Did UK buyout funds outperform the FTSE100 during the period 1986-2013?

(1)

BSc Economics and Business Specialization: Finance

“Did UK buyout funds outperform the FTSE100 during the period

1986-2013?”

Bachelor Thesis

Luuk de Wildt Student number: 10166548

Date: 1 July 2014

(2)

I. Introduction

Leveraged buyouts are often mentioned as value creating processes (Metrick and Yasuda, 2010). The largest leveraged buyout deal to date is the buyout of TXU Corporation. In this deal KKR & Co, TPG Capital and Goldman Sachs Capital Partners bought this company for about 48 billion dollars in 2007 (Bloomberg). This deal failed whit the bankruptcy of TXU Corporation, resulting in a 8,3 billion dollar loss in equity provided by the three buyout firms.

Over the past few years the private equity industry is investigated heavily by both researchers and investors (Mason and Harrison, 2002). The total private equity industry grew from about ten billion dollars in 1985 (Kaplan and Schoar, 2005) to over a trillion (Metrick and Yasuda, 2010).

The private industry is well-known for numerous previous highly profitable deals. However, there is a lot of discussion about the performance of private equity funds. The actual profitability measurement is discussed heavily in current literature, for example by Harris et al. (2013). Given the private nature of the data and different assumptions made by researchers on sample selection, findings on private equity returns are mixed and uncertain (Harris et al. 2012).

In Europe, the UK is the country with the highest private equity activity. It is Europe’s leading country based on private equity funds as well as on private equity investments made. Moreover, the UK forms the second largest private equity industry worldwide. Nevertheless, previous research focuses predominantly on the US private equity market. Very little is known about the UK buyout industry. This research investigates UK buyout funds during the period 1986-2013. The central question in this research is: “Did UK buyout funds outperform the FTSE100 during the period 1986-2013?”

The structure of this paper is as follows. Part II discusses relevant literature about buyout funds and the leveraged buyout process. Part III describes the hypotheses and methodology used to test these hypotheses. Moreover it presents an overview of the UK buyout sample data. Part IV presents an analysis of the data to test the hypotheses. Expectations and actual results are compared and discussed. Part V provides a conclusion and summary of the results found in this paper. It ends with a discussion and recommendation for further research.

(4)

II. Literature review

In this section the theoretical background on buyout funds is discussed. An explanation for common used terms is given which are used throughout this paper. This section focuses on the process of LBO transactions and the structure of buyout funds.

1. Overview of buyout industry

In a leveraged buyout, a firm is bought by a buyout fund. A buyout fund is an investment fund raised by a private equity firm. This private equity firm acquires a company purely as a financial investment. If a public firm is bought, a premium of 15 to 50 percent is paid on the current value of the stocks (Kaplan, 1989). Usually 60 to 90 percent of the transaction is financed with debt, while the remaining 10 to 40 percent is financed with equity (Kaplan and Stromberg, 2008). This is the reason for calling it a leveraged buyout. By increasing the amount of leverage, the private equity firm increases her return on the investment (all else equal). However, by increasing the amount of leverage, bankruptcy risk increases as well. Various different kinds of debt are used in a LBO. Large part of this debt is senior debt, which is arranged by a bank. In recent years, more of this debt is bought by institutional investors including hedge funds (Kaplan and Schoar, 2008). Next to senior debt, the debt usually includes junior debt. This part is not secured by a bank and consists of mezzanine debt or high-yield bonds (Kaplan and Schoar, 2008).

Private equity funds are typically closed-end limited partnerships (Kaplan and Stromberg, 2008). Closed-end means that investors are unable to withdraw their capital until the fund is closed. The closing of a fund occurs when the fund exits the investment. This can be done in several ways. The three most common exit routes are: a sale of the company to a strategic buyer, a sale of the company to another private equity firm (secondary LBO) and listing the company on a public stock exchange through an IPO (Kaplan and Stromberg, 2008).

The private equity firm serves as General Partner (GP) of the fund. The general partner is manager of the fund. The GP has to agree on a time period with the investors in their fund, called limited partners (LPs), in which the capital provided by the LPs will be invested. Usually, this is a period of five years (Kaplan and Schoar, 2005). The general partner usually provides around 1 percent of the total capital (Kaplan and Stromberg, 2008). By doing this, the GP aligns his objectives to the objectives of the LPs. Besides this, the GP has an agreed

(5)

period in which the capital will be returned to the LPs. This period is usually between ten and twelve years (Kaplan and Schoar, 2005). This means the LPs make an illiquid investment which lasts for ten to twelve years.

