Master Thesis The Effects of Internal R&D and External M&A on Firm Performance
The Effects of Internal R&D and External M&A
on Firm Performance in the Pharma Industry
University of Groningen Faculty of Economics and Business
Abstract: This study investigates the effects of different strategies on firm performance. The samples are from 21 countries from 1990 to 2015. According to the results, there is a significant difference in effects of strategies on firm performance. In addition, external M&A strategy is more productive and effective to improve firm performance than internal R&D growth strategy. The findings suggest that firm size could moderate the relationship between different strategies and firm performance. Also, if companies are from different types of markets, i.e., emerging markets and developed markets, the effects of growth strategies on firm performance will also be different.
Key words: M&A, R&D, emerging markets, firm performance, firm size.
Author: Lu Li
Student number: s3425681
Supervisor: Dr. Wolfgang Bessler Assessor: Dr. R.O.S (Raymond) Zaal
1 1. Introduction
Today a large number of innovation-active companies do not rely solely on internal growth; they also obtain knowledge from the outside to develop their innovations (Rigby and Zook, 2002). In addition to developing internal growth strategy, namely doing internal research and development (R&D), there is another typical way obtaining knowledge, which is external R&D through mergers and acquisitions (M&A), R&D outsourcing (Granstrand et al., 1992). Although it is not denied that a large number of firms conduct internal and external knowledge-acquisitions simultaneously, the analysis and comparison of these two types of growth strategies (internal R&D and external R&D through M&A) are little in the academic literature. Therefore, this study will concentrate on this question, namely the comparison of internal R&D growth strategy and external R&D through M&A growth strategy. Because the requirements of complementing two strategies regarding to managerial skills and organizational structures are different, it is possible that two strategies have a different impact on firm performance (Penrose, 1959; Delmar et al., 2003; Dalton, 2006).
With this respect, while there is a great work of literature on the implications of external growth strategy on firm performance, the comparison of two strategies are neglected and overlooked. Therefore, I am interested in the question that whether the different strategies (internal vs. external) has a differential impact on the firm performance, and it is also the focus of this thesis. At the same time, the companies in pharma industry rely on innovation to grow, so the R&D spending is essential for these companies. This factor, together with the increasing number of M&As in pharma industry, makes the pharma industry relevant to this study. Overall, the aim of this study is to fill this gap by examining which strategy is more beneficial for improving firm performance using a sample of listed pharma firms. It may be useful for pharma companies to make investment decisions of whether they should pay more attention to internal growth or external growth. More importantly, this study also analyses and compares the firm performance across different countries. I examine that if the country plays an important role in deciding the relation between firm growth strategies and firm performance. Additionally, most of M&A analysis is studying U.S. firms. Therefore, this study shed some lights on the influence of countries on the relationship between strategies and firm performance based on a sample of firms from 21 countries. The importance of countries will be detected and investigated in the following analysis.
1.1 Background
2
number of M&As. The pattern has become common that larger firms acquire smaller firms to acquire some knowledge and technology, and this pattern is especially prominent in the industries with high technology, such as pharmaceutical (pharma), internet et al. The pharma industry, as an R&D intensive industry, has to focus on innovation and investments in R&D and R&D productivity. However, Rafols et al. (2012) observed a decline of the total number of research publications including published papers by large pharma firms. It should be noted that publication activities of pharma companies do not only provide insight into the R&D processes but also shed some light on dynamics in the areas of R&D efforts (Rafols et al., 2012). Therefore, the decrease in publications means that these pharma companies less concentrate on R&D efforts. At the same time, they show a relative increase of pharma companies’ external collaborations and these companies have engaged in a series of M&A activities to acquire smaller innovative drug firms. It suggests a tendency for pharma companies to either do joint research through mergers or obtain research (R&D) through acquisitions and the propensity to external R&D rather than internal research. Therefore, this phenomenon raises the question that whether M&A decision is more beneficial than internal strategy or not.
On the one hand, for pharma companies, the internal or organic growth strategy is mainly concentrating on investment on internal R&D (research and development). In addition, internal R&D investment is mainly focusing on establishing companies’ R&D systems, such as employing more researchers and developing new drugs on their owns. Firms derive their expansion internally by relying on internal strategy (Dalton, 2006). The external strategy, on the other hand, aims to search for external sources of advanced knowledge and technologies normally through M&A (merger and acquisition) activities to grow companies. Many scholars suggest that both internal and external growth strategies are important, and they could be beneficial to grow companies and create shareholder’s value (Bessler et al., 2017; Aktas et al., 2008). However, based on the different patterns and requirements of two strategies, the effects of two strategies on firm performance are likely to be different although they are both important for the growth of companies. As Aktas et al. (2008) suggested, these two types of growth strategies may have a different impact on firm performance as the requirements of strategies regarding managerial skills and organizational structures are different. Therefore, one of the goals of this thesis is to find that whether internal R&D and drug development or external R&D through M&As are more productive. In another word, which strategy will make companies grow faster and increase more shareholder value.
3 1.2.1 Pharma industry.
There are many papers focusing on the relationship between M&A and R&D, and they also describe the effects of them on firm performance (Bertrand, 2005; Cassiman, 2005). However, most of them deal with many industries, so one delimitation of this paper is focusing on a specific industry, namely pharma industry. There are several reasons why I chose pharma industry as the main research object.
First important reason is that pharma industry is an R&D intensive industry. Because the topic of this thesis is the comparison between the effects of external M&A growth strategy and internal R&D growth strategy, M&A and R&D should be comparable. In another word, both of them should have a significant impact on firm performance. Pharma industry is the industry that meet this condition. On the one hand, pharma industry is the industry that invest the most in R&D. According to the report of The Pharmaceutical Industry and Global Health (International Federation of Pharmaceutical Manufacturers & Association, 2017), the annual R&D spending by the pharma industry is greater than other high-technology industries. For example, it is 5 times greater than that of the chemicals industry, and 1.8 times more than that of the internet and software industry. In addition, R&D spending is a large part of the total costs in pharma companies, accounting for approximately 18% of revenues (Managing innovation pharma, 2013). Consequently, innovation activities and R&D spending will be important indicators to analyse pharma companies. On the other hand, M&A is also an important approach for the growth of pharma companies (Danzon, 2004). For example, the M&A deal value of pharma companies in 2017 is over 180 billion, which means that M&A has become an important investment activity in pharma industry (PwC Deals). In addition, a larger number of pharma executives suggest that employing highly skilled and motivated people to research innovative products is a big challenge for them (PwC, 2013). In this situation, some companies have to consider other ways to obtain innovation. Danzon et al. (2007) suggest that large pharma firms acquire smaller companies due to patent expirations and gaps in a firm’s product pipeline. Thus, M&A is likely to become an alternative solution when companies face innovation problems. To conclude, M&A and R&D are important investment activities for the pharma companies.
