Internal capital markets in cross-border mergers and
acquisitions; a financial market development perspective
Floris Joost Valk*
Rijksuniversiteit Groningen, Groningen, The Netherlands
Master in International Financial Management Double Degree Uppsala University January 2018
Abstract
Cross-border M&A’s have been gaining momentum in the past decades. As growth opportunities are becoming scarce, more and more multinational corporations seek their expansion opportunities across the border through M&A’s. Whether these cross-border M&A’s add value has been a heavily debated topic. This research takes a value adding
perspective by showing the effect of financial market development and capital dependence on the abnormal returns of acquiring firms from the US. Our results show that the effects of financial development and capital dependence are statistically significant, but their financial significance is small.
Field Key Words: mergers, acquisitions, external capital dependence, internal capital markets, financial market development
Introduction
Mergers and acquisitions (M&A’s) have been gaining momentum in the past decades.
According to Hitt, Harrison and Ireland (2001) the total amount of acquisitions in 1997 alone was more then the total amount of the 1980s combined. A total of 4 trillion dollar was spent on M&A’s between 1998 and 2000, more than the total amount of the 30 years preceding. Although M&A’s have not continued this pace that exponentially up to the present, the total amount of M&A’s each year did continue to growing due to an increase in international trade and the consolidation of industries (Shimizu, Hitt, Vaidyanath & Pisano, 2004).
Cross-border M&A’s have seen a rapid growth as well. In the periods 1990-1995 and 1996-2003 the value of cross-border deals grew from 2,5 billion US dollars to 17,6 US dollars. Globalization and technological development contributed greatly to the growing popularity of cross-border M&A’s (Shimizu et al., 2004). M&A’s used to mostly involve firms from the same country, however between 1999 and 2000 over 40 percent of all M&A’s involved firms originating from two different countries (Hitt et al., 2001). In this period the cross-border M&A literature picked up pace as well and showed that cross-border acquisitions can be problematic from a value adding perspective. Post-merger integration turned out to be especially challenging for cross-border M&A’s (Child, Faulkner & Pitkethly, 2001). Yet cross-border M&A’s did not show signs of slowing down.
A lot of literature has been written on the value adding effects of cross-border M&A’s, however, the global playing field is ever changing and these new conditions ask for additional theory in order to make sense of them. In the current situation, growth opportunities are becoming scarce, resulting in more and more firms and multinational corporations (MNC’s) seeking their expansion opportunities across the border further and further away home. Globalization has not slowed down either and a lot of countries that used to be in financial need are now spanning their own MNC’s. However, there are still a lot of countries where the financial system has not reached the level of developed countries yet. Remarkably, it is to these underdeveloped financial markets where we have seen a significant rise in target firms from 41 between 1990-1995 to 174 in 1996-2003 from the United States (US) alone (Francis, Hasan & Sun, 2008). Most of these M&A deals still went to and from financially developed markets despite the puzzling fact that some effects hinder the
Not a lot of research has been done on the added value of mergers to less developed financial countries, mostly because most of these financially less developed countries used to have restrictions on foreign direct investment (henry, 2000). It was not until the mid 1990s that these countries opened up enough for MNC’s to enter. Research that does exist on this has been performed mostly around the period where these restrictions were just lifted and might not be representable for these markets today (Francis, 2008). A re-examination of the value adding effect of these segmented markets is thereforee interesting and an important topic for two reasons.
Firstly, as shown earlier, cross-border acquisitions have become an increasingly important source of investment. More and more money is spent on these cross-border M&A’s, so for investors it is important to understand these transactions and how they might add or destroy value in order to value them correctly and make well informed investment decisions.
Secondly, the understanding of financial development beyond what is possible by looking at domestic investments in countries that went through financial liberalization can be increased. The literature that did address similar issues remarkably showed that M&A’s targeting firms in these less financially developed markets show higher abnormal returns (Ma, Pagan & Chu, 2009; Barbopoulos, Marshall, MacInnes & McColgan, 2014; Francis et al., 2008). Aybar and Ficici (2009) looked at the reverse effect, namely the abnormal return of emerging market multinationals (EMNC’s), and found that abnormal returns are negatively correlated with the acquisition of target firms in developed financial markets. It is widely accepted that emerging market countries and most countries with less developed financial markets carry certain risks like political and economic risks (Diamonte, Liew, & Stevens, 1996; Perotti & Van Oijen, 2001). Yet, investors seem to look beyond this risk and assign a premium to the acquisition of these firms.
This gives rise to the question as to why abnormal returns are perceived to be in acquisitions going to less developed financial markets then they are going to developed financial markets. It is this question that this paper will try to answer.
How does financial development influence abnormal returns in mergers and acquisitions?
implies that this research cannot conclude that acquiring a firm in a less financially developed market adds financially significant positive value.
The remainder of this paper will be as follows. In section 2 we will discus the literature surrounding the topic and present our main hypotheses. In section 3 we will elaborate on our research method, used variables and sample construction. In section 4 we will discuss the results of our research. In section 5 will draw some conclusions flowing from our research. Finally, in section 6 we will discuss some limitations of this paper and make
recommendations for future research.
Theory
This section will describe existing literature on M&A theory, cross-border literature and will build up to the hypothesis proposed in this paper.
M&A’s
To understand M&A strategy it is important to know why firms take part in M&A’s. There are several reasons mentioned in the literature, which according to Haleblian, Devers, McNamara, Carpenter and Davison (2009) can be divided in four sub categories namely value creation, managerial self-interest (value destroying), environmental factors and firm characteristics. We will briefly describe some of the motivations for each category.
According to the literature acquisitions can be motivated by value creation. Some examples for an acquisition motivations for value creation that are mentioned in the literature are; gaining market power (Kim & Singal, 1993), achieving efficiency through cost sharing (Banerjee & Eckard, 1998) and long-term productivity planning (McGuckin & Nguyen, 1995), horizontal acquisitions as means to redeploy assets (Capron, Dussauge & Mitchell, 1998), acquisitions as a mean to discipline inefficient managers (Jensen, 1986)
Environmental factors have also been described by the literature as motives to engage in acquisitions. More specifically; regulation uncertainty and environmental uncertainty are argued to increase the likelihood of an acquisition in order to cooperate and overcome these uncertainties (Folta, 1998; Schilling & Steensma, 2002), acquisitions in order to obtain (scarce) needed resources (Pfeffer, 1972); network ties would increase the likelihood of acquisitions if an interlock directorship is present as firms with these ties would imitate the firms that they were tied to (Haunschild, 1993).
Lastly, firm characteristics have been argued to motivate M&A’s. Previous acquisition experience has shown to increase the likelihood of subsequent acquisitions (Amburgey & Miner, 1992) and when firms struggle with a “strategic hurdle” that another firm can solve they are more likely to be acquired (Graebner & Eisenhardt, 2004)
As aforementioned, M&A theory has been a heavily debated topic. Haleblian, Devers, McNamara, Carpenter and Davison (2009) concluded in their review of merger and
acquisitions that previous M&A’s suggests that acquisitions do not enhance firm value. This applies to both short term measurement (Asquith, 1983; Jarrell & Poulsen, 1989; Malatesta, 1983) and long-term performance measurement (Agrawal, Jaffe, & Mandelker, 1992; Asquith, 1983; Loderer & Martin, 1992).
Cross-border M&A’s
Our paper is interested in the value adding perspective specifically, we will discuss the literature on cross-border in the next part of the section.
There are theoretical arguments that pleat in favour of cross-border M&A’s as well such as the synergies argument (Doukas, 1995; Morck & Young, 1992). Another theoretical argument in favour of M&A’s is that investors would be willing to pay a premium for globally diversified firms. They are willing to do so because this offers them an opportunity to diversify at a lower cost. Normally, when an investor wants to diversify his country specific risk, he has to select stocks from different countries. But when a firm operates in many different countries the stock itself is already globally diversified. The Investor then no longer has to diversify himself and would be willing to pay a premium for this (Denis, Denis & Yost, 2002). Lowering risk has also been recognised by the literature as plausible
explanation for cross-border M&A’s. Pantzalis (2001) and Hudson and Laing (2014) have argued that through operational hedging foreign exchange exposure can be reduced and through this process value is created in cross-border acquisitions. However, the opposite has been argued by Reeb and Kwok (1998; 2000) who find that foreign exchange exposure and systematic risk increase when acquiring firms in an emerging market, who per definition do not have fully developed financial markets yet.
