1 Introduction

In addressing perhaps the most central question in the scholarly field of entrepreneurship—why an individual chooses to become an entrepreneur while others do not, most theories have focused on characteristic variation across individuals (Acs and Audretsch 2010). Do they possess propensities towards risk taking, proclivities for independence and autonomy, gender, experience, or access to crucial resources such as finance, or social and human capital? The ensuing research agenda would seemingly deliver an answer to the age-old debate on nature versus nurture, that is, whether entrepreneurs are born or made.

More recently, a different approach has suggested that there may be alternative perspectives shedding a different light, not just on why some people choose to become entrepreneurs while others do not, but also how and why entrepreneurship is a critical issue in regards to improving economic performance. According to the knowledge spillover theory of entrepreneurship (KSTE), the context in which decision-making is derived can influence one’s determination to become an entrepreneur. In particular, a context that is rich in knowledge generates entrepreneurial opportunities from those ideas created. By commercializing the ideas that evolved from an incumbent organization but commercialized independent of this organization via the creation of a new firm, the entrepreneurs not only serve as a conduit for the spillover of knowledge, but also for the ensuing innovative activity and enhanced economic performance (Acs et al. 2009).

The theory brings together contemporary theories and thoughts of entrepreneurship with prevailing theories of economic growth. In particular, this approach advances the microeconomic foundation of the endogenous growth theory by providing a new framework explaining the unobserved heterogeneity of growth rates between regions and nations. While keeping constant the primary research question of intrinsic motivation among entrepreneurs, the knowledge spillover theory is concerned with the contextual variables that shape entrepreneurship. Considering that entrepreneurship and new venture creations are not a recent phenomenon, and by keeping the intrinsic motivations of entrepreneurs constant, the observed increase in the rate of startups and entrepreneurial activities should reflect a change in the costs and benefits of creating a new venture based on changes in the costs and benefits of the operating context. Knowledge created by incumbent firms and research organizations, which is underexploited and not fully commercialized for purposes of economic gain, then spills over to other economic agents—entrepreneurs, and is identified as the primary factor in resource allocation.

To this end, we propose the Knowledge Incubator—a private firm, non-profit organization, government, university, or research institution which has, through its own labor and resources, developed new knowledge with potential in the commercial markets but has, for various reasons such as uncertainty, opted not to commercialize and exploit said knowledge. Economic agents, which are able to absorb knowledge spillovers and convert them into economic knowledge, are not required to bear the full costs of the knowledge development. This specific type of entrepreneur, who utilizes knowledge spillover, but does not bear the full costs of the newly developed knowledge, is referred to as a—high-impact Entrepreneur (Acs 2010). This individual, or group of individuals, is unique from other entrepreneurs in the sense that they are utilizing the spillover from the knowledge incubator, commercializing this knowledge by founding a new firm, entering the marketplace, and converting the new knowledge into economic knowledge. Consequently, the expected benefits increase via exploitation of the knowledge spillovers, which are ideally converted into economic knowledge and ultimately foster economic growth.

The theory contributes to the existing body of knowledge by explaining how and why knowledge spills over, and in what manner entrepreneurship acts as the mechanism by which knowledge evolves into economic knowledge within a given framework. While new ventures in the tech-sectors have led to the well-known erosion of previously existing industries (Schumpeter 1934), such as Facebook Inc., technological change and progress have been considered drivers of economic growth since the first industrial revolution. Since then, new venture creation and technological change was considered to be almost entirely exogenous. The most influential factors in the last few decades can be characterized by a shift in the costs and benefits of the contextual factors, which have led to innovations and new venture creation far beyond exogenous factors. Not only innovations but also new ventures and entrepreneurial firms often fall like manna from heaven.

The theory thus shifts the unit of analysis away from firms and endowment assumed to be exogenous and focuses attention on individual agents, high impact entrepreneurs, who possess new knowledge endowments characterized as knowledge that has been captured from spillovers. This suggests a strong relationship between knowledge spillovers on the one end of the spectrum, and entrepreneurial activities on the other, while both have influence on the growth rates of regions and countries. Unobserved heterogeneity between regions and countries is based on differences in the knowledge spillover endowment and the ability to foster entrepreneurship (entrepreneurship capital); or in other words, unobserved heterogeneity in growth rates is due to differences in the costs and benefits of the knowledge endowments.

The following section of this paper explains in detail concepts of the knowledge spillover of entrepreneurship, and why knowledge spillover of entrepreneurship matters for economic performance. The paper then identifies what has been gained from the literature about the knowledge spillover theory of entrepreneurship; the next sections discuss the role of competition and examine absorptive capacity. The final section of the paper provides a summary and conclusions.

2 Defining the knowledge spillover theory of entrepreneurship

Like any theory of entrepreneurship, the theory explains why some people take action, while others opt for inaction, and when an entrepreneurial opportunity presents itself. The theory is based on the proposition that entrepreneurial behavior is a response to profitable opportunities from knowledge spillovers. The reason people start entrepreneurial firms is because they have access to knowledge spillovers. In particular, the potential for taking advantage of a knowledge spillover creates the entrepreneurial opportunity, which then drives knowledge spillover entrepreneurship. Thus, theory focuses on entrepreneurial behavior within the context of knowledge spillovers, and by doing so, links traditional entrepreneurship theory to the theory and literature on knowledge spillovers.

The traditional theoretical approach that has become well established throughout the scholarly literature on entrepreneurship has been focused on the role of opportunities, both recognizing or creating opportunities, as well as acting upon or exploiting those opportunities.Footnote 1 As Sarasvathy et al. (2003, p. 142) point out, “An entrepreneurial opportunity consists of a set of ideas, beliefs and actions that enable the creation of future goods and services in the absence of current markets for them.” They decompose entrepreneurial opportunities into three types—opportunity recognition, opportunity discovery, and opportunity creation.

Much of the entrepreneurship literature has focused on differences across individuals, or individual specific characteristics as explaining the ability of some individuals to either discover or create entrepreneurial opportunities or else to act upon or exploit those entrepreneurial opportunities. According to Krueger (2003, p. 105), “The heart of entrepreneurship is an orientation toward seeing opportunities,” which frames the research questions: What is the nature of entrepreneurial thinking and what cognitive phenomena are associated with seeing and acting on opportunities?

