DP14132 Beveridgean Unemployment Gap
|Author(s):||Pascal Michaillat, Emmanuel Saez|
|Publication Date:||November 2019|
|JEL(s):||E24, E32, J63, J64|
|Programme Areas:||Labour Economics, Public Economics, Monetary Economics and Fluctuations|
|Link to this Page:||cepr.org/active/publications/discussion_papers/dp.php?dpno=14132|
This paper measures the unemployment gap (the difference between actual and efficient unemployment rates) using the Beveridge curve (the negative relationship between unemployment and job vacancies). We express the unemployment gap as a function of current unemployment and vacancy rates, and three sufficient statistics: elasticity of the Beveridge curve, recruiting cost, and nonpecuniary value of unemployment. In the United States, we find that the efficient unemployment rate started around 3% in the 1950s, steadily climbed to almost 6% in the 1980s, fell just below 4% in the early 1990s, and remained at that level until 2019. These variations are caused by changes in the level and elasticity of the Beveridge curve. Hence, the US unemployment gap is almost always positive and highly countercyclical---indicating that the labor market tends to be inefficiently slack, especially in slumps.