IESET.
Hypotheses·institutional quality·corruption_state_allocation_growth_interaction

Across a broad panel of economies 1980-2020, state allocation of resources — measured by government consumption share, state- enterprise share of output, and public-investment share — has negative long-run effects on total-factor-productivity growth when corruption is high (WGI Control of Corruption below median), but neutral or positive effects when corruption is low (above median).

The pre-registered claim is that the interaction term between state-allocation intensity and corruption is negative and significant at p<0.05, and that the marginal effect of state allocation is negative and significant in the high-corruption sub-sample but not significantly negative in the low-corruption sub-sample.

PARTIALengine/runs/corruption_state_allocation_growth_interaction

PARTIAL — coef=+0.001013, p=0.729 (above α=0.05); 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

Over a long period, do more market-oriented institutions translate into higher income or productivity, once the comparison looks beyond a single success story?

plain answer

The evidence is suggestive but not decisive. coef=+0.001013, p=0.729 (above α=0.05); direction inconclusive

why it matters

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

how the test works

It compares 55 country or place units from 1980 to 2020, using a panel fe design, with fixed effects for country and year.

what was measured
What changed
  • State allocation intensity
  • Government consumption share
What we checked
  • Productivity growth 10yr
  • Real income per capita growth
  • Investment efficiency
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

9 input datasets, 2 unresolved missing series, provenance status: incomplete.

Results

engine/runs/corruption_state_allocation_growth_interaction
1007550250198020002020USAGBRDEUFRAITAESPNLD
illustrative sketch · run pending
No coefficients yet. When the model fires, this chart will show tfp_growth_10yr across 55 sampled countries over 1980–2020.
The shapes above are stylised — none of the lines are real data.
Placeholder for corruption_state_allocation_growth_interaction. Published chart will be generated from engine/runs/corruption_state_allocation_growth_interaction/chart_data.json.

Who has skin in the game — schools predicting on this

9 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.

Austrian Business Cycle Theory (ABCT)wins if MIXEDpartial/no-callChicago School Monetarismwins if MIXEDpartial/no-callClassical Liberalwins if MIXEDpartial/no-callStructuralist / Developmentalistwins if SUPPORTEDpartial/no-callEmpirical Pragmatist (the framework's house position)wins if SUPPORTEDpartial/no-callNew Institutional Economicswins if SUPPORTEDpartial/no-callNew Keynesian (mainstream synthesis)wins if SUPPORTEDpartial/no-callOrdoliberalism (Freiburg School)wins if SUPPORTEDpartial/no-callSocial Democraticwins if SUPPORTEDpartial/no-call

Pre-registration

pre-registered
first-spec commit 5ce4495 · 2026-05-02T19:11:20Z
run generated · 2026-06-29T17:53:17Z

Across a broad panel of economies 1980-2020, state allocation of resources — measured by government consumption share, state- enterprise share of output, and public-investment share — has negative long-run effects on total-factor-productivity growth when corruption is high (WGI Control of Corruption below median), but neutral or positive effects when corruption is low (above median). The pre-registered claim is that the interaction term between state-allocation intensity and corruption is negative and significant at p<0.05, and that the marginal effect of state allocation is negative and significant in the high-corruption sub-sample but not significantly negative in the low-corruption sub-sample.

Falsification criterion — what would disprove this

set before the run · honoured after

This hypothesis is considered falsified if:

Not supported if (a) the interaction coefficient is not negative and significant at p<0.05, OR (b) the marginal effect of state allocation is negative and significant even in the low-corruption sub-sample (no conditioning), OR (c) the marginal effect is positive and significant in the high- corruption sub-sample (reverse of predicted). A "state allocation is always bad" reading wins if (b) holds; a "corruption doesn't matter" reading wins if the interaction is insignificant and the main effect of state allocation dominates.

formal test & threshold
test:      panel_fe_state_allocation_corruption_interaction_on_tfp
threshold: panel_FE_beta(state_allocation × corruption) < 0 at p<0.05 AND marginal_effect_state_allocation_at_high_corruption < 0 at p<0.05 AND marginal_effect_state_allocation_at_low_corruption NOT < 0 at p<0.10

Method

Template
panel_fe
Fixed effects
country, year
Clustering
country
Sample
55 countries · 1980 – 2020
Evidence type
associational

Panel FE with interaction: state_allocation × corruption. Primary test: sign and significance of interaction. Sub- sample analysis: high-corruption vs low-corruption countries. Robustness: use Fraser EFW government-size instead of constructed state-allocation index; use V-Dem corruption index instead of WGI; exclude oil exporters (where state allocation is rent-driven, not policy-driven).

