Please summarize the attached journal article. The summary should…

Question Answered step-by-step Please summarize the attached journal article. The summary should… Please summarize the attached journal article. The summary should discuss the main finding(s), the methods used, the implications of the work, and assess the pros and cons of the work.  Executive summary Financing decisions taken by start-ups are known to determine their failure or success, and thus are considered to be among the most important decisions investigated in entrepreneurship research (Cassar 2004). A number of studies have tried to explain start-ups’ financing decisions by relying on Pecking Order Theory (Myers 1984), which states that firms always give priority to the cheapest type of financing, being internal financing, and only when this is exhausted, refer to costlier external debt. External equity, given its even higher costs, is considered only as a last resort. While many studies support the pecking order rationale within the context of newly established firms, others present opposing evidence that start-ups prefer external equity over debt. In order to dispel this confusion, our study extends the Pecking Order Theory by putting forward entrepreneurial orientation (EO) – i.e. a start-up’s propensity to engage in innovative activities, willingness to take risks, and proactiveness relative to market opportunities – as an important yet overlooked determinant of start-ups’ financing decisions, and respectively investigates the conditions under which this EO matters. The theorized relationships are examined using observations from 4456 German start-ups surveyed between 2014 and 2016. Consistent with theoretical arguments, we find empirical support for our main propositions. Specifically, our findings suggest that start-ups are more likely to follow the traditional pecking order, i.e. to prefer external debt over external equity financing, when their EO is low, but less so when their EO is high. While these findings hold for start-ups founded in less risky industries, they are altered if start-ups are active in riskier ones, as external (debt) providers then favor start-ups that demonstrate a greater fit to this uncertain environment. Lastly, our study shows that start-ups’ EO as well as the fit between their EO and industry-level risk have a stronger effect on their financing strategies in early stages of their development than later on. The findings of this study offer important contributions to both theory and practice. By taking an entrepreneurial strategy and contingency perspective, our study primarily contributes to the Pecking Order Theory. Whereas it was already known that corporate finance theories cannot entirely explain start-ups’ financing decisions and several studies have advanced a firm strategy as a potential https://doi.org/10.1016/j.jbusvent.2019.01.006 Received 6 March 2018; Received in revised form 21 January 2019; Accepted 31 January 2019 ⁎ Corresponding author. E-mail addresses: E..e@UGent.be (E. Vaznyte), P..s@UGent.be (P. Andries). Journal of Business Venturing 34 (2019) 439-458 Available online 14 February 2019 0883-9026/ © 2019 Elsevier Inc. All rights reserved. T determinant (e.g., Fraser et al. 2015), our study is the first to integrate insights on entrepreneurial orientation with Pecking Order Theory. In particular, we extend Pecking Order Theory by suggesting that a start-up’s EO together with two boundary conditions – a start-up’s development stage and industry-level risk – will affect the costs and benefits associated with external debt and equity capital, and therefore will determine a start-up’s use of these respective forms of financing. In this way, we not only test alternative variables predicting start-ups’ financing decisions (see call by Hanssens et al. 2016; Huyghebaert and Van de Gucht 2007; Mueller 2008), but also outline the necessity to investigate the boundary conditions under which Pecking Order Theory is most likely to hold. Our study also makes an important contribution to EO research. First, while many scholars identified EO as a key driver for firms’ financial performance, we demonstrate its merits for explaining firms’ financing decisions. Consequently, we advance financing decisions as important intermediary mechanisms through which EO may affect financial performance, thereby improving our understanding of how and why EO affects firm performance (see call by Wales et al. 2013, 2015). Additionally, by jointly considering start-ups’ EO, as well as organizational and environmental characteristics, we contribute to the previous call in the literature for a richer characterization of different contexts which helps to better understand the manifestation and consequences of EO (e.g., Miller 2011). With respect to practical implications for start-up founders, we demonstrate which type of external financing best fits their level of EO and under which conditions (in particular, the uncertainty of their business environment and the development stage of their startup). This can help to better understand and target proper sources of financing. As for financiers, we suggest to consider also start-ups’ entrepreneurial strategy along other funding criteria, as this may reduce the level of information asymmetries and thus enhance the funding decision process. 1. Introduction Access to entrepreneurial finance is a crucial success factor for start-ups (Cassar, 2004; Cole et al., 2016; Coleman, 2000; Ko and McKelvie, 2018) as it allows them to bridge the valley of death and grow. As a firm’s failure or success depends heavily on its initial financing decisions (Åstebro and Bernhardt, 2003; Mueller, 2008; Robb and Robinson, 2014; Vanacker and Manigart, 2010), ample studies have investigated start-ups’ access to financing. Many of them build on well-known corporate finance theories, among which Pecking Order Theory figures prominently. Pecking Order Theory states that internal financing is less costly and therefore more preferred than external financing; with external debt being cheaper and more preferable than external equity (Myers, 1984; Myers, 1984). A number of empirical studies find support for Pecking Order Theory in explaining the financing practices of start-ups (e.g., Cassar, 2004; Cosh et al., 2009; Giudici and Paleari, 2000; Mina et al., 2013; Robb and Robinson, 2014). However, several others observe that, contrary to theoretical predictions, start-ups first approach private equity investors before seeking external debt financing (e.g., Carpenter and Petersen, 2002; Paul et al., 2007). Yet, it remains unclear why certain start-ups follow the traditional pecking order while others do not. Several authors have argued that this lack of understanding is due to corporate finance theories’ failure to account for start-ups’ strategic goals, which they suggest to be crucial drivers of start-ups’ financing decisions (Barton and Gordon, 1987; Chaganti et al., 1996; Fraser et al., 2015; Matthews et al., 1994; McMahon and Stanger, 1995). Cassar (2004), Cosh et al. (2009) and Mina et al. (2013) made a first attempt at unravelling this relationship by testing the importance of start-ups’ growth objectives in explaining their tendency to seek external financing. We advance their insights further by investigating to what extent a start-up’s strategic posture affects its decision to attract different types of external financing, namely external debt and external equity capital. While debt and equity are the two most commonly used forms of external financing (Berger and Udell, 1998), they are different in essence (e.g., Audretsch and Lehmann, 2004; Cole et al., 2016; Ueda, 2004; Winton and Yerramilli, 2008). This makes a start-up’s decision to finance business activities through external debt or equity, or a combination of both, complicated and relevant to research further. Additionally, although several studies have already included environmental conditions (e.g., Cosh et al., 2009; MacKay and Phillips, 2005) and start-ups’ organizational characteristics such as development stage (e.g., Chaganti et al., 1996; Ko and McKelvie, 2018) as relevant factors directly influencing capital structure, we advance these characteristics as important yet neglected contingencies in the relationship between strategic goals and external financing decisions. In particular, we put forward a start-up’s entrepreneurial orientation (EO), a strategic construct that captures a firm’s entrepreneurial strategy-making practices, management philosophies, and behaviors (Anderson et al., 2009), as an important determinant of external debt and external equity financing decisions. EO has been defined as a firm’s propensity to “engage[] in product-market innovation, undertake[] somewhat risky ventures, and […] come up with ‘proactive’ innovations, beating competitors to the punch” (Miller 1983: 771). It has been extensively used in explaining firm financial performance, arguing that firms adopting a more entrepreneurially oriented posture perform better (Rauch et al., 2009; Shahzad et al., 2016; Wales, 2016). Yet its implications for other outcome variables such as strategic and tactical decisions have received little investigation so far (Lumpkin and Dess, 1996; Wales et al., 2013). In particular, very little is known about the relationship between EO and start-ups’ financing decisions. While the study by Wiklund and Shepherd (2005) investigates the moderating role of access to capital in explaining the relationship between EO and small firm performance, it does not explain how EO affects access to (different) resources in the first place. And while the research by Mousa et al. (2015) illustrates the importance of EO in attracting financial resources at the time of IPO, it opens further questions about whether EO is important for initial financing decisions. This lack of understanding is striking, since how business start-ups are financed is a fundamental question (Cassar, 2004), and a start-up’s financing decisions are among the most important determinants of its failure or success. Given the vital role these new companies play in our economy (i.e. contribution to employment growth, competition, and innovation) (Coleman, 2000; Colombo and Grilli, 2007; Mina et al., 2013), it is undoubtedly a meaningful area to be researched. The current study therefore aims to clarify (1) to what extent a start-up’s level of EO explains its use E. Vaznyte and P. Andries Journal of Business Venturing 34 (2019) 439-458 440 of external debt and external equity in meeting its total financing needs, and (2) to what extent this relationship is contingent on environmental and organizational factors. Extending Pecking Order Theory, we propose that a start-up’s choice of external debt