We explored the domain boundaries of two different views of the internationalization of entrepreneurial firms. Supporting New Venture Internationalization Theory (NVI), we found significant influences of the initial resource endowment on internationalization patterns during the early stages of internationalization. Supporting Process Theory of Internationalization (PTI), we found significant influences of the internationalization experience on perceived cost of further internationalization and on growth during later stages of internationalization. An integrated theory for the internationalization of entrepreneurial firms should thus be possible.

During the past few years, new life has been injected into research on the internationalization of small and medium sized firms (SMEs). In particular, the long established ‘Process Theory of Internationalization’ (PTI) has been increasingly challenged for not fully explaining all aspects of the international growth processes of small and medium-sized firms. This challenge has been particularly acutely thrown by researchers building on entrepreneurship research tradition, whose ‘New Venture Internationalization Theory’ (NVI) challenges some of the core behavioral assumptions of the PTI (Oviatt & McDougall, 1994; McDougall & Oviatt, 1997). Questions at issue include: Should the internationalization process of firms always start late and proceed incrementally, as PTI suggests? Under which conditions can ‘born global’ firms be created, and is an early initiation of international operations a recommendable strategy for entrepreneurial firms? How does the initial resource endowment influence internationalization patterns of new firms?

The PTI essentially builds on the behavioral theory of the firm (Cyert & March, 1963) and includes elements of Penrose’s Theory of the Growth of the Firm (1959). This view, developed by the Uppsala Group (Johanson & Vahlne, 1977, 1990), emphasizes the incremental nature of the various change processes that a firm undergoes. In this view, the internationalization process of the firm is initiated in a reactive mode, as the firm reacts to unsolicited export orders. Once initiated, the process proceeds slowly and incrementally, regulated by the gradual accumulation of ‘foreign organizing knowledge’. The recommendation derived from this view is that firms should initiate their internationalization fairly late and proceed slowly, so that the inevitable mistakes made during the process would not destroy the firm (Eriksson, Johansson, Majkgråd & Sharma, 1997). The proponents of the NVI emphasize the importance of entrepreneurial vision and the initial resource endowment of the firm in allowing early internationalization decisions. This view states that particularly in knowledge-intensive industries, where time windows tend to be short, internationalization processes should be started early in a firm’s life. The key for successful internationalization is the competence and vision possessed by the entrepreneurial management of the firm. Rapid internationalization is also made possible by the highly mobile nature of knowledge-intensive assets, as compared to more fixed, tangible assets (McDougall, Shane, & Oviatt, 1994; McDougall & Oviatt, 1997).

There is some controversy between the two views, mainly regarding the optimal timing of the initiation of internationalization process. At first sight, the two views appear to be in conflict, in that one implicitly favors late initiation, while the other emphasizes the need to get your feet wet early. Proponents of both approaches have called for empirical verification of their ideas, but little empirical evidence exists to support or reject either view (see Eriksson et al. 1997; McDougall et al. 1994b, Oviatt & McDougall, 1997). The existing work has been primarily small-sample case studies, and theory advancement has been partially stalled by the difficulty of generalizing from the limited evidence. Some steps toward testing the empirical validity of each view have already been taken (Autio, Sapienza, & Almeida, 1999).

The aim of our paper is to address the empirical validity of the two competing views of the internationalization processes of small and medium-sized firms. We do this by analyzing a large sample of internationalized SMEs, using constructs derived from both the Process Theory of Internationalization and from the New Venture Internationalization Theory. By comparing the explanatory validity of these in different sub-sections of our sample, our aim is to contribute to a better understanding of the domain boundaries of the two competing views.

Our sample consists of 230 British technology-intensive firms. These firms were surveyed in 1997, using a mailed questionnaire survey. Hierarchical regression analysis is used for testing the theoretical models hypothesized. Our analysis shows that each of the two views have predictive validity in different sub-sections of our empirical sample, suggesting that the two competing views are complementary, rather than contradictory. In conclusion, recommendations are made for further research.

Process Theory of Internationalization

The dominant process models of international growth claim explicit intellectual descent from Penrose, using her resource (or knowledge) accumulation-business expansion framework as the basis from which to analyze international growth (Johanson & Vahlne, 1977, 1990; Calof & Beamish, 1995; Eriksson et al. 1997). In these models, the internationalization process is depicted as a learning process. The number of realistic opportunities available for the firm is considered a product of the firm’s resources and its ability to exploit these resources. As many resources are cumulative, e.g., experiential knowledge, the internationalization process is essentially viewed as incremental.

