2009 Conference Papers

Sandip Basu, California State/East Bay & Corey Phelps, Washington/Seattle: The Role Of Exploratory Subunits In Organizational Ambidexterity: An Inductive Examination Of Corporate Venture Capital Units

Ambidextrous firms effectively balance exploration and exploitation and can undertake long-term renewal without compromising short-term performance. To achieve ambidexterity, established firms often set up subunits that engage primarily in exploration. However, exploratory units often become isolated from the mainstream organization, unable to leverage the resources of the parent effectively to carry out their exploratory mission and unable to diffuse their exploratory learning to other units. Such isolation reduces organizational ambidexterity. Relatively little is known about how exploratory units overcome these challenges and effectively contribute to organizational ambidexterity. To address this question, we examine a particular type of exploratory subunit often employed by established firms corporate venture capital (CVC) units. We conducted an inductive, qualitative study of 13 active CVC units and 4 disbanded units. We identify and describe numerous mechanisms that our sample CVC units used to improve the efficacy of their exploratory search and their integration with the broader organization. We show that although the CVC unit usually initiated these mechanisms, they involved multiple levels within the organization. Given the divergent nature of these activities, our results suggest that effective exploratory units are ambidextrous. We also identify and describe the subset of these multilevel mechanisms that were particularly effective in enhancing an exploratory units ambidextrous performance. Finally, we develop a model of CVC unit ambidexterity that theoretically integrates these mechanisms. We discuss the implications of our results for research on organizational ambidexterity and corporate venture capital.

Chris Bingham, UNC: Socratic Dialogue: Bidirectional Questioning And The Micro-Foundations Of Interorganizational Learning

Interorganizational learning is of fundamental interest in the organizations and strategy literature. Yet surprisingly, most of the research on interorganizational learning does not show how such learning actually happens. Instead, learning is generally inferred from performance outcomes. Using an inductive research logic and in-depth data from entrepreneurial firms, we address this gap. Our primary contribution is helping to open the black box of interorganizational learning. Specifically, we find that key to the process of interorganizational learning is bidirectional questioning and loss aversion. We find that focal firms learn by intentionally asking different types of questions to different types of other firms. But, we find that focal firms learn more from other firms that highlight the potential for a loss. Further, we find that focal firms learn through the unexpected questions posed by different types of other firms. However, we find that focal firms learn more from other firms that are highly risk averse. Overall, while most literature on interorganizational learning contributes by emphasizing important behavioral or sociological foundations we contribute by bringing to light important psychological micro-foundations.

Alan Boss, Maryland: Entrepreneurial Resilience: How New Venture Founders Recover and Re-Start After Failure

Everyone experiences failure at some point. Entrepreneurs, especially, have a high incidence of failure, with estimates that over 60% fail within six years. Yet, a high percentage of failed entrepreneurs recover and start another business. How do entrepreneurs recover from failure? Based upon fundamental theories of human behavior and recent inquiries that have influenced the entrepreneurship literature, we draw upon four areas of research, (1) entrepreneurial self-efficacy, (2) emotion regulation, (3) practical intelligence, and (4) self-leadership, to propose a path to recovery. We propose that these areas of research may enhance our knowledge of how entrepreneurs recover from failure.

Sea-Jin Chang, NUS & Brian Wu, Michigan: Post-Liberalization Industry Shakeout in China, 1998-2006

The most important event in Chinas progress in the past three decades is the reform and open door policy, as adopted by the Communist government in 1978. As the reform process deepened in recent years, many traditional state-owned enterprises (SOEs) and collectives were dismantled and transformed into a modern ownership structure. At the same time, many foreign multinationals entered China via foreign direct investments. This liberalization process provides an ideal setting for us to examine how the interaction of different firm types reshapes the Chinese industries. In so doing, we develop a theoretical framework that combines the industry evolution model that formalizes the competitive dynamics of firms with different levels of efficiency with the organizational ecology model that emphasizes organizational diversity and institutional environment. The empirical analysis in this study, based on panel data on more than 280,000 firms in China from 1998 to 2006, shows that the timing of entry and the choice of ownership structure significantly determine firm efficiency level and the likelihood of exit.

