IESET.
Hypotheses·trade·trade_openness_long_run_income_convergence

Across a broad panel of countries 1960-2019, higher trade openness predicts faster long-run convergence of real GDP per capita toward the global frontier (the United States) than industrial-policy intensity does.

Convergence is measured by β-convergence — the coefficient on initial log GDP per capita becomes less negative (or more positive in a conditional-convergence specification) when interacted with trade openness — and by σ-convergence, the declining cross-sectional standard deviation of log GDP per capita over 20-year windows. The hypothesis tests the Smithian-Ricardian claim that openness is a more reliable convergence engine than dirigiste industrial targeting.

PARTIALengine/runs/trade_openness_long_run_income_convergence

PARTIAL — coef=+6.729e-18, p=0.00881; effect magnitude effectively zero

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

policy briefMixed or noisy

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

The evidence is suggestive but not decisive. coef=+6.729e-18, p=0.00881; effect magnitude effectively zero

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 62 country or place units from 1960 to 2019, using a panel fe design, with fixed effects for country and year.

what was measured
What changed
  • Trade openness pct income
  • Industrial policy intensity index
Possible pathway
  • Import penetration ratio
  • Export diversification index
What we checked
  • Real income per capita
  • Convergence speed beta
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

6 input datasets, 4 unresolved missing series, provenance status: incomplete.

Results

engine/runs/trade_openness_long_run_income_convergence
1007550250196019902019ARGAUSAUTBELBGDBRACAN
illustrative sketch · run pending
No coefficients yet. When the model fires, this chart will show real_gdp_per_capita across 62 sampled countries over 1960–2019.
The shapes above are stylised — none of the lines are real data.
Placeholder for trade_openness_long_run_income_convergence. Published chart will be generated from engine/runs/trade_openness_long_run_income_convergence/chart_data.json.

Who has skin in the game — schools predicting on this

3 schools list this hypothesis as a test of their position. The chips below are school-level scoreboard outcomes, not a second hypothesis verdict.

hypothesis verdict vs scoreboard outcome

The banner verdict judges this hypothesis as written. The scoreboard asks whether each school's polarity-corrected prediction was right. Raw status is not a school win: SUPPORTED supports schools that needed SUPPORTED, but refutes schools that needed REFUTED.

Chicago School Monetarismwins if SUPPORTEDpartial/no-callClassical Liberalwins if SUPPORTEDpartial/no-callEmpirical Pragmatist (the framework's house position)wins if SUPPORTEDpartial/no-call

Pre-registration

pre-registered
first-spec commit 5ce4495 · 2026-05-02T19:11:20Z
run generated · 2026-06-29T17:54:43Z

Across a broad panel of countries 1960-2019, higher trade openness predicts faster long-run convergence of real GDP per capita toward the global frontier (the United States) than industrial-policy intensity does. Convergence is measured by β-convergence — the coefficient on initial log GDP per capita becomes less negative (or more positive in a conditional-convergence specification) when interacted with trade openness — and by σ-convergence, the declining cross-sectional standard deviation of log GDP per capita over 20-year windows. The hypothesis tests the Smithian-Ricardian claim that openness is a more reliable convergence engine than dirigiste industrial targeting.

Falsification criterion — what would disprove this

set before the run · honoured after

This hypothesis is considered falsified if:

SUPPORTED if the interaction coefficient β3 (initial_log_gdp × openness) is positive and significant at p<0.10, while the corresponding interaction for industrial-policy intensity is either negative, insignificant, or materially smaller in magnitude. PARTIAL if both interactions are positive and significant but openness has the larger coefficient. REFUTED if the openness interaction is negative and significant at p<0.10, or if the industrial-policy interaction is positive and significantly larger.

formal test & threshold
test:      panel_fe_convergence_interaction_trade_vs_industrial_policy
threshold: β3_trade_openness > 0 at p<=0.10  AND |β3_trade_openness| > |β3_industrial_policy|  AND σ-convergence (20yr window SD of log GDP pc) declines more in high-openness decades than in high-industrial-policy decades.

Method

Template
panel_fe
Fixed effects
country, year
Clustering
country
Sample
62 countries · 1960 – 2019
Evidence type
associational

Two-way fixed-effects panel with Driscoll-Kraay standard errors. Primary specification: growth of log GDP pc = β0 + β1*initial_log_gdp + β2*openness + β3*(initial_log_gdp × openness) + controls + country FE + year FE. The interaction term β3 is the convergence test: if β3 > 0, openness accelerates convergence. A parallel specification replaces openness with industrial-policy intensity. Robustness: 20-year non-overlapping windows to avoid serial correlation; entrepôt outliers dropped or downweighted.

