Corporate performance and exit strategy of PE investors

(1)

Corporate Performance and Exit Strategy of PE Investors

Master Thesis

Master in International Finance, Amsterdam Business School Prepared by: Stanislav Butorin

Supervisor: dr. J.K. (Jens) Martin Date: 31 August 2016

(2)

2

Abstract

This thesis examines the influence of private equity and venture capital investors on the performance of their portfolio companies. A substantial sample of 278 companies was used in the research. The array of data covers the period between 2000 and 2009. A simple cross-sectional OLS regression was employed for econometric analysis. The study confirmed positive short-term influence of the retained ownership stake held by PE/VC investors on the performance of their portfolio companies, while the presence of the long-term effect was not discovered. In contrary to the expectations, the period of ownership is found to be strongly insignificant in all variations of the model and series. The size of the company is negatively correlated with its performance which can be explained in terms of the negative effect of scale. The aforementioned dependencies remain after splitting the sample into two subsamples based on the industry factor.

(3)

3

INTRODUCTION

Private equity fund is one of the forms of collective investments in equity and debt of predominantly non-public corporations. A PE firm, establishing such a fund, defines an investment strategy, attracts professional investors (so called “limited partners”), and selects perspective projects for allocating money, also known as portfolio companies. Venture capital funds are a subset of PE funds which invest primarily in startups and enterprises in early stages of the corporate life cycle, working mainly in innovative industries. Such investments are characterized by taking a very significant risk and, consequently, required rates of return are substantially higher compared to those of investing in shares or bonds of stable, mature companies.

Writing about startups and hi-tech companies, we should mention such a category of investors as business angels. A business angel is an individual who invests in innovative startups and receives shares in exchange for his contribution. Unlike funds, a business angel is an individual investor who risks his own money and does not bear any fiduciary responsibility. This peculiarity is one of the reasons of differences in investment strategies of funds and angels which influence portfolio companies.

Important similarity of different types of PE investors is investing in firms which usually cannot attract traditional financing through bank loans or issuing bonds or shares in capital market at reasonable costs. This is explained by higher risks of projects being realised, as well as by early stages of life cycle of such companies, when the cash flow from operations (CFO) is negative, so the company just absorbs money. This state of things puts the shareholders before the choice: either to find investors being ready to assume substantial risks but placing strict requirements (significant share stake or active participation in managing the company), or to use limited internal resources and restrain development of the enterprise.

The aim of this thesis is to identify the relationship between corporate performance and receiving PE investments. In other words, the hypothesis to be checked is that attracting PE investors contributes to firm’s achieving higher financial results and market value and outperforming its peers.

The solution of the aforementioned problems can be divided into the following consequent steps:

(5)

5

1. Analysing related scientific articles and preparing a literature review in the field of private equity and, in general, ownership structure, embracing the following main questions:

a. The relationship between the type of the main shareholder

(families/individuals, institutional investors, foreign investors, insiders, government) and the corporate performance.

b. The influence of institutional investors of different types and their behavioural patterns on the performance of the investees.

c. What actually changes in a company which gets funds from a PE investor – capital structure, governance, innovation activity?

2. Develop an econometric model to examine corporate performance in relation to different variables, characterizing a PE investor and the firm.

3. Prepare an array of historical data from different databases.

4. Conduct an empirical test of the developed model and proposed hypotheses based on the collected data.

Ownership structure and its influence on companies has been under scrutiny for a very long period of time. This topic is probably one of the oldest in empirical corporate finance. The first study in this field is believed to be the book of Bearle, Means, published in 1932. Last years we have seen a lot of papers which analyse the influence of share ownership of different investors on the value of portfolio companies, including studies based on data from developing markets, for which the lack of reliable data is a very well-known problem. We can state that the selected topic for this thesis is actual and widely discussed in the scientific literature.

The novice of this thesis is studying a specific type of institutional investors and their influence on their investees. The object of the research is the performance of various portfolio companies.

(6)

6

1. EMPIRICAL STUDIES EXAMINIG THE RELATIONSHIP BETWEEN

OWNERSHIP STRUCTURE AND CORPORATE PERFORMANCE

Corporate strategy and performance are being influenced by a lot of different factors. The theory of corporate finance identifies three fundamental components:

 Ownership structure – the distribution of the cash flows (payouts) among the shareholders, as well as the rights to vote on key strategic issues which are in the competence of the company’s general meeting.

 Capital structure – the ratio of shareholders’ capital (equity) and borrowed funds (debt). Increase in the proportion of debt reduces cost of capital (WACC) but increases financial risks, as far as significant volatility in revenues may result in insolvency and bankruptcy as the result of inability to satisfy the creditors’ requirements.

 Corporate governance, which directly influences the quality of solving various agency conflicts that inevitably arise in the day-to-day functioning of any enterprise. The higher is the quality, the better is the company’s ability to balance interests of different stakeholders and mitigate risks.

The combination of the aforementioned factors is called financial architecture. The term was first introduced by Myers in 1999, and the novice of this approach is that it suggests to consider and study all these components together, not apart, when treating them as separate and independent qualities. In this research the accent is on the first of the elements of financial architecture – ownership structure. Academic interest is in studying the relationship between financial indicators/firm value and the concentration of ownership in the hands of a limited number of holders, the type of the controlling shareholder, managerial ownership, and the presence in the capital of investors of a specific type. This chapter is dedicated to the overview of empirical studies concerning the interconnection between established ownership structure and corporate performance. Additionally, studies in the field of capital structure and governance will also be mentioned.

