Skip to main content

Table 9 Instrumental variable estimation: employment, hours worked and remittances

From: Remittances and labor supply in post-conflict Tajikistan

 

Panel A: Full sample

Panel B: Conflict region

 

IV-2SLS

IV-Tobit

IV-2SLS

IV-Tobit

 

Employed last 40 days

Hours worked last 14d

Employed last 40 days

Hours worked last 14d

 

men

women

men

women

men

women

men

women

 

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

Ln (total remittances)

−0.071***

−0.017

−11.509***

−3.821***

−0.068***

−0.016

−7.562***

−0.517

(0.026)

(0.019)

(1.590)

(1.328)

(0.025)

(0.031)

(1.972)

(1.804)

N

6552

7351

7119

7402

3624

4238

3982

4255

R squared

0.17

0.12

  

0.19

0.12

  

Log likelihood

  

−39903.35

−36687.28

  

−21875.84

−20931.03

N of left-censored obs

  

2661

3949

  

1549

2305

Robust F (1, 62) (ln(remittances) is endogeneous)-first stage

6.60

0.22

  

4.52

0.22

  

p-value

0.01

0.64

  

0.04

0.64

  

First-stage summary stats endogenenous regressor: ln (remittances)

        

R-squared

0.09

0.09

  

0.09

0.09

  

Adj. R-squared

0.08

0.08

  

0.08

0.08

  

Partial R-squared

0.02

0.03

  

0.02

0.03

  

Robust F (1, 62) (Natwork=0)

23.92

30.10

  

11.20

15.07

  

p-value

0.00

0.00

  

0.00

0.00

  

Weak instrument test

2SLS

2SLS

  

2SLS

2SLS

  

Min.eigenvalue statistic=F-stat.“network=0”

161.64

235.94

  

75.36

116.17

  

2SLS Size of nominal 5% Wald test

16.38

16.38

  

16.38

16.38

  

IV-Tobit

        

Wald test of exogeneity

  

37.18

4.39

  

8.90

0.22

p-value

  

0.00

0.04

  

0.00

0.64

  1. Notes: * significant at 10%; ** significant at 5%; *** significant at 1%. All regressions controls for age, age squared, education level, non-wage income, household dependency ratio, indicators for being married, female household head and rural residence, residence in ‘conflict region’ and a constant term. All regressions include district-level fixed effects. Robust standard errors in parenthesis.
  2. Source: As for Table 2.