IESET.
Hypotheses·trade·green_industrial_policy_global_chip_race_2022_2026

The 2022-2026 wave of major-economy industrial-policy programmes — US IRA + CHIPS, EU Critical Raw Materials Act + Net-Zero Industry Act, EU Chips Act, Japan Green Transformation (GX, ¥150tn / ~$1tn announced), Korea K-Chips + Korean New Deal 2.0, China 14th Five-Year Plan + Made-in-China-2025-2.0 with semiconductors and clean energy as national-security frontier — represents the largest coordinated wave of industrial-policy spending in the post-1970s OECD record.

The comparative-effectiveness question (which programmes delivered MW deployed / fab construction / clean-manufacturing capex per dollar spent) is testable. The hypothesis predicts that programmes which combine a clear technology target, a price-signal complement (carbon pricing or feed-in tariff), and an export-discipline mechanism (China WTO-era model) outperform pure-subsidy programmes (US IRA leans subsidy-heavy with weak export discipline; EU NZIA mixes subsidy with weak target-clarity). Net global clean-energy + semiconductor capacity rose materially but with substantial geographic redistribution rather than aggregate creation.

INCONCLUSIVEengine/runs/green_industrial_policy_global_chip_race_2022_2026

INCONCLUSIVE_DATA_PENDING — insufficient observations after listwise deletion (20)

confidence cueResult card produced; verdict unclassified.

policy briefCoverage too thin

In ordinary language

When countries open more of the economy to trade and competition, do people end up with better long-run income or productivity outcomes?

plain answer

This test cannot make a firm call yet. insufficient observations after listwise deletion (20)

why it matters

This matters because trade claims should change belief only when they survive a pre-declared empirical test.

how the test works

It compares 14 country or place units from 2018 to 2027, using a panel fe decomposition design, with fixed effects for country and year.

what was measured
What changed
  • Program design score
  • Post 2022 dummy
What we checked
  • Log program announced disbursement usd
  • Log program realised disbursement usd
  • Log country clean energy mw added
what this does not prove

A single test is not the whole truth. It narrows the claim under a specific sample, time period, and method. Strong policy conclusions need the pattern to survive nearby tests, alternative data, and serious objections.

verification

No evidence packet has been generated yet.

Results

engine/runs/green_industrial_policy_global_chip_race_2022_2026
1007550250201820232027USADEUFRAJPNKORCHNGBR
illustrative sketch · run pending
No coefficients yet. When the model fires, this chart will show log_program_announced_disbursement_usd across 14 sampled countries over 2018–2027.
The shapes above are stylised — none of the lines are real data.
Placeholder for green_industrial_policy_global_chip_race_2022_2026. Published chart will be generated from engine/runs/green_industrial_policy_global_chip_race_2022_2026/chart_data.json.

Pre-registration

registration ordering unverified
first-spec commit 4c8ce8e · 2026-07-18T22:11:21Z
run generated · 2026-06-29T17:54:39Z
Run timestamp predates this path's first git-add commit (rebase, rename, or pre-git local run). Spec hash is still the path's first-add commit — not repository HEAD — but ordering is not a clean pre-registration proof.

The 2022-2026 wave of major-economy industrial-policy programmes — US IRA + CHIPS, EU Critical Raw Materials Act + Net-Zero Industry Act, EU Chips Act, Japan Green Transformation (GX, ¥150tn / ~$1tn announced), Korea K-Chips + Korean New Deal 2.0, China 14th Five-Year Plan + Made-in-China-2025-2.0 with semiconductors and clean energy as national-security frontier — represents the largest coordinated wave of industrial-policy spending in the post-1970s OECD record. The comparative-effectiveness question (which programmes delivered MW deployed / fab construction / clean-manufacturing capex per dollar spent) is testable. The hypothesis predicts that programmes which combine a clear technology target, a price-signal complement (carbon pricing or feed-in tariff), and an export-discipline mechanism (China WTO-era model) outperform pure-subsidy programmes (US IRA leans subsidy-heavy with weak export discipline; EU NZIA mixes subsidy with weak target-clarity). Net global clean-energy + semiconductor capacity rose materially but with substantial geographic redistribution rather than aggregate creation.

Falsification criterion — what would disprove this

set before the run · honoured after

This hypothesis is considered falsified if:

Not supported if (a) program_design_score has zero or negative effect on effectiveness ratios at p<0.10, OR (b) the rank-order of programmes by effectiveness does not match author-predicted rank (Korea K-Chips + China composite outperform US IRA + EU NZIA), OR (c) global clean-energy / semiconductor capacity addition exceeds pre-2022 trend by less than 10% (would refute the "step-change" framing — programs would just be redistribution), OR (d) the country-level effectiveness measure is dominated by GDP / pre-existing industrial base rather than program-design (no policy lesson).

formal test & threshold
test:      global_industrial_policy_design_panel
threshold: program_design_score β > 0 at p<0.10 on effectiveness ratios AND Korea + China rank in top 3 of 8 by spend-per-MW + spend-per-kWspm AND global clean-energy capacity 2022-2026 > 1.20x pre-2022 trend extrapolation AND program-design-score retains predictive power after netting GDP and pre-2022 baseline.