and external equity does not only depend on the costs, but also on the benefits these financing forms entail. Consequently, we argue that a start-up’s entrepreneurial strategy, and in particular its level of EO, differently affects the costs and benefits associated with external debt and equity, and therefore also its use of these respective forms of financing. We hypothesize that because of these differences in terms of costs and benefits, start-ups with low levels of EO will, in line with traditional Pecking Order Theory, use more external debt and less external equity financing; while start-ups with high levels of EO will rely more on external equity capital and less on external debt capital to meet their financing needs. Moreover, knowing that the relationship between strategy (and EO in particular) and various organizational outcomes is contextspecific, i.e. depends on environmental and organizational factors (Covin and Slevin, 1989, 1991; Dess et al., 1997; Wiklund and Shepherd, 2005), we further investigate this EO-external financing relationship through a contingency approach (Drazin and Van de Ven, 1985). In particular, we argue that the effect of EO on the costs and benefits associated with external debt and equity will depend on the uncertainty of the business environment (i.e. the industry-level risk) and the life-stage the start-up is in (i.e. whether or not it has reached the break-even point). As such, we also expect that the combination of a start-up’s entrepreneurial orientation, the riskiness of the industry it operates in, and its life-stage will jointly determine its use of external debt and equity financing. We test our hypotheses using pooled cross-sectional survey data on 4456 German start-ups. Compared to traditional data on capital structure, this unique survey dataset allows us to attain subtle insights on the role of start-ups’ entrepreneurial strategy in explaining their external financing decisions, i.e. the degree to which a start-up meets its total financing needs in a given year through external debt and external equity financing. In support of our main propositions, we find that start-ups’ EO is related to these external financing decisions, and that this relationship is dependent on start-ups’ development stage and industry-level risk. Supplementary analyses of follow-up survey data allow us to draw causal inferences between EO and financing decisions, pointing not only to the robustness of our main findings but also suggesting an enduring effect of start-ups’ initial entrepreneurial orientation on later external financing decisions. By conducting this study, we primarily contribute to the emerging debate regarding the applicability of corporate finance theories – and Pecking Order Theory in particular – for explaining financing decisions of newly established firms. It was already known that traditional corporate finance theories cannot fully explain the financing decisions of start-ups, and several studies have suggested to look at firm strategy, and environmental and organizational characteristics as potential determinants of start-ups’ financing decisions. Our study is the first, as far as we know, to conceptually integrate these emerging insights with Pecking Order Theory by arguing (a) that EO affects the costs and benefits associated with external debt and equity, and (b) that these effects depend on environmental and organizational characteristics. In particular, we demonstrate that environmental (i.e. industry-level risk) and organizational (i.e. start-ups’ development stage) characteristics interact with EO in determining the costs and benefits of external debt and equity, and thus are important boundary conditions modifying the relationship between EO and start-ups’ financing decisions. As such, we respond to the previous call in the literature to test alternative predictor variables of financing decisions (Hanssens et al., 2016; Huyghebaert and Van de Gucht, 2007), and also point to the importance of boundary conditions for the traditional pecking order to hold. Secondly, we advance the EO literature in multiple ways. First, we demonstrate that EO is instrumental for explaining not only a firm’s financial performance, but also its financial decisions. As such, we underline the role of financial decisions as an intermediary mechanisms between EO and venture performance – a research area currently lacking solid theoretical and empirical evidence (Wales et al., 2013). Second, we provide empirical evidence supporting the merits of a contingency approach and respectively a joint investigation of start-ups’ strategic, environmental and organizational characteristics (e.g., Covin and Lumpkin, 2011; Miller, 2011; Titus and Anderson, 2018), which contributes to a better understanding of financial decision making of a newly established firms. Besides theoretical contributions, our findings also have important implications to practice, and in particular to start-up founders and their financiers. 2. Theoretical argumentation 2.1. New Ventures’ financing decisions: does pecking order theory apply? Deciding which different financial resources to attract and how to combine them is not a straightforward process (Chaganti et al., 1996), especially when considering informationally opaque start-ups. For a long time, start-ups’ external financing decisions have been explained by the Pecking Order Theory (Myers, 1984; Myers, 1984), which states that costs associated with information asymmetries arising between the entrepreneur and outside financiers determine financing decisions. That is, outside financiers are confronted with an adverse selection problem as they cannot directly assess the quality of a firm, and as a result require “lemons” premiums (Akerlof, 1970). Since internal financing is not subject to information asymmetries (Myers, 1984), start-ups should prefer it over more costly external financing in general. However, as most start-ups have limited internal funds, they often need to attract external financing in order to develop and survive (Bruns and Fletcher, 2008; Eckhardt et al., 2006; Fryges et al., 2015), and external debt and equity financing are most commonly used in this respect (Berger and Udell, 1998). Debt financiers require a collateral and have priority rights to a start-up’s assets in case of its bankruptcy. As a result, the value of these debt securities will be only slightly affected when the start-up’s actual performance is revealed to the market (Myers, 1984, 2001). Debt financiers hence face relatively low information asymmetries, and E. Vaznyte and P. Andries Journal of Business Venturing 34 (2019) 439-458 441 thus charge relatively low risk premiums. Equity financiers, on the other hand, do not require any guarantee, but demand ownership rights and high “lemons” premiums for their risky investments (Carpenter and Petersen, 2002; Vanacker and Manigart, 2010). As the cost of external debt is hence typically lower than that of external equity, Pecking Order Theory predicts that firms will issue debt securities first, and equity securities only as a last resort (e.g., Cosh et al., 2009; Frank and Goyal, 2003; Landström, 2017; Walthoff- Borm et al., 2018). While some studies indeed find evidence for this pecking order, others observe a preference of start-ups for external equity over external debt, which is in contrast to theoretical predictions. The advantages of external equity have been promoted as the main reason for this finding. In particular, it has been argued that the benefits of external equity provided by sophisticated investors, such as business angels and venture capitalists, are substantially larger than those of external debt, thereby compensating for the larger cost that external equity entails (e.g., Carpenter and Petersen, 2002; Garmaise, 2001; Fama and French, 2005; Frank and Goyal, 2003; Paul et al., 2007). Equity providers are considered to be better at assessing start-ups’ commercialization prospects and to offer higher value-added than banks (Carpenter and Petersen, 2002; Cosh et al., 2009). On the one hand, private equity providers are better at coping with information asymmetries surrounding new ventures, as they are more prone to employ non-traditional methods in assessing firms’ future value where they consider not only financial start-ups’ characteristics, but also non-financial ones (business prospect, alliances, top management team characteristics, etc.) (Audretsch and Lehmann, 2004; Baum and Silverman, 2004; Gompers and Lerner, 2001; Maxwell et al., 2011; Ueda, 2004; Winton and Yerramilli, 2008). On the other hand, private equity investors do not only provide the financial means, but also enable start-ups to join their networks, serve as a sound board, and guide and support them with marketing, managerial and technical advice (Bertoni et al., 2011; Block et al., 2018; Cole et al., 2016; Manigart and Struyf, 1997; Paul et al., 2007; Zacharakis and Meyer, 2000). Through these activities, equity investors support the start-up in introducing new products or services into the market as well as in establishing a competitive position (Gompers and Lerner, 2001). Moreover, a start-up’s ability to attract external equity capital is often regarded as a sign of quality (Bellavitis et al., 2017; Bertoni et al., 2011; Carpenter and Petersen, 2002) and helps start-ups to secure additional funding under more beneficial terms and valuations (Paul et al., 2007; Cole et al., 2016; Stuart et al., 1999). Although the latter stream of studies clearly suggests that start-ups do not only take the costs, but also the benefits of different forms of financing into account, the circumstances under which these benefits outweigh their concomitant costs remain unclear. We therefore extend Pecking Order Theory by proposing that a start-up’s strategy, in particular its level of EO, in combination with two contingencies – environmental and organizational characteristics – will affect the costs and benefits associated with external debt and equity capital, and therefore its use of these respective forms of financing. 