(7)

7

Theoretical framework, financial architecture, and the relationship between the type of the main shareholder and corporate performance

One of the fundamental studies about the subject of ownership structure is the article of Jensen, Meckling, published in 1976. The authors develop the theory of agency conflict between company’s management and its shareholders. According to this point of view, the company's management is interested primarily in enhancing their own well-being, including to the detriment of the shareholders. Incompleteness of contracts does not allow to create the necessary incentives and to concentrate the efforts and experience of management on maximizing the company's value. One way to create these incentives, which is discussed by the authors, is debt financing. The company which receives a bank loan or issues bonds is obliged to pay the interest and at the end of the period to pay off the creditors. Failure to fulfil its obligations leads to bankruptcy, which is accompanied by the change of the management of the company. Seeking to avoid such a situation, the management is forced and interested to undertake maximum efforts to improve the quality of running the business. On the other hand, the severity of agency conflict is reduced if the managers are also co-owners of the company, i.e. have a share in the capital. In any case, an important objective of shareholders is the oversight, monitoring of the activities of company’s management, so the key financial indicators of the enterprise, the efficiency of doing business, and, ultimately, its market value depend on how well they cope with their responsibilities. Another reason for the importance of ownership structure is distinctions of strategic goals and motives of different categories of shareholders. Thomsen and Pedersen (2000) adhere to the following typology of investors and provide their characteristics:

 Families and individuals, often holding managerial positions in addition to their shares. Usually they are the founders or their descendants. For this category of owners, the business on the one hand is the guarantee of their welfare, and on the other hand it is the family pride. They are less prone to risking and raising capital by issuing shares due to dilution of their stake. The preferred strategy is maintaining a small, but stable long-term growth.

 Corporations. When having a net of subsidiaries and dependent companies, operating and financial synergies come to the fore, as well as the tasks of creation of vertically integrated structures, marketing and any other

(8)

8

consideration not directly related to the maximization of the value of the investment object. Thus, corporations are often mediocre investors.

 Government. It is the least effective owner because is normally focused not on the economic efficiency but on solving various political or socially important tasks (e.g., utilities sector, energy, national defence, etc.).

 Institutional investors and banks, which are sometimes treated as a separate group. The first priority for them is financial results that can only be achieved by maximizing the market value of the firm. This gives birth to the hypothesis that institutional investors are the most effective owners.

Among the issues discussed in papers on the subject of the ownership structure, one is the problem of endogeneity of ownership structure. The discussion is conducted about the nature of causation, whether the ownership structure is an independent, exogenous variable, or, on the contrary, is determined by economic performance of the company. Below are the arguments for both points of view.

An explicit or implicit premise of many studies is the hypothesis that company value is a function of its ownership structure. Scholars believe that the motivation and skills of the shareholders of a particular firm in the field of mitigating agency conflicts lead to achievement of a certain level of corporate performance. In their article, McConnell and Servaes (1990), among others, pay special attention to the problem of endogeneity and conclude that ownership structure is exogenous, i.e., is the result of external events.

According to the opposite point of view, the basic, underlying reason is economic efficiency of the enterprise. For example, Demsetz and Villalonga (2001), analysing a sample of U.S. companies, confirmed the lack of a stable statistical relationship between differences in ownership structure and performance. The obtained results are interpreted by the authors as a sign of investors’ rationality. According to this concept, making decisions about investing, particularly the purchase or sale of shares, all investors maximize the return on their investments. Thus, any transaction is possible only under the condition that both the seller and the purchaser consider it beneficial to themselves. Consequently, the set of shareholders of a particular company is the result of many transactions, and in each of them the counterparts sought to increase their own well-being, accordingly, this motive is the root cause, and the ownership structure is the result.

(9)

9

In this thesis, we will stick to the point of view about the exogeneity of ownership structure, as far as companies having shareholders of a certain type are to be compared with those which do not have such owners, i.e. the ownership structure is considered to be given, predetermined. We now turn to the overview of papers that analyse the relationship of business efficiency with the type of the key shareholder. For enterprises in which the ownership is separated from managing day-to-day operations, an important problem is the control of the management team by the shareholders. The interest of each shareholder and his willingness to expend energy and resources for such a monitoring are highly dependent on the size of his stake. The same can be said about the ability to influence decisions, which is realised as the right to vote at shareholders’ general meetings and to form the Board of Directors. Academic literature reveals a positive or at least bell-shaped (positive for small stakes, negative after reaching a breakpoint) correlation of ownership concentration with corporate performance of the enterprise. The effect of the type of the key shareholder on corporate performance is also under scrutiny. For example, Thomsen and Pedersen (2000) found that this relationship is statistically significant. Let us consider this question in detail.

A common way of motivating managers to work on increasing the value of the company is using stock options or stock payments, whereby key executives essentially receive a share in the equity. The stake owned by the management team and, in some cases, members of the Board of Directors is referred to in the academic literature as the “insider ownership”. In different papers, authors come to the conclusion that the insiders’ stake has a bell-shaped relationship with performance of companies. For example, McConnell and Servaes (1990) analysing U.S. companies discovered that the Tobin’s-Q ratio increased with the rise of the stake to the interval of 40-50%, but further reduced. Han, Suk (1998) based on the sample of U.S. companies revealed a stable positive relationship between performance and the share of insiders, until the latter reaches the level of 41.8%. In his article, Cho (1998) investigated companies from the Fortune 500 list and obtained similar results, although the breakpoint value turned out to be smaller.

The study of Russian companies conducted by Kokoreva and Stepanova in 2011 and covering data for the period 2005-2010 identified foreign investors from countries with developed capital market as a separate category. Having done a cluster analysis, the authors revealed three stable in time clusters of firms with significantly different

(10)

10

characteristics. The companies that were the most efficient and kept the steady growth even during the financial crisis of 2007-2008 were characterized by the highest level of foreign ownership, whilst the level of debt financing was the lowest, roughly 28% in stable periods. In contrast, the companies with mainly domestic shareholders and the minimal stake of foreign investors in the period of economic recession demonstrated worse financial results. The debt burden is higher and typically in the range between 30% and 40%. The result is consistent with the view that the influx of shareholders from developed capital markets increases the quality of corporate governance, transparency and ultimately business efficiency. Ferreira and Matos (2008) come to the same conclusion. Their data sample covers firms from 27 countries in the period 2000-2005. The authors conclude that the presence of foreign investors ultimately enhances the welfare of all shareholders. There is also the opposite point of view. The paper by Mykhayliv, Zauner (2013) analysed data for Ukrainian companies in 2003-2007 and found that the higher share of foreign investors in the capital led to stricter budget constraints that resulted in underinvestment and, as a consequence, the rejection of profitable projects. Thus, foreign investors reduce efficiency of the enterprise by depriving it of potential value growth.