Liquidity
Reeb and Kwok (2000) also argue that firms from emerging markets acquiring firms in developed markets have lower risk due to access to more liquid and well-functioning markets. This liquidity and functionality of financial markets argument has only recently been shed light on by the finance academic society. Bruno and Shin (2014) showed the importance of global liquidity by examining that firms which are dependent on external finance are effected more by global liquidity and benefit from more permissive financial conditions. Moreover, Rajan and Zingales (1996) stated that the development of financial markets has an effect on the growth of industries, especially if they depend more on external capital. In other words, firms that depend on outside capital are positively affected by higher global liquidity in more financially developed markets. It is this line of thinking that this paper will elaborate on.
Functionality of financial markets
in more- and in less-developed financial markets amongst different countries. Moreover, functionality of a financial market gives us the opportunity to look at the internal market hypothesis (Stein, 1997). Stein (1997) states that small firms have more difficulty raising capital than larger firms. When these smaller firms merge or are acquired by larger firms they can solve their capital constraint problem through the use of their internal capital market and take on positive net present value (NPV) projects that were out of reach before. When
applying this argument on financial market development, firms from better functioning (more developed) financial markets could use their internal capital market to take away the capital constraints for firms in less well-functioning financial markets making more projects with a positive NPV possible. Hence, through the use of the internal capital market, cross-border M&A’s should add value for a firm originating from a financially developed market acquiring a firm located in a less developed financial market.
Bekaert and Harvey (1997) argued that market liberalization of the financial market helps lowering the volatility in emerging markets, and thereby lowering risk and increasing the correlation between the local market return and the world market return. In other words, the more financially developed a market gets, the more it acts like the world market, which pushes the cost of capital towards a more globally determined figure. Stulz (1999), argued that this is due to the non-optimal diversification effect that is present in markets that are not fully financially developed. As these markets do not have full access to the financial world market, they cannot diversify optimally and investors require a higher return for that risk. When market liberalization takes places and restrictions are removed, the cost of capital should drop with it.
Listing
Moreover, Piotroski and Srinivasan (2007) and Li (2007) have shown that since the
introduction of the Sarbanes Oxly act the amount of cross-listings on US stock exchanges has dropped considerably, especially for firms from less developed financial markets. This is due to the increased standards of capital requirements, firm accountabilities and reporting
liberalization process are still only partially integrated into the world market after
liberalization. To arrive at a higher level of integration with the world market, more time is needed for the liberalized country to adjust (Errunza & Miller, 2000). According to Errunza and Miller (2000) firms from markets that were liberalized in the 1990s, still face a higher cost of capital compared to firms from developed financial markets.
Continuing this line of thinking, when liberalization is unsuccessful or at least not successful enough to lead to the preferred cost of capital, the need to accept an acquisition by other firms that do have the access to this preferred level of cost of capital should remain. Firms that find themselves in this position do have another option at their disposal, listing on a stock
exchange in a developed financial market. Hail and Leuz (2009) found that firms that cross-list on US exchanges experience a decrease of cost of capital between 70 and 120 basis points and conclude that these effects remained after the Sarbanes Oxly act was introduced.
However, the option to list might not be available to a significant part of these firms. A reason for this could be that the requirements of certain indici are too demanding for these firms. In the case of the US, both the NASDAQ and the NYSE state that in order to be applicable for a listing the firm needs to have had an average pre-tax income of 100 million dollar for the three years preceding the listing request. This is simply out of reach for most firms, especially private firms (Hail and Leuz, 2006). These firms could also consider listing on smaller exchanges like the pink sheets or the over the counter market, but these options do not provide the same cost of capital, meaning this would defeat the purpose (hail et al., 2006). Moreover, for private firms the cost of an initial public offering would also have to be
considered, as Francis, Hasan, Lothian and Sun (2010) have shown that the underprizing in initial public offerings is even more pressing for emerging market firms. This leads to the following hypothesis:
H1: Firms originating form developed financial markets acquiring firms form less developed
financial markets has a positive impact on abnormal returns
Capital dependence
Continuing this line of thinking, Fuller, Rajan and Zingales (1996) found that financial development lowers the cost of external capital for firms as well. They arrived at this
the pace of development of these industrial sectors and found a disproportional faster growth for the sectors that are situated in less financially developed markets (Rajan & Zingales, 1998). Therefore, although the need for capital for firms in these less financially developed countries is higher due to faster expansion, the supply of capital is less efficient due to less well-functioning financial markets which leads to a higher cost of capital for these firms.
Campbell, Dhaliwal and Schwartz (2011) find that a high cost of capital for firms is an intervening variable that in combination with internal financial constrains leads to foregoing investment opportunities meaning that financial market restrictions have a real impact on the economic activity and corporate investments. Rauh (2006) also found that firms which are experiencing financing constrains based on variables such as credit ratings see a decline in capital expenditure. Although Rauh’s (2006) argumentation follows from his observation of firms that are financially constrained due to mandatory contributions to pension plans, the same principal can be applied when looking at other forms of financial constraints. Almeida and Campello (2007) also found that financial constraints in less developed financial markets lead to lower investment cashflows.
Internal capital market and capital dependence
If some firms in the industrial sector are indeed more reliant on external financing and thereforee more sensitive to the degree of development of financial markets as stated by Rajan and Zingales (1998), and the internal capital argument of Stein (1997) holds for cross-border M&A’s originating from developed financial markets to less developed financial markets, this could be of interest for MNC’s that are looking for new investment or growth opportunities. Logically this should mean that firms who are strongly dependent on external capital and are situated in less financially developed countries have a higher cost of capital. This then leads them to face higher cost of capital and thereforee the passing up of growth or investment opportunities that their foreign competitors operating in more financially
developed markets could have taken on. Being acquired by MNC’s might be the solution for these financially less developed capital deprived firms. When MNC’s acquire these firms, the freshly acquired firms can then use the internal capital market of their acquirer and thus get some relive of their capital shortage and high cost of capital.
the acquisition. The degree of dependence on external finance should intensify this relation as firms that are more external capital dependent should experience greater relieve from the acquisition. This leads to the following hypothesis.
H2: Capital dependence of target firms positively moderates the effect of financial
development on the abnormal returns
With this paper we hope to contribute to the current literature though providing a better understanding of internal capital markets and their added value in less developed financial markets. The dependence on the external finance moderator sheds an interesting light on the effect of capital constraints in these less developed financial markets and how internal capital markets can help to create value through overcoming these capital constraints.
In the next section we will describe our research method by explaining the variables and control variables that are used. In this section we will also explain how the sample for this research has been constructed.
Research methodology and data
Methodology
This paper looks at M&A announcement returns for acquirers from developed markets and targets from different levels of market development. If the hypothesized effects presented in this paper hold, then target firms from less developed financial markets should add more value then target firms from a higher level of financial market development due to a larger reduction of capital constraints and a lower cost of capital. This should allow the target firms to take on more positive NPV projects that were not possible before the merger. This should add value to the target firm of which the acquirer owns the residual income rights, therefore we should see an increase in the stock price of the acquirer upon announcement of the acquisition. The abnormal returns of the acquirers are the reflection of the added value.
This study will use a two-step approach to test its hypothesis. First we will generate the cumulative abnormal return for each acquiring firm in our sample. Secondly we will regress the cumulative abnormal return on a set of variables and control variables. To do so, we use an ordinal least squared (OLS) regression based on acquisitions in 32 countries in both developed and less developed financial markets.
