Essays in Private Equity

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

By Ji-Woong Chung Graduate Program in Business Administration

The Ohio State University 2010

Dissertation Committee: Professor Isil Erel Professor Berk A. Sensoy Professor Michael S. Weisbach, Advisor

Copyright by Ji-Woong Chung 2010

Abstract

This first essay, Leveraged Buyouts of Private Companies, studies the motivations and the consequences of leveraged buyouts of privately held companies. Over the last two decades, the number (enterprise value) of leveraged buyout transactions involving privately held targets totals 10,013 ($855 billion), accounting for 46% (21%) of the worldwide leveraged buyout market. Yet the vast majority of academic studies focus on the buyouts of publicly held targets. This chapter investigates the effects of leveraged buyouts on privately held targets. I find that, unlike the corporate restructuring process of public firms after the buyouts, private targets sponsored by private equity firms grow substantially after the buyouts. The overall evidence suggests that private equity firms, through leveraged buyouts, facilitate private targets growth by alleviating targets investment constraints.

In the second essay, Incentives of Private Equity General Partners from Future Fundraising which is co-authored with Berk Sensoy, Lea Stern, and Mike Weisbach, we model and estimate the total incentives facing private equity general partners. Incentives from the explicit fee structure (two and twenty) of private equity funds understate the actual incentives facing private equity general partners because they ignore the rewards stemming from the effect of current performance on the ability to raise larger funds in the future. We evaluate the importance of these implicit incentives in the context of a ii

learning model in which investors use current performance to update their assessments of a general partners ability, and, in turn, decide how much capital to allocate to the partners next fund. Our estimates suggest that implicit incentives from expected future fundraising are about as large as explicit incentives from carried interest in the current fund. This implies that the performance-sensitive component of revenue is about twice as large as suggested by previous estimates based only on explicit fees. Consistent with the model, we find that these implicit incentives are stronger when abilities are more scalable and weaker when current performance is less informative about ability. Overall, the results suggest that implicit incentives from future fundraising have a substantial impact on general partners welfare and are likely to be an important factor in the success of private equity firms.

In the last chapter, I study performance persistence in the private equity industry. Contrary to what has been known in the literature, I find that performance persistence in private equity is short-living. Current fund performance is positively and significantly associated with the first follow-on fund performance, but not with the second or third follow-on funds. Even the statistically significant association between two consecutive funds performance is not economically large. The returns of the best performing quartile portfolio drops by about half, and those of the worst performing portfolio improve substantially from one fund to the next fund. There is no difference in the performance of the second (and after) follow-on funds of current top and bottom performing quartile portfolios. Performance converges in the long run. The commonality of relevant market iii

conditions between two consecutive funds largely explains performance persistence. Also, excessive fund growth conditional on past performance erodes performance and reduces persistence.

iv

Dedication

To my parents, Dong-Jo Chung and Wol-Sun Kim

Acknowledgments

I am grateful to my advisor, Mike Weisbach, and the members of my committee, Isil Erel and Berk Sensoy, for their constant encouragement, support and guidance. I also thank my colleague Ph.D. students for their companionship and having spent countless hours discussing with me on various matters, and the faculty in the Department of Finance for their guidance. Lastly, I thank the Fisher College of Business for providing generous financial support.

vi

Vita

2003................................................................B.A. Economics and Applied Statictics, Yonsei University, South Korea 2003-2005 ......................................................Graduate Associate, Department of Business Administration, Yonsei University, South Korea 2005 to present ..............................................Graduate Associate, Department of Finance, The Ohio State University

Fields of Study

Major Field: Business Administration

vii

Table of Contents

Abstract ............................................................................................................................... ii Dedication ........................................................................................................................... v Acknowledgments.............................................................................................................. vi Vita.................................................................................................................................... vii Fields of Study .................................................................................................................. vii Table of Contents ............................................................................................................. viii List of Tables .................................................................................................................... xii List of Figures .................................................................................................................. xiv Chapter 1: Leveraged Buyouts of Private Companies ....................................................... 1 1.1. Introduction .............................................................................................................. 1 1.2. Hypothesis development .......................................................................................... 7 1.3. Data and summary statistics ..................................................................................... 9 1.3.1. Data sources and some institutional background .............................................. 9 1.3.2. Sample selection .............................................................................................. 13 1.3.3. Construction of control sample........................................................................ 14 viii

1.3.4. Ownership structure of private targets............................................................. 16 1.4. Post-buyout growth of target firms ........................................................................ 18 1.5. Analysis of Deal rationales .................................................................................... 21 1.6. Pre-buyout investment constraints and post-buyout growth .................................. 23 1.7. Operating performance after buyouts ..................................................................... 25 1.8. Conclusion.............................................................................................................. 27 Chapter 2: Incentives of Private Equity General Partners from Future Fundraising ........ 29 2.1. Introduction ............................................................................................................ 29 2.2. Model ..................................................................................................................... 36 2.2.1. Setup ................................................................................................................ 37 2.2.2. Cross-sectional implications ............................................................................ 38 2.2.3. Lifetime compensation of GPs ........................................................................ 42 2.3. Data ........................................................................................................................ 47 2.4. The Empirical Relation between todays Returns and Future Fundraising ........... 52 2.4.1 Calculating indirect incentives for different types of funds ............................. 52 2.4.2. Indirect incentives of older and younger partnerships..................................... 55 2.4.3. Indirect incentives and fund size ..................................................................... 56 2.5. General Partner Incentives Implied by the Regression Estimates.......................... 56 2.5.1. Basic results ..................................................................................................... 56 ix

