Table 3.
Effect of Unemployment Rate on Monthly SSDI Claims, Base Model vs. Distributed Lag Specification.
| Number of Applications |
Number of Allowances, All Decision Levels |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Base model (1) |
Distributed lag model: AIC/BIC optimizing lag (2) |
Base model (3) |
Distributed lag model: AIC/BIC optimizing lag (4) | ||||||||
| Contemporaneous | 4,455*** | (978) | 3,642*** | (345) | 1,860*** | (555) | 2,195*** | (286) | |||
| L1 | 1,319*** | (101) | 423* | (245) | |||||||
| L2 | 31 | (184) | −415* | (246) | |||||||
| L3 | −528*** | (184) | −318 | (276) | |||||||
| L4 | −615*** | (133) | |||||||||
| L5 | −444*** | (106) | |||||||||
| L6 | −180 | (129) | |||||||||
| L7 | 57 | (144) | |||||||||
| L8 | 196 | (129) | |||||||||
| L9 | 210** | (105) | |||||||||
| L10 | 119 | (131) | |||||||||
| L11 | −8 | (183) | |||||||||
| L12 | −60 | (184) | |||||||||
| L13 | 126 | (96) | |||||||||
| L14 | 755*** | (329) | |||||||||
| Total Effect | 4,455*** | (978) | 4,619*** | (96) | 1,860*** | (555) | 1,885*** | (50) | |||
| Mean DV | 1,35,945 | 1,35,945 | 73,751 | 73,751 | |||||||
| Pct. Change | 3.3 | 3.4 | 2.5 | 2.6 | |||||||
| Elasticity | 0.25 | 0.26 | 0.19 | 0.19 | |||||||
| Sum of Positive Lags | 6,455*** | (503) | 2618*** | (193) | |||||||
| Sum of Negative Lags | −1,836*** | (472) | −733*** | (181) | |||||||
| Percent Shifted Forward | 28%*** | (5.2%) | 28%*** | (4.9%) | |||||||
| AIC | 54,534 | 54,475 | 49,802 | 49,761 | |||||||
| BIC | 54,540 | 54,512 | 49,808 | 49,786 | |||||||
| R-squared | 0.987 | - | 0.987 | - | |||||||
| N | 3,825 | 3,825 | 3,825 | 3,825 | |||||||
Notes:
***
Standard errors in parentheses. Significant at the 1% level.
**
Significant at the 5% level.
*
Significant at the 10% level. In all models, N = 3825 state-month observations of the number of SSDI claims filed. Estimates are the coefficient on the number unemployed multiplied by the number of workers equivalent to one% of the national labor force, and thus indicates the number of additional applications of a given outcome type induced by a one-point increase in the unemployment rate.