IESET.
Hypotheses·fiscal·ira_2022_clean_energy_investment_decomposition

The Inflation Reduction Act (IRA, signed August 2022) produced a measurable step-change in US clean-energy investment, manufacturing reshoring, and fiscal cost over 2022-2026 relative to a pre-IRA trajectory and to non-US comparators (EU, China, Japan, Korea).

Specifically, cumulative tax-credit outlays exceeded the original CBO estimate of ~$369bn (10-year), utility- scale clean-energy MW deployed accelerated above the 2018-2021 trend, announced battery + EV + solar manufacturing capex relocated to the US at a rate that materially raised the US share of new global clean-energy manufacturing investment, and clean-energy-related job postings rose. The decomposition tests which of these channels (deployment, manufacturing, jobs, fiscal cost) materialised at the magnitudes claimed by Treasury and by the IRA's defenders, and which fell short.

PARTIALengine/runs/ira_2022_clean_energy_investment_decomposition

PARTIAL — coef=+1.688e+04, p=0.886 (above α=0.1); direction inconclusive

confidence cueThe result is useful, but not decisive. Treat it as a clue, not a settled conclusion.

policy briefMixed or noisy

In ordinary language

In plain terms, this asks whether us post ira dummy is actually linked to better or worse log us cumulative clean energy tax credit outlay usd from 2015 to 2026.

plain answer

The evidence is suggestive but not decisive. coef=+1.688e+04, p=0.886 (above α=0.1); direction inconclusive

why it matters

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

how the test works

It compares 12 country or place units from 2015 to 2026, using a panel fe decomposition design, with fixed effects for country and year.

what was measured
What changed
  • Us post ira dummy
  • Us post ira eligible subsector interaction
What we checked
  • Log us cumulative clean energy tax credit outlay usd
  • Us utility scale clean energy mw added
  • Log us clean manufacturing announced capex usd
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/ira_2022_clean_energy_investment_decomposition
1007550250201520212026USADEUFRAGBRJPNKORCHN
illustrative sketch · run pending
No coefficients yet. When the model fires, this chart will show log_us_cumulative_clean_energy_tax_credit_outlay_usd across 12 sampled countries over 2015–2026.
The shapes above are stylised — none of the lines are real data.
Placeholder for ira_2022_clean_energy_investment_decomposition. Published chart will be generated from engine/runs/ira_2022_clean_energy_investment_decomposition/chart_data.json.

Pre-registration

pre-registered
first-spec commit 098ce96 · 2026-04-30T12:57:33Z
run generated · 2026-06-29T17:52:23Z

The Inflation Reduction Act (IRA, signed August 2022) produced a measurable step-change in US clean-energy investment, manufacturing reshoring, and fiscal cost over 2022-2026 relative to a pre-IRA trajectory and to non-US comparators (EU, China, Japan, Korea). Specifically, cumulative tax-credit outlays exceeded the original CBO estimate of ~$369bn (10-year), utility- scale clean-energy MW deployed accelerated above the 2018-2021 trend, announced battery + EV + solar manufacturing capex relocated to the US at a rate that materially raised the US share of new global clean-energy manufacturing investment, and clean-energy-related job postings rose. The decomposition tests which of these channels (deployment, manufacturing, jobs, fiscal cost) materialised at the magnitudes claimed by Treasury and by the IRA's defenders, and which fell short.

Falsification criterion — what would disprove this

set before the run · honoured after

This hypothesis is considered falsified if:

Not supported if (a) the USA × post-2022 effect on clean-energy MW added is within the 95% CI of the pre-IRA trend extrapolation (no deployment acceleration), OR (b) US share of global clean-manufacturing capex does not rise above its 2018-2021 mean by at least 5 percentage points by end-2025, OR (c) clean-energy employment is flat or declines relative to total manufacturing employment, OR (d) total tax-credit outlays fall below the original CBO 10-year estimate (suggesting take-up failure). If only the fiscal-cost channel materialises (deficit rises) without deployment / capex / jobs response, classify as "fiscal cost without industrial response."