The most extensive test will be the one with all control variables for which the formula is: 𝐶𝐴𝑅𝑖 = 𝛼𝑖 + 𝛽1𝐹𝐷. +𝛽2𝐻𝑃. +𝛽3𝐴𝐶𝑄𝑆𝐼𝑍𝐸. +𝛽4𝑇𝑂𝐵𝐼𝑁𝑆𝑄. +𝛽5𝐶𝐹. +𝛽6𝐿𝐴𝑉.
Here 𝐶𝐴𝑅𝑖 is the cumulative abnormal return for stock i, 𝛼𝑖 is the constant for stock i, 𝐹𝐷 is the financial development of the targets country, 𝐻𝑃 is the capital dependence of the target, 𝐴𝐶𝑄𝑆𝐼𝑍𝐸 is the size of the acquirer, 𝑇𝑂𝐵𝐼𝑁𝑆𝑄 is the Tobin’s Q of the acquirer, 𝐶𝐹 is the free cashflow of the acquirer, 𝐿𝐴𝑉 is the leverage of the acquirer, 𝐷𝐸𝐴𝐿𝑆𝐼𝑍𝐸 is the size of the deal, 𝐼𝑁𝐷𝑅𝐿𝑁 is the industry relatedness between the target and the acquirer, 𝐺𝐿𝑂𝐷𝐼𝑉 is the global diversification of the acquirer, 𝑌𝐸𝐴𝑅 is the time fixed effect, 𝐼𝑁𝐷 is the industry fixed effect and 𝜀𝑖 is the error term.
In this paper we will be looking at the cumulative abnormal returns as the dependent variable. The abnormal returns are obtained in a window of 5 days [-2+2] , taken from the
announcement date and will be expressed in percentage points. By summing up the abnormal returns within the window we obtain the cumulative abnormal return for each firm. We use the stock prices of publicly traded US firms, hereby we assume that the US stock market is fully efficient and the stock information gives a fair representation of the added value of the mergers and acquisitions.
The abnormal returns are calculated as follows:
𝐴𝑅𝑖 = 𝑅𝑖𝑡− 𝐸[𝑅𝑖𝑡]
Here the 𝐴𝑅𝑖 is the abnormal return for firm i, 𝑅𝑖𝑡 is the Return of firm i at time t, and 𝐸[𝑅𝑖𝑡] is the expected return of firm i at time t.
With 𝑅𝑖𝑡 formulated as shown here:
𝑅𝑖𝑡 = 𝛼𝑖 + 𝛽𝑖 ∗ 𝑅𝑚𝑡+ 𝜀𝑖𝑡
Here the 𝛼𝑖 represents the constant of stock i, 𝛽𝑖 is the coefficient of the sensitivity of stock i to the movement of the market return, 𝑅𝑚𝑡 is the market return at time t and 𝜀𝑖𝑡 is the error
term for stock i at time t.
To calculate the expected return of stock i we need to calculate the β for stock i and multiply it with the expected market return. However Brown and Warner (1980) found that in studies when observing abnormal returns, simple methodology based on the market model performs well. More importantly, Brown and Warner (1980) find that using the market model does not underperform risk adjusted methods in a short time frame. The formula for the expected return on stock i on time t (𝐸[𝑅𝑖𝑡]) that we use for our calculation is therefore as follows:
Here 𝐸[𝑅𝑖𝑡] is the expected return on stock i at time t, 𝛼̂𝑖 is the expected constant for stock i,
𝛽̂𝑖 is the expected coefficient of the sensitivity of stock I to the movement of the market return, 𝑅𝑚𝑡 is the market return at time t.
𝐴𝑅𝑖 = 𝑅𝑖𝑡− 𝐸[𝑅𝑖𝑡]
As a benchmark market, we used the S&P 500 as our sample consists of US multinationals that operate globally. The market beta is constructed for the S&P 500 index for every firm within a window of 180 days, namely 185 days before the announcement until 5 days before the announcement.
With the announcement being the event at time t=0 we can formulate the formula of our dependent variable of our sample, namely the cumulative abnormal returns, which is defined as follows:
𝐶𝐴𝑅𝑖 = ∑ 𝐴𝑅𝑖𝑡 2
𝑡=−2
The independent variable is a measure of how financially developed the country of the target firm is. Beck (2002) came across the problem that accurate and adequate indicators for financial development across a longer time span and across countries are not available. To solve this, he proposed a proxy variable for financial development. Beck (2002) argues that the main function of the financial sector is the mobilization of savings and allocation of resources to the private sector. He therefore constructed a proxy that reflects this function. For this proxy variable he used the private credit by deposit money banks and other financial institutions to gross domestic product in percentage points. This proxy measures a part of mobilized savings in a country and measures the part that is channelled to private firms. Although this measure is not perfect in capturing direct efficiency, it does capture a
The moderator variable is whether the target firm is more dependent on external finance or if it is not. Following Hadlock and Pierce (2010) we measure dependence on external finance as described in the formula below.
𝐻𝑃 = −0.737 ∗ log(𝑇𝐴) + 0.043 ∗ (log(𝑇𝐴))2− 0.04 ∗ 𝑎𝑔𝑒
The moderator will be constructed by multiplying the capital dependence measure by the financial development measure.
Control variables
Besides the variables of interest, we will also include control variables in our OLS regression. With these control variables we attempt to make sure that our results are not influenced by generally accepted theories in the field.
Moeller, Schlingemann and Stulz (2004) research on firm size and the gains from acquisitions found that small firms gain from acquisitions, where as large firms have a tendency to overpay for their targets. This holds for both public and private firms. Small firms exceed the abnormal announcement return of large firms by 2.24 percentage points in this research. We therefore include the acquirers size as a control variable. Size of the acquirer is defined as follows:
𝑆𝑖𝑧𝑒 = log(𝑏𝑜𝑜𝑘 𝑣𝑎𝑙𝑢𝑒 𝑎𝑐𝑞𝑢𝑖𝑟𝑒𝑟)
When looking at the relationship between abnormal returns and the Tobin’s Q, Servaes (1991) found that the takeover returns (abnormal returns) are higher when the bidder has a high Tobin’s Q score. With this research, Serveas (1991) confirmed the results of an earlier research by Lang, Stulz and walking (1989) who only looked at tender offers but found similar results. Therefore it is important to control for the Tobin’s Q as multiple researches have confirmed that the Tobin’s Q effects abnormal returns. Tobin’s Q is defined as follows:
𝑇𝑜𝑏𝑖𝑛′𝑠 𝑄 =𝑡𝑜𝑡𝑎𝑙 𝑚𝑎𝑟𝑘𝑒𝑡 𝑣𝑎𝑙𝑢𝑒 𝑎𝑐𝑞𝑢𝑖𝑟𝑒𝑟
𝑡𝑜𝑡𝑎𝑙 𝑏𝑜𝑜𝑘 𝑣𝑎𝑙𝑢𝑒 𝑎𝑐𝑞𝑢𝑖𝑟𝑒𝑟
Jensen (1986) argues that the availability of excess cash to management is value destructive. The reasoning behind this is that there is no need for the raising of equity or debt for
which are value destructive. For these reasons we include free cashflow as a control variable. We define free cashflow as follows:
𝐶𝐹 =𝑜𝑝𝑒𝑟𝑎𝑡𝑖𝑛𝑔 𝑖𝑛𝑐𝑜𝑚𝑒 𝑏𝑒𝑓𝑜𝑟𝑒 𝑑𝑒𝑝𝑟𝑒𝑐𝑖𝑎𝑡𝑖𝑜𝑛 − 𝑖𝑛𝑡𝑒𝑟𝑒𝑠𝑡 𝑒𝑥𝑝𝑒𝑛𝑠𝑒𝑠 − 𝑖𝑛𝑐𝑜𝑚𝑒 𝑡𝑎𝑥 𝑏𝑜𝑜𝑘𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑡𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠
Jensen (1986) argues that having debt reduces the agency cost of free cashflow. By having to pay interest, managers are pressured to undertake positive NPV projects and have less wiggle room to pursue private gains and empire building. Moreover, debt holders are an extra
monitor for the management which should improve the firms performance and therefore ad value to the firm. Shareholders know this and will perceive an acquisition to be well analysed by lenders. This will enhance their confidence in the acquisition as being good for the
acquiring firm. A higher leverage therefore is hypothesized to have a positive effect on the abnormal returns for the acquirer and should be controlled for. We define leverage as follows:
𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒 =𝑡𝑜𝑡𝑎𝑙 𝑒𝑞𝑢𝑖𝑡𝑦 𝑡𝑜𝑡𝑎𝑙 𝑑𝑒𝑏𝑡
Alexandridis, Fuller, Terhaar and Travlos (2013) wrote a paper on merger premia and acquisition deal size. They found that deal size is negatively correlated with merger returns, meaning that larger deals are more likely to have negative merger return. Moeller,
Schlingemann and Stulz (2004) found similar results. We thereforee expect that the bigger the deal size the lower the abnormal returns for the acquirer and should control for this. We define deal size as follows:
𝐷𝑒𝑎𝑙 𝑆𝑖𝑧𝑒 = 𝑎𝑐𝑡𝑢𝑎𝑙 𝑡𝑟𝑎𝑛𝑠𝑎𝑐𝑡𝑖𝑜𝑛 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑡ℎ𝑒 𝑎𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑚𝑎𝑟𝑘𝑒𝑡 𝑐𝑎𝑝𝑖𝑡𝑎𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑡ℎ𝑒 𝑎𝑞𝑢𝑖𝑟𝑒𝑟
Scanlon, Trifts & Pettway (1989) researched the impacts of relative size and industrial relatedness on returns to shareholders of acquiring firms. They found that industrial
unrelatedness has a negative impact on merger return, especially when they were relatively big in size. We thereforee control for industry relatedness as well. We use a dummy variable to control for this. The dummy is 1 when the industries are related and 0 when they are not. Industries will be classified as related when the first two digits of their SIC codes are the same (Francis et all, 2008).
