[Working papers]
“From Mass to Motion: Conceptualizing and Measuring the Temporal Dynamics of Industry Clusters” (R&R at Strategic Management Journal)
Min Jung Kim, Myles Shaver, and Russell Funk. 1st Essay of Dissertation
This article outlines a temporal dynamics approach to the study of industry clusters. Despite extensive work on cluster size (or “mass”), little attention has been paid to their temporal dynamics (or “motion”). We propose that understanding cluster dynamics is important, because clusters are seldom stable, and cluster dynamics may have strategic implications not accounted for by existing approaches. We introduce a novel measure of cluster motion. Applying this measure to data on establishments in the U.S. computer and semiconductor industries, we document wide variation in cluster mass over time, both within and across regions. Furthermore, utilizing patent data, we find that cluster motion correlates with localized knowledge spillovers in ways different from cluster mass, suggesting that our approach may offer novel insights to strategy researchers.
[Emerging Hot Spot Analysis of the US Computer Industry from 1974 to 2014]
“Cluster Temporal Dynamics, Boundary-Crossing Resource Mobility, and the Nature of Technological Innovation”
Min Jung Kim. 2nd Essay of Dissertation (Job Market Paper)
This study examines the relationship between the temporal dynamics of clusters—namely, the patterns of growth or decline of an industry cluster over time—and firms’ knowledge creation. I examine how one particular form of cluster dynamics, sustained growth, influences firm innovation. I suggest that firms in clusters experiencing sustained growth will be more likely to generate innovation beyond their technology, industry, and geography relative to firms in clusters of comparable size but experiencing stable or declining periods. This is because sustained growth implies that employees are increasingly coming in from elsewhere, which constitutes inflows of knowledge from different geographies or industries. A period of sustained growth also reflects that local institutions are increasingly supportive of local firms to understand and exploit such boundary-crossing knowledge. This is because local institutions—in particular, relevant organizations, such as suppliers and research institutes—are likely to pay more attention to growing clusters and then these organizations also likely come in following these employee inflows, and this repeated influx creates a collective knowledge base that facilitates the spillovers of boundary-crossing knowledge. I test my arguments in the context of the U.S. medical device industry from 1974 to 2016. Controlling for cluster size, I find that firms in clusters experiencing sustained growth are likely to generate more exploratory innovation relative to firms in clusters experiencing a stable or declining period. Interestingly, aligned with the findings in the first chapter, cluster dynamics correlate with disrupting innovation, but in different ways than cluster size alone does. I further examine a key underlying mechanism (i.e., resource mobility across cluster boundaries) by tracing the previous industries and geographies of inventors and by comparing the heterogeneous effects of cluster dynamics between large established firms and entrepreneurial firms.
“Substitutes or Complements? Performance Implication of Joint Decisions on FDI Location and Entry Mode”
Min Jung Kim, Jon Moon, Chris Chung, and Jingoo Kang.
This study investigates how firms approach joint decisions on location (whether to locate in clusters) and operating mode (whether to operate with local firms), and evaluates the performance implications of these decisions. This study is based on the premise that what firms are currently doing may not be what they should be doing. To test that these two decisions have simultaneous and interdependent relationships, we run two-stage probit least squares models using firms in China over the period 1998–2009. In addition, to test the implications of firms’ joint decisions for firm performance, we employ propensity score matching and a difference-in-differences approach. Our findings suggest that firms perceive locating in an industry cluster and operating with a local partner firm as substitutes in gaining local knowledge. However, when we investigate post-entry financial performance, our findings suggest that firms should regard locating in a cluster and having a local partner as complements, not substitutes.
[Work in progress]
“Are Shrinking Clusters Bad for Everyone? The Positive Externalities of Shrinking Clusters for Entrepreneurial Companies’ Resource Acquisition”
Min Jung Kim. Data collection stage
I examine how novice or minority entrepreneurs can benefit from the positive externalities owing to shrinking industry clusters, particularly in the aspect of resource acquisition. A period of decline indicates that firms are increasingly leaving the focal cluster as they relocate to other geographies, change industries, or shut down businesses. When they leave, firms likely free up some resources that are specific to the focal cluster’s geography or industry, making the resources more accessible to others. These freed-up resources include human capital and relationships with institutions (e.g., university labs) that are not easily transferrable to other geographies or industries. Considering these dynamics, I expect that shrinking clusters can be beneficial for ventures founded by novice or minority entrepreneurs (e.g., female or ethnic minorities), which otherwise may not be able to easily access resources, such as talented inventors or opportunities to collaborate with university labs. I explore such conjecture by adopting both qualitative and quantitative approaches in the context of the U.S. computer industry. I conduct case studies of three major computer clusters—Minneapolis-St. Paul, Boston, and San Francisco. For quantitative analyses, I use data on patents and research collaborations.
“Technology Creation and Commercialization in Cluster Dynamics”
Min Jung Kim. Data collection stage
I examine how cluster dynamics influence the commercialization of innovation. In high-tech industries with high levels of governmental regulation, such as the medical device industry, firms require substantial inputs and support from local institutions to successfully commercialize their inventions. Given that cluster dynamics imply changes in local institutions, firms’ commercialization process and performance partly depend on how clusters change over time. I test these relationships by creating a unique dataset that links patent-level data and device-level data. In general, identifying which patents have been used for a particular device is challenging as the information is often not readily available—in particular, in the medical device industry. I have been building the dataset by manually collecting data from archival documents, including registrations and product labels, supported and directed by the Earl E. Bakken Medical Devices Center and medical device companies in Minneapolis-St. Paul.