The impact of government interventions on firm value and how
ATPs influence hostile takeovers
Olaf Schoenmaker s2378191 MSc Finance University of Groningen
Supervisor: Drs. Boris van Oostveen January 2018
ABSTRACT
This study addresses the role governments play during hostile takeover attempts by analysing hostile takeover bids between 2000 and 2016 in Continental Europe and the Anglo-Saxon countries. Anti-Takeover Provision (ATP) application by targets’ management seems to enhance negotiation power of the target’s shareholders while it does not enable management to block a hostile takeover attempt. In line with the stockholder interest hypothesis targets’ management do not use ATPs to protect themselves but rather to increase the bid. Even installing multiple ATPs management cannot halt the takeover attempt. At this point, concluded is that (in)direct
government interventions do not lead to an increase in the value of the firm.
Interestingly, the longer the bid is outstanding, the higher the likelihood of success of a hostile takeover attempt. Findings suggest that shareholder protection and
takeover regulation can be limiting the openness of the domestic hostile takeover market.
Keywords: Anti-takeover provisions (ATPs), hostile takeover, government
intervention, shareholder value, institutional context, duration.
1
1. Introduction
Governments sometimes support companies that use ATPs to prevent being taken over in a hostile manner. After the offer of Kraft Heinz on Unilever, mid-February 2017, which was rejected by Unilever, the share price surged directly. Unilever declined the bid because it did not see financial or strategic reason for its
shareholders. Additionally, Unilever believed the offer was highly undervaluing the company. The attempt has raised some questions on the ability of Unilever to defend itself, since Unilever, until now, does not have Anti-Takeover Provisions (ATPs) installed. Therefore, there are two outcomes, either Unilever acquirers a big company itself and by doing so becomes too big to become acquired, or it will be acquired in the future. In the meantime, the shareholders are expecting Unilever to defend itself in a proper manner while the share price increased right after the unfriendly offer. The situation Unilever currently finds itself in, is one of several in which big
international firms are positioned. The Dutch government takes a prominent role in the debate by addressing the importance of such multinationals and aims to
counteract takeovers of firms with a large societal impact. In the present takeover landscape, there is a lot of activity, approximately 50.000 transactions with an accumulated value of more than 3.5 trillion USD in 20161. This also motivates the reason for this study. It is engaging to investigate the relation between ATPs, government intervention and the shareholder value.
Deploying ATPs can be explained by the conflict of interest between managers and shareholders. Having ATPs installed can lead to an increase in the takeover price due to the increase in bargaining power (Comment and Schwert, 1995). This can obviously translate into an increase in the value of the company in the situation of a successful takeover. However, the general belief is that ATPs negatively impact firm performance and shareholder value due to limitations in terms of the market effect of corporate control. It is argued that managers seek to entrench themselves and thus protect their position by deploying ATPs. This can imply that shareholders see the value of the firm decreasing. This can be explained following the principal-agent problem between managers, shareholders and the government. Target’s
shareholders may be interested in return on investment maximization, whereas managers are looking to keep their jobs and governments do not want the big firms to downsize or potentially leave the country. This misalignment problem is especially
2 present in the case of hostile takeover attempts (Armour and Skeel, 2007; DeAngelo and Rice, 1983 and Jensen and Meckling, 1976).
In the hostile takeover context, the role governments play is particularly important since governments create and implement takeover regulation and hence define the playing field for hostile takeovers. (In)Direct government intervention in the case of corporate takeovers is often subject to national interests (Dinc and Erel, 2013).
Recent developments in Europe show the influence governments appropriate. The loi Florange, in France, automatically grants double voting rights to shares that are registered for more than two years, unless two-thirds of the shareholders vote to repeal it. In addition, the mandatory Board Neutrality Rule (BNR) is eliminated. This neutrality rule provides that from start to end of the offer the target board may not take any ‘frustrating action’ which might cause the offer to fail. However, seeking alternative bidding companies is allowed under the BNR. Furthermore, the country enabled managers to use ATPs without shareholder approval. In the Netherlands, the government wants to implement a cool down period of one year in the case of a hostile takeover. The idea behind this is to give companies more space and time to consider the takeover. Depending on the evaluation, the company can deny or accept the offer. At the same time, the shareholders can also assess the offer. Both countries were triggered to change legislation after a hostile takeover threat of one of the large and important national companies. Moreover, the European Takeover Directive was introduced by the European Parliament and came into force in 2004 in order to harmonize rules related to: ‘the conduction of takeovers bids and
shareholder protection in the EU’. However, it allows member states to opt-out from the mandatory board neutrality. Overall, governments have the ability to influence takeovers directly through interventions and indirectly via takeover regulation (Dinc and Erel, 2013).
The discrepancy in institutional context differently impacts the influence ATPs have on shareholder value. The Anglo-Saxon system has long been viewed as superior, however, the stakeholder oriented, Continental European system gains momentum. Supporters of the stakeholder oriented view argue that large shareholder monitoring deals better with the conflict of interest between shareholders and managers
compared to market monitoring. Reasoned is that the stakeholder oriented
3 This study answers the question what the impact of government interventions is on the relation between ATPs and firm value during, and after hostile takeovers, this is done in six sequential steps. Hostile takeover bids are useful in this case since they pronounce the conflict of interest between protection of the domestic industry and corporate governance. First, the impact of having ATPs installed is analysed. By investigating the likelihood of increasing bids, final takeover premium, and bid success, it is examined whether managers entrench themselves. Secondly, the impact of (in)direct government interventions on hostile takeovers and the implications on the outcome of the takeover attempt are examined. Thirdly, the impact of a takeover attempt on the firm value at/one month after
completion/withdrawal date is analysed. Fourthly, the implications of the presence of BNR is tested. Specifically, does the existence of BNR affect the openness of the takeover market. Fifthly, analysed is whether the duration of the takeover attempt has an impact on the probability of takeover success. Lastly, the impact of the institutional context on the usage of ATPs is analysed.
The contribution of this study is fourfold. First, the impact of a hostile takeover attempt on the value of the target firms is measured. Considering the share price of the target firm as firm value has not been done in previous research. It is
accomplished by looking at the share price at the completion/withdrawal date and one month after the completion/withdrawal date. Secondly, this is one of the first studies to analyse the impact of duration of the takeover attempt on its success probability. Thirdly, this study includes an analysis of the different institutional contexts. The differences between ownership and control structures, in the specific institutional contexts, differently impact shareholder value due to the dissimilar influence ATPs have on shareholder value. Hence, the analysis is based on countries being either Anglo-Saxon or Continental European. More specifically, Australia, Canada, Ireland, New Zealand, United Kingdom and the United States are compared to the Continental European countries. Lastly, since it is an extension of the research conducted by Rowoldt and Starke (2014) it adds value because it covers a longer time period with more observations and therefore decreases the significance problem. The study by Rowoldt and Starke (2014) has a small sample of 9 negative interventions in hostile takeovers in the period 2000-2014. This is mainly because direct interventions are rare events.
4 on the impact ATPs have are highlighted. The second section also discusses the impact of government intervention during takeover attempts. Additionally, the institutional characteristics of Continental European and Anglo-Saxon countries are explained. After that the methodology is described in the third part. The fourth section describes the dataset that is used and the fifth part presents the results. Finally, the sixth section provides a conclusion and rounds up the analysis by addressing the limitations and suggestions for further research.
