Pre-registration
The cumulative Venezuelan real-GDP contraction over the 2013- 2023 Maduro era (>70% peak-to-trough from the 2013 peak) is decomposed into four channels: (a) the exogenous 2014-2016 oil price shock, (b) the 2013-2019 monetary-fiscal-fusion channel culminating in the 2017-2019 Cagan-threshold hyperinflation, (c) the post-2015 US sanctions escalation (SDN 2017, PDVSA designation January 2019, secondary sanctions on tanker operators 2019-2020), and (d) post-2017 political destabilisation (2017 constitutional crisis, 2019 Guaidó parallel-government episode, 2024 election dispute). The pre-registered claim is that channels (b) and (d) — the domestic monetary-fiscal-fusion and political-destabilisation channels — together account for a larger share of the cumulative loss than channels (a) + (c) — the exogenous oil-shock and sanctions channels — combined. A falsification would indicate the collapse is more exogenously driven than the framework's institutional-policy coding implies.
Falsification criterion — what would disprove this
This hypothesis is considered falsified if:
Not supported if EITHER (a) the combined variance share of channels (b) + (d) is less than 40% of the total explained variance in the VEN-donor-pool gap, OR (b) the oil-shock channel (a) alone accounts for more than 50% of the explained variance (which would indicate the collapse is predominantly exogenous, not policy-driven), OR (c) the sanctions channel (c) alone accounts for more than 40% of the explained variance (which would indicate the post-2015 sanctions escalation is the dominant force and the framework's institutional-quality coding is overstated).
formal test & threshold
test: shapley_decomposition_channel_variance_shares threshold: variance_share(b + d) > 0.40 AND variance_share(a) < 0.50 AND variance_share(c) < 0.40
Method
- Template
panel_fe_decomposition- Fixed effects
country, year- Clustering
country- Sample
- 7 countries · 2005 – 2023
- Evidence type
- causal
Panel FE decomposition with four treatment intensities entered linearly (plus interaction terms for monetary_fiscal_fusion × sanctions). The estimator returns share-of-variance-explained estimates for each channel under Shapley-style decomposition (Fields 2003; Shorrocks 2013), reporting each channel's marginal contribution to the observed VEN-vs-donor-pool log-GDP-pc gap. Robustness: synthetic-control weighting on donor pool fitted to VEN pre-2013 dynamics; channel-specific counterfactuals holding one channel at donor-pool mean while allowing others to vary. Permutation inference on donor pool for significance.
Data
| Variable | Source | Transform |
|---|---|---|
real_gdp_pc_ppp outcome | world_bank_wdi:NY.GDP.PCAP.PP.KDtier 2 | log_level_annual |
real_gdp_level outcome | world_bank_wdi:NY.GDP.MKTP.KDtier 2 maddison:mpd2020tier 3 | log_level_annual |
cumulative_venezuela_contraction_2013_peak_to_year outcome | derived | log(GDP_VEN_t) - log(GDP_VEN_2013) |
maduro_era_indicator treatment | constructed:= 1 for VEN from 2013 onward; 0 otherwisetier 5 | binary |
venezuela_oil_shock_exposure treatment | constructed:(oil_price_change) * (oil_export_share_venezuela)tier 5 | interaction |
monetary_fiscal_fusion_intensity treatment | constructed:1 once BCV direct financing of PDVSA reaches a threshold OR monthly inflation first breaches 20% m/m, whichever earlier;tier 5 | graded_indicator |
sanctions_regime_intensity treatment | constructed:stepwise intensity — 0 pre-2015, 1 post-SDN-individuals 2017, 2 post-PDVSA-SDN Jan 2019, 3 post-secondary-sanctions Aug tier 5 | ordinal |
political_destabilisation_intensity treatment | constructed:composite of (i) UCDP event counts for Venezuela, (ii) V-Dem electoral-integrity index inverse, (iii) WGI political stabtier 5 | composite_index_rescaled_0_1 |
commodity_terms_of_trade control | world_bank_wdi:TT.PRI.MRCH.XD.WDtier 2 | log_level |
oil_export_share_pre_shock control | world_bank_wdi:TX.VAL.FUEL.ZS.UNtier 2 | average_2010_2013 |
wgi_rule_of_law control | wgi:GOV_WGI_RL.ESTtier 4 | annual_level |
● ready · ● pending · ● reconstruct-needed
Detailed result card
Result card — maduro_era_venezuelan_collapse_decomposition_2013_2023
Verdict: INCONCLUSIVE_DATA_PENDING — treatment 'maduro_era_indicator' has no within-country variation under country fixed effects
Pre-registration
- Claim: The cumulative Venezuelan real-GDP contraction over the 2013- 2023 Maduro era (>70% peak-to-trough from the 2013 peak) is decomposed into four channels: (a) the exogenous 2014-2016 oil price shock, (b) the 2013-2019 monetary-fiscal-fusion channel culminating in the 2017-2019 Cagan-threshold hyperinflation, (c) the post-2015 US sanctions escalation (SDN 2017, PDVSA designation January 2019, secondary sanctions on tanker operators 2019-2020), and (d) post-2017 political destabilisation (2017 constitutional crisis, 2019 Guaidó parallel-government episode, 2024 election dispute). The pre-registered claim is that channels (b) and (d) — the domestic monetary-fiscal-fusion and political-destabilisation channels — together account for a larger share of the cumulative loss than channels (a) + (c) — the exogenous oil-shock and sanctions channels — combined. A falsification would indicate the collapse is more exogenously driven than the framework's institutional-policy coding implies.
