NextFin News - HSBC bought at least $3 billion of Indian government bonds after a flood of foreign-currency deposits left the bank holding more rupees than it wanted, converting the Reserve Bank of India's $72.8 billion campaign to rebuild external buffers into a forced bid for local debt. The trade is the clearest signal yet that the RBI's deposit drive has spilled out of the balance-of-payments ledger and into the bond market - and it raises a question the recent rally has not answered: is this durable foreign demand for India, or a liquidity wave that reverses the moment the deposit window shuts?
The Deposit Boom That Became a Bond Bid
The mechanics are straightforward, and that is what makes them powerful. In June, the RBI opened a special USD-INR swap facility so that banks could raise foreign-currency non-resident (FCNR(B)) deposits from overseas Indians and hedge the currency risk with the central bank at a concessional rate. Under the facility, a bank takes in dollars from a depositor, hands those dollars to the RBI under a swap, and receives rupees on its domestic books at a pre-agreed rate for the life of the deposit. The objective was to rebuild India's external buffers after a bruising 2025-26, when the rupee fell 11% and the capital surplus shrank to 0.1% of GDP. The scheme worked faster than policymakers expected.
As of 21 August, inflows under the facility reached $72.848 billion, the RBI said in a press release dated 22 August and signed by chief general manager Brij Raj. FCNR(B) deposits alone accounted for $65.397 billion, with overseas foreign-currency borrowings at $4.86 billion and external commercial borrowings at $2.591 billion. The pace forced the central bank's hand: on 14 August it advanced the closure of the deposit-hedging window by a month, to 31 August, because inflows had already exceeded $50 billion. India's foreign-exchange reserves climbed to $716.9 billion in the week to 14 August, their highest level in about six months, up nearly $10 billion in a single week.
Here is where the bond trade enters. When a bank such as HSBC raises a dollar deposit from a non-resident Indian and swaps it with the RBI, it receives rupees on its domestic books. Those rupees cannot sit idle earning nothing; a bank flush with local liquidity must deploy them, and the most natural buyer-of-first-resort is government paper. HSBC, State Bank of India and ICICI Bank together mobilized half of all inflows under the scheme, official data showed - foreign banks booked $8.4 billion, private lenders $10.7 billion and public-sector banks $8.8 billion. HSBC's $3 billion bond purchase is the redeployment leg of that chain: dollars in, rupees created, bonds bought.
The rupee, the intended beneficiary, has barely moved. Since the scheme launched on 8 June, the currency has depreciated just 0.7%, compared with a 6% slide in the first five months of the year. On 24 August the RBI reference rate stood at 95.726 per dollar, little changed from 95.747 three days earlier. Stability is the point - but stability purchased with a wall of hedged inflows is not the same thing as stability earned through structural current-account strength.
Why This Is a Liquidity Wave, Not a Regime Shift
The bond market has treated the inflow surge as evidence that foreign demand for Indian debt has turned a corner. That read confuses the plumbing with the demand. The transmission runs in three steps: an FX deposit inflow creates swapped rupee liquidity at the bank; the bank, holding excess local funding, buys local assets; yields fall as a mechanical consequence of that forced buying. The buyer is not a pension fund making a strategic allocation to India; it is a bank managing a balance sheet swollen by a policy-designed, time-limited facility.
Three facts mark this as cyclical rather than structural. First, the window closes on 31 August for FCNR(B) deposits. RBI Governor Sanjay Malhotra said at a post-policy briefing on 5 August that there was "no proposal under consideration to close the scheme prematurely," yet the central bank pulled the deadline forward nine days later - a reminder that the facility is a tap, not a permanent source of funding.
"We have got robust flows as you have mentioned and we do hope to get good healthy flows going forward. But as of now, there is no proposal under consideration to close the scheme prematurely. We will keep you posted on that," Malhotra said, adding that the country's balance of payments was expected to "register a healthy surplus this year."
Second, the deposits are hedged: the currency risk sits with the RBI, not the depositor, so the inflow is a synthetic dollar loan to the sovereign rather than a conversion of foreign savings into rupee assets. Third, the liquidity is acknowledged as temporary even by the central bank, which has flagged eventual normalization beyond September as currency in circulation rises and forward-book maturities resume.
The structural case rests on a different foundation, and it is not without merit. Foreign portfolio investor holdings of government securities reached 3.75 trillion rupees, or 3.34% of the outstanding stock, by 12 May, with the Fully Accessible Route accounting for 6.74% of its segment. The government has widened the FAR, exempted FPI interest income from certain taxes and simplified investment norms - changes that lower the cost of owning Indian debt for good. In June alone, overseas investors poured a record 55,518 crore rupees into Indian bonds.
But those reforms operate on a different timescale than a $65 billion deposit sprint. The right way to separate the two forces is to treat the deposit wave as the cyclical leg - mean-reverting as the swap book rolls off - and the FAR reforms plus record reserves as the structural leg that raises the floor for foreign ownership. HSBC's trade sits in the cyclical leg. It is a liquidity arbitrage, not a conviction allocation.
And the market already knows the window is closing. The rally has persisted not because investors believe the flow is permanent, but because they believe the liquidity overhang will take time to drain. That belief is the fragile link in the chain.
