Where the India-US Tax Relationship Actually Stands in 2026: The Remittance Tax That Landed, the Totalisation Gap That Did Not Close, and What Is Real Versus Wishful

In the first half of 2026 three things were true at once for an Indian living and working in the United States, and the news cycle blurred all of them together. A new 1% federal tax on money sent out of the country became law and took effect on 1 January 2026. Indian workers on H-1B and L-1 visas kept paying 7.65% of their wages into a US Social Security system most of them will never collect a cent from, because India and the US still have no totalisation agreement. And the treaty that governs how the two countries divide the right to tax your income, signed in 1989, stayed exactly where it has been for thirty-six years, including the much-misunderstood 25% cap on dividends. Around all of this swirled talk of a "treaty renegotiation" that, on the evidence, is not actually happening.

The 30-second answer: Three real things changed or stayed in 2026. First, the 1% remittance excise tax under IRC Section 4475 is now law, effective 1 January 2026, but it only hits transfers funded by cash or physical instruments; money wired from a US bank account is exempt, so most NRIs pay nothing. Second, there is still no India-US totalisation agreement, so H-1B and L-1 workers keep losing the 6.2% Social Security plus 1.45% Medicare they pay, an estimated 4 billion dollars a year in forfeited contributions. Third, the 1989 DTAA is unchanged: its 25% dividend cap for individuals stands, the 15% rate applies only to corporate holders of 10%-plus stock, and a full treaty renegotiation is rumoured, not real.

This is a news-analysis piece, not a how-to. The job here is to separate what actually became law from what is being talked about, put real numbers on what each item costs you, and tell you where to spend your attention. If you want the mechanics of claiming treaty relief or sending money home efficiently, the dedicated guides linked at the end go deeper. What follows is the lay of the land in 2026, honestly drawn.

The remittance tax landed, but smaller and narrower than the headlines

Start with the item that generated the most panic in the NRI WhatsApp groups, because it is also the one where the panic was most overdone.

The original proposal, in early 2025, was a 5% excise tax on money sent abroad by non-citizens. That number genuinely would have stung. During the legislative process it was cut to 3.5%, then cut again, and the version that the One Big Beautiful Bill Act actually made law, when the bill was signed on 4 July 2025, is 1%. It is codified as a new excise tax under IRC Section 4475 and applies to remittance transfers made on or after 1 January 2026.

Here is the part the headlines mostly buried, and it is the part that matters for you. The 1% applies only to remittance transfers funded with cash, a money order, a cashier's cheque, or a similar physical instrument. Transfers funded from an account at a US financial institution, or paid with a US-issued debit or credit card, are exempt. The tax is collected by the remittance provider and remitted quarterly to the IRS, and Treasury and the IRS issued proposed regulations and even penalty relief for providers during early 2026 as the machinery came online.

Read that exemption carefully, because it is the whole story for a salaried NRI. If you send money to India the way almost every white-collar Indian in the US already does, by an ACH pull or a wire from your US checking account, or through Wise, Remitly, ICICI Money2India or your bank funded from that account, you are exempt. The 1% is designed to fall on cash sent through storefront remitters, which is overwhelmingly used by undocumented and lower-income migrants, not by the H-1B engineer wiring money to an NRE account.

Put a number on it so the scale is unmistakable. Suppose Anjali, a software engineer in Seattle, sends Rs 20 lakh equivalent, about 24,000 dollars, to her NRE account over the course of 2026, all by wire and online transfer funded from her US bank. Her remittance excise tax for the year is zero dollars, because every one of those transfers is account-funded and exempt. Now the counterfactual that some commentary implied: had the 1% applied to all her transfers, she would have paid 240 dollars for the year. And had the original 5% proposal applied to all of it, 1,200 dollars. The gap between the feared outcome and the real one, for a typical digital sender, is the difference between 1,200 dollars and nothing.

