Entrepreneurship and Prior Work Experience: The Role of Firm and Peer Quality

• Applicant: Dr. Martin Friedrich
• Subject Area: Economic Policy, Applied Economics ; Statistics and Econometrics
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 558011730

This project seeks to expand our understanding of the determinants of successful entrepreneurship by exploring, for the first time, the influences of previous employer productivity and previous workplace peer productivity. Existing literature indicates that most successful entrepreneurs gain several years of labor market experience before starting a business. Building on a model of knowledge spillover at the workplace, I propose that work experience is more valuable for future entrepreneurs when acquired in highly productive firms and alongside highly productive peers. The research addresses two questions: (1) How does the productivity of previous employers and workplace peers influence the likelihood of transitioning from employment into entrepreneurship? (2) How does the productivity of an entrepreneur’s previous employer and workplace peers affect the success of her newly established business? Additionally, the study aims to differentiate the effects of previous employer size and productivity, which have likely been conflated in past research, as I demonstrate with preliminary evidence. I will also explore mechanisms that may explain the outlined relationships, such as the higher incidence of training in more productive firms or the opportunity to build a network of more productive individuals. The analysis will utilize a novel dataset that I will create: the linked MEP-IEB data. This dataset will include the universe of German entrepreneurs from the Mannheim Enterprise Panel (MEP) and their social security records from the IAB’s Integrated Employment Biographies (IEB). Furthermore, I will incorporate the social security data of all workplace peers the entrepreneurs were exposed to during their employment. The analysis comprises two main steps. First, I will decompose wages into worker and firm fixed effects by estimating the AKM model on the full IEB data, yielding precise measures of each worker’s (and future entrepreneur’s) productivity, as well as the productivity of her former employer and peers. These variables are typically unobserved in common data sources like surveys and have not been previously analyzed as determinants of entrepreneurship. Second, I will regress entrepreneurship entry and success simultaneously on individual productivity, employer productivity, and workplace peer productivity. Entrepreneurship success outcomes will include metrics such as business survival rates, revenue, and employment growth. My novel methodology will not only enable the measurement of previously unobserved predictors of entrepreneurship, but it will also address the endogenous sorting of higher-productivity individuals into more productive and larger workplaces, as well as more productive peer groups—the key estimation challenge in the literature on peer effects. The estimates I aim to provide will be unbiased by this sorting and thus internally valid. They will also be representative of the average entrepreneur and therefore externally valid.

• DFG Programme: Research Grants
• Co-Investigator: Professor Dr. Michael Oberfichtner