Episode 62 · Impact · 44 min

The fifteen-day loan nobody makes

India files 63 million businesses under one word, MSME, and every scheme that opens stacks at the top. Arthan Finance and the Michael & Susan Dell Foundation are arguing for the roughly 10 million nano entrepreneurs underneath: firms turning over ₹10 lakh to ₹1 crore whose real need is ₹15,000 for fifteen days, not a twelve-month term loan. Their claim is that this borrower already repays better than retail. She simply has no credit file to prove it.

GG
Geeta Goel and Pravash Dash
MD, MSDF India, and MD & CEO, Arthan Finance, Michael & Susan Dell Foundation and Arthan Finance · with Vishal Krishna
The fifteen-day loan nobody makes · episode thumbnail
44:07
Said in this episode
▶ 9:42
10 million
Nano entrepreneurs in India
A sub-segment of the 63 million MSMEs, sized on air at roughly 10 million firms, most with bank accounts and many with smartphones.
▶ 8:57
₹10L–₹1 cr
The nano turnover band
Annual turnover from ₹10 lakh (under ₹1 lakh a month) up to ₹1 crore; the lower end is street-level informal trade, the upper end masons and carpenters employing a few people.
▶ 6:40
16,000 / 5,000
Arthan loans disbursed and active customers
Sixteen thousand-plus loans written to date against five thousand-plus active customers, across smart branches, MFI partnerships and fintech tie-ups.
▶ 31:12
$30M → $150M
Foundation capital and the capital it unlocked
About $30 million committed to this segment in two post-COVID years unlocked roughly $150 million for partners through guarantees and equity, reaching over 150,000 clients.
▶ 33:12
20–24% vs 12%
What this borrower pays, versus the goal
NBFCs lend to the segment at twenty to twenty-four percent; Goel's stated aspiration is that a proven credit score gets the same customer closer to a bank's twelve.
▶ 14:23
₹15,000
The loan a kirana store actually asks for
Against ₹2 lakh of turnover on a fifteen-day cycle, enough for a few boxes of biscuits owed to a supplier, not the six- or twelve-month product lenders offer.
The brief

The argument in sixty seconds

Geeta Goel's claim is that India's credit machinery cannot see the businesses it most needs to fund. Sixty-three million enterprises are bucketed into a single category, MSME, and when a scheme is unlocked it stacks at the top, so little reaches the roughly 10 million nano entrepreneurs below, the ₹10 lakh to ₹1 crore turnover band that includes the chole-bhature cart selling fifty plates at ₹60 and the mason who employs three people. Half of them own a smartphone, fewer than a tenth of their transactions are digital, and almost none carry a bureau score. Pravash Dash's counter-claim is that the risk here is perceived rather than real: small-ticket repayment beats large-ticket, and microfinance runs above 99% while ordinary retail lending carries one-and-a-half to two percent delinquency. What is missing is the product. A kirana store with ₹2 lakh of turnover on a fifteen-day cycle needs ₹15,000 for a few boxes of biscuits, not a six- or twelve-month loan, so Arthan underwrites the transaction rather than the borrower, using the post-Jio API stack across three channels: digital-first branches in Odisha, Maharashtra and Andhra Pradesh, tech stacks handed to local MFIs, and real-time invoice discounting for the dealers of ₹10–200 crore anchors. Behind it sits the foundation's catalytic logic (about $30 million committed in two years, roughly $150 million unlocked for partners, guarantees of 10–15% so lenders can loosen their norms) on the bet that proof of concept pulls commercial capital in behind it. The stakes are the price of money: this borrower pays 20–24% today, and the entire exercise is about earning the way down towards twelve.

Worth your time if you are

Impact investors weighing catalytic capital against returns
Fintech founders underwriting borrowers with no credit file
NBFC and small-finance-bank credit heads
Policy people who assume an MSME scheme reaches the bottom
Anyone who thinks the kirana store is the whole informal economy
Episode map

Where the conversation travels

Every block is a chapter, coloured by what it's about. Click any of it to jump straight to that minute on YouTube.

