Analysis of the survival determinants of University Spin-Offs in a crossnational sample

Analysis of the survival determinants of University Spin-Offs in a crossnational sample David Rodeiro-Pazos Department of Finance and Accounting Universidade de Santiago de Compostela Avda. do Burgo, s/n., 15782 Santiago de Compostela, Galicia, Spain. Email: david.r[email protected] Sara Fernández-López Department of Finance and Accounting Universidade de Santiago de Compostela Avda. do Burgo, s/n., 15782 Santiago de Compostela, Galicia, Spain. Email: [email protected] María Jesús Rodríguez-Gulías Department of Business Universidade da Coruña Campus Elviña, 15701, A Coruña, Galicia, Spain. Email: mar[email protected] Christian Corsi Università degli Studi di Teramo, Via R. Balzarini 1, 64100 Teramo, Italy E-mail: c[email protected] Antonio Prencipe Università degli Studi di Teramo, Via R. Balzarini 1, 64100 Teramo, Italy E-mail: a[email protected] 1 Title: Analysis of the survival determinants of University Spin-Offs in a crossnational sample Abstract: An understanding of the dynamics related to university spin-off’s survival is a critical point for many scholars and policy makers, especially in the technology transfer process. Drawing on the population and organizational ecology perspectives and the resource-based view of the firm, this paper aims to investigate the survival determinants of University Spin-Offs (USOs). After applying event analysis techniques in a crossnational dataset of 1,275 Italian and Spanish USOs over the period 2005-2013, the findings show that firm size, firm age, industrial partners, and firm efficiency positively influence USOs’ survival, whereas patent activity decreases USOs’ chances to survive. Additionally, a U-shaped relationship between financial leverage and USOs’ survival emerges. Based on the emerging evidence, some policy and managerial implications are provided in order to help USOs to cross their “Valley of Death”. Keywords: university spin-off; firm survival; event analysis, cross-national, resourcebased theory. 1 Introduction Translating knowledge and technologies into marketable goods and services by converting discoveries and ideas into successful innovations represents a common and increasing dilemma in several countries (Barr et al., 2009; Good et al., 2019); this is significantly recognized in Europe (European Commission, 2017). In this regard, the knowledge-based economy has considerably underlined the role of universities in commercializing the knowledge produced in their research activities (Hayter and Link, 2018; Wu et al., 2017) through the creation of academic start-ups or University Spin-Offs 2 (USOs), among other ways (Rasmussen et al., 2014; Fini et al., 2017; Rodríguez-Gulías et al., 2018). Indeed, USOs play a critical knowledge bridging function by placing otherwise unexploited technologies and know-how into industry in order to create socioeconomic value (Zahra et al., 2007; Miranda et al., 2018). Nevertheless, not all of the USOs reach this goal and a significant part of them fail. The missing linkage in shifting from a new technology to setting up a market-driven and profitable business has been indicated by the literature as the “Valley of Death” (Barr et al., 2009; Markham et al., 2010; Midler, 2019). Concerning USOs, projects and ventures originating from university research that have potential to produce marketable products and services fail in commercial application because they have not yet been developed enough to be profitable (Abereijo, 2015). Consequently, the effective technology transfer from university to industry is prevented, negatively affecting the USOs’ growth and survival (Clayman and Holbrook, 2003; Lundqvist, 2014; Ries, 2011; Hirsch and Tambuyzer, 2015;). The nature of USOs as mainly knowledge-based firms (Miranda et al., 2018) leverages the common problems linked to the launch of a new venture with the difficulty of innovation development (Berbegal-Mirabent et al., 2015). As a result, USOs are very sensitive to several types of market failures, mainly through the start-up phase (Hall and Rosenberg, 2010; Parmentola and Ferretti, 2018), which threaten their survival. In detail, the embryonic nature of USOs’ knowledge, together with the extensive time lag between the research phase, product development and its launch to the market (Rasmussen and Rice, 2011; Wu et al., 2017), create significant risks and uncertainties in commercialization. This prevents investors from financing emerging early-stage innovation, putting a substantial brake on further development and the survival prospects of the USOs. 3 Additionally, USOs are typically resource-constrained and not familiar with forecasting the resources needed to overcome the liabilities of both newness and smallness (Skute, 2019; Novotny, 2020). Clearly, USOs show several limitations in attaining the finance needed to support their growth and also their survival (Sørheim et al., 2011; Galati et al., 2017). In addition, university entrepreneurs general show limited managerial capabilities (Oliveira et al., 2013; Lundqvist, 2014) as well as little prior expertise in industry (Rasmussen and Borch, 2010; Drivas et al., 2018). This emerging setting calls for a better understanding of the survival determinants of USOs. Up to now, most of the literature on the field has focused on USOs’ growth and creation determinants (Hesse and Sternberg, 2017; Ramaciotti and Rizzo, 2015; Bock et al., 2018). Although this approach is useful to understand the success of university entrepreneurship, the study of factors affecting the extent to which USOs survive is still limited (Wennberg et al., 2011) and fragmented (Gilsing et al., 2010; Miranda et al., 2018; Prokop et al. 2019). This study aims to fill these gaps. Mainly drawing on population and organizational ecology perspectives and the resource-based view of the firm, this paper analyses the firm-level determinants of USOs’ survival in a cross-national sample of 1,275 Italian and Spanish USOs over the period 2005-2013. Both Mediterranean European countries have intensively developed policies geared towards boosting the creation of these kinds of firms in the last two decades (Yagüe and March, 2012; Ramaciotti and Rizzo, 2015; Miranda et al., 2017). In spite of this, the USOs phenomenon has not been fully explored, and a widespread study about the effectiveness of universities at creating surviving firms is needed. This paper contributes to the literature on the effectiveness of university technology transfer trough USOs, which it is still rather limited (Kroll and Liefner, 