The performance of private equity funds
during the latest private equity wave
and the financial crisis
Bachelor thesis
July 2014
Author:
Jenk Lemmink
Student number: 10189823
Content
1 Introduction……… 3
2 Literature……… 4
2.1 Private Equity, how does it work? ……… 4
2.2 Private equity fund performance………... 5
2.3 Fund performance during economic up- or downturns………. 6
3 Methodology……… 8 3.1 Hypotheses.………... 8 3.2 Methodology..……… 9 3.3 Dataset……..………. 11 4 Results………. 13 4.1 Leveraged buyouts………. 13 4.2 Venture Capital……….. 15 4.3 Limitations………. 16 5 Conclusion……….. 17 6 Bibliography……… 19 Appendix……… ……... 21
1 Introduction
Private equity as an asset class has grown spectacularly over the last decades. Institutional investors and rich individuals have invested large amounts of money in private equity funds in order to generate excess returns over the public equity markets. Does investing in private equity bring the level of excess return expected by the investors? Several studies have tried to answer this question over the last few years, but they all report different results due to the various datasets and performance measures used. All these studies focus on the performance of private equity funds during the first and second private equity wave. This study however, includes the latest private equity wave and the performance data during the financial crisis in the dataset. Additionally, this study looks at the performance of leveraged buyouts and venture capital funds in periods when the pubic stock markets have positive or negative excess returns. Therefore, this study will give new insights into the performance of private equity as an asset class.
The main purpose of this paper is to examine the performance of venture capital and leveraged buyout funds during the latest private equity wave and the financial crisis, and compare it to the performance of the public market indices during the same time period. Venture capital funds are compared with the Nasdaq, while leveraged buyout funds are compared against the S&P 500.
The results clearly show that leveraged buyout funds outperform the S&P 500 in the period from 1988 till 2013 by approximately 2.5% per quarter. These results are in line with the results found in earlier research by Harris, Jenkinson & Kaplan (2013), Phalippou & Gottschalg (2009) and Ljungqvist & Richardson (2003). Venture capital funds on the other hand, underperform the Nasdaq index, while this underperformance is not statistically
significant from zero. Leveraged buyout and venture capital funds both outperform the public stock market indices during times when the public stock market indices generate negative excess returns. Leveraged buyout funds underperform the S&P 500 during periods when the public stock market generates positive excess returns, while venture capital funds outperform the Nasdaq in these periods.
This paper is structured as follows. Section 2 gives an overview of the present
literature on private equity. Section 3 describes the methodology and data. Section 4 presents the results. Finally, a conclusion is drawn in section 5.
Literature
This section gives an overview of the present literature on private equity. First, there will be an explanation as to what private equity funds are and how they generate returns. This is followed by a quick overview of the existing literature on private equity fund performance. Finally, the literature on the private equity fund performance during economic up- or downturns is described.
2.1 Private Equity, how does it work?
Private equity by definition is an equity investment in a company which is not quoted on a stock exchange. There are two main types of private equity, namely “buyout” and “venture capital”. Buyout transactions are generally equity participations in large and well established companies. Another feature of buyout transactions is that it is largely financed with debt. Venture capital on the other hand, does typically not include debt and involves participations in young and sometimes start-up firms (Fraser-Sampson, 2010).
Investing in private equity can be distinguished by “fund investing” and “direct investing”. Private equity investors commonly invest in a private equity fund, as the fund in turn invests in the underlying portfolio companies. In such a fund, the investors are called “LP” (Limited Partner) and the portfolio manager is called “GP” (General Partner). These limited partners are institutional investors, like pension funds and insurance companies, as well as rich individuals. A private equity fund is a closed-end fund and has a finite life of approximately 10 to 12 years. At the start of the life cycle of the fund, all the investors set their amount of committed capital. This is the maximum amount of capital which an investor has promised toprovide. During the life cycle of the fund, a certain amount of the committed capital will be requested by the private equity fund and paid by the investors. This is called drawdown capital and includes both capital to be invested in companies as well as money required for fees and expenses. During the first 3 years of the fund, the capital is raised and the GP makes the investments. After this 3 year period the investors expect to receivethe benefits from their investments and they are looking forward thatthe returns will be
distributed. Therefore, the returns of private equity funds will be negative in its early years as money is paid into the fund. However, as distributions start to flow back to the investor, a point will be reached where the amount of cash-inflows matches the amount of cash-outflows
and finally the total return becomes positive. This effect is known as the J-curve (Fraser-Sampson, 2010).
At the end of the funds lifetime the process will be finished and the
equity-participations will be sold. Exiting the investment is an important step in the entire process. Kaplan & Strömberg (2009) show in their research that the sale of the company to a strategic buyer is the most common exit route. Another usual exit route is selling the firm to another private equity fund. Listing the company on a public stock exchange through an initial public offering is also a commonly used exit.
