The Influence of R&D Collaboration
on the Scale and Depth of the
Internationalization of Spanish
Manufacturers:
The moderating role of Access to Capital
and Product Diversification
Derrick Z.O. de Nes
11422467
August 2017
MSc Business Studies: International Management
Supervisor:
Second Reader:
Dr N. Pisani
Ms M. Mihalache
VERKLARING EIGEN WERK
Hierbij verklaar ik, Derrick de Nes, dat ik deze scriptie zelf geschreven heb en dat ik de volledige verantwoordelijkheid op me neem voor de inhoud ervan.
Ik bevestig dat de tekst en het werk dat in deze scriptie gepresenteerd wordt origineel is en dat ik geen gebruik heb gemaakt van andere bronnen dan die welke in de tekst en in de referenties worden genoemd.
De Faculteit Economie en Bedrijfskunde is alleen verantwoordelijk voor de begeleiding tot het inleveren van de scriptie, niet voor de inhoud.
ABSTRACT
Research on the internationalization effort of firms has led to suggest that businesses innovate to overcome liability of foreignness (LoF). One of the main methods to achieve innovation is by research and development (R&D). R&D can sometimes require more resources than a firm can or dares to commit; in these cases, collaboration may be an outcome. This thesis aims to analyze the relationship between R&D collaboration and a firm’s scale or depth of internationalization. The lack of access to capital is believed to be a driver for collaboration, while product diversification is thought to facilitate R&D. The variables access to capital and product diversification will, therefore, serve as moderators when testing the relationships between R&D collaboration, and the scale and depth of internationalization.
A Spanish panel survey (ESEE), conducted on business strategies, by the SEPI Foundation in cooperation with the Spanish Ministry of Industry, for the year 2011, has been used. The regression analysis has shown that R&D collaboration has an adverse effect on the depth of internationalization of the sample. The moderators have shown no proof for significance. The findings have academic and managerial value and will be discussed near the end of the thesis.
Keywords: R&D Collaboration · Scale and Depth of Internationalization · Access to Capital · Product Diversification · Manufacturing
TABLE OF CONTENT 1. INTRODUCTION...5 2. LITERATURE REVIEW...9 2.1 Innovation...9 2.2 Technological Cooperation... 11 2.3 Internationalization...13 3. THEORETICAL FRAMEWORK... 17
3.1 Development of R&D Collaboration and the Scale, and Depth of Internationalization... 17
3.2 The Moderating Effect of Access to Capital... 20
3.3 The Moderating Effect of Product Diversification... 21
4. METHODS... 24 4.1 Data Collection...24 Dependent variables... 25 Independent variables... 25 Moderating variables... 27 Control variables... 28 4.2 Data Analysis... 31 4.3 Results... 33 5. DISCUSSION... 39 5.1 Academic relevance... 40 5.2 Managerial implications... 41
5.3 Limitations and suggestions for future research... 42
6. CONCLUSSION... 45
1. INTRODUCTION
In our current day and age, internationalization has become an integral part of our everyday life. It has entered the cultural, economical, and political domain and everyone regardless of one's wealth, living rural or urban, or the development of the society they live in will encounter some degree internationalization. Producers and consumers of goods and content to a certain extent all partake in cross-border exchanges. Visible reminders of such exchanges are the ubiquitous global presence of strong brands. The glass of water one drinks out of when going out for dinner may very well be designed in down town Stockholm, the sodium carbonate of which it might be made of can come from China, the glass itself might be produced in Vietnam, while the mineral water it contains may come from Northern Italy.
A telling example of our global connectedness and internationalization is the omnipresence of Coca Cola. It has been reported that their red and white logo is recognized by 94% of the world's population (Bhasin, 2011). Furthermore, this soft drink can be found worldwide in grocery stores in all but two countries, Cuba and North Korea. Due to a US trade embargo, Coca Cola is prohibited from conducting business in either country, although sources have indicated that contraband Coca Cola can be found in the North Korean capital (Hebblethwaite, 2012). Besides, not even the staunchest isolationist and self-proclaimed autarkic economies can completely disconnect from the outside world as is shown by Pyongyang’s dependence of Beijing for their basic food and energy needs (Albert, 2017). Remnants of internationalization, however small, are even to be found within the hermit kingdom (The Economist, 2017).
The fact is that internationalization has always been a part of human life ever since the dawn of time; the difference now is that it has been democratized and therefore has become more widespread. This trend has continued to such an extent that nowadays consumers in free market societies are often even unaware of what the origins of the imported products are that they purchase are (Conaway, 2015), as buying such goods has become standard practice.
This, however, has not always been the case; in the past, only expensive non-perishable goods were able to justify the high logistical costs related to long distance and risk. As time passed, the economic principle of supply and demand, led by the insatiable apatite of the ever more refined taste of the consumer base, encouraged the development of technological capabilities and colonization, to more efficiently transport and secure goods. The Dutch East India Company (VOC) is often considered to be the world’s first truly transnational corporation (Brook, 2008; Shorto, 2013). The period that followed would be later known as the Dutch Golden
Age in which commerce and culture flourished in the Low Countries. Business was
good, and the financial figures could back this statement. The VOC paid an averaged of 18% dividend over the course of its 200-year existence (Van Lent & Sgourev, 2013). Many countries followed suit, and whole continents were to be carved up (e.g.,
Scramble for Arica during the 1884-85 Berlin Conference) (St. John’s College, 2014).
Empire grew, and colonial powers initially almost solely extracted from these territories (unidirectional resource dispersion). Later on, cotton would be shipped from India to Great Britain, where it would be milled into cloth, to be sold back to the India. This was made possible by the increase in productivity, which was driven by the Industrial Revolution (Broadbury & Gupta, 2005). Empire would however not last forever, other great (European) continental powers came to prominence after their
unifications (e.g., Germany and Italy) which would eventually lead tension on the continent, resolving into the First World War (WWI). After WWI the colonial powers had been reduced to a shadow of their former glory as their cities and economies lied in ruin and the seed of independence and decolonization within the colonies had been spread. WWII was the fatal blow, and independence of former territories followed in quick succession. Former colonizers had to grind their rent-seeking behavior to a halt (Engerman & Sokoloff, 2005) and start to create value by innovation.
Skip forward to the 21st
century and operating across borders has never before been as accessible for firms as it currently stands. Especially small and medium-sized enterprises have benefited from the democratization of long-distance transportation and instant communication, which have been led by technological advancements and economy of scale. This phenomenon has accelerated the ongoing trend of globalization ever more. Technological cooperation has particularly been prevalent in recent years as information has become easily transferable, and research and development (R&D) can be strenuous, yield insufficient new findings, and requires substantial investment capital. The act of seeking new opportunities has led firms to geographically diversify and pursue internationalization, however, this requires considerable resources and the ability to overcome one’s foreignness. Knowledge may partially be an outcome for this seemingly disadvantage, and perhaps the cost savings due to R&D Collaboration and the rewards of Internationalization may offset the costs related to Internationalization?
