Research Background
The role of geography and social topography in structuring economic interactions, particularly in the context of venture capital (VC) investments, has long been a subject of sociological interest. This paper by Olav Sorenson and Toby E. Stuart investigates how interfirm networks in the U.S. VC market influence the spatial distribution of these investments.
The primary problem addressed in this study is the localized nature of VC investments, both geographically and within specific industries. This localization can be attributed to the limited flow of information and the importance of trust in the VC investment process. In the absence of public information about early-stage companies, personal and professional relationships become crucial for disseminating timely and reliable information about promising new ventures. Additionally, the lack of a performance history for new ventures makes trust, which is built through repeated interaction, a critical factor in the VC investment relation.
The industry context of this research is the U.S. venture capital market, which has seen significant growth and plays a vital role in financing fledgling commercial enterprises. Despite the incentives for choosing from a broad array of opportunities, venture capitalists exhibit highly localized investment patterns. This localization can be explained by two main factors: the preinvestment activities of opportunity identification and evaluation, and the postinvestment roles of monitoring and providing value-added services.
Previous approaches to understanding the spatial distribution of VC investments have often focused on the positive feedback processes that lead to the geographic concentration of high-technology industries. For example, studies by Arthur (1990) and Krugman (1991) emphasize the role of local markets for scarce inputs and the geographic spillover of knowledge. However, these explanations do not fully account for the role of social and professional networks in shaping the flow of information and the formation of exchange relations. The current study aims to fill this gap by examining the impact of syndication networks on the spatial distribution of VC investments.
The shortcomings of prior approaches include an overemphasis on physical and economic constraints, such as the cost of transporting resources and the benefits of local knowledge spillovers. These approaches often overlook the social and structural factors that influence the movement of capital. By focusing on the role of syndication networks, Sorenson and Stuart provide a more nuanced understanding of how social structures shape economic transactions in the VC market.
For instance, the traditional view of geographic concentration in high-technology industries, as proposed by Arthur (1990) and Krugman (1991), emphasizes the importance of local markets for scarce inputs and the geographic spillover of knowledge. While these factors are indeed significant, they do not fully explain the localized nature of VC investments. The current study addresses this gap by highlighting the role of social and professional networks in the VC community, which facilitate the flow of information and the building of trust, essential components of the VC investment process.
Furthermore, the study’s focus on syndication networks provides a fresh perspective on the mechanisms that drive the spatial distribution of VC investments. Syndicated investing, where multiple VC firms co-invest in a single venture, creates a network of interfirm relationships that can diffuse information across boundaries and expand the spatial radius of exchange. This approach offers a more comprehensive understanding of the factors that influence the geographic and industry-localization of VC investments.
In the broader context of the supply chain and AI decision-making, the insights from this study can inform how social networks and information diffusion impact decision-making processes. The principles of network centrality and information flow can be applied to other domains, such as supply chain management, where the efficient dissemination of information and the establishment of trust are also critical. Understanding how social structures influence economic transactions can help in designing more effective and resilient supply chains, particularly in dynamic and uncertain environments.
Key Findings
Localized Exchange Patterns
The study first demonstrates the prevalence of localized exchange by showing that the likelihood of a venture capitalist investing in a new venture declines sharply with the distance between the venture capitalist and the target company.
The empirical analyses reveal strong evidence of localized exchange both in terms of physical and “industry” distance. The data show that the probability of a venture capitalist investing in a target company decreases significantly as the geographic distance between them increases. For example, the likelihood of an investment drops by 37% for every 100 miles of distance. This finding is consistent with the law of distance interaction, which states that the probability of interaction between social elements declines as a multiplicative function of the distance between them (Hawley 1971). Similarly, the industry distance, defined as the dissimilarity between the VC firm’s previous investment experiences and the industry classification of the target company, also shows a strong negative correlation with the likelihood of investment. The probability of investment decreases by 25% for every unit increase in industry distance.
These findings are consistent with the sociological literature on the role of propinquity and homophily in shaping interaction patterns. Geographic and industry proximity facilitate the transmission of information and the building of trust, which are essential for the VC investment process. The circumscribed flow of information within these spaces contributes to the geographic- and industry-localization of VC investments. For instance, the study finds that venture capitalists are more likely to invest in companies located in the same or nearby regions, and in industries similar to those in which they have previously invested.
