The impact of geographical distance on the relationship between alliance intensity and firm performance

(1)

The impact of geographical distance on the

relationship between alliance intensity and firm

performance

By

Frank Hofman

s2327287

University of Groningen

Faculty of Economics and Business

MSc Business Administration

Strategic Innovation Management

July 2018

Supervisor: Thijs Broekhuizen

Co-assessor: John Dong

(2)

2 Abstract

(3)

3

1. Introduction

Strategic alliances are one of many strategic tools in a firm's arsenal to improve its competitive advantage and are driven by the benefits of complementary abilities. They have become the key to success in industries such as the high-tech and pharmaceutical sector (Yoon, Rosales and Talluri, 2018). Prior to the 1970’s, the pharmaceutical industry was led by and dominated by big firms. However, there has been an increasing need for highly specialized capabilities, which allowed room for smaller start-ups to thrive and play an important role in the new drug development industry. According to Rojakkers and Hagedoorn (2006), the high costs of drug development and the need for venture capital motivate pharmaceutical startups to work together with big pharmaceutical incumbents. These strategic alliances between firms have thus increased over the past years.

The transfer of knowledge between alliance partners plays an important role in the new drug development process. During the innovation process different sources of knowledge are combined (Schumpeter, 1993; Fleming, 2001). Alliance networks provide firms with a valuable source of external knowledge (Phelps, 2010) that are important to innovation (Gallego, Rubalcaba and Suárez, 2013). In addition to external knowledge, alliances provide firms with technical and market information which in turn improves firm performance (Freeman, 1991). The capability to transfer knowledge between alliance partners plays a crucial role in the success of these alliances (Schulze, Brojerdi and Krogh, 2014). Furthermore, smaller firms may choose to ally with a larger, established and more reputable firm to improve its credibility through endorsement effects (Hitt, Dacin, Levitas, Arregle and Borza, 2000). Partnering with a reputable partner improves a firm's position in international networks, increases the firm’s ability to mobilize resources in these networks and strengthens bargaining power (Chen and Chen, 2002).

According to Korbi and Chouki (2017), the ability to transfer knowledge between alliance partners is affected by the degree of specialization of the partners, the divergence of their managerial cultures, the nature of the transferred knowledge, and language and translation barriers. A common barrier that negatively affects the transfer of knowledge is the geographic distance between alliance partners (Capaldo and Messeni Petruzzelli, 2015). This is due to the higher likelihood of firms to take on opportunistic behavior and commit to ‘learning races’. Another possible non-opportunistic explanation could be decreased ease of communication such as less face to face contact. These issues are detrimental to the transfer of knowledge between partners.

(5)

5

of geographic distance. The table shows several studies that cover the influence of geographic distance on firm practices. The first two studies shown in the table have taken a knowledge-based view and study the effect of alliances on product innovation and patent citations. The third and the fourth paper study the effect of alliance experience on firm financial performance. First, this study will cover the gap between these studies by increasing the scope of the performance metric of the first two studies by focusing on firm financial performance instead of innovation. Since, geographic distance between alliances partners influences more than just factors that influence innovation performance. A larger geographic distance can result in increased risks and transaction costs, that influence the firm and are not limited to one department such as R&D. Second, by focusing on alliance intensity instead of alliance experience (3rd_{and 4}th_{study), this paper covers another construct of alliances. Alliance experience}

covers the ability of a firm to successfully learn and obtain knowledge from alliances. Alliance intensity covers the absolute number of alliance partners and the possible increasing amount of knowledge and benefits that an increase brings.

To study this research gap, this paper will explore the relationship with a firm performance outcome: return on assets. Since, the ultimate bottom line is to improve firm performance, and not the intermediate outcome of innovation. This more abstract and overarching focus is relevant for managers who want to know how the intensity of alliance partners affects firm performance, and whether the distance of such alliances matter. Previous research has not yet covered the relationship between alliance partner proximity and firm financial performance. Hence, the research question is the following:

Does the geographical distance between alliance partners influence firm performance?

The current study contributes to alliance network theory by answering the question whether the performance of firms, who participate in alliance networks, is influenced by the average geographical proximity between alliance partners. Additionally, this will provide managers with more extensive knowledge about decision making in relation to alliance partners. This study aims to provide managers with knowledge on whether firm performance increases when the number of alliance partners increases. Furthermore, by incorporating geographic distance between alliance partners, this study provides managers with additional information on which characteristics of possible future partners are important.

(6)

6

Table 1: Previous studies on geographical distance and alliance’s impact on firm performance

Authors, year

Paper focus and results Alliance vs. geographical distance study

Industry Method Inclusion of alliance intensity Performance measure Ganesan, Malter and Rindfleisch, 2005

Firms located closer together engage in more face-to-face communication, but this has little effect on the types of knowledge that enhance new product outcomes Geographical distance study U.S. Optics industry Cross-sectional survey and a longitudinal follow-up survey No Product level: New product outcomes - new product creativity - new product development speed Capaldo and Petruzzelli, 2014

The effect of partner geographic proximity on the innovative performance of knowledge-creating alliances.

Findings: geographic distance between partner firms and their affiliation with the same business group negatively influences the alliance innovative performance.