Government is a highly specific owner. Government ownership, according to the literature, comprises shares owned by governmental institutions (ministries, agencies, etc.) and municipalities. They have the main common property: the government as the owner is focused primarily on the solution of various political and social problems, and achieving high financial results is a minor goal. The most typical example of state-owned companies is large, systemically important banks, as well as natural monopolies and enterprises from utilities sector (power plants, water treatment plants, railways and other transportation companies, etc.). It is clear that the government gives the heads of such entities the task of providing public goods of acceptable quality at affordable prices to the general public (electricity and gas, passenger transport, etc.), and is ready to tolerate mediocre financial results. Now let us consider in more detail the study on the influence of the government on state-owned companies.

The article by Duprey (2015) compares business models of private and public banks analysing the sample of 366 firms for the period of 1990-2010. The author concludes that the lending activity in the banks with state participation is much less cyclical (subject to reduction during economic crises) than in private banks. This phenomenon can be explained by the fact that in the times of recession the government imposes on

(11)

11

state-owned banks the function of stimulating recovery of the national economy forcing them to actively lend to the real sector avoiding the decline in the GDP. The results are higher level of risk and lower profitability. The study Xiao, Zhao (2012) based on the World Bank’s data is also dedicated to state-owned banks but considers a more complicated relationship. The authors conclude that the development of the banking sector has a negative impact on innovation activity of companies in countries with a high level of state ownership in the banking sector. The explanation of such an unusual conclusion is the following. The higher is the development of the banking sector in the country, the bigger role banks play in lending companies compared to the capital market. State-owned banks are not inclined to fund high-risk, innovative projects preferring a more conservative lending. As a result, the level of innovation in the real sector decreases, which negatively affects the efficiency of the business.

Writing about natural monopolies and utilities, we should pay attention to a few studies. In the article by See and Coelli (2012) the authors analysed thermal power plants in Malaysia in the period 1998-2005. The scholars examined operational efficiency of these companies using their specially designed index and came to the conclusion that it is more than 20% lower for the public sector. There is suggestion that the reasons for this difference are both underutilization of the capacities and excessive bureaucratization of business processes in the companies. In their article, Romano and Guerrini (2014) examine Italian water utilities. It was found that water utilities with private or mixed ownership demonstrate higher level of profitability compared to the “pure” state-owned ones. However, in another work in 2014 based on the analysis of Italian public utilities Monteduro found that the greatest efficiency is attributed to the mixed, state-private ownership.

We should also consider studies of countries with developing economies because there may be some peculiarities. In the work Mykhayliv and Zauner (2013), the authors note that state ownership, as well as the presence of foreign investors, leads to a drop in the level of investment and reduces business efficiency. A number of articles dedicated to the Chinese market consider the problem of state influence on business in the context of privatization of state-owned companies actively conducted since the beginning of 2000s. In their article of 2012, Zhang, Tang, and He found that privatized enterprises were characterized by increased production and operating efficiency, however, their profitability fell down. The reason lies in the “partial” privatization of many companies: despite transferring large share stakes to private owners, the state

(12)

12

still remains the controlling shareholder, i.e., the status quo in fact has not changed. The article by Kang and Kim (2012) offers a more in-depth analysis of Chinese firms with state ownership, emphasizing the fact that these companies are heterogeneous and highlighting the so-called “market-oriented state-owned enterprises”. This term refers to firms that are de jure owned by the state structures, however, focused on the economic results of their business. The analysis of this subgroup shows that their efficiency is higher than that of other state-owned companies. Another scholar, Yu (2013) comes to a different conclusion. Analysing the sample for the period of 2003-2010 the author infers that the government share has a U-shaped relationship with the performance, i.e. after a breakeven point the presence of the state in the capital of the company positively affects economic performance. This is due to the general positive influence of the ownership concentration which was discussed above.

Concluding the consideration of the role of the government, it is worth noting the work of Su and Fung (2013) which is also dedicated to the Chinese market. The authors identify as a separate factor of business performance the so-called “political connection” – the participation in the Board of Directors and the Board Management of officials of various levels and influence, which provides both private and state-owned companies access to low-cost financing, government guarantees, and government contracts. It turned out that the existence of such ties increased the performance of both types of firms.

Based on the abovementioned papers, we can draw some important conclusions: 1. The question about the causal relationship between ownership structure and

performance of the company is still not solved in the academic literature (the so-called “problem of endogeneity of ownership structure”). For the purposes of this work the ownership structure is considered as given from outside, i.e. exogenous.

2. Corporate performance has an inverse U-shaped relationship with managerial ownership, i.e. starts to decrease after reaching a breakeven point. Thus, the dominance of managers in equity has a negative impact on the company. 3. Foreign investors have a positive impact on the company as they promote best

business practices, increase transparency, and improve the quality of corporate governance. However, setting hard budget constraints for the managers, they

(13)

13

can be the cause of underinvestment by the company in prospective, lucrative projects.

4. In developed markets, excluding the utilities sector, government is an inefficient owner due to the priority of solving various socio-political problems over profitability of the business. For developing countries, there is a U-shaped relation between performance and state ownership because of the positive ownership concentration effect.

The influence of institutional investors on corporate performance

Many studies examine the impact of institutional investors on the company's performance. This category refers to various funds (PE, venture capital, pension, family offices, etc.), insurance companies, and banks. As mentioned above, for them the primary purpose of investing is to maximize the return on invested capital. Let us consider different papers testing empirically the hypothesis about efficiency of institutional investors as business owners.

In the article of McConnell and Servaes (1990) this problem is examined on the example of American companies’ data for 1976 and 1986. The results obtained are consistent with the hypothesis of efficient monitoring which states that the increase in the stake of institutional investors has a positive effect on the value of the company. However, modern research show that the problem is much more complicated and the conclusions are not so clear, particularly due to the fact that institutional investors are highly heterogeneous and pursue different strategies. The work of Ferreira, Matos (2008) includes data for 27 countries for the period of 2000-2005. The authors draw attention to the fact that institutional investors are playing an increasing role in the economy, their weight and involvement in different sectors is becoming more significant than ever. These investors prefer, most often, large firms with high quality of corporate governance. However, the positive effect on the value of portfolio companies is confirmed only in the case of foreign or independent institutional investors; otherwise there is no connection. The reason is that for independent and foreign investors it is easier to influence portfolio companies forcing them to work better because they have no other business relations and interdependencies.

(14)

14

Cornett, Marcus, Saunders, and Tehranian (2007) come to similar conclusions. They divide institutional investors into two categories – sensitive and pressure-resistant. The first group includes investors who have, or are likely to have, business ties with portfolio companies (banks, insurance companies), the second – those who have no such ties. In the first case, the authors believe, the monitoring function will be weakened because shareholders will not intervene in the companies’ operations fearing to spoil relations with the management which will negatively impact their own business (e.g., a company may change its serving bank or insurance company). Accordingly, the best option from the point of view of the monitoring efficiency is the share ownership of any pressure-resistant investors. However, an important aspect is the size of their stake because if it is small, on the one hand, there are no sufficient incentives and, on the other hand, the investor can relatively quickly and easily sell out his block of shares if the investment did not meet the expectations. After testing the hypothesis on the array of the companies from the S&P100 index over the period of the 1990s the authors come to the conclusion that only independent, pressure-resistant investors contribute to the growth of the company value; dependent investors are more worried about maintaining and strengthening business relationships and cannot be trusted monitoring agents. Adhering to the aforementioned classification, Bhattacharya and Graham (2009) got different results for Finnish companies for the period of 1993-2000. They reveal that the best performing companies are those with an approximately equal distribution of proportion of both types of investors but not those with the dominance of the “independent group”, and thus the direct relationship of ownership concentration with performance was not confirmed.

The article of Woidtke (2002) also considers institutional investors as a heterogeneous group. Pension funds, being in the spotlights, are divided into private and public. Examining the sample of companies from the S&P500 index for 1989-1993 the author concludes that only private pension funds contribute to the value growth of the firms, while the public ones’ ownership leads to the opposite result. The author suggests an interesting explanation: in private funds the managers’ remuneration system is structured in such a way that they are interested in the maximum return on investment. Public pension funds tend to offer lower salaries and bonuses for executives that is reflected both in their qualification and incentives. In addition, if a fund is owned and managed by the government it focuses not so much on the profitability but rather on the solution of political or social challenges. This situation generates conflicts between

(15)

15

the shareholders which often come out to the public. The result is easy to predict: the portfolio companies’ value goes down.

Another example of the negative impact of institutional investors on the firm value is found in the work of Al-Najjar (2015) about the tourism industry of Jordan. Having mentioned the low quality of corporate governance inherent to the companies from his data sample, the author suggests that institutional investors are opportunistic and tend not so much to control the work of the management team but to negotiate with them to extract the maximum benefit from this “cooperation”. Such a motive is the stronger, the greater block of shares is owned by the institutional investor. Testing on the data for 2005-2012, the author confirms the correctness of his hypotheses. However, it was found that mutual funds and foreign investors are not aligned with this rule; for them the dependence was not revealed.

Another factor related to institutional investors which affects portfolio companies is the stability of their investors’ ownership. The study of this question was conducted in the article of Elyasiani and Jia (2009). The authors focus on such factors as the time variation of the total company’s ownership by institutional investors and the period during which a specific investor owns shares. It is argued that the longer the investor has a stake in the company, the better he gets to know the situation from the inside thereby reducing information asymmetry and agency costs, and the greater are incentives to monitoring. On the other hand, long-term investments allow to focus on strategic results which are not taken into account in short-term horizons. Finally, institutional owners with stable presence can have a severe impact on corporate governance, changing the management compensation system to the one that would encourage them to focus on the long-term performance of the company. The assumptions made were fully confirmed based on the sample of more than 1500 companies over the period of 1992-2004. Thus, the ownership stability of institutional investors increases firm value.

Let us now consider in more detail the influence on portfolio companies of such a category of institutional owners as private equity investors. In this thesis, the most important aspect is their influence on firm performance. The article of Guo and Jiang (2012) is dedicated to the Chinese market. The authors use the sample of 1998-2007 data of 2500 companies. After analysing a number of indicators such as sales growth, ROE, ROS, labour productivity, the authors conclude that firms which were invested by PE funds demonstrate better results. Comparing figures for the same companies

(16)

16

before PE investing and thereafter, one can mention significant improvement. In another work published in 2013 Acharya, Gottschalg, Hahn, and Kehoe analyse 395 transactions of 37 major European PE funds made in 1991-2007. Examining the growth of EBITDA margin in the period of PE investors’ ownership and EV/EBITDA after their exit, the authors found that private equity investors contribute to creation of the excess return and consequently additional increase of the firm value. Special attention is paid to the contribution of the human capital of PE investors. Most general partners of funds from the sample were former consultants or top managers with extensive experience in related industries.

The importance of “immersion” of a PE investor in operations of portfolio companies is the subject of the article by Landau (2014). It examines the role of investment funds’ representatives in the Boards of Directors. The author draws attention to the fact that it is not confined only to the managers’ supervision (so-called “monitoring approach”) and should be realised in the form of consulting, analytical, and expert support – so-called “resource-based approach” according to which a PE investor brings a certain intellectual capital in the company. After analysing the data obtained through a survey of CEOs and CFOs of 315 European companies which were subjects to buyout deals, the author argues that the contribution of monitoring to value creation is comparable to competence, management skills, and experience provided to the company by the Board of Directors. It is worth noting an earlier work which presents a different point of view about the influence of PE investors on portfolio companies. Burton, Filatotchev, Chanine, Wright (2010) examined 224 companies over the period of 1996-2002, half of them were British, and another half were French. The reason for choosing these countries is the difference in legal systems – case law in the UK and civil law in France – as well as the different level of development of venture capital industries. The authors suggested that this factor should affect the strategy used by investors as well as their financial results. Additionally, they subdivided the sample depending on whether the investor was a VC fund or a business angel. These categories of shareholders have different investment horizons, sources of funds, and sometimes even different goals of investment which should generate dissimilar agency problems. For example, VC investors are themselves agents in relation to their “depositors” and have obligations regarding payout time as well as the minimum level of profitability. Violation of these commitments will affect the reputation of the team that runs the fund and they will encounter problems with attracting funding for next rounds which is definitely not in

(17)

17

their interests. At the same time, business angels do not have such restrictions. They invest their own money at their own discretion, do not report to anyone and, thus, are more “patient” owners. The authors conclude that although the ownership of PE investors and the size of their stake in general is positively correlated with efficiency, the relationship is stronger if the investor is a business angel.