2. Literature review and Hypotheses
This section highlights the preceding literature and includes the hypotheses. The literature can be grouped into four parts. The first part introduces the topic of ATPs, describes why they are implemented and illustrates the two conflicting hypotheses: the managerial entrenchment hypothesis and the stockholder interest hypothesis. The second part highlights the impact of direct and indirect interventions on acquisition outcome. After this, the final part evolves around the divergent
implications differences in ownership and control structures, between institutional contexts have on ATPs and its usage.
2.1 Anti-takeover provisions
ATP deployment during hostile takeover attempts can either have a positive or negative effect on acquisition outcome. The fact that managers do not always make acquisitions that are value maximizing for the shareholders is widely recognized. Morck et al. (1989) present various acquisitions that hurt shareholders while at the same time give managers substantial benefits. In order to mitigate the agency problem between the manager and shareholder several corporate governance
5 attempt. Which translates into higher expected costs of the takeover and a decrease in the probability of success (Bebchuk et al., 2007; Field and Karpoff, 2002).
DeAngelo and Rice (1983) present a theoretical and empirical analysis of two
contrasting explanations for the implications of ATPs on shareholder value. By doing so, they were one of the earlier researchers who argued that ATPs can have an influence on shareholder wealth. The first view is the managerial entrenchment hypothesis. They reasoned that ATPs provide protection of incumbent managers at the expense of shareholders. Hence, the existing agency conflict between
shareholders and managers increases due to ATPs. On the other hand, the
stockholder interest hypothesis states that the dominant effect of ATPs is beneficial to stockholders by increasing the expected gains from transferring corporate control.
Previous studies investigating the impact of ATPs on acquisition outcome find
contrasting results. The fact that ATPs increase bargaining power and therefore lead to an increase in the takeover premium, which is beneficial to shareholders is widely supported (Comment and Schwert, 1995; Gaughan, 2011; Grossman and Hart, 1980; Heron and Lie, 2006; Holmén et al, 2012 and Varaiya 1987). In contrast, DeAngelo and Rice (1983) find weak evidence in support of the managerial
entrenchment hypothesis. This is supported by Bebchuk and Cohen (2005), Bebchuk et al. (2009), Cremers and Nair (2005) and Gompers et al. (2003) who all find that ATPs enable managers to deter hostile bids and harm the disciplinary mechanism. Focussing on the long term, a negative relationship between ATPs and firm value as well as long term share performance is found (Bebchuk and Cohen, 2005; Bebchuk et al., 2009).
Managers may benefit from deploying ATPs to prevent hostile takeovers by keeping control of the firm. The managerial entrenchment hypothesis relies on four crucial assumptions (DeAngelo and Rice, 1983). The first assumption is that control is
6 replaced. Thirdly, assumed is that this market for corporate control is the best
available governance mechanism to deal with managerial incompetence. Put
differently, DeAngelo and Rice (1983) assume that the market for corporate control is more adequate, as governance mechanism, in coping with the agency conflict
compared to shareholder monitoring or incentive based compensation. The final assumption is that ATPs are protective measures, installed by managers, coming at the costs of shareholders. This because the likelihood of a takeover, aimed at correcting managerial incompetence, decreases. In other words, managers prevent shareholder maximizing takeovers via ATPs. If all four assumptions hold, the
disciplinary role of the market for corporate control is weakened by ATPs and it can lead to behaviour that is not value maximizing. This implies that in the end, ATPs come at the expense of shareholders (Jensen and Meckling, 1976). Chawla (2015) finds that when ATPs interact, overcoming them, and thus taking over a target
becomes very difficult. Bebchuk et al. (2007), Pound (1987) and Subramanian (2003) focus on actual hostile takeover bids and provide further evidence supporting the management entrenchment hypothesis. They show that ATPs do not increase the premium offered but instead decrease the likelihood of completion of the transaction.
Shareholders may benefit from hostile takeovers due to the strategy realignment in the case a takeover attempt fails or a more efficient resource allocation when the takeover succeeds. The threat of a takeover itself may be sufficient to force managers to adjust the strategy in order to realign its interests closer to the
shareholders’ interests. The potential takeover party monitors the managers and the takeover possibility serves as a threat to minimize non-value maximizing behaviour (Armour and Skeel, 2007; Deakin and Slinger, 1997; Easterbrook and Fischel, 1981; Jensen and Ruback, 1983; Manne 1965; Scharfstein, 1988 and Van Apeldoorn and Horn, 2007). Besides this, the mere threat may discipline managers because a takeover could imply that the respective managers will be replaced following the takeover (Armour and Skeel, 2007; Martin and McConnell, 1991). In addition, a successful (hostile) bid can increase a company’s performance via a more efficient use of the available resources. This can be achieved by the created synergies between the two companies (Burkart, 1999; Deakin and Slinger, 1997 and Manne, 1965).
7 might lead to increases in bids and increasing takeover premiums. In contrast, the management entrenchment hypothesis states that ATPs benefit entrenched targets’ managers since ATPs can defeat takeover attempts and hence harm shareholders’ interests. In this regard, tested is what impact ATPs have on acquisition outcome. In order to measure acquisition outcome three aspects are considered. 1) an increase in the bid, 2) the final takeover premium offered and 3) the probability of the success of a hostile bid. Also, the impact of having multiple ATPs and the number of ATPs are analysed accordingly.
H1: The presence of ATPs positively impacts acquisition outcome.
2.2 (In)Direct government interventions
The mixed results, on the impact ATPs have on acquisition outcome, found in the literature could be attributable to the (in)direct impact the government has on hostile takeovers. Governments directly and indirectly impact hostile takeover attempts. Directly via opposing specific hostile takeover attempts and indirectly via takeover regulation. First, the focus will be on direct interventions and afterwards, implications of indirect interventions are described.
Direct government intervention is likely to follow protectionist reasons. Following Dinc and Erel (2013) it is expected that the probability of a negative government
intervention increases when the hostile takeover bid is from outside the home country. Furthermore, it is expected that direct interventions have a positive impact on firm value (Dinc and Erel, 2013). This is since the intervention would only be necessary in the situation that the company is rather important for the country. Additionally, Rowoldt and Starke (2016) argue that direct interventions are done as ultimo ratio and hence decrease the probability of takeover success.
8 bidders, in the form of BNR (Aktas et al. 2007; Davies et al. 2010; Hill, 2010 and Hopt, 2014). Specifically, the presence of BNR in takeover regulation increases the probability of cross border hostile bids.
Following the news that a minister of the Dutch government proposed to implement a cooling down period after a hostile takeover attempt was announced (June 2017), the duration of a bid is also analysed. As described before, the idea of this period is to make sure all aspects of the takeover attempt are clear to all stakeholders. In
addition, a company can defend itself more firmly against a hostile takeover attempt in this period of time. Renneboog and Zhao (2014) investigate the relationship between corporate networks and takeover process. They find that when companies have one or more directors in common, the duration of the process is lower and the success probability is higher. In addition, Calcagno et al. (2017) analyse the
determinants of takeover process duration. However, both studies do not focus on the impact of duration on success. The only study so far in this respect is executed by Dickson, Gibson and Tsakalotos (2002). They find that when the duration is longer the probability of a successful takeover decreases. It is expected that an increase in the duration of a takeover attempt leads to a decrease in the success of this attempt. The Dutch minister suggested a cooling down period of one full year. In this
research, looked is what the impact of duration on success is, and the impact of an attempt that lasted for a period of more than 100 days is analysed.
H2: Government intervention positively impacts acquisition outcome.
H3: Government intervention increases the value of the target firm at/ one month after completion or withdrawal date.
H4: The presence of BNR in takeover legislation increases the probability of a cross border hostile bid.
H5: The probability of success of a takeover attempt decreases with the duration (over 100 days) of a bid.
2.3 Institutional context
9 argue that the difference in ownership between Continental Europe and the Anglo-American model implies that corporate governance mechanisms and ATPs’ influence on shareholder value are unequal due to the difference in institutional environment.
Continental Europe's institutional context is oriented towards large shareholders whereas the Anglo-Saxon countries are more market oriented (Schleifer and Vishny, 1997). Anglo-Saxon countries are characterized by low concentration of
shareholders, more specifically no shareholder holds more than 5% of the
outstanding shares. This implies that shareholders generally do not have significant power. Thus, management has the power to decide. Inversely, in Continental
Europe, a few shareholders hold large percentages of the firm’s outstanding shares. These shareholders will be able to exercise their power and control the firm.
Summarizing, Continental Europe is known for relatively weak managerial power and strong shareholders whereas on the contrary Anglo-Saxon countries have relatively strong managerial power and weak shareholders. Goergen et al. (2005) argue that the main advantage of the shareholder oriented system is the way it addresses misalignment between shareholders and managers. In Anglo-Saxon countries, the market monitors and regulation forces managers to follow shareholders’ interest. On the other hand, continental European countries rely mainly on large shareholder, employee or creditor monitoring. Since Anglo-Saxon countries are characterized by strong managers, expected is that they are more likely to use ATPs, compared to managers in Continental European countries.
H6: The probability of using ATPs is lower in Continental European countries.
3. Research Methodology
This section explains the models used to analyse the impact of government interventions on acquisition outcome. First, calculated is what the actual takeover premium is.
In accordance with Rossi and Volpin (2004) and Subramanian (2003) the takeover premium is measured using the following formula:
(1) !"#$%&$' )'$*+,*- = /0-12 34456789165 :60;589165 :60;5 <
< = 100
10 The Final Offer expresses the ultimate takeover price that is paid per share in US Dollar. Share Price means the share price of the target firm at time t, one week prior to the announcement. Information leakage can have a substantial impact on the share price, prior to the bid (Rowoldt and Starke, 2016). In order to avoid this, the share price one week prior to the announcement is considered.
A hostile bid is considered successful if the acquisition is ‘completed’ and the acquirer holds more than 50% of the shares after the acquisition, i.e. the bidder gained control. This is true for 33,1%of the hostile takeover attempts.
(2) @"A,$ BCD'$"E$- = -5F 89165 :60;589165 :60;5G789165 :60;5<
< = 100
(3) @"A,$ BCD'$"E$1- = -5F 89165 :60;5H89165 :60;5I789165 :60;5<
< = 100
Using the same line of reasoning, to look at the value of the target company after the takeover attempt, looked is at the Value Increase. The Value Increase is the
difference in the share price, four weeks prior to the announcement and at the
completion/withdrawal date, as a percentage. The new Share Price is the share price of the target firm at time z, at the completion/withdrawal date. Share Price means the share price of the target firm at time t, four weeks prior to the announcement.
Furthermore, the effect of the actual completion or withdrawal is captured by looking at the share price one month after completion or withdrawal, Value Increase1. The Value Increase1 is the difference in the share price, four weeks prior to the
announcement and four weeks after the completion/ withdrawal date, as a
percentage. The new Share Price1 is the share price of the target firm at time x, one month after the completion/withdrawal date. Again, Share Price means the share price of the target firm at time t, four weeks prior to the announcement.
All transactions, withdrawn and completed, are included in this study. This is done to analyse the impact ATPs, negative government intervention and duration have on success. The results separating completion and withdrawal date are reported in the appendix, but are not further discussed.
11 amount of observations exceeds 1.000, which is not the case. Thus, the probit model is preferred because it accounts for constant error variation and the errors of the probit model follow a normal distribution. Furthermore, it is the more widely used method in economics. The dependent variable Acquisition outcome consists of three parts; Increase in bid, Takeover premium and Success. Increase in bid, Takeover premium and Success are different characteristics related to the outcome of the takeover attempt and all have an impact on the value of the firm. The dependent variables ‘Increase in bid’ and ‘Success’ are binominal, they can take one of two possible values, one in the case if ‘true’ and zero in the case of ‘non-true’ (formula 4). As described above, Takeover premium is calculated as a percentage and hence is not restricted to range between zero and one. Therefore, ordinary OLS is the
preferred method (formula 5). The independent variable ATPs also consists of three (binary) elements; J!)H is the fact whether an ATP has been used, J!)K whether multiple ATPs have been used and J!)L how many ATPs have been used.
(4) M+'*N$'O%'*"CD$0P = s KQH $7H/K(GT<
U s)
with W0P = bH +bKJ!)E0P+bL +C Y+'$DZ[%&$'C*$CZBCZ$'&$CZ+%C0P + b\Y+'$DZ[%&$'C*$CZBCZ$'&$CZ+%C=J!)0P+ b]^%CZ'%AE0P + D0 + &P+Î0P
Where, Firmperformance is the binary dependent variable that consists of Increase in bid or Success, bH is a constant, bK is the coefficient for the independent variables ATPs, bL is the coefficient for the independent variable
(in)directGovernmentIntervention, b\ is the coefficient for the independent variable directGovernmentInterventionxATP, b] is the coefficient for the control variables, D0 is
the industry dummy, &P is the year dummy and Î0P is the error term.
(5) !"#$%&$' )'$*+,*- = "H+ "KJ!)E0P+ "L +C Y+'$DZ[%&$'C*$CZBCZ$'&$CZ+%C0P+
"\ Y+'$DZ[%&$'C*$CZBCZ$'&$CZ+%C=J!)0P+ "]^%CZ'%AE0P+ D0 + &P+Î0P (6) @"A,$ BCD'$"E$- 1 = "H+ "KJ!)E0P +
12 !"#$%&$' )'$*+,*-, @"A,$ BCD'$"E$- and @"A,$ BCD'$"E$1- are calculated following formula 1, 2 and 3. Equation 5 and 6 show the respective OLS regression models. Where aH is a constant, "K is the coefficient for the independent variables J!)E, "L is
the coefficient for the independent variable +C Y+'$DZ[%&$'C*$CZBCZ$'&$CZ+%C, "\ is
the coefficient for the independent variable Y+'$DZ[%&$'C*$CZBCZ$'&$CZ+%C=J!) and "] is the coefficient for the control variables. D0 is the industry dummy and &P is the
year dummy and Î0P is the error term. Equation 6 considers both @"A,$ BCD'$"E$
-and @"A,$ BCD'$"E$1-.
4. Data
This section describes the dataset used in this study. The sample is presented in terms of country, industry and whether ATPs have been used. Furthermore, the control variables are defined. They can be grouped into country, deal and target level control variables.
4.1 Data description
This study addresses the role governments play during hostile takeover attempts by analysing 314 hostile takeover bids between 2000 and 2016 in Continental Europe and the Anglo-Saxon countries. The takeover attempts are provided by the Thomson One Mergers & Acquisitions database. A takeover attempt is classified as ‘hostile’ if the board officially rejects the offer but the acquirer persists with the takeover. The database provides data on 325 interfirm hostile bids. Excluded is one bid that is intended but a formal offer never followed, and ten transactions that are still pending are also omitted. Thus, the final sample constitutes of 314 hostile takeover bids. The amount of hostile takeover bids varies widely over the years; however, a downward trend is noticed. The median amount of transactions is 17, whereas the lowest and highest are 3 and 51 respectively. The greater part, 187 transactions, are withdrawn (59.6%) whereas 127 transactions are completed (40.4%). 116 takeover attempts (36,9%) are from a foreign acquirer and 198 attempts are national. The majority (73.9%) of bids is targeted at Anglo-Saxon countries (231) whereas the remaining (26.1%) is aimed at taking over Continental European firms (82). As expected, the American market is the main hostile takeover market (Table 1). Additionally, table 2 shows the sample distribution by industry.
