Determinants of premium for control of the emerging markets

Размещено на http://www.allbest.ru/

ФЕДЕРАЛЬНОЕ ГОСУДАРСТВЕННОЕ АВТОНОМНОЕ ОБРАЗОВАТЕЛЬНОЕ УЧРЕЖДЕНИЕ

ВЫСШЕГО ОБРАЗОВАНИЯ

"НАЦИОНАЛЬНЫЙ ИССЛЕДОВАТЕЛЬСКИЙ УНИВЕРСИТЕТ

"ВЫСШАЯ ШКОЛА ЭКОНОМИКИ"

Факультет экономических наук

Образовательная программа "Стратегическое управление финансами фирмы"

МАГИСТЕРСКАЯ ДИССЕРТАЦИЯ

Детерминанты премии за контроль на развивающихся рынках

Выполнил: Муравский Виктор Петрович

Научный руководитель:

проф. д.э.н. Ивашковская Ирина Васильевна

Москва

2016



Content

Introduction

1. Theoretical Evidence of Control Premium

2. Sample and Methodology

3. Empirical Results

Concluding Comments

References

Appendix



Introduction

At the current stage of business valuation practices are particularly important issues related to the need to correct recognizing control premiums in the value of holdings shares and stakes.

M&A deals play an important role in the modern financial world. The estimation of fair value becomes important, when the company acquires another company. Lately the offer price was reconsidered repeatedly. At first, it was only an assessment in a classical reason that is an assessment by means of discounting of cash flows of the company. Later this approach improved in respect of adjustment of fundamental value on premiums and discounts. In this research, we will consider one of amendments, called control premium.

There is a set of research papers in this area for developed countries, but on Emerging markets, it is significantly less. In addition, there are unified databases, which accumulate control premium information.

We may note that control premium differ between countries and today is important to real business to understand what fair size of control premium and what key factors which may defined control premium.

For example, the average size of control premium for developed countries is approximately 22,5 % in 2014 ( in 2015 - 17%), however for Emerging markets we may observed 70 % size of control premium (lower bound) Control Premium Study 3RD Quarter 2014, Available at: http://www.bvmarketdata.com/pdf/CPS3q14.pdf.

The size of control premium during the crisis differs with the period of economic stability. For example, average size of control premium during the crisis going to 43,4%, that significantly higher than the size control premium in 2014-2015.

Unlike the developed countries, where the issues of control premium is in detail studied (exist the practice of application of these approaches for deals), in a case with Emerging countries it is difficult to understand how much control premium worth in the company. Overall, most adjustments appear to be made to the market value of equity. However, the highest adjustment made to the multiple in a control premium scenario.

Figure 1. An application of control premium on Developed market Australian Valuation Practice 2015, Available at: https://www.kpmg.com/AU/en/IssuesAndInsights/ArticlesPublications/valuation-practices-survey/Documents/valuation-practices-survey-2015.pdf

As you would expect, the reverse principle prevails with control premium (see figure above): the premium increased as the size of stake increases. Participants are far cleaner about applying a premium when a controlling stake is involved, with 85 % of those using the market approach opting to do so.

For emerging markets, we have not relevant statistic information, but we assume that using control premium approaches are widespread. In order to prove this statement we showed growth of M&A deal by region during the 2014-2015 period.



Figure 2. M&A activity dynamic Thomson Reuters and PWC analysis

As we can see from the figure above, the growth rate of M&A deals is more in Emerging markets, than in developed that shows prospect of growth М&A of the market. It means that the importance of control premium in emerging markets is higher than on developed markets. Therefore, this subject of research deserves attention for following investigation.

In theoretical part of work considered definition of control premium, types of control and the determinant, which may influence on size of control premium.

Theoretically, control premium should be estimated as the sum of each advantage produced by the controlling position. In Finance literature there are two main approaches regarding investors' reasons for taking control over a company: (i) the potential new controlling shareholder could change the general company's policy in order to achieve a better performance and, based on that, he / she is expecting in the future an increase in the shares' prices (Shleifer and Vishny, (1986)). Not only controller shareholders will benefit of this price increase, but, also the smaller (minority) shareholders; (ii) the potential new controlling shareholders expect to take control over the corporation in order to obtain higher private benefits. In this case, the increase in wealth for controller shareholder is equivalent to a decrease of smaller shareholders' wealth (expropriation of minority shareholders) (Fama and Jensen, (1983)).

Determinants of control premium allow determining the fair size of premium. Therefore, business is interested in the solution of this problem.

The aim of the present research is to reveal the key determinants, which have an effect on control premium on emerging capital markets.

In our investigation, we consider following objectives:

i. To find determinants of strategic control premium

ii. To compare results from strategic and financial controls premium

iii. To compare the determinants of control premium in China and India

iv. To test control premium for crisis period

v. To develop hypotheses for empirical research

vi. To determine the data and methodology for empirical research

vii. To get empirical results

Control premium is one of the key factors playing fundamental role in valuing a firm. It arises during acquisition, share purchase of publicly traded companies etc. In this research, we have tested 13 OLS-models for Emerging markets such as Brazil, India, China and Republic of South Africa. The first model checks control premium for sample in general, the others models use dummy variables for fixing of separate features, such as:

(i) Accounting of crisis period;

(ii) Type of the investor;

(iii) Residency of Target Company;

(iv) Cross dummies.

