Table 2

Spatial scope of human capital spillovers – dependent variable is the number of new establishments per cell of 1 km2

# of workers with college-or-more(1)(2)(3)(4)
0 to 1 km3.07e−04***2.77e−04***2.78e−04***2.96e−04***
(3.61e−05)(3.92e−05)(3.88e−05)(4.56e−05)
1 to 5 km6.44e−05***3.01e−05**7.15e−05***1.45e−05
(2.07e−05)(1.35e−05)(1.99e−05)(1.28e−05)
5 to 10 km1.43e−05−3.15e−05***6.79e−06−4.22e−05***
(1.18e−05)(6.80e−06)(1.09e−05)(9.57e−06)
10 to 20 km9.56e−06−3.22e−05**−1.62e−05***−4.36e−05**
(1.22e−05)(1.64e−05)(5.41e−06)(2.08e−05)
Cell controlsNoYesYesYes
Municipal controlsNoYesYesYes
Linear time trendNoNoYesYes
District FENoNoYesNo
IVNoNoNoYes
Pseudo R20.03250.07120.17630.0729
Pseudo-LL−74975.73−71973.41−63830.97−71842.56

Note(s): Table 2 shows the results for the estimation of distinct specifications of Equation (1) when we aggregate all rings in a unique explanatory variable. All models were estimated using 50,422 observations. Column (1) shows results without inclusion of controls. Column (2) includes cell and municipal controls (see Table A2) plus a linear trend. Column (3) includes all the controls plus the 627 district fixed effects. Column (4) presents the results when considering all the controls and an instrumental variable (IV) strategy through a control function. Significance levels: ***p < 0.01, **p < 0.05

[ViewLarge]
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal