Risk and Return of Listed Private Equity

1

Risk and Return of Listed Private Equity

A Theoretical and Empirical Exploration of the Listed

Private Equity Market

Master Thesis by Pieter-Bob Houpt

*

Abstract

This paper investigates the risk and return characteristics of Listed Private Equity (LPE). An extensive theoretical framework is presented, in which a complete picture of relevant LPE topics is drawn. Based on the theoretical framework, the LPE market is further empirically explored, using a worldwide sample of 109 liquid LPE companies for the period June 1994 to December 2010. It is concluded that LPE is a very heterogeneous asset class, with large differences in performance between various types of LPE. The largest difference is that between buyout and venture capital investment styles. Whereas buyout LPEs show significant outperformance, venture capital‟s performance is dramatic. Overall, an equally-weighted (EW) portfolio strategy is to be preferred over a value-weighted (VW) strategy, which still exhibits large firm-specific risk and negative CAPM Alphas over the MSCI World Index. EW portfolios show on average positive Alphas, but they are not significant. Betas are considerably greater than 1, and correlation with other asset classes is higher than often assumed, especially in periods with low returns. This is also supported by the finding that LPE does not seem to have any unique exposure to macroeconomic risk factors, besides those incorporated by the market. LPE is positively related to the Size factor, whereas the influence of the Book-to-market factor depends on the type of LPE. Finally, I confirm earlier research, which found that LPE returns are higher in the first half of the year, compared to the second half.

JEL Codes: G11, G12, G24, G32

Keywords: Listed Private Equity, Risk and Return, Capital Asset Pricing Model, Risk Factors, Portfolio Strategy, Investment Styles

*

2

Table of contents

1. Introduction ... 3

1.1 Cause of study ... 3

1.2 Relevancy LPE ... 4

1.3 Means of research ... 4

2. Theoretical framework ... 6

2.1 Terms and Definitions ... 6

2.1.1 Definition of Private equity ... 6

2.1.2 Investment styles ... 8

2.1.3 Historical development of Private Equity ... 9

2.1.4 Why Private Equity? ... 11

2.1.5 Investing in traditional Unlisted Private Equity ... 12

2.1.6 Listed Private Equity... 14

2.2 Literature Review ... 19

2.2.1 Literature on Unlisted Private Equity ... 19

2.2.2 Literature on Listed Private Equity ... 20

2.3 Research focus and relevant variables ... 27

2.3.1 Overall risk and return ... 27

2.3.2 LPE behaviour through time ... 27

2.3.3 Heterogeneity within the LPE universe ... 28

2.3.4 Influence of macroeconomic global risk factors on LPE ... 30

2.3.5 Influence of Size and Book-to-market risk factors on LPE ... 31

3. Data ... 32

3.1 Origination of the database ... 32

3.2 Potential sample biases... 33

3.3 Sample and its characteristics ... 35

4. Methodology ... 39

4.1 Portfolio construction ... 39

4.2 Risk, return and correlation estimators ... 40

4.3 LPE behaviour through time ... 43

4.4 Global risk factors model ... 44

5. Empirical results ... 46

5.1 Risk and return overall LPE universe ... 46

5.1.1 Performance of LPE through overall time series and subperiods ... 46

5.1.2 Seasonal behaviour of overall LPE universe ... 50

5.2 Heterogeneity within LPE ... 51

5.3 LPE’s exposure to risk factors ... 55

5.3.1 Exposure of LPE to global macroeconomic risk factors ... 55

5.3.2 Exposure of LPE to Size and Book-to-market factors ... 57

3

1. Introduction

In this first part, the Listed Private Equity market will be introduced, along with a brief discussion on the relevancy and methods of this research. In section 1.1, the cause for this research is described, followed by the relevancy of LPE as a subject of research in section 1.2. Finally, the methods and structure of this study are presented in section 1.3.

1.1 Cause of study

Not even two years ago, the market for Private Equity firms listed on a stock exchange looked like it could default altogether. The Financial Times1 reported that shares of 3i (Europe biggest listed buy-out firm) had fallen below the flotation level of 14 years earlier and were trading at a 75% discount to the market estimates of the value of 3i‟s underlying assets. The Dutch financial newspaper, Financieel Dagblad, using again 3i as an example, estimated that surviving chances for the venture capital industry were slim2. Investors apparently believed that reported Net Asset Values (NAV) were meaningless and indeed a wave of write-downs was expected3. Although the Listed Private Equity (LPE) market was indeed hit hard (3i, for example, was forced to undertake an rights issue), the sector showed its resilience in the aftermath. Since the all-time low in the beginning of 2009, the LPE market bounced back and the LPX50 TR4 index reported a return of 41,3% for 2010, thereby outperforming traditional stock markets.

This rollercoaster ride of the rather unknown listed part of the very large global private equity (PE) universe, is a good starting point to explore this market further. LPE is the term used to describe all Private Equity firms that are listed on an exchange, as opposed to the Limited Partnership structure that the vast majority of PE funds exhibits. With a market capitalization of roughly $ 100 Billion for the LPE market covered by the LPX Group, it is just a small fraction of the Trillion Dollar unlisted PE (UPE) market. However, both from an investor and academic point of view, LPE possesses attractive characteristics.

1

3i shares fall below flotation level of 14 years ago, Financial Times, 16/12/2008. 2

Overlevingskansen durfkapitaal nemen sterk af, Het Financieele Dagblad, 14/01/2009. 3

Private Equity assets melting away, Financial Times, 02/02/2009 4

4

1.2 Relevancy LPE

Private equity is often believed to deliver high risk-adjusted returns, independent of the traditional stock market. This forms an attractive proposal to investors seeking exposure to excess returns independent of the market (Alpha). Investing through an UPE fund has some shortcomings though. Minimum committed investment amounts are often high, fixed investment horizons are long, performance measurement is difficult and cash flow patterns are unstable and unpredictable. Investing in LPE potentially solves all these issues, while still maintaining exposure to underlying PE portfolios.

In contrast to the lack of academic research attention the LPE market has received, there is a vast body of academic work on UPE. The lack of reliable market data on this asset class presents a major challenge for researchers that study risk and return characteristics. Reported results are ambiguous and hard to compare, due to the use of different datasets, performance measurements and assumptions (Phalippou, 2007). Since LPE firms are listed on an exchange, reliable data on returns and volatility can easily be obtained. Hence, researchers have the opportunity to apply standard financial risk and return techniques to measure the performance of LPE. That makes it surprising that the field of LPE is still largely under-researched (Mathonet and Meyer, 2008). Bauer et. al. (2001) were the first to study a large sample of 124 liquid LPEs. They find high overall risk-adjusted performance and low correlation to other asset classes, although heterogeneity among different investment styles is large. In a follow-up study, Bilo et. al. (2005) again find superior performance for 114 LPEs. However, correcting for severe biases caused by the relative illiquidity of LPEs stocks, has dramatic consequences for the return premium, which almost disappears. The third study to investigate a large sample of 274 LPEs is that of Lahr and Herschke (2009). They find no overall excess returns for LPE and a market beta higher than 1, but results vary amongst different organizational structures.

