Select Research portfolio

Abstract: When adolescent ventures choose their scaling strategy for longer-run innovation outcomes, is organic- or acquisition-led-growth preferable? The multifaceted reasons managers may choose an acquisitive path, only some of which are observed and measured, makes this question difficult to address. We study U.S. firms undergoing an initial public offering (IPO) between 1975 and 2016 and track the extent to which they conduct acquisitions, pre- and post-IPO. We use firms’ patenting activity as proxies for innovation. We address endogenous selection of acquisition strategies by employing difference-in-differences and instrumental variable methods and estimate a 6 - 10% boost in innovation for acquisition relative to organic scaling. These results contrast with naïve analyses, which suggests a negative or null effect of acquisitions on innovation.

What geographic pull factors shape where individuals engage in startup activity? The related literatures discuss how agglomeration forces shape the geographic locus of innovation, characterize the business environment influencing individual mobility, and suggest that local spinoffs can lead to industry clustering. To improve our understanding of why some individuals move geographies (bypassing their home social capital) to start new ventures, we evaluate three candidates: venture capital access, specialized human capital access, and co-location to access knowledge spillovers. Using startups in the artificial intelligence (AI) industry, we examine the relative empirical importance of these factors in the 2001-2018 timeframe. We find that our measures of knowledge spillover capital and human capital best explain geographic pull factors for entrepreneurship, especially after a major technological advance.

This paper examines how early exploration strategies, sequential or parallel, affect financial and innovation outcomes of deeptech ventures. Entrepreneurial strategy research emphasizes experimentation as a means of resolving uncertainty, with lean methods promoting early commercial exploration for learning. While this work has largely focused on asset-light contexts like software, deeptech firms address multiple dimensions of uncertainty, including the underlying technology. They also rely on physical assets to commercialize, incurring significant opportunity costs for search. Therefore, the choice of early exploration strategy is particularly consequential in deeptech contexts. Using a novel longitudinal dataset of hardware firms, the study finds that parallel exploration improves the likelihood of early commercial outcomes by mitigating information frictions with potential resource providers. In contrast, sequential exploration fosters organizational learning, yielding better long-run financial and technological performance.