B2B stablecoin payments hit $6bn a month. In India, the same transfer is an FEMA violation waiting to happen.
The global infrastructure story and India’s regulatory one are diverging fast, and finance teams are caught in the middle.
B2B stablecoin payments hit $6 billion a month, and rising
Between early 2023 and mid-2025, monthly B2B stablecoin payments volume went from under $100 million to more than $6 billion, a jump of roughly 733% year on year. B2B transactions now account for around 60% of all stablecoin payment volume, ahead of consumer remittances and trading combined. The more interesting number is who is moving the money. It is not crypto exchanges. It is ship brokers, steel traders and import-export firms settling invoices that used to take three to five days to clear through correspondent banking.
That is the part most coverage of this trend gets right and then stops at. The part it usually skips is what happens once this story reaches a market where the central bank has decided, in public testimony, that the underlying asset class should not exist in legal form at all. That market is India, and the gap between the global adoption curve and the domestic legal one is wide enough to matter to any finance team currently being pitched a faster settlement rail.
What actually cleared the runway elsewhere
For most of the last decade, stablecoins for B2B settlement were a demonstration, not infrastructure. Two things changed that over the past year. First, regulatory clarity: the GENIUS Act, passed in the US in July 2025, gave stablecoin issuers a federal licensing framework with monthly independent reserve audits, and issuers with more than $50 billion in outstanding tokens must now publish annual audited financial statements. The EU’s MiCA and Singapore’s MAS framework did comparable work in their own jurisdictions.
Second, and more decisive: integration. SAP S/4HANA and NetSuite both extended their treasury modules this year to let finance teams configure stablecoin settlement, mostly in USDC, alongside existing wire and ACH rails, inside the same dashboards their controllers already use for month-end close. A trade financier does not switch settlement rails because a law passed somewhere. They switch once the option shows up as a configuration inside the ERP they already run their books through, next to the wire transfer they were already going to send.
The economics behind that switch are not subtle. Correspondent banking on a mid-sized cross-border invoice commonly costs three to five percent in combined transfer and foreign-exchange fees, on a settlement window measured in days rather than seconds, and the fee climbs further on corridors that route through two or three intermediary banks. That is the gap stablecoin rails are closing. For a firm moving large volumes on thin trade margins, saving even a couple of percentage points is a straightforward commercial reason to pilot the rail, independent of whatever a stablecoin issuer’s marketing team says about the future of money.
Where India sits: the RBI is building a wall, not a bridge
None of that transfers cleanly to India, because the country’s central bank has taken close to the opposite regulatory position from the US, the EU and Singapore. On 2 July 2026, the RBI told a Parliamentary Standing Committee that virtual digital assets, the category that includes stablecoins, should not be legalised. Government assessments circulating around that testimony describe the RBI’s position as having hardened toward outright prohibition over the preceding year, not softened toward a licensing regime of the kind the US and EU have built.
The FEMA gap that turns a payment into a case
The practical problem sits one level below the RBI’s stated preference, inside the Foreign Exchange Management Act itself. Section 2(h) of FEMA defines currency, and stablecoins do not fall inside that definition. They are neither currency notes nor have they been separately notified as currency by the RBI. That is not a technicality. It means a stablecoin payment received from an overseas customer is not currently treated as a recognised foreign exchange inflow under the law that governs how Indian businesses are allowed to receive money from abroad.
“A payment that reads as an efficiency win in Singapore or Dubai can be the exact fact pattern an FEMA enforcement notice is built on in Mumbai.”
That gap stopped being theoretical on 17 June 2026, when the Enforcement Directorate raided five Bengaluru-based crypto-payment firms over an alleged ₹2,500 crore in unauthorised cross-border stablecoin transfers. The firms involved were reportedly routing trade payments through stablecoin rails precisely because they were faster and cheaper than the correspondent banking alternative, the same argument driving adoption everywhere else. In India, that argument currently runs straight into a currency definition that has not caught up with it.
FEMA contraventions are civil rather than criminal, but the penalty structure is not symbolic. Section 13 of the Act allows a penalty of up to three times the sum involved in the contravention, or up to ₹2 lakh where the sum cannot be readily quantified, with a further daily penalty for as long as the contravention continues. For a firm that has routed meaningful trade volume through stablecoin rails over several months, that multiplier is what turns a compliance debate into a board-level problem rather than a footnote in the finance team’s minutes.
