A factor-augmented new Keynesian Phillips curve for the European Union countries

In this paper, a factoraugmented version of the hybrid New Keynesian Phillips curve (NKPC) is assessed using a data set comprised of a large panel of European Union (EU) member countries. The factoraugmentation is natural given that countrylevel inflati on rates are highly comoving. The presence of unattended common factors is important because it raises the issue of omitted variables bias, as the real marginal cost, which is a regressor of the NKPC, is likely to load on the same factors as inflation. On e possibility here is to employ the regular instrumental variables approach. However, if the external instruments are subject to the same factors as those in the error term of the NKPC, the instruments would be invalid and the approach would therefore be i nappropriate. We propose a novel econometric approach to estimate the hybrid NKPC, which allows for very general forms of factor dependencies and endogeneity, and should as a result lead to improved identification. Our main findings provide support for the hybrid NKPC when the presence of unknown common factors as well as external instruments are accounted for, although the results differ depending on the countries included in the estimation. More specifically, the evidence is stronger when the full sample of EU or Euro Area countries is used, rather than solely the new EU member countries which joined the EU in 2004 or later.