So I modelled it, using official country-wise trade data pulled directly from the Commerce Ministry's TradeStat database for FY2016-17 through FY2025-26. The arithmetic produces one number that makes the case look overwhelming ($228.5bn in BRICS-bloc trade exposure) and a second number that quietly destroys it ($93.4bn actually reroutable to rupee settlement).

Taking the ten BRICS members and partners for which bilateral data exists, here is FY2025-26 — the year just closed:

The headline number: $321.8 billion, or 41.5% of India's entire merchandise import bill of $774.98 billion. That is the size of the prize. The BRICS bloc also accounts for $226.1 billion of bilateral deficit — 67.8% of India's total merchandise trade deficit. On the face of it, no other policy lever touches so much of India's dollar demand at once.

Bilateral local-currency settlement is not a payment technology problem. It is an accounting identity problem.

When India pays China in rupees, China ends up holding rupees. It can do three things with them: buy Indian goods, invest in Indian assets, or sit on them. India's exports to China are $19.5 billion. Its imports are $131.6 billion. So after the trade nets out, China is left holding $112.1 billion of rupees a year with no corresponding Indian goods to spend them on.

That gives the model its two critical quantities. The self-financing portion is the volume that both legs can cover — the smaller of imports and exports, country by country. Across BRICS that is $93.4 billion. Everything above it is the residual: $228.5 billion a year of rupees that the counterparty must be willing to accumulate and hold.

This is the point most commentary misses. Settling a balanced trade flow in local currency genuinely removes dollar demand and costs nobody anything. Settling a deficit in local currency does not remove the obligation — it converts it into a rupee claim on India held by a foreign government. That is a capital inflow. It is economically identical to China or Russia buying Indian government bonds, and it is subject to exactly the same question: do they want to?

S2 is real policy. It is achievable, it is sustainable, and at 12% of the import bill it would be a genuine and material achievement — roughly five times what India manages today.

S3 and S4 are where the arithmetic breaks. To put $228.5 billion a year in perspective: India's entire foreign exchange reserves are $682.2 billion. Three years of S3 would create foreign rupee claims on India roughly equal to those reserves. Five years would create claims exceeding them by two-thirds.

Ten years of data reveal something the annual snapshot conceals. India's imports from BRICS grew 124% between FY2016-17 and FY2025-26, from $143.4 billion to $321.8 billion. Its exports to BRICS grew only 52%, from $63.0 billion to $95.8 billion. The deficit grew 181%.

The single most important line in this analysis is the coverage ratio — exports as a share of imports, which is what determines how much trade can be settled in local currency without anyone accumulating unwanted claims. In FY2016-17 India's exports covered 44% of its BRICS imports. By FY2025-26 that had fallen to 29.8%.

The de-dollarisable share of India's BRICS trade is not growing. It is shrinking, and it has been shrinking for a decade.

This is not a theoretical concern, because India has already tested it at scale. After 2022, sanctions pushed India–Russia trade substantially out of dollars, and Russian crude went from about 2% of India's oil imports to roughly a third. Russian exporters duly accumulated rupees they could not spend, because India sells Russia almost nothing — $4.5 billion against $55.4 billion of purchases in FY2025-26, of which around 86% was crude oil.

Sergey Lavrov said the quiet part out loud in Goa in May 2023: this is "a problem," and the money needed to be converted into something else. Reported standing balances of roughly $7–8 billion in late 2023 were run down to about $3–3.5 billion by September 2024 — but the mechanism matters, and it is not the flattering one. The balance fell largely because the oil trade migrated to dirhams and yuan, not because Russia spent the rupees on Indian goods. The chronology settles it: the RBI only permitted surplus vostro balances to be invested in government securities in August 2025, and the Sberbank-sponsored Nifty-linked fund for Russian investors followed in December 2025 — both well after the balance had already halved, so neither can explain a drawdown that preceded them. That fund, in any case, holds structured notes issued by Sberbank CIB rather than Indian shares, so it moves no capital into India at all. Faced with a choice between holding rupees and using a third currency, the largest creditor chose the third currency.

There is a further problem that changing the payment rail does not touch. Russian crude is priced off Brent. Chinese electronics are priced in dollars. Paying in rupees at a dollar-referenced price leaves the underlying dollar exposure exactly where it was — it simply moves the conversion to a different point in the chain.

This connects to the finding in the previous article on exchange rates and exports: under the dominant-currency paradigm, pass-through to the dollar runs at roughly 0.78 while pass-through to actual trading-partner currencies is about 0.16. The dollar is doing the pricing work whether or not it does the settling work. The RBI's own Inter-Departmental Group put dollar invoicing at about 86% of both India's exports and imports. Settlement reform attacks the smaller half of the problem.

Here the honest answer deflates the whole thesis. India's merchandise imports of $775 billion spread over roughly 250 trading days is about $3.1 billion a day of dollar demand. Global USD/INR turnover was around $185 billion a day in the BIS Triennial Survey of April 2025.

So India's entire import-related dollar demand is about 1.7% of daily USD/INR turnover. Removing all BRICS imports from the dollar — the maximalist S3 — would remove roughly 0.70% of daily turnover. The rupee's level is set by capital flows, positioning and the dollar's global cycle, not by the settlement currency of trade invoices.

And there is a deeper reason the exchange-rate benefit is smaller than advertised: local-currency settlement does not shrink the current account deficit. India still consumes more real resources than it produces and must still finance the gap. Changing the currency of the invoice changes the denomination of the obligation, not its size. If anything, the partner's accumulated rupees are a claim that can later be sold — deferred depreciation pressure rather than avoided depreciation pressure.

What does improve is genuine but narrower: fewer forced conversions, lower transaction costs (the RBI cited 5–6% conversion costs when nudging banks toward rupee-dirham settlement), a smaller gross intervention footprint for the RBI, and real sanctions resilience. As the central-bank levers piece showed, every dollar the RBI sells drains about ₹88 of domestic liquidity that then has to be replaced through open-market operations. Reducing the gross flow reduces that treadmill. That is a plumbing benefit, not an exchange-rate benefit.

In January 2026 the RBI proposed linking BRICS members' central bank digital currencies — India's e-rupee, Brazil's Drex and their counterparts — into an interoperable network for cross-border settlement, explicitly not a single BRICS currency. It is on the 2026 summit agenda.

Read against this model, the proposal is well-targeted but addresses only one of the two constraints. Linked CBDCs attack friction: correspondent banking costs, settlement time, SWIFT dependence, and the compliance timidity that has hobbled the existing rupee mechanism. That is exactly the barrier holding back scenario S2, the achievable one.

What linked CBDCs cannot do is make China want to hold $112 billion of rupees a year. No payment technology solves an accounting identity. If the counterparty does not want the currency as a store of value, faster rails simply deliver unwanted balances more efficiently.

It is worth noting that the bloc itself is candid about this. Putin said in November 2024 that a common currency was premature and "we do not have such goals among ourselves"; India has opposed a shared currency; South Africa's own representative called de-dollarisation "not practical or economically viable." The 2025 Rio declaration encouraged local-currency trade without a single quantified target.