A survey of alternative methodologies for estimating potential output and the output gap

In this paper, we survey some techniques proposed in the literature to measure potential output. Given the reported shortcomings of univariate approaches and mechanical filters, we focus on three simple multivariate methodologies: the multivariate Beveridge-Nelson methodology (MBN), Cochrane's methodology (CO), and the structural VAR methodology with long-run restrictions (LRRO). These methodologies are presented and then applied to U.S. data. The results show that the LRRO estimates provide significant evidence that permanent shocks have more complex dynamics than the random walk assumed in CO and MBN approaches. As in other studies, estimates of the output gap remain imprecise.