Thus, the traditional theories of entrepreneurship consider the context external to the individual as given or constant and then focus on the cognitive decision made by the individual in deciding whether or not to enter into, or remain in, entrepreneurship. Theories of entrepreneurship most typically focus on characteristics specific to the individual. According to Shane and Eckhardt (2003, p. 187), “We discussed the process of opportunity discovery and explained why some actors are more likely to discover a given opportunity than others.”

As Parker (2009) identifies in his meticulous review of the literature, the specific characteristics of individuals influencing the entrepreneurial decision range from the willingness to incur risk and confront failure, preferences for self-directed and autonomous decision making, and access to scarce and valuable resources, including financial capital, human capital, social capital, and experiential capital.

By contrast, the knowledge spillover theory of entrepreneurship has an alternative starting point. Rather than focus on the heterogeneity of individuals, the starting point is instead the heterogeneity of contexts with which individuals find themselves. The salient characteristic of the context involves the creation of new knowledge and ideas. As Arrow (1962) emphasized, decisions involving economic knowledge have three characteristics that distinguish them from decisions with normal economic goods. The first is the high degree of uncertainty. As Alvarez et al. (2010) argue, neither the outcomes nor any associated likelihood of distributions can be associated with decisions involving uncertainty. While decision-making under risky conditions have known outcomes, which are likely to occur with known probabilities, the outcomes based on knowledge involve inherent uncertainty where there are no certain outcomes.

The second condition distinguishing the uncertainty inherent in knowledge from risky decision-making involving information is the asymmetric nature of knowledge and ideas. The valuation of any uncertain idea or knowledge will vary across individuals due to moral differences between those individuals. The third condition is that the cost of transacting those differences leading to heterogeneous valuation of knowledge and ideas is non-trivial.

Taken together, these three conditions contribute to disparities in the valuation of new ideas across individuals in general, and in particular, within decision-making hierarchies. Such disparities in the value of a new idea create commensurate disparities about whether to act upon such an idea, that is, to develop that idea and commercialize it in the form of an innovative activity. The asymmetries in knowledge across individuals also create asymmetries in opportunities across individuals. In the case where an individual places a higher valuation on a new idea, as opposed to a decision-making hierarchy within an incumbent firm or organization, this individual is confronted by an entrepreneurial opportunity because they perceive an opportunity while the incumbent organization does not. The entrepreneurial opportunity is derived from the creation of knowledge that has not been fully appropriated within the incumbent organization from which that knowledge originated.

Thus, what distinguishes this theory from other theories of entrepreneurship is that the source of the entrepreneurial opportunity involves knowledge spillovers. The knowledge spillover theory of entrepreneurship explains the entrepreneurial act—why certain people become entrepreneurs while others abstain from entrepreneurship—as a response to knowledge spillovers.

Simply because knowledge spillovers provide the ability to generate or facilitate an entrepreneurial opportunity does not necessarily indicate that the induced entrepreneurship is particularly important or notable. In fact, there are both compelling theoretical reasons and empirical evidence suggesting that knowledge spillover entrepreneurship does indeed play a key role in the economy and in particular, in generating innovative activity, economic growth, employment, and competitiveness in global markets. Since the early beginnings of the Industrial Revolution, technological progress not only leads to economic growth but also welfare related externalities for society. Technological progress simply leads to standardized work and labor division. Additionally, the negative externalities of technological progress and change lead to mass production, poverty of workers, and child labor among others, and the benefits for society are often neglected, such as increased income, medical progress, etc. Technological progress was mainly born by one invention—the steam engine. This invention led to some endogenous technological innovations by replacing human labor and workforce with the steam engine, as well as spinning machines, mills, ships, railways (horses), among others. It was the second industrial revolution, which changed the world in this regard, that the innovations made are, more or less, falling from heaven like money, with innovations in the energy market, chemistry, medicine, biology, food, communication, and the invention of the internal combustion engine.

These inventors, who recognized an opportunity and made risky decisions to invest in a new venture, and to commercialize their ideas, made the majority of these inventions. Most of them continue to exist, named after their initial founder and entrepreneur, like Daimler, Siemens, Johnson & Johnson, or Eli Lilly, among others. At the end of the 19th century, the number of entrepreneurial and new ventures that entered the stock market via an Initial Public Offering (IPO) reached a level that was only exceeded in the mid 20th century. In Germany, the number of IPOs during that time exceeded the number of IPOs of the dotcom era. Two world wars, the great recession, political revolutions, and the end of colonialism led to a drastic decrease in entrepreneurial activities. Instead of technological change and knowledge production, economic growth was driven by the reconstruction efforts and, in particular, the demand for consumer goods in the United States.

Even in the Solow (1956) model of economic growth, knowledge, or what was termed as constituting “technical change,” was a driving force of economic performance. However, knowledge is viewed as being outside of the model and reflected in the residual, and thus is beyond the reach of policy influence. As Solow famously pointed out, such knowledge “falls like manna from heaven,” which would seemingly put it out of the reach of entrepreneurship or any other particular organizational form or behavior.

By contrast, in the endogenous growth models introduced by Romer (1990) and Lucas (1993), knowledge is not only included within the model’s parameters, but has a particularly potent impact on growth, because of its strong propensity, as Arrow (1962) made clear, to spill over from the firm or organization creating knowledge to other third-party firms accessing that knowledge for a cost less than its value. While the Romer (1990) model assumes that knowledge spills over automatically, Acs et al. (2004), (2012), Audretsch et al.(2006), and Braunerhjelm et al. (2010) suggest that instead, the automatic spillover of knowledge from its source is impeded by what they term as the knowledge filter. The knowledge filter prevents or at least impedes knowledge from automatically spilling over for innovation and commercialization. Regulations and legal restrictions may account for some of the knowledge filter. However, the broadest and most prevalent source contributing to the knowledge filter are the conditions inherent in knowledge—uncertainty, asymmetries, and high costs of transaction. Knowledge spillover entrepreneurship is important and significant because it provides a conduit penetrating the knowledge filter and serves as a catalyst for the commercialization of knowledge and ideas created in one organizational context but generating innovative activity in a the context of a new firm, which ultimately contributes to economic growth, employment creation, and global competitiveness.

3 The emergence of the knowledge spillover theory of entrepreneurship

Although several theoretical streams and thoughts from a variety of disciplines in the social sciences shape the theory, its primary roots can be traced back to early thoughts on how to create value and wealth, and how to distribute it within society. Long before Adam Smith, philosophers, policy makers and practitioners were concerned about seeking answers to these questions, and to the present day the body of knowledge is continuing to expand. The KSTE is perhaps the most recent theoretical structure in providing answers to the aforementioned questions. The answer, according to KSTE, is the individual entrepreneurs.