13 the Thomson One database and partly collected from Datastream. Still, not all values are encountered.
Table 1: Distribution of sample by target country and board neutrality
Country Total No BNR BNR
Number % Number % Number %
Australia* 32 10,2 0 0,0 32 22,2 Canada 43 13,7 43 25,3 0 0,0 Ireland 3 1,0 0 0,0 3 2,1 New Zealand 4 1,3 4 2,4 0 0,0 United Kingdom 60 19,2 0 0,0 60 41,7 United States 90 28,8 90 53,6 0 0,0 Anglo Saxon subtotal 232 73,9 137 80,6 95 66,0 Belgium 1 0,3 1 0,6 0 0,0 Czech Republic 2 0,6 0 0,0 2 1,4 Denmark 2 0,6 2 1,2 0 0,0 Finland 2 0,6 0 0,0 2 1,4 France 9 2,9 0 0,0 9 6,3 Germany 9 2,9 9 5,4 0 0,0 Hungary 1 0,3 1 0,6 0 0,0 Italy 3 1,0 0 0,0 3 2,1 Luxembourg 1 0,3 1 0,6 0 0,0 Netherlands 8 2,6 8 4,8 0 0,0 Norway 8 2,6 8 4,8 0 0,0 Poland 3 1,0 3 1,8 0 0,0 Portugal 2 0,6 0 0,0 2 1,4 Spain 14 4,5 0 0,0 14 9,7 Sweden 10 3,2 0 0,0 10 6,9 Switzerland 7 2,2 0 0,0 7 4,9 Cont. EU subtotal 82 26,1 33 19,4 49 34,0 Total 314 100,0% 170 100,0% 144 100,0%
This table presents the sample by target country and whether the country follows the board neutrality rule. *Australia does not have BNR as the other countries have. However, it allows defences to a certain degree, poison pills can be used as a defence mechanism. It is therefore considered to be BNR.The existence of BNR is important in the situation of hostile bids since it determines whether target managers can use ATPs without shareholder approval. In this regard, BNR is a crucial
component of takeover regulation. The European takeover directive allows member states to withdraw from the BNR. In addition, a reciprocity decree allows targets to differ from mandatory BNR if
differences between the target’s and bidder’s takeover regulation are present.
In order to analyse the impact of indirect government interventions different indexes of takeover regulation are used as measures of the jurisdiction. The revised number of anti-director rights index constructed by Djankov et al. (2008) focusses on the protection of minority shareholders against self-dealing transactions that are
14 managers and directors. Board neutrality and the legal environment of a country are correlated, isolating the causal effects of the BNR on hostile takeover bids is hard. This causation may imply that the results are driven by general level of shareholder protection or the existence of a BNR. The respective indexes account for the different level of shareholder protection.
Table 2: Distribution of sample by industry.
Fama French 17 industries Total
n %
Food 13 4,14%
Mining and Minerals 36 11,46%
Oil and Petroleum products 19 6,05%
Textiles, Apparel & Footware 3 0,96%
Consumer durables 6 1,91%
Chemicals 4 1,27%
Drugs, Soap, Perfumes & Tobacco 17 5,41% Construction and Construction Materials 16 5,10%
Steel Works 5 1,59%
Fabricated products 11 3,50%
Machinery and Business Equipment 9 2,87%
Automobiles 2 0,64%
Transportation 13 4,14%
Utilities 13 4,14%
Retail Stores 20 6,37%
Banks, Insurance Companies, and Other Financials 44 14,01%
Other 83 26,43%
Total 314 100,00%
This table shows the distribution per industry based on Fama and French’s 17-industry allocation.
Shareholder protection is worse in civil law countries and the civil law proxy proves a useful measure of shareholder protection. Continental European countries are known for high concentration of ownership, unusually high according to La Porta et al.
(1998). They find results that suggest that concentration of ownership is an adaptation to the poor legal protection in these countries. In order to protect
15 Following Rossi and Volpin (2004), controlling for macroeconomic and regulatory differences in jurisdictions is done by using the GDP growth rate, the unemployment rate and GNP per capita2. GDP growth is used as a measure of the change in macroeconomic conditions, unemployment rate as performance indicator and GNI per capita as proxy for a country’s wealth.
In the situation of a hostile takeover attempt the government of the target firm has three options. Firstly, the government can support the bid, secondly, the government can oppose the bid or finally, it can do nothing. Since the focus of this study is on negative government intervention, only negative reactions are monitored. Data on negative government intervention is hand collected from LexisNexis database, which is based on a list of keywords such as: government, hostile, takeover, intervention, anti-takeover provision (ATP), minister, national interest, icon, strategic, public interest, national symbol, patriotic, politics, protectionism and economic nationalism (Dinc and Erel, 2013). Government intervention is analysed for each transaction individually. Based on these keywords negative government interventions can be identified. Intervention is considered as such when government officials react, by means of quotes or public statements, against the transaction. Statements regarding anti-competition issues are neglected since it is outside the scope of this study. The takeover attempts during which the respective government intervened are presented in table 3.
Table 3: Negative direct government interventions. Year Target company Target
country
Bidder company Bidder
country
2003 Iberdrola SA Spain Gas Natural SA Spain
2006 Endesa SA Spain E.ON AG Germany
2006 Scania Ab Sweden MAN SE Germany
2007 MOL Groip Hungary OMV AG Austria
2008 Aer Lingus Group PLC Ireland Coinside Ltd. Ireland 2010 Potash Corp. Of
Saskatchewan
Canada BHP Billiton PLC UK
2012 Azoty Tarnow SA Poland Norica Holding S.à. R.l. Luxembourg
2012 Rona Inc. Canada Lowe's Comp. Inc. USA
2014 Metso Oyj Finland Weir Group PLC UK
2015 Norfolk Southern United States
Canada Pacific Railway Canada This table shows the takeover attempts on which the government of the target company opposed to a hostile bid.
2
16 Management is most likely to use a Poison Pill as ATP to prevent a hostile takeover attempt. In line with Pound (1987) and DeAngelo and Rice (1983) deployment of ATPs is seen as defending against an unwanted takeover attempt. A total of 91 ATPs has been used during the period, to prevent 71 takeover attempts. This implies that in 15 out of 314 hostile takeover attempts several ATPs were used. A list of ATP definitions can be found in the Appendix (Table 16: ATP definitions, usage and whether it is preventive or reactive). The top three most used ATPs are Poison Pill (37), White Knight Defence (24) and Proxy Fight (15).
ATPs can be categorized as reactive or preventive, ATPs can be implemented as a reaction to a specific takeover attempt or can be installed beforehand as a
precautionary measure to prevent takeover attempts in general. Both preventive (pre-bid) and reactive (post-(pre-bid) ATPs are considered. See the Appendix for the
classification of the described ATPs. Share repurchases can be labelled as pre- and post-bid defence. The pre-bid situation would occur when managers buy back shares in the market because they believe the intrinsic value is lower compared to the
current market value. Whereas post-bid would be the case when managers buy shares from the market to decrease its liquidity on the market, making it more
expensive for the acquiring party to buy outstanding shares. Share repurchase is the only pre-bid ATP that is used, once, the remainder, 88 ATPs where used as a post-bid defence mechanism. Managers of target firms deployed ATPs in 29,2% of the hostile takeover attempts. Table 4 and 5 show that ATPs are not used in all countries and present the application of ATPs per industry respectively.