2.5.2. Indirect incentives over the partnerships life ................................................. 60 2.5. Discussion and Conclusion .................................................................................... 62 Chapter 3: Performance Persistence in Private Equity Funds .......................................... 65 3.1. Introduction ............................................................................................................ 65 3.2. Data ........................................................................................................................ 70 3.3. Testing Performance Persistence ........................................................................... 72 3.3.1. Transitional Probabilities................................................................................. 74 3.3.2. Correlation between current fund performance and follow-on fund performance ............................................................................................................... 76 3.3.3. Multivariate regression .................................................................................... 77 3.3.4. Subsequent performance of initial performance quartile ................................ 80 3.3.5. Robustness of the results ................................................................................. 83 3.4. Why (Not) Performance Persists? .......................................................................... 84 3.4.1. Fund flows and fund performance ................................................................... 84 3.4.2. The effect of fund flows on performance persistence ..................................... 87 3.4.3. The effect of time gap on performance persistence ......................................... 89 3.4.4. Common market conditions............................................................................. 91 3.5. Conclusion.............................................................................................................. 95 References ......................................................................................................................... 97 x

Appendix A: Tables for Chapter 1 .................................................................................. 110 Appendix B: Tables for Chapter 2 .................................................................................. 124 Appendix C: Tables for Chapter 3 .................................................................................. 141

xi

List of Tables

Table A.1. Distribution of leveraged buyout transactions in the U.K .............................111 Table A.2. Distribution of final sample of leveraged buyouts .........................................113 Table A.4. Logistic regression to predict the likelihood of being a leveraged buyout target .................................................................................................115 Table A.5. Ownership characteristics of privately held targets .......................................116 Table A.6. Changes in firm growth after leveraged buyouts of public targets ................117 Table A.7. Changes in firm growth after leveraged buyouts of private targets with private equity................................................................................................................................118 Table A.8. Acquisitions and disposals of businesses and operations after leveraged buyouts .............................................................................................................................119 Table A.9. Pre-buyout investment constraints and post-buyout growth ..........................120 Table A.10. Operating performance after a buyout: Private equity sponsored targets ....121 Table B.1. Descriptive Statistics ......................................................................................125 Table B.2. Committed Capital by Type of Fund and Fund Sequence .............................126 Table B.3. Fund Growth ..................................................................................................128 Table B.4. Fund Performance and Time between Successive Funds ..............................129 Table B.5. Future Fundraising and Current Performance ................................................130 Table B.6. Future Fundraising, Current Performance, and Fund Sequence ....................132 Table B.7. Future Fundraising, Current Performance, and Fund Size.............................133 xii

Table B.8. Future Revenue and Current Performance .....................................................134 Table B.9. Future Revenue, Current Performance and Fund Sequence ..........................137 Table C.1. Private equity fund performance and fund raising by vintage year ...............142 Table C.2. Transition probabilities from current funds performance quartiles to followon funds performance quartiles ......................................................................................144 Table C.3. Pearson and Spearman correlations between current fund performance and follow-on fund performance ............................................................................................145 Table C.4. Cross sectional regression of current performance on past performance ......146 Table C.5. Subsequent fund performance (unadjusted IRRs) by quartile portfolios based on current fund performance ............................................................................................147 Table C.6. Current fund performance and follow-on fund growth ..................................150 Table C.7. Fund growth and follow-on fund performance ..............................................151 Table C.8. The effects of fund growth and time gap on performance persistence ..........152 Table C.9. The effects of similar market conditions on performance persistence ...........153

xiii

List of Figures

Figure A.1. Typical corporate structure after a buyout. ...................................................122 Figure B.1. Importance of incentives from future fundraising over the partnership's life ..........................................................................................................................................140 Figure C.1. Private equity fund performance and fund raising by vintage year. .............161 Figure C.2. Performance of quartile portfolios ranked on current fund performance. ....162 Figure C.3. Cash flows of a private equity fund over its life. ..........................................163

xiv

Chapter 1: Leveraged Buyouts of Private Companies

1.1. Introduction This paper examines the effects of leveraged buyouts on the investment and performance of privately held targets. The leveraged buyout market for privately held firms is large. Over the last two decades there have been more than 10,000 acquisitions of private firms through leveraged buyouts, for an aggregate deal value exceeding $850 billion (Strmberg, 2007). 1 Despite the importance of these transactions in terms of the frequency and the size, academic studies have devoted little attention to private-to-private transactions. 2 Importantly, the economic forces driving these private-to-private