- Component orders fell 25% in Q3 2025 and stayed below 2024 for the year. The 24% rise in H1 2026 is a recovery to roughly 2024 levels, not expansion.
- A3 names the mechanism in its own Q1 release, unquoted by the trade press: components lag OEM ordering cycles by six to eighteen months.
- Component units rose 28.1% but value only 15.5%. Price per robot fell about 10%, which is a cost-constrained buyer meeting a programme obligation.
The headline went around in August. North American robot orders up again, and inside them a split that looked like a story: automotive OEM orders down 25 per cent in the first half, component supplier orders up 24 per cent. The reading everyone published was that suppliers are automating while the OEMs sit on their hands.
If you run a tier supplier, that probably did not match your week. Supplier bankruptcy filings through December 2025 were higher than any year in the past decade. Marelli went through Chapter 11, Accuride went through Chapter 11, and Automotive News started a dedicated tracker for supplier layoffs and closures in May. A sector filing at that rate is not on an automation spending spree.
Both things are true. The order numbers are correct. The interpretation is not.
What the baseline does to the number
A3 publishes percentages, not levels, and a percentage without its base is close to meaningless. Here is the base.
Automotive component orders fell 25 per cent in the third quarter of 2025, and A3's own year-end summary said the segment "remained below 2024 levels" for the full year. So the 24 per cent rise in the first half of 2026 is a recovery from a trough. It puts component orders back at roughly where they were in 2024. That is not expansion.
The OEM number works the same way in reverse. Automotive OEM orders were up 34 per cent in the first half of 2025 and up 68 per cent in the third quarter. The comparison base for 2026 is that surge. A3 called the current decline cyclical and pointed at the size and timing of vehicle programmes.
Put the two series next to each other and they are not diverging. They are swapping places.
The sentence nobody quoted
A3 explained the mechanism itself, in the first quarter release, and neither of the trade write-ups picked it up. Component supplier orders rose in Q1, "reflecting the segment's typical lag behind OEM ordering cycles."
Anyone who has sat through a programme award already knows this. The OEM commits capacity first. Suppliers tool for the same vehicle six to eighteen months later, once the award is signed and the volumes are in the contract. OEMs ordered heavily through 2025, so suppliers are ordering in 2026 against those awards, and they would be ordering whether or not the balance sheet enjoyed it.
Which means the component number is not a signal about supplier confidence. It is an echo of a decision an OEM made last year.
The price tells you the rest
Look at units against revenue. In the first quarter, component orders were up 28.1 per cent in units but only 15.5 per cent in value. Average price per robot fell about 10 per cent. Suppliers are buying more machines and cheaper ones.
That is what a cost-constrained buyer meeting a fixed obligation looks like. The flexible cell with vision and force control that somebody wanted becomes the smallest robot that will hold the cycle time in the contract. Anyone who has been handed a capital number and told to make the programme work will recognise the exercise.
The OEM side has its own tell. Orders fell 35.1 per cent in units but 48.2 per cent in value, so price per unit dropped around 20 per cent there too, which is what you would expect if the money is moving out of body-in-white. It is. GM has directed $830 million into three Midwest powertrain plants for transmissions and castings. Nissan cancelled a $500 million EV conversion at Canton. VW ended EV production at Chattanooga. Powertrain machining and casting uses far fewer robots per dollar than a body shop, so the same capex buys a smaller unit count.
It is also worth ruling out the obvious alternative. The rise is not battery and e-drive newcomers. The Dallas Fed counted roughly ten North American battery projects cancelled or stalled, over $10 billion of investment, after federal EV purchase subsidies ended in September 2025. If battery suppliers sit inside that component bucket, they are pulling it down rather than lifting it.
What to do with this
The distress and the orders describe different companies. BCG's 2026 supplier study puts average supplier EBIT margin at about 5.7 per cent, ahead of OEMs at 4.8 per cent, with a wide spread between the top quartile and the bottom. The headlines come from the bottom of that distribution. The robot orders come from somewhere else in it.
So three things are worth checking before your own capital committee meets.
First, know which half of that distribution your programme awards put you in, because the sector average will tell you nothing useful.
Second, if you are tooling in 2026 for a programme awarded in 2025, the volume assumption in that contract was made in a different market. Run the automation business case at the volume you now believe, not the volume in the award, and find out what the payback looks like if the programme underdelivers by twenty per cent.
Third, watch the third quarter print. If component orders stay positive while OEM orders recover, the lag reading holds. If components roll over as the 2025 echo fades, this was a base effect and nothing more. Either way, do not let a growth rate in a trade publication stand in for what is actually happening in your own order book.