The survival of academic spinoff companies: An empirical study of key determinants

Abstract

The formation of university spinoff companies has been studied extensively, yet limited attention has been devoted to their survival. Consequently, little is known about spinoff’s later stage developments. Spinoff companies exist in university networks where they access resources through different types of actors. However, it remains unclear on which actors specifically these firms should focus their networking efforts, especially in relation to their success. It is also poorly understood how the regional economic environment affects spinoff survival. This article examines the core determinants of survival of academic spinoff companies. The article analyses a unique sample of 870 UK spinoff companies from 81 universities formed between 2002 and 2013. The results show that spinoff company survival is dependent on three core university network actors: investors, external entrepreneurs and technology transfer offices (TTOs). In addition, spinoff companies born into less industrially diversified regions enjoy greater probability of survival.

Keywords

entrepreneurship firm survival networks university spinoffs

Introduction

There is growing evidence that university spinoff companies are important to economic development (Vincett, 2010), as they constitute one of the modes of translating academic research into societal and economic impacts. Consequently, over the past two decades, scholars have devoted significant efforts to improve understanding of how to increase the efficiency of this commercialisation process to form more spinoff companies (Fini et al., 2017; Lockett and Wright, 2005; Miranda et al., 2018; Mustar, 1997).

Recent studies have concentrated on understanding the later stages of spinoff companies, especially post-start-up performance (Iacobucci and Micozzi, 2015; Visintin and Pittino, 2014) and their success (BVCA/Library House, 2005; Hayter, 2016a). Performance of spinoffs is typically measured as growth or change in turnover and/or employment (Scholten et al., 2015; Visintin and Pittino, 2014). While success has been mostly treated as reaching Vohora et al.’s (2004) sustainability stage – where a firm achieves stable returns, other studies define success as progression through Vohora et al.’s critical junctures (Hayter, 2016a), which refers to any form of a firm’s achievement in terms of operational persistence.

Another approach used in the field of industrial organisation is to observe the survival of a firm’s operational existence or persistence over time, which offers an insight into the factors that lead to its sustainability or broadly defined success. Firm survival is regarded as a comprehensive indicator of a firm’s performance related to profitability (Alchian, 1950), efficiency (Jovanovic, 1982) or simply higher productivity levels (Klepper and Simons, 2000), based on an optimal arrangement of its resources and capabilities (Kogut and Zander, 1992). Until now, only a small number of studies have investigated the survival of spinoff companies, with a very narrow scope for any generalisations given their focus on either a single university and industrial concentration (Nerkar and Shane, 2003), or a single region and human capital (Criaco et al., 2014).

It is acknowledged that spinoffs exist in networks of actors that contribute resources to their successful development (Hayter, 2016b; Rasmussen et al., 2015). The literature on spinoffs typically identifies four types that play critical roles in their lifetimes: investors, technology transfer offices (TTOs), business incubator, and experienced entrepreneurs. While much of the engagement of these actors is typically ascribed to the early stages of spinoff development, it remains unknown whether they influence the survival of these firms. Furthermore, the spatiality of spinoff companies has attracted limited interest (e.g. Iacobucci and Micozzi, 2015), particularly how diverse regional economic structures relate to their survival resulting from predominantly small sample studies. This is especially important as it makes it difficult to establish what policy instruments, at national or regional levels, would best serve and support spinoffs, given their significant return on public investment in underlying research (Vincett, 2010).

This article aims to answer the following research questions: First, which network actors determine the survival of spinoff companies? Second, how are regional economic development conditions related to spinoff survival? To answer these questions, a sample of 870 UK spinoff companies from 81 universities formed between 2002 and 2013 is examined in terms of their survival. The academic spinoff company is defined here as a firm founded by academic staff to exploit university-generated intellectual property.

Our findings show that the university network plays a critical role in spinoff company survival, particularly, when it is composed of investors, external entrepreneurs and TTOs. Furthermore, the survival of the spinoff is also related to its network structure and regional industrial character. The major implication of these findings stems from the importance of the networks and their specific architectures, pointing to greater survival prospects of spinoffs with less constrained network positions.

This article is structured as follows: First the literature on survival, networks and spinoffs is introduced, followed by a detailed outline of the methodology, and the presentation and discussion of results, with the conclusions offered in the final section of the article.

Firm survival and networks

Firm survival

In the lifecycle theory of the firm and its industrial sector, there are four key struggles that define firm success. These are liabilities of newness, liabilities of adolescence, liabilities of senescence and liabilities of obsolescence (Coad, 2017). Liability of newness posits that younger firms are more likely to fail, as they have not had enough time to develop their unique resources and capabilities, resulting in decreased external credibility and legitimacy (Bower, 2003). Some evidence for this can be observed from studies on diversification strategies typically employed by older firms (Borghesi et al., 2007). In contrast to this, the liability of adolescence acknowledges that firms experience a ‘honeymoon effect’ (Hudson, 1987) in their first years of operations, due to sunk costs (Murray, 1988), which leads to higher survival rates in the early years, and an increase in failure in the subsequent years.

The remaining two liabilities can be described best as issues of inertia (Hannan and Freeman, 1984). Over time, firms develop their human capital, procedures, routines, technologies and become reluctant to change when the conditions in the industry present competition-type challenges. This leads to a loss of market or productivity disadvantage, and over time results in increased risk of failure. For example, these processes of inertia have been linked to the evolution of oligopolies (Klepper and Simons, 2000), where only the adaptable firms remained in the market. However, firms that enjoy a dominant position in their industry may avoid such existential threats, even despite organisational rigidity (Van Witteloostuijn, 1998).

These liabilities present a set of non-linear characteristics of firm’s age, where at some point age changes the direction of its relationship with firm’s survival (Coad, 2017). For example, Jelfs (2016) and Lawton Smith et al. (2014) observed that UK spinoffs experienced limited failure rates in their first three years (‘honeymoon’), only increasing afterwards, suggesting spinoff’ liability of adolescence. As such, the survival of a firm is dependent on its human capital, the characteristics of the firm, and the broader macro-environment. There is an established body of literature that finds a generally positive relationship between human capital and firm survival (e.g. Gimmon and Levie, 2010). Spinoffs benefit from the commercial (Wennberg et al., 2011) and management skills of their external entrepreneurs (Grandi and Grimaldi, 2003). For example, Criaco et al. (2014) found that spinoffs whose management teams had entrepreneurial education or included an academic founder had improved survival rates. Conversely, Nerkar and Shane (2003) observed previous start-up experience to be positively associated with a spinoff’s failure, suggesting a very complex interplay of human capital factors explaining spinoff survival.

Firm characteristics, such as demographics and innovativeness, are also critical to a firm’s survival. In particular, a firm’s age (Coad, 2017), size (Holmes et al., 2010) or the number of company founders (Cressy, 1996) are positively associated with its survival. The importance of human capital to firm success extends into its innovative activities, given the need for greater absorptive capacity (Cohen and Levinthal, 1990) in the form of a more educated workforce (Mata and Portugal, 2002). In the case of spinoffs, Nerkar and Shane (2003) found that their technological radicalness and broader patent scope are positively related to survival, pointing to the strong role of a firm’s innovativeness.

Being embedded in their regional environment makes firms dependent on unique socio-economic and spatial conditions (Bristow, 2010). A number of studies found a relationship between firm survival and its location (Howell, 2017; Huggins et al., 2017). For example, firm survival is related to the economic conditions of its territory, whether measured by output growth (Huggins et al., 2017; Jensen et al., 2008), industrial structure (Tavassoli and Jlienwatcharamongkhol, 2016) or industry concentration (Nerkar and Shane, 2003). Consequently, it is impossible to disentangle the firm from its broader environment without taking a more network-based perspective (Huggins and Prokop, 2016).

Firm survival and networks

Firm success is dependent on its ability to configure its resources and capabilities in a way that makes them valuable and difficult to reproduce by other firms (Barney, 1991; Teece et al., 1997; Wernerfelt, 1984). This is particularly important as the creation of such unique firm characteristics involves the use of networks (Kogut, 2000) to appropriate and process knowledge into a firm’s operational and strategic decision-making (Jovanovic, 1982; Sidney and Winter, 1964), and products or services. Such knowledge encompasses business management, finance, markets and product or service innovation. Although networks are found to play a critical role in firm survival (Bosma et al., 2004; Littunen, 2000; Singh and Mitchell, 1996; Wilson et al., 2013; Zhao and Burt, 2018), they are not commonly studied in industrial organisation, as information on firm networks is generally difficult to capture.

Networks exert a dual influence on firm activities (Borgatti and Foster, 2003), one side stemming from network structure (Burt, 1992) and the other derived from what is transmitted in the networks (Coleman, 1990). The structural aspects relate to the composition of a network, which can be strongly interconnected (approaching closure) or sparsely, where actors may exist on a spectrum between strong homogeneous structures or weaker diverse configurations. These characteristics have a particular set of effects on firm performance. Firms that are part of strong homogeneous groups exhibit problems of inertia, path dependence and homophily (Hite and Hesterly, 2001), limiting their performance due to a lack of risky ties with newer diverse actors. Conversely, firms that exist in poorly connected networks remain less constrained in developing ties to heterogeneous actors, allowing them to exploit network opportunities (Burt, 1992) by increased proactivity in searching for new ties (Hite and Hesterly, 2001). This has been observed by Franklin et al (2001) in their study of spinoff formation, which found that UK universities with more open networks generate more spinoffs. Furthermore, Capaldo (2007) highlighted that actors located in between the two extremes enjoy greater innovation performance.

The resources exchanged in networks allow understanding of the nature of particular networks. Specifically, the type of resource transmitted defines whether the networks have a more social character (e.g. friendship), or economic (Hite and Hesterly, 2001; Huggins, 2010) (e.g. innovation). This resource access is important to firm performance. Firms unable to form ties struggle to satisfy their growth needs (e.g. financial capital, specialist skills), while those that increase their connectivity improve their success prospects (Borgatti and Foster, 2003). Hite and Hesterly (2001) indicate that such strong resource access is particularly critical at the emergence stage, when firms resort to their established identity-based ties that form socially cohesive sets (Borgatti et al., 1990). At later growth stages, firms tend to form ties with a select group of actors that bring in specific resources as and when they are required (Hite and Hesterly, 2001; Huggins, 2010). Such heterogeneous resource access requires structural network adjustments beyond the comforts of a homophilous set (Lin, 2001), essentially combining the structural network aspects with the resource ones. Grandi and Grimaldi (2003) point out that the first networks that spinoffs build are inherited from their parent organisations, indicating the cohesiveness and embeddedness of their initial networks. As spinoffs develop, their parent-centred networks shift towards commercially oriented contacts (Hayter, 2016b), suggesting the transition from ascribed social to purposively acquired network capital (Hite and Hesterly, 2001; Huggins, 2010).