2.2 Private equity fund performance
The capital committed to private equity funds hasincreased dramatically over the last few decades. Phalippou & Gottschalg (2009) show for example, that capital committed to US private equity funds increased from $5 billion in 1980 to more than $300 billion in 2004. Investors commit capital to private equity funds in order to generate excess returns over the public stock markets. Does investing in private equity bring the level of excess return expected by the investors? The last few years, several studies have tried to answer this question by calculating the “market adjusted” private equity returns. Such an analysis
integrates an opportunity-cost benchmark into the measurement of return, due to the fact that the cash flows to private equity are not risk-free, but require adjustment by an opportunity cost. The outcomes of these studies are surprisingly different. These differences occur due to the different datasets and different performance measures used.
Ljungqvist & Richardson (2003) are one of the first researchers who analyze the cash flows, returns and risk characteristics of venture capital and leveraged buyout funds. They use a relative small sample of just 73 funds during the period from 1981 till 1993 and compare the returns net of fees on a value weighted basis to the returns of the S&P 500. They come to the conclusion that both venture capital and buyout funds outperform the S&P 500 by
approximately 6 percent per year. According to them, this excess return is partly a compensation for the extreme illiquidity of private equity.
Kaplan & Schoar (2005) on the other hand, use a larger dataset of 746 funds in the period from 1980 till 2001, which gives different results. They find a public market equivalent (PME) of 1.21 on a value weighted basis for venture capital. This indicates that 1 dollar invested in a venture capital fund discounted at the rate of return of the S&P 500, would generate 1.21 dollar over the life of the fund. Thus, on value weighted basis venture capital
outperforms the S&P 500. However, on equally weighted basis venture capital has a PME of 0.96 and underperforms the S&P 500. Kaplan & Schoar (2005) also report underperformance for buyout funds compared to the S&P 500. On an equally weighted basis Buyout funds have a PME of 0.97 and on a value weighted basis the PME is 0.93.
A third important study on the performance of private equity funds is done by
Phalippou & Gottschalg (2009). Using a sample of 852 funds during the period from 1980 till 1993, they find that in most cases private equity funds do not outperform the S&P 500.
Venture capital underperform the S&P 500 both on equally weighted basis as well as on value weighted basis, with PMEs of respectively 0.76 and 0.95. Buyout funds, however,
underperform the S&P 500 on equally weighted basis with a PME of 0.97, but at the same time outperform the S&P 500 on value weighted basis with a PME of 1.06.
The most recent study on the performance of private equity funds is done by Harris, Jenkinson, & Kaplan (2013). This studyuses the largest dataset of all studies mentioned in this paper. The sample includes nearly 1400 buyout and venture capital funds in the period from 1984 till 2008. They report outperformance of buyout funds by approximately 3% on equally weighted basis annually, compared to the S&P 500. Venture capital outperform the S&P 500 in the 1990s, while underperformance is reported for the 2000s.
2.3 Fund performance during economic up- or downturns.
Little empirical research has been done into the performance of private equity funds during economic up- or downturns. Phalippou & Zollo (2005) are one of the few researchers who actually test the influence of market- and business cycles on the performance of private equity funds. They conclude that the returns off private equity funds are pro-cyclical. The
performance of private equity funds improves significantly when the investments take place during periods with high GDP growth, high public market performance, high return on call options as well as low returns on put options. Moreover,they show that private equity funds, like hedge funds, bear significant right-tail risk. The excess returns are mainly realized when the public stock-market has high returns. When the public stock market returns are below a certain bound, the private equity performance is very low, regardless of how low the public stock market returns are. A reason for the pro-cyclical performance might be the high level of debt in buyout investments and the venture capital participations in small-growth firms, which are risky investments that are likely to perform better during economic upturns. An important shortcoming of this research is that there is no distinction made between the performance of
leveraged buyouts and venture capital. Both types of private equity are completely different, but this study looks at the performance of the private equity sector as a whole.
Kaplan & Strömberg (2009) show that the capital committed to private equity funds is cyclical. In addition, it is easier to lend in boom times due to the favorable credit market environment. So, according to them, private equity deals are easier to undertake during economic upturns. Because of this, there will be some overinvesting in unprofitable deals during these times as well. Therefore, they suggest that on average, deals undertaken during economic upturns will underperform deals undertaken during economic downturns.
Kaplan & Schoar (2005) agree with this andshow that private equity funds raised in boom times are less likely to raise follow-on funds, indicating that these funds are more likely to perform poorly. In addition to the argument mentioned above, they also argue that this is caused by the lack of skills of the general partners. It may be difficult for a successful fund to hire partners who are as skilled as the existing partners.
3 Methodology
This section provides an overview of the methodology and data. First, the testable hypotheses are stated. Second, the methodology is presented. Finally, the dataset and the sample selection process is described.
3.1 Hypotheses
The main purpose of this paper is to examine the performance of venture capital and leveraged buyout funds during the latest private equity wave and the financial crisis, and compare it to the performance of the public market indices during the same time period. The aim of this section is to discuss the hypotheses tested in this paper.
