Random or non-random matching? Implications for the use of the UV 234 curve as a measure of matching effectiveness

The instability of the Beveridge curve in Britain since the mid-50s has been interpreted as revealing a detioration of labour market effectiveness in matching vacancies to unemployed job-seekers. This paper repeats the stability analysis of the matching technology, having relaxed some of the restrictions generally imposed in deriving a Beveridge curve. We explore the robustness of the curve under an essentially continuous-time non-random matching technology. The non-random nature of the matching process is derived from a theoretical standpoint of stock/flow matching technology that excludes interactions between the existing stocks of unmatched unemployed and vacancies. The empirical analysis based on this premise leads to a re-thinking of standard conclusions on matching effectiveness.