The Tariff That Was Never Lawful — What India Conceded, and What It Should Learn

Pritiraj BrahmaFounding Partner, Pritiraj & Partners · 14 May 2026

India negotiated relief from a 50 per cent US tariff in February 2026. Thirteen days later the US Supreme Court held the instrument behind it was never authorised. On the narrow question of the additional duty, the settlement bought seventeen days — and the refunds accrue to American importers, not Indian exporters. A lesson in negotiating against legal durability.

ProvenancePublished 14 May 2026; revised 28 August 2026 to add the counter-case section the essay lacked, two findings and two figures. Tariff dates and rates are from the executive orders of 27 August 2025 and 20 February 2026, the Customs and Border Protection collection notice, and statements of the Minister of Commerce and Industry of 7 February 2026. The judgment is Learning Resources, Inc. v. Trump, decided 6–3 on 20 February 2026; refund estimates of $166 billion to $200 billion are as reported and the question was remanded and remains unresolved. Trade figures are monthly merchandise data for January and February 2026. The countervailing duty determination on crystalline silicon photovoltaic cells is preliminary, dated 25 February 2026.

India traded durable concessions for relief from a tariff that a court then held had never been lawful. On the narrow question of the additional duty, the February settlement bought seventeen days. The instrument was temporary; the reorientation of crude sourcing and procurement commitments is not. That asymmetry is the lesson worth extracting — and the strongest objection to drawing it is that no negotiator can be asked to bet on how a foreign court will rule.

Between 27 August 2025 and 7 February 2026, most Indian exports to the United States carried an additional 25 per cent duty, imposed by executive order under the International Emergency Economic Powers Act in response to India’s purchases of Russian crude. Stacked on the existing reciprocal tariff, the effective rate on much of India’s trade reached approximately 50 per cent — a level that touched well over half of roughly $87 billion of annual goods exports to the United States, while pharmaceuticals, semiconductors, energy and critical minerals were exempted.

On 2 February 2026 an interim trade understanding was announced, and on 7 February the Commerce and Industry Minister confirmed the effective tariff had come down to 18 per cent, with the punitive duty rescinded and zero-duty access secured for a list of agricultural exports. India’s side, as reported, involved reducing Russian crude purchases, expanding purchases of American goods, and movement on tariff and non-tariff barriers. On 20 February the United States Supreme Court held, 6–3, in Learning Resources, Inc. v. Trump, that the Act does not authorise the President to impose tariffs at all.

Six findings anchor this analysis:

1. The pressure instrument was legally fragile from inception, and that was knowable. The Supreme Court affirmed a Federal Circuit ruling from August 2025 — a ruling handed down within weeks of the India tariff taking effect. The legal risk to the instrument was public, litigated, and on the record throughout the period in which India negotiated against it.
2. The concessions and the relief have different half-lives. A tariff can be terminated by executive order overnight. Reorientation of crude sourcing, procurement commitments and market-access changes are structural, multi-year, and politically costly to reverse. Any negotiation that trades the second for the first should be priced accordingly.
3. The pressure did not end; it changed statute. On 11 and 12 March 2026 the United States Trade Representative initiated new Section 301 investigations — one concerning structural excess capacity in manufacturing, covering sixteen economies including India, and another concerning enforcement against goods produced with forced labour, covering some sixty economies. Section 301 is the durable instrument: statutorily grounded, procedurally slow, and far harder to strike down.
4. A trade remedy overrode the settlement within days, for at least one sector. On 25 February 2026 the Department of Commerce announced preliminary countervailing duties of roughly 126 per cent on crystalline silicon photovoltaic cells and modules from India. The headline tariff had just fallen to 18 per cent and then to nothing; for that sector neither number mattered. Trade remedies operate independently of tariff diplomacy and are decided on documentation rather than on relationships.
5. The restitution accrues to United States importers, not to Indian exporters. The refund question — with estimates in the range of $166 billion to $200 billion — was remanded to the lower courts, where it remains unresolved. The duty was collected from importers of record in the United States, so whatever is repaid is repaid to them. Indian exporters absorbed the competitive damage of six months at an elevated rate and have no claim in that proceeding.
6. The macro picture did not improve with the headline. India’s merchandise trade deficit was $34.68 billion in January 2026, with exports up just 0.6 per cent year on year; in February 2026 the deficit was $27.10 billion, with imports up 24 per cent and exports down 0.8 per cent. Tariff relief arrived into flat export volumes.

