Decoupling the Illusion: Domestic Liquidity Cannot Substitute Global Capital

Summary: India’s equity market has increasingly relied on domestic institutions and household savings to absorb foreign portfolio investor selling, but the article argues that this should not automatically be treated as evidence of structural market resilience. It traces the shift from record FPI inflows in FY24 to persistent outflows through FY25 and FY26 and further selling in FY27, while noting temporary buying in July and August 2026 before renewed selling in September. The analysis attributes the pressure to a combination of rupee depreciation, higher global yields, relative opportunities in North Asian technology markets, elevated Indian valuations and domestic transaction costs. It argues that SIP and DII flows can cushion volatility but cannot determine the global cost of equity or eliminate currency risk. The article also examines increases in listed-equity capital gains rates and Securities Transaction Tax (STT) on derivatives, alongside SEBI’s enhanced FPI disclosure framework, whose size threshold was increased from ₹25,000 crore to ₹50,000 crore in April 2025 while the 50% concentration criterion remained. The proposed reform package would reverse the April 2026 F&O STT increase, restore earlier listed-equity capital gains rates, permit STT set-off against capital gains tax, and recalibrate FPI disclosure thresholds and exemptions. It also proposes a review mechanism based on FPI secondary-market flows, derivative open interest and India’s MSCI EM weight. The article acknowledges counterarguments that domestic flows have reduced volatility and that global factors, oil prices and currency movements account for much of the pressure, but maintains that policy should distinguish genuine domestic capital formation from merely absorbing offshore selling.

Domestic Liquidity Cannot Substitute Global Capital: India’s FPI Outflows, Tax and Disclosure Reforms

Introduction

Foreign capital has been leaving Indian equities since FY25. Celebrating domestic SIPs as a “shock absorber” mistakes a transfer of risk for resilience, and the fix lies in tax and disclosure policy, not in more patriotic rhetoric.

The commonly told story of Indian equities is one of resilience. Foreign portfolio investors sell, domestic institutions buy, the index wobbles and recovers, and regulators cite the outcome as proof of market maturity. SEBI’s own annual report says domestic institutions took in a record ₹8.43 lakh crore in FY26, more than offsetting foreign equity outflows of ₹1.8 lakh crore and lifted DII ownership of NSE-listed companies to an all-time high of 17% while FPI ownership fell to a 15-year low of 15.8%. Read as accounting, that is true. Read as a verdict on market health, it is a category error. When the marginal seller is a price-sensitive global allocator and the marginal buyer is a household on autopilot, the absorption is not a firewall. It is a hand-off of risk.

The argument of this piece is narrower than “foreigners are fleeing”, and it is worth stating what the data supports. FPI selling since FY25 is structural in scale and persistent in direction, but not uninterrupted: foreigners returned in July and August 2026 before turning sellers again in September. The reversal matters. It shows that capital leaves on price and comes back on price, which is what a cost-of-equity story predicts and what a “loss of trust” story alone does not. The policy point survives either reading. India is layering transaction costs and disclosure friction on top of a currency that has done much of the damage already, and it is asking retail savers to be the buyer of last resort.

1. The Four-Year Liquidation Arc

FY24: The Peak

FY24: the peak. NSDL data show FPIs put about ₹2.08 lakh crore into Indian equities in 2023–24, and roughly ₹3.39 lakh crore across equity, debt and hybrid instruments. SEBI described the combined figure as the highest since 1992–93. Any narrative that starts the “exodus” in FY24 is wrong. FY24 is the baseline against which everything after should be judged, and it is what makes the subsequent reversal so sharp.

FY25: The Turn

FY25: the turn. Government data put FPI equity outflows at ₹1.27 lakh crore in FY25, though another tally suggests that it was ₹1.31 lakh crores. October 2024 was then the worst month on record, with about ₹94,017 crore (roughly $11.2 billion) withdrawn, more than March 2020’s Covid outflow of $7.9 billion, and the Nifty fell about 6.2% that month. Domestic institutions absorbed nearly all of it; DII net buying in calendar 2024 had already topped a record ₹4 trillion by mid-October.