Aaron Chatterji, Duke & Kira Fabrizio, Duke: How Do Firms Use Users?: Collaborative Innovation In The Medical Device Industry

Prior research on corporate innovation has highlighted the importance of accessing external knowledge from other firms and universities. However, survey evidence indicates that product users are perhaps the most important source of external knowledge. Interestingly, previous studies on corporate knowledge sourcing have not theoretically considered the distinct nature of user-based knowledge and we have scant empirical evidence as to whether accessing user knowledge enhances firm innovation. We focus on the potential innovative benefits of engaging users in the innovation process by studying collaborations between medical device companies and physicians. We propose that firms benefit from these collaborations by accessing unique user knowledge that enables the firm to both generate higher quality inventions and develop more (FDA approved) product innovations. Our analysis of a panel of medical device firms yields evidence that supports our propositions.

Rafael Corredoira, Maryland: Social Capital Accessibility: An Empirical Study of Generative Mechanisms and Redundancy of Inter-Firm Knowledge Access through Mobility, Alliance and Geographic Proximity

Organizations learn from others through a variety of mechanisms (i.e., alliances, employee mobility, vicarious learning, etc.). This paper explores if and how two generative mechanisms provide access to knowledge across firm boundaries that result in a form of innovation: technical solutions granted patents. In doing so, it addresses two aspects: 1) the role of enduring social relationships and attention-focusing routines in supporting knowledge flows through mobility ties, and 2) the redundancy of such mechanisms (between them and in geographically contained areas), while controlling for alliances and geographic proximity. Its purpose is twofold: first, to identify and test underlying generative mechanisms in the innovative process based on network social capital access (Lin, 2001); and second, to move beyond the events defining the ties between firms (mobility, alliances and geographic proximity) and, by building on the generative mechanisms, provide support for the redundancy of access mechanisms for geographically proximate firms. This paper departs from extant literature on inter-organizational knowledge transfer, which has mainly studied the impact of alliances (attributing the transfer to social interaction and organizational learning routines) and employee hiring (attributing the transfer to portable human capital or social ties of the mobile employee), by actually testing the generative mechanisms underlying the knowledge access phenomenon.

Jason Davis, MIT: Network Dynamics Of Exploration And Exploitation: Pruning And Pairing Processes In Collaborative Innovation

While some research suggests that social networks, organizational behavior, and innovative outcomes influence each other in a co-evolutionary process, few studies have explored these processes in detail because of the difficulties of collecting longitudinal network data. Using a multi-case, inductive study of eight technology collaborations between ten firms in the computing and communications industries, this paper examines the organizational processes that enable managers to influence network dynamics and generate collaborative innovations. Comparisons of more and less innovative collaborations show that rather than leveraging existing ties that bridge organizational boundaries, managers of innovative collaborations use pruning to remove brokers and bridging ties that can become information bottlenecks in the emerging collaboration network. Second, rather than relying on spontaneous social processes to rewire these networks, managers of innovative collaborations follow pruning with pairing to form new boundary spanning ties based on knowledge criteria. By contrast, mangers of less successful collaborations either do not prune prior ties, fail to make new ties, or make new ties based on other criteria. Taken together, the sequence of pruning and pairing processes rewire networks to accelerate knowledge transfer and recombination search mechanisms that underlie innovative exploration and exploitation. The primary theoretical contribution is to understand how managers shape network dynamics to achieve important organizational objectives like innovation. The study highlights the differential impact of different sequences of network processes on global network structure, including the ironic finding that in order to increase network density it may be advisable to first prune some ties. Moreover, network processes increase the local structural alternatives available to managers including the creation and destruction of structural holes and cohesive sub-groups as they search for innovations. Overall, this research links network dynamics with theories of global and local search that are relevant to innovation.

Gary Dushnitsky, Wharton: A Cross-Country Study Of Entrepreneurial Optimism And Valuations

Optimism is a well-documented entrepreneurial characteristic. To date, the literature has mostly focused on the (erroneous) actions of the individual optimistic entrepreneur, yet, less attention was given to the effect this characteristic has on the interaction between entrepreneurial ventures and prospective resources providers. The following question motivates our study: How does optimism affect entrepreneur-investor interaction and what are the implications to ventures valuations? We conjecture that contingent-pay contracts (e.g., preferred shares) can deter charlatans but may be less effective in screening optimists, thus resulting in an optimism discount. We further conjecture that IPR regime may minimize the discount: an entrepreneur can attract higher valuation by disclosing her invention, but she would do so only when disclosure is not vulnerable to imitation (e.g., if patents are effective). Analyses of a sample of international venture capital investments between 1990 and 2006, using a novel measure of cross-country optimism levels, support our hypotheses. The results advance our understanding of entrepreneurial resource assembly, and particularly the nature of the interactions between entrepreneurs and resources providers.