2.2. Strategy perspective: the role of entrepreneurial orientation Entrepreneurial Orientation (EO) embodies entrepreneurial strategy-making processes pertaining to planning, decision-making and strategic management (Lumpkin and Dess, 1996) that are used by start-up founders in order to achieve their firm’s organizational objectives, fulfil their vision and establish a competitive advantage in the marketplace (Rauch et al., 2009). According to Miller (1983) and Covin and Slevin (1989), the entrepreneurial orientation of a firm reflects its (1) innovativeness, i.e. its propensity to engage in creativity and experimentation through the introduction of new products/services, (2) risk-taking, i.e. its proclivity to take bold actions, and (3) proactiveness, i.e. the degree to which it seeks novel opportunities and aims to be ahead of competitors. A shared variance between these three characteristics determines the extent to which a firm is entrepreneurial (Covin et al., 1990; Covin and Wales, 2012, 2019). In what follows, we argue that a start-up’s level of EO will affect the costs and benefits associated with external debt and equity capital differently, and therefore also its use of these financing forms. When looking at start-ups with low levels of EO, it can be argued that the cost of external debt is indeed lower than the cost of external equity, as argued by the Pecking Order Theory. First of all, external debt financing is often considered as a relatively cheap and plentiful source of financing (Landström, 2017; Vanacker and Manigart, 2010; Robb and Robinson, 2014), and thus may be appealing to start-ups with a more conservative strategic posture. External debt capital is associated not only with lower asymmetric information costs, and thus lower threat of mispricing (Myers, 2001), but also with lower opportunity costs of the search for private investors (Bertoni et al., 2011; Cosh et al., 2009) than external equity capital. Second, external equity financing involves a dilution of ownership shares (Harris and Raviv, 1991) and imposes the investors’ decision power on a firm’s strategy (Mueller, 2008; Winton and Yerramilli, 2008), which is known to be a great concern to risk-averse founders, who generally have a desire to retain control of the firm (Miller, 1983). Finally, more conservative firms with lower levels of EO may be reluctant to disclose their proprietary information to external investors as they may fear expropriation of their business idea (Ueda, 2004; Mina et al., 2013). Whereas venture capitalists may pursue the project without its original founder or copy it “by passing the project content to a firm in which the venture capitalist has already been investing and having that firm undertake the project”, banks usually do not have such intentions (Ueda 2004: 604). The contractual terms with banks are better-defined than the ones with private equity investors, who can for example terminate agreements on a relatively short notice and without any explanations (Stuart et al., 1999). While the cost of external equity seems to exceed the cost of external debt for ventures with low levels of EO, the benefits of external equity and of external debt do not appear to differ significantly for these ventures, as the networks and professional competencies of equity investors may not be particularly important for start-ups that do not pursue a strategy of innovation, risk-taking and proactiveness (Wiklund and Shepherd, 2005). As a result, the value of external equity for low EO start-ups will be lower than that of external debt financing. This situation reflects the traditional pecking order, where external equity capital is perceived only as a last resort as described in corporate finance theories. However, it can be argued that the value of external debt financing will differ for start-ups with higher levels of EO. Start-ups with E. Vaznyte and P. Andries Journal of Business Venturing 34 (2019) 439-458 442 high levels of EO “are willing to take on high-risk projects with chances of very high returns, and are bold and aggressive in pursuing opportunities” (Covin and Slevin 1991: 7-8). We expect that this strategic posture will affect the cost of external debt financing. First, debt financiers providing a loan or a credit in exchange for fixed debt related payments do not reap higher returns in case of borrowers’ success, and therefore avoid funding firms that undertake risky and innovative activities (Bruns and Fletcher, 2008; Carpenter and Petersen, 2002; Deakins and Hussain, 1994), making it more difficult for firms with high levels of EO to obtain external debt capital. Second, external debt financing typically reduces cash flow (Paul et al., 2007) and can impose credit rationing or covenants restricting firm behaviour (Carpenter and Petersen, 2002; Simerly and Simerly and Li, 2000). Whereas these restrictions may not hinder non-innovative, reactive and risk-averse start-ups (Covin et al., 1990), they may pose an important threat to entrepreneurially oriented start-ups, hindering them to pursue innovative and risky opportunities (Hutchinson, 1995). As a result, the cost of external debt for high EO firms is likely to exceed that for low EO firms. On the other hand, we do not expect the benefits of external debt to differ between high and low EO firms. This implies that the value of external debt financing for high EO start-ups will be lower than for low EO start-ups. This leads us to the following hypothesis: Hypothesis 1. Start-ups’ EO is negatively associated with the use of external debt financing. Also the value of external equity financing can be expected to change for start-ups with higher levels of EO. On the one hand, we expect the costs of external equity that stem from information asymmetries to be lower for firms with high levels of EO. Evidence in prior studies suggests that private equity providers, and especially sophisticated private equity providers such as business angels and venture capitalists, prefer start-ups that are proactive, innovative, and risk-taking. First of all, entrepreneurially oriented start-ups can provide them with a desired window on novel technologies (Dushnitsky and Lenox, 2005; Manigart and Struyf, 1997; Titus and Anderson, 2018) and high expected returns as a compensation for their risky investment (e.g., Audretsch and Lehmann, 2004; Hall, 2010; Sapienza et al., 1996). The fact that private equity providers are more willing to invest in highly uncertain start-ups and are more prone to employ advanced valuation methods (e.g., Amit et al., 1998; Ueda, 2004; Ko and McKelvie, 2018; Winton and Yerramilli, 2008), makes access to equity capital very likely for these firms (Carpenter and Petersen, 2002; Garmaise, 2001). At the same time, start-ups with a conservatively oriented posture may not suffice the demands of highly selective external equity providers (Bellavitis et al., 2017), and as a result experience a greater cost of external equity capital than start-ups with high level of EO. Furthermore, we expect the benefits of external equity for high EO firms to be higher than for low EO firms. As previously mentioned, private equity providers are associated with higher value-adding services to their investee firms than debt providers (e.g., Block et al., 2018; Cosh et al., 2009; Paul et al., 2007). Access to networks (including relationships with suppliers, customers, legal advisors, etc.) and professional competencies by equity investors may be especially important for entrepreneurially oriented start-ups seeking to differentiate themselves from other firms and to acquire a competitive advantage (Wiklund and Shepherd, 2005). The fact that managers of entrepreneurially oriented start-ups, contrary to conservatively oriented ones, avoid organizational constraints and provide their employees with decision-making autonomy (Miller, 1983), suggests that they are less concerned with keeping control over their firms and are open to insights of other individuals. As such, they may also perceive the participation of these equity investors in the strategic decision-making of the start-up (Manigart and Struyf, 1997; Mueller, 2008; Winton and Yerramilli, 2008) as an advantage rather than a disadvantage. Moreover, as EO is considered to be a resource-intensive strategy, entrepreneurially oriented start-ups may also acknowledge the financial advantages associated with external equity capital. It is well known that activities related to innovation, R&D, and technological leadership are inherently risky (Miller and Friesen, 1982; Hughes and Morgan, 2007), and require daring resource commitments to business activities with uncertain outcomes (Mousa et al., 2015; Covin and Slevin, 1991). As such, start-ups with high levels of EO are likely to value the larger amounts of funding that equity providers contribute (Colombo and Grilli, 2007; Manigart and Struyf, 1997). Taking these arguments together implies that the value of external equity for high EO start-ups will be higher than for low EO start-ups. As a result, start-ups with higher levels of EO will have a higher preference for external equity than start-ups with lower levels of EO: Hypothesis 2. Start-ups’ EO is positively related to the use of external equity financing. Together, these arguments can explain why some start-ups prefer external equity over external debt financing, as observed by Fama and French (2005), Frank and Goyal (2003), Carpenter and Petersen (2002), and Paul et al. (2007). Particularly, start-ups with high levels of EO may find that the cost of external debt exceeds that of external equity, while debt financiers may provide fewer benefits than equity providers for these highly entrepreneurial ventures, essentially turning external equity into “first resort” of external financing. 2.3. Contingency approach: the role of environmental and organizational characteristics After acknowledging the importance of the strategy perspective, we now turn to potential contingencies. Entrepreneurial strategy scholars argue that the relationship between EO and various organizational outcomes is context-specific, i.e. depends on environmental and organizational factors (Covin and Slevin, 1989, 1991; Dess et al., 1997; Lumpkin and Dess, 1996; Wiklund and Shepherd, 2005), and that a better appreciation of different environmental and organizational contexts is thus essential to fully understand the manifestation of EO and its consequences (Miller, 2011). As such, a contingency approach may prove to be extremely valuable as it helps to examine “the conditions or boundaries in which particular structures and processes hold” (Van de Ven et al. 2013: 396). As we will argue below, a fit or congruence between strategy and key variables, such as environmental and organizational factors, is not only crucial for reaching optimal performance (Drazin and Van de Ven, 1985; Lumpkin and Dess, 1996; Rauch et al., 2009), but also for financial decision making. Although the potential moderating role of these characteristics has been ignored in traditional Pecking E. Vaznyte and P. Andries Journal of Business Venturing 34 (2019) 439-458 443 Order Theory, we will outline below that they interact with EO in determining the costs and benefits of external debt and equity and thereby affect the relationship between EO and financing decisions. 