Geographical distance study 10 fortune 500 companies operating in the electronic equipment industry Negative binomial regression No Patent level: Innovation performance measured as number of patent citations Sivakumar, Roy, Zhu and Hanvanich, 2010

The effect of alliance expertise and alliance governance on firm performance in cross-border alliances. Findings: they find that prior alliance experience has a positive influence on innovation generation, diversity of partners has a negative relationship.

Alliance study U.S. Pharmaceut ical industry Fixed effect panel regression No, alliance experience Financial level: Tobin’s Q Zaheer and Hernandez, 2011

Distance between technology alliance partners and subsidiary hurt MNC performance Geographical distance study Fortune 500 firms that performed technology related activities Dynamic panel GMM No, alliance experience Financial level: ROA, lagged one year

R&D intensity

Qi Dong et al. 2017)

When the goal is breakthrough innovation, collaborating with more partners that are more central in alliance networks the better. But to a certain extent.

Alliance network study U.S. Pharmaceut ical industry Negative binomial regression Yes, as a control variable Patent level: Innovation performance measured as the number of patent citations that received forward citations above the 97th percentile of all patents within a

patent class.

(breakthrough innovations) This paper The moderating effect of

geographic distance between alliance partners on the relationship between alliance intensity and firm performance. This study found no effect of alliance intensity on firm performance. Additionally, no support was found for the moderating effect of geographic distance. both Pharmaceut ical industry Fixed effect panel regression

Yes Financial level:

(7)

7

2. Literature review

2.1 Conceptual model

The conceptual model that will be studied in this paper is based on the knowledge transfer perspective. Where resource-based theory is focused on the strategic resources of a firm that can offer a competitive advantage, the knowledge transfer perspective focusses on individual and organizational knowledge as an increasingly more critical resource of the firm (Grant, 1996). The ability to transfer and integrate knowledge has a significant effect on the performance of a firm (Martin and Salomon, 2003; Smale, 2008). This perspective is underlying to the conceptual model. The ability to transfer knowledge between alliance partners is inherent for its success. First for the XY relationship, when firms partner with more firms there is more knowledge available for transfer, which arguably improves firm performance. Second, this conceptual model poses that this relationship is influenced by the geographic distance between these firms. The success of the transfer of knowledge between firms is influenced by the geographic distance between them.

This paper builds upon the work done by Qi Dong, McCarthy and Schoenmakers (2017). They study how a firm’s position in an alliance network influences their breakthrough innovation. Instead of focusing on the relative position between firms in an alliance network, this paper studies the absolute distance between firms. Furthermore, this paper chooses a financial metric as a measurement of performance instead of a R&D metric.

2.2 Alliance intensity and firm performance

This paper defines alliance networks as ‘groups of companies linked by alliances in different forms (including loose arrangements), which compete in the market with other networks as well as single companies’ (Sroka and Hittmar, 2013:20).

Alliances provide firms with several benefits. They can provide firms with access to resources, especially when under time pressure (Teece, 1992). These resources can improve a firm's strategic position in difficult market situations (Eisenhardt and Schoonhoven, 1996). These firms are succumb to several competitive pressures, the increased speed of technological development and shortened product development lifecycles (Powell, Koput and Smith-Doerr, 1996). According to Baum, Calabrese and Silverman (2000), alliances can provide firms, particularly startups which lack access to resources and stable relationships, the tools to overcome the liabilities of newness. Additionally, partnering with

Geographic distance

(8)

8

an established firm can be a way to commercialize innovations since they possess complementary assets like distribution channels, manufacturing and marketing. Mature firms also benefit from alliances, especially in technologically-intense industries. Small firms can provide them with highly specialized capabilities. Incumbents can obtain differentiated resources through alliances, which provides them with greater flexibility and access to resources that would be too expensive to create internally (Smit and Trigeorgis, 2004). Finally, both new entrants and incumbents can partner with another reputable firm that spreads positive referrals to other firms, which can reduce the risk of partners behaving opportunistically through social control (Goerzen and Beamish, 2005; Granovetter, 2005).

Firms can benefit more from alliances when there are some technological similarities between partners. However, there should still be differences between them otherwise there is nothing to learn from the partner. Additionally, when firms are too different, partners struggle to learn from each other (Sampson, 2007). Furthermore, a firm’s internal R&D becomes more productive when they form a technology alliance with firms that operate in the same industry. When partnering with firms in unrelated industries, the internal R&D productivity can decrease. (Noseleit and De Faria, 2013).

Laursen and Salter (2006) pose the idea of search breadth. They argue that external search breadth (the number of external sources that firms rely on in their innovation activities) influences innovation performance. The higher the search breadth the better innovation performance of a firm. But they argue, based on Koput (1997), that firms can over search and that eventually the innovation performance decreases due to diminishing returns. This paper will control for the effect of diminishing returns via a robustness check.

Previous studies have found evidence that a greater alliance intensity has a positive influence on firm survival (Baum and Oliver, 1991). They argue that this is due to that partnering with more institutions offers increased stability, social support, legitimacy and access to resources. Furthermore, Powell et all. (1996), found that a greater alliance intensity positively influences firm growth rates. They argue that this is due to experience and organizational learning which improve the capabilities related to utilizing knowledge.