1.3 Means of research

5

only be important for a better understanding of the LPE universe, but also provide insights in the risk and return characteristics of private equity in general (multiple earlier studies, including that of Huss, 2005 and Jegadeesh et. al., 2009, find that LPE can be used as a proxy for UPE).

This study is structured as follows. In part II, I present an extensive theoretical framework, in which I will define and describe all relevant subjects, provide a review of earlier literature and finally, will discuss the research focus and relevant variables. In part III, a description of the data and its characteristics is provided, followed by the methodology used to test the risk and return profiles in the LPE universe in part IV. The empirical results will be shown in part V. In part VI, the study will be concluded; implications and recommendations for further research and practice will be presented.

6

2. Theoretical framework

In the theoretical framework part, I will embed my research into an appropriate academic framework. I will start with defining and describing all relevant terms and definitions. This is followed by an overview of earlier academic research on the subject. Finally, I will determine the focus of this study and formulate a number of relevant variables that will be explored in this thesis.

2.1 Terms and Definitions

In this first section of the theoretical framework, I will define and describe the relevant terms and definitions on the subject of my research. I will start with a chapter on private equity in general and unlisted private equity in particular. This is followed by a description of listed private equity and finally, I will present an overview of the differences between LPE and UPE.

2.1.1 Definition of Private equity

Private equity is a very broad term for a very heterogeneous asset class. Multiple definitions for PE are given in existing literature. Bance (2004: 2) provides the broadest definition: “investing in securities through a negotiated process”. Smit (2002) states that what distinguishes PE from other equity is that PE cannot be sold in the financial markets. Weidig and Mathonet (2004: 6) use a similar definition5: “Private equity provides equity capital to enterprises not quoted on a stock market”. Bergmann et. al. (2010: 3) add that PE “provides long-term, committed share capital, to help unquoted companies grow and succeed”.

These definitions help explain the “Private” part in the term “Private Equity”, namely that PE stakes are not traded on a public financial market. However, they still don‟t tell much about the kind of investors and the type of private investments these investors engage in. The definition, thus, needs to be further explored to come up with a definition that is the most relevant to research on Listed Private Equity.

Achleitner and Klöckner (2005: 9) already provide a more comprehensive image: “PE is an asset class that focuses predominately on actively investing in and supporting businesses with the potential for high growth. These businesses are typically not listed on the stock market. The aim of investors is to help them grow and create value over several years by providing advice,

5

7

incentives, networking and knowledge through a range of specific investment structures. The private equity/venture capital firm makes investments through a negotiated process with a company or an entrepreneur in order to achieve growth objectives and returns on behalf of their investors. For the company or entrepreneur, the investment is made over a limited timeframe to help grow the business or to make it more efficient and, thus, to further its development”.

This definition makes clear that PE investors are active investors, who use a range of investment structures, and seek return in a limited timeframe. However, it is still not clear who these active investors are and what the “Equity” part stands for in the term “Private Equity”.

In my opinion, Bauer et. al (2001: 3) provide the most specific definition of PE that fits my research best: “Private equity investments are characterized by professionally managed investments in unregistered securities of mainly private companies, i.e. the securities are not traded on a public exchange market. The investments are predominantly in the form of equity, but also hybrid structures of debt and equity. Private equity managers usually acquire large ownership stakes and pursue an active role monitoring and advising portfolio companies. Thereby the investment horizon of the private equity managers is limited by an exit strategy with the aim of realizing a return on investment”.

Using this definition means that the “Equity” part in “Private Equity” does not stand exclusively for plain equity stakes, but can also include mezzanine financing6 (a relevant specification, as will be seen later in this study). Furthermore, this definition clearly states that investments are professionally managed, and thus effectively excludes the so-called Business Angels7. Finally, the limited investment horizon and the exit strategy are the characteristics that distinguish PE from business conglomerates (diversified holding companies).

6

Bergmann et. al. (2010) define mezzanine capital as “any capital between equity and debt e.g. subordinated debt, convertible debt or loans with equity kickers”.

7

8

2.1.2 Investment styles

In general, Private Equity can be divided into three investment styles, which do not differ between UPE and LPE. These investments styles are Buyouts, Venture capital and Growth capital. Consistent with Bergmann et. al. (2010) and the fact that Mezzanine financing can be observed across all the above investment styles, I do not regard mezzanine capital as an investment style, but rather as a financing style (besides regular equity).

Venture capital is, according to Bance (2004: 3) “the business of building business”. It focuses on companies that have undeveloped or developing products or revenue. Venture capital funds/firms invests in companies which haven‟t reached the economic state of maturity yet. Every earlier state (seed, start-up, expansion) of the economic life-cycle can be of interest to venture capital. Venture capital plays an important role in any economy, by acting as facilitator of growth and entrepreneurship. Typically, venture capital invests in start-up companies with possible high potential, but also with high uncertainty and low or negative current earnings and assets. As Huss (2005: 3) describes it: “venture capital provides capital to firms that otherwise might have difficulties in attracting finance”. As lending is almost impossible for these high-risk companies, venture capital financing is almost always through equity stakes. These equity stakes are sometimes being compared with out-of-the-money call options. To increase the chance of this call option ending up in-the-money, venture capital invests in a large number of companies and often supports companies and entrepreneurs with management support, not to be confused with management control.

Buyout funds/firms, in contrast to venture capital, focus on mature companies with low growth possibilities, but with steady cash flows. Buyout funds/firms typically provide large/majority stakes of equity capital in transactions in which companies or parts of companies (e.g. business unit) are acquired from the current shareholders. This is often done in a joint effort with the current management team, who also acquire equity stakes. In that case the term “management buyout” is used. If the buyout fund/firm works together with an outside management team, the term “management buy-in” is used.

9

To compensate for the moderate risks and thus moderate expected returns of buyout investments, they are typically financed with high levels of debt to optimize the cost of capital, hence the often used term “leveraged buyout”. High leverage is only possible because the acquired company generates large steady free cash flows8 to cover the interest payments, which in turn provide an attractive tax-shield. As Tolkamp (2007: 17) states: “The companies repay the debt used for their own acquisition”.

