India is reported to have drawn up a list of 1,272 products worth $189 billion of annual imports to replace with domestic manufacturing. The list itself is not public. What is public is the trade data underneath it — and it says the target is 24 per cent of the import bill, 41 per cent of the imports manufacturing could plausibly touch, and 74 per cent of the four sectors the plan actually names.
You Cannot Substitute Crude Oil or Gold. That Is 40% of India’s Import Bill.
The short version.
- India imported $776 billion in FY2025-26 (DGCI&S, all 98 populated chapters).
- The reported $189 bn substitution target is 24.4 per cent of that.
- But mineral fuels ($203 bn) and pearls, precious stones and gold ($109 bn) are 40 per cent of imports and cannot be manufactured away. Strip them and the target is 40.8 per cent of what is left.
- Against the four sectors the plan actually names — chemicals, electronics, machinery, speciality steel — which import $256 bn between them, $189 bn is 73.7 per cent.
- Electronics, the largest of those, has doubled since FY2018-19. The biggest target is the one moving fastest in the wrong direction.
What is being claimed, and what is not published
A centre-state exercise has reportedly identified 1,272 products, worth close to $189 billion of annual imports, as candidates for domestic manufacturing. The stated filter is products importing more than $50 million a year that are either not made in India at all or made in inadequate quantity. The sectors named are chemicals, electronics, machinery and speciality steel.
One thing should be said before any analysis. The list is not published. It has been reported in the business press; no Press Information Bureau release setting out the 1,272 products or the $189 billion could be located for this piece, and searching PIB for the associated department returned only defence indigenisation lists and an unrelated year-end review. So the 1,272 and the $189 billion are treated here as reported figures, not as established ones.
The list cannot be reproduced without knowing which products were picked. But the filter can be, because DGCI&S publishes import data at the 8-digit commodity level. That is done below, and it turns out to reconcile on one of the two numbers and not the other.
What that chapter data can do is size the ambition. If someone tells you they intend to remove $189 billion of imports, the useful question is not which 1,272 products. It is: out of what?
Out of what: the actual import bill
India imported $776,014 million in FY2025-26 — about $776 billion — against $721 billion the year before, a rise of 7.6 per cent. Exports were $441.7 billion, leaving a goods deficit of $334.3 billion. Here is where the imports sit.
| Chapter | Imports FY2025-26 $ million | Share per cent | Cumulative per cent |
|---|---|---|---|
| 1. HS 27 — Mineral Fuels, Mineral Oils And Products Of | 203,415 | 26.2 | 26.2 |
| 2. HS 71 — Natural Or Cultured Pearls,Precious Or Semip | 109,430 | 14.1 | 40.3 |
| 3. HS 85 — Electrical Machinery And Equipment And Parts | 104,853 | 13.5 | 53.8 |
| 4. HS 84 — Nuclear Reactors, Boilers, Machinery And Mec | 74,019 | 9.5 | 63.4 |
| 5. HS 29 — Organic Chemicals | 25,411 | 3.3 | 66.6 |
| 6. HS 39 — Plastic And Articles Thereof | 22,234 | 2.9 | 69.5 |
| 7. HS 15 — Animal Or Vegetable Fats And Oils And Their | 19,777 | 2.5 | 72.1 |
| 8. HS 72 — Iron And Steel | 15,753 | 2.0 | 74.1 |
| 9. HS 90 — Optical, Photographic Cinematographic Measur | 15,375 | 2.0 | 76.1 |
| 10. HS 31 — Fertilisers | 14,580 | 1.9 | 77.9 |
| 11. HS 28 — Inorganic Chemicals; Organic Or Inorganic Co | 14,180 | 1.8 | 79.8 |
| 12. HS 88 — Aircraft, Spacecraft, And Parts Thereof | 13,775 | 1.8 | 81.5 |
| 13. HS 74 — Copper And Articles Thereof | 12,780 | 1.6 | 83.2 |
| 14. HS 26 — Ores, Slag And Ash | 10,403 | 1.3 | 84.5 |
| 15. HS 76 — Aluminium And Articles Thereof | 10,002 | 1.3 | 85.8 |
| All 98 chapters | 776,014 | 100.0 | 100.0 |
DGCI&S commodity-wise import data via the TradeStat EIDB portal, 2-digit HS chapter level, FY2025-26 provisional, source last updated 19 May 2026. Shares computed here. Fifteen chapters carry 85.8 per cent of the bill.
