A balance of payments moving into surplus is read as evidence that the world has repriced a country’s assets upward. Sometimes it is. Sometimes it is evidence that the country has paid enough to borrow the difference, and the two look identical in the headline number.
India’s external accounts through 2026 offer an unusually clean case for separating them. Reserves sit near a record, gross foreign direct investment has hit a fifteen-year high, and a special swap window pulled in $136 billion in under three months. What follows concerns the composition of those flows rather than their size - specifically, which of them can be recalled, and on whose decision.
Gross FDI Is at a Fifteen-Year High. Net FDI Is Not.
India recorded $30.7 billion of gross FDI in April-June 2026, the strongest quarter in fifteen years, against a net FDI print of $7.8 billion. Roughly three-quarters of the gross figure did not stay.
Both numbers are real and they describe different things. Gross FDI measures interest in the asset class. Net FDI measures what is actually financing the external account. A commentary cycle that leads with the fifteen-year high and mentions the net figure in a subordinate clause is not wrong on the facts, but it is reporting the more flattering of two series that have diverged.
Repatriation Has Roughly Doubled in Four Years
The gap between the two is repatriation, and it is not a one-quarter artefact. Gross FDI repatriation outflows rose from around $20 billion a year in 2021 to roughly $40 billion a year by 2025. Repatriation and disinvestment in June 2026 alone ran at $5.8 billion, 36% higher than the same month a year earlier.
This is a preference, not a failure. Foreign investors taking profit on Indian positions against a valuation gap with other markets are behaving rationally, and the same openness that lets capital in is what lets it out. The point is narrower: a rising repatriation run-rate means each incremental dollar of gross inflow finances less of the external account than the dollar before it did.
The Swap Window Changed the Financing Mix, Not the Current Account
Against that backdrop the RBI’s FCNR(B) swap facility drew $136.38 billion between June and August 2026. That is a large number against a current account deficit running near 1.5% of GDP, and it has visibly steadied the rupee.
It is also deposits. The money arrives on a three-to-five-year tenor, pays a coupon, and leaves on a schedule known at inception. Set beside net FDI of $7.8 billion in the same quarter, the composition of India’s external financing has shifted toward the recallable end - not because the equity flows fell away, but because a much larger volume of dated money arrived alongside them.
Reserves Near a Record Say Less Than They Appear To
Headline reserves crossed $701 billion in January 2026, close to the all-time peak, and cover roughly eleven months of imports. On a balance-of-payments basis, however, reserves were drawn down by $23.6 billion across 2025-26, against $5.0 billion the year before.
The two statements are compatible: valuation effects and swap-funded accretion can lift the stock while the flow is negative. But they answer different questions. The stock answers what is on hand. The flow answers whether the external account is self-financing at the current exchange rate. In 2025-26 it was not, and the gap was bridged.
What This Means for Allocators
Read net FDI, not gross, when assessing the durability of the external account. The two series have moved apart far enough that they now support different conclusions.
Separate reserve stock from reserve flow. An adequacy ratio built on a stock that includes swap-funded accretion is answering a narrower question than it appears to, and the maturity schedule underneath it is knowable rather than uncertain.
And treat a surplus as a question rather than an answer. Surplus is a signal. Whether it signals strength or subsidy depends entirely on what financed it - and for now the honest answer is some of each.
Back to Writing