†Euro Area Macroeconomic Developments, European Central Bank, Frankfurt amMain. The views expressed

Preliminary version, not to be quoted without authors ’ permission In this paper we propose a model-based approach to estimate potential output and the output gap for the euro area. The approach combines the determination of poten-tial output according to a production function with a Phillips-type relationship between inflation and the output gap, and is formulated within the general statistical class of unobserved components (UC) models. Potential output is derived from the permanent components of the factors of production consistent with stable inflation, whereas the out-put gap results from the combination of the transitory components. Various alternative statistical specifications are considered. The main advantages of this approach are that it allows to measure the contribution of the various factors of production and productivity to potential output, and that it is possible to estimate uncertainty bands. The models based on the production function approach are compared with univariate and bivariate UC models of output and inflation, in terms of inflation forecasting performance and the degree of uncertainty of estimates of the output gap. On the basis of these two criteria, results suggest the existence of a trade-off: while bivariate models tend to outperform multivariate models in terms of forecast accuracy, the latter yield estimates characterised by a lower degree of uncertainty, i.e. narrower confidence bands