What DPIIT actually built

In December 2025 — the copy circulating carries the annotation "Revised & Updated version as on 16.12.2025" — the Department for Promotion of Industry and Internal Trade (DPIIT) published a "Guidebook on Mapping of Harmonized System of Nomenclature (HSN) Codes," an exercise to assign every one of India's tariff codes to the ministry or department responsible for the policy governing it. India classifies goods at the 8-digit "tariff line" level, one level more granular than the 6-digit standard the World Customs Organisation sets internationally, and the guidebook works at that 8-digit level: 12,174 codes in total, sourced from the CBIC Tariff Manual.

The method DPIIT used to assign each code an owner is not new judgement calls dressed up as a fresh exercise — it is explicitly grounded in two documents every ministry is already bound by: the Allocation of Business Rules, 1961, which is the actual statutory instrument dividing the government's subject-matter responsibilities among its ministries and departments, and the CBIC Tariff Manual, published by the Central Board of Indirect Taxes and Customs, which gives the detailed goods descriptions customs officers use for classification. DPIIT layered a "value-chain analysis" of each product's use-case on top, then ran the results through formal Inter-Ministerial Consultations — meaning every ministry got a chance to accept or decline the codes proposed to it, not just receive a list handed down.

Out of that process, 11,679 codes were formally accepted by one of 31 ministries and departments. The remaining 495 were not accepted by anyone, and DPIIT grouped them into what the guidebook itself calls the "Residual List."

The residual list isn't a classification gap — it's an acceptance failure

The natural assumption about 495 unassigned tariff codes is that nobody could figure out which ministry they belonged to — genuinely ambiguous, multi-use, or newly created products falling through a real classification gap. That is not what DPIIT's own document shows. Every single one of the 495 residual rows carries a column headed "Probable Ministry/Department": DPIIT (or the consultation process) proposed a likely home for each and every code. What's missing is not a proposal — it's formal acceptance.

Some of the specific products stuck in that gap: mate (the beverage), bark of African cherry, jojoba seed, gum arabic and asafoetida — each with a "probable" ministry (Commerce, Ayush, or Agriculture) already named next to it, unaccepted regardless. A cluster of Chapter 49 codes — printed books, dictionaries and encyclopaedias, children's picture books, globes — are all proposed to the Department of School Education and Literacy, a ministry that does not appear anywhere in the formal 31-ministry table of codes actually accepted; it exists in this document only as an unaccepted suggestion. And at least one code makes the ambiguity explicit rather than implicit: HS 49119910, "hardcopy (printed) of computer software," is tagged with two probable ministries at once — Electronics and Information Technology, and School Education and Literacy — a single product DPIIT itself could not settle between two plausible regulators.

The 495 codes are not evenly spread across India's ministries either. Roughly 62 per cent of them have a probable-ministry suggestion pointing to just two: the Ministry of Heavy Industries (around 201 of the 495) and the Department of Chemicals and Petro-Chemicals (around 108). That is not a coincidence of alphabetical clustering — those two ministries already carry two of the largest caseloads in the formal, accepted table (see below), which suggests the residual pile isn't a set of genuine edge cases so much as overflow from the two ministries whose product remit is already the broadest and most heterogeneous in government.

The other problem, hiding in DPIIT's own footnote

The 495-code gap is a vacuum. There is a second, opposite problem sitting inside the "successfully mapped" 11,679 figure itself: DPIIT's table of codes accepted by all 31 ministries carries a footnote reading "Certain HS Codes have been accepted by multiple Ministries/Departments." Summing every ministry's individually stated tally in that table comes to 12,912 — not 11,679. The gap, 1,233 codes, is the scale of double- and multiple-claiming going on even among the codes DPIIT counts as resolved. Put plainly: over one in ten of the "accepted" codes are being formally claimed by more than one ministry at the same time.

India's tariff schedule, in other words, has both failure modes at once in the same 12,174-code universe: a 495-code pile nobody will claim, and a roughly 1,233-code overlap where more than one ministry already has. Neither is the kind of gap a company or an import-substitution planner can just work around by picking the "obvious" ministry — the obvious ministry may be one of two competing claimants, or may be the one DPIIT proposed and nobody accepted.

Who owns the most, and the least

DPIIT holding the largest single caseload of any ministry is itself a structural quirk worth naming: the department that ran the entire mapping exercise is also its single biggest claimant, ahead of Textiles and Chemicals. At the other end, the Ministry of New and Renewable Energy formally owns just two HS codes nationally — despite still being named as the probable owner of at least one residual code (solar-energy-based equipment and gadgets) it apparently never got around to formally accepting.

Where this connects to the rest of the policy machinery

This is not the first time this blog has found India's regulatory map more fragmented than a single-ministry story would suggest — the same pattern of split, overlapping and sometimes absent ownership runs through this blog's earlier count of India's 37 separate quality regulators, and through the PLI-scheme landscape, where nine different ministries each run their own production-linked incentive scheme with their own company lists and reporting cadences. A tariff code with no accepted ministry, or with more than one, sits directly upstream of both problems: it is hard to design a quality standard, or a PLI eligibility list, around a product category the government's own paperwork hasn't settled who is responsible for.

What would fix this

• DPIIT should publish why each of the 495 residual codes was not accepted — a one-line reason per code (declined, no response, disputed) turns a bare list into something a ministry, a company, or Parliament can actually act on.
• The Cabinet Secretariat should set a public deadline for resolving the 495-code residual list, with a default-assignment rule (e.g., the "probable" ministry becomes the owner of record if no formal objection is raised within a fixed window) so the list doesn't simply persist indefinitely across future revisions of this same guidebook.
• The 1,233-code overlap found in DPIIT's own ministry-tally arithmetic deserves the same public accounting as the residual list. A footnote acknowledging multiple claims is not the same as naming which codes are contested and resolving who leads on each.
• A ministry like School Education and Literacy, named as probable owner for real product categories (books, globes, dictionaries) but absent from the formal 31-ministry table entirely, should be asked to formally accept or explicitly decline — leaving it unlisted is a third, quieter failure mode alongside outright rejection and unresolved overlap.

These are proposals from this blog, not recommendations any agency has adopted.

Sources and caveats

About this article: Researched, written and edited by Umashankar Triplicane Dwarakanathan, with AI research assistance; every figure is meant to trace to the primary source cited. See the Editorial Policy for how sourcing, AI use and corrections work.