4
novel drugs but also doing clinical practices. Otherwise, they cannot achieve high sales of me-too drugs when the faced superior marketing of big pharma (Bartfai and Lees, 2013). Consequently, pharma companies need to have higher innovation capabilities, and innovation is the key determinant of the competitive advantage for the success of pharma firms. A good example to demonstrate is that in order to improve competitiveness, these companies employ a large number of university graduates to make research, and the portion of graduates is the largest in all large industries (Bartfai and Lees, 2013). It can be concluded that pharma companies emphasis on the innovation and innovative personnel. As referred in managing innovation pharma report of PwC (2013), there is a correlation between innovation and growth in pharma companies. Apart from this, the most innovative companies will have a higher growth rate than other less innovative companies. Most companies even indicate innovation as a competitive necessity. As a result, M&A and R&D play important roles in improving innovation and thereby are important to the success of pharma companies.
Next, another important reason for choosing pharma industry is some specific characteristics that other industry do not have. One of the most prominent characteristic of pharma industry is the high risk and high uncertainties. For pharmaceutical companies, they have to have their own innovative products to maintain competitive in markets. However, the process of new drug research is high-risky and the possibility of failure is uncertain (DiMasi et al., 2016). What’s more, the cost of developing a new drug is huge (Danzon et al., 2004). Once failed, companies will have no gain and no profits from the new drug development process. Nevertheless, M&A activities will moderate high-risk, allowing firms to have lower research risks (2016). Ernst and Young (2011) suggest that the cost of developing an original and new drug approaches to 900 million dollars, which is high amount of fund for pharma companies. However, they do not need to solely undertake this high R&D activity by implementing M&As. Moreover, once acquiring companies succeed in taking over acquired companies, they will acquire knowledge or patents that acquired firms hold and do not be necessary to undertake huge risks to do research. The firms could increase the innovations in short term (Jung, 2002). Consequently, M&A activities become extremely common in the pharma industry due to the motivation for acquiring knowledge and patents. The question of whether external M&A is a more productive growth strategy is worth exploring and discussing. Finally, these three reasons are the main motivations to select the pharma industry as a research object.
1.2.2 Firm size
5
are likely to become target firms. Acquirers, however, are enterprises that pay attention to long-term growth and have low R&D expenses. Therefore, I conjecture that there will be the significant difference between large growth companies and other small companies in terms of the M&A behaviours and R&D spending. Walker and Petty (1978) suggest that large growth companies are likely to have strong financial power and reinvest retained earnings rather than divide profits to shareholders. In another word, it is a decision that how companies take advantage of their earnings, and large growth companies tend to spend money on expanding companies. In contrast, small companies are likely to distribute dividends to shareholders rather than reinvest internally or externally (Chan and Chen, 1991). At the same time, if companies intend to grow for a long time, M&As are good channels for them to reinvest. Therefore, I conjecture that company’s long-term growth strategy is likely to be different between large firms and small firms. For these reasons, another important topic of this thesis is to study whether firm size can moderate the relationship between strategies and firm performance.
1.2.3 Emerging markets and developed markets
My study also suggests that pharma companies will have different firm performance if they are from different types of economies, i.e., emerging markets and developed markets. Although there is not a precise definition of emerging markets, economists usually define emerging markets as less advanced developed markets. While emerging markets typically have a physical financial infrastructure that could meet the demand of economic development, there are still gaps in terms of market efficiency between emerging markets and developed markets (Lebedev et al., 2014). However, these gaps are important for the growth of companies. Subsequently, the relationship between strategies and firm performance will be affected by these gaps. There are some studies analysing differences of M&A motivation and post-M&A performance in emerging markets and in developed markets (Lebedev et al., 2014). They suggest that since there are significant differences in institutional environments, corporate governance practices between emerging markets and developed markets, firms among emerging markets and developed countries will perform differently. Therefore, I conjecture that pharma firms will perform differently in emerging markets and developed markets.
1.3 Outline
6
In the same section, the incentives for selecting M&A as a strategy is argued in order to get an understanding for what motivates pharma companies to implement this strategy. This is followed by the analysis of the role of firm size and countries.
Next, section 3 provides the description of sample and methodology design. It begins with the description of research design and data. Because M&A is an “event”, and R&D is not an “event”, which is a lengthy process that takes place in time, the method and study are not straightforward. To solve this problem, I select companies with high R&D as internal growth firms, and firms with high M&A as external growth firms. After that, the explanation of empirical models is included. The following section 4 gives the results of the empirical study and the result of the robustness test. Finally, the conclusions and discussions are presented in section 5.
2. Literature review and hypothesis development
7
complement M&A since financial and economic performance will increase over the more emphasis on external growth strategy. Boston Consulting Group (2004) also suggests that M&A growth strategy creates superior shareholder returns by investigating M&A activities of U.S firms.
It is not surprising that both internal and external growth strategies have pros and cons (Aktas et al., 2008). For pharma companies, in-house R&D is a type of traditional growth strategy. Several rationales may explain the advantages of internal growth strategy. First, internal growth strategy could make companies have more corporate control, which may have a positive effect on encouraging internal entrepreneurship (Aktas et al., 2008). To explain, in these companies, managers normally have a better understanding and knowledge of companies’ advantages and disadvantages (Hess and Kazanjian, 2006). Besides, the investment plan will be conducted under the control of top managers. Therefore, they could make an internal investment decision based on their understandings. Second, internal growth could protect organizational culture. There is a substantive body of theories and research suggesting the negative effect of cultural differences on the post-acquisition firm performance (Dauber, 2012; Weber and Pliskin, 1996; Krishnan et al., 1997; Harris and Ravenscraft, 1991). A good case to demonstrate is the case of Pharmacia & Upjohn Merger. Because there are large cultural differences between two companies, Europe and the United States, merged entity perform a dramatic fall in the revenues (Gupta Vivek and Perepu Indu, 2007). However, internal growth companies will not be affected by problems of culture difference. In addition, internal growth is manifested in improving customer relationships and enhancing innovation capacity, leading to the improvements of an organization’s competencies and capabilities (Kling et al., 2009). Finally, Dalton and Dalton (2006) suggest that M&A is a risky activity since these activities have very different outcomes for the acquiring firms. Moreover, for the acquiring companies, they have a pay a premium demanded by the acquired firm shareholders. M&A strategy thus may be costly to exploit (Giudici and Bonaventura, 2018). In contrast, internal investment may be more controllable and less risky.
8
designs or competitive positions (Quinn, 2000). The R&D project is likely to be inconsistent with the market trend. In this case, they will be in a dilemma. On the one hand, if they give up the continuing R&D projects to make reactions according to the development of pharma markets, it means that the prior investment is useless. On the other hand, if they continue R&D projects, they may suffer greater losses due to the changes in the environment of drug markets. Therefore, internal growth companies need to tap the capabilities of external knowledge leaders for the continuous innovation and evolution of ideas, so that they can keep themselves at the frontier of the industry (Quinn, 2000). Otherwise, they will have more losses. Therefore, compared with companies that complement external M&A strategy, internal R&D companies may experience more uncertainties resulting from the changes in market environments and other factors (Bartfai and Lees, 2013).
