IESET.
Hypotheses·labour·minimum_wage_employment_effect_us_states

US state-level minimum-wage increases over 1990-2024 produce small, statistically uncertain effects on teen and low-wage employment when compared to contiguous counties or neighbouring states that did not raise their minimum.

The Card-Krueger 1994 / Dube-Lester-Reich 2010 research design is the baseline: the hypothesis predicts that the estimated employment elasticity with respect to the minimum wage falls in the range [-0.15, +0.05] with the 95% confidence interval straddling zero, consistent with the post-1990 border-discontinuity literature but inconsistent with Neumark-Wascher-style nationwide panel findings of elasticities < -0.2. Pre-registered direction of effect: minimum-wage elasticity is close to zero and typically not significantly negative at the 5% level on the primary outcome.

INCONCLUSIVEengine/runs/minimum_wage_employment_effect_us_states

inconclusive — data gap on bls:LAU_state_teen_employment_population_ratio_panel, bls:CES_state_NAICS722_employment_panel. The spec's primary Callaway-Sant'Anna staggered DiD on the 1990-2024 US-state panel cannot be estimated without the state-level BLS LAU teen employment-to-population series and the USDOL state minimum-wage history. On-disk BLS vintages currently include only national-level series (LNS*, CES05*, CUUR*); the state fan-out (SMS*, LAUST*, ENU*-county QCEW) and the USDOL state-history table have not been fetched. No coefficients computed.

confidence cueResult card produced; verdict unclassified.

policy briefNot enough data

In ordinary language

When minimum wages rise high relative to normal local pay, do lower-skill workers keep their jobs, or does hiring fall at the margin?

plain answer

This test cannot make a firm call yet. data gap on bls:LAU_state_teen_employment_population_ratio_panel, bls:CES_state_NAICS722_employment_panel.

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 1 country or place units from 1990 to 2024, using a did callaway santanna design.

what was measured
What changed
  • State minimum wage log level
What we checked
  • Teen employment to population ratio
  • Low wage sector employment
  • Unemployment duration teen
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

2 input datasets, 5 unresolved missing series, provenance status: incomplete.

Results

engine/runs/minimum_wage_employment_effect_us_states
Loading chart…

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.

Democratic Socialism (Bernsteinian revisionism + modern)wins if SUPPORTEDpartial/no-call

Pre-registration

pre-registered
first-spec commit 4c8ce8e · 2026-07-18T22:11:21Z

US state-level minimum-wage increases over 1990-2024 produce small, statistically uncertain effects on teen and low-wage employment when compared to contiguous counties or neighbouring states that did not raise their minimum. The Card-Krueger 1994 / Dube-Lester-Reich 2010 research design is the baseline: the hypothesis predicts that the estimated employment elasticity with respect to the minimum wage falls in the range [-0.15, +0.05] with the 95% confidence interval straddling zero, consistent with the post-1990 border-discontinuity literature but inconsistent with Neumark-Wascher-style nationwide panel findings of elasticities < -0.2. Pre-registered direction of effect: minimum-wage elasticity is close to zero and typically not significantly negative at the 5% level on the primary outcome.

Falsification criterion — what would disprove this

set before the run · honoured after

This hypothesis is considered falsified if:

PRIMARY (dispositive): the hypothesis is SUPPORTED if the Callaway-Sant'Anna staggered-DiD elasticity of teen employment-to-population with respect to the log state minimum wage over the 1990-2024 US-state panel falls in [-0.15, +0.05] AND the 95% CI crosses zero. REFUTED if the elasticity is significantly below -0.15 at the 5% level (Neumark-Wascher direction) OR significantly above +0.10 at the 5% level (positive-employment direction; the claim is "close to zero," not "positive"). SECONDARY (also dispositive): the Dube-Lester-Reich contiguous-county-pair elasticity on QCEW NAICS-722 (accommodation/food services) county employment must fall in [-0.20, +0.05]. INFORMATIVE: leave-one-state-out sign stability of the CS-ATT coefficient; pre-trend test on the 4 years preceding each state's first minimum-wage hike. METHOD_VALID: state-level BLS LAU teen E/P available for at least 40 of 50 states with >= 20 years of coverage; USDOL state minimum-wage history available with at least one state-level binding change in at least 30 states; QCEW NAICS-722 county-level employment available for the border- pair specification.

formal test & threshold
test:      staggered_did_plus_border_pair_both_in_range
threshold: PRIMARY: CS_ATT_elasticity in [-0.15, +0.05] AND 95% CI crosses zero AND CS_ATT not significantly < -0.15 AND not significantly > +0.10. SECONDARY: border_pair_elasticity in [-0.20, +0.05].

Method

Template
did_callaway_santanna
Clustering
state
Sample
1 countries · 1990 – 2024
Evidence type
causal

Primary: Callaway-Sant'Anna staggered DiD exploiting state-level minimum-wage changes 1990-2024 against never-/late-changing states. Secondary: Dube-Lester-Reich contiguous-county-pair design using QCEW accommodation-and-food-services county employment around state-border discontinuities. The border-pair design is the cleaner identification but has narrower outcome coverage.

