Cross Border Acquisition, Joint Venture and Firm Value:

Page 1

Cross Border Acquisition, Joint Venture and Firm Value:

Evidence from EU and NAFTA firms Acquiring and entering in JV with firms from GCC

Research Paper June, 2015 Name: Dawlat Bik (s2342529)

Study program: MSc. IFM Supervisor: Dr. H. Vrolijk Assessor: Prof. Dr. N. Hermes

Abstract

This paper studies the effect of Cross-border Acquisition and JV on stock prices. The samples were chosen from EU+NAFTA firms acquiring and entering in JV with firms from GCC countries in the period 1997-2014. The study shows that cross-border Acquisition affects the stock prices positively and significantly, while the cross-border JV does not have significant effect on the stock prices. Further, the sub-analyses shows that the deals which had taken place in 1997-2007 had negative effect, though not significant, and the deals which took place in 2008-2014 had positive effect on the stock prices. Also this paper studies the Effect of Acquisition and JV for the long-run period and finds that, the effect remains relatively stable in the post-event period as well, which this confirms the efficient market hypothesis. Based on this study, cross-border Acquisition creates more value than the cross-border JV for the shareholders.

Keywords: Cross-border acquisition, cross-border joint-venture, event study, foreign

market

Page 2 Table of Contents

1. Introductions

………...3

2. Overview of the literature

….…..………..………..……..5

2.1. Acquisition ………..………..……….5

2.1.1. Previous studies on cross-border Acquisition ………..…………..………...6

2.2. Joint Ventures (JV) ………..……….………...9

2.2.1. Previous studies on cross-border JV ……….………….……….10

2.3. Choosing the strategy, Acquisition vs. JV ………...13

3. Data and methodology

…...16

3.1. Data ……….……....16

3.2. Event study methodology ……….…………..17

3.3. Significance test………..21

4. Results and discussions

…...22

4.1. Descriptive results ………...………22

4.2. Statistical results ………...25

4.2.1. Short-run impact ………..25

a. All Acquisition and JV samples (1997-2014) ………25

b. Sub-samples study (1997-2007, and 2008-2014) ……….………..26

4.2.2. Long-run impact ……….…………...27

4.3. Discussions ………...29

5. Conclusions

………...33

Appendix- I (Agency Theory)……….……….………..35

Appendix- II (significant test methods) ……….…….………...36

Appendix-III (Tables)...……….……….…………...37

References……….……….…………...48

Page 3 1. Introduction

With the globalization era of economic and business, firms internationalize via models such as export, joint venture, green-field investment or acquisitions to expand their businesses and reach new potential markets beyond their own countries of origin.

This paper will study cross-border acquisitions and joint ventures to find which strategy is better for entering a foreign market or expand the existing business in a foreign market. The focus of study will be on three economic regions: EU and NAFTA (North America Free Trade Agreement) firms as acquirers and GCC (Gulf Cooperation Council) firms as the targets.

International business strategy literature is rich from topics about cross-border Acquisition and joint venture (JV) but the outcomes are not conclusive. To the best of my knowledge, there is no paper which studies empirically both cross-border Acquisition and JV. Cross-border Acquisition and JV are two important strategies for the firms to enter into a new market or expand their existing businesses in the market. Firms, when internationalizing or expanding their existing businesses in a foreign market, may face the question of which strategy should be chosen, Acquisition or JV? From these two strategies, which one will create more value for the firm?

This paper will study the effect of EU and NAFTA firms acquiring and entering in JV with Gulf Cooperation Council (GCC) firms. GCC as an economic region has not been explored by the academics yet. GCC as a regional intergovernmental political and economic union was established in 1981 and it consists of all Persian-Gulf-Arab monarchy states. Its member countries are: Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and United Arab Emirates (UAE).

This economic union has around 50 million populations, a GDP of 1.63 trillion in current US$

and an average per capita of 34 thousand dollar (World Bank, 2013). Although Persian Gulf

countries are popular for the oil production and Petroleum economies, since last two decades,

they have been diversifying their economies. For example, United Arab Emirates has been able

to diversify its economy, which, now around 71 % of its GDP comes from non-petroleum

products. GCC states have been trying to improve their living standards by making mega-plans

which are very attractive for the foreign firms. According to Economist news website (Jan 10th

2015) Saudi Arabia is planning to invest 500 billion US$ by 2020 in infrastructure to spur

economic development and this will include making new cities, new financial centers, metro,

universities , schools and so on. This union is learning from EU experiences to integrate its

Page 4

members by creating new railway transport systems which are underway, economic and financial institutions. One advantage of GCC is its location which is between east and west. That might be a reason which Dubai international airport became the busiest international hub in 2014. As we can see in Graph-1 the FDI inward to GCC between years 2000 and 2013 have had 14 times growth (in the same period, the growth for the developed economies was 2.8 times, developing economies 4.8 times and transition economies 16 times)

. The main reason for this growth may have been the economic mega-plans, economic diversification plans and improvement in the business facilities for the foreign investor firms in GCC.

Graph-1

This paper is different from previous studies in the following senses: first of all, in this study the Acquisition and JV strategies will be compared to each other. Secondly, their effects on stock prices will be seen in the short-run as well as in the long-run. Thirdly, the target group firms are chosen from an economic union of countries, GCC, which are similar in culture, language, religion, government and economic structures. These target countries are all monarchies and in contrast to other states in the region have been relatively stable and their security has been important for the western countries. Fourth, although these target countries are rich in income and ranked high-income economies by the World Bank, their government and economic institutions still need a lot of improvements; which this requires technology and work skills transfers from developed economies such as EU and NAFTA. A special characteristic of this economic union is: from per-capita income point of view, GCC countries resemble developed

1 Within this classifications, GCC member countries are classified as the developing economies (UNCTADSTAT,2014)

Page 5

countries, and from institutional progress, technology and labor skills point of view to the developing countries. For example, per capita income of GCC is identical with that of EU (EU per capita national inc: US$34,290). And last, to the best of my knowledge, up to now, there has not been any studies made concerning developed countries firms acquiring or entering in JV with GCC firms.