Since firms in their infancy suffer from resource-starvation and legitimacy problems, they would naturally refer to their networks to overcome such issues. For example, networking allows spinoffs to build credibility, legitimacy (Bower, 2003; Bruneel et al., 2012), secure necessary funding and further opportunities for growth (Grandi and Grimaldi, 2003) and survival (Mustar, 1997). In other words, academic spinoffs develop networks primarily to access resources otherwise not available to them (Hayter, 2013a, 2016b). Furthermore, the close homophilous links are found importantly related to firm survival (Littunen, 2000), indicating that firms need to maintain their early developed network in order to build new connections, especially to heterophilous actors (Hite and Hesterly, 2001; Lin, 2001). As such, early stage firms rely more on social capital for survival (Bosma et al., 2004), particularly that of their founders (Hite and Hesterly, 2001; Huggins, 2010). Therefore, new firms with larger boards of directors exhibit lower risk of failure (Singh et al., 1986). However, the effect of social capital on firm survival is also visible among more established firms (Pennings et al., 1998). Furthermore, a firm established within a region with which its founder is familiar enjoys a greater probability of success, thanks to the regional embeddedness explained by greater accumulation of social capital (Dahl and Sorenson, 2012).

As firms grow, they switch from their reliance on social capital towards more instrumental and calculative network capital (Huggins, 2010), that provides specific and unique resources unavailable from homophilous relations. In essence, such firms start decreasing their network constraint (Burt, 1992), by connecting with new and heterophilous actors, resulting in improved entrepreneurial and survival opportunities. However, firms remaining in embedded sets that are not proactive in such network building processes may suffer from poorer survival prospects as their connections reconfigure their own positions by adding new ties (Singh and Mitchell, 1996). Such overreliance on the cohesion of a network leads to decreased structural holes, resulting in inferior performance (Zaheer and Soda, 2009).

University spinoff survival and network actors

The academic entrepreneurship literature has devoted much attention to the conditions underlying the birth of spinoff companies (Miranda et al., 2018). Although this is a salient issue, it neglects the stylised facts of the industrial organisation, for example, the strong and positive correlation between firm formation and failure (Geroski, 1995). Some limited voice was given to this matter when the Lambert’s Review (2003) concluded that universities should focus on the quality of spinoffs, rather than their numbers. While universities may have acted in line with such suggestions, little scholarly effort has been expended on understanding the factors that contribute to spinoff survival, with the notable exceptions of Nerkar and Shane (2003) and Criaco et al. (2014). This is critical, given that spinoffs originate from non-commercially oriented individuals and organisations of charitable status, endowing them with a distinctive set of features in comparison with the remaining populations of firms, as evidenced in studies of Swedish university spinoffs compared to corporate spinoffs (Wennberg et al., 2011), US venture-capital-backed university spinoffs compared to other venture-capital-backed firms (Zhang, 2009) and Italian spinoffs compared to other high technology firms (Civera et al., 2019). Furthermore, UK studies reported that university spinoffs do not reflect the characteristics of high-growth firms, but rather struggle financially (Jelfs, 2016) and could be classified as lifestyle companies (Harrison and Leitch, 2010), making their survival particularly important to understand. In addition, a spinoff’s survival may be linked to the character of the parent university (although not ownership level as observed by Ferretti et al. (2019)), especially its eminence (Lawton Smith et al., 2008), traditional science focus (Shane, 2004), or patenting activity (De Nicola et al., 2018), but also with the seniority and gender of the academic founders, given that research suggests that academic entrepreneurship is more prevalent among older academics (Hewitt-Dundas, 2015) with a strong male bias (Lawton Smith et al., 2017).

Networks are critical to understanding spinoffs in a university-centred environment (Hayter, 2013a). Of particular importance here are the networks with TTOs, investors, external entrepreneurs and business incubators, which constitute a heterogeneous set of key actors with linkages to spinoff companies. Specifically, investors and external entrepreneurs have a heterophilous character, with the TTOs and especially on-campus-based business incubators being homophilous in nature, suggesting that the former would be positively related to a spinoff’s survival, while the latter could have detrimental or no effect on its survival.

Overall, networks encompassing a range of actors can be recognised as critical to a spinoff’s success (Hayter, 2016a) as they are expected to provide them with greater access to a wider set of actors and resources. Specifically, an academic spinoff’s network capability is significant for its growth, performance and sustainability (Walter et al., 2006), suggesting that networks are equally important at every stage of the spinoff’s development (Hayter, 2016b). In fact, spinoffs need to adapt their networks by building and restructuring their network position, aligning it with changing needs and expectations over time (Rasmussen et al., 2015), especially as a spinoff’s strong network position is related to its performance (Scholten et al., 2015) and survival (Mustar, 1997).

A specific set of key actors forming university network is important to a spinoff’s success. As such it is essential to offer clarity between what constitutes a network and a university network. The term network is used here as a generic expression meaning groups of actors or ties between actors and could represent both social and network capital (Huggins, 2010). Conversely, university network has a specific meaning in terms of its relational composition. When using this term, the article considers university, TTO, business incubator, external entrepreneurs and investors. As university network includes actors that have a more deterministic function, this embeds them in a network capital conceptualisation (Huggins, 2010).

Investors

Much of the literature on academic spinoffs has focused on the role played by investors, primarily in the form of business angels (Mosey and Wright, 2007), public funds (Huggins, 2008), university venture funding (Munari et al., 2018) or the private venture capital industry (Bonardo et al., 2011). These actors contribute vital resources that add to the development of a spinoff company and ensure its growth (Clarysse et al., 2011b). In exchange for equity, investors offer finance to otherwise cash-starved businesses (Van Geenhuizen and Soetanto, 2009), allowing them to process university knowledge into technologies and finally develop revenue-generating products or services. The receipt of investment not only influences a spinoff’s growth (Rodríguez-Gulías et al., 2017) but also signals its credibility (Vohora et al., 2004), quality (Fini et al., 2017) and entrepreneurial orientation (Diánez-González and Camelo-Ordaz, 2016). This is because investors are often highly risk-averse (Mason and Stark, 2004) and engage in high-selectivity of spinoff ventures (Clarysse et al., 2005). Such signalling is often used as a prerequisite for other investors engaging with the venture, especially as spinoff companies require multiple rounds of funding to fuel their growth (Wright et al., 2006) in order to reach the stage of sustainability (Vohora et al., 2004) or scalability (Bigdeli et al., 2015). Furthermore, there is a tendency for investors, especially venture capitalists (VCs), to co-invest in order to reduce information asymmetries related to higher risk companies (Brander et al., 2002) and improve own investment performance associated with greater survival prospects of firms funded by such syndicates (Hochberg et al., 2007). Consequently, spinoffs that can attract a greater number of investors undergo more frequent selectivity evaluations, signalling their better survival prospects:

Hypothesis 1. There is a positive relationship between the number of investors holding equity stakes in a spinoff company and its survival.

External entrepreneur

Although academics are found to be entrepreneurial (Fini et al., 2010), only a proportion of them have the right mind-set, skills or networks to develop their own company (Goethner et al., 2012). The successful creation of an academic spinoff (Hayter, 2013b) and further growth (Lundqvist, 2014) are positively influenced by engaging an experienced entrepreneur. These effects are especially visible in the strategies employed at the universities most successful at creating spinoff companies (Lockett et al., 2003) recognised as more flexible and open to such actors. The core benefits experienced entrepreneurs bring to the venture include business experience, networks (Franklin et al., 2001) and commercial (Wennberg et al., 2011) and management skills (Grandi and Grimaldi, 2003). In fact, the presence of external entrepreneurs is positively related to entrepreneurial orientation of academic spinoffs (Diánez-González and Camelo-Ordaz, 2016). These entrepreneurs are crucial in securing investment (Vohora et al., 2004) and adding credibility to their management teams (Wright et al., 2006). In addition, Visintin and Pittino (2014) observe that it is the diversity of the founding team (including academic and non-academic entrepreneurs) that is critical to a spinoff’s performance. This was also noted by Criaco et al. (2014) who found that the university experience of a member of the founding team (or the presence of an academic founder) is positively related to spinoff survival. Therefore, a spinoff company’s survival is expected to be associated with the presence of such experienced entrepreneurs:

Hypothesis 2. There is a positive relationship between the presence of experienced external entrepreneur(s) and spinoff company survival.

TTOs

While the main purpose of the TTOs is to commercialise university-generated intellectual property (Fitzgerald and Cunningham, 2016), there is a growing body of literature which suggests that they are necessary for other knowledge commercialisation activities (Bourelos et al., 2012), beyond forming spinoff companies (Aldridge and Audretsch, 2011). This is particularly important as TTOs are traditionally associated with creating spinoff companies (Shane, 2004). Clarysse et al. (2011a) indicate that faculty entrepreneurial intentions are independent of the presence of TTOs; therefore, it is unlikely that they have any major impact on academic entrepreneurship. This concurs with Vohora et al. (2004), who found that it is frequently the inventor who recognises the commercial opportunity of a piece of research.

A further complication of the TTO’s role arises from the findings of Fini et al.’s (2017) study, suggesting that the quality of the spinoff companies in attracting first round of VC funding is negatively affected by the presence of a TTO. However, TTO’s initial seed investment in the spinoff is positively associated with the receipt of venture capital (Gubitta et al., 2016).

Since TTOs are subject to resource constraints and organisational rigidities present at universities, it is expected that the TTOs that have been operating for longer are able to accumulate the knowledge of supporting spinoffs along with the evidence to justify to the university administrators the need for greater endowments. As such, they can translate these resources into improved support for the spinoffs, increasing their survival prospects:

Hypothesis 3. There is a positive relationship between the experience of the TTO and spinoff company survival.

Business incubator

Business incubators are critical to spinoff formation and widely discussed (Degroof and Roberts, 2004), considered as one of the key modes of organisational or governmental intervention in nurturing entrepreneurship (M’Chirgui et al., 2018). These facilities offer a list of support services: office space, administration, training, investment, consultancy and professional business management support (Bruneel et al., 2012; Grimaldi and Grandi, 2005; Wynarczyk and Raine, 2005). Their key role is to accelerate business development and reduce the chance of a firm failing by isolating it from immediate market risks (Carayannis and von Zedtwitz, 2005). However, in studies of university spinoff rates (Gonzalez-Pernia et al., 2013), they were not found to be significantly influenced by the university access to business incubators. This may result from the fact that the majority of the spinoff companies are not based in any business incubators (Hewitt-Dundas, 2015). Consequently, it remains difficult to expect business incubators to shield spinoffs from failure:

Hypothesis 4. There is no relationship between access to business incubators and spinoff company survival.

Spinoff’s network position

Although much of the academic spinoff literature stresses the role of networks (e.g. Bourelos et al., 2012), little comprehensive empirical evidence exists. Specifically, while access to particular university network actors conveys important information on a spinoff company’s ability to ensure its successful development (Hayter, 2016b), the structure of the concomitant network and the position of a spinoff within it is important to understanding its survival. For example, Scholten et al. (2015) found that bridging ties of Dutch spinoff companies played a role in explaining their early growth. Spinoffs occupying more central positions (Freeman, 1978) may enjoy greater network benefits compared to those with poor centrality, which may result in greater access to knowledge and resources (Hayter, 2016a).