account when researching cross-border bidder return. We define global diversification as follows:
𝐺𝑙𝑜𝑏𝑎𝑙 𝑑𝑖𝑣𝑒𝑟𝑠𝑖𝑓𝑖𝑐𝑎𝑡𝑖𝑜𝑛 = 𝑓𝑜𝑟𝑒𝑖𝑔𝑛 𝑠𝑎𝑙𝑒𝑠 𝑡𝑜𝑡𝑎𝑙 𝑠𝑎𝑙𝑒𝑠
Besides the control variables mentioned above, we will also control for time fixed effects and industry fixed effects. We will do so by constructing a dummy variable for every observed year and another dummy variable for every observed industry looking at the first two digits of the SIC codes.
With the research method established, the following part of section will describe how the data set has been build and what restrictions have been taken into account.
Data
In order to conduct this research, we have collected a sample of US affiliated cross-border acquisitions that took place between 1984 and 2015 from the Zephyr database provided by the University of Groningen.
This paper aims to show the effect of financial market development on the value created by cross-border M&A’s and how this effect is affected by the degree of dependence on external capital within firms. To do so, the acquiring firms needs to be in a fully developed financial market. It is widely accepted that the United States of America have the most liquid and one of the, if not the, best functioning financial market (Rajan & Zingales, 2001). Moreover, the US has the largest foreign direct investment cap in the world and should therefore provide a relatively large sample size (Demirgüç-Kunt & Levine, 2004). Due to these considerations this paper looks at US acquirers only.
From the Zephyr database we also gathered the following information of these M&A’s; the first four primary SIC code digits of both target and acquirer, the nationality of the targets, the total assets of the target, age of the target, the value of the transaction and the
announcement date. The remainder of the variables have been obtained from Reuters data stream.
acquisitions where the total amount of shares owned by the acquirer after the acquisition equalled 100%. Like the 1 million dollar restriction, we use this restriction to make sure that the impact on the acquirer is substantial enough. Moreover, firms might be less willing to use their internal capital markets when other shareholders might freeride on this. Furthermore, we will exclude utility industries (SIC codes 4000-4999) and government related industries (SIC codes 9111-9999). We exclude utility and government related industries in accordance with a research conducted by Swanstrom (2006). He found that acquisitions in these industries are often related to government or regulatory intervention in order to save distressed firms that, when bankrupt, would harm the local or even global economy. Moreover, these industries operate in special regulatory environments that could make them behave differently from other industries. Lastly, we require all deals to have a completed deal status and are not listed on a stock exchange outside of their own country.
In the next part of this section we will look at the descriptive statistics of the sample.
Descriptive statistics
Table 1 Descriptive statistics
N Minimum Maximum Mean Standard
deviation CAR 579 -0,4574 0,3500 0,0021 0,0859 Size 579 3,420 8,180 5,869 0,800 Tobins_Q 579 0,170 76,495 4,061 6,054 Leverage 579 -39,280 16,060 0,487 3,086 Deal Size 579 0,20 3835,93 186,60 376,50 Industry Relatedness 579 0 1 0,575 0,495 Global Diversification 579 0 100 35,23 23,47 Financial development 579 10,67 203,12 122,35 38,0542 Capital dependency 579 -2,460 0,300 -1,166 0,487 moderator 579 -28,01 497,41 141,02 73,32 CF 557 -686000 46749000 1348898 4306893
Table 1 shows the descriptive statistic of the sample used in this paper. It describes the following statistics for the variables; the number of observations N, the minimum value measured, the maximum value measured, the mean of the values measured and the standard variation within the observed values. From top to bottom the variables are the cumulative abnormal return (CAR), the size of the acquirer (Size), the Tobin’s Q of the acquirer (Tobin’s_Q), the debt to equity ratio of the acquirer (Leverage), the amount payed for the acquisition (Deal Size), whether the industries of the target and acquirer are related (1) or not (0) (Industry relatedness), the level at which the acquirer was already globally diversified (Global diversification), the level of financial development in the target country (financial development), how capital dependent the target was (capital dependency), the
Table 1 shows the descriptive statistics of the sample. When looking at leverage we see some extreme values of which the minimum is the most remarkable. Normally the leverage for individual firms is positive as neither debt nor equity will be negative. However in reality some firms do report negative equity values and when they do these values are often very small. This results in highly negative leverage values for the acquirer. When looking at financial development we see that, although the minimum and maximum suggest a broad range in variation, the mean is skewed to the right. This suggests that a driving part of the sample comes from target firms in more financially developed countries, while our hypothesized relationship is expected to be weakest here since the difference in financial market development is smaller. To analyse this in more detail we will look at the set of countries in more detail later on in this section.
Graph 1 shows the amount of cross-border M&A’s originating from the US on the vertical axes and the years on the horizontal axes.
