India's August Export Figures Are Published in Dollars. That Changes the Rupee Argument

The number, and the unit it is published in

The Ministry of Commerce and Industry released India's trade figures for August 2026 in mid-September. Merchandise exports for the month were USD 43.81 billion against USD 34.74 billion in August 2025, a rise of about 26 percent. Merchandise imports were USD 70.67 billion against USD 61.96 billion. Cumulatively for April to August 2026-27, merchandise exports were USD 215.91 billion against USD 183.21 billion a year earlier, a rise of 17.85 percent.

Two lines in that release are ours. Cotton yarn, fabrics, made-ups and handloom products rose 13.79 percent, from USD 0.99 billion in August 2025 to USD 1.12 billion in August 2026, and the release names that category among the major drivers of the month's export growth. Man-made yarn, fabrics and made-ups rose 9.92 percent. On the import side, textile yarn, fabric and made-up articles coming into India fell 8.12 percent.

The unit matters more than the size. Every figure above is denominated in US dollars. That single fact does most of the work in this post, because the argument a buyer is most likely to have heard about Indian fabric this year runs the other way.

Why the currency the figure is published in decides what it proves

The argument a buyer hears is roughly this: the rupee has been weak, so Indian export numbers look good, so Indian cloth must be getting cheaper and a quote should come down. We took that argument apart in a weaker rupee does not make Indian fabric cheaper for you, on the ground that a comparatively higher tariff at the buyer's own border can absorb whatever the currency gives.

There is a second and simpler reason the argument fails, and this release is where it shows. A weaker rupee mechanically inflates an export figure quoted in rupees, because the same cloth converts into a larger rupee number. It does nothing of the kind to a figure quoted in dollars. If the rupee falls and a mill ships the same cloth at the same dollar price, the dollar export line is flat and only the rupee line moves.

So a 13.79 percent rise in the dollar value of cotton yarn, fabric and made-up exports is not a currency artefact. It is more cloth, dearer cloth, or both. That is a materially different claim from the one a currency headline supports, and it is the reason we went to the release rather than to the coverage of it.

This is the same split that runs through our earlier reading of India's cotton-heavy export mix. Full-year 2025-26 data had man-made textiles rising in rupee terms while falling in dollar terms, which is exactly what a currency effect without volume growth looks like. August 2026 does not look like that.

The same August, counted twice, in two currencies

A second reading of the same month was published on 17 September 2026, and it is the cleanest illustration of this argument we have seen, because it is the same cloth in the same August counted in the other currency. The Ministry of Textiles put August 2026 textile exports up 16.1 percent year on year and April to August cumulative textile exports up 10.3 percent. Those are rupee-denominated series, and they are primary rather than press-reported: the figures are read off the ministry release of 17 September 2026, which is linked at the foot of this post.

Set the two commodity lines side by side and the gap does the talking. Cotton yarn, fabrics, made-ups and handloom products rose 24.1 percent in August on the rupee series, against 13.79 percent on the dollar series in the release quoted above. Man-made yarn, fabrics and made-ups rose 19.9 percent on the rupee series, against 9.92 percent on the dollar one. Same product groups, same month, roughly ten percentage points apart on both lines.

That spread is not a discrepancy and neither series is wrong. It is the currency, measured. The rupee number is the dollar number plus the depreciation over the comparison period, which is precisely the mechanism described above, and here it is large enough to be about two fifths of the rupee-denominated headline on the cotton line.

One more reading of the same period was published on 19 September 2026, by the Confederation of Indian Textile Industry, and it is the one that most directly answers a buyer who has heard that Indian textile exports are falling. On the industry body's figures, in dollar terms, August 2026 textile and apparel exports together rose 6.39 percent, with textiles up 13.03 percent and apparel down 2.74 percent. For April to August 2026 the split is wider still: textile exports up 6.94 percent, apparel exports down 9.10 percent, and the two combined marginally lower, by 0.24 percent year on year. These are an industry association's figures rather than a government release, so we are attributing them by name and not linking them.

That cumulative minus 0.24 percent is the number behind most headlines saying Indian textile exports are flat or falling, and it is a garment number wearing a textile label. The half of the basket a fabric buyer actually sources from rose over the same five months. It is the export-side version of the output-side split we described in the looms are busy and the sewing floor is not, arrived at from different data, which is the reason we trust it more than either reading alone. A buyer opening a negotiation with a falling-exports headline is quoting the sewing floor at a mill.

The buyer's conclusion follows directly. Growth of 24.1 percent sounds like an industry with more demand than it can serve, which is a poor setting in which to ask for a discount. Growth of 13.79 percent in the currency you actually pay in is a different and more modest picture. Whichever figure a negotiation opens with should be the one denominated in the currency of the contract, and for almost every buyer we quote, that is the dollar one.