The people who do pay are the people who fund transfers with physical cash. A construction worker walking 500 dollars in cash into a money-transfer storefront pays 5 dollars in excise tax on that transfer. It adds up over a year for someone sending small cash amounts frequently, but it is structurally a tax on cash, not a tax on Indians, and certainly not a tax on the salaried NRI the panic was aimed at. US citizens are exempt entirely, regardless of how they fund the transfer, which tells you the politics of the measure more than its tax logic does.

The honest framing on the remittance tax: it is real, it is law, and for the reader of this site it is a non-event as long as you keep funding transfers from a US bank account rather than with cash. The one behavioural change worth making is simply never to fund an India transfer with cash or a money order from 2026 onward. Beyond that, carry on. For the mechanics of moving money home cleanly, see sending money to India.

The totalisation gap is the expensive one nobody made law about

If the remittance tax got the headlines, the totalisation gap deserves them, because it quietly costs Indian workers far more and nothing happened in 2026 to fix it.

Here is the mechanism. From your first US paycheque, whether you are on H-1B, L-1 or any other work visa, 6.2% of your wages goes to Social Security and 1.45% goes to Medicare, a combined 7.65% (and your employer matches it). To draw a US Social Security retirement benefit you generally need 40 quarters of covered earnings, ten years. An Indian professional who comes over, works six or seven years, and returns to India, or whose H-1B simply does not get extended, walks away having paid in for years and qualifying for nothing.

A totalisation agreement is the instrument that fixes exactly this. Where two countries have one, a worker can combine the social-security credits earned in each to reach the qualifying threshold, contributions are not duplicated across both systems during a temporary posting, and benefits become portable. The US has these agreements with more than 30 countries, including the UK, Canada, Germany, Japan and South Korea. India is the conspicuous exception among the large sources of skilled migration, despite Indian officials pressing for an agreement since 2007.

The cost is not abstract. Industry estimates put the Social Security and Medicare collected from Indian nationals who will never qualify to claim at roughly 4 billion dollars a year. At the individual level the loss is just as concrete.

Put it on one worker. Take Vikram, on H-1B in Texas, earning 150,000 dollars a year, who works in the US for seven years and then returns to Bengaluru. His Social Security contribution alone is 6.2% of 150,000, or 9,300 dollars a year. Over seven years that is 65,100 dollars of his own contributions (set aside the employer match, which is also gone). Because he never reaches 40 quarters and there is no agreement to let him combine his US quarters with his Indian EPF service, that 65,100 dollars is simply forfeited. Add the Medicare 1.45%, another 15,225 dollars over the period, and the total he pays into US social insurance and will never see is roughly 80,000 dollars.

Now the counterfactual that shows what an agreement would be worth. Had a totalisation agreement existed on the German or Canadian model, Vikram's seven years of US credits could have been combined with his Indian contributory service to qualify him for a (proportional, "totalised") benefit, or at minimum he would have been spared the Social Security leg entirely during a defined detachment period, putting that 9,300 dollars a year back in his pocket. The difference between the two worlds, for one worker over seven years, is in the tens of thousands of dollars. Multiply across the H-1B population and you arrive at the 4-billion-a-year figure.

What actually happened on this in 2026? Nothing concrete. As of the most recent ministerial commentary, the Indian government said it would continue to pursue an agreement but that the issue was "not currently on the agenda" and would be taken up "once there is a new Congress in place". That is diplomatic language for "no progress". A totalisation agreement requires US legislative cooperation, and it has not been a priority for any recent administration. So the honest read here is bleak: this is the single most expensive feature of the India-US tax-and-social-insurance relationship for the working NRI, it is entirely unaddressed, and you should plan as if it will not change during your US stint. The defensive moves, maximising retirement savings you can actually keep and understanding what does and does not survive your return, are in social security and totalisation agreements.

The 1989 treaty did not change, and its 25% dividend cap is widely misunderstood

The third pillar is the Double Taxation Avoidance Agreement itself, the 1989 India-US treaty that entered into force on 21 December 1990. Thirty-six years on, it is unchanged, and the most common mistake NRIs make about it concerns dividends.