On the other hand, M&A also has advantages. For pharma companies conducting external M&A strategy, many scholars suggest that synergies are the main motivation for M&As, and synergies can add value by combining the two company resources, which will create more opportunities that would not be available to these firms operating individually (Cassiman, 2005; Cloodt et al., 2006; Danzon, 2007; Gaughan, 2002). As suggested by Chondrakis (2016), synergies exist in technology acquisitions. What’s more, for the case of high-tech firms, the relationship between synergy effects of innovation and firm performance is positive (Lee et al., 2017). Chondrakis (2016) explains the reasons for these synergy effects are the acquiring firms can benefit from patent ownership and generate synergies with target firms. Besides, M&A can be beneficial for acquiring some specific resources (Danzon, 2007). Sometimes it is difficult for acquiring firms to copy or replicate many resources that could be useful to offer acquiring firms a competitive advantage (Capron and Dussauge, 1998), and this difficulty is more serious and more pronounced for the pharma industry (Demirbag, 2007). M&As, however, could contribute to solving this difficulty since one of the advantages of M&As is bidders could take over the specific resources of targets, acquiring a richer or broader knowledge base (Ahuja and Katila, 2001). Therefore, acquiring companies own the innovation results of the targets and could take advantage of these patents and innovation results. As a result, once bidders have held competitive patents, they do not be necessary to invest in R&D to do innovation activities. Hence, acquiring firms or merging with other companies will reduce costs by avoiding useless duplication (Ornaghi, 2009).
9
first evidence of negative impact of M&A is that M&A may intervene in the R&D process by influencing the “championing culture”, which could be beneficial for fostering innovation activities (Burgelman, 1986). Therefore, the managers’ incentives for developing novel and new product would be lower. Accordingly, the R&D inputs and outputs are also reduced, and then the firm’s long-term performance will also decline (Hitt et al., 1991). The second evidence is that the technological M&A activities can also destroy original organizational routines, and then this disruption is likely to affect the technological subsystem of the firm (Ahuja, 2001), which means that one M&A activity may have a long-term impact on the innovation activities of companies. Furthermore, it is uncertain whether this change in organizational routines are beneficial for the growth of companies or not. Another drawback of M&A is about organizational management. Datta (1991) concludes that the organizational differences between bidders and targets of M&As have an impact on post-M&As performance. He suggests that different styles of top management between bidders and targets may cause many diversifications, such as different risk-taking propensities, different approaches to decision-making and control and communication patterns. These differences may contribute to “cultural ambiguity” (Lewis, 1985), a situation in which whether the culture of bidders or that of targets will dominate. Generally, the bidders management will dominate and impose their style on the management of the targets. This can lead to a loss of identity among targets and also result in declining productivity and poor post-acquisition performance of bidders (Hirsch and Andrews, 1983; Ivancevich et al., 1987).
Based on statements above, the two types of growth strategies are likely to place different demands on managers, and these paths to growth may also have a differential impact on firm performance. Delmar (2003) suggest that because of the heterogeneous nature of two growth strategies (internal and external), these two patterns of growth can differ significantly and have different causes. Thus, it can also be inferred that different strategies have its advantages and disadvantages that the other do not possess. From this follows the first hypothesis:
Hypothesis 1: Different growth strategies will have different effects on firm performance in the pharma industry.
10
M&A and non-M&A activities, and they found that firms with M&A perform relatively better than their non-M&A rivals. Based on the argument of Demirbag (2004), I will only focus on some characteristics of pharma industry and compare these two strategies. Because internal growth is not an “event”, and the research is not straightforward, there is little research focusing on it (Aktas, 2008). In contrast, a lot of empirical studies have been made on the impacts of M&A. Therefore, I will first figure out the motivations of M&A and then analyse the reasons and opportunities of M&A to present the advantages of M&As that are hard to be achieved by performing internal strategy for pharma companies.
At first, the motivations of M&As for pharma companies primarily include increasing revenues through economies of scales and/or scope, patent expirations, and market power (Danzon, 2004). For the first commonly cited rationale for M&A, namely the economies of scale and scope, the main benefits arise from sharing fixed costs and the opportunity to exploit knowledge (Henderson et al., 1996). The economies of scale also exists in R&D and in sales and marketing (Danzon, 2004). In addition, M&A may offer cost reductions in duplicative functions, offsetting the negative impact of reduced revenues on net profits and then generating economies of scale in the long term (Malik, 2009). A good example to demonstrate this is the phenomenon that merged or acquiring firms are likely to low layoff quality and/or duplicate employees, which may lead to the reduction of labour costs (Kyeong Hun Lee et al., 2018). Furthermore, the added knowledge as well as the merged research laboratories and product pipelines from M&A could enable pharma firms to increase the likelihood of the emergence of potential blockbuster drugs (Demirbag, 2007).
11
Finally, another considerable motive is that pharma companies intend to get access to new markets or to obtain more market power through M&As, which is an effective way to deal with the competitive pressure from the global market. However, the hypothesis that the incentive of M&A is market power has been tested and rejected by several studies (Eckbo, 1983). The study of Devos et al. (2008) has proved that merger gains are obtained through better resource allocation rather than through the decline in tax payments or the increase in the market power. In addition, it is implausible for the pharma industry to pursue market power while the overall level of concentration of one industry low (Danzon, 2004). To explain, the increase in market power is normally important and effective for the industry with the high level of concentration. Although concentration is higher in some categories of pharma industry (e.g., cardiovascular), it is advocated and required by many authorities that the research about drugs should be diverse as the M&A might cause the lessen competition or even monopoly that is forbidden by many countries (Danzon, 2004). Therefore, it seems that the incentive of market power is not that important, and these theories regarding market power cannot provide support for the horizontal M&A between pharmaceutical firms.
Based on the motives of M&A analysed before, I will analyse the impact of M&A on firm performance and present some advantages of M&A that internal R&D strategy lacks. First, there is little doubt that M&As will make pharma companies save costs and value can be created by exploiting cost-based synergies (Capron, 1999). The staff costs, some fixed costs, and some common costs will accordingly be reduced (Ornaghi, 2008). Subsequently, lower costs will result in the increase of revenues, which then lead to a better performance of the firms (Angwin, 2007). For pharma industry, the largest benefit of cost reductions is reflected in the decreased duplicate costs such as redundant overhead (Demirbag, 2007). In addition, external growth companies can have synergistic benefit from M&A. For M&As in the pharma industry, the synergistic benefits mainly reflect in the combination of different knowledge inputs to be able to achieve efficiencies that are hard to obtain previously. For example, companies could share knowledge and even find inter-firm linkages, which will allow companies to achieve higher performance (Berchicci, 2010). As a result, companies could combine capabilities of two companies to create more knowledge and broaden their existing technology base, which will be helpful to improve the innovation capabilities and firm performance. However, internal growth firms cannot have cost-based synergies and synergies benefits.
12
pharma companies. However, M&As allow pharma companies to obtain patents, so that companies are less influenced by the problem of patent expirations (Danzon et al., 2007). In addition, the R&D investment costs have been increasing during recent years (DeMasi, 2016), and R&D efficiency is often higher after M&A (Cohen and Levin, 1989; Roller et al., 2001). Therefore, if companies perform M&A due to the lack of patents, pharma companies not only can avoid the high R&D investment costs as well as improve R&D efficiency, but also can hold patents to keep growing, which may enable companies to become more competitive regarding the development of new drugs and generate higher profits. Conversely, if internal growth pharma companies will soon lose patent protection on key drugs or promising products in late-stage, they have to make reactions in advance. It is because R&D activities are always considered as long-term investments, and R&D process is marked by high financial risk (DiMasi, 2016). Therefore, internal growth companies have to anticipate patent expirations, and they should respond the patent expirations by exploiting new R&D projects in advance, in order to maintain net revenue growth (Danzon, 2004). In addition, these new R&D projects may fail to result in a marketed products or products that do not succeed in obtaining marketing approval from regulatory authorities (DiMasi, 2006). Consequently, internal growth companies may face more financial risks than M&A firms if they face the problem of patent expirations.
