Skip to main content

Table 3 The impact of the reform, difference-in-differences estimates (in days)

From: Can a search model predict the effects of an increase in the benefit duration? Evidence from the Portuguese unemployment insurance reform

 

Treatment groups

 
 

15–24

30–34

40–44

 

(1)

(2)

(3)

(4)

(5)

(6)

After

−15.1

16.8

−13.2

34.4

−13.2

21.2

 

(1.6)

(3.0)

(2.9)

(5.0)

(3.4)

(6.4)

Treat

−56.8

−38.9

−95.1

−66.7

−34.6

−53.7

 

(2.2)

(2.7)

(3.7)

(4.6)

(4.8)

(6.1)

After × treat

47.9

42.8

76.1

74.4

114.9

110.2

 

(2.5)

(2.5)

(4.1)

(4.1)

(5.4)

(5.3)

Log of previous wage

 

1.8

 

34.3

 

57.2

  

(1.2)

 

(1.6)

 

(2.1)

Female

 

0.4

 

14.4

 

34.5

  

(1.0)

 

(1.7)

 

(2.3)

Age

 

−0.1

 

31.0

 

−9.3

  

(2.9)

 

(7.6)

 

(11.7)

Age2

 

0.1

 

−0.4

 

0.2

  

(0.1)

 

(0.1)

 

(0.1)

Dummies

      

Regional

No

Yes

No

Yes

No

Yes

Month of unemployment

No

Yes

No

Yes

No

Yes

Year of unemployment

No

Yes

No

Yes

No

Yes

Observations

60,449

60,449

52,738

52,738

40,786

40,786

  1. The group 25–29 is the control group for the treatment group 15–24. The group 35–39 is the control group for the treatment groups 30–34 and 40–44. Age at the beginning of the subsidized period. The coefficient on After × treat is the difference-in-differences estimate of the impact of the increase in the maximum duration of benefits on the unemployment duration of benefit recipients. Standard deviations in parentheses