Skills, earnings, and employment: exploring causality in the estimation of returns to skills

Large-scale Assessments in Education

An IERI – International Educational Research Institute Journal

Table 5 Robustness of returns to ICT skills in sample of municipalities without own main distribution frames.

Dependent variable: log gross hourly wage
	OLS (municipality level)		2SLS (second stage)		OLS
	(1)	(2)	(3)	(4)	(5)
ICT skills	.209*** (.079)	.271*** (.087)	.405*** (.204)	.521*** (.213)
Municipality characteristics	X	X	X	X	X
Individual characteristics		X		X	X

Regressions weighted by sampling weights (giving same weight to each municipality). Least squares estimations with variables aggregated at the municipality level in columns (1)–(2); two-stage least squares estimations in columns (3)–(4); least squares estimations in column (5). Sample: West German employees aged 20–65 years, no first-generation immigrants, only municipalities without an own main distribution frame (MDF). ICT and numeracy skills are standardized to SD 1 within country. Threshold: binary variable indicating whether a municipality is more than 4200 m away from its MDF (1 lower probability of DSL availability), and 0 otherwise. Distance calculations are based on municipalities’ geographic centroid. Municipality characteristics are unemployment rate in 1999 (i.e., share of unemployed individuals in the working-age population aged 18–65 years) and population share of individuals older than 65 in 1999. Individual characteristics are quadratic polynomial in work experience and gender. Column (5) controls for ICT skills. Robust standard errors, adjusted for clustering at the municipality level, in parentheses. Significance levels: * p < 0.10, ** p < 0.05, *** p < 0.01