The incremental cycle is regulated by two main factors that are connected to each other through a positive feedback loop. Once an initial resource commitment to international markets has been made, this triggers a learning process, as the firm is exposed to foreign market conditions. This learning process results in the experiential accumulation of ‘foreign organizing knowledge’ (Johanson & Vahlne, 1990). As this knowledge accumulates in the firm, it increases both the speed and size of further resource commitments to foreign markets, resulting in further accumulation of foreign organizing knowledge. As experiential knowledge accumulates slowly, it takes time for the feedback loop to gather significant momentum. The model is rooted in the behavioral theory of the firm (Cyert & March, 1963), and it therefore emphasizes the importance of the perceptions of the firm’s management for internationalization decisions.

Process theories of internationalization are largely descriptive rather than prescriptive. However, some appear to suggest that firms are better off delaying international growth. Eriksson et al. (1997:353) conclude: “many small mistakes in gradual internationalization allow management to form more realistic perceptions than do a single great mistake in a leap-frogging approach to internationalization. This should, however, be investigated empirically.” Also the gradual, momentum- building nature of the internationalization process is emphasized: “the structures and routines [geared to supporting international growth] must be built gradually as a consequence of learning both a  firm’s capabilities and foreign market needs.”

The focus of the process theory being on explaining how the internationalization process enfolds, the theory does not say much about how the process actually gets initiated. It is only stated that the process begins when the firm reacts to unsolicited export orders, reflecting a fairly passive and reactive mode of internationalization Johanson & Vahlne, 1990). This view is in contrast with the proactive, entrepreneurial picture painted by the New Venture Internationalization Theory. It also suggests a different empirical domain for the two views, the main focus of the PTI being on how the internationalization process gradually gathers momentum, once it has been initiated.

New Venture Internationalization Theory

The New Venture Internationalization Theory (NVI) was prompted by the observation that many new firms, particularly in knowledge-intensive industries, go international immediately or soon after their inception. The main thrust of the whole theory has largely been geared to explaining why this happens.

The developers of the theory, Oviatt & McDougall (1994) and McDougall et al. (1994)  emphasize the importance of the entrepreneur as a central decision-maker in his or her firm. Their review of existing internationalization theories concludes that none of these sufficiently explains the formation of international new ventures. They note several trends that contribute to an increasingly early internationalization: the increasing speed and efficiency of international communication and transportation, the increasing homogenization of many markets, the emergence of international financing opportunities, and the emergence of increasingly internationally mobile human capital. They argue that in such business domains it may be increasingly attractive for new ventures to go international at or soon after inception. Such trends, they point out, may be particularly relevant for knowledge-intensive firms, who can enjoy foreign location advantages by combining highly mobile knowledge resources with less mobile resources and opportunities in foreign markets.

Similarly to Johanson & Vahlne (1990), NVI argues that early internationalization establishes a learning process that enables early accumulation of experiential knowledge in the firm and shapes the structures and routines of the firm so as to support further internationalization. Thus, as observed by Autio, Sapienza, & Almeida (1999), the two theories are actually quite similar to each other where the unfolding of the internationalization patterns is concerned. The key difference appears to concern the implications of an early versus late initiation for the later development of the firm. McDougall et al. (1994:484) emphasize the imprinting influence of an early initiation of internationalization for the later development: “In the light of the path-dependence of competence development, new venture founders should consider whether establishing a domestic new venture with plans to later internationalize will be as successful a strategy as establishing a new venture that is international from inception.” The PTI, assuming that most SMEs start their internationalization process fairly late, warn that changing the direction of the firm later in its life is bound to take time. At a fundamental level, both theories thus acknowledge the path-dependent nature of firm development, and that this path-dependency is induced by experiential learning.

Defining the Domains of the Two Views

Even though there appears to be controversy between the two views, particularly where the optimal timing of internationalization is concerned, this may be due more to appearances than substance. Johanson & Vahlne (1977) developed their original model in Sweden, based on a number of case studies focusing on Swedish low- to mid-tech manufacturing SMEs. The business environment of the 1970’s was definitely different from that of the 1990’s: currency flows were still tightly regulated, the Warsaw Pact still existed, international operations of firms were much more limited than they are now, and international management competence was much more scarce than today. Thus, the reactive picture of the internationalization process, in which the firm first entered markets characterized by low psychic distance, made much sense, particularly given that the sample consisted of manufacturing SMEs.