13
What’s more, internal growth companies may face the strong inertia forces and do not adjust themselves to their changing environments, so that they are likely to be constrained actions (Capron, 1999; Capron and Mitchell, 1998). Therefore, the firm performance of internal growth companies will be negatively effected in the face of inertia forces (Rumelt, 1995).
To conclude, based on the incentives and advantages of M&As analysed above, I expect that external M&A strategy will be more effective than internal R&D strategy for pharma industry. Thus, I have the following hypothesis:
H2: External M&A strategy is more effective than internal R&D strategy in improving firm performance in pharma industry.
Phillips and Zhdanov (2013) suggest that large firms and small firms show different propensities and reasons when performing M&A activities in the pharmaceutical and biotech industry. Danzon et al. (2004) show that large firms tend to do M&A activities when facing the difficulty in patent expirations and gaps in a product pipeline. Small companies, however, merge other companies due to financial difficulties (Danzon, 2007). In addition, Moeller et al. (2003) present the existence of a size effects in firm’s acquisition announcement returns. Therefore, I conjecture that due to the different motivations to perform M&A between the large firms and the small firms, the impact of M&A on firm performance will also be different, and thereof the effects of strategies on firm performance will accordingly change over firm size. In order to answer this question, I will firstly analyse M&A strategies and internal strategies of both small firms and large firms, and then I will compare them and give possible explanations.
14 and technologies (Acs and Audretsch, 1987).
On the other hand, large companies present a different scenario. Because large pharma companies are less likely to face financial problems, they could choose the investment strategy (internal R&D or external M&A) that is most beneficial for their growth (Danzon, 2004). To explain, without the limitations of capital, firms have a wider choice to organize activities in which they utilize their specialized ability, and then they make investment decisions and projects that will generate the largest profits (Chatelain et al., 2003). Furthermore, in contrast with small companies, it is easier for larger companies to achieve economies of scale and scope as they have many competitive advantages in terms of market capital, management and market experiences (Moeller et al., 2004). Consequently, I conjecture that large companies and small companies may have different investment strategies. As a result, the effects of strategies on firm performance will also be different.
According to the investigation of Moeller et al. (2004), large firms pay more for acquisitions than small firms, and acquisitions are more profitable for shareholders of small companies. In a word, the premium paid for the acquisitions will increase with the increase in firm size. Therefore, they infer that small firms have higher abnormal returns and an increase in stock price after M&A. Penrose (1959) shows that M&A may be a response to trouble, but they are not an effective solution for large firms. Moreover, Byun and Ahn (2007) also suggest that small companies are better acquirers than large firms. Hence, I also infer that small firms gain more from M&A than large firms.
After speculating that small firms are better acquirers than large firms, I examine the possible explanations for the existence of this size effects. First, the problem of free-rider may cause a higher payment for an acquisition activity (Oliver, 1980). More specifically, because large companies are likely to acquire public targets compared with small companies, and the free-rider problem will be identified in the acquisition of a public firm, the shareholders of public firms have a better deal when the firm is acquired (Grossman and Hart, 1980). Nevertheless, this problem does not exist in acquisitions of acquiring a private firm. Therefore, the increase of the wealth of targets’ shareholders will cause the decrease in the firm value of acquiring firms. In addition, shareholders of bidder are likely to lose when buying a public firm but gain when purchasing a private firm or subsidiary (Fuller et al., 2002). To conclude, large companies may have a lower firm performance resulting from the higher propensity of acquiring public firms as well as the negative impact of acquiring a public firm.
15
that acquisitions through tender offers could create gains for both targets and bidders. In addition, these increases in wealth are not stemming from the market’s reassessment of previously undervalued securities (Bradley et al., 1988). Moeller et al. (2004) show that small firms exhibit positive synergistic gains by testing the sample of U.S. M&As between 1980-2001. Therefore, small companies may obtain more gains than large firms from M&As.
Finally, from the strategic management perspective, M&As present significant management challenges for companies. The problem of management is more pronounced in large companies since there are more complex management structure in large companies, and companies may be not able to effectively utilize companies’ resources if they lack managerial abilities (Munjal, forthcoming). In addition, as firms grow by size, there will be more formalization and bureaucratic control problems, which may contribute to organizational management problems (Levie,1997). In contrast, small companies are dynamic and flexible and easy to be managed (Penrose, 1959). Hence, I can infer that large companies are likely to face more complex management problems after M&A. As a consequence, there problems are likely to negatively influence firm performance of large companies.
Based on these reasons, I expect that external M&A strategy may not be a more effective strategy for large companies in comparison with small companies. As a consequence, I propose the following hypothesis:
H3. The positive impact of M&A strategy on firm performance will be moderated by firm size, and this positive impact will be weakened as the increase in firm size.
While the research of M&A in developed markets has a long tradition, empirical research on M&A in emerging markets has started in early 2000s (Peng et al., 1999). The analysis and comparisons regarding the impacts of M&A between developed markets and emerging markets are also rarely. Plus, the literature that is related with internal R&D strategy is much less. However, there are some research concentrating on the M&A in emerging markets and developed markets. As a result, the analysis about the relationship among strategies, firm performance and countries is mainly based on the effects of external M&A strategy. Some scholars have suggested that firm’s strategic decisions such as internal R&D and external M&A are significantly affected by the institutional environments that may influence rules, contracts, and personal relations (Peng et al., 2008; Peng and Heath, 1996). Therefore, it can be speculated that the impacts of strategies on firm performance are also likely to be different among emerging markets and developed markets.
16
investigate the effect of institutional environment on the acquisition activity in the United States and China. They speculate that firms are likely to derive benefits from acquisitions if they are in the stable and highly developed institutional environment, such as United States. But if companies are in a less stable and less predictable markets like China, these benefits may be more short-lived. Furthermore, Lin et al. (2009) analysed the importance of structural holes in acquisitions. Because structural holes positions provide firms in developed markets an opportunity to enjoy the informational benefits and the control benefits, these firms can act as brokers in M&A and then strategically manipulate the resources for the maximization of their interest (Burt, 1992; Lin et al., 2009). Therefore, these broker firms are more easily access to private information about other firms (Zaheer and Bell, 2005), and they are able to find under-priced targets in the market (Lin et al., 2009). However, such advantage from the structural position may not exist or even be negative for firms in the emerging markets, which make the obtain of the information about targets more risky and costly (Wright et al., 2005). As a consequence, firms in developed markets can enjoy more information of targets, thereby enjoying the maximum interest from acquisitions, but in emerging markets, the interest from acquisitions might be reduced.