One of the problems with measuring the performance of leveraged buyout and venture capital funds is selecting an appropriate benchmark. Anson (2007) argues that the Nasdaq over the counter stock index is the best benchmark for venture capital funds. It ismainly technology and biotech firms thatare listed on the Nasdaq, which, in terms of risk, are
comparable to the young, start-up firms in which venture capital funds participate. Leveraged buyouts funds, on the other hand, often participate in large, well established companies that suffer from some form of inefficiency. Therefore, Anson (2007) argues that a stock market index with larger capitalized stocks, like the S&P 500, is a good benchmark for leveraged buyout funds. The results of table 3 in the appendix show, that venture capital fund returns correlate best with the returns of the Nasdaq, while leveraged buyout fund returns correlate best with the returns of the S&P 500. This paper, therefore, chooses the Nasdaq index as the performance benchmark for venture capital funds and the S&P 500 as the performance benchmark for leveraged buyout funds.
As discussed in the previous section, several studies test the performance of private equity funds and compare it to the returns of a public market index. However, the results are different due to the variable datasets used. For instance, the dataset used in the study by Harris et al. (2013) includes the latest private equity wave and therefore is the best comparable research. They conclude that buyout funds outperform the S&P 500 and report that venture capital funds underperform the S&P 500 in the 2000s. In line with these results, this paper hypothesizes the following:
Hypothesis 1: Leveraged buyout funds generate excess returns over the S&P 500 in the
period from12/31/1998 to 31/12/2013.
Hypothesis 2: Venture capital funds generate lower returns than the Nasdaq in the period
from 12/31/1998 to 31/12/2013.
In addition, several theoretical studies argue that private equity funds underperform the public market indices during economic upturns. Phalippou & Zollo (2005) examine the performance of the whole private equity sector during economic up- or downturns in their empirical study. They come to the conclusion that private equity returns are pro-cyclical. This paper, therefore, hypothesizes the following:
Hypothesis 3: Leveraged buyout funds generate excess returns over the S&P 500 in periods
when the S&P 500 has positive excess returns.
Hypothesis 4: Venture capital funds generate excess returns over the Nasdaq in periods when
the Nasdaq has positive excess returns.
Hypothesis 5: Leveraged buyout funds generate lower returns than the S&P 500 in periods
when the S&P 500 has negative excess returns.
Hypothesis 6: Venture capital funds generate lower returns than the Nasdaq in periods when
the Nasdaq has negative excess returns.
3.2 Methodology
The single factor private equity asset pricing model is used to test the performance of
leveraged buyout and venture capital funds and compare it to the returns of the S&P 500 and the Nasdaq. This model applies the Capital Asset Pricing Model to private equity funds, making the assumption that the returns of private equity funds are related to market risk and therefore are not risk-free, but require adjustment for systematic risk. The historical returns from the private equity funds are regressed on the historical returns of the public market indices. As a result, the regression takes the following form:
𝑟!,!− 𝑟!,! = 𝛼 + 𝛽!∗ 𝑟!,! − 𝑟!,! + 𝜖!,! (1.1) Where:
𝑟!,! is the historical capital weighted average internal rate of return net of fees of the fund at time t.
𝑟!,! is the yield of the 10-year US-treasury bill at time t.
𝑟!,! is the historical internal rate of return of the selected public stock market index at time t. 𝜖!,! is a residual term which is not explained by the data.
𝛽! measures the systematic risk of all the leveraged buyout or venture capital fund returns with respect to the selected public stock market index.
𝛼 measures the risk-adjusted excess return of the private equity funds in comparison to the selected public stock market index returns.
Woodward (2004) shows in her research, that the quarterly returns reported by the general partners come from a mixture of current and “stale” valuations, and thus are not an accurate measure of the true change in value of the portfolio for that given quarter. The presence of staleness in GP valuations implies that the quarterly changes in value reported by GPs are not only related to current returns on the stock market, but also to past returns. This would result in a underestimation of the market related risk (𝛽!) and an overestimation of the excess returns (𝛼) generated by private equity funds. Therefore, a second regression is run, which includes the multi-period pricing effect. The returns of leveraged buyout and venture capital funds in period t are regressed against the contemporaneous returns of the public stock market indices, as well as the returns of the public stock market indices from previous periods t-1, t-2, t-3, t-4 etc. This results in the following regression equation:
𝑟!,!− 𝑟!,! = 𝛼 + ! 𝛽! ∗
!!! 𝑟!!!− 𝑟!,!!! + 𝜖!,! (1.2)
Where:
𝑟!,! is the historical capital weighted average internal rate of return net of fees of the fund at time t.
𝑟!,!!! is the yield of the 10-year US-treasury bill at time t-i, where i=0,1,2,3,4.
𝑟!,!!! is the historical internal rate of return of the selected public market index at time t-i,
𝛽! measures the systematic risk of all the leveraged buyout or venture capital fund returns at time t-i with respect to the selected market index returns at time t-i, where i=0,1,2,3,4. 𝜖!,! is a residual term which is not explained by the data.
𝛼 measures the risk-adjusted excess return of the private equity funds in comparison to the selected public stock market index returns.