You will read, on a dozen finance blogs, that the treaty gives you a 15% rate on US dividends. For an individual investor that is wrong. Article 10 of the treaty caps US withholding on dividends at 15% only where the beneficial owner is a company that owns at least 10% of the voting stock of the company paying the dividend. Everyone else, which means every individual retail investor regardless of holding size, falls into the 25% bucket. So when your US brokerage pays you a dividend on Apple, Microsoft or an S&P 500 fund, the default treaty-reduced withholding for you as an Indian tax resident is 25%, not 15%.

This matters in two directions depending on which side of the relationship you sit. If you are an Indian resident (or an RNOR) holding US stocks, the US withholds 25% and you claim that as a foreign tax credit in India against the Indian tax on the same dividend, using Form 67, so you are not taxed twice. If you are a US-resident NRI holding Indian stocks, India withholds on your Indian dividends and you credit that against your US tax. Either way the treaty's job is to prevent double taxation, not to make the tax cheap, and the 25% individual rate is the number to plan around.

See the size of the misconception on a real figure. Suppose Ravi, an Indian resident, holds US dividend stocks paying 5,000 dollars of dividends in a year. At the correct 25% treaty rate, US withholding is 1,250 dollars. Had the mythical 15% rate applied, it would have been 750 dollars. The 500-dollar gap is not money he loses, because he credits the US tax against his Indian liability, but it is 500 dollars of his money parked with the US Treasury until he files and claims the credit, exactly the kind of float that catches people who budgeted for 15%. The fix is not to chase a rate that does not exist for individuals; it is to file Form 67 correctly and on time so the credit actually lands. The mechanics are in the India-US DTAA deep dive and foreign tax credit and Form 67.

There are other parts of the 1989 treaty that experienced NRIs argue about, and they too are unchanged in 2026: the treatment of US retirement accounts and whether Roth distributions are protected under Article 17 is genuinely debated, and the saving clause lets the US tax its own residents and citizens largely as if the treaty did not exist for many items. None of that moved this year. The treaty is the same instrument it was in 1990, and anyone telling you a new dividend rate or a new pension rule arrived in 2026 is mistaken.

FATCA reporting got faster, and the reciprocity that was promised still has not arrived

One piece of the relationship genuinely did tighten, just not in your favour, and it is worth understanding because it changes the risk calculus on under-reporting.

Under the FATCA Model 1 Intergovernmental Agreement that India signed in 2015, Indian banks and fund houses report details of accounts held by US tax residents to India's Central Board of Direct Taxes, which passes them to the IRS; the US is meant to provide reciprocal information back to India. What changed by 2025-26 is the speed: the digital exchange between the CBDT and the IRS is now close to real-time. The practical implication for a US-resident NRI is simple and worth stating bluntly. Your Indian accounts, deposits and mutual-fund holdings are visible to the IRS, and the lag that once made non-disclosure on an FBAR or Form 8938 a low-detection-risk gamble has largely closed. If you are a US tax resident with Indian assets, treat full disclosure as the only safe posture, and be aware separately that US tax law treats most Indian mutual funds as PFICs with punishing reporting, a problem the DTAA deep dive addresses.

The reciprocity side, by contrast, is where promise and reality diverge. The Model 1 IGA contemplated the US sending India comparable information about Indian residents' US accounts. Years on, full reciprocity has not materialised to the degree India was led to expect, because delivering it would require US domestic legislation that has not passed. So the flow is, in practice, lopsided: India reports comprehensively to the US, and gets back less than the symmetric arrangement implied. This is not a 2026 development so much as a 2026 continuation, but it belongs in any honest map of the relationship, because it is the clearest example of an India-US tax commitment that exists on paper and underperforms in fact.

Where it all nets out, by your situation

The relationship looks different depending on which kind of NRI you are, so here is the practical net for each.