17
Table 4: Distribution of sample by target country and ATP deployment.
Country ATP No ATP
n % n % Australia 4 5,33% 28 11,72% Canada 22 29,33% 21 8,79% Ireland 0 0,00% 3 1,26% New Zealand 0 0,00% 4 1,67% UK 6 8,00% 54 22,59% US 31 41,33% 59 24,69%
Anglo Saxon subtotal 63 84,00% 169 70,71%
Belgium 1 1,33% 0 0,00% Czech Republic 0 0,00% 2 0,84% Denmark 1 1,33% 1 0,42% Finland 0 0,00% 2 0,84% France 1 1,33% 8 3,35% Germany 1 1,33% 8 3,35% Hungary 0 0,00% 1 0,42% Italy 0 0,00% 3 1,26% Luxembourg 0 0,00% 1 0,42% Netherlands 3 4,00% 5 2,09% Norway 2 2,67% 6 2,51% Poland 0 0,00% 3 1,26% Portugal 0 0,00% 2 0,84% Spain 0 0,00% 14 5,86% Sweden 1 1,33% 9 3,77% Switzerland 2 2,67% 5 2,09% Cont. EU subtotal 12 16,00% 70 29,29% Total 75 100,00% 239 100,00%
This table presents the ATP deployment per target country.
Table 5: Distribution of sample by Fama and French 17 Industries and ATP deployment.
Fama French 17 industries ATP No ATP
n % n %
Food 2 2,67% 11 4,60%
Mining and Minerals 12 16,00% 24 10,04%
Oil and Petroleum products 8 10,67% 11 4,60%
Textiles, Apparel & Footware 0 0,00% 3 1,26%
Consumer durables 1 1,33% 5 2,09%
Chemicals 2 2,67% 2 0,84%
Drugs, Soap, Perfumes & Tobacco 6 8,00% 11 4,60%
Construction and Construction Materials 7 9,33% 9 3,77%
Steel Works 3 4,00% 2 0,84%
Fabricated products 5 6,67% 6 2,51%
Machinery and Business Equipment 1 1,33% 8 3,35%
Automobiles 0 0,00% 2 0,84%
Transportation 3 4,00% 10 4,18%
Utilities 2 2,67% 11 4,60%
Retail Stores 5 6,67% 15 6,28%
Banks, Insurance Companies, and Other Financials 4 5,33% 40 16,74%
Other 14 18,67% 69 28,87%
Total 75 100,00% 239 100,00%
18
4.2 Control Variables
Several control variables are used, they can be categorized as; country, deal and target level variables. In order to control for macroeconomic differences between target nations controlled is for the GDP growth rate, GNI per capita and the
unemployment rate. In line with Rossi and Volpin (2004) these numbers are collected from the World Bank. By using the logarithm of GNI per capita and unemployment the possibility of biased results due to outliers is minimized. In accordance with Rowoldt and Starke (2016) controlled is for deal level variables by considering ATPs, multiple ATPs, the number of ATPs, multiple bidders, number of bidders, cross border, increase in bid, success, transaction value (in mil. $), (in)direct government intervention, takeover premium and the value increase at (one month after)
completion/withdrawal date are considered. In order to minimize the impact of outliers, the logarithm of transaction value is taken.
Lastly, target characteristics are considered in order to control on target level. Safieddine and Titman (1999) conclude that target firms increase their leverage position succeeding a takeover attempt that was unsuccessful. Committing to make improvements that might have been executed by the acquiring party. Leverage is measured as the target company’s last 12 months average of liabilities divided by the target company’s last 12 months average total assets. The amount of leverage a firm has is considered as characteristic because it gives firms a way to become less (more) attractive. Pinkowitz (2000) investigates the relationship between cash levels and the probability of becoming a target. He finds that the probability of being
19 correlation matrix can be found in the appendix (table 23: Pairwise correlation
20
Table 6: Variable definitions and related data sources
Variable Definition Source
Country level variables
Board Neutrality Rule Equals one if, according to regulatory framework of the target’s jurisdiction, the board of a hostile target company may not take any frustrating action against the offer and zero otherwise. Commission of the European communities (2007) Revised Number of anti-director rights
Measures the level of investor protection in a country, computed as an index consisting of six protection rules. When each of the following rules apply, a one is added to the total. (1) Shareholders are allowed to mail their proxy vote to the firm, (2) shareholders are not required to deposit their shares before the shareholders
meeting, (3) proportional representation of minorities in the board or cumulative voting is allowed, (4) a mechanism to oppress minorities is in place, (5) the minimum percentage of shares a shareholder needs to be able to call for an extraordinary shareholders meeting is set at no more than 10%, (6) only a shareholders vote can waive the pre-emptive rights shareholders have.
Djankov et al. (2008)
Shareholder protection
Measures the target country’s level of investor protection, calculated as the product of the revised number of anti-director rights and the rule of law index.
Rossi and Volpin (2004), World Justice Project
Civil law Equals one if the legal setting of the country is civil law (Continental Europe) and zero if common law (Anglo-Saxon).
La Porta et al. (1998)
Log GNI per capita The logarithm of Gross National Income per capita based on Purchasing Power Parity.
Worldbank
GDP growth The annual growth rate of a country’s GDP, at market prices, based on constant local currency, in percentage.
Worldbank
Log Unemployment The logarithm of the ratio of people that are unemployed, but available for work as percentage of the total labour force.
Worldbank
Target level variables
Leverage Measured as the average of the total liabilities on the announcement date and one year prior to it, scaled by the total amount of assets (Heron and Lie, 2006).
Datastream
Cash Measured as the average of cash and equivalents on the announcement date and one year prior to it, scaled by the total amount of assets (Comment and Schwert, 1995; Pinkowitz, 2000).
Datastream
EBIT Measured as the average of the total liabilities on the announcement date and one year prior to it, scaled by the total amount of assets.
Datastream
Deal level variables
ATPs Takes the value of one if management defended against a hostile takeover attempt (DeAngelo and Rice, 1983; Pound, 1987).
ThomsonOne
Multiple ATPs Takes the value of one if management defended against a hostile takeover attempt using multiple ATPs.
ThomsonOne
Number of ATPs Presents the number ATPs that have been used by the target company. ThomsonOne
Multiple Bidders Takes the value of one if there are multiple companies bidding for the same target company (Comment and Schwert, 1995; Heron and Lie, 2006).
ThomsonOne
Cross Border Equals one if the bidder is not from the same country as the target firm (Dinc and Erel, 2013; Rossi and Volpin, 2004).
ThomsonOne
Bid Increase Takes the value of one if the acquirer increased its offer after the initial offer (Heron and Lie, 2006).
ThomsonOne
Success Equals one if the acquiror gained control over the target; the bidder owns more than 50% of the shares after the transaction (Holmén et al., 2012; Pound, 1987; Schwert, 2000).
ThomsonOne
Log Transaction value Used to control for size effect of the transaction, measured as logarithm (million USD) ThomsonOne
Premium (1 day) Measures the final premium paid by the bidder in relation to the share price one day prior to the announcement (Subramanian, 2013).
ThomsonOne/Datastream
Premium (1 week) Measures the final premium paid by the bidder in relation to the share price one week prior to the announcement (Subramanian, 2013).
ThomsonOne/Datastream
Value Increase Measures the difference in share price one month prior to the attempt and at (one month after) completion/withdrawal date.
ThomsonOne/Datastream
Government intervention
Negative interventions are found in quotes or statements that indicate negative government attitude and direct actions against takeover attempts. Interventions with respect to anti-competition issues are neglected (Dinc and Erel, 2013).