Growth capital is an investment style that provides minority equity capital to facilitate growth opportunities of a mature company. According to Bergmann et. al. (2010: 6), “growth opportunities include acquisitions, increasing production capacity, market or product development, turnaround opportunities, shareholder succession and change of ownership situations”. So in contrast to buyout capital, growth capital is not about taking control over a company, but is directed at providing capital to the existing company‟s growth opportunities. The minority equity stake that characterizes growth capital, means that divestments are less complicated and less predictable compared to buyout capital. This calls for a flexible variable investment horizon. According to Bergmann et. al. (2010: 6), LPE is more suited for these kind of flexible investment horizons, because “an evergreen LPE company is not obliged to distribute the money back to the limited partners after a specified period of time”.

2.1.3 Historical development of Private Equity

Although Private Equity didn‟t receive much attention from the general public until the buyout wave in the 80s, the industry has been around since the late 40s. In fact, 3i, which today is one of the largest PE firms in the world, was founded in 1945 by the UK government to facilitate growth by providing small start-up companies with long-term capital (Bance, 2004). Throughout the years, this venture capital industry displayed modest growth, but still consisted mainly of wealthy individuals and (semi-) government programs.

The status quo continued until the 70s, when governments in both Europe and the U.S. approved regulatory changes concerning investments in alternative asset classes by institutional investors such as banks and pension funds. This less strict investment legislation, combined with the introduction of the Limited Partnership9 structure, caused a massive growth of institutional investment activity in the PE asset class. This trend coincided with another trend: the

8

Free cash flow is the after-tax cash flow from operations available to all investors: debt holders and equity holders (Koller, Goedhart and Wessels, 2005).

9

10

conglomerate business model was becoming less and less efficient in the economic circumstances of the late 70‟s and 80‟s. Holding companies were looking to divest large business units unrelated to the core business (Smit, 2002).

The large amount of available capital and the large amount of mature companies seeking new owners led to a rapidly shifted focus from small venture capital investment to large buyouts. After the introduction of junk bonds10in the mid 80‟s, the leveraged buyout wave became even bigger. PE funds could now finance the acquisition of large companies, using only small equity stakes, and large amounts of debt (Smit, 2002). At the end of the 80‟s, the popularity of PE resulted in an imbalance of available funds and available investment opportunities. The competition between PE funds led to absurd prices of target companies. PE funds winning these kind of bidding contests were confronted with the „winners curse‟ phenomenon11

, ultimately leading to the rapid downfall of the leverage buyout industry in the early 90‟s.

From 1992-2001, the PE industry grew again. In contrast to the buyouts of the 80‟s, the focus of PE funds did not lie on restructuring mature companies, but shifted to companies exhibiting high growth opportunities. In the late 90‟s, the PE industry boomed as the high-tech industry offered very attractive investment opportunities. Ultimately, these attractive investment opportunities turned out to be the dot-com bubble and PE funds were forced to write-off many of their investments.

In the aftermath of the dot.com bubble, the PE industry experienced a few tough years. For example, the LPE industry did not show any real growth until the second half of 2003. As the trend towards globalization led to worldwide macroeconomic growth, PE became an attractive investment category again. The seemingly endless economic growth, combined with very easily attracted debt financing, made the PE industry highly profitable. However, at the end of 2007 it became apparent that worldwide high debt levels, were no longer sustainable. As it turned out, many companies and banks had overplayed their hands in their no limit debt game. Ultimately, this led to great distrust between banks and the complete dry up of available credit, better known as the financial crisis. The highly leveraged PE industry was hit hard, as refinancing old debt and attracting new debt for transactions became virtually impossible. This downfall was highly visible in the LPE industry, as LPE traded at huge discounts to their Net Asset Values (NAV). Shares of

10

Junk bonds arebonds that are rated below investment grade at the time of purchase. These bonds have a higher risk of default or other adverse credit events, but typically pay higher yields than better quality bonds in order to make them attractive to investors. These kind of bonds are important for LBOs to attract enough debt financing (besides securitized bank loans).

11

An excellent and famous example of the beginning of the end of the leveraged buyout wave is the 1989 acquisition of RJR Nabisco by KKR for $25 Billion, excellently described in the book Barbarians at the

11

3i (Europe biggest listed buy-out firm), for example, fell below their flotation level of 14 years earlier and were trading at a 75% discount at the end of 200812.

Although many thought that the financial crisis would lead to the total collapse of PE, the industry showed its resilience in the last couple of years. Although new debt financing was and is still not easily attracted, underlying portfolio companies mostly survived the tough years rather well, thus validating net asset values of PE firms. In a study on behalf of the Private Equity Growth Capital Council13, Thomas (2010: 1) even shows that “during the “Great Recession” of 2008-2009 private equity-backed businesses defaulted at less than one-half the rate of comparable companies”.

2.1.4 Why Private Equity?

The existence and increased popularity of private equity financing over the years, is worth exploring further. Seen from the (institutional) investors who invest in PE, the popularity of PE is easy to understand: PE is believed to be an asset class with high historical performance and low correlation to other asset classes. As correlations between international markets have more than doubled during the 90‟s (Bilo, 2002), investors seek diversification in alternative assets. The high PE performance, independent of other markets (Alpha) provides institutional investors with a way to outperform their benchmarks.

But why does PE financing exist? As Smit (2002: 22) states: “this is a legitimate question”. In a world of perfect information and highly efficient financial markets, one would assume that all positive NPV projects can be easily financed. Public equity would then be the preferable choice, as it provides more liquidity than PE does. However, according to Smit (2002), there are two very fundamental real world bottlenecks that explain the existence of private equity:

1. In reality, there is no such thing as perfect financial markets. Information asymmetries between relevant parties in the financial market have serious consequences. Because the public has not the same inside information that management has, “bad” firms can pretend to be “good” and consequently may be overvalued. On the other hand, “good” firms may be perceived “bad” and consequently be undervalued. This latter situation is often a reason to take a firm private. As management has the inside information that their firm is “good”, but undervalued, they opt for a different financing structure: private equity.

12

Financial Times, December 16th, 2008. 13

12

Management, in return for higher valuation, provide PE funds with more in-depth (inside) information on their investments.

2. Agency problems between the public capital market and management limit managerial discipline. In the public equity markets, there is separation of ownership (shareholders) and control (management) of the company. This relationship is explained by Jensen and Meckling (1976), as a principal (shareholders) – agent (management) arrangement, in which the principal hires the agent to run the company on his behalf thereby delegating a certain degree of authority. As the non-perfect financial market doesn‟t provide enough discipline power, the agent is tempted to abuse his decision power and pursue his own objectives rather than those of the principal (moral hazard). This inefficiency in the management of the company is reflected in the stock price. This is where PE comes in. In contrast to the scattered public markets, PE typically acquires a large controlling stake in the target company, which gives PE the power to improve the management of the company. This is done in two ways. Firstly, PE typically aligns its own interest with that of management, by distributing equity to management, thus providing a high incentive for management to create value. Secondly, PE transactions are financed with relatively high levels of debt, which act as a disciplinary factor. Because interest payments are high, the free cash flow of a company is significantly reduced, thereby limiting the opportunities for the management to abuse company resources.