This paper will focus on “The Influence of R&D Collaboration on the Scale
and Depth of the Internationalization of Spanish Manufacturers” as previous
innovation (Hagedoorn, 2002) but not how this influences the internationalization efforts of firms.
2. LITERATURE REVIEW
To determine "The Influence of Technological Cooperation on the
Internationalization of Firms," one should first have an extensive insight concerning the relevant topics, which are to be discussed. The literature review will help the reader reach this deeper understanding by dissecting the research question and by elaborating thoroughly on the three most important concepts of which the research question is comprised. Recently published findings on these three core concepts will be critically assessed and contradicting stance will be given a platform for rebuttal.
2.1 Innovation
The ongoing mutation of the modern economic paradigm will incessantly be the driving force for the decay of the old, and the creation of the new economic reality. This phenomenon was coined creative destruction and was first discussed by Schumpeter in his 1942 paper Capitalism, Socialism, and Democracy. It explained the vicious cycle of the rise and fall of new production units replacing outdated ones, the inevitability of change, and the need for constant innovation to remain relevant within the contemporary context. Innovation can be described as the "production or
adoption, assimilation, and exploitation of a value-added novelty in economic and social spheres; renewal and enlargement of products, services, and markets; development of new methods of production; and the establishment of new management systems. It is both a process and an outcome" (Crossan & Apaydin,
2010). The rate in which innovation occurs is exponentially increasing (Hagel et al., 2013), consequently also its importance. The main driver for contemporary innovation has been the trend of information becoming ever more readily available due to the rise of the Internet. The mass distribution of knowledge via the Internet has
granted vast populations access to knowledge and has allowed them to partake in the innovation process. Simultaneously mass computing power has furthermore increased the pace of innovation, as it has been able to solve complex calculations in a far superior speed and accuracy compared to its human counterparts. Mass computing power has granted humanity access to previously literary inconceivable worlds, namely the quantum realm and the universe.
"Innovation has become a core driver of growth, performance, and valuation" (Barsch et al. 2008) mainly because it has become easier to scale up and therefore exploit new findings. Companies all over the world are increasingly heavily investing in perceived future disruptive technologies in order maintain relevance by hopefully being part of the next paradigm shift. Such technologies include; Artificial Intelligence (AI), Blockchain, Greentech, Internet of Thing (IoT), Big Data, Genetics (CRISPR), and Augmented Reality (AG), just to name a few. To put things in perspective, large tech companies like Alphabet (previously known as Google) and Facebook have been known for splurging hundreds of millions, sometimes billions of dollars, on acquiring new technologies (Opam, 2014; Welch, 2014). The trend of laying an increased emphasis on innovation can also be seen on a macroeconomic scale as is shown by the rise of the average gross domestic R&D expenditure in OECD countries 12.3% or 123 bps from 2.124% in 2000, to 2.403% in 2015 (OECD, 2017). Innovation, reinvention, and the ability to adapt are key characteristics a firm should have in the modern day economy, and firms unable to keep by will encounter major hurdles. Competition is killing, and firms, which are not able to maintain relevance within the ever shifting and perpetually accelerating economic context, will suffer from Schumpeter’s creative destruction.
2.2 R&D Collaboration
R&D Collaboration can be defined as an agreement between multiple independent
agents with the aim to increase their competitive advantages by sharing resources and, developing and executing technological processes. The type of agreement will depend on environmental contingencies (Gulati, 1998; Arranz & Fdez. de Arroyabe, 2009).
The pressure to innovate is high, as are the costs and the level of uncertainty during the R&D process regarding results and final market response (Arranz & Fdez. de Arroyabe, 2009). R&D Collaboration may help overcomes these challenges as resources and risk can be shared, as well as knowledge. Also due to the interdependence of collaborating partners, general strategic decisions can be made which are beneficial to both parties.
Parallels, concerning possible beneficial outcomes, can be drawn between R&D Collaboration and embarking on a joint venture (JV) as both revolve arround sharing of knowledge and risk, and the decrease in one-party investments a mentioned by Kogut (1991). Research shows that R&D Collaboration can overall positively affect the innovation process regardless of the size of the involved parties (Pittaway et al., 2004) as well as ensure an increase in added value per employee (Belderbos et al., 2004).
The traditional view of the relationship between R&D and collaboration is predominantly based on the Absorptive Capacity Theory, which states that firms with higher internal R&D activity will regard collaboration with outside agents more relevant than their low internal R&D counterparts (Cohen & Levinthal, 1990). Literature shows that there is a high emphasis on obtaining and creating new knowledge by branching out to new partners and collaborating with external parties
(Rothwell, 1994; Lundvall, 1992). Empirical studies have shown that innovation rarely occurs in isolation (Christensen & Lundvall, 2004).
According to Child and Faulkner (1998), formal cooperation is vital in assuring innovation to take place as it enhances synergy and reduces transaction costs (Arranz & Fdez. de Arroyabe, 2009). Clear agreements are necessary within the cooperative innovation processes as knowledge, technology, and know-how (Chesbrough et al., 2006) may become the subject of unintentional one-directional information spillover. This is especially likely when there is a strong initial information asymmetry as limitless sharing will inevitably lead to information equilibrium.
Literature shows that smaller firms rely heavily on large firms during their collaborative innovative activities, as development of innovation in isolation of large firms is regarded impossible. “The interdependence of large firms and small firms
during the evolution of new technologies and industries is best described as dynamic complementarities.” (Rothwell & Dodgson, 1994) The dependence of smaller firms
on R&D Collaboration can be seen in the Rothewell and Dodgson’s paper in which SMEs employ relatively more people in relationship to their internal R&D expenditure and spend “almost twice as much of their R&D expenditures towards
R&D collaboration” (Narula, 2001) compared to their larger counterparts. Either way,
innovation by collaboration, due to limitations in resources, remains the single most important reason for firms, regardless of size, to collaborate. Smaller firms, in general, are able to leverage their limited R&D resources more efficiently (Nooteboom, 1994) than their larger, more resourceful, counterparts. This is in line with the following statement made by Luptáčik (2010): “efficiency implies the
resources.” Narula’s (2001) findings show that successful smaller firms are able to
compete with larger competitors, when relatively more resources are allocated to
R&D Collaboration.
Collaboration, however, does have its drawbacks. Firm engaging in R&D
Collaboration effort will need in-house resources, which can integrate R&D findings,
as findings may be differently interpreted across firms. Furthermore, even though the risk is shared, the failure rate still remains high, Narula’s (2001) sample regarded a failure rate of 50% to be “very good indeed”. For a large firm, failure is easier to accept, as often they will have multiple researches running simultaneously. Failure for smaller parties will have significantly more implications as they are often less diversified due to them having fewer resources to commit to R&D. The importance of the outcome of R&D Collaborations will surely influence the power dynamic between the smaller and larger R&D partners.