Based on these two papers and the previously mentioned benefits that alliances bring to a firm, this paper argues that when a firm has more alliance partners their performance increases.

(9)

9

2.3 Moderating influence of geographic proximity

According to Rosenkopf and Almeida (2003), knowledge is localized within technological and geographic context. This is due to firms being limited in their search for new knowledge by organizational and relational constraints (Jaffe, Trajtenberg and Henderson, 1993). Building upon this study, Kwon, Lee and Lee (2017) find that the localization effects of knowledge spillovers continue to grow overtime. This is unexpected when considering the rapid globalization and improvement in communication technologies over the past few years. They argue that this could be due to the recent growth in knowledge production being accompanied by a larger specialization. Therefore, closer contact might be necessary to transfer specialized knowledge since face to face contact is easier to realize. This shows that geographic distance still matters especially for hard to codify, tacit knowledge (Ganguli, Lin and Reynolds, 2017).

At the regional level, geographical distance between firms has a negative influence on knowledge flows. Knowledge flows are easier to attain within regions that share the same language than between regions which are separated by language barriers. When knowledge spillovers are limited between regions, firms must rely on smaller knowledge bases for R&D. This in turn can lead to differences in economic prosperity between regions (Maurseth and Verspagen, 2002).

When zooming in to the firm level, Torre and Rallet (2005) pose the idea of temporary geographical proximity. This idea entails that actors do not need to be in geographical proximity to each other constantly and that during their collaboration meetings, visits and temporary co-location might be adequate. It is also argued that close collaboration is only necessary during certain periods of the collaboration (Gallaud and Torre, 2004). Small firms benefit less easily from temporary geographical proximity since they have less resources to afford the high transportation costs and their inefficient human resources (Torre, 2008).

Transaction cost literature shows that transaction costs increase with increasing geographic distance between firms (Grote and Umber, 2006). Additionally, the further firms are located apart, the more time consuming and costly exchanges of goods and workers between them become.

Agency theory shows that geographic distance between firms increases monitoring and agency costs (McCarthy and Aalbers, 2016; Böckerman and Lehto, 2006).

Previous research has shown that when the geographic distance between a portfolio of alliance partners increases, the transaction costs increase while the transfer of knowledge decreases. Therefore, this paper argues that a closer geographic distance between alliance partners has a positive influence on their performance.

(10)

10

3. Methodology

3.1 Empirical setting

The empirical setting of this study is the pharmaceutical industry (SIC code: 2833-2836). The pharmaceutical industry is a knowledge-intensive industry in which alliances have been extensively studied (Baum et al.,2000; Rothaermel, 2001). Research has shown the important role that strategic alliances play in this industry and that they have been a key to success (Yoon et al., 2018). Alliances are important and frequently used in this industry, therefore making it more interesting to research in this context. Additionally, the focus is on the pharmaceutical industry as a single industry, this allows for control of industry differences. For the alliance sample, the Thomson Reuters’ Securities Data Company (SDC) Platinum Database was used.

3.2 Data and sample

The sample combined two archival data sources and one online source. First, this paper uses the Thomson Reuters’ Securities Data Company (SDC) Platinum database for the alliance sample. This database includes all alliances with an US target since 1979. This resulted in 132831 alliances between 1985 and 2010. Alliances with more than 2 partners and alliances of which none of the partners was active in the pharmaceutical industry were removed, resulting in a sample of 4852 alliances. Subsequently, this study filtered out alliances of which the geographic location was unknown. The net sample consists of 3882 pharmaceutical alliances. For these remaining 3882 alliances, firm-level financial data was gathered for each of the alliance partners. The Standard and Poor’s Compustat database was used to gather this data. The Compustat database provides financial, statistical, and market information on active and inactive global companies throughout the world (Qi Dong et al., 2017). This resulted in a firm level sample that consists of 669 firms that were operating in the pharmaceutical industry from 1985 until 2005.

(11)

11

3.3 Key measures

3.3.1 Dependent variable - Return on assets

This paper relies on return on assets (ROA) to measure the focal firms’ performance (Dze and Soldi, 2011; Lin, Yang and Arya, 2009). The use of ROA as a performance metric allows for capture of the objective economic performance of a firm and for easier comparison with previous research. Following prior research (Rust, Moorman and Dickson, 2002; Luo, Rindfleisch and Tse, 2007), this paper measures financial performance as return on assets lagged by one year, as the impact of alliances takes time to take effect. The ROA was calculated by dividing the operating income before depreciation by total assets. The return on assets are analyzed as t+1 (i.e., one year after the start of the alliance). In this paper, a financial performance indicator was chosen instead of an innovation performance indicator. This is due to the different types of alliances (strategic, marketing, manufacturing, licensing and funding) being included and that these possibly will influence more than just innovation performance (Bos, Faems and Noseleit, 2017; Jiang, Tao and Santoro, 2010).

