Things are further complicated by the fact that the so-called 'natural rate' of unemployment is also sometimes used to denote a NAIRU. This is confusing because the term 'natural rate' is a more general one used to describe the hypothetical unemployment rate that would emerge if certain equilibrium conditions were met. Groenewold and Hagger (1998), for instance, estimate the natural rate by constructing a model that enables them to estimate the rate of unemployment that would emerge in the absence of labour mismatch induced by differences in sectoral growth rates in the economy. Fahrer and Pease (1993) do not model the equilibrium unemployment rate itself, but the so-called Beveridge curve, which describes an equilibrium relationship between the job-vacancy rate and the unemployment rate. Holding the vacancy rate constant, shifts in this curve can be used to infer changes in factors such as the efficiency of job search, and skill and locational mismatches between job vacancies and job seekers. Shifts in the natural or equilibrium rate of unemployment can be identified in this way. These differences in conceptual framework can result in large differences in estimates of the natural rate – in 1985 the Groenewold and Hagger (1998) estimate is, at around 9 per cent, 2 percentage points higher than that obtained by Fahrer and Pease (1993).
Finally, we doubt that the emphasis on estimates of equilibrium unemployment or single-number estimates of demand pressure have, by themselves, much direct relevance to the deliberations of those who set monetary policy. While the equations in (8) are proposed by Gruen, Pagan and Thompson (1999) as an organising framework for discussing Australian research, there is a striking similarity between these equations and the way in which the Reserve Bank discusses inflationary prospects in its regular Statement on Monetary Policy. These presentations follow a regular pattern – first, there is a discussion of recent trends in consumer and producer prices, followed by a section on labour costs and inflation expectations, and a concluding section provides views on the inflationary outlook. Clearly, demand pressure has a role to play in this analysis, but it would appear that policy-makers are – whether formally or informally – analysing inflation using the whole system described by (8), rather than emphasising the NAIRU or SIRG that can be derived as a special case of it. In the view of the present writer this is as it should be, because useful information is necessarily lost in only considering special (equilibrium) cases of forecasting equations.
What can be concluded from Australian macro research on equilibrium unemployment? There is a general consensus that the NAIRU and other 'natural rate' measures of unemployment have risen significantly in Australia since the early 1970s. Although there is a good deal of uncertainty as to how much it has risen, it is generally believed that the extent to which aggregate demand-management can reduce unemployment below 5 or 6 per cent is limited. Other labour-market policies will be required to make significant inroads into unemployment.