The Output Gap: Measurement, Related Concepts and Policy Implications

This paper compares the various approaches to the measurement of potential output and the NAIRU, and identifies practical policy implications. The respective advantages and disadvantages of each method – the Okun’s Law approach, the trend-fitting method, the production function approach, the simultaneous equations system method and the stochastic filter methods – are discussed in detail. It is noted that the production function approach, despite obvious advantages in comparison with the trend and the Okun’s Law methods, has the drawback of relying on exogenous or trend values of NAIRU to get equilibrium values of labour inputs. Some of the weakness of this approach is noted to be present even in the ‘fully structural’ system approach, which does not overcome the limitations of its individual component equations. In search of alternatives, the stochastic univariate and multivariate filter methods are compared to the structural methods. It is reasoned that the extended multivariate filter methods in which additional structural information replaces a priori restrictions of univariate methods may be better-equipped to produce good estimates of potential output. On the NAIRU, a recommendation of the paper is that open-economy influences should not be neglected in its estimation, so that the real exchange rate is given its due role. An implication is that in a situation where a current account surplus that leads to allows a fall in the endogeneous relative price of imports prevails, the NAIRU is lowered, allowing the possibility of beneficial policy action using such a trade-off.