IESET.
Hypotheses·labour·natural_rate_hypothesis_long_run_phillips_vertical

Cross-country Phillips-curve data post-1970 shows no stable unemployment-inflation tradeoff in the long run, consistent with Friedman's natural-rate hypothesis.

REFUTEDengine/runs/natural_rate_hypothesis_long_run_phillips_vertical

REFUTED - cpi_inflation_yoy, unemployment_rate: no-cointegration share=10.0%, cointegration share=90.0%, tested countries=20

confidence cueThis test cuts against the claim as written or misses its pre-declared threshold.

policy briefNeeds review

In ordinary language

In plain terms, this asks whether ten year inflation expectations is actually linked to better or worse cpi inflation yoy from 1970 to 2023.

plain answer

The data did not support the prediction. cpi_inflation_yoy, unemployment_rate: no-cointegration share=10.0%, cointegration share=90.0%, tested countries=20

why it matters

Labor-market rules often help some workers while risking job loss or slower hiring for others. This test looks for that tradeoff in observable employment or unemployment data.

how the test works

It compares 20 country or place units from 1970 to 2023, using a cointegration vecm design.

what was measured
What changed
  • Ten year inflation expectations
  • Nairu estimate
What we checked
  • Cpi inflation yoy
  • Unemployment rate
  • Rich-country cpi all items
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

0 input datasets, 0 unresolved missing series, provenance status: no input vintages recorded.

Results

engine/runs/natural_rate_hypothesis_long_run_phillips_vertical
1007550250197019972023USAGBRFRADEUITAESPNLD
illustrative sketch · run pending
No coefficients yet. When the model fires, this chart will show cpi_inflation_yoy across 20 sampled countries over 1970–2023.
The shapes above are stylised — none of the lines are real data.
Placeholder for natural_rate_hypothesis_long_run_phillips_vertical. Published chart will be generated from engine/runs/natural_rate_hypothesis_long_run_phillips_vertical/chart_data.json.

Who has skin in the game — schools predicting on this

1 school 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 SUPPORTEDschool refuted

Pre-registration

registration ordering unverified
first-spec commit 4c8ce8e · 2026-07-18T22:11:21Z
run generated · 2026-05-04T20:08:00Z
Run timestamp predates this path's first git-add commit (rebase, rename, or pre-git local run). Spec hash is still the path's first-add commit — not repository HEAD — but ordering is not a clean pre-registration proof.

Cross-country Phillips-curve data post-1970 shows no stable unemployment-inflation tradeoff in the long run, consistent with Friedman's natural-rate hypothesis.

Falsification criterion — what would disprove this

set before the run · honoured after

This hypothesis is considered falsified if:

PRIMARY (dispositive): run a country-by-country Johansen trace test on CPI inflation year-over-year and unemployment for the registered OECD sample. SUPPORTED if no stable rank-0 cointegrating vector is found at the 10% critical value in >=80% of tested country series. REFUTED if cointegration is found in >=50% of tested country series. PARTIAL if at least 8 countries are tested but neither threshold clears. METHOD_VALID if each tested country contributes at least 20 paired inflation/unemployment observations and at least 8 countries are tested.

formal test & threshold
test:      Country-level Johansen rank-0 trace-test screen on unemployment-inflation pairs across OECD 1970-2023.
threshold: [object Object]

Method

Template
cointegration_vecm
Fixed effects
Clustering
country
Sample
20 countries · 1970 – 2023
Evidence type
associational

VECM / cointegration test on long-run Phillips relationship in OECD post-1970 panel. Tests for stable cointegrating vector between unemployment and inflation; failure to cointegrate supports Friedman natural-rate / vertical-long-run-Phillips view.