Three questions worth asking before piloting this
Vendors pitching stablecoin settlement to a finance team tend to lead with the speed number and stop there. Three questions cut through that faster than reading another vendor-produced guide. First: which legal entity actually holds the tokens between settlement and conversion to fiat, and under whose regulatory umbrella does that entity sit? A well-run programme minimises the time value spends in an unregulated wallet rather than a licensed custodian. Second: what does the transaction record look like when an auditor asks for it cold, not when a vendor demonstrates it in a sales call? A reconciliation dashboard built for a demo and a reconciliation process that survives a year-end statutory audit are not the same product. Third, and specific to India: does any leg of the transaction cross the border in stablecoin form, or does the stablecoin conversion happen entirely offshore, with only a fiat rupee settlement touching an Indian bank account? Under the current FEMA reading, that distinction is close to the whole question, and it is usually the one a vendor’s sales team is least prepared to answer precisely.
What changes on 1 October 2026, and what does not
The RBI has separately notified the Foreign Exchange Management (Export and Import of Goods and Services) Regulations, 2026, taking effect on 1 October 2026. They consolidate export and import rules into a single instrument for the first time, extend realisation timelines, and formalise service export declarations. It is a genuine modernisation of how Indian exporters and importers are expected to report and settle cross-border trade. It says nothing about stablecoins. Businesses hoping the reform would quietly extend legitimacy to stablecoin settlement alongside it will be disappointed: the consolidation tightens the framework around recognised currency and authorised banking channels, not around assets sitting outside that framework.
The problem that shows up even where it is legal: reconciliation
Set the legal question aside for a moment and assume an Indian business operating through a jurisdiction where stablecoin settlement is unambiguously permitted. A second, less discussed problem does not go away: matching on-chain settlement against the accounting system. A transfer that finalises in under a second on Solana, or a couple of seconds on Tron, is complete before most ERP systems have registered that it happened. Fifty stablecoin payouts a month can still be reconciled by hand in a spreadsheet. Five thousand payouts a month, spread across several chains with different finality times and fee structures on each leg, is where that manual process breaks down, and where most of the vendor-produced guides on this topic quietly stop offering detail, because the answer is a reconciliation product they happen to sell.
| Market | Legal status | Practical bottleneck |
|---|---|---|
| United States | Licensed under the GENIUS Act (2025) | Reconciling near-instant settlement with legacy ERP and treasury workflows |
| European Union | Licensed under MiCA | Consistent cross-border treatment between member states |
| Singapore | Licensed under the MAS framework | Multi-chain reconciliation at scale |
| India | Not recognised as currency under FEMA Section 2(h); RBI opposes legalisation | The underlying legal question, before reconciliation is even relevant |
What this means for a finance team right now
For a finance lead at an Indian B2B SaaS company or exporter, the honest answer in 2026 is that the global stablecoin payments story is not yet a story about India, whatever a vendor’s sales deck implies. If an overseas customer proposes settling an invoice in USDC because it is faster than a wire, that proposal carries FEMA exposure that a five-day correspondent-banking delay does not. A faster and cheaper rail does not change what an authorised dealer bank and an auditor need to see, and it does not change what Section 2(h) currently recognises as a foreign exchange receipt. The digital rupee pilot and the RBI’s own signalling are the indicators worth tracking here, not the global stablecoin volume charts. In this market, the second is not yet a leading indicator for the first.
Frequently asked questions
Related reading
RBI's digital lending rules aren't a compliance patch. They're four permanent states in your loan engine.
RBI's digital lending framework reads like a disclosure checklist. In a loan-servicing codebase it's four states: who can touch the money, who's allowed on the platform, and two clocks it can't ignore.
OCEN was pitched as 'UPI for credit'. Here's why that never happened
OCEN launched in 2020 to do for credit what UPI did for payments. NPCI and Sahamati publish their numbers every month. OCEN publishes none, and that absence explains more than any feature gap.
The labour codes' 50% wage rule isn't a policy tweak. It's a payroll data-model rewrite.
India's four labour codes have been in force since November 2025. The hard part for payroll teams isn't the law itself — it's that CTC structures and HRIS systems weren't built for its biggest changes.