Clusterization of regional labour markets in view of NAIRU estimates

Changes in the labor market have a significant impact on economic processes, including monetary policy. However, decision-making in this policy is complicated by the significant regional heterogeneity of not only employment, wages, migration flows, but also inflationary processes. The aim of this study is to analyze the heterogeneity of regional labor markets in relation to the business cycle. Estimates of the non-accelerating inflation rate of unemployment (NAIRU) derived from unobservable component models are an integral part of the modern Phillips curve and allow this relationship to be established. As a result of evaluating the models based on quarterly data for the period 2011-2018, conclusions were drawn about the existence of a significant relationship between the unemployment gap and inflation for 52 Russian regions. Taking into account the NAIRU estimates, the use of modern cluster analysis algorithms made it possible to identify three groups of regional labor markets that clearly differ in socio-economic indicators, which made it possible to identify the main problems in the social and demographic spheres of the territories in relation to their economic development. Methods of k-means and spectral clustering have been identified as the most suitable for grouping regions, suggesting that the resulting regional clusters have a hyperspherical shape. The results obtained can be useful for supporting decision-making on monetary policy in terms of improving the quality of economic analysis, taking into account the specific features of regional labor markets.