Data

VariableSourceTransform
real_gdp_per_capita
outcome
pwt:rgdpotier 3
log_per_capita
convergence_speed_beta
outcome
constructed:20-year rolling regression of log GDP pc growth on initial log GDP pctier 5
rolling_window_estimate
trade_openness_pct_gdp
treatment
world_bank_wdi:NE.TRD.GNFS.ZStier 2
level
industrial_policy_intensity_index
treatment
constructed:composite of state ownership share, directed credit, and sectoral subsidy intensity (CRDF/ITC + Fraser EFW subcomponentstier 5
z_score_composite
import_penetration_ratio
channel
wits:import_valuetier 2
ratio_to_gdp
export_diversification_index
channel
un_comtrade:export_product_diversitytier 2
theil_index
initial_log_gdp_per_capita
control
pwt:rgdpotier 3
log_initial_5yr_window
human_capital_index
control
pwt:hctier 3
level
rule_of_law
control
wgi:RL.ESTtier 4
level
population_growth
control
world_bank_wdi:SP.POP.GROWtier 2
level
investment_share_gdp
control
world_bank_wdi:NE.GDI.TOTL.ZStier 2
level

● ready  · ● pending  · ● reconstruct-needed

Detailed result card

Result card — trade_openness_long_run_income_convergence

Verdict: PARTIAL — coef=+6.729e-18, p=0.00881; effect magnitude effectively zero

Pre-registration

  • Claim: Across a broad panel of countries 1960-2019, higher trade openness predicts faster long-run convergence of real GDP per capita toward the global frontier (the United States) than industrial-policy intensity does. Convergence is measured by β-convergence — the coefficient on initial log GDP per capita becomes less negative (or more positive in a conditional-convergence specification) when interacted with trade openness — and by σ-convergence, the declining cross-sectional standard deviation of log GDP per capita over 20-year windows. The hypothesis tests the Smithian-Ricardian claim that openness is a more reliable convergence engine than dirigiste industrial targeting.
  • Falsification rule: SUPPORTED if the interaction coefficient β3 (initial_log_gdp × openness) is positive and significant at p<0.10, while the corresponding interaction for industrial-policy intensity is either negative, insignificant, or materially smaller in magnitude. PARTIAL if both interactions are positive and significant but openness has the larger coefficient. REFUTED if the openness interaction is negative and significant at p<0.10, or if the industrial-policy interaction is positive and significantly larger.
  • Falsification test: panel_fe_convergence_interaction_trade_vs_industrial_policy

Estimate

  • Method: linearmodels.PanelOLS
  • Coefficient (treatment): +6.729e-18
  • Std error: 2.565e-18
  • p-value: 0.00881
  • Observations: 1281, countries: 61
  • Within R²: 1
  • Fixed effects: entity=True, time=True
  • Clustering: country

Variables resolved

  • pwt:rgdpo → real_gdp_per_capita (outcome, publisher=pwt, n=10399)
  • world_bank_wdi:NE.TRD.GNFS.ZS → trade_openness_pct_gdp (treatment, publisher=world_bank_wdi, n=10714)
  • world_bank_wits:import_value → import_penetration_ratio (decomposition_channels, publisher=wits, n=14395)
  • pwt:rgdpo → initial_log_gdp_per_capita (controls, publisher=pwt, n=10399)
  • pwt:hc → human_capital_index (controls, publisher=pwt, n=8637)
  • wgi:RL.EST → rule_of_law (controls, publisher=wgi, n=5296)
  • world_bank_wdi:SP.POP.GROW → population_growth (controls, publisher=world_bank_wdi, n=16672)
  • world_bank_wdi:NE.GDI.TOTL.ZS → investment_share_gdp (controls, publisher=world_bank_wdi, n=10428)

Variables missing data

  • constructed: 20-year rolling regression of log GDP pc growth on initial log GDP pc (outcome, name=convergence_speed_beta) — vintage not on disk
  • constructed: composite of state ownership share, directed credit, and sectoral subsidy intensity (CRDF/ITC + Fraser EFW subcomponents) (treatment, name=industrial_policy_intensity_index) — vintage not on disk
  • un_comtrade:export_product_diversity (decomposition_channels, name=export_diversification_index) — vintage not on disk

Generated by scripts/run_panel_fe.py at 2026-06-29T17:54:43+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: - PWT rgdpo, hc (ready) - WDI NE.TRD.GNFS.ZS, NE.GDI.TOTL.ZS, SP.POP.GROW (ready) - WGI RL.EST (ready from 1996; pre-1996 backfill flagged) - Industrial policy intensity composite (pending — Fraser EFW used as v1 proxy) - UN Comtrade / World Bank WITS export diversity (ready)

Authored framework. Read the transparency note.