USING FILTERS AND PRODUCTION FUNCTION METHOD FOR ESTIMATING OUTPUT GAP AND POTENTIAL GDP FOR ROMANIA

Univariate filters and production function methods are among the most widely recommended methods for estimating potential output and output gap. In this paper we apply the production function method on annual time series covering years 1991-2012 and employ the Hodrick-Prescott filter, Baxter- King and Christiano-Fritzgerald band- pass filters on quarterly data of real GDP covering the period 2000- 2012 to extract potential GDP and output gap for Romania. The production function we use is Cobb-Douglas type with constant returns to scale and Hicks neutral technological progress and the potential labor is computed considering both constant NAIRU, estimated with Ball-Mankiw's method (2002), and variable NAIRU resulted from a version of Phillips curve expectation augmented model (Greenslade, Pierse (2003)). Potential output and output gap are calculated using two different values for production elasticity with respect to capital, and both constant and variable NAIRU. Regarding the empirical applications of the filters, we conclude that the CF filter performed better, and that all the methods used show the economic crisis that affected the Romanian economy during 2008-2010.