Update: US Strategic Subsidy Targeting Hits 76% as Big Three Converge on Chokepoints

What changed since the August update vintage

In August we refreshed this theme with GTA’s 2025 NIPO briefings on motives and green instruments: Western security justifications jumped 26% → 63%, climate fell to 12%, and inside green measures import barriers surged from under 4% to 48%. That update answered the framing and toolkit question — are governments justifying industrial policy as security, and are they coercing more than they subsidise? This Q3 vintage answers the targeting question the newest official print forces: where do the subsidies actually land, and have China, the EU, and the United States converged on the same chokepoints?

Two Global Trade Alert releases supply the refresh. Zeitgeist #88 (April 2026), Subsidising the Chokepoint, tracks the share of subsidy-based industrial actions covering dual-use or advanced-technology products through 2025–26. The companion long panel, Industrial Policy as Market-Shaping Competition, extends a consistent China–EU–US NIPO ledger through December 2024 and documents ten stylised facts on activity levels, green catch-up, subsidy races, and instrument persistence. Neither replaces the August motive table; together they show what the prior print could only imply — that security language and coercive tools are concentrating on the same narrow product space.

The dashboard above is built as a vintage delta: diverging change bars, a three-bloc strategic path, prior→new dumbbells, green catch-up bars, a subsidy-follow band, selective-activity levels, persistence comparisons, and a strategic-versus-green scatter. Use the bloc and delta group controls to isolate China, the EU, or the United States.

The headline table: prior window vs newest print

The story is not “industrial policy paused after the August security spike.” Selective action counts plateau near 1,900 a year — roughly 800 above the pre-2019 mean — while subsidy dollars and notices concentrate on semiconductors, critical minerals, batteries, and advanced electronics. The August print said governments talk security; this print says they spend and restrict on the same chokepoints.

Strategic targeting: the US catch-up is the headline delta

Toggle Delta group → Strategic. The United States is the largest vintage move on the board: strategic-sector coverage of subsidy-based industrial actions rises from 33% in 2009–16 to 76% in 2025–26 — a +43 percentage-point swing. ZG #88 notes the path was already near 76% by 2020–22, coinciding with Section 301 escalation, Entity List designations, CHIPS, and the IRA, then stabilising through the newest window.

China started where others are trying to arrive. Strategic sectors absorbed 85% of Chinese subsidy actions in 2009–16 and never fell below 90% thereafter, reaching 98% in 2025–26. The EU is the compressed catch-up story: 50% → 70% over the same long windows, with a ~75% peak in 2023–24 after the European Chips Act, Critical Raw Materials Act, Net-Zero Industry Act, and multiple IPCEIs, before settling at 70% in 2025–26.

Read that against the August update carefully. Security motives at 63% among G7+KR+AU peers told you the justification. Strategic shares at 70–98% among the Big Three tell you the product map. Convergence is real, but the paths are not: China built a decade-plus architecture; the US pivoted sharply after 2017; the EU compressed the same turn into a few post-2020 statute cycles.

Activity stays high: a regime, not a spike

Filter Delta group → Activity. The Market-Shaping panel’s selective industrial-action series averages about 1,100 measures a year from 2009–19, jumps to roughly 1,800 in 2020–21, and plateaus near 1,900 in 2022–24. That is the stock-context partner to our H-NIPO research ledger, which already showed 34,248 historical interventions and a 2023 census of 2,580 measures: the flow did not unwind when pandemic emergency frameworks expired.

Instrument mix still diverges even as sectoral targeting converges. In China and the EU, domestic subsidies and export incentives remain roughly 85–97% of recorded industrial actions across sub-periods. The United States keeps a more trade-exposed toolkit — subsidies paired with import barriers, localisation rules, and eligibility conditions — which is consistent with the August finding that Western green measures shifted toward import barriers even while the long H-NIPO stock stayed subsidy-heavy.

Green catch-up blurs into chokepoint policy

Switch Delta group → Green. China already directed about 40% of subsidy-based industrial actions to low-carbon technologies in 2009–16 and kept that share high. The EU and United States sat near or below one-fifth through 2009–19, then surged past 50% by 2023–24. Batteries, power electronics, grid components, and critical-mineral inputs sit in both the “green” and “dual-use / advanced” buckets, so the green catch-up is not a separate climate silo — it is how Western capitals entered the same chokepoint race China had already been running.