Table 2
Financial development per country
Countries Mean N Std. Deviation Minimum Maximum
Japan 182,84 5 5,39 175,64 189,03 Spain 152,86 4 21,67 121,45 167,26 Switzerland 151,96 12 8,38 142,41 168,05 United Kingdom 140,68 232 29,64 82,56 202,2 Denmark 138,13 12 67,03 32,23 203,12 Hong Kong 136,70 1 136,70 136,7 Canada 136,03 101 34,67 74,17 183,83 South Africa 133,67 2 17,77 121,10 146,23 New Zealand 127,31 1 127,31 127,31 China 120,55 2 6,56 115,91 125,18 Ireland-Rep 111,85 6 28,64 70,57 152,67 Singapore 109,40 2 8,74 103,22 115,58 Thailand 106,87 2 13,76 97,14 116,6 Sweden 106,55 24 21,76 64,28 129,39 Netherlands 104,89 7 14,40 82,97 115,47 Malaysia 103,43 2 9,19 96,93 109,92 Australia 102,12 26 24,14 69,47 138,12 Germany 97,38 40 11,87 77,52 116,33 Norway 96,88 8 25,85 69,63 128,44 Austria 94,04 1 94,04 94,04 South Korea 89,15 3 4,69 86,11 94,55 Luxembourg 87,47 1 87,47 87,47 France 85,16 34 7,68 74,45 96,2 Chile 83,45 4 18,45 67,84 103,86 Italy 80,42 14 14,21 57,64 90,01 Israel 74,76 15 8,88 64,56 88,38 Finland 64,27 4 15,23 52,53 85,32 Belgium 59,90 7 2,99 56,20 65,11 Brazil 47,67 1 47,67 47,67 Mexico 20,37 2 6,04 16,10 24,64 Poland 18,74 3 4,05 14,87 22,94 Argentina 10,67 1 10,67 10,67 Total 122,35 579 38,05 10,67 203,12
Table 2 shows the rank of financially developed markets per country by the mean of observations of that particular country. Countries at the top of the list had observations of financial development that gave the highest means and the countries at the bottom had the lowest means. The table further describes the following statistics for financial
When examining Table 2, we can see that financial development changes quite a lot over time. By looking at the standard deviation of each country we see that some countries have a very small variation, like Japan, Belgium and Poland. Other countries have a lot more variation, like Denmark, Canada and the United Kingdom. This can be explained by the fact that in some countries there are only a few observations and for some countries there is no standard deviation at all as they only have one observation. As financial development changes over time more observations in different years will show a financial development increase or decline with high values for standard deviation. To clarify, a country that only has two observations in 2014 and 2015 will show a relatively low standard deviation but the same country with observations in 1990 and 2015 could show a relatively high standard deviation. The same line of thinking explains the difference in the minimum and maximum. This implicates that the financial development variable moves slowly over time. This is not surprising as financial development of a country does not change overnight but is indeed a gradual and slow process.
Next we look at the mean cumulative abnormal returns per country as shown in table 3. Table 3 is structured from the highest financially developed country mean to the least financially developed country mean. We must note that this gives a slightly deformed view on which countries are the most financially developed, as the mean does not necessary say something about the financial development at the year of each observation. This means that table 3 might give a somewhat distorted view on cumulative abnormal returns per country.
Nevertheless, it does provide a crude indication of cumulative abnormal returns magnitude in order of financial development.
According to hypothesis 1 we would expect countries with a higher financial development to have lower means then those of less financially developed countries. However, the highest cumulative abnormal return mean is rather high on the list, namely Hong Kong, and the lowest as third last, namely Mexico. Although we must note that these two countries have only three observations when combined, it is a remarkable figure. Upon further examination of table 3 we see that positive and negative means appear rather random and do not seem to behave according to this ranking. We will elaborate on this further in the results section.
Table 3
Cumulative abnormal returns per country
Countries Mean N
Std.
Deviation Minimum Maximum
Japan 0,0221 5 0,0708 -0,0892 0,0915 Spain 0,0625 4 0,0515 -0,0136 0,0975 Switzerland -0,0109 12 0,0886 -0,1707 0,0952 United Kingdom 0,0072 232 0,0826 -0,2778 0,3500 Denmark 0,0035 12 0,0616 -0,1341 0,0988 Hong Kong 0,1090 1 0,1090 0,1090 Canada -0,0059 101 0,0936 -0,4574 0,2872 South Africa 0,1367 2 0,2579 -0,0457 0,3191 New Zealand -0,0053 1 -0,0053 -0,0053 China 0,0277 2 0,0066 0,0230 0,0324 Ireland-Rep 0,0281 6 0,0326 -0,0274 0,0551 Singapore -0,0748 2 0,1171 -0,1575 0,0080 Thailand -0,0097 2 0,1119 -0,0888 0,0695 Sweden 0,0046 24 0,0756 -0,1314 0,1807 Netherlands 0,0201 7 0,1454 -0,1642 0,3073 Malaysia 0,0045 2 0,0150 -0,0061 0,0151 Australia -0,0156 26 0,1304 -0,3515 0,2584 Germany -0,0040 40 0,0506 -0,1620 0,1346 Norway -0,0241 8 0,0585 -0,1027 0,0888 Austria 0,0138 1 0,0138 0,0138 South Korea 0,0023 3 0,0149 -0,0149 0,0116 Luxembourg -0,0254 1 -0,0254 -0,0254 France -0,0020 34 0,0816 -0,2181 0,1923 Chile -0,0635 4 0,1019 -0,1956 0,0205 Italy 0,0058 14 0,0675 -0,0647 0,1932 Israel -0,0247 15 0,0889 -0,3001 0,0682 Finland 0,0574 4 0,1006 -0,0514 0,1762 Belgium 0,0729 7 0,0820 -0,0144 0,2309 Brazil -0,0025 1 -0,0025 -0,0025 Mexico -0,0974 2 0,2134 -0,2483 0,0535 Poland -0,0126 3 0,0113 -0,0228 -0,0004 Argentina 0,0674 1 0,0674 0,0674 Total 0,0021 579 0,0859 -0,4574 0,3500
Results
In this section we will discuss the results of this research. One assumption of an OLS
regression is that there is no perfect collinearity between variables. To see if this is an issue in our sample, we have constructed a correlation matrix shown in table 4.
Table 4
Correlation table
Size Tobins_Q Leverage Deal_Size Ind_Rel Glo_Div Fin_Int Cap_Dep CAR CF Size -0,136 -0,009 -0,246 -0,025 0,149 0,071 -0,523 -0,098 0,531 Tobins_Q -0,013 -0,097 0,005 -0,013 -0,064 0,081 0,114 -0,013 Leverage -0,049 0,048 0,037 -0,032 -0,006 -0,023 -0,051 Deal_Size 0,064 0,018 0,012 -0,211 0,015 -0,106 Ind_Rel 0,024 -0,014 -0,055 -0,019 -0,035 Glo_Div 0,057 -0,003 -0,02 0,029 Fin_Dev 0,09 0,027 0,004 Cap_Dep 0,023 -0,225 CAR -0,048 CF
Table 4 shows the correlation matrix of the sample used in this paper. From top to bottom and left to right the variables are; size is the log of the total assets of the acquirer, Tobins_Q is Tobin’s Q of the acquirer, leverage is the debt to equity ratio of the acquirer. Deal_Size is the log of the deal value of the M&A, Industry_Relatedness is a dummy variable that shows whether or not the target and acquirer operate in the same industry, Global_Div is a measure of global diversification of the acquirer, CF are the free Cashflow of the acquirer, Financial_Dev is the financial development of the host country. Cap_D is the capital dependency of the target, Financial development * Capital
dependence is the moderator. The corresponding values show the degree to which variables correlate with each other.
In table 4 we can see that, on average, the correlation between the variables is relatively small. Non of the variables show a correlation of more than 0,6, which is the threshold value set for this research at which we would drop a variable due to multicollinearity. Howeverwe must note that we do see a few higher correlation values. Size and capital dependence show a correlation value of -0,523 which is close to our threshold value. Moreover, size correlates with free cashflows relatively well too, with a value of 0,531. This implies that in our sample, larger acquirers also have larger free cashflows to their disposal as we correct free cashflow for the book value of the acquirer. Yet both of these values are not high enough to drop them from our model as they do not exceed the threshold. Furthermore, capital dependence shows a high correlation value with size and capital dependence is the variable that is hypothesized to have a moderating effect on the relationship between financial development and
target being capital dependent. This would make sense as bigger firms have a larger internal capital market and could be better capable of supporting capital dependent targets.
Next, we will look at the relationship between financial development, abnormal returns and the moderating effect of capital dependence on this relationship. We will do this through an OLS regression where the cumulative abnormal return is the dependent variable.
