The cyclical behavior of short -term and long -term job flows

Every quarter many manufacturing plants create jobs and many other manufacturing plants destroy jobs. On average in the U.S. manufacturing sector, over 5 percent of existing jobs are destroyed and 5 percent of existing jobs created every quarter. These two gross flows, job creation and job destruction, together determine the cyclical behavior of net employment changes and also describe the shifting fortunes of the individual plants that compose the manufacturing sector. As a result, understanding job creation and job destruction enables economists to better understand the causes behind movements in aggregate employment. Of fundamental importance in trying to understand these causes is the question of whether these plant-level changes in employment are long-lasting or, not.;Using time series econometric techniques, I identify the temporary and permanent components of plant level employment movements to answer the above question of whether job creation and job destruction are primarily long-lasting or not, and to investigate how short-term and long-term job flows separately account for important aspects of the cyclical behavior of aggregate job flows. I find that long-term job flows, those flows most likely to be associated with the permanent reallocation of labor across plants, are related to the cycle, with long-term job destruction rising sharply in recessions. Further, job flows across sectors are an important component of long-term job flows. I also find that short-term job flows are larger in magnitude than long-term job flows and are primarily responsible for the cyclical volatility of job flows. Motivated by the facts that both short-term and long-term job flows contribute significantly to the cyclical pattern of overall job flows and that cross sectoral job flows are an important component of long-term job flows, I build two models of job flows-one that includes within sector and across sector long-term job flows and one that include short-term and long-term job flows—and show how these models can help account for some important stylized facts about the cyclical behavior of job flows.