We managed to mark out distinctions. There are results, which contradict with standard theory in this area. It only says that there is a need to investigate this direction to receive other results confirming or disproving previous. Only thus, there is an opportunity to receive steady results, so necessary for the real companies.



1. Theoretical Evidence of Control Premium

A. Control Premium: a Definition for Listed Firms

Estimation of control premium is one of the main problems in the financial literature. This reason related with the fact that the control premium is widely used in corporate finance, business valuation, corporate governance and valuation of controlling shareholding. In addition, different approaches to the estimation of control premium include synergistic effects, agency costs, and share of market power, the degree of liquidity and other factors, which may help us to determine the determinants of control premium. From this reason of view, we are interested to know the determinants of control premium for developed and emerging markets, as well as for different sectors and periods.

There are many various definitions of control premium in financial literature, main of them we show in the Table I.

Table I

The Definitions of Control Premium in Financial Literature

Number	Definition	Reference
1	The monetary cost that people are willing to incur in order to retain control over their own payo?s.	The Control Premium: A Preference for Payo? Autonomy (Owens, Grossman, Fackler (2012))
2	Sum of each advantage produced by the controlling position.	Estimation of control premium: The case of Romanian listed companies (Dragotг et al (2007))
3	An amount paid to gain enough ownership interest to control a corporation or other entity. This would typically be an amount in excess of the simple fair market value of the shares sought to be purchased;	The Complete Real Estate Encyclopedia (Denise L. Evans, JD & O. William Evans, JD. Copyright © 2007 by The McGraw-Hill Companies, Inc.
4	An average difference between the price of the target firm's stock a specified number of days before the first announcement of a control transaction (the unaffected stock price) and the price at which a deal is consummated.	Mark-up pricing in mergers and acquisitions (Schwert, William, G., (1996))

Dyck and Zingales (2004) explained the existence of control premium with the help of two reasons: the increase in the value of the company because of new management and private benefits after taking control. The second reason we may see on Fig. 3. In particular, the final price formed from two components: market capitalization of the company, and premiums. The premium in this deal is the cost of benefits such as private benefits of control, synergy and other improvements which buyer may receive by paying some price premium for controlling stake. Finally, we accept the strong definition from International Valuation Standards. According to them, control premium is the additional value inherent in the control interest as contrasted to a minority interest, which reflects its power of control. Moreover, control is the power to direct the management and policies of a business International Valuation Standards, Available at: http://www.romacor.ro/legislatie/20-gn6.pdf.

Figure 3. Control premium: definition for listed companies.Source: Antonio Salvi “Control Premium in tender offers. Evidence from Italy and France”

The investor has additional opportunities by purchasing a controlling stake in the company, through the mechanism of voting to make decisions that maximize their well-being. Moreover, the size of control premium depends on potential demand of such benefits, because otherwise the investor is not profitable to buy a controlling stake.

Moreover, a synergistic effect has manifested in revenue growth, cost reduction, tax cuts and reducing the cost of capital (Ross et al. (2005)). Such an approach is the subject of a huge amount of research in business valuation and corporate Finance. In addition, major shareholders, controlling managers, can benefit minority shareholders (Shleifer and Vishny (1986)).

The real impact of Discount for Lack of Control (Minority interest discount) and Discount for Lack of Marketability for a business valuation represented with using diagrams of different levels of business value. These diagrams have undergone in the last 10-15 years, some changes associated with the evolution of ideas about the allocation of these levels.

Let us define some types of value, which we need little beat later. According to International Standards, fair value - the estimated price for the transfer of an asset or liability between identified knowledgeable and willing parties that reflects the respective interests of those parties. Additionally, investment value - the value of an asset to the owner or a prospective owner for individual investment or operational objectives. Moreover, synergistic value - an additional element of value created by the combination of two or more assets or interests where the combined value. Australian Valuation Practice 2015, Available at: https://www.kpmg.com/AU/en/IssuesAndInsights/ArticlesPublications/valuation-practices-survey/Documents/valuation-practices-survey-2015.pdf

Figure 4. Levels of value. SOURCE: Z. Christopher Mercer, “Understanding and Quantifying Control Premiums: The Value of Control vs. Synergies of Strategic Advantages, Part II,” The Journal of Business Valuation (Proceedings of the Fourth Joint Business Valuation Conference of The Canadian Institute of Chartered Business Valuators and the American Society of Appraisers, September 24 and 25, 1998). Toronto: The Canadian Institute of Chartered Business Valuators, 1999, p. 51.

As you can see on figure, there is two approaches for determination levels of value: traditional and updated. The "traditional" levels of value chart has three levels: the control level, the marketable minority (or "as-if freely traded") level, and the nonmarketable minority level. As we will see, there are two key valuation discounts and one premium that enable "movement" between levels on the chart. It also shows conceptual discounts and a premium that enable appraisers (and courts) to move from one level to another. In second approach has an addition level of value - Strategic control value.

Thus, when it comes to a complete acquisition (100% of the shares or shares), the value of the company for a particular investor comes from the point of view that is different from the market consensus, then we move away from the standard of fair market value (consensus value) and proceed to investment value. It should be remembered, that the premium rates that reflect the benefits to such special customers, brings to the situation element of the investment costs, in contrast to the net fair market value. The aim is to clarify the schema definition of the levels of value in the complete acquisition.

