Skip to Main Content
Table 5

OLS and IV regressions of log labor force participation rate and hourly wages of 50- to 59-year-olds on log relative cohort sizes for 16- to 24-year-olds and 25- to 49-year-olds, 1984–2016

LFPHourly Wages
OLSIVOLSIV
Both Sexes 
 ln(cohort size aged 50–59 / cohort size aged 16–24) –0.020 –0.061 0.019 –0.114 
[0.021] [0.058] [0.023] [0.132] 
 ln(cohort size aged 50–59 / cohort size aged 25–49) 0.038 0.255** –0.030 –0.318** 
[0.020] [0.085] [0.035] [0.099] 
 First-stage F statistic     
   Dependent variable = ln(cohort size aged 50–59 / cohort size aged 16–24)  12.32  12.32 
   Dependent variable = ln(cohort size aged 50–59 / cohort size aged 25–49)  11.61  11.61 
 Durbin-Wu-Hausman test for endogeneity (p value)  .001  .002 
Men 
 ln(cohort size aged 50–59 / cohort size aged 16–24) –0.027 –0.103 0.023 –0.179 
[0.014] [0.088] [0.018] [0.210] 
 ln(cohort size aged 50–59 / cohort size aged 25–49) 0.030 0.190 –0.014 –0.336* 
[0.019] [0.120] [0.029] [0.157] 
 First-stage F statistic     
   Dependent variable = ln(cohort size aged 50–59 / cohort size aged 16–24)  5.83  5.83 
   Dependent variable = ln(cohort size aged 50–59 / cohort size aged 25–49)  7.76  7.76 
 Durbin-Wu-Hausman test for endogeneity (p value)  .056  3.026 
Women 
 ln(cohort size aged 50–59 / cohort size aged 16–24) –0.019 –0.011 0.013 –0.092 
[0.029] [0.100] [0.030] [0.136] 
 ln(cohort size aged 50–59 / cohort size aged 25–49) 0.074* 0.392** 0.002 –0.293* 
[0.031] [0.116] [0.040] [0.134] 
 First-stage F statistic     
   Dependent variable = ln(cohort size aged 50–59 / cohort size aged 16–24)  12.31  12.31 
   Dependent variable = ln(cohort size aged 50–59 / cohort size aged 25–49)  10.71  10.71 
 Durbin-Wu-Hausman test for endogeneity (p value)  .002  .002 
LFPHourly Wages
OLSIVOLSIV
Both Sexes 
 ln(cohort size aged 50–59 / cohort size aged 16–24) –0.020 –0.061 0.019 –0.114 
[0.021] [0.058] [0.023] [0.132] 
 ln(cohort size aged 50–59 / cohort size aged 25–49) 0.038 0.255** –0.030 –0.318** 
[0.020] [0.085] [0.035] [0.099] 
 First-stage F statistic     
   Dependent variable = ln(cohort size aged 50–59 / cohort size aged 16–24)  12.32  12.32 
   Dependent variable = ln(cohort size aged 50–59 / cohort size aged 25–49)  11.61  11.61 
 Durbin-Wu-Hausman test for endogeneity (p value)  .001  .002 
Men 
 ln(cohort size aged 50–59 / cohort size aged 16–24) –0.027 –0.103 0.023 –0.179 
[0.014] [0.088] [0.018] [0.210] 
 ln(cohort size aged 50–59 / cohort size aged 25–49) 0.030 0.190 –0.014 –0.336* 
[0.019] [0.120] [0.029] [0.157] 
 First-stage F statistic     
   Dependent variable = ln(cohort size aged 50–59 / cohort size aged 16–24)  5.83  5.83 
   Dependent variable = ln(cohort size aged 50–59 / cohort size aged 25–49)  7.76  7.76 
 Durbin-Wu-Hausman test for endogeneity (p value)  .056  3.026 
Women 
 ln(cohort size aged 50–59 / cohort size aged 16–24) –0.019 –0.011 0.013 –0.092 
[0.029] [0.100] [0.030] [0.136] 
 ln(cohort size aged 50–59 / cohort size aged 25–49) 0.074* 0.392** 0.002 –0.293* 
[0.031] [0.116] [0.040] [0.134] 
 First-stage F statistic     
   Dependent variable = ln(cohort size aged 50–59 / cohort size aged 16–24)  12.31  12.31 
   Dependent variable = ln(cohort size aged 50–59 / cohort size aged 25–49)  10.71  10.71 
 Durbin-Wu-Hausman test for endogeneity (p value)  .002  .002 

Notes: Data source is described in notes to the figures, and specification details are described in notes to Table 1. The table reports estimates of Eq. (2). Regression is weighted by average state population through the sample period. Standard errors are clustered by state. The two IVs used are total number of births for 50- to 59-year-olds divided by the total number of births for 16- to 24-year-olds by state and year and the total number of births for 50- to 59-year-olds divided by the total number of births for the 25- to 49-year-olds by state and year. We exclude certain years and states with missing birth data for the cohorts because only a handful of states started reporting births in 1915. N = 1,326. (Note that we could use more observations in this table than in Table 6 for 60- to 69-year-olds because the absence of early birth data is less of a constraint. However, we keep the samples the same in the two tables to make the estimates most comparable).

p < .10; *p < .05; **p < .01

Close Modal

or Create an Account

Close Modal
Close Modal