Table A.1

Labor Supply Elasticity to the Firm by Task Intensities (TI)—Exponential Model

	RTI	NRMTI	NRCTI
Separation rate to employment
log wage (ϵ^e _sw mean TI)	-1.454^***	-1.376^***	-1.420^***
	(0.011)	(0.011)	(0.011)
log wage × TI	-0.333^***	-0.195^***	0.383^***
	(0.009)	(0.009)	(0.009)
ϵ^e _sw (high TI)	-1.787	-1.571	-1.037
ϵ^e _sw (low TI)	-1.121	-1.181	-1.803
Observations	2,998,063	2,998,063	2,998,063
Separation rate to nonemployment
log wage (ϵⁿ _sw mean TI)	-1.849^***	-1.802^***	-1.816^***
	(0.008)	(0.008)	(0.008)
log wage × TI	-0.255^***	-0.106^***	0.266^***
	(0.007)	(0.007)	(0.007)
ϵⁿ _sw (high TI)	-2.104	-1.908	-1.550
ϵⁿ _sw (high TI)	-1.594	-1.696	-2.082
Observations	5,460,312	5,460,312	5,460,312
Hiring probability from employment
log wage	1.725^***	1.724^***	1.717^***
	(0.010)	(0.010)	(0.010)
log wage × TI	-0.114^***	-0.098^***	0.160^***
	(0.008)	(0.008)	(0.009)
ϵ_θw (high TI)	1.052	1.085	1.045
ϵ_θw (mean TI)	1.066	1.069	1.082
ϵ_θw (low TI)	1.059	1.028	1.104
Observations	979,514	979,514	979,514
Share of hires from employment (θ)
with high TI	0.347	0.333	0.443
with mean TI	0.382	0.380	0.370
with low TI	0.424	0.436	0.291
Firm-level labor supply elasticity (ϵ_Lw)
with high TI	2.729	2.282	1.314
with mean TI	2.086	1.947	2.008
with low TI	1.455	1.625	2.700

Source: SIAB and BHP 1985–2014. Authors’ calculations.
Notes: Clustered standard errors at the person level in parentheses. Routine task intensity (RTI), nonroutine manual task intensity (NRMTI), and nonroutine cognitive task intensity (NRCTI) are standardized with mean zero and standard deviation one. Thus, for instance, workers with low RTI are workers with RTI one standard deviation below the mean, and workers with high RTI are workers with RTI one standard deviation above the mean. Same control variables as in Table 2. Significance:
* p < 0.10,
** p < 0.05,
↵*** p < 0.01.