We can make some generalizations concerning the influence of institutional investors on portfolio companies:

1. Institutional investors are capable of monitoring the management and having a positive impact on corporate performance only when they are pressure-resistant.

2. Institutional investors’ ownership is usually a characteristic of efficient companies, however, the effect on performance can be negative in the case of opportunism or government affiliation (e.g., a state pension fund).

3. When a private equity investor becomes a shareholder of a company, the firm efficiency goes up which is reflected in the increase of its market value and better financial results.

Private equity investors and their impact on portfolio companies

A significant block of research in the academic literature is dedicated to initial public offerings (IPOs) of companies with PE investors among initial shareholders. Does a firm doing an IPO meet larger or smaller increase in stock price in the first day of trading compared to its peers which were not backed by PE? Is the cumulative abnormal return (CAR) observed after 1-2 years of trading positive? A sharp soaring in the first trading day indicates a significant underestimation of the offering value of the company. According to the signal theory, the share ownership of private equity investors is a sign of high quality, a kind of certification indicating that the company has an estimate close to the intrinsic value. However, the strength of such a signal largely depends on a specific PE-investor. The factors that determine the quality of this signal and its impact on the market are examined in the paper by Hopkins and Ross (2013). Based on the results of the survey of Australian financial buy-side analysts conducted in 2008 the authors found that for market participants the main priority is the reputation of the private equity investor which is perceived subjectively in some cases. The second

(18)

18

priority is the stake retained by the investor after the IPO, and finally the involvement in the governance and period of ownership. The survey showed that the market pays attention to the retained stake not only because of the effect of monitoring but also considering this fact as a sign of stability of the financial indicators, i.e. the PE investor is confident about the prospects of the company. In the study of Hearn (2013) the research object is 86 companies from the Maghreb conducted IPOs between 2000 and 2013. Local and foreign PE investors are addressed separately. It was found that the participation of “foreigners” is perceived by the market as a positive signal that reduces the underpricing of portfolio companies while domestic investors are correlated with greater consequent revaluation. The author interprets this finding as the feature of the economies of the Middle East, weak development of the institutions, and the constraints faced by local PE investors which tend to be closely associated with local elites and business groups and do not have proper incentives to execute monitoring functions. At the same time, foreign investors are relatively independent and resilient to such pressure.

Minardi, Ferrari, and AraujoTavares (2012) examine Brazilian companies which conducted IPOs in 2004-2008 – a total of 108 firms. The object of the study is the cumulative abnormal return (CAR) over the period of 1 calendar year. The results are consistent with those obtained in the above-mentioned articles. CAR was found to be higher for PE-backed companies than for those which had never had investments from private equity. Because of the global financial crisis, the authors further subdivided the considered time period into 2 parts – 2004-2006 and 2007-2008. Eventually, they came to conclusion that in both subperiods firms with private equity investors demonstrated higher return, although the 2008 crisis affected all selected companies. The authors also draw attention to a sustainable and positive correlation of performance of portfolio companies with the value of the intellectual capital provided by a PE investor.

Having found out what is the influence of private equity investors on the financial performance of portfolio companies, let us now consider in more detail the reasons for such changes. As mentioned before, PE/VC investors forming the portfolio pick up companies which are able to achieve significant growth in a short period of time, usually 3-5 years. Venture capital investors and business angels, in principle, invest only in innovative enterprises with the highest business risk. It is clear that to achieve the desired return on capital the firm has to spend a lot of money on R&D projects aiming to develop innovations and improve the quality and competitiveness of its

(19)

19

products. On the other hand, traditional debt financing is not appropriate for R&D projects due to a high uncertainty of the results. The required funding can be obtained only in the form of equity capital from investors willing to take substantial risks. Hence we come to the idea that the emergence of venture capitalists has a positive effect on the innovative activity of the enterprise.

Empirical confirmation of this hypothesis can be found in the paper by Cumming and Johan (2014). Examining a number of Australian companies, the authors conclude that the share purchase made by VC funds leads to an increase in R&D spending and the number patents registered by the firm. Popov and Roosenboom (2012) based on a broader sample of panel data over 1991-2005 in 21 European countries obtained similar results. The difference, however, is that this relationship is statistically significant only in countries with highly developed VC industry. The best result in terms of innovations’ development is demonstrated in the countries with minimum entrepreneurship barriers, favourable conditions for doing business, and relatively low taxes. The study by Brown and Floros (2012) also based on an extensive sample from 1995-2008 confirms the existence of a direct relationship between innovation activity and access to financing from private equity investors. At the same time, no difference depending on the type of investor was revealed. The authors also make an assumption that companies which increased innovative activity are more successful in the long-term.

Research in this direction is continued by the article of Guo and Jiang (2012). The intensity of innovation activity is measured as the ratio of R&D expenditures to revenues. As found in previous studies, a VC investor leads to the growth of innovation activity. Completing the topic, we consider the article by Faria and Barbosa (2014). Their sample includes data from 17 EU countries in 2000-2009. Sticking to the number of patent applications filed by firm, the authors conclude that VC investors do have a positive effect on innovative activity, however it is realised at later stages of development of portfolio companies. This confirms the hypothesis that the role of VC investors is more in commercialization of already developed prototypes of innovative products but not in their creation.

We find the study of another occurrence of the influence of PE investors on their investees in the article by Gemson, Gautami, and Rajan (2012). The authors analysed more than 2800 of infrastructure projects realized in the period between 1990 and 2009 in 83 developing and 34 developed countries. Splitting the sample into two parts from

(20)

20

1990 to 1999 and from 2000 to 2009, one can see that since the beginning of 2000s the number of projects with the involvement of PE investors has grown more than three times. The authors conclude that PE investors often fund projects in developed countries and such projects are larger in scale and attract more co-investors.