01Cold open: 12 million kiranas, and beyond 0:00 Vishal sets up an episode about a segment he says has no reports written about it (push-cart sellers, small vendors, masons) and introduces the Dell foundation's India head alongside the founder of a fintech that has disbursed 16,000 loans. 02A family foundation's India ledger 2:15 Michael Dell's family foundation has run in India since 2006 on the principle of accelerating human opportunity, deploying about $250 million across education, livelihoods and financial inclusion, claiming impact on over 30 million children and families, and measuring every dollar because it considers itself small. 03Why credit is the livelihoods engine 4:36 Financial inclusion is framed not as charity but as the fuel line: nano entrepreneurs need working and growth capital exactly as large corporates do, and Arthan is one of the foundation's equity investments made to open that market. 04What Arthan Finance actually is 6:40 Pravash Dash describes a fintech built to be the preferred financial partner of the most underserved small businesses, 16,000-plus loans disbursed and 5,000-plus active customers in, with job creation as the stated impact metric. 05Defining the nano entrepreneur 8:11 India lumps 63 million businesses into MSME; the foundation carves out the nano band at ₹10 lakh to ₹1 crore annual turnover, roughly 10 million firms, half with smartphones, from the chole-bhature seller doing fifty plates at ₹60 to masons and carpenters who employ a few people. 06Why formal lenders stay away 10:45 Dash breaks the exclusion into supply-side and demand-side failures: one-size underwriting that assumes steady cash flows, expensive last-mile reach, low financial literacy, no documented income and no bankable collateral, all of which pushes the borrower to the local moneylender. 07After Jio: APIs and the biscuit loan 13:01 The post-Jio data ecosystem and stable third-party APIs let Arthan run purpose-based, transaction-level underwriting through AI and ML scoring, so a kirana with ₹2 lakh of turnover and a fifteen-day cycle gets ₹15,000 for boxes of biscuits rather than a twelve-month term loan. 08They do pay back 14:53 Vishal presses on why anyone would choose this segment, and Dash answers with fifteen-plus years inside it and the numbers: small-ticket repayment runs higher than large-ticket, microfinance sits above 99%, while ordinary retail lending carries one-and-a-half to two percent delinquency. 09What the lockdown did to the file 17:21 Lockdown drained cash flows and savings, working capital went into household consumption, defaults dented bureau scores, and the big banks and NBFCs consolidated and stopped writing new loans exactly when businesses needed to restart. 10Three channels, one balance sheet 19:05 Arthan runs digital-first smart branches in Odisha, Maharashtra and Andhra Pradesh where competition is thin, hands its tech and product stack to local MFIs upgrading group borrowers to individual loans, and discounts ₹500-to-₹10,000 invoices for dealers of ₹10–200 crore anchors within two hours. 11Catalytic capital and the perceived risk 22:38 Goel argues the opportunity is large and lucrative once risk is genuinely mitigated, which is why catalytic impact capital goes first, proving the concept so investors who will not take that risk today follow, the same playbook used in urban microfinance and micro-mortgages. 12Alternative data, scoring as a service, guarantees 25:30 With no collateral and no bureau history, the answers are alternative data (electricity and mobile payments, even how often you change your number) credit models sold as a service so every lender need not build one, and pooled guarantees that transfer the residual risk. 13The three levers, and the last mile 28:37 Goel names her three systemic shifts: digitisation that cuts servicing cost and customises tenor, ecosystem-level risk assessment, and partial guarantees that raise an NBFC's risk appetite. She concedes the last mile is still not fully digital and runs on field partners. 14What $30 million actually bought 30:39 Two post-COVID years of about $30 million committed unlocked roughly $150 million for partners through guarantees and equity and reached over 150,000 clients with ₹1–5 lakh loans, among them a health worker buying an operational bed and businesses growing revenue tenfold on a refrigerator. The deeper win is a credit footprint that should pull rates down from 20–24% towards a bank's twelve. 15Ten million, five years, an option on credit 33:58 Goel's target is not that everyone borrows but that all 10 million have the option within five years. Roughly a tenth already access credit through formal supply chains, and UPI, smartphones, account aggregators and new data stacks make the other ninety percent reachable faster than Jan Dhan's decade. 16₹2,000 crore, 15 percent, and the series ahead 37:59 Dash reads the nano segment as microfinance fifteen years ago, with group borrowers migrating to individual loans and footprints accumulating, and targets two million loans and ₹2,000 crore of AUM in five years, while Goel aims to unlock fifteen percent of the 10 million directly and Vishal trails a series on NBFCs, credit scoring and aggregator platforms.
Timeline

The arc, briefly

How a foundation and a small NBFC arrived at the ten million businesses India's credit machinery cannot see.