The two chapters that end the argument
The largest import chapter is HS 27, mineral fuels and oils, at $203,415 million. On its own it is larger than the entire $189 billion target. The second is HS 71, pearls, precious and semi-precious stones and gold, at $109,430 million.
Neither can be substituted by building factories. India imports crude because it does not have crude; refining capacity is already large and is not the constraint. Gold is imported because households and investors buy gold, which is a savings behaviour, not a manufacturing gap. You can tax gold imports, and India does. You cannot manufacture your way out of them.
Remove them and the addressable import bill is $463,168 million. Against that base, $189 billion is not 24 per cent. It is 40.8 per cent.
Reconstructing the filter from public data
The reported filter is specific enough to test: products importing more than $50 million a year. DGCI&S publishes exactly that, at 8-digit HS level. Applying the threshold to FY2025-26 import data gives this.
| Threshold | HSN-8 lines | Their imports $ million | Share of all imports per cent |
|---|---|---|---|
| ≥ $10 million | 3,069 | 763,676 | 98.4 |
| ≥ $25 million | 1,990 | 746,307 | 96.2 |
| ≥ $50 million | 1,314 | 722,253 | 93.1 |
| ≥ $100 million | 786 | 685,254 | 88.3 |
| ≥ $250 million | 354 | 618,249 | 79.7 |
| ≥ $500 million | 180 | 558,470 | 72.0 |
Computed here from DGCI&S 8-digit commodity-wise import data, FY2025-26. The 8-digit sum for the year is $775,712 million, within 0.04 per cent of the chapter-level total used elsewhere in this piece, which is the internal consistency check.
That is a genuinely close match, and it suggests the reported filter is real and applied to something very like this data. But the value does not follow. Those 1,314 lines import $722 billion between them, not $189 billion — a factor of 3.8. Excluding fuels and gems and gold still leaves 1,249 lines importing $411 billion, a factor of 2.2.
So the widely-repeated framing — 1,272 products accounting for $189 billion of annual imports — cannot be right as stated. If the 1,272 products are the lines above $50 million, their imports are nearly four times $189 billion.
What $189 billion probably is
There is a reading that fits. The claim implies an average of $148.6 million per product. Test that against bands of the actual data:
| Band | Lines | Total $ billion | Average per line $ million |
|---|---|---|---|
| $50 million to $500 million | 1,134 | 163.8 | 144 |
| $50 million to $1 billion | 1,232 | 232.1 | 188 |
| All lines above $50 million | 1,314 | 722.3 | 550 |
| Implied by the reported claim | 1,272 | 189 | 148.6 |
The claim’s implied average of $148.6 million per product sits almost exactly on the $50 million to $500 million band, whose actual average is $144 million across 1,134 lines. Widen the band to $1 billion and you get 1,232 lines — within 3 per cent of 1,272 — averaging $188 million.
In other words the list is almost certainly a mid-tail programme: products importing more than $50 million but not the giants. That makes sense on its own terms. Crude oil, gold and smartphones are not candidates for a “not made in India or made in inadequate quantity” list — the first two are resource and savings imports, and the third is already assembled here at scale.
This reading is inference, not confirmation, and it should be labelled as such. But it is a considerably more sensible plan than the headline implies, and it changes what the target means: not 24 per cent of the import bill, but close to all of a specific band that sits below the largest lines.