Sun et al. (2012) also point out that institutions are important determinants of M&As. They report that institutional environment can shape financial markets, and financial markets can influence acquisition entry. Therefore, the acquisition activities can strongly effected by institutional environment. Emerging markets usually exhibit a higher degree of uncertainty, as well as a lack of transparency. For this reason, transaction costs of M&As, such as negotiations and post-acquisition integration, will be higher in emerging markets than developed markets. Therefore, the costs of M&A is higher for acquirers in emerging markets (Sun et al., 2012). In addition, from the transition economies perspective, acquirers from developed markets can enjoy convenient transformation process, which can significantly improve the acquirers’ firm performance (Estrin, 2009). It can be inferred that M&As may have more positive impacts on firm performance in developed markets than in emerging markets due to the lower M&A transaction costs and more convenient transformation process.
17
of corporate governance in a country. In addition, developed countries normally perform better in institutional development and corporate governance.
Therefore, I speculate that M&A strategy for companies in developed markets may be more effective than in emerging markets. Hence, I propose following hypothesis:
H4: The positive effect of external M&A strategy on firm performance is weakened in emerging markets.
3. Data and Methodology: 3.1 Data selection
To investigate the relation between strategies and firm performance, I need to first obtain the M&A sample of pharma firms since M&A activities represent firms’ external strategy. The sample of M&As comes from Zephyr database. I select M&As of pharma companies (SIC code: 2834, 2835, 2836) with announcement dates between January 1, 1990, and December 31, 2015. The requirements of the selection of M&A deals are as below:
(1) the acquirer owns less than 50% of the target firms’ equity before the bid, and own more than 50% of the target firm;
(2) the transaction is completed;
(3) the minimum deal value is at least $1 million and this restriction is to eliminate the influence of unimportant deals; the definition of deal value in Zephyr is actual value plus the equity value, enterprise value, a modelled enterprise valued.
(4) the acquirers are listed companies because related information has to be included in different databases, namely Thomson Datastream.
The reason for selecting companies as this criteria is that I aim to restrict my sample of M&As that have a large impact on firm’s research, development, marketing and/or sales processes. After collecting these M&As, I eliminate M&As with missing ISIN number in Zephyr database and acquirers with less than 10 million US dollars. What’s more, because my analysis includes country-dimension analysis, companies, and countries with only one M&A or two M&As are also excluded in order to reduce the influence of unimportant countries. It is also because if there is only one pharma company or only one M&A in one country, it will be difficult to find the comparable companies in this country. Overall, these requirements yield a sample of 1,461 successful M&As and 301 acquirers. Because the analysis is based on firms, these acquirers form the group of external growth companies.
18
the group of external growth companies, I further randomly select some similar-sized companies without M&A but with a high R&D intensity, and the group of internal growth companies consists of these companies. In addition, the companies with only one M&A during 25 years but with a higher R&D intensity are seen as internal growth companies. High R&D is measured by R&D intensity. R&D intensity, defined as the ratio of R&D expense and total sales of a company, reflects the internal investment on R&D. Overall, there are 330 pharma companies with a high R&D but without M&A. In total, there are 669 companies in the sample. The period of data is from January 1, 1990, to December 31, 2015. Finally, I collect 6,108 observations.
Figure 1: The distribution of M&As by years
The sample contains completed mergers and acquisitions between 1990 and 2015 in pharma industry of 21 countries where the pharma industry relatively developed better. All acquirers are publicly traded pharma companies that gains control of a public, private, or subsidiary target whose transaction value is at least $1 million.
First, I analyse the number of M&As by year. The results are shown in Figure 1. It shows that although the number of M&A averagely does increase through time, there is a significantly different patterns in the number of M&As each year between different periods. For the period of the first 10 years, namely from 1990 to 2001, the number of M&As changes steadily and there is no significant increase or decrease (below 20). However, the number of M&As has increased largely since 2002. In addition The pharma industry has been undergoing a profound restructuring since 2000s. This restructure stems from the surge in M&A activity in late 1990s (Demirbag, 2007). Therefore, according to the Figure 1 and the findings of Demirbag (2007), the number of M&As after 2000 is dramatically larger than in the 1990s. Thereby I speculate the effects of M&A on firm performance may be different between these two periods due to the M&A waves. Therefore, because of the huge difference in M&A activity between before 2000s and after 2000s, the sample will be divided into two groups and analysed,
0 50 100 150 200 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Th e n u m b er o f M & A years
19 separately.
20
Table 1: The distribution of companies and M&A by countries
The sample contains internal growth companies and external growth companies through M&A between 1990 and 2015 from 21 countries. In 21 countries, 13 countries belong to developed markets and 8 countries belong to emerging markets. What’s more, there are all completed M&A between 1990 and 2015 listed on Zephyr where the public acquiring firm gains control of a public, private, or subsidiary target whose transaction value is at least $1 million.
Country
The strategy
Total sample Sum of M&A deals Internal R&D External M&A
Developed markets:
Australia 10 4 14 20
Ireland 6 5 11 28
Canada 8 9 17 48
Japan 22 26 48 101
Hong Kong, SAR 1 2 3 10
Denmark 7 2 9 7 France 7 4 11 15 Germany 8 4 12 72 Sweden 9 3 12 9 Switzerland 3 1 4 33 Spain 2 3 5 18 United Kingdom 30 17 47 158 U.S. 93 38 131 281 Total (Developed) 206 118 324 800 Emerging markets: Bermuda 4 5 9 9 Malaysia 1 2 3 5 India 15 34 49 122 Rep. of Korea 34 25 59 48 South Africa 0 3 3 18 Indonesia 0 2 2 7 China 53 107 160 440 Taiwan 17 5 22 12 Total (Emerging) 124 183 307 661 Total sample 330 301 631 1461
3.2 Main and control variables 3.2.1 Dependent variable
21
of how efficiently one company allocate the capital to profitable investments (Macalpine, 2016). Both internal R&D strategy and external M&A strategy are important investment decisions. Therefore, ROIC can evaluate the success or failure of investments, and I use ROIC to measure the firm performance. The data of ROIC is obtained from Datastream database in which the calculation of ROIC is shown as below:
(𝑁𝑒𝑡 𝑖𝑛𝑐𝑜𝑚𝑒 − 𝐵𝑜𝑡𝑡𝑜𝑚 𝐿𝑖𝑛𝑒 + ((𝐼𝑛𝑡𝑒𝑟𝑒𝑠𝑡 𝐸𝑥𝑝𝑒𝑛𝑠𝑒 𝑜𝑛 𝑑𝑒𝑏𝑡 − 𝐼𝑛𝑡𝑒𝑟𝑒𝑠𝑡 𝐶𝑎𝑝𝑖𝑡𝑎𝑙𝑖𝑧𝑒𝑑) ∗ (1 − 𝑇𝑎𝑥 𝑅𝑎𝑡𝑒))) ∗ 100
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑜𝑓 𝐿𝑎𝑠𝑡 𝑌𝑒𝑎𝑟′𝑠 𝑎𝑛𝑑 𝐶𝑢𝑟𝑟𝑒𝑛𝑡 𝑌𝑒𝑎𝑟′𝑠 (𝑇𝑜𝑡𝑎𝑙 𝐶𝑎𝑝𝑖𝑡𝑎𝑙 + 𝑆ℎ𝑜𝑟𝑡 𝑇𝑒𝑟𝑚 𝐷𝑒𝑏𝑡&𝐶𝑢𝑟𝑒𝑒𝑛𝑡 𝑃𝑜𝑟𝑡𝑖𝑜𝑛 𝑜𝑓 𝑙𝑜𝑛𝑔 𝑡𝑒𝑟𝑚 𝑑𝑒𝑏𝑡)
3.2.2 Independent variable
22
Figure 2: The R&D spending in companies that complement different growth strategies.
