Skip to main content

Table 15 Houseing prices and immigration—MS-regions from 1990-2016: LOO instrument.

From: How does immigration affect housing costs in Switzerland?

 

IV regressions

First-stage regressions

Dep. var.:

\(\Delta \, \text {ln} \, HP_{it}\)

\(\Delta \, \text {ln} \, AP_{it}\)

\(\frac{\Delta I_{it}}{Pop_{it-1}}\)

\(\frac{\Delta I_{it}}{Pop_{it-1}} \times {\textbf {1}} (t \ge 2002)\)

 

(1)

(2)

(3)

(4)

(5)

(6)

(7)

\(\frac{\Delta I_{it}}{Pop_{it-1}}\)

2.498\(^{***}\)

-2.285

4.487\(^{***}\)

0.846

   
 

[0.003]

[0.230]

[0.000]

[0.563]

   

\(\frac{\Delta I_{it}}{Pop_{it-1}} \times {\textbf {1}} (t \ge 2002)\)

 

6.664\(^{**}\)

 

5.074\(^{**}\)

   
  

[0.019]

 

[0.024]

   

\(\Delta \, u_{it-1}\)

− 0.057

− 0.283

0.106

− 0.065

− 0.031

-0.035

− 0.000

 

[0.714]

[0.107]

[0.571]

[0.705]

[0.402]

[0.334]

[0.997]

\(\frac{\Delta \hat{I}_{it}^{LOO}}{Pop_{it-1}}\)

    

0.210\(^{***}\)

0.158\(^{***}\)

− 0.000

     

[0.003]

[0.002]

[0.997]

\(\frac{\Delta \hat{I}_{it}^{LOO}}{Pop_{it-1}} \times {\textbf {1}} (t \ge 2002)\)

     

0.084

0.242\(^{**}\)

      

[0.322]

[0.011]

Year FE

Yes

Yes

Yes

Yes

Yes

Yes

Yes

\(Obs\)

2754

2754

2754

2754

2754

2754

2754

\(R^{2}\)

0.449

0.371

0.537

0.531

0.411

0.412

0.536

K-P F-stat

9.42

4.94

9.42

4.94

   

S-W F-stat, immigration

 

9.91

 

9.91

   

S-W F-stat, interaction

 

11.69

 

11.69

   
  1. p-values in brackets. Standard errors are clustered by MS-regions. The regressions are run at the MS-regional level for the period from 1990-2016. \(\Delta\) indicates first differences. The instrument is the change in the predicted stock of foreigners divided by initial population. The stock of foreigners is predicted by the settlement pattern of immigrants in 1980. In contrast to before, here the respective MS-region i is not taken into account in \(I_{c,t}\) when estimating the predicted stock of foreigners (LOO: leave-one-out estimator). The K-P F-stat is the Kleibergen-Paap rk Wald F statistic. The S-W F-stat is the Sanderson and Windmeijer (2016) multivariate F-test of excluded instruments for weak identification of each endogenous regressor separately; immigration and interaction refer to endogenous regressors \(\Delta I_{it}/Pop_{it-1}\) and \(\Delta I_{it}/Pop_{it-1} \times {\textbf {1}} (t \ge 2002)\), respectively
  2. Significance levels: \(^{**}\) \(p<0.05\), \(^{***}\) \(p<0.01\)