2.3.1. Environmental contingencies: the moderating role of industry-level risk Prior studies have shown that environmental conditions moderate the relationship between a firm’s EO and various organizational outcomes (Covin and Lumpkin, 2011; Lumpkin and Dess, 1996; Wales et al., 2013; Rauch et al., 2009). They suggest that the importance of EO for firm performance is increasing with increasing levels of environmental uncertainty (e.g., Covin and Slevin, 1989; Miller, 1983; Zahra and Covin, 1995). The underlying reason is that uncertain business environments threaten firms’ viability and future performance, and therefore only firms with high levels of EO are able to gain or maintain their competitive advantage (Covin and Slevin, 1989; Miller, 1983; Miller, 1983). Surprisingly however, environmental conditions have received little attention in the traditional finance literature (see critique by Matthews et al., 1994; Simerly and Li, 2000) and have been investigated mainly as a control variable (Cosh et al., 2009; Colombo and Grilli, 2007; Eckhardt et al., 2006; MacKay and Phillips, 2005). Specifically, it has been argued that start-up environments characterized by high industry-level risk reflect a higher degree of information asymmetries and thus may indirectly lead to higher costs related to financial distress, including a higher probability of default, and higher selection, monitoring and contracting costs (Myers, 1984), which in turn may determine the availability and price of external financing (Eckhardt et al., 2006; Simerly and Li, 2000). We will argue below that in addition to this direct effect, environmental characteristics also play an important moderating role. In particular, we will argue that the fit or misfit between the industry-level risk (Dencker and Gruber, 2015) and a start-up’s level of EO, will have a significant impact on the costs of external debt and equity financing, altering the relationship between the level of EO and the use of these two financing forms. In developing our Hypothesis 1, we argued that in general, debt financiers prefer secure projects over start-ups that are innovative, risk-taking and proactive, making it more costly for firms with higher levels of EO to obtain external debt capital. However, practitioner oriented work argues that firms’ “ability to recognize and respond to changing conditions, and their ability to develop and implement effective strategies” (Berger 1997/1998: 73) are important evaluation criteria for debt financiers. As such, start-ups with a strong entrepreneurial posture, who have in general a good fit with uncertain environments (Wiklund and Shepherd, 2005), may be more positively perceived by debt providers when the industry-level risk is high. On the other hand, debt capital providers are likely to acknowledge that start-ups with a conservative orientation may be unable to adapt to this uncertain environment (Covin and Slevin, 1989), and as a result may charge higher debt-related costs. So, although we argued above that highly entrepreneurial firms are penalized by debt financiers for their innovative and risky activities (e.g., Carpenter and Petersen, 2002), we expect this premium to be lower when the environment is risky. In other words, we expect the difference between the cost of external debt capital for startups with high EO and that for start-ups with low EO to be much smaller when the industry-level risk is high than when the industrylevel risk is low. As we do not expect the effects of industry-level risk on the benefits of external debt to differ between high EO and low EO firms, this implies that: Hypothesis 3. The relationship between EO and debt financing is moderated by industry-level risk such that the relationship between EO and debt financing is relatively more negative in less risky versus riskier industries. Equity providers on the other hand, are generally more interested in riskier industries due to their potential of unsettled opportunities and risk premium (Amit et al., 1998; Baum and Silverman, 2004; Zacharakis and Meyer, 2000). Therefore, start-ups that originated from uncertain business environments may have a greater possibility of attracting this type of funding, and consequently may incur lower opportunity costs. It can be argued however that this will be particularly true when the entrepreneurial orientation of the start-ups fits this turbulent business environment. MacMillan et al. (1985) already demonstrated that equity providers only consider an investment when there is a fit between the entrepreneur and the environment, in the sense that the entrepreneur not only has familiarity with the business environment but is also able to adequately cope with the risk inherent to that environment. We know that whereas high levels of EO reflect the proactiveness, innovativeness, and risk-taking that are needed under uncertainty, low levels of EO are associated with a decisive risk aversion by the top managers (Covin and Slevin, 1989). And as previously argued, conservative entrepreneurs may hesitate to reveal their proprietary information to external parties, fearing expropriation of their business idea (Ueda, 2004; Mina et al., 2013), whereas co-creation and collaboration are especially important in uncertain environments (Sarasvathy, 2001). We therefore expect that as the industry-level risk goes up, start-ups with higher levels of EO better fit this environment, and therefore will attract equity capital more easily than start-ups with lower EO levels. Or in other words, we expect that, as the industry-level risk goes up, the decrease in the cost of external equity capital for start-ups with high EO will be higher than that for start-ups with low EO. As we do not expect the effects of industry-level risk on the benefits of external equity to differ substantially between high EO and low EO firms, this implies that the value of external equity will increase more for firms with high levels of EO than for firms with low levels of EO. As such, we hypothesize that: Hypothesis 4. The relationship between EO and equity financing is moderated by industry-level risk such that the relationship between EO and equity financing is relatively more positive in riskier versus less risky industries. 2.3.2. Organizational contingencies: passing the break-even point The extent to which firms are inclined to follow pecking order when funding their business activities depends on the availability of internal funds and the severity of information asymmetries when accessing external funds (Myers, 1984; Myers, 1984). While most start-ups have limited internal resources (e.g., Eckhardt et al., 2006), access to external funds may be especially challenging for earlystage start-ups since they lack historical data and a valuable reputation, and thus may be more disadvantaged than start-ups with a E. Vaznyte and P. Andries Journal of Business Venturing 34 (2019) 439-458 444 track record (Huyghebaert and Van de Gucht, 2007). Therefore, several scholars have argued that studies on firms’ financing decisions should control for business development stage, since the issue of information asymmetries is the most acute in early stages of firms’ life-cycle than later on (Berger and Udell, 1998; Chaganti et al., 1996; Chittenden et al., 1996; Frank and Goyal, 2003; Ko and McKelvie, 2018). We will argue below that, in addition to this potential direct effect, whether or not a start-up has reached the breakeven point (i.e. the point in time when total costs are compensated by total revenues) will also interact with its EO in affecting its financing decisions. The break-even point is a good indicator of a start-up’s development stage, as it shows a start-up’s progress towards survival and profitability (Delmar et al., 2013; Walter et al., 2014), and thus represents a reduction in information asymmetries. It can be argued that when a start-up passes the break-even point this will affect the costs and benefits of debt and equity financing, and these effects will differ for start-ups with low versus high levels of EO. In developing our Hypothesis 1, we argued that in general, debt financiers prefer secure projects over more entrepreneurial ones, making it more costly for start-ups with higher levels of EO to obtain external debt capital. However, we expect these costs to be even higher for the start-ups who are not yet break-even. More precisely, whereas before breaking-even, start-ups with high levels of EO are associated with severe issues of information asymmetries, reaching profitability serves as a signal of quality (e.g., Bruns and Fletcher, 2008; Cassar, 2004; Cosh et al., 2009). This in turn reduces information asymmetries associated with entrepreneurial startups, and thereby the cost incurred when attracting external debt capital (Harris and Raviv, 1991). Whereas before break-even, high levels of EO caution debt financiers, their concerns become much alleviated once these entrepreneurial start-ups reach the break-even point. So, although we expect that profitability will reduce the cost of external debt for both low and high EO start-ups, we expect this difference to be larger for highly entrepreneurial start-ups. Concurrently, also the benefits of external debt capital would be more important before than after break-even, but we do not expect this effect to differ substantially between start-ups with low versus high levels of EO. This implies that: Hypothesis 5. The relationship between EO and debt financing is moderated by whether or not a start-up has reached the break-even point such that the relationship between EO and debt financing is relatively more negative before break-even versus after break-even. In developing our Hypothesis 2, we argued the costs of external equity to be lower for start-ups with high levels of EO than for start-ups with low levels of EO. At the same time, we reasoned that the benefits of external equity are higher for start-ups with high levels of EO than for those with low levels of EO. When contemplating on these effects in relation to the break-even point of the startup, we believe that after break-even, the cost of equity financing will go down as equity financiers among projects with high information asymmetries still prefer those projects where selection or monitoring costs are less severe (Amit et al., 1998), but this decrease will not differ substantially between start-ups with high levels of EO and those with low levels of EO. Or in other words, the difference between the cost of equity capital for start-ups with high EO and that for start-ups with low EO will not change drastically, as equity financiers still have a preference for start-ups with high levels of EO after break-even (e.g., until they “go public”) (Amit et al., 1998; Berger and Udell, 1998). On the other hand, we do expect to see important implications of breaking-even with respect