The skilled u-shaped Europe: is it really and on which side does it stand?

Abstract

This study estimates a quadratic sectoral wage equation for the member countries of the enlarged EU where wages are a function of each country's geographical location with respect to market size (new trade/New Economic Geography (NEG) effect), human capital (HOS endowment effect), and productivity (Ricardian effect). Wages react to these variables according to a U-shaped curve. They react negatively to market size and productivity of neighbouring countries, but react positively to skilled labour in neighbouring countries. However, due to non-linearity, further increases in these variables will induce a reversal in the outcome. Wages react more to market size, less to skill endowments, and even less to productivity, making economies of scale an important mechanism in explaining why firms can afford to pay higher wages in larger markets. In addition, EU real wages are significantly determined by country-specific characteristics other than geography that push the real wages of old EU members upward and pull the real wages of new EU members downward.

Notes

¹ For thorough surveys of empirical work within the NEG framework see, for example, Overman et al. (2003) and Head and Mayer (2004).

² As wages differ sectorally, it is implicitly assumed that human capital is sector-specific. However, no assumption is placed on inter-country mobility. Hence, access to human capital can be interpreted as a consequence of both migration and skilled content embodied in imported intermediate inputs.

³ Industrial aggregates (ISIC Rev. 2): chemicals (35), leather products (323, 324), machinery (382, 383), metals (37, 381), minerals (36), textiles and clothing (321, 322), transport equipment (384), and wood products (33); ISIC Rev. 3: chemicals (24), leather products (19), machinery (29, 30, 31), metals (27, 28), minerals (26), textiles and clothing (17, 18), transport equipment (34, 35), and wood products (20, 36). Sector codes: Chemicals (chem), Leather and Footwear (leat), Machinery (mach), Metals (meta), Minerals (mine), Textiles and Clothing (text), Transport Equipment (trans) and Wood Products (wood). Country codes: Austria (aus), Belgium (bel), Czech Republic (cze), Denmark (dk), Estonia (est), Finland (fin), France (fra), Germany (ger), Greece (gre), Hungary (hun), Ireland (ire), Italy (ita), Netherlands (ned), Poland (pol), Portugal (por), Spain (spa), Sweden (swe), United Kingdom (uk). Eastern countries such as Bulgaria, Latvia, Lithuania, Romania, Slovakia, and Slovenia are excluded from the sample because productivity data were not available.

⁴ There are no sectoral dummies as the regressions are run for each sector separately.

⁵ The eight groups of industries are Chemicals (chem), Leather and Footwear (leat), Machinery (mach), Metals (meta), Minerals (mine), Textiles and Clothing (text), Transport Equipment (trans) and Wood Products (wood).