Update: West Installations Fall 8–10% While China Hits Record 295k — Cobots +12%

What changed since the research vintage

Our IFR World Robotics 2025 research post mapped where factory robots land: 542,076 global installations in 2024, Asia’s 74% share, China’s 54% of demand, and electronics edging automotive. That print answered the level question. This update answers the delta question the same vintage forces once you stop ranking markets and start ranking changes: who is still scaling physical automation, and who is cycling down?

The dashboard above is built as a vintage delta — diverging YoY bars, decade industry dumbbells, flow-vs-stock dual axis, cobot path, density leaders, and an installations×YoY scatter — with Market region and Delta group controls. The shareable frame is narrow: world flow held sideways while operational stock rose 9% to 4.66 million, Asia installations rose 5%, Europe fell 8%, and the Americas fell 10%. Collaborative robots grew 12% to 64,500 units — about 12% of all industrial installs.

The headline table: levels vs YoY

Read the table against the research post carefully. The research print was not wrong that Asia dominates levels. The update adds that Asia is also the only major region still growing installations, while Western markets are cutting units even as their density leaders remain among the world’s densest factories.

YoY divergence: China and India up; West and Japan down

Toggle Market region → All on the left panel. The diverging bars separate growers from cutters without drowning in absolute scale. China (+7%) and India (+7%) sit alone among large markets on the positive side. Japan (−4%), Korea (−3%), and Germany (−5%) are mild contractions from high bases. United States (−9%), Italy (−16%), France (−24%), Canada (−12%), and the UK (−35%) are sharper — the UK’s plunge is the hangover from a 2023 tax-credit spike, not a structural exit from automation.

Filter to Europe and the regional story collapses into one sentence: nearshoring talk did not prevent an 8% installation drop from Europe’s 2023 all-time high of ~92k units. Filter to Americas and the same pattern appears at −10%, with the US accounting for about 68% of regional installs and still declining. Asia’s +5% is enough to keep the global total flat because Asia is three-quarters of the flow.

That is the first update punchline versus the research vintage: concentration is not only a stock fact — it is a growth fact. When three-quarters of new robots land in Asia and Asia is the only region expanding, “global factory automation” increasingly means Asian factory automation with Western cyclical noise around it.

Stock +9% while flow is flat — the compounding story

The research post noted the plateau. The update’s dual-axis panel makes the implication visible: installations can sit sideways while the operational fleet keeps growing. World stock rose from roughly 4.28M to 4.66M (+9%). China’s stock alone crossed 2.03 million — about 43% of world stock — after years of 20%+ average growth.

Why the asymmetry? Robots stay on the floor for years. A flat flow still adds hundreds of thousands of units to stock each year, and retirements do not cancel the entire cohort. For planners, that means labor displacement and maintenance markets track stock, while OEM order books track flow. Reading only the 542k annual number understates how much automation is already embedded — and overstates how “done” Western factories are when their density remains high even as their new orders fall.

Cobots +12%: the slice still accelerating

Collaborative robots are the structural exception inside a flat aggregate. IFR’s WR 2025 collaborative chapter / press series put 2024 cobot installations at 64,500, up 12%, restoring double-digit growth and lifting cobots to roughly 12% of all industrial robot installations. Over five years, cobot installs more than doubled from the ~30k class of 2020.

That does not mean cobots are replacing high-payload welding arms on auto body lines. It means lighter, safer, easier-to-integrate systems are finding SKUs in packaging, electronics handling, and mixed-product lines where traditional fencing was the barrier. When the dashboard’s Delta group → Structure filter lights up cobot YoY next to stock YoY and China domestic share, you see three accelerating stories inside a flat headline: stock compounding, cobot mix, and Chinese supplier localization.

Decade mix: general industry 36% → 53%; auto 43% → 23%

The research post’s electronics-vs-auto duel for 2024 (24% vs 23%) was a one-year photo finish. The update’s decade dumbbell is the longer movie. Automotive’s share of installations fell from 43% in 2014 to 23% in 2024. General industries (ex-auto) rose from 36% to 53%. Electronics edged from 21% to 24%. Installations in general industries more than tripled over the decade; auto rose only about a third.