to the benefits obtained from equity financing. Whereas during their early development, entrepreneurially oriented start-ups need to rely more heavily than their conservative counterparts on the network and competencies of their equity providers, these additional benefits become less important once they have developed their own competencies, developed entrepreneurial processes, recruited necessary team members and are generating sustainable returns (e.g., Cole et al., 2016; Sapienza et al., 1996). Moreover, whereas a start-up’s ability to attract equity financing is often regarded as a sign of quality (Bellavitis et al., 2017; Bertoni et al., 2011; Carpenter and Petersen, 2002) and helps start-ups secure additional funding under more beneficial terms and valuations (Paul et al., 2007; Cole et al., 2016; Stuart et al., 1999), this quality sign becomes less important once the start-up has demonstrated the ability to generate returns and profitability. In other words, we expect the difference between the benefits of equity capital for start-ups with high EO compared to those for start-ups with low EO to become much smaller when the start-up has passed the break-even point. Together, our expectations in terms of the costs and benefits of external equity imply that: Hypothesis 6. The relationship between EO and equity financing is moderated by whether or not a start-up has reached the breakeven point such that the relationship between EO and equity financing is relatively more positive before break-even versus after break-even. 2.3.3. A joint consideration of EO, environmental uncertainty and break-even Finally, we extend the previous contingency hypotheses into a more comprehensive multivariate combination allowing to better understand their complex interrelations (Dess et al., 1997; Titus and Anderson, 2018). In particular, we propose that the fit between a start-up’s level of EO and the characteristics of its environment is more important in early development stages than later on. As previously argued, start-ups with a conservative strategic posture are more inclined to follow traditional pecking order and fund their business activities (when needed) with external debt capital, as its costs and benefits (and thus the overall value) are more advantageous than those of external equity capital, especially when operating in less risky industries. We argued that this relationship becomes less pronounced if start-ups operate in riskier industries, in which they need to demonstrate their ability to adapt to external conditions in order to attract external debt capital at a lower cost (Berger, 1997/1998). In other words, when attracting external debt capital, the fit between a start-up’s strategic posture and the characteristics of its environment is important. We can argue that this fit between a start-up’s EO and industry-level risk is even more important for securing external debt capital in its early development stages than later on. Whereas debt providers base their decisions mainly on financial considerations such as profitability (e.g., Bruns and Fletcher, 2008; Mason and Stark, 2004), they face more uncertainty when funding start-ups with little or no financial track record. As such, in the period before a start-up is breakeven, debt financiers are more prone to rely on other start-up qualities in order E. Vaznyte and P. Andries Journal of Business Venturing 34 (2019) 439-458 445 to reduce the extent of information asymmetries associated with a start-up (Fletcher, 1995), among which we argue the fit between its entrepreneurial orientation and industry-level risk would be crucial. Nevertheless, these fit considerations would be less of a concern when more solid financial information becomes available, i.e. a start-up passes the break-even point. Thus, by jointly considering start-ups’ strategic, environmental and organizational characteristics, we hypothesize that: Hypothesis 7. The moderating effect of industry-level risk on the relationship between EO and debt financing is stronger before break-even versus after break-even. On the other hand, we proposed that start-ups with high levels of EO are more interested in funding their business activities with external equity capital, and that the value of external equity for entrepreneurial start-ups is especially pronounced in riskier industries (as opposed to less risky ones), since equity providers are known to cluster around the industries with high-return opportunities, and thus give priority to start-ups that pose a greater fit to this uncertain environment (MacMillan et al., 1985; Shepherd and Zacharakis, 1999). It can be argued that this strategy-environment fit will be even more important before start-ups’ break-even (when the issue of information asymmetries is the most acute), as then equity providers are most likely to consider other signals reflecting start-ups’ quality (Hsu, 2007; Ko and McKelvie, 2018). We also expect that the importance of fit will be lower once start-ups are break-even, as investors can then rely on start-ups’ financial information along other start-up qualities, which respectively reduces their investment uncertainty (Amit et al., 1998; Mason and Stark, 2004). We therefore hypothesize that: Hypothesis 8. The moderating effect of industry-level risk on the relationship between EO and equity financing is stronger before break-even versus after break-even. 3. Methods 3.1. Data Our study builds on three recent survey waves from the IAB/ZEW Start-Up Panel survey.1 In a given year, the IAB/ZEW Start-Up Panel selects and contacts a sample of legally independent German companies up to three years old, stratified by industry sector and year of foundation (‘first-time survey’). The following year, the ones that responded are contacted again (‘follow-up’), while a new random stratified sample of companies up to three years old is added (‘first-time survey’). This process is repeated year after year.2 Computer-aided telephone interviews are conducted with (one of) the start-up’s founder(s). This is important since these founders are determining the firm’s strategy, and are hence the first and principal source of information about the start-up. Respondents are promised confidentiality, which increases data quality. This confidentiality entails that no individual firm data can be published, but does not inhibit the research team to link data on a specific firm to firm-level information from other data sources. While each year key information on the start-ups, their founders, and their financing sources is collected in the survey, only the waves of 2014, 2015 and 2016 included questions on EO (see Fryges et al., 2010, for more information on the survey structure) and are used in the current study. In the first-time and follow-up survey of 2014, respectively 11325 and 5650 start-ups were contacted and inquired about their EO. The question module was also included in the first-time survey of 2015 and the first-time survey of 2016, targeting 10836 and 15602 start-ups respectively. In total 43413 start-ups were thus contacted and 9092 responded,3 resulting in an acceptable external survey response rate of 21% (e.g., Sarkar et al., 2001). As the question module on EO was not included in the 2015 and 2016 follow-up survey, there is no overlap between the start-ups in the three survey waves, and we use a pooled cross section of the data. From these 9092 respondents, we excluded 1938 start-ups from the 2014 follow-up survey that were more than three years old at the time of the survey. We did this in order to have a homogenous sample in terms of start-up age and limit the survivorship bias (e.g., Cassar, 2004). This resulted in a sample of 7154 different start-ups. After controlling for missing values for all the variables used in our analyses (i.e. listwise deletion), we obtained a final sample of 4456 start-ups. We compared this final sample with the total sample of 7154 start-ups and found only minor differences in the mean comparison tests. Following guidelines by Podsakoff et al. (2003), we also inspected for a common method bias, which does not appear to be a significant issue in our dataset. Finally, we combined this dataset with Creditreform data to obtain an industry-specific credit risk rating, as explained below. 1 This survey was established in 2008 by the Centre for European Economic Research (ZEW) together with the KfW Bankengruppe (the stateowned promotional bank in Germany) and Creditreform (the largest credit rating agency in Germany) and was known as the KfW/ZEW Start-Up Panel. Since 2015 the KfW Bankengruppe has been replaced by a new partner, the Research Institute of the Federal Employment Agency (IAB). The survey data are available to external researchers under certain conditions. 2 The IAB/ZEW Start-up Panel survey is similar to the Kauffman Firm Survey (KFS), which is broadly used by researchers analysing capital structure decisions by newly established firms in the U.S. (e.g., Cassar, 2014; Robb and Robinson, 2014). The IAB/ZEW Start-up Panel survey conducts follow-up interviews until start-ups reach an age of eight years, and unlike the KFS, adds a new sample of new firms (up to three years old) each year. 3 The firms that did not participate in the survey either could not been reached (e.g., wrong contact details, no founder was available to be interviewed, founder refused to participate), or did not fulfil the criteria of the IAB/ZEW Start-up Panel survey (e.g., were subsidiaries or branches). More details regarding the survey’s response rates are available at each year’s technical reports (https://www.gruendungspanel.de/). E. Vaznyte and P. Andries Journal of Business Venturing 34 (2019) 439-458 446 3.2. Variables 3.2.1. Dependent variables We analyse a start-up’s external financing decisions based on the proportion of its total financing needs in the year prior to the survey that was met by external debt and equity financing. In line with Fryges et al. (2015) and Giudici and Paleari (2000), we define a start-up’s total financing needs as its total expenditures on investments and operating costs. In particular, the survey design asked start-ups to indicate the amounts of (a) retained earnings, (b) founders’ own means (i.e. personal savings or business deposits), and (c) external funding that were used to cover all capital investments and operating costs in the year prior to the survey. We summed up these three amounts to calculate a start-up’s total financing needs. Furthermore, the questionnaire defined eight different sources of external funding, namely (1) equity capital (including private equity, venture capital, business angels, and subscription of shares by third parties), (2) overdraft credit, (3) loans from banks, (4) promotional loans