Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We estimate the effects of industrial localization on the spatial persistence of employment in the software industry, using establishment data from Texas for the 2000–2006 period. Locations with an initial concentration of software employment retain an excess number of employees, beyond that expected from job turnover and job persistence at the establishment level. This is not driven by differential establishment growth or survival, but it is due to (a) the retention by establishments in a location of jobs lost by other establishments in that location, and (b) the propensity of software establishments to enter in locations with prior software establishment presence. These findings are more consistent with labor channel effects than with disembodied knowledge spillovers.