Academic journal article National Institute Economic Review

Using Real-Time Output Gaps to Examine Past and Future Policy Choices

Academic journal article National Institute Economic Review

Using Real-Time Output Gaps to Examine Past and Future Policy Choices

Article excerpt

Alternative measures of the UK output gap are considered for 1984-2007. The real-time series is strongly affected by the rolling-time estimation of the trend, and produces a picture of the business cycle which is not consistent with contemporary perceptions of the large fluctuations of the late 1980s and early 1990s. A new, 'nearly-real', measure developed here may be better for estimating historical reaction functions. In the context of the current recession, none of these mechanically derived measures of the output gap are useful. Policymakers should make careful estimates of the likely fall in potential output on the basis of other information.

Keywords: Output gap: real-time; quasi-real; nearly real: Taylor rule JEL Classifications: E52; E58

I. Introduction

The real-time output gap - roughly, the difference between actual and estimated potential output as observed in real time--is a convenient way of summarising the state of the economy. It can be and has been used in analyses of how policy was made in the past, notably in regressions of Taylor rules and other reaction functions. It can also be used as an input into the making of current and future policy. Indeed, much of the contemporary debate on the use of real-time economic data has been concerned with efficient estimation of recent, current and future economic conditions--so-called 'backcasting', 'nowcasting' and 'forecasting' (e.g. Diron, 2008; Garratt et al, 2008; Giannone et al, 2008; Mitchell, 2009; Planas and Rossi, 2004; Castle, Fawcett and Hendry, 2009), in which the policymaker makes use of both the latest statistical data and information on how such data have been revised by the statistical authorities in the past.

In trying to understand the behaviour of policymakers in the past, it seems clearly desirable to use exactly the data they had available at the time when they made their decisions; the use of different data might lead to incorrect estimates of their behaviour: However, whilst the logic of using real-time data seems clear, a number of authors, including Orphanides and van Norden (2002, 2005) in the context of the US economy, Runstler (2002) for the Euro area, and Bernhardsen et al (2005) for Norway--all of whom have been concerned mainly with future policymaking--have raised concerns about the unreliability of real-time measures of the output gap in particular. Part of the problem lies with the effect of subsequent revisions to output data, but the greater concern is with the implications of using real-time data releases for the contemporaneous estimation of unobservable potential output, around which the output gap is measured. In fact, with few exceptions, the literature on estimating Taylor rules has tended not to address this issue (see, for example the work by Clarida, Gali and Gertler, 1998, 2000; Nelson, 2000; Adam, Cobham and Girardin, 2005; Hayo and Hoffman, 2006). (1)

In this paper we discuss the properties of alternative measures of the output gap in the UK over the past 25 years, in particular their ability to identify the timing of turning points in the business cycle. We then consider how policymakers might take account of such measures as the UK begins to move out of the current recession. We suggest that the use of real-time output gap measures to analyse the historical behaviour of the monetary authorities--for example in the estimation of Taylor Rules--may result in a seriously misleading representation of the actual policymaking process, especially at times when the economy is subject to large variations in actual output. We show that the underlying trends of output generated by real-time estimation are not prima facie plausible, and we show that the turning-points in real-time output gaps have a much less good fit than the full-sample turning-points with the perceptions of policymakers at the time. We also introduce a new output gap measure which may sometimes provide a more accurate historical representation of the authorities' assessment of the true output gap than conventional real-time or full-sample measures, especially in relatively turbulent times. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.