Wirtschaft und Gesellschaft 42. Jahrgang (2016), Heft 4 spending separately and exclude union density from specifications (7) and (8) since it’s strongly correlated with other country-level variables (negative correlation below –0.9 for Austria). We find robust significantly negative effects of globalisation, measured by intermediate import penetration and outward FDI on the wage share in specifications (1) to (6), while the effect of the variables accounting for technological change is not robust and does not always have the expected sign: total capital stock as a ratio to value added is insignificant in all specifications while when capital is disaggregated as ICT and non-ICT capital, ICT capital services as a ratio to value added has a negative effect and non-ICT capital services as a ratio to value added has a positive effect. With regard to the control variables at the country level, we find a positive but not robust effect of union density, while social government spending turns out to be insignificant for the determination of the wage share in Austria. We furthermore include two specifications augmented by additional variables measuring migration, financialisation and person income inequality.53 Among our financialisation variables, household debt and financial income and payments are significantly negative and robust to changes in the sample when the first difference estimator is applied. Furthermore we find positive effects of the share of migrant workers in total labour force and negative effects of the Gini coefficient although the statistical significance of these two variables varies. Besides robustness tests using different estimation techniques and different measures of the wage share as described in section 3, we estimated our specifications for different sub-pools, i. e. only manufacturing or only service sectors, as well as for high- and low skilled sectors within manufacturing and services separately. This not only allows us to test the robustness of our results, but at the same time provides insights with regards to the variables that have potentially contrasting effects for manufacturing and services or across skill groups. However, since our cross sections are limited to 20 sectors for the 1-digit level estimations the estimations across skill groups can only provide indicative evidence. 5.1 Globalisation Among our globalisation variables intermediate import penetration appears to have a negative impact on the wage share across all skill groups within the manufacturing sectors given that it is negative and significant for high and low skilled sectors alike. In the services sectors our data for intermediate import penetration is limited to one sector (recycling), but our results for the total economy are robust to the exclusion of this sector. This finding is also robust when different estimation methodologies are used. Intermediate import penetration is significant in specifications (1) to (3) 576