2008). In 4 particular, the paper provides new insights about the success determinants of USOs, with particular focus on firm survival. By combining theoretical arguments from the resourcebased view with those from population ecology perspectives, it provides a better understanding about the mechanisms generating differential survival expectations in USOs. Further, unlike previous studies in the topic, the use of a larger dataset englobing firms from two national contexts over an eight-year period, as well as a new set of financial and innovation variables, spreads the understanding about USOs´ survival and potentially improves the reliability and generalizability of the emerging findings. This paper is organized as follows: Section 2 introduces the literature background and presents the research hypotheses. In Section 3, the methodology is described. Section 4 presents the findings of the study. In Section 5 the main results are discussed, while Section 6 provides some managerial and policy implications. Section 7 presents the main limitations and conclusions of the study. 2 The determinants of USOs’ survival and research hypotheses development The elements affecting firm survival, especially new entrants, have been widely investigated in the literature related to Industrial Organization (see Josefy et al. (2017) for a recent review). This trend has not similarly pervaded studies focusing on the specific case of USOs. Besides some general descriptive research into the survival rates of USOs (e.g., Lowe 2002), little evidence is available on the determinants of USOs’ survival. Shane and Stuart (2002) and Nerkar and Shane (2003) examine respectively how organizational endowments (i.e., social capital) and patents affect the probability of survival of new technology ventures originating at the Massachusetts Institute of Technology. Conceiçao and Faria (2014) place the emphasis on a few firm-level characteristics (i.e., size, age, and parent university reputation) and on regional factors. Rodríguez-Gulías et al. (2016) pay attention to the effect of USOs’ financial performance 5 on firm survival, whereas Prokop et al. (2019) focus on the role played by the “externals” (i.e., investors, entrepreneurs….). Finally, Fernández-López et al. (2020) analyse how the interaction between innovation and export activities affects a USO’s survival. Despite these studies, there is no widely accepted theoretical approach in the analysis of USOs’ survival. In contrast, literature on USOs has shown that the resource-based view of the firm provides a critical framework in the study of spin-out process effectiveness by universities (Vinig and Van Rijsbergen, 2010; Rodeiro-Pazos et al., 2012; Berbegal-Mirabent et al., 2015). Stemming from this approach, the failure risk may be lower for firms with a systematic and well-established set of resources and capabilities which encompass the initial and critical start-up phase and help buffer against hard luck during the rest of their entrepreneurial life (Bruderl and Schussler, 1990). Also, from the organizational and population ecology perspectives, firm survival is a function of the amount and type of resources a firm controls (Chung and Cheng, 2019; Hodgson et al., 2017). Concerning the study of USOs’ survival, both theoretical perspectives have been observed to support the understanding of the evolution of the university technology transfer to industry (Cardozo et al., 2011). Hence, the theoretical framework used in this study to explain USOs’ survival is based on the arguments related to both the resource-based view of the firm and the population and organizational ecology perspectives. In detail, the focus of the analysis is on nine firm-level characteristics widely researched as survival drivers in the literature on firm survival, but these are frequently overlooked by the scarce studies on USOs’ survival. In so doing, the specific nature of USOs and how these characteristics may differently affect firm survival has been taken into account. These firm-level characteristics are (1) firm 6 size, (2) firm age, (3) patent activity, (4) high-technology sector, (5) venture capital, (6) industrial partnership, (7) firm profitability, (8) firm efficiency, and (9) financial leverage. Previous studies about the survival of innovative firms claim the relevance of the firm size and its impact on the organizational outcomes. This is an issue related to the "liability of smallness" from the population ecology perspective (Audretsch and Mahmood, 1991; Aldrich and Auster, 1986). In this regard, scholars remark on different logic behind the superior survival rates for larger firms as compared to their smaller counterparts (Agarwal and Audretsch, 2001; Mas-Verdú et al., 2015; Esteve-Pérez and Mañez-Castillejo, 2008), which may not be able to gain the adequate critical resources required to survive (Chang et al., 2008). Larger firms have superior resources to allocate to innovative activities and tacit knowledge transfer activities (Tamer et al., 2003). They are also expected to hire more employees with specialized know-how in science and engineering as well as in management (Laursen and Salter, 2004), which are critical for knowledge-based firms such as USOs. Similarly, literature points out that larger firms have greater availability to achieve a minimum efficiency threshold under which firms typically fail (Fritsch et al., 2006). In this respect, Giovannetti et al. (2011) highlight that in industries characterized by a high level of technological innovation, such as those related to USOs, or that are at an early stage of their life cycle, firm size represents a critical and vital competitive advantage. In brief, the superior survival rates of larger firms are linked with the presence of economies of scale, which reduce the average cost of the firm, together with improved access for both financial markets and human resource capital (Mata et al., 1995; Geroski et al., 2010). Although the topic has been well-recognized in the literature on firm survival, the analysis gains relevance in the USOs’ context since scholars claim that these firms tend to remain 7 small in size for long time with a significant deficit of resources that impedes their growth (Galati et al., 2017). The “size deficit” affecting a USO’s growth may also create obstacles to their survival prospects. Without an adequate resource endowment, both the growth and existence of the firm is compromised (Colombo and Piva, 2008). Thus, Conceiçao and Faria (2004) and Fernández-López et al. (2020) found a positive effect of size on a USO’s survival, while