Pre-registration
Canada's 1990-1996 unemployment-insurance reforms (1990 Bill C-21 entrance-requirement tightening, 1996 EI Act with hours-based eligibility and intensity rule) lowered the Canadian NAIRU by at least 1.5 pp by 2000 relative to a synthetic control of OECD peers, with the largest gains in high-frequent-claimant regions (Atlantic provinces).
Falsification criterion — what would disprove this
This hypothesis is considered falsified if:
SUPPORTED if synth-DiD gap on Canadian NAIRU < -1.5 pp by 2000 (estimated as Hodrick-Prescott-trended unemployment rate as proxy) AND the gap is concentrated in Atlantic provinces. REFUTED if NAIRU gap is insignificant OR distributed evenly across provinces (no region-heterogeneity consistent with frequent-claimant story).
formal test & threshold
test: Synth-DiD on Canadian unemployment rate (HP-trended) 1990- 2000 vs OECD donor pool with US-cycle control, plus province- level event study; placebo permutation at p<0.10.
Method
- Template
synth_did- Fixed effects
country, year- Clustering
country- Sample
- 7 countries · 1985 – 2002
- Evidence type
- associational
Data
| Variable | Source | Transform |
|---|---|---|
unemployment_rate outcome | world_bank_wdi:SL.UEM.TOTL.ZStier 2 | level |
nairu_imf_estimate outcome | imf:NAIRUtier 2 | level |
long_term_unemployment_share outcome | oecd:DSD_LFStier 2 | level |
canada_ui_ei_reform treatment | constructed:indicator for 1990-Q4 (Bill C-21) then 1996-Q3 (EI Act) intensificationtier 5 | indicator |
us_unemployment_rate control | fred:UNRATEtier 1 | level |
gdp_per_capita_real control | world_bank_wdi:NY.GDP.PCAP.KDtier 2 | log |
trade_openness control | world_bank_wdi:NE.TRD.GNFS.ZStier 2 | level |
● ready · ● pending · ● reconstruct-needed
Detailed result card
Result card - labour_reform_canada_1990s_ui_reform_nairu
Verdict: REFUTED - Canadian unemployment is above the synthetic control through 2000
JST Synthetic-Control Benchmark
- Donor weights: USA=0.917, NLD=0.083.
- Mean 1990-2000 unemployment gap: 3.67 pp.
- End-2000 unemployment gap: 2.94 pp.
- Placebo p-value: 0.143.
The registered regional heterogeneity leg remains unavailable locally, so this is a national unemployment benchmark.
Generated by engine/runs/labour_reform_canada_1990s_ui_reform_nairu/replication.py at 2026-05-17T18:10:06+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
Treatment dated 1990-Q4 then intensified 1996-Q3. The 1990s Canadian NAIRU debate (Fortin, Setterfield) makes Canada a testbed case for whether UI reform shifts the structural unemployment rate. Region-heterogeneity spec captures the Atlantic-Canada concentration of frequent-claimant behaviour.