Now we should focus on the conditions of the definition of this new category of business valuation - synergistic (strategic) premium. On the quantitative basis of the size of synergistic (strategic) premium, there is no statistically significant estimates. Empirical research that would help to determine the size of discounts (for lack of absolute control), are absent, but usually its size is in the range from 5% to 15%. Control Premium Study 3RD Quarter 2014, Available at: http://www.bvmarketdata.com/pdf/CPS3q14.pdf

It is easy to calculate the range of control premiums based on submitted expert valuation discounts for lack of strategic control (100%). Then the range of expert estimates strategic (synergistic) premium for the total acquisition going to 5.3%-17.6 % Control Premium Study 3RD Quarter 2014, Available at: http://www.bvmarketdata.com/pdf/CPS3q14.pdf. Such significant difference in the quantitative values of this premium had explained as follows: 100% of the acquisition provides a strategic opportunity management. A synergistic effect may appear (or not appear) in the implementation of the strategic opportunities (after the event) additional events to increase company value. The effect of these events should have expected with a certain time lag is about 1-2 years.

The lower bound of the range of strategic (synergistic) premium for full (100%) acquisition of the company has characterized by the appearance of additional strategic opportunities management. With regard to the upper boundary of this range (17.6 %), it is likely the assessment of the possible (but not mandatory) the emergence of synergies. By itself, the synergistic effect may not appear automatically in the total acquisition of the company.

As result, it is necessary to separate the concept of a strategic premium and synergistic premium for 100% buyout of the company.

The strategic premium is a more general category. It occurs usually at 100% acquisition by the owner of a company. Obvious advantages of this acquisition - complete freedom of action of the owner in the conduct of personnel policy within the company, and to release the acquired company's non-core assets, as well as in the implementation of other activities, associated with the company's reorganization and diversification of its activities.

Moreover, premium has determined by the fact that in a certain situation develops a greater effect than the sum of its parts. In particular, this effect can occur when combining a variety of commercial enterprises to increase income and faster growth than can be achieved by each of them separately.

With regard to the synergistic of the premium, it has linked to the potential and not with the real possibilities of increasing the cost of acquiring the company. Therefore, the strategic premium is one category, and synergistic premium - different category. Although both types of premiums have the same basis: they occur only when exist 100% acquisition of the company.

There are only potential opportunities to benefit from the expansion of the business after the acquisition - increase in cash flow generated new value drivers that appear after this acquisition. However, the increase in cash flow, and hence increase the value of the company has achieved immediately, but after a certain time. And not just by one or other of the orders, and as a result, as a rule, the implementation of specially prepared activities and the investment of additional funds, the returns from which, as expect, will be rather high, but not immediately, but after 1-2 years.

The distinction between the strategic premium and synergistic premium defines a very significant range of values. The lower bound of this range (5.3 %) is associated only with strategic control, the top -with a synergistic effect (17.6 %).

In addition, in the works of other authors have presented and a higher range of values of premiums to the market price of companies in transactions for the acquisition and acquisition. For example, Damodaran provide empirical data on these premiums in the range of 20-35%.

Now we going to describe the main research papers in control premium valuation by methods.

A.1. M&A Method

Positive acquirer returns and dollar value gains appear unique to the transfer of control in M&A of Emerging-markets. Acquirers can improve the target value by sharing better institutional and corporate governance practices such as legal and accounting standards. The importance of contract enforceability is likely to matter more in contract-intensive activities such as R&D or other intangible asset production. A stockholder would only be willing to pay a premium if he or she believes in the potential of the acquisition to increase the value of the firm. We research several articles, which applied M&A method for control premium study (see Table below).



Table II

Investigation of Control Premium Across the Country

Author, Year	Country	Main Inference and Distinguishing Feature
Komiak (2007)	Cross-country	Location of target firm and acquisition premium. Control premium higher when there is domestic acquisitions
Dracota (2013)	Romania	Take into account bargaining power of different agents. Control premium for Romanian listed companies are similar to those revealed for other countries
Varaiya (1987)	Cross-country	Observed premiums are predicted to be positively related to the magnitude of the acquiring firm's estimate of acquisition gains and the acquired firm's relative bargaining strength

From the table you can see investigation of control premium across county. Some authors concluded that control premium higher when there is domestic acquisitions, in another case control premium for Romanian companies are similar to those revealed for other countries.

A.2. Block Transaction Method

It involves a significantly large number of shares or bonds being traded at an arranged price between parties, outside of the open markets. Considered approach is very similar to M&A one, but the latter implies substantial premiums, while the first case is not necessarily so.

In this case, we consider Matrix of premiums made by Damodaran (See Table below)

Table III

Matrix of Premiums

Type	Capital Allocation	Premiums Distribution	Type	Capital Allocation	Premiums Distribution
1	90+10	35+5	4	50+50	20+20
2	80+20	32+8	5	51+49	25+15
3	80+10+2+1	32+5+1+0,5	6	49+49+2	17,5+17,5+5

In the table above we may see that control premium vary across equity stakes. For example, for the first type of capital allocation, 90% of the company equal 35% th size of control premium, and for 10% - 5% accordinglly. Also we can note that with decreasing of a controlling stake decrease premium of control.

