1. Introduction

Over the years, entrepreneurship has been considered a key element in the development of countries. Classic authors (^{Schumpeter, 1934}; ^{Leibenstein, 1968}) through to contemporary authors (^{Zsuzsanna & Herman, 2012}; ^{Toma, Grigore, & Marinescu, 2014}; ^{Liñán & Fernandez-Serrano, 2014}) have understood the importance of entrepreneurship in well-being through employment creation, income growth and innovation.

A special type of international entrepreneurship is known as born global firms, which are defined as firms that present an accelerated process of internationalization, meaning they initiate their commercial activities immediately in international markets or do so in a period relatively close to their foundation (^{McDougall, Shane & Oviatt, 1994}; ^{Coviello & Munro, 1995}; ^{Autio, Sapienza & Almeida, 2000}; ^{Knight & Cavusgil, 2004}). These firms present different characteristics to the gradual internationalization model, or Uppsala model, which asserts that firms develop an incremental internationalization process (^{Johanson & Vahlne, 1977}). Thus, in several academic studies a growth in global born ventures has been indi-cated, which can be explained by factors linked to more customized products or services, advances in transportation, greater market information, sturdier international networks or even more and greater financing options (^{Knight & Cavusgil, 1996}; ^{Madsen & Servais, 1997}; ^{Oviatt & McDougall, 1997}; ^{Autio, 2005}; ^{Smolarski & Kut, 2011}).

However, not only has the speed of the internationalization of firms been a subject of study for born globals. Choosing the destination economies for exports has also been a subject of scientific interest. Some studies have emphasized the role that geographical and psychic distances can play, particularly in terms of how any cultural differences existing between the country in which a born global firm is founded and the countries in which it decides to export may affect the internationalization process of new firms (^{Cancino, 2014}; ^{Ojala, 2015}; ^{Bhowmick, 2018}; ^{Magnani, Zucchella & Floriani, 2018}). Nevertheless, the previous research does not show the same results depending on the economy analysed. In the case of emerging countries, some studies show that there are differences in the characteristics and distances of customer countries from born globals firms compared to developed countries (^{Ciravegna, Lopez & Kundu, 2014}). This suggests that although the phenomenon of born global firms is prevalent throughout the world, it would not be homogeneous among different types of countries and could change according to the economic, social and cultural context (^{Bruton, Ahlstrom, & Obloj, 2007}). Thus, it is not clear which factors are related to existance and success of this type of entrepreneurships. Consequently, understanding the behaviour of born global firms can be very relevant in terms of public policies, considering the impact they can have on the development potential of countries.

As a result of the significant positive influence that a greater number of born global firms have on an economy, which are not only active internationally, but also generate a series of profit spills in the domestic market (^{Cancino, 2014}), a series of public productive support programs have been developed in each country that seek to strengthen the founding and subsequent development of firms with international activities. The success of these programs and the effect they have on boosting the development of born globals firms is not always evaluated.

This paper aims to evaluate the factors that determine the success of new Chilean firms supported by CORFO¹’s productive promotion programs in Chile in their venture into international markets. For this purpose, a Probit regression model was conducted in two stages, which seeks to show the factors and characteristics that impact on the probability of export success of Chilean fims supported by CORFO. Given the fact that, in general, entrepreneurs supported by CORFO value internationalization as a growth strategy even before being approached by CORFO, the two-stage method used in this work intends to correct the possible self-selection bias existing in the sample (^{Heckman, 1976}).

While there is a wide literature examining the phenomenon of born global firms and the way public programs that support them work, there is less evidence that considers the inherent possible selection bias that exists when generating models that explain the characteristics and factors that determine a firm’s internationalization process. This bias would come from the fact that when one tries to measure a born global firm through variables such as the level of international sales or participation in foreign markets, it is not considered that these firms made a decision beforehand, which is to decide to participate in the international market. This previous decision, which is not random, generates a possible self-selection bias in the sample, that is, firms that have already decided to participate internationally. This must be corrected to reach more reliable estimates. The two-stage Probit method that we performed in this study is quite innovative and aims to give value to research.