1999Michael Dell's family foundation is set up. Its stated principle is accelerating human opportunity, and its mission is the lives of children growing up in urban poverty.
2006It starts operating in India. About $250 million invested since, across education, livelihoods and financial inclusion, through grants to non-profits, equity and debt into for-profit social enterprises, guarantees, and work with state governments. Goel puts the reach at over 30 million children and families and insists on measuring every dollar, on the grounds that a small foundation in a country this size only moves the needle by being ruthless about impact.
2020Lockdown empties the credit file. Cash flows collapsed, savings went, working capital was eaten by household consumption, and the defaults that followed dented bureau scores that had taken years to build. Nobody wrote fresh loans. When the unlocking came and these businesses needed money to restart, the large banks and NBFCs were consolidating branches and lending to nobody new.
2020-22$30 million in, about $150 million unlocked. In 2019 nano entrepreneurs were not a significant part of the foundation's work; after COVID it redirected funding to them. Two years and roughly $30 million of commitments pulled about $150 million more into partners through guarantees and equity, and reached over 150,000 clients on loans of one to five lakh. A health worker who could not get a loan bought a mini operational bed and set up a unit in her community. Others grew revenue tenfold on a refrigerator that stopped the stock spoiling.
TodayThe loan is ₹15,000 and the cycle is fifteen days. India files 63 million businesses under one word, MSME, so schemes stack at the top and little reaches the roughly 10 million nano firms turning over ₹10 lakh to ₹1 crore. Half hold a smartphone, few of their transactions are digital and almost none carry a bureau score, so they pay the local moneylender's price. After Jio built the data ecosystem and stable third-party APIs arrived, Arthan could underwrite the transaction instead of the borrower: a kirana store with ₹2 lakh of turnover wants ₹15,000 for boxes of biscuits it already owes a supplier, not a twelve-month term loan. 16,000 loans and 5,000 active customers in, run through digital-first branches in Odisha, Maharashtra and Andhra Pradesh, tech stacks handed to local MFIs, and invoice discounting settled within two hours. Dash's answer to the risk question is that small-ticket repayment beats large-ticket: microfinance runs above 99 percent while ordinary retail lending carries one and a half to two percent delinquency.
In 5 yearsTen million people with the option, whether or not they take it. Goel's target is not universal borrowing but universal access, with about fifteen percent of the ten million reached directly through the foundation's own platform. Dash is aiming at two million loans and ₹2,000 crore of assets under management, and reads the segment as microfinance ten or fifteen years ago, with group borrowers migrating to individual loans and a footprint accumulating behind each one. The number that decides whether any of it worked is the price of the money: 20 to 24 percent today, against the twelve a bank charges.
Takeaways

Ideas to carry out of this hour

01

Nano is a segment, not a rounding error

India buckets 63 million businesses into the single word MSME, so when a scheme is unlocked for MSMEs it stacks at the top, at the small and medium end, and little flows down. Goel carves out the nano band deliberately: annual turnover of ₹10 lakh to ₹1 crore, a sub-segment of micro enterprise, roughly 10 million firms. Her phrase for why they are fundable is precise: neither too small nor too invisible, and yet not large enough to be seen by the schemes written for MSMEs.

02

The credit-worthy and the credit-scored are different people

The perceived risk of the segment is high; the observed risk is not. Dash says small-ticket repayment runs materially higher than large-ticket in his own book, and microfinance sits above 99%, while ordinary retail lending invariably carries one-and-a-half to two percent delinquency. Vishal's framing is the same from the other side: the intent to repay is there, and the absence of a platform reaching these borrowers is the opportunity, not the risk.