Also we provide literature review of control premium on Emerging markets (see table below).

Table IV

Literature Review of Control Premium Across Country

Author, Year	Country	Main Inference and Distinguishing Feature
Jurfest (2015)	Chile	Studies the change in private benefits after the switch from the Market takeover Rule to an Equal Opportunity Rule (it grants all shareholders the right to participate in a takeover transaction on equal terms). Private benefits decreased significantly.
Nenova (2003)	Cross -country	The value of control-block votes is an important part of firm value and varies widely across countries. Average value of control-block votes ranges from 48% of firm value in South Korea to 2.88% (assuming the controls are at their mean values) in Hong Kong.
Nenova (2005)	Brazil	Firm control value is directly affected by changes in the legal protection of minority shareholders (i) increased after Law 9457/1997 that weakened the rights (ii) decreased after Instruction 299/1999, which restated the protection
Byrka-Kita (2015)	Poland	Private benefits of control are much lower than on mature capital markets. Reason: law liquidity and information imperfectness
Bian et. al. (2012); Fan et al. (2012)	China	Both articles concluded that block trades are priced at a sizeable discount of close to 4%. The price discount of block transactions is smaller when the firm is large, when there are multiple block transactions on the same day, or when the buyer is represented by China International Capital Corporation. As a result state regulation (in 2008), there was a dramatic increase in the volume of transactions on the large trade platform.
Dyck, Zingales (2004)	Cross -country	The value of control-block votes spans from 2% in South Africa to 65% in Brazil. While the average in emerging countries is 19%. Therefore, there are plenty of institutional factors, which affect control premium, such as: accounting standards, legal protection of minority shareholders, law enforcement, intense product market competition, level of diffusion of the press and rate of tax compliance.

Literature review show that many articles investigate control premium in Emerging markets applied Block transaction method. Average value of control-block votes ranges from 48% of firm value in South Korea to 2.88% (assuming the controls are at their mean values) in Hong Kong. The value of control-block votes spans from 2% in South Africa to 65% in Brazil. While the average in emerging countries is 19%. This study confirmed that control premium vary across country and region.

A.3. Dual-class ownership method

The right to control a corporation - particularly valuable. The controlling party receives private benefits of pecuniary and nonpecuniary type. Obligatory condition for the data sample: the firms chosen for the research possess common voting and preferred non-voting shares outstanding that differ primarily in their voting rights, hence, control privileges.

Table V

Investigation of Control Premium on Emerging Markets

Author, Year	Country	Main Inference and Distinguishing Feature
Saito (2003)	Brazil	Except for traditional factors examines the influence of legislation alterations in the sphere of voting minority shareholders' rights protection and determination
Chung, Kim (1999)	Republic of Korea	Come to a conclusion, that larger firms exhibit smaller voting premiums
Berezinets et al (2011)	Russia	U-shaped relationship between CEO age, CEO ownership concentration and private benefits of control
Guadalupa (2007)	22 countries (Brazil, Chile, Colombia, Mexico, Peru, South Africa, Venezuela)	Product market competition improves governance -> increase private benefits of control estimated using voting premium.

Investigation of control premium shows that legislation alteration, CEO age, competition may influence on valuing premium of control.

B. Determinants of Control Premium

During the analysis, it was found that nearly each article about determinants of control premium contains different determinants. That is why all detected determinants were grouped into several groups. This classification is shown on the graph.



Figure 5. Group of determinants of the control premium

Now let us consider other side of control premium. The existence of private benefits of control does not necessarily lead to a redistribution of value from majority shareholders to minority (Holderness (2003)). These benefits will have a different shape and cost (Dyck and Zingales (2004)). For example, this benefit has reflected in the fact that the company occupies a higher social position and confidence. In opposite case, when minority shareholders will suffer, then the company loses these advantages. Therefore, in some cases Benet from owning a controlling stake can produce benefits to both groups of shareholders (Holderness, (2003)).

A control premium has explained by the unrest of moderately controlling stake in the loss of part of their well-being because of the conduct of the holders of a controlling stake (Barclay and Holderness (1989); Dyck and Zingales, (2004)). The same situation is observed in the works for Bulgaria (Atanasov (2005)), Poland (Trojanowski (2008)), Romania (Dragotг et al. (2009)) and other countries. The definition of a control premium is sometimes more difficult for these countries and for the developing and Command economies financial markets (Atanasov (2005), for Bulgaria; Trojanowski (2008), for Poland, Filatotchev et al. (1999) and Chernykh (2011), for Russia, etc.). Such differences can be explained not only economic or financial variables, but also others, such as psychological, the difference in the level of financial education of investors, which definitely effects on size of control premium.

Control premium calculated in two ways. For example, Nenova (2003) measuring the benefits of corporate control as a cost that the owners of the dominant shares of the company redistribute with their favor. In addition, Nenova (2003) insists that the control premium can be measured by fluctuations in the price of the package during the auction. This has own drawbacks, because it does not consider the characteristics of financial markets in different countries.

Moreover, the estimation of control premium based on two classes of shares that are not wealthy to developing countries. This is primarily because the markets have low informational efficiency, and there is a shortage of relevant data. Therefore, it is difficult to measure private benefits of control for developing countries and this leads to the displacement of estimates.