LexisNexis
21 Table 7: Descriptive statistics
Variable n Mean Std. Dev. Min. Max.
BNR 314 0.459 0.499 0.000 1.000
Revised number of anti-director
rights 314 3.774 0.844 2.000 5.000
Shareholder protection 303 2.976 0.727 1.136 4.044
Civil law 314 0.258 0.438 0.000 1.000
Log GNI per capita 309 10.618 0.184 9.989 11.080
GDP growth (in %) 314 0.023 0.021 -.056 0.263
Log Unemployment (in %) 314 -2.824 0.327 -3.693 -1.512
Deal level variables
ATP 314 0.239 0.427 0.000 1.000 Multiple ATPs 314 0.051 0.220 0.000 1.000 Number of ATPs 314 0.303 0.610 0.000 4.000 Multiple Bidders 314 0.344 0.476 0.000 1.000 Number of Bidders 314 1.398 0.596 1.000 4.000 Cross Border 314 0.369 0.483 0.000 1.000 Increase in bid 314 0.519 0.500 0.000 1.000 Success 314 0.331 0.471 0.000 1.000
Log Transaction Value (in mil. $) 290 6.141 2.450 -2.104 11.881 Government intervention 314 0.032 0.176 0.000 1.000 Government intervention x ATP 314 0.003 0.056 0.000 1.000 Takeover premium (4weeks) 92 0.456 0.341 -0.217 1.817 Takeover premium (1week) 269 0.422 0.463 -0.954 5.280 Takeover premium (1day) 267 0.408 0.539 -0.954 6.572 Value Increase at Completion/
Withdrawal date 272 1.604 12.365 -1.000 161.500
Value Increase one month after
Completion/ Withdrawal date 272 1.630 12.730 -1.000 150.389 Value Increase on Completion
date 105 1.811 11.799 -1.000 120.583
Value Increase on Withdrawal
date 167 1.473 12.741 -1.000 161.500
Value Increase one month after
Completion date 105 2.047 14.113 -1.000 144.449
Value Increase one month after
Withdrawal date 167 1.360 11.815 -1.000 150.389
Duration combined 308 127.604 125.900 0.000 1102.000
Duration combined (>100days) 308 0.435 0.497 0.000 1.000
Target level variables
Liabilities/Assets 173 0.533 0.245 0.007 1.446
Log Cash/Assets 173 -2.854 1.630 -9.323 -.033
EBIT/Assets 170 0.017 0.207 -1.928 0.405
22
5. Results
This section discusses the results from the conducted tests. The drafted hypotheses are evaluated. Year and industry dummies are included in all models. Additionally, controlled is for different measures of shareholder protection as well as country, deal and target level variables in all models. This is done in order to limit the concerns regarding correlation between takeover regulation and country level effects.
Furthermore, in order to avoid the dummy variable trap, one observation for the year (2016) and one observation for the industry (Automobiles) are excluded from all models.
5.1 ATPs and bid outcome
The relation between ATP deployment and corporate governance is analysed by considering the impact of ATPs on bid outcome. Preceding literature did not show conclusive evidence. On one side, Comment and Schwert (1995), Heron and Lie (2006) and Varaiya (1987) show that ATP deployment increases the takeover premium, and therefore is beneficial to target’s shareholders. On the other side, successful completion of hostile takeover attempts may be prevented by ATPs, without having a positive impact on the takeover premium. In that sense, ATPs can be considered as a tool for managers of targets to exploit their privileges (Bebchuk et al., 2007; Pound, 1987; Subramanian, 2003).
Following the stockholder interest hypothesis ATPs should benefit shareholders via increases in bids and the final takeover premium offered. Firstly, the relationship between ATPs and Increase in bidding and the takeover premium are analysed. Following H1 expected is that ATPs should increase the possibility of upward bid adjustments and the takeover premium offered. Since shareholders’ bargaining power increases by the deployment of ATPs it should translate into higher bids and higher takeover premiums. Looking at the takeover premium has the advantage that adjustments due to negotiation (because of ATP deployment) are included. The takeover premium is calculated based on difference between the share price one week prior to the announcement and the final offer of the acquiring firm. The stockholder interest hypothesis - which states that the dominant effect of ATPs is beneficial to stockholders by increasing the expected gains from transferring
23 The relationship between ATPs and increase in bids is positive and significant,
however, there is no evidence of a significant relation relationship between ATPs and takeover premium. The results indicate that, ATP used, the use of multiple ATPs and the number of ATPs used have a positive and significant effect on the dependent variable bid increase. The relationship is significant and positive at the 5%, 5% and 1% level respectively (β=0.745; β=1.325; β=0.673). Implying that ATPs do lead to bid increases and that the shareholders use ATPs as a means of bargaining power. In contrast to the stockholder interest hypothesis there is no evidence of a positive and significant relationship between ATPs and takeover premium. Similar regressions with the final takeover premium based on the share price one day before
24
Table 8: Relation between ATPs and increase in bid.
Dependent variable: Increase in bid
No government intervention Indirect government intervention Direct government intervention
(1) (2) (3) (4) (5) (6) (7) (8) (9) ATP used 0.745** 0.651* 0.854** (0.357) (0.380) (0.363) Multiple ATPs 1.325** 1.118* 1.330** (0.533) (0.603) (0.545) Number of ATPs 0.673*** 0.569** 0.729*** (0.212) (0.242) (0.214) BNR 0.764 0.783 0.794 (0.589) (0.580) (0.585) Revised number of anti-director rights -1.767 -2.269* -2.022* (1.240) (1.224) (1.216) Shareholder protection 0.429 (1.257) 1.026 (1.235) 0.724 (1.226) Civil law -3.504*** -3.311*** -3.401*** (0.734) (0.713) (0.747) Govern. Interv. -0.987 -1.116 -1.026 (0.685) (0.711) (0.702) Govern. Interv. x ATP 0.000 0.000 0.000 (.) (.) (.)
Log GNI per
capita -0.697 (0.849) -0.678 (0.862) -0.994 (0.876) (2.103) -6.510*** -6.327*** (2.162) -6.641*** (2.113) (0.870) -0.864 (0.875) -0.768 -1.157 (0.898) GDP growth 0.830 -2.198 -0.308 39.774** 33.442* 37.134* 2.473 -1.696 0.638 (8.419) (6.366) (6.923) (19.350) (19.122) (19.294) (11.007) (7.450) (8.520) Log Unemploy. 0.250 (0.482) 0.355 (0.480) 0.288 (0.481) (1.049) 2.696** 2.835*** (1.021) 2.786*** (1.047) (0.487) 0.256 (0.484) 0.398 0.306 (0.485) Liabilities/ Assets -0.255 (0.601) -0.163 (0.608) -0.133 (0.607) (0.720) -0.160 -0.141 (0.724) -0.074 (0.722) (0.608) -0.178 (0.614) -0.109 -0.050 (0.616) Log Cash/Assets 0.072 (0.088) 0.077 (0.088) 0.064 (0.090) (0.101) 0.026 0.039 (0.097) 0.022 (0.102) (0.091) 0.073 (0.089) 0.077 0.067 (0.093) EBIT/Assets -0.298 -0.289 -0.332 -0.894 -0.726 -0.845 -0.345 -0.365 -0.394 (0.705) (0.712) (0.716) (0.683) (0.686) (0.688) (0.746) (0.749) (0.759) Log Trans. Value 0.121* (0.063) 0.133** (0.061) 0.120* (0.063) (0.070) 0.115 0.128* (0.067) 0.115* (0.069) (0.063) 0.158** (0.062) 0.168*** 0.161** (0.063) Multiple bidders 0.192 (0.305) 0.311 (0.294) 0.208 (0.302) (0.451) 1.073** 1.186*** (0.427) 1.093** (0.440) (0.327) 0.033 (0.307) 0.172 0.052 (0.323) Cross Border 0.201 0.214 0.234 0.607 0.479 0.568 0.322 0.328 0.363 (0.274) (0.274) (0.275) (0.376) (0.359) (0.369) (0.280) (0.281) (0.283) Constant 16.435* 16.491* 19.546** 90.978*** 89.848*** 92.679*** 18.600** 18.123** 21.850** (8.631) (8.806) (8.889) (24.203) (24.972) (24.467) (8.898) (8.957) (9.173) N 147 147 147 140 140 140 146 146 146
Year dummy Yes Yes Yes Yes Yes Yes Yes Yes Yes
Industry
dummy Yes Yes Yes Yes Yes Yes Yes Yes Yes
pseudo R-sq 0.194 0.200 0.213 0.349 0.349 0.358 0.221 0.221 0.240
25
Table 9: Relation between ATPs and Takeover Premium.