The LPs are the main providers of the capital to the fund and typically include institutional investors (Kaplan and Stromberg, 2008). They are not involved in the day-to-day activities of the fund and have little to say about what investments are made by the GPs with their capital (Kaplan and Stromberg, 2008). To provide protection for the LPs, basic covenants of the fund agreement have to be followed by the GPs. These covenants can pose restrictions on debt levels, percentage of capital invested in one specific company and the type of securities in which the fund is allowed to invest (Kaplan and Stromberg, 2008).

For managing the fund, the GPs are compensated with various fees paid by the LPs. Four common fees paid are management fees, carried interest, transaction fees and monitoring fees. Management fees are fixed and are independent on performance of the fund. Carried interest is a share in the profits of the fund, which in almost all cases equals 20% (Metrick and Yasuda, 2010). Transaction fees are comparable to M&A advisory fees and are independent on fund performance. Monitoring fees are dealt to the portfolio of companies a fund invests in. This is a compensation for the time spent in working with these companies (Metrick and Yasuda, 2010).

2. Performance of buyout funds

This sections starts with an overview of the performance determinants of buyout funds. After this it presents an overview of the most commonly applied performance measures and their limitations. It ends with a presentation of empirical results on buyout fund performance and

buyout cyclicality.

2.1 Determinants of performance

The performance of private equity funds is determined by several factors. One of these factors is skills of the GPs (Kaplan and Schoar 2005). Another factor investigated by researchers is the size of a buyout fund. Findings on the relationship between fund size and fund performance are mixed. Metrick and Yasuda (2010) and Kaplan and Schoar (2005) find a positive relationship between fund size and fund performance. When LPs see this good performance as a result of high skills, the demand for a new fund increases. GPs usually

(6)

increase the size of these so called follow up-funds. These larger funds lead to higher revenue per partner in the fund (Metrick and Yasuda, 2010).

Humphery-jenner (2011) finds a negative relationship between fund size and fund performance which he calls the “size effect”. He predicts that large funds should earn smaller returns than small funds. One of the key drivers of this size effect is the investments in small companies (Humphery-jenner, 2011). Large funds investing in small companies earn significant lower returns.

Next to firm size and GP skills, Jensen (1989) finds that by investing in specific industries funds can increase their profitability. He finds that slow growth, mature industries are more suitable for private equity investments. Buyout activity in these industries is found to be larger (Jensen, 1989). He states that in more mature industries, more value is destructed than in new, young industries. This creates a lot of profitable investment opportunities for buyout funds in these mature industries.

To improve performance, the new management of the company bought usually contributes a small fraction of capital as well (Kaplan and Stromberg, 2008). This gives the new management incentives to improve the company and generate higher profits (Jensen, 1989).Kaplan and Schoar (2005) find a strong persistence of performance across funds of the same partnership. Recent research by Harris et al. (2013) confirms these past findings. Successful private equity firms raise a new fund every 3 to 5 years (Metrick and Yasuda, 2010).

Next to the performance factors mentioned here, fund performance is influenced by general economic conditions. Market cyclicality has an impact on overall private equity performance and is investigated regularly. Kaplan and Stromberg (2008) suggest that the returns of buyout funds are affected by market cycles. Robinson and Sensoy (2011) find similar movements in private equity returns. They find a high correlation between private equity performance and public equity performance.

Besides this relationship between market cyclicality and fund performance, the relationship between market cyclicality and fund raising is a topic of interest. Kaplan and Stromberg (2008) find a counter-cyclical relation between fundraising and fund performance. During good times, interest rates are low, which makes it easy to attract capital. However, this increase in the amount of funds raised has a negative effect on fund performance.

(7)

2.2 Performance measures and limitations

Various metrics are applied to calculate the profitability of buyout funds. Each performance measure has its own advantages and disadvantages. The most commonly applied performance measures are discussed below.

Internal rate of return (IRR)

The IRR is the principal measure of performance in the private equity industry (Gottschalg et al., 2003). The IRR is the discount factor for which the NPV of an investment is equal to zero. It is a money-weighted performance measure. This performance measures assumes that interim capital distributions are reinvested at the same rate of return. The IRR calculations include the residual net asset value (NAV) as of the last date reported as a final cash flow, until all cash is returned to the investors (Harris et al., 2012). This ending NAV is included as a capital inflow, while the beginning NAV is included as a capital call. Using the IRR in a benchmarking process, one has to take into account the fact that it is a money-weighted return. Benchmarking these money-weighted returns with public market returns is not directly possible.

