摘要:This study uses county-level manufacturing data (1985-1994) from Pennsylvania to investigate the spatial variation of relationships among manufacturing firm births, deaths, and certain local economic variables suggested by the theoretical literature. To do this, a panel data vector autoregressive method is employed, which, unlike the more customary models of studies of this nature, allows for, among other things, full interdependence among all variables and makes adjustment for omitted-variable bias related to "area-specific" effects. The results, which in some cases confirm and in others contradict those of some of the leading studies on the issue, reveal a rich network of interactions among the variables that indicate that firm births and deaths and local economic conditions are mutually driven. The main conclusion drawn from the results is that firm birth rates and, hence, economic growth, are highest in those counties with: (i) a relatively high proportion of small firms; (ii) lower unemployment; (iii) a growing market demand; (iv) a lower-than-average school property tax rate; (v) a higher-than-average quality of public education; and (vi) some urban agglomeration advantages. Some of the implications of the findings for local economic development strategy are considered.