So the plateau is not “automation paused.” It is customer industry reweighting. Auto capex cycles in Europe, Japan, and North America pulled Western installs down in 2023–24. Electronics and metal/machinery (16% share, +2 pp) offset enough of that decline to keep the world total flat. Anyone still narrating industrial robotics as an auto story is a decade late.

Density: Korea 1,220 vs world 177 — intensity ≠ flow

Pair this update with our manufacturing robot density chart. WR 2025 puts world average density at 177 robots per 10,000 manufacturing employees in 2019–24’s endpoint year. Asia averages 204 (12% CAGR since 2019), Europe 148 (7% CAGR), Americas 131 (6% CAGR). Country leaders remain South Korea (1,220), Singapore (818), then China (567) — density that now sits above Germany (449) and Japan (446).

The update’s density panel is the corrective to “US is falling behind because installs fell 9%.” The US can cut new units while still running a large stock; Korea can install only ~31k units a year and still lead the world on intensity. Flow answers who is buying this year. Density answers how automated the floor already is. Both matter; they are not substitutes.

China suppliers: 28% → 57% home-market majority

The research print flagged China’s domestic supplier share at 57%. The update treats the +29 percentage-point decade climb from ~28% in 2014 as its own structural delta. For the first time, Chinese manufacturers sold more robots in China than foreign suppliers. That reshapes who captures margin when Asia supplies three-quarters of new units — and who is exposed if Chinese OEMs begin exporting more aggressively into Asia’s second tier (India, Southeast Asia).

IFR still expects China demand to support roughly 10% average annual growth potential into 2028, with Asia installations near 435,000 in the 2025 outlook year. Global path: 575,000 in 2025 (+6%), then a trajectory past 700,000 by 2028. Those forecasts assume the Asian engine keeps running; they do not require Western installs to recover immediately.

Who is exposed under the new vintage

Exposed: Western robot OEMs and integrators keyed to auto-capex cycles in the US, Germany, Italy, and France; narratives that treat “global robotics growth” as synonymous with Western factory employment stories; policymakers who read a flat 542k year as proof automation stalled rather than relocated and reweighted.

Relative winners under current rules: Chinese domestic robot makers with home-market majority; Asia-based electronics and general-industry integrators; cobot vendors still on a double-digit unit path; maintenance/service businesses tied to stock rather than flow.

What would rewrite the story: a 2025–26 Western auto reinvestment wave that restores Europe/Americas positive YoY; a China growth miss below the ~10% narrative; or cobot share stalling as safety/throughput limits bite. IFR’s own forward path still points up — but the geography of the upcycle remains Asian until Western YoY signs flip.

Caveats and methodology

• Same IFR WR 2025 vintage as the research postthis is a lens update (YoY, density, cobots, decade mix), not a newer statistical year. Full 2025 installation counts await WR 2026.
• YoY % for some mid-tier markets back out from disclosed 2024 units and YoY; those 2023 levels are marked estimated in the data module.
• Cobot series before 2024 interpolates IFR’s “more than doubled in five years” narrative around the disclosed 64,500 / +12% / ~12% share endpoints.
• Density uses manufacturing employment denominators that IFR revises with lag; country ranks can shuffle on employment revisions without install changes.
• Spain YoY is coded ~0 where IFR disclosed units without a clear YoY; treat as placeholder.
• General industry is IFR’s ex-automotive manufacturing bucketnot identical to “all non-auto customers” in every national account.
• Forecast years (2025–2028) are IFR outlook paths, not observed installs.

The shareable takeaway

Versus our research print’s level map, the YoY update says physical factory automation is still scaling in Asia while Western installs contract: Asia +5%, Europe −8%, Americas −10%, China record 295k (+7%), world flow flat at 542k, stock +9% to 4.66M, cobots +12% to 64.5k. Automotive’s decade share collapse (43% → 23%) and China’s domestic supplier majority (57%) are the structural footnotes. For density intensity see manufacturing robot density; for the geography levels see the research vintage.