DGTR recommends, the Finance Ministry decides

A domestic producer watches Chinese API prices collapse by 40 to 50 percent in a single quarter and begins to sell below cost. The trade remedies architecture is real and operable, and the relief's earliest credible date is twelve to eighteen months after injury became severe. The Chinese exporter can reset the price twice in that window. DGTR investigates and recommends; the Ministry of Finance decides whether the duty is imposed and at what level. The petitioner who runs a clean investigation may still find the recommended 30 percent duty notified at 12 percent. What does this reveal, and where must engagement actually happen?

BIS controls what enters the market. NPPA controls what it costs. DGTR controls who competes. The trade remedies architecture is petition-driven and injury-looking-back. No DGTR investigation begins without a domestic industry petition; no petition clears the preliminary filter without standing; no standing is credited without audited evidence of dumping, injury, and causal link across a defined investigation period. The architecture is real and operable, and the domestic producer who waits until margins have collapsed to start assembling the file discovers that the remedy's earliest credible date is twelve to eighteen months after injury became severe. The Chinese exporter can reset the price twice in that window. The petition's quality is the variable that most determines how fast the relief arrives; the quality is set by work that happens before the petition is filed, not during the investigation.

The legal frame sits across three instruments. The Customs Tariff Act 1975, particularly Section 9A, provides the substantive power to levy anti-dumping duty where exports are sold into India at less than normal value and cause or threaten material injury to domestic industry. Section 9 carries the parallel countervailing duty power for subsidised imports; Section 8B carries the safeguard power for import surges irrespective of dumping. The Anti-Dumping Rules 1995, notified under the Customs Tariff Act, provide the procedural architecture through which DGTR conducts investigations. DGTR itself sits under the Department of Commerce, separated from the Department of Revenue which eventually notifies the duty, and the separation is institutionally deliberate because the fact-finding function and the policy-decision function are handled by different ministries. The investigation is quasi-judicial in character; both the domestic industry (the petitioner) and the foreign exporters (the respondents) submit evidence, participate in hearings, and receive disclosure of essential facts before the final determination.

Standing is the petition's first live question, and it is where many petitions are parked before they are ever investigated. Rule 2(b) defines domestic industry as producers whose collective output constitutes a major proportion of the total domestic production of the like article. The petition must be supported by producers accounting for at least 25 percent of total domestic production, and of the producers who express an opinion, those supporting the petition must account for more than 50 percent of the expressed production. Producers related to exporters or importers of the allegedly dumped product, or who themselves import the product, are excluded from the domestic industry definition. A hybrid Indian manufacturer that both manufactures the API and imports the same API from China to balance its capacity cannot petition against the very import it relies on; the related-party exclusion removes it from the standing calculation entirely. This single Rule 2(b) exclusion disqualifies a material share of the Indian pharmaceutical and chemical sector from trade remedies self-help in their own categories, and the producers who remain eligible are often sub-scale for the standing threshold.

The evidentiary burden on the petitioner runs across three proofs. Dumping is established by comparing normal value in the exporting country (typically ex-factory price in the exporter's domestic market, or constructed value where the domestic market is distorted or in a non-market economy) with the export price to India, with appropriate adjustments for freight, insurance, commissions, and conditions of sale. For China, the non-market economy treatment persists for most sectors, which allows the petitioner to propose a surrogate country's cost structure as the normal-value anchor. Selection of the surrogate is a contested sub-proceeding of its own; the exporter typically proposes a lower-cost surrogate, the petitioner proposes a higher-cost one, and DGTR adjudicates. Injury is established by examining volume of dumped imports (absolute and relative to domestic production), price effects (undercutting, depression, suppression), and impact on domestic industry across fifteen-plus parameters including capacity utilisation, market share, sales, profits, cash flow, employment, inventories, and return on investment. Causal link connects the two proofs: the injury must flow from the dumping and not from other factors such as demand contraction, technological obsolescence, or self-inflicted margin compression.

The period of investigation is typically the twelve months preceding the petition, with a three-year injury period for trend analysis. Evidence must be producible in audited form, segregable to the specific HS code or product basket under investigation, and defensible under DGTR's verification visit. The verification is not symbolic. DGTR officers visit the petitioner's plant, examine production records, cross-check cost data against the Statutory Auditor's certification, and review inventory and sales registers. Evidence that reads well in the petition and cannot survive verification is worse than weak evidence, because it undermines the petitioner's credibility on the dimensions that do hold. A petition that would have cleared at 60 percent of its original claim quantum is parked entirely when verification uncovers a single material misstatement.