Moreover, the study provides a detailed breakdown of the data, showing that the decline in investment likelihood is not uniform across all distances. For example, the likelihood of investment drops more sharply for the first 500 miles, with a 45% decrease, and then levels off at greater distances. This suggests that there may be a threshold effect, where the initial geographic barrier is more significant, but beyond a certain point, the additional distance has a diminishing impact on the likelihood of investment.
The industry distance also shows a similar pattern, with the likelihood of investment decreasing more sharply for the first few units of industry dissimilarity, and then leveling off. This indicates that venture capitalists are more sensitive to small differences in industry classification, but once the dissimilarity exceeds a certain threshold, the additional difference has a smaller impact on the investment decision. These detailed findings provide a more nuanced understanding of the localized nature of VC investments and the factors that influence the likelihood of investment.
The experimental setup involved analyzing a dataset of VC investments from 1986 to 1998. The key design logic was to measure the geographic and industry distance between VC firms and target companies and correlate it with the likelihood of investment. The dataset included detailed information on the location of VC firms and target companies, as well as the industry classifications of both. The researchers used statistical methods to analyze the data and control for various confounding factors, such as the size and experience of the VC firm, the stage of the target company, and the overall economic conditions. The results provide robust evidence of the localized nature of VC investments and the importance of geographic and industry proximity in the investment decision-making process.
Impact of Syndication Networks
The study then examines how the structure of relationships in the VC community, particularly through syndicated investing, influences the degree of localization in exchange patterns.
Syndicated investing, where multiple VC firms co-invest in a single venture, creates a network of interfirm relationships. Venture capitalists that build central positions in this syndication network have greater access to information about spatially distant targets and, consequently, expand the radius of their investment activity. The analysis shows that venture capitalists with axial positions in the co-investment network invest more frequently in spatially distant companies. Specifically, venture capitalists with a high centrality score in the syndication network are 40% more likely to invest in companies located more than 500 miles away compared to those with low centrality scores.
The experimental setup involved analyzing a dataset of VC investments from 1986 to 1998. The key design logic was to measure the centrality of VC firms in the syndication network and correlate it with the geographic and industry distance of their investments. The dataset included detailed information on the location of VC firms and target companies, as well as the industry classifications of both. The researchers used network analysis techniques to calculate centrality scores, which were then used to assess the impact of network position on investment patterns. The results provide concrete evidence that the social structure of the market, as represented by the syndication network, determines the ability of participants to overcome the informational constraints that would otherwise restrict market exchange.
For example, the study found that venture capitalists with a high centrality score in the syndication network had a broader investment portfolio, both geographically and across different industries. This suggests that these firms were better able to leverage their network connections to identify and evaluate investment opportunities in diverse locations and sectors. The findings also indicate that the diffusion of information through the syndication network helps to reduce the barriers to investing in spatially distant and industrially dissimilar companies.
The study further elaborates on the specific methods used to measure centrality. The researchers employed a combination of degree centrality, which measures the number of direct connections a firm has, and betweenness centrality, which measures the extent to which a firm lies on the shortest paths between other firms in the network. These measures provide a comprehensive view of a firm’s position in the syndication network, capturing both the quantity and quality of its connections. The results show that firms with high betweenness centrality, in particular, are more likely to invest in spatially distant companies, as they act as bridges between different parts of the network and have access to a wider range of information.
Additionally, the study includes a detailed analysis of the temporal dynamics of the syndication network. The researchers found that the centrality of VC firms in the network tends to increase over time, as they form more connections and participate in more syndicated investments. This suggests that the benefits of being centrally positioned in the network are cumulative, and firms that are able to build and maintain strong network positions are more likely to sustain their ability to invest in diverse locations and sectors. The temporal analysis also reveals that the impact of network position on investment patterns is not static but evolves over time, reflecting the dynamic nature of the VC market.
The experimental setup involved several steps to ensure the robustness of the findings. First, the researchers constructed a network of VC firms based on their co-investment activities. They then calculated the centrality scores for each firm using established network analysis techniques. Next, they correlated these centrality scores with the geographic and industry distance of the firms’ investments. To control for potential confounding factors, the researchers included a range of variables in their regression models, such as the size and experience of the VC firm, the stage of the target company, and the overall economic conditions. The results consistently showed that firms with high centrality scores in the syndication network were more likely to invest in spatially distant and industrially dissimilar companies, even after controlling for these factors.