2.3 Internationalization
With an increase in competition, companies will eventually, especially companies operating in a price-sensitive market (Porter, 1986), be forced to expand operations abroad to reach economies of scale and thus a competitive edge. Other rational is to stake a claim to a larger slice of the global market. Transaction Cost Theory state the form and competitiveness of the international operations of a multinational enterprize (MNE) depend crucially upon firm specific advantages (FSAs), country specific advantages, and the corresponding internalization advantages (Rugman & Verbeke, 1992). Adam Smith’s theory of Absolute Cost Advantage states that one should enter a market when one is capable of producing more than the competitor due to FSAs with the same amount of resources (Ingham, 2003). The Global Strategy Framework gives three generic approaches how companies can create value abroad; Adaption
(tailor to the local situation), Aggression (economies of scale or scope), and Arbitrage (performance enhancement). The approach will be selected based on the firm’s characteristics and the host country’s situational context (Ghemawat, 2008). Other ideas on international expansion may focus on cultural distance; one such concept is the Uppsala Model by Johanson and Vahlne which claims that firms tend to expand to culturally similar countries to minimize their liability of foreignness (LoF) (Zaheer, 1995). Benito and Gripsund (1992) on the other hand claim that foreign direct investment (FDI) location choices are purely based on the expected outcome (risk/return).
Internationalization is a term, which is often used in our day and age, but till what geographic extent does internationalization transpire? Yes, some MNEs are known for maintaining an Arbitrage strategy, in which they cover certain aspects of their value chain abroad by exploiting local opportunities (Ghemawat, 2007), but what extent does this occur and how much do these cost-reducing solutions impact the overall profitability of the firm? More often than not, MNEs operate domestically and to a lesser extent on semi-global/regional basis thus making them not as global as they may initially seem. The reason why one cannot speak of a fully globalized world is that the world has not been homogenized when it comes to all factors of distance except for direangular distance (geographical). Full globalization has not occurred as, among other things, "the international integration of markets for goods, services, and
capital" (Garrett, 2000), has not taken place on a global scale. Developments, to a
greater extent, towards such integration occurred on a regional scale, most notably within the European Union (Arabic Knowledge@Wharton, 2012).
Multiple studies suggest that the degree of globalization among MNEs is highly overestimated, Pisani managed to capture this in his 2016 publication “The
changing global footprints of the Fortune Global 500” by analyzing the
majority-owned affiliates of Fortune 500 companies. His research showed that 46% of majority-owned affiliates were based domestically, 17% regionally, and 37% globally. US and Canadian MNEs have been shown to be the most (65%), whereas Western European MNEs, with a mere 30%, being the least domestically oriented. This void has unsurprisingly, to a large degree, been covered by the Western European MNEs' majority-owned regional affiliates (26%). These findings coincide with both the publication of Ghemawat and Pisani in 2014 and statements made in Ghemawat and Altman’s 2016 report on Global connectedness for DHL in which they state: “Europe remains the world’s most globally connected region." Also, Ghemawat has recently proposed two laws of globalization in the DHL Global connectedness report;
• The law of semiglobalization: International interactions, while non-negligible, are significantly less intense than domestic interactions.
• The law of distance: International interactions are dampened by distance along cultural, administrative, and geographic dimensions and are often affected by economic distance as well.
Ghemawat's second proposed law of globalization implies that there is a negative correlation between distance and international engagement. The reasoning substantiating the above mentioned proposed law is the following; as distance increases (e.g., being cultural, administrative), so does the monetary and non-monetary costs involved with bridging these differences, often resulting in more risk and a smaller margin thus a lower likelihood of international engagement. An often referred to model concerning increased barriers due to distance is Ghemawat's (2001) CAGE distance framework in which Ghemawat identifies Cultural, Administrative, Geographic, and Economic differences to be key factors to focus on when considering foreign expansion. Johanson and Vahlne’s 1977 landmark article, often coined The
Uppsala Model, proposes that MNEs are likely to expand abroad to countries with
whom they have the least psychic distance. How can a firm become less sensitive to these so-called psychic distances and thus become more likely to seek foreign endeavors? One such way is to become less dependent on current location bound firm specific advantages (LB FSAs) is by pursuing non-location bound firm specific advantages (NLB FSAs). These NLB FSAs are transferable across borders and will ensure some advantage, thus making internationalization process more profitable for a firm because they can, to a certain degree, exploit their current capabilities. NLB FSAs enables firms to scale-up and increase the scope of their activities (Rugman & Verbeke, 2008a). To acquire NLB FSAs, a firm requires a higher degree of resource recombination, thus making it more costly (Rugman & Verbeke, 2008b).
There are many barriers to overcome when one enters the internationalization process. The entire market dynamic may change due to distance; the one constant asset firms may be able to transfer abroad are NLB FSAs (e.g., knowledge). Acquiring such advantages requires time, capital, assets, and the right labor force. This can be very costly, especially in an innovation intensive firm (Trigo & Vence, 2012). R&D Collaboration across firms may bring an outcome as cross pollination of ideas can take place, and assets can be used at a higher rate thus improving the overall return on investment on them.
A lot has been written about the internationalization of R&D, however to my knowledge; little research has been conducted on the interaction between R&D Collaboration and the internationalization process of firms. Hence my research question for this paper will be:
What is the Influence of R&D Collaboration on the Scale and Depth of Internationalization? What are the moderating effects of Access to
3. THEORETICAL FRAMEWORK
3.1 Development of R&D Collaboration, and the Scale and Depth of Internationalization
The literature review suggests that R&D Collaboration enhances a firm’s level of internationalization. The question is how and to what extent this assumption holds. The added value derived from perpetual R&D Collaboration, when it comes to knowledge exchange, may at a certain juncture no longer offset the corresponding costs. The learning curve, as
depicted in Figure 1
(Haylock, 2014), of such an endeavor, might eventually
plateau due to near
information symmetry
(Sampson, 2002) while the inevitable increase in costs associated with the deepening one’s understanding of a complex matter will no longer justify further participating in the collaboration effort. A parallel can be seen between the predicted sigmoid relation, and the widely publicized learning curve (Hirschmann, 1964) as both relationships entail a pursuit for knowledge in exchange for time and effort. There will be an initial start-up phase in which mutual unfamiliarity will have to be overcome, followed up by a growth phase in which the vast majority of knowledge will be transferred in a relatively short period, and eventually, there will be an inevitable decline in knowledge transfer as symmetry is near. Similar to natural resource extraction, knowledge which is easily accessible will be gained rather quickly and cost efficiently, while knowledge which requires a greater understanding will force firms to delve deeper, like in a mine, and will require them to allocate more
resources which tend to lead to a lower cost efficiency (Farrow, 1985). As testing a sigmoid relationship has proven to be difficult, I have opted to hypothesize for a positive correlationhip, as a sigmoid relationship is, in fact, a positive relationship with two inflection points.
