"A peculiar rebound in global industry in 2025"

Global manufacturing in 2025 did not follow the script of “trade conflict = industrial recession.” JPMorgan breaks the resilience of this year into several clues: global industrial production (IP) rebounded after the sluggish period of 2022-2024, and during the most intense phase of trade friction, the goods sector outperformed services.

According to Chasing Wind Trading Desk, JPMorgan’s global economics research team’s Joseph Lupton stated in a report on the 24th that industrial performance in 2025 “broke the historical close relationship with GDP.” This is the main anomaly the report seeks to explain: why growth didn’t collapse, inventories didn’t drag, and it's not just technology holding things up.

Their answer falls on three variables: First, the demand side is dominated by capital expenditures (especially equipment investment); second, the AI boom did indeed boost tech output, but more crucially, non-tech sectors returned from two consecutive years of contraction to positive growth; third, inventory shifted from being a “drag” to being “lean,” leaving room for flexible growth going forward.

Looking ahead, the report's outlook is not radical: with end demand stable, inventory low, and demand possibly spreading, global industrial output may still achieve 2%-3% annualized growth in the coming months. But risks are “bi-directional”—tech returning to a more sustainable pace, and labor market stagnation restraining retail, could push growth back down; tariff-related judicial changes are seen as unlikely to alter the mainline of US “trade conflict.”

2025's 2.4% was not achieved in one shot: strong front-loaded Q1, but no “hangover” in H2

The report sets 2025’s tone with one number: global industrial output grows 2.4% y/y (4Q/4Q). Of this, 1.6 percentage points occur in February and March, annualized to 9.4%. The report tends to link this to “concerns about global trade conflict” causing early production and pre-purchasing.

Worth noting, JPMorgan originally expected this “short stimulus” would stall in H2 2025, but the opposite happened: growth did slow, but maintained 1.7% annualized growth. Part of the explanation for why it “didn't collapse” is attributed to the pace of inventory adjustment—inventory decline gradually slowed in 2025 and provided a floor for output.

AI is the obvious driver, but 2025 is more like a “non-tech recovery”: from -0.8% to +1.2%

The tech sector’s strength is undisputed. The report states: excluding China, global tech output in 2025 grew 9.1% y/y, higher than 4.7% in 2024, driven by AI enthusiasm and hyperscaler capital expenditure.

But the report stresses: tech has been strong for years, and alone can’t explain the shift in 2025 from “2022-2024 slump” to “reacceleration.” The crucial variable is that non-tech output returned to positive growth—globally (excluding China) non-tech output in 2025 was +1.2% y/y, compared to -0.8% the previous two years. This isn’t a high prosperity reading, but the shift is enough to reshape thinking on “industrial cycle spread.”

Developed markets didn’t recover via autos: US and EU jumped in Q1 then leveled, with “multi-point fixes”

Regionally, the report points to developed economies (DM) manufacturing revival as a key theme of 2025: DM manufacturing contracted 1.4% in 2024, returning to +1.4% y/y in 2025. The most obvious improvements are in the US and Eurozone: in the two years to Q4 2024, US/EU manufacturing output shrank by -0.9% annualized and -2.9% annualized respectively; in 2025, growth was 1.7% and 1.8% respectively.

The report does not avoid the reality of “growth concentrated in Q1, then flat,” but also highlights: the anticipated “post-front-load reversal” did not appear.

At the sector level, autos were not the engine of this recovery. The report notes: autos performed stagnantly in developed markets, and US autos remained weak (related to trade friction impact). But the range of recovery was wider than “tech + pharma”: besides tech and pharma, the US also saw improvements in aerospace and machinery; in the Eurozone, machinery, autos and other heavyweight sectors rebounded from their 2024 weaknesses. Using report statistics, 14 of 20 major US manufacturing IP categories improved in 2025 vs. 2024, while in the Eurozone 9 of 11 categories improved.

Emerging markets: aggregate OK, but “Asia strong = tech strong,” and not evenly strong

The report views emerging markets (EM) as “steady but differentiated”: EM commodity production still saw 3.8% y/y growth (down from 4.6% in 2024), with recent strength mainly driven by Asia, and Asia’s strength highly concentrated in tech.

Among the economies listed, Singapore (+25%) saw one of the standout increases in tech output, with tech accounting for a larger share in manufacturing; Singapore also experienced disruption from year-end pharma front-loading. Conversely, in regions where tech drive is weaker and non-tech weighs more, 2025 was not “bright”—manufacturing output in Korea and Thailand shrank by 4Q/4Q, linked by the report to trade conflict drag.

Watch capex on the demand side: equipment investment lifts non-tech, but structural blind spots remain

The report sees “end demand uplift” as the base for the industrial rebound, with capital expenditures (capex) the standout. Global commercial equipment investment grew 6.5% y/y in Q3 2025, the fastest in three years; their capex nowcaster shows about 4.4% annualized growth in Q4, tracking a monthly pace of about 6% annualized entering the current quarter.

An easily overlooked point: equipment investment’s rise did not only occur in the US. The report’s tracking shows that, excluding the US and China, global equipment investment growth in the first three quarters of 2025 rose to 5.9% annualized, from 3.6% in 2024.

But the report admits visibility is limited: whether the rebound in non-tech output comes more from non-tech capex outside the US, or changes in non-tech inventories, is hard to dissect from the data. At least in the US, 2025 capex acceleration was mainly tech-driven, with other categories more mixed; by year-end, durable goods orders and shipments showed more “non-tech” improvement, seen as an early signal of demand spreading.

Inventory from “drag” to “lean”: upside risk comes from here

Inventory is a repeated “explainer” for the report. Early 2025 saw end demand (especially capex) jump ahead of output, which meant inventory growth slowed; output then caught up, and by mid-year inventory contributed slightly positively to manufacturing growth. The report thinks by year-end 2025, inventory drag has basically disappeared, and there is reason to think inventory levels are “lean.”

This is one reason why they price risk “bi-directionally”: If demand stays at current levels, lean inventories may force extra restocking, giving industrial output elasticity above demand itself.

Judicial changes to tariffs have “little impact”: mainline trade conflict persists

The report mentions the latest US Supreme Court overturning IEEPA tariffs; while it appears positive, JPMorgan concludes: This is more a constraint on policy tools than a shift in trade stance. The government subsequently announced 15% Section 122 tariffs and multiple exemptions; overall, the impact on effective tariff rates and “trade conflict” trajectory is limited, unlikely to disrupt recovering business confidence.

Coming months: 2%-3% annualized window, and “tech/employment” two-sided risk

With “lean inventory + fair end demand + potential spreading from tech to non-tech,” the report expects global industrial output may still achieve 2%-3% annualized growth in the next few months.

But they write the two downside lines clearly: first, tech growth returns to more sustainable levels; second, labor market stagnation suppresses retail and consumer goods demand. The report’s baseline assumption is that firm caution dissipates and hiring picks up, letting consumption replace investment as a more “income-driven” expansion; if this doesn't materialize, the “industry outpacing services” structural anomaly seen in 2025 may converge again.

~~~~~~~~~~~~~~~~~~~~~~~~

The above comes from Chasing Wind Trading Desk.

For more detailed interpretation, including real-time analysis and frontline research, please join [Chasing Wind Trading Desk Annual Membership]

Risk DisclaimerMarkets have risks, investment should be cautious. This article does not constitute personal investment advice, nor does it consider the unique investment goals, financial situation, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article are appropriate for their circumstances. Investing accordingly is at your own risk.