3.3.2 Independent variable - Alliance intensity

The independent variable, the firm’s alliance intensity, is measured as the number of alliances a firm has started in a given year (Hoffmann, 2007). This paper measures alliance intensity per year instead of all ongoing alliances because of a lack of data about the termination date of the alliances. A control for this measurement was performed with a robustness test which measured alliances intensity of active alliances.

3.3.3 Moderating variable - Geographic proximity

The geographic proximity between two alliance partners will be calculated with the haversine formula (Chopde and Nichat, 2013). This formula shows the shortest distance between two points on the earth’s surface and takes into account the curvature of the earth’s surface.

Geographic distance = r × arcos [sin(latacq) × sin(lattgt) + cos(latacq)× cos(lattgt) × cos(lontgt− lonacq)]

a = sin²(Δφ/2) + cos φ 1 ⋅ cos φ 2 ⋅ sin²(Δλ/2) c = 2 ⋅ atan2( √a, √(1−a) )

d = R ⋅ c

Where: φ= latitude, λ= longitude, R= the earth’s radius

(12)

12

measures for geographic distance are assessed. For instance, control for other compositions of geographic distance namely, using the shortest distance and longest distance between partners.

3.3.4 Control variables

Several control variables were selected that could possibly impact firm performance. The inclusion of control variables provides a more robust test of the hypotheses. Firstly, according to Vithessonthi and Racela (2016), a higher R&D intensity negatively influences the firm’s performance due to high degrees of uncertainty and risk associated with these activities. Therefore, controlling for R&D intensity, measured as the R&D expenditure divided by total assets. Second, the effect of financial leverage on firm performance has been widely debated. With authors arguing for a positive influence on firm performance (Lubatkin and Chatterjee, 1994), while others arguing for a negative effect (Balakrhishnan and Fox, 1993; King and Santor, 2008) and some for no significant effect (Phillips and Sipahioglu, 2004). The financial leverage is calculated by dividing the total debt by total assets. Third, this paper controls for a firm's prior performance. This is measured as the operating income before depreciation divided by the total assets. The fourth control variable is firm size since it has a significant positive effect on firm performance (Pervan and Višić, 2012). Firm size was measured by the natural logarithm of total sales. Lastly, to capture the aggregate effect of time this paper controls for the fixed effect of time.

3.4 Analysis strategy

The independent variable is a count variable and the dependent variable is a continuous variable. Since the data is collected over time, over the same entities and that the observations are not independent, a panel data regression analysis was used to test the hypothesis. Panel data allows control of variables which cannot be observed or measured, like differences in business practices across companies (Baltagi, 2008). It accounts for individual firm heterogeneity. For testing the first hypothesis, ROA was included as the DV, alliance intensity as the IV and the control variables. To test the second hypothesis which includes the moderator, alliance intensity was regressed on ROA and include geographic distance (MOD) and the interaction term (MODxIV).

Table 2 shows a summary of the variables and their correlations. All the variables are included. It shows that the firms vary widely in size, number of alliance partners and distance between alliance partners. Interesting to note is the significant positive correlation between ROA and alliance intensity. It shows that when a firm has more alliance partners their ROA increases simultaneously. Additionally, all 4 control variables show a significant correlation with ROA. Lastly, there appears to be no significant correlation between the geographic distance between alliance partners and ROA .

(13)

13

The multicollinearity of the variables is low which means that there is no strong inter-association among the independent variables.

Before testing the hypothesis, it is needed to determine if a fixed effect model or a random effect is more appropriate. To test this, a fixed effect panel regression and a random effect panel regression were performed. The results of which were stored in Stata. Following these analysis, a Hausman test was performed. The results of this test are shown in Appendix 1. When looking at these results, it can be determined whether to perform a random or fixed effects model. When using a level of significance of 5%, the null hypothesis of random effects was rejected, since prob>chi = 0,0021. Therefore, a fixed effect panel regression analysis is preferred. The benefit of a fixed effect model is that it can better control for endogeneity.

For testing the first hypothesis, a panel data regression analysis including ROA, alliance intensity and the control variables was performed. For testing the second hypothesis, another panel data regression analysis was performed. However, now including the average geographic distance and the interaction term. The interaction term was computed by multiplying the average geographic distance between alliance partners by alliance intensity.

Table 2: descriptive statistics and correlations

(14)

14

4. Results

4.1 Hypothesis testing

Table 4 reports the results of the fixed effect model that includes for the first model ROA, alliance intensity and the control variables: R&D intensity, financial leverage, prior performance and firm size. The fixed-effects model controls for all time-invariant differences between the firms, so the estimated coefficients of the fixed-effects models cannot be biased because of omitted time-invariant characteristics (Kohler and Kreuter, 2005). This analysis is used to test the first hypothesis: The more

alliance partners a firm has, the higher the firm’s performance. The model shows that the variable

alliance intensity is insignificant (p = 0.579). This means that the coefficients are not statistically significantly different from zero. Since the two-tail p-value of alliance intensity is insignificant, there was no significant relationship found between alliance intensity and return on assets, and therefore the first hypothesis was rejected.