Research Question: Which strategy, entering in joint venture or acquiring of GCC countries

firms by the EU and NAFTA firms create more value for the stockholder of the acquiring firms, in the short-run and in the long-run?

The remaining of this paper will be as the followings: section 2 will explain literature of crossed-border Acquisition and JV. Section 3 will explain data and methodology which have been used in this paper. Section 4 explains the results and discussion and is followed by section 5 which gives the conclusion.

2. Overview of the Literature

This section describes the related literature of cross-border Acquisition and Joint-ventures (JV).

Firstly I look to the literature of Acquisition and JV separately and then I will compare them to see which strategy would be better for investing in developing economies such as GCC countries.

2.1. Acquisition

Merger and Acquisition (M&A) are two strategic management terms which generally mean

consolidation of firms. Although they are mostly used together, they are technically and legally

two different things. Merger happens when two firms combine together to form a new entity,

whereas acquisition is the purchase of one firm by another one. In case of merger the two firms

which merge together do not exist anymore under their previous identities; they are a new entity

which is a consolidation of those two firms. In case of Acquisition the two firms, the acquirer

and the target, may still keep their identities. But the acquirer firm as the new owner has the right

to control and organize the acquired firm. In this paper I use only the term ‘acquisition’ and by

this I mean purchase of one firm by another one-meaning purchase of more than 50% stocks of

one firm by another firm.

Page 6 2.1.1 Previous studies on cross-border acquisition:

Cross-border acquisition literature is relatively rich and it has been studied by scholars from different perspectives. Eun et al. (1996) studies cross-border acquisition synergy for 103 paired samples of US targets and foreign acquirers for the period 1979-1990. The paper finds positive significant abnormal return for the acquirers as well as for the target firms, which confirms synergy gain hypothesis. The paper concludes that the gains are not the same for all acquirer countries. As it is shown in their subsample analyses, for example, while the Japanese firms got the maximum gains as acquirers, there was no gain for the British firms. The paper further argues that the main reason for the Japanese firms’ higher gain is because of having necessary skills to

“arbitrage international market imperfection” and create synergy.

Cakici (1996) studies cross-border acquisition for a sample of 195 foreign firms which acquired target firms in the US and a control sample of 112 US firms which acquired target firms in foreign countries for the period 1983-1992. The paper finds significant positive abnormal returns for the foreign firms which acquired US target firms but no value effect for the US firms which acquired foreign firms.

Morosini et al. (1998) study cultural distances and cross-border acquisition performances on 52 sample, which in every acquisition, one partner either acquirer or target was from Italy and other partners from USA or other developed EU countries, for the period 1987-1992. The performance has been measured by percentage growth in the sales for the two years following acquisition. The paper finds that cultural distances has positive effect on the performance of the cross-border acquisition. This finding is in opposite with the hypothesis which says that distant cultural difference may cause extra cost for the acquirers because of the integration of the acquirer and target firms and therefore may do not create value for acquirer shareholders.

Bris and Cabolis (2008) study a sample of 506 acquisitions from 39 countries for the period

1989-2002. The paper finds that investor’s protection and accounting standards of the acquirer

firm’s country has positive effect on the abnormal return of the cross-border acquisition. They

come to the conclusion that when the acquirer firm comes from well-protected shareholders

rights country and acquires a target in a weak-protected shareholders right country, the deal will

create premium for the acquirer. They explain that corporate governance might be a motive for

Page 7

the cross-border acquisition. Bris and Cabolis (2008) findings has been also confirmed by Martynova and Renneboog,( 2008) findings who hypothesize the “positive spillover by law” and argue that part of the synergy gain in cross-border acquisition is made by the imposition of strong governance regulations which is spilled from the acquirer to the target.

Chari et al. (2010) study a sample of 594 acquisitions which were made by developed country firms as acquirers and emerging country firms as targets, a sample of 1624 acquisitions which were made by developed countries as acquirers and targets and a sample of 900 acquisitions which were made by the emerging country firms as acquirers and targets. The paper finds that cross-border acquisition brings positive abnormal return for the developed country firms when acquiring emerging country firms, and emerging country firms when acquire emerging country firms. The acquisitions by the developed country firms do not create significant return when they acquire targets in developed countries. They further find that the gain of developed country firms are more than that of emerging country acquirers when both acquiring emerging country firms.

Corhay and Rad (2000) study a sample of 111 cross-border acquisitions which were made by Dutch firms for the period 1990-1996. The targets were US and European firms and the paper finds weak evidence that cross-border acquisition creates value for the shareholders, especially in US.

Since last two decades, there has been an attempt to study the cross-border acquisitions which were made by emerging country firms. Cross-border acquisition made by emerging country firms is in rise and as it looks from the literature, they choose mostly developed country firms as their targets. Bhagat et al (2011) study a sample of 698 cross-border acquisitions which were made by emerging country firms for the period 1991-2008. The paper finds positive abnormal return for the acquirer firms. According to this paper most of the targets were in the developed countries and according to their explanation better corporate governance in target countries is positively related to the acquirers’ returns.

Gubbi et al. (2010) study a sample of 425 cross-border acquisitions which were made by Indian firms for the period 2000-2007. The paper finds positive abnormal return for the shareholders.

They further find that the level of economic and institutional advancement of the target country

is positively correlated with the market expectation of the shareholders. They argue that cross-

Page 8

border acquisition is an instrument for the emerging market firms to gain strategic assets from different markets to enhance their international capabilities.

The summary characteristics and results of the above papers are presented in Table-1. It shows that the positive effect of cross-border acquisition is dominant; especially when the targets are in developing countries. GCC countries, though rich by per capita income, are still in developing stages. Therefore I may hypothesize:

Hypothesis 1: the EU+NAFTA firms acquiring firms from GCC countries create value for the shareholders.

