42. Jahrgang (2016), Heft 4 Wirtschaft und Gesellschaft If not otherwise mentioned in the text our baseline results are confirmed by these robustness tests. We aim at using our variables at the most disaggregated level for which data are available. While our dependent variable is available at the two digit level of ISIC 4 (International Standard Industrial Classification of All Economic Activities), most of our explanatory variables are available at the 1-digit level with the exception of total capital stock and intermediate import penetration which are available at a 2-digit level. For this reason we switch between the two and one digit level according to the specification as explained in the next section. Estimation period differs due to data availability depending on the variables used in each specification and country. While data for the wage share at sectoral level is available for 1970-2011, the data for the FDI starts only in 1985 and detailed data on imports disaggregated as intermediate and final imports starts in 1995. The estimation period for Austria and most other countries is 1996-2010 for specifications including intermediate imports and 1986-2010 for specifications including FDI, with the exception of Denmark where our sample finishes in 2011. Furthermore, data for our measures of financialisation starts in 1995 for Austria and most other countries with the exception of France where data is available from 1970. It is mostly data on the capital stock that constrains the last year of our sample period, although for same countries, like the US, data for the sectorlevel wage share also ends in 2010. We exclude the Agriculture, Hunting, Forestry and Fishing, and Mining and Quarrying sectors as well as mostly publicly owned sectors (Public Administration and Defence, Compulsory Social Security, Education, Human Health and Social Work Activities) from the reported estimations, as these sectors’ wage setting behaviour may constitute an outlier and may not be determined by the same forces as in other sectors, but results are robust to the inclusion of these sectors. 5. Estimation Results Table 1 shows estimation results for Austria for the total sector pool, while the reader is referred to Guschanski and Onaran (2016a) for estimation differentiated by skill group and manufacturing and service sectors. We estimate specifications (1) to (3) at the 2-digit level while specifications (4) to (8) is estimated at the 1-digit level. We separately estimate the effect of increasing import penetration and outward FDI on the wage share, while controlling for union density and individual government spending at the country level in specifications (1) to (6). To avoid multicollinearity we estimate specifications with union density and government 573