The IRR as a performance measure has several advantages over other performance measures. First of all, IRR’s take the timing of cash flows into account, which multiples don’t. Since private equity cash flows occur throughout the year on an irregular basis, the timing of cash flows is important to account for. Secondly, IRR’s are relatively easy to calculate and are easy to interpret.

Besides these advantages of using the IRR, there are some disadvantages as well. First of all, it is highly questionable whether interim capital distributions can be reinvested at the same rate of return. Secondly the IRR is relatively easy to manipulate (Brown et al., 2013). This can be done, for example, by overstating the NAV (Brown et al., 2013). Assumptions about NAVs are of big influence on the IRR’s calculated for recent vintages, because the proportion of realized investments falls for recent vintages (Harris et al., 2012). It is hard to determine the value of the assets hold by buyout funds. However, these NAVs have a great impact on the reported IRR’s of the buyout funds. By manipulating the NAV of their fund, fund managers can adjust their returns upwards (Brown et al., 2013). Recently Brown et al. (2013) investigate the reported NAVs in the private equity industry. They find a limited overstatement of NAVs during high fundraising activity periods. However, the funds boosting their NAVs are found to be firms that are not able to raise subsequent funds. This indicates

(8)

that investors actually look through this manipulation of NAVs. For the best performing funds, conservative reporting of returns is found (Brown et al., 2013).

Profitability index (PI)

The profitability index compares the performance of private equity funds with public market performance. The PI can be calculated by dividing the present value of distributed cash flows by the present value of invested cash flows (Phalippou and Zollo, 2005). For active funds, the final NAV is used as a cash inflow (Phalippou and Zollo, 2005). In this metric, a discount rate is applied, which is usually the return on a benchmark index. A value (PI) higher than one indicates outperformance of the index, while a PI lower than one indicates underperformance. An advantage of the PI measure is that a direct comparison between buyout fund returns and public market returns can be made (Phalippou and Zollo, 2005). Another advantage is that the PI is easy to interpret.

Besides these advantages, the PI contains some disadvantages as well. First of all, when using this method, one implicitly assumes a beta equal to one (Ljungqvist and Richardson, 2003). Previous findings on betas of buyout funds show different betas. Therefore it is questionable whether assuming a beta of one is appropriate. Another disadvantage of using the PI is that it is relatively easy to manipulate. The reason for this is, just like in the IRR calculation, the final NAV is included as a cash inflow. Public market equivalent (PME)

Given the problems with benchmarking private equity returns, Kaplan and Schoar (2005) constructed the public market equivalent (PME). The PME is a ratio of the discounted cash outflows of the fund to the discounted cash inflows of the fund (Kaplan and Schoar, 2005). To make a comparison with market returns, the total return on the market index is used as discount factor. The PME can be calculated as follows:

Where, Dt are capital distributions, Ct are capital calls and rt is the return on the public market index used as benchmark (Robinson and Sensoy, 2011). A PME greater than one indicates

(9)

outperformance of the index, while a PME lower than one indicates underperformance.

Just like the profitability index, the PME is easy interpretable. For example, a PME of 1,20 indicates a 20% outperformance over the fund’s life. Harris et al. (2012) find another advantage of the PME measure. They find that the PMEs are relatively insensitive to higher betas. They test this by implying betas of 1,5 and 2. Their results are very comparable to the situation in which they make no specific beta assumptions. Therefore, they reason that using the PME benchmarking method, no specific assumptions about the beta have to be made (Harris et al., 2012). This makes it more reliable than the profitability index.

A disadvantage of the PME is that it is relatively easy to manipulate. Just like in the IRR and PI calculation, final NAVs are included which are easy to manipulate.

Multiples

Besides the performance measures mentioned above, various multiples exist which provide insight in the performance of buyout funds. The multiples most commonly used are the distributions to paid-in capital (DPI), the residual value to paid-in capital (RVPI) and the total value to paid-in capital (TVPI). The DPI gives insight in the realized return, while the RVPI gives insight in the unrealized return. The TVPI is the sum of the DPI and RVPI and gives insight in the total return. One has to keep in mind that the longer the holding period is, the higher the multiples have to be to attain the same IRR (Fraser-Sampson, 2011).

An advantage of these multiples is that they are easy to calculate and easy to interpret. For this reason, some researchers choose to include multiples in their performance analysis. However, a severe disadvantage is that these multiples don’t take the timing of cash flows into account. For that reason multiples are only used as first insight in private equity performance.