2.1.5 Investing in traditional Unlisted Private Equity

Ever since the 70‟s, the large majority of Private Equity funds are constructed as a limited partnership. Today, this is still by far the most common organizational structure in the PE universe.

Because of the usual large minimum investment amount of more than $ 5 million, investors in these limited partnerships are typically institutional investors like pension funds, banks and insurance companies. These investors are the Limited Partners (LP), who commit a certain amount of capital to buy an interest in the newly founded fund. The fund is being managed by a professional private equity firm, who is called the General Partner (GP). Funds usually have a target size in terms of commitment, and when this is reached, the fund effectively starts operations and is closed for further investments.

13

it “calls” the required portion of the committed capital from the LPs. When the investments are sold again in the second half of the fund‟s life, the cash proceedings are returned to the LPs. Sometimes, intermediate returns are earned through dividends of underlying portfolio companies. GPs are compensated in two ways. First, they charge the LPs a management fee of 1,5% – 2% over committed capital or the amount invested (note the important difference, especially over the first years when committed capital and amount invested differentiate enormously). Second, they receive a performance fee through a share of the profits, called “carried interest”. This can be computed over each investment separately or over the sum of all investments (again, note the important difference14). Typically, the carried interest amounts to 20% of the cash flows and is only paid out when a specified minimum return on investment is reached: the “hurdle rate” which is usually around 8%.

Although the limited partnership has been the dominant way of PE investing for decades, a number of serious shortcomings of the unlisted limited partnership structure can be identified. First, Brown and Kräusl (2010) state that diversification is very difficult for an LP to achieve. This is because limited partnerships are often very focused and invest in a limited number of companies. Moreover, the high minimum investment amount makes diversification through various funds a very expensive exercise only available for large institutions.

Second, Investors are faced with unpredictable cash flows. Although they have committed a certain portion of their portfolio to PE, that doesn‟t mean that this commitment is immediately called by the fund. Moreover, Brown and Kräusl (2010) state that the investment and divestment phase of a fund may overlap, effectively meaning that before the total commitment is invested, part of it is already being returned. These phenomena leave an investor with a smaller amount actually invested in PE than desired: the so-called “cash drag”. Investors try to minimize this cash drag by over-committing capital to the PE class. However, the unpredictable cash flows make it very difficult to determine and maintain the “right” level of over-commitment.

Third, Returns for LPs follow a so called “J-curve” pattern. During the first years of the fund, management fees and early write-downs of bad investments cause negative returns for the investors. Only a after around 6 years, investors “break-even” (Toledano, 2006). For pension funds, this phenomenon poses a difficulty in matching their investments with obligations.

Fourth, Performance measurement is only reliable when all investments returns are realized at the end of the fund‟s life. Ex ante, an investor can only judge a GP‟s prior performance by the

14

14

stated IRR in the prospectus, which can be easily manipulated, does not take the timing of cash flows into account and questionably assumes that intermediate cash flows are reinvested at the same rate of return (Huss, 2005). Comparison between funds is therefore difficult and time-consuming. During the fund‟s life, GPs do provide an interim estimate of the value of the portfolio, but this arbitrary estimate doesn‟t necessarily reflect the true value.

Fifth and finally, PE investments are per definition illiquid. Investors in limited partnerships commit their capital to a fund for a fixed amount of time and exiting before the agreed date is impossible or heavily penalized by the GP. Investments in PE can be sold in the secondary market, but according to Brown and Kräusl (2010: 5), this is a “relatively new phenomenon that cannot provide a guaranteed exit strategy prior to the natural termination of the fund”.

According to Brown and Kräusl (2010), unlisted funds of funds (FoFs) can eliminate some of the above issues, because they invest in a number of PE funds with different investment styles, vintage years and sector and geographic focus. By investing in these FoFs, investors automatically diversify and effects of cash drags and J-curve returns will be smoothed. However, issues with liquidity and performance measurement are still very much present, and moreover, a FoF charges additional management fees on top of the fees charged by the underlying funds‟ GPs. This creates a double fee structure.

2.1.6 Listed Private Equity

In comparison with its unlisted counterpart, listed private equity (LPE) is still a fairly new, unknown and small asset class to investors. Bilo (2002: 36) classifies vehicles as LPE, “if the underlying business is PE investing, but the funds themselves are quoted on an exchange”. Although LPE did roughly experience the same historical development as UPE, the scale of this development has been a fraction of that of UPE. The buyout wave of the 80‟s and the growth period leading to the dot.com bubble resulted in a steady but not explosive growth of LPE. However, since the second half of 2003 the growth of the asset class accelerated with a wave of new listings in 2006 and 2007. New listings came to an abrupt end when the financial crisis began and LPEs were hit hard. Today, investors can choose from 109 liquid LPE companies (sample used in this research, based on earlier work of Bergmann et. al., 2010), and a several times larger number of less liquid LPE companies15. In terms of assets under management, the LPE class amounts to roughly 10% of the UPE universe. Market capitalization of the group of 109 liquid

15

15

LPE vehicles amounts to $ 86 billion as per December 2010, compared to a low of $ 24 billion in March 2009 and an all-time high of $ 121 billion in July 2007.

The relative unpopularity of the LPE market seems surprising, considering the fact that LPE does eliminate almost all of the shortcomings of the unlisted limited partnership structure. The fact that all LPE vehicles are traded at stock exchanges, means investing in this asset class can be very attractive.

First and most important is the advantage of LPE over UPE in providing liquidity for investors. Shares of LPEs can be bought and sold any time an investor wishes. This means that investors can actually use private equity for tactical asset allocation instead of only strategic asset allocation (Ibbotson Associates, 2007). Institutional investors, for instance, can fine tune their overall exposure to PE by buying and selling LPE stakes. Another advantage of daily trading is the fact that an LPE stock can often be bought with a discount to its Net Asset Value (NAV), in contrast to UPE where investors always buy at cost. Furthermore, performance measurement becomes more reliable and straightforward, as LPE prices can be observed on a daily basis (thus time-weighted) making all the statistical financial techniques available to PE. Second, diversification is easy to achieve without the need for high capital commitments. Every investor, even small ones, can simply buy a diversified portfolio of shares of a number of LPE companies, which provide them with exposure to hundreds of private companies across different industries, countries and investment styles. Third, investors can easily eliminate cash flow pattern issues that unlisted funds exhibit. LPEs do not have a fixed life, which lessens the J-curve effect (see Lahr and Kaserer, 2010 for evidence on that). More importantly, investors can buy shares of mature diversified LPE companies thereby avoiding any cash drag or J-curve returns. Brown and Kräusl (2010) state that new listings of LPEs often consist of already built mature UPE portfolios which are then sold to the LPE company right after its IPO. This avoids any cash drag or J-curve effect of a new vehicle. In 2008, an LPEQ Preqin16 survey showed that 58% of 100 interviewed institutional investors agreed that established LPEs are and attractive way of investing after the J-curve. Fourth, due to the fact that LPEs are listed on an exchange, they are bound to strict governance legislation and are obliged to regularly publish extensive information to investors and the general public. This leads to the LPE market being more transparent and its reported data being more reliable compared to the traditional limited partnership. Moreover, investments bank often publish their own research on public stocks, providing investors with valuable information (Bilo, 2002)