According to the first Probit regression model results, it is demonstrated that there is indeed a selection bias in the sample. In the case of the sampled firms, all are defined as firms whose international focus is part of the core values. Nevertheless, the distances or differences with their export destination markets is not consistent. The findings suggest that new firms selling over a long psychic distance would be more likely to succeed in international markets than their peers exporting to nearby countries in the region. Choosing Asian, North American, and European markets is critical to growth and export success. In this sense, being global and not regional (not Latin America countries in this case) is essential to achieve export success. Considering the importance of psychic distances in explaining the success of the internationalization of new companies, we also developed a second Probit regression model in two stages, where the factors that influence the probability of deciding whether or not to sell at a high psychic distance are determined. The results of this second model show that, although there does not seem to be a selection bias that could affect the estimates, the fact of having received private capital in the domestic market has a negative impact on the decision to sell long distance, while having more experience in previous ventures has a positive and significant impact on this probability. Namely, raising capital in Chile fosters the growth of commercial activities in the region’s nearby markets over markets that are psychically distant. It would appear that the domestic investor prefers higher short term returns, although these may be less than the adventure of marketing products in distant markets, but where the probabilities of risk may be greater. In contrast, the past experience of entrepreneurs positively influences the choice of psychically distant markets, where the experience of creating companies, with successes or failures, is a fundamental resource for the international strategy of the born globals founders.

The findings described could be very important in supporting policy-making in Chile, or in countries of the region with a similar level of development, where the national strategy is oriented towards the large markets of Asia, North America and Europe. It is precisely these economies that require important products from the region, and the participation of new companies is essential.

The work is structured as follows. After this introduction, section two presents the literature review and research hypotheses. Section three presents the research methodology. Section four presents the results.

Section five presents the discussions of the study. A final section presents the findings of the study.

2. Literature Review and Hypothesis

There is ample literature related to born globals companies and the internationalization process that underlies these firms. In this respect, there are studies that have tried to explain the behaviour and determinants that define a born global and differentiate it from a company that performs a gradual internationalization process, both in emerging and developed economies (^{Sapienza, Autio, George, & Zahra, 2006}; ^{Persinger, Civi & Vostina, 2007}; ^{Cancino, 2014}; ^{Cancino & Coronado, 2014}; ^{Cumming, Fischer & Peridis, 2014}). More recent research has emphasized the importance of geographical and psychic distances in choosing the markets to participate in and their impact on the internationalization of SMEs (^{Håkanson & Ambos, 2010}; ^{Ojala, 2015}; ^{Magnani, Zucchella & Floriani, 2018}; ^{Bhowmick, 2019}).

Smolarski and Kut (²⁰¹¹) conducted an analysis of small and medium-sized enterprises with fewer than 100 employees. To do so, they considered a sample of 186 Swedish firms that received financing or some form of private capital. According to an Ordered Probit regression model, an attempt was made to explain the firms’ performance and internationalization level, using the sales rate and export ratio as proxies, respectively. The findings of this research show that there would be a significant relationship between the type of financing received, which could be incremental, lump sum or syndicated and the performance and level of internationalization of a firm. This relationship would be positive when firms acquire funds incrementally and syndicated on a separate basis and negative when the firm acquires both at the same time.

Meanwhile, Cancino and Coronado (²⁰¹⁴) used a logistic regression model on a sample of 115 small and medium Chilean companies that decided to internationalize, in order to explain their behaviour and understand what variables define that these firms can be considered born globals. The findings of this study show, among other things, that networks with international capital and the cultural distance from countries that receive exports have a positive impact on the likelihood a firm could be considered a born global. Along the same lines, Cancino (²⁰¹⁴) conducted an analysis of 112 Chilean SMEs with regular exports. The results of this study are similar, finding a positive influence from national and international networks and the psychic distance with respect to the probability of behaving as a born global, and also the absence of companies participating in the high technology sector within these types of companies. The latter is a particular case of international born globals.