It is a misconception to believe that internationalization will always ensure a firm’s profitability; it may increase the potential market share and revenue but will inevitably correspond with new barriers, and therefore new costs, which will first have to be offset. SMEs for OECD countries have reported that they encounter these 10 internationalization barriers when embarking on foreign endeavors; “(1) Shortage
of working capital to finance exports, (2) Identifying foreign business opportunities, (3) Limited information to locate/analyze markets, (4) Inability to contact potential overseas customers, (5) Obtaining reliable foreign representation, (6) Lack of managerial time to deal with internationalization, (7) Inadequate quantity of and/or untrained personnel for internationalization, (8) Difficulty in matching competitors’ prices, (9) Lack of home government assistance/incentives, and (10) Excessive transportation costs” (OECD, 2009). R&D Collaboration among firms may enable
them to partially overcome the barrier of insufficient financial means as resources can be shared thus drastically reducing costs (Harrigan, 1988; Mowery, 1988; Hagedoorn, 2002). Knowledge can be transferred, which will significantly increase its potential return on investment, thus making it overall more cost efficient. Furthermore, facilities can be shared enabling newfound access to resources which were not previously even considered, and resources can be used more intensively and efficiently thus making it easier to spread fixed costs (Barnes et al., 2006).
In addition to the R&D related benefits, firms may also be collaborating with partners, which are directly capable of decreasing a firm’s LoF. This is possible if the
partner firm has previous experiences with foreign parties, or what would be even more significant, the partner firm itself may be foreign. In such a case, a firm with foreign aspiration could partially overcome internationalization barriers like identifying foreign business opportunities, having limited information to locate/analyze markets, and/or being unable to contact potential overseas customers (OECD, 2009). Having said this, I believe collaborating partners will only comply with such knowledge request if the markets they operate in are dissimilar enough for the other not to be regarded as a potential competitor. On the other hand, embarking on an R&D Collaboration means that information flows in both directions. R&D Collaboration can be problematic for a firm’s internationalization ambitions to a country where its collaborative partner comes from if the markets they operate in are similar and more information is flowing out than coming in.
Although both the dependent variable in hypotheses 1 and 2 are proxies for internationalization, there are distinct differences. Where Scale of Internationalization is determined by foreign sales divided by total sales, Depth of Internationalization is comprised out of a more complex calculation including the concept of distance. Having said this, I do not hypothesize that the Scale and Depth of Internationalization will differ to such an extent that they will act counter opposite of one another in regards to their relationship with R&D Collaboration.
Overall the literature points into the direction of suggesting that R&D Collaboration provides more advantages than disadvantages when seeking internationalization, hence the following hypotheses;
Hypothesis 1: The Relation between the Development of R&D Collaboration and a Firm’s Scale of Internationalization is Positive
Hypothesis 2: The Relation between the Development R&D Collaboration and a Firm’s Depth of Internationalization is Positive
3.2 The Moderating Effect of Access to Capital
Access to financial capital is one of the key facilitating resources for conducting business activities (Kelley et al., 2012). Kuntchev et al. (2012) state that access to capital is a crucial factor that can catalyze a firm’s entrepreneurial activities. According to research conducted by Dileep Kumar (2012) on the Entrepreneurial
Orientation of Internationalization of SME’s, “entrepreneurial orientation plays a significant role on [an] entrepreneur’s decision to go global." Klonowski (2012) even
dares to state that high levels of leverage may resolve into superior SME performance, while Acharya begs to differ and warns for the underperformance of firms led by high indebtedness. The overall consensus, however, agrees with Kuntchev, Dileep Kumar, and Klonowski when it comes to the crucial role access to capital plays in a firm’s overall performance (Wiklund & Shepard, 2005; Mazanai & Fatoki, 2011). It is fair to state that access to capital plays a significant role when seeking internationalization.
A lack of operational capital seems to be a reason why firms do not allocate additional financial capital to their R&D departments. Multiple researches have shown that R&D expenditure is positively related to firm size (Zhang & Li, 2008; Chai & Gu, 2015), thus by collaborating one might increase the resources at its disposal. Lerner’s et al. (2003) research shows that small U.S.-based biotechnology companies indeed tend to “fund their R&D through alliances with larger firms when
financial markets are disadvantageous." In these situations, the larger firm often pays
for the vast majority of the joint R&D and in return gets control. Risks involved with this kind of structure are that the smaller of the two becomes very reliant on the other and becomes a prime target for being taken over (Mitchell & Sign, 1996). The presumption is that R&D Collaboration can partially make up for the lack of financial
assets when pursuing NLB FSAs, yet how will the access to capital influence R&D Collaboration/Internationalization dynamic?
The access to capital alleviates the need to rely on R&D Collaboration solely and will provide firms with an alternative way to resolve former capital related challenges. For this reason, I would like to hypothesize the following:
Hypothesis 3: The access to capital markets negatively moderates the relationship hypothesized in Hypotheses 1 and 2
3.3 The Moderating Effect of Product Diversification
The second moderating effect to be suggested is product diversification. “Diversification may arise for a variety of reasons: to take advantage of
complementarities in production and existing technology; to exploit economies of scope; to reduce exposure to risk; to stabilize earnings and overcome cyclical business conditions; etc.” (Khemani & Shapiro, 2003). The moderating effect of
product diversification is especially relevant for the relationship between R&D
Collaboration and the Sale and Depth of Internationalization if the subject firm is to
choose for Ghemawat’s Adaptation strategy (2007) when entering foreign markets. The Adaption strategy helps foreign firms with overcoming cultural, administrative, geographic, and economic distance (2001) by adapting to local needs to maximize local relevance. In the process of adapting to local needs, it is not unforeseeable that firms will diversify to cater to these requirements. Going back to the 2003 quote made
R&D Collaboration
Access to Capital
The Scale and Depth of Internationalization
-‐
by Khemani and Shapiro, R&D Collaboration will increase the number of commentaries (production and technology) of which collaborating firms can take advantage of, it this therefore likely this may have a positive moderating role on the relationship between R&D Collaboration and the different proxies for internationalization.
On the other hand, research conducted by Baysinger and Hokisson (1989) indicates that product diversification can have a negative effect on investments in R&D, partially because diversification might cause a firm to drift away for its core competencies. Furthermore, Hitt et al. (1991) state that the “primary mode of product
diversification is through acquisitions of other firms that are often financed by debt”
thus suggesting that Product Diversification will not necessarily lead to more R&D. Kehmani and Shapira’s (2003) paper also states that there is mounting evidence that
related diversification may be more profitable than unrelated diversification."