Against the sectors actually named, it is three-quarters
The plan names four sectors. Mapped to HS chapters as closely as 2-digit data allows, they are:
| Chapter | Imports FY2025-26 $ million | Share of all imports per cent |
|---|---|---|
| HS 85 — Electrical machinery, electronics and parts | 104,853 | 13.5 |
| HS 84 — Nuclear reactors, boilers, machinery and mechanical appliances | 74,019 | 9.5 |
| HS 29 — Organic chemicals | 25,411 | 3.3 |
| HS 39 — Plastics and articles thereof | 22,234 | 2.9 |
| HS 72 — Iron and steel | 15,753 | 2.0 |
| HS 28 — Inorganic chemicals | 14,180 | 1.8 |
| Named-sector total | 256,450 | 33.0 |
| $189 bn target as a share of these | 73.7 per cent | |
Chapter mapping is this article’s, not the plan’s. “Speciality steel” is a subset of HS 72, so including all of HS 72 is generous to the target; “electronics” and “machinery” span HS 84 and 85 with some overlap into HS 90. A different reasonable mapping would move the denominator by perhaps $20–30 billion, which changes the percentage but not the conclusion.
To take $189 billion out of a $256 billion group of sectors is to replace roughly three-quarters of what India currently buys from abroad in chemicals, electronics, machinery and steel. Not to grow domestic output by three-quarters of that figure — to replace the imports themselves, while domestic demand keeps rising.
And the biggest target is growing fastest
Electronics is the largest single component of the addressable bill, and its trajectory runs against the plan.
| Chapter | FY18-19 $ mn | FY19-20 $ mn | FY20-21 $ mn | FY21-22 $ mn | FY22-23 $ mn | FY23-24 $ mn | FY24-25 $ mn | FY25-26 $ mn | Change per cent |
|---|---|---|---|---|---|---|---|---|---|
| HS 85 — Electrical machinery, electronics | 52,049 | 49,188 | 46,692 | 62,490 | 67,637 | 79,301 | 88,596 | 104,853 | +101 |
| HS 84 — Machinery, mech. appliances | 43,840 | 43,368 | 37,032 | 50,587 | 54,377 | 57,409 | 64,302 | 74,019 | +69 |
| HS 71 — Pearls, precious stones, gold | 64,720 | 54,494 | 55,199 | 81,664 | 73,930 | 78,474 | 88,976 | 109,430 | +69 |
| HS 27 — Mineral fuels and oils | 167,872 | 153,646 | 99,703 | 194,857 | 260,921 | 219,054 | 218,456 | 203,415 | +21 |
Same DGCI&S series. Change is FY2018-19 to FY2025-26. FY2020-21 is the pandemic trough and distorts any two-point comparison that uses it as a base.
Electronics imports went from $52,049 million in FY2018-19 to $104,853 million in FY2025-26 — slightly more than double. Machinery rose from $43,840 million to $74,019 million. This happened through the period of the production-linked incentive schemes, the phased manufacturing programme for mobile phones, and the semiconductor mission.
That is not necessarily a policy failure, and it is important to be precise about why. Assembly-led electronics manufacturing pulls in components: a phone assembled in India imports its display, its chipset and its camera module. Output rises and so do imports, at least for a while. The published critique of counting substitution by output value is exactly this — mobile-phone assembly is reported at around 18 to 20 per cent domestic value addition, meaning four-fifths of the value of an Indian-made phone is still bought abroad.
What the target would actually require
Three things follow from the arithmetic, none of which requires knowing the 1,272 products.
It is a decade-scale target, not a plan-period one. Removing 41 per cent of addressable imports while domestic demand grows means building capacity faster than consumption rises, in the specific sectors where India currently buys most. No timeline has been reported alongside the $189 billion, which is the first thing a reader should ask for.
The value-addition denominator matters more than the product count. A list of 1,272 products tells you where imports are. It does not tell you what share of each product’s value could realistically be domestic. On the phone example, substituting the assembly step moves 18 to 20 per cent of the value and leaves the rest imported — the import line barely moves.