Table 5
variables model 1 model 2 model 3 model 4 model 5
(Constant) 0,067** 0,057* 0,075** 0,067* 0,018 (0,033) (0,034) (0,035) (0,035) (0,044) Size -0,001** -0,011* -0,014** -0,015** -0,015** (0,005) (0,006) (0,007) (0,007) (0,007) Tobins_Q 0,001** 0,001** 0,001** 0,001** 0,001** (0,001) (0,001) (0,001) (0,001) (0,001) Leverage 0,001 0,001 0,001 0,001 0,001 (0,001) (0,001) (0,001) (0,001) (0,001) Deal_Size 0,000 0,000 0,000 0,000 0,000 (0,000) (0,000) (0,000) (0,000) (0,000) Ind_Rel -0,005 -0,005 -0,005 -0,005 -0,006 (0,007) (0,007) (0,007) (0,007) (0,007) Glo_Div 0,000 0,000 0,000 0,000 0,000 (0,000) (0,000) (0,000) (0,000) (0,000) CF 0,000 0,000 0,000 0,000 0,000 (0,000) (0,000) (0,000) (0,000) (0,000) Industry fixed control Yes Yes Yes Yes Yes
Year fixed control Yes Yes Yes Yes Yes Fin_Dev 0,000 0,000 0,000** (0,000) (0,000) (0,000) Cap_Dep -0,007 -0,009 -0,050** (0,009) (0,009) (0,024) Fin_Dev * Cap_Dep 0,000* (0,000) adjusted R squared 0,030 0,031 0,032 0,034 0,039 F-statistic 2,412** 2,180** 2,257** 2,107** 2,233** N 579 579 579 579 579
Table 5 shows the OLS regression results with the variables of interest added in phases. Every phase is indicated as a model. Model 1 does not contain any of the variables that this paper has hypothesized to have an effect on the cumulative abnormal returns, it consists of the control variables alone. In model 1 can be seen that two variables show significance, namely size of the acquirer and the Tobin’s Q, both at a 5% level. However both of these variables have very low coefficients of 0,001 meaning that they explain a very limited movement of the abnormal return. We also observe that model 1 has an adjusted R squared which is relatively small with a value of 0,030. We then add the financial development variable to the model in model 2 and see that size, Tobin’s Q and the constant still remain significant but the constant and size have dropt from a 5% significance level to 10%. Where size has gained 0,010 in its coefficient the constant has lost that same amount. Financial development failed to add any significance or explanatory power on its own, which is not in line with our hypothesis. We also see that the explanatory power of the model gained a bit as the R squared increased by 0,001. In model 3, size, Tobin’s Q and the constant all regain their 5% level significance and the coefficient of the constant increases to 0,075. Capital
dependence failed to explain any significant changes in abnormal returns but is only hypothesized to do so as a moderator on financial development. In model 4 we have added both our variables of interest but do not see any significance for both capital dependence and financial development. Moreover, we see a rather comparable picture in model 4 as we have seen in model 1 with the notable difference that the coefficient of size has grown to -0,015. Finally when we look at the complete model 5, where we include the moderator. It is interesting that in model 5 financial development and capital dependence both become significant at a 5% level while Tobin’s Q and size remain significant at a 5% level as well. The constant has however lost all of its significance. It is however important to note that both the moderator and financial development have a coefficient of 0,000 so despite their
statistical significance, their financial significance is negligible. However, capital dependence shows the highest coefficient of all variables with a value of -0,050. The adjusted R squared of model 5 is also the highest in explanatory power as it possesses the highest adjusted R squared of all models, namely 0,039. Although this is the highest R squared of these models, it is still relatively low. Concluding, this paper cannot support hypothesis 1 or 2 form a financial perspective. Hypothesis 1; firms originating form developed financial markets
acquiring firms form less developed financial markets has a positive impact on abnormal returns, does have statistical significance in model 5, but due to the size of the coefficient for
This implicates that in our sample, the acquisition of a firm situated in a financially less developed country did not result in higher abnormal returns for the acquirer. This is in contrast with the findings of Ma et al. (2009), Barbopoulos et al. (2014) and Francis et al. (2008). Hypothesis 2; capital dependence of target firms positively moderates the effect of
financial development on the abnormal returns of the acquirer, also does have statistical significance in model but the size of the coefficient is not sufficient to support the hypothesis from a financial perspective either. This implicates that capital dependence does not make the
effect of financial development more profound, meaning that abnormal returns for the
acquirer do not seem to be influenced by capital dependence of the target firm in combination with the financial development of the country in which the target is situated.
One of the possible reasons that we do not see a financially significant effect of financial development and capital dependency could be that investors do not know or do not value this effect. A big assumption that is made when working with abnormal returns is that the stock market is fully efficient and therefore reflects the price that firms are actually worth when taking both risk and return into account. However, the literature often argues that stock markets are not perfect and investors cannot be classified as what the literature calls homo economicus (Kaletsky, 2009; Fox & Sklar, 2009). If this is true, then investors have imperfect information and as investors cannot trade on information they do not possess the price is not a true reflection of all information there is. When examining the investors of the acquiring firms, these investors do have access to the information of the acquisition that is about to take place as we measure abnormal returns with a window around the public announcement of the acquisition. Similarly, they have access to information on where the target firms are located and the price that will be paid for the target. The investors might not have the information on how capital dependent these target firms are. In this paper we use a proxy for capital
In the next section we will test, through a couple of robustness tests, whether our results are influenced by the usage of a different cumulative abnormal return window or different subsamples.
Robustness
As we have seen in the descriptive statistics section, Great Britain takes up a large part of our sample. Great Britain alone has 232 observations of the 579 in total, which is over 40% of our sample. To see whether this influences our results, we ran the regression again but this time with the exclusion of Great Britain. Before doing so, we tested for multicollinearity, but found no correlation between variables above the threshold value of 0,6. Table 6 shows the results of the regression. Here we see that throughout the models the only variable that is consistently significant is the Tobin’s Q at a 1% level. All other variables remain
insignificant until model 5. In model 5 we see the same thing happening as in our regression of the whole sample. The moderator, financial development and capital dependence, becomes significant at a 10%, 5% and 5% level. Moreover, the coefficients of financial integration and the moderator remain small just as the coefficient of the Tobin’s Q. Here too we see that capital dependence has the highest coefficient with a value of -0,060. Finally the explanatory power of this model is the same as our whole sample regression with an R squared of 0,039. So we see that, compared to our previous regression, the constant and size have lost their significance but the rest of the models behave rather the same as our previous regression.