Furthermore, spinoffs may gain opportunities from exercising their favourable network positions by being the main linking actor between its most immediate connections, if these connections are poorly or not interlinked (Burt, 1992). This is especially important when firms reconfigure their network positions by closing non-redundant connections (Gulati, 1999) to allow them to bypass structural limitations of their networks. Some evidence of this has been observed in the evolution of the spinoff company networks (Rasmussen et al., 2015). Therefore, spinoffs that have a lower degree centrality constraint (DCC) (i.e. high level of non-redundant connections based on degree centrality rather than Burt’s (1992) strength of ties) are expected to have greater prospects of survival:

Hypothesis 5. There is a negative relationship between spinoff’s DCC and its survival.

Regional industrial structure

Much of the research on spinoff companies has been aspatial in nature, largely due to poor availability of data and focus on the entrepreneurial processes without considering the macro-environment within which these processes take place. Studies that control for the regional aspects considered the macro-activity of R&D intensity in the region (Fini et al., 2017; Lockett and Wright, 2005), the formation context (Fini et al., 2011) or the importance of spinoffs to the regional economy (Benneworth and Charles, 2005; Berggren and Dahlstrand, 2009), yet the determinants of spinoff survival may differ from those of spinoff formation. In particular, the regional context is clearly important to spinoffs (Garnsey and Heffernan, 2005; Gonzalez-Pernia et al., 2013; Rodríguez-Gulías et al., 2018; Lawton Smith et al., 2014) and to firm survival (Tavassoli and Jlienwatcharamongkhol, 2016), given each region’s unique organisation of actors and resources. For example, Iacobucci and Micozzi (2015) found that spinoffs with greater turnover were based in more developed and industrially diversified Northern Italian regions (Mameli et al., 2012). In the United Kingdom, stocks of existing entrepreneurship are positively related to industrial diversity and specialisation, yet new venture registrations are greater in less industrially diverse regions (Huggins and Thompson, 2016), indicating that firms born into less industrially diverse regions like London (O’Donoghue, 2016) would enjoy higher survival rates, with this effect reversing once the firms become more established. It is, therefore, anticipated that spinoffs born in industrially diverse and specialised regions will be characterised by lower survival outcomes:

Hypothesis 6a. The industrial diversity of the region in which a spinoff is born is negatively related to its survival.

Hypothesis 6b. The industrial specialisation of the region in which a spinoff is born is negatively related to its survival.

Methods

Data collection

The university spinoff data were collected from an Internet service dedicated to UK spinoffs (www.spinoutsuk.co.uk) on 12 January 2014, which covered the company name and its parent university(/ies) for 1303 companies founded from the year 2000 onwards. In addition, the names of spinoff companies were obtained from all UK university websites. This helped ensure that the robustness of the dataset was maximised. A similar approach was employed by Hewitt-Dundas (2015). As virtually all UK spinoffs take the legal form of a limited company at the start-up stage, the list of company names was further enriched with information on the spinoffs from a Bureau van Dijk’s FAME (Financial Analysis Made Easy) database, which holds detailed demographic and financial data on registered UK companies obtained from Companies House, as used in similar research (e.g. Lawton Smith et al., 2008). This step provided information on the spinoff’s demographics: (1) status–live/deregistered, with exact dates of incorporation and dissolution; (2) size, based on employment information or the type of filed accounts; (3) sector classification according to Standard Industrial Classification 2007 coding (Office for National Statistics (ONS), 2009); (4) address; (5) a list of directors and their details; and (6) shareholders. The list of shareholders was used to construct a network of UK spinoffs and their connections.

The final sample frame for which data were found consisted of 1331 companies and is the closest reflection of the population of such firms studied in the United Kingdom, details of which are largely unknown and unreported in the literature. The consequent and unfortunate limitation of this is the restricted comparability of the sample’s characteristics to those of the census of academic spinoff companies to assess its representativeness. To partially overcome such restriction, a comparison is made across 87 universities present in the sample frame and their active spinoff company numbers with data reported in HEFCE’s (Higher Education Funding Council for England) HE-BCIS (Higher Education-Business and Community Interaction Survey) (Appendix 1). A Wilcoxon signed-rank test was performed, indicating no statistically significant difference between the two samples (Z = −1.156, p = 0.248, r = −0.124).

The data were complemented with university-specific information obtained from HE-BCIS, which provided data on universities related to their Third Mission activity and measures for the university network actors. Furthermore, based on the addresses of firms, respective regional data from Office for National Statistics (ONS) were added to the database to describe the industrial structure of regions in which those firms were based. The geographical unit utilised is based on the European Union’s NUTS (Nomenclature of Territorial Units for Statistics) 1 level, which covers 12 UK regions. In addition, university data were collected from HESA (Higher Education Statistics Agency) to capture university size.

As the data on spinoff companies need to be matched across a number of datasets, there is a natural limitation on the availability of such data across time. While the sample frame captured companies formed between 1959 and 2013, with survival event recorded on 1 May 2014, the official HESA statistics were only available for 2002–2014 at university level. ONS data are less restricted in time with regional employment by sector data available from 1998 to 2008 from Annual Business Inquiry and 2009 to 2016 from Business Register and Employment Survey. Consequently, the analysis presented here focuses on spinoff companies formed between 2002 and 2013, while the data itself cover the period 2002–2014. This limitation reduces the number of original spinoff companies studied from 1331 to 870, while the number of universities is reduced from 87 to 81. These changes, although reflected in the descriptive statistics presented in Table 1 – especially seen through reduced mean (M = 17.01, N = 1331; M = 11.89, N = 870) and standard deviation (SD) values (SD = 22.68, N = 1331; SD = 14.01, N = 870) of spinoffs formed, do not impose any further major changes to the characteristics of the final sample, with both skewness and kurtosis values only slightly reduced. The sample is representative of the sample frame across both regions (χ²(11, N = 2201) = 4.08, p = 0.97, V = 0.04) and sectors (χ²(6, N = 2201) = 2.74, p = 0.82, V = 0.04) of firms, but not age or size, for obvious reasons.

Table 1.

Number of spinoffs generated: descriptive statistics of sample frame (1959–2013) and sample (2002–2013).

	Sample frame	Sample
Mean	17.01	11.89
SD	22.68	14.01
Skewness	2.28	2.15
Kurtosis	5.71	5.33
Minimum	1	1
Maximum	106	67
Quartiles
Low	2.00	2.00
Median	7.00	6.00
Top	24.00	17.50
N	87	81

SD: standard deviation.

Variables

The survival of spinoff companies is represented by a binary variable that records the status of each firm (Criaco et al., 2014). The status represents whether the company is registered with Companies House (i.e. live firm) or not (i.e. deregistered firm) (Wennberg et al., 2011). While this is a frequently applied approach in industrial organisation literature that focuses on the typical liquidation (Huggins et al., 2017), it does not capture the full variability related to different motivations for exit or deregistration (e.g. withdrawal of university support). However, the analysis controls for two key alternative forms of exit: (a) related to personal circumstances of the company directors (age of directors) and (b) related to acquisitions (successful exit). Given that spinoffs are typically dependent on external investments in order to develop the underlying IP (Clarysse et al., 2011b; Huggins, 2008), the exits would be anticipated to take the form of liquidation. In particular, when the company is formed through the careful selection of disclosures (Clarysse et al., 2005), there is an expectation of long-term commitment from the university to maintain the firm alive (due to sunk costs; Murray, 1988). Consequently, dormant firms are also treated here as live firms, as they are typically younger spinoffs (M = 5.31, SD = 3.53) than non-dormant (M = 6.69, SD = 3.18) firms (U = 9961.5, Z = −2.47, p < 0.05), and their exclusion from the sample does not affect the results (Appendix 2). Within the sample, 82.87% of spinoff companies were still registered as live firms on 1 May 2014, confirming a high survival rate of university-born firms (Vincett, 2010), in line with results typically reported for other European countries (Bolzani et al., 2014; Jelfs, 2016). Given the conceptual model tested in this article, the data on the university network actors are sourced from FAME and HE-BCIS.

In order to measure whether spinoffs received investment, considered a firm quality indicator (Fini et al., 2017), a measure is used that captures the number of equity-type investors at each spinoff company. This came from shareholder information, where firms and organisations providing entrepreneurial finance (e.g. VCs, business angel networks, public seed funds) were identified through online searches.

To measure the presence of experienced external entrepreneurs (Visintin and Pittino, 2014), a proxy variable is composed from the directors’ information. The variable uses an average of the number of directorships held by spinoff company’s directors, with an expectation that the higher values of the measure would be indicative of the presence of experienced external entrepreneurs.

Given that all universities in the sample have a TTO, the commercialisation experience is measured by establishing the difference in years between spinoff’s birth and the formation of the TTO unit (Fini et al., 2017). While information on a spinoff company’s use of business incubators is very difficult to capture, especially since these are not widely used by spinoffs (Hewitt-Dundas, 2015), a university’s provision of access to on-campus and off-campus business incubators, and science parks is used instead. Dummy variables are used to describe whether the spinoff had access to each type of incubator at its birth.

The network position of a spinoff is explored with DCC constructed from shareholder data of spinoffs, typically used in social network studies of listed firms (e.g. Li et al., 2016). The meaning of DCC is closely aligned with Burt’s (1992) structural holes. Structural holes measure the power of an actor that depends on the extent to which her or his network is unconnected (non-redundant) (Burt, 1992), with that actor enjoying greater benefits when the network is more unconnected, leaving the actor the main connecting node. The DCC departs from the structural holes at the measurement of the strength of ties, where instead it relies on the centrality of actors.

The premise behind structural holes is that actor $i$ benefits most if $i$ can negotiate between $j$ and $k$ , when $j$ and $k$ cannot negotiate directly. Once $j$ and $k$ develop a relation/link, $i$ ’s relations become redundant and structural hole advantage is diminished. Structural holes depend on the strength of ties; hence, if $i$ is connected strongly to $j$ and $k$ , while $j$ has a weak relation with $k$ , $i$ still maintains advantage in the network. Structural holes are measured as a constraint $C$ of $j$ on $i$ , where $i$ is more constrained in a network with more redundant ties, as it has few or no structural holes to benefit from (Burt, 1992). Therefore, the aggregate constraint $A G$ of actor $i$ is defined as follows

A G_{i} = \sum_{j} C_{i j}

where

C_{i j} = {(p_{i j} + \sum_{k, i \neq k \neq j} p_{i k} p_{k j})}^{2}

and

p_{i j} = \frac{a_{i j} + a_{j i}}{\sum_{k} (a_{i k} + a_{k i})}

with $p_{i j}$ being the proportion of relation values $a$ from actors $i$ to $j$ , and $j$ to $i$ , out of the sum of all relation values $\sum_{k} (a_{i k} + a_{k i})$ of actor $i$ (Burt, 1992). $A G_{i}$ takes a value of 1 for unconnected actors (with degree centrality of 0) in Pajek. For actors with 1 connection (i.e. degree centrality of 1), the aggregate constraint also takes value of 1, as the term $\sum_{k, i \neq k \neq j} p_{i k} p_{k j}$ would be equal to 0 reducing $C_{i j}$ to ${(p_{i j})}^{2}$ .