Method

Template
panel_fe_decomposition
Fixed effects
country, year
Clustering
country
Sample
14 countries · 2018 – 2027
Evidence type
causal

Primary specification: country-year panel. Outcome: log clean-energy MW added, log semiconductor capacity, log clean-manufacturing capex. Treatment: program_design_score (continuous, 0-9) interacted with post-2022 dummy. Test: does higher design-score predict more output per dollar? Effectiveness ratios: log spend / output. Compare ratios across countries, ranked. Decomposition asks: of the major programmes (US IRA, US CHIPS, EU CHIPS, EU CRMA, EU NZIA, JPN GX, KOR K-Chips, CHN composite), how do they rank on (a) MW deployed per $bn, (b) fab capacity per $bn, (c) clean-manufacturing capex per $bn? Net-global vs redistribution test: aggregate global clean-energy capacity additions 2022-2026 vs counterfactual extrapolation of 2018-2021 trend; share attributable to incremental capacity vs geographic redistribution from one country to another. Known limitations: (1) Program-design scoring is author-coded and contested; sensitivity to alternative codings reported. (2) Realised disbursement data lags announcements by 1-3 years; 2022-2024 data is mostly announcement, 2025-2027 will start showing realised. v1 reports announced + early realised; v1.1 updates with full realised by 2027. (3) Export-discipline channel is hard to measure cross-country in a comparable way; using qualitative score rather than continuous elasticity. (4) Counterfactual is "no major industrial policy 2022-2026" which is implausible; better counterfactual is 2018-2021 trend extrapolation (used here).

Data

VariableSourceTransform
log_program_announced_disbursement_usd
outcome
constructed:cross-country comparative tabulation: US Treasury / Commerce CHIPS PO / DOE LPO; EU Commission CRMA / NZIA tracker; METItier 5
log
log_program_realised_disbursement_usd
outcome
constructed:same sources but realised flow (obligated / paid) rather than announced. Manual-drop pending.tier 5
log
log_country_clean_energy_mw_added
outcome
irena:capacitytier 2
log
log_country_semiconductor_capacity_kwspm
outcome
constructed:SEMI World Fab Forecast country-level. Manual-drop pending.tier 5
log
log_country_clean_manufacturing_capex_usd
outcome
constructed:BloombergNEF / IEA Energy Technology Perspectives — country-level clean-manufacturing capex. Manual-drop pending.tier 5
log
spend_per_mw_clean_energy_usd
outcome
constructed:realised disbursement / MW added. Effectiveness ratio.tier 5
log
spend_per_kwspm_semiconductor_usd
outcome
constructed:realised disbursement / kWspm capacity added. Effectiveness ratio.tier 5
log
program_design_score
treatment
constructed:ordinal score per country-program on three dimensions: (a) target clarity (single-technology vs portfolio); (b) price-sitier 5
level
post_2022_dummy
treatment
constructed:indicator = 1 from country-specific program enactment date onwards. US: 2022-08; EU CHIPS: 2023-09; EU CRMA: 2024-04; EUtier 5
indicator
log_real_gdp
control
world_bank_wdi:NY.GDP.MKTP.KDtier 2
log
real_interest_rate_country
control
constructed:imf:IFS_PMP / fred for USA / boe / boj / ecb. Composite.tier 5
level
pre_2022_industrial_policy_baseline_spend_share_gdp
control
constructed:oecd or country-specific tabulation of industrial-policy spending pre-2022, normalised by GDP. Used to net out pre-existtier 5
level

● ready  · ● pending  · ● reconstruct-needed

Detailed result card

Result card — green_industrial_policy_global_chip_race_2022_2026

Verdict: INCONCLUSIVE_DATA_PENDING — insufficient observations after listwise deletion (20)