FY26: A Quieter, Larger Bleed

FY26: a quieter, larger bleed. Calendar 2025 outflows were about ₹1.66 trillion. For FY26 itself, SEBI’s annual report gives ₹1.53 lakh crore, which shows that in the undercurrent, that gradually money is pulled out of the Indian Markets.

FY27: A Record and a Reversal

FY27: a record and a reversal. Foreign selling in calendar 2026 has already passed all of 2025: about ₹2.32 to ₹2.45 trillion against ₹1.66 trillion last year. The cadence was lumpy. January saw ₹35,962 crore of outflows. February brought net buying of ₹22,615 crore, the highest monthly inflow in 17 months, and then March delivered a record ₹1.17 lakh crore withdrawal. April added ₹60,847 crore. Financial services carried the burden, with net outflows reported above ₹1 lakh crore in January–August.

The July–August Reversal

The July–August reversal. FPIs bought ₹20,200 crore in July and ₹29,630 crore in August (CDSL). Roughly two months of buying reversed within weeks: September net FPI equity outflow stands at ₹17,131 crore as of 27 September (₹25,682 crore of exchange selling, partly offset by ₹8,551 crore invested in IPOs and other primary issues), and foreigners were net sellers in each of the last five weeks while DIIs bought throughout.

The IPO detail deserves attention. Foreigners are still willing to buy at issue price. What they are shedding is secondary-market exposure, where prices are set by the flows described above.

The Anatomy of a Derating

The anatomy of a derating. On 28 September the Sensex fell 1.52% to 72,771.72 and the Nifty 1.56% to 22,780.25, its lowest close since 2 April, after a setback in US-Iran diplomacy pushed crude higher. Both indices had fallen for seven consecutive weeks, shedding nearly 7% in two months. The Nifty’s 52-week range is 22,182.55 to 26,373.20. Midcap 100 and Smallcap 100 fell 1.63% and 1.85% on the day. One estimate has the Nifty down about 13% and the Sensex about 15% this year, while an earlier Reuters-cited figure put the Nifty down to 10.1%.

Two complications should be admitted rather than buried. First, the earnings line is not where the damage is. HDFC AMC reports that large-cap profitability ex-oil and gas grew about 20% year on year in Q1 FY27, which does not describe a wave of EPS downgrades, so the compression is more cost-of-capital and flows than earnings. Second, the multiple tells a mixed story. The Nifty trades near 19x earnings, while at the start of FY26 MSCI India carried a forward multiple of 24.3x against 14.9x for MSCI EM. India still commands a premium; it is just a smaller one, and it is being paid on the way down by whoever is left holding.

The breadth story has also flipped the usual script. Through the first half of FY27, small and midcaps were reported to have outperformed the Nifty by a wide margin (Smallcap 250 up about 23%, Midcap 150 about 13%, Nifty about 2%), with smallcaps at about 34x earnings, a 25% premium to their five-year average. Even so, the direction is the point: retail money is concentrated where valuations are richest, while foreigners exit large-cap financials.

Pull and Push

Pull and push. The external drivers are clear. US yields recently touched newer highs with the 10-year yield is cited near 5.227%, Brent crude traded above $100 for much of May and has stayed elevated, and global capital has chased Korean and Taiwanese hardware, a theme India barely offers. The domestic pushes are valuation, a currency in structural decline, and transaction costs that have risen in every recent Budget.

2. The Decoupling Illusion: Dollar Returns and Relative Underperformance

An offshore allocator does not earn rupee returns. The rupee closed at 95.98 per dollar on 28 September, and reports cite a record low near 96.96. If the rupee was near 83 in early FY24, the currency has lost roughly 13 to 14% of its dollar value over the arc. That single line explains more of the “flat or negative real dollar return” than any tax or disclosure rule. One report says the Nifty in dollar terms is back near September 2021 levels, even though it is much higher in rupee terms.

