P2P Lending in India: The Risk Behind the 12% Return
Remember the "12% returns" tab inside Cred, BharatPe or MobiKwik apps some years back? Put money in, take it out whenever you want, earn more than an FD.…

Remember the "12% returns" tab inside Cred, BharatPe or MobiKwik apps some years back? Put money in, take it out whenever you want, earn more than an FD. Sounded great.
Here's the thing. That wasn't a deposit. That was you lending your money to strangers. The app sat in the middle, so you didn't always see the borrowers falling behind or the losses building up under the headline number.
The simplest way to think about it: P2P isn't a better FD. It's unsecured lending to individuals, packaged through an app. Once you hold that idea, everything else in this piece makes sense.
RBI saw the gap between what was being sold and what it actually was. In August 2024 it rewrote the rules. P2P lending itself didn't stop, it's still a licensed, RBI-regulated activity. But the version most of us saw inside payment apps was taken apart. What's left is smaller, slower, and much more honest about what it is. This piece explains how it works now, what changed, what happened to lenders' money, and how to size it up. It's not telling you what to do with your money.
What you're actually buying

Say Rahul in Pune needs ₹40,000 for six months. His bank says no, or says yes at a rate he doesn't like. So he goes to a P2P app like LenDenClub, IndiaP2P or Lendbox. The app checks his documents and credit score, gives him a risk grade, and puts his loan up.
You're in Bangalore with ₹50,000 lying idle. You put it on the same app. But you don't lend all of it to Rahul. The app cuts your ₹50,000 into tiny bits, sometimes ₹10 each, and spreads it across hundreds of people like Rahul. Rahul pays 16% to 20% a year. You're told you could get around 9% to 12%. The difference covers the app's fees and the cost of running the whole thing. What it does not cover is Rahul not paying. Under RBI's rules, the app can't take on your credit losses. Those are yours.
That's the whole product. You're the lender. The app is a matchmaker with an RBI licence (it's called an NBFC-P2P). It doesn't lend its own money, and it isn't allowed to promise you'll get yours back. In fact, RBI now makes you sign a declaration saying you understand you could lose your entire principal.
Two things make this very different from an FD or a debt fund. One, there's no bank and no fund manager between you and Rahul. If he doesn't pay, you lose, not the app. Two, your money is stuck. A loan runs for its full time. You get paid as borrowers pay, and there's no "withdraw anytime" button anymore.
What RBI changed in August 2024
RBI allowed P2P in 2017 as a marketplace for borrowing and lending. What some apps built by 2023 looked more like a deposit product. RBI's own words when it revised the rules on 16 August 2024: some platforms were offering tenure-linked assured returns and liquidity options, acting "like deposit takers and lenders instead of being a platform", breaching the fund-transfer rules, and matching lenders and borrowers inside closed groups run by partner apps. The revised rules went after each of those. Here's what matters to you as a lender:

A week later RBI fined LenDenClub ₹1.99 crore and LiquiLoans ₹1.92 crore for earlier violations. In December 2024 it sent eight platforms an 18-point list of questions on where the money was sitting and how their loans were doing.
The limits on how much you can lend were already there and still apply. ₹50 lakh in total across all apps. If you go above ₹10 lakh, you need a CA certificate showing your net worth is at least ₹50 lakh. And you can't lend more than ₹50,000 to any one borrower.
What did this do to the business? The math here is brutal. Take away the promised return, the anytime withdrawal, and the Cred/BharatPe distribution and you have taken away the three reasons most people showed up. Business Today reported in October 2024 that volumes fell by up to 90% across platforms. Lendbox's co-founder put his own drop at 90%. LiquiLoans stopped giving new loans and focused on collecting old ones. Cred, BharatPe and MobiKwik left. By January 2025, RBI had 26 licensed P2P companies but only 10 or 11 actually running, on about ₹11,000 crore of total loans (Business Standard). A year on, The Alt Investor's September 2025 check found lending at 15% to 30% of what it used to be, slowly climbing, with LenDenClub and IndiaP2P as the ones who rebuilt their products to fit the new rules.
The number to look at: bad loans

