Table 3

Estimated Impact of Any Husband’s Displacements on the Probability of Having an Additional Child

model: reporting:Logit Odds Ratios
(1)		Logit Marginal Effects
(2)		LPM Marginal Effects
(3)		Logit Odds Ratios
(4)		LPM Marginal Effects
(5)
Displacement year − 21.160
(0.127)		0.011
(0.009)		0.019
(0.015)		1.103
(0.145)		0.012
(0.018)
Displacement year − 11.046
(0.105)		0.003
(0.007)		0.006
(0.013)		0.983
(0.124)		− 0.005
(0.016)
Displacement year1.299***
(0.116)		0.020***
(0.008)		0.036***
(0.013)		1.194
(0.146)		0.028*
(0.017)
Displacement year + 11.167*
(0.104)		0.011
(0.007)		0.020*
(0.012)		1.125
(0.139)		0.013
(0.016)
Displacement year + 2–31.225***
(0.083)		0.015***
(0.005)		0.025***
(0.009)		1.279**
(0.148)		0.024*
(0.013)
Displacement year + 4–50.858*
(0.073)		− 0.010*
(0.005)		− 0.013*
(0.008)		0.990
(0.135)		−0.014
(0.013)
Displacement year + 6–70.745***
(0.084)		− 0.018***
(0.006)		− 0.023***
(0.008)		0.918
(0.143)		− 0.020
(0.013)
Displacement year + 8 +0.389***
(0.042)		− 0.049***
(0.004)		− 0.038***
(0.003)		0.570***
(0.094)		− 0.023*
(0.012)
Average Total Effect on the Treated− 0.145***
(0.040)		− 0.145***
(0.040)		− 0.149***
(0.040)		—− 0.098
(0.117)
Individual fixed effectsnononoyesyes
Individual observations4,1424,1424,1423,2383,238

  • Notes: The dependent variable is an indicator for having a child in a given year. All columns include age fixed effects and a quadratic in the year. The omitted category is never having a displaced husband or three or more years prior to a husband’s displacement. Standard error estimates, clustered on the individual, are displayed in parentheses. Regressions are weighted using the family weight from the last year women are observed with their first husband. The estimation of the average total effect is described in Section V.C. The odds ratios in columns 1 and 4 are eψ—an odds ratio of one indicates no effect, greater than one indicates increased odds, and less than one indicates reduced odds.

  • ↵* significant at 10 percent;

  • ↵** significant at 5 percent;

  • ↵*** significant at 1 percent