Table-1: Previous studies on cross-border acquisition

Paper Sample

period

Sample size Acquirer country (type)

Target country (type)

Methodology used

Results*

Bhagat et al (2011) 1991-2008 698 Developing All sorts Market model +

Cakici (1996) 1983-1992 195 112

Developed US

US

Not specified

Market model Market model

+ 0

Chari et al. (2010) 1986-2006 594 1624

900

Developed Developed

developing

developing Developed developing

Market model Market model Market model

+ 0

+

Corhay and Rad (2000) 1990-1996 111 Netherlands US & EU Market model 0

Eun et al. (1996) 1979-1990 103 (mainly) developed countries

US Mean Adjusted

model

+

Gubbi et al. (2010) 2000-2007 425 India All sorts Market model +

Markides and Oyon (1998) 1975-1988 236 US Canada & EU model not specified

+

Note: this table presents characteristics of the research papers which study the effect of cross-border Acquisition on stock prices.

*The sign (+) represents significant positive abnormal return; sign (-) represents significant negative abnormal return and the sign (0) represents not significant abnormal return.

Page 9 2.2. Joint Venture:

Joint Venture (JV) is defined as a distinct legal and organizational entity which is formed by two or more parent organizations by means of investing financially as well as other resources to reach a particular objective (Liu et al., 2014). In JV two or more firms bring a portion of their resources to form a common legal organization which is, principally, one of the alternative modes to transact (Kogut, 1988). According to Contactor and Lorange (1988), as was cited by Park and Kim (1997), JV is a strategic move to extend the competitive capability of a firm by utilizing the complementary resources which is provided by each partner firm or getting access to the firm-specific assets of another partner firm.

Finnerty et al. (1986) defines JV as, “partial business combinations wherein two or more firms form a profit-motivated entity after negotiating the financial and legal terms of the combination .”

Following the above definitions, in this paper, JV is defined as a distinct legal entity which is formed by two or more firms by collective contribution of required resources to complement the production process and reach a specific objective, by sharing the costs as well as the profits which are earned by the entity.

Firms always search for new projects with maximum possibility of high profits; but earning high

profit indulges, mostly, high risks as well as high investment requirements. Some projects may

seem too risky for a single firm to accept and run lonely and they may become acceptable to the

board of directors as well as to the financial market when such projects are shared by two or

more firms (Jones and Danbolt, 2004). That is why the firms form a JV to share risks, reduce the

investment costs and get access to other firms’ specific competences (Park and Kim, 1997). JV

as a mechanism gives the firms the option to expand into risky markets (Kogut, 1991), and learn

the markets well before making a full commitment to the market. As Bowman and Hurry (1993)

argue that JV is an initial and trial investment when entering into a new market or country. The

paper gives the example of Japanese firms when entering to US market, they began, often, with

JV and after getting enough experience in the market, they preferred acquisition.

Page 10 2.2.1. Previous studies on Cross-border Joint Ventures (CJV):

As it looks from the literature, there has been much research done about cross-border JV but most of the studies have been about economical and legal aspects of the JV. Unlike Acquisition, there has not been much study done about JV from financial perspective. Previous researchers have assumed that JVs have the same financial impact to the f irms’ stock prices as that of Acquisition (Park and Kim, 1997). The crossed-border JV is assumed to create value for shareholders because of the expectation of complementary technologies, reduction of costs and risks and economies of scale which are gained by the firm.

According to the previous studies there are three conflicting hypothesis: a. the announcement of cross-border Joint-venture (CJV) positively affects the stock prices; b. the announcement of CJV negatively affects the stock prices. C. the announcement of CJV has no effect on the stock prices.

a. Positive effect:

Crutchley et al. (1991) study 146 Joint ventures between US and Japanese firms for the period 1979-1987 and find significant positive abnormal returns for both countries firms. The paper argues that overcoming of restrictions for the flow of goods and services may be the motive for the international joint venture.

Frohls et al. (1998) Study CJV with a sample of 320 cases where at least one partner was US firm and the other from emerging economies (including eastern European communist countries and China) and industrialized G7 countries, for the period of 1987-1992. The paper finds, on average, abnormal returns for the announcement of CJV. The paper concludes that there is significant positive reaction for the announcement of CJV with emerging economies but insignificant positive reaction with the industrialized economies. Frohls et al. (1998) findings have been also confirmed by the Irwanto et al. (1999).

Jones and Danbolt (2004) study CJV for a sample of 158 UK firms for the period 1991-1996,

and find significant positive abnormal returns for the firms. They add that the returns are

significantly lower when the JV is undertaken by large firms or the project is located in Asia and

returns are significantly higher when the project is larger in contrast to the firm and the project is

either domestic or located in EU.

Page 11

Liu et al. (2014) study 394 CJV cases from different countries for the period of 2005-2010 and the paper finds that CJV, on average, creates value for the shareholders.

Chen et al. (1991) study 88 CJV by the US firms in China for the period of 1979-1990. The paper finds significant positive abnormal returns for the shareholders.

b. Negative effect:

Lee and Wyatt (1990) study 109 US firms CJV for the period of 1974-1986. They divide data for 3 groups of developed countries, newly industrialized countries (NIC) and less developed countries (LDC). Their result shows significant negative losses for the CJV for the developed and newly industrialized countries but insignificant positive for the less developed countries. The paper explains that the negative effect might be because of agency cost cash flows. According to agency cost cash flow hypothesis, when the managers have cash flows in excess of their normal needs, they use it for their own desires rather than benefiting the shareholders. Therefore investing in cross-border JV might be one example of such agency cost cash flows. Lee and Wyatt (1990) finding has been confirmed by the Chung et al. (1993) who study 230 CJV formed by 173 US firms for the period 1969-1989.

c. No effect:

Borde et al. (1998) study a sample of 100 US firms CJV for the period 1979-1994 and the joint partners are from different countries around the world. The paper comes to the conclusion that, on average, there is no immediate impact on the stock prices. The paper adds that CJV is more favorable when the JV is with the Asian countries and less favorable when with lower risk developing countries. The paper does not mention which countries is lower-risk and which countries higher-risk developing countries. This creates ambiguity for the reader of the paper.

Chen and Hu (1991) study 46 cases of CJV between US and east European countries and find no significant effects of CJV announcement on the stock prices.

Finnerty et al. (1986) study 208 US domestic as well as international JVs for the period 1976- 1979. The paper comes to the conclusion that there is no significant effect on the stock prices.