03

The product is the tenor, not the interest rate

A kirana store turning over ₹2 lakh on a fifteen-day cycle does not need a six- or twelve-month loan. It needs ₹15,000 today for a few boxes of biscuits it must pay a supplier for. Arthan's answer is purpose-based, transaction-level underwriting on an AI and ML scoring system rather than document-heavy, long-tenor cash-flow assessment. The claim is that nobody is currently able to give this segment that shape of product, which is why the demand looks invisible.

04

Jio rebuilt the underwriting stack before it rebuilt the customer

Dash dates the change to the arrival of a mass digital ecosystem, which produced both a data-driven borrower and a market of stable third-party APIs delivering micro-services into a lender's system. That is what makes small-ticket underwriting economically possible: the information now flows in rather than being collected by feet on the street. Goel's parallel point is on the collections side: disbursement and recovery over UPI replace the person who used to be sent to collect cash.

05

Catalytic capital is priced to be copied

Goel's stated role is not to fund the segment but to de-risk it in public. The foundation takes risks that other investors will not take today, on the argument that once proof of concept exists the market follows, the same model it ran in urban microfinance and micro-mortgages, catalysing the early years before commercial capital arrived. The leverage claim is explicit: about $30 million committed in two years unlocked roughly $150 million for partners through guarantees and equity.

06

Assessing risk is not enough; someone has to absorb it

Even a good alternative-data score is still a score on a customer with no credit history, so Goel's third lever is risk transfer. Pooling diverse MSME and nano loans allows a partial guarantee to sit over the portfolio; the foundation wrote 10–15% guarantee protection with partners during the COVID stress, which let the lender relax its credit norms rather than freeze. The companion move is credit scoring sold as a service, so that not every institution has to build a complex model of its own.

07

COVID broke the credit file, not the borrower

Lockdown exhausted savings and turned working capital into household consumption; the resulting defaults dented bureau scores, and no one wrote fresh loans. Then the large NBFCs and banks consolidated their locations and stopped lending precisely when businesses were trying to restart. Goel's argument is that protecting a thin credit score through a single shock is itself an impact outcome, because these customers are creditworthy and one bad year should not price them out for years.

08

The real target is the interest rate, not the loan

NBFCs serving this segment lend at twenty to twenty-four percent, and Goel is candid that this is the cost of illegibility rather than of default. The stated aspiration is that a documented repayment record and a real credit score let the same borrower walk into a bank and get somewhere closer to twelve. Every intervention described (digital footprint, scoring, guarantees) is ultimately an argument about compressing that spread, and she concedes there is a long way to go.

The numbers, drawn

What the episode measures

Every figure below was said on air, with timestamps included and caveats kept.

How digital a nano entrepreneur already is

%
Own a smartphone50Transacted digitally25Payments gone digital10
As stated in conversation, citing a survey: about half own a smartphone and more than a quarter have transacted digitally, but under a tenth of their transactions are digital. The first two bars are shares of entrepreneurs; the third is a share of transactions. The point is a footprint thin enough to be missed and real enough to underwrite.▶ 9:13

The price of having no credit file

% interest
NBFC today, low end20NBFC today, high end24Bank rate aspired to12
Rates as stated on air: NBFCs serving nano entrepreneurs lend at twenty to twenty-four percent, against Goel's stated goal of getting this borrower closer to the twelve percent an established borrower can access.▶ 33:12

What two years of catalytic capital moved

$ million
Foundation committed30Unlocked for partners150
Goel's numbers for the two post-COVID years: roughly $30 million committed to nano entrepreneurs unlocked about $150 million for partners through instruments from guarantees to equity, servicing more than 150,000 clients. Goel herself calls the figures small against the scale of the segment.▶ 31:12

The loan she needs, against the loan on sale

tenor, drawn to scale
What the kirana asks for ₹15,000, for 15 days, to pay for biscuits What the market sells A term loan of six to twelve months Drawn to scale: 15 days against 12 months. So underwrite the transaction, not her
Dash's worked example. The shop turns ₹2 lakh on a fifteen-day cycle and owes a supplier for a few boxes of biscuits, which makes the product on the shelf the wrong shape rather than the wrong price. Nobody, he says, currently offers this segment that shape.▶ 14:23
Worth keeping

Lines that stay

They're neither too small nor too invisible — and yet they're not large enough. When a scheme gets unlocked for MSMEs it stacks at the top, at the small and medium enterprises, and little flows through to the nano.