As far back as history will allow us to look, individuals have acted as opportunity seekers in order to improve their own wealth status and general well-being. Changes and developments in the environment have led to new challenges and thus new opportunities, which have consequently led to a new type of player entering society in recent decades—the entrepreneur as an economic agent. While the individual incentives to start up a firm may be kept rather constant over time, technological, political, and social factors endogenously changed over time. KSTE is concerned with these changes and integrates them into a new framework, which goes beyond explaining why some people choose to become entrepreneurs while others do not, but how and why entrepreneurship matters crucially in improving economic performance. The theory integrates the context in which decision-making is undertaken in order to explain how value and wealth is generated in societies and how it should be distributed across the society.

Thus, roots of KSTE could be traced back to early thoughts on economic history up until recent developments on macro- and micro economic levels. Furthermore, the theory shifts the unit and lens of analysis from the state or society as a whole to the single entrepreneur as the critical economic agent in generating wealth and value. Understanding the roots and history of the development of the theory also helps explain recent political systems in promoting economic growth, which range from pure absolutistic systems, over governmental and centralistic interventions, to decentralized, market-based policy approaches in both generating wealth and value and its allocation and distribution among members of a society.

3.1 From mercantilism to the knowledge production function

One of the earliest concept of explaining economic growth dates back to the 16th century with the expansion and growth of trade companies—mercantilism, named later on after the Latin name mercator and the French adverb mercantile. The 16th century was characterized by absolutistic policy regimes, such as in France, the UK, Spain and Portugal, their expansions across the new world, and the growth of trade companies. Economic value was generated by export activities under a strong protection (subsidies, monetary policy, and import restrictions) of the absolutistic monarchies. Economic value was thus distributed among the actors, i.e. the monarchy and the trade companies. Regarding the societal and political environment, mercantilism could be treated as an efficient way to produce wealth and distribute said wealth among members of a society. Changes in the technological and political environment increased the costs of governmental protectionisms and increased the advantages of division of labor and specialization, leading to a complete rethinking and reformulating of economic strategy.

In the middle of the 20th century, economists began to formulate models directed towards explaining economic growth and distribution. Growth and wealth in a society could be simply expressed by something such as gross domestic production, and the main questions are concerned with how to increase gross domestic production (GDP) and which factors are the most relevant. Harrod (1939) and Domar (1946) showed that GDP growth depends primarily on the stock of capital, the investment made, and the savings ratio. Economic growth thus increases c. p. with the investments made and investments strongly increase with the saving rate. However, production growth and income are also shaped by the costs of production, i.e. the amount of capital needed to produce a single unit of GDP. Thus, wealth is generated by the production of goods and thus by the actors who make the investments and thus the people who save money (instead of consuming) and thus are rewarded by interest rates. Regarding the industrial landscape at this time, it is characterized by large firms producing mainly standardized products in tailored-style mass production (automobile, steel, railways, weapons, food, energy), whereby the main drivers of GDP on the supply side are investments in plants and transportation and the only scarcity is access to financial resources and therefore the money saved and spent on banks or equity invested in the stock market. Observed variations on the demand side, e.g. a decrease in consumption, should be removed by the forces of the free market (which is the neoclassical point of view) or policy inventions (Keynesianism). The production process from inputs to outputs was approximately best described by production functions, based on capital and labor as the main inputs, and only varying in the elasticity in the final production function (Cobb-Douglas-Type). Economic wealth is distributed among society by the interest rates (compensation for not consuming and thus saving money) and the wage for providing labor force.

After WW2, new inventions led to product differentiation, new products and production processes. In the following, also the input factors vary from standardized inputs like unskilled to skilled labor workforce, different types of services and other inputs. Product differentiation, endogenously driven by investments not only in plants or transportation but also in new production processes and products and exogenously driven by inventions, like the first computer, leads to a rethinking of the existing models explaining growth and the production process. One of the first taking this development into account was Robert Solow, reformulating the existing model of growth by explicitly implementing technological progress. In steady-state equilibrium, investments in the capital equal its depreciation rate (or growth of population). Additional growth rates could only be received by technological progress. Solow himself finds empirical evidence for his revised growth model by testing a revised version of the Cobb-Douglas production function. In the following, considerable research has proceeded based on Solow’s theoretical article, linking physical capital and (unskilled) labor to growth rates (see Nelson 1981). Also Solow acknowledged that technological change contributes to economic growth, it was considered as exogenous, an unexplained residual, which falls like manna from hell.

However, knowledge is not only exogenously given but also the result of investments in education and R&D (Mansfield et al. 1977a, b). Knowledge spills over, like a reinforcing processes, and thus the unexpected residual in explaining growth rates increased over time. Also, the political landscape evolved, leading to openness between countries with increases in trade and foreign investments. The Solow-Model, based on exogenously given technology within a closed economy and mass production, leaves a great amount of variation of growth rates unexplained and leads to the question of where these unexplained growth rates come from, if not falling from heaven.

This was the starting point of economists like Romer (1986, 1990), Lucas (1988), Aghion and Howitt (1992), among others. Romer effectively rearranged the industrial sector as it was considered by Solow by explicitly including the research and development sector, the role of patents and trademarks, the sector for intermediate inputs and the consumer sector, thus using endogenous technology as the main driver of economic performance. Now, technological change became the central role in explaining economic growth and the rate of per capita GDP growth equals the rate of technological change on the steady-state growth path. “And the efficiency of technology and knowledge production is enhanced by the—historically developed—stock of scientific-technology knowledge” (Acs et al. 2009, p. 16). Basic to the Romer model are the assumptions that first, all knowledge is economic knowledge and secondly, that knowledge spills over. Arrow (1962) emphasized knowledge as inherently different from traditional factors of production, resulting in a gap between new knowledge and what he termed economic knowledge. While new knowledge leads to opportunities that can be exploited commercially, economic knowledge holds commercial opportunity. Economically useful scientific-technological knowledge consists of non-rival, partially excludable elements of knowledge like codified knowledge published in books, papers, or patent documentations, and rival, excludable knowledge. The latter contains elements of tacit knowledge of individuals, like experience, insights and individual learning. Knowledge spillovers are the results of inter-temporal spillovers, which yield the endogenous growth. In other words, investment in R&D in the present period automatically generates returns in future periods.