The seventeen days

A negotiated framework has value beyond the rate: predictability, agricultural carve-outs, a standing channel, and a relationship banked for the next dispute. Those are real gains and a government is entitled to weigh them. But they should be entered on the ledger as what they are — relationship assets — rather than as the price of tariff relief that arrived on its own three weeks later.

The generalisable lesson is procedural rather than political: when a trading partner applies pressure through a contested legal instrument, the first analytical question is not what to concede, but how long the instrument can survive. That question is answerable. It requires a standing capability to read the counterparty’s domestic litigation as a leading indicator of its trade policy — a capability India’s trade establishment does not visibly maintain, and which is inexpensive relative to what it protects.

What actually determines export performance

It is tempting to read the fall from 50 to 18 per cent as the restoration of competitiveness. The trade data through February suggests otherwise, and the reason is instructive. Tariffs are one term in a landed-cost equation that also contains freight, insurance, financing cost, lead time and compliance overhead. Through this same period, exporters were adjusting shipment schedules and routing in response to higher logistics costs and disruptions in West Asia, with rising insurance premia. A seven-point tariff improvement is easily consumed by a freight and insurance shock, and neither is within a trade negotiator’s control.

Nor does a headline rate describe access. Sanitary and phytosanitary standards, technical barriers, and product-specific trade remedies operate independently of it — which is what the photovoltaic determination demonstrates.

The counter-case, honestly stated

Three arguments cut against this analysis, and the third is the one that should give any critic pause.

First, a government cannot negotiate on the assumption that a foreign court will rule its way. Sovereign risk management requires acting against the instrument in force, not the instrument as it might later be struck down. Had India waited and the Court gone the other way — and a 6–3 division indicates a genuinely contested question, not a foregone one — exporters would have carried an effective 50 per cent rate for an indefinite further period. Negotiating against a live tariff is not a failure of analysis; it is the job.

Second, the seventeen-day framing understates what was secured, and in one respect inverts our own argument. The zero-duty agricultural access for spices, tea, coffee and several fruits was not contingent on the emergency tariff and did not terminate with it. That is a durable gain obtained during a temporary crisis — the reverse of the asymmetry this essay alleges. A fair ledger records it on the same side as the concessions.

Third, and most seriously: the central claim rests on a counterfactual this essay cannot test. We do not know what the reciprocal tariff would have been in the absence of the February understanding, whether the punitive duty would have been re-imposed under a different statute in the interim, or how the relationship would have priced a refusal to engage. Criticism of negotiators conducted with the judgment in hand is cheap, and an essay whose own recommendation is a legal-durability assessment should concede that such an assessment yields probabilities rather than answers. The recommendation survives because it is cheap and improves decisions at the margin — not because it would have told anyone in January 2026 what the Court would do in February.

What we would do

1. Build a legal-durability assessment into every trade negotiation. Before conceding, establish which statute the pressure rests on, what litigation is pending against it, and what the realistic timeline to invalidation is. It yields a probability rather than an answer, which is exactly what a negotiator should be given. This is a small research function with disproportionate leverage.
2. Reorient defensive preparation toward Section 301 and trade remedies. The excess-capacity and forced-labour investigations initiated in March 2026 are where the next round of exposure sits. They are evidence-driven proceedings, which means they are winnable with documentation — and lost by default when firms and ministries fail to file.
3. Prepare sector-level dossiers now, not on receipt of a notice. The photovoltaic determination shows how quickly a single sector can lose its market irrespective of the headline tariff. Any sector with subsidy exposure should have its cost, subsidy and supply-chain documentation assembled in advance, because a countervailing duty proceeding is decided on what can be produced in weeks.
4. Separate the freight problem from the tariff problem in export policy. Through this cycle, logistics cost and route disruption plausibly did more to Indian export competitiveness than the tariff differential. It receives a fraction of the policy attention because it has no negotiating counterpart and therefore no ministerial announcement.
5. State the concessions publicly and track their cost. Commitments on energy sourcing and procurement carry a measurable fiscal and current-account cost. If they were worth making, they are worth measuring — and the measurement is what makes the next negotiation better than this one, which is the only defensible use of hindsight.

India negotiated capably under real pressure and secured a genuine improvement, including access that outlasted the instrument. It also paid in durable currency for relief from a tariff with a short and visible legal life, and its exporters hold no claim in the refund proceeding that follows. The correction is not to negotiate less. It is to know, before sitting down, how long the other side’s leverage can lawfully last — and to say out loud that the answer will be a probability.