Due to resource limitations, the university network studied here has undefined values for relationships, which by default take value of 1 (i.e. $a = 1$ ), consequently excluding the theoretical considerations of the strengths of ties. As a result the $C_{i j}$ depends on degree centrality $D C$ of network actors

C_{i j} = {(\frac{1}{D C_{i}} + \frac{1}{D C_{j}})}^{2}

p_{i j} = \frac{a_{i j} + a_{j i}}{\sum_{k} (a_{i k} + a_{k i})} = \frac{1 + 1}{D C_{i} (1 + 1)} = \frac{1}{D C_{i}}

therefore

C_{i j} = {(p_{i j} + \sum_{k, i \neq k \neq j} p_{i k} p_{k j})}^{2} = {(\frac{1}{D C_{i}} + D C_{i} \frac{1}{D C_{i} D C_{j}})}^{2} = {(\frac{1}{D C_{i}} + \frac{1}{D C_{j}})}^{2} = {(p_{i j} + p_{k j})}^{2}

where the aggregate constraint is a product of degree centrality of $i$ and $j$ . Therefore, the variable is assumed to measure the DCC, rather than Burt’s (1992) original aggregate constraint. The variable is calculated for all network actors (i.e. universities, spinoffs, and their shareholders); however, the values used here are only those that describe the network position of spinoff companies. The approach employed here focuses on ego-centred formalised networks with clear engagement expressed through equity ownership.

The regional industrial structure is depicted with two variables that originate from Theil’s (1972) work on employment entropy to measure industrial diversity and an adopted dissimilarity index to measure industrial specialisation. These measures are widely established in economic geography literature (e.g. Fotopoulos, 2014; Huggins and Thompson, 2016). Here, they are used from employment at NUTS1 regions at two-digit SIC (Standard Industrial Classification) sectors. The industrial diversity $D I V_{r}$ of region $r$ is defined as follows

D I V_{r} = \frac{\sum_{i} \frac{p_{r i}}{p_{r}} l n \frac{p_{r}}{p_{r i}}}{l n (I)}

where

p_{r i} = \frac{E_{r i}}{\sum_{r} \sum_{i} E_{r i}}

and

p_{r} = \sum_{i} p_{r i}

With $E_{r i}$ representing employment in region $r$ and industry $i$ , $p_{r i}$ defining a proportion of employment in industry $i$ to a sum of such employments across all $I$ industries for each region $r,$ $p_{r}$ being sum of proportions $p_{r i}$ across all $I$ industries for each region $r$ , while $I$ representing the total number of industries considered, in this case 15, as in Fotopoulos (2014) (Appendix 3). The variable is bound within [0, 1] interval, where higher values represent more industrially diverse regions.

The industrial specialisation $S P E C_{r}$ of region $r$ is measured as follows

S P E C_{r} = \frac{1}{2} \sum_{i} | \frac{E_{r i}}{E_{r}} - \frac{E_{n i}}{E_{n}} |

where $E_{r}$ is the total employment in region $r$ , $E_{n i}$ is employment in industry $i$ at the national level, while $E_{n}$ is the total employment at the national level. The variable takes values within a [0, 1] interval, where higher values signify more industrially specialised regions. Both variables describe regional industrial structure at a spinoff company’s birth in its parent university’s region.

A number of variables are also used to control for the effects of the characteristics of the spinoff and the university. The firm controls capture age and its quadratic form (Coad, 2017), size expressed as a dummy if the firm is small (Harrison and Leitch, 2010), sector (Criaco et al., 2014) represented by four dummies (detailed composition of variables is available in Appendix 4), successful exit dummy capturing whether the spinoff underwent exit in any form: initial public offering (IPO) (Mustar et al., 2008) or acquisition (Nerkar and Shane, 2003); number of directors measuring an attempt at executive team building to incorporate experienced individuals (Vohora et al., 2004); age of directors controlling for personally motivated exits (Becker, 1965); nationality of directors representing more entrepreneurially minded non-native academics (Krabel and Mueller, 2009) or external entrepreneurs (Fairlie and Lofstrom, 2015); and the region’s size measuring employment in a region spinoff company is born into (Tavassoli and Jlienwatcharamongkhol, 2016). The number of parent universities is controlled for, as spinoff companies may have more than one parent institution (Lockett et al., 2003). Finally, the university characteristics represent university size (Fini et al., 2017) expressed as total university income, counts of patents (Gonzalez-Pernia et al., 2013) and science bias (SB) (Shane, 2004).

Based on the Research Assessment Exercise (RAE) 2008 and the Research Excellence Framework (REF) 2014, SB reflects university’s focus on high-quality scientific research. The traditional science fields typically generate more spinoffs due to more tangible IP protection method of patenting (Shane, 2004), and therefore the focus is on REF’s (and respective fields in RAE 2008) Panels A and B (www.ref.ac.uk). The approach taken here is to focus on the overall quality profile and outputs deemed world-leading (rated 4*). High-end research (HER) transforms proportion-expressed RAE and REF scores into total submission numbers for every ith university (separately for RAE and REF)

H E R_{i} = \sum_{j} R O_{i j} \times N O S_{i j}

where HER is the high-end research (submission numbers of 4* quality); RO is the percentage of research outputs with overall quality score of 4*; NOS is the number of submissions; i is the university; and j is the research field.

SB is a proportion of $H E R$ in Panels A and B over $H E R$ in all Panels (A–D)

S B = \frac{H E R_{A B}}{H E R_{A B C D}}

The variable composed here uses an average of SB measure for RAE 2008 and REF 2014 covering a timeline of 2001–2013.

Table 2 offers a brief description of variables. The descriptive statistics for the variables are presented in Tables 3 for continuous variables and Table 4 for categorical variables.

Table 2.

Description of variables.

Dependent variable
Spinoff survival	A binary variable taking value of 1 if the company was still registered as a live firm on CH on 1 May 2014
Independent variables
Investment	The number of institutional investors (seed funds, business angels, VCs) that were recorded as spinoff company’s shareholders on 1 May 2014
External entrepreneur	An average of the number of directorships held by spinoff company directors on 1 May 2014
Commercialisation experience	The age of a TTO since or until its creation at spinoff’s birth. In cases with multiple university parents, the age of the oldest parent is used
On-campus incubator	A binary variable taking value of 1 if the spinoff company had access to on-campus business incubator at its birth
Off-campus incubator	A binary variable taking value of 1 if the spinoff company had access to off-campus business incubator at its birth
Science park	A binary variable taking value of 1 if the spinoff company had access to science park at its birth
Degree centrality constraint	A modified version of Burt’s (1992) structural holes where aggregate constraint is dependent solely on degree centrality and not on the strength of ties. It measures the position of spinoff companies within their networks of shareholders
Industrial diversity	Industrial diversity of a spinoff’s region at company’s birth
Industrial specialisation	Industrial specialisation of a spinoff’s region at company’s birth
Control variables
Age	Spinoff’s age
Age²	A quadratic term for spinoff’s age
Small firm	A binary variable taking value of 1 if the spinoff company was a small firm (0–49 employees) on 1 May 2014
Manufacturing sector	A binary variable taking value of 1 if the spinoff company registered its main activities in the manufacturing sector
Information and communication	A binary variable taking value of 1 if the spinoff company registered its main activities in the information and communication sector
Professional, scientific and technical sector	A binary variable taking value of 1 if the spinoff company registered its main activities in the professional, scientific and technical sector
Human health and social care sector	A binary variable taking value of 1 if the spinoff company registered its main activities in the human health and social care sector
Successful exit	A binary variable taking value of 1 if the spinoff company was acquired by another firm or underwent an IPO
Number of directors	The number of company directors on 1 May 2014
Age of directors	Average age of spinoff’s directors on 1 May 2014
Nationality of directors	The nationality is expressed as a proportion of spinoff’s directors that are British out of all directors
Parent universities	The counts of parent institutions involved in the formation of a spinoff company
University size	Total university income at spinoff’s birth. For spinoff companies with multiple university parents, the sum of total incomes is used. The variable is expressed in billions of pound sterling
Patents	The number of university granted patents at spinoff company’s birth. Where spinoff companies had more than one UK parent institution, the values of the patents were added together. The number of patents is expressed in thousands
Science orientation	Proportion of RAE or REF outputs of 4* quality that are in traditional science fields (REF’s Panels A–B) out of 4* outputs in all academic fields (REF’s Panels A–D) of a university corresponding to spinoff’s birth year. In cases when a spinoff was formed by more than one university, the sum of 4* outputs of all universities involved was first established in order to calculate the percentage
Region’s size	The size of the spinoff’s region at company’s birth expressed in total employment across all sectors. The variable is expressed in millions

VC: venture capitalist; TTO: technology transfer office; IPO: initial public offering; RAE or REF: Research Assessment Exercise or Research Excellence Framework.

Table 3.

Descriptive statistics of continuous variable (N = 870).

Variable	Mean	Std. deviation	Minimum	Maximum
Investment	1.2	1.96	0	14
External entrepreneur^a	3.41	3.87	0	50
Commercialisation experience	16.19	10.08	−2.77	44.32
Degree centrality constraint	0.21	0.21	0.01	1
Industrial diversity	0.75	0.03	0.66	0.81
Industrial specialisation	0.07	0.04	0.03	0.16
Age	6.64	3.21	0.42	12.28
Number of directors	4.48	2.41	1	18
Age of directors	47.06	6.91	21.86	68.54
Nationality of directors	0.86	0.24	0.00	1.00
Parent universities	1.13	0.39	1	4
University size	£0.42	£0.33	£0.02	£2.26
Patents	0.39	0.62	0	3.69
Science orientation	0.54	0.17	0	1
Region’s size	2.53	0.95	0.67	4.45

N = 869.

Table 4.

Properties of categorical variables (N = 870).

Variable	Yes	No
Spinoff survival	721	149
On-campus incubator	753	117
Off-campus incubator	723	147
Science park	636	234
Small firm	857	13
Manufacturing sector	144	726
Information and communication sector	147	723
Professional, scientific and technical sector	451	419
Human health and social work sector	30	840
Successful exit	81	789

Table 5 presents correlations of all variables, with a number of interesting relationships with dependent variable uncovered. Six of the eight independent variables have significant correlations with the dependent variable, notably Investment, External entrepreneur, Commercialisation experience, On-campus incubator, DCC and Industrial diversity all of expected coefficient signs. Despite the sample size (N = 870), there are two pairs of variables that could indicate presence of multicollinearity: Industrial diversity and Industrial specialisation (r = 0.71), and University size and Patents (r = 0.76). Only the first pair presents abnormal behaviour when entered together, with inflated coefficients and significance of Industrial specialisation compared to a separate entry. To further test for multicollinearity, variance inflation factors (VIFs) were obtained by fitting an ordinary least squares (OLS) regression. When the two industrial structure variables are entered together, their VIFs are 8.0 for industrial diversity and 4.3 for industrial specialisation, confirming moderate multicollinearity issues. To limit any imprecision in the estimation of the coefficients, the variables are entered in separate models.