The performance of leveraged buyout and venture capital funds during periods when the public stock markets have positive or negative excess returns, is tested using the following methodology. A dummy variable is used to split the dataset in “two states of the world”: the dummy variable has a value of 1 when 𝑟!,!− 𝑟!,! ≥ 0 and a value of 0 when 𝑟!,! − 𝑟!,! < 0. To test the performance of leveraged buyout and venture capital funds during periods when the public stock markets generate positive excess returns, regression (1.2) is run for all data, where the dummy variable equals 1. To test the performance during periods when the stock markets have negative excess returns, regression (1.2) is run for all data, where the dummy variable is 0. This way, we can test the performance of leveraged buyout as well as venture capital funds during up and down public stock markets separately.
3.3 Dataset
Private equity fund performance data is obtained from the Thomson One Private Equity database. This database contains extensive performance data on all private equity funds based in the United States, Europe, Middle East and Africa since 1969. This paper uses performance data of worldwide buyout and venture capital funds in the period from 12/31/1998 to
31/12/2013. This dataset is chosen because it covers the latest private equity wave and the credit crisis of 2008-2009.Only liquidated funds are used in the sample because of the J-curve effect, as explainedin section 2.1. The internal rate of return is used as the performance measure. The IRR is calculated by taking the cash inflows (from LP to fund), distributions (from fund to LP) as well as the residual value, reported by the funds. The returns are net of fees and quarterly reported .
For the purpose of the analysis, the fund returns are compared against the S&P 500, Dow Jones, Russell 3000, Nasdaq and FTSE 100. The performance data of these public stock market is also obtained from the Thomson One Private Equity database. The IRR of the public market comparators is calculated by taking the private equity inflows and outflows as
buying and selling shares into the index. The private equity inflows are treated as the purchase amount into the index, while private equity distributions are treated as the selling off that value in shares in the index. This study is interested in the performance of all leveraged buyout and venture capital funds separately and therefore looks at the capital weighted average returns in a certain period. This capital weighted average accounts for fund size and multiplies each underlying fund IRR by ratio (individual fund size/capitalization of the sample). This way, large funds do not have the same weighting as small firms, which is the case if the average returns are taken.
The returns of private equity funds are not risk free and, hence, require an adjustment, when there is looked at the performance of private equity as an asset class. Berk & DeMarzo (2011) claim that most large firms and financial analysts use the yields of long-term (10- to 30 year) US Treasury bills as an approximation of the risk-free rate. Since private equity
investments usually have a holding period of approximately 10 to 12 years, the yields on 10-year US Treasury bills serve as the risk free rate in this research. These yields in the period from 12/31/1998 to 31/12/2013 are obtained from the Federal Reserve Bank Reports database.
4 Results
This section provides an overview of the results of the statistical analysis. First, the results on the performance of leveraged buyout funds are presented. This is followed by the results on the performance of venture capital funds. Finally, some limitations of this research will be discussed.
4.1 Leveraged buyouts
The summary statistics in table 1 of the appendix show, that leveraged buyouts on average had higher returns than the S&P 500 in the period from 1998 till 2013. However, we have to look at risk adjusted returns when we are comparing the performance of two different asset classes.
Regressing the returns of leveraged buyout funds on the S&P 500 index returns using the single factor capital asset pricing model, to adjust for market risk, leads to an alpha of 2.03 (see table 4). This implies that leveraged buyout funds generate 2.03% risk adjusted excess returns per quarter over the S&P 500. The alpha has a t-value of 2.04 and is therefore statistically significant from zero at a 5% significance level. Hence, in the period from 12/31/1998 till 31/12/2013, leveraged buyout funds generated excess returns over the S&P 500, using the single factor capital asset pricing model. This result is in line with the existing literature by Harris et al. (2013), Phalippou and Gottschalg (2009) and Ljungqvist &
Richardson (2003), who also compare the returns of leveraged buyout funds on value weighted basis to the returns of the S&P 500.
Regressing the returns of leveraged buyouts funds on the returns of the S&P 500 in the contemporaneous period, as well as on the return in the periods t-1, t-2, t-3 and t-4 to correct for the “stale” pricing effect in private equity valuation, results in an alpha of 2.76 (see table 4). This alpha has a t-value of 2.42, which is statistically significant from zero. Therefore, we conclude that leveraged buyout funds generated excess returns of 2.76% per quarter over the S&P 500 in the period from12/31/1998 till 31/12/2013, using the multi-period capital asset pricing model. These results are also similar to the results of Harris et al. (2013), Phalippou and Gottschalg (2009) and Ljungqvist & Richardson (2003), but contradicts the results found by Kaplan & Schoar.
Including multi periods to the single factor capital asset pricing model leads to higher market related risk (sum of all the βs) and higher market adjusted excess returns (𝛼).
Woodward (2004) claims, however, that including multi periods to the single factor
regression leads to higher market related risk (sum of all βs) and lower market adjusted excess returns (𝛼) generated by the private equity funds. This study shows that this is not necessarily the case. These differences with the findings by Woodward (2004) can be the result of the different dataset used. In addition, table 6 shows that none of the lagged betas is significant from zero at a 10% significance level. Thus, there is no evidence for the stale pricing effect in the valuation of leveraged buyout funds. Nonetheless,the multi-period model has a higher adjusted R-squared than the single factor model. This implies that the data is better explained by the expanded model.
