Estimating the time-varying NAIRU for Germany and policy implications

Abstract

This article extends Ball and Mankiw (2002) and applies the Hodrick–Prescott filter (1997) to estimate the time-varying NAIRU for Germany. The slope estimate of the Phillips curve is insignificant when the widely used lagged inflation rate π _{t −1} is selected as a proxy for the expected inflation rate. The average inflation rate in past 4 years is a better proxy for the expected inflation rate. Estimated NAIRUs with λ = 1000 show a steady rising trend. The estimated NAIRU of 9.84% with λ = 1000 in 2005 is slightly higher than the actual unemployment rate of 9.70% in 2005, suggesting that monetary easing or large deficit spending may cause the inflation rate to accelerate.

Notes

¹The related subject has been studied by Ball (2000), Ball and Mankiw (1995), Barsky (1984), Barro (1977), Blanchard and Perotti (1999), Blanchard and Summers (1986), Camarero et al. (2006), Friedman (1968), Hatton (2002), Lucas (1996, 1973), Phelps (1968, 1967), Staiger et al. (1997), Stock and Watson (1999), Summers (1986), and others.