In the same context of emerging economies, Dib, da Rocha and da Silva (²⁰¹⁰) seek to examine the phenomenon of born globals in Brazil. To do so, they conducted a logistic regression model on a sample of 79 software firms, which were divided between companies with born global characteristics and firms with a gradual process of internationalization. The results found in this study show that some variables such as innovation and customer orientation may have a positive impact on the type of process that the firm undergoes to internationalize. In addition, in a qualitative analysis Persinger, Civi and Vostina (²⁰⁰⁷) studied the reasons, characteristics and success factors that may be related to born global in emerging economies. These factors may be associated with variables such as the entrepreneurs’ personality and previous international experience, risk propensity, among others. With regard to previous experience, studies such as Hollender, Zapkau and Schwens (²⁰¹⁷) show consistent results with the above. The authors use a hierarchical linear regression to measure the international performance of firms, finding a positive relationship between previous international experience and the firm’s performance abroad. Similarly, Colombo and Grilli (²⁰⁰⁵) found a positive relationship between prior entrepreneurial experiences and firm growth for new technology-base firms. The above is consistent with other studies which propose firms managed by prior experienced entrepreneurs would be more probable to persue an early and proactive mode of internationalization (^{Sapienza, Autio, George, & Zahra, 2006}; ^{Federico, Kantis, Rialp, & Rialp, 2009}).

Other writers, in turn, have presented extensions to the born global phenomenon. One such case is the one that literature calls born again global (^{Bell, McNaughton & Young, 2001}; ^{Bell et al., 2003}; ^{Jantune, Nummela, Puumalainen & Saarenketo, 2008}; ^{Schueffel, Baldegger & Amann, 2014}). Bell, McNaughton and Young (²⁰⁰¹) define it as firms that are well established and have had a previous focus on their domestic market, but suddenly have a rapid internationalization process. Qualitative studies such as Bell et al. (²⁰⁰³) and Schueffel, Baldegger and Amann (²⁰¹⁴) point out that events that provide firms with both financial and human resources could lead to a company becoming a born again global, as well as having a reduced local market and identifying opportunities in other markets. Bell et al. (²⁰⁰³) also indicate that the main competitive advantage of this type of company would be related to a greater knowledge base.

On the other hand, in a quantitative analysis, Jantune et al. (²⁰⁰⁸) found differences in the factors that would explain the international performance of born globals, born-again globals and traditional companies. The authors use a linear regression model for each of the three groups mentioned above. Their results suggest that the learning orientation would be related to greater performance of born again globals, while there would be a moderate positive relationship between business orientation and the born again globals performance.

Other recent research has analysed the implications of geographical, cultural and psychic distances on the phenomenon of international business. Håkanson and Ambos (²⁰¹⁰) conducted a background study on psychic distance. To perform their analysis, the authors used a sample of 1,414 observations from 25 countries, which consider both developed and emerging countries. The results of this study show that most of the variability in the perception of psychic distance would be explained by geographic distance, to the detriment of cultural distance, which does not seem to have a strong relationship. Consequently, the authors recommend giving greater emphasis to greographic distance when attempting to model situations in which psychic distance may play an important role.

Magnani, Zucchella and Floriani (²⁰¹⁸) conducted a comparative analysis to complement previous studies related to the internationalisation process of companies. In their research, the authors compared the experience of three Brazilian firms that have internationalized in Italy and three Italian firms that have internationalized in Brazil. Some of their conclusions suggest that a key factor for companies in the decision to access a foreign market would be the firm’s specific strategic objectives. To this extent, if the firm considers that entering a particular market is part of a strategic decision, that objective may even counteract the perception of high psychic distance. Additionally, psychic distance should not necessarily be seen as a barrier to international markets, but rather as an appealing factor as it represents an opportunity and fits the firm’s strategy, which could potentially play an important role in the internationalization process (^{Bhowmick, 2019}).

3. Data and Research Methodology

4. Results

In order to understand the variables which could be most important for ventures to succeed in international markets, a two-stage probit regression model was estimated. The results are shown in Table 5 and 6. In addition, some performance measures of the model can be seen in Appendix A and B.

