The Looms Are Busy and the Sewing Floor Is Not

One number that is easy to check, and surprising

India's Index of Industrial Production moved to a 2022-23 base year this June, so the index reads as a straight comparison against 2022-23 output. On that base, in June 2026, Manufacture of textiles stood at 136.4 and Manufacture of wearing apparel stood at 95.6.

Read those two numbers slowly. Fabric and yarn production is running about 36 percent above the 2022-23 level. Garment production is running about 4 percent below it. Not below last year, below the base year the whole index is measured against.

Year on year the same table gives textiles up 13.7 percent and wearing apparel down 6.9 percent. For the April to June quarter it is textiles up 14.3 percent and apparel down 7.4 percent, so this is a quarter-long pattern rather than one odd month. General industrial production grew 7.3 percent over the same month and manufacturing 7.8 percent, which means textiles beat the manufacturing average by a wide margin while apparel went backwards against it.

These figures are quick estimates and get revised, and the weights are worth knowing when you read them: textiles carries 3.275 in the index and wearing apparel 1.970.

China's official data says the same thing from a different direction

The same split turns up in China, measured in money rather than in volume. Figures released by China's National Bureau of Statistics on 27 August 2026, as reported, put revenue at above-scale textile companies for January to July up 2.2 percent year on year, with profits up 7.9 percent. Apparel and clothing revenue over the same period was down 3 percent.

Cloth up, garments down. Two countries, two completely different statistical instruments, one shape.

We should be straight about how we know this. Those Chinese figures reached us through a news agency report of a report, and we could not open China's statistics bureau from here to read them in the original, so they are attributed and not linked. The Indian figures below and above are different: they come from the government press release itself, which is linked at the foot of this post, and you can check every number in this post against it.

The story that came with those Chinese numbers, and why we are not repeating it

The reason the Chinese data was in the news at all was a claim that roughly 1,500 companies left China's official industrial statistics during the first half of 2026, with the count of above-scale textile companies down 389 from a year earlier. That was reported as contraction.

It is worth being careful here, because the same article carries the facts that complicate its own headline. "Above-scale" is a reporting threshold, not a measure of health: it means annual revenue above a set level, and the report itself notes that leaving the category can mean a company shrank below the threshold, restructured, or closed. Meanwhile revenue and profit for the companies still in the category went up.

Fewer firms, more revenue, more profit. That is the shape of consolidation, not collapse. It may be uncomfortable for the firms that dropped out, and it is a real event, but it is not the same claim as an industry shrinking, and a buyer who acts on the second reading when the first is true will make a bad decision.

We are not in a position to settle which reading is right. We are in a position to say that the numbers quoted in support of the contraction story do not, on their own, establish it.

What this does not mean: India is not simply winning

The tempting next sentence is that China is shrinking and India is taking the work. We are not writing that sentence, because the 2026 trade evidence does not support it.

India's busy looms are a production fact, not an export-share fact. Trade reporting through 2026 describes Indian textile and apparel exports to the United States falling sharply in the first half of the year, on tariff changes rather than on anything to do with capability, with Bangladesh drawing level in US shipments. We are not putting a number on that here because the figures we could reach sit behind a paywall that masks the digits, and a number we cannot read is not a number we will publish.

So the honest 2026 picture is narrower and more useful than the headline version: Indian fabric output is strong, Indian garment output is weak, and Indian export share in at least one major market has gone backwards. Those three things are all true at once, and any story that keeps only the first is selling something.

Why the fabric stage is the strong one, and what it means for a quote

This site has been describing the same divergence from a third series all year, and it is worth putting the three side by side because they were reached independently. India's Wholesale Price Index has Manufacture of Textiles running at 12.63 percent year on year in August 2026 against Manufacture of Wearing Apparel at 3.12 percent, which is the pass-through problem in official form. India's output index has fabric volume up and garment volume down. China's revenue data has cloth up and apparel down.

Price, volume and revenue, three instruments, all saying the cloth stage is carrying the chain and the garment stage is not keeping up.

For a buyer the practical reading is about where cost pressure is sitting and who is absorbing it. If you buy finished uniforms, your garment maker is working in the weak half of that chain, buying from the strong half. That is precisely the squeeze that produces requests to reopen a price, shorten a quote validity window, or substitute a cheaper cloth late in a programme. Knowing which stage the pressure is actually coming from is what lets you tell a real cost problem from a negotiating position.

If you buy cloth directly, the same fact reads differently: the stage you are buying from is busy. Busy is good for availability and bad for the assumption that a mill will discount to fill a loom.

One thing India cannot tell you, and it is worth knowing why

China can produce a story about firms leaving a statistical category because it maintains that category and updates it. India cannot produce the equivalent, because the official count does not exist in current form.

The Office of the Textile Commissioner's state-wise table of mills and installed capacity is the national reference for how many mills, spindles and looms India has. It is dated 30 September 2017. There is no 2026 equivalent. The Annual Survey of Industries reports textiles as a share of factories and of persons employed rather than publishing a current headline count in its summary.

So if you ever read that a specific number of Indian mills closed or opened this year, ask where the number came from, because there is no official series that produces it. This is not a criticism of anyone; it is a limit on what can be said honestly, and it is the reason this post talks about output volume rather than about how many mills exist.

What a buyer can actually do with this

• Separate the two stages when you read sourcing news. A claim about the garment industry is not a claim about the fabric industry, and in 2026 they are pointing in opposite directions in both India and China.
• Treat "the industry is contracting" claims as a question rather than a fact. Ask whether the evidence is firm counts or revenue, because those two have been telling different stories.
• If your garment maker asks to reopen a price, the squeeze they describe is real and measurable, and it is not evidence that your cloth supplier is the one who moved. Ask which input actually changed.
• Do not plan a programme on the assumption that busy mills will discount. Fabric output running well above its base year is not the condition in which capacity gets sold cheaply.
• Fix the specification early if a decision is near. Blend, count, weave, GSM and shade settled in advance is what lets you accept a quote inside its window, which matters more when the stage above you is under pressure. Planning the purchase calendar is the cheapest protection available.