Endogenous growth theory stimulated the emergence of fruitful, promising new fields in academia, not only in economics and management but also geography, sociology, and finance. In particular, endogenous growth theory refocuses the light away from capital and labor as the main drivers of wealth and growth towards technology, education, and various forms of financial resources, social developments, and different forms of public policy. Stimulating growth could thus not only be done in a Keynesian way but also by fostering firm R&D, education policy or by increasing the market forces to stimulate agents to invest in knowledge and technology. The introduction of knowledge into macroeconomic growth models leads to the formulation of knowledge production function. As before, physical and labor were econometrically linked to growth, but now in a revised form by considering knowledge and knowledge spillovers. One of the most prevalent models found in the literature of technological change is Griliches (1979) model of the knowledge production function. Firms exist exogenously and engage in the pursuit of new economic knowledge as an input into the process of generating innovative activity and thus economic growth. The most decisive input into the knowledge production function is new economic knowledge. Subsequent to Griliches’ article, a plethora of studies empirically tested the knowledge production function. Though the economic concept of innovative activity does not automatically lend itself to exact measurements (Griliches 1990), a number of measurements are developed such as patent inventions, new product introductions, market growth of new products, growth in productivity, or export performance of new products as proxies for innovative output. The ensuing literature that empirically tested the knowledge production function model generates a series of econometrically robust results substantiating Griliches’ view. Several refinements are made subsequently. Jaffe (1989), coming back to findings from Mansfied et al. (1977a, b), considers that R&D spillovers constitute unambiguous positive externalities. Cohen and Levinthal (1990) developed the capacity to adapt new knowledge and ideas in other firms and to absorb external knowledge. This key insight implied that by investing in R&D, firms could develop the absorptive capacity to appropriate at least some of the returns accruing to investments in new technology made external to the firm.

While considerable empirical evidence supports the knowledge production function model, linking knowledge inputs to innovative output, this relationship apparently becomes stronger as the unit of observation becomes increasingly aggregated, particularly at the country level of analysis (Acs and Audretsch 1990; Audretsch 1995). This raises several questions, in particular, whether knowledge is only firm specific, automatically spills over, is only imbedded in large and established firms (incumbents), and why the relationship between knowledge inputs and innovative outputs in empirical studies is strongest on the country level. Providing answers to these questions are in the core of the knowledge spillover theory (KST).

3.2 Knowledge spillover theory and the role of spatial context

Globalization combined with technological change, in particular the information and communication technology breakthroughs, have rendered obsolete the comparative advantage in low technology and even traditional moderate-technology industries. This revolution has also brought two developments that were largely unanticipated; the first involves economic geography and the organizational context of the sources of knowledge production and absorption. The fact that countries differ in large according to the economic growth of their regions is a well-known phenomenon. Differences in climate, natural resources, and culture, among others played a major role in explaining different levels of growth and path dependencies. The question regarding external sources of knowledge spillovers turned academic attention to one of the largest inventions of mankind—the university. Since their first appearance in the 13th century, universities played a crucial role in developing regions by the spillovers of the knowledge produced within their boundaries. Thus, regions and geographic proximity have (re)emerged as important spatial units of economic activity. While knowledge generated and produced within large and incumbent firms was analyzed in large and also that knowledge of incumbent firms may spill over to other firms (Mansfield et al. 1977a, b; for the chemical industry), the role of universities and research industries as sources of knowledge was rather neglected. With the observation that some regions, in particular the Silicon Valley of the United States, as well as the “Route 128,” areas around San Antonio, Raleigh, and Durham among others, are not only the center of commercial innovation showing above average growth rates, but instead they are also located around research intensive and prestigious universities. Due to this phenomenon, new streams of literature began analyzing the geographic proximity to universities and knowledge spillover. Jaffe (1986, p. 957) states that it is “certainly plausible that the pool of talented graduates, the ideas generated by faculty, and the high quality libraries and other facilities of research universities facilitate the process of commercial innovation in their neighborhood…” For example, as Carlsson and Friedh (2002) examines, only half of the invention disclosures in US universities result in patent applications, from which only about half result in actual patents. Only one third of these patents are licensed, and only 10–20 % of licenses yield a significant income. Thus, Braunerjhelm et al. (2010, p. 107) state, “only 1 or 2 % of inventions are successful in reaching the market.” However, we ask, what about the other 98 % of uncommercialized ideas? Only about 25 % of the inventions resulted in patents as codified knowledge—the overwhelming amount rests in tacit knowledge. The higher the level of R&D activities, the more knowledge is produced, the greater the level of absorptive capacity as well as the pool of tacit knowledge which remains unexplored and could potentially be exploited and transformed into economic knowledge. The ability to transform knowledge into economic knowledge involves not only a set of skills and insights, but also local proximity to the source of the knowledge. The matching mechanism between inventors and economic agents who commercialize the inventions works best when both parties have access to R&D and entrepreneurial skills (Michelacchi 2003).

As one of the first who systematically and empirically analyzed this phenomenon, Jaffe (1986) finds compelling evidence to support the theory of geographic proximity and knowledge spillovers of universities (Acs et al. 1994; Audretsch and Feldman 1996; Audretsch and Stephan 1996; Audretsch and Lehmann 2005a, b). The underlying “trigger” or the mechanism between geographical proximity and knowledge spillovers is tacit knowledge. While codified knowledge as patents, academic articles, and books among others, could easily be transmitted over long distances with rather low costs, tacit knowledge is bound to the individual as the source of knowledge (Kogut and Zander 1992). The “transport” mechanism, according to Jaffe (1986, p. 957), is informal conversation, and thus geographic proximity to the spillover source may not only be helpful but may also even be necessary in capturing the spillover benefits. University research as a source of spillovers has been measured in various studies by quantity and quality. Quantity is often measured by the amount of money spent on R&D, the number of articles published in scientific and academic journals, the number of employees engaged in research or the number of patents (Henderson et al. 1998; McWilliams and Siegel 2000; Varga 2000; Hall et al. 2003), while quality effects are expressed by the number of citations of patents and articles, the specific human capital of researchers or the position in national and international research rankings. While earlier research was more concerned on quantity effects, more recent work highlights the importance of quality and the nature of spillovers and differentiates among the social sciences and the natural sciences or tacit and codified knowledge (Audretsch et al. 2004, 2005).

The emergence of innovative clusters around universities and the explosion of new firms within these clusters questioned the dominant role of large and established firms as the main and sole source of knowledge and innovation. Most of the budget spent in R&D was made to increase production efficiency via investments in automatization and therefore the replacement of human labor by machines, logistics, and firm infrastructure. The amount of capital spent in R&D today leads to an increase in profits tomorrow and thus economic growth. These inter-time knowledge spillovers are captured in the Solow model. While large-scale firms have been observed as the backbone of industry, welfare and growth, mass production serves as the primary guarantee for employment (see Chandler 1977).

