The Effect of Early Flood Warnings on Mitigation and Recovery During the 2010 Pakistan Floods - Reducing Disaster: Early Warning Systems for Climate Change - page 254

Table 13.7 Logit: dependent variable is 1 if household has recovered/replaced household
possessions by the time of the survey

(1) (2) (3) (4) (5)

Has
recovered
household
possessions

Has
recovered
household
possessions

Has
recovered
household
possessions

Has
recovered
household
possessions

Has
recovered
household
possessions Explanatory variables

Age −0.0338 −0.0599 * −0.0353 −0.0459 −0.0438 *

(0.0291) (0.0348) (0.0253) (0.0317) (0.0263)

Female −1.733 ** −2.097 * −1.676 * −2.473 ** −1.710 *

(0.730) (1.139) (0.878) (1.025) (0.932)

Monthly income −2.63e-05 *** −2.47e-05 *** −2.62e-05 ** −2.15e-05 ** −1.83e-05 *

(9.69e-06) (8.21e-06) (1.12e-05) (8.46e-06) (1.06e-05)

Received any
assistance with
moving possessions

2.605 ***
(0.938)

2.931 ***
(1.069)

2.563 **
(1.055)

2.840 **
(1.326)

2.655 ***
(0.976)

Perceives self to be
able to recover
faster than others
from loss

0.182
(1.340)

Took any mitigation
action before 2010
floods

1.403 ***
(0.518)

Learned new mitigation
techniques due to
2010 floods

0.262
(0.676)

Time preference 1.590 *

(0.950)

Changed precautions
since 2010 floods

1.109
(0.699)

Constant 1.158 1.655 1.148 −0.0722 0.741

(1.030) (1.147) (1.006) (1.143) (1.014)

Observations 56 56 56 56 56

Robust standard errors in parentheses
*** p < 0.01; ** p < 0.05; * p < 0.1

warnings, our data show that flood warnings prompted many households to take
mitigation action, especially when given preparation time. Our findings also indi-
cate, however, that the type of warning is important, as face-to-face warnings sig-
nificantly increased mitigation, while remote warnings did not, and official warnings
from government or mosque announcements were more important for the costliest
mitigation investments. Furthermore, our results suggest that such early warnings
and mitigation actions matter for household outcomes, as those households that
received warnings and did mitigation activities had significantly lower losses of
household structure value, and taking mitigation action increased the likelihood of
having replaced lost household possessions.

Acknowledgments The authors are grateful for financial support from The British Academy, the
Lahore School of Economics, The Wharton School, and the Travelers Foundation.