Brent Goldfarb, Maryland & David Kirsch, Maryland: When Are There Bubbles In New Industries?

New technologies and markets are important sources of economic growth and yet are fraught with uncertainty that is only reduced through market experiments (Rosenberg, 1992). New industries often witness rapid entry followed by shakeouts. We investigate two questions related to the inter-relation of financing of market experiments associated with new industry formation: 1) why do we find financial market bubbles in some sets of economic experiments and not in others? 2) what are the implications of fluctuations in investor sentiment on the speed of technological diffusion and technological trajectories of new industries?

There is significant variation in the appearance of bubbles with the birth of new industries. For example, the late 1990s saw a tenfold increase in venture capital outlays, the lions share of which were focused on the IT sector; public market valuations of internet companies fell, on average, close to 90% (Goldfarb, Kirsch & Miller, 2007). In contrast, our preliminary research finds little evidence of a financial bubble during the birth of the laser industry, a classic General Purpose Technology (Bresnahan & Trajtenberg, 1995) that one might associate with speculative investment behavior. At times, there appears to be intra-industry variation as well. During the birth of the electric arc lighting industry, there was not a bubble in its birthplace, Cleveland, while there was one in London.

To explain these differences, we hypothesize that the likelihood of bubbles increases under four conditions: a) easily understood benefits accompanied by little understanding of how one might appropriate returns upon provision of products or services that supply these benefits, and / or b) a systematic underestimation of the time necessary to develop these products or services and c) the presence of uninformed investors, d) event(s) which coordinates beliefs around the potential benefits of this new technology.

Ben Hallen, Maryland: Can Entrepreneurs Believe Their Impressions of Venture Capitalists? Does it Matter? Entrepreneur Evaluations and Network Safety Nets

This paper examines the ability of entrepreneurs to accurately evaluate potential organizational exchange partners. Although there has been significant research on how experienced actors use social network ties, status, and information signals to accurately evaluate potential partners, relatively little attention has been given to the abilities of less-experienced actors to accurately evaluate potential partners. I attempt to redress this gap by integrating two distinct logics. First, building on cognitive psychology research into the limitations of novice problem solvers, I argue that entrepreneurs are generally poor evaluators of an alters partnership abilities. Second, and building on social network research into the knowledge and influence of network brokers, I then argue that entrepreneurs may nonetheless be guided towards better alters by their network brokers. I test and find support for these logics in a study of 9,722 entrepreneurs posting anonymous evaluations of 569 venture capitalists to an online forum. Overall, I find that even though entrepreneurs may often be limited in their ability to directly evaluate potential partners, intermediating brokers may act as network safety nets that save entrepreneurs from their own limitations.

Shon Hiatt, Cornell: The Enabling and Constraining Effects of Attention Structures on Entrepreneurial Activity in the U.S. Biodiesel Industry

I examine the role that attention structures can play in influencing entrepreneurial processes in the institutional environment. Empirically, I investigate the impact of trade associations, serving as attention-regulating actors, on biodiesel producer foundings and technology adoption in the United States. By directing potential entrepreneurs attention to new kinds of organizational forms and available resources, and by channeling the attention of resource-endowed actors to the new industry, trade association members increased biodiesel producer foundings. Additionally, the degree of heterogeneity of trade association members in a given region affected entrepreneurial innovation and technological diversity.

Mukti Khaire, Harvard & R. Daniel Wadhwani, Univ of the Pacific: Changing Landscapes: The Construction of Meaning and Value in A New Market Category Modern Indian Art

Stable category-meanings facilitate market exchange by providing bases for comparison and valuation. Yet little is known about how stable meanings are generated for new categories or how they translate into value hierarchies. We address this gap through a multi-method study of a new market category, modern Indian art. Discourse analysis revealed that firms manipulated discursive elements to generate intersubjective meanings, which enhanced commensurability. This established the bases for aesthetic comparisons and valuation. Analysis of auction data covering 1995 to 2007 revealed that valuations within the category became less volatile and more convergent after intersubjective meanings were established. The paper improves our understanding of the relationship between categories, shared meanings, and valuation in market settings.