The relative record is poor. MSCI India returned about 4.29% in dollars in 2025, while the S&P 500 gained about 16.39%, the DAX about 22% and the Nikkei about 27%, and MSCI EM outperformed India by more than 30 percentage points. India’s weight in MSCI EM has reportedly slipped from 16% to about 11.3%, which matters mechanically: benchmark-tracking funds sell as the weight falls.

The claim that India can decouple from global cost of capital was always a claim about who sets the marginal price. When foreigners were buying, domestic growth was the story. When US real yields rose and North Asian technology offered better momentum, the marginal price setter changed its mind and the “decoupling” vanished. Domestic institutions can dampen volatility. They cannot set the cost of equity for an asset class whose valuation premium is being repriced by the global one.

3. The Shock-Absorber Fallacy

The domestic build-out is real and large. SIP inflows hit a record ₹32,297 crore in August 2026, up 14% from ₹28,265 crore a year earlier, from about 10.2 crore contributing accounts (sources range from 10.2 to 10.75 crore). Equity fund inflows were ₹29,329 crore in August, with small-cap funds taking a two-year-high ₹7,973 crore and midcap funds ₹6,989 crore. Demat accounts stood at 22.5 crore at the end of FY26. In the first quarter of calendar 2026, DIIs put in about $17.2 billion and absorbed nearly 90% of foreign outflows.

Key metric. FY26: DII net inflow ₹8.5 lakh crore vs FPI outflow ₹1.8 lakh crore. Mutual funds contributed about ₹6.4 lakh crore. DII ownership 17% vs FPI ownership 15.8% (though a second source cites 14.7%, likely on a different basis).

The policy establishment treats this as a triumph. It should be treated as a warning. A household setting up a monthly SIP is not making a valuation call; it is saving, and the saving is routed by the product. When that routing coincides with foreign selling, the household is the counterparty to a professional who has already decided the price is too high. It is worse when the flow tilts toward small and mid-cap funds, which took roughly half of August’s equity fund inflows, at a time when foreigners were leaving financials in the first fortnight of September (₹6,204 crore of financial-sector outflows, the largest of any sector).

The distinction the celebration erases is between compounding and absorption. A retail investor who buys and holds a diversified portfolio of businesses that grow earnings is a wealth creator. A retail investor whose savings arrive on a fixed schedule, irrespective of price, and are used to clear the sell orders of offshore funds is ballast. The first outcome is what policy should engineer. The second is what a “domestic firewall” narrative celebrates. The test is simple: would we describe the flow as healthy if the buyer were a pension fund of the same size and the seller a distressed foreign one? We would call that a rescue, and we would ask who was being rescued at whose expense.

None of this means SIP investors are being deceived, and it does not mean equities are a poor long-run asset. Domestic flows have arguably prevented a disorderly fall in 2024, 2025 and 2026. The claim is about incentives and framing. A system that measures success by how much domestic money it can pour into secondary markets has confused liquidity with capital formation.

4. Self-Inflicted Wounds: Fiscal Escalation and Regulatory Chokeholds

The Tax Stack

The tax stack. Budget 2024 (July 2024) lifted long-term capital gains tax on listed equity from 10% to 12.5% and short-term from 15% to 20%. In October 2024, STT on options premium rose from 0.062% to 0.1% and on futures from 0.0125% to 0.02%. Budget 2026 raised them again from 1 April 2026, to 0.15% on options and 0.05% on futures, and moved buyback taxation into shareholders’ hands as capital gains. Delivery STT was unchanged. Across the two rounds, futures STT has quadrupled and options STT has risen about 2.4 times.