Table 5.

Correlations between variables (N = 870).

		1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22	23	24	25
1	Spinoff survival	1.00
2	Investment	0.07	1.00
3	External entrepreneur	0.10	0.14	1.00
4	Commercialisation experience	0.13	0.03	0.06	1.00
5	On-campus incubator	0.07	0.04	0.05	0.16	1.00
6	Off-campus incubator	0.06	0.09	−0.04	0.01	0.10	1.00
7	Science park	0.00	−0.01	−0.12	−0.02	−0.02	0.09	1.00
8	Degree centrality constraint	−0.15	−0.42	−0.04	−0.05	−0.02	−0.14	−0.03	1.00
9	Industrial diversity	–0.08	0.10	0.01	−0.15	−0.04	–0.07	0.30	0.03	1.00
10	Industrial specialisation	−0.05	−0.04	−0.01	−0.04	–0.08	0.05	−0.37	–0.07	−0.71	1.00
11	Age	0.08	0.27	0.07	−0.22	−0.15	0.04	−0.03	−0.30	0.40	0.02	1.00
12	Age²	0.15	−0.03	−0.04	0.06	−0.05	0.00	0.03	0.13	0.11	0.02	0.02	1.00
13	Small firm	−0.01	−0.17	−0.02	–0.07	0.01	0.00	0.01	0.09	−0.03	0.00	−0.12	−0.05
14	Manufacturing sector	0.02	0.04	−0.06	−0.04	−0.05	0.03	−0.03	−0.10	0.00	0.07	0.10	0.00	0.00	1.00
15	Information and communication sector	0.00	−0.05	−0.10	−0.06	0.04	0.00	0.02	0.02	0.06	−0.06	0.00	0.00	0.00	−0.20	1.00
16	Professional, scientific and technical sector	0.07	0.10	0.10	0.09	0.01	0.06	0.00	–0.08	−0.03	0.00	−0.03	0.06	−0.02	−0.46	−0.47	1.00
17	Human health and social work sector	0.00	−0.06	0.00	−0.04	0.00	−0.05	−0.01	0.11	−0.03	0.02	−0.03	−0.06	0.02	−0.08	−0.09	−0.20	1.00
18	Successful exit	0.14	0.01	0.04	0.08	0.01	−0.02	−0.01	−0.07	0.00	−0.05	0.08	0.02	−0.09	0.00	−0.05	0.02	0.03	1.00
19	Number of directors	−0.35	0.40	0.02	0.03	−0.07	0.01	0.01	−0.26	−0.02	0.08	0.11	−0.15	−0.19	0.04	−0.09	0.02	−0.01	−0.03	1.00
20	Age of directors	0.12	0.07	0.11	0.09	0.02	0.01	0.00	−0.03	−0.06	−0.03	−0.05	−0.03	−0.03	0.03	−0.17	0.11	0.07	0.04	0.11	1.00
21	Nationality of directors	–0.08	0.00	0.09	–0.08	−0.06	−0.04	−0.05	0.07	0.11	−0.04	0.05	−0.01	0.04	0.02	−0.03	0.01	0.00	−0.17	0.05	0.13	1.00
22	Parent universities	0.05	0.25	0.04	0.14	0.08	0.11	0.10	−0.19	0.00	0.03	0.15	−0.03	−0.11	0.01	−0.04	0.03	−0.05	−0.02	0.17	0.09	−0.02	1.00
23	University size	0.16	0.12	0.00	0.45	0.18	0.15	0.16	−0.16	−0.33	0.01	−0.22	0.04	−0.07	−0.02	−0.06	0.11	−0.03	0.06	0.07	0.09	−0.16	0.40	1.00
24	Patents	0.11	0.03	−0.02	0.33	0.17	0.16	−0.06	−0.12	−0.55	0.26	−0.27	0.03	−0.01	−0.03	−0.08	0.09	0.02	0.06	0.08	0.05	−0.11	0.25	0.76	1.00
25	Science orientation	0.08	0.07	0.01	0.21	0.11	0.14	−0.05	−0.13	−0.47	0.34	−0.10	−0.02	−0.03	−0.02	−0.02	0.02	−0.02	0.05	0.06	0.01	−0.10	0.07	0.36	0.44	1.00
26	Region’s size	−0.01	0.05	−0.01	0.10	0.01	0.00	−0.14	−0.11	−0.59	0.26	−0.03	–0.08	−0.03	−0.03	−0.06	0.06	0.01	0.04	0.12	0.08	−0.04	0.04	0.33	0.46	0.43

Correlations of external entrepreneur are available for 869 observations only. Correlations in italics are significant at 5% level (two-tailed); correlations in bold are significant at 1% level (two-tailed). Correlations between continuous variables are Pearson’s, correlations between continuous and categorical dummy variables are point-biserial and correlations between categorical dummies are phi coefficients.

The dependent variable has a binomial distribution, taking a value of 0 for spinoff companies that were dissolved, and a value of 1 for firms that were still live on 1 May 2014, with a mean clearly falling between these two values (Hosmer et al., 2013), suggesting a logit regression model, with a similar approach found in Criaco et al. (2014) in their study of Catalan spinoffs.

The model fitted to explain university spinoff company survival $S_{i}$ takes the following form

S_{i} = β_{0} + β_{1} I_{i} + β_{2} E E_{i} + β_{3} C E_{i} + β_{4} B I_{i} + β_{5} D C C_{i} + β_{6} R I S_{i} + β_{7} F C_{i} + β_{8} U C_{i} + β_{9} R C_{i} + ε_{i}

where $I_{i}$ is the equity investment, $E E_{i}$ represents external entrepreneur, $C E_{i}$ is the commercialisation experience, $B I_{i}$ captures the business incubator types, $D C C_{i}$ measures the DCC, $R I S_{i}$ reflects regional industrial structure, $F C_{i}$ are firm controls, $U C_{i}$ are university controls, while $R C_{i}$ are region controls. Out of 870 observations, the analysis in the full model focuses on 869 spinoff companies, given the presence of an outlier in a predictor.

Results and discussion

Table 6 presents a general overview of survival of the UK spinoff companies by formation cohort. While the formation of the spinoff companies has decreased since 2002, there is a clear pattern signifying higher survival rate of spinoffs in their first three years with typically at least approximately 95% of firms in each cohort surviving. In the subsequent two years reported here (i.e. five-year survival), the survival falls at an accelerated rate, suggesting the liability of adolescence (Coad, 2017) or presence of the ‘honeymoon’ effect (Hudson, 1987), observed in smaller samples by Jelfs (2016) and Lawton Smith et al. (2014).

Table 6.

Survival patterns of UK spinoff companies.

Start year	Births	Three-year survival (%)	Five-year survival (%)
2002	102	99.02	97.06
2003	85	100.00	92.94
2004	90	100.00	93.33
2005	81	98.77	87.65
2006	84	98.81	86.90
2007	85	96.47	89.41
2008	63	95.24	90.48
2009	59	94.92	86.96
2010	75	98.67	−
2011	62	96.67	−
2012	64	−	−
2013	20	−	−

N = 870.

The regressed models explaining university spinoff company survival are presented in Table 7, with complementary VIF values in Table 8. The five models presented are an improvement over an intercept-only model (as given by Omnibus test) significant at 1% level, with model 4 offering the best fit as observed from information criteria or pseudo R² measures. Classification plots indicate that full models estimate predicted probabilities correctly for circa 88% of fitted values, in particular, the sensitivity tests present that more than 96% of surviving spinoff companies were correctly predicted by the models. Models 1, 4 and 5 pass the goodness of fit test; however, caution is suggested in interpreting the results of models 2 and 3.

Table 7.

Logit of spinoff survival.

	Model 1	SE	Model 2	SE	Model 3	SE	Model 4	SE	Model 5	SE
Investment	−0.004	0.057					0.309***	0.092	0.259***	0.089
External entrepreneur	0.111***	0.039					0.125**	0.049	0.120**	0.049
Commercialisation experience	0.032***	0.010					0.028**	0.014	0.023	0.014
On-campus incubator	0.333	0.245					0.323	0.343	0.105	0.331
Off-campus incubator	0.202	0.236					−0.312	0.327	−0.281	0.318
Science park	0.068	0.210					0.315	0.290	−0.131	0.290
Degree centrality constraint	−1.582***	0.424					−2.407***	0.592	−2.430***	0.586
Industrial diversity			−7.822**	3.341			−33.451***	7.165
Industrial specialisation					−3.653	2.303			−3.410	3.715
Age							0.267***	0.055	0.159***	0.048
Age²							0.585***	0.155	0.503***	0.153
Small firm							−1.208	1.489	−1.435	1.397
Manufacturing sector							0.628	0.429	0.846**	0.420
Information and communication sector							0.941**	0.418	0.918**	0.408
Professional, scientific and technical sector							0.619*	0.349	0.627*	0.340
Human health and social work sector							1.308*	0.738	1.252*	0.714
Successful exit							3.204***	1.086	3.283***	1.099
Number of directors							−0.686***	0.070	−0.648***	0.066
Age of directors							0.075***	0.018	0.076***	0.018
Nationality of directors							−0.438	0.616	−0.799	0.595
Parent universities							0.058	0.402	−0.005	0.392
University size							1.904***	0.733	1.334*	0.735
Patents							−0.303	0.402	0.404	0.416
Science orientation							−0.161	0.821	0.664	0.828
Region’s size							−0.656***	0.190	−0.211	0.144
Intercept	0.642*	0.39	7.443***	2.515	1.855***	0.201	25.603***	5.791	0.911	2.009
df	8		2		2		25		25
N	869		870		870		869		869
Nagelkerke R²	0.085		0.011		0.005		0.505		0.473
McFadden R²	0.057		0.007		0.003		0.394		0.364
Percentage correct	82.85%		82.87%		82.87%		88.38%		87.80%
Sensitivity	99.44%		100.00%		100.00%		96.67%		96.67%
Specificity	2.68%		0		0		48.32%		44.97%
−2LL	750.864		790.888		794.277		482.564		506.501
AIC	766.864		794.888		798.277		532.564		556.501
BIC	805.003		960.100		963.489		651.748		675.685
Hosmer and Lemeshow test	10.111		17.403**		17.074**		4.235		5.900

SE: standard error; df: degrees of freedom; AIC: Akaike information criterion; BIC: Bayesian information criterion.

***

Significant at 1% level; **5% level; *10% level.

Table 8.

Variance inflation factors.

