Original collectionUnder development

Peter Hull’s original collection. This paper belongs to the original 22-paper collection, following the criteria established in Peter Hull’s twfe-replication repository. The original collection is retained in full, separately from the community extension.

The Rise and Fall of Local Elections in China — expanded estimators

Archive index · Regression tests · Results & settings (JSON) · Methods & code

Martinez-Bravo et al. (2022), AER. https://doi.org/10.1257/aer.20181249

4 specifications. Effects are in outcome standard deviations; intervals are 95%.

Treatment adoption

S12-P001: seeded sample of 13 of 13 eligible units. Purple untreated; yellow treated; white unobserved.

Jump to specification Select…LOCAL-ELECTIONS-T5-C3-LAND-EXPROPRIATION — Indicator for expropriation of village landLOCAL-ELECTIONS-T5-C3-ONE-CHILD — Number of one-child-policy exemptions granted by the village governmentLOCAL-ELECTIONS-T5-C3-PUBLIC-GOODS — Public-goods expenditure funded by villagers (10,000 constant RMB)LOCAL-ELECTIONS-T5-C3-VC-UNILATERAL — Sum of policies unilaterally decided by the elected village committee chairman

Indicator for expropriation of village land

LOCAL-ELECTIONS-T5-C3-LAND-EXPROPRIATION · Main-text Equation (1), Table 5 column 3, panel E · Data & fit details

195 observations · 13 units · 13 clusters · 11 treated / 2 control units.

Controls: latest-adopting cohort; analysis ends before period 16.

Estimate comparison

Estimator	Estimate (SDs)	SE	95% CI	Averaging
Static TWFE	-0.014	0.055	[-0.134, 0.106]	Coefficient
Dynamic TWFE	-0.015	0.083	[-0.195, 0.166]	Treated cells
BJS / did2s	0.086	0.080	[-0.087, 0.259]	Treated cells
Sun–Abraham	0.086	0.000	[0.086, 0.086]	Treated cells

Event studies

Periods −10 to +10; pointwise 95% intervals where available.

Number of one-child-policy exemptions granted by the village government

LOCAL-ELECTIONS-T5-C3-ONE-CHILD · Main-text Equation (1), Table 5 column 3, panel D · Data & fit details

195 observations · 13 units · 13 clusters · 11 treated / 2 control units.

Controls: latest-adopting cohort; analysis ends before period 16.

Estimate comparison

Estimator	Estimate (SDs)	SE	95% CI	Averaging
Static TWFE	-0.135	0.216	[-0.606, 0.336]	Coefficient
Dynamic TWFE	0.301	0.296	[-0.345, 0.947]	Treated cells
BJS / did2s	0.319	0.197	[-0.110, 0.747]	Treated cells
Sun–Abraham	0.455	0.101	[0.234, 0.675]	Treated cells

Event studies

Periods −10 to +10; pointwise 95% intervals where available.

Public-goods expenditure funded by villagers (10,000 constant RMB)

LOCAL-ELECTIONS-T5-C3-PUBLIC-GOODS · Main-text Equation (1), Table 5 column 3, panel B · Data & fit details

195 observations · 13 units · 13 clusters · 11 treated / 2 control units.

Controls: latest-adopting cohort; analysis ends before period 16.

Estimate comparison

Estimator	Estimate (SDs)	SE	95% CI	Averaging
Static TWFE	-0.000	0.017	[-0.038, 0.037]	Coefficient
Dynamic TWFE	-0.004	0.030	[-0.069, 0.061]	Treated cells
BJS / did2s	0.023	0.022	[-0.024, 0.070]	Treated cells
Sun–Abraham	0.024	0.009	[0.004, 0.045]	Treated cells

Event studies

Periods −10 to +10; pointwise 95% intervals where available.

Sum of policies unilaterally decided by the elected village committee chairman

LOCAL-ELECTIONS-T5-C3-VC-UNILATERAL · Main-text Equation (1), Table 5 column 3, panel A · Data & fit details

195 observations · 13 units · 13 clusters · 11 treated / 2 control units.

Controls: latest-adopting cohort; analysis ends before period 16.

Estimate comparison

Estimator	Estimate (SDs)	SE	95% CI	Averaging
Static TWFE	0.189	0.157	[-0.154, 0.531]	Coefficient
Dynamic TWFE	0.038	0.111	[-0.204, 0.280]	Treated cells
BJS / did2s	0.044	0.070	[-0.109, 0.198]	Treated cells
Sun–Abraham	-0.027	0.166	[-0.389, 0.334]	Treated cells

Event studies

Periods −10 to +10; pointwise 95% intervals where available.