For scale: at the time of Budget 2024, the Revenue Secretary put the additional yield from the capital gains rate changes at about ₹15,000 crore (a contemporaneous estimate, not an outturn). Set that against ₹2.3–2.45 trillion of calendar-2026 FPI outflows and ₹8.5 lakh crore of FY26 DII inflows, and the revenue at stake is small relative to the flow it may be deterring.

Set against a currency move of the order of 13 to 14%, the capital gains changes are second order in isolation. But two things make them matter. First, tax is the one variable Delhi controls, while the rupee and US yields are not. Second, for a multi-jurisdictional allocator running a hurdle rate, every 100 basis points of friction on a marginal decision counts, and the derivative-side STT hikes hit precisely the hedging and arbitrage activity that foreign funds use to manage the currency and index exposure. In an environment of thin dollar returns, marginal costs decide marginal flows.

The Disclosure Overhang

The disclosure overhang. SEBI’s August 2023 circular required granular look-through disclosure from FPIs with more than 50% of equity AUM in a single group or more than ₹25,000 crore of Indian equity AUM. In April 2025 SEBI doubled the size threshold to ₹50,000 crore, citing more than a doubling of cash-market volumes since FY23, kept the 50% single-group trigger, and exempted broad-based pooled vehicles and government-related investors subject to conditions. The brief’s premise of a ₹25,000 crore trigger is therefore out of date, and the chilling effect on sovereign funds is weaker than often claimed, since they are exempted. Though, there is no empirical evidence that any sovereign fund or endowment has withdrawn because of the rule. What remains is a compliance burden and operational risk for large, multi-manager platforms above the threshold, and a regulatory posture that reads to an outsider as suspicion. That signal costs capital even when the rule is workable.

5. The Reform Blueprint: A Costed Rollback and a Few Regulatory Fixes

A reform is worth proposing only if it changes the marginal decision of a global allocator at a cost the exchequer can name. Three facts from the sections above constrain the design. First, the rupee and US yields are outside Delhi’s control and explain most of the dollar shortfall, so no tax change alone restarts the flows. Second, the increases India has made since mid-2024 fall on precisely what foreign funds do: realise gains and hedge with derivatives. Third, the revenue at stake is modest and, for STT, unreliable. FY26 STT collections were ₹57,522 crore against a Budget estimate of ₹78,000 crore, a shortfall of about 26%, which is consistent with a levy whose base is a trading volume that reacts to the levy. The proposal is therefore a rollback, with its cost stated.

The Package, With Static Annual Revenue Cost

The package, with static annual revenue cost. Move 1, reverse the April 2026 F&O STT hike: about ₹10,000 crore. Move 2, restore 10% long-term and 15% short-term capital gains rates: about ₹15,000 crore. Move 3, STT set-off: not costed. Combined, roughly ₹25,000 crore a year. Both figures were estimates made when the Budgets were presented (a market-expert estimate for STT, the Revenue Secretary’s estimate for capital gains) and these numbers must be re-run on current volumes and gains before you rely on them.

Move 1: Reverse the 2026 F&O STT Hike

Move 1: reverse the 2026 F&O STT hike (futures back to 0.02%, options premium back to 0.10%). Futures STT has risen fourfold in under two years, from 0.0125% to 0.05%. For a foreign fund the derivative is the hedge, and the tax lands on the instrument used to manage exactly the currency and index risk this piece has documented. The stated aim was to curb retail speculation and losses. That aim is better served by conduct rules, such as the upfront premium collection and removal of expiry-day calendar-spread benefits already introduced, than by taxing the hedger. The cost is real and probably larger than the Budget-day figure: collections through 17 September were ₹40,214 crore, up about 53% year on year, against a FY27 Budget estimate of ₹73,700 crore. The trade-off is that retail F&O losses continue unless the conduct rules are enforced harder.