Academic researchers increasingly criticize this perspective, instead arguing that innovation output is linked to firm size, and they promote compelling evidence for the importance of small and medium-sized firms within the innovation process (Acs and Audretsch 1988, 1990; Audretsch 1995). This also leads to a renaissance of Schumpeter’s work on the role played by entrepreneurial firms in society. Instead of living in a neoclassical world without entrepreneurial firms in the steady-state equilibrium, entrepreneurial firms are not real and existing; they also play an increasing role expressed by employment rates, growth rates and innovation (Audretsch 1995). Entrepreneurial activity has long been observed to vary not only across industries but also across geographic space (Reynolds et al. 1994; Audretsch and Fritsch 2002).

This puts the interest of research towards the relationship between knowledge spillovers from scientific institutes and entrepreneurial activities. Audretsch and Stephan (1996, 1999) use joint articles written together by scientists working in the industry and university context. They show that the spillover of knowledge to a new startup firm facilitates the appropriation of knowledge for the individual scientist but not necessarily for the organization that was the originator of that new knowledge. Linking universities and startups in the biotech sector, Zucker et al. (1998) find evidence suggesting that it is not spillovers per se but rather the intellectual capital of prominent scientists that plays a major role in shaping both location and timing of the new firm entry into the market. Shane (2001a, b) finds compelling evidence that universities create technological spillovers that are exploited by new firms. Audretsch et al. (2004, 2005) show that the kind of knowledge matters for entrepreneurial activities. They differentiate between knowledge in the natural science and in the social science and show that geographic proximity plays a greater role in accessing and absorbing university spillovers when knowledge is tacit.

Although the relationship between research output on the one hand and entrepreneurial activities on the other could be confirmed empirically, the empirical link is strongest the less aggregated the unit of observations becomes, which is why we observe the strongest relationship in studies characterized with individual cases. This creates an interesting dilemma since the production of knowledge as measured by patents, patent inventions, or academic and scientific research articles has increased drastically since the 1990s (see Kortum and Lerner 1997, p. 1).Footnote 2 Thus, a large part of the new ideas and inventions are not turned into knowledge that necessarily constitutes a commercial opportunity. As suggested by Arrow (1962), not all of the knowledge produced is economically useful. Since the production of knowledge is far away from being costless, the production of knowledge that could not be converted into economic knowledge is a sunk-loss of capital and other resources. This may lead to competitive disadvantages in particular to the new emerging countries like China, Russia, and India with a more centralized and governed R&D policy, avoiding duplication of research and thus less squandering of resources. Therefore, the question arises, which factors prevent or constrain spillovers that limit the efficient conversion of new knowledge into economic knowledge. The answer to this puzzle is provided by the concept of the knowledge filter leading to the knowledge spillover theory of entrepreneurship.

3.3 The knowledge filter

In their early framework, Acs et al. (2004) develop the concept of a knowledge filter. They conceptualized the combination of factors that function as barriers limiting the total conversion of innovative knowledge into economic knowledge to be utilized in the market via new products, processes, and organizations. They argue that the knowledge filter “must be penetrated for knowledge to be appropriated, packaged, modified, and enhanced for it to ultimately contribute to economic growth” (Acs and Plummer 2005, p. 442). Those willing and able to penetrate the filter, in order to enable and absorb knowledge spillovers, are either incumbent firms—or new ventures. While incumbent firms are endowed with the capacity to recognize, evaluate, and absorb knowledge from external resources, they also account for about 25–50 % of the patents produced, which are not used in an economic sense. This sheds some light on the importance of new ventures as a means to penetrate the knowledge filter. The initial contribution of Acs et al. (2004) influenced several studies testing the importance of new ventures and entrepreneurship in penetrating the knowledge filter (Acs and Plummer 2005; Mueller 2006). They all confirm that the stock of knowledge is mainly transformed to economic knowledge by new ventures. It is the entrepreneur who acts as the opportunity seeker because the arbitrage of knowledge resources is a particular specialty of alert and motivated entrepreneurs (Kirzner 1979, 1997).

These agents actively penetrate the knowledge filter and incur the costs of doing so with the expectation that they will reap the returns. New ideas and knowledge are characterized by uncertainty, and the inertia inherent in decision-making under uncertainty within incumbent organizations reflects the knowledge filter (Acs et al. 2004; Audretsch et al. 2006).

Not only does holding the individual attributes constant but also varying the knowledge context gives rise to the KSPE. However, the consideration of entrepreneurship as an endogenous response to the incomplete commercialization of new knowledge resulting in entrepreneurial activity provides the missing link in recently developed economic growth models. By serving as a conduit of knowledge spillovers, entrepreneurship serves as an important source of economic growth that otherwise will remain missing (Audretsch 1995). Therefore, entrepreneurship is the mechanism by which society more fully appropriates its investments into the development of new knowledge, such as research and education. The knowledge spillover theory of entrepreneurship posits one source of entrepreneurial opportunities: new knowledge and ideas (Acs and Armington 2004; Acs et al. 2004; Audretsch et al. 2006; Audretsch and Keilbach 2004, 2007). As shown in the previous sections, ideas and knowledge created within one organizational context such as an incumbent firm or university, but left uncommercialized as a result of the uncertainty inherent in knowledge, serves as a source of knowledge generating entrepreneurial opportunity (Audretsch and Keilbach 2007).

4 Localized competition

Three core conjectures derive from the KSPE. First, the knowledge hypothesis states that, “ceteris paribus, entrepreneurial activity will tend to be greater in [spatial] contexts where investments in new knowledge are relatively high, since the new firm will be started from knowledge that has spilled over from the source producing that new knowledge” (Audretsch et al. 2006, p. 44). Second, the commercialization efficiency hypothesis predicts the “the more efficiently incumbents exploit knowledge flows, the smaller the effect of new knowledge on entrepreneurship” (Acs et al. 2009, p. 17). Finally, since acquiring knowledge requires spatial proximity, the localization hypothesis predicts that, “knowledge spillover entrepreneurship will tend to be spatially located within close geographic proximity to the source of knowledge actually producing that knowledge” (Audretsch et al. 2006, p. 29).