Dependent variable: Takeover Premium (1 week)
No government intervention Indirect government intervention Direct government intervention
(1) (2) (3) (4) (5) (6) (7) (8) (9) ATP used -0.219 -0.256 -0.214 (0.179) (0.181) (0.184) Multiple ATPs -0.145 -0.202 -0.146 (0.213) (0.228) (0.216) Number of ATPs -0.132 -0.167 -0.129 (0.126) (0.131) (0.128) BNR -0.008 -0.011 -0.024 (0.211) (0.209) (0.201) Revised number of anti-director rights 0.397 0.500 0.473 (0.288) (0.323) (0.309) Shareholder protection -0.615 -0.728 -0.703 (0.436) (0.508) (0.477) Civil law -0.076 -0.060 -0.088 (0.128) (0.129) (0.129) Govern. Interv. -0.017 0.018 0.007 (0.228) (0.215) (0.220) Govern. Interv. x ATP -0.200 -0.375 -0.271 (0.283) (0.278) (0.275)
Log GNI per
capita -0.236 (0.520) -0.331 (0.561) -0.235 (0.493) (0.621) -0.449 -0.523 (0.653) -0.445 (0.599) (0.556) -0.250 (0.595) -0.334 -0.242 (0.527) GDP growth -1.972 -1.420 -1.780 -10.027 -9.148 -9.187 -1.960 -1.390 -1.752 (1.993) (1.832) (1.913) (7.523) (7.201) (7.375) (2.049) (1.855) (1.956) Log Unemploy. -0.099 (0.257) -0.126 (0.276) -0.115 (0.264) (0.480) -0.480 -0.547 (0.526) -0.520 (0.496) (0.255) -0.102 (0.276) -0.133 -0.120 (0.263) Liabilities/ Assets -0.060 (0.200) -0.056 (0.210) -0.079 (0.214) (0.227) -0.002 -0.003 (0.238) -0.013 (0.234) (0.204) -0.060 (0.213) -0.055 -0.078 (0.217) Log Cash/Assets 0.052 (0.035) 0.047 (0.035) 0.052 (0.037) (0.029) 0.051* 0.045 (0.029) 0.053* (0.031) (0.034) 0.051 (0.034) 0.047 0.051 (0.036) EBIT/Assets -0.042 -0.051 -0.036 -0.017 -0.035 -0.013 -0.044 -0.048 -0.035 (0.238) (0.242) (0.234) (0.209) (0.218) (0.209) (0.247) (0.251) (0.242) Log Trans. Value 0.037 (0.026) 0.031 (0.024) 0.036 (0.026) (0.024) 0.018 0.012 (0.023) 0.016 (0.024) (0.027) 0.039 (0.026) 0.034 0.037 (0.027) Multiple bidders 0.102 (0.100) 0.074 (0.106) 0.092 (0.101) (0.100) 0.124 0.080 (0.104) 0.106 (0.100) (0.107) 0.096 (0.115) 0.068 0.087 (0.109) Cross Border 0.067 0.078 0.065 0.157 0.170 0.162 0.071 0.083 0.069 (0.095) (0.097) (0.091) (0.114) (0.122) (0.116) (0.097) (0.100) (0.094) Constant 2.576 3.505 2.546 4.556 5.081 4.379 2.707 3.489 2.596 (4.913) (5.268) (4.631) (6.440) (6.631) (6.165) (5.330) (5.655) (5.018) N 160 160 160 154 154 154 160 160 160
Year dummy Yes Yes Yes Yes Yes Yes Yes Yes Yes
Industry
dummy Yes Yes Yes Yes Yes Yes Yes Yes Yes
pseudo R-sq 0.231 0.217 0.231 0.300 0.284 0.303 0.232 0.220 0.232
26 Next, the impact ATPs have on the probability of success is analysed. In line with the managerial entrenchment hypothesis, the impact ATPs have on the probability of success is evaluated. This view argues that ATPs provide protection of incumbent managers at the expense of shareholders and actually reduce shareholder wealth (DeAngelo and Rice, 1983). ATPs may benefit targets’ management via preventing the completion of hostile takeover bids. In this situation, ATP deployment would lead to an increase in managerial power and thus harm corporate governance. Combining various ATPs can serve as such a strong defence mechanism that makes
overcoming them too difficult. Table 10 reports the results with Success as dependent variable and ATPs as independent variables in order to analyse the impact ATPs have on Success.
The application of ATPs does not block hostile takeover attempts; it rather increases the offer for the firm. The results do not indicate an expected significant negative relationship between success and ATPs. In none of the different models a significant relationship is reported. Hence, there are no signs that ATPs decrease the likelihood of a successful hostile takeover attempt. This finding is in opposition to the previous results on the managerial entrenchment hypothesis (Bebchuk et al., 2007; Pound, 1987; and Subramanian, 2003). It is however in line with the previously described results following the stockholder interest hypothesis. The implementation of multiple ATPs does not have a significant impact on takeover success. Having multiple ATPs does not have an effect on the likelihood of success in none of the models.
Therefore, the hypothesis that multiple ATPs lead to a decrease in the probability of success is not confirmed. Adding an extra ATP hence may give target’s managers the idea that they are better protected against hostile takeovers. However, it does not imply the company is actually less likely to be taken over.
Summarizing, ATP application by targets’ management seems to enhance negotiation power of the target’s shareholders. It does not seem to enable
27
Table 10: Relation between ATPs and Success.
Dependent variable: Success
No government intervention Indirect government intervention Direct government intervention
(1) (2) (3) (4) (5) (6) (7) (8) (9) ATP used 0.245 0.189 0.274 (0.356) (0.367) (0.358) Multiple ATPs 0.126 0.173 0.100 (0.539) (0.541) (0.543) Number of ATPs 0.064 0.061 0.065 (0.208) (0.210) (0.209) BNR -0.507 -0.488 -0.497 (0.474) (0.471) (0.471) Revised number of anti-director rights -1.417 -1.536 -1.501 (0.951) (0.949) (0.944) Shareholder protection 1.597 1.732 1.692 (1.072) (1.082) (1.071) Civil law 0.283 0.290 0.284 (0.421) (0.420) (0.421) Govern. Interv. 0.666 0.609 0.618 (0.774) (0.771) (0.770) Govern. Interv. x ATP 0.000 0.000 0.000 (.) (.) (.)