Testing the performance of leverage buyout fund in periods when the S&P 500 generated positive excess returns results in an alpha of -3.40 (see table 4). This means that leveraged buyout funds underperformed the S&P 500 by -3.40% per quarter during these times. The alpha, however, has a t-value of -0.89 which implies that alpha is not statistically significant from zero. Therefore, there is not enough evidence to conclude that leveraged buyout funds under- or outperform the S&P 500 in periods when the S&P 500 generates positive excess returns.
During times when the S&P 500 has negative excess returns an alpha of 0.38 (see table 4) is found. This indicates that leveraged buyout funds outperformed the S&P 500 by 0.38% per quarter during these times. A t-value of 0.29 is reported, so the alpha again is not statistically significant from zero. Because of this, no conclusions can be drawn about the performance of leverage buyout fund during times when the S&P 500 generates negative excess returns.
The large difference in performance of leveraged buyout funds during up or down public stock markets, might be the result of the conservative valuation by private equity managers. According to Anson (2007), private equity managers value their portfolios worse than they really are in up stock markets and better than they really are in down stock markets. The results are also in line with the theories by Kaplan & Strömberg (2009) and Kaplan & Schoar (2005), who argue that there will be some overinvesting during economic upturns. Leveraged buyouts will therefore outperform during economic downturns and underperform during economic upturns, according to them.
4.2 Venture Capital
Table 5 in the appendix shows that venture capital funds underperformed the Nasdaq in the period from 12/31/1998 till 31/12/2013, when the single factor capital asset pricing model is applied to venture capital performance. An alpha of -1.15 is found, which implies that venture capital funds underperform the Nasdaq by 1.15% per quarter. This reported underperformance is in line with the studies done by Phalippou & Gottschalg (2009) and Harris et al. (2013). The alpha, however, is not statistically significant from zero with a t-value of -1.06 and a 10% significance level. For this reason, there is not enough evidence to conclude that venture capital funds under- or outperformed the Nasdaq in the period from 12/31/1998 till 31/12/2013.
Expanding the single factor private equity asset pricing model, with multiple periods to correct for the stale pricing effect in private equity valuation results in an alpha of -0.25 (see table 5). Thus, venture capital funds underperforms the Nasdaq by -0.25% per quarter, which is also in line with prior research of Phalippou & Gottschalg (2009) and Harris, et al. (2013). The alpha again is not statistically significant from zero with a t-value of -0.23, so no conclusions can be drawn.
Including multiple periods to the single factor capital asset pricing model leads to an increase of the adjusted R-squared, which means more data is explained by the expanded model. Table 7 reports that the lagged β2, which measures the systematic risk of the venture
capital returns at time t with respect to the Nasdaq returns at time t-1, is statistically significant from zero at a 5% significance level. The lagged betas, β3 and β4, are also
statistically significant, but at a 10% significance level. β5, on the other hand is, is not
statistically significant. Thus, market returns from 3 quarters back, have significant influence on the returns of venture capital funds. These results show therefore, that there is enough evidence for the stale pricing effect in venture capital fund valuation and correspond to the results of Woodward (2004). However, the reported alpha and the sum of the betas for venture capital funds are different from what Woodward’s (2004) theory expects. We find higher market risk when we correct for the stale pricing effect, as Woodward (2004) predicts, but a less negative alpha in comparison to the single factor model results. These differences in value of the coefficients can be the result of the different datasets used.
Table 5 in the appendix shows that venture capital funds outperform the Nasdaq in periods when the Nasdaq generates positive excess returns. An alpha of 1.35 is reported, which means that venture capital funds outperformed the Nasdaq by 1.35% per quarter. This
alpha is not statistically significant from zero, so there is not enough evidence to conclude that venture capital funds under or outperform the Nasdaq during these times.
Venture capital funds also outperform the Nasdaq in periods when the Nasdaq has negative excess returns. The reported outperformance during these times is 1.35% per quarter. The alpha again is not statistically significant from zero, so no further conclusions can be drawn.
4.3 Limitations
There are some limitations to this research. First of all, the data acquired from the Thomson One database is self-reported by the private equity funds and for this reason potentially subject to positive selection bias. This occurs due to the fact that that the performance of private equity funds is not publically available. Well performing funds will report
performance data earlier to Thomson One than the worse performing funds. This could have influenced the results.
To test the performance of leveraged buyout and venture capital funds in comparison to the public market returns, this paper applies the capital asset pricing model to private equity fund performance. The capital asset pricing model assumes that the returns of private equity funds are related to market risk and does not take into account other factors that may
influence the performance of private equity funds. The model assumes that all investors can buy and sell all their securities, at competitive market prices at any time. Investments in private equity, however, are extreme illiquid. This could have influenced the returns of private equity funds because investors want to be compensated for this type of risk. Another factor that could have influenced the returns to private equity is the size of the funds. Larger funds might have performed betterthan smaller firms. Larger firms might diversify their portfolio better, because they have easier access to capital. Therefore, the size of the funds could also have asignificant influence on the performance of private equity funds.