	Model 1	Model 2	Model 3	Model 4	Model 5
Investment	1.2			1.5	1.5
External entrepreneur	1.0			1.1	1.1
Commercialisation experience	1.0			1.3	1.4
On-campus incubator	1.0			1.1	1.1
Off-campus incubator	1.0			1.1	1.1
Science park	1.0			1.3	1.3
Degree centrality constraint	1.2			1.4	1.4
Industrial diversity		1.0		2.9
Industrial specialisation			1.0		1.5
Age				1.8	1.4
Age²				1.1	1.1
Small firm				1.1	1.1
Manufacturing sector				2.2	2.2
Information and communication sector				2.2	2.2
Professional, scientific and technical sector				2.9	2.9
Human health and social work sector				1.3	1.3
Successful exit				1.1	1.1
Number of directors				1.3	1.3
Age of directors				1.1	1.1
Nationality of directors				1.1	1.1
Parent universities				1.4	1.4
University size				3.5	3.6
Patents				3.4	3.3
Science orientation				1.5	1.5
Region’s size				1.9	1.5

The regression fitted in model 1 introduces variables that define university networks. Models 2 and 3 describe the regional industrial structure of a spinoff. Models 4 and 5 present the full specification, including a set of controls for the firm, university and regional characteristics. The number of investors has a positive association with spinoff survival, confirming Hypothesis 1. Hypothesis 2 is confirmed by the positive and significant coefficient of external entrepreneur. The TTO’s experience of commercialisation is positively associated with spinoff survival, in line with Hypothesis 3, but only in specification with industrial diversity. As expected, no form of business incubation explains spinoff survival (Hypothesis 4). Spinoff’s lower DCC is associated with its greater survival prospects, lending support to Hypothesis 5. Finally, it is found that spinoffs born in industrially diversified regions (Hypothesis 6a) have decreased probability of survival. No support is found for Hypothesis 6b, although the coefficient of the industrial specialisation is of the expected sign.

From the firm-level controls, it is found that a spinoff’s survival is related to age with the non-linear term entering the equation significantly; however, it is independent of its size. All sector dummies are positively and significantly related to spinoff’s survival, except for manufacturing in model 4. As could be anticipated, there is a positive relationship of a spinoff’s survival to its successful exit. Interestingly, the number of company directors has a negative and significant coefficient, suggesting diseconomies of management. At the same time, exits are more prevalent among younger directors. Nationality of the directors is unrelated to spinoff survival. Furthermore, the number of parent universities has no association with spinoff survival. Only one university control is significant – the size of the university, with a positive coefficient. No evidence was found for SB or the importance of patents. Region’s size is only significant in industrial diversity specification.

The fact that the number of investors is positively related to spinoff survival (Hypothesis 1) confirms a wide array of literature praising the role of VC (Fini et al., 2017) but also seed capital, public funds (Huggins, 2008) and business angels (Mosey and Wright, 2007). Clearly, university spinoffs struggle to survive without external capital to maintain and accelerate their development. Such non-organic forms of growth promote quicker transition from a small company to a more established one, improving spinoff’s credibility and distinguishing them from lifestyle-type firms (Harrison and Leitch, 2010). The findings may indicate the viability of universities holding their own VC funds (beyond current seed capital or challenge funds) to reduce risks to potential private investors through matched/joint funding.

Although the engagement of an external entrepreneur in a spinoff has been widely discussed (Hayter, 2013a), limited empirical evidence exists that links the role of external entrepreneurs to spinoff performance (Visintin and Pittino, 2014). It is clear that these actors are also critical to its survival (Hypothesis 2). The important role of such entrepreneurs stems from experience, commercial (Wennberg et al., 2011) and management skills (Grandi and Grimaldi, 2003) and developed networks (Franklin et al., 2001). More generally, entrepreneurs experience is a vital asset to a firm’s survival (Bosma et al., 2004; Taylor, 1999). In addition, external entrepreneurs could also act as facilitators of investment-driven growth as suggested by Vohora et al. (2004), as their introduction at spinoffs could relieve the academic founders’ time from searching for capital to technology development.

The partial results confirming Hypothesis 3 contribute a fresh insight into the role of TTOs, questioned in formation studies (Bourelos et al., 2012). Some scholars argue that TTOs could improve if better resourced and able to employ experienced professionals (Gonzalez-Pernia et al., 2013), while Degroof and Roberts (2004) suggest that TTOs could collaborate for resource access, or as pointed out by Leitch and Harrison (2005), concentrate efforts on research of commercial potential only. Although such implications are critical at the foundation stage, the experienced TTOs also engage in activities that support the survival of spinoff companies.

Business incubators have a very prominent place in the literature on academic spinoffs (Bourelos et al., 2012), yet no evidence has been found of any significant contribution to their survival (Hypothesis 4). The findings may appear somewhat contradictory, given business incubators’ purpose of shielding firms from failure (Grimaldi and Grandi, 2005). However, on the account of recent evidence on the UK spinoffs indicating that they are typically not based in business incubators (Hewitt-Dundas, 2015), the results appear rather unsurprising.

The significance of spinoff’s DCC (Hypothesis 5) lends evidence to the well-advanced debate in the literature on the importance of networks (Rasmussen et al., 2015). The network position a spinoff occupies, defined by dynamic and evolutionary network processes (Hayter, 2016b; Rasmussen et al., 2015), is crucial to its activities as a network actor (Burt, 1992). It is important for spinoffs to learn to recognise their network positions and transform this knowledge into a practical approach, where they can be more calculative and strategic (Huggins, 2010) in using their contacts to pursue competitive advantage. This clearly produces observable results of reduced probability of failure for spinoff companies.

Regional industrial structure at a spinoff’s birth suggests that firms born into less industrially diverse regions have an increased chance of survival (Hypothesis 6a). However, evidence of the industrial specialisation could not be observed (Hypothesis 6b), confirming similar findings reported in Huggins and Thompson (2016) or Tavassoli and Jlienwatcharamongkhol (2016). Regions that have lower variety of knowledge are fertile grounds for highly innovative new firms, which may benefit from Marshallian externalities in terms of better access to labour or non-traded inputs (e.g. local supply chains for healthcare spinoffs in London (Lawton Smith et al., 2014)).

Conclusion

This article examines university networks and regional industrial structure in relation to spinoff survival. It regards the spinoff as a firm embedded in a multi-actor network, where connections have deterministic functions (Huggins, 2010). The analysis reports five core findings. First, the number of investors is positively associated with spinoff survival, suggesting the importance of credibility signalling (Vohora et al., 2004) and risk-reduction strategies exercised by investors (Clarysse et al., 2005). Second, the presence of an experienced external entrepreneur in the management team is critical to a spinoff’s survival, indicating the importance of the acquisition of a set of entrepreneurial capabilities (Wennberg et al., 2011) and company rather than its technology building orientation. Third, TTOs are involved in the later stages of spinoffs revitalising the debate about the competencies of the TTOs from merely early stage administration (Shane, 2004). Fourth, the network position of spinoffs is critical to their survival, suggesting the need for dynamism and strategy in firm’s network development (Huggins, 2010; Rasmussen et al., 2015). Fifth, the regional industrial structure is essential to understanding spinoff survival (Tavassoli and Jlienwatcharamongkhol, 2016), indicating a spatial asymmetry of successful development opportunities for spinoffs.

From a broader perspective, the results indicate that not only are university networks important to spinoff survival, but a particular composition of such networks is critical. Spinoff companies rely on network connections for access to particular resources (Hayter, 2016a). Yet, key roles in such networks are played by only three actors: investors, external entrepreneurs and TTOs. It is noteworthy that the university network configuration critical to spinoff company survival might be different than one calibrated for spinoff formation.

It is clear that the network paradigm is important to understanding firm survival in the industrial organisation literature. Specifically, works on network evolution (Hite and Hesterly, 2001) and network capital (Huggins, 2010) address the issues related to the role of networks in firm survival, where successful firms are more instrumental and calculative in how they configure their ties to access particular resources. Furthermore, it transpires that the survival of the firm is not just a function of the accumulation of generic social capital (Pennings et al., 1998), but rather specific types of resources based on identified growth needs, whether finance, managerial skills or other forms of capital.

Spinoff survival cannot be separated from the considerations of its regional economic environment. Given that the greater survival prospects are in less industrially diversified regions, potentially pertaining to positive externalities of specialisation, policymakers need to reconsider overcoming spatial asymmetries of outcomes. In particular, national policymakers should contemplate designing a redistributive fund for universities or spinoff companies, based in more industrially diversified regions like West Midlands or Scotland, to assist in alleviating issues related to industrial structure. At a regional level, there appears a scope for greater collaboration between universities to pool resources. Finally, at the local level, a concerted effort by universities, investors and entrepreneurs is required to develop strong networks to support spinoffs, as improved network structure could overcome the complex economic context.

The analysis presented here is not free from imperfections. First, the modelling has a cross-sectional character, as there was no possibility of collecting annual records of the data underlying the external entrepreneurs and the DCC constructs. This leads to another issue, where the evolutionary development of a spinoff company’s network position could not be captured. Second, using the number of investors may be quite problematic as it does not portray the value of the investments received. However, such data are difficult to obtain, limiting the depth of the analysis. Finally, shareholder networks are imperfect as not every actor in spinoff networks may hold an equity stake in the firm. It is acknowledged here that this limits the types of actors captured in the analysis, given a breadth of actor types engaged in spinoff success (Hayter, 2016a).

Future studies should focus on longer time frames, given that the growth of spinoff companies decelerates after around three decades (Vincett, 2010), indicating that this article only captures short- to mid-term development. It is also important to recognise the heterogeneity of spinoff types. While this study attempted to control for different types of firms, future research could specifically highlight diverse reasons, motivations and development paths of dormant, lifestyle and growth-oriented spinoffs. While a differential regional industrial structure was investigated here, the spatial configurations of university networks are unknown. These configurations require further elucidation in terms of university network setups at spinoff company formation and survival.

Footnotes

Appendix

Appendix 4.

Classification of firm sectors based on 2007 Standard Industrial Classification.

Sector categories	Alphabetical SIC sectors	Numerical SIC sectors
Manufacturing	Manufacturing	10821, 10822, 13910, 13990, 14120, 20140, 20411, 20420, 20590, 20600, 21100, 21200, 23110, 23440, 23690, 23990, 24450, 25110, 25210, 25500, 25730, 25990, 26110, 26200, 26301, 26309, 26511, 26512, 26513, 26600, 26701, 26702, 26800, 27110, 27120, 27310, 27320, 27900, 28110, 28132, 28250, 28910, 28960, 28990, 29201, 32200, 32300, 32500, 32990, 33140, 33200
Information and communication	Information and Communication	58110, 58210, 58290, 59111, 61200, 61900, 62011, 62012, 62020, 62030, 62090, 63110, 63120, 63990
Professional, scientific and technical	Professional, Scientific and Technical	70100, 70229, 71111, 71121, 71122, 71129, 71200, 72110, 72190, 72200, 73120, 73200, 74100, 74209, 74300, 74901, 74909
Human health and social work	Human Health and Social Work	86102, 86210, 86220, 86230, 86900
Administrative and support and other	Administrative and Support Agriculture, Forestry and Fishing Mining and Quarrying Electricity, Gas, Steam and Air Conditioning Supply Water Supply Sewerage, Waste Management and Remediation Activities Construction Wholesale and Retail Trade Repair of Motor Vehicles and Motorcycles Transportation and Storage Accommodation and Food Service Activities Financial and Insurance Activities Public Administration and Defence Compulsory Social Security Education Arts, Entertainment and Recreation Other Service Activities	77310, 81300, 82990 1110, 1610, 2400, 3210, 9100, 35110, 36000, 37000, 38210, 38320, 39000, 41100, 43210, 43999, 46170, 46180, 46440, 47190, 47749, 47789, 47890, 47910, 49319, 52290, 55900, 26702, 26800, 27110, 27120, 27310, 27320, 27900, 28110, 28132, 28250, 28910, 28960, 28990, 29201, 32200, 32300, 32500, 32990, 33140, 33200, 64999, 66190, 66210, 84110, 84120, 85310, 85320, 85421, 85590, 85600, 90020, 90030, 93199, 95110, 96090

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research supporting this paper was funded by the Economic and Social Research Council, Ref: 1231191.