First of all, from the defined selection model, it can be observed that the coefficient of the Rho parameter was significant. This suggests that there is indeed a selection bias in the sample and that the estimators obtained from the model presented in Table 6 would be more reliable compared to those obtained using a one-stage estimation method. It can also be noted that the selection variable Program was significant for most of the categories. This leads us to understand that there would certainly be important differences in the probability of internationalizing depending on the type of program to which the enterprise belongs. In the case of SSAF Innovacion there would not be a significant difference, probably because the ventures of this program have a high degree of innovation as well as those of the category Others (Seed Capital and StartUp Chile).

Based on the model presented in Table 6, four variables are statistically significant: Workers, Accum domestic sales, Psycological distance and International fundraising. Therefore, the results suggest that firms with more workers would be more likely to succeed. This occurs once the threshold of ten workers is exceeded. That is to say, larger enterprises would have a better chance of being successful in the international market. On the other hand, ventures that have higher accumulated domestic sales would have a better chance of success. This finding suggests that, in the case of Chilean ventures supported by CORFO, their previous performance in the domestic market would be an important condition in explaining their possible success in venturing internationally. As for psycological distance, the relationship is positive, indicating that firms venturing to export at a high piscological distance would have a greater chance of international success. This result confirms Hypotheses 1, where either because of the appeal of the international market or as result of a strategic decision, entrepreneurships whose sales come at least in a small proportion from psycological distant markets, they have a higher probability of success.

As for the second model, in Table 7, concerning the selection model, we can see that there is no selection bias that significantly affects the results of the Probit model. Additionally, two variables were significant in the subsequent estimation: Previous entrepreneurship and Private fundraising. These results suggest that, on the one hand, a variable that could positively explain the decision to sell at a high psychic distance would be previous entrepreneurship, that is, the previous experience that entrepreneurs have in this type of activity would be important when venturing to countries that are psychically distant from Chile. It is consistent with Hypotheses 2, which claims that less risk averse and more experienced entrepreneurs would have a higher probability to participate in high psychically distant markets.

5. Discussion of the Results

Quite often entrepreneurial literature talks about different types of ventures. It clasiffies entrepreneurships depending on their features and the proccess they carry out to boost their activities, either in a domestic level or in an international level. In this respect, the last decades evidence has shown a special type of companies, which are characterized by a accelerated internationalization proccess, well-known as born globals. This evidence goes in the oposite direction to the classic internationalization Üppsala"model.

Despite the fact there are plenty of studies who try to explain the features to define born globals existance, which might be associated with financing received, geografical or psicological distance, previous experience, among other variables, not too many studies have analized what are the determinants of success for these kind of firms. Under the reasonable assumption born globals might behave in a different way than ventures with a slow or more traditional internationalization proccess, it appears not only proper, but also neccesary study those determinants of success to facilitate the best decisions by policymakers. Besides the above, due controls must be taken to prevent possibles estimation biases that may lead to ambiguous results.

Enterprises that venture into selling in markets with a high psychic distance have greater chances of success in the international market. This result confirms hypothesis 1 and may imply that the decision by firms to sell at greater psychic distances would be the result of a strategic decision arising from an identified opportunity in the market, which would be aligned with the competitive advantages of the organization. This could explain the greater probability of success for these firms when they internationalize.

In the Chilean economy, the markets located close to the domestic market, that is, the Latin American markets, are not only close in terms of the products and services they commercialize, but also in terms of development levels. On the other hand, the attractive markets of the USA, Canada, Asian countries and Europe are all psychically distant, and these markets require Chilean products worldwide, being able to pay more for our products. Hence, exporting to psychically distant countries has greater value for our companies, and allows them to be more successful. Our productive promotion programs should not only promote business internationalization, but also channel efforts so that our entrepreneurs visualize their growth opportunities in the most important international markets, which are all psychically distant (^{Cancino & Coronado, 2014}; ^{Cancino, 2014}).

Similarly, previous experience in ventures seems to be a determining factor in deciding to sell at a high psychic distance. This result confirms hypothesis 2 and is consistent with the literature, which indicates that previous experience in the international market and risk propensity would be positively related to the decision to internationalize and access distant markets. Although the variable used does not differentiate between previous national or international experience, it can still be a proxy of both the effect of this variable and the risk propensity.