Move 2: Restore 10% LTCG and 15% STCG

Move 2: restore 10% LTCG and 15% STCG, keeping the ₹1.25 lakh exemption. The illustrative arithmetic in Section 4 shows the 2024 rate rise trimmed post-tax gains by only about 2.8% (long-term) and 5.9% (short-term), so the case is not that this alone brings foreigners back. The case is the signal: three separate increases on equity between July 2024 and April 2026, and none reversed, is a pattern an allocator prices as a rising risk premium on India. A full reversal also keeps a five-point gap between short and long-term rates, so the incentive to hold remains. The cost is about ₹15,000 crore statically, on a gains base that has since grown. The political trade-off is plain: a tax cut on market gains sits badly beside retail losses in derivatives, which is why Move 1 and the conduct rules must travel together.

Move 3: Allow STT Paid to Be Set Off Against Capital Gains Tax

Move 3: allow STT paid to be set off against capital gains tax. STT is levied on turnover regardless of profit, then capital gains tax is levied on top. Set-off, capped at the tax payable, removes the stacking. I could find no public data on STT paid by taxpayers with net gains, so this cannot be costed from outside the Finance Ministry; the ministry should publish the estimate. Treat it as the second-phase structural fix.

Sequencing and the Test

Sequencing and the test. Move 1 first, because it is the newest, the most directly linked to hedging depth, and the cheapest to reverse. Moves 2 and 3 in the Budget that follows. Attach a two-year review: if FPI net secondary flows, derivative open interest and India’s MSCI EM weight do not stabilise while the rupee does, the rollback has failed its own test and should be reassessed. What it will not do is offset a rupee near 96 to the dollar or US yields near 5.227%. Pretending otherwise would repeat the mistake of the decoupling story.

Regulatory Fix A: Index the Disclosure Threshold to Turnover

Regulatory fix A: index the disclosure threshold to turnover. SEBI raised the look-through threshold from ₹25,000 crore to ₹50,000 crore in April 2025 because cash-market volumes had more than doubled since FY23. That logic implies a fixed rupee threshold decays every year. Reset it annually to a fixed multiple of average daily cash turnover, so it does not depend on a fresh board decision each time.

Regulatory Fix B: Trigger on Behaviour, Not Size

Regulatory fix B: trigger on behaviour, not size. SEBI’s stated concern was circumvention of minimum public shareholding norms and Press Note 3. Keep the 50% single-group test. Replace size-only look-through with a company-level trigger for an FPI, together with related entities, whose holding in one company’s free float crosses a level SEBI sets. I do not propose the level; that is a calibration for the regulator. This targets the risk the rule was written for and leaves large diversified funds alone.

Regulatory Fix C: Make the Exemptions Predictable

Regulatory fix C: make the exemptions predictable. Broad-based pooled vehicles and government-related investors are already exempt subject to conditions. The remaining cost is uncertainty. Codify the conditions and set a fixed deadline for SEBI to rule on exemption claims.

One Uncosted Idea Worth Testing

One uncosted idea worth testing. Foreigners put ₹8,551 crore into primary issues in September while selling ₹25,682 crore in the secondary market, and public equity raised a record ₹2.35 lakh crore in FY26. A lower long-term rate for shares acquired in primary issues and held for several years would tilt both foreign and household money toward capital formation instead of bidding up existing shares.

The Counter-Case

The counter-case. A fair sceptic would say the following. Domestic buying has reduced volatility and lowered India’s dependence on hot money. Earnings growth remains intact. The outflows track a global rotation into AI hardware and a rupee driven by oil, not by Indian policy. Foreigners came back for two months this summer, so capital is not permanently lost. And the government’s rationale for the F&O hike, retail losses, is a real problem. Each of these has force. What they do not answer is the distributional question: who bears the risk when foreigners de-risk and prices fall? Until policy treats retail savers as owners to be protected, not as liquidity to be harvested, the answer will always remain the household.

Sources used: NSDL, CDSL, AMFI, SEBI annual report and circulars, Union Budget 2024 and 2026 coverage, STT collection reports, Business Standard, Reuters-cited market reports, HDFC MF and HDFC AMC research.