Pre-registration

pre-registered

first-spec commit c48dfd6 · 2026-05-13T10:43:00Z

Countries with larger PMR declines from 2018 to 2023 saw faster PDB labour-productivity growth in the following window.

Falsification criterion — what would disprove this

set before the run · honoured after

This hypothesis is considered falsified if:

Estimate `lp_growth_2018_2024 ~ pmr_decline_2018_2023` on the landed OECD PDB vintage; expected sign: positive. SUPPORTED requires the pre-registered sign at p<0.10 or the named dominance criterion; otherwise the result is partial or weak.

formal test & threshold

test:      oecd_pdb_batch03_oecd_pdb_market_reform_productivity_compounder
threshold: p<0.10 with pre-registered sign, except dominance tests compare the named contribution magnitudes

Method

Template: descriptive
Fixed effects
Clustering: country
Sample: 31 countries · 2018 – 2024
Evidence type: associational

Local first-pass throughput screen; upgrade robustness before scoreboard conversion.

Data

Variable	Source	Transform
lp_growth_2018_2024 outcome	oecd:OECD.SDD.TPStier 2	growth_or_level_as_named
pmr_decline_2018_2023 treatment	oecd_pmr:OECD.ECO.GCRDtier 4	lagged_level_or_change_as_named

● ready · ● pending · ● reconstruct-needed

Detailed result card

PMR reform productivity compounder

Verdict: REFUTED_OR_WEAK

Claim: Countries with larger PMR declines from 2018 to 2023 saw faster PDB labour-productivity growth in the following window.

Test: lp_growth_2018_2024 ~ pmr_decline_2018_2023

Sample: n=31, countries=31, years=None–None.

Key coefficients

pmr_decline_2018_2023: beta=-1.827, p=0.198, 90/95 CI approx [-4.612, 0.9568]

Data: oecd:OECD.SDD.TPS,DSD_PDB@DF_PDB,2.0 from data/vintages/oecd/DSD_PDB@2026-05-12T133454Z.parquet.

Strongest opposing argument

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