Data

VariableSourceTransform
teen_employment_to_population_ratio
outcome
bls:LNS12300012tier 1
bls:LAUtier 1
level_pct
low_wage_sector_employment
outcome
bls:CEStier 1
log_level
unemployment_duration_teen
outcome
bls:LNS12000000tier 1
median_weeks_unemployed
state_minimum_wage_log_level
treatment
bls:state_minimum_wage_history (USDOL)tier 1
log_nominal_state_minimum_wage
state_gdp_growth
control
bls:QCEWtier 1
fred:NYNGSPtier 1
yoy_pct_change
state_unemployment_rate_adult
control
bls:LAU_state_unemploymenttier 1
level_pct

● ready  · ● pending  · ● reconstruct-needed

Detailed result card

US state minimum-wage employment effects, 1990-2024

Verdict: inconclusive — data gap on bls:LAU_state_teen_employment_population_ratio_panel, bls:CES_state_NAICS722_employment_panel. The spec's primary Callaway-Sant'Anna staggered DiD on the 1990-2024 US-state panel cannot be estimated without the state-level BLS LAU teen employment-to-population series and the USDOL state minimum-wage history. On-disk BLS vintages currently include only national-level series (LNS*, CES05*, CUUR*); the state fan-out (SMS*, LAUST*, ENU*-county QCEW) and the USDOL state-history table have not been fetched. No coefficients computed.

Summary

  • The hypothesis tests the post-1990 Card-Krueger / Dube-Lester-Reich consensus that state-level minimum-wage elasticities are small (in the [-0.15, +0.05] band) against the Neumark-Wascher claim of elasticities < -0.2.
  • Primary statistic: Callaway-Sant'Anna staggered-DiD elasticity of teen employment-to-population ratio with respect to log state minimum wage on the 1990-2024 US-state panel.
  • Secondary: Dube-Lester-Reich contiguous-county-pair elasticity on QCEW NAICS-722 (accommodation/food services) county employment.
  • Required series: 2 outcome, 1 treatment, 2 border-pair, 2 controls = 7 total.
  • Found on-disk: 2 of 7.
  • Missing primary outcome: ['bls:LAU_state_teen_employment_population_ratio_panel', 'bls:CES_state_NAICS722_employment_panel'].
  • Missing treatment: (none).
  • Missing border-pair: ['manual:vaghul_zipperer_county_minimum_wage'].
  • Missing controls: ['bls:LAU_state_unemployment_rate_panel', 'fred:state_real_gdp_panel'].

Method

Pre-registered specification (50-state panel, 1990-2024, excluding 2020 COVID and federal-floor-binding state-years):

log(teen_E/P)_{s,t} = beta * log(min_wage)_{s,t}
                    + alpha_s + tau_t + X_{s,t}'gamma + e

with state and year fixed effects, Callaway-Sant'Anna 2021 estimator using never- and late-treated states as controls. Standard errors clustered by state. Border-pair robustness uses Dube-Lester-Reich 2010 contiguous-county-pair fixed effects on QCEW NAICS-722 employment.

Falsification thresholds (dispositive): PRIMARY: CS_ATT_elasticity in [-0.15, +0.05] with 95% CI crossing zero → SUPPORTED. CS_ATT_elasticity < -0.15 significant at 5% → REFUTED (Neumark-Wascher direction). CS_ATT_elasticity > +0.10 significant at 5% → REFUTED (positive-effect direction). SECONDARY: border_pair_elasticity in [-0.20, +0.05] required for full SUPPORTED verdict.

Data

Required (per spec):

  • bls:LAU_state_teen_employment_population_ratio_panel — missing
  • bls:CES_state_NAICS722_employment_panel — missing
  • usdol:state_minimum_wage_history — available
  • bls:QCEW_county_NAICS722_employment_panel — available
  • manual:vaghul_zipperer_county_minimum_wage — missing
  • bls:LAU_state_unemployment_rate_panel — missing
  • fred:state_real_gdp_panel — missing

Promotion verdict: inconclusive (method-validity gate fails on data availability — state-level BLS series are not on disk; the BLS fetcher currently exposes only national LNS/CES/CUUR series). Per HANDOFF_TO_RUN_AGENT.md a data gap is NOT a refutation — the scoreboard treats this as neutral. Re-run when the BLS state fan-out (LAU state teen E/P, SMS state CES, ENU county QCEW) and USDOL state minimum-wage history fetchers are wired and the Vaghul-Zipperer county-minimum dataset is dropped into data/manual/.

Strongest opposing argument

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

Notes

Data-gated but BLS state-level series are mostly ready. CPS microdata requires IPUMS download; automated via ipums-cps package when added to fetchers. Vaghul-Zipperer county-level minimum-wage dataset is a manual drop in data/manual/ for the border-pair robustness spec.

Authored framework. Read the transparency note.