Your situation	Remittance tax	Totalisation gap	1989 treaty	The move that matters
H-1B / L-1 salaried in the US	None, if you wire from a US bank	Costs you 7.65% you will likely forfeit	25% on US dividends, credit in India	Save in vehicles you keep; never fund India transfers with cash
Returning to India after a US stint	None on your transfers	US Social Security largely lost below 40 quarters	FTC on dividends both ways via Form 67	Plan around forfeited contributions; close FBAR exposure cleanly
US-resident NRI with Indian assets	None on bank-funded transfers	Same loss as above	Indian income taxable, credit in US	Full FATCA disclosure; watch the PFIC trap on Indian MFs
Indian resident holding US stocks	Not applicable	Not applicable	25% US dividend withholding, Form 67 credit	File Form 67 on time; do not budget for the mythical 15%

The pattern is clear once you line them up. The remittance tax is a near-non-issue for everyone on this list. The totalisation gap is the real and uncompensated cost for anyone who works in the US and does not stay a decade. The treaty is unchanged and its dividend rule is routinely misquoted.

Edge cases and the things being said that are not quite true

"They are renegotiating the whole treaty." This is the central piece of wishful thinking, so be precise about it. There is no public evidence in 2026 of a formal renegotiation of the 1989 DTAA, no announced protocol, no signed amendment, no negotiating round on the official record. Commentary that frames a "new India-US tax treaty" as imminent is speculation dressed as news. Could it happen eventually? Of course; treaties get updated. But as of now it is rumour, and you should make no decision, on where to hold assets or when to realise income, on the assumption that the treaty is about to change. Plan around the treaty that exists.

The totalisation agreement and the treaty are different things. People conflate them. The 1989 DTAA is an income-tax treaty; a totalisation agreement is a separate social-security instrument negotiated through different channels. Progress or stagnation on one tells you nothing about the other. Both are stuck in 2026, but for different reasons and on different tracks.

The remittance-tax cash exemption could narrow. The 1% currently spares account-funded transfers, but the proposed regulations were still being finalised through early 2026, and the scope of "similar physical instrument" is the kind of definitional edge that can shift. The safe behaviour, funding everything from a US bank account, is robust to most plausible tightening, but it is worth watching if you ever use prepaid cards or unusual funding methods.

"US citizens of Indian origin are affected too." On the remittance tax, no. US citizens are exempt from the excise tax regardless of how they fund a transfer. The measure targets non-citizens, and an Indian-origin US citizen sending money to family in India pays nothing under Section 4475.

The closing read

The honest read on the India-US tax relationship in 2026 is that the loud development was the least important and the quiet one was the most. The remittance tax that filled the group chats with alarm is, for the salaried NRI who sends money digitally, a non-event; keep funding transfers from a US bank account and your bill is zero. The development that actually costs you real money, the absence of a totalisation agreement, generated almost no noise precisely because nothing happened, yet it quietly takes 7.65% of your US wages and, if you leave before ten years, gives nothing back. And the treaty that governs everything else did not move at all, which means the people promising you a shiny new 15% dividend rate or an imminent renegotiation are selling a story, not reporting a fact.

So commit to the two things within your control. First, stop ever funding an India transfer with cash from 2026 onward, and the remittance tax disappears from your life. Second, if you are working in the US on a temporary visa, build your retirement around vehicles you will actually keep, your own savings and India-side EPF and NPS, and treat the US Social Security you are forced to pay as a tax rather than a benefit, because for most returnees that is exactly what it is. Everything else, the renegotiation chatter especially, is noise until a signed document says otherwise. If your situation involves a large US retirement account, a planned return to India, or a six-figure FATCA disclosure gap, that is the point to pay a cross-border specialist, not to act on a news cycle, this guide included.

Related guides

• The India-US DTAA, deep dive
• Social security and totalisation agreements
• Budget 2026: LRS and TCS changes
• Foreign tax credit and Form 67
• Sending money to India
• All News and analysis
• All Taxation guides
• All Jobs and careers guides

This guide is news analysis and is educational and general in nature. It is not individual tax advice. The remittance excise tax regulations were still being finalised in early 2026, the totalisation position can change with US legislation, and your treaty and FATCA outcomes depend on your exact residency, holdings and filings, so confirm your specific position with a qualified cross-border tax adviser before acting.