Baysinger and Hokisson however also stated “early gains obtained from limited
product diversification (e.g., related) are offset by continued product diversification further from the core business (e.g., unrelated)”, thus overall agreeing with Kehmani
and Shapira.
Furthermore, Hitt et al. (1994) have stated that internationalization is positively related to firm innovation. This was believed to be so because "international markets provides opportunities for greater returns on innovations and
reduces the risk of failure due to the additional number of markets in which the innovation may be applied."
As diversifying one's product range requires a wider array of inventory, more R&D, and thus more capital, it has been hypothesized that the broadening of one's
product line may have a positive moderating effect on the R&D Collaboration / Internationalization dynamic. Hence the following:
Hypothesis 4: The firm’s product diversification positively moderates the relationship hypothesized in Hypotheses 1 and 2
R&D Collaboration
Product Diversification
The Scale and Depth of Internationalization
+
4. METHODS
4.1 Data Collection
The empirical study is based on an extensive survey of Spanish manufacturing firms (ESEE) conducted by the SEPI Foundation in cooperation with the Spanish Ministry of Industry. The Survey on Business Strategies, Encuesta sobre Estrategias
Empresariales (ESEE), reoccurs annually and has an average sample size of
approximately 1800. ESEE’s population ranges across industries and segment size. Contributing firms are presented with a questionnaire, which touches on a multitude of topics ranging from the firm’s productivity to the number of patents it holds abroad, and from the number of employees the firms have, to whether it is family run. As the data utilized for this research is derived from a single year survey (ESEE 2011), and therefore also from a single moment in time, it is fair to say that this study will only give a snapshot insight into the relationship between R&D Collaboration and internationalization.
To measure the relationship between these two variables the dataset that has been used need to be manipulated as not all necessary information was readily available or in certain cases vital information was missing. Of the 1816 partaking firms, nearly a third the rows within the dataset did not include data on foreign sales. This meant that the initial sample size had to be reduced to 1221 (67.24%) as some of the variables were to be calculated based on ESEE provided data. Missing data would misrepresent reality and resolve in erroneous data points. These data points (#DIV/0! or #Value!) have been replaced with "." and have been later on, altogether with the remaining data points of the row, completely removed using listwise deletion to enhance validity. As a result, the sample size decreased furthermore, with 52.33% to
582. Although the sample size is not even a third of what it used to be, it is believed that the heavily diminished sample size is statistically seen still acceptable.
ESEE has proven its legitimacy as it has been conducted by an OECD governmental institution and has been applied by much economic based research including articles in scientific journals, PhD dissertations and MA theses (Cassiman & Golovko, 2007; Merino, 2014).
All the variables, which have been used to test the hypothesized statements, are shown in Table 1 and will be further elaborated on in the forthcoming pages.
Table 1 Operationalization of variables in the models.
Variable Operationalization
Dependent
Int’l Scale Foreign Sales divides by Total Sales
Int’l Depth (EU% Sales + Latin America% * 2 + OECD * 3 + Rest of the World *4)
divided by (Int’l Scale * 4) Independent
R&D Collaboration External R&D Expenses divided by Total R&D Expenses
Moderator
Gearing Debt divided by Equity
Diversification Number of product segments operative in
Interaction terms
R&D Collaboration x Gearing Interaction variable between R&D Collaboration and Gearing
R&D Collaboration x
Diversification Interaction variable between R&D Collaboration and Diversification
Control
Age Number of years since inception
Employees Number of Employees
Foreign Patents Number of foreign held patents
Family Ownership Firm is run by a family (1 = Yes; 0 =No), dummy variable use in models
Listed Listed on the Stock Exchange (1 = Yes; 0 =No), dummy variable use in models
Standardization Products are standardized (1 = Yes; 0 =No), dummy variable use in models
Dependent variables
Scale of Internationalization: Quantifying a complex concept as internationalization
is not as straightforward as one would initially presume. This is because many interpretations can be made of what the concept should constitute. Firms can increase their level of internationalization by many means; this can be achieved by among other things establishing subsidiaries in foreign markets or outsourcing labor (Kafouro, Buckley, Sharp, & Wang, 2008). As a result, researchers use a vast array of
measures to proxy internationalization, including foreign sales to total assets and the number of countries a company is operative in. However following the many studies that have preceded mine (e.g. Grant, 1987; Kotabe, et al., 2002; Pisani et al. 2016) and recommendations made by Sullivan (1994), who wrote a journal piece solely on the topic, this thesis quantifies the scale of internationalization by the ratio of foreign sales to total sales (FSTS).
Depth of Internationalization: Although FSTS has enjoyed a widespread
adaptation as a proxy for internationalization this does not mean that the discussion revolving the topic of how to quantify internationalization has seized to exist (Pangarkar, 2008; Cerrato & Piva, 2012). Even Sullivan (1994) acknowledges that FSTS is not perfect and that combining multiple variables will provide a more reliable and accurate representation of a firm’s degree of internationalization. Taking this in mind and lending inspiration from “A Technological Contingency Perspective on the
Depth and Scope of International Outsourcing" (Mol et al., 2003), I have devised a
simple yet elegant method in which the psychic distance of internationalization and the degree of exposure to it has been quantified. The further a firm operates, combined the extent of exposure (measured in sales) to the designated distance the firm encounters will determine the so-called depth of internationalization.
ESEE measures five possible spheres in which partaking firms, of the survey, can geographically (partially) operate. The first sphere is the national market, in the case of ESEE that would be Spain, option number two is the EU single market. The remaining options, in order of distance, are Latin America, residual OECD countries, and the Rest of the World. As the distance increases so does the multiplier related to the specific distance. Because sales conducted within the national borders do not constitute as international trade these sales are therefore multiplied with a
factor 0. The percentage of sales in the EU single market is granted a factor 1, the percentage of Latin American sales are granted a factor 2, residual OECD sales a factor 3, and sales in the Rest of the World a factor 4. DEPTH is calculated by summing up all individual calculations (sum ranges from 0 to 4) and subsequently divided by 4 in order to get a measure, which has a maximum value of 1. To illustrate how this seemingly complex formula works an exemplary hypothetical firm has been devised, with certain export characteristics, of which the depth can be calculated.
Firm Alpha derives 30% of its sales from within Spain, 20% for the EU, 10% from Latin America, 30% for remaining OECD countries, and 10% from the Rest of the World. According to previously conceived formula for DEPTH the degree of internationalization of Firm Alpha should be calculated in the following manner;
(30% * 0 + 20% * 1 + 10% * 2 + 30% * 3 + 10% * 4) / 4 = (0 + 0.2 + 0.2 + 0.9 + 0.4) / 4 =
1.7 / 4 = 0.425
Independent variables
R&D Collaboration: The measure of R&D Collaboration has been determined by
External R&D Expenses divided by Total R&D Expenses, as this will show the extent on which the firm relies on collaboration when it comes to R&D compared to not collaborating.