from publicly financed funding programs, (5) financial means from the Federal Employment Office, (6) gifts and loans from family and friends, (7) mezzanine capital and (8) other funds. Start-up founders were asked to specify whether or not they use any of these respective funds, and if so, how much. The information for the first of these eight funding sources allows us to construct our dependent variable equity financing (ranging between 0 and 100%) as the amount of equity capital divided by the total financing needs. We construct a second dependent variable, debt financing (ranging between 0 and 100%), by adding up the amounts of (1) overdraft credits, (2) loans from banks and (3) promotional loans from publicly financed funding programs, and dividing this by the total financing needs. Both measures indicate funding activities for the calendar year prior to the survey year. The average start-up in our sample obtains approximately 61% of its total financing needs from retained earnings, 26% from founders’ own means, and the rest from external funding. In line with the findings in prior literature stating that start-ups rely heavily on external debt financing and on a lower degree on external equity financing (Cosh et al., 2009; Fryges et al., 2015; Robb and Robinson, 2014), we find that the average start-up in our sample obtains (54% of its external funding or) 7% of its total financing needs from debt financing, and (11% of its external funding or) 2% of its total financing needs from equity financing, and the remaining 4% of its total financing needs from other sources of external financing (e.g., family and friends). In essence, the descriptive statistics reveal the prevalence of a pecking order among German start-ups, especially among the ones characterized by a lower level of entrepreneurial orientation. 3.2.2. Independent variables For measuring a start-up’s entrepreneurial orientation (EO), we endorse the notion by Miller (1983) and Covin and Slevin (1989) that EO is reflected by the shared variance between innovativeness, risk-taking, and proactiveness. We employ a questionnaire built on the scale proposed by Covin and Slevin (1989), which respectively inquires about the start-up’s fundamental strategic orientation since its inception. We calculate entrepreneurial orientation as the average score over the six different items (i.e. two questions for innovativeness, proactiveness, and risk-taking respectively), varying from 1 (conservative) to 5 (entrepreneurial). Although the conceptualization and measure of a unitary EO construct is well recognized in entrepreneurship research (Covin and Lumpkin, 2011; Covin and Wales, 2012, 2019), we validate the reliability of our scale. We conduct (1) an exploratory factor analysis, which shows that all questions regarding EO load onto three latent factors (χ2=2517.26, p < 0.01), (2) a confirmatory factor analysis that suggests a good model fit (χ2=135.94, p < 0.01), and finally we measure (3) Cronbach's alpha, which reveals an acceptable scale reliability (α=0.62) (e.g., Wiklund and Shepherd, 2005). The second independent variable represents the environmental uncertainty or industry-level risk at the time of a start-up's foundation and is measured by the average degree of credit risk for all start-up companies founded in a specific industry in that specific year (see Dencker and Gruber, 2015). According to Dencker and Gruber, it provides a good approximation of the riskiness of the entrepreneurial opportunity and reflects possibilities or constraints faced by start-up founders when exploiting opportunities in particular settings. We constructed this variable using a database of Creditreform - the largest credit rating agency in Germany - that allowed us to assess the average credit risk rating of each industry based on a five-digit level classification. Credit risk rating varies from 100 to 600, where an excellent or good (100-208) credit risk rating implies a non-existent or very low probability of default, and thus reflects a low industry-level risk; a poor credit risk rating (356-600) implies a very high probability of default, and hence a very high industry-level risk. Each start-up received the industry-level credit risk value of the five-digit level sector in which it is active (NACE Rev. 2). Lastly, we capture a start-up's organizational characteristic by measuring whether or not the firm has surpassed the break-even point. Start-up founders were asked to indicate whether their firm realized (a) profits, (b) losses or (c) an even amount of revenues and costs (zero outcome) before taxes in the calendar year preceding the survey year. Based on their answers, we assigned start-ups into two groups: (1) start-ups that were profitable (55% of sample), and (2) start-ups that incurred a loss or a zero outcome (45% of sample).4 3.2.3. Control variables In line with previous studies, we employ a number of firm-level and entrepreneur-level variables that are important determinants 4 We ran robustness checks where we made the distinction between (1) start-ups that were profitable or incurred a zero outcome, and (2) start-ups that incurred a loss. Our main results remain. However, we believe it is more appropriate to take the start-ups with a zero outcome together with the ones that make losses, as a "project that simply breaks even on an accounting basis gives you your money back but does not cover the opportunity cost of the capital tied up in the project. A project that breaks even in accounting terms will surely have a negative NPV" (Brealey et al. 2001: 475). E. Vaznyte and P. Andries Journal of Business Venturing 34 (2019) 439-458 447 of a start-up's financing decisions (e.g., Cassar, 2004; Cosh et al., 2009; Hall, 2010; Huyghebaert and Van de Gucht, 2007; Eckhardt et al., 2006; Mina et al., 2013). As for firm-level characteristics, we control for start-up age, size and level of sales since these have been shown to be positively related to the amount of external financing attracted by the start-up. We measure start-up age by the number of years from its foundation until the reference year, size by the number of employees at the time of business foundation, and sales by the total sales in euros generated from business activities in the reference year, i.e. the year before the survey. Furthermore, a start-up's ability to attract funds depends on its asset tangibility and level on R&D spending. That is, firms containing little tangible assets and investing heavily in R&D are associated with higher information asymmetries, and hence are more likely to be financed by external equity than debt capital (e.g., Hall, 2010). As the survey design does not include any information on start-ups' capital stocks or financial assets, we follow Fryges et al. (2015) and use the ratio of expenditure on capital investments over a firm's total financing needs in the reference year as an indicator of tangibility. We also control for R&D spending per employee in the reference year as an indication of firm level of innovativeness. Whereas higher R&D expenses are associated with higher information asymmetries, number of patents can serve as an external certification mechanism reducing agency risks (Mina et al., 2013) or even as a collateral (Bellavitis et al., 2017), and can thus increase the probability of receiving external funding. In this respect, we control for a number of patents at the time of a start-up's foundation. Additionally, we control for three binary variables: business plan (Eckhardt et al., 2006; Mason and Stark, 2004), collateral (Bruns and Fletcher, 2008), and ownership (Baum and Silverman, 2004) as they have been shown to affect access to external funds. The variable business plan indicates whether a founder had prepared a business plan before a start-up's foundation, collateral specifies whether a founder used any private resources for the business establishment, and finally, ownership indicates whether in the reference year a start-up's shares are held by any other company (i.e. whether it holds 1-75% of a start-up's shares). A second set of control variables reflects founders' characteristics. Unlike established firms, start-ups are associated with greater information asymmetries, thus start-ups' human capital and social networks play an important role as additional signalling mechanisms when securing external debt and equity funds (Baum and Silverman, 2004; Cassar, 2004; Ko and McKelvie, 2018). Specifically, prior studies have indicated that start-up founders' experience and education are associated with the quality and reliability of human capital, and may be positively perceived by external fund providers (e.g., Bruns and Fletcher, 2008; Colombo and Grilli, 2007; Eckhardt et al., 2006; Hsu, 2007; Ko and McKelvie, 2018). Therefore, we include the continuous variable industry experience measured in years. If more than one entrepreneur was included in start-up foundation, we use information on the founder having the most industry experience. Next, we include the binary variable entrepreneurial experience that is equal to 1 if one of the founders has established an enterprise before, and zero otherwise. Additionally, we include the binary variable education indicating whether or not any of the founders hold a university degree. Lastly, we account for founding team size, as larger teams may carry greater human and social capital (Baum and Silverman, 2004; Hsu, 2007). Since several control variables are highly skewed, namely age, size, sales, R&D spending, patents and industry experience, we include them as log-transformations to the empirical analysis (for variables with zero values we add the smallest positive value before transforming). Finally, in order to account for specific industry and unobserved year effects, we include a set of industry and time dummy variables. 3.3. Estimation approach Since only 20% of the start-ups in our sample use debt financing, and only 5% use equity financing in order to fulfil their total financial needs, we have to control for the excess of zero values that represent the firms not using these types of financing. Thus, the nature of both dependent variables suggests to use a corner solution model, known as the Tobit model. It takes into account the excess of zero values, and hence is suitable for the partially continuous dependent variables debt financing and equity financing (Wooldridge, 2002). Important to note, this model accounts both for a start-up's decision to use debt (equity) financing and for the proportion of debt (equity) financing used. 4. Results 4.1. Descriptive statistics Table 1 presents the main descriptive statistics and a correlation matrix of all variables, which does not indicate any issue of multicollinearity (VIF=1.64). The correlation matrix reveals a positive, but insignificant correlation between EO and debt financing, and a positive correlation between EO and equity financing significant at the 1% level. These descriptive statistics provide initial support for the second hypothesis, but appear to go against the first one. Overall, they suggest that EO may play a more important role for securing equity financing than debt financing. 