Philip Kim, Wisconsin-Madison & Mingxiang Li, Wisconsin-Madison: Seeking Riches amid Uncertainty: Institutional Incentives for Entrepreneurial Entry in Emerging Economies

Classical sociological theories predict that well-functioning regulatory institutions promote stable conditions necessary to start businesses. We question whether this assumption holds in emerging economies by explaining how regulatory institutions may not directly affect founders organizing efforts. Instead, we predict that because of expropriation threats, individuals may be less inclined to start businesses in regions with stronger legal environments. We also argue that as an alternative to regulatory institutions, social trust can provide assurances to individuals pursuing entrepreneurship and indicate how their interrelationship shapes founding conditions. Findings from cross-national analyses confirm our predictions that legal quality and social trust have opposite effects on the likelihood for entrepreneurship. Moreover, generalized trust in strangers and particularized trust from voluntary group membership moderate the effects of legal quality in complementary ways. Our explanations contribute to the entrepreneurship, institutions, and social trust literature.

Daniel Malter, Maryland: Learning From Success, Failure, or Observation In Opportunity Identification And Capture? Evidence For Entrepreneurial Alertness

We investigate whether economic actors benefit from success, failure, or observation experiences in an online auction market. Our results indicate that observing the market alerts bidders to opportunities and enables them to win auctions more cheaply. By contrast, bidders do not benefit from prior success or failure experiences. The results suggest that "know-when" can be more important than "know-how" in markets with sequential, clearly identifiable opportunities. Further, for observation to enable opportunity identification, bidders must search and process information about the auctions. Consequently, our results also support that information search and processing are directly linked to entrepreneurial alertness. Finally, since a winners bid increases the final auction price, entrepreneurial alertness helps smoothing out interauction price fluctuations. Thus, alert bidders engage in intertemporal arbitrage that drives the market toward equilibrium. Overall, our study provides evidence for the relevance and function of entrepreneurial alertness in this market.

Lourdes Sosa, London Business School: From Old Competence Destruction to New Competence Access: Evidence from the Comparison of Small- and Large-Molecule Targeted Anti-Cancer Drug Discovery
Traditional creative destruction theories distinguish disruptions as competence-destroying or competence-enhancing to incumbents’ capabilities, with the former case resulting in incumbents’ loss of competitive advantage in in-house R&D performance (even if complementary assets aid in retaining final market share). In this paper, I propose that attention to the extent of competence destruction is necessary but not sufficient for analyses of competitive advantage in R&D through a technological discontinuity. A full analysis requires the comparison of the value added and ease of access of all capabilities, old and new. In other words, an analysis of competition during a discontinuity requires assessment not only of how many of the old capabilities became obsolete but also what it takes to acquire the new ones. I find evidence for this proposition in data from the transition of anti-cancer drug discovery from standard chemotherapy to targeted therapies. Among targeted therapies, I compare two variants, small- vs. largemolecule drugs, which though equally competence-destroying to chemotherapy-based drug discovery, differ in that large-molecule drugs require one new capability that is inaccessible to incumbents: expertise in biopharmaceutical technology. By tracing the origin, evolution, and added value of biopharmaceutical technology, I can show contrasting results in the two competence-destroying changes: incumbents retained leadership in small-molecule targeted anti-cancer drug discovery; but in large-molecule targeted anti-cancer drug discovery, incumbents fell behind those few entrants that had access to the high-value difficult-to-access expertise in biopharmaceutical technology. I discuss implications for theories on creative destruction.

Vivek Tandon, Michigan: Where Do Schumpeterian Combinations Come From? The Role Of Incentives And Structure In Codetermining The Location Of Combinations In Established Firms

At least since Schumpeter, scholars have conceptualized the job of entrepreneurs as discovering new combinations of means of production (Schumpeter, 1942). Consequently, in the context of corporate entrepreneurship, this conceptualization draws attention towards the locus of discovery of these combinations within firms and therefore towards understanding the factors that explain the search behavior of firms (Nelson and Winter, 1982). Most prior research on search treats firms as monoliths and not as organizations with hierarchy wherein actors have differing motivations (Levinthal, 2005). Thus, how internal processes and systems guide search is less understood. The little empirical research that considers internal organization focuses on the organizational structure but ignores the motivation of searchers. Studies of motivation, on the other hand, ignore firm-level heterogeneity in structure. However, structure and incentives jointly influence search patterns within firms (Cohen, 1984; Siggelkow and Rivkin, 2005). In this paper, I propose a theoretical model to understand how providing incentives through tournament (by increasing payoffs to promotion) and a firms product market portfolio jointly impact the entrepreneurship process within firms. I argue that tournaments direct the attention of division managers to other managers activities because of internal competition and increased strategic value of internal information. This increases managers awareness of a firms capabilities. On the other hand, tournaments also reduce lateral cooperation among the managers. However unrelated diversifiers benefit more from the positive impact of increased awareness and suffer less from the negative impact of reduced cooperation than related diversifiers. This is because the relatedness of activities implies greater internal awareness even without tournaments on one hand and demands greater cooperation on the other. Therefore, I predict that unrelated diversifiers are more likely to search internally for innovations than related diversifiers in the presence of tournaments.