2.3 Empirical results on buyout performance

Several studies have been conducted regarding private equity performance. Most of these studies focus on the performance of the complete private equity industry, while this research focuses only on buyout performance. The findings of these studies are mixed and can be ordered in three different outcomes: market outperformance, market underperformance and performance equal to market returns.

Kaplan and Schoar (2005) report a slight underperformance net of fees of buyout funds compared to the S&P500. However, gross of fees they find an outperformance of the S&P500 for buyout funds. They investigate the performance of private equity partnerships

(10)

during the period 1980-1997 using the VentureXpert database. They find a large heterogeneity between funds. Fund performance is compared to the S&P500 using the PME measure. This study finds that fund size and GP’s experience are positively related to the performance of the fund.

Another underperformance for buyout funds is found by Phalippou and Gottschalg (2006). They find that accounting values (NAVs) stated by buyout funds are seriously biased upwards. After correction for this bias, they find an underperformance net of fees of 3,83% per year relative to the S&P500 (Phalippou and Gottschalg, 2006). These researchers use both the IRR and profitability index to measure buyout fund performance.

Robinson and Sensoy (2011a) investigate the characteristics of private equity cash flows and private equity performance. They find an outperformance of buyout funds relative to the S&P500 of 18% over the fund’s life. Robinson and Sensoy (2011a) show that their results are nearly the same for different beta assumptions. In a beta range of 1,5 – 2,5 they find a flat relation between relative performance (PME) and buyout beta. A co-movement between buyout fund returns and market returns is found. Even during bad economic periods, buyout funds outperform the S&P500 (Robinson and Sensoy, 2011a).

The most recent large scale research on buyout fund performance is conducted by Harris et al. (2012). They dismiss some of the previous findings on private equity performance based on some fundamental problems with the VentureXpert database (Stucke, 2011). This database is used in many of the previous studies. In their study, Harris et al. (2012) use data of nearly 1400 venture capital and buyout funds located in the United States. They use new data provided by Burgiss.

In contracts with many of the previous studies, they include all funds in their sample. Previous studies only included only liquidated funds and very mature funds. Since private equity activity increased heavily during recent years, Harris et al. (2012) choose to include funds with recent vintage years as well. In their study, they find a larger outperformance than has been documented before. They find an outperformance vis-à-vis the S&P500 of 20-27% over the life of a fund; more than 3% per year. No significant relation is found between fund size and fund performance.

The measure Harris et al. (2012) use to compare public market with private equity returns is the PME measure. Using various benchmark indices, similar PMEs are found. This indicates that the results are robust, regardless of the chosen public market index. Besides this, Harris et al. (2012) try to make adjustments for risk. However, they find that their results are relatively insensitive to these beta assumptions. This is similar to the findings by Sorensen

(11)

and Jagannathan (2013). They state that advantage of PMEs is that public and private returns can be compared this way, without having to make assumptions about risk.

By investigating the relationship between PME and IRR’s and multiples, Harris et al. (2012) develop a model which makes it possible to estimate PMEs without the use of underlying cash flows. This could be of great value in further research.

2.4 Emperical results on buyout cyclicality

Buyout cyclicality is investigated by various researchers. The results they find are uniform. The main findings on buyout cyclicality are outlined below.

One of the studies regarding buyout cyclicality is conducted by Kaplan and Schoar (2005). They find fund performance to be procyclical. More partnerships are started after a period of good industry performance. However, the ability of the GP’s of these funds to raise a follow-on-fund is lower. This indicates that these funds, which are started during booming periods, perform worse (Kaplan and Schoar, 2005). A possible explanation for this is that the number of entrants is negatively related to fund performance. Older funds are less affected by the number of entrants. This is caused by underlying differences between the funds regarding skills of the GP’s; more experienced GP’s have access to better investment opportunities (Kaplan and Schoar, 2005). The performance of the GPs is consistent across different funds. Using a more recent dataset, Robinson and Sensoy (2011a) investigate the cyclicality of US private equity firms. Their results on cyclicality of buyout performance are in line with previous research by Kaplan and Schoar (2005). Funds that raise capital during hot periods perform worse in terms of IRR (Robinson and Sensoy, 2011a). However, the PMEs of these funds are not lower, which indicates that relative performance versus the S&P500 index is not lower for funds started during hot periods.

(12)

III. Methodology and data

1. Objectives of the study and hypotheses

Below the objectives and hypotheses of this research are outlined. Based on the previous findings and literature, expectations are formulated regarding each hypothesis.

