs — rent for a registered address, tax obligations, accounting requirements. Her competitive advantages were flexibility, low overhead, and trust-based credit. Formalization strips those away. The registered version of her business is less competitive than the informal version. She is being asked to trade a system that works for one that serves the state's need for legibility.

Scott's most elegant example is Brasilia. Designed as the ultimate modernist planned city — rigidly zoned, built for automobiles, monumental scale. No street life. The city functioned only because of the unplanned informal settlements that sprang up around it where workers actually lived, socialized, and built functioning neighborhoods. The informal city made the formal city possible.

I see the same dynamic in every market we operate in. The formal transport system — licensed taxis, registered vehicles, official routes — serves a fraction of demand. The informal system — shared minibuses, motorcycle taxis, unlicensed operators — moves the majority. The informal system is chaotic, unsafe by formal standards, and unmeasured. It is also the only thing keeping the city's economy moving.

Building on the stack, not replacing it

The businesses that work in these markets are the ones that add a layer on top of the informal stack without trying to replace it. mPharma builds on top of the pharmacist-as-doctor pattern — it does not try to replace the pharmacist with a clinic. It makes the pharmacist more effective. The financing loop works because it makes an existing informal asset — a vehicle operated by a transport company — scoreable to a bank without requiring the transport company to become a formal corporation. The data layer converts de Soto's dead capital into live capital without the 289-day registration process.

This is the design principle. Do not formalize the informal economy. Instrument it. Add the data layer that makes informal assets visible to formal capital without destroying the metis — the trust networks, the social credit, the relational commerce — that makes the informal economy functional.

A transport company with 50 vehicles operating on a tech platform generates real-time data — utilization, revenue per trip, maintenance patterns, driver performance. A bank can underwrite that. The transport company does not need to become a different kind of organization. It keeps its ground-level operations, its local knowledge, its relationships with drivers. The platform adds legibility without imposing the full cost of formalization.

The $9.3 trillion question

De Soto's number — $9.3 trillion in dead capital — is over two decades old. The actual figure today is certainly larger. The question is what unlocks it.

The first wave of emerging market tech tried to replace the informal stack with formal digital systems — e-commerce platforms with checkout flows, digital banks with KYC requirements, logistics companies with formal driver contracts. Most struggled. They were building Brasilia — legible from above, hostile at street level.

The second wave is building differently. Platforms that match existing behavior rather than imposing new behavior. Payment systems that layer on top of how people already move money. Marketplace businesses that read a city's existing patterns and accelerate them rather than redesigning them.

This approach has limits. Extractive institutions eventually block further progress — a government that benefits from opacity will resist the transparency platforms create. And eventually, legal reform is necessary. You cannot build a mortgage market on WhatsApp messages. But the operational insight is sequencing. Build the data layer first. Make informal assets legible to capital. Let the economics pull the legal system forward rather than pushing formalization onto an economy that has survived by being invisible.

Scott called this kind of intervention a "thin simplification" — just enough structure to unlock value without destroying the local knowledge that makes the system work. The platform that instruments the informal economy without replacing it is exactly that. Enough legibility to unlock capital. No more.