The investigation sequence runs on a statutory calendar. Initiation notification is published in the Gazette within sixty days of the Director General's satisfaction on prima facie grounds. Questionnaires are issued to known exporters, importers, the domestic industry, and user industries within days of initiation. Responses are typically due within thirty to forty-five days. A public hearing is convened, usually six to eight months into the investigation, where all interested parties present their positions on record. Verification visits to exporters are conducted where the exporter has cooperated; where the exporter declines or is uncooperative, best-available-information is applied and the exporter's margin is typically adverse. A disclosure statement is issued to all interested parties before final finding, setting out DGTR's essential factual findings and allowing written comments. The final finding, published as a Gazette notification, contains DGTR's determination on dumping, injury, and causal link, along with the recommended duty quantum per exporter. The full cycle from initiation to final finding is statutorily capped at twelve months, extendable by six months with MoF concurrence, under Section 9A(6) of the Customs Tariff Act.

Provisional duty is the mechanism through which the architecture can compress the relief timeline. DGTR may issue a preliminary finding sixty days after initiation, and if the preliminary finding is affirmative, provisional duty can be imposed. In practice, provisional duty arrives five to nine months after initiation for well-prepared petitions, and twelve months after initiation or not at all for petitions where the evidence required supplementation. The petitioner who can yield audited dumping and injury evidence on the day of filing materially shortens the time to provisional duty; the petitioner who has to build the evidentiary base after initiation runs out the clock before preliminary finding becomes defensible.

The institutional bifurcation between DGTR and the Ministry of Finance is where the process reveals its most consequential dynamic. When DGTR concludes an investigation with an affirmative finding and recommends a definitive anti-dumping duty, the recommendation is transmitted to the Ministry of Finance, specifically to the Department of Revenue. Finance has the authority to accept, modify, or decline the recommendation. This is not a rubber stamp. Finance evaluates the recommendation against a wider institutional calculus that DGTR's mandate does not encompass: the impact on downstream industries that use the product as an input, the trade policy implications of imposing duties on a specific country (particularly where bilateral trade negotiations are sensitive), the effect on consumer prices, and the diplomatic consequences. A DGTR recommendation that is procedurally sound and evidentiarily supported may not result in a duty if Finance determines that the broader policy cost outweighs the protection benefit.

At the Department of Revenue, the Tax Research Unit prepares the duty notification in coordination with the Department of Commerce. The final duty may be accepted in full, reduced, imposed with a form change (specific, ad valorem, reference-price-based), delayed, or declined in the rare cases where broader policy considerations override the investigation record. This policy-override stage is where downstream-consumer-interest arguments, foreign policy considerations, and revenue-implication arguments are heard, often through representations to the Secretary of Revenue and the Secretary of Commerce made by exporter-country diplomatic channels, user-industry associations, and line ministries of downstream sectors. The petitioner who has run a clean investigation but not engaged at the notification stage can discover that the recommended 30 percent duty has emerged as a 12 percent duty, or as a reference-price-based duty set at a level that the exporter can price-cross without triggering.

This institutional separation means that the outcome of an anti-dumping investigation is not determined by the investigation alone. The domestic industry that files the petition and prevails before DGTR may still not receive the protection it sought if Finance exercises its discretion differently. The foreign exporter that expects the investigation to be the final word may discover that the duty was never imposed, or was imposed at a lower rate than DGTR recommended. The decision authority is distributed across two institutions with different mandates, and the point at which the outcome is determined shifts from DGTR's investigation room to Finance's evaluation of the recommendation.

Sunset reviews explain why some duties persist for decades. Anti-dumping duties in India carry an initial five-year tenure but are routinely extended through sunset reviews. The domestic industry files for continuation; DGTR investigates again; the duty is renewed. Some duties have been in place for over fifteen years across multiple review cycles, creating a permanent cost structure for importers that was originally designed as a temporary remedy.

Anti-circumvention is the architecture's acknowledged vulnerability. Once an anti-dumping duty is in force, exporters commonly respond through three routes: transshipment via a third country, marginal product modification that moves the HS code or the DGTR product scope, or relocation of final processing to a country not covered by the duty. Rule 25 of the Anti-Dumping Rules provides for anti-circumvention investigations, but Rule 25 is not self-executing. It requires a fresh petition from the domestic industry, with fresh evidence of the circumvention pattern, and it runs on its own investigation timeline. The domestic producer who secured relief against Chinese API imports may find, eighteen months later, that the same economic flow has reappeared as imports from Vietnam or Thailand or as a slightly modified HS classification, and the remedy architecture requires a second full cycle to address the second wave.

The architecture's reactive design is not a defect; it is the WTO-compliant frame within which it must operate. The petitioner who reads that frame correctly builds the evidentiary base before injury becomes existential, runs industry-association coordination for standing before the related-party exclusions fracture it, retains specialist trade-remedies counsel before the questionnaire phase, and engages with the Department of Revenue at the notification stage rather than after. The petitioner who waits until margins have collapsed to start assembling the file discovers that the architecture works precisely as designed, and that the design includes the interim period during which the injury continues.