McDougall et al. (1994), on the other hand, base their model on an empirical sample of knowledge-intensive firms in the 1990’s. These firms rely less on fixed assets than do manufacturing SMEs. Their management is likely to be much more sophisticated than that of a typical manufacturing SME. The business environment of the 1990’s is totally different than that of mid-1970’s. The similarities and differences of the two views are summarized in Table 1.

The above discussion has highlighted similarities and differences between the two views of the internationalization of small and medium-sized firms. The discussion also suggests that the two views may be complementing rather than competing. Focusing more on the initiation of internationalization, the NVI emphasizes the importance of the firm’s initial resource endowment. Thus, the empirical validity of the NVI should be greatest in the early stages of internationalization. Focusing more on how the internationalization process unfolds once initiated, the PTI emphasizes the feedback cycle between experiential learning and resource commitments. Because it takes time for this feedback cycle to gather momentum, it may be more applicable in later stages of internationalization. Therefore, our first and main hypothesis in this paper is:

Hypothesis 1a: The early stage of the internationalization process will be regulated by the characteristics of the firm’s initial resource endowment, suggesting a greater empirical validity of NVI in this stage.

Hypothesis 1b: The later stages of the internationalization process will be regulated by the experience accumulation—resource allocation cycle, suggesting a greater empirical validity of PTI in these stages.

In the following, we will derive more detailed hypotheses from both the PTI and NVI.

New Venture Internationalization Theory

The main focus of the NVI is on early stages of internationalization. The entrepreneurial vision of the firm and the knowledge-intensity of its initial resource endowment allow the firm to initiate its internationalization early and to expand its international operations rapidly. The effect should be particularly visible in the early internationalization stages of the firm:

Hypothesis 2: The knowledge-intensity of the firm’s initial resource endowment is negatively associated with the perceived cost of further international expansion in firms that are in the early stage of internationalization.

In the early stage of internationalization, the accumulation of experiential knowledge will not have gathered momentum, as yet, and the internationalization process should be regulated more by its initial resource endowment than by its resource allocation—experience accumulation cycle

Hypothesis 3: The knowledge-intensity of the firm’s initial resource endowment will have a direct positive influence on its international growth in early stages of internationalization.

Process Theory of Internationalization

The PTI emphasizes the importance of experiential learning as a regulator of the speed and size of resource allocations to foreign markets. Eriksson et al. (1997) showed that such learning lowers the perceptions of the management regarding the cost of further international expansion, thereby increasing the speed of resource commitments to foreign markets.

It is likely that the resource allocation—experience accumulation cycle needs to have been operation for some time before it starts exercising significant and visible influences on the firm. This is because the firm will have gathered an established momentum in the domestic market, and changing the established direction does not happen quickly. This is an essential tenet of the PTI. Therefore, we expect that:

Hypothesis 4: Experiential learning will be negatively associated with the perceived cost of further international expansion in firms that have passed their initial phase of internationalization.

The low perceived cost of internationalization will be reflected in international growth through increased speed of resource commitments:

Hypothesis 5: The perceived cost of international expansion will be negatively associated with growth in firms that have passed their initial phase of internationalization.

Following the tenets of the behavioral theory of the firm, we also postulate that the influence of internationalization experience on international cost is mediated through the perceived cost of further internationalization:

Hypothesis 6: The influence of internationalization experience on international growth is mediated through the perceived cost of further internationalization in firms that have passed their initial phase of internationalization.

Sample

Because NVI emphasizes the importance of knowledge-intensive resources for early internationalization decisions, we focus our study on technology-intensive new firms. For the purpose of this study, a technology-intensive new firm is defined as being a legally independent company, not older than ten years, which operates in one or more high-technology sectors. An operational definition of high-technology sectors in the UK has been proposed by Butchart (1987); whose definition of high-tech industries is based on the two ratios of R&D expenses to sales and employees working in R&D to total employees. On the basis of this definition, 17 four-digit NACE codes were identified as having above average expenditures for R&D. These were selected as the target industry sectors for the empirical study.

The addresses were drawn from a Dun & Bradstreet database. We included all firms that had at least three employees in 1997 and that had been founded between 1987 and 1996. 1 800 firms were chosen using a stratified random sampling process (i.e. stratified by size class and service/manufacturing categorization).

Data Collection

The data was compiled by means of a mailed survey, using a four page questionnaire. Six pilot case studies were carried out to test whether or not the questions in the survey instrument were valid, easy to understand and unambiguous to the target respondents. As a result, the questionnaire was modified to take into account any expressed concerns. An introductory letter and questionnaire, followed by three reminders where appropriate, was sent to the managing directors of the 1 800 firms. Altogether 454 firms responded, and consistency checks confirmed that 230 firms met our sample selection criteria.