One of the private equity research areas is the role of screening (selection of portfolio companies) compared to value creation activity. As far as PE-backed firms are indeed more financially successful, is it the result of PE investors’ involvement in running the company, i.e. a kind of “value added investments”, or the success secret is in the careful selection of portfolio companies? In their work, Croce, Marti, and Murtinu (2013) review a number of earlier articles, most of which stick to the position of value creation. The authors draw attention to the VC investors’ possession of a specific experience on “cultivation” of innovative companies and professional network that help in business development of portfolio companies. Nearly 700 companies from 6 European countries which received financing from VC funds in the period 1995-2004 were selected for empirical testing of the hypothesis. The authors conclude that the main reason of value increase and success of portfolio companies is the non-financial activity of the investor associated with implementation of improvements in the company operations but not the selection of the best. The fact that firms received VC financing demonstrate better results was confirmed in the aforementioned work of Guo and Jiang (2012) which is based on the Chinese market, but additionally the researchers found that the major role belongs to the selection “component”. It can be seen from the examined data sample that investment is usually provided to those companies that managed to achieve outstanding performance in previous periods. As for the ex post effects, it is important whether the venture capital is of foreign origin or not. In the latter case, the effect of value creation was not observed.

One of the peculiarities of venture capital investors is the limited period of ownership. Having invested in the selected company, such an owner waits for achieving the required rate of return on the investment and then sells his share and finds another perspective firm. Thus, in addition to the role of value creation for portfolio companies, it is interesting to know the duration, or durability, of this positive effect. Croce, Marti, and Murtinu (2013) revealed the existence of the so-called “imprinting effect”, which refers to the fact the competence introduced by the investor as well as the productivity augment do not disappear with his exit and retain for at least two years. The study of this issue can also be found in the article by Melesa, Monferra, and Verdoliva (2014).

(21)

21

The authors confirmed that portfolio companies keep their “superiority” after the exit of PE investors, however, the effect is long-term only if the investor was a venture capitalist. Another important observation was the inverse U-shaped relationship between the holding period and the subsequent operating efficiency. Thus, too lasting ownership of a PE investor can be a signal of the investee’s low quality. If a PE investor is a part of a bank conglomerate, the “superiority” effect turns out to be particularly long-lasting. This is because the bank is interested in the long-term growth of the company which will provide the synergy with its core business. Finally, the increase in performance is less significant and more short-term if the company becomes public after the exit of PE investor.

A variety of transactions involving private equity investors are leveraged buyout (LBO) and management buyout (MBO). The question of whether the management intending to repurchase the company call PE investors was raised in the paper by Fidrmuc, Palandri, Roosenboom, and van Dijk (2013). The authors discovered that it was the last resort: the situation when the management team had neither enough own funds nor a chance to borrow money. After analysing a sample of 221 British firms over the period of 1997-2003, the authors draw attention to the fact that the efficiency of the companies increased both after MBO transactions as well as after “normal” PE investments. In the second case the growth was evident rather quickly but was delayed in the first one. For this reason, the authors conclude that in MBO transactions private equity investors are only in the role of funds supplier and do not intervene in managing the portfolio company. Another variety of transactions, LBO, are covered in the work of Sannajust, Arouri, and Chevalier (2015) which is performed on the data for Latin American companies in 2000-2008. The authors analyse the performance of companies after the leveraged buyout. As far companies are delisted after such transactions, one can observe only indicators from the accounting reports but not the market price. It was found that there is an increase of ROA and profit per employee after LBO. The authors also draw attention to the positive relationship between efficiency augment and ownership concentration.

Discussing the topic of MBO and LBO transactions, one should consider their implications from the point of view of financial instability (bankruptcy). Transactions of both types lead to a significant increase of financial leverage that can achieve the ratio of 9:1, which increases the likelihood of insolvency. In the research conducted in 2012, Cressy and Farag analysed about 200 British companies which faced financial

(22)

22

problems in 1995-2008, and half of them had PE owners. The results showed that firms with PE owners were able to recover twice as often as those which did not have such shareholders. Tykvova and Borell (2012) came to similar conclusions. Their sample covers firms from 15 European countries over the period of 2000-2008. The authors infer that buyout companies are less likely to face bankruptcy and the more experienced PE investor participated in the transaction, the less is the probability of such negative events.

Summarizing studies about the impact of PE investors on the portfolio companies, we emphasize the following:

1. PE investor’s ownership is a sign of company’s high quality which ultimately reduces the underpricing during the IPO. The positive signal strength significantly depends on the investor’s reputation.

2. Investments from PE/VC funds positively affect innovation activity of the firm leading to higher R&D expenditures and the growth in the number of patents. 3. High performance of PE investors’ portfolio companies is the joint result of

careful screening (selection of investees) and value creation (functions of monitoring combined with providing intellectual capital).

4. Despite the fact that a number of transactions involving private equity is conducted with a very high financial leverage, the risk of financial instability for companies with PE investors is even lower than without them.

Overall conclusions from the literature review

This chapter presents an overview of the academic literature dedicated to studying the influence of ownership structure on financial performance and business efficiency as well as peculiarities of the company's operations. Two other components of the financial architecture, capital structure and corporate governance, are also discussed. We can highlight the following key conclusions:

1. The main reason why ownership structure matters is the agency conflict between the firm owners and the managers which creates the need for the shareholders’ control of the management team. The ability of control, incentives,

(23)

23

as well as the strategy and motives of the owners vary greatly depending on their type, origin, and the size of their stake.

2. The most efficient owners are institutional investors that do not have business ties with the investee. The worst performance, with some exceptions, is demonstrated by state-owned companies. The management turns out to be an efficient owner until insider ownership reaches a breakeven point.

3. Receiving investments from private equity becomes the cause of major changes in the firm – business processes reorganization, improvement of corporate governance, increase in innovation activity. Finally, this leads to a significant augment in the efficiency of the company.

4. PE investors’ ownership has an additional, purely signalling role as far as the market interprets this fact as the confirmation of the company’s high quality. 5. Among the variety of companies characterized by different combinations of the

components of the financial architecture certain groups, or clusters, of firms with similar parameters and performance can be distinguished.

In the next chapter we will consider the construction of econometric model and the formulation of hypotheses to test on historical data.

(24)

24

2. ECONOMETRIC MODEL FOR STUDYING PERFORMANCE OF

PORTFOLIO COMPANIES

In this thesis an econometric model is used to identify statistically significant relationships between performance of companies and the ownership of private equity investors. The current chapter is dedicated to choosing of explained, explanatory, and control variables, designing the model, and formulating hypotheses to be verified on historical data.