While the commercialization efficiency hypothesis has yet to be tested directly, the evidence that does exist is inconclusive. Even if commercialization efficiency can be properly measured and analyzed, the focus on such efficiency as a constraint on the entrepreneurial opportunities available for discovery causes the KSTE to overlook and obscure the factors that reduce an entrepreneur’s incentive to exploit the opportunity she does discover.

First, building on a regional knowledge production framework, Plummer and Acs (2012) start from the premise that within a region all the knowledge created by private industry and universities (as well as other public entities such as federal labs) is subject to potential discovery by alert entrepreneurs. Second, whether an opportunity is exploited depends on the degree of competition for opportunity in the region. Integrating this second premise into the KSTE involves nothing more than expanding Jane Jacobs’ view of “localized competition” already naturally imbedded into the KSTE (Audretsch and Keilbach 2004; Audretsch et al. 2006; Audretsch et al. 2012).

4.1 The KSTE with localized competition

To extend and generalize the KSTE, Plummer and Acs (2012) start by establishing the knowledge production function by which new knowledge is created in a region. The most common conceptual framework for analyzing the geographic spillover of new knowledge is a regional knowledge production function whereby the knowledge output in a given region is the product of the research and development performed by private industry and the research performed by universities given local economic conditions (Jaffe 1986; 1990). A key premise of the knowledge production framework is that university research creates both new ideas for industry research to adopt and develop and supplies industry R&D labs with key human capital in the form of scientific and technical expertise (Feldman 1999). Thus, both the prevailing theory and empirical evidence suggests that local university research and industry research spatially interact in a way that “boosts” the knowledge output of the region (Anselin et al. 1997; Acs and Varga 2005; Klarl 2013). Moreover, a key assumption of the knowledge production framework is that knowledge spillovers are more prevalent in regions with greater knowledge investments (Audretsch and Feldman 1996).

Plummer and Acs (2012) extend the theory by positing that, given broad institutional constraints, knowledge-driven entrepreneurship reflects the attempt by individual entrepreneurs to profit by exploiting new knowledge produced in the region conditional on the intensity of the “local” competition for opportunity. The knowledge-based opportunities available for entrepreneurial exploitation are the product of both industry research and university research given local economic conditions. Consistent with established theory, an individual chooses to become an entrepreneur and start a new venture when the expected profit of the opportunity exceeds the sum costs of needed resources and the anticipated wage earned by working for an incumbent enterprise.

In this theory, local competition for opportunity is related to the number of agents pursuing the same opportunity as the entrepreneur. If other potential agents move to exploit the same opportunity, such competition would reduce the focal entrepreneur’s entrepreneurial profit to the point at which she no longer has an incentive to act (Fiet 2002). Moreover, as Casson (2005b) points out, the concurrent discovery of the opportunity by other agents has no bearing on the focal entrepreneur’s recognition of the opportunity; instead, it is the concurrent exploitation of the opportunity by other agents that reduces the incentive for the focal entrepreneur to act. Indeed, the focal entrepreneur becomes aware of the competition for opportunity only when she seeks to exploit the opportunity and finds it difficult to obtain the necessary resources at a favorable price (Casson 2005b). This point may seem trivial, this logic suggests that the “agents” that reduce the incentive for the focal entrepreneur to act are not other individual entrepreneurs who have merely discovered the same opportunity, but rather the organizations that have marshaled the resources needed to exploit the opportunity. Therefore, this suggests that the competition for opportunity represents the number of existing firms in the region.

If localized competition for knowledge threatens an incumbent’s viability and thus drives it to be more innovative, then such localized competition must also be a major influence on entrepreneurs’ decisions to exploit opportunities manifest in new knowledge. Given this, the effect localized competition has on regional entrepreneurial activity is two-fold. First, as Feldman and Audretsch (1999) find in their analysis, greater localized competition will pressure incumbents to be more innovative, which in turn will tend to expand the pool of opportunities for entrepreneurs to discover. Second, because localized competition by definition means the “ruthless” adoption and imitation of a firm’s innovations by other neighboring firms, it also follows that greater localized competition will tend to reduce the likelihood that knowledge-based opportunities will be exploited by entrepreneurs. As discussed, the reason is that localized competition makes it less likely that, in the entrepreneur’s judgment, the rewards for pursuing the given opportunity will outweigh the costs of doing so (Casson 2005a). Thus, in sum, it follows that localized competition will accelerate the rate of innovation in the region, which in turn begets more entrepreneurial activity; however, it also follows that with greater localized competition less new knowledge is commercialized by new ventures. Three core hypotheses emerge from this argument.

4.2 Density

Extant versions of the KSPE contend that the creation of new knowledge will result in entrepreneurial activity conditional on how efficiently incumbent firms exploit or commercialize new knowledge (Acs et al. 2009). Plummer and Acs’ findings are broadly consistent with Jacobs’ (1969) concept of localized competition. First, using simultaneous equations they find that greater localized competition leads to more knowledge and to higher rates of knowledge-driven entrepreneurship. Second, they find that the positive relationship between new knowledge and high-tech firm birth rates is negatively moderated by greater localized competition. This evidence supports the premise that localized competition has a two-fold effect on knowledge-driven entrepreneurship by increasing the pool of opportunities for entrepreneurs for discovery, but also reducing the share of these opportunities exploited by entrepreneurs.

Moreover, the finding concerning the effect of population density fits the conjecture central to agglomeration theory that dense concentrations of firms, households, and people blunts the forces of competition that would otherwise erode a firm’s competitive advantage and performance (cf. McCann and Folta 2008). In particular, the effect of density operates by reversing the effect localized competition has on negatively moderating the relationship between new knowledge and high-tech firm births. Density neither has a direct effect on entrepreneurial activity nor moderates the respective relationships between localized competition and new knowledge and the rate of high-tech firm births. This suggests that regions with prodigious capacities for new knowledge creation and a large number of incumbent firms tightly packed in a relatively small geographic area are particularly conducive to knowledge-driven entrepreneurship because the negative effects of localized competition cannot take hold. More to the point, this finding buttresses the view that localized competition is an important driver of the knowledge spillover entrepreneurship process.

5 Absorptive capacity

Qian and Acs (2013) extend the KSPE by introducing absorptive capacity. First, the term “entrepreneurial absorptive capacity” is introduced, based on which an absorptive capacity theory of knowledge spillover entrepreneurship is developed. Second, deviating from exogenously assumed knowledge, the new model posits that new knowledge is endogenously created. Reflecting this assumption in the theory, the KPF is integrated into the knowledge spillover theory of entrepreneurship. These efforts enable a better understanding of the relationships between knowledge, new knowledge and entrepreneurship, thus further unveiling the mechanism of knowledge spillover entrepreneurship.