1.1 Outperformance

Previous results on the performance of private equity funds are mixed. Kaplan and Schoar (2005) and Robinson and Phalippou and Gottschalg (2006) find an underperformance of US buyout funds relative to the public market. However, more recent results of research conducted by Robinson and Sensoy (2011a) and Harris et al. (2012) shows an outperformance of US buyout funds. Based on these recent results for the US buyout market, the following hypothesis is formed:

On average, UK buyout funds outperform the public market index FTSE100 during the period 1985-2013.

1.2. Buyout market cyclicality

Findings by Kaplan and Strömberg (2008) and Robinson and Sensoy (2011) show that buyout performance and buyout fundraising is affected by market booms and busts. They find a pro-cyclical relationship for both. Based on these findings, the following hypotheses are formed:

Buyout performance is pro-cyclical. Fundraising is pro-cyclical

1.3 Buyout fund size

Previous findings on the relationship between fund size and fund performance are mixed. Metrick and Yasuda (2010) find a positive relation, while Humphery-jenner (2011) finds a negative relation between the two. Based on these mixed findings the following hypothesis can be formed:

(13)

There exists no significant difference in performance between small, medium, large and mega UK buyout funds.

1.4 Buyout fund industry

As stated by Jensen (1989) mature industries are most profitable for buyout funds. He finds buyout activity to be the highest in these industries as well. Based on his findings, the following hypothesis is formed:

LBO activity and returns in the UK will be higher in low-growth, mature industries. 2. Methodology

Just like Harris at al. (2012), I include both active and liquidated funds in my sample of funds. They notice a large expansion of buyout funds in recent years and therefore also include active funds in their sample. It is hard to compare fund performance between funds. One of the reasons for this is the large heterogeneity between private equity firms. Usually a distinction is made based on vintage years. A vintage year is the year in which the buyout fund delivers the first capital to an investment. In this research funds with same vintage years are compared to each other.

During this research, private equity IRR’s are compared to public market returns to test whether UK buyout funds outperformed the public market. As outlined before, the IRR is a money-weighted performance measure. Public market returns are calculated on a time-weighted basis. These two different performance measures cannot be compared directly. Besides this, public markets and buyout funds have different risk characteristics. In previous literature, researchers tried to estimate buyout fund betas. Buyout fund betas of around one are found or assumed in a lot of previous papers. Beta calculations can be found in previous research. Different researchers find different betas and therefore it is questionable whether assuming a beta of one is appropriate (Harris et al., 2012).

Given the uncertainty about buyout fund risk, some researchers apply risk adjustment methods. However, different methods are used and no common way for doing this has been found yet. To overcome these risk related problems, the public market equivalent is used to compare buyout fund returns with public market returns. The PME calculation makes it possible to compare these returns without making assumptions about the systematic risk, as measured by beta (Sorensen and Jagannathan, 2013). Another advantage of the PME is that

(14)

IRR’s and market returns can be compared, without have problems with the different type of performance measures. The PME is calculated by discounting the returns of buyout funds, using the public market return as a discount factor.

The PME’s are calculated on a yearly basis using quarterly cash flows. Harris et al. (2012) suggest that results can be dependent on the public market index chosen. To test whether this is the case, Harris et al. (2012) use various market indices. I compare the buyout fund performance to both the FTSE100 and the FTSE250. These are two major UK stock indices. The FTSE 100 contains the 100 major companies listed on the London Stock

Exchange. The FTSE 250 is an index containing the 101st_{to 350st largest companies listed on}

the London Stock Exchange.

Different PME’s can be compared the public market return; the median and the average PME. Previous research finds that the average PME is the correct one to compare with public market returns. Findings by Harris et al. (2013) and Sensoy et al. (2013) indicate that the typical limited partner has access to the average buyout fund. Because of these findings, I compare average yearly PME’s with yearly public market returns. Public market returns are calculated using data from yahoo finance.

Cyclicality of returns and counter-cyclicality are analyzed using the data found in the ThomsonOne database. Graphs and tables of these are presented in the data section. By comparing fund raising and performance during boom and bust times I can see whether these patterns are pro- or counter-cyclical.

ThomsonOne makes a distinction between different fund sizes. Although the exact fund sizes are not presented, a distinction is made between small, medium, large and mega-buyouts. The sizes are determined as follows:

T-tests will be used to test whether there are significant differences in performance between these different fund sizes.

Besides the relationship between fund size and fund returns, the relationship between fund return and fund industry is examined. Some funds invest a significant part of their capital in one specific industry. The measure used to test this, is the VEIC measure. This VEIC measure indicates whether a fund invests at least 60% of its capital in one specific industry.

