> ## Content Index
> Fetch the complete content index at: https://research.finbox.in/llms.txt
> Use this file to discover other available public pages before exploring further.

# India paid for this lesson. Southeast Asia can have it for free.
- URL: https://research.finbox.in/blog/india-paid-for-this-lesson-southeast-asia-can-have-it-for-free/
- Published: 2026-08-26T10:03:35.000Z
- Updated: 2026-08-26T10:03:35.000Z
- Description: Southeast Asia's platforms are posting fast growth and 1% defaults. India's numbers looked identical two years ago — until the loans aged. Here's the playbook India's lenders paid for, and what's worth doing now, while the numbers still look good.
- Author: Rajat Deshpande
- Tags: #BigPicture, #digitallending, defaults, SEA, FinTech, Credit, Digital Credit Infrastructure, NPA, NPAs

Hi,

Southeast Asia's biggest platforms just posted a set of lending numbers that look almost too good to question. Sea's Monee book [reached](https://www.morningstar.com/news/business-wire/20260810021050/sea-limited-reports-second-quarter-2026-results?ref=research.finbox.in) $11.1 billion in Q2 2026, up 62.5% year-on-year, with non-performing loans at just 1.0%. GoTo's loan book [grew](https://www.investing.com/news/transcripts/earnings-call-transcript-goto-posts-second-straight-profit-in-q2-2026-93CH-4820309?ref=research.finbox.in) 58% to IDR 11 trillion, and its fintech arm out-earned its ride-hailing business for the first time. Grab's portfolio nearly [tripled](https://s205.q4cdn.com/179588156/files/doc%5Ffinancials/2026/q2/Grab-Q2-2026-Earnings-Remarks.pdf?ref=research.finbox.in) to $2.3 billion, part of that through folding in its Superbank stake. Fast growth, low defaults, rising profits — the full set.

 India's digital lenders posted numbers that looked just like these two years ago. Unsecured lending was [compounding](https://ris.org.in/en/node/4091?ref=research.finbox.in) at 28% a year, more than twice the pace of secured credit, and by March 2024 unsecured loans had climbed to over 25% of all bank lending. Delinquencies stayed low the entire way up. Then growth slowed, the loans aged, and the stress arrived fast — microfinance loans 31 to 180 days overdue [jumped](https://www.business-standard.com/finance/personal-finance/microfinance-sector-shows-signs-of-stress-as-delinquency-rates-rise-125011000612%5F1.html?ref=research.finbox.in) from 2.15% to 4.30% in six months, and stressed assets in the sector kept [climbing](https://www.business-standard.com/amp/finance/news/nbfc-asset-quality-worsens-to-5-9-amid-rising-write-offs-rbi-fsr-125063001232%5F1.html?ref=research.finbox.in), to 5.9% by March 2025 from 3.9% six months earlier. 

The low defaults on the way up were never proof that the lending was sound. They were just a sign that the loans were young. A book growing 60% a year is mostly loans written in the last few months, and a loan written last month has not yet had the chance to go bad. The number looks its best exactly when the book is least tested.

India lived through the full version of this cycle, and the lenders who came through it in the best shape did a specific set of things while their numbers still looked good. That response is the closest thing Southeast Asia has to a playbook. Here is what it looked like. 

### Pull back on first-time borrowers before you have to

When stress showed, Indian lenders cut new-to-credit borrowers first, and cut them hard. In late 2024, consumption-loan originations for new-to-credit consumers [fell](https://newsroom.transunioncibil.com/41-of-first-time-borrowers-are-gen-z-according-to-transunion-cibils-latest-cmi-report/?ref=research.finbox.in) 21% year-on-year, against just a 2% decline for borrowers with an existing credit history. The new-to-credit share of all originations kept [falling](https://www.business-standard.com/amp/finance/news/new-to-credit-loan-share-dips-to-16-in-q1-as-lenders-stay-cautious-report-125092401219%5F1.html?ref=research.finbox.in) — from 20% in 2023 to 16% by mid-2025 — as lenders stayed cautious.

The logic transfers directly. The borrowers who look cheapest to acquire during a boom are the ones a lender knows least about, and they are the first to buckle when conditions tighten. 