A slight shift in the model can be seen when excluding Great Britain. This raises the question if we would see different results when we exclude more “highly” financially developed countries. When we exclude the 6 countries with the highest mean for financial development, which are; Japan, Spain, Switzerland, United Kingdom, Denmark and Hong Kong, we are left with about half our original sample size with 279 observations. Once again no variables exceed the multicollinearity threshold of 0,6, meaning there is no need to drop any variables. In table 7, the results of this regression are shown. In this subsample, Tobin’s Q is significant at a 1% level throughout the models too, but still of little financial significance, as the
coefficient is 0,002. Once again we see that only when the moderator is added to the model financial development, capital dependence and the moderator become statistically significant. In model 5 of table 7 we do see an increase in the R squared to 0,069, which is higher than the other regressions this paper has produced in which these regressions both had an R
Table 6
variables model 1 model 2 model 3 model 4 model 5
(Constant) -0,009 -0,012 -0,006 -0,009 -0,093 (0,048) (0,049) (0,051) (0,051) (0,065) Size 0,003 0,002 0,002 0,001 0,004 (0,008) (0,008) (0,010) (0,001) (0,01) Tobins_Q 0,002** 0,002*** 0,002*** 0,002*** 0,002*** (0,001) (0,001) (0,001) (0,001) (0,001) Leverage 0,001 0,001 0,001 0,001 0,002 (0,002) (0,002) (0,002) (0,002) (0,000) Deal_Size 0,000 0,000 0,000 0,000 0,010 (0,000) (0,000) (0,000) (0,000) (0,000) Ind_Rel -0,002 -0,003 -0,003 -0,003 0 (0,010) (0,010) (0,010) (0,010) (0,000) Glo_Div 0,000** 0,000** 0,000** 0,000** 0,000* (0,000) (0,000) (0,000) (0,000) (0,000) CF 0,000 0,000 0,000 0,000 0,000 (0,000) (0,000) (0,000) (0,000) (0,000) Industry fixed control Yes Yes Yes Yes Yes
Year fixed control Yes Yes Yes Yes Yes Fin_Dev 0,000 0,000 0,001** (0,000) (0,000) (0,000) Cap_Dep -0,003 -0,003 -0,063** (0,014) (0,014) (0,029) Fin_Dev * Cap_Dep 0,001** (0,000) adjusted R squared 0,054 0,054 0,054 0,054 0,069 F-statistic 2,210** 1,938* 1,931* 1,722* 2,008** N 279 279 279 279 279
Table 7
variables model 1 model 2 model 3 model 4 model 5
(Constant) -0,009 -0,012 -0,006 -0,009 -0,093 (0,048) (0,049) (0,051) (0,051) (0,065) Size 0,003 0,002 0,002 0,001 0,004 (0,008) (0,008) (0,010) (0,001) (0,01) Tobins_Q 0,002** 0,002*** 0,002*** 0,002*** 0,002*** (0,001) (0,001) (0,001) (0,001) (0,001) Leverage 0,001 0,001 0,001 0,001 0,002 (0,002) (0,002) (0,002) (0,002) (0,000) Deal_Size 0,000 0,000 0,000 0,000 0,010 (0,000) (0,000) (0,000) (0,000) (0,000) Ind_Rel -0,002 -0,003 -0,003 -0,003 0 (0,010) (0,010) (0,010) (0,010) (0,000) Glo_Div 0,000** 0,000** 0,000** 0,000** 0,000* (0,000) (0,000) (0,000) (0,000) (0,000) CF 0,000 0,000 0,000 0,000 0,000 (0,000) (0,000) (0,000) (0,000) (0,000) Industry fixed control Yes Yes Yes Yes Yes
Year fixed control Yes Yes Yes Yes Yes Fin_Dev 0,000 0,000 0,001** (0,000) (0,000) (0,000) Cap_Dep -0,003 -0,003 -0,063** (0,014) (0,014) (0,029) Fin_Dev * Cap_Dep 0,001** (0,000) adjusted R squared 0,054 0,054 0,054 0,054 0,069 F-statistic 2,210** 1,938* 1,931* 1,722* 2,008** N 279 279 279 279 279
Table 7 shows the OLS regression results with the variables of interest added in phases. Every variable shows two values the top one being the coëficient and the bottom one being the standard deviation. P-value significance at a 1%, 5% and 10% level is indicated by ***, **, * respectively and shown behind the coeficient.
Table 8
variables model 1 model 2 model 3 model 4 model 5
(Constant) 0,057 0,060 0,037 0,040 -0,044 (0,034) (0,035) (0,036) (0,036) (0,046) Size -0,009 -0,009 -0,003 -0,002 0,001 (0,006) (0,006) (0,007) (0,007) (0,007) Tobins_Q 0,002*** 0,002*** 0,002*** 0,002*** 0,002*** (0,001) (0,001) (0,001) (0,001) (0,001) Leverage 0,001 0,001 0,002 0,002 0,002 (0,001) (0,001) (0,001) (0,002) (0,001) Deal_Size 0,000 0,000 0,000 0,000 0,000 (0,000) (0,000) (0,000) (0,000) (0,000) Ind_Rel -0,001 -0,001 0,000 0,000 -0,001 (0,007) (0,007) (0,007) (0,07) (0,007) Glo_Div 0,000 0,000 0,000 0,000 0,000 (0,000) (0,000) (0,000) (0,000) (0,000) CF 0,000 0,000 0,000 0,000 0,000 (0,000) (0,000) (0,000) (0,000) (0,000) Industry fixed control Yes Yes Yes Yes Yes
Year fixed control Yes Yes Yes Yes Yes Fin_Deb 0,000 0,000 0,001** (0,000) (0,000) (0,000) Cap_D 0,017* 0,018* -0,042* (0,010) (0,010) (0,022) Fin_Dev * Cap_Dep 0,001*** (0,000) adjusted R squared 0,090 0,091 0,073 0,102 0,130 F-statistic 3,841*** 3,383*** 3,755*** 3,377*** 4,005*** N 579 579 579 579 579
Table 5 shows the OLS regression results with the variables of interest added in phases. Every variable shows two values the top one being the coëficient and the bottom one being the standard deviation. P-value significance at a 1%, 5% and 10% level is indicated by ***, **, * respectively and shown behind the coeficient.
Conclusion
In this paper we integrated theories form the finance literature to look closer at the added value of mergers and acquisitions to less financially developed markets by looking at shareholder’s returns around the announcement dates of these acquisitions. The research of this paper focusses primarily on the degree of financial development of the host country, and capital dependence of the individual firms and their effect on abnormal returns as an
indication of added value. The results show significance for both hypothesis 1; firms
originating form developed financial markets acquiring firms form less developed financial markets has a positive impact on abnormal returns and hypothesis 2; capital dependence of target firms positively moderates the effect of financial development on the abnormal returns,
but they do so with very small coefficients meaning that their financial significance is small. These coefficients do seem to be influenced in our robustness tests albeit very little, which could imply that the obtained sample for this research was skewed too much to developed markets. When looking at the robustness tests where the top 6 financially developed countries are excluded, not only is there an increased significance but also higher coefficients albeit still to small to argue that value is added by a financially significant amount. Apart from the aforementioned, we do see rather consistent results throughout our robustness tests. These results imply that this research cannot conclude that acquiring a firm in a less financially developed market adds financially significant value. This is an unexpected result as our hypothesized effect based on literature did predict that value would be added through these acquisitions. The same has to be concluded for the moderating effect.
In the next section we will address some of the limitations of this paper and we will make a couple of recommendations for future research.
Limitations and future research
Another limitation of this paper is that the only a small proportion of the data on target firms came from financially less developed countries or segmented financial markets as described by Francis et al. (2008). Of our 579 observed target firms, only 17 firms came from these segmented countries. This is due to fewer M&A’s from the US to these countries, but more so due to the sporadic data availability on target firms form these countries. This paper did not use a dummy like Francis et al. (2008) did in their paper but a proxy for financial
development. However it is important to note that these least developed financial markets or financially segmented countries as Francis et al. (2008) refers to them, are of specific interest to MNC’s when arguing from a financial development point of view. For future research it would be interesting to see how the relationship between financial development and
cumulative abnormal returns reacts to a better balanced sample with respect to the proportion of target firms in these financially less developed countries.
We make another assumption in this paper, namely that the cost of capital stays the same for the acquiring company after the acquisition. Although we do control for size of the acquiring firm, we cannot directly perceive the effect of the consolidation of the target firm on the cost of capital of the acquirer. If the acquired target would have an effect on the cost of capital of the acquirer, this could lead to an increase in the cost of capital which would partially diminish the effect (Bekaert & Harvey, 1997) which we hoped to show with this paper.
An additional limitation is that financial development is very complex and hard to measure. In this paper we use the private credit by deposit money banks and other financial institutions to gross domestic product in percentage points as used by Beck (2002), but it is important to note that this is only a proxy. Although there are a lot of suggestions on the calculation of financial development, none of these measures has been widely adopted by the literature, as all have their downsides. We therefore strongly encourage future research to come up with an alternative measure for financial development, so research regarding this topic can be
As pointed out in the results section, information asymmetries or lack of investor knowledge could influence the abnormal returns measured in this paper. Therefore it would be
interesting to see how these firms perform in the period after the acquisition. In this paper we argued that the access to the internal capital market of the acquirer would provide lower cost of capital for the target which would make more positive NPV projects possible, so we should see these acquired firms take on more or bigger projects in the period after the acquisition. Due to consolidation of the firms balance sheet and income statement by the acquirer, this information is unfortunately often not available. Moreover, a lot of firms acquired in our sample were not listed and therefore did not have to make their balance sheet and income statement public. This information is necessary to compare the firms previous performance with its performance after the acquisition. The collection of this data was
unfortunately outside of the scope of the paper, but we strongly recommend future research to invest the time and effort to collect this data as we think it would give a more complete picture of the internal capital argument.