Pre-registration

  • Claim: The 2022-2026 wave of major-economy industrial-policy programmes — US IRA + CHIPS, EU Critical Raw Materials Act + Net-Zero Industry Act, EU Chips Act, Japan Green Transformation (GX, ¥150tn / ~$1tn announced), Korea K-Chips + Korean New Deal 2.0, China 14th Five-Year Plan + Made-in-China-2025-2.0 with semiconductors and clean energy as national-security frontier — represents the largest coordinated wave of industrial-policy spending in the post-1970s OECD record. The comparative-effectiveness question (which programmes delivered MW deployed / fab construction / clean-manufacturing capex per dollar spent) is testable. The hypothesis predicts that programmes which combine a clear technology target, a price-signal complement (carbon pricing or feed-in tariff), and an export-discipline mechanism (China WTO-era model) outperform pure-subsidy programmes (US IRA leans subsidy-heavy with weak export discipline; EU NZIA mixes subsidy with weak target-clarity). Net global clean-energy + semiconductor capacity rose materially but with substantial geographic redistribution rather than aggregate creation.
  • Falsification rule: Not supported if (a) program_design_score has zero or negative effect on effectiveness ratios at p<0.10, OR (b) the rank-order of programmes by effectiveness does not match author-predicted rank (Korea K-Chips + China composite outperform US IRA + EU NZIA), OR (c) global clean-energy / semiconductor capacity addition exceeds pre-2022 trend by less than 10% (would refute the "step-change" framing — programs would just be redistribution), OR (d) the country-level effectiveness measure is dominated by GDP / pre-existing industrial base rather than program-design (no policy lesson).
  • Falsification test: global_industrial_policy_design_panel

Estimate

  • Error: insufficient observations after listwise deletion (20)

Variables resolved

  • constructed: cross-country comparative tabulation: US Treasury / Commerce CHIPS PO / DOE LPO; EU Commission CRMA / NZIA tracker; METI GX bond + execution; Korean MOTIE / KIIT; China NDRC + national IC fund disbursements. Manual-drop pending under data/manual/derived/. → log_program_announced_disbursement_usd (outcome, publisher=constructed, n=140)
  • irena:capacity → log_country_clean_energy_mw_added (outcome, publisher=irena, n=5848)
  • constructed: ordinal score per country-program on three dimensions: (a) target clarity (single-technology vs portfolio); (b) price-signal complement (carbon price level, FIT design); (c) export discipline (Korean / Chinese model = high; US IRA = low). 0-9 ordinal scale, hand-coded with documented rubric. → program_design_score (treatment, publisher=constructed, n=140)
  • constructed: indicator = 1 from country-specific program enactment date onwards. US: 2022-08; EU CHIPS: 2023-09; EU CRMA: 2024-04; EU NZIA: 2024-06; JPN GX: 2023-02; KOR K-Chips: 2023-03; CHN: continuous. → post_2022_dummy (treatment, publisher=constructed, n=140)
  • world_bank_wdi:NY.GDP.MKTP.KD → log_real_gdp (controls, publisher=world_bank_wdi, n=12104)
  • constructed: imf:IFS_PMP / fred for USA / boe / boj / ecb. Composite. → real_interest_rate_country (controls, publisher=constructed, n=140)

Variables missing data

  • constructed: same sources but realised flow (obligated / paid) rather than announced. Manual-drop pending. (outcome, name=log_program_realised_disbursement_usd) — vintage not on disk
  • constructed: SEMI World Fab Forecast country-level. Manual-drop pending. (outcome, name=log_country_semiconductor_capacity_kwspm) — vintage not on disk
  • constructed: BloombergNEF / IEA Energy Technology Perspectives — country-level clean-manufacturing capex. Manual-drop pending. (outcome, name=log_country_clean_manufacturing_capex_usd) — vintage not on disk
  • constructed: realised disbursement / MW added. Effectiveness ratio. (outcome, name=spend_per_mw_clean_energy_usd) — vintage not on disk
  • constructed: realised disbursement / kWspm capacity added. Effectiveness ratio. (outcome, name=spend_per_kwspm_semiconductor_usd) — vintage not on disk
  • constructed: oecd or country-specific tabulation of industrial-policy spending pre-2022, normalised by GDP. Used to net out pre-existing trajectory. (controls, name=pre_2022_industrial_policy_baseline_spend_share_gdp) — vintage not on disk

Generated by scripts/run_panel_fe.py at 2026-06-29T17:54:39+00:00

Strongest opposing argument

Every hypothesis ships with its charitable opposing argument. The framework earns credibility by handling objections at their strongest, not weakest.

Notes

Data readiness: - IRENA capacity panel: ready - WDI: ready - SEMI / BloombergNEF / IEA: manual-drop pending - Country-level program disbursement tabulation: manual-drop pending — single biggest data dependency - Program design rubric: author-coded YAML to be added under hypotheses/derived/ Run when program-disbursement manual-drops + design-rubric YAML are populated. This is the umbrella hypothesis; individual-program decompositions live in ira_2022_clean_energy_investment_decomposition and chips_act_2022_semiconductor_capacity_2024_2027.

Authored framework. Read the transparency note.