Business Age, another Source of Employment Growth

This paper contributes new evidence using recent administrative data to advance what is known about an important, but still controversial, topic among economists—the expected contributions of old and new businesses to employment growth. Improved understanding of the relative place-specific contributions of old and new businesses to employment growth is needed to help policy makers make efficient and effective decisions to hasten and sustain recovery from the current recession. While not an either-or decision, federal, state and local governments are encouraged by competing advocacy groups to favor new incentives for hiring by existing businesses, or to prefer offering of incentives for new business start-ups. Successful employment growth policies will remain complex, place specific and contentious, but new recent information about the origins of employment growth can add a common starting point for consideration of available policy options. Our analysis uses Quarterly Census of Employment and Wages (QCEW) data at the county level for one state, covering business employment dynamics between January 2004 and September 2008. New transformations of business establishment start and end dates and employment sizeclass designations are introduced. Innovative transformations of published local unemployment rates incorporate both magnitude and propensity. Based on estimates from multistage econometric modeling with inverse use of instrument variable technique and hierarchical panel data with autocorrelation correction, policy relevant findings include a positive employment growth advantage for older business establishments in more than half of local industry sectors. Established large businesses are typically found in a different spatial pattern and lower count than the distribution of smaller, often newer establishments. This means that the former locales are more dependent upon the success of fewer businesses, including merger and acquisition transactions. Local unemployment rates impact employment growth. However, it is the unemployment rate trend that consistently associates with employment growth, not the magnitude of the rate per se. Local demographics matter too, for both labor supply and consumer behavior reasons. The local level administrative data source, the unique transformations and measurement, and the model specifications and statistical results presented in this paper offer a new guide for future researchers and policy decision-makers. Local emergence from the U.S. recession that began in December 2007 will be uneven in timing, magnitude and composition. The design and implementation of successful interventions to accelerate and advance local employment growth will depend in part on an improved understanding of the current local economic composition. 2 We thank Maryland Department of Labor, Licensing and Regulation (DLLR), Office of Workforce Informat ion and Performance staff for their patient explanation of data field definitions and appropriate data processing steps for