Testing the performance of private equity funds and including these two factors to the model, could give different results and is a suggestion for future research.
5 Conclusion
This paper analyzes the performance of venture capital and leveraged buyout funds during the latest private equity wave and the financial crisis, and compares it to the performance of the public stock markets during the same time period.
The results show that leveraged buyout funds outperformed the S&P 500 in the period from 1998 till 2013 by 2.03% per quarter. Corrected for the “stale” pricing effect in private equity valuation, the outperformance was 2.76% per quarter. These results are in line with the results found in earlier research by Harris et al. (2013), Phalippou & Gottschalg (2009) and Ljungqvist & Richardson (2003). Leveraged buyout funds underperform the S&P 500 by -3.40% in periods when the S&P 500 generates positive excess returns. They outperform the S&P 500 by 0.38% in periods when the S&P 500 reports negative excess returns. Both alphas, however, are not statistically significant from zero. The large differences between the
performance in up or down public stock markets can be the result of the rule of conservatism in private equity valuation by the general partners, mentioned by Anson (2007).
Venture capital funds underperform the Nasdaq index by -1.15% per quarter. Corrected for the “stale” pricing effect the underperformance is -0.24% per quarter. These alphas, however, are not statistically significant from zero. Therefore,there is not enough evidence to conclude that venture capital funds out- or underperformed the Nasdaq in the period from 1998 till 2013. During periods when the Nasdaq has positive excess returns, an outperformance of 1.35% is reported. Venture capital funds outperform the Nasdaq in periods when the Nasdaq generates negative excess returns by 0.73% per quarter. Again, both alphas are not statistically significant from zero.
Expanding the single factor private equity asset pricing model with multiple periods to correct for the stale pricing effect leads to higher market related risk (sum of β’s), but also to higher market adjusted excess returns (α). These findings are remarkable because Woodward (2004) predicts that correcting for the “stale” pricing effect in private equity valuation results in higher market related risk and lower market adjusted excess returns. In addition, the results of this paper show that there is no evidence for the stale pricing effect in leveraged buyout fund valuation. However, there is enough evidence to conclude that market returns from 3 quarters back have significant influence on the returns of venture capital funds and hence there is a stale pricing effect in venture capital fund valuation.
Historical performance is no guarantee for future performance. However, the results of this paper could give investors more information about whether or not to invest in leveraged buyout and venture capital funds.
6 Bibliography
Anson, M. (2007). Performance Measure in Private Equity : Another Look. The Journal of
Private Equity, 7–22.
Asness, C. S., Krail, R. J., & Liew, J. M. (2001). Do Hedge Funds Hedge? The Journal of
Portfolio Management, 28(1), 6–19. doi:10.3905/jpm.2001.319819
Berk, J., & DeMarzo, P. (2011). Corporate Finance (Second Edi., pp. 381–382). Harlow: Pearson Education Limited.
Cumming, D. (2010). Private Equity: Fund Types, Risks and Returns and Regulation. Hoboken, New Jersey: John Wiley & Sons, Inc.
Fraser-Sampson, G. (2010). Private equity as an asset class. Padstow: John Wiley & Sons, Ltd.
Harris, R. S., Jenkinson, T., & Kaplan, S. N. (2013). Private Equity Performance : What Do We Know ? Journal of Finance, Forthcoming.
Higson, C., & Stucke, R. (2012). The Performance of Private Equity. SSRN Electronic
Journal. doi:10.2139/ssrn.2009067
Jegadeesh, N., Kräussl, R., & Pollet, J. (2009). Risk and Expected Returns of Private Equity Investments : Evidence Based on Market Prices. NBER Working Paper, 1–44.
Jensen, M. C. (1989). Eclipse of the public corporation. Harvard Business Review, 17(3), 560–593.
Kaplan, S. N., & Strömberg, P. (2009). Leveraged Buyouts and Private Equity. Journal of
Economic Perspectives, Winter, 121–146.
Kaplan, S., & Schoar, A. (2005). Private Equity Performance : Returns , Persistence and Capital Flows. Journal of Finance, 60(4), 1791–1823.
Lerner, J., Sorensen, M., & Strömberg, P. (2011). Private Equity and Long-Run Investment : The Case of Innovation. Journal of Finance, 66(2), 445–447.
Ljungqvist, A., & Richardson, M. (2003). The Cashflow, Return, and Risk Characteristics of Private Equity, Working Paper No. 9495, NBER.
Phalippou, L., & Gottschalg, O. (2009). The Performance of Private Equity Funds. Review of
Financial Studies, 22(4), 1747–1776.
Phalippou, L., & Zollo, M. (2005). What Drives Private Equity Fund Performance ?
Unpublished Working Paper.
Woodward, S. E. (2004). Measuring Risk and Performance for Private Equity. Sand HIill
Appendix
Table 1: Summary statistics: all leveraged buyout funds.