That reframes the August green-instrument story. Import barriers rising to 48% of Western green measures looked like climate policy turning coercive. In the Q3 print, the same products are also the strategic-subsidy core. Coercion and credits are increasingly aimed at overlapping HS codes, which is why a firm can face a production credit in one jurisdiction and a barrier in another for the same battery chemistry or wafer node.

Subsidy races and export lock-in

Filter Delta group → Interaction. Same-product (HS6) subsidy follow within twelve months already ran one-third to two-thirds of Big Three subsidy actions in 2009–16. After 2020, following rates frequently sit in the 60–80% band, especially when China or the United States leads. The dashboard’s midpoint series (50% → 70%) is a readable summary of that disclosed range — not a claim that every pair hits exactly seventy.

Persistence completes the vintage. Export restrictions that once retained about half of measures after twelve months now show 80–90%+ retention in recent China and US cohorts, with the EU approaching full persistence. Import barriers remain more contingent; subsidies remain near-permanently on the books by design. Pair that with the August security-motive spike: security framing plus durable export controls is a qualitatively stickier regime than climate-framed production subsidies that finance ministries can sunset.

For the dollar ledger rather than intervention shares, keep our US industrial subsidies vs tariffs over 30 years open beside this post. NIPO counts a tariff notice and a multi-year grant as comparable actions; fiscal exposure is a different question.

Who is exposed under the Q3 vintage

Exposed: mid-sized economies that cannot match Big Three subsidy follow speed on the same HS6 products; exporters into markets where strategic subsidies arrive bundled with rival-directed import barriers; green-tech investors who underwrote Western catch-up as cooperative climate policy and now face chokepoint logic; fabs and battery projects that win announcement subsidies but still lack supplier depth — ZG #88 flags cancelled or delayed flagship semiconductor projects even after large state-aid pledges.

Relative winners under current rules: firms already inside dual-use and advanced-technology product lists in all three blocs; capitals that can stack multi-year subsidy commitments and persistent export controls; upstream critical-mineral and equipment suppliers that sit at the intersection of green and security targeting.

What would change the story: a 2026–27 NIPO window where strategic subsidy shares diverge again (EU retreat from 70%, US fall from 76%); a fiscal brake that actually cuts follow-on subsidies below the 60% band; or successful ecosystem builds that make chokepoint subsidies look less like announcement theatre. None of those reverse signals appear in the April 2026 chokepoint briefing or the 2009–2024 panel.

Caveats and methodology

• Windows differ. The August update used GTA 2023–24 vs 2025 motive/instrument shares and H1’24 vs H1’25 activity. This Q3 print emphasises ZG #88 2009–16 vs 2025–26 strategic subsidy shares and the Big Three panel through Dec 2024.
• Shares ≠ dollars. Intervention counts treat a licensing rule and a multi-billion grant as comparable observations.
• Path points between disclosed endpoints (2017–19, 2020–22, 2023–24 on the strategic line chart) are estimated from narrative waypoints in ZG #88 and Market-Shaping Fact 5; endpoints 33/50/85 → 76/70/98 are disclosed.
• Subsidy-follow and export-persistence midpoints summarise disclosed ranges (60–80%, 80–90%+); charts show bands where the source gives them.
• Green late shares for EU/US are coded near 52% to represent the disclosed “exceeding 50%” bandnot a false precision point estimate.
• Geography is Big Three–centric here; August’s G7+KR+AU and non-Western series remain valid for motive comparisons but are not restated in ZG #88’s subsidy-targeting figure.
• Cancelled megaprojects in ZG #88 are qualitative caveats about capability, not NIPO field revisions.

The shareable takeaway

Versus our August motive-and-instrument vintage, the newest GTA print shows Big Three industrial policy converging on chokepoints, not cooling off: US strategic subsidy targeting jumps 33% → 76%, the EU reaches 70%, and China sits at 98%. Selective actions hold near 1,900 a year, Western green subsidy shares cross 50%, same-product subsidy races crowd into a 60–80% follow band, and export controls lock in at 80–90%+ twelve-month retention. Security motives told you the story in August; dual-use product maps tell you where the money and barriers now land.

Related reading: August fiscal & industrial policy update and Fiscal & industrial policy research (H-NIPO/2023).