formal test & threshold
test:      ira_decomposition_panel
threshold: USA × post-2022 coefficient on clean-energy MW > 1.5x pre-IRA trend AND USA share of global clean-manufacturing capex rises ≥ 5pp by 2025 AND clean-energy employment growth > total manufacturing employment growth AND realised tax-credit outlays within ±50% of original CBO estimate (≥3 of 4 outcomes hit thresholds for support; 2 = mixed; ≤1 = refuted)

Method

Template
panel_fe_decomposition
Fixed effects
country, year
Clustering
country
Sample
12 countries · 2015 – 2026
Evidence type
causal

Primary specification: country-year panel with USA × post-2022 interaction on each outcome (deployment, manufacturing capex, jobs, fiscal cost, global-share). Compare effect sizes against pre-IRA trend extrapolation (2018-2021 fitted line) and against synthetic-USA (donor pool: DEU, GBR, JPN, KOR, AUS). Decomposition asks: of the six outcomes, how many show a >1-sigma acceleration above pre-IRA trend in 2023-2026? Which fall short of the Treasury / Goldman / Credit Suisse 2022-2023 projections? Known limitations: (1) IRA tax-credit costs are uncapped; "actual cost" is a moving estimate with CBO reestimates rising substantially through 2024. (2) Announced capex ≠ realised capex; manufacturing-reshoring claims depend on plant operations, not announcement counts. (3) Trump administration post-Jan-2025 actions may modify or partially repeal IRA — v1 windows 2022-2026, v1.1 updates if statutory landscape changes. (4) Counterfactual is hard: 2022-2024 also saw EU CRMA, Japan GX, Korea IRA-equivalent — global clean-energy capex would have risen anyway.

Data

VariableSourceTransform
log_us_cumulative_clean_energy_tax_credit_outlay_usd
outcome
constructed:US Treasury / OMB outlay tabulations cross-checked against CBO 2022/2023/2024 reestimates. fred:FYONET (deficit baselinetier 5
log
us_utility_scale_clean_energy_mw_added
outcome
irena:capacitytier 2
level
log_us_clean_manufacturing_announced_capex_usd
outcome
constructed:BEA NIPA private fixed investment in clean-energy manufacturing + DOE LPO loan-guarantee tracker + BloombergNEF announcetier 5
log
us_clean_energy_employment
outcome
bls:CES1021100001tier 1
log
us_federal_deficit_share_gdp
outcome
fred:GFDEGDQ188Stier 1
level
comparator_clean_manufacturing_capex_share_global
outcome
constructed:BloombergNEF / IEA Energy Technology Perspectives — share of global clean-energy manufacturing capex by location (USA, Etier 5
level
us_post_ira_dummy
treatment
constructed:indicator = 1 for USA from 2022-08 (IRA enactment) onwards; 0 otherwise.tier 5
indicator
us_post_ira_eligible_subsector_interaction
treatment
constructed:interaction of post-IRA dummy with subsector eligibility (battery / EV / solar / wind eligible = 1; non-eligible manufactier 5
indicator
log_real_gdp
control
world_bank_wdi:NY.GDP.MKTP.KDtier 2
log
real_interest_rate
control
fred:DFII10tier 1
level
brent_oil_log
control
imf_pcps:POILBREtier 1
log
china_clean_manufacturing_capex_share
control
constructed:same BloombergNEF / IEA tracker, China share. Used to control for global clean-energy capex cycle independent of IRA.tier 5
level

● ready  · ● pending  · ● reconstruct-needed

Detailed result card

Result card — ira_2022_clean_energy_investment_decomposition

Verdict: PARTIAL — coef=+1.688e+04, p=0.886 (above α=0.1); direction inconclusive