Reference list
Agrawal, A., & Walkling, R. A. (1994). Executive careers and compensation surrounding takeover bids. The Journal of Finance, 49(3), 985-1014.
Agrawal, A., Jaffe, J. F., & Mandelker, G. N. (1992). The post‐merger performance of acquiring firms: a re‐examination of an anomaly. The Journal of finance, 47(4), 1605-1621. Alexandridis, G., Fuller, K. P., Terhaar, L., & Travlos, N. G. (2013). Deal size, acquisition premia and shareholder gains. Journal of Corporate Finance, 20, 1-13.Almeida, H., &
Amburgey, T. L., & Miner, A. S. (1992). Strategic momentum: The effects of repetitive, positional, and contextual momentum on merger activity. Strategic Management
Journal, 13(5), 335-348.
Asquith, P. (1983). Merger bids, uncertainty, and stockholder returns. journal of Financial
Economics, 11(1-4), 51-83.
Banerjee, A., & Eckard, E. W. (1998). Are mega-mergers anticompetitive? Evidence from the first great merger wave. The Rand Journal of Economics, 803-827.
Barbopoulos, L., Marshall, A., MacInnes, C., & McColgan, P. (2014). Foreign direct investment in emerging markets and acquirers’ value gains. International Business Review, 23(3), 604-619.shareholder gains. Journal of Corporate Finance, 20, 1-13.
Beck, T. (2002). Financial development and international trade: Is there a link?. Journal of international Economics, 57(1), 107-131.
Bekaert, G., & Harvey, C. R. (1997). Emerging equity market volatility. Journal of Financial economics, 43(1), 29-77.
Bjorvatn, K. (2004). Economic integration and the profitability of cross-border mergers and acquisitions. European Economic Review, 48(6), 1211-1226.
Brown, S. J., & Warner, J. B. (1980). Measuring security price performance. Journal of financial economics, 8(3), 205-258.
Campbell, J. L., Dhaliwal, D. S., & Schwartz Jr, W. C. (2011). Financing constraints and the cost of capital: Evidence from the funding of corporate pension plans. The Review of
Financial Studies, 25(3), 868-912.
Campello, M. (2007). Financial constraints, asset tangibility, and corporate investment. The Review of Financial Studies, 20(5), 1429-1460
Capron, L., Dussauge, P., & Mitchell, W. (1998). Resource redeployment following horizontal acquisitions in Europe and North America, 1988-1992. Strategic management
journal, 631-661.
Chang, S. (1998). Takeovers of privately held targets, methods of payment, and bidder returns. The Journal of Finance, 53(2), 773-784.
Child, J., Faulkner, D., & Pitkethly, R. (2001). The management of international acquisitions. Oxford University Press.
Demirgüç-Kunt, A., & Levine, R. (Eds.). (2004). Financial structure and economic growth: A cross-country comparison of banks, markets, and development. MIT press.
Denis, D. J., Denis, D. K., & Yost, K. (2002). Global diversification, industrial diversification, and firm value. The Journal of Finance, 57(5), 1951-1979.
Diamonte, R. L., Liew, J. M., & Stevens, R. L. (1996). Political risk in emerging and developed markets. Financial Analysts Journal, 52(3), 71-76.
Doukas, J. (1995). Overinvestment, Tobin’s Q and gains from foreign acquisitions. Journal of Banking & Finance, 19(7), 1285-1303.
Edison, H. J., & Warnock, F. E. (2003). A simple measure of the intensity of capital controls. Journal of Empirical Finance, 10(1), 81-103.
Errunza, V. R., & Miller, D. P. (2000). Market segmentation and the cost of the capital in international equity markets. Journal of Financial and Quantitative analysis, 35(4), 577-600.
Folta, T. B. (1998). Governance and uncertainty: The trade-off between administrative control and commitment. Strategic management journal, 1007-1028.
Francis, B. B., Hasan, I., & Sun, X. (2008). Financial market integration and the value of global diversification: Evidence for US acquirers in cross-border mergers and
acquisitions. Journal of Banking & Finance, 32(8), 1522-1540.
Francis, B. B., Hasan, I., Lothian, J. R., & Sun, X. (2010). The signaling hypothesis revisited: Evidence from foreign IPOs. Journal of Financial and Quantitative Analysis, 45(1), 81-106.
Fuller, K., Netter, J., & Stegemoller, M. (2002). What do returns to acquiring firms tell us? Evidence from firms that make many acquisitions. The Journal of Finance, 57(4), 1763-1793.
Graebner, M. E., & Eisenhardt, K. M. (2004). The seller's side of the story: Acquisition as courtship and governance as syndicate in entrepreneurial firms. Administrative Science
Quarterly, 49(3), 366-403.
Hadlock, C. J., & Pierce, J. R. (2010). New evidence on measuring financial constraints: Moving beyond the KZ index. The Review of Financial Studies, 23(5), 1909-1940.
Hail, L., & Leuz, C. (2009). Cost of capital effects and changes in growth expectations around US cross-listings. Journal of financial economics, 93(3), 428-454.
Haleblian, J., Devers, C. E., McNamara, G., Carpenter, M. A., & Davison, R. B. (2009). Taking stock of what we know about mergers and acquisitions: A review and research agenda. Journal of Management, 35(3), 469-502.
Haunschild, P. R. (1993). Interorganizational imitation: The impact of interlocks on corporate acquisition activity. Administrative science quarterly, 564-592.
Henry, P. B. (2000). Stock market liberalization, economic reform, and emerging market equity prices. The Journal of Finance, 55(2), 529-564.
Hitt, M. A., Harrison, J. S., & Ireland, R. D. (2001). Mergers & acquisitions: A guide to creating value for stakeholders. Oxford University Press.
Hutson, E., & Laing, E. (2014). Foreign exchange exposure and multinationality. Journal of Banking & Finance, 43, 97-113.
Jarrell, G. A., & Poulsen, A. B. (1989). The returns to acquiring firms in tender offers: Evidence from three decades. Financial management, 12-19.
John, K., Freund, S., Nguyen, D., & Vasudevan, G. K. (2010). Investor protection and cross-border acquisitions of private and public targets. Journal of Corporate Finance, 16(3), 259-275.
Kaletsky, A. (2009). Goodbye, homo economicus. real-world economics review, (58), 151-156.
Kaplan, S. N., & Zingales, L. (1997). Do investment-cashflow sensitivities provide useful measures of financing constraints?. The quarterly journal of economics, 112(1), 169-215.
Kwok, C. C., & Reeb, D. M. (2000). Internationalization and firm risk: An upstream-downstream hypothesis. Journal of International Business Studies, 31(4), 611-629.
Lang, L. H., Stulz, R., & Walkling, R. A. (1989). Managerial performance, Tobin’s Q, and the gains from successful tender offers. Journal of financial Economics, 24(1), 137-154.
Loderer, C., & Martin, K. (1992). Postacquisition performance of acquiring firms. Financial
Management, 69-79.
Ma, J., Pagan, J. A., & Chu, Y. (2009). Abnormal returns to mergers and acquisitions in ten Asian stock markets. International Journal of Business, 14(3), 235.
Malatesta, P. H. (1983). The wealth effect of merger activity and the objective functions of merging firms. Journal of financial economics, 11(1-4), 155-181.
McGuckin, R. H., & Nguyen, S. V. (1995). On productivity and plant ownership change: New evidence from the longitudinal research database. The RAND Journal of Economics, 257-276.
Moeller, S. B., & Schlingemann, F. P. (2005). Global diversification and bidder gains: A comparison between cross-border and domestic acquisitions. Journal of Banking & Finance, 29(3), 533-564.
Moeller, S. B., Schlingemann, F. P., & Stulz, R. M. (2004). Firm size and the gains from acquisitions. Journal of Financial Economics, 73(2), 201-228.