This table shows the summary statistics of the returns for all variables used in this paper. All returns are quarterly reported and net of fees in the period from 12/31/1998 till 31/12/2013. Return LB (in %) is the IRR of all leveraged buyout funds. This IRR is calculated by taking the cash inflows (from LP to fund), distributions (from fund to LP) as well as the residual value, reported by the funds. S&P 500, Dow Jones, Russell 3000, Nasdaq and FTSE 100 are the returns (in %) of each index. These IRR’s are calculated by taking the leveraged buyout fund inflows and outflows as buying and selling shares into the index. The leveraged buyout fund inflows are treated as the purchase amount into the index, while leveraged buyout fund distributions are treated as the selling off that value in shares in the index.This causes the differences in the market returns for leveraged buyout and venture capital funds.
Mean (1) Std. Dev. (2) Min (3) Max (4) N (5) Return LB 4.283 9.728 -‐11.715 44.082 60 S&P 500 1.626 8.755 -‐21.941 16.014 60 Dow Jones 2.151 8.211 -‐18.913 14.975 60 Russell 3000 1.655 8.825 -‐23.269 16.238 59 Nasdaq 2.135 14.193 -‐32.618 48.309 60 FTSE 100 1.638 9.886 -‐26.336 26.176 60 All data are obtained from Thomson One.
Table 2: Summary statistics: all venture capital funds.
This table shows the summary statistics of the returns for all variables used in this paper. All returns are quarterly reported and net of fees in the period from 12/31/1998 till 31/12/2013. Return VC (in %) is the IRR of all venture capital funds. This IRR is calculated by taking the cash inflows (from LP to fund), distributions (from fund to LP) as well as the residual value, reported by the funds. S&P 500, Dow Jones, Russell 3000, Nasdaq and FTSE 100 are the returns (in %) of each index. These IRR’s are calculated by taking the venture capital fund inflows and outflows as buying and selling shares into the index. The venture capital fund inflows are treated as the purchase amount into the index, while venture capital fund distributions are treated as the selling off that value in shares in the index. This causes the differences in the market returns for leveraged buyout and venture capital funds.
Mean (1)
Std. Dev. (2) Min (3) Max (4) (5) N Return VC 2.167 9.779 -‐17.333 47.103 60 S&P 500 1.567 8.797 -‐22.167 16.111 60 Dow Jones 2.092 8.277 -‐18.904 14.959 60 Russell 3000 1.596 8.876 -‐23.515 16.345 59 Nasdaq 2.137 14.205 -‐32.578 47.438 60 FTSE 100 1.937 9.748 -‐26.535 26.299 59 All data are obtained from Thomson One.
Table 3: Correlations and R2.
This table reports the correlations between the returns of all leveraged buyouts funds as well as all venture capital fund returns with the returns of the S&P 500, Dow Jones, Russell 3000, Nasdaq and FTSE 100. It also reports the R2 of the single factor capital asset pricing model. The returns of the
leveraged buyout or venture capital funds are regressed on the returns of each public stock market index in this model.
Leveraged Buyouts Venture Capital
R2 (1) Correlation (2) R 2 (3) Correlation (4) S&P 500 0.4539 0.6424 0.2021 0.3474 Dow Jones 0.3820 0.5834 0.1250 0.2164 Russell 3000 0.4182 0.6235 0.1828 0.3702 Nasdaq 0.2315 0.4388 0.2998 0.4645 FTSE 100 0.4264 0.6243 0.1309 0.3135
Table 4: Results all leveraged buyout funds.
This table shows the results of the 4 different OLS regression, where the returns of all leveraged buyout funds on a capital weighted average are compared to the returns of the S&P 500 in the period from 12/31/1998 till 31/12/2013. The returns of all leveraged buyout funds are regressed on the returns of the S&P 500, using the single period capital asset pricing model in model (1). Model (2) is the multi-period capital asset pricing model, where the historical returns of leveraged buyout funds in period t are regressed on the contemporaneous returns of the S&P 500, as well as the returns of the S&P 500 from previous periods t-1, t-2, t-3, t-4. Model (3) is the multi-period capital asset pricing model in periods when the S&P 500 generated positive excess returns. Model (4) is the multi-period capital asset pricing model in periods when the S&P 500 generated negative excess returns.
(1) (2) (3) (4) α 2.031** (0.106) 2.761** (1.143) (3.839) -‐3.404 (1.344) 0.384 Β1 0.739*** (0.998) 0.784*** (0.110) 2.027*** (0.557) 0.473*** (0.148) Β2 0.062 (0.112) (0.201) 0.103 (0.116) 0.137 Β3 0.002 (0.110) (0.211) 0.016 (0.105) 0.085 Β4 0.102 (0.109) (0.238) 0.468* (0.093) 0.046 Β5 0.043 (0.110) (0.223) -‐0.139 (0.098) 0.054 N 60 56 23 33 Adj.-‐R2 0.445 0.482 0.313 0.452
Note: ***, **, and * indicate statistical significance at the 1%, 5% and 10% (two-tail) test levels, respectively.
Table 5: Results all venture capital funds.