Pre-registration

  • Claim: The Inflation Reduction Act (IRA, signed August 2022) produced a measurable step-change in US clean-energy investment, manufacturing reshoring, and fiscal cost over 2022-2026 relative to a pre-IRA trajectory and to non-US comparators (EU, China, Japan, Korea). Specifically, cumulative tax-credit outlays exceeded the original CBO estimate of ~$369bn (10-year), utility- scale clean-energy MW deployed accelerated above the 2018-2021 trend, announced battery + EV + solar manufacturing capex relocated to the US at a rate that materially raised the US share of new global clean-energy manufacturing investment, and clean-energy-related job postings rose. The decomposition tests which of these channels (deployment, manufacturing, jobs, fiscal cost) materialised at the magnitudes claimed by Treasury and by the IRA's defenders, and which fell short.
  • Falsification rule: Not supported if (a) the USA × post-2022 effect on clean-energy MW added is within the 95% CI of the pre-IRA trend extrapolation (no deployment acceleration), OR (b) US share of global clean-manufacturing capex does not rise above its 2018-2021 mean by at least 5 percentage points by end-2025, OR (c) clean-energy employment is flat or declines relative to total manufacturing employment, OR (d) total tax-credit outlays fall below the original CBO 10-year estimate (suggesting take-up failure). If only the fiscal-cost channel materialises (deficit rises) without deployment / capex / jobs response, classify as "fiscal cost without industrial response."
  • Falsification test: ira_decomposition_panel

Estimate

  • Method: linearmodels.PanelOLS
  • Coefficient (treatment): +1.688e+04
  • Std error: 1.171e+05
  • p-value: 0.886
  • Observations: 132, countries: 12
  • Within R²: 0.00219
  • Fixed effects: entity=True, time=True
  • Clustering: country

Variables resolved

  • irena:capacity → us_utility_scale_clean_energy_mw_added (outcome, publisher=irena, n=5848)
  • fred:GFDEGDQ188S → us_federal_deficit_share_gdp (outcome, publisher=fred, n=60)
  • constructed: BloombergNEF / IEA Energy Technology Perspectives — share of global clean-energy manufacturing capex by location (USA, EU, CHN). Manual-drop under data/manual/iea/. → comparator_clean_manufacturing_capex_share_global (outcome, publisher=constructed, n=144)
  • constructed: indicator = 1 for USA from 2022-08 (IRA enactment) onwards; 0 otherwise. → us_post_ira_dummy (treatment, publisher=constructed, n=144)
  • world_bank_wdi:NY.GDP.MKTP.KD → log_real_gdp (controls, publisher=world_bank_wdi, n=12104)
  • fred:DFII10 → real_interest_rate (controls, publisher=fred, n=288)
  • imf_pcps:POILBRE → brent_oil_log (controls, publisher=imf_pcps, n=444)

Variables missing data

  • constructed: US Treasury / OMB outlay tabulations cross-checked against CBO 2022/2023/2024 reestimates. fred:FYONET (deficit baseline) + Treasury Green Book line items. Manual-drop pending. (outcome, name=log_us_cumulative_clean_energy_tax_credit_outlay_usd) — vintage not on disk
  • constructed: BEA NIPA private fixed investment in clean-energy manufacturing + DOE LPO loan-guarantee tracker + BloombergNEF announced-capex tracker. Manual-drop under data/manual/derived/. Cross-checks: fred:PNFI (private nonresidential fixed investment) and fred:INDPRO (manufacturing IP). (outcome, name=log_us_clean_manufacturing_announced_capex_usd) — vintage not on disk
  • bls:CES1021100001 (outcome, name=us_clean_energy_employment) — vintage not on disk
  • constructed: interaction of post-IRA dummy with subsector eligibility (battery / EV / solar / wind eligible = 1; non-eligible manufacturing = 0). Identifies whether eligible subsectors accelerate relative to non-eligible. (treatment, name=us_post_ira_eligible_subsector_interaction) — vintage not on disk
  • constructed: same BloombergNEF / IEA tracker, China share. Used to control for global clean-energy capex cycle independent of IRA. (controls, name=china_clean_manufacturing_capex_share) — vintage not on disk

Generated by scripts/run_panel_fe.py at 2026-06-29T17:52:23+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 (data.fetchers.irena) - BLS CES: ready (data.fetchers.bls) - FRED deficit + interest: ready - WDI GDP: ready - IMF PCPS oil: ready - BEA / DOE manufacturing-capex tracker: manual-drop pending under data/manual/derived/ - BloombergNEF / IEA share-of-global tracker: manual-drop pending under data/manual/iea/ - Treasury IRA outlay tabulation: manual-drop pending; cross-check against CBO score Run when manual-drop manufacturing-capex tracker + Treasury outlay table are populated.

Authored framework. Read the transparency note.