By Ramachandran Rajeev Kumar — 2026-09-19

The strongest argument against everything below is the one most of Delhi already holds, and it deserves its best version before anyone argues with it.

It runs like this. The Lindsey O. Graham Sanctioning Russia and Iran Act is an American statute regulating what India buys from a third country. The moment India applies to Washington for relief from it, India has conceded that its energy sourcing is a matter on which Washington holds a vote. A tariff is a price, and prices can be paid. A precedent is not a price. It compounds. Whatever Delhi hands over to get out from under this law becomes the floor beneath the next demand. That demand will be about defence procurement or about Iran, and it will come from an administration that has already learned the answer is negotiable. On that reading India should ask for nothing, sign nothing, absorb the hit, and let the cost of tariffing a country of 1.4 billion people be discovered by the people who chose to impose it.

Most of that argument survives contact with the statute. The part that does not survive is its central assumption, which is that seeking relief is a single act with a single price. The law contains two separate exits. They cost India entirely different amounts. The distance between them is the whole of what Delhi should be negotiating about this month, and on the public record so far, Delhi is negotiating for the wrong one.

What the law actually does

The House cleared the bill 262 to 159 on 16 September, after the Senate passed it 86 to 11 on 7 August. It now sits with the President, who has indicated he will sign.

Section 112 is the part aimed at Moscow. It directs the President to raise duties on all goods from the Russian Federation to as much as 500 per cent within thirty days of enactment, covering oil, gas, refined products and coal, stacked on top of existing duties. No floor is specified.

Section 113 is the part aimed at India. It authorises tariffs of up to 100 per cent on countries that fall into either of two baskets: the top five importers of Russian crude or natural gas after enactment, and the top five countries facilitating evasion of Russian oil sanctions, which includes shadow-fleet operations. The President shall impose these within thirty days. The qualitative judgement sits in deciding who belongs in the basket, not in whether to act at all. The list is reassessed every 180 days, and Congress must receive a written justification ten days before any imposition or modification.

One exception is written into the gas criterion. A country importing less than 15 per cent of Russia's gas exports, and taking significant steps to reduce, is exempt. No equivalent exception exists for crude. A statute that rewards reduction in gas and refuses to reward it in oil is telling its readers something about what it was drafted to do.

India has no realistic route out of the basket at first assessment. The Global Trade Research Initiative puts India's FY2026 Russian crude purchases at $40.8 billion, close to a third of total crude imports. Whatever the current monthly share, that is the annual figure a Washington spreadsheet will use. India will be named.

Two exits, and they are not the same thing

Section 115 lets the President waive the tariffs on submitting a written certification that the waiver is in the national interests of the United States, with an explanatory report that may carry a classified annex.

Section 117 lets the President terminate secondary tariffs on certifying that the relevant government has ceased the offending activity and has provided reliable assurances of future compliance. Congress then has thirty days to pass a joint resolution of disapproval.

The two sentences describe different transactions. A section 115 waiver is an American document about American interests. It asserts nothing about Indian conduct, requires no Indian signature, contains no Indian commitment and creates no Indian obligation. If an American president certifies that keeping Indian refineries supplied serves American interests during a Gulf war, the sentence is true, and it is his sentence to write. India appears in it as a circumstance, not as a party.

A section 117 termination is a different animal. It requires India to have ceased, and to have given assurances about the future. That document is the one the autonomy argument is properly frightened of. Assurances of future compliance are a forward commitment on sovereign purchasing, made to a foreign legislature that can review them, and once given they become the baseline from which every subsequent conversation starts. It is the difference between a neighbour deciding not to complain and signing a covenant with him about what you may build.

The objection with which this piece opened treats both routes as surrender, when the surrender is entirely in the second one.

The refiners are asking for the wrong thing

India's public position has been steady. The Ministry of External Affairs said on 17 September that the measures could affect bilateral ties, and restated that energy security would be protected. Officials indicated purchases would continue through diversified sourcing and on the basis of evolving market dynamics, a formula that permits volumes to fall without any policy being reversed.

The refiners have gone further, and in the wrong direction. Reporting on their approach to the government describes them seeking a wind-down period for transactions already arranged, which is reasonable, and a quota for future Indian purchases of Russian crude, which is not.

