Table 4 Robustness check: balanced sample and other clustering strategy.

From: Plugging in for cities: the impact of power infrastructure on urban agglomeration

 

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

 

Balanced Sample

Clustering at Provincial Level

 

Area

Pop

Density

Height

Area

Pop

Density

Height

Plant

−0.1675***

0.0185**

0.0650***

0.1446***

− 0.2018***

0.0186*

0.0755***

0.1865***

 

(0.0282)

(0.0083)

(0.0191)

(0.0266)

(0.0339)

(0.0099)

(0.0254)

(0.0333)

ControlVariable

CityFE

Year*CountryFE

Observation

744736

875173

670173

209356

1235480

875662

741535

235762

R2

0.9604

0.9685

0.9567

0.9251

0.9572

0.9682

0.9593

0.9234

  1. ***, **, * denote significance at 1%, at 5%, and at 10%. “ControlVariable” includes two time-varying variables: temperature and rainfall. We include city fixed effect and country × year fixed effect into each regression. “Area” is an abbreviation for Built-up Area; “Pop” is an abbreviation for Population; “Density” is an abbreviation for Population Density. All dependent variables are logarithmic. The standard errors are reported in parentheses and clustered at the city level.