Thereby, the main results of this research, show that international psicologically long distant markets seem to be more atractive for ventures who are looking for international success. It means entrepreneurships with international presence, should try look for developed markets, where they can find a greater demand. However, entrepreneur features could be crucial into this natural internationalization proccess, where the results suggest more experienced entrepreneur would be more likely to venture into psychically more distant markets. Thus, previous experience or any obervable variable that might explain the entrepreneurial propensity, would be a good sign, not only for the venture itself, but also for who must decide where to place resources to support this type of companies.

In addition, an interesting finding is linked to accumulated domestic sales, which would be strongly related to the probability of success, despite having controlled for firm size. At first glance this result may seem counterintuitive, considering that it contradicts the theory of born globals, according to which born global companies are firms that rapidly move to international markets without the need to have any background in the local market. However, as noted above, some authors have given an extension to the phenomenon of born globals, by referring to companies that while behaving like a born global in the sense of rapid internationalization, have previously had experience in the domestic market, which they have named as born again global (^{Bell, McNaughton & Young, 2001}; ^{Bell et al., 2003}; ^{Jantune, Nummela, Puumalainen & Saarenketo, 2008}; ^{Schueffel, Baldegger & Amann, 2014}). Consequently, it seems that the Chilean firms that have been supported by CORFO present the behaviour of a born again global rather than a born global.

Finally, when studying the performance of firms in the international market, it was confirmed that caution should be exercised with regard to the possible selection bias in the sample that may lead to unreliable results (^{Campa & Kedia, 2002}; ^{Constant & Massey, 2003}; ^{Chaney, Jeter & Shivakumar, 2004}; ^{Kirkeboen, Leuven & Mogstad, 2016}). The results showed that the effect of selection bias is significant and that results may differ when using a two-stage regression model, which seeks to correct bias, from using a single-stage model.

6. Conclusion

This study intends to explain the phenomenon of the born global in the context of Chilean ventures, particularly those ventures that have been supported by CORFO’s public programs. The objective was to find factors that could be related to the success of new Chilean companies in the international market, in order to understand the behaviour of these types of companies and how such behaviour may resemble or differ from that exposed by literature and international experience.

In order to do this, we tried to consider variables related to the characteristics of the company and variables related to the operating context, as well as possible problems related to the sample and the inherent biases it might have. Consequently, the contribution of this work is twofold: on the one hand, it is methodological, according to which an attempt was made to correct the possible selection bias that exists when working with truncated samples that represent a part of the population. On the other hand, this paper contributes to better understanding the phenomenon of small and medium enterprises internationalization in the context of emerging economies.

From the findings obtained, we also expect to contribute to the extension of the study of born globals in the reality of developing countries. As a result, it can be observed that Chilean companies supported by CORFO would adopt the same behaviour as born again globals, where variables such as previous experience in local markets can be important when defining success in the current internationalization of a business. At the same time, ventures that decide to sell at a high psychic distance, would provide a favorable message that a strategic move would increase the chances of success, especially when choosing the markets of Asia, North America and Europe as the destination of our exports. In this line, such a decision would also be related to the entrepreneur’s risk tolerance.

This may be very important both for future public policies when making decisions related to promoting innovation and economic development, as well as for future research, which should take into account potential problems associated with the samples they use when generating models that explain firm performance within a specific context. On the one hand, policymakers or organizations as CORFO might use this results to guide his decision and improve resource allocation when they try to foster international entreprenurships participation. On the other hand, some limitations should be considered when analizing the results. First of all, Psychic Distance variable was created under the assumption every market outside the continent might be considered psychically distant from any chilean venture. Further that, this research consider cross section data, overlooking possible endogeneity between variables as Psychic Distance and Success. Thus, it is unknowing if ventures which export to psychically distant markets are more successful or successful ventures has propensity to participate in psychically distant markets.

This study aims to contribute to international born global literature, the results might be interesting to undertand the chacaracteristics of born global in an emerging economies context. It is proposed for future research to deepen in the secondaries findings of this work, which would suggest born globals in Chile would tend to behave as born again globals and deal with possible endogeneity problems usind panel data.