16

16

However, LPE does have some shortcomings of his own. Firstly, the fact that an LPE vehicle does not have a fixed life, means proceeds from divestments are usually retained to be reinvested again. As divestment and new investment do almost never take place at the same time, an LPE company can get stuck with a large amount of cash on its balance sheet and no opportunity to invest. These cash positions are a concern, because they dilute the performance of the fund (Bilo, 2002). Second, the fact that the information LPEs publish is in fact accessible for everyone, and companies do not wish to share commercial sensitivities, means that the depth of this information is limited. Hence, investors in LPE do not receive the same amount of inside information that an investor in UPE receives. (Brown and Kräusl, 2010). Finally, whereas liquidity may be higher than that of UPE, it is still not comparable to normal stocks, as several studies on LPE have shown17. Of course, a number of large LPE companies are exceptions, but overall liquidity is still an issue, posing difficulties for investors to buy or sell large stakes of equity. Together with the fact that LPE companies are exposed to financial market turbulence on a daily basis and with no delay, investors can easily get stuck with a stake that can only be sold for a very large NAV discount (Brown and Kräusl, 2010).

There are a few categorical factors that distinguish investing in the LPE market from investing in the unlisted universe, which is dominated by the limited partnership structure. The LPE universe is more heterogeneous in terms of legal structure. Over the years, earlier academic research has categorized LPE in different ways18. In this thesis, I will use the classification that Bergmann et. al. (2010) propose. In my opinion, this choice reflects the underlying value and risk drivers best. Moreover, the database on which my own empirical tests will be based, is constructed along the same categorization.

The first category cannot be found in the UPE universe, thus is unique for LPE. Investors in LPE have the opportunity to invest in a listed fund manager. Fund management companies do not actually invest in PE, but are the GPs of mainly unlisted funds. Although there are only a few of these present in the universe, The Blackstone Group and KKR are two of the best known PE companies in the world. The second category is based on the distinction between LPEs that invest direct or indirect in private equity. Thus funds that directly invest in portfolio companies or funds of funds (FoFs) that invest in a number of (typically unlisted) PE funds managed by others. The

17

See for example studies of Lahr and Herschke (2009), Bilo and Zimmermann (2005), Bauer et. al. (2001), which show that imposing moderate liquidity constraints already has a big impact on their sample sizes.

18

17

final two categories both consist of direct investment companies, but are divided by the style their investments are financed with. Direct private equity companies mainly provide equity financing to underlying portfolios companies. Direct private mezzanine companies mainly provide mezzanine financing to underlying portfolio companies.

18

Figure 1

19

2.2 Literature Review

In this section of the theoretical framework, I will describe earlier academic literature relevant to my thesis. I will start by shortly discussing a selection of earlier work on unlisted private equity. This selection is partly based on relevancy to my thesis and partly on the general importance to the field of UPE. I will continue with a discussion on the most relevant literature on LPE. After a table is presented that provides an overview, each paper will be discussed in more detail. Since the field of LPE is still fairly small and under researched, my goal was to draw a complete and in-depth picture of the entire academic universe regarding this asset class. Therefore, I also included a table and description of earlier academic papers that were less relevant to this thesis (Appendix A).

2.2.1 Literature on Unlisted Private Equity

In contrast to the academic field on LPE, UPE has been the subject of academic research fairly often. However, Phalippou (2007) states that most research focused on the relationship between the PE fund and its portfolio company. Only since the turn of this century, academics have switched to the LP side of PE, thus focusing on the performance of the fund. A few things have to be kept in mind when reviewing earlier work on UPE. As data on UPE is not publicly available, almost all researchers use different databases. These databases often suffer from selection biases, and require researchers to apply different sets of assumptions. Moreover, due to the illiquid nature of UPE, researchers cannot apply standard financial methods to study the risk and return profiles, which leads to heterogeneity of methods used. Results of the various academic studies are thus ambiguous and hard to compare. Below, I will discuss the most important papers regarding the risk and return profiles of UPE19.

Gompers and Lerner (2000) show that inflows of capital into venture funds increase the valuation of these funds‟ new investments. These higher values are not reflected in higher exit values. This is consistent with competition for a limited number of attractive investments being responsible for rising prices. Lungqvist and Richarson (2003b) confirm these findings. Diller and Kaserer (2009) report that European private equity returns are driven by fund inflows, unrelated to stock market returns and negatively correlated with the development economy as a whole. This is in line with the „money chasing deals‟ phenomenon that Gompers and Lerner introduced.

19

20

Ljungqvist and Richardson (2003) document that private equity generates excess returns on the order of five to eight percent per annum relative to the aggregate public equity market. Betas, computed by finding a matching public market company for the PE portfolio company, are around 1. Kaplan and Schoar (2005) find average fund returns (net of fees) approximately equal the S&P 500, although substantial heterogeneity across funds exists. PE performance is positively related to size, past performance and is procyclical. However, funds raised during boom times are less likely to raise follow-on funds, suggesting that these funds perform worse.

Cochrane (2005) developed a sophisticated method to measure the risk and return of venture capital gross of fees, by evaluating the performance of individual financing rounds. He shows that the main problem for measuring PE performance, is to overcome selection bias. By controlling for this bias, his reported results are lowered dramatically. He reports negative Alphas for log returns, but high arithmetic mean returns and Alphas.

Phalippou and Gottschalg (2009), show that performance estimates of previous research (e.g. that of Ljungqvist and Richardson) are overstated. Samples are biased towards better performing funds and values for non-exited investments are overstated. Correcting for these biases, they report a substantial 3% underperformance net of fees for PE. Gross of fees, they still report outperformance by 3%, which empathizes the importance of fee structures in computing PE returns.

Cumming and Walz (2009) study the reporting of performance by fund managers across many countries for a dataset containing mainly venture capital. They show that larger, syndicated funds, that use convertible securities are less inclined to overstate the value of their investments and perform better. Less stringent accounting standards and weak legal systems facilitate overstating of portfolio values by fund managers. This overvaluation lead tot reputational damage and discrepancy between reported NAV and stock price. Lerner and Schoar (2005) report higher gross of fees PE performance in common-law countries than civil-law and socialist countries.