The sample includes 631 pharma companies from 21 countries. Two values are mean R&D intensity of each group, namely external growth firms and internal growth firms. The R&D intensity is R&D expenses divided by the total sales of a company. Therefore, the R&D intensity reflects the internal investment in R&D of one company.
3.2.3 Control variable
The moderator firm size is measured by the logarithm of total assets in US dollar (million). This data is collected from Datastream database and using firm’s total assets in U.S. dollarsallows us to compare firms across countries. The variable firm size is expected to have a positive impact on ROIC (Moeller, 2003). One of the control variables is leverage, which is the total debt divided by total assets, and it reflects the capital structure of companies. The leverage will positively or negatively affect the firm performance. Some scholars suggest that there is a U-shaped relationship between the leverage and firm performance (Bae, 2017). Therefore, the effect of leverage on firm performance is uncertain. The control variable cash flow to sales ratio reflects the profitability of companies and the ability of turn cash flow into cash. The expected relationship between them is positive as companies’ ability to make profits will positively influence firm performance. Another explanatory variable is sales per employee and it is also a type of profitability ratio. There is no much doubt that innovative staff is essential for the growth of companies. therefore, if average sales generated by one employee is high, companies will perform better. The detailed meanings of control variables are shown in the table 2. 395 2478 0 1000 2000 3000
external growth firm internal growth firm
23
Table 2: Variable Description
Variables Description Expected impact on ROIC
Dependent variables:
ROIC The ratio of return and the invested capital,
Independent variable:
External_strategy The dummy variable, equal to 1 for companies mainly implementing external R&D, equal to 0 for companies implementing internal R&D
Positive
Moderator:
Firm_size The logarithm of a company’ total assets in US dollars million
Positive Emerging_markets The dummy variable, equals to 1 if firm is
from emerging markets, otherwise 0
Negative
Control variables:
Sales_per_employee The total sales of a company divided by the number of employees
Positive Cash_flow/Sales The ratio between cash flow and total sales Positive
Leverage The total debt divided by total assets Positive/negative
3.3 Regression models
To investigate the hypothesis if different strategies have different effects on firm performance this study uses a dummy variable approach. I test my four hypotheses by running a fixed-effect (year) regression model. OLS regression will be used to test the hypothesis.
First, the first and second hypothesis will be analysed by estimating regression equation designed as bellow:
𝑅𝑂𝐼𝐶𝑖,𝑡 = 𝛼 + 𝛽1𝐸𝑥𝑡𝑒𝑟𝑛𝑎𝑙_𝑠𝑡𝑟𝑎𝑡𝑒𝑔𝑦𝑖+ 𝛽2𝐹𝑖𝑟𝑚_𝑆𝑖𝑧𝑒𝑖,𝑡+ 𝛽3(𝐸𝑥𝑡𝑒𝑟𝑛𝑎𝑙_𝑠𝑡𝑟𝑎𝑡𝑒𝑔𝑦𝑖∗
𝐹𝑖𝑟𝑚_𝑆𝑖𝑧𝑒𝑖,𝑡) + 𝛽4 𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖,𝑡+ 𝛽5𝑆𝑎𝑙𝑒𝑠_𝑝𝑒𝑟_𝑒𝑚𝑝𝑙𝑜𝑦𝑒𝑒𝑖,𝑡+ 𝛽6𝐶𝑎𝑠ℎ_𝑓𝑙𝑜𝑤/𝑠𝑎𝑙𝑒𝑠𝑖,𝑡+ 𝛾𝑇 + 𝜀𝑖,𝑡
(1)
Where the dependent variable 𝑅𝑂𝐼𝐶𝑖,𝑡 indicates return on invested capital (ROIC) in year t, 𝛼 is the constant term, T is a series of time dummies for the period 1990-2015 and 𝜀 is a random disturbance term. The independent variable 𝐸𝑥𝑡𝑒𝑟𝑛𝑎𝑙_𝑠𝑡𝑟𝑎𝑡𝑒𝑔𝑦𝑖 is defined as a dummy
variable that equals to 1 if the firm i implement external R&D strategy during the corresponding period, otherwise 0. Other control variable including firm size, leverage, sales per employee, cash flow to sales for firm i in year t are also included in the regression.
24
internal growth companies by testing whether the coefficient 𝛽1 are significantly different from zero. In addition, I can investigate whether the coefficient 𝛽1 is positive or negative to test the
effects of different strategies on ROIC. If it is positive and significant, it means that external growth strategy is more effective than internal growth strategy, and vice versa. 𝛽3 reflects that moderating role of firm size on ROIC. if it is significant, it means that firm size can moderate the relationship between different strategies and ROIC. The coefficients of control variables including sales per employee, cash flow to sales and leverage have the same meanings as the coefficient of External_strategy. In a word, if they are positive and significant, it means that they have positive impacts on firm performance, and vice versa.
To investigate the country effects, firstly I will divide the sample into two groups, i.e., companies from emerging markets and companies from developed markets, and then I will test the hypothesis separately. The aim to do this is observing if there is any difference between two groups. Besides this method, I will test the regression model as following to examine the effects of different markets:
𝑅𝑂𝐼𝐶𝑖,𝑡 = 𝛼 + 𝛽1𝐸𝑥𝑡𝑒𝑟𝑛𝑎𝑙_𝑠𝑡𝑟𝑎𝑡𝑒𝑔𝑦𝑖+ 𝛽2𝐹𝑖𝑟𝑚_𝑆𝑖𝑧𝑒𝑖,𝑡+ 𝛽3(𝐸𝑥𝑡𝑒𝑟𝑛𝑎𝑙_𝑠𝑡𝑟𝑎𝑡𝑒𝑔𝑦𝑖∗
𝐹𝑖𝑟𝑚_𝑆𝑖𝑧𝑒𝑖,𝑡) + 𝛽4 𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒𝑖,𝑡+ 𝛽5𝑆𝑎𝑙𝑒𝑠 𝑝𝑒𝑟 𝑒𝑚𝑝𝑙𝑜𝑦𝑒𝑒𝑖,𝑡+ 𝛽6𝐶𝑎𝑠ℎ_𝑓𝑙𝑜𝑤/
𝑠𝑎𝑙𝑒𝑠𝑖,𝑡+𝛽7𝐸𝑚𝑒𝑟𝑔𝑖𝑛𝑔 𝑚𝑎𝑟𝑘𝑒𝑡𝑠 + 𝛽8(𝐸𝑚𝑒𝑟𝑔𝑖𝑛𝑔 𝑚𝑎𝑟𝑘𝑒𝑡𝑠 ∗ 𝐸𝑥𝑡𝑒𝑟𝑛𝑎𝑙_𝑠𝑡𝑟𝑎𝑡𝑒𝑔𝑦𝑖) + 𝛾𝑇 + 𝜀𝑖,𝑡
(2)
Where the variable emerging markets equals to 1 if companies are from emerging markets, otherwise equals to 0. If 𝛽8 is negative or positive and significant, it means that the emerging markets plays a negative or positive moderating role on the positive relationship between external growth strategy and the return of company.