The Second-Order Effect: A Rally That Complicates the RBI's Job
The first-order effect of the bond buying is obvious: yields fall and prices rise. The 10-year benchmark yield ended at 6.8343% on the final Friday of July, one basis point higher for the week after an 11-basis-point rise over the preceding two weeks - a reminder that the direction of travel has not been one-way. The second-order effect is what matters for the next six months, and it cuts against the comfort the rally provides.
A forced domestic bid for government paper suppresses the term premium the market would otherwise demand for India's fiscal and external risk. That makes sovereign borrowing look cheaper than it would in an unhedged market, and it narrows the space for the RBI to cut rates further without reigniting the very currency pressure the deposits were meant to relieve. The central bank is caught in a familiar trap: it invited the dollars to defend the rupee, and the rupee stability those dollars bought has made domestic rates look rich, which invites more carry-seeking flow - until the hedge cost or the window closure breaks the loop. This matters because the RBI has already cut the repo rate by 125 basis points since February 2025, taking it from 6.50% to 5.25%; each additional cut now carries a higher currency cost than it did before the inflow wave arrived.
There is also an expectation gap building, and it shows up in the positioning data. Foreign investors net sold almost $600 million of Fully Accessible Route bonds in six sessions on uncertainty over oil prices and index-inclusion timing, even as they remained net buyers of $3.9 billion of FAR bonds between 1 June and 31 July. The market has priced the inflow wave as if it were durable foreign demand for Indian bonds. It is not - and the short-term outflows are the tell. When the FCNR(B) window closes on 31 August, the flow that created the liquidity surplus stops, and the liquidity-absorption question the RBI deferred returns.
HDFC Bank captured the conditional nature of the moment in a post-policy note: "There was no hint towards the need for any durable liquidity-absorption measures as of now," the bank wrote, but added that the room for such measures "would be determined by the extent of dollar flows that come in."
That conditional is the whole story: the liquidity is a function of the flow, the flow is a function of the window, and the window closes on 31 August. The index-inclusion disappointment underscores the point - even as the deposit facility pulled in tens of billions, the index provider deferred adding Indian bonds, signaling that reforms must be sustained rather than merely announced before structural money commits.
Pranjul Bhandari, HSBC's chief India economist, said in a July interview that FCNR deposits would bring in about $50 billion and that external commercial borrowings and other channels would add another $10 billion to $15 billion over a three-month period. Those forecasts were met - and then exceeded. But a forecast about a time-bound facility is a forecast about a wave, not a tide.
The Counter-Thesis: Why the Bulls Could Be Right
The strongest case against the cyclical reading is that the deposits are stickier than the facility that attracted them. FCNR(B) deposits under the scheme carry maturities of three to five years and a mandatory one-year lock-in. The $65.4 billion is not overnight money; it is committed for years, and the reserves it built - $716.9 billion, a six-month high - buy the RBI credibility that outlasts the window. A central bank with that much dry powder can defend the rupee through a shock, and that credibility itself attracts the structural investors the cyclical argument says are absent.
The balance-of-payments math also improved faster than expected. India's FY27 surplus was revised to $40 billion from $25 billion, with the current-account deficit estimated at 1.7% of GDP. After two years in which the capital surplus shrank to 0.4% and then 0.1% of GDP, a $40 billion surplus is not a rounding error - it is a regime repair. If the deposit base proves sticky and the FAR reforms keep pulling in real money after the window closes, today's forced buying will be remembered as the moment the market noticed India again.
This counter-thesis is substantial, and it is why the structural leg of the story deserves weight. But it does not change the nature of HSBC's $3 billion trade. Even if the deposits prove sticky and the reserves prove durable, the specific bond purchase was triggered by a temporary liquidity surplus at the bank, not by a reassessment of India's long-run credit. The counter-thesis wins the medium-term argument about foreign ownership; it does not win the near-term argument about what forced this trade.
The signal that would falsify the cyclical call is precise. If net FPI holdings of Indian bonds fail to grow in the 60 days after the 31 August window closure, and the 10-year yield rises above 7.1% as liquidity absorption begins, then the rally was a deposit artifact, not a demand shift. Conversely, if foreign ownership keeps climbing after the window shuts and the yield holds below 6.6%, the structural bulls are vindicated and the cyclical label was too harsh.
What Comes Next: Three Horizons
Short term (weeks): the bond rally has room to run while the liquidity surplus persists, but the 31 August closure is a hard date to watch. Any sign that the RBI begins durable liquidity absorption - larger reverse repos or open-market sales - would be the first evidence that the forced-bid phase is ending. Beneficiaries are holders of duration; the exposed are banks that raised deposits they cannot redeploy at a spread.
Medium term (quarters): the test is whether FPI ownership of government securities climbs above the 3.34% share recorded in May without the swap facility pulling new money. The base case is a pause: yields stabilize in the high-6% range as the liquidity overhang drains. The upside case is continued FAR-driven inflows that push the 10-year toward 6.5%. The downside case is a reversal above 7.1% if the deposit book rolls off faster than structural demand replaces it.
Long term (years): the structural leg - tax exemptions, a wider Fully Accessible Route, and reserves large enough to deter speculative attacks on the rupee - is the part of this story that will outlast the headlines. If India can convert a temporary deposit wave into a permanently higher foreign-ownership floor, the forced buying of August will be remembered as the inflection point.
The rupee's calm and the bond rally are real, but they were bought with a facility that has an expiration date. HSBC's $3 billion purchase tells you where the money went; the weeks after 31 August will tell you whether it stays.
Explore more exclusive insights at nextfin.ai.