Now the focus is shifted to advancing the KSPE. The absorptive capacity theory of knowledge spillover entrepreneurship argues that, the level of knowledge spillover entrepreneurship depends not only on the speed of knowledge creation, or the level of new knowledge, but also on entrepreneurial absorptive capacity. Entrepreneurial absorptive capacity is defined as the ability of an entrepreneur to understand new knowledge, recognize its value, and subsequently commercialize it by creating a firm.

Entrepreneurial absorptive capacity varies among people who are potential entrepreneurs. It basically has two dimensions. On the one hand, it involves scientific knowledge the individual should have in order to understand what a new invention really is and further to recognize its market value. On the other hand, it relies on market or business knowledge with which the individual can successfully create and further operate a new firm. Both types of knowledge are indispensable for knowledge spillover entrepreneurship. In the typical case of the knowledge spillover theory of entrepreneurship, as discussed by Acs et al. (2009) and Audretsch (1995), the inventor who develops a new technology already has the scientific knowledge, and thus her/his success in commercializing the new technology to a great extent depends on the market knowledge she/he bears to start up and operate a business (Audretsch et al. 2009).

Introducing entrepreneurial absorptive capacity makes two major contributions to the KSPE. First, it connects new knowledge and the entrepreneurial action of starting a new firm. The act of entrepreneurship involves not only where the opportunity is but also the process of discovering and exploiting the opportunity. The theory, while clarifying new knowledge as one source of entrepreneurial opportunities, has not well explained whether entrepreneurs can discover and exploit entrepreneurial opportunities. The single conduit of knowledge spillover suggested by the theory is the previously-discussed inventors’ career choice model. In this case, entrepreneurial discovery is less important since the inventor simply needs to estimate the potential market value of the invention, and does not need to learn what the invention exactly is (because she/he already knows). The inventors’ career choice model may suggest whether to exploit a new invention, but sheds little light on whether the inventor is capable of doing it. Entrepreneurial absorptive capacity addresses such capability especially in terms of market knowledge to undertake entrepreneurial discovery and exploitation.

Second, non-inventors’ entrepreneurial actions to appropriate the market value of new knowledge developed in large firms and universities can be integrated into the knowledge spillover theory of entrepreneurship through entrepreneurial absorptive capacity. With strong absorptive capacity, an entrepreneur will have sufficient scientific and market knowledge to understand a new invention developed by others, recognize its market value, and apply it at the commercial end by creating a new firm. For non-inventor entrepreneurs who are generally businessmen and already have certain market knowledge and resources, scientific knowledge must be obtained at the start so as to perceive an entrepreneurial opportunity embedded in a new invention. Entrepreneurial absorptive capacity is important for both inventors’ and non-inventors’ entrepreneurial activity. In the former case, the inventors’ capacity to commercialize new knowledge (i.e., market knowledge) is more needed. In the latter case, by contrast, the entrepreneurs’ capacity to identify and understand new knowledge (i.e., scientific knowledge) is more desired.

The second effort to extend the KSTE by Quin and Acs is consistent with Agarwal et al. (2007), to endogenize the knowledge production process, which contributes to a better clarification of the mechanism by which knowledge influences entrepreneurship. In this specific context, endogenizing knowledge production is also desired because knowledge is closely associated with entrepreneurial absorptive capacity that has been introduced. To do that, the KPF developed by Romer (1990) is adopted, but is simplified as a human capital model:

$$ d(A) = f(H) $$

where new knowledge is simply a function of human capital. There are two reasons why this simplification works. First, human capital refers to knowledge and skills embodied in people, thus capturing not only the knowledge stock but also R&D workers, both of which are input factors in Romer’s KPF. Second, even for Romer’s KPF, new knowledge is ultimately reliant on human capital, since the knowledge stock is the inter-temporal accumulation of new knowledge.

Qian and Acs (2013) identifies two conduits through which human capital or knowledge embodied in people influences entrepreneurship. The first one is via the creation of new knowledge that contains entrepreneurial opportunities, and the second one is via the building of entrepreneurial absorptive capacity which allows the entrepreneur to successfully commercialize new knowledge by starting a new firm. This dual-conduit mechanism can better explain the role of knowledge in entrepreneurial activity, or in other words, provide better insights on knowledge-based entrepreneurship than can explain the knowledge spillover theory of entrepreneurship.

6 The papers

Despite the emergence of a vast and compelling literature identifying the existence and impact of knowledge spillover entrepreneurship, a number of crucial questions and answers remain unanswered. To shed light on these issues, we assembled a meeting of leading scholars of entrepreneurship and innovation from around the world for a conference on “The Knowledge Spillover Theory of Entrepreneurship” at the University of Augsburg in August 2012. The following papers were selected as the most salient, adding the most value and insight to a dynamic literature in not just entrepreneurship but also spanning a broad spectrum of academic fields, such as economic growth, innovation, regional studies, organizational behavior, and strategy. The first two papers extend and enlarge the existing theoretical models on KSTE. The following papers consist of empirical and anecdote frameworks highlighting topics in KSTE which, until know, have not been explored.

Entrepreneurship scholars following Schumpeter see entrepreneurship as the “missing link” in converting knowledge into economically relevant knowledge. Acs et al. (2004, 2009) and Braunerhjelm et al. (2010) advanced the microeconomic foundations of endogenous growth theory by developing a knowledge spillover theory of entrepreneurship (KSTE). Knowledge created by incumbent firms’ R&D results in knowledge spillovers to entrepreneurs who identify and exploit these opportunities for profit. Acs and Sanders (2013) contribution ‘Knowledge Spillover Entrepreneurship in an Endogenous Growth Model’ enlarges the microeconomic foundation of KSTE by developing a general equilibrium steady state growth model. They add to the literature on KSTE by fully specifying and endogenizing a more general knowledge spillover process in a dynamic general equilibrium framework that enables analysis of welfare implications of KSTE. They present a model in which knowledge generation and commercialization are two separate and costly activities that both need to be properly rewarded for the private sector to engage in them. They separate entrepreneurship from profit motivated corporate R&D and embed the core idea of the knowledge spillover theory of entrepreneurship in established knowledge-based growth models by enriching their knowledge spillover structure. Introducing knowledge spillovers drives a wedge between the optimal and market allocation of resources between new knowledge creation and commercialization. The optimal ratio of R&D to entrepreneurship then depends exclusively on the relative strength of knowledge spillovers and concludes that policy can improve welfare over the market outcome.