4. Empirical result 4.1 Descriptive analysis
25
Table 3. Descriptive statistics
The sample include 631 firms in 21 countries from 1990 to 2015. The dependent variable ROIC represent the return on invested capital. The External strategy is the independent variable and it is a dummy variable, representing the strategy that firms mainly use. The firm size is the logarithm of the total assets in U.S. dollar. The cash flow/ sales means the ratio of the cash flow and total sales of a company. Sales per employee is sales divided by the total employees.
Variable Obs. Mean Median Std. dev Min Max ROIC 5,117 -6.21 5.730 87.232 -3,520.180 3,117.750 Year 5,117 2007 2008 6.246 1991 2015 External strategy 5,117 0.5396 1 0.498 0 1 Firm size 5,117 12.905 12.669 1.972 9.217 19.17 Cash flow/Sales 5,117 -1,441.536 11.000 41,607.07 -2,585,810 22,200 Leverage 5,117 24.905 16.150 177.530 -9,020.370 8,168.560 Sales per employee 5,117 125,916.8 590.000 1,381,150 0 50,629,364
26
Table 4: The comparison in ROIC, firm size, R&D intensity between emerging markets and developed markets
ROIC denotes the return on invested capital. Firm size is defined as the total assets in U.S. dollars, and the unit is million. The R&D intensity is R&D expense divided by the total sales of a company. The final row for each sub-group lists the number of observations. The difference test are based on t-tests for equality in means and a Wilcoxon-text for equality of medians. Median values are in brackets. *,**,*** mean 10%, 5%, 1% significance level.
All (1) Emerging markets (2) Developed markets (3) Difference (2)-(3) ROIC -6.21*** [5.8]*** 8.80*** [9.135]*** -14.74*** [3.28]*** 23.54*** [5.855]*** Firm size 3,410*** [253]*** 402*** [175]*** 5,119*** [332]*** -4,716*** [-157]*** R&D intensity 1,355*** [9.83]*** 2,117*** [2.94]*** 14.31*** [18.24]*** 2,102.69*** [-15.3] n 5,826 2,110 3,715
Table 5 displays the correlation matrix of the independent variable and control variables that are used in the regression model. The correlation matrix allows us to understand the relationship between different variables.
Table 5. Correlation matrix for variables
Correlations (1) (2) (3) (4) (5)
(1) M&A 1.000
(2) Leverage 0.003 1.000
(3) Sales per employee 0.051 0.003 1.000
(4) Cash flow/Sales 0.035 0.006 0.003 1.000
(5) Firm size 0.344 -0.002 0.029 0.009 1.000
4.1 Full-period analysis
27
the Eq. (1) for the full period. I used the time fixed effect, and the results can answer the question if external M&A strategy leads to a greater firm performance. The adjusted R-squared value is 0.056, which implies that 5.56 percent of the variation in the dependent variable is explained within this model. According to the results presented in column (1), the coefficient of External_strategy is significant at 1% significance level. Therefore, the hypothesis 1 is supported, and the finding in this study is consistent with the argument in prior studies (Delmar, 2003), where the researchers suggest that the effects of strategies on firm performance are different in pharma industry. In addition, because the coefficient of External_strategy is positive and significant, the hypothesis 2 is also supported, which is that external M&A strategy is more effective than internal R&D strategy to improve firm performance. This findings is consistent with the finding of Demirbag (2007) who suggests that pharma companies that complement M&A perform better than non-M&A firms. Moreover, the interaction between Firm_size and External_strategy is negative, which means that the firm size can negatively moderate the positive relationship between external growth strategy and firm performance. This finding is consistent with the arguments of Moeller et al. (2004) and Byun and Ahn (2007), who suggest that M&A strategy is more effective for small firms than large firms in improving firm performance.
The results of other control variables in this study are interpreted as follow. The coefficients of firm size is significant at 1% level, which is in conformity with the argument of Moeller et al. (2003), who find the existence of size effects on M&A and firm performance. Similarly, the coefficient of cash flow to sales is also significant at 10% significance level, which indicate that there is a positive relationship between cash flow to sales and firm performance. Regarding to the coefficients of leverage, sales per employee, none of them shows significant impact on firm performance. Lastly, combining all the findings together, the coefficients of External_strategy, External_strategy*Firm_size present significant impact on firm performance. Thus, Hypothesis 1, 2, 3 are supported for the whole sample between 1990-2015.
4.2 sub-period analysis
28
performance. Although there is related literature that could support the findings, I speculate the possible reason for the insignificant results is likely to be that the characteristics of M&A waves are likely to be different between these two periods (Demirbag, 2007; Gaughan, 2010). Also, the restructure of pharma company stemming from M&A waves in late 1990s lead to the different impacts of M&A on firm performance. In conclusion, in the period of 1990-2001, there is significant difference in the impacts of different strategies on firm performance for pharma industry. Thus, the Hypothesis 1, 2 are rejected during this period.
Table 6: regression of strategies on firm performance
The sample includes 631 pharma companies worldwide from 1990 to 2015. The dependent variable is firm performance, which is measured by return on invested capital (ROIC). The independent variable is External strategy is a dummy variable, which equals to 1 when firms are external growth companies and equals to 0 when firms are internal growth companies; Firm size is the logarithm of the total assets in U.S dollar of a company; Sales per employee is the sales divided by the number of employees; leverage is the total debt divided by total assets; cash flow to sales is the cash flow divided by sales of a company. *,**,*** means that the significance level are 0.1, 0.05, 0.01 respectively.
Variables Coefficients Full-period (1) 1990-2001 period (2) 2002-2015 period (3) C -190.897*** (13.402) -98.644*** (8.050) -207.071 (17.676) External strategy 148.930*** (17.072) -12.143 (17.437) 170.329*** (22.057) Firm size 14.257 *** (1.103) 7.495*** (0.619) 15.591*** (1.474) Sales per employee 0.000
(0.000) 0.000 (0.000) 0.000 (0.000) Leverage -0.004340 (0.00669) -0.038*** (0.004) 0.019 (0.0098) Cash flow to sales 0.000*
(0.000)
0.003*** (0.0003)
0.000 (0.000) Firm size*external strategy -10.700***
(1.349) 0.777 (1.231) -12.460*** (1.768) Adjusted R-squared 0.056 0.290 0.050 Observations: 5117 1074 4043
29
the same conclusion with the full-period analysis. Similarly, I find that for the period of 2002-2015, the external M&A growth strategy has a positive effect on the firm performance. In addition, firm size can negatively moderate the positive relationship between external M&A strategy and firm performance.
According to result in Table 6, it can be concluded that firm performance is positively affected by external strategy of pharma firms. In another word, compared with investing internally, pharma firms can improve firm performance by investing externally. However, the conclusion is different for the period of 1990-2001. In this period, there is no significantly different impact of strategies on firm performance.