The paper does not mention how many of those studied JVs were domestic and how many were

international. The paper argues that lack of information may be the reason which splits the

Page 12

reaction of the market and the positive reactions are neutralized by the negative reactions. It means that when the JV is announced the market does not know fully about the advantages or disadvantage of the JV project. It might be too early to judge about it. This may cause a split in interpretation of positive or negative gains from the project.

The summary characteristics and results of the above papers are presented in Table-2.

Table-2: Previous studies on cross-border JV

Paper Sample period Sample size Acquirer country (type)

Target country (type)

Methodology used

Results*

Borde et al. (1998) 1979-1994 100 US Developed and

developing

Market model 0

Chen and Hu (1991) 1987-1989 46 US Eastern Europe Market model 0

Chung et al. (1993) 1969-1989 230 US Developed and

developing

Scholes and Williams(1977)

-

Crutchley et al. (1991) 1979-1987 146 US and Japan Japan and US Market Adjusted model

+

Frohls et al. (1998) 1987-1992. 102

218

US

US

China & East Europe Western Europ.

Market model

Market model

+

0

Irwanto et al. (1999) 1972-1993 300 US Developing Market model +

Jones and Danbolt (2004) 1991-1996 158 U.K. Developing and

developed

Market adjusted model

+

Lee and Wyatt (1990) 1974-1986 109 US Developed and

developing

Scholes and Williams(1977)

-

Liu et al. (2014) 2005-2010 394 Developing and

developed

Developing and developed

Market model +

Note: this table presents characteristics of the papers which studied the effect of cross-border JV on stock prices.

*Here, sign (+) represents significant abnormal return, sign (0) represents not significant abnormal returns and sign (-) represents negative abnormal returns.

Page 13

As it looks in the above table, positive effect of CJV has been more dominant; especially when the joint partner is chosen from a developing country. Therefore, for the GCC, I may hypothesize that:

Hypothesis-2: the EU+NAFTA firms entering in JV with firms from GCC countries create value for stockholders.

2.3. Choosing the strategy, JV vs. Acquisition:

When a firm wants to enter into a new market, outside of its home country, the first question which it will face is: which strategy or mode of entry should be chosen? In general, this will depend on the characteristics of the firm’s business; political, economical and social characteristics of the concerned market’s country; internationalization skills of the firm; and financial, organizational and technological capability of the firm. A firm will choose its final entry mode after finding a balance between the potential financial gains and the level of risk related to the project.

To know which strategy to choose, whether JV or Acquisition, we need to give some information about other alternatives as well. Entering to a foreign market via exporting and licensing/franchising are the cheapest modes of entry but in terms of profitability and control of the assets, they may not be the ideal ones. When entering a foreign market via the export, there comes the problem of distribution. If the distribution task is given to the local firm, there arises the problem of agency costs as well as the extra restrictions which is mostly imposed by some governments because of saving their own internal products. Licensing/franchising is cheaper in contrast to the Greenfield investment (which involves the construction of the business facilities from the scratch), JV and Acquisition modes. But the problem with the licensing/franchising arises because of having less control on the assets, and there may be risk of losing the assets. The licensee may steal the technology by bringing some small changes to it and branding it by its own name. However, still it depends to the characteristics of the firm and its products, and in some cases still these two strategies may be the ideal ones. Three other strategies which are more important and they contain more risk and cost, are: Greenfield investment, Acquisition and JV.

In Greenfield investment the firm commits its investment in a foreign market from scratch and

according to its own needs. The main benefit of this strategy, in contrast to Acquisition and JV,

Page 14

is that by this strategy the firm will keep its full control over the functions of the newly established foreign branch and scratch it the way which satisfies the firm. But the main problem with this strategy is the cost, time and extra risk because of not having experience and familiarity with the market. Acquisition and JV in contrast to Greenfield investment are cheaper and they take less time to enter a market. They provide the firm the facilities to exploit the opportunities in the foreign market on time and give the firm the power, in addition to entering the market, to expand its business as well. The big advantage of a foreign acquisition over setting a new plant investment (Greenfield) is the fast and proper entry to the foreign market and benefitting from local labor force, culture and the ties which already exists between the foreign firm and the local institutions (Harzing, 2002; Hennart and Park, 1993).

As it looks from the literature, the main factors which affect the strategy of the firm for choosing among three main alternatives of Greenfield, Acquisition and JV are the profit expectation of the project, the risk related to the project, level of control over the project, financial capability of the firm and organizational and technological skills of the firm. Unlike Greenfield investments, JV and Acquisition create synergy for the firm (McConnell and Nantell, 1985; Jones and Danbolt, 2004; Raff et al., 2009). But here is also the probability of opposite to synergy as well; firms may face losses because of integration problems in case of acquisition and loss of their firms-specific assets in case of JV.

Raff et al. (2009) argue that the possibility of Greenfield investment affects the choice for choosing for Acquisition or JV. If from market perspective the Greenfield is a viable option, then this will give the firm the incentive to prefer JV over Acquisition and Acquisition over Greenfield-if the fixed costs are lower than the Greenfield investment. The reason is that if the Greenfield is profitable, this profitability of Greenfield will reduce the cost of Acquisition and will give the local firms the incentive to agree for JV. The paper further argues that if Greenfield is less profitable than the exporting, the local firms will not be ready to participate in JV and there remain only Acquisition and exporting as options for the firm.

JV is superior to Acquisition when it is difficult to separate the unneeded assets of the selling

firm from the rest of the firm (Park and Kim, 1997). This means some target firms may have

some assets which are not needed by the acquirer. In contrast, JV is made from scratch it may

give some easiness for the partner firms to choose their assets.

Page 15

Although, JV in contrast to the Acquisition may give more “synergistic gains at minimum cost under conditions of information asymmetry and complementary benefits” the partners are exposed to opportunistic exploitation of their specific assets (Park and Kim, 1997), specially, when the JV is between two rival firms.

According to Lopez-Duarte and Vidal-Suarez, (2013) findings, firms prefer JV over Acquisition and Greenfield investment when the political risk and cultural distances are high.

