The equilibrium rate of unemployment : a theoretical discussion and an empirical evaluation for six OECD countries

This paper examines theoretically and empirically the controversial notion of "equilibrium rate of unemployment" (ERU) which opposes proponents of the Phillips curve to those of the WS/PS models. Theoretically, this paper shows that the specification of the wage equation has a significant influence on the ERU. It also shows that an "asymmetric" wage-price setting using a Phillips curve and a price equation in level allows a clear distinction between the medium run ERU and the long run ERU that can be seen as a reconciliation between WS/PS models and the Phillips curve. Moreover, this paper calculates ERU for 6 OECD countries (the United States, Germany, France, the United Kingdom, Spain and the Netherlands) for the period 1970-late 1990's by using a structural approach. As there are little studies that test directly the impact of institutional variables in a Phillips curve, the first empirical objective is to test this effect. The results in this matter are not really conclusive. The second objective is to calculate medium run and long run ERU. The evolutions of the medium run ERU explain relatively well those of the actual unemployment rate until the late eighties. Its determinants are the slowdown of productivity growth, the 2 oil shocks, the increase of employer's social contribution and the change in the inflation target of the monetary authorities. In the nineties, the actual unemployment lies clearly above its equilibrium level suggesting that a big part of the European unemployment is due to an excess supply.