Balagopala Vissa, INSEAD: With Whom do Entrepreneurs Connect? A Matching Theory of Tie Formation and Success

This study advances our understanding of interpersonal network dynamics by applying matching theory to examine how entrepreneurs add new ties to their personal network. I propose that task priorities and social similarity are important matching criteria in entrepreneurs interpersonal tie formation efforts, and test whether good matches increase the likelihood of an exchange relationship being formed between the focal entrepreneurs venture and his contacts organization. The novel research design reveals effects of matching on Indian technology entrepreneurs intentions to form new interpersonal ties and offers evidence that matched interpersonal ties improve the odds of forming new inter-firm exchange relationships.

David Waguespack, Maryland: Network Positions, Network Signals, and New Tie Formation: the et al Effect

Why do individuals with better network positions become more attractive collaborators? One common explanation is that individuals with "better" network positions gain access to superior resources and information, making them intrinsically more appealing partners. An alternative explanation, seen as particularly relevant for newcomers to a social network, is that network connections broadcast extrinsic signals to potential partners. Two signals are particularly salient: connections to high status alters, and a clear identity as demonstrated by membership within a cohesive subgroup. Observing whom an individual has interacted with thus influences assessments of that person's quality (have good people chosen to work with her?) and identity (what camp is she in?), and in turn the propensity to initiate a partnership.

One obvious problem with these assertions, as Podonly (2001) pointed out, is that the intrinsic and extrinsic explanations are simultaneously loaded on a single observable. Podolny notes that in the literature networks are portrayed both as pipes which carry information and resources to the focal actor, and as prisms that shade third party perceptions of the focal actor. This conflation complicates drawing conclusions about what precisely drives future tie formation patterns. If a new entrant with a better initial network position subsequently garners more collaborators than less well positioned newcomers, how much of that variance is due to superior resources and information versus superior signals sent to potential future partners? Moreover, the optimal mix may be positions and signals that are quite different. For instance, an individual would presumably intrinsically benefit from tapping diverse parts of the network to create novel combinations of knowledge (Fleming), and extrinsically benefit from a signal that nonetheless portrays a cohesive identity (Campbell).

A second serious problem with position and signal arguments is that, under normal circumstances, neither value is experimentally assigned. Instead, individuals largely choose a network position and the signal that comes with it. Any correlation between position/signal and tie formation is therefore potentially incidental. The simpler hypothesis is that both position/signal and tie formation are caused by variation in individual quality and preferences. If quality and preferences were fully controlled for the relationship between position/signal and tie formation may look quite different or simply disappear.

To address these issues, in this paper we examine the use of "et als" on references to technology related working papers. We propose that the "et al" effect potentially offers leverage on the position vs. signal issues for two reasons. First, in essence the use of "et al" means that there are two networks, the actual network of collaborators and the readily visible network of collaborators. "Et al" works to obscure some relationships, meaning some individuals will have network signals that are distinct from their real network position. Second, the use of "et al" on references is out of the control of the affected parties, and hence becomes a natural experiment on network signals. While network position remains an endogenous choice, variation in the usage of "et als" randomly alters the visible network. We regard the "et al" effect as a stringent test of hypotheses related to network signals. If signals do not matter then random variation in revealed social network position will have no influence on future tie formation. By contrast, if signals do matter we have a test where there is greater confidence that unobserved quality and/or preferences are not driving future tie formation patterns. Furthermore, an implication of identity related arguments is that relationships relatively remote from the focal actor, such as the presence or absence of connections between friends of your friends, influence future tie formation. Taken at face value, the identity argument means that even individuals with fully revealed ego networks (no "et als" on their collaborations) are affected when relationships among third parties are hidden.