A quota is a section 117 instrument wearing commercial clothes. It writes a number into an understanding with Washington, converts every barrel above that number into a diplomatic event, and hands the United States an agreed ceiling to lower at the next reassessment. The refiners are asking for it because a quota is bankable and a waiver is not, and from a treasury manager's desk that is the correct preference. From the national position it inverts the cost. Delhi should give the refiners their wind-down, refuse them their quota, and explain why in private.

The clock nobody is reading

Enactment starts a thirty-day fuse on imposition. If the bill is signed in late September, tariffs land in late October. That is the part everyone is watching.

The 180-day reassessment is the part that matters more and is being discussed less. India does not have to stop buying Russian crude. India has to not be among the top five buyers when the list is next examined, which is a volume-management problem with a date attached rather than a question of principle. Russian crude's share of Indian imports had already fallen under 20 per cent by January 2026 after the Rosneft and Lukoil designations of November 2025 and the reciprocal tariffs that followed. It rose again for reasons that had nothing to do with preference. The number moves. It has moved before, in both directions, inside a single year.

What the moment is actually worth buying

The American position has an awkwardness in it that Delhi has been too polite to press.

India did not increase its Russian intake in 2026 because Russian crude was cheap. India increased it because the Strait of Hormuz stopped working. Roughly 45 per cent of India's crude, half its LNG and 90 per cent of its LPG moved through that single passage. When Iran restricted it in the spring, the Indian crude basket went from $69 to above $114 a barrel within weeks. India rerouted about 70 per cent of its crude away from Hormuz, up from 55 per cent, and it did so in significant part by absorbing Russian barrels that travelled under temporary American licences. The United States Treasury issued those licences. India now buys from around forty countries, including Venezuelan heavy at 383,000 barrels a day in August, Nigerian Cawthorne landing at Sikka for the first time in April, and Angolan Pazflor through intermediaries.

A law that penalises India for the sourcing pattern American waivers produced is not a coherent policy. Delhi should say so once, calmly, on the record, and then stop saying it, because the argument is worth more as a premise for a section 115 certification than as a grievance.

What India should be extracting from this moment is unglamorous and durable. India holds about seven days of strategic petroleum reserve against the IEA's 90-day benchmark, and a 2020 memorandum with Washington on expanding it that has never been operationalised. Mundra commissioned its first very large crude carrier terminal in January. A one-year agreement already covers 2.2 million tonnes of American LPG, close to a tenth of Indian demand. Each of these reduces the exposure that made India coercible in the first place, and none of them requires India to promise anything about Russia. Each is also something an American administration can announce as a win, which is the quality that gets things signed.

That is the trade to be seeking in the ten days before Congress receives its justification letter, rather than a quota.

The part that is India's own doing

Seven days of cover is not bad luck. It is the accumulated result of two decades in which strategic reserve capacity lost every budget argument it entered, because the barrels always arrived and the cheapest insurance is the one never bought. The country that imports a record 88.7 per cent of its crude through two straits, one of which is now closed and the other of which is held at its southern end by a militia that has been taking Red Sea islands through September, chose to hold a week and a half of reserve.

The tariff, if it lands, will cost something on the order of the $11 billion GTRI estimates it would add to the import bill, on top of an effective 18 per cent duty already applied to Indian goods after February's interim framework cut it from 50. That is painful and survivable. What is not survivable in the long run is a structural position in which any serious actor who can reach a sea lane can set the price of India's next budget.

What the objection keeps

The autonomy argument with which this began is right about the thing that matters most, which is that precedent is the expensive currency and Delhi has a long history of spending it for short-dated relief. The ask will not stop here, and an administration that secures one concession prices the next one against it.

Where it goes wrong is in refusing the cheap exit alongside the expensive one, and there is a cost to that refusal too. A country that declines every instrument on principle ends up with no instruments, and pays list price for everything, and calls the bill sovereignty.

The distinction is available in the text of the law itself. Take the waiver, which asks nothing. Refuse the assurances and the quota, which ask for the future. Spend the window on reserve capacity and non-Hormuz infrastructure, which are useful whichever way the certification goes. And accept the tariff if it comes, for the number of months it takes to fall out of the top five, which on the record of the last twelve months is not many.

Xi Jinping arrives in Washington on 24 September. India will not be the most important file on the desk that week. Countries that are not the most important file are usually better off having decided what they want before the meeting than after it.