Data

VariableSourceTransform
tfp_growth_10yr
outcome
pwt:rtfpnatier 3
log_diff_10yr
real_gdp_per_capita_growth
outcome
world_bank_wdi:NY.GDP.PCAP.KD.ZGtier 2
level
investment_efficiency
outcome
constructed:gdp_growth_per_unit_of_investmenttier 5
level
state_allocation_intensity
treatment
constructed:0.4×gov_consumption_gdp + 0.3×soe_output_share + 0.3×public_investment_gdptier 5
level
government_consumption_share
treatment
world_bank_wdi:NE.CON.GOVT.ZStier 2
level
control_of_corruption
treatment
wgi:CC.ESTtier 4
level
high_corruption_dummy
treatment
constructed:indicator = 1 if wgi:CC.EST < mediantier 5
indicator
log_initial_gdp_pc
control
world_bank_wdi:NY.GDP.PCAP.KDtier 2
log
human_capital_index
control
pwt:hctier 3
level
trade_openness
control
world_bank_wdi:NE.TRD.GNFS.ZStier 2
level
rule_of_law
control
wgi:RL.ESTtier 4
level
financial_development
control
world_bank_wdi:FS.AST.PRVT.GD.ZStier 2
level

● ready  · ● pending  · ● reconstruct-needed

Detailed result card

Result card — corruption_state_allocation_growth_interaction

Verdict: PARTIAL — coef=+0.001013, p=0.729 (above α=0.05); direction inconclusive

Pre-registration

  • Claim: Across a broad panel of economies 1980-2020, state allocation of resources — measured by government consumption share, state- enterprise share of output, and public-investment share — has negative long-run effects on total-factor-productivity growth when corruption is high (WGI Control of Corruption below median), but neutral or positive effects when corruption is low (above median). The pre-registered claim is that the interaction term between state-allocation intensity and corruption is negative and significant at p<0.05, and that the marginal effect of state allocation is negative and significant in the high-corruption sub-sample but not significantly negative in the low-corruption sub-sample.
  • Falsification rule: Not supported if (a) the interaction coefficient is not negative and significant at p<0.05, OR (b) the marginal effect of state allocation is negative and significant even in the low-corruption sub-sample (no conditioning), OR (c) the marginal effect is positive and significant in the high- corruption sub-sample (reverse of predicted). A "state allocation is always bad" reading wins if (b) holds; a "corruption doesn't matter" reading wins if the interaction is insignificant and the main effect of state allocation dominates.
  • Falsification test: panel_fe_state_allocation_corruption_interaction_on_tfp

Estimate

  • Method: linearmodels.PanelOLS
  • Coefficient (treatment): +0.001013
  • Std error: 0.002919
  • p-value: 0.729
  • Observations: 775, countries: 42
  • Within R²: 0.531
  • Fixed effects: entity=True, time=True
  • Clustering: country

Variables resolved

  • pwt:rtfpna → tfp_growth_10yr (outcome, publisher=pwt, n=6407)
  • world_bank_wdi:NY.GDP.PCAP.KD.ZG → real_gdp_per_capita_growth (outcome, publisher=world_bank_wdi, n=13897)
  • world_bank_wdi:NE.CON.GOVT.ZS → government_consumption_share (treatment, publisher=world_bank_wdi, n=9133)
  • wgi:CC.EST → control_of_corruption (treatment, publisher=wgi, n=5201)
  • constructed: indicator = 1 if wgi:CC.EST < median → high_corruption_dummy (treatment, publisher=constructed, n=2255)
  • world_bank_wdi:NY.GDP.PCAP.KD → log_initial_gdp_pc (controls, publisher=world_bank_wdi, n=12104)
  • pwt:hc → human_capital_index (controls, publisher=pwt, n=8637)
  • world_bank_wdi:NE.TRD.GNFS.ZS → trade_openness (controls, publisher=world_bank_wdi, n=10714)
  • wgi:RL.EST → rule_of_law (controls, publisher=wgi, n=5296)
  • world_bank_wdi:FS.AST.PRVT.GD.ZS → financial_development (controls, publisher=world_bank_wdi, n=9562)

Variables missing data

  • constructed: gdp_growth_per_unit_of_investment (outcome, name=investment_efficiency) — vintage not on disk
  • constructed: 0.4×gov_consumption_gdp + 0.3×soe_output_share + 0.3×public_investment_gdp (treatment, name=state_allocation_intensity) — vintage not on disk

Generated by scripts/run_panel_fe.py at 2026-06-29T17:53:17+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

SOE output-share data is patchy outside OECD and major emerging markets. Government consumption and public investment from WDI are more complete but conflate allocation with other functions. The composite index is flagged as a measurement limitation.

Authored framework. Read the transparency note.