ORCID iD

Daniel Prokop

Author biographies

Daniel Prokop is a Lecturer in Economic Geography at the School of Geography and Planning, Cardiff University, UK. His research interests are in the area of regional economic development, in particular around entrepreneurship, innovation and networks.

Robert Huggins is a Professor of Economic Geography and Director of Centre for Economic Geography at the School of Geography and Planning, Cardiff University, UK. His areas of interest highlight regional economic development, in particular the study of clusters and competitiveness, knowledge flows, entrepreneurship, innovation and inter-organisational networks.

Gillian Bristow is a Professor of Economic Geography at the School Geography and Planning and Dean of Research for the College of Arts, Humanities and Social Sciences, Cardiff University, UK. Her research interests are in the areas of regional economic development, regional economic resilience, local and regional competitiveness, and regional policy.

References

Alchian

(1950) Uncertainty, evolution, and economic theory. Journal of Political Economy 58(3): 211–221.

Aldridge

Audretsch

(2011) The Bayh-Dole act and scientist entrepreneurship. Research Policy 40(8): 1058–1067.

Barney

(1991) Firm resources and sustained competitive advantage. Journal of Management 17(1): 99–120.

Becker

(1965) A theory of the allocation of time. The Economic Journal 75(299): 493–517.

Benneworth

Charles

(2005) University spin-off policies and economic development in less successful regions: Learning from two decades of policy practice. European Planning Studies 13(4): 537–557.

Berggren

Dahlstrand

ÅL

(2009) Creating an entrepreneurial region: Two waves of academic spin-offs from Halmstad University. European Planning Studies 17(8): 1171–1189.

Bigdeli

Shi

(2015) Sustainability and scalability of university spinouts: A business model perspective. R&D Management 46(3): 504–518.

Bolzani

Fini

Grimaldi

et al . (2014) Fifteen years of academic entrepreneurship in Italy: Evidence from the TASTE Project. Technical Report, University of Bologna, Italy, 1 July.

Bonardo

Paleari

Vismara

(2011) Valuing university-based firms: The effects of academic affiliation on IPO performance. Entrepreneurship Theory and Practice 35(4): 755–776.

10.

Borgatti

Foster

(2003) The network paradigm in organizational research: A review and typology. Journal of Management 29(6): 991–1013.

11.

Borgatti

Everett

Shirey

(1990) LS sets, lambda sets and other cohesive subsets. Social Networks 12: 337–357.

12.

Borghesi

Houston

Naranjo

(2007) Value, survival, and the evolution of firm organizational structure. Financial Management 36(3): 5–31.

13.

Bosma

van Praag

Thurik

et al . (2004) The value of human and social capital investments for the business performance of startups. Small Business Economics 23: 227–236.

14.

Bourelos

Magnusson

McKelvey

(2012) Investigating the complexity facing academic entrepreneurs in science and engineering: The complementarities of research performance, networks and support structures in commercialisation. Cambridge Journal of Economics 36(3): 751–780.

15.

Bower

(2003) Business model fashion and the academic spinout firm. R&D Management 33(2): 97–106.

16.

Brander

Amit

Antweiler

(2002) Venture-capital syndication: Improved venture selection vs. the value-added hypothesis. Journal of Economics & Management Strategy 11(3): 423–452.

17.

Bristow

(2010) Resilient regions: Re-‘place’ing regional competitiveness. Cambridge Journal of Regions, Economy and Society 3: 153–167.

18.

Bruneel

Ratinho

Clarysse

et al . (2012) The evolution of business incubators: Comparing demand and supply of business incubation services across different incubator generations. Technovation 32(2): 110–121.

19.

Burt

(1992) Structural Holes – The Social Structure of Competition. London: Harvard University Press.

20.

BVCA/Library House (2005) Creating Success from University Spin-outs. London: British Venture Capital Association.

21.

Capaldo

(2007) Network structure and innovation: The leveraging of a dual network as a distinctive relational capability. Strategic Management Journal 28: 585–608.

22.

Carayannis

von Zedtwitz

(2005) Architecting gloCal (global–local), real-virtual incubator networks (G-RVINs) as catalysts and accelerators of entrepreneurship in transitioning and developing economies: Lessons learned and best practices from current development and business incubation practices. Technovation 25(2): 95–110.

23.

Civera

Meoli

Vismara

(2019) Do academic spinoffs internationalize? Journal of Technology Transfer 44(2): 381–403.

24.

Clarysse

Tartari

Salter

(2011a) The impact of entrepreneurial capacity, experience and organizational support on academic entrepreneurship. Research Policy 40(8): 1084–1093.

25.

Clarysse

Wright

Van de Velde

(2011b) Entrepreneurial origin, technological knowledge, and the growth of spin-off companies. Journal of Management Studies 48(6): 1420–1442.

26.

Clarysse

Wright

Lockett

et al . (2005) Spinning out new ventures: A typology of incubation strategies from European research institutions. Journal of Business Venturing 20(2): 183–216.

27.

Coad

(2017) Firm age: A survey. Journal of Evolutionary Economics 28(1): 13–43.

28.

Cohen

Levinthal

(1990) Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly 35(1): 128–152.

29.

Coleman

(1990) Foundations of Social Theory. Cambridge, MA: Harvard University Press.

30.

Cressy

(1996) Are business startups debt-rationed? The Economic Journal 106(438): 1253–1270.

31.

Criaco

Minola

Migliorini

et al . (2014) ‘To have and have not’: Founders’ human capital and university start-up survival. Journal of Technology Transfer 39: 567–593.

32.

Dahl

Sorenson

(2012) Home sweet home: Entrepreneurs’ location choices and the performance of their ventures. Management Science 58(6): 1059–1071.

33.

De Nicola

Prencipe

Corsi

(2018) The effect of university patent activity on the performance of university spin-offs – A path dependency perspective. International Journal of Business and Social Science 9(2): 91–98.

34.

Degroof

J-J

Roberts

(2004) Overcoming weak entrepreneurial infrastructures for academic spin-off ventures. Journal of Technology Transfer 29(3–4): 327–352.

35.

Diánez-González

Camelo-Ordaz

(2016) How management team composition affects academic spin-offs’ entrepreneurial orientation: The mediating role of conflict. Journal of Technology Transfer 41: 530–557.

36.

Fairlie

Lofstrom

(2015) Immigration and entrepreneurship. In: Chiswick

Miller

(eds) Handbook of the Economics of International Migration, Volume 1B. Oxford: North-Holland Publishing Company, pp.877–911.

37.

Ferretti

Ferri

Fiorentino

et al . (2019) Neither absent nor too present: The effects of the engagement of parent universities on the performance of academic spin-offs. Small Business Economics 52(1): 153–173.

38.

Fini

Mathisen

et al . (2017) Institutional determinants of university spin-off quantity and quality: A longitudinal, multilevel, cross-country study. Small Business Economics 48(2): 361–391.

39.

Fini

Grimaldi

Santoni

et al . (2011) Complements or substitutes? The role of universities and local context in supporting the creation of academic spin-offs. Research Policy 40(8): 1113–1127.

40.

Fini

Lacetera

Shane

(2010) Inside or outside the IP system? Business creation in academia. Research Policy 39(8): 1060–1069.

41.

Fitzgerald

Cunningham

(2016) Inside the university technology transfer office: Mission statement analysis. Journal of Technology Transfer 41: 1235–1246.

42.

Fotopoulos

(2014) On the spatial stickiness of UK new firm formation rates. Journal of Economic Geography 14: 651–679.

43.

Franklin

Wright

Lockett

(2001) Academic and surrogate entrepreneurs in university spin-out companies. Journal of Technology Transfer 26: 127–141.

44.

Freeman

(1978) Centrality in social networks – Conceptual clarification. Social Networks 1: 215–239.

45.

Garnsey

Heffernan

(2005) High-technology clustering through spin-out and attraction: The Cambridge case. Regional Studies 39(8): 1127–1144.

46.

Geroski

(1995) What do we know about entry? International Journal of Industrial Organization 13: 421–440.

47.

Gimmon

Levie

(2010) Founder’s human capital, external investment, and the survival of new high-technology ventures. Research Policy 39(9): 1214–1226.

48.

Goethner

Obschonka

Silbereisen

et al . (2012) Scientists’ transition to academic entrepreneurship: Economic and psychological determinants. Journal of Economic Psychology 33(3): 628–641.

49.

Gonzalez-Pernia

Kuechle

Peña-Legazkue

(2013) An assessment of the determinants of university technology transfer. Economic Development Quarterly 27(1): 6–17.

50.

Grandi

Grimaldi

(2003) Exploring the networking characteristics of new venture founding teams. Small Business Economics 21(4): 329–341.

51.

Grimaldi

Grandi

(2005) Business incubators and new venture creation: An assessment of incubating models. Technovation 25(2): 111–121.

52.

Gubitta

Tognazzo

Destro

(2016) Signalling in academic ventures: The role of technology transfer offices and university funds. Journal of Technology Transfer 41: 368–393.

53.

Gulati

(1999) Network location and learning: The influence of network resources and firm capabilities on alliance formation. Strategic Management Journal 20(5): 397–420.

54.

Hannan

Freeman

(1984) Structural inertia and organizational change. American Sociological Review 49(2): 149–164.

55.

Harrison

Leitch

(2010) Voodoo institution or entrepreneurial university? Spin-off companies, the entrepreneurial system and regional development in the UK. Regional Studies 44(9): 1241–1262.

56.

Hayter

(2013a) Conceptualizing knowledge-based entrepreneurship networks: Perspectives from the literature. Small Business Economics 41: 899–911.

57.

Hayter

(2013b) Harnessing university entrepreneurship for economic growth: Factors of success among university spin-offs. Economic Development Quarterly 27(1): 18–28.

58.

Hayter

(2016a) A trajectory of early-stage spinoff success: The role of knowledge intermediaries within an entrepreneurial university ecosystem. Small Business Economics 47: 633–656.

59.

Hayter

(2016b) Constraining entrepreneurial development: A knowledge-based view of social networks among academic entrepreneurs. Research Policy 45: 475–490.

60.

Hewitt-Dundas

(2015) Profiling UK university spin-outs. ERC Research Paper No.35, Enterprise Research Centre, Birmingham, July.