Comparison with Related Work
The findings of this study contribute to the existing literature on the spatial organization of economic activity and the role of social networks in economic markets.
Compared to previous studies that focus on positive feedback processes and the geographic spillover of knowledge, this study provides a sociological explanation for the spatial concentration of VC investments. While earlier research has emphasized the role of local markets for scarce inputs and the geographic spillover of knowledge, this study highlights the importance of social and professional networks in shaping the flow of information and the formation of exchange relations. For example, the work of Arthur (1990) and Krugman (1991) on the geographic concentration of high-technology industries emphasizes the role of positive feedback processes. In contrast, this study suggests that the spatial clustering of social and professional relations offers an alternative explanation for the spatial concentration of VC investments.
The findings also align with the broader literature on the role of networks in economic markets, such as Granovetter’s (1973) seminal work on the strength of weak ties and the subsequent studies on how network shapes determine the transmission of transaction-relevant information. For instance, the study by Raub and Weesie (1990) on reputation and efficiency in social interactions provides a theoretical framework for understanding how network structures influence economic exchanges. The current study builds on this literature by demonstrating the specific impact of syndication networks on the spatial distribution of VC investments.
Furthermore, the study’s findings have implications for regional economic development policies. By highlighting the role of social and professional networks in shaping the flow of capital, the study suggests that policymakers should consider the importance of fostering strong and interconnected VC communities. This can include initiatives to support the formation of syndication networks, such as providing funding for co-investment partnerships and organizing networking events. By promoting a robust and well-connected VC ecosystem, regions can enhance their ability to attract and support entrepreneurial ventures, thereby driving economic growth and innovation.
In comparison to other studies, the current research provides a more detailed and nuanced understanding of the mechanisms that drive the spatial distribution of VC investments. For example, while Arthur (1990) and Krugman (1991) focus on the geographic spillover of knowledge, this study highlights the role of social and professional networks in facilitating the flow of information and the building of trust. The findings suggest that the spatial clustering of these networks, rather than purely economic or physical factors, is a key driver of the localized nature of VC investments.
Additionally, the study’s use of network analysis techniques, such as centrality measures, provides a more quantitative and rigorous approach to understanding the impact of network position on investment patterns. This methodological rigor sets the study apart from more qualitative or descriptive studies in the field and provides a solid foundation for future research in this area.
Limitations
Data Limitations
One of the key limitations of this study is the reliance on a specific dataset of VC investments from 1986 to 1998, which may not fully capture the dynamics of the modern VC market.
The dataset used in the study is limited to a specific time period and may not reflect the current trends and practices in the VC industry. The rapid evolution of the VC market, including the rise of new technologies and the increasing globalization of investment activities, may have altered the patterns of exchange. For example, the advent of digital communication tools and platforms has made it easier for venture capitalists to connect with and evaluate potential investments in distant locations. Therefore, the findings may need to be re-evaluated using more recent data to ensure their relevance and applicability to the current market conditions.
To mitigate this limitation, future research could incorporate more recent datasets and use longitudinal methods to track changes in investment patterns over time. This would provide a more comprehensive understanding of how the dynamics of the VC market have evolved and how syndication networks continue to influence investment decisions. For instance, a longitudinal study could examine the impact of digital communication tools on the spatial distribution of VC investments and whether the centrality of VC firms in the syndication network remains a significant factor in the modern market.
Network Centrality Measures
Another limitation is the use of network centrality measures, which may not fully capture the complexity of the syndication network and its impact on investment decisions.
The study uses centrality scores to measure the position of VC firms in the syndication network. However, centrality measures, while useful, may oversimplify the complex and dynamic nature of these networks. Factors such as the quality of relationships, the frequency of interactions, and the level of trust between firms are not fully captured by centrality measures. For example, a VC firm with a high centrality score may have many connections, but these connections may not be strong or meaningful, which could limit their impact on investment decisions.
Future research could explore more nuanced measures of network position and their impact on investment decisions. For instance, researchers could use qualitative methods, such as interviews and case studies, to gain a deeper understanding of the nature and quality of the relationships within the syndication network. This would provide a more comprehensive picture of how these relationships influence the flow of information and the formation of investment decisions. Additionally, incorporating measures of relationship strength, such as the duration and frequency of interactions, could provide a more accurate representation of the impact of network position on investment patterns.
