On 14 August 2026 the Reserve Bank of India told banks that its special FCNR(B) swap facility would accept fresh deposits only through 31 August, not 30 September as first set, and advanced the swap-execution cutoff to 11 September from 16 October. Nine days earlier the RBI Governor had said there was no proposal to close it early. Then it closed.
The window had done exactly what it was built to do. By 13 August banks had mobilised $52.3 billion in Foreign Currency Non-Resident (Bank) deposits, with total inflows across the linked facilities near $56.85 billion and still climbing toward $65 billion as the deadline neared. Reserves crossed $707 billion. A scheme that works is not usually cut short. This one was, and the official line is a strong response. The more useful reading is in what the RBI stopped buying, and why.
What the subsidy actually was
The facility let banks swap fresh three-to-five-year FCNR(B) dollar deposits with the RBI at effectively zero hedging cost, and it exempted those deposits from cash reserve ratio and statutory liquidity ratio requirements. That is a large subsidy stacked on a single funding line. It made the deposits cheap for banks to book and lucrative for non-resident savers, whose dollar rates were pushed well above the ordinary 3 to 4 percent band that FCNR(B) money earns.
Every subsidised dollar did two things the central bank has to net against the reserves it gained. It added rupee liquidity to a banking system the RBI is otherwise trying to keep tight, and it parked the hedging cost of the swap on the RBI’s own book. The window, in effect, was a way to buy reserves on credit. The price rose with volume.
The cost shows up on the banks first
The bank-level maths cuts both ways, and the direction depends on where you look. On the FCNR(B) book itself, the CRR and SLR exemption is worth a lot: analysts at Systematix estimated fresh deposits could lift the margin on that slice by as much as 60 basis points. Across the wider balance sheet the sign flips. Business Standard reported estimates that the same deposits would dilute overall net interest margins by 3 to 15 basis points, because banks are paying up for multi-year dollar money they now have to reinvest or refinance as the subsidy lapses.
The window was cheap to enter and expensive to carry. That asymmetry explains the early exit better than “strong response” does. Demand had not dried up; inflows were still rising. The RBI chose to stop, and it tightened two deadlines in the same notice. A regulator that simply liked what it saw does not do that.
The window was cheap to enter and expensive to carry.
A crisis tool used without a crisis
The precedent is the tell. The last time India reached for FCNR(B) swaps at this scale was 2013, when Raghuram Rajan used them to arrest a run on the rupee and pulled in roughly $34 billion across the deposit and overseas-borrowing windows combined. That was an emergency instrument deployed in an emergency.
Reaching for the same instrument in 2026, without an acute crisis, and then withdrawing it a month early, says the RBI wanted the reserves cushion but was unwilling to keep paying the carry to build it past a level it had already decided was enough. It bought insurance, capped the premium, and moved on.
What it leaves behind
For non-resident savers the arithmetic is now plain. Concessional swap-backed pricing was available only on deposits booked by 31 August. Fresh FCNR(B) money booked from September earns standard rates, which drift back toward the 3 to 4 percent band that held before June. The elevated returns were a policy instrument, not a market signal, and the instrument has been switched off.
For the banks that leaned hardest on the window, the next few quarters carry a modest but real margin drag as the subsidised book is funded at full cost. Three to 15 basis points is not large in isolation, but it lands on lenders already fighting tight deposit competition, and it concentrates on whoever mobilised the most.
For the rupee, the signal is the one worth keeping. India now holds more than $707 billion in reserves, a number that reads as ample strength. But the way the last $52 billion arrived, through a time-limited subsidy the RBI felt it had to cap, shows the reserves line is not free-floating. Part of it was bought, on terms the central bank chose to stop extending the moment the bill outgrew the comfort. The reserves headline is a stock. The early close is the price of the flow, and it is the more honest number.
Discussion
Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.