If a meaningful share of a platform's growth is coming from first-time borrowers, that is the part of the book to slow down on first — before the data forces the decision. 

### Slow the whole book down while the numbers still look good

This is the hardest move on the list, because nothing in the data tells you to make it. Indian lenders made it anyway.NBFC unsecured loan growth was allowed to [fall](https://www.icra.in/CommonService/OpenMediaS3?Key=cba5a221-d409-44a1-a7ed-69cfd2f1d3cd&ref=research.finbox.in) from 30% in FY24 to a projected 10 to 12% in FY25\. S&P [credited](https://www.business-standard.com/amp/finance/news/nbfcs-loan-growth-to-moderate-to-18-in-fy25-s-p-global-ratings-124092400813%5F1.html?ref=research.finbox.in) the survivors' asset quality to a deliberate focus on lower-risk customers and generally low loan approval rates — in other words, choosing to approve fewer loans while the approving was still easy.

For a platform whose loan book is doubling, the equivalent is letting growth run slower than it could, on the understanding that the current default rate cannot yet tell you whether the pace is safe. 

There are early signs some Southeast Asian lenders are already thinking this way. 

GoTo [told investors](https://www.investing.com/news/transcripts/earnings-call-transcript-goto-posts-second-straight-profit-in-q2-2026-93CH-4820309?ref=research.finbox.in) it had "opted for a prudent approach" and tightened its weakest risk cohorts, and Sea's credit-loss provisions [rose](https://quartr.com/companies/sea-limited%5F3991?ref=research.finbox.in) 71.5% year-on-year even as its reported NPL held at 1.0%. Those are the moves of lenders provisioning for a book they expect to age, not just for the losses on the books today. 

### Keep a secured anchor in the mix

As unsecured credit tightened in India, both lenders and borrowers moved toward collateral. Gold loans [grew](https://www.icra.in/CommonService/OpenMediaS3?Key=cba5a221-d409-44a1-a7ed-69cfd2f1d3cd&ref=research.finbox.in) 18% while unsecured growth collapsed. Collateral is what a lender reaches for when it can no longer assume a clean repayment record will survive a downturn. Southeast Asia's platforms are built around unsecured, high-frequency credit, which makes this harder to replicate — but the principle holds. A book with no secured component has nothing to fall back on when the unsecured side turns. 

### Take the losses early and honestly

The Indian lenders who recovered fastest were the ones who recognised the damage and cleaned their books rather than letting bad loans sit. Write-offs at large NBFCs [surged](https://www.business-standard.com/amp/finance/news/nbfc-asset-quality-worsens-to-5-9-amid-rising-write-offs-rbi-fsr-125063001232%5F1.html?ref=research.finbox.in) to 72.9% of the relevant book by March 2025, up from about 50% in mid-2022\. Writing off is painful, and it dents the reported numbers, but a book that reflects reality is one a lender can actually manage. A book that hides its losses only delays the reckoning.

### Watch what the borrower owes everyone, not just you

India's stress was not mainly a story of individual lenders underwriting badly. It was a story of the same borrowers taking loans from many lenders at once. Two-thirds of small-ticket borrowers had more than [three active loans](https://ris.org.in/en/node/4091?ref=research.finbox.in) at the time of origination, and nearly half of unsecured borrowers already held another live retail loan. Every lender's own book looked healthy. The borrower underneath was carrying far more than any single lender could see.

Southeast Asia is structurally exposed to exactly this. Its lending runs through a handful of super-apps drawing on overlapping pools of users. When several platforms lend to the same person against the same income, each one can report a clean 1% default rate while the borrower stacks obligations that none of them sees in full. The single number every platform publishes says nothing about the total load the borrower is carrying — and in India, that total load was the thing that broke.

### The window is now

Southeast Asia's platforms are not India's NBFCs, and their books may well hold. Several are already provisioning early and tightening their weaker cohorts, which is the right instinct. But every item on this list is something a lender can only do while the numbers still look good — pull back on unproven borrowers, slow growth before the data demands it, build a secured anchor, take losses early, and underwrite against a borrower's total leverage rather than its own slice of it. Once the cycle turns, the options narrow to damage control.

India paid for this playbook over multiple hard years. Southeast Asia can read it before the bill comes due.

  
Cheers,  
Rajat