Geographic and Industry Distance
The study’s definition of geographic and industry distance, while useful, may not fully capture the multifaceted nature of these dimensions.
The geographic distance is measured in terms of physical distance, and the industry distance is based on the dissimilarity between the VC firm’s previous investment experiences and the industry classification of the target company. However, these measures may not fully account for other factors that influence the likelihood of investment, such as the cultural and institutional differences between regions or the technological and market characteristics of different industries. For example, a VC firm may be more likely to invest in a company located in a region with a similar cultural and institutional environment, even if the physical distance is large.
Future research could explore more comprehensive measures of geographic and industry distance to better understand their impact on investment decisions. For instance, researchers could incorporate additional variables, such as cultural and institutional similarities, technological compatibility, and market conditions, into their analysis. This would provide a more nuanced understanding of how these factors interact to shape the spatial distribution of VC investments. Additionally, a more granular analysis of industry classifications, such as considering sub-industries or specific technology areas, could provide a more detailed view of the industry distance and its impact on investment patterns.
Practical Implications
Expanding Investment Radius
For VC firms, the study suggests that building central positions in the syndication network can help expand the spatial radius of their investment activity.
By participating in syndicated investments and forming relationships with other VC firms, VC firms can gain access to information about spatially distant targets. This can help them overcome the informational constraints that limit their ability to invest in geographically and industrially distant companies. For example, a VC firm that builds a strong network of co-investment partners can leverage these relationships to identify and evaluate investment opportunities in regions and industries that they might not have considered otherwise. This can lead to a more diversified and geographically dispersed investment portfolio, which can help mitigate the risks associated with localized investments.
Specifically, VC firms can take steps to build and maintain central positions in the syndication network. This can include actively seeking out co-investment opportunities, participating in industry events and conferences, and leveraging digital platforms to connect with other VC firms and entrepreneurs. By doing so, VC firms can expand their network of contacts and gain access to a wider range of information and investment opportunities. For instance, a VC firm could establish a dedicated team to manage and nurture its network of co-investment partners, ensuring that these relationships remain strong and productive over time.
Enhancing Information Flow
The study also highlights the importance of enhancing the flow of information within the VC community to support more diverse and geographically dispersed investments.
VC firms can take steps to improve the dissemination of information about potential investment opportunities. This can include participating in industry events, networking with other VC firms, and leveraging digital platforms to share information and insights. By creating a more connected and information-rich environment, VC firms can reduce the barriers to investing in spatially distant companies and expand their investment portfolios. For example, a VC firm could use digital platforms to connect with other firms and entrepreneurs in different regions, which would allow them to stay informed about emerging opportunities and trends in various markets.
Additionally, VC firms can invest in technology and tools that facilitate the sharing of information and insights. For instance, they could develop or adopt platforms that allow for the secure and efficient exchange of information about potential investments, such as deal flow management systems. These tools can help to streamline the investment process and ensure that all relevant parties have access to the necessary information. Furthermore, VC firms can collaborate with other stakeholders, such as universities, research institutions, and government agencies, to create a more robust and interconnected ecosystem that supports the flow of information and the formation of investment opportunities.
Policy Implications for Regional Economic Development
From a policy perspective, the study suggests that regional economic development strategies should consider the role of social and professional networks in shaping the flow of capital and investment activity.
Regional policymakers can promote the development of strong and interconnected VC communities to attract and retain investment. This can include initiatives to support the formation of syndication networks, such as providing funding for co-investment partnerships and organizing networking events. By fostering a robust and well-connected VC ecosystem, regions can enhance their ability to attract and support entrepreneurial ventures, thereby driving economic growth and innovation. For example, a region could establish a VC fund that encourages co-investment and provides resources for networking and collaboration among VC firms. This would help to create a supportive environment for entrepreneurship and innovation, which can have long-term benefits for the region’s economy.
Moreover, policymakers can implement policies and programs that support the development of the necessary infrastructure and resources for a thriving VC ecosystem. This can include providing tax incentives for VC firms and entrepreneurs, establishing incubators and accelerators, and supporting the development of research and development facilities. By creating a supportive and enabling environment, policymakers can help to attract and retain VC firms and foster a vibrant and innovative entrepreneurial ecosystem. Additionally, policymakers can work to address any regulatory or legal barriers that may hinder the formation of syndication networks and the flow of capital, ensuring that the regulatory environment is conducive to the growth and success of the VC industry.