Choosing variables to characterize performance of portfolio companies

The authors of the studies mentioned in the literature review (Chapter 1) use a variety of dependent variables designed to identify the functional link with the factors of ownership structure. They can be divided into two large groups:

1. Accounting measures. These include return on assets (ROA), return on equity (ROE), return on sales (ROS), EBITDA margin, sales growth, free cash flow (FCF). Undoubted convenience of this group of indicators is the simplicity of use because they can be easily obtained from accounting reports. The main drawback – distortions and conventions related to the rules of accounting and preparation of financial statements. This includes, for example, accruals and non-cash expenditures – depreciation, reserves, and provisions. In addition, applied accounting standards may vary over countries, so one company may demonstrate different financial when using, for example, IFRS and US GAAP. To obtain comparable results, it is highly desirable to use figures calculated using the same accounting standards which is not always possible.

2. Market measures. First of all, it is the Tobin's Q ratio, calculated by dividing market capitalization over book value of assets. The higher is the score, the more optimistic the market is about this firm. Another market measure is the total shareholder’s return (TSR) defined as the sum of received dividends and appreciation of stock price over a specific time period (e.g., one year since IPO). The cumulative abnormal return (CAR) equal to the sum of excess of actual return over a benchmark is also used by scholars. Unlike accounting measures, market indicators are much less prone to distortion and are more objective.

(25)

25

Their main disadvantage – the firm may not be listed, so there is no market price for its stocks, or may be illiquid, then the price will be volatile and will not reflect the real value of the company.

Now let us consider which of the above-mentioned variables are more appropriate for using in this study. The main research object is companies that received funding from private equity investors at various stages as well as those which became targets in LBO/MBO transactions. Usually, only companies which are expected to quickly generate a significant (multiple) increase of market value are spotted by PE investors. Such firms operate in high-risk industries – such as HiTech. We can make several assumptions:

 For most of the companies which will be included in the experimental sample, the largest part of the market value will be attributed not to the assets on the balance sheet (fixed assets, inventory, etc.) but to the intellectual capital and intangible assets. Highly qualified personnel, patents, trademarks, inventions, profitable long-term contracts, and loyal customers appear at the forefront but not buildings or equipment. In this situation, the book value of assets may be substantially underestimated, and as a result such an indicator as ROA is non-representational, it does not show the real picture, so using it in the model is impractical.

 Return on equity (ROE) seems to be more adequate because the value of equity capital is combined of owners’ initial contributions (usually made in cash) and retained earnings, so this figure takes into account financial resources which are actually tied in the business. ROE is calculated as follows:

Equity

NI

ROE 

,

where

NI

is net income and

Equity

is the book value of equity. However, private equity transactions are often legally structured in the form of providing a loan and appear in the balance sheet as debt instead of equity, consequently, the ROE may be very distorted.

 Another accounting measure, return on sales (ROS), could also be of interest. Although due to the peculiarities of accounting rules it is also a subject to distortions, it reflects the operating costs control. A decrease in ROS can be the

(26)

26

consequence of an unjustified increase of administrative costs or deterioration of the production technology that leads to increased COGS. Here is the formula:

Sales

EBIT

ROS 

,

where

EBIT

is operating profit and

Sales

is the period revenue. The problem, however, is that, as revealed in several studies, the presence of PE investors in the capital increases the intensity of innovation activities, hence, increasing R&D expenditures. Capitalization of such costs is prohibited by accounting rules, therefore, they would reduce operating profit. This fact may be interpreted as that private equity investors are the reason of profitability reduction.

Now let us consider the next class of metrics – market measures:

 The vast majority of analysed studies dedicated both to developed and developing markets, use as one of business efficiency measures Tobin’s-Q ratio. This metric reflects market expectations regarding the prospects of the company. Guided by the efficient market hypothesis one can assume that the observed market price should reflect the firm’s intrinsic value, at least on average. The following formula is used:

Assets

EV

Q 

,

where

EV

(enterprise value) is the sum of the market capitalization and debt,

Assets

is the book value of assets.

 Another market measure, TSR, instead of expectations, reflects the accomplished fact – the increment of the shareholders’ wealth, expressed in the achieved increase in the value of shares (capital gain) and payouts, usually in the form of dividends. TSR is calculated as follows:

Price

Payouts

Price

TSR







,

where



Pr

ice

is the stock price increment during the considered period,

Payouts

for cash inflows during the same time, and

Price

is the price at the beginning of the period.

A common drawback of both indicators is their absolute nature, i.e. the obtained values are not compared with any benchmarks. For example, should a realized profitability of

(27)

27

11% per annum be treated as a good investment result? If so, is that applicable to all companies? A more appropriate approach allowing to compare the performance of companies with the peers and with the market is the analysis based on the cumulative abnormal return (CAR). A “normal” return which is determined based on one of the market models like CAPM is calculated for each company; the excess of actual return over the expected one is computed afterwards. This method “normalizes” the return given the difference in the levels of risk and as a result makes it possible to draw a conclusion whether some companies managed to outperform the market or not.

Regression model, independent and control variables

The next important issue that must be addressed when designing an econometric model is the choice of the independent, or explanatory, and control variables. In this work, as may be concluded from the topic, independent variables should characterize the ownership structure. Based on the reviewed studies we can select the two following independent variables which are expected to influence the performance of the firm:

 Retained ownership. Even after the exit from the investment project and the sale to third parties of the most of its shares, a PE investor tends to keep a (small) stake. There may be several reason for this strategy; the literature focuses on the signal effect of this behaviour.

 Time before the exit of PE investors, or, in other words, the period during which they were among the shareholders. Given the significant contribution of PE investors in the company's transformation and improvement, we can assume that the aptitude of the changes will depend on the duration of their ownership. Furthermore, because various additional factors inside the firm may influence its performance, we should add several control variables. A number of firm-specific variables are commonly used in the analysis of financial performance as controls (e.g., Berger and Ofek, 1995; Dushnitsky and Lenox, 2006).

 Firm size is calculated as the natural logarithm of firm’s market capitalization. As we know from the literature, “growth” companies and “value” companies tend to have different return patterns and are usually treated separately (e.g., Fama-French three-factor model specifically distinguishes among them).

