Phillips curve or wage curve? Evidence from West Germany, 1980-2004

"This paper reconsiders the West German wage curve using the employment statistics of the Federal Employment Services of Germany (Bundesanstalt für Arbeit) over the period 1980-2004. This updates the earlier study by Baltagi and Blien (1998) by 15 years for a more disaggregated 326 regions of West Germany. It is based on a random sample of 417,426 individuals drawn from the population of employees whose establishments are required to report to the social insurance system. We find that the wage equation is highly autoregressive but far from unit root. This means that this wage equation is not a pure Phillips curve, nor a static wage curve, and one should account for wage dynamics. This in turn leads to a smaller but significant unemployment elasticity of -0.02 up to -0.03 rather than the -0.07 reported in the static wage curve results reported by Baltagi and Blien (1998)." (author's abstract