- Falsification rule: Not supported if EITHER (a) the combined variance share of channels (b) + (d) is less than 40% of the total explained variance in the VEN-donor-pool gap, OR (b) the oil-shock channel (a) alone accounts for more than 50% of the explained variance (which would indicate the collapse is predominantly exogenous, not policy-driven), OR (c) the sanctions channel (c) alone accounts for more than 40% of the explained variance (which would indicate the post-2015 sanctions escalation is the dominant force and the framework's institutional-quality coding is overstated).
- Falsification test: shapley_decomposition_channel_variance_shares
Estimate
- Error: treatment 'maduro_era_indicator' has no within-country variation under country fixed effects
Variables resolved
world_bank_wdi:NY.GDP.PCAP.PP.KD→ real_gdp_pc_ppp (outcome, publisher=world_bank_wdi, n=8325)world_bank_wdi:NY.GDP.MKTP.KD; maddison:mpd2020→ real_gdp_level (outcome, publisher=world_bank_wdi, n=12104)constructed: = 1 for VEN from 2013 onward; 0 otherwise→ maduro_era_indicator (treatment, publisher=constructed, n=133)constructed: (oil_price_change) * (oil_export_share_venezuela)→ venezuela_oil_shock_exposure (treatment, publisher=constructed, n=133)constructed: 1 once BCV direct financing of PDVSA reaches a threshold OR monthly inflation first breaches 20% m/m, whichever earlier; scaled by intensity post-threshold→ monetary_fiscal_fusion_intensity (treatment, publisher=constructed, n=133)world_bank_wdi:TT.PRI.MRCH.XD.WD→ commodity_terms_of_trade (controls, publisher=world_bank_wdi, n=6478)world_bank_wdi:TX.VAL.FUEL.ZS.UN (fuel exports % merch)→ oil_export_share_pre_shock (controls, publisher=world_bank_wdi, n=9329)wgi:GOV_WGI_RL.EST→ wgi_rule_of_law (controls, publisher=wgi, n=5296)
Variables missing data
derived(outcome, name=cumulative_venezuela_contraction_2013_peak_to_year) — vintage not on diskconstructed: stepwise intensity — 0 pre-2015, 1 post-SDN-individuals 2017, 2 post-PDVSA-SDN Jan 2019, 3 post-secondary-sanctions Aug 2019(treatment, name=sanctions_regime_intensity) — vintage not on diskconstructed: composite of (i) UCDP event counts for Venezuela, (ii) V-Dem electoral-integrity index inverse, (iii) WGI political stability inverse(treatment, name=political_destabilisation_intensity) — vintage not on disk
Generated by scripts/run_panel_fe.py at 2026-06-29T17:52:54+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
Data-gated: VEN post-2013 national-accounts data are partial (BCV publication gaps 2015-2019, partial resumption 2020+). IMF estimates, Hanke-Krus inflation series, and opposition- calibrated estimates (Observatorio Venezolano de Finanzas) are all used with provenance-tracked. Pre-registration is binding; missing-data imputation rules are pre-specified in the v1 spec and cannot be post-hoc adjusted. This is the Maduro-era-specific decomposition. The broader venezuela_chavismo_framework_validation hypothesis tests the 1999-2020 aggregate gap; this hypothesis tests the 2013-2023 sub-period with explicit channel decomposition.