(28)

28

 Firm age is calculated as the natural logarithm of the period of time since its foundation. Performance (financial results) of a company strongly depends on the stage of its life cycle, so adding this variable is important.

 Firm’s financial structure (leverage) is captured by the capital ratio calculated as the ratio of its debt to total assets.

 Firm productivity, measured as labour productivity, is the ratio between the logarithm of sales and the number of employees.

 Capital productivity is the logarithm of sales divided by fixed assets.

Differences may be found when accounting for industries in which companies operate. As far as each company has its assigned SIC-code, one can easily employ this classification. At the same time, industry is a nominative characteristic opposed to the quantitative ones, and it would be better to somehow split the original data sample into subsamples and compare the results.

Summarising, corporate performance will be analysed using the following regression equation: it it oductivity Cap it oductivity Lab it CapRatio it Age it MV it t YearsToExi it e tainedStak it CAR                   Pr * 7 Pr * 6 * 5 * 4 * 3 * 2 Re * 1

Here is the short description of the variables which will be used in the regression:  CAR – cumulated abnormal return for a specific company and number of years.  RetainedStake – residual ownership of a PE/VC investor for a specific company

and moment in time.

 YearsToExit – number of years during which PE investors owned shares of the company (time since inception and till the exit).

 MV – firm size calculated as the natural logarithm of its market capitalization at the same moment when the retained stake is taken.

 Age – firm age calculated as the natural logarithm of the period since the foundation and till the same moment when the retained stake is taken.

 CapRatio – firm’s debt divided to total assets. Both values are related to the same moment in time when the retained stake is taken.

(29)

29

 LabProductivity – firm’s labour productivity. The natural logarithm of revenues divided by the number of employees. Both values are related to the same moment in time when the retained stake is taken.

 CapProductivity – firm’s capital productivity. The natural logarithm of revenues divided by fixed assets. Both values are related to the same moment in time when the retained stake is taken.

Econometric hypotheses

The last step before running the regression analysis is the formal stating of the hypotheses – predictions which are to be verified based on the collected historical data. Taking into consideration the reviewed literature, we can formulate the following hypotheses.

Hypothesis 1: residual ownership has a significant positive short-term effect on corporate performance.

Hypothesis 2: residual ownership has a significant positive long-term effect on corporate performance but the influence is lower (the coefficient is expected to be smaller compared to the one found during testing the Hypothesis 1).

Hypothesis 3: PE/VC investors’ ownership period has a significant positive effect on corporate performance.

Hypothesis 4: firm size has a significant positive effect on corporate performance. Hypothesis 5: positive and negative effects discovered while testing Hypotheses 1-4 remain stable after splitting the sample into two subsamples depending on firm’s industry.

(30)

30

3. EMPIRICAL ANALYSIS OF HISTORICAL DATA

This chapter is dedicated to the key stage of the research – performing the empirical analysis and testing the hypotheses. The sample was drawn from the universe of companies which did IPO in 2000-2009. The period embraces two significant economic turmoils – the “dotcom bubble” and the financial crisis of 2007-2008. In total, the sample is made up of 278 companies from 125 different (sub)industries.

Descriptive statistics and general analysis

The selection of cumulated abnormal return (CAR) as a dependent variable for the regression model imposed certain limitations on the set of companies which could be used in the data sample. CAR is estimated based on market prices, so the companies which are picked up in must be publicly traded during the considered period. This makes impossible to use firms which, say, did IPO but became private again in the following up to three years. The same problem was for PE-backed enterprises which experienced exits in the form of secondary buyouts, trade sales, or selling to a strategic investor.

The starting point was finding the list of IPOs which were done in the period under review. Several databases and scientific research tools were used to prepare the resulting set of data. The list of sources of information with some important remarks is presented below:

1. The primary fetch of PE-backed IPOs was retrieved from the Thomson ONE system. It was important to avoid including in the list those companies which did not have PE investors prior to IPO.

2. The next step was revealing CUSIP codes for the selected companies. CUSIPs are widely used in financial databases to identify companies but they were not provided in the primary fetch – tickers were used for identification. In the result of the step 2, companies without assigned CUSIPs were eliminated from the draw, so all the remaining ones had such a code.

3. Standard Industrial Classification (SIC) codes were retrieved from the CRSP database. They were used to group the firms according to the industries which

(31)

31

they operate in. Although there are several hundreds of industries classified by the SIC system, they all may combined into 10 “supergroups” – Agriculture, Forestry and Fishing; Mining; Construction; Manufacturing; Transportation, Communications, Electric, Gas and Sanitary service; Wholesale Trade; Retail Trade; Finance, Insurance and Real Estate; Services; Public Administration. As far more than 42% of the companies in the sample are attributed to the Manufacturing industries, and 32% are from the Services sector, it was decided to make only 2 categories – manufacturing and non-manufacturing firms. This aggregative approach will be used in the research, as far as there is no need for a very detailed distinction.

4. Information about ownership stakes was found in the section “Peer Analysis Momentum” of the Thomson ONE. It allows to setup filters and get specifically the list of PE/VC owners for a particular company and date (ends of quarters). For the purposes of this research, for each firm I collected data at the ends of its lockup period and 10 consequent quarters after the IPO year. This will help to observe the evolution of PE/VC ownership in their investees.

5. CARs were calculated by the Eventus system which is a part of the Wharton Research Data Services (WRDS). This is a professional tool for performing event study analysis. CARs were calculated for each of the firms for the windows (0; 12), (0; 24), (0; 36), (0; 48), and (0; 60). So, the subsequent analysis will be conducted based on the cumulated returns for 1, 2, 3, 4, and 5 years since the IPO.

6. The CRSP database also provided information about firms’ market capitalization. It is worth mentioning that the figures were collected for the last days of each quarters, so they are fully consistent with CARs and ownership data.

7. Data for calculating control variables, except for market capitalization, such as the number of employees, assets, fixed assets, debt, and revenues, were retrieved from the Compustat Annual Updates database.

The empirical analysis employs several regressions based on data for different time periods. Below one can find the descriptive statistics for all series used for testing in various combinations:

Corporate performance and exit strategy of PE investors