Moderating variables
Gearing: Gearing is a measurement of how much debt a firm has accumulated compared to the amount of equity it holds (Kochhar, 1996). The Gearing ratio indicates a firm's access to capital as it shows to what extent lenders are willing to partake in the firm’s business endeavors. Also, it is a good indication concerning a firm’s outlook on its own future as this shows their willingness to amass debt and pay
interest, to finance present undertakings. Furthermore, the values of Gearing have undergone a log transformation to reduce the highly skewed distribution skewness 20.318 to 0.297 and the initial standard deviation from 1111.183 to 0.480.
Diversification: The variable Diversification constitutes a firm’s degree of product diversification and is measured by the number of product segments a firm is operative in. Literature suggests that product diversification yields superior performances compared to firm’s with a less diversified product-line (Bettis & Hall, 1982; Palepu, 1985) as “the resource-based view and transaction cost theory suggests
that internationalization offers the opportunity to successfully leverage a firm’s strategic resources across different levels of product diversification” (Bausch &
Krist, 2007). Also, internationalization requires firms to adapt to local needs (Nasir & Altinbasak, 2009; Powers & Loyka, 2010) thus possessing the capability of being able to manufacture and develop a wide product range ought to be a useful ability one can have when aspiring or conducting internationalization. Also presumably because firms with a highly diversified product-range are suggested to yield superior performance compared to their counterparts, one might presume that such firms are more likely to possess the resources needed for internationalization.
Control variables
Age: Age is calculated by years of inception (Delios & Henisz, 2003). A firm’s age
can be an indicator of whether a firm is dynamic and flexible or stagnant. Of course other variables than age including industry, company culture, and nature of the firm’s resources will have an influence to the extent of a firm’s dynamism but it is fair to say that young companies are hardwired and purposely structured to encounter the quickly shifting contemporary economic reality (Thornhill & Amit, 2003). Unsurprisingly young ventures seem to more easily internationalize compared to their
elder counterparts as research has shown (Autio et al., 2000). The fact is that “flexibility plays a central role in the internationalization process” (Bausch & Krist, 2007) as entering the unknown often corresponds with risk and uncertainty. Also, young firms often still display entrepreneurial behavior (Penrose, 1959) whereas elder firms more frequently opt for uncertainty avoidance (Sapienza et al., 2003). It is for this reason why it is expected that age will be negatively correlated with either measure of internationalization.
Employees: The number of employees can be seen as a measure of the size of a firm. The firm’s size is a good indication of the amount of resources a firm has to its exposal (George et al., 2005; Lu & Beamish, 2001) and how willing a firm to take on risk, e.g. international expansion and confronting LoF (Coviello & McAuley, 1999; Kirby & Kaiser, 2003). Previously researched Malaysian SMEs has shown that in some circumstances there can be a positive relationship between the size of a firm and the degree of internationalization it experiences (Chelliah et al., 2010). Furthermore, the values of Employees have undergone a log transformation (Verwaal & Donkers, 2002) to reduce the highly skewed distribution skewness 7.080 to 0.254 and the initial standard deviation from 1110.824 to 0.564.
Foreign Patents: The ownership of foreign patents in itself is already an indication of international activity, let alone when one takes into consideration what a firm does or is to do with such a patent. Foreign patents can show what the intent of firms are abroad and to what extent they are committed to these endeavors (Cantwell & Piscitello, 2005; Branstetter et al., 2006; Allred & Park, 2007). Also, the values of Foreign Patent have undergone a log transformation to reduce the highly skewed distribution skewness 17.694 to 6.464 and the initial standard deviation from 8.557 to 0.206.
Family Ownership: A firm can be regarded family-owned when a single family
controls the majority of the shares and when at least one of the family members holds a managerial position within the firm (Cerrato & Piva, 2012). As being or not being a family-owned business is a binary option, firms that have been categorized are assigned value 1 whereas firms that have not been are assigned value 0, this is in accordance with Fernández and Nieto (2006). Family ownership is expected to have a negative influence on internationalization as family-run firms tend to be more conservative and are thus more risk averse (Caspar et al., 2010), thus presumably less likely to undertake international endeavors. This line of thought is consonance with findings done among Taiwanese firms’ ownership structure and the level of internationalization they encounter (Liu et al., 2011).
Listed: Being listed on the Stock Exchange vastly increases a firm’s ability to access
capital thus making international expansion more likely due to the high costs related to such endeavors. The extent of which being listed will influence internationalization will depend on the alternative modes of accessing capital, in other words, how developed is the local financial sector (Gonenc & de Haan, 2014)? Furthermore, not every firm can get listed; such companies will need to comply with strict governmental regulations to maintain access to the stock market. This in itself can give an indication of the overall health of the firm and its abilities for international expansion.
Standardization: Opposite of having a diversified product range, standardization implies not a firm’s ability to adapt to local needs but to achieve economies of scale by maintaining no or minor increases in fixed costs due. Economies of scale can be an excellent strategy for keeping costs down but will most likely not contribute to internationalization efforts.
4.2 Data analysis
Table 2 exhibits the descriptive statistics with the corresponding correlations and significance levels. It is noteworthy to mention that the issue of missing dependent variable data has been resolved by a listwise deletion to keep the integrity of the research. As a result, the sample size has decreased from 1816 to a still respectable 582. Multicollinearity has unsurprisingly been established between both dependent variables, Scale of Internationalization and Depth of Internationalization as the corresponding correlation coefficient was slightly higher than 0.8 and there was a p-value of lower than 0.01, thus suggesting high significance (Signori, 2016).
Although Scale of Internationalization and Depth of Internationalization are highly correlated their means seem to differ significantly, 0.370 vs. 0.191. Also, the mean of R&D Collaboration is 0.324, which would indicate that on average the sample spends 32.4% of their R&D expenses externally. The average age of the sample is 34.69 years, and the eldest firm within the sample is 119 years old. Family-run firms represent 43% of the sample, and surprisingly 58% of the firms are listed. Additionally, 96% of the 582 firms regard their products to be standardized, whereas the average number of product segments firms branch and diversify out to is 0.270.