The second model reports the results of the fixed effect model that includes ROA, alliance intensity, geographic distance and the interaction term. This model is used to test the second hypothesis:

the geographically closer alliance partners are located to each other, the higher the positive effect of alliance intensity on firm (financial) performance. The model shows that the interaction term

(geographic proximity x alliance intensity) is not statistically significant (p = 0.741). This proves that the coefficients are not statistically significantly different from zero and that there is no moderating effect of geographic distance. Therefore, the second hypothesis must be rejected and can be concluded that the geographic distance between alliance partners does not significantly influence the relationship between alliance intensity and return on assets.

(15)

15

4.2 Robustness checks

Several robustness checks were performed to test whether the coefficients are reliable and robust, this will provide evidence of structural validity (Lu and White, 2014). In the conceptual model, a linear relationship is proposed between alliance intensity and return on assets. However, several authors argue that when the number of alliance partners increases the benefits that they bring will eventually decrease (Goerzen & Beamish, 2005; Deeds & Hill, 1996; Grant & Baden-Fuller, 2004). To test for the possibly diminishing returns (knowledge overlap, limited absorptive capacity) and increased coordination costs. A test for a possible inverted U-shape relationship between alliance intensity and alliance performance was performed. The variable alliance intensity^2 was created and inserted to test for this relationship. Therefore, re-testing hypothesis 1, where alliance intensity^2 replaced alliance intensity. The outcome of this regression was not significant (p=0.886). Therefore, it can be concluded that there is no inverted U-shape relationship between alliance intensity and alliance performance.

The second robustness test incorporates the logarithmic function of geographic distance. Therefore, simplifying the model and the interaction term. For this test, the variable LNgeographic distance was created. This new variable replaced geographic distance in the second fixed effect panel regression model. The outcome shows that there is no significant effect for LNgeographic distance: p

= 0.391 and p = 0.731 for the new interaction term.

The third robustness tests only includes firms with less than three alliance partners. This allows for control of the influence of a firm's alliance portfolio size. Geographic distance is more impactful on firms with less alliance partners since the influence of one located further away partner diminishes when the number of closer partners increases. To test this, a fixed effect panel regression was performed with firms that have only 1 or 2 alliance partners. The results of these tests were insignificant and shows that there is no influence of geographic distance and that this is the same for firms with many partners as for firms with few partners.

The fourth robustness test covers the duration of alliances. The average alliance duration is 36 months (Pangarkar, 2003; Phelps, 2003). This is of course heavily subjected to change between firms, but it provides an estimate. When substituting alliance intensity with the cumulative alliance variable in both fixed effects panel regression models the outcome is not significant. Model 1: p = 0.492 and p

= 0.720 for the second model. These results show that alliance intensity per year and cumulative alliance

intensity does not influence firm financial performance.

(16)

16

smallest firms from the sample. Firms size was measured as total sales in a year. Two fixed effect panel data regression analysis where performed with these new samples. For both the groups the outcome is not significant. The p-values for the 20% smallest firms was p = 0.282 for the moderator and p = 0.844 for the interaction term. For the 20% largest firms the p-values were p = 0.775 for the moderator and p

= 0.929 for the interaction term. Therefore, there was no difference in effects for the smallest and largest

firms, both of their ROA are not influenced by alliance intensity and the average geographic distance between alliance partners.

The sixth robustness test consists of two parts. The first part covers the shortest distance between alliance partners per year and the second part covers the longest distance between alliance partners per year. These tests were performed to study a possible positive relationship between the shortest distance between partners on ROA which would not come forward when measuring the average distance. Furthermore, also to study a possible negative influence of the largest distance between partners on ROA, possibly due to more time spend on the partner and increased expenses. The outcome of both these tests was insignificant. The model that incorporated the shortest distance resulted in p =

0.866 and p = 0.761 for the new interaction term. The model that included the longest distance resulted

(17)

17

5. Discussion

The aim of this study was clarifying the relationship between alliance intensity and firm performance and to study the moderating effect of geographic distance between alliance partners on this relationship. The mechanisms underlying these relationships have been set out and elaborated. Previous studies found either a positive influence (Stuart, 2000; Sarkar, Echambadi and Harrison, 2001) or negative influence (Callahan and Smith, 2006) of alliances on firm financial performance. Previous research mostly covers alliance experience, this study covers distance between alliance partners. It appears that when you take the average distance of the portfolio of alliance partners started the previous year, the alliance intensity has no impact on firm performance. This is the same for firms that have partners located further away as for firm that have partners located close by. Remarkably, no support was found for either of these effects. This allows for answering the research question. In this setting geographic distance between alliance partners does not influence firm financial performance.

Several reasons could exist for the insignificance of intensity on firm performance. A reason for this lack of influence could be the goal of the alliance, within our sample there could be relatively many alliances of which the goal was not to improve firm performance. As discussed in the literature review, firms partake in alliances for numerous reasons: obtaining complementary assets, gathering knowledge, increase their reputation and resources (Baum et al, 2000). It could be the case that increasing financial firm performance by partaking in alliances was not the key goal of the firms in the sample. Another possible explanation is that the alliance simply does not contribute, so firm performance is not influenced.