2.2.2 Literature on Listed Private Equity

21

Table I

Chronologically presented overview of earlier academic work on LPE, most relevant to my research.

Year Author(s) Research Focus Data Methods Results

2001 *Bauer, Bilo and

Zimmermann

Risk and return of LPE

Weekly, 1986-2000, 124 LPEs

Equally-weighted portfolio, Sharpe ratio, Correlation matrix, CAPM with global risk factors (MSCI World)

Heterogenous, overall high performance, correlation with other assets low , sensitive to IPO volume credit and Ted spread, half year anomaly in returns 2005 *Bilo,

Christophers, Degosciu and Zimmermann

Risk and return of LPE

Weekly, 1986-2003, 114 LPEs

Equal + Value portfolios, Sharpe ratio, CAPM (MSCI World), Dimson model for autocorrelation

Heterogenous, overall high performance, after 2000 underperformance, lower performance after correcting for bid-ask spread and autocorrelation

2005 Huss Difference in

behaviour between UPE and LPE 849 UPEs vs LPX50 TR LPE index Public Market Equivalent approach

UPE and LPE behave almost identically, hence LPE can serve as proxy for UPE 2009 Jegadeesh, Kräussl and Pollet Risk and expected return of LPE funds and FoFs (proxy for UPE)

1994-2008, 26 FoFs and 129 funds

Event study for expected abnormal return and four-factor CAPM (MSCI World)

FoFs (UPE) expected small abnormal return, LPE funds 0, both betas of 1, positive on GDP and Size, negative on credit spread, LPE can predict UPE returns, thus can serve as proxy 2009 Lahr and

Herschke

Risk and return of LPE and its categories Weekly, 1986-2008, 274 LPEs Equal + Value portfolios, Sharpe ratio, International CAPM (MSCI World), Dimson model for autocorrelation

LPE overall no Alpha, beta between 1.2-1.7,

heterogeneity between the 4 LPE categories, internally managed LPEs have higher betas 2010 Bergmann, Christophers, Huss and Zimmermann Empirical overview and characteristics of LPE universe 122 LPEs, LPX50 TR index for empirical work Explorative research of LPE categories, mean returns and SDs, correlation matrix

Multiple categories identified, LPEs indeed mainly invest in PE, LPE outperforms market in boom times, underperforms in bad times, correlation with other assets not that low

2010 Phalippou Methods to measure risk and return in general PE

Weekly, 1993-2008, 19 Buyout LPEs

CAPM using Dimson multiple regression model to correct for autocorrelation

Heterogenous, average Alpha 6%, beta 1.5, first look at VC LPE suggests low performance 2010 Kaserer, Lahr, Liebhart and Mettler Time-varying risk of LPE

Same as Lahr and Herscke (2009)

Betas over rolling window, stabilty of betas by Pearson + Spearman correlations and Markov transition matrix

LPE betas are unstable, covariance with market increases with uncertainty, apparently LPE has some specific risk factors, hard to reliably predict future returns 2010 Brown and Kräussl Performance and characteristics of European LPE market

73 European LPEs Analysis of individual LPE's characteristics, and performance during credit crunch downmarket

Overview individual LPE's characteristics (fees, styles, leverage, NAV premia, dividends etc), and (dis)advantages of LPE compared to UPE, underperformance of LPE in recent downturn

22

Bauer, Bilo and Zimmermann (2001) were the first to investigate an extensive sample of LPE companies in the period 1986-2000. They find 229 LPEs worldwide. After imposing a set of liquidity constraints, they are left with a sample of 124 instruments. This sample is highly heterogonous in terms of size and investment style.

Using equally-weighted portfolios, they conclude that the performance of the LPE market as measured by the Sharpe ratio, is higher than that of other asset classes. However, these results vary significantly among the various investment styles. With a correlation of 0,3 - 0,5 to normal stocks, they state that LPE has indeed diversification potential that is not (or not only) related to illiquidity.

With regard to the influence of global risk factors on LPE returns, Bauer et. al. come to the following findings: LPE returns are significantly influenced by the stock market (NASDAQ) credit spread (business cycle risk), the TED spread20 (health and confidence in the financial system), and the trading volume on the global IPO market. They find that LPE reacts differently to these variables than do traditional stock markets, hence LPE should be regarded as a separate asset class. They also report an anomaly in the returns between the first and second half of the year, which could be related to the release of audited information on the value of the companies.

In a follow-up study, Bilo, Christophers, Degosciu and Zimmermann (2005) use roughly the same database. They now find 287 LPE companies in the period 1986-2003. After imposing slightly different liquidity constraints than in 2001, they come up with a sample of 114 liquid LPE stocks. In contrast to the 2001 paper, the authors now construct a value-weighted (buy and hold), and two equally weighted (one buy and hold, one fully rebalanced) portfolios to measure the performance of LPE. Hereby, they address the important weighing and rebalancing biases. In general, only the fully rebalanced equally-weighted portfolio shows superior risk-adjusted (Sharpe ratio and Alpha) performance compared to the MSCI World index over the whole sample period (results of the other portfolio strategies are dramatically affected by the downmarket after 2000). Betas of the equally-weighted portfolios (0.6-0.7) are significantly lower than that of the value-weighted portfolio (1.2). Because of the relative illiquidity of LPE compared to the regular stock market, the study also addressed three performance biases:

First, they state that LPE shows high artificial autocorrelation and thus volatility estimates are strongly downwards biased. Adjusting for this bias leads to significantly higher volatility, higher betas and lower Sharpe ratios. However, the fully rebalanced equally weighted portfolio still

20

23

outperforms the MSCI World index. Second, they also address the so-called bid-ask bias which significantly affects the rebalancing investment style. An investor implementing an rebalancing strategy is forced to buy at the ask price and sell at the bid price (not the observable arbitrarily closing price). Controlling for this bias leads to dramatic results: the return premium of the rebalancing strategy almost entirely disappears. Finally, the study controls for the survivorship bias by including eight non-surviving companies in their sample. Surprisingly and in contrast with literature, including these non-surviving companies leads to a slightly positive bias. They conclude that correcting for this bias, is not as important as correcting for autocorrelation and the bid-ask spread.

Huss (2005) studies the differences in behaviour of unlisted and listed PE over a long period of time. He uses a comprehensive dataset for the unlisted part of his research and uses the LPX-50 TR index21 as a proxy for the LPE market. This is a global value-weighted total return index, which covers the 50 largest LPE companies. He states that UPE‟s outperformance is mainly attributed to a lack of liquidity and transparency. As LPE doesn‟t possess those characteristics, he hypothesizes that LPE should tend to perform as public equity. Huss uses a Public Market Equivalent approach (PME) to compare the performance of listed and unlisted PE. The PME approach has the advantage over IRR, that it takes into account the actual timing of cash flows and assumes intermediate cash flows to be reinvested in the public benchmark instead of in the (closed-end) fund, which is far more feasible.