Operationalization of Variables

Our two dependent variables, the perceived cost of international sales and sales growth, were operationalized in the following way. We asked respondents to state the level of costs of accessing information on foreign markets, of identifying and forming commercial relationships, of setting up foreign sales channels, of overseas product launch, operating costs of the sales channel and the costs of monitoring foreign activities. These six items were measured on a five point Likert scale and were subsequently combined into a single summary scale (alpha = 0.89). In order to operationalize sales growth, we asked respondents to state the level of sales during their last financial year and the expected year-end sales forecast for their current financial year. In the regressions, we measure the expected year-end sales, controlling for sales in the end of last financial year.

Independent Variables

Our independent variables were measured as follows. Internationalization experience of the firm was measured as the number of years that a firm had been active in international markets at the time of the survey. Earlier internationalization experience of the founders was operationalized as a dummy variable that indicated whether the founders had lived abroad before starting their venture. The knowledge-intensity of the firm’s resources was measured in two ways. First, we asked respondents to indicate the percentage of employees that had scientific or technical education at degree level. Second, we asked the founders whether they had experienced any shortage of management skill. This was operationalized using a five point Likert scale for reporting a shortage of marketing, sales, financial management and general management skills. These four items were combined into a single scale (alpha = 0.75).

Control Variables

As control variables, we used the age of the firm measured in years and the size of the firm as the log of employees at the time of the survey.

Stage of Internationalization

We have framed our domain hypotheses using the internationalization stage as the differentiating underlying variable. In this study, we used the extent of international sales of total sales of the firm, the most commonly used measure of the degree of internationalization of the firm (Sullivan, 1994), as a proxy for internationalization stage. All firms generating less than 25% of their total sales from international operations were considered to be in early stages of their internationalization.

The tests of the NVI hypotheses are shown in Tables 2–3. Table 2 shows that, as predicted, NVI variables emerge as significant influences on perceived cost of further internationalization, supporting hypothesis 2. We can also note that PTI variables are not indicated significant influences, suggesting a greater empirical validity for NVI during early stages of internationalization. Thus, hypothesis 1a is supported.

The test of hypothesis 3 is shown in Table 3. We can observe a predicted, positive influence of knowledge-intensity on sales growth. Note that we measure sales growth by controlling for sales at the end of last financial year. None of the PTI variables emerges as a significant influence, providing further support for hypothesis 1a.

The tests of the PTI hypotheses are shown in Tables 4–6. Table 4 shows that when inserted alone, internationalization experience emerges as a significant influence on perceived cost of internationalization, supporting hypothesis 4. However, the equation as a whole does not achieve a significant level of explanatory power.

The influences with both PTI and NVI variables included in the equation are shown in Table 5. Again, internationalization experience emerges as an almost significant influence, but the equation as a whole is not significant. Thus, this equation gives only weak support for hypothesis 1b.

The tests of hypotheses 5 and 6 are shown in Table 6. Because hypothesis 6 predicts a mediating influence, two models are tested. In model 1, the direct influences on sales growth are shown. Apart from control variables, none of the influences emerges as significant. In model 2, we observe a direct negative influence of perceived cost of further internationalization on sales growth, supporting hypothesis 5. However, we have only weak support for hypothesis 6: even though we are able to observe a direct influence of internationalization experience on perceived cost in Table 5, the direct influence of internationalization experience on sales growth in model 1 (Table 6) is non-significant. Furthermore, the models in Table 6 fail to provide support for hypothesis 1b. The results of the analysis are summarized in Table 7.

In this paper, we set out to examine the domain boundaries of two currently debated theories on the internationalization of small- and medium-sized firms. By so doing, we have tried to contribute to a better understanding of this increasingly important phenomenon. By examining the similarities and differences between the PTI and NVI, we have sought to reconciliate some of the recent controversies between the two views. Specifically, we suggest that the two views should be seen as complementary rather than competing, because both emphasize learning and path-dependencies on the international growth of small and medium-sized firms. We also suggested that the New Venture Internationalization Theory may be better suited to explain the early internationalization patterns of technology-intensive new firms, whereas the Process Theory of Internationalization might be better suited to explain internationalization patterns in more advanced stages of internationalization.