The hypothesis was posed that an average larger geographic distance between alliance partners has a negative effect on the relationship between alliance intensity and firm financial performance. This paper argued that this is due to limited knowledge flows (Maurseth and Verspagen, 2002) and increasing transaction costs (Grote and Umber, 2006) that come with a higher geographic distance. The analyses have shown no support for this relationship and therefore that there is no impact of geographic distance between alliance partners on the relationship between alliance intensity and firm financial performance.

(18)

18

5.1 Theoretical implications

This paper adds incrementally to the literature by examining the relationship between alliance intensity and firm financial performance and the influence of geographic distance on this relationship by studying 669 firms in the pharmaceutical industry. Even though the results were not statistically significant, this paper still contributes to theory on several areas. First, previous research suggested that partaking in alliances would be beneficial for firm performance. No evidence was found for this claim in this setting. Additionally, this paper has shown that geographic distance does not play a role in the financial success of alliances in the pharmaceutical industry. A possible explanation for this could be that pharmaceutical firms are able to overcome cross national language barriers that arise when partnering with firms abroad. Employees in the pharmaceutical industry are highly educated and therefore, might be able to deal with these barriers. Another possible explanation for the insignificance of the findings might be the codifiability of industry specific knowledge. Industry specific knowledge in the pharmaceutical industry consists of the development of drugs which is based on knowledge of chemistry and biology. It could be that expertise knowledge on these areas is easy to codify, decode and transmit, and therefore easy to transfer across larger geographic distances.

Second, robustness checks to control for several different compositions of the dependent and moderator variables were performed. This was done to ensure that the findings are not due to different measurements of these variables. These tests have shown that the number of active alliance partners did not influence the financial performance of firms. This was confirmed by performing a robustness check with firms that have less than three partners.

Third, a robustness check to test for an inverted U-shape relationship between alliance intensity and firm performance was performed, which was posed by several authors (Goerzen & Beamish, 2005; Deeds & Hill, 1996). This paper contributes to the literature by finding no statistically significant proof for a linear or inverted u-shape relationship between alliance intensity and firm performance in the pharmaceutical industry.

5.2 Managerial implications

(19)

19

such as language barriers, limited knowledge flows and increased transaction costs (Maurseth and Verspagen, 2002; Grote and Umber, 2006).

5.3 Limitations and directions for further research

There are several limitations to this study which can inspire future research. Given that the focus was on one industry and one type of collaboration the generalizability of this study is limited. Future research is needed to further explore the influence of geographic proximity between firms. First, scholars could focus on different knowledge intensive industries to study if they find similar results, like the biotechnology sector. Second, to broaden the scope of this subject future research could focus on different interfirm cooperation structures. Firms have several ways for gathering external knowledge such as mergers or joint ventures, R&D collaborations, cross-patenting. By doing so it can provide managers with additional knowledge on which cooperation structure would suit them best and if geographic distance between firms plays a role in any of these inter firm co-operations.

In terms of measurement, First, the variable geographic distance between firms was measured as the absolute distance between the cities in which the firm originates as done by previous studies. However, including geographic distance as a variable that is measured as the travel time between locations might provide additional closure about the insignificance of the results. Absolute distance between partners can be the same but the travel time can vary widely due to infrastructure, access to airports and border policies. By doing so, future research further explores the potential influence of geographic distance between partners.

This number of variables used in this study were limited. The main argument for the negative influence of geographic distance between alliance partners on firm performance are barriers to the transfer of (tacit)knowledge. Previously, the codifiability of knowledge in this industry was discussed. In this study, it was neither included as a variable nor did it investigate multiple industries that vary in terms of tacitness. Future research could include multiple industries that vary in terms of knowledge tacitness such that the influence of geographic distance can be assessed under different conditions. By doing so, industry differences in relation to codified and tacit knowledge can be studied. In addition to testing the codifiability of knowledge as a possible mechanism underlying the relationship between geographic distance and firm performance.

In terms of measurement of the variables, future research could focus more on intermediate goals of alliances such as innovation performance (e.g., measured by patents, or number of product innovations introduced) instead of focusing on bottom line financial goals.

(20)

20

6. References

Balakrishnan, S., & Fox, I. (1993). Asset specificity, firm heterogeneity and capital structure.

Strategic Management Journal, 14(1), 3-16.

Baltagi, B. (2008). Econometric analysis of panel data. John

Wiley & Sons.

Baum, J. A., Calabrese, T., & Silverman, B. S. (2000). Don't go it alone: Alliance network

composition and startups' performance in Canadian biotechnology. Strategic

management journal, 267-294.

Baum, J. A., & Oliver, C. (1991). Institutional linkages and organizational

mortality. Administrative science quarterly, 187-218.

Böckerman, P., & Lehto, E. (2006). Geography of domestic mergers and acquisitions (M&As):

Evidence from matched firm-level data. Regional Studies, 40(8), 847-860.

Bos, B., Faems, D., & Noseleit, F. (2017). Alliance Concentration in Multinational Companies:

Examining Alliance Portfolios, Firm Structure, and Firm Performance. Strategic

Management Journal, 38(11), 2298-2309.

Callahan, C. M., & Smith, R. (2006). Firm Partnerships and Alliances: The Impact of

Partnering Relationships on Operating Risk and Financial Performance.