(15)

Returns of funds with different industry VEICs will be compared to see which industries are most profitable for buyout investments. Data of the biotechnology, communications and media, consumer related, industrial/ energy and medical/ health industry from ThomsonOne are used to compare these different industries.

Besides a normal t-test, the Welch’s t-test is used in this paper. This specific test is used, because I find a difference in the variance of performance for funds of different sizes. This Welch’s t-test statistic is calculated as follows:

, where = Industry1 IRR, = Industry2 IRR

The degrees of freedom can be calculated by applying the following formulae:

The T-test is used to test whether two means are significantly different. In this paper, various mean returns are compared. The Welch’s t-test helps interpreting the results found.

3. Data

Discussion of the dataset

In this section, a presentation and discussion of the data is presented. The sample contains a total of 236 buyout funds over the time period ranging from 1985-2013. In order to test the above hypotheses, data from the TomsonOne private equity database is used. ThomsonOne collects information about private equity funds on a quarterly basis. Over 38,000 venture, buyout and mezzanine funds are included in this database. Both active and liquidated funds are reported. The data in this database is voluntarily reported by the GPs and LPs of the funds. Since the data is private, it is not possible to validate the consistency of the data (Kaplan and Schoar, 2005). However, since data are provided by both GPs and LPs, ThomsonOne claims that the possibility for inconsistent reporting is limited.

(16)

included. From the nineties onwards, the buyout industry started to grow rapidly and so did the number of fund observations in this database. Data about fund size, fund sequence, fund location and various performance measures are available. The performance measures reported are the internal rate of return, cumulative total value to paid-in capital and distributed total value to paid-in capital. The reported data are net of fees, so they reflect what the returns for the LPs of the funds.

Stucke (2011) finds some significant problems with this dataset provided by ThomsonOne. He states that there is a downward bias in the returns presented in this dataset. However, this database is still used by private equity researchers. Harris et al. (2012) include it in their research as well. They find an outperformance of the S&P500 based on this dataset. The fact that this dataset is still used in research makes it a suitable dataset for this paper. Data overview

This section presents an overview of the sample data used in this research. Table 1 shows the buyout funds included in the sample. All funds are included in the sample, so also the recent increase in buyout activity is reflected. This significant increase in buyout fund activity can be found in fund capitalization as well. The enormous growth, from 2000 onwards, is in line with the growth of the whole private equity industry.

(17)

Table 2 shows the capital weighted IRR’s of the funds included in the sample. The capital weighted IRR’s are calculated from inception of the investment. Capital weighting is used to reflect for the differences in fund sizes. The capital weighted IRR’s reflects best what a one pound investment yields. As is done in previous research, vintage years are used to group funds.

(18)

IV. Results and analysis

In this section, they data is analyzedin order tofind an answer to the hypotheses reported in the

previous section. The first hypothesis is:

H1: On average, UK buyout funds outperform the public market index FTSE100 during the period 1986-2013.

In order to make a comparison between UK buyout funds and the returns of the FTSE100, PMEs are calculated. Table 3 presents the PMEs for the period 1986-2013. The year 1985 is excluded from this analysis, since for this year no cash flow data is available.

A t-test is used to test whether buyout fund performance is significantly better than market performance during the period 1986-2013. Quarterly PMEs for both indices are compared with a PME equal to 1. I find T-values of 11,45 and 6,47 for the PMEs based on the FTSE100 and FTSE250 respectively. These values indicate a significant outperformance at the 5% level for buyout funds relative to the public market indices. All of the PMEs are at least equal to one during this period. In comparison to the FTSE100 the average PME for this time period is 1,21. This indicates and outperformance of 21% relative to the FTSE100 over the life of a fund. When buyout fund performance is compared to performance of the FTSE250, an average PME of 1,17 is found. This indicates an outperformance of 17% relative to the FTSE250 during 1986-2013.

Harris et al. (2012) state that performance is usually compared to multiple market indices. For this reason, the PMEs are calculated for both the FTSE100 and FTSE250. The PMEs reported in table 3 show comparable results for both comparisons. For both indices, all PMEs exceed one. The average PME for the FTSE250 is slightly lower; 1,17 versus 1,21 for the period 1986-2013.

The fact that the PMEs are quite similar, using two different market indices indicates that PMEs are only slightly dependent on the choice of benchmark. These results are in line with the results of Harris et al. (2012). They find comparable PMEs for US buyout fund performance using four different market indices. The PMEs they find during the period 1984-2008 are between 1,16 and 1,22. These are very close to the PMEs I find of respectively 1,17 and 1,21.