Chang et al. (2012) argues from two perspectives: First, it is difficult and costly to monitor, control and settle the disputes in a wholly owned subsidiary (WOS). Therefore, sharing the project via JV with a local firm will decrease the cost of monitoring, control and settling of the disputes. Secondly, if the governing quality of the target country is very poor, that will give rise to the opportunism which entering in JV will be costly because of monitoring of the partner and negotiation. In such situations WOS will be better choice than the JV.

In entering to a developing country, in which the economic, political and cultural system and regulation are very different than that of the firm, entering by JV mode will be more effective than Acquisition or Greenfield (Meschi, 2004).

The firms which make small investment in China (via JV) will have the flexibility to exploit the opportunities if it arises and still keep their loss probabilities to the level of their investments. On the other hand, the flexibility for expansion of firms with big investments (WOS) will be small and their downside risks still remains big (Cheng et al, 1991).

Based on the above literature, it seems that JV is a better strategy for entering to markets such as GCC. So, I may hypothesize:

Hypothesis 3: when entering to GCC market, JV strategy will be a better choice than Acquisition

and creates more value for stockholders.

Page 16 3. Data & Methodology

3.1. Data

This paper studies whether Cross-border Acquisition (CBA) as an international business strategy or Cross-border Joint Ventures (CJV) will benefit the NAFTA+EU firms when investing in GCC countries. To get the required data, I used ‘Zephyr database’ as a source for getting information regarding the CBA and CJV of EU+NAFTA firms with the GCC firms (from Jan.1997-Dec.2014 period). This database is provided by Bureau van Dijk and it contains detailed information about M&A and JV. DataStream database, which is provided by Thomson Financial, is used as a source for firms’ daily stock returns and market indexes. For each firm the local market index is used. In some cases, when the firms have been registered in more than one stock exchange, I use the market index, where the firm originally belonged.

The choosing criteria of the sample data were as the followings:

i. Listed acquirer firms from NAFTA and EU countries ii. Listed and unlisted target firms from GCC countries iii. Minimum value of target firm: 0.5 million Euro iv. The deals have been completed as on Dec. 2014.

v. The deals should have been taken place between Jan. 1997 and Dec. 2014

3.1.1 Cross-border Acquisition: there were totally 72 Acquisition deals which complied with the above research criteria. Out of these deals, 1 deal was dropped because the acquirer’s ISIN no.

was not mentioned by the Zephyr, 7 deals which belonged to the banking and finance and 6 deals which their data were not available in Datastream were also dropped from the list. The final processed Acquisition deals were 58. The list of all 72 deals is available in Appendix-III (Table- 12).

3.1.2 Cross-border Joint Ventures(JV): The total JV deals which complied with the fore

mentioned criteria were 155 deals; which 3 deals were not with the GCC firms (though they were

in the GCC countries) and therefore they were dropped from the list. 20 deals which belonged to

the banking and finance industry were dropped from the list. In 29 deals which the NAFTA+EU

firms ISIN no. was not mentioned were dropped from the list. And in 7 deals the data was not

Page 17

available in DataStream website; therefore, they were also dropped from the list. The final processed samples were 96 JV deals. The list of all 155 deals is available in Appendix-III (Table- 13).

3.2. Event Study Methodology (ESM)

Event study methodology measures the reaction of stock prices to a firm’s specific event announcement. Events such as earnings announcements, joint venture announcements, change of CEO, M&A and stock split announcements have been some popular example of events which have attracted attention of the researchers to find if such events have had an effect on the stock prices. Event study methodology’s history goes back to the 1930s. James Dolley (1933), who studied the effect of stock split on the stock prices, is believed to be the first published evidence of the method. Through the 20

century, the methodology got more attention and in 60s, 70s and 80s there were developments on the sophistication of it. The works of Fama et al. in 1969 “The adjustment of stock prices to new information ”, and that of Brown & Warner, in 1980

“Measuring security price performance” are two popular examples of the method.

Event Study Methodology consists of three main models which are as the followings:

i. Mean Adjusted Returns(MAR)

In MAR model the firm is expected to have a stock return equal to average return of its estimation period. In this model, it is assumed that the mean return of a given stock is constant over the time. The abnormal return is derived by subtracting the average return of estimation period from the average return of event period:

AR= R

- K

AR is the abnormal return, R is the return of security i in event-window in time t and K is the predicted return on the basis of the average return of estimation period. Although the MAR is the simplest model of event study methodology, it gives almost the same result as that of the sophisticated models (Brown & Warner, 1980). This is because the variance of abnormal return is not much reduced with the use of more sophisticated models (MacKinley, 1997).

Page 18

ii. Market Returns

In this model expected return is measured by using capital asset pricing model. It is assumed that every security has its own risk and expectation, and th ey should be found through finding α and β and link them with that of the market. Therefore this model takes into account the market trend as well as the firm’s risk. In this model the expected return is:

E(R

,

) = α

+ β

R

+ ε

, (2)

The E(Ri,t) is the expected return of i security in time t; a shows the average return of firm compared to that of the market; β shows the sensitivity of i firm’s return to that of the market and ε is the random variable which is expected to be zero.

According to this model abnormal return is found by deducting the expected return from the actual return of the event period.

AR

= R

– E (R

) (3) iii. Market Adjusted Returns

In this model the Abnormal Return (AR) is defined as the difference between the actual returns on event period and the market index. This model takes into account the movement of market and assumes that the ex ante expected returns and risks are equal across all securities. According to Brown & Warner (1980), expected return should be equal to that of the market because the market portfolio of risky assets is actually a linear combination of all securities.

For any security i the market-adjusted model is:

2 According to MacKinlay (1997), ‘market adjusted return’ model should be used only when the data is limited and the pre-event data is not feasible. And further he recommends that the biases which may arise, by imposition of restrictions by this model, should be considered. However, there are some scholars who favor ‘market adjusted model’ over the market model (at least in case of acquisition announcements) and they have their own arguments. Doukas et al. (2013) and Bae et al (2013) argue that the

‘market adjusted return’ model can be used to circumvent some drawbacks of the ‘market model’. They say that market model estimation is based on a time series of returns which are preceding the announcement date and in case of samples containing firms with more target acquisitions, there will be the risk that prior acquisitions will be included in the estimation period.

