Count trade supported, not credit lent.
FromMore lending→ToMore trade per unit of risk capital
The measure of an open finance layer isn’t how much it lends. It is how much productive, inclusive trade each unit of risk capital supports. This shift asks what shared measures would show whether the layer is working.
By loan volume, little has changed
Over a year, the creation capital that first funded Meera’s batch has been released and redeployed five times illustrative, into her second product and four other creators. The collective, which had never borrowed, finances raw millet through the network. Arjun carries three new categories.
By loan volume, little is visible. By trade supported per unit of risk capital, all of it is.
From volume to velocity
How much was lent?
How much trade did the same capital make possible?
| Today | On an open network |
|---|---|
| Volume lent | Trade supported per unit of risk capital |
| Portfolio size | Cycles completed, capital recycled |
| Access for established firms | Access for new participants |
Don’t count what was lent. Count what the same capital made possible.
How it would work
Network metrics. Six shared measures, visible to participants, providers and public stakeholders:
| Measure | What it shows |
|---|---|
| Inclusion | Previously constrained participants obtain sustainable finance |
| Allocation | Losses and exposure match risk-bearing capability |
| Speed | The time from evidence to capital falls |
| Cost | The risk-adjusted cost falls without losses rising |
| Velocity | The same risk capital supports more trade cycles |
| Recycling | Capital is released and redeployed earlier |
The principle: maximise the productive trade supported per unit of risk capital, while keeping losses and governance within acceptable bounds.
Go deeper
What already exists
Financial inclusion is already measured by more than volume. The Reserve Bank’s Financial Inclusion Index combines access, usage and quality, and stood at 67.0 in March 2025, up from 64.2 a year earlier[1]. Account Aggregator reporting already shows its share of lending: an estimated 8.4 per cent of retail and MSME lending by value in FY26[2]. TReDS platforms report invoices financed and participants onboarded[3]. None of these measures trade per unit of risk capital.
Scope. We haven’t defined “risk capital” precisely or tested any of the six measures on real data. The design proves none of the gains. Each has to be demonstrated.
Rules
Published measures change behaviour: what is measured gets chased. Participants’ own data stays theirs, and only aggregated measures are public. Where regulators already require reporting, the layer reports through those channels.
Getting it right
Pair every gain with its cost: velocity with loss rates, inclusion with sustainability.
Compare against a baseline and a comparison group, so that a gain can be told from a trend.
Publish the method as well as the numbers, so that others can run a corridor of their own.
Who it helps. Providers can see where capital is doing the most work. Public stakeholders can see whether new participants are gaining access. Brands like Meera’s can see how the layer is serving them.
What could go wrong
- Chasing the metric.If velocity is rewarded, capital may be pushed round too fast, so velocity is always read with loss rates.
- Inclusion that isn’t sustainable.Finance a new participant can’t service isn’t inclusion, so “sustainable” is part of the definition.
- Measures nobody trusts.Every measure needs a baseline and a comparison group.
- Too many numbers.Six measures, read together, are enough.
How we’d test it
Set a baseline in one corridor before anything starts, and pick a comparison group that doesn’t get the layer.
- The test
- Trade per unit of risk capital
- What we would measure
- Measure whether more trade is supported per unit of risk capital and whether more new participants are financed, without losses rising.
- What would end the idea
- The corridor doesn’t support more trade per unit of risk capital than the comparison group, or losses rise.
Sources
- [1]Reserve Bank of India, Financial Inclusion Index for March 2025 (22 July 2025)Checked; the Reserve Bank’s press release
- [2]IANS, India’s Account Aggregator ecosystem drives Rs 3.82 lakh crore loans in FY26 (27 August 2026), reporting a Sahamati reportChecked; Sahamati’s own estimates, as reported by a news agency
- [3]CXOToday, RXIL TReDS Crosses Landmark ₹3,00,000 Crore Milestone in Cumulative Invoice Financing (10 April 2026)Checked; figures reported by the company in a media release
Spotted a source we missed or a rule we misread? Tell us what we got wrong.
If each piece can be tested, what would it take to build them all as one layer?