By including recent years in my analysis, the performance of funds in most recent vintages depends on the future performance of the investments made by those funds (Harris et al., 2012). Given the J-curve pattern in the returns of buyout funds, I have reason to assume

(19)

this indeed is the case. A suggestion this indeed is the case comes from recent research by Harris et al. (2013) who find higher PMEs for recent vintages than Harris et al. (2012) because of the use of a more recent dataset. Since the main focus of this research is not on

(20)

The second hypothesis is: H2: Buyout performance is pro-cyclical.

In order to see whether the performance of UK buyout funds is pro-cyclical, the capital weighted average IRR’s are calculated for the period 1985-2013. These are presented in table 2. The correlation between market returns and buyout returns I find is 0,053 and is presented in table 4. This indeed indicates a positive correlation, although the correlation I find is very small and not significant at the 5% level.

What is remarkable is that the dot.com bubble had little to no impact on the performance of UK buyout funds. Since I don’t have fund level data, I am not sure what the reason for this could be. My suggestion is that UK buyout funds had relatively little invested in the technical companies which were affected significantly during the dot.com bubble. As Jensen (1989) finds, buyout activity is highest in mature industries. This supports my suggestion that buyout funds didn’t invest heavily in technological start-ups before the dot-com bubble.

The third hypothesis is:

H3: Fundraising is pro-cyclical.

Table 5 presents the amount of capital raised by UK buyout funds during the period 1985-2013. The correlation I find between market returns and fundraising is 0,6406, which is significant at the 5% level (see table 4). During the recent economic crisis, a decrease in fundraising is found in 2007. However, fundraising increased during 2008, indicating funds adjusted to the new economic situation. In the years preceding the recent crisis, there is a boom in fundraising. Again, the dot.com-bubble had little influence on the amount of funds raised.

This significant positive correlation is in line whit my expectations. Looking at table 5, it shows three private equity fundraising waves (1997-2001, 2005-2006 and 2011-2013)

(21)

which occur during good economic periods. These results are in line with previous results on the US private equity market. As Kaplan and Stromberg (2008) point out, a fund raising boom reduces the performance of buyout funds. This is pointed out by Kaplan and Schoar (2005) as well. They find that when many new entrants enter the buyout market, performance decreases.

This might seem in contrast with the pro-cyclical movement of private equity returns, since most funds are raised during boom periods. However, there is a time lack between the boom in fund raising and the actual effect of this on buyout fund performance (Kaplan and Schoar, 2005).

The fourth hypothesis is:

H4: There exists no significant difference in performance between small, medium, large and mega UK buyout funds.

Previous results about the relation between buyout fund performance and fund size are mixed. In table 6 the capital weighted IRR’s for funds of different sizes presented for the period 1985-2013. The largest part of the funds falls into the small category, with a market capitalization below £200 million. Large performed best in the UK during the period 1985-2013, with a capital weighted average IRR of 16,74%. Mega funds showed the worst performance, with a capital weighted IRR of 8,86%. The standard deviation is the lowest for

(22)

mega funds, indicating there are relatively small difference between different funds in this group. The performance of small and medium funds falls in between the performance of large and mega funds.

In order to test whether the differences in performance are statistically significant, I use the Welch’s T-test. Table 7 presents the outcomes of this t-test. The only statistically significant difference in performance is found between large and mega funds; large funds significantly outperformed the mega funds. Although the others findings are not statistically significant, one has to keep in mind the economic significant differences between returns. For example, a difference in performance of 4,56% is quite big in an economic perspective, although it is not statistically significant.

To some extent, the performance of the fund increases with fund size. Large fund perform better than small and medium funds, while medium funds outperform the small funds. However, mega funds show worst performance, indicating that there is a limitation on the relationship between fund size and fund performance for the UK buyout funds investigated.

(23)

The fifth hypothesis is: H5: LBO activity and returns will be higher in low-growth, mature industries.

Table 8 shows the performance of UK buyout funds per industry. The funds included in this table invest at least 60% of their capital in one specific industry. Only 46 buyout funds of the total 236 buyout funds invest more than 60% of their capital in one specific industry. This indicates that most funds spread their investments across various industries. Spreading investments makes these funds less prone to problems in one specific industry.

Largest activity during the period 1985-2013 is shown in the consumer related industry, while the least activity is shown for the industrial energy industry. The capital weighted average is highest for funds investing in the consumer related industry, while this lowest in the medical/ health industry.

These results are in line with my hypothesis and previous results of Jensen (1989). Largest activity and best fund performance are shown in the most mature industry; consumer related.