Page 19

AR

= R

– R

,

AR

is defined as the market adjusted abnormal return on i stock at time t; Ri,t is stock return of i at time t and R

is the local market index at time t .

According to Brown and Warner (1980) experimental findings, the three above models work well in finding the event effects on the securities. The mean adjusted model which is the simplest one gives almost the same conclusion as the other two models but they also conclude that if the returns are not normally distributed the Mean Adjusted Return Model should not be used because it may not work well in such conditions.

In this paper, the announcement of acquisitions and joint ventures of NAFTA+EU firms with the GCC firms will be considered as an event. I will use three days event period (-1, 0, +1) and 120 days estimation period as well as post event period. The estimation period will begin from -121

event day and will include up to -2

event day. In this paper I study post-event period as well. In the research question I use t he word “long-run” and with that I mean post event-period. In case of Acquisition and joint venture, although it may take years before the dealing firms integrate their businesses, it is difficult to study the effects after years of transactions. That is because the firm may have or face different other types of events which neutralizes the concerned acquisition and JV deals. Therefore, I choose 4 months post-event period to study whether the effect of the deals are only temporary for the event period or they may last in the post event periods as well. I assume in the 4 months post-event period there may not be other events beyond the acquisition or JV deals. This post-event study will also test the market efficiency theory which says that stock prices always reflect the available information and no one will outperform the market. This theory says that when an event is announced the market will adjust itself to the new information and will stay so up to announcement of any new event. Therefore, if acquisition and JV will have any effect on the stock prices, the market adjusts itself to the announcement and will remain so until the announcement of any new event. The post-event period will begin from +10

event day and will include up to +130

. As for as I know, this paper is the first one which studies the post- event period within 120 days context.

I have chosen here the post-event from the +10

day to give the market some extra day in

addition to the event-period to digest the effect of the event if that has any.

Page 20

To find the answer to my research question, I will use two models of the event study methodology; the “market model”, which is the most sophisticated model and the “mean adjusted model” which is the simplest one. The ‘market model’ is actually an improvement of the other two models. In this model, the variances of abnormal return are reduced and that increases the possibility to detect the events effect (MacKinley, 1997). With “market model” I will try to find the answer to the question whether the NAFTA+EU firms entering in Acquisition and JV with GCC firms will create value in the short-run. At the same time I will use the “mean ad justed model” for the short-run as a robust model

To find the answer to the question whether the deals in the long-run will benefit the acquirer firms, I will use only the “mean adjusted model”. I think it would be a more suitable model for the measuring of the long-run effect. Here, with use of mean adjusted return, I will compare the average return of the post event period with that of the estimation period as well as that of the event window period. After finding the answer for the short-run as well as long-run result of the Acquisition and JV, I will compare the results and will reach to the conclusion which strategy would be the best one to choose for investing in GCC countries.

To rewrite the complete formulas, the market model expected return:

E (R

) = αi + β

R

+ ε

(5) Abnormal return (AR) can be found:

AR

= R

– E (R

) (6)

Cumulative abnormal return (CAR):





 ²

1

) , (_{1 2}

t

t t

it

i t t AR

CAR

(7)

The average abnormal return for each period would be:





 ^N

i it

t AR

AR N

1

1

, (8)

Finally, the cumulative average abnormal return would be:

� � , � =

∑

� � , � (9)

Page 21

3.3. Significance test:

To choose which statistics test is the most suited, one should first test the normality distribution of the data. If the data are normally distributed, a test from parametrical test family should be chosen and if not normally distributed from non-parametric tests. A good test for the normality distribution of the data is the Jarque-Bera test.

Jarque-Bera test is defined as the following:

=

[ +

− 3 ²] (10) In the above formulae, n represents the number of events, s Skewness and for the Kurtosis of the observations.

Table-3:

Statistical results of EU+NAFTA firms acquiring and entering in Joint-ventures with GCC firms for the years 1997-2014.

Particulars Acquisitions

(n=58)

JV (n=96)

Mean 0.0070 0.0063

Median 0.0073 -0.0004

Maximum 0.0973 0.9844

Minimum -0.2202 -0.1943

Standard Dev. 0.0407 0.1108

Skewness -2.7869 7.3257

Kurtosis 16.7145 65.4065

JB Test 529.6274 16608.1316

The data was processed by using market model of event-study methodology. The event-window was three days (-1, 0, 1) and estimation window 120 days.

As it looks in the above table, the JB Test result 529.6 is for the Acquisition and 16,608.13 for the JV. In both cases the numbers are larger than 5.99 and it means they are not normally distributed. Therefore, I choose Wilcoxon Signed Rank test for the significance test of the outcomes and at the same time t-test will be used as a robust. The Wilcoxon Signed Rank test has been also used by Jones and Danbolt (2004), Chari et al. (2010) and McConnel et al. (1985).

The formulae for the Wilcoxon Signed Rank test will be:

W� = ∑

Rank A�, t (10)

3 Detailed explanation about statistical test methods is available in the Appendix-I.

Page 22

Rank (Ai,t) represents the positive rank of the absolute value of abnormal return for firm i at time t.

� =

The letter n represents the number of samples.

4. Results and Discussions 4.1. Descriptive results

As it was given by the Zephyr database, in the period Jan.1997-Dec.2014, there were a total of 167 joint-ventures and 77 acquisition deals between NAFTA+EU listed firms and GCC firms (which 5 acquisition deals and 12 joint-venture deals have been up to present, March 2015, in announced status). I chose the beginning date as Jan. 1997 because the data of Acquisition and JV prior to this date are not available in Zephyr database. I put here completed status as well as announced status deals to see the general yearly trend of all transactions. As it looks from the data, in contrast to the Acquisitions, it takes for the joint-ventures more time to be completed.