Since the early 1980, researchers tried to determine the value of the right of vote and determinants by one country (Lease et al. (1983), Horner (1988), Megginson (1990), Zingales (1994, 1995), Hoffmann Burchardi (1999), Goetzmann et al. (2002), Linciano (2002), Neumann (2003)) and then by countries (Nenova (2003)).

First researched the size of the premium and determinants in developed markets, and after for emerging markets. For example, Barclay and Holderness (1989) analyzed rates of 63 stakes from 1978 to 1982. They found that average premiums going to 20%. These premiums has perceived a private benefit of control. Meeker and Joy measured control premium for package of shares for private banks and found a large amount of the premiums between 50% and 70%.

In general, studies of determinants of control premium can distinguish the following groups of factors: control variables, financial structure, economic factors and legal protection of the business environment. In the financial architecture group includes variables: the governance structure, ownership structure and capital structure. The following table show the set of control variables (see also Appendix):

Table VI

Control Variables

Variable	Study	Impact
Total assets	Karen Simonyan (2014) Dyck and Zingales (2002), Barclay and Holderness (1989), Nenova (2003), Gaspar et al. (2005), GrzegorzTrojanowski (2003) Nicodano and Sembenelii (2000), Franks et al (2001), Chong-en Bai et al (2003), Eugene Nivorozhkin et al (2012),	-
Total capitalization	Barclay and Holderness (1989), Nenova (2003), Dyck and Zingales (2004), Gaspar et al. (2005), Dyck and Zingales (2002), Dittmann et al (2004), Victor Dragota et al (2013)	+
Prior firm performance	Grzegorz Trojanowski (2003), Barclay and Holderness (1989)	+
Target total assets divided by bidder total assets	M.F.Thraya, J.Hagendorf (2010), Karen Simonyan (2014), Alexandridis, Fuller, Terhaar, andTravlos (2013) Vijay B. Gondhalekar et al (2004)	+
Ratio of dividends to net income	M.F.Thraya, J.Hagendorf (2010)	_
Volatility of the common stock return	Barclay and Holderness, (1989), Karen Simonyan (2014), Mario Massari et al (2004)	+/-
Dual class shares	Mario Massari et al (2004), Bianchi et al (1996),	+
Profitability	Grzegorz Trojanowski (2003), Nicodano and Sembenelii (2000), Dyck and Zingales (2002)	+
Tangibility	Sonia P. Jurfest et al (2015)	-
Tobin's Q	Nivorozhkin et al (2012), Vijay B. Gondhalekar et al (2004)	+

The first two variables, namely, total assets and market capitalization, serve as a proxy for firm size. Initially, the relationship between firm size and the control premium is entirely ambiguous. Moreover, larger companies are more susceptible to control by the government, stock exchange or institutional investors, which reduces the ability to extract private benefits at the expense of minority shareholders. On the other hand, large companies potentially provide more opportunities for extraction benefits. However, despite this logic, the authors observed a positive correlation between variables.

The next two variable focused on the agency conflict between controlling shareholders and minority shareholders, namely: the dividend payout ratio and the ratio of cash flow to net sales. The authors use the dividend payout ratio for last 12 months. A negative relationship between size of control premium and dividend payout ratio has explained as follows: the company, in order to pay high dividends are constrained from spending funds for other needs and in particular to divert money from the company.

Dual class shares. Although the issue of voting shares allow for controlling shareholder to implement a simpler and more stable control of the company restricting the rights of small shareholders. For this variable, the authors use a measure of the proportion of non-voting shares in the structure of all issued shares and find a negative relationship with the size of control premium.

According to Grzegorz Trojanowski (2003), Nicodano and Sembenelii (2000), Dyck and Zingales (2002) profitability positively influences on size of control premium, because more benefits can receive the controlling shareholder, ceteris paribus.

The high share of fixed assets in the total assets of the company affects on size of control premium. Because for these assets require more time for their sale or removal of assets of the company. Therefore, controlling shareholders easier to steal cash and cash equivalents because it be done quite quickly (Sonia P. Jurfest et al (2015)).

The last group of control variables is Q-Tobin's. Nivorozhkin et al (2012), Vijay B. Gondhalekar et al (2004) explain a positive correlation with the control premium because firms with high Tobin's Q potentially can extract more resources and the value of its shares increases.

In the variable financial architecture group (Table VII, see also Appendix), we considered the indicators of the governance ownership structure, and capital structure.



Table VII

Financial Architecture Variables

Variable	Study	Impact
Leverage	Jensen (1986), HarrisandRaviv (1988), Stulz (1988), BarclayandHolderness (1989), VijayB. Gondhalekaretal (2004), Palepu (1986), Billett (1996), Barberetal. (1999), Nuttall (1999), CuddandDuggal (2000), BarakandLauterbach, (2012), Albuquerque&Schroth, (2010), Nicodano&Sembenelli, (2004)	+/-
HHI	Duggal and Millar (1999), Nenova (2003), Gaspar et al. (2005), Luypaert and Huyghebaert (2010), Victor DRAGOTГ et al (2013), McConnel and Servaes (1990), Faccio and Lasfer (1999), Short and Keasey (1999), Yu Yen and Andrea (2007), La Porta et al (1998), Shleifer and Vishny (1997)	+
Type of the major shareholder (domestic or foreign)	Hanousek et al. (2007), Victor Dragotг et al (2013)	+
Buyer's type (corporation or individual)	Barclay and Holderness (1989), Kand and Kim (2008), Ragozzino and Reuer, (2011), Hanousek et al. (2007), Victor DRAGOTГ et al (2013)	+
Percentage of share purchased in the transaction	Walkling and Edmister (1985), Barclay and Holderness (1989), Dyck and Zingales (2004), Victor DRAGOTГ et al (2013)	+
Bidder's origin	Dyck and Zingales (2004), Massari et al. (2006)	+/-