(24)

(25)

V. Conclusions

This research gives an overview of the most important characteristics of the performance of UK buyout funds. By using the ThomsonOne private equity database, the performance of these funds is investigated. Previous research focused on the US buyout market and most of these papers did not include recent vintage years. Most of the results in this paper are in line with the results of previous research on the US private equity market.

First of all, the results show an outperformance of buyout funds versus the public market. The largest outperformance is found relative to the FTSE100. Calculated PMEs show an outperformance of approximately 21% over the life of the fund. This level of outperformance seems relatively insensitive to the choice of benchmark. As I stressed before, the ultimate level of funds in recent vintage years depends on the final realization of the current investments of these funds (Harris et al., 2012).

Secondly, the cyclicality of the UK buyout market is investigated. Performance of UK buyout fund is pro-cyclical. The dot-com bubble seems to have no impact on buyout performance. A possible explanation for this is the probably low exposure of funds to the technological industry. Buyout fund raising is pro-cyclical as well.

Thirdly, the relationship between fund size and fund performance is investigated. My results on this are mixed. Large funds show better performance than medium and small funds. However, mega funds show worst performance. This may suggest that there is a limitation on the positive effect of fund size on fund performance. The fact that large funds perform better, is usually found because follow-on funds are generally larger funds (Kaplan and Schoar, 2005). This would be an interesting topic for further research in the UK.

Fourthly, the activity and performance of funds in specific industries is investigated. The results show that performance and activity is highest in the consumer related industry. This industry is the most mature of the industries investigated and so these industry specific results are in line with previous research of Jensen (1989).

It would be interesting to see what actually causes the market outperformance of UK buyout funds. This currently forms a puzzle for many researchers. I leave this as a recommendation for further research, for the UK en US buyout markets in specific.

(26)

Bibliography

http://www.finance.yahoo.com (accessed on 11 June 2014)

Axelson, U., Strömberg, P. and Weisbach, M., (2009). Why Are Buyouts Levered? The Financial Structure of Private Equity Funds. The Journal of Finance, (4), 1549-1582. Brown, G. W., Gredil, O., & Kaplan, S. N. (2013). Do private equity funds game

returns. Available at SSRN 2271690.

Fraser-Sampson, G. (2011). Private equity as an asset class. John Wiley & Sons.

Gottschalg, O., Phalippou, L. and Zollo, M., (2003). Performance of Private Equity Funds: Another Puzzle? Insead working paper.

Harris, R., Jenkinson, T. and Kaplan, S., (2012). Private Equity Performance: What Do We Know? Working Paper. Forthcoming: Journal of Finance.

Harris, R. S., Jenkinson, T., Kaplan, S. N., & Stucke, R. (2013). Has persistence persisted in private equity? Evidence from buyout and venture capital funds. Evidence from

Buyout and Venture Capital Funds (April 1, 2013).

Humphery-Jenner, M. (2011). Private equity fund size, investment size, and value creation. Review of Finance, rfr011.

Jensen, M., (1989). Agency Costs of Free Cash Flow, Corporate Finance and Takeovers. The American Economic Review, (2), 323-329.

Kaplan, S. N., & Schoar, A. (2005). Private equity performance: Returns, persistence, and capital flows. The Journal of Finance, 60(4), 1791-1823.

Kaplan, S. and Strömberg, P., (2009). Leveraged Buyouts and Private Equity. Journal of Economic Perspectives, (4), 121-146.

Ljungqvist, A., and Richardson, M., (2003). The Cash Flow, Return and Risk Characteristics of Private Equity. NBER Working Paper No. 9454.

Metrick, A., and Yasuda, A., (2010). The Economics of Private Equity Funds. Review of Financial Studies, (6), p. 2303-2341.

Phalippou, L. and Gottschalg, O., (2009). The Performance of Private Equity Funds. Review of Financial Studies, (4), 1747-1776.

Phalippou, L., & Zollo, M. (2005). What drives private equity fund performance. Unpublished

working paper.

Robinson, D. T., & Sensoy, B. A. (2011). Cyclicality, performance measurement, and cash

(27)

Robinson, D. T., & Sensoy, B. A. (2011a). Private equity in the 21st century: Liquidity, cash flows, and performance from 1984-2010. NBER Working Paper, (17428).

Sensoy, B. A., Wang, Y., & Weisbach, M. S. (2013). Limited partner performance and

the maturing of the private equity industry (No. w18793). National Bureau of

Economic Research.

Sorensen, M., & Jagannathan, R. (2013). The Public Market Equivalent and Private Equity Performance. Available at SSRN 2259261.

Did UK buyout funds outperform the FTSE100 during the period 1986-2013?