A shopkeeper in Nepalgunj wants sixty thousand rupees to buy stock before Dashain. She has run her business for six years. She has never missed a payment to her suppliers. She does not own land.
She will not get the loan. Not because the bank has no money. Nepal’s banking system is currently holding roughly one trillion rupees it cannot lend, with interest rates at their lowest in living memory. Last year, despite all that idle cash, credit to the private sector grew just 6.5 percent against a target of 12 percent. Nepal Rastra Bank cut rates. It flooded the system. Nothing moved.
The shopkeeper will not get her loan because the bank has no way of knowing she is good for it. This is the strange arithmetic at the center of Nepali finance, and it is worth stating plainly. According to the World Bank, the financing gap for micro, small and medium enterprises in Nepal stands at $6bn, around 14 percent of our entire economy. These are not marginal businesses. They produce more than 90 percent of Nepal’s industrial output and around 70 percent of what we export. They are, in the most literal sense, the productive economy.
And only 17 percent of them use banks to finance their investments. We have therefore built a financial system in which the money and the need are in the same country, in the same city, sometimes on the same street, and cannot find each other. Why? Because of what a Nepali bank asks for before it lends.
Roughly three-quarters of all bank lending in Nepal is secured against fixed assets, overwhelmingly land and buildings. This is not greed. It is ignorance, in the technical sense. A bank that cannot judge whether a borrower will repay demands instead something it can seize if they do not. Collateral is what lending looks like when you have no information. So the question becomes: why does the bank have no information?
Here the numbers are genuinely startling. The Credit Information Center collects only negative information. It records who has defaulted. It does not record who has faithfully repaid. And its coverage extends to roughly 10 percent of the adult population, against 21 percent across South Asia. It cannot see your utility bills, your insurance, your tax record. There is no single identifier tying any of it together.
Read that again and the six-billion-dollar gap stops being a mystery. A financial system that knows only who has failed, and knows that about one adult in ten, cannot lend on character or cash flow. It can only lend against land. It is not choosing collateral over judgement. It has no judgement to exercise.
The cruelty of this is precisely distributional. The person least likely to have inherited land is the same person least likely to have a borrowing record. The young founder. The returnee migrant with savings and an idea. The woman whose family property sits in her brother’s name. Only 20 percent of Nepali women borrow from a formal institution at all. The system is not merely declining to serve them. It cannot see that they exist.
Into this frustration arrives an appealing idea: peer-to-peer lending. Cut out the bank. Let savers lend directly to borrowers through an online platform. The government has now put P2P lending and crowdfunding in the budget, and Nepal Rastra Bank’s Monetary Policy for FY 2083-84 (2026-27) commits to exploring it.
I have spent a good deal of my working life on this question, and I want to say something that will disappoint the enthusiasts. A P2P platform does not solve the problem. It relocates it. Think about what such a platform actually is. It is a venue. It introduces a lender to a borrower. It does not tell the lender whether the borrower will repay, because nothing in Nepal currently tells anyone that. So instead of one bank unable to assess our shopkeeper, you now have four thousand ordinary savers unable to assess her, each risking money they cannot afford to lose, none with the faintest capacity to price the risk.
The World Bank, incidentally, listed three barriers to MSME credit in Nepal: inadequate collateral, absence of credit history, and limited financial literacy. That third one is usually read as a problem afflicting borrowers. But in a P2P market, the person being asked to make a sophisticated credit judgement is the lender, and the lender is an ordinary Nepali saver. We would be building an instrument to solve a crisis of financial literacy on the assumption that thousands of people will suddenly display expert financial literacy.
Other countries have run this experiment. Those that opened P2P markets before building the underlying information infrastructure did not democratize credit. They produced platform collapses, retail investors wiped out, and eventually a regulatory crackdown that killed the sector for a decade. China is a notorious case, but it is not alone.
So the enthusiasm is misplaced, but the instrument is not wrong. It is simply out of sequence. P2P lending is a distribution technology. It changes how credit reaches people. It cannot change whether credit can be priced. And distribution innovations only work once the pricing problem has been solved. Which brings me to the part of this story nobody is talking about, and the reason I am moderately hopeful.
In January this year, the World Bank approved a $95m operation for Nepal called the Sustainable and Inclusive Finance Project. Most coverage treated it as another lending programme. It is not. Read the components. Nearly $8m goes to rebuilding our credit information system. It will move the credit bureau from recording only failures to recording performance, the fact that our shopkeeper has paid her suppliers on time for six years. It will pull in non-traditional data: mobile transactions, utility payments. It will build credit scoring, including models designed to see women borrowers. And it will install a centralized KYC system, so that a Nepali proves who they are once, rather than to every institution separately, forever.
The Bank’s own document describes the purpose in a phrase I have not been able to stop thinking about. The aim, it says, is to deliver “improved reputational collateral for thin-file borrowers.” Reputational collateral. Your good name made it legible. Your record of keeping your word, converted into something a stranger can price.
That is the answer to our shopkeeper. Not a website. A credit history. And it explains why the Monetary Policy did not simply announce P2P lending. It committed to exploring P2P lending based on individual credit scoring. Those last four words are not decoration. They are the entire design.
A credit score is not a piece of technology. It is public infrastructure, like a road or a land registry. It converts a person’s past into something a lender who has never met them can evaluate. Once it exists, and is trusted, and reaches beyond the fortunate 10 percent, an enormous range of things become possible. Banks can lend on cash flow. Guarantees can be priced properly. And yes, peer-to-peer platforms become viable, because at last there is something for the lender to read.
Without it, P2P lending is a marketplace where nobody can see the price tags. And we will have handed our least protected savers the job that our banks have never managed to do. So, my plea, as this moves from budget speech to regulation, is for patience of a very specific kind.
Build the identifier. Build the bureau and make sure it reaches the woman who has never held a bank account. Get the credit scores working. Sort out the boundary between Nepal Rastra Bank and SEBON before licenses are issued, not afterwards in a public quarrel with people's money caught in the middle. Then run a small pilot, with a handful of platforms, tight exposure limits, and honest permission to fail.
Only then open the market. This will feel unbearably slow. There will be pressure to show results, to launch something, to declare the gap closed. That pressure should be resisted, because the alternative is not a faster bridge. It is a bridge with no foundations, and we know exactly who falls when those give way. It will not be the platform. We have spent decades asking Nepalis to prove they are creditworthy by owning land. We are finally building a system that could let them prove it by being trustworthy instead.
That is worth waiting a few years to get right. Giri writes on financial regulation and payment systems. The views expressed are personal.