Among researchers, there is no consensus about the impact of financial leverage on the size of control premium. First, the debt limits of access free cash flow, indicating a negative impact of private benefits of control (Albuquerque and Schroth (2010), Barak and Lauterbach (2012)). On other hand, debt allows company to expand company's assets, which undoubtedly has a positive effect on the size of control premium (Barclay and Holderness (1989); Nicodano and Sembenelli (2004)).

Ownership structure also has a significant impact on control premium, the authors found that if the target company has one majority shareholder (more than 50% of shares), then a control premium is higher. In the case where the ownership structure is becoming more diversified, (there are shareholders who have more than 33 % but not more than 50% of the capital) correlation becomes lower and negative when the shareholder owns less than 33 % of the shares.

If the target company have domestic owners, then a control premium higher. Because domestic owners can offer more from a position of control than just the benefits of the company. Local environment, social capital, communication and this are something that cannot be valued in monetary terms. Indeed, Victor Dragota investigated this issue and noticed that domestic ownership before transaction were associated with a higher control premium.

Also Hanousek et al. (2007) and Victor Dragotг et al (2013) tested the effects of state ownership on the size of the control premium. However, the results were inconclusive. First, such companies in the sample a bit, in which controlling stake belongs to the state, and secondly, when the authors took into account these companies, the results were statistically insignificant.

A control premium is higher if the buyer of corporation's shares not the individual investor. Because the corporation, which buy a controlling, stake of company, may take a strategic decision to expand business. The corporation is ready to pay much higher for a controlling stake in contrast to the individual investor, who intends to preserve the capital. In addition, a control premium is directly dependent on the size of stake, because a larger package gives more votes for decision-making.

The influence buyer origin of the country on the size of the control premium is not clear. Some studies, such as Dyck and Zingales (2004) find a positive correlation. They argue that companies that are attractive targets for foreign companies are subject to greater competition in the market for corporate control; therefore, the increase in investment demand leads to an increase in the control premium. Others, such as Massari et al. (2006) found a negative relationship, suggesting that foreign buyers have faced with great difficulties in extracting private benefits from control.



Table VIII

Legal Environment Variables

Article	Nenova (2003)	Nenova (2005)	Jurfest (2015)	Guadelupe (2007)	Reason
Sample / Factor	30 countries	Brazil	Chile	22 countries
Investor protection	-	-			Good general investor protection laws increase the likelihood of a lawsuit and therefore a monetary or reputation loss to a controlling party who is in a position to extract private benefits.
Law enforcement	-		-	-	The strength of legal regulation weak private benefits of control and leads to reduction in Takeover law - - control premium
Takeover law	-			-
Competition law				-

La Porta et al. (1998) and Nenova (2003) analyzed this relationship for a large group of countries. In addition to other variables they used in the model, variables that characterize the quality of the legal environment and the level of protection of investor rights: Rule of law, Investor protection, Takeover regulations.

Investor protection. Averages six indicator variables of general investor protection (1):

(i) If shareholders are allowed to mail their proxy votes,

(ii) If the shares are blocked before a general shareholder meeting,

(iii) If cumulative voting is allowed in the law

(iv) If oppressed minority provisions are incorporated in the law

(v) If shareholders enjoy a preemptive right over future issues, and

(vi) If the vote percentage required to convene an extraordinary shareholder meetings at or below 10%. Higher values indicating better protection Nenova (2003), La Porta et al. (1998).

Takeover regulations. Averages three variables of investor protection during a corporate control contest (1):

(i) If the legal code requires a control contestant to offer all classes the same tender price, zero otherwise;

(ii) If the legal code requires a buyer of a large or majority block to pay minority shareholders the same price as for the block shares by share class, zero otherwise; and

(iii) Minus the level of ownership at which a dominant vote-owner is legally required to make an open market bid for all shares, zero if there is no such provision in the law. Higher values indicate stricter takeover law.

Figure 6. External determinants: Competition

Rule of law. Assessment of the law and order tradition in the country produced by country risk-rating agency (International Country Risk for 1982-1995). Rescaled 0 to 1, with higher values indicating stricter rule of law (La Porta et al. (1998)).

By including these variables in the model, the authors concluded that the value of controlling stake significantly depends on the mechanism of corporate governance and investor protection. Rule of law, Takeover regulation, and Investor protection together explain 68% of systematic differences in the value of the block shares. Moreover, the results indicate the importance of a legal framework in curbing private benefits from control of the company. On the other hand, the authors show that the rights of minority shareholders have very effectively supported by the presence of a good legal framework that shows the relevance of this study.



