Bangladesh
$100,000 runs one experiment here, start to finish, because everyone a problem touches already works within reach of everyone else.
The arithmetic
Buys one working experiment, in the field for a year and measured from the first day.
The result goes public, including a result that tells us to stop.
Bangladeshis abroad send more money home every year than the country receives in foreign aid and foreign investment combined. Almost all of it feeds households, as it should, and almost none of it is aimed at a problem.
Point a small share of it at experiments. $100,000 buys one working thing in the field for a year, measured from the first day, with the result published either way.
Run ten a year and the country gets ten answers, including the ones that say stop.
The diaspora fund
Remittances feed households, and they should keep going. This is the other pocket: money aimed at a problem, on terms anyone can check.
$100,000 runs one experiment for one year. Small enough that a family office, a foundation or a group of professionals can write it without a committee.
Ten tickets a year, each on a different problem from the list below, and none sized to hurt the person writing it.
We agree the measures before the money moves and instrument the work from the first day. The result goes into 81:26 whatever it shows, including a result that tells us to stop.
Nobody needs our permission to copy a result that works.
Capability
The garment industry carries most of the country's export earnings. The industrial base, the logistics and the compliance discipline already exist, and they already work.
Grameen Bank has lent on terms designed in the field for half a century. BRAC has run programs at national scale for nearly as long. The method won a Nobel Peace Prize.
What is missing is a table where all of them sit down on the same problem for longer than one funding cycle.
Proximity
An operator, a financier, a clinician, a regulator and a supplier all work within a short radius of Dhaka. In most countries getting those five in one room takes a year of asking.
That proximity is the advantage. It makes designing with people rather than for them practical here in a way it is not most places.
It is also why an experiment costs $100,000 instead of a multiple of it, and cheap experiments are what make publishing failures affordable.
Structure
SRI, MITRE and USC's Information Sciences Institute did this work for decades. They convened the people a problem touched, built the proof, and handed it on. All three are nonprofits.
We kept the shape and changed the money. CEI is a Delaware C-Corporation with a public-benefit election in progress. It can hold equity in what it builds, buy buildings, and carry a program to its first paying customer without waiting on a grant.
A grant ends. Rent does not. The chapter is built to own its ground, so the ground pays for the work done on it.
Stakeholders
A hard problem always has the same cast, and not one of them can move first alone. Each is waiting on somebody else, which is why the problem is still here. The institute is meant to hold a table steady long enough for all of them to commit at once.
Government and regulators
Ministries, the city corporations, Bangladesh Bank and the securities regulator. They decide what is legal before anybody finds out what works.
Banks, MFIs and capital
Grameen, BRAC Bank and the MFI network. They have priced risk in these communities for decades.
Industry and operators
Garment groups, pharmaceutical firms, logistics operators and the family businesses that employ most people. A pilot runs on their supply chains or it runs on paper.
Universities and research
BUET, Dhaka University, BRAC University and the agricultural institutes. Measurement has to be done by someone who does not need the answer to come out well.
Communities and unions
Drivers, farmers, garment workers, settlement associations. They live with the result, so they decide whether it holds.
The diaspora
Bangladeshis abroad writing $100,000 tickets, alongside development finance institutions and long-horizon foundations. Patient money on terms that match how long the problem takes.
Design with them, not for them. If they are not in the room, we have not designed with them.
Three problems
Each needs more parties than one company can convene, which is why it is still open. Each carries the experiment $100,000 would buy. We start with one.
EV (BD) Equity
The battery three-wheeler is already how the country moves. The driver owns none of it.
He rents the vehicle by the day, and often the battery on top of that. The margin from his own labor goes to whoever holds the asset. Charging, batteries, finance and road rules each sit with a different owner, and none of them has a reason to make him one.
This is the front-runner. Every piece it needs already exists and only the financing is missing. Put a driver on a path to owning his vehicle and his battery, and the same trip that earned him a day's wage starts earning him equity.
Affordable Housing
A worker can pay rent her whole working life and own nothing at the end of it.
Rent gets collected every month by somebody who owns the building. Ownership needs land, a lender, a builder, a municipal approval and a clean title, and those pieces sit with different parties who each wait for the others to move.
So the families who keep a city running have no claim on any of it. What is missing is not the will to build. It is a way to get those parties onto one table and keep them there long enough to close.
Plastic Alternatives: Jute and Water Hyacinth
Bangladesh banned the thin plastic bag before any other country did. The bags are still everywhere.
A ban without a substitute at the same price is a wish. Nothing arrived that a shopkeeper could hand across the counter for the same money, so the old bag came back through the side door.
Jute grows here. Water hyacinth chokes the waterways for free and somebody has to pull it out anyway. The raw material was never the constraint. The price on the shelf is.
How a chapter works
One company holds the site. Another runs the programs on it and pays rent. Between sessions the space is let for convenings, which pays for holding it. McDonald's did the same with the land under its restaurants.
The first year
- First
Stand up the chapter
We register the entity in Bangladesh and bring in local operators to lead it. Nothing else starts before this does.
- Then
Secure the site
One company buys and holds the ground, so the chapter has a permanent address instead of a booked room.
- Within the year
Run one pilot
One problem from this page. EV (BD) Equity is the front-runner, because only the financing is missing. The measures are agreed before it starts, and the result is published either way.
The ask
Nothing on this page is an offer of any security, and no capital terms appear here. If you want to talk about capital, that conversation happens through definitive documents rather than a website.
Ten tickets at $100,000. We would rather have ten partners who each fund one experiment than one who funds the fund.
A site in Dhaka. We would rather have a partner who contributes the ground than one who writes a check against it.
People from Bangladesh who have run something here and carried its payroll through a bad year.
Tell us who has it, what it costs them today and what you already know about it. If a year in the field could answer it, we want to hear from you.