The segment split in the same release is worth one more line, because it is a composition point rather than a currency one. In August, cotton yarn, fabrics, made-ups and handloom grew 24.1 percent and man-made yarn, fabrics and made-ups 19.9 percent, while ready-made garments of all textiles grew 6.1 percent, carpets 15.0 percent and handicrafts excluding handmade carpets 41.4 percent. Cumulatively for April to August the same ordering holds, with cotton textiles at 18.0 percent and man-made textiles at 13.6 percent against carpets at 11.9 percent. Even before the currency is stripped out, the growth is concentrated in yarn, cloth and made-ups rather than in finished garments, which is the half of the basket a fabric buyer is sourcing from.

Volume against value, and who said which

A dollar figure still cannot separate more cloth from dearer cloth on its own. The Commerce Secretary, Rajesh Agrawal, was asked directly at the release whether the month was a currency story, and his answer as reported in the Indian business press was that 68 of 168 principal commodities showed growth in both volume and value, with a further 39 showing value growth alone.

We are attributing that to the press coverage rather than to a document, because it was a spoken answer at a briefing and we have not found it in the published release. It is the right shape of answer to the question and we are repeating it as an official's characterisation, not as a measured series. A buyer who wants the volume split for a specific fabric line will not find it there either way, because the principal-commodity grouping is far broader than a fabric quality.

The reported destination detail is also secondary rather than primary. Coverage of the same release put exports to the United States up about 22 percent to USD 8.4 billion despite the additional 10 percent line we described in the Section 301 post. If that holds, it is the more interesting number in the release, because it says the additional duty did not stop the trade. It does not say who absorbed it.

The import line is the quieter half

Textile yarn, fabric and made-up articles imported into India fell 8.12 percent year on year in August 2026. That line rarely gets written up and it is worth a paragraph, because it closes off a reading that would otherwise be available.

If Indian textile exports were rising while textile imports rose alongside them, a reasonable buyer could conclude that more cloth was simply passing through the country, and that an Indian address on an invoice was becoming a weaker signal about where the cloth was made. Exports up and imports down does not support that reading for the month.

This matters at the paperwork stage rather than the price stage. Where a buyer claims a preference under an agreement, the question is never where the invoice was raised, it is whether the consignment passes the origin test, which for woven cloth usually turns on where the weaving happened. Our India-UK origin rules post sets out what that test actually asks for.

What none of this does to your quote

Nothing in a monthly trade release reaches a poly-viscose quote, and we would rather say so plainly than let a good month imply otherwise.

A quote for suiting or shirting moves on the polyester feedstock chain, on dyeing and processing charges, on freight, and on the duty payable at the buyer's own border. A national export aggregate contains all of those and resolves none of them. It is also a backward-looking figure for a month that closed before the enquiry was raised.

The one legitimate use of a release like this is directional. It tells you that the trade you are buying into is growing rather than contracting in the currency you are paying in, which is a reasonable thing to want to know about a supply base before committing a programme to it. It is not an input to a price negotiation, and a buyer who opens one with it is arguing from an aggregate against a quotation.

If you want the figures that do decide a quote, the shorter route is to ask for them. What to send a mill on day one is the list, and a stated blend, GSM band, width and shade reference will tell you more about a price than a trade release ever will.

The September release, read the same way

15 October 2026. The Ministry of Commerce and Industry released the September 2026 figures today. Merchandise exports were USD 44.40 billion against USD 36.08 billion in September 2025, a rise of 23.04 percent, and cumulative April to September merchandise exports were USD 259.89 billion, up about 18.5 percent. The merchandise trade deficit narrowed year on year to USD 30.21 billion from about USD 33 billion, though it widened from August. The Commerce Secretary attributed the month to electronics, petroleum products, engineering goods, chemicals and agricultural products.

The release itself names three textile lines. Cotton yarn, fabrics, made-ups and handloom products rose 20.76 percent, from USD 0.93 billion to USD 1.12 billion, and are again listed among the month's major drivers. Man-made yarn, fabrics and made-ups rose 11.23 percent, against 9.92 percent in August, and ready-made garments of all textiles rose 13.68 percent. These are dollar figures again, so the same reading applies as for August: more cloth, dearer cloth, or both, and not a currency artefact. One caution. September 2025 was a weak base, which flatters any year-on-year rate for the month, so the direction matters more than the size: both fabric lines grew faster in September than in August, and garments are growing too. That is consistent with the fuller mill order books described in our post on the textile stocks rally, though one month cannot confirm it.

What we are not saying

One month is one month. April to August cumulative growth of 17.85 percent in dollar terms is the steadier figure and we would treat that as the reading rather than the August number on its own, particularly against a full year 2025-26 that was reported down 2.2 percent for textiles and garments.

We are also not treating the release as settling the tariff question. The comparatively higher duty Indian goods face in some markets against other Asian suppliers is a separate fact and this data does not overturn it. Both things can be true: the trade grew in dollar terms in August, and the duty a buyer pays at their own border is still the largest single item a headline about India will not show them.

We have left out the rupee level entirely, as we do throughout this site, and we have left out the composite services estimate in the same release because it has nothing to do with cloth.