2. Sample and Methodology

In order to identify the determinants of control premium on emerging markets, we have used the Zephyr database and uploaded the deals of public companies. It should initially recognize the fact that the BRICS countries are completely different in terms of culture, traditions, institutional and financial infrastructure. In connection with these deals, it is important to consider analysis not only within the same group, but also for different countries.

Data collection except the financial sector, because there occurs specificity evaluation. We assume that we will get a heterogeneous sample, since the survey has conducted in different countries and industries. The data uploaded for 2008 to 2010. We expect to receive different result for this period.

There are following sectors for our analysis : Primary Sector (agriculture, mining, etc.), Food, beverages, tobacco, Textiles, wearing apparel, leather, Wood, cork, paper, Publishing, printing, Chemicals, rubber, plastics, non-metallic products, Metals & metal products, Machinery, equipment, furniture, recycling, Gas, Water, Electricity, Construction, Wholesale & retail trade, Hotels & restaurants, Transport, Post and telecommunications, Other services, Public administration and defense, Education, Health. All deals provided by public companies.

We also examined the effect of the country on control premium. Therefore, we divided our sample into groups by country label. Moreover, we used dummy variables to reflect this influence on the dependent variable. Descriptive statistics for each country presented in the appendix and on the figure 7.



Figure 7. Dataset structure

The average amount of control premium for whole sample going to 3.2%, separately China - (-10%), India - 13%, Brazil - 12.7%, and the Republic of South Africa - 12.4%. In general, some differences are observable in the sample; in particular, China stands out, as the control premium is negative, which indicates that in average stakes are selling at the price below the market and in this case control worth nothing.

Figure 8. Correlation matrix



Correlation matrix shows for us that many variables have weak relationships between groups of variables; however, some determinants have strong relation power (dark colors).

For further analysis, we have put seven hypotheses. In addition, they we divided into several groups. The first group refers to the explanatory variables, and the last testing the differences between countries in the size of the control premium in context of different periods. In case, we define control premium that occurs at 50-100% acquisition, here, obviously, we are talking about strategic plans for the company, but not short-term extraction of private benefits. To detect difference two premiums, we introduce a dummy variable to buy blocks of 50-100%. This variable allows understanding, which variables will influence on size of strategic control premium, and which will be not statistically significant.

We formed following hypothesis:

H1.Control premium tend to be lower when the tangibility of the company's assets is higher.

We expect that high level of fixed assets in the total assets of the company negatively effect on size of control premium. Because these assets require more time for their sale or removal of assets of the company. Therefore, controlling shareholders can easier tunnel cash and cash equivalents because it can be done quite quickly (Sonia P. Jurfest et al (2015))

H2.The size of the stake purchased in transaction has a positive effect on the control premium.

The shareholder in a company is active or passive; the number of purchased shares may affect on size of control premium. A large share of ownership in a company gives the controlling shareholder not only some private advantages, but also some private spending (for example, a diversified portfolio) (Barclay and Holderness (1989)). For this reason, the relationship between the size of the stake (ACQ_STAKE) and the control premium is not obvious. However, in the case of the emerging capital markets, we expect that private benefits exceed the private costs.

H3. The level of financial leverage has negative effect on control premium.

Among researchers, there is no consensus about the impact of financial leverage on size of control premium. First, the debt limits access of free cash flow, indicating a negative impact on the private benefits of control (Albuquerque and Schroth (2010), Barak and Lauterbach (2012)). On the other hand, debt allows the company to expand company's assets, which undoubtedly has a positive effect on the size of control premium (Barclay and Holderness (1989), Nicodano and Sembenelli (2004)).

In our case, we expect a negative relationship, because investors in emerging markets have a more short-term investment views, and, respectively, leverage will limit the possibility of obtaining private benefits.

H4. Control premium is higher, when acquirer is strategic investor.

For the strategic investor the acquired package used investment value and for ordinary investor - fair market value. We expect that a control premium in this case will be higher.

H5. The size of the acquired company has inversely related with the control premium.

Both large and small companies are the subject of transactions in the market of mergers and acquisitions. We expect that the ownership structure has more dispersed in a large company than in smaller ones. In the case of small companies, ownership can be more concentrated and the shareholders will not want to sell their shares. This factor has used in majority of studies relating to the size of control premium. However, the relation between firm size and private benefits from control ambiguous (Massari et al. (2006)). As a proxy for firm size, we used the capitalization of the company before transaction (Nenova (2003); Gaspar et al. (2005)).

H6. Global Financial Crisis 2008 to 2010 effected on control premium determinants.

During the financial crisis, many economic indicators change. M&A activity differ for period of stable economy and crisis, and therefore differs a control premium.

We expect to get a result that is different for the sample in a stable period. It is possible that some factors will be more significant in the crisis period. Also interesting to see what amount of differences observed between two periods.

H7. Control premium determinants vary in China and India.

We expect that some countries will have different set of significant variables as it is in other studies. However, it is important to understand what factors affect on size of the control premium in different countries, and why a control premium is different in countries.

