IESET.
Hypotheses·regulatory·startup_density_frontier_prosperity

Among high-income economies 2000-2020, startup density (new high- growth firms per 1000 working-age population) predicts frontier prosperity — measured by real GDP per capita growth and productivity growth — more strongly than industrial-policy spending as a share of GDP.

The pre-registered claim is that, in a horse-race regression, the coefficient on startup density is larger in absolute t-statistic than the coefficient on industrial- policy spending, and that countries in the top quartile of startup density show at least 0.35 percentage points higher annual productivity growth than countries in the bottom quartile.

PARTIALengine/runs/startup_density_frontier_prosperity

PARTIAL — coef=-6.218e-06, p=0.386 (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=-6.218e-06, p=0.386 (above α=0.05); direction inconclusive

why it matters

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

how the test works

It compares 24 country or place units from 2000 to 2020, using a panel fe design, with fixed effects for country and year.

what was measured
What changed
  • Startup density
  • Venture capital invested pct income
What we checked
  • Real income per capita growth
  • Labour productivity growth
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/startup_density_frontier_prosperity
1007550250200020102020USAGBRDEUFRAITAESPNLD
illustrative sketch · run pending
No coefficients yet. When the model fires, this chart will show real_gdp_per_capita_growth across 24 sampled countries over 2000–2020.
The shapes above are stylised — none of the lines are real data.
Placeholder for startup_density_frontier_prosperity. Published chart will be generated from engine/runs/startup_density_frontier_prosperity/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.

Austrian Business Cycle Theory (ABCT)wins if SUPPORTEDpartial/no-callClassical Liberalwins if SUPPORTEDpartial/no-callOrdoliberalism (Freiburg School)wins if SUPPORTEDpartial/no-call

Pre-registration

pre-registered
first-spec commit 5ce4495 · 2026-05-02T19:11:20Z
run generated · 2026-06-29T17:54:34Z

Among high-income economies 2000-2020, startup density (new high- growth firms per 1000 working-age population) predicts frontier prosperity — measured by real GDP per capita growth and productivity growth — more strongly than industrial-policy spending as a share of GDP. The pre-registered claim is that, in a horse-race regression, the coefficient on startup density is larger in absolute t-statistic than the coefficient on industrial- policy spending, and that countries in the top quartile of startup density show at least 0.35 percentage points higher annual productivity growth than countries in the bottom quartile.

Falsification criterion — what would disprove this

set before the run · honoured after

This hypothesis is considered falsified if:

Not supported if (a) the startup-density coefficient is not positive and significant at p<0.05, OR (b) in the horse-race the industrial-policy-spending coefficient has a larger absolute t-statistic than startup density, OR (c) the top-quartile vs bottom-quartile productivity growth gap is below 0.20 pp/year. A developmentalist / industrial-policy reading wins cleanly if (b) holds — i.e. state-directed spending outperforms private startup formation in predicting prosperity.

formal test & threshold
test:      panel_fe_horserace_startup_density_vs_industrial_policy
threshold: panel_FE_beta(startup_density → productivity_growth) > 0 at p<0.05 AND |t_stat(startup_density)| > |t_stat(industrial_policy_spending)| in horse-race AND top_quartile_mean_gap >= 0.35 pp/yr

Method

Template
panel_fe
Fixed effects
country, year
Clustering
country
Sample
24 countries · 2000 – 2020
Evidence type
associational

Panel FE horse-race: both startup density and industrial-policy spending entered simultaneously. Primary test is on absolute t-statistic comparison. Quartile comparison as robustness. Robustness: drop US (high startup density outlier); use only European sub-sample; alternative industrial-policy measure from Criscuolo et al. OECD state-aid database.