Capaldo, A., & Petruzzelli, A. M. (2014). Partner geographic and organizational proximity and

the innovative performance of knowledge‐creating alliances. European Management

Review, 11(1), 63-84.

Capaldo, A., & Messeni Petruzzelli, A. (2015). Origins of knowledge and innovation in R&D

alliances: a contingency approach. Technology Analysis & Strategic Management,

27(4), 461-483.

Chen, H., & Chen, T. J. (2002). Asymmetric strategic alliances: A network view. Journal of

Business Research, 55(12), 1007-1013.

Chopde, N. R., & Nichat, M. (2013). Landmark based shortest path detection by using A* and

Haversine formula. International Journal of Innovative Research in Computer and

Communication Engineering, 1(2), 298-302.

Deeds, D. L., & Hill, C. W. (1996). Strategic alliances and the rate of new product

development: An empirical study of entrepreneurial biotechnology firms. Journal of

(21)

21

Dong, J. Q., & Yang, C. H. (2015). Information technology and organizational learning in

knowledge alliances and networks: Evidence from US pharmaceutical industry.

Information & Management, 52(1), 111-122.

Dze, C. J., & Soldi, A. (2011). Strategic Alliances: Performance Measurement in the Financial

Service Industry. Case study: The Beneficial Life Insurance SA and Microfinance

Institutions in Cameroon, Var-terminen/Spring.

Eisenhardt, K. M., & Schoonhoven, C. B. (1996). Resource-based view of strategic alliance

formation: Strategic and social effects in entrepreneurial firms. Organization

Science, 7(2), 136-150.

Fleming, L. (2001). Recombinant uncertainty in technological search. Management science,

47(1), 117-132.

Freeman, C. (1991). Networks of innovators: a synthesis of research issues. Research policy,

20(5), 499-514.

Gallaud, D., & Torre, A. (2004). Geographical proximity and circulation of knowledge through

inter-firm cooperation. In Academia-Business Links (pp. 137-158). Palgrave Macmillan

UK.

Gallego, J., Rubalcaba, L., & Suárez, C. (2013). Knowledge for innovation in Europe: The role

of external knowledge on firms' cooperation strategies. Journal of Business Research,

66(10), 2034-2041.

Ganesan, S., Malter, A. J., & Rindfleisch, A. (2005). Does distance still matter? Geographic

proximity and new product development. Journal of Marketing, 69(4), 44-60.

Ganguli, I., Lin, J., & Reynolds, N. (2017). The Paper Trail of Knowledge Spillovers: Evidence

from Patent Interferences (No. 17-44).

Goerzen, A., & Beamish, P. W. (2005). The effect of alliance network diversity on

multinational enterprise performance. Strategic Management Journal, 26(4), 333-354.

Gomes-Casseres, B. (1994). Group versus group: How alliance networks compete. Harvard

Business Review, 72(4), 62-66.

Grant, R. M. (1996). Toward a knowledge‐based theory of the firm. Strategic management

journal, 17(S2), 109-122.

Grant, R. M., & Baden‐Fuller, C. (2004). A knowledge accessing theory of strategic alliances.

Journal of Management Studies, 41(1), 61-84.

Granovetter, M. (1985). Economic action and social structure: The problem of embeddedness.

(22)

22

Grote, M. H., & Umber, M. P. (2006). Home biased? A spatial analysis of the domestic merging

behavior of US firms (No. 161). Working Paper Series: Finance & Accounting.

Hitt, M. A., Dacin, M. T., Levitas, E., Arregle, J. L., & Borza, A. (2000). Partner selection in

emerging and developed market contexts: Resource-based and organizational learning

perspectives. Academy of Management Journal, 43(3), 449-467.

Hoffmann, W. H. (2007). Strategies for managing a portfolio of alliances. Strategic

management journal, 28(8), 827-856.

Howells, J. R. (2002). Tacit knowledge, innovation and economic geography. Urban studies,

39(5-6), 871-884.

Jaffe, A. B., Trajtenberg, M., & Henderson, R. (1993). Geographic localization of knowledge

spillovers as evidenced by patent citations. The Quarterly journal of Economics,

108(3), 577-598.

Jiang, R. J., Tao, Q. T., & Santoro, M. D. (2010). Alliance portfolio diversity and firm

performance. Strategic Management Journal, 31(10), 1136-1144.

King, M. R., & Santor, E. (2008). Family values: Ownership structure, performance and capital

structure of Canadian firms. Journal of Banking & Finance, 32(11), 2423-2432.

Knoben, J., & Oerlemans, L. A. (2006). Proximity and inter‐organizational collaboration: A

literature review. International Journal of Management Reviews, 8(2), 71-89.

Kohler, U., & Kreuter, F. (2005). Data analysis using Stata. Stata press.

Ku, Y. Y., & Yen, T. Y. (2016). Heterogeneous effect of Financial Leverage on Corporate

Performance: A quantile regression analysis of Taiwanese companies. Review of

Pacific Basin Financial Markets and Policies, 19(03), 1650015.