For example, in 2014 all joint venture deals have been on announced status and have not been completed yet. Here, joint-venture deals have been made more than twice of the acquisition deals. While for the same time period, the number of cross-border Acquisitions which were made between the EU and NAFTA firms as the acquirers and targets have been 14.9 times more than JV; and the number of Acquisitions which were made by the EU+NAFTA as acquirers with the developed countries (excluding EU & NAFTA countries) were 6.4 times more than the JV deals;

and the number of Acquisitions which were made by the EU+NAFTA as the acquirers with that of BRICS countries were 1.5 times more than the JV

. As we can see in the graph-2, both types of deals have had growth from 1999 onwards but it shows that joint-ventures had more growth than Acquisitions. Joint-venture looks to have had its highest growth in year 2008 and Acquisitions in year 2011. It shows that both types of deals have been affected by the 2008 financial crisis but the effect has been more on joint-ventures. It is interesting that joint-venture

4 No. of deals between EU and NAFTA firms as the acquirers and targets for the period 1997-2014: Acquisitions 6,493, JV 436 No. of EU+NAFTA deals with the developed countries(excluding EU+NAFTA) for the period 1997-2014: Acquisitions 2,157, JV 339; No of EU+NAFTA deals with the BRICS countries for the same period: Acquisitions 2,059, JV 1,337(Ref. Zephyr)

Page 23

deals which had been twice of acquisition deals between years 2003 and 2010, have been exceeded by acquisition deals in 2014.

Graph-2:

It seems that, as a general, acquisitions have had almost a sustainable growth in contrast to the JV deals.

From the acquirers, 66% of joint-ventures and 61% of acquisition deals were made by EU countries and 34% joint-ventures and 39% of acquisitions by the NAFTA countries. From the EU countries, UK, France and Germany were the top ranking EU countries which made the highest number of deals. These three countries together accounted for 47 % of all joint-venture and 40 % of all acquisition deals. From NAFTA countries, USA accounted for 30% of all joint- venture and 35% of all acquisition deals. Canada accounted for 4.2% of all joint-venture and acquisition deals. Mexico, which is also a member of NAFTA, had no deals.

As it can be seen in the acquirer countries list in the Appendix-III (Table-9), some of these acquirer countries had JV as well as Acquisition deals with the GCC countries but some had only JV or only Acquisitions. It shows that Netherlands, Spain, Belgium and Cyprus had only JV deals and Poland and Ireland only Acquisition deals.

From the target countries, as we can see in Table-4, Saudi Arabia and United Arab Emirates

(UAE) are at the top of the list. Saudi Arabia accounted for 30.6 % of all acquisitions and 36.1 %

of all joint-ventures. UAE, which is in contrast to Saudi Arabia smaller in terms of population

and GDP, accounted for 55.6 % of all acquisition and 33.5 % of all joint-venture deals.

Page 24

It is may be because of its location advantages, which gives facility of access to other GCC countries; and also its, relatively open society, in contrast to other GCC member countries, which has become now a hub in the Middle East.

Table-4:

NAFTA+EU firms Acquisition and JV deals with GCC firms b/w 1997 and 2014

Target Country Acquisitions Joint-Ventures

deals Percentage deals Percentage

U.A.E. 40 55.6% 52 33.5 %

Bahrain 3 4.2% 8 5.2 %

Kuwait 3 4.2% 7 4.5 %

Oman 1 1.4% 11 7.1 %

Qatar 3 4.2% 21 13.5 %

Saudi Arabia 22 30.6% 56 36.1 %

Total 72 100 % 155 100 %

Note: the list contains only the completed-status deals of Acquisitions and joint-ventures b/w EU+NAFTA firms as acquirers and GCC firms as the targets. There are 6 announced acquisition and 12 joint-ventures which up to March,215 have not been completed. (source: Zephyr data base)

Industry targets:

The major industry targets of EU+NAFTA firms, when acquiring GCC firms, were: Personal, Leisure & Business Services (25.6%); Banking, insurance and financial services (8.9 %); Metal and Metal products (8.9%); Computer-IT-Internet services (7.8 %) and Wholesaling (7.8%) The major industry targets for the JVs were: Personal, Leisure & Business Services (15.4 %);

Chemicals, Petroleum, Rubber & Plastic (12.8 %); Banking, insurance and financial services (10.8 %); Mining & Extraction (9.2 %); and Industrial, Electric & Electronic Machinery (6.2 %).

In samples for study, I have dropped ‘banking, insurance and financial’ industry. I mentioned it

in the above to be able to see all the major targets. The complete list of industry target is

available in the Appendix-III section of the paper (table-10).

Page 25

4.2 Statistical results

To make the analyses easier, I put the findings for the short-run and long-run under two separate titles.

4.2.1 Short-run impact

To have a better study of acquisition and joint-ventures I study the samples in two stages. One is all the sample-deals which happened from 1997 up to 2014. The second one, I divide the sample to 1997-2008 and 2008-2014 to see if the financial crisis in 2008 had any effect on the deals as well.

a. All Acquisition and joint-venture samples (1997-2014)

One part of this papers research question was the impact of cross-border acquisition and joint- ventures on share prices in the short-run. With the short-run it is meant the immediate impact of the deals on the price after the announcement. The statistical results of the acquisitions and joint- ventures which were between EU+NAFTA firms and GCC firms are presented in Table-5.

Table-5:

Statistical results of EU+NAFTA firms acquiring and entering in Joint-ventures with GCC firms for the years 1997-2014.

Particulars Acquisitions

(n=58)

Joint-ventures (n=96)

Mean 0.0070 0.0063

Median 0.0073 -0.0004

Maximum 0.0973 0.9844

Minimum -0.2202 -0.1943

Standard Dev. 0.0407 0.1108

Skewness -2.7869 7.3257

Kurtosis 16.7145 65.4065

JB Test 529.6274 16608.1316

R-Square 0.2333 0.2854

Z-Wilcoxon test -2.4190**

(0.0071)

-0.6249

T-test

1.3057* 0.5555

Notes: *** Significant at 1 %, **significant at 5 %, *significant at 10 %

The data was processed by using market model of event-study methodology. The event-window was three days (-1, 0, 1) and estimation window 120 days.

Page 26

As it looks from these results, the mean of abnormal returns, on the event-window, for the acquisition and joint venture are almost identical. The standard deviation of joint-ventures is more than twice size of acquisition deals standard deviation. It shows that joint-venture deals are more volatile than the acquisition deals.