Data

VariableSourceTransform
real_gdp_per_capita_growth
outcome
world_bank_wdi:NY.GDP.PCAP.KD.ZGtier 2
level
labour_productivity_growth
outcome
pwt:rgdpo_per_emptier 3
log_diff
startup_density
treatment
world_bank_wdi:IC.BUS.NREGtier 2
per_1000_working_age
venture_capital_invested_pct_gdp
treatment
oecd:venture_capital_investedtier 2
per_gdp
industrial_policy_spending_share_gdp
treatment
constructed:state_aid + subsidies + directed_credit_guaranteestier 5
level
log_initial_gdp_pc
control
world_bank_wdi:NY.GDP.PCAP.KDtier 2
log
rd_spending_share_gdp
control
world_bank_wdi:GB.XPD.RSDV.GD.ZStier 2
level
human_capital_index
control
pwt:hctier 3
level
trade_openness
control
world_bank_wdi:NE.TRD.GNFS.ZStier 2
level
product_market_regulation
control
oecd_pmr:pmr_compositetier 4
level

● ready  · ● pending  · ● reconstruct-needed

Detailed result card

Result card — startup_density_frontier_prosperity

Verdict: PARTIAL — coef=-6.218e-06, p=0.386 (above α=0.05); direction inconclusive

Pre-registration

  • Claim: Among high-income economies 2000-2020, startup density (new high- growth firms per 1000 working-age population) predicts frontier prosperity — measured by real GDP per capita growth and productivity growth — more strongly than industrial-policy spending as a share of GDP. The pre-registered claim is that, in a horse-race regression, the coefficient on startup density is larger in absolute t-statistic than the coefficient on industrial- policy spending, and that countries in the top quartile of startup density show at least 0.35 percentage points higher annual productivity growth than countries in the bottom quartile.
  • Falsification rule: Not supported if (a) the startup-density coefficient is not positive and significant at p<0.05, OR (b) in the horse-race the industrial-policy-spending coefficient has a larger absolute t-statistic than startup density, OR (c) the top-quartile vs bottom-quartile productivity growth gap is below 0.20 pp/year. A developmentalist / industrial-policy reading wins cleanly if (b) holds — i.e. state-directed spending outperforms private startup formation in predicting prosperity.
  • Falsification test: panel_fe_horserace_startup_density_vs_industrial_policy

Estimate

  • Method: linearmodels.PanelOLS
  • Coefficient (treatment): -6.218e-06
  • Std error: 7.162e-06
  • p-value: 0.386
  • Observations: 224, countries: 19
  • Within R²: 0.192
  • Fixed effects: entity=True, time=True
  • Clustering: country

Variables resolved

  • world_bank_wdi:NY.GDP.PCAP.KD.ZG → real_gdp_per_capita_growth (outcome, publisher=world_bank_wdi, n=13897)
  • pwt:rgdpo_per_emp → labour_productivity_growth (outcome, publisher=pwt, n=9529)
  • world_bank_wdi:IC.BUS.NREG → startup_density (treatment, publisher=world_bank_wdi, n=2370)
  • world_bank_wdi:NY.GDP.PCAP.KD → log_initial_gdp_pc (controls, publisher=world_bank_wdi, n=12104)
  • world_bank_wdi:GB.XPD.RSDV.GD.ZS → rd_spending_share_gdp (controls, publisher=world_bank_wdi, n=3140)
  • 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)
  • oecd_pmr:pmr_composite → product_market_regulation (controls, publisher=oecd_pmr, n=105)

Variables missing data

  • oecd:venture_capital_invested (treatment, name=venture_capital_invested_pct_gdp) — vintage not on disk
  • constructed: state_aid + subsidies + directed_credit_guarantees (treatment, name=industrial_policy_spending_share_gdp) — vintage not on disk

Generated by scripts/run_panel_fe.py at 2026-06-29T17:54:34+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

OECD Entrepreneurship at a Glance and the Criscuolo-OECD high-growth firm database are preferred sources. Venture-capital data from OECD Entrepreneurship Finance database. Industrial- policy spending is the weakest-measured variable; EU State Aid Scoreboard provides the best coverage for European countries.

Authored framework. Read the transparency note.