The “natural rate of unemployment,” upon which so much of modern macroeconomics is based, may be much more flexible than is commonly believed, say American economists Robert Hall and Marianne Kadlyak, who have studied changes in this indicator during recovery from economic downturns. If their assumptions are further confirmed, the “Phillips curve” – the assumed relationship between inflation and unemployment, which is quite problematic in economic theory, will go into the archives.
The “natural rate of unemployment” (NUR) in the neo-Keynesian branch of economic theory is not a directly observable, but intuitive indicator on which quite a lot of macroeconomists’ calculations are based. Milton Friedman and Edmund Phelps, who first used the concept of the EUB in 1968, insisted that it should be understood as the level of unemployment at which, in the achieved equilibrium, it itself does not affect inflation. In theory, the properties of the EUB (it is described as a very conservative indicator, depending on the history of a particular labor market, its structure, cultural characteristics of society, and therefore, as is commonly believed, can change over decades) are described by the “Phillips curve” – empirically identified in the 20th century codependence of the price level and current unemployment. This, in turn, through the connections between unemployment and output in the economy, is the basis for most neo-Keynesian macromodels.
Robert Hall of Stanford and Marianne Kadlyak of the Federal Reserve, in a preprint of an article for NBER, “The Active Role of the EUB During Cyclical Economic Recovery,” question the properties of the EUB, the very definitions of the “natural level” of employment, and the nature of the “Phillips curve.” The shape of the curve has been a subject of debate among economists since the 2000s, with the widespread belief that in modern economies it is flat. Hall and Kadlyak, based on their studies of the behavior of unemployment and estimates of the EUR in the ascending phase of economic cycles, show that the theoretically calculated “natural rate” is highly correlated with the current level of unemployment – the EUR in their estimates is flexible and changes quickly. The authors suggest that the methods for calculating the EUR in central bank models, which almost always estimate the slope of the Phillips curve to be near a constant value, are explained by problems in the methods of regression analysis in which the EUR appears to be de facto stable. The authors give their own definition of the “true EUB” – this is equilibrium unemployment when inflation is close to the “anchored” levels of inflation expectations, which, other things being equal, does not create independent inflationary pressure. That is, Hall and Kadlyak suggest that the flat “Phillips curve” is an illusion, the movements of the EUB themselves, as defined by the authors, matter, inflation and unemployment are in fact related.
The ideas of Hall and Kadlyak, first expressed in works of 2022, are purely econometric; they are quite far from discussing the structure of the labor market, the speed and features of adaptation of this market to the dynamics of prices that change during crises. At the same time, the assumption about the stability and conservatism of the EUB is usually transferred without basis to the description of the behavior of employers and employees in labor markets – the flat shape of the “Phillips curve” is usually explained precisely by changes in recent decades in sectoral employment, a reduction in the share of wages in GDP, and the development of new institutions in employment. However, the authors’ ideas, if accepted, should accelerate research in this area, and not just in macroeconomic models: the dynamism of the “natural rate of unemployment” in relation to price dynamics indicates that labor market flexibility is greatly underestimated – and The target of full employment, usually chosen by governments as the most important goal of economic policy, generally needs clarification and additional definitions.