Pre-registration
The Sep 2021 introduction of Bitcoin as legal tender plus the Chivo state-wallet launch and $30 per-adult BTC buy-in failed to produce a statistically significant change in El Salvador's total remittance volume or in the aggregate price of remittance services (World Bank remittance-cost series) relative to Central American and Caribbean peer countries. The pre- registered claim is that in a synthetic-control design, SLV post-2021 remittance volume and remittance cost do not deviate from the counterfactual donor pool by more than a small magnitude, and that the share of remittances flowing through BTC/Chivo rails remained below 10% in any rolling 6-month window through 2024 per BCR monthly remittance reports. This hypothesis stands against the ex-ante proponent claim (Bukele government, Strike, Jack Mallers) that BTC legal tender would reduce remittance costs by replacing high-fee Western Union / MoneyGram corridors and would raise total remittance volume by enabling previously-unbanked recipients.
Falsification criterion — what would disprove this
This hypothesis is considered falsified if:
The hypothesis is 'Bitcoin legal tender produced no material change in remittance volume or cost.' It is falsified (i.e., rejected in favour of the proponent story) if any of: (a) synthetic-control CATT on log(remittance_inflows_usd) over 2021Q4-2024Q4 is positive at p_perm < 0.05 and magnitude > 10% of pre-treatment level, OR (b) synthetic-control CATT on remittance_cost_usd_200_avg is negative at p_perm < 0.05 and magnitude > 2 percentage points, OR (c) the Chivo-share- of-remittances exceeds 10% in any rolling 6-month window 2022-2024. Support requires none of (a)(b)(c) to be met.
formal test & threshold
test: synthetic_control_cost_and_volume_plus_chivo_share_mechanism threshold: CATT_2021Q4_2024Q4(log_remittance_inflows) NOT > 0.10 at p_perm < 0.05 AND CATT_2021Q4_2024Q4(remittance_cost) NOT < -2pp at p_perm < 0.05 AND max_rolling_6m(chivo_remittance_share) < 0.10
Method
- Template
synthetic_control- Clustering
country- Sample
- 9 countries · 2010 – 2025
- Evidence type
- causal
Primary: synthetic control (Abadie 2010) with SLV treated 2021-09-07 onward, Central American and Caribbean donor pool weighted on 2010-2021 remittance-inflow and remittance-cost pre-trends. Secondary: Callaway-Sant'Anna DiD with monthly remittance data. Mechanism test: the Chivo-share-of- remittance metric is a direct behavioural variable; if it stays < 5% throughout 2021-2024, the ex-ante proponent claim is directly refuted regardless of aggregate outcome.
Data
| Variable | Source | Transform |
|---|---|---|
remittance_inflows_usd outcome | world_bank_wdi:BX.TRF.PWKR.CD.DTtier 2 sv_bcr:Remesastier 2 | log_level |
remittance_share_gdp outcome | world_bank_wdi:BX.TRF.PWKR.DT.GD.ZStier 2 | level |
remittance_cost_usd_200_avg outcome | world_bank_remittance_prices:RPWtier 2 | level |
chivo_remittance_share outcome | sv_bcr:Remesastier 2 | level |
bitcoin_legal_tender_active treatment | constructed:binary = 1 for SLV from 2021-09-07 through 2025-01-29; 0 otherwisetier 5 | binary |
chivo_active treatment | constructed:binary = 1 for SLV from 2021-09-07 onwardtier 5 | binary |
us_employment_hispanic control | fred:LNS12000009tier 1 | log_level |
us_gdp control | fred:GDPC1tier 1 | log_level |
oil_price_brent control | fred:DCOILBRENTEUtier 1 | log_level |
recipient_country_gdp control | world_bank_wdi:NY.GDP.MKTP.KDtier 2 | log_level |
● ready · ● pending · ● reconstruct-needed
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 on BCR remittance series (monthly since 2010, includes billetera-digital decomposition since Oct 2021) and World Bank Remittance Prices Worldwide quarterly series. Donor pool peers have comparable US-corridor remittance dependency and Central American / Caribbean migration patterns. Pre- registration date for v1: first git-commit timestamp of this file supersedes prior draft.