The Unreliability of Output-Gap Estimates in Real Time

Abstract

Most research on monetary policy assumes availability of information regarding the current state of economy, at the time of the policy decision. A key challenge for policy-makers is to find indicators that give a clear and precise signal of the state of the economy in real time—that is, when policy decisions are actually taken. One of the indicators used to asses the economic condition is the output gap; and the estimates of output gap from real time data misrepresents the true state of economy. So the policy decisions taken on the basis of real time noisy data are proved wrong when true data become available. Within this context we find evidence of wrong estimates of output gap in real time data. This is done by comparing estimates of output gap based on real time data with that in the revised data. The quasi real time data are also constructed such that the difference between estimates of output gap from real time data and that from quasi real time data reflects data revision and the difference between estimates of output gap from final data and that from quasi real time data portray other revisions including end sample bias. Moreover, output gap is estimated with the help of five methods namely the linear trend method, quadratic trend method, Hordrick-Prescott (HP) filter, production function method, and structural vector autoregressive method. Results indicate that the estimates of output gap in real time data are different from what have been found in final data but other revisions, compared to data revisions, are found more significant. Moreover, the output gap measured using all the methods, except the linear trend method, appropriately portray the state of economy in the historical context. It is also found that recessions can be better predicted by real time data instead of revised data, and final data show more intensity of recession compared with what has been shown in real time data. JEL Classification: E320 Keywords: Data Uncertainty, Measurement Uncertainty, Output Gap, Business Cycle, Economic Activity

Impact of Monetary Policy Shocks on the Output Gap in Nigeria

“…Despite the criticisms, the current study aligns with the position that an empirical model should at least replicate co-movement in the data generating process of the group of variables and avoid methods of estimating the output gap that are sensitive to end-sample observations (Zeng, 2001;Satti & Malik, 2017). Moreover, the accuracy or otherwise of the measurement of the potential output will be absorbed in the monetary policy random shock as a measurement error, and this can be tested in the course of evaluating the model efficiency (Cochrane, 2011).…”

“…In order to ensure the accurate approximation of potential output, the emerging literature like Borio, Disyatat, and Juselius (2013) and Scalone (2014), further refined it to include financial cycle and liquidity information, while others like Palumbo (2014) propose the setting of policy targets on unemployment rather than potential growth. Potential output is usually generated by de-trending or removing the cyclical component of the actual output using five methods: the linear trend method, quadratic trend method, Hordrick-Prescott (HP) filter, production function, and structural VAR (Satti & Malik, 2017).…”

Nexus Between Asset Class Volatility and the Output Gap in Nigeria: A Bayesian Var Approach

“…The output gap would serve as a better objective than economic growth because of the estimation of the potential gap which the normal economic growth, doesn't possess. (Satti and Malik, 2017) believe that the five approaches of estimating output gap are the Hordrick-Prescott (HP) filter, production function, linear trend method, quadratic trend method, and structural VAR are typically used to de-trend, or remove the cyclical component of the actual output, to generate potential output. Pedro and Adesina-Uthman.…”