Rodríguez-Gulías et al. (2016) documented an inverted U-shaped relationship between both variables, and non-significant results were reported by Prokop et al. (2019). Drawing on previous arguments, the following is assumed: H1: Firm size is positively related to USOs’ survival probability. Age also has a substantial effect on firms’ survival prospects (Phillips and Kirchhoff, 1989; Sapienza et al., 2006). From the organizational ecologist perspective, newcomers are exposed to the “liability of newness” (Susarla and Barua, 2011). Thus, during the development-stage of a firm, a learning process takes place that allows the business to unfold (e.g., carry out investments, develop specific knowledge and routines, build social networks, etc.) (Mudambi and Zahra, 2007). As time goes by, firms are expected to have more advanced routines and well-known processes than their younger counterparts (Salimath and Jones, 2011). Additionally, firms gain external legitimacy, increasing their likelihood of survival. From the resource-based view, limited resources and know-how can be risky for young firms (Wiklund et al., 2010; Carr et al., 2010), making the commercialization of research knowledge a significantly hazardous activity and increasing the premature mortality of the firm. The above-mentioned arguments are especially relevant to USOs, which usually develop their core technology early during their lifespan (Gübeli and Doloreux, 2005; Zahra et al., 2007). This period of time is fundamental to effectively boost the learning process and overcome the “liability of newness”. In fact, Conceição and Faria (2014) report that the 8 failure rate increases after the first year of activity and considerably grows until the third year, supporting that USOs fail in the early stages of their life. In contrast, the likelihood of failure for the USOs that survive the first 2/3 years of activity falls gradually with the firm’s age. Hence, the following hypothesis is advanced: H2: Firm age is positively related to USOs’ survival probability. According with the resource-based view, patents represent an intangible resource that support firms in acquiring competitive advantages in their industries. The availability of patents signals that an organization has an inimitable set of resources protected from imitators. This “direct” benefit from patenting is especially relevant for small newcomers whose activity is subject to high uncertainty and asymmetric knowledge (Shane, 2002; Geuna and Muscio, 2009). Thus, previous research reports a positive association between patenting and firms’ survival rates (Cockburn and MacGarvie, 2006; Cockburn and Wagner, 2020). Additionally, patenting produces “indirect” effects that may significantly improve the competitive advantage of newcomers. Indeed, literature on NTBFs reveals that patent activity also serves to gain access to external funding and reputation, which in turn positively influence the survival dynamics of the firm (Löfsten, 2016). In spite of the above-mentioned arguments, the links between patent activity and USOs’ survival remain unclear as the only study concerning this issue has found a negative relationship between patenting and firm survival (Fernández-López et al., 2020). Stemming from the theoretical arguments, the following is advanced: H3: Patent activity is positively related to USOs’ survival probability. The firm’s sector also influences the firms’ survival rate (Colombelli et al., 2013; Tavassoli and Jlienwatcharamongkhol, 2016). Particularly, the differences in firm survival between low-, moderate- and high-tech sectors have been analysed from the 15 and March, 2011). Thus, the dependent variable was the survival time of the USOs in years (_t), calculated as the time elapsed between the firm birth date1 and the moment in which the firm failed, nuanced by a dummy variable that indicates whether the event (failure) has taken place (_d)2. The firms’ legal status, available in SABI and Aida databases, are used to define the USOs as failed (event). For Spanish firms, 9 SABI categories are considered to be firm failure: bankruptcy, state of insolvency, extinct, dissolved, closing of the register, provisional closing of the register, inactive, probably inactive, and untraceable according to sources. For Italian ones, 4 Aida categories are considered to be firm failure: closed, closed (in liquidation), closed (closure due to bankruptcy), and in liquidation. The opposite group, the reference group, are active firms which included merged firms (fusion), following Zhang (2009). The survival time is censored to the right on December 2013 because an exit event is not observed for all USOs (i.e., for continuing firms). In order to test the hypotheses, we defined a set of explanatory variables that are widely used in the literature on firm survival (Table 1). In addition, a country dummy variable was included in order to capture institutional factors that may influence a USO’s survival. INSERT Table 1 Definitions of independent variables and predictions HERE 3.3 Estimation and model specification Since firm failure is the result of strategies over time, it should be considered to be a continuous process, even though the appraisal of failure occurs at a certain time (Dimitras 1 For 54 Italian USOs the exact birth date was not available, while their birth year is available. Hence, we decided to approximate their birth date as July 1 for all of them in order to avoid discarding these firms. 2Inthiswork,thedataisintheformofapanel,therefore,each companypresents asetofrecords, specificallyoneforeachperiodoftime(t0,t1].Attheendofeachtimeperiod(t1),theeventvariable(_d), dummyvariablecensored,willtakethevalue1ifthecompanyhasexperiencedafailureand0otherwise. Thus,eachcompanywillshowavectorofzerosthatcanendina1ifthecompanyfinallyfails,or0ifit hasnotfailedinthestudyperiod(censorshipontheright).Inourcase,thefailureeventisexclusive,that is,oncethecompanypresentstheevent,itleavesthesampleandceasestobeobserved(absorbingor culminatingevent). 