61.

Hite

Hesterly

(2001) The evolution of firm networks: From emergence to early growth of the firm. Strategic Management Journal 22(3): 275–286.

62.

Hochberg

Ljungqvist

(2007) Whom you know matters: Venture capital networks and investment performance. The Journal of Finance 62(1): 251–302.

63.

Holmes

Hunt

Stone

(2010) An analysis of new firm survival using a hazard function. Applied Economics 42(2): 185–195.

64.

Hosmer

Jr Lemeshow

Sturdivant

(2013) Applied Logistic Regression (3rd edn). Hoboken, NJ: Wiley.

65.

Howell

(2017) Marshallian source of relatedness and their effects on firm survival and subsequent success in China. Economic Geography 93(4): 346–366.

66.

Hudson

(1987) The age, regional, and industrial structure of company liquidations. Journal of Business Finance & Accounting 14(2): 199–213.

67.

Huggins

(2008) Universities and knowledge-based venturing: Finance, management and networks in London. Entrepreneurship and Regional Development 20(2): 185–206.

68.

Huggins

(2010) Forms of network resource: Knowledge access and the role of inter-firm networks. International Journal of Management Reviews 12(3): 335–352.

69.

Huggins

Prokop

(2016) Network structure and regional innovation: A study of university-industry ties. Urban Studies 54(4): 931–952.

70.

Huggins

Thompson

(2016) Socio-spatial culture and entrepreneurship: Some theoretical and empirical observations. Economic Geography 92(3): 269–300.

71.

Huggins

Prokop

Thompson

(2017) Entrepreneurship and the determinants of firm survival within regions: Human capital, growth motivation and locational conditions. Entrepreneurship and Regional Development 29(3–4): 357–389.

72.

Iacobucci

Micozzi

(2015) How to evaluate the impact of academic spin-offs on local development: An empirical analysis of the Italian case. Journal of Technology Transfer 40: 434–452.

73.

Jelfs

(2016) Financial performance analysis of spin-off companies from a UK ‘regional’ university: A case study of the University of Birmingham. International Journal of Entrepreneurship and Small Business 29(2): 271–286.

74.

Jensen

Webster

Buddelmeyer

(2008) Innovation, technological conditions and new firm survival. Economic Record 84(267): 434–448.

75.

Jovanovic

(1982) Selection and the evolution of industry. Econometrica 50(3): 649–670.

76.

Klepper

Simons

(2000) The making of an oligopoly: Firm survival and technological change in the evolution of the U.S. Tire industry. Journal of Political Economy 108(4): 728–760.

77.

Kogut

(2000) The network as knowledge: Generative rules and the emergence of structure. Strategic Management Journal 21(3): 405–425.

78.

Kogut

Zander

(1992) Knowledge of the firm, combinative capabilities, and the replication of technology. Organization Science 3(3): 383–397.

79.

Krabel

Mueller

(2009) What drives scientists to start their own company? Research Policy 38(6): 947–956.

80.

Lambert Review (2003) Lambert Review of Business-University Collaboration. Norwich: HMSO.

81.

Lawton Smith

Chapman

Wood

et al . (2014) Entrepreneurial academics and regional innovation systems: The case of spin-offs from London’s universities. Environment and Planning C: Government and Policy 32: 341–359.

82.

Lawton Smith

Etzkowitz

Meschitti

et al . (2017) Female academic entrepreneurship and commercialisation: Reviewing the evidence and identifying the challenges. In: Henry

Nelson

Lewis

(eds) The Routledge Companion to Global Female Entrepreneurship. London: Routledge, pp.78–92.

83.

Lawton Smith

Romeo

Bagchi-Sen

(2008) Oxfordshire biomedical university spin-offs: An evolving system. Cambridge Journal of Regions, Economy and Society 1(2): 303–319.

84.

Leitch

Harrison

(2005) Maximising the potential of university spin-outs: The development of second-order commercialisation activities. R&D Management 35(3): 257–272.

85.

Ren

Feng

et al . (2016) Network of listed companies based on common shareholders and the prediction of market volatility. Physica A: Statistical Mechanics and Its Applications 462: 508–521.

86.

Lin

(2001) Social Capital: A Theory of Social Structure and Action. Cambridge, MA: Cambridge University Press.

87.

Littunen

(2000) Networks and local environmental characteristics in the survival of new firms. Small Business Economics 15: 59–71.

88.

Lockett

Wright

(2005) Resources, capabilities, risk capital and the creation of university spin-out companies. Research Policy 34(7): 1043–1057.

89.

Lockett

Wright

Franklin

(2003) Technology transfer and universities’ spin-out strategies. Small Business Economics 20: 185–200.

90.

Lundqvist

(2014) The importance of surrogate entrepreneurship for incubated Swedish technology ventures. Technovation 34: 93–100.

91.

Mameli

Iammarino

Boschma

(2012) Regional variety and employment growth in Italian labour market areas: Services versus manufacturing industries. CIMR Research Working Paper Series, Working Paper No. 4, March 2012, pp.1–23. London: University of London.

92.

Mason

Stark

(2004) What do investors look for in a business plan? International Small Business Journal 22(3): 227–248.

93.

Mata

Portugal

(2002) The survival of new domestic and foreign-owned firms. Strategic Management Journal 23(4): 323–343.

94.

M’Chirgui

Lamine

Mian

et al . (2018) University technology commercialization through new venture projects: An assessment of the French regional incubator program. Journal of Technology Transfer 43(5): 1142–1160.

95.

Miranda

Chamorro

Rubio

(2018) Re-thinking university spin-off: A critical literature review and a research agenda. Journal of Technology Transfer 43: 1007–1038.

96.

Mosey

Wright

(2007) From human capital to social capital: A longitudinal study of technology-based academic entrepreneurs. Entrepreneurship Theory and Practice 31(6): 909–935.

97.

Munari

Sobrero

Toschi

(2018) The university as a venture capitalist? Gap funding instruments for technology transfer. Technological Forecasting & Social Change 127: 70–84.

98.

Murray

(1988) An intertemporal model of industrial exit. Quarterly Journal of Economics 103(2): 333–344.

99.

Mustar

(1997) How French academics create hi-tech companies: The conditions for success or failure. Science and Public Policy 24(1): 37–43.

100.

Mustar

Wright

Clarysse

(2008) University spin-off firms: Lessons from ten years of experience in Europe. Science and Public Policy 35(2): 67–80.

101.

Nerkar

Shane

(2003) When do start-ups that exploit patented academic knowledge survive? International Journal of Industrial Organization 21: 1391–1410.

102.

O’Donoghue

(2016) Exploring the links between employment clusters and economic diversity in the British urban system. Modern Economy 7: 743–757.

103.

Office for National Statistics (ONS) (2009) UK Standard Industrial Classification of Economic Activities 2007 (SIC 2007). Basingstoke: Palgrave Macmillan.

104.

Pennings

Lee

van Witteloostuijn

(1998) Human capital, social capital, and firm dissolution. Academy of Management Journal 41(4): 425–440.

105.

Rasmussen

Mosey

Wright

(2015) The transformation of network ties to develop entrepreneurial competencies for university spin-offs. Entrepreneurship and Regional Development 27(7–8): 430–457.

106.

Rodríguez-Gulías

Fernández-López

Rodeiro-Pazos

et al . (2018) The role of knowledge spillovers on the university spin-offs innovation. Science and Public Policy 45(6): 875–883.

107.

Rodríguez-Gulías

Rodeiro-Pazos

Fernández-López

(2017) The effect of university and regional knowledge spillovers on firm’s performance: An analysis of the Spanish USOs. International Entrepreneurship Management Journal 13: 191–209.

108.

Scholten

Omta

Kemp

et al . (2015) Bridging ties and the role of research and start-up experience on the early growth of Dutch academic spin-offs. Technovation 45–46: 40–51.

109.

Shane

(2004) Academic Entrepreneurship. University Spinoffs and Wealth Creation. Cheltenham: Edward Elgar.

110.

Sidney

Winter

(1964) Economic ‘natural selection’ and the theory of the firm. Yale Economic Essays 4(1): 225–272.

111.

Singh

Tucker

House

(1986) Organizational legitimacy and the liability of newness. Administrative Science Quarterly 31(2): 171–193.

112.

Singh

Mitchell

(1996) Precarious collaboration: Business survival after partners shut down or form new partnerships. Strategic Management Journal 17: 99–115.

113.

Tavassoli

Jlienwatcharamongkhol

(2016) Survival of entrepreneurial firms: The role of agglomeration externalities. Entrepreneurship and Regional Development 28(9–10): 746–767.

114.

Taylor

(1999) Survival of the fittest? An analysis of self-employment duration in Britain. Economic Journal 109(454): C140–C155.

115.

Teece

Pisano

Shuen

(1997) Dynamic capabilities and strategic management. Strategic Management Journal 18(7): 509–533.

116.

Theil

(1972) Statistical Decomposition Analysis: With Applications in the Social and Administrative Sciences. London: North-Holland Publishing Company.

117.

Van Geenhuizen

Soetanto

(2009) Academic spin-offs at different ages: A case study in search of key obstacles to growth. Technovation 29(10): 671–681.

118.

Van Witteloostuijn

(1998) Bridging behavioural and economic theories of decline: Organizational inertia, strategic competition, and chronic failure. Management Science 44(4): 501–519.

119.

Vincett

(2010) The economic impacts of academic spin-off companies, and their implications for public policy. Research Policy 39: 736–747.

120.

Visintin

Pittino

(2014) Founding team composition and early performance of university-based spin-off companies. Technovation 34: 31–43.

121.

Vohora

Wright

Lockett

(2004) Critical junctures in the development of university high-tech spinout companies. Research Policy 33(1): 147–175.

122.

Walter

Auer

Ritter

(2006) The impact of network capabilities and entrepreneurial orientation on university spin-off performance. Journal of Business Venturing 21(4): 541–567.

123.

Wennberg

Wiklund

Wright

(2011) The effectiveness of university knowledge spillovers: Performance differences between university spinoffs and corporate spinoffs. Research Policy 40(8): 1128–1143.

124.

Wernerfelt

(1984) A resource-based view of the firm. Strategic Management Journal 5(2): 171–180.

125.

Wilson

Wright

Altanlar

(2013) The survival of newly-incorporated companies and founding director characteristics. International Small Business Journal 32(7): 733–758.

126.

Wright

Lockett

Clarysse

et al . (2006) University spin-out companies and venture capital. Research Policy 35(4): 481–501.

127.

Wynarczyk

Raine

(2005) The performance of business incubators and their potential development in the North East region of England. Local Economy 20(2): 205–220.

128.

Zaheer

Soda

(2009) Network evolution: The origins of structural holes. Administrative Science Quarterly 54: 1–31.

129.

Zhang

(2009) The performance of university spin-offs: An exploratory analysis using venture capital data. Journal of Technology Transfer 34: 255–285.

130.

Zhao

Burt

(2018) A note on business survival and social network. Management and Organization Review 14(2): 377–394.