The Real-time Forecasting Performance of Phillips Curves

Abstract

Analysts typically use a variety of techniques to forecast inflation. These include both ‘bottom-up’ approaches, for near-term forecasting, as well as econometric methods (such as mark-up models of inflation, which have been found to perform quite well for Australia – see de Brouwer and Ericsson (1998)). One of the econometric approaches to inflation forecasting which is sometimes considered is the use of Phillips curves based on estimates of the output gap. This paper suggests, however, that the real-time capacity of such Phillips curves to forecast inflation is limited, relative even to such simple benchmark forecasting approaches as an autoregressive (AR) model of inflation or a random walk assumption. It appears that the lack of precision with which output-gap-based Phillips curves can be estimated in real time limits their usefulness as a means of forecasting inflation in isolation. Phillips curve-based forecasts may, however, perform a little better than AR model-based ones in at least predicting whether inflation will increase or decrease from its current level. Moreover, combining Phillips curve-based forecasts with those from simple, alternative approaches does seem to offer some scope for improving the real-time forecast accuracy of the latter. These observations suggest that, in spite of their generally disappointing performance as a means of forecasting inflation in isolation, output-gap-based Phillips curves may continue to be useful in real time – as a tool for conditioning gap estimates within a multivariate filtering framework, and as a possible complement to other, alternative inflation forecasting approaches.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 16 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.