4.3 full-sample analysis and sub-sample analysis
To investigate the country effects, I test the Eq. (2). The results are shown in column (3) Table 7. Firstly, I test the Eq.(1) for the sub-group of firms in emerging markets and developed markets, and the results shown in column (1) and (2), respectively. The adjusted R-squared are 17.5 percent, 6.3% percent and 7.5 percent. Therefore, the model (1) is the best model among three models since it has the largest R-squared. According to the results of column (1), the coefficient of External strategy is significant at 10% significant level. Also, it is positive. The results are in conformity with the findings of Chi et al. (2011), who report the positive relationship between external M&A and firm performance exists in emerging markets. Additionally, the column (2) shows that the coefficient of External strategy is also positive and significant at 1% significance level for firms in developed markets. This finding is consistent with the argument of Chari et al. (2010), who report the positive relationship between external M&A and firm performance in developed markets. Therefore, it can be concluded that whether companies are from developed markets or emerging markets, external M&A strategy is more effective than internal R&D strategy for improving firm performance.
30
Table 7: The regression results of Eq. (2).
The sample includes 631 pharma companies worldwide from 1990 to 2015. The dependent variable is firm performance, which is measured by return on invested capital (ROIC). The independent variable is External strategy is a dummy variable, which equals to 1 when firms are external growth companies and equals to 0 when firms are internal growth companies; Firm size is the logarithm of the total assets in U.S dollar of a company; Sales per employee is the sales divided by the number of employees; leverage is the total debt divided by total assets; cash flow to sales is the cash flow divided by sales of a company; the variable emerging market is a dummy variable, which equals to 1 if firms are in emerging markets and equals to 0 if firms are in developed markets. *,**,*** means that the significance level are 0.1, 0.05, 0.01 respectively.
Variables
Results in emerging markets
Results in developed
markets all companies
(1) (2) Constant -42.113*** (7.466) -261.288*** (24.550) -241.482*** (17.772) External strategy 16.459* (8.864) 196.224*** (31.822) 180.006*** (22.956) Firm size 4.699*** (0.638) 18.870*** (2.038) 17.272*** (1.464) Cash flow/sales 0.005 (0.001) 0.000 (0.000) 0.000 (0.00) Leverage -0.273 (0.019) 0.0002 (0.0001) 0.018 (0.069) Sales per employee 0.000
(0.000)
-4.87 (0.54)
0.000 (0.000) Firm size*external strategy -1.356*
(0.739) -14.266*** (2.473) -12.784*** (1.785) Emerging markets 47.032*** (4.715) Emerging markets*external strategy -31.820*** (6.310596) Adjusted R-squared 0.175 0.063 0.075 Observations 1563 2480 4043 4.4 Robustness test
31
results of prior models in Table 6, the R-squared are higher than that in Table 8. Therefore, the possible reason for the inconsistency is that the models of robustness test are not that well-designed and cannot explain the data well.
Table 8: The regression results of robustness test
The sample includes 631 pharma companies worldwide from 1990 to 2015. The dependent variable is firm performance, which is measured by operating profit margin. The independent variable is External strategy is a dummy variable, which equals to 1 when firms are external growth companies and equals to 0 when firms are internal growth companies; Firm size is the logarithm of the total assets in U.S dollar of a company; Sales per employee is the sales divided by the number of employees; leverage is the total debt divided by total assets; cash flow to sales is the cash flow divided by sales of a company. *,**,*** means that the significance level are 0.1, 0.05, 0.01 respectively.
Variables Sub-period of 1990-2001 (1) Sub-period of 2002-2015 (2) Full period Constant -3229.182 (942.881) 8505.344 (10603.79) 3110.607 (7791.862) External strategy 173.025 (1293.964) -8796.490 (13153.59) -3597.578 (9907.536) Firm size 221.319 (73.533) -1083.522 (884.350) 576.063 (782.422) Leverage 0.303 (0.375) -2.794 (5.835) -1.5033 (3.871) Sales per employee 0.000
(0.000)
-4.87 (0.54)
0.000 (0.000) Firm size*External strategy -17.243
(97.583) 1098.342 (1055.035) -12.784*** (1.785) R-squared 0.027 0.005 0.004 Observations 1105 4016 5121
Note: *,**, and *** denoted significance at the 10%, 5%, and 1% level, respectively.
32
aims to eliminate the influence of unimportant companies with only a few M&As. In addition, I removed the companies that only acquiring private targets because I need to analyse the characteristics of targets. However, if targets are unlisted companies, it is difficult to find financial data of target. Hence, it is also difficult to examine the R&D expense and differentiate whether targets are R&D intensive companies or not. Hence, the independent variable in Eq.(1) External_strategy is replaced by the variable Target R&D, which equals to 1 when the bidders are companies that focus on acquiring R&D intensive companies, otherwise equal to 0. Therefore, we can analyse if there is the difference in firm performance between companies that focus on acquiring high R&D intensity targets and low R&D intensity targets by observing the coefficient of Target R&D. The results are shown in Table 9.
Table 7: Robustness regression results
The sample includes in total 99 companies and 1512 observations. The dependent variable is ROIC , i.e., the return on invested capital, which represents the firm performance.The independent variable is Target R&D is a dummy variable, which equals to 1 when firms acquire R&D intensive companies and equals to 0 when firms acquire low R&D intensity firms; Firm size is the logarithm of the total assets in U.S dollar of a company; Sales per employee is the sales divided by the number of employees; leverage is the total debt divided by total assets; cash flow to sales is the cash flow divided by sales of a company. *,**,*** means that the significance level are 0.1, 0.05, 0.01 respectively.
Variables coefficient Constant 53.693** (25.363) Target R&D -131.803*** (35.466) Firm size -2.266 (1.795) Cash flow/sales 0.003*** (0.001) Leverage -0.426*** (0.083)
Sales per employee 0.000
(0.000)
Firm size*Target R&D 8.656***
(2.438)
R-squared 0.054
Observations 1376
33
companies. In another word, the strategy of acquiring high R&D companies is less effective than the strategy of acquiring low R&D companies for the growth of pharma companies. Nevertheless, the results in Table 9 are not consistent with the previous studies. The reasons for the inconsistent results are likely to be biased data since I found companies that acquire R&D intensive firms spend a lot on R&D, and the spending of these companies on R&D are even higher than that of companies that acquire companies with low R&D spending (shown in Figure 3). Therefore, the group of companies that acquire low R&D firms cannot represent companies that complement internal R&D strategy as a result of a lower internal R&D spending. Consequently, it may cause the inconsistency.
Figure 3: The comparison of R&D intensity of bidders
The low R&D targets means bidders that acquire low-R&D companies, and high R&D targets means bidders that acquire high-R&D companies. In addition, the figure shows the average of R&D intensity of different groups. It is obvious that there is a significant difference in R&D intensity of different groups, and companies that acquire high-R&D companies usually have a higher R&D intensity.
5. Conclusion and discussion
This article not only contributes to the small but expanding literature on the topic of the effects of different strategies on firm performance, but is also focusing on comparing and contrast these strategies between emerging markets and developed markets.
On the one hand, this study is of some significance. First, the discussion that whether pharma companies should invest internally or externally has not been studied broadly since internal growth is not an “event” (Aktas, 2008). However, the measurement of external growth is M&A activities, which is an “event”. Therefore, it is not that straightforward to compare the effects of different strategies. To solve this problem and answer the question that which strategy is more effective for the growth of companies, I regard firms with many M&As as companies that complement external growth strategy. Companies that do not complement M&A but pay
0 200 400 600 800 1000 1200 1400
low R&D targets high R&D targets