In our study, we assessed control premium as the difference between the sale price and the price of the shares prior to announcement of the transaction (Barclay and Holderness (1989); Dragotг et al. (2007)). Thus, a control premium (CP) we estimate as follows:

, where (1)

Po,t - offer price of the block

Pt-1 - price of the shares before the bidding moment

Model

CP = Xb + e, where (2)

X - Matrix of variables;

b - Vector of coefficients

e - Residual perturbation;

EQUITY-Target shareholders equity last twelve months;

T_MCAP - The total capitalization of the target-company before the transaction;

LEVERAGE - Book value of Financial Debt / EQUITY of the target one year before transaction;

ACQ_STAKE - The size of the stake purchased in the transaction;

MB - Tobin's Q;

DEAL_VALUE -

D_RESIDENT - Dummy variable for the buyer's type (1 if the buyer is a domestic company, 0 if the buyer is a foreign company;

DSTRATEGIC - Dummy variable (1 if investor buy 50-100%, 0 otherwise);

D_CRISIS - Dummy variable (1 if year for estimation is 2008, 2009 or 2010, 0 otherwise).

D_CN - Dummy variable (1 if target is in China 0 otherwise);

D_IN - Dummy variable (1 if target is in India 0 otherwise);

And also we use cross dummy (D_IN:D_CRISIS, D_IN:D_RESIDENT, D_CN:D_CRISIS, D_CN:D_RESIDENT, D_CN:D_STRATEGIC, D_RESIDENT:D_STRATEGIC , D_CRISIS:D_RESIDENT

Empirical Results

Econometric analysis results are shown in Table IX. With six variables, we tried to explain the size of control premium, not only for the whole sample and in the context of the country, the financial crisis, the type of investor.

Table IX

Empirical Results of the Investigation

Model/Variable	1	2	3	4	5	6	7	8	9	10	11	12	13
(Intercept)	1.26	1.27	1.23	1.01	1.29	1.39	1.34	1.36	1.23	1.29	0.93	1.10	1.25
ACQ_STAKE	-0.30	-0.31	-0.30	-0.21	-0.18	-0.29	-0.28	-0.31	-0.29	-0.27	-0.18	-0.23	-0.31
MB	-0.19	-0.19	-0.19	-0.18	-0.08	-0.18	-0.19	-0.16	-0.17	-0.17	-0.19	-0.18	-0.20
LEVERAGE	-0.10	-0.09	-0.09	-0.08	0.28	-0.08	-0.08	-0.08	-0.07	-0.08	-0.07	-0.08	-0.09
DEAL_VALUE	0.31	0.32	0.32	0.28	-0.16	0.31	0.26	0.33	0.27	0.25	0.25	0.32	0.32
T_MCAP	-0.17	-0.17	-0.17	-0.16	-0.12	-0.18	-0.17	-0.20	-0.16	-0.15	-0.14	-0.18	-0.18
T_EQUITY	-0.14	-0.14	-0.14	-0.12	0.10	-0.13	-0.09	-0.12	-0.11	-0.11	-0.12	-0.14	-0.14
D_CRISIS		0.07					0.15		0.07				0.05
D_RESIDENT			0.04					0.10		-0.05		0.04	0.04
D_STRATEGIC				-0.25							-0.02	-0.22
D_CN					-0.29				0.15	-0.07	0.15
D_IN						-0.08	-0.10	0.04
D_IN:D_CRISIS		Sagn. Level				-0.11
D_IN:D_RESIDENT			0.05					-0.14
D_CN:D_CRISIS			0.00						-0.09
D_CN:D_RESIDENT			0.00							0.25
D_CN:D_STRATEGIC			0.01								-0.52
D_RESIDENT:D_STRATEGIC												-0.02
D_CRISIS:D_RESIDENT													0.01
# observations	350	93	286	44	147	155	26	126	59	127	27	35	78
R^2	0.60	0.60	0.60	0.61	0.55	0.61	0.59	0.60	0.57	0.52	0.57	0.62	0.61
Adj. R^2	0.59	0.59	0.59	0.61	0.53	0.60	0.58	0.58	0.56	0.51	0.55	0.61	0.59
p-value	< 2.2e-16	< 2.2e-16	< 2.2e-16	< 2.2e-16	< 2.2e-16	< 2.2e-16	< 2.2e-16	< 2.2e-16	< 2.2e-16	< 2.2e-16	< 2.2e-16	< 2.2e-16	< 2.2e-16

As a result, 13 regression models were constructed. Some hypothesis was confirmed, and the rest were rejected. All models are adequate and significant, R-squared is about 60%. Thus, we used only six variables and explained more than half of the variance. It should be noted that during the crisis, control premium increases by 7% and correspond dummy variable is significant at the 0.05 level. Thus, we can say that there is a significant difference in the value of buying a package in times of crisis and in times of economic stability. The fact of whether the target of the acquirer are residents in one country does not affect the size of the control premium. The following results will be presented in section six explanatory variables and dummy variables.

Figure 9. Relationship ACQ_STAKE with dummy variables

H1 was rejected, because we got a negative relationship between CP and ACQ_CTAKE. This can be explained as the company on emerging markets, which buy large packages, have not only private benefits of control, but also bear the extra expense. In addition, the average costs exceed benefits.

The graph shows the values of the elasticity coefficient for ACQ_CTAKE for different models. Dummy variables D_RESIDENT, D_IN: D_RESIDENT, D_RESIDENT: D_STRATEGIC and D_CRISIS: D_RESIDENT are not significant. Moreover, we may conclude that there is no difference in the size of control premium for this case, where target and acquirer are resident in the same country.