This is the part most people miss. While apps were advertising 10% to 12%, bad loans across the sector were growing much faster than that.
Capitalmind filed an RTI with RBI and got the numbers. Bad loans (NPAs) in P2P went from ₹14.7 crore in FY19 to ₹472 crore in FY23 to ₹1,163 crore in FY24. Business Standard's December 2024 report on that data put FY24 NPAs at over 17% of all P2P lending.
Now, what does 17% actually mean for you? Not that every lender lost 17%. An NPA is a loan where the borrower has stopped paying on time. It's not the same as money gone for good. Some of it gets recovered later, in part or in full. Actual losses depend on the platform, its borrower mix, how good its collections are, and how long loans stay stuck. And this is the whole sector added up; default rates differ a lot by platform and by borrower grade.
What this means for your actual return

The mistake most of us make is treating the rate on the app like an FD rate. It isn't. Here's the real equation:
What you actually earn = interest you received + principal you got back − principal you lost − platform fees − tax
P2P interest is taxed at your slab, same as an FD. In the 30% bracket a 12% headline is 8.4% before a single default. Then defaults. Splitting ₹50,000 across 500 borrowers protects you from one Rahul. It doesn't protect you from a bad year when a lot of Rahuls stop paying together which is exactly what the NPA chart above shows happening.
How it stacks up against the usual options

The row that matters most is the third one. Everything else on this list has something standing between you and a loss: a bank, a fund's spread of issuers, a company's balance sheet. P2P has nothing. That's what the higher advertised rate is paying you for.
Where the risk really sits
Default risk is the obvious one, and the numbers above say it's not small. But two more are worth naming.
Platform risk. The app doesn't hold your money but it does hold the borrower relationship and does the collecting. If a platform stops taking new business like LiquiLoans did your loans still exist, but getting paid back now depends on a company with no fresh income. Money in escrow is safe by design. Money already lent out is only as good as the collection team chasing it.
Rule risk. RBI has now reshaped this sector twice, in 2019 with the ₹50 lakh cap and in 2024 with everything above. It's still sending platforms questions and notices.
Before you put money in, ask these five questions
What's the platform's current NPA ratio? It has to publish this monthly now. If you can't find it easily, that's your answer.
How much principal have lenders on this platform actually lost, historically? Not NPAs, losses. The disclosure has to include this too.
How is my money spread? How many borrowers, what risk grades, and is any one grade or partner channel a big chunk of the book?
What exactly happens when a borrower defaults? Who chases the money, for how long, and what's the platform's recovery track record?
How long could my money realistically stay locked? Not the loan tenure on the screen. The real answer if borrowers pay late or the platform stops new business.
If a platform can answer all five clearly then you're at least looking at the real product. If it keeps steering you back to "up to 12%", you're looking at the old one.
Conclusion
P2P lending is you lending unsecured money to individuals through a licensed matchmaker. No guarantee, no easy exit, and a return that's only as real as the defaults underneath it. The version that looked like a 12% savings account is gone, and the numbers that came out afterward show why. What's left is a legit, RBI-regulated way to take on retail credit risk directly, in a sector where bad loans crossed 17% in the last year we have data for. Whether that's worth any part of your money is your call. But now you know what the deal actually is.
References
Regulatory: RBI, Master Direction – NBFC-P2P Lending Platform (Reserve Bank) Directions, 2017, as amended 16 August 2024 (lender exposure caps, escrow and T+1, ban on assured returns and credit enhancement, monthly NPA disclosure, lender declaration on loss of principal)
RBI press release of 16 August 2024 on violations observed at some platforms
RBI, December 2019 revision of lender exposure caps. Industry and NPA data: Capitalmind RTI to RBI on P2P NPAs FY19–FY24, as reported by Business Standard (December 2024)
Business Standard (January 2025: ₹11,000 crore AUM, 26 licensed / 10–11 active); Business Today (October 2024: volume declines, Lendbox quote, RBI penalties on LenDenClub and LiquiLoans)
The Alt Investor (September 2025 state of the industry); Blume Ventures, "The New Era of P2P Lending". Advertised yields: LenDenClub and Lendbox platform pages, as of September 2026.
Responses (0)
Leave a response
Related Articles
ULI vs OCEN: Why RBI Built a New Lending Platform When One Was Already There?
India has a interesting problem right now. We have not one but two digital systems trying to fix the same thing i.e. get…
How ML Models Are Replacing FICO/CIBIL Scores in Credit Underwriting
How ML Models Are Replacing FICO/CIBIL/Experian Scores in Credit UnderwritingCredit scoring has worked the same way for …
India's Embedded Finance Graveyard Has One Survivor: The LSP Model
Every fintech pitch deck in 2021 had the same slide. A giant market size number. A simple diagram. And the promise that …