16 et al., 1996). An appropriate approach to analyse the dynamics of firm failure is the use of event analysis techniques. These techniques consider “time to failure” as an integral factor in firm failure analysis (Kleinbaum and Klein, 2005; Chancharat et al., 2007). Therefore, event models study the effect of a set of covariates on the time elapsed before the event (failure). The choice of this methodology is based on two issues. First, event models are adequate to cope with censured samples, which is the case of our right-censored sample since some firms were still active in 2013. Second, they allow including time-varying covariates which not only overcome the limitation of considering characteristics previous to the time of a firm’s entry in the dataset as the unique determinants of its survival probability (Esteve and Mañez, 2008) but also deal with the deterioration in those variables which involves different effects during the firm’s failure process (Luoma and Laitinen, 1991). Following Conceiçao and Faria (2014), Rodríguez-Gulías et al. (2016), and Fernández- López et al. (2020), we apply the semiparametric Cox proportional hazards model specification (Cox, 1972): ℎ𝑡|𝑥ℎ 󰇛𝑡󰇜 𝑒  where h0(t) is the baseline hazard, xj is a vector of covariates, and βx is the vector of their coefficients. Under this specification, the effect of a unit of change of the explanatory covariates is a constant parallel shift of the baseline function. The statistic based on the Schoenfeld residuals (Schoenfeld, 1982; Grambsch and Therneau, 1994) is used to test this proportionality assumption. Therefore, the following equation gives the basic specification of our model: 17 h (t|xj) = h0 (t) exp (β 1LN_ASSETSij + β 2LN_AGEij + β 3PATij + β 4HIGH_TECHi + β5VENT_CAPi + β 6IND_SHARi + β 7ROAij + β 8TOT_TUR ij +β 9LEV_Rij + β10LEV_RSQUARij + β11SPAINi) There are two main advantages of the Cox proportional hazards model (Wennberg et al., 2011). Firstly, any assumption regarding the duration dependence is required. Secondly, the Cox model allows flexible handling of the non-linear relations and the time-varying covariates. Since no assumption of the baseline hazard function is required, the semiparametric Cox model is less efficient than the correct parametric specification. Nevertheless, misspecification of the hazard rate h0(t) may lead to inconsistent estimates (Cleves et al., 2008). 4 Results 4.1 Non-parametric and descriptive analysis Of the 1,275 USOs analysed, 1,046 survived for the period 2005–2013. Thus, the survival rate is 82.04% (or the failure rate is about 17.96%). By using the Kaplan-Meier estimate procedure, Figure 1 depicts the non-parametric estimate of the survival function. Accordingly, the estimate of the survival functions indicates that 25% of the USOs did not survive after 9.027 years; while the estimated median, the time beyond which 50% of USOs are expected to survive, is 14.81 years. INSERT Figure 1 Kaplan-Meier estimate of the survival function HERE The smoothed hazard rates are shown in Figure 2. The failure hazard of USOs increases at a steady pace with age during the first 12 years of activity, but after then, failure rates increase more rapidly. This shape may reflect that the effect of the parent university’s support received by USOs during their early stages is exhausted as time goes by, 18 suggesting that the failure risk of the USOs is not at its peak during the initial years of activity, as it could be expected. INSERT Figure 2 Smoothed hazard estimates HERE Table 2 shows the descriptive statistics of the explanatory variables. INSERT Table 2 Descriptive statistics of independent variables HERE The average total assets of the sampled USOs are 835 thousand euros. The firms have an average annual age of 5.18 years. The percentage of observations with patent activity is 6.8%, and 57.3% of the USOs operate in the medium- and high-tech industries. The percentage of observations of venture-backed USOs is 9.1% and with industry shareholders is 31.4%. The mean of the returns on assets (ROA) is negative (−3.3%), the average total asset turnover is 0.911, and the mean leverage ratio is about 61.7%. Finally, Table 3 shows the correlation matrix of independent continuous variables. The correlation matrix shows that there are no critical correlations which we have to hold in account. Thus, the firm’s age is positively associated with ROA, total asset turnover, and leverage ratio. Whereas firm profitability (ROA) is negatively correlated to total asset turnover and the firm’s indebtedness (LEV_R). Firm size also shows a positive correlation with firm age and a negative one with total asset turnover. Surprisingly, no significant correlation is found between firm size and indebtedness. In this regard, the literature observes that firms with a high value of assets also have improved borrowing capability, suggesting a positive relationship between firm asset and leverage (Booth et al., 2001; Rajan and Zingales, 1995). This is especially the case when a significant part of such assets are tangible assets and can act as collateral for debt. In contrast, USOs generally exhibit a low asset tangibility due to their technology transfer activity (Greco 19 et al., 2013), which may partly explain the lack of a significant correlation between USOs’ assets and leverage. INSERT Table 3 Correlation matrix HERE 4.2 Semiparametric analysis: the Cox model The results of estimating the proportional hazards Cox models are shown in Table 4. In order to test the hypotheses, different empirical models were estimated. Model 1 included the most widely used independent variables in the literature (LN_ASSETS, LN_AGE) as well as the control variable (SPAIN). Model 2 added the variables related to patent activity and the technological level of industry (PAT and HIG-TECH). In Model 3 and Model 4, variables referring to the strategic partners (VENT_CAP and IND_SHAR) and financial performance (ROA, TOT_TUR, LEV_R and LEV_RSQUAR) were included, respectively. Finally, all of the variables in Model 5 were considered simultaneously. In order to test the assumption of proportionality, a test based on Schoenfeld residuals was performed for each model. In all of them, the null hypothesis that the hazard rate is proportional is not rejected, suggesting that the models are correctly specified. INSERT Table 4 Cox estimation HERE The results of the Cox estimations show a significant inverse effect of both the firm size (LNASSETS), more smoothly, and the firm age (LN_AGE) on failure hazard. Thus, firm size and firm age have a positive effect on USOs’ survival, and the failure hazard is reduced as firm size or firm age increase. These findings are consistent with those by Conceiçao and Faria (2014) and Fernández-López et al. (2020). Hence, Hypothesis 1 and 2 are validated. Contrary to expectations, patent activity (PAT) shows a significant positive relationship with failure hazard, that is, it reduces the survival probability. Although this finding does 20 not validate Hypothesis 3, it is similar to that of Fernández-López et al. (2020). Similar to Rodríguez-Gulías et al. (2016) and Fernández-López et al. (2020), operating in hightech sectors (HIGH_TECH) seems to have no effect on a USO’s failure propensity. Hypothesis 4 is not validated. No effect of the presence of venture capital partners (VENT_CAP) on failure hazard was found. Hence, the Hypothesis 5 cannot be validated. Having industrial partners (IND_SHAR) seems to be negatively related to failure hazard, validating Hypothesis 6. This result is consistent with that by Prokop et al. (2019). Concerning the financial ratios, the results do not indicate any effect of return on assets (ROA) on failure hazard, similar to Rodríguez-Gulías et al. (2016) and Fernández-López et al. (2020). In this way, the Hypothesis 7 is not validated. Meanwhile, asset efficiency (TOT_TUR) seems to be negatively related to firm hazard failure in USOs, supporting the Hypothesis 8. This result is consistent with that by Rodríguez-Gulías et al. (2016). In turn, the leverage ratio (LEV_R) has an inverted U-shaped relationship with the failure hazard, contrary to the results expected in Hypothesis 9 as well as the findings of Fernández-López et al. (2020). Finally, the country’ institutional factors also matter in determining a USO’s chances to survive; being located in Spain (SPAIN) significantly increases the failure hazard of USOs. 5 Result Discussion From the empirical findings of this study, it emerges that old and large USOs survive more than their smaller and younger counterparts. This is in line with the organizational population perspectives and resource-based view arguments, remarking that the wellestablished set of resources and expertise of large and old USOs helps them in achieving 21 a competitive advantage position for surviving the difficulties related to the development of innovation and its commercialization, allowing a more resilient organizational advance. From the organizational ecology perspective, the concepts of the “liability of smallness” and “liability of newness” also support the superior organizational processes and routines of larger and older USOs that let them efficiently conduct their business, decreasing the firm failure rate. The findings also reveal that patenting USOs face lower chances of survival. This unexpected result might be partly explained by the type of technology developed by USOs. Thus, USOs tend to patent radical technology to a greater extent than other firms (De Coster and Butler, 2005; Stephan, 2014). However, radical innovation is characterized by high risks of failure (Scaringella et al., 2017; Bock et al., 2018) and requires prior generation of valuable organizational resources (Nerkar and Shane, 2003). Without this “cushion” of resources, radical technology may decrease the firm’s chances to survive. Further, given that USOs are formed by researchers and other academics, they are usually focused on the technology development rather than on its effective translation into marketable and profitable products. As previously remarked, USOs lack managerial capabilities and prior expertise in industry, which leads them to undervalue the entrepreneurial orientation and economic returns needed for firms (Neves and Franco, 2018; Prokop et al., 2019). This creates an emphasis on patenting that is not counterbalanced with clear business planning and product development. High-tech USOs do not exhibit superior chances to survive compared to non-high-tech USOs, suggesting that the technological level of the industries where USOs operate is not relevant in determining their survival chances. In this respect, other industry’s features, such as the degree of industry concentration (Nerkar and Shane, 2003), may have more relevance for firm survival. Nevertheless, this requires further research. 22 Similarly, venture-backed USOs have no higher chances to survive compared to nonventure-backed ones. The evidence does not support the existence of screening, signalling or monitoring effects of venture capitalists on USOs’ survival. This counterintuitive result may be related to the low maturity of the venture capital industry in Continental European countries compared to Anglo-Saxon ones. Indeed, significant differences in the institutional contexts may lead the former to underperform in terms of firm growth and survival compared to the later (Pommet, 2017; Bruton et al., 2010). Again, this issue requires further study. Industrial partners increase a USO’s chances to survive. This evidence is in line with the arguments which claim that industrial links improve access to critical skills and resources, helping newcomers to overcome the so-called liability of “newness in the sector”. In this respect, industrial partnerships may help academic founders in developing managerial capabilities and networking, contributing to firm survival. Firm profitability seems to have no effect on a USO’s survival prospects, suggesting the irrelevance of this economic indicator in predicting a USO’s survival. This result, although theoretically counterintuitive, is partly consistent with the empirical studies that report the low financial performance reached by USOs, which does not prevent them from having higher survival rates than non-USOs (Cantner and Goethner, 2011; Criaco et al., 2014; Rodríguez-Gulías et al., 2016). This can be partly explained by the support provided by parent universities that keeps USOs alive (Su and Sohn, 2015; Civera and Meoli, 2018), sometimes in an artificial way. In contrast, asset efficiency increases a USO’s survival prospects. This finding, together with that referred to firm profitability, suggests that the direct way in which firm efficiency may influence firm survival is more important than the indirect one (i.e., 23 through firm profitability). In other words, USOs that are more efficient in the use of their assets to generate income are expected to increase their survival prospects. Contrary to expectations, the findings reveal a U-shaped relationship between USOs’ survival and their indebtedness. Thus, the estimated coefficients indicate that, initially, financial leverage decreases a USO’s chances to survive, however there is a turning point after which the leverage enables the firms’ survival prospects. This finding speaks in favour of a signalling effect of financial leverage about the quality of the business; if the USO is able to support a certain level of indebtedness, it may gain access to additional resources for organizational development. Also, the findings show that USOs located in Italy survive more than those located in Spain, suggesting that country-level institutional factors play an important role in a USO’s survival. 