One central assumption in the KSTE is that the set of technological opportunity is endogenously created by investment in new knowledge. One seminal source of the creation of knowledge is universities and research laboratories (Link and Walsh 2013). In their paper ‘Knowledge Spillovers, Collective Entrepreneurship, & Economic Growth: The Role of Universities’, Leyden and Link (2013) shift the lense towards the role universities play in facilitating the transmission of knowledge to private-sector business enterprises so as to generate economic growth. They build on the Knowledge Spillover Theory of Entrepreneurship to develop a formal model of university-with-business enterprise collaborative research partnerships in which the outcome is both mutually desirable and feasible. Their model shows that if a university seeks to act as a complement to private-sector collaborative R&D so that it will be attractive to both incumbent firms and startup entrepreneurs, it needs to structure its program so that business enterprise revenues increase and business enterprise R&D costs rise by a smaller proportion than revenues increase, if they rise at all (and a fall would be better). Such a structure is consistent with both business enterprise and university interests but is only likely to be feasible if the university is subsidized to cover the cost of such public-private collaborative research partnerships. In the absence of such support, the university will have to cover its costs through a fee charged to participating business enterprises and that will result in the university being seen as a substitute rather than a complement to private-sector collaborative R&D, and thus the university will be seen as an unattractive partner for many business enterprises.

Previous studies on knowledge spillover entrepreneurship treat opportunities as endogenous and generally focus on opportunities recognized and acted upon by entrepreneurs in creating new firms. In their paper ‘The Missing Pillar: Creativity Theory of Knowledge Spillover Entrepreneurship’, Audretsch and Belitski (2013) argue that knowledge spillover entrepreneurship depends not only on ordinary human capital, but more importantly on creativity embodied in creative individuals and a diverse urban environment that attracts creative classes. They extend the KSTE by introducing creativity as an additional source of knowledge spillover of entrepreneurship that has not been adequately researched in the literature. Their theoretical extension of Romer’s (1990) knowledge production function is supported by empirical evidence on European cities over the 1999–2010 period, sheds light on the impact of creativity on entrepreneurship. They interpret their results as a kind of self-selection process of creative individuals into entrepreneurship that enables them to recognize creativity and commercialize their creativity.

The paper ‘University Specialization and New Firm Creation Across Industries’ examines how the scientific specialization of universities impacts on new firm creation across industries at the local level. In accordance with the Pavitt-Miozzo-Soete taxonomy, Bonaccorsi et al. (2013) consider eight industry categories, which reflect the characteristics of firms’ innovation patterns and, ultimately, the knowledge inputs they require. Their paper adds to the KSTE by offering a comprehensive answer to two interrelated research questions: how does the allegedly positive effect of university knowledge on new firm creation at the local level vary across industries, and how does this effect depend on the scientific specialization of universities. Using data on new firm creation in Italian provinces, they estimate negative binomial regression models separately for each industry category to relate new firm creation to the scientific specialization in basic sciences, applied sciences and engineering, and social sciences and humanities of neighboring universities. They confirm that universities specialized in applied sciences and engineering have a broad positive effect on new firm creation in a given province, with this effect especially strong in services. Conversely, the positive effect of university specialization in basic sciences is confined to new firm creation in science-based manufacturing industries, even if this effect is of large magnitude. Universities specialized in social sciences and humanities have no effect on new firm creation at the local level whatever industry category is considered.

Based on detailed information about the regional knowledge base, particularly about universities, Fritsch and Amoucke (2013) add to the literature on KSTE in several ways. Their paper ‘Regional Public Research, Higher Education, and Innovative Start-ups—An Empirical Investigation’ analyzes the role played by regional knowledge, particularly academic knowledge, in the emergence of innovative start-ups in Germany. Their study makes several significant contributions to the KSTE. First, they provide an explicit focus on start-ups in innovative and knowledge-intensive industries. Second, they identify a strong link between start-ups and higher education institutions as a source of knowledge spillovers which allows identifying those parts and features of a region’s academic knowledge base that are the most relevant for innovative start-ups. While nearly all the earlier studies are based on pure cross-sections, their analysis uses relatively long time series data that allow employing panel estimation techniques. They find that regional public research and education have a strong positive impact on new business formation in innovative industries but not in industries classified as non-innovative. They also find compelling evidence that measures for the presence and size of public academic institutions have more of an effect on the formation of innovative new businesses than do indicators that reflect the quality of these institutions. The relatively weak evidence for interregional spillovers of these effects clearly demonstrate the importance of localized knowledge and, especially, of public research for the emergence of innovative new businesses.

In his empirical study ‘Knowledge and Entrepreneurial Employees: A Country Level Analysis’, Stam (2013) argues that the existing empirical evidence on the KSTE widely ignores entrepreneurial activities of employees within established organizations, although in multiple advanced capitalist economies entrepreneurial employee activity is more prevalent than independent entrepreneurial activity. Stam thus shifts the perspective on KSTE away from the traditional entrepreneur towards entrepreneurial activities within established organizations. His paper presents the outcomes of a large scale international study into entrepreneurial employee activities, highlighting that innovation indicators are positively correlated with the prevalence of entrepreneurial employee activities, but are not or are even negatively correlated to the prevalence of independent entrepreneurial activities. This would provide further evidence on the relevance of entrepreneurial activities within established organizations and thus adds to the KSTE. Since most policy attention until now has been focused on stimulating individuals to become independent entrepreneurs, investments in innovation in established organizations might as well be the source of opportunity recognition and pursuit by entrepreneurial employees.

Hayter (2013) argues that efforts to promote and support knowledge-based entrepreneurship as a vehicle for economic development are increasingly focused on the importance of networks to entrepreneurial success. In his paper ‘Conceptualizing Knowledge-based Entrepreneurship Networks: Perspectives from the Literature’, he reviews the extant empirical literature and finds a striking consensus among multiple disciplinary perspectives in that networks' characteristics mediate resources important to entrepreneurial performance. He criticizes that current conceptual frameworks do not adequately account for the unique nature of knowledge spillovers and their role in innovation and economic dynamism. He adds to the literature by suggesting that scholars should embrace the nascent Knowledge Spillover Theory of Entrepreneurship to guide future empirical research on entrepreneurship networks and focus intently on their impact on entrepreneurial performance—and therefore economic growth.