However, since the data for the samples were not normally distributed, here I have chosen Wilcoxon Signed Rank test which relies on Median. Here, the T-test is used just as a robust. If we see the acquisition median, it shows 0.73 % gain for the shareholder for the announcement of the deal. This number is significant at 5 % level of significance. But for the joint-venture deals the median value is -0.04 % which is not significant. It means the announcement of joint-venture does not have significant impact on share prices in the short-run period.

b. Sub-sample studies (1997-2007 and 2008-2014)

To have a deeper study of the subject, I divided the deal period to 1997-2007 and 2008-2014 to see if 2008 financial crisis may have had any effect on the deals. As it looks from the results in Table-6, we can see a difference of impact on the acquisition and joint-ventures. The impact of acquisition for the period 1997-2007 looks not to be significant for the short-run.

Table-6: Short-run impact (sub-study)

Statistical results of EU+NAFTA firms acquiring and entering in Joint-ventures with GCC firms b/w years 1997-2014.

Particulars Acquisitions Joint-Ventures

1997-2007 (n=15)

2008-2014 (n=43)

1997-2007 (n=42)

2008-2014 (n=54)

Mean -0.0104 0.0130 -0.0092 0.0182

Median 0.0024 0.0093 -0.0066 0.0036

Maximum 0.0671 0.0973 0.1330 0.9844

Minimum -0.2202 -0.0446 -0.1943 -0.1358

Standard Dev. 0.0655 0.0261 0.0502 0.1405

Skewness -2.4680 0.9883 -0.8784 6.3348

Kurtosis 8.2006 2.3636 5.4528 44.1313

JB Test 32.1323 7.7251 5.6891 1157.6907

Z-Wilcoxon test -0.0568 -2.9830*** -1.6000 -0.4262

T-test -0.6128 3.2813** -1.1874 0.9531

Notes: ***Significant at 1 %, **significant at 5 %, *significant at 10 % level.

The data was processed by using market model of event-study methodology. The event-window was three days (-1, 0, 1) and estimation window 120 days.

Page 27

The impact of acquisition for the period 2008-2014 looks to be positively significant at 1 % level of significance. It means that the announcement of acquisition for the period 2008-2014 creates an average gain of 1.30 % (0.93 % median) for the shareholders in the short-run period.

The impact of joint-venture announcement also shows some differences before the 2008 financial crisis and after that but not significantly. The joint-ventures which were made in the period 1997-2007 had negative impact on the share prices in the short-run. On the other hand, joint-ventures which were made in the period 2008-2014 had positive impact on the share prices.

Since for the join-ventures the significant power tests are not significant, therefore it can be said that the announcement of joint-venture deals for both periods did not affect share prices significantly.

The above findings have been also confirmed with the robust study of mean adjusted model (ref.

Appendix-II, table-11)

4.2.2 Long-run impact

To find the long-run impact of the acquisition and joint-ventures on share prices, I compared post event average returns with that of pre-event period and event-period, using mean adjusted return model. To make the analyses easier I put the statistical results in two different tables.

a. Acquisition deals: as it looks from Table-7 results for the acquisitions, the mean and median

returns are higher for the event-period. We can see that also for the maximum (minimum) return

which is higher (lower) considerably in case of event-period. The difference between post-

returns and pre-event returns are very low; which means the post-event return trends are as that

of the pre-event. The difference between post-event period returns and event-period return is

considerable but it is not significant. It means there have been some changes in the share prices

but not significantly. This finding complies with the efficient market hypothesis.

Page 28 Table-7: Long-run Impact (Acquisition)

Statistical results of EU+NAFTA firms acquiring GCC firms between years 1997-2014.

1 2 3 3-1 2-3 Particulars Pre-event

period

Event-period (CAAR)

Post-event period

∆ post-event &

pre-event

∆ event-period

& post-event

Mean 0.0005 0.0075 0.0004 -0.0002 0.0072

Median 0.0006 0.0105 0.0001 -0.0002 0.0108

Maximum 0.0184 0.1366 0.0164 0.0155 0.1397

Minimum -0.0098 -0.2171 -0.0083 -0.0118 -0.2336

Standard Dev. 0.0038 0.0480 0.0034 0.0043 0.0497

Skewness 1.6293 -1.3932 1.5109 0.2545 -1.6024

Kurtosis 10.0335 8.0405 8.3381 3.6800 9.3053

JB Test 145.2165 80.1623 90.9322 1.7435 120.9001

Z-Wilcoxon test -0.5300 -0.3738

T-test -0.2985 1.0355

Notes: *** Significant at 1 %, **significant at 5 %, *significant at 10 % level.

The data was processed by using Mean Adjusted Return model of event-study methodology. The event-window was three days (-1, 0, 1) and estimation and post-event windows were 120 days.

Therefore, we may conclude that acquisition affects the share prices positively and its impact remains relatively stable in the long-run (post-event period).

b. Joint-venture deals: The difference between post-event and pre-event period, and event-period and post-event periods are not much different. As it can be seen in Table-8, the differences in

Statistical results of EU+NAFTA firms entering in Joint-ventures with GCC firms between years 1997- 2014.

1 2 3 3-1 2-3 Particulars Pre-event

period Event-period

Post-event period

∆ post-event &

pre-event

∆ event-period

& post-event

Mean 0.0007 0.0061 0.0004 -0.0003 0.0057

Median 0.0005 -0.0010 0.0007 -0.0001 -0.0003

Maximum 0.0116 0.9719 0.0156 0.0124 0.9563

Minimum -0.0068 -0.1962 -0.0059 -0.0113 -0.1943

Standard Dev. 0.0025 0.1107 0.0027 0.0035 0.1093

Skewness 0.6953 7.0707 1.5875 0.2699 7.0033

Kurtosis 5.5606 62.0860 9.1643 2.6228 61.3326

JB Test 34.3167 14918.3449 194.3192 1.7342 14395.4988

Z-Wilcoxon test -0.8770 -0.5546

T-test -0.7323 0.5156

Notes: *** Significant at 1 %, **significant at 5 %, *significant at 10 %

The data was processed by using Mean Adjusted Return model of event-study methodology. The event-window was three days (-1, 0, 1) and estimation and post-event windows were 120 days.