El Salvador's homicide rate fell from 52 per…

El Salvador's homicide rate fell from 52 per 100,000 (2019) to 2.4 per 100,000 (2023) — a 95% reduction — under Bukele's Estado de Excepción security crackdown beginning March 2022.

This security improvement is associated with measurable GDP and FDI gains: El Salvador grew approximately 3.5% real in 2022 and 4.0% in 2023 per IMF NGDP_RPCH, outperforming Honduras and Guatemala — the structural comparators — for the first time since 2010. The pre-registered causal claim is that crime reduction via security crackdown (regardless of rule-of-law costs) produced positive economic returns through three channels: tourism recovery, FDI attraction in light manufacturing, and domestic consumer confidence, detectable as a positive cumulative average treatment effect on the treated (CATT) in a synthetic-control design with HND, GTM, NIC, and DOM as donor countries. The rule-of-law deterioration — mass imprisonment of approximately 75,000 persons, habeas corpus suspension, credible reports of wrongful detention — is a confounding factor coded as a separate variable and is not subsumed into the economic growth claim. The hypothesis tests only the economic-returns mechanism of crime reduction, not the normative legitimacy of the enforcement method.

Pre-registration

registration ordering unverified

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

run generated · 2026-04-28T12:41:54Z

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.

El Salvador's homicide rate fell from 52 per 100,000 (2019) to 2.4 per 100,000 (2023) — a 95% reduction — under Bukele's Estado de Excepción security crackdown beginning March 2022. This security improvement is associated with measurable GDP and FDI gains: El Salvador grew approximately 3.5% real in 2022 and 4.0% in 2023 per IMF NGDP_RPCH, outperforming Honduras and Guatemala — the structural comparators — for the first time since 2010. The pre-registered causal claim is that crime reduction via security crackdown (regardless of rule-of-law costs) produced positive economic returns through three channels: tourism recovery, FDI attraction in light manufacturing, and domestic consumer confidence, detectable as a positive cumulative average treatment effect on the treated (CATT) in a synthetic-control design with HND, GTM, NIC, and DOM as donor countries. The rule-of-law deterioration — mass imprisonment of approximately 75,000 persons, habeas corpus suspension, credible reports of wrongful detention — is a confounding factor coded as a separate variable and is not subsumed into the economic growth claim. The hypothesis tests only the economic-returns mechanism of crime reduction, not the normative legitimacy of the enforcement method.

Falsification criterion — what would disprove this

set before the run · honoured after

This hypothesis is considered falsified if:

The hypothesis is falsified if the synthetic-control CATT on real GDP growth for SLV over 2022-2024 is not statistically positive at p_perm < 0.10 relative to the donor pool counterfactual, OR if the CATT on log GDP per capita is not positive and economically meaningful (> 2% cumulative above synthetic control), OR if FDI inflows show no differential increase for SLV post-2022 relative to donor countries.

formal test & threshold

test:      synthetic_control_gdp_growth_plus_fdi_mechanism
threshold: CATT_2022_2024(real_gdp_growth) > 0 at p_perm < 0.10 AND CATT_2022_2024(log_gdp_per_capita) > 0.02 cumulative AND FDI_inflows_SLV_2022_2024 > FDI_synthetic_control_2022_2024

Method

Template: synthetic_control
Sample: 5 countries · 2010 – 2024
Evidence type: causal

Primary: synthetic control (Abadie 2010) with SLV as treated unit, treatment date 2022-03-27, donor pool HND/GTM/DOM weighted on 2010-2021 GDP growth, GDP per capita, remittance share, and FDI pre-trends. CATT evaluated on real GDP growth and log GDP per capita 2022-2024. Secondary: Callaway-Sant'Anna DiD using annual panel. Mechanism test: two-stage regression of homicide_rate on estado_de_excepcion_active (first stage), then GDP outcomes on fitted homicide reduction (second stage), to estimate the economic return per unit of crime reduction. Confound separation: rule_of_law and freedom_house_score entered as controls to test whether GDP gains survive institutional-quality deterioration controls.

Data

Variable	Source	Transform
gdp_growth_real outcome	imf:NGDP_RPCHtier 2 world_bank_wdi:NY.GDP.MKTP.KDtier 2	pct_change
gdp_per_capita_ppp outcome	world_bank_wdi:NY.GDP.PCAP.PP.KDtier 2	log_level
fdi_inflows_pct_gdp outcome	world_bank_wdi:BX.KLT.DINV.WD.GD.ZStier 2	level
tourism_arrivals outcome	world_bank_wdi:ST.INT.ARVLtier 2	log_level
estado_de_excepcion_active treatment	constructed:binary = 1 for SLV from 2022-03-27 onward; 0 for all donor countriestier 5	binary
homicide_rate_per_100k treatment	unodc:intentional_homicidetier 2 constructed:slv_pnc_officialtier 5	level
us_remittance_conditions control	fred:LNS12000009tier 1	log_level
current_account_pct_gdp control	imf:BCA_NGDPDtier 2	level
rule_of_law control	wgi:GOV_WGI_RL.ESTtier 4	level
freedom_house_score control	freedom_house:fiw_all_countriestier 4	level
military_spending_pct_gdp control	sipri:milex_share_gdptier 3	level
population_total control	world_bank_wdi:SP.POP.TOTLtier 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.