In contrast to his hypothesis, he finds that LPE behaves almost identically to UPE. He concludes that illiquidity is apparently not the main driver for PE performance. He claims that LPE can be used as a substitute for UPE, from both the investor‟s as the academic‟s point of view.

Jegadeesh, Kräussl and Pollet (2009) investigate the expected return (ex-ante) and risk characteristics of LPE funds and listed funds of funds (FoFs) that invest in UPE funds (as proxy for UPE). Instead of focusing on the ex-post returns of PE, they develop a model that predicts the ex-ante return of PE in the long run.

By calculating the difference between the market value (corrected for fees) and book value (after IPO) of a fund, they find that UPE, in contrast with earlier literature, is expected to have a small abnormal return of only 0% - 2% and listed PE is not expected to have any abnormal return.

21

24

Further risk calculations, using a

four-factor CAPM model by Carhart (1997)

22, show that both FoFs and LPEs have betas close to 1, and are positively related to the size factor (SMB) of Fama and French (1993). Furthermore, both LPE and FoFs are positively related to global GDP growth, implying that the increase in the existing investments values, offsets the increased competition between funds that lead to the so called “money chasing deals” phenomenon. Additionally, both FoFs as LPEs are negatively correlated with the credit spread. Finally, market returns of FoFs and LPE can predict future changes in the self-reported book values of UPE funds, thus support the so-called delayed adjustment hypothesis implying that performance of listed PE can be used as a proxy for future returns of UPE.

In 2009, Lahr and Herschke published their research on the performance LPE. Since Bilo et.al. (2005), this is the most comprehensive study on LPE to date. They built upon the work of Bilo et. al. and find 274 liquid LPE entities in the period from 1986 to 2008. Whereas Bilo et. al. investigate the LPE market as a whole, Lahr and Herschke create subsamples to investigate differences between the various organizations forms (internally or externally managed companies and single funds or funds of funds).

For all organizational forms, they construct 3 weekly rebalanced portfolios: one value-weighted and two equally-weighted (one adjusted for bid-ask spreads). They calculate Sharpe ratios, both unadjusted and adjusted for autocorrelation. They also compute Jensen‟s Alphas, using an international CAPM (correcting for currency risks) and a Dimson regression model (correcting for autocorrelation issues). MSCI World is chosen as the market index.

The LPE market as a whole shows no significant excess return and a Dimson‟s beta of 1.7 for the value weighted portfolio and 1.2 for the bid-ask spread adjusted equally-weighted portfolio. On an organizational level, Jensen‟s Alpha varies strongly between different portfolios, estimation periods and organizational forms. Internally managed companies show higher betas than do externally managed funds, which can be caused by riskier cash flow streams of the former (carried interest). This finding implies that LPE should not be used as proxy for UPE, without taking precautions.

In an empirical overview of the LPE market, Bergmann, Christophers, Huss and Zimmermann (2010) explore the characteristics of LPE and compare these to UPE. Their data is provided by the earlier mentioned company LPX GmbH, and consist of a database of 122 liquid LPEs. Bergmann et. al. conclude that LPE does not differ much from UPE in terms of investment

22

25

style (buyout, growth and venture), investment decision making, investment financing and portfolio management. The heterogeneous mix of organizational forms present in the LPE industry (as described in chapter 2.1.6) is the main difference between LPE and UPE.

The authors find that 84% of assets on an aggregate LPE‟s balance sheet are committed to private equity. Using the LPX-50 TR index for their empirical work, they show that LPE performance relative to the market (MSCI World) varies greatly between good and bad times. They find outperformance in good years, but underperformance in bad years, especially during the credit crunch. Risk, as measured by volatility is relatively high, and increases in bad years. Correlation with other asset classes is higher than generally assumed.

In 2010, Phalippou touches on the LPE market in a paper23 on the different methods to measure risk and return in the general PE market. He arbitrarily constructs his own small empirical database of 19 vehicles. These companies are at least listed since June 2005, are not venture capital firms, and have a certain level of liquidity. Using weekly returns and the Dimson (1979) regression model to account for autocorrelation due to remaining illiquidity, he calculates betas per company relative to local market indices.

He finds an average beta of 1.5 and an average alpha of 6% per annum. If the analysis is restricted to the last 3 years (starting June 2005, minimizing selection issues), the average beta is 1.4 and the average alpha 5%. He warns for drawing too strong conclusions of this sample, but on first glance past performance looks good and risk relatively low. Although investors cannot predict beforehand that companies won‟t go bankrupt and have high liquidity, Phalippou remarks that he is not aware of an LPE firm going bankrupt and liquidity levels seem stable through the years. Finally, he takes a quick look at listed venture capital firms and finds that, in contrast to his buyout/growth sample, many of them went bankrupt and surviving companies perform poorly.

In 2010, Kaserer, Lahr, Liebhart and Mettler publish an article on the time-varying risk of LPE, based on the study and database of Lahr and Heschke (2009). Kaserer et. al. study the aggregate and individual market risk and its variability over time for the LPE market, in order to generalize their results to the general PE market. Where Lahr and Heschke (2009) only focus on the lifetime beta of LPE vehicles, Kaserer et. al. go further and compute betas over a rolling window of one and two years and total lifetime. They also examine the stability of these betas, by computing correlations of cross-sections for consecutive years, using the Pearson correlation

23

26

matrices, which are checked on robustness by calculating the Spearman rank correlations. Finally, the authors analyze the transition probabilities between different risk classes, using the Markov transition matrix.

Kaserer et. al. find that aggregate market risk of LPE varies strongly over time and is positively correlated with the market return variance. CAPM betas are highly unstable and can only be predicted for 2 - 3 years in the future. High- and low-risk companies are more likely to keep their risk levels than do medium risk companies. The authors conclude that PE apparently possesses some industry specific market risk factors such as acquisitions and divestments that rebalance portfolios, information scarcity on portfolio companies, and rapid changes within these portfolio companies. This makes it hard for investors to reliably predict long-term risk of PE investments, making portfolio allocation difficult. Furthermore, the large increase in covariance between LPE and the market in times of uncertainty, casts doubt on the diversification potential of PE during crises.

27

2.3 Research focus and relevant variables

In this final section of the theoretical framework, I will define and describe the focus of my research and the corresponding relevant variables that form the base for my empirical work. This section is structured along four separate research topics.