Log GNI per
capita -0.774 (0.815) -0.689 (0.774) -0.708 (0.807) (0.995) -2.193** -2.170** (0.979) -2.173** (0.997) (0.826) -0.650 (0.791) -0.556 -0.582 (0.822) GDP growth 23.163 21.379 21.932 36.255 33.710 34.810 24.682 22.834 23.279 (18.453) (18.414) (18.620) (26.999) (26.278) (26.711) (18.039) (18.080) (18.265) Log Unemploy. -0.040 (0.462) -0.023 (0.457) -0.026 (0.458) (0.763) 0.151 0.176 (0.764) 0.170 (0.763) (0.468) -0.066 (0.462) -0.044 -0.048 (0.463) Liabilities/ Assets -0.515 (0.663) -0.545 (0.652) -0.534 (0.658) (0.747) -0.695 -0.724 (0.740) -0.709 (0.741) (0.658) -0.489 (0.647) -0.528 -0.514 (0.653) Log Cash/Assets -0.090 (0.106) -0.085 (0.106) -0.087 (0.106) (0.109) -0.030 -0.025 (0.108) -0.026 (0.108) (0.106) -0.087 (0.106) -0.082 -0.084 (0.106) EBIT/Assets -0.154 -0.138 -0.148 -0.201 -0.177 -0.189 -0.116 -0.105 -0.114 (0.632) (0.624) (0.629) (0.657) (0.650) (0.654) (0.641) (0.633) (0.637) Log Trans. Value 0.028 0.034 0.034 0.082 0.086 0.086 0.026 0.033 0.032 (0.069) (0.068) (0.068) (0.072) (0.072) (0.072) (0.069) (0.068) (0.069) Multiple bidders -0.029 0.018 0.003 0.017 0.072 0.049 -0.009 0.041 0.027 (0.317) (0.314) (0.319) (0.336) (0.319) (0.327) (0.318) (0.313) (0.320) Cross Border -0.035 -0.062 -0.053 -0.307 -0.333 -0.323 -0.036 -0.065 -0.056 (0.287) (0.290) (0.289) (0.345) (0.348) (0.346) (0.286) (0.287) (0.287) Constant 1.641 0.856 1.018 17.829 17.799 17.754 -0.207 -1.071 -0.832 (8.508) (8.118) (8.426) (11.437) (11.382) (11.503) (8.696) (8.387) (8.670) N 145 145 145 137 137 137 144 144 144
Year dummy Yes Yes Yes Yes Yes Yes Yes Yes Yes
Industry
dummy Yes Yes Yes Yes Yes Yes Yes Yes Yes
pseudo R-sq 0.170 0.168 0.168 0.239 0.238 0.238 0.170 0.168 0.168
28
5.2 (In)Direct government interventions
In order to analyse whether the inconclusive relation between ATPs and acquisition outcome is caused by (in)direct government interventions, the impact is analysed here. The BNR and different measures of shareholder protection are considered to be indirect interventions. Direct interventions alone are included as well as direct interventions multiplied by ATPs used. This in order to incorporate the effect direct interventions have on firm value and the impact on the relation between ATPs and firm value.
5.2.1 Bid outcome
The impact of (in)direct government interventions on the outcome of the takeover attempt is analysed. Expected is that a negative government interventions would lead to an increase in the value of the firm and decline the probability of success.
Hypothesized is that (in)direct government intervention increase acquisition outcome (H2). Tables 8, 9 and 10 present the results of (in)direct government interventions on Increase in bid, Takeover Premium and Success.
(In)Direct government intervention is positively related to a bid increase, however, there is no evidence of a significant relation relationship between (in)direct
government intervention and takeover premium and success. The results show that, when looking at indirect government interventions, ATPs used, the use of multiple ATPs and the number of ATPs used are positive and significant, at the 10%, 10% and 5% level respectively (β=0.651; β=1.118; β=0.569). Considering direct
government interventions, ATPs used, the use of multiple ATPs and the number of ATPs used all are positive and significant, at the 5%, 5% and 1% level (β=0.854; β=1.330; β=0.729). Furthermore, there is no evidence of a significant relationship between (in)direct government interventions and takeover premium and success. Moreover, noted should be that out of the ten negative government interventions two still succeeded while eight failed. As mentioned before, the generalisability therefore may be rather limited at this point.
5.2.2 Value Increase
29 Hypothesized is that a(n) (in)direct government intervention increases the value of the firm (H3). The value of the firm is calculated using formula 2 and 3. Hence, the value of the firm is measured at two different moments in time, at
completion/withdrawal date or one month after completion/withdrawal date.
Completion and withdrawal are combined in order analyse the increase in value in general. Separate calculations are reported in the appendix (tables 19-22). The dependent variables thus are increase in value at completion/withdrawal date and increase in value one month after completion/withdrawal date. Since we are dealing with a dependent variable that is not binary a probit model does not suit, ordinary OLS is executed here (equation 6). Table 11 and 12 show the results of (in)direct government intervention on Increase in value, at/one month after completion date.
30
Table 11: Relationship between increase in value and government intervention, at completion/withdrawal
date.
Dependent variable: Increase in value at completion/withdrawal date
Indirect government intervention Direct government intervention
(1) (2) (3) (4) (5) (6) ATP used 0.056 -0.306 (4.005) (3.920) Multiple ATPs -1.952 -1.810 (4.708) (4.435) Number of ATPs -0.869 -0.795 (2.269) (2.091) BNR -3.627 -3.761 -3.764 (3.518) (3.610) (3.668) Revised number of anti-director rights 8.155 8.400 8.131 (7.903) (8.739) (8.382) Shareholder protection -5.484 -5.826 -5.528 (7.937) (9.001) (8.643) Civil law 0.248 0.078 0.009 (2.838) (2.875) (2.776) Govern. Interv. -5.108 -5.079 -5.124 (6.219) (5.904) (5.981) Govern. Interv. x ATP 9.119 8.820 9.469 (7.968) (7.597) (7.810)
Log GNI per capita 2.045 2.486 2.701 -2.952 -2.524 -2.255
(10.532) (9.845) (10.447) (9.070) (7.668) (8.638) GDP growth -148.947 -142.147 -144.974 -9.034 -9.277 -10.928 (141.133) (133.491) (137.136) (39.612) (33.445) (36.548) Log Unemployment -5.060 -5.208 -4.984 -2.184 -2.302 -2.171 (4.911) (5.084) (4.881) (3.807) (3.513) (3.590) Liabilities/ Assets -6.162 -6.208 -6.236 -7.506 -7.665 -7.709 (6.284) (6.319) (6.344) (6.579) (6.670) (6.759) Log Cash/Assets 2.001 2.051 2.069 1.687 1.715 1.727 (1.480) (1.514) (1.518) (1.330) (1.352) (1.352) EBIT/Assets 7.542 7.585 7.640 6.983 7.038 7.060 (7.839) (7.953) (7.930) (8.065) (8.099) (8.067)
Log Trans. Value -2.194 -2.157 -2.148 -1.938 -1.912 -1.907
(1.380) (1.365) (1.354) (1.240) (1.219) (1.217) Multiple bidders -0.566 -0.614 -0.453 -0.380 -0.436 -0.307 (1.652) (1.656) (1.637) (1.664) (1.471) (1.581) Cross Border 5.545 5.540 5.492 4.811 4.757 4.708 (5.718) (5.844) (5.753) (4.509) (4.644) (4.504) Constant -22.306 -27.338 -28.774 53.499 48.785 46.383 (113.871) (109.680) (114.594) (92.211) (79.019) (87.773) N 151 151 151 157 157 157
Year dummy Yes Yes Yes Yes Yes Yes
Industry dummy Yes Yes Yes Yes Yes Yes
pseudo R-sq 0.301 0.302 0.302 0.285 0.286 0.286
