Table 2.

Contrast estimates for party control in state legislatures and the governor’s party on logged state IMR, NMR, and PNMR

Contrast	Estimate	LnIMR	LnNMR	LnPNMR
	Legislature Transitions
All Dem to Lower House Rep	Coeff	.033	.034	.033
	p-value	.039	.080	.056
	95% CI	.002 – .065	−.004 – .072	−.001 – .067
All Dem to Upper House Rep	Coeff	.025	.007	.062
	p-value	.103	.752	.000
	95% CI	−.005 – .056	−.035 – .048	.033 – .091
All Dem to Both Houses Rep	Coeff	.042	.019	.081
	p-value	.007	.292	.000
	95% CI	.012 – .072	−.017 – .056	.039 – .124
	Governor Transitions
All Dem to Governor Rep	Coeff	.001	−.000	.006
	p-value	.859	.967	.567
	95% CI	−.014 – .017	−.018 – .018	−.015 – .027
	Legislatures and Governor Transitions
All Dem to All Rep	Coeff	.043	.019	.087
	p-value	.003	.202	.000
	95% CI	.015 – .071	−.010 – .048	.041 – .133

Note: All effects are derived from estimates of equation 1. The standard errors are clustered at the state level to account for arbitrary forms of within state serial correlations. All regressions are weighted using time-constant, averaged total number of births in the states. All states included with the exception of Nebraska (n=49), which has a unicameral system. Controls are females average age, unemployment rate, the birth rate, and time and state fixed effects. Boldface indicates statistical significance (p<0.05).