Pre-registration
Politicised development-bank lending that increases around election cycles predicts weaker subsequent total-factor-productivity growth and higher non-performing loan ratios in a broad panel of emerging and developing economies during 1990-2020.
Falsification criterion — what would disprove this
This hypothesis is considered falsified if:
Falsified if election-year development-bank lending surges do not show a statistically significant negative association with subsequent TFP growth and a positive association with NPL ratios after controlling for private credit growth and institutional quality.
formal test & threshold
test: event_study_election_credit_on_tfp_and_npls threshold: [object Object]
Method
- Template
event_study- Clustering
country- Sample
- 90 countries · 1990 – 2020
- Evidence type
- causal
Event-study around election years with development-bank lending surges. Panel FE with election-year interaction as robustness.
Data
| Variable | Source | Transform |
|---|---|---|
total_factor_productivity_growth outcome | pwt:rtfpnatier 3 | annual_pct |
nonperforming_loan_ratio outcome | world_bank_wdi:GFDD.DM.04tier 2 | level_pct |
real_gdp_growth outcome | world_bank_wdi:NY.GDP.MKTP.KD.ZGtier 2 | annual_pct |
election_cycle_credit_surge treatment | constructed:dev_bank_lending_in_election_year_minus_off_yeartier 5 | level_pct_gdp |
election_year_dummy treatment | constructed:binarytier 5 | binary |
private_credit_growth control | world_bank_wdi:GFDD.DM.02tier 2 | annual_pct |
institutional_quality control | wgi:RL.ESTtier 4 | level |
commodity_price_index control | constructed:terms_of_trade_indextier 5 | level |
● ready · ● pending · ● reconstruct-needed
Detailed result card
Result card — politicised_credit_election_cycle_growth_drag
Verdict: PARTIAL — shape=TWFE, coef=+3.326e+12, p=0.455 (above α=0.10)
Pre-registration
- Claim: Politicised development-bank lending that increases around election cycles predicts weaker subsequent total-factor-productivity growth and higher non-performing loan ratios in a broad panel of emerging and developing economies during 1990-2020.
- Falsification rule: Falsified if election-year development-bank lending surges do not show a statistically significant negative association with subsequent TFP growth and a positive association with NPL ratios after controlling for private credit growth and institutional quality.
- Falsification test: event_study_election_credit_on_tfp_and_npls
- Event year: (not extracted)
Estimate
- coefficient: 3326068214352.472
- std_error: 4454135868554.447
- p_value: 0.45522230934702135
- n_obs: 1344
- n_countries: 64
- r_squared_within: 0.6517467163383956
- fe_entity: True
- fe_time: True
- cluster: country
- method: event-study TWFE fallback (linearmodels failed: 'election_cycle_credit_surge')
- shape: multi_country_twfe
- dropped_controls_due_to_overlap: []
Variables resolved
pwt:rtfpna→ total_factor_productivity_growth (outcome, publisher=pwt, n=6412)world_bank_wdi:NY.GDP.MKTP.KD.ZG→ real_gdp_growth (outcome, publisher=world_bank_wdi, n=13897)constructed:dev_bank_lending_in_election_year_minus_off_year→ election_cycle_credit_surge (treatment, publisher=constructed, n=2790)wgi:RL.EST→ institutional_quality (controls, publisher=wgi, n=5296)
Variables missing data
world_bank_wdi:GFDD.DM.04(outcome, name=nonperforming_loan_ratio)constructed:binary(treatment, name=election_year_dummy)world_bank_wdi:GFDD.DM.02(controls, name=private_credit_growth)constructed:terms_of_trade_index(controls, name=commodity_price_index)
Generated by scripts/run_event_study.py at 2026-05-02T19:20:21+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
Development-bank lending data from national sources, OECD DBS, or World Bank FinDev gateway. Election dates from NELDA, V-Dem, or national electoral commissions.