Predicting Firm Creation in Rural Texas: A Multi-Model Machine Learning Approach to a Complex Policy Problem

Abstract

What factors predict firm creation in rural America? Policymakers asking this question face two obstacles. First, research on firm creation centers on high-tech, urban firms. Second, entrepreneurship research stretches across disciplines, often using econometric methods to identify the effect of a specific variable, rather than comparing the predictive importance of multiple variables. In this paper, we apply three machine learning methods (subset selection, lasso, and random forest) in addition to linear regression to a novel dataset to examine what social and economic factors are predictive of firm growth in rural Texas counties from 2008-2018.

Results suggest that some factors commonly discussed as promoting entrepreneurship (e.g., access to broadband and patents) may not be as predictive as socioeconomic ones (age distribution, ethnic diversity, social capital, and migration patterns). We also find that the strength of specific industries (oil, wind, and healthcare) predicts firm growth, as does the number of local banks. Most factors predictive of firm growth in rural counties are distinct from those in urban counties, supporting the argument that rural entrepreneurship is a distinct phenomenon worthy of distinct focus. We also discuss how this multi-model approach can offer initial, focusing guidance to policymakers seeking to address other complex policy problems.

Keywords: Entrepreneurship, rural entrepreneurship, entrepreneurship policy, machine learning

JEL Classification: R58, R5, R11, O38, O3, O1, L26, L53