Further significant findings to be derived from Table 2 are the following; there is a negative correlation between Depth of Internationalization and R&D
Collaboration, namely -0.071 (p<0.10). Notable is the fact that there is no significant
correlation between R&D Collaboration and the other proxy for internationalization, namely Scale of Internationalization. Not only was Depth of Internationalization significantly negatively correlated with R&D Collaboration, so were Diversification with -0.028 (p<0.10) and the act of being Listed with -0.091 (p<0.05). Age and
Ta bl e 2 De sc ri pt iv e st at is tic s an d lis tw is e co rr el at io ns (N= 58 2) !! !! 1 2 3 4 5 6 7 8 9 10 11 12 13 1 In t'l S ca le 1 !! !! !! !! !! !! !! !! ! ! !! !! 2 In t'l D ep th 0. 819 ** 1 ! ! ! ! ! ! ! ! ! ! ! 3 R& D Co lla bo ra tio n -0. 039 -0. 071 † 1 ! ! ! ! ! ! ! ! ! ! 4 Ge ar in g 0. 055 0. 028 0. 069 1 ! ! ! ! ! ! ! ! ! 5 R& D Co lla bo ra tio n x G ea ri ng 0. 020 0. 033 0. 009 -0. 019 1 ! ! ! ! ! ! ! ! 6 Di ve rs ifi ca tio n -0. 024 -0. 019 -0. 028 † 0. 019 0. 009 1 7 R& D Co lla bo ra tio n x D iv er si fic at io n 0. 010 -0. 015 -0. 053 -0. 001 0. 00 3 -0. 058 † 1 8 Ag e 0. 062 0. 042 0. 009 -0. 076 † -0. 036 0. 055 -0. 062 † 1 9 Em pl oy ee s 0. 244 ** 0. 121 ** 0. 011 0. 193 ** 0. 047 0. 074 † -0. 007 0. 181 ** 1 10 For ei gn Pa te nt s 0. 080 † 0. 057 -0. 052 0. 012 -0. 012 0. 071 † -0. 036 0. 128 ** 0. 166 ** 1 11 Fa m ily O w nersh ip -0. 125 ** -0. 092 * 0. 030 -0. 108 ** -0. 049 -0. 023 0. 022 0. 025 -0. 190 ** 0. 064 1 12 Li st ed -0. 149 ** -0. 117 ** -0. 091 * -0. 155 ** -0. 036 0. 018 0. 035 0. 098 * 0. 024 0. 043 0. 092 * 1 13 St an da rdi za tio n -0. 121 ** -0. 085 * -0. 05 -0. 023 0. 035 * -0. 031 0. 032 -0. 042 -0. 130 ** 0. 001 0. 081 † 0. 012 1 Me an 0. 370 0. 191 0. 324 2. 058 0. 012 0. 270 -0. 006 34. 690 2. 219 0. 045 0. 430 0. 580 0. 960 St d. de v. 0. 310 0. 210 0. 360 0. 480 0. 172 0. 618 0. 208 21. 615 0. 564 0. 206 0. 495 0. 495 0. 191 Mi n 0. 000 0. 000 0. 000 0. 300 -0. 560 0. 000 -0. 720 1. 000 0. 780 0. 000 0. 000 0. 000 0. 000 Ma x 1. 000 1. 000 1. 000 4. 410 1. 170 2. 000 1. 020 119. 000 4. 110 2. 260 1. 000 1. 000 1. 000 † p< 0. 10 * p< 0. 05 ** p< 0. 01
likely to have a higher degree of debt to equity compared to their elder counterparts. The control variable Employees is all positively correlated with Scale of
Internationalization 0.244 (p<0.01), Depth of Internationalization 0.121 (p<0.01), Gearing 0.193 (p<0.01), Diversification 0.074 (p<0.10), and Age 0.181 (p<0.01). The
number of Foreign Patents have shown to have a positive correlation within the sample with Scale of Internationalization 0.080 (p<0.10), Diversification 0.071 (p<0.10), Age 0.128 (p<0.01), and Employees 0.166 (p<0.01). Being a family-run firm seems to have a contra productive effect on internationalization as the negative correlation with; Scale of Internationalization -0.125 (p<0.01), and Depth of
Internationalization -0.092 (p<0.05), but also Gearing -0.108 (p<0.01), and Employees -0.190 (p<0.01) would suggest. Being as Listed is not much better as do
the correlations with Scale of Internationalization -0.149 (p<0.01), Depth of
Internationalization -0.117 (p<0.01), and Gearing -0.155 (p<0.01) imply. On the
other hand being listed does positively correlate with Age 0.098 (p<0.05) and Family
Ownership 0.092 (p<0.05). Finally, Standardization appears to negatively correlate
with both dependent variables, Scale of Internationalization -0.121 (p<0.01) and
Depth of Internationalization -0.085 (p<0.05), and Employees -0.130 (p<0.01). A
positive correlation to a lesser extent can be found between Standardization and
Family Ownership 0.081 (p<0.10).
4.3 Results
A regression analysis has been conducted based on the Ordinary Least Squares (OLS) method of which the findings have been depicted in Table 3 and 4. Each table presents 3 models. Table 3 attends to the relationship between R&D Collaboration and Scale of Internationalization, while Table 4 covers the relationship between R&D
considers the control variables. Model 2 adds the dependent variable of internationalization (Scale for Table 3 and Depth for Table 4) to the equation. The full set of hypotheses (1-4), as previously mentioned, are tested in last remaining model, which includes the full sample. This model includes both the moderating variables
Gearing and Diversification as well as the interaction variables R&D Collaboration x Gearing and R&D Collaboration x Diversification. The Listed and Standardization
effects have been included in both regression analyses as a mean of control.
Based on the findings as depicted in both tables no all variables have a significant effect on the dependent variable. In Model 3, in both Table 3 and 4, the moderators and interaction variables seem not to have scored a P-value of less than 0.10, thus not been able to pass the most liberal of thresholds for significance. Although the additional variables of both Model 3, in both Table 3 and 4, did not
seem to have added value due to the lack of significance, it did increase the R2
ever so slightly from 0.102 to 0.104. However, because there were no significant contributions made by the added variables of either Model 3, in both Table 3 and 4,
one might contribute the nonsignificant increase of R2
to this.
Table 3 R&D Collaboration and Scale of Internationalization
Model 1 Model 2 Model 3
R&D Collaboration -0.048 (-0.034) -0.048 (-0.034)
Gearing -0.009 (-0.027)
Gearing x R&D Collaboration -0.006 (0.072)
Diversification -0.024 (0.020)
Diversification x R&D Collaboration 0.029 (0.059)
Age 0.000 (0.001) 0.000 (-0.001) 0.000 (0.001) Employees 0.115** (0.023) 0.116** (0.023) 0.119** (0.024) Foreign Patents 0.079 (0.061) 0.075 (0.061) 0.080 (0.061) Family Ownership -0.043† (0.026) -0.041 (0.026) -0.042† (0.026) Listed -0.095** (0.025) -0.098** (0.025) -0.100** (0.025) Standardization -0.138** (0.065) -0.142* (0.065) -0.145* (0.065) R2 0.099 0.102 0.104 Adj. R2 0.089 0.091 0.087 F-stat 10.476** 9.275** 6.047** † p<0.10 * p<0.05 ** p<0.01
Briefly looking a the control variables, Table 3 leads to suggest that the
are the prime predictors of a firm’s Scale of Internationalization. Furthermore, all control variables but Age and the number of Foreign Patents seem to be significantly related to the proxy of internationalization. Both Family Ownership and the act of being Listed appear to have a negative coefficient on a firm’s internationalization.