Our hypotheses receive partial support in a sample of 230 technology-intensive British firms operating in 17 different industry sectors. In particular, we find support for the greater empirical validity of the New Venture Internationalization Theory in explaining internationalization patterns in the early internationalization stages of technology-intensive new firms. In particular, consistent with the NVI, we find that the knowledge-intensity of the firm’s initial resource endowment is a potentially robust regulator of the early internationalization patterns of technology-intensive new firms. On the other hand, we received only weak and mixed support for the empirical validity of the Process Theory of Internationalization in explaining internationalization patterns in the more advanced internationalization stages of technology-intensive new firms.

Many factors may have contributed to the weak support of the PTI in our analysis. First, the PTI was originally developed using a sample of medium- to low-tech firms, and our sample consisted mainly of technology-intensive new firms. Second, our sample was highly heterogeneous, including both service and manufacturing firms from 17 different industry sectors. Johanson & Vahlne (1990) have themselves acknowledged that their model does not necessarily explain the internationalization patterns of services firms very well. Third, we have used data that has originally been gathered for another purpose, so some of the construct operationalizations used here may not have been optimal for addressing the type of issues explored in this paper. In spite of these shortcomings, we have still been able to observe influences that are consistent with previous research (Eriksson, Johanson, Majkgård, & Sharma, 1997). One the other hand, given the non-optimal nature of our construct operationalizations, the significant influences observed for knowledge-intensity constructs provide a rather impressive testimony of the salience of the knowledge-based view in understanding growth and development patterns of technology-intensive new firms.

In general, we interpret our results as supporting the central tenet of our argumentation, namely, that the PTI and NVI should be seen as complementary rather than competing. Our analysis also highlights the need for further theory development and empirical research, particularly to broaden the domain of empirical validity of the New Venture Internationalization Theory. The significant changes in the operating environment of technology-intensive firms have served to increase the urgency of such work. The observed significant influences of the knowledge-intensity measures suggest that the knowledge-based and learning views may provide potentially fruitful avenues for further theory development. These views have recently been successfully applied into research on the internationalization of firms (Autio, Sapienza, & Almeida, 1999; Barkema & Vermeulen, 1998).

In addition to more theory development, for which we see an urgent need, more empirical work is needed in order to explore the domain boundaries of the Process Theory of Internationalization and of the New Venture Internationalization Theory. The analysis presented here represents only a first step toward this direction, and it has been carried out as a response to recent calls for large-sample tests of the empirical validity of the two views (Oviatt & McDougall, 1997; Eriksson et al., 1997). Clearly, much more empirical is required to sort out the open issues highlighted in recent debates. We hope that this study inspires further empirical research along such lines, and that the collective efforts will lead to the emergence of a more robust, empirically validated theory on the international growth of technology-intensive new firms.

CONTACT: Erkko Autio, Helsinki University of Technology, Institute of Strategy and International Business, P O Box 9500, 02015 TKK, Espoo, Finland; (T) +358-9-451 087; (F) +358-9-451 3095; erkko.autio@hut.fi

Autio, E., H. Sapienza, & J. Almeida. (2000) “Influences of age at entry, knowledge-intensity and imitability on growth.” Academy of Management Journal, 41 (in print).

Barkema, H. G., & F. Vermeulen. (1998) “International expansion through start-up or acquisition: A learning perspective.” Academy of Management Journal, 41: 7–26.

Cyert, R. M., & J.G. March. (1963) A behavioral theory of the firm. Englewood Cliffs, NJ: Prentice-Hall.

Eriksson, K., J. Johanson, A. Majkgård, & D. Sharma. (1997) “Experiential knowledge and cost in the internationalization process.” Journal of International Business Studies, 28: 337–360.

Johanson, J., & J-E. Vahlne. (1977) “The internationalization process of the firm—A model of knowledge development and increasing foreign market commitments.” Journal of International Business Studies, 8: 23–32.

Johanson, J., & J-E. Vahlne. (1990) “The mechanism of internationalization.” International Marketing Review, 7(4): 11–24.

McDougall, P.P., S. Shane, & B. Oviatt. (1994) “Explaining the formation of international new ventures: The limits of theories from international business research.” Journal of Business Venturing, 9: 469–487.

Oviatt, B. M., & P.P. McDougall. (1997) “Challenges for internationalization process theory: The case of international new ventures.” Management International Review, Special Issue 2, 37: 85–99.

Penrose, E. (1959) The theory of the growth of the firm. New York, NY: John Wiley.

Sullivan, D. (1994) “Measuring the degree of internationalization of a firm.” Journal of International Business Studies, 25: 323–342.