6 Policy and managerial implications The findings of this study have some implications for policy makers and management practice. First, given the difficulties that small and young USOs must overcome to survive, policy makers need to better design acceleration programs to effectively boost the USO’s development until it reaches a certain firm size and maturity. In this regard, improvement of the existing business incubation programs and science parks is required. They should offer services that are tailored to significantly enable USOs’ growth rather than focus on supporting business modelling and exploitation of technology (Soetanto and Jack, 2016). Second, the negative effect of patent activity on USOs’ survival concerns the type of technologies patented. Not all patents are easily transformed into successful entrepreneurial opportunities. Particularly, radical innovation, even when protected, 24 requires a long time-to-market as well as slack resources, putting a major constraint on firm survival. This constitutes important challenges for academic entrepreneurs, who need to assess a priori the feasibility of the nascent business against realizing the patent’s value by licensing or selling it to external investors. Third, the non-effect of venture capital calls for reflection on the potential drawbacks of the venture capital industry in contexts where it is still not well-developed, as in the case of Spain and Italy (Acevedo et al., 2016). In this regard, policy schemes targeted to stimulate both the national market and to attract global venture capitalists are required. Yozma, the Israeli venture capital program requiring the co-investment of university and industry, has proved to be a good practice in generating resilient incentives to invest in talented innovative high-tech start-ups, improving their growth and survival prospects (Avnimelech and Teubal, 2006) Fourth, the positive effect of industrial partners calls policy makers to actively take part in building innovative networks involving both public and private actors with the aim of effectively performing the technology transfer process from university to industry. Sharing resources, knowledge, expertise, and risks between universities, start-up firms, industry, and governmental agencies would allow increasing the competitive advantage, growth, and the chances of survival for all of the actors involved. Additionally, the findings speak in favour of embedding managerial practices in USOs. In this regard, it is possible to alleviate the frequent lack of managerial skills of academic entrepreneurs through (1) the appointment of surrogate entrepreneurs and (2) the involvement of professional investors. Fifth, USOs may be competitive in using investment assets to generate income. It is worth noting that the asset turnover ratio tends to vary among sectors. Thus, USOs in technology-based and hard science sectors (such as biotech) are characterized by large 31 Conceição, O. and Faria, A. P. (2014) Determinants of research-based spin-offs survival. Working paper series, NIPE WP 21/ 2014. Corsi, C., Prencipe, A., Rodríguez-Gulías, M.J., Fernández-López, S. and Rodeiro- Pazos, D. (2017) ‘The effect of parent university on firm growth: An analysis of the Spanish and Italian USOs’, Journal of Management Development, Vol. 36, No. 2, pp. 233–249. Cox, D.R. (1972) ‘Regression models and life-tables’, Journal of the Royal Statistical Society, Series B (Methodological), pp.187-220. Criaco, G., Minola, T., Migliorini, P. and Serarols-Tarrés, C. (2014) ‘To have and have not”: founders’ human capital and university start-up survival’, The Journal of Technology Transfer, Vol. 39, No. 4, pp. 567-593. Davidsson, P., Steffens, P. and Fitzsimmons, J. (2009) ‘Growing profitable or growing from profits: putting the horse in front of the car?’, Journal of Business Venturing, Vol. 24, No. 4, pp. 388-406. De Coster, R. and Butler, C. (2005) ‘Assessment of proposals for new technology ventures in the UK: characteristics of university spin-off companies’, Technovation, Vol. 25, No. 5, pp. 535-543. Delmar, F., McKelvie, A. and Wennberg, K. (2013) ‘Untangling the relationships among growth, profitability and survival in new firms’, Technovation, 33(8-9), 276-291. Dimitras, A.I., Zanakis, S.H. and Zopounidis, C. (1996) ‘A survey of business failures with an emphasis on prediction methods and industrial applications’, European Journal of Operational Research, Vol. 90, No. 3, pp.487−513. Dosi, G., Pugliese, E. and Santoleri, P. (2017) ‘Growth and survival of the" fitter"? Evidence from US new-born firms’, (No. 2017/06) ‘LEM Working Paper Series. 32 Drivas, K., Panagopoulos, A. and Rozakis, S. (2018) ‘Instigating entrepreneurship to a university in an adverse entrepreneurial landscape’, The Journal of Technology Transfer, Vol. 43, No. 4, pp. 966-985. Ebert, T., Brenner, T. and Brixy, U. (2019) ‘New firm survival: the interdependence between regional externalities and innovativeness’, Small Business Economics, Vol. 53, No. 1, pp. 287-309. Esteve-Pérez, S. and Mañez-Castillejo, J.A. (2008) ‘The resource-based theory of the firm and firm survival’, Small Business Economics, Vol. 30, No. 3, pp. 231-249. European Commission (2017) ‘https://ec.europa.eu/growth/industry/key-enabling- technologies_en (March 8, 2017). Fernández-López, S.; Rodríguez-Gulías, M.J.; Dios-Vicente, A.; Rodeiro-Pazos, D. (2020) ‘Individual and joint effect of patenting and exporting on the University Spin-Offs' survival’, Technology in Society, No. 62, 101326. Fini, R., Fu, K., Mathisen, M.T., Rasmussen, E. and Wright, M. (2017) ‘Institutional determinants of university spin-off quantity and quality: a longitudinal, multilevel, cross-country study’, Small Business Economics, Vol. 48, No. 2, pp. 361-391. Fritsch, M., Brixy, U. and Falck, O. (2006) ‘The effect of industry, region, and time on new business survival–a multi-dimensional analysis’, Review of industrial organization, Vol. 28, No. 3, pp. 285-306. Fryges, H., Vogel, A. and Wagner, J. (2015) ‘The impact of R&D activities on exports of German business services enterprises: First evidence from a continuous treatment approach’, The World Economy, Vol. 38, No. 4, pp. 716-729. 