Essays on Entrepreneurship: The Role of Complexity of Innovation and Efficient Hierarchies

Thumbnail Image
Ding_umd_0117E_23327.pdf(1.94 MB)
No. of downloads:
Publication or External Link
Ding, Yuheng
Braguinsky, Serguey
Agarwal, Rajshree
Entrepreneurial activities have been on the decline across a broad range of sectors in the U.S. during the past few decades. This decline (sometimes also called “declining business dynamism”) is reflected in the decreasing rate of new firm entry, the share of young firms (usually defined as those five years of age or less) in the total number of firms and/or the share of employment at young firms in total employment, and so on (e.g., Decker et al. 2014; Akcigit and Ates, 2021). All of the above have exhibited a secular decline, not just in the U.S. but in other advanced economies as well. The underlying causes of these trends, however, are not yet clear with a broad array of explanations suggested in the literature (Akcigit and Ates, 2019; 2021; Decker et al., 2016; Hopenhayn et al., 2018; Karahan et al., 2019; Andrews et al., 2016). There also appears to be a lot of heterogeneity in how strongly the decline in entrepreneurial activities (business dynamism) is pronounced in various industries and sectors of the economy. In particular, the evidence in Haltiwanger et al. (2014) suggests that high-tech industries could be affected more than other sectors of the economy. High-tech sectors have been the driving force of growth in recent decades, so uncovering the reasons for declining business dynamism in those sectors is a task of first-order importance. In the first chapter, I employ the restricted-use data on the science and engineering workforce in the U.S. to investigate whether the increasing burden of knowledge is a growing concern for science-based entrepreneurship. Results show that since 1997, the rate of startup formation has precipitously declined for firms operated by U.S. Ph.D. recipients in science and engineering. The decline in startup formation is accompanied by an earnings decline, increasing work complexity in R&D, and more administrative work for science-based founders. With limited access to efficient knowledge hierarchies, founders of science-based startups must shoulder the burden of knowledge by doing more tasks by themselves. Workers at established firms, on the other hand, could better mitigate the burden of knowledge by narrowing the span of control and increasing the depth of hierarchy. Moreover, less experienced founders were hit harder than more experienced founders as the increasing burden of knowledge led to increasing returns to labor experience. While in the first chapter I use individual-level work data, in the second chapter I utilize firm-level data from the U.S. Census Bureau to develop the analysis further. I adopt the abductive approach and leverage matched employee-employer Census data between 2000-2014 to investigate how a growing burden of knowledge (measured as knowledge interdependence) in the most innovative firms affects potential entrepreneurs’ decisions to start their own business ventures. I show that higher knowledge interdependence in incumbent firms is negatively associated with employee entrepreneurship, and the negative effect is pronounced even stronger among the highest-performing employees. Moreover, higher knowledge interdependence has a positive selection effect on the quality of “spinouts”, and this effect is significantly stronger if the startup is formed by individuals ranked highest in the human capital distribution. These results suggest that knowledge interdependence does not merely raise the barrier for entry into entrepreneurship by imposing higher costs of knowledge transfer. It also changes the functioning of the internal labor market inside the firms. In the third chapter, I further investigate the mechanism underlying the relationship between knowledge interdependence and employee entrepreneurship. I propose a formal theoretical framework that reconciles all empirical findings. The theory suggests firms that rely on higher knowledge interdependence should share “rent” with their employees by paying wage premia if the profit from higher knowledge interdependence is high enough. As a result, within-firm earning dispersion would always be larger in firms relying on higher knowledge interdependence. I find supporting evidence in the data for this alternative explanation. Overall, these findings have important implications for declining entrepreneurial activity, rising income inequality, and technological change in the U.S. economy. While the conventional wisdom might view the declining entrepreneurial activity in the U.S. as the demise of economic growth, it is possible that as innovation becomes more complex, large established firms start to substitute the role of start-ups in pushing forward the technological frontier and driving economic growth as the efficient knowledge hierarchy could better deal with complex knowledge needed in the production process (Garicano, 2000; Garicano and Rossi-Hansberg, 2004). If this is the case, the declining business dynamism might just be a reflection of technological change and efficient (re)allocation of resources but not necessarily detrimental to technological advancement and economic growth. Whether this is true remains an avenue for future research.