4.2. Multivariate analysis Table 2 provides the results of the Tobit model regarding debt financing and equity financing respectively. In contrast to our first hypothesis, the results indicate that EO, ceteris paribus, has a positive but insignificant effect on the share of debt financing, thus rejecting Hypothesis 1 (Model 2: β=1.923, p > 0.10). In line with Hypothesis 2, the positive coefficient on EO indicates that more entrepreneurially oriented start-ups attract a higher share of equity financing (Model 5: β=13.909, p < 0.01). Specifically, results presented in Table 3, we see that for low levels of industry-level risk the relationship between EO and debt financing is negative as wewere expecting in our Hypothesis 1. For instance, if the industry-level risk equals to 275, start-ups' increase in EO by 0.1 wouldrepresent a decrease in a share of debt capital by 1.28% (please note, the marginal effect is insignificant). However, if the riskincreases, and for example equals to 288, then the increase in start-ups' EO would represent an increase in a share of debt capital by6.33%. Our results, therefore, suggest that debt providers are positively assessing the EO of their lenders, especially when makingfinancing decisions in riskier industries. In model 6 (Table 2), we examine whether the industry-level risk is a moderating factor in therelationship between EO and equity financing. Contrary to our Hypothesis 4, results indicate a negative but insignificant interactioneffect (Model 6: β=−0.314, p > 0.10). As such, our results suggest that, regardless of industry-level risk, start-ups with high levels ofEO can secure more equity financing. Opposite to debt providers, equity providers are known to select start-ups only from the fewindustries in which they can leverage their specialized expertise (Manigart and Struyf, 1997; Winton and Yerramilli, 2008), and ourresults suggest that they apply the same selection criteria when evaluating start-ups in more or less risky industries.In order to test the remaining hypotheses, we classify start-ups into two groups depending on whether or not they reached thebreak-even point.6 As noted by Hoetker (2007), this approach provides a better approximation than a simple inclusion of an interactionterm with a binary variable since we cannot assure that unobserved variation is the same across the comparative groups innon-linear (in this case, Tobit) models. The respective results are presented in Table 4.With respect to debt financing, we cannot support the moderating role of start-ups’ break-even, and have to reject Hypothesis 5(Model 7 vs Model 9: χ2EO=0.42, p > 0.10). However, when we jointly investigate start-ups’ break-even point together with their EOand industry-level risk, we do find compelling evidence supporting its moderating role, and thus respectively Hypothesis 7. The resultingChow test statistics indicate that indeed the conditional effect of EO on debt financing is stronger for start-ups that have not yetreached the break-even point than for start-ups that are beyond it (Model 8 vs Model 10: χ2EO⁎Industry-level risk=4.65, p < 0.05). Specifically,the negative relationship between EO and debt financing is more pronounced in less uncertain environments (e.g., -2SDbelow the mean) when a start-up is not break-even, but less so when the break-even point is surpassed. At the same time, thisrelationship becomes more positive when the industry-level risk is high (e.g., +2SD above the mean) and the start-up is not break-even,however less so when the break-even point is surpassed (Fig. 1).With respect to equity financing, we do find support for Hypothesis 6 stating that start-ups' high EO is more important in securingexternal equity financing when start-ups are not break-even, but less thereafter (Model 11 vs Model 13: χ2EO=5.64, p < 0.05). Yet,when investigating start-ups' strategic, environmental and organizational characteristics simultaneously, we cannot find evidencesupporting Hypothesis 8 (Model 12 vs Model 14: EOχ2=0.51, p > 0.10). The respective interaction plot presented in Fig. 2 clearlyillustrates the strong effect of start-ups’ break-even point in securing equity financing, however at the same time it shows that the slopeconcerning EO⁎Industry-level risk curve does not change with increasing level of industry risk (i.e., the change is minimal, and similarto the size effects presented in Table 3).4.3. Robustness checksWe performed several robustness checks to validate our findings. First, we conducted a two-step Tobit model with endogenousvariable approach (Wooldridge, 2002, Ch. 16.6.2) in order to address potential endogeneity problems (Miller, 2011). As instrumentalvariables we used (1) founders’ age (Cruz and Nordqvist, 2012) and (2) whether a founder established a new venture in order tocommercialize a specific idea (versus reasons including work independence, income levels, etc.), which reflects a founder’s motivationand promotes EO (Covin and Slevin, 1991). We observed that both instrumental variables have a strong effect on EO (in areduced form equation) and result in reliable F-test statistics (F-stat=39.94) (Steiger and Stock, 1997), and thus offer a good fit forfurther investigating endogeneity issues. The results of this two-step approach were fully in line with our key findings. Second, we exploited follow-up survey data (available only for start-ups interviewed in 2014 and 2015) and introduced a one year time lagbetween start-ups’ EO and their subsequent debt and equity financing decisions, which lent support for a causal interpretation of theproposed relationships. Additionally, in order to reduce potential unobserved heterogeneity, we included seven variables representingwhether or not start-ups received financial support from various governmental agencies, as this may reduce informationasymmetries and have a certification effect when securing additional external capital (Mina et al., 2013). The inclusion of theseadditional control variables (in our main model and in our two-step model with instrumental variables) confirmed our main findings.Since our main empirical model investigates the financing start-ups obtained conditional on that they sought for it (similar toVanacker and Manigart, 2010), we also inspected for a potential selection bias. We employed Heckman’s (1979) two-step selectionapproach, with the first step indicating a start-up’s likelihood to seek for external financing, and the second step indicating theamount of external financing (debt or equity) obtained. As an exclusion restriction we use information on whether or not the founderwas unemployed before founding the start-up. Even after controlling for the selection effects (which appeared to be significant fordebt financing), our initial findings are confirmed.Finally, we validated our key findings by imposing certain sample restrictions, using different EO specifications (i.e. a fivedimensional construct; Lumpkin and Dess, 1996) and controlling for additional variables (e.g., including the break-even dummy variable instead of splitting the sample). Overall, different estimation methods as well as model specifications present similar results, and thus support the robustness of our key findings (all robustness checks are available from the authors upon request). 5. Discussion 5.1. Contributions and managerial implications Our study offers several key contributions to the literature. Firstly, it contributes to the emerging debate regarding the applicability of corporate finance theories, and Pecking Order Theory in particular, for explaining financing decisions of newly established firms. Whereas it was already known that the financing decisions of start-ups cannot be solely explained by corporate finance theories and several studies have suggested to look at firm strategy as a potential determinant (e.g., Chaganti et al., 1996; Fraser et al., 2015; McMahon and Stanger, 1995), our study is the first to our knowledge to conceptually discuss and empirically test the role of EO as an explanation for start-ups’ external financing decisions. We extend Pecking Order Theory by proposing that (a) a start-up’s choice of external debt and external equity does not only depend on their costs but also on their benefits, (b) the costs and benefits of these two financing forms are differently affected by the start-up’s entrepreneurial orientation, and (c) this effect of entrepreneurial orientation is contingent on environmental and organizational characteristics. We empirically show that start-ups are more likely to follow the traditional financing hierarchy as put forward by Pecking Order Theory when their EO is low, but less so when their EO is high. As such, we provide robust evidence concerning the importance of entrepreneurial strategy in explaining start-ups’ financing decisions. Moreover, we address the importance of contingency effects, which have been neglected when investigating the applicability of Pecking Order Theory within a start-up context. Specifically, we show that the industry-level risk and start-ups’ development stage interact with EO in determining the costs and benefits of external debt and equity, and thus are important boundary conditions modifying the relationship between EO and start-ups’ financing decisions. In particular, we demonstrate that the fit between start-ups’ EO and environment is more important when securing external (debt) funds in early stages of start-ups’ development than later on when additional financial information becomes available. In this way, we also contribute to the previous call in the literature to test alternative predictor variables of financing decisions (Hanssens et al., 2016; Huyghebaert and Van de Gucht, 2007; Mueller, 2008) and also help to explain heterogeneity regarding the financing practices of new entrepreneurial firms. Secondly, we advance the entrepreneurial strategy literature by showing that EO is not only instrumental to explain a firm’s financial performance, but also its financing decisions. Besides responding to the previous request in the literature to investigate the relationship of EO with other outcome variables than financial performance (Lumpkin and Dess, 1996; Wales et al., 2013; Wales, 2016), we also advance financial decisions as an intermediary mechanisms through which EO may influence performance. This extends our understanding of how