33 Galati, F., Bigliardi, B., Petroni, A. and Marolla, G. (2017) ‘Which factors are perceived as obstacles for the growth of Italian academic spin-offs?’, Technology Analysis & Strategic Management, Vol. 29, No. 1, pp. 84-104. George, G. (2005) ‘Slack resources and the performance of privately held firms’, Academy of Management Journal, Vol. 48, No. 4, pp. 661-676. Geroski, P.A., Mata, J. and Portugal, P. (2010) ‘Founding conditions and the survival of new firms’, Strategic Management Journal, Vol. 31, No. 5, pp. 510-529. Geuna, A. and Muscio, A. (2009) ‘The governance of university knowledge transfer: A critical review of the literature’, Minerva, Vol. 47, No. 1, pp. 93-114. Gilsing, V.A., Van Burg, E. and Romme, A.G.L. (2010) ‘Policy principles for the creation and success of corporate and academic spin-offs’, Technovation, Vol. 30, No. 1, pp. 12-23. Gimeno, J., Folta, T.B., Cooper, A.C. and Woo, C. Y. (1997) ‘Survival of the fittest? Entrepreneurial human capital and the persistence of underperforming firms’, Administrative Science Quarterly, Vol. 42, No.4, pp. 750-783. Giot, P. and Schwienbacher, A. (2007) ‘IPOs, trade sales and liquidations: Modelling venture capital exits using survival analysis’, Journal of Banking & Finance, Vol. 31, No. 3, pp. 679-702. Giovannetti, G., Ricchiuti, G. and Velucchi, M. (2011) ‘Size, innovation and internationalization: a survival analysis of Italian firms’, Applied Economics, Vol. 43, No. 12, pp. 1511-1520. Gompers, P.A. (1995) ‘Optimal investment, monitoring, and the staging of venture capital’, The Journal of Finance, Vol. 50, No. 5, pp. 1461-1489. 34 Good, M., Knockaert, M., Soppe, B. and Wright, M. (2019) ‘The technology transfer ecosystem in academia. An organizational design perspective’, Technovation, Vol. 82, pp. 35-50. Grambsch, P.M. and Therneau, T.M. (1994) ‘Proportional hazards tests and diagnostics based on weighted residuals’, Biometrika, Vol. 81, No. 3, pp.515−526. Greco, M., Grimaldi, M., Scarabotti, L. and Schiraldi, M. M. (2013) ?The sources of competitive advantage in University Spin-Offs: a case study?’, Journal of Technology Management & Innovation, Vol. 8, No. 3, pp. 139-151. Hall, B.H. and Rosenberg, N. (2010) ‘Handbook of the Economics of Innovation’, (Vol. 1) ‘Elsevier. Harrison, R.T. and Leitch, C. (2010) ‘Voodoo institution or entrepreneurial university? Spin-off companies, the entrepreneurial system and regional development in the UK’, Regional Studies, Vol. 44, No. 9, pp. 1241-1262. Hayter, C.S. and Link, A.N. (2018) ‘Why do knowledge-intensive entrepreneurial firms publish their innovative ideas?’, Academy of Management Perspectives, Vol. 32, No. 1, pp. 141-155. Helmers, C. and Rogers, M. (2010) ‘Innovation and the Survival of New Firms in the UK’, Review of Industrial Organization, Vol. 36, No. 3, pp. 227-248. Hesse, N. and Sternberg, R. (2017) ‘Alternative growth patterns of university spinoffs: why so many remain small?’, International Entrepreneurship and Management Journal, Vol. 13, No. 3, pp. 953-984. Hirsch, C. and Tambuyzer, E. (2015) Innovative Entrepreneurship Via Spin-offs of Knowledge Centers. KVAB Press. Hodgson, G., Herman, S. and Dollimore, D. (2017) ‘Adaptability and survival in small-and medium-sized firms’, Industrial and Corporate Change, dtx 039. 35 Huynh, K.P., Petrunia, R.J. and Voia, M. (2012) ‘Duration of new firms: The role of startup financial conditions, industry and aggregate factors’, Structural Change and Economic Dynamics, Vol. 23, No. 4, pp. 354-362. Iacobucci, D., Iacopini, A., Micozzi, A. and Orsini, S. (2011) ‘Fostering entrepreneurship in academic spin-offs’, International Journal of Entrepreneurship and Small Business, Vol. 12, No. 4, pp. 513-533. Jensen, M.C. (1986) ‘Agency costs of free cash flow, corporate finance, and takeovers’, The American Economic Review, Vol. 76, No. 2, pp. 323-329. Josefy, M.A., Harrison, J.S., Sirmon, D.G. and Carnes, C. (2017) ‘Living and dying: Synthesizing the literature on firm survival and failure across stages of development’, Academy of Management Annals, Vol. 11, No. 2, pp. 770-799. Kleinbaum, D.G. and Klein, M. (2005) Survival Analysis: A Self-Learning Approach, 2nd ed., Springer Science+Business Media, Inc., New York. Krishnan, C.N.V., Ivanov, V.I., Masulis, R.W. and Singh, A.K. (2011) ‘Venture capital reputation, post-IPO performance, and corporate governance’, Journal of Financial and Quantitative Analysis, Vol. 46, No. 5, pp. 1295-1333. Kroll, H. and Liefner, I. (2008) ‘Spin-off enterprises as a means of technology commercialisation in a transforming economy—Evidence from three universities in China’, Technovation, Vol. 28, No. 5, pp. 298-313. Laursen, K. and Salter, A. (2004) ‘Searching high and low: what types of firms use universities as a source of innovation?’, Research Policy, Vol. 33, No. 8, pp. 1201-1215. Löfsten, H. (2016) ‘Business and innovation resources: Determinants for the survival of new technology-based firms’, Management Decision, Vol. 54, No. 1, pp. 88- 106. 36 Lundqvist, M.A. (2014) ‘The importance of surrogate entrepreneurship for incubated Swedish technology ventures’, Technovation, Vol. 34, No. 2, pp. 93-100. Luoma, M. and Laitinen, E. (1991) Survival analysis as a tool for company failure prediction’, Omega, Vol. 19, No. 6, pp.673−678. Mahmood, T. (2000) ‘Survival of newly founded businesses: A log-logistic model approach’, Small Business Economics, Vol. 14, No. 3, pp. 223-237. March-Chordà, I. and Yagüe-Perales, R.M. (2011) ‘Biopharma business models in Canada’, Drug Discovery Today, Vol. 16, No. 15-16, pp. 654-658. Markham, S.K., Ward, S.J., Aiman‐Smith, L. and Kingon, A.I. (2010) ‘The valley of death as context for role theory in product innovation’, Journal of Product Innovation Management, Vol. 27, No. 3, pp. 402-417. Mas-Verdú, F., Ribeiro-Soriano, D. and Roig-Tierno, N. (2015) ‘Firm survival: The role of incubators and business characteristics’, Journal of Business Research, Vol. 68, No. 4, pp. 793-796. Mata, J., Portugal, P. and Guimaraes, P. (1995) ‘The survival of new plants: Startup conditions and post-entry evolution’, International Journal of Industrial Organization, Vol. 13, No. 4, pp. 459-481. Midler, C. (2019) ‘Crossing the Valley of Death: Managing the When, What, and How of Innovative Development Projects’, Project Management Journal, Vol. 50, No. 4, pp. 447-459. Miranda, F.J., Chamorro, A. and Rubio, S. (2017) ‘Determinants of the intention to create a spin-off in Spanish universities’, International Journal of Entrepreneurship and Innovation Management, Vol. 21, No. 4-5, pp. 299-317. 