This table shows the results of the 4 different OLS regression, where the returns of all venture capital funds on a capital weighted average are compared to the returns of the Nasdaq in the period from 12/31/1998 till 31/12/2013. The returns of all venture capital funds are regressed on the returns of the Nasdaq, using the single period capital asset pricing model in model (1). Model (2) is the multi-period capital asset pricing model, where the historical returns of all venture capital funds in period t are regressed against the contemporaneous returns of the Nasdaq, as well as the returns of the Nasdaq from previous periods t-1, t-2, t-3, t-4. Model (3) is the multi-period capital asset pricing model in periods when the Nasdaq generated positive excess returns. Model (4) is the multi-period capital asset pricing model in periods when the Nasdaq generated negative excess returns.
(1) (2) (3) (4) α -‐1.153 (1.085) -‐0.247 (1.098) 1.347 (4.056) 0.731 (1.108) Β1 0.373*** (0.075) 0.363*** (0.077) (0.502) 0.074 0.469*** (0.072) Β2 0.151** (0.069) (0.147) 0.243 (0.054) 0.029 Β3 0.128* (0.070) (0.161) 0.124 (0.050) 0.093* Β4 0.138* (0.070) (0.250) 0.083 0.139*** (0.043) Β5 0.119 (0.071) (0.151) 0.163 (0.053) 0.091* N 60 56 26 30 Adj.-‐R2 0.288 0.3732 -‐0.0021 0.7286
Note: ***, **, and * indicate statistical significance at the 1%, 5% and 10% (two-tail) test levels, respectively.
Table 6: Lagged pricing effect in leveraged buyout fund returns.
This table shows the results of including the lagged pricing effect to the single period capital asset pricing model for leveraged buyout funds quarterly returns in the period from 12/31/1998 till 31/12/2013. Model (1) is the single period capital asset pricing model where the returns of the
leveraged buyout funds are regressed on the returns of the S&P 500 in the contemporaneous period. In model (2) is an additional period added to model (1). The returns of the leveraged buyout funds are now regressed on the returns of the S&P 500 in the contemporaneous period as well as on the market returns in period t-1. In model (3) are the returns of the leveraged buyouts funds regressed on the returns of the S&P 500 in periods t, t-1 and t-2. In model (3) are the returns of the leveraged buyouts funds regressed on the returns of the S&P 500 in periods t, t-1, t-2, t-3. And finally in model (4) are the returns of the leveraged buyouts funds regressed on the returns of the S&P 500 in periods t, 1, t-2, t-3, t-4. . (1) (2) (3) (4) (5) α 2.031** (0.998) 2.312** (1.042) 2.224** (1.097) 2.279** (1.131) 2.761** (1.143) Β1 0.739*** (0.106) 0.719*** (0.109) 0.714*** (0.110) 0.739*** (0.111) 0.784*** (0.110) Β2 0.126 (0.110) 0.122 (0.113) 0.105 (0.113) 0.062 (0.112) Β3 0.007 (0.112) -‐0.013 (0.112) 0.002 (0.110) Β4 0.095 (0.112) (0.109) 0.102 Β5 0.043 (0.110) N 60 59 58 57 56 Adj.-‐R2 0.445 0.447 0.434 0.447 0.482 Note: ***, **, and * indicate statistical significance at the 1%, 5% and 10% (two-tail) test levels, respectively.
Table 7: Lagged pricing effect in venture capital fund returns.
This table shows the results of including the lagged pricing effect to the single period capital asset pricing model for venture capital fund quarterly returns in the period from 12/31/1998 till 31/12/2013. Model (1) is the single period capital asset pricing model where the returns of the venture capital funds are regressed on the returns of the Nasdaq in the contemporaneous period. In model (2) is an
additional period added to model (1). The returns of the venture capital funds are now regressed on the returns of the Nasdaq in the contemporaneous period as well as on the market returns in period t-1. In model (3) are the returns of the venture capital funds regressed on the returns of the Nasdaq in periods t, t-1 and t-2. In model (3) are the returns of the venture capital funds regressed on the returns of the Nasdaq in periods t, t-1, t-2, t-3. And finally in model (4) are the returns of the venture capital funds regressed on the returns of the Nasdaq in periods t, t-1, t-2, t-3, t-4.
(1) (2) (3) (4) (5) α -‐1.153 (1.085) (1.074) -‐0.857 (1.069) -‐0.684 (1.065) -‐0.454 (1.098) 0.247 Β1 0.373*** (0.075) 0.360*** (0.073) 0.349*** (0.072) 0.355*** (0.069) 0.363*** (0.077) Β2 0.169** (0.073) 0.157** (0.072) 0.146** (0.070) 0.151** (0.069) Β3 0.148** (0.072) 0.136* (0.071) 0.128* (0.070) Β4 0.147** (0.070) 0.138* (0.070) Β5 0.119 (0.071) N 60 59 58 57 56 Adj.-‐R2 0.228 0.3354 0.3677 0.4045 0.3732
Note: ***, **, and * indicate statistical significance at the 1%, 5% and 10% (two-tail) test levels, respectively.