Kwon, H. S., Lee, J., Lee, S., & Oh, R. (2017). Knowledge spillovers and patent citations:

trends in geographic localization, 1976-2015 (No. CWP55/17). Centre for Microdata

Methods and Practice, Institute for Fiscal Studies.

Lin, Z. J., Yang, H., & Arya, B. (2009). Alliance partners and firm performance: resource

complementarity and status association. Strategic Management Journal, 30(9),

921-940.

Lu, X., & White, H. (2014). Robustness checks and robustness tests in applied economics.

Journal of Econometrics, 178, 194-206.

(23)

23

Luo, X., Rindfleisch, A., & Tse, D. K. (2007). Working with rivals: The impact of competitor

alliances on financial performance. Journal of marketing research, 44(1), 73-83.

Martin, X., & Salomon, R. (2003). Knowledge transfer capacity and its implications for the

theory of the multinational corporation. Journal of International Business

Studies, 34(4), 356-373.

Maurseth, P. B., & Verspagen, B. (2002). Knowledge spillovers in Europe: a patent citations

analysis. The Scandinavian journal of economics, 104(4), 531-545.

McCarthy, K. J., & Aalbers, H. L. (2016). Technological acquisitions: The impact of geography

on post-acquisition innovative performance. Research Policy, 45(9), 1818-1832.

Pangarkar, N. (2003). Determinants of alliance duration in uncertain environments: The case

of the biotechnology sector. Long Range Planning, 36(3), 269-284.

Pervan, M., & Višić, J. (2012). Influence of firm size on its business success. Croatian

Operational Research Review, 3(1), 213-223.

Phelps, C. C. (2003). Technological exploration: A longitudinal study of the role of

recombinatory search and social capital in alliance networks (Doctoral dissertation,

New York University, Graduate School of Business Administration).

Phelps, C. C. (2010). A longitudinal study of the influence of alliance network structure and

composition on firm exploratory innovation. Academy of management journal, 53(4),

890-913.

Phillips, P. A., & Sipahioglu, M. A. (2004). Performance implications of capital structure:

evidence from quoted UK organisations with hotel interests. The Service Industries

Journal, 24(5), 31-51.

Powell, W. W., Koput, K. W., & Smith-Doerr, L. (1996). Interorganizational collaboration and

the locus of innovation: Networks of learning in biotechnology. Administrative science

quarterly, 116-145.

Qi Dong, J., McCarthy, K. J., & Schoenmakers, W. W. (2017). How central is too central?

Organizing interorganizational collaboration networks for breakthrough innovation.

Journal of Product Innovation Management, 34(4), 526-542.

Roijakkers, N., & Hagedoorn, J. (2006). Inter-firm R&D partnering in pharmaceutical

biotechnology since 1975: Trends, patterns, and networks. Research policy, 35(3),

431-446.

Rosenkopf, L., & Almeida, P. (2003). Overcoming local search through alliances and mobility.

(24)

24

Rothaermel, F. T. (2001). Incumbent's advantage through exploiting complementary assets via

interfirm cooperation. Strategic management journal, 22(6‐7), 687-699.

Rust, R. T., Moorman, C., & Dickson, P. R. (2002). Getting return on quality: revenue

expansion, cost reduction, or both?. Journal of marketing, 66(4), 7-24.

Sampson, R. C. (2007). R&D alliances and firm performance: The impact of technological

diversity and alliance organization on innovation. Academy of management journal,

50(2), 364-386.

Sarkar, M. B., Echambadi, R. A. J., & Harrison, J. S. (2001). Alliance entrepreneurship and

firm market performance. Strategic management journal, 22(6‐7), 701-711.

Schulze, A., Brojerdi, G., & Krogh, G. (2014). Those who know, do. Those who understand,

teach. Disseminative capability and knowledge transfer in the automotive industry.

Journal of Product Innovation Management, 31(1), 79-97.

Schumpeter, J. A. (1939). Business cycles (Vol. 1, pp. 161-74). New York: McGraw-Hill.

Sivakumar, K., Roy, S., Zhu, J., & Hanvanich, S. (2011). Global innovation generation and

financial performance in business-to-business relationships: the case of cross-border

alliances in the pharmaceutical industry. Journal of the Academy of Marketing Science,

39(5), 757-776.

Smale, A. (2008). Global HRM integration: a knowledge transfer perspective. Personnel

Review, 37(2), 145-164.

Smit, H. T., & Trigeorgis, L. (2004). Quantifying the strategic option value of technology

investments. Montreal: 8th Annual International Real Options Theory.

Sroka, W., & Hittmár, Š. (2013). Management of alliance networks: Formation, functionality,

and post operational strategies. Springer Science & Business Media.

Stuart, T. E. (2000). Interorganizational alliances and the performance of firms: a study of

growth and innovation rates in a high‐technology industry. Strategic Management

Journal, 21(8), 791-811.

Teece, D. J. (1992). Competition, cooperation, and innovation: Organizational arrangements

for regimes of rapid technological progress. Journal of Economic Behavior &

Organization, 18(1), 1-25.

Torre, A. (2008). On the role played by temporary geographical proximity in knowledge

transmission. Regional Studies, 42(6), 869-889.

(25)

25

The impact of geographical distance on the relationship between alliance intensity and firm performance