37 Miranda, F.J., Chamorro, A. and Rubio, S. (2018) ‘Re-thinking university spin-off: A critical literature review and a research agenda’, The Journal of Technology Transfer, Vol. 43, No. 4, pp. 1007-1038. Motohashi, K. (2005) ‘University–industry collaborations in Japan: The role of new technology-based firms in transforming the National Innovation System’, Research Policy, Vol. 34, No. 5, pp. 583-594. Motta, G.D.S., Garcia, P.A.D.A. and Quintella, R.H. (2015) ‘A patentoscientometric approach to venture capital investment prioritization’, Journal of the Association for Information Science and Technology, Vol. 66, No. 4, pp. 765-777. Mudambi, R. and Zahra, S.A. (2007) ‘The survival of international new ventures’, Journal of International Business Studies, Vol. 38, No. 2, pp. 333-352. Myers, S.C. (1984) ‘The capital structure puzzle’, The Journal of Finance, Vol. 39, No. 3, pp. 574-592. Nerkar, A. and Shane, S. (2003) ‘When do start-ups that exploit patented academic knowledge survive?’, International Journal of Industrial Organization, Vol. 21, No. 9, pp. 1391-1410. Neves, M. and Franco, M. (2018) ‘Academic spin‐off creation: barriers and how to overcome them’, R&D Management, Vol. 48, No. 5, pp. 505-518. Novotny, A. (2020) ‘Bridging or isolating? The role of the university Technology Transfer Office in the start-up incubation ecosystem. In Research Handbook on Start-Up Incubation Ecosystems. Edward Elgar Publishing. Novy-Marx, R. (2013) ‘The other side of value: The gross profitability premium’, Journal of Financial Economics, Vol. 108, No. 1, pp. 1-28. Oliveira, M.A., Ferreira, J.J.P., Ye, Q. and Van Geenhuizen, M. (2013, September) ‘Spin-up: A comprehensive program aimed to accelerate university spin-off growth. 38 In Proceedings of the 8th European Conference on Innovation and Entrepreneurship (ECIE 2013), Vol. 1, pp. 34-44. Parmentola, A. and Ferretti, M. (2018) ‘Stages and trigger factors in the development of academic spin-offs’, European Journal of Innovation Management, Vol. 21, No. 3, pp. 478-500. Phillips, B.D. and Kirchhoff, B.A. (1989) ‘Formation, growth and survival; small firm dynamics in the US economy’, Small Business Economics, Vol. 1, No. 1, pp. 65-74. Pommet, S. (2017) ‘The impact of the quality of venture capital financing and monitoring on the survival of IPO firms’, Managerial Finance, Vol. 43, No. 4, pp. 440- 451. Prokop, D., Huggins, R. and Bristow, G. (2019) ‘The survival of academic spinoff companies: An empirical study of key determinants’, International Small Business Journal, Vol. 37, No. 5, pp. 502-535. Rajan, R. G. and Zingales, L. (1995) ‘What do we know about capital structure? Some evidence from international data?’, The Journal of Finance, Vol. 50, No. 5, pp. 1421-1460. Ramaciotti, L. and Rizzo, U. (2015) ‘The determinants of academic spin‐off creation by Italian universities’, R&D Management, Vol. 45, No. 5, pp. 501-514. Rasmussen, E. and Borch, O. J. (2010) ‘University capabilities in facilitating entrepreneurship: A longitudinal study of spin-off ventures at mid-range universities’, Research Policy, Vol. 39, No. 5, pp. 602-612. Rasmussen, E. and Rice, M.P. (2011) ‘A framework for government support mechanisms aimed at enhancing university technology transfer: the Norwegian case’, International Journal of Technology Transfer and Commercialisation, Vol. 11, No. 1-2, pp. 1-25. 39 Rasmussen, E., Mosey, S. and Wright, M. (2014) ‘The influence of university departments on the evolution of entrepreneurial competencies in spin-off ventures’, Research Policy, Vol. 43, No. 1, pp. 92-106. Revest, V. and Sapio, A. (2012) ‘Financing technology-based small firms in Europe: what do we know?’, Small Business Economics, Vol. 39, No. 1, pp. 179-205. Rodeiro-Pazos, D., Fernández-López, S., Otero-González, L. and Rodríguez- Sandiás, A. (2012) ‘A resource-based view of university spin-off activity: New evidence from the Spanish case’, Revista Europea de Dirección y Economía de la Empresa, Vol. 21, No. 3, pp. 255-265. Rodríguez-Gulías, M.J., Rodeiro-Pazos, D., & Fernández-López, S. (2016) ‘Is university-based entrepreneurship successful? The Spanish case’, International Journal of Globalisation and Small Business, Vol. 8, No. 4, pp. 373-390. Rodríguez-Gulías, M.J., Rodeiro-Pazos, D., Fernández-López, S., Corsi, C. and Prencipe, A. (2018) ‘The role of venture capitalist to enhance the growth of Spanish and Italian university spin-offs’, International Entrepreneurship and Management Journal, Vol. 14, No. 4, pp. 1111-1130. Salimath, M.S. and Jones III, R. (2011) ‘Population ecology theory: implications for sustainability’, Management Decision, Vol. 49, No. 6, pp. 874-910. Sapienza, H.J., Autio, E., George, G. and Zahra, S.A. (2006) ‘A capabilities perspective on the effects of early internationalization on firm survival and growth’, Academy of Management Review, Vol. 31, No. 4, pp. 914-933. Scaringella, L., Miles, R. E. and Truong, Y. (2017) ‘Customers involvement and firm absorptive capacity in radical innovation: The case of technological spin-offs’, Technological Forecasting and Social Change, Vol. 120, pp. 144-162. 40 Schoenfeld, D. (1982) ‘Partial residuals for the proportional hazards regression model’, Biometrika, Vol. 69, No. 1, pp. 239−241. Shane, S. (2002) ‘Selling university technology: patterns from MIT’, Management Science, Vol. 48, No. 1, pp. 122-137. Skute, I. (2019) ‘Opening the black box of academic entrepreneurship: a bibliometric analysis’, Scientometrics, Vol. 120, No. 1, pp. 237-265. Soetanto, D. and Jack, S. (2016) ‘The impact of university-based incubation support on the innovation strategy of academic spin-offs’, Technovation, Vol. 50, pp. 25-40. Sørheim, R., Øystein Widding, L., Oust, M. and Madsen, Ø. (2011) ‘Funding of university spin-off companies: a conceptual approach to financing challenges’, Journal of Small Business and Enterprise Development, Vol. 18, No. 1, pp. 58-73. Stephan, A. (2014) ‘Are public research spin-offs more innovative?’, Small Business Economics, Vol. 43, No. 2, pp. 353-368. Su, D.J. and Sohn, D.W. (2015) ‘Roles of entrepreneurial orientation and Guanxi network with parent university in start-ups’ performance: evidence from university spinoffs in China’, Asian Journal of Technology Innovation, Vol. 23, No. 1, pp. 1-19. Susarla, A. and Barua, A. (2011) ‘Contracting efficiency and new firm survival in markets enabled by information technology’, Information Systems Research, Vol. 22, No. 2, pp. 306-324. Tamer Cavusgil, S., Calantone, R.J. and Zhao, Y. (2003) ‘Tacit knowledge transfer and firm innovation capability’, Journal of Business & Industrial Marketing, Vol. 18, No. 1, pp. 6-21. Tavassoli, S. and Jienwatcharamongkhol, V. (2016) ‘Survival of entrepreneurial firms: the role of agglomeration externalities’, Entrepreneurship & Regional Development, Vol. 28, No. 9-10, pp. 746-767. 47 Figure 1 Kaplan-Meier estimate of the survival function 0.00 0.25 0.50 0.75 1.00 Survival probability 0 5 10 15 Analysis time (years) Kaplan-Meier survival estimate 48 Figure 2 Smoothed hazard estimates .02 .04 .06 .08 .1 Harad rate 0 5 10 15 Analysis time (years) Smoothed hazard estimate