Employing the Getis-Ord local indicator of spatial autocorrelation and a combination of publicly available and restricted-access data on new business formation in Sweden, we document a novel set of results to extend the existing discourse on native and immigrant entrepreneurship. First, we find that the weak global interconnectedness in the data obscures the otherwise strong underlying local dynamics, with Swedish startups forming spatial clusters around the country’s largest economic hubs of Stockholm, Gothenburg, and Malmö. Second, we find that spatial dependence appears slightly stronger among immigrants than among natives. We rationalize our observations using the selection and agglomeration theories of economics and the disadvantage and cultural-capital theories of entrepreneurship research.
For the insightful comments and enthusiastic methodological assistance, we thank Keisuke Kondo, Fredrik Garli, Kusum Mundra, Kingsley Haynes, Karl Wennberg, and the audiences at the Western Regional Economic Association Conference and Eastern Economic Association Conference. We declare that the views expressed here are our own and do not reflect those of our affiliated institutions or of any agencies in any way. Swedish business formation data are obtained from the Swedish Agency for Growth Policy Analysis. Data on immigrant entrepreneurship are provided by Statistics Sweden via a special written agreement. Interested researchers are welcome to contact us for guidance on obtaining the data, or to contact the above agencies directly. Software code to reproduce all our tables and figures are available upon request. None of the authors has any conflict of interest to declare.