(1) Yi = β0 + γX + ε
X = all control variables
Model 2 includes the independent variable R&D Collaboration and shows there is a non-significant negative relationship with a coefficient of -0.048 to the Scale of
Internationalization.
(2) Yi = β0 + β1R&D Collaboration + γX + ε
Model 3 furthermore includes the moderators and interactions terms which all seem to be non-significant and have a negative coefficient. Only Diversification x R&D
Collaboration has a positive coefficient. Thus I failed to show a significant coefficient
between R&D Collaboration and Scale of Internationalization. Also, Gearing and
Diversification have a nonsignificant negative coefficient with internationalization,
unlike previously assumed.
(3) Yi = β0 + β1R&D Collaboration + β2Gearing + 0Gearing x R&D
Collaboration + β3Diversification + 1Diversification x R&D
Collaboration + γX + ε
As the given value for R&D Collaboration in Table 3 has a P-value of more than 0.10 and therefore is not significant, there is no other option than to conclude that I have failed to reject the hypothesis due to a lack of significance.
In line with the result of Table 3, a Curve Fit graph (Figure 3) has been
constructed within SPSS,
which depicts a nonsignificant negative relationship between
both dependent and
independent variable in a linear, quadratic, and cubic manner. The absence of a substantial slope coincides
with the low R&D
Collaboration coefficient as shown in Model 2 and 3.
Table 4 R&D Collaboration and Depth of Internationalization
Model 1 Model 2 Model 3
R&D Collaboration
-0.048* (0.024) -0.049* (0.024)
Gearing
-0.004 (0.019)
Gearing x R&D Collaboration
0.023 (0.050)
Diversification
-0.012 (0.014)
Diversification x R&D Collaboration
-0.009 (0.041) Age 0.000 (0.000) 0.000 (0.000) 0.000 (0.000) Employees 0.033* (0.016) 0.034* (0.016) 0.035* (0.016) Foreign Patents 0.049 (0.043) 0.044 (0.043) 0.046 (0.043) Family Ownership Listed -0.050-0.027 ** (0.018) (0.017) -0.053-0.025 ** (0.018) (0.018) -0.053-0.025 ** (0.018) Standardization -0.072 (0.045) -0.076† (0.045) -0.075† (0.046) R2 0.040 0.047 0.048 Adj. R2 0.030 0.035 0.030 F-stat 3.984** 4.016** 2.632** † p<0.10 * p<0.05 ** p<0.01
Table 4 presents the regression analysis concerning the other proxy for internationalization, namely Depth of Internationalization. It seems that the coefficient of the control variables in Table 4 are less outspoken than their Table 3 counterparts, as a matter of fact, the variables Family Ownership and Standardization are no longer significant (p>0.10) although they do remain negative. The coefficient of Age has decreased from 0.115 to 0.033 compared to Table 3 and has fallen in
significance (p<0.05 instead of P<0.01). Listed has become the only negative coefficient (-0.050) with significance in Model 1.
Model 2 also shows differences compared to Table 3 as R&D Collaboration, although having the same coefficient of -0.048 as in Table 3, can be regarded significant (p<0.05). The negative coefficient suggests that R&D Collaboration, in fact, has a negative influence on the Depth of Internationalization of a firm. Furthermore, Standardization has become significant (p<0.10) in Model 2 and has become ever so slightly more outspoken (β = -0.076 instead of -0.072).
Like in Table 3, the additional variables of Model 3 do not show any significance what so ever, the only thing, which can be deemed noteworthy, is that the independent variable R&D Collaboration has remained significant and has become slightly more outspoken (β = -0.049 instead of -0.048).
I expected to find a significant positive coefficient for hypothesis 2, however the results show a significant negative coefficient.
As for Table 3, a Curve Fit graph (Figure 4) has been constructed for Table 4. However, the coefficient for
the independent variable is significant whereas in Table 3 it was not. The graph on the right depicts the linear,
quadratic, and cubic
interpretations of the
significant negative
coefficient. The extremely gentle slope in the graph coincides with the minor significant negative coefficient of Depth of Internationalization (β = -0.049).
The coefficients of the investigated relationships, which have been put forward for analyses in both Table 3 and 4, have shown small βs and often no significance. This indicates that there are other factors which have a greater and more significant contribution to a firm's internationalization effort than R&D Collaboration and the used moderator and control variables.
Due to the fact that in Tables 3 and 4 both moderators and the corresponding interaction terms have not shown a sufficient significance, one might say there is no indication that the moderating variables Gearing (Access to Capital) and
Diversification (Product Diversification) have a significant moderating effect. I find
no evidence to support my hypotheses 3 and 4, which states that there is a moderating effect.
To conclude the result section, a clear overview (Table 5) has been made including the significance on whether the coefficient is positive or negative regarding the variables, which have been discussed in the hypotheses.
Table 5 Hypotheses Findings
Scale (Table 3) Depth (Table 4)
Hypothesis 1 No significant coefficient (-) N/A
Hypothesis 2 N/A Significant coefficient (-)
Hypothesis 3 No significant coefficient (-) No significant coefficient (+)
5. DISCUSSION
The hypothesized relationships between R&D Collaboration and the Scale and Depth
of Internationalization have been tested in this paper. The effect of the moderator
variables and their corresponding interactions terms on these relationships have subsequently also been tested. Statistical evidence has indicated that there is a significant negative correlation when it comes to R&D Collaboration and the Depth
of Internationalization, resulting in the rejection of hypothesis 2, which is "The Relation between the Development R&D Collaboration and a Firm's Depth of Internationalization is Positive." This suggests that an increase in R&D Collaboration
collates with a decrease in Depth of Internationalization of the firms within the sample. Both coefficients of either proxy for internationalization turn out to be both -0.048 however; only the coefficient for Depth seemed to have a significant coefficient resulting in the fact that it is not possible to present conclusive evidence concerning the relationship between R&D Collaboration and the Scale of Internationalization thus no definitive outcome has been presented for hypothesis 1.
When it comes to the effect of the moderating variables the statistical evidence suggests that neither Gearing nor Diversification has a significant impact on the Scale
of Internationalization of firms. This implies that neither a higher debt to equity ratio
nor being operative in a greater number of product segments has a beneficial effect on the Scale of Internationalization. The lack of significance and the negative coefficient values corresponding to the interaction variables did contradict my previous thoughts and interpretation of the existing literature on the subject.
Different findings were found for the relationship between R&D
