Improving the reliability of real-time output gap estimates using survey forecasts

Measuring economic activity in real-time is a crucial issue both in applied research and in the decision-making process of policy makers; however, it also poses intricate challenges to statistical filtering methods that are built to operate optimally when working with an infinite number of observations. In this paper, we propose and evaluate the use of survey forecasts for augmenting such methods, in order to reduce the end-of-sample uncertainty that is observed in the resulting gap estimates. We focus on three filtering methods that are employed commonly in business cycle research: the Hodrick-Prescott filter, unobserved components models, and the band-pass filter. We find that the use of surveys achieves powerful improvements in the real-time reliability of the economic activity measures associated with these filters, and argue that this approach is preferable to model-based forecasts due to both its usually superior accuracy in predicting current and future states of the economy and its parsimony.