and why EO affects firm performance – a research area currently lacking solid theoretical and empirical evidence (Wales et al., 2013, 2015). Moreover, we extend EO research by conceptually arguing for and empirically demonstrating the advantages of a contingency approach, and in particular a joint investigation of start-ups’ strategic, environmental and organizational characteristics (e.g., Covin and Lumpkin, 2011; Miller, 2011; Titus and Anderson, 2018). Our study is the first to demonstrate that a start-up’s EO, as well as the EO-environment fit, are particularly relevant in the early stages of its lifecycle as initially proposed by Lumpkin and Dess (1996). As we measure the development stage based on the break-even assumption, we also add to a call by Wales (2016) on testing the boundary conditions that can serve for the effective manifestation of EO. This has important implications for future studies which should be careful to take firms’ development stage into account when investigating the effect of EO. Furthermore, by investigating the effect of EO with respect to different types of external financial resources, we provide empirical evidence supporting the argument raised by Wales (2016: 9) that “greater EO is not universally more beneficial than a more conservative strategic orientation.” Lastly, our study offers valuable practical implications for start-up founders and fund providers. As for start-up founders, our results show which type of external financing best fits their level of EO and under which conditions (in particular, the industry-level risk and the development stage of their start-up). As it happens quite often that start-ups are unable to secure the type of financing they would prefer (Cosh et al., 2009), our findings may allow them to better understand, target, and obtain proper sources of financing. For instance, if a start-up is highly entrepreneurial, then the best option might be to look for equity rather than debt financing. However, if that same start-up is active in a risky industry, then it could also consider debt financing. In addition, our study points out that start-ups need to think about this fit very early on, as (a) the strategy-financing fit is particularly important before break-even, and (b) initial financing practices are known to be the foundation for future financing decisions (Eckhardt et al., 2006; Hanssens et al., 2016), which may even help to reduce the rate of young ventures’ failures (Chaganti et al., 1996; Matthews et al., 1994; Zacharakis and Meyer, 1998). Furthermore, our findings also point to the usefulness of writing a business plan. Recently, both academic work (e.g., Sarasvathy, 2001; Ries, 2011) and popular literature (e.g., Guttman, (2015): “Don’t Write Business Plans”, Velho (2017): “Bin the business plan”) have started to argue that business planning may inhibit rapidly evolving start-up activities in adapting to volatile environments. Our findings suggest that the development of a business plan is equally important for low as well as high EO start-ups, and for risky as well as stable environments. Writing a business plan forces entrepreneurs to thoroughly think about their strategy, and our research demonstrates that this is crucial for obtaining the proper type of external financing. As for debt and equity providers, our study suggests they should take information on start-ups’ strategy into account while evaluating funding proposals. Specifically, business plans may serve as a primary source of information not only on objective firm characteristics, but also on a firm’s level of EO. This additional information may reduce information asymmetries surrounding the new business, and help E. Vaznyte and P. Andries Journal of Business Venturing 34 (2019) 439-458 454 to advance the funding decision process. 5.2. Limitations and future research directions Even though we feel confident about the findings presented in this study, there are several limitations that need to be addressed in future research. First, although we made every effort to control for it, we cannot fully rule out the issue of endogeneity. In our robustness checks, we controlled for potentially endogenous explanatory variables of start-ups’ EO, introduced additional variables potentially reducing space for unobserved heterogeneity, and addressed potential causality issues by introducing time lags between EO and start-ups’ financing decisions. The results indicated that there are no problems of endogeneity, and confirm the robustness of our main findings. Nevertheless, future research may consider alternative approaches including panel data methods in order to address the causal direction between EO and start-ups’ financing decisions. Second, in our robustness checks, we also made an attempt to control for a selection effect by constructing a sample of start-ups seeking potential financing. The results of this verification were in line with our key findings, yet they could not fully reject the presence of a selection bias (at least for debt financing). Therefore, we encourage future research to investigate these financing decisions by taking into account the selection effect between start-ups seeking and obtaining external financing. Third, unfortunately we do not know whether the start-ups in our sample received any external financing before business foundation, whereas this could have an impact on their further financing decisions (Eckhardt et al., 2006). Although we try to overcome this limitation by controlling for several financial support mechanisms associated with the certification effect (Mina et al., 2013), and are able to reproduce our main findings, future research may account the financing acquired prior to start-up foundation. Fourth, we do not differentiate between different types of external debt and equity capital providers. As it is generally accepted that venture capitalists, for example, differ from business angels with respect to contractual agreements, extent of additional support, etc. (Mason and Stark, 2004; Vanacker et al., 2013), and commercial banks differ from non-bank creditors with respect to financing costs or the way they cope with asymmetric information (Berger and Udell, 1998; Carey et al., 1998), future research could provide additional insights into the fit between EO and specific types of investors and lenders. Moreover, more precision could be added when considering intermediary financing sources, such as convertible debt or preferred stock. In this study we could only make a distinction between formal debt and equity capital without having possibility to discern a class of hybrid securities. According to Pecking Order Theory, this source of funding is considered as an intermediary external financing source typically associated with higher issues of information asymmetries than debt securities, and yet less than equity capital (Myers, 1984, 2001). However, these hybrid securities may carry higher transaction costs (Fama and French, 2005), and thus may be approachable by both more conservatively and more entrepreneurially oriented start-ups. Therefore, future studies could investigate to what extent start-ups’ EO can explain the costs and benefits of hybrid securities. Additionally, while our study draws upon and respectively extends the pecking order model, future studies could investigate alternative models, such as for example Static Trade-Off. Static Trade-Off Theory, which is often regarded as a competing view to Pecking Order Theory, proposes that firms search for an optimal capital structure (i.e. debt-to-value ratio), and borrow up to the point where the marginal benefits of debt (e.g., tax savings) are just offset by its marginal costs (e.g., bankruptcy costs) (Myers, 1984, 2001). Although Static Trade-Off Theory has received little empirical support in the context of start-ups (see Atherton, 2012; Landström, 2017), its predictive value may be improved when taking strategic posture, as well as environmental and organizational characteristics into account. In particular, it would be worthwhile investigating whether and to which extent start-ups’ EO, industrylevel risk and development stage jointly affect their debt-to-value ratio. Lastly, there are several studies arguing that entrepreneurial orientation may change over time, for example, when significant changes in the firm’s management team occur (Lyon et al., 2000), when team members become overly passive or overly risk-taking (Lumpkin and Dess, 1996) or when firm simply needs to adjust its strategy due to growth and changes associated with it (Wales et al., 2011). Future longitudinal studies could explore how changes in EO affect the financing decisions of young firms. 6. Conclusions This study intended to dispel confusion regarding the applicability of Pecking Order Theory, in view of the contradictory empirical findings on start-ups’ financial decision making, by incorporating both an entrepreneurial strategy and contingency perspective. In particular, it aimed to improve our understanding of the role of a start-up’s entrepreneurial orientation for its financial decision making and the conditions under which this entrepreneurial orientation matters. Empirical analysis of 4456 German startups reveals that entrepreneurial orientation is an important determinant explaining start-ups’ external financing decisions, and shows how this relationship is moderated by industry-level risk and by a start-up’s development stage. This study advances the literature on entrepreneurial finance by extending Pecking Order Theory with insights on entrepreneurial orientation and its contingencies, and adds insight in the relationship between entrepreneurial orientation and firm performance. It also provides valuable practical implications for start-up founders and external financiers. Acknowledgments We greatly acknowledge the Editor Justin W. Webb and two anonymous reviewers for significantly contributing to the improvement of this work. Special thanks to prof. Tom Vanacker and prof. Sophie Manigart for valuable comments on earlier version of this manuscript. Also, we sincerely thank Sandra Gottschalk and others from the Centre for European Economic Research (ZEW) in E. Vaznyte and P. Andries Journal of Business Venturing 34 (2019) 439-458 455 Mannheim for providing data access. We are also grateful for suggestions from participants attending the 78th Annual Meeting of the Academy of Management 2018 (Chicago), the Entrepreneurial Finance Conference 2018 (Italy), the International ZEW Conference on the Dynamics of Entrepreneurship 2018 (Germany), the DRUID Academy 2018 (Denmark) and RENT: Research in Entrepreneurship and Small Business 2017 (Sweden)-  Business Entrepreneurship MBA-FP 6006 Share QuestionEmailCopy link Comments (0)