— Geeta Goel ▶ 9:28

What he asks for is ₹15,000 to buy a few boxes of biscuits that he has to pay his supplier for immediately. Nobody is able to give that type of product to this segment today.

— Pravash Dash ▶ 14:23

These customers are credit deprived, but it's not that they don't pay well. On the small-ticket loans we give, the repayment rate is much higher than on the larger tickets.

— Pravash Dash ▶ 16:23

If somebody else gets a loan at twelve percent, this customer should also be able to go to the bank and get a little closer to that.

— Geeta Goel ▶ 33:29

Can we make it a loan on demand — you click here, a lot of information opens up, you get a credit score, you get an instant loan?

— Geeta Goel ▶ 42:25
Clips that travel

Short on time? Start here

Policy people who assume an MSME scheme reaches the bottom

Who is a nano entrepreneur, exactly

The definition the rest of the episode rests on: 63 million MSMEs, the ₹10 lakh to ₹1 crore carve-out, and why schemes stack at the top.

8:11 → 10:45 · 3 min ▶ Watch clip
Fintech founders underwriting borrowers with no credit file

Why the ₹15,000 loan doesn't exist

The clearest statement of the product gap: fifteen-day cash cycles, biscuit boxes, and transaction-level underwriting instead of document-based term loans.

13:01 → 14:53 · 2 min ▶ Watch clip
NBFC and small-finance-bank credit heads

Three channels, one balance sheet

Arthan's actual operating model: digital-first branches in thin-competition geographies, tech stacks lent to MFIs, and two-hour invoice discounting.

19:05 → 22:38 · 4 min ▶ Watch clip
Impact investors weighing catalytic capital against returns

De-risking a segment nobody will price

Alternative data, credit scoring sold as a service, and 10–15% partial guarantees, the mechanics of making an unbankable book bankable.

25:30 → 28:37 · 3 min ▶ Watch clip
Anyone who wants impact stated as numbers

What $30 million actually bought

The leverage claim, the 150,000 clients, the health worker's operational bed, and the honest admission that these numbers are still small.

30:39 → 33:58 · 3 min ▶ Watch clip
Glossary

The jargon, unpacked

Nano entrepreneur
A business with annual turnover between ₹10 lakh and ₹1 crore. A sub-segment of micro enterprise, numbering roughly 10 million in India, sitting above the self-employed and below anything a bank calls an SME.
MSME
Micro, small and medium enterprises: the single 63-million-strong bucket into which Indian policy files almost every business, which is why relief and credit schemes tend to be captured at the larger end.
NBFC
Non-banking financial company: a lender without a banking licence, the vehicle most small-ticket credit in India actually runs through, including Arthan's own book.
Catalytic impact capital
Money deployed to prove a market rather than to maximise return, using grants, debt, equity and guarantees to take the first risk so commercial investors will follow once the concept is shown to work.
Partial guarantee
A backstop covering the first slice of losses on a pooled loan book, quoted here at 10–15%, which lets a lender loosen its credit norms without putting its own capital at risk.
Transaction-based underwriting
Assessing a specific purchase or cash cycle rather than the borrower's overall creditworthiness, so a loan can be sized and timed to one fifteen-day working-capital need instead of a year.
Supply-chain financing
Discounting a dealer's or distributor's invoices against a larger anchor company's credit, here pulled down to anchors of ₹10–200 crore and tickets as small as ₹500, settled within two hours.
Account aggregator
India's consent-based framework for sharing financial data between institutions, cited as one of the stacks that could make thin-file borrowers legible to lenders without physical documentation.
Connections

If this resonated, go here next

Full transcript

The whole conversation, searchable

173 segments

Auto-generated captions, lightly cleaned. Click a timestamp to open that moment on YouTube.