Pre-registration
Argentina's 2019 PASO shock generated an immediate official-FX break, reserve loss, and inflation pass-through; the 2020 base-money expansion was followed by a lagged inflation pickup by Q4.
Falsification criterion — what would disprove this
This hypothesis is considered falsified if:
SUPPORTED if at least 3 of 4 metrics pass: PASO-week devaluation exceeds 25 percent, reserves fall by more than 10 percent in about 30 days, Aug-Sep 2019 monthly inflation averages above 4.5 percent, and 2020 base money grows more than 20 percent from March to September while Q4 monthly inflation exceeds Q2 by more than 1pp. REFUTED if 2 or fewer pass.
formal test & threshold
test: argentina_paso_2019_base_money_lag_four_metric_window threshold: SUPPORTED if >= 3 metrics pass; REFUTED if <= 2 pass.
Method
- Template
multi_metric_checklist- Clustering
none- Sample
- 1 countries · 2019 – 2020
- Evidence type
- canonical_case_multi_metric
Compact national event-window replication from cached ONS/INE/BCRA vintages.
Data
| Variable | Source | Transform |
|---|---|---|
official_fx_retail outcome | bcra:4tier 1 | daily ARS/USD |
international_reserves outcome | bcra:1tier 1 | daily USD millions |
monthly_cpi_inflation outcome | bcra:27tier 1 | monthly percent |
monetary_base outcome | bcra:15tier 1 | daily ARS millions |
paso_primary_shock treatment | constructed:2019-08-11 PASO primarytier 5 | event indicator |
pandemic_base_money_expansion treatment | constructed:2020 monetary-base expansion windowtier 5 | event indicator |
● ready · ● pending · ● reconstruct-needed
Detailed result card
Result card - argentina_paso_2019_fx_reserves_inflation_base_money_lag
Verdict: SUPPORTED - 3/4 metrics passed (support >= 3; refute <= 2).
Claim
Argentina's 2019 PASO shock generated an immediate official-FX break, reserve loss, and inflation pass-through; the 2020 base-money expansion was followed by a lagged inflation pickup by Q4.
Metrics
| Metric | Value | Threshold | Pass | Details | |---|---:|---|:---:|---| | paso_fx_devaluation | 33.577 | >25% devaluation within one week | yes | 46.55 to 62.18 ARS/USD | | reserves_drawdown | 24.461 | >10% reserve fall within about 30 days | yes | 66309 to 50089 USD mn | | inflation_pass_through | 4.950 | Aug-Sep 2019 average monthly CPI >4.5% | yes | 4.95% average monthly | | base_money_lagged_inflation | 1.933 | base money +20% Mar-Sep and Q4 inflation > Q2 by 1pp | no | base 2292083 to 2393695; CPI avg 1.73% to 3.67% |
Interpretation
This is a compact predeclared event-window verdict using local cached national-statistics vintages. It is strong for timing and magnitude, but not a full causal structural decomposition.
Provenance
See manifest.yaml for exact vintage files and SHA-256 hashes. Re-run with replication.py.
Strongest opposing argument
Every hypothesis ships with its charitable opposing argument. The framework earns credibility by handling objections at their strongest, not weakest.
Notes
Generated by scripts/generate_national_event_wave.py from local cached vintages; no network fetch required.