The two biggest chapters will keep the headline import number high regardless. Even if every rupee of the $189 billion target were achieved, India would still import mineral fuels and gold worth $313 billion at current levels. Any political claim that substitution will close the trade deficit runs into that floor. The deficit was $334.3 billion in FY2025-26; fuels and gold alone are $313 billion of the import side.
What would make this checkable
- Publish the list. 1,272 products with their HS codes and import values would let anyone reconcile the $189 billion against DGCI&S in an afternoon. Until then the figure can only be quoted, not verified.
- State the timeline. A substitution target without a target date is not a plan, and the share of addressable imports it represents makes the date the single most informative missing number.
- Report domestic value addition per product, not output. The 18–20 per cent phone figure is the reason output-based reporting and import-based targets diverge, and it is measurable.
The honest summary
$189 billion is a real amount of money and a defensible thing to aim at. It is 24.4 per cent of a $776 billion import bill, which sounds ambitious but achievable. The trouble is that the two largest chapters in that bill — crude and gold, 40 per cent between them — are not manufacturing problems, and once removed the target becomes 40.8 per cent of what remains and 73.7 per cent of the sectors actually named.
Those are not impossible numbers. They are, however, a completely different order of ambition from the headline, and the difference is entirely a matter of choosing the denominator.
The reconstruction above suggests the plan itself is more sensible than its presentation. A $50 million threshold really does produce about 1,272 lines in the public data, and the implied average per product points at a mid-tail band rather than at the whole import bill. If that reading is right, the programme is targeted at the products where domestic manufacturing plausibly could substitute, and deliberately not at crude, gold or the largest assembled goods. That is a defensible design. It is simply not what “$189 billion of imports” sounds like, and the arithmetic only becomes legible once someone divides the number by the right denominator.
Sources and caveats
All import figures are from DGCI&S commodity-wise import data retrieved from the TradeStat EIDB portal (tradestat.commerce.gov.in/eidb), at 2-digit HS chapter level, covering FY2018-19 to FY2025-26 with FY2025-26 provisional and the source last updated 19 May 2026. Ninety-eight of the ninety-nine possible chapters are populated. The FY2025-26 total of $776,014 million, the FY2024-25 total of $721,200 million, exports of $441,746 million and the resulting goods deficit of $334,268 million are sums computed here across all populated chapters, not figures quoted from a release; they may differ slightly from headline Ministry of Commerce totals depending on treatment of unclassified residuals. The 1,272-product count, the $189 billion figure, the $50 million per-product threshold and the four named sectors are as reported by Business Standard (16 July 2026) and Policy Circle. No primary government document setting out this list was located. A PIB search returned only defence indigenisation material and an unrelated DPIIT year-end review. These figures should be treated as press-reported, and this article deliberately does not reproduce or endorse the product list itself. The 8-digit reconstruction — 1,314 lines above $50 million importing $722,253 million, the threshold ladder, and the $50 million to $500 million and $50 million to $1 billion bands — is computed here from DGCI&S commodity-wise import data at 8-digit HS level for FY2025-26, covering 12,273 codes of which 9,561 carry value. The 8-digit annual sum of $775,712 million agrees with the chapter-level total to within 0.04 per cent. The inference that the 1,272-product list corresponds to a mid-tail band is this article’s own and is not confirmed by any source; it rests on the implied average per product matching the observed band average, which is suggestive rather than conclusive. The mapping of the four named sectors onto HS chapters 84, 85, 28, 29, 39 and 72 is this article’s own and is stated as such in the table note; “speciality steel” is a subset of HS 72 and including the whole chapter is generous to the target. The 18 to 20 per cent domestic value-addition figure for mobile-phone assembly is as reported by Policy Circle and is not independently verified here. The judgement that HS 27 and HS 71 are not substitutable by manufacturing is this article’s analytical claim, not a figure from a source: crude oil imports reflect resource endowment and gold imports reflect household savings behaviour. Readers who disagree with that exclusion should use the 24.4 per cent figure instead of 40.8 per cent; both are given above. Nothing in this piece is investment, trade or policy advice.
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.