2.3.1 Overall risk and return

As stated earlier, private equity is believed to be an asset class showing high historical returns and low correlation with traditional asset classes. As an important alternative asset class, investors seek exposure to PE for diversification purposes and to be exposed to the market independent excess returns (Alpha) that PE is supposed to deliver. As shown in the literature review, previous academic work on UPE risk and return is ambiguous to say the least. Due to incomparable databases, severe sample biases, large differences in applied assumptions and difficulties with performance measurement, results vary significantly and are inconclusive. Because LPEs are listed on a stock exchange, risk and return measurement is much more straightforward. Standard statistical techniques can be easily applied and results compared to traditional asset classes. Several studies24 on LPE state that characteristics and behaviour of LPE and UPE are quite similar, hence LPE can serve as a proxy for its unlisted counterpart. Earlier studies on the performance of LPE by Bauer et. al. (2001) and Bilo and Zimmermann (2005) show that overall risk adjusted returns are indeed relatively high and correlation with other asset classes is low. Lahr and Herschke (2009) come to the conclusion that LPE does not deliver any risk adjusted excess return. The ambiguous results of earlier academic work makes it worthwhile to re-examine the historical return and diversification potential of LPE again. So I hypothesize that LPE delivers risk adjusted excess returns compared to traditional stock markets and moreover, shows low correlation to other asset classes.

2.3.2 LPE behaviour through time

A very interesting issue to investigate is the risk and return pattern of the LPE portfolios through different periods in time. Kaserer et. al. (2010) find that market risk of LPE varies strongly over time and individual CAPM betas are highly unstable. As the diversification potential of LPE is one of the main reasons to invest in the asset class, it is essential to find out how the LPE market

24

28

actually behaves in times of crisis, the time where this diversification is needed most. Brown and Kräusl (2010) show that the highly debt dependent LPE market was hit hard in the recent financial crisis, and NAV discounts widened extremely. They also state that UPE has actually performed better than expected during the same crisis, even outperforming public markets. The fact that LPEs are always left out in the open and are never able to fly under any radar, could be a unique shortcoming in times of crisis. In the same light it wouldn‟t be a surprise if liquidity levels in the form of bid-ask spreads would further decrease in down markets. Since liquidity is already low compared to other public markets, it could well turn into a serious issue in bad times. I hypothesize that LPE‟s overall average risk and return pattern does not tell the whole story, and expect to find very different risk and return patterns between good and bad times.

An additional interesting case to examine is the half year anomaly in LPE returns that Bauer. et. al. found in 2001. Bauer et. al find that almost the entire annual return of LPEs is achieved in the first half of the calendar year. In 14 of the 15 investigated years, the first half of the year exceeds the second half. This difference is statistically significant. Bauer et. al. state that this could be related to the timing of the publication of annual reports. From an investor‟s point of view, this could be an important phenomenon. LPE could have its own version of the “sell in May and return in September” investment strategy, where investors anticipate this half year anomaly.

2.3.3 Heterogeneity within the LPE universe

29

et. al. (2009) find that FoFs have similar betas to other LPEs, but are expected to earn a slightly higher abnormal return. When it comes to the listed fund managers, there is unfortunately not enough data available for testing, as there are currently only four of these companies present in LPE sample, of which three have been listed for less than five years.

Next, I will examine the differences between the investment styles: Buyout, Venture and Growth. As said earlier in section 2.1.6, only the direct private equity category exhibits enough investments styles across companies to be able to examine differences. Although other categories do include a small number of venture and growth funds, their number is too small to be used for further research. Intuitively, venture capital investments are the riskiest (highest chance of bankruptcy), followed by buyout and then growth. On the other hand, buyout funds use the highest levels of debt, which should expose them more to market risk (Lahr and Herschke, 2009). Risk/return patterns should reflect these differences. Bauer et. al. (2001) state that listed venture capital indeed shows the highest volatility, and also the highest mean return. Risk adjusted, listed buyout performs better. Phalippou (2010) reports that, at first glance, listed buyout funds show superior results compared to listed venture funds. Huss (2005) reports slightly underperformance

Another part of the heterogeneity between organizational structures that may be worthwhile to investigate is the way how LPEs are managed. Specifically, whether or not differences exists between funds that have internal and external management. If these differences are indeed found, a further distinction between internal managed funds with and without additional fund management business may be necessary to better pinpoint the source of this difference. Lahr and Herschke (2009) find that internal managed LPEs show higher exposure to market risk. They suggest that this could be caused by the riskier cash flows of carried interest. Alternatively, one could also hypothesize that management fees provide a less risky stable cash flow and combined with carried interest offers diversification potential, hence exposure to market risk should be lower. Bergmann et. al. (2010: 13) state that “additional fund management business is a major advantage since it gives an investor the opportunity to participate in a steady cash flow stream”. of unlisted buyout funds compared to unlisted venture funds.

30

continental civil-law countries. In a study, partly based on the work of La Porta et. al., Cumming and Walz (2008) show that PE fund managers in countries with weaker accounting standards, tend to overvaluate their investments causing these values to be less reliable. Hence, information asymmetries between funds and its investors increase. Intuitively, one would thus assume LPE companies in common-law countries to show higher long-term performance, a hypothesis confirmed by Lerner and Schoar (2005) for UPE funds.

2.3.4 Influence of macroeconomic global risk factors on LPE

The next part of my empirical investigation focuses on the exposure of the LPE universe to global macroeconomic risk factors. Results on these risk factors can tell us something about what drives LPE performance, and to which extent. It will be interesting to see which factors are relevant for the LPE market and to see if LPE has risk exposure that isn‟t seen in regular stock markets, thus is unique for (L)PE. Several studies have investigated these kind of risk factors for (L)PE in the past. In 2001, Bauer et. al. found that LPE has significant positive relationships with the global

stock market, the global IPO volume and surprisingly, the credit spread. A less significant

positive relationship is found for the TED spread25 and global GDP growth. They find a significant negative relationship with stock market volatility. Compared to the MSCI World index, LPE shows the same signs (+/-), but the magnitude of the signs differs substantially, with reactions to IPO volume, credit and TED spread and volatility being much stronger for LPE. Jegadeesh et. al (2009) find that both FoFs and other LPEs are positively related to GDP growth and negatively to the credit spread. The latter finding is in contrast with the finding of Bauer et. al (2001), but is in line with both intuition and other literature on LPE (Lahr and Kaserer, 2010) and UPE. That both Bauer et. al. and Jegadeesh et. al. find LPE to be positively related to GDP growth seems no surprise. However, Gompers and Lerner (2000), Diller and Kaserer (2005) and Kaplan and Schoar (2005) find that GDP growth increases competition between unlisted funds, causing higher valuation of investments and lower fund performance in the aftermath. For my own research, the risk factors and their expected relationship with LPE are presented below:

25