Data

VariableSourceTransform
cpi_inflation_yoy
outcome
world_bank_wdi:FP.CPI.TOTL.ZGtier 2
level
unemployment_rate
outcome
world_bank_wdi:SL.UEM.TOTL.ZStier 2
level
oecd_cpi_all_items
outcome
oecd:OECD.SDD.TPStier 2
log_yoy
ten_year_inflation_expectations
treatment
fred:T10YIEtier 1
level
nairu_estimate
treatment
imf:NAIRUtier 2
level
output_gap
control
oecd:OutputGaptier 2
level
oil_price_brent
control
imf_pcps:POILBREtier 1
log_yoy
policy_rate
control
fred:FEDFUNDStier 1
level

● ready  · ● pending  · ● reconstruct-needed

Detailed result card

Result card - natural_rate_hypothesis_long_run_phillips_vertical

Verdict: REFUTED - cpi_inflation_yoy, unemployment_rate: no-cointegration share=10.0%, cointegration share=90.0%, tested countries=20

Pre-registration

  • Claim: Cross-country Phillips-curve data post-1970 shows no stable unemployment-inflation tradeoff in the long run, consistent with Friedman's natural-rate hypothesis.
  • Falsification rule: PRIMARY (dispositive): run a country-by-country Johansen trace test on CPI inflation year-over-year and unemployment for the registered OECD sample. SUPPORTED if no stable rank-0 cointegrating vector is found at the 10% critical value in >=80% of tested country series. REFUTED if cointegration is found in >=50% of tested country series. PARTIAL if at least 8 countries are tested but neither threshold clears. METHOD_VALID if each tested country contributes at least 20 paired inflation/unemployment observations and at least 8 countries are tested.
  • Falsification test: Country-level Johansen rank-0 trace-test screen on unemployment-inflation pairs across OECD 1970-2023.

Cointegration Screen

  • Method: Johansen trace test, rank-0 rejection at 10pct critical value
  • Pair: cpi_inflation_yoy and unemployment_rate
  • Tested countries: 20
  • No-cointegration share: 10.0%
  • Cointegration share: 90.0%

| Country | n | trace r=0 | cv90 | cointegrated at 10pct | |---|---:|---:|---:|---| | AUS | 33 | 18.1 | 13.4 | True | | AUT | 33 | 13.3 | 13.4 | False | | BEL | 33 | 27.3 | 13.4 | True | | CAN | 33 | 18.7 | 13.4 | True | | CHE | 33 | 17 | 13.4 | True | | DEU | 33 | 16.1 | 13.4 | True | | DNK | 33 | 15.1 | 13.4 | True | | ESP | 33 | 19.1 | 13.4 | True | | FIN | 33 | 22.2 | 13.4 | True | | FRA | 33 | 14.9 | 13.4 | True | | GBR | 33 | 21.8 | 13.4 | True | | IRL | 33 | 22.9 | 13.4 | True | | ITA | 33 | 15 | 13.4 | True | | JPN | 33 | 20.2 | 13.4 | True | | KOR | 33 | 20.7 | 13.4 | True | | NLD | 33 | 21.6 | 13.4 | True | | NOR | 33 | 13.3 | 13.4 | False | | NZL | 33 | 15.5 | 13.4 | True | | SWE | 33 | 32.2 | 13.4 | True | | USA | 33 | 18.8 | 13.4 | True |

Variables resolved

  • world_bank_wdi:FP.CPI.TOTL.ZG -> cpi_inflation_yoy (outcome, publisher=world_bank_wdi, n=9066)
  • world_bank_wdi:SL.UEM.TOTL.ZS -> unemployment_rate (outcome, publisher=world_bank_wdi, n=8071)
  • oecd:OECD.SDD.TPS,DSD_PRICES@DF_PRICES_ALL,1.0 -> oecd_cpi_all_items (outcome, publisher=oecd, n=2908)
  • fred:T10YIE -> ten_year_inflation_expectations (treatment, publisher=fred, n=24)
  • oecd:OutputGap -> output_gap (controls, publisher=oecd, n=3331)
  • imf_pcps:POILBRE -> oil_price_brent (controls, publisher=imf_pcps, n=740)
  • fred:FEDFUNDS -> policy_rate (controls, publisher=fred, n=1460)

Variables missing data

  • imf:NAIRU (treatment, name=nairu_estimate)

Generated by scripts/run_cointegration_vecm.py at 2026-05-04T20:08:00+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

Stub seeded from a Chicago-monetarist school prediction (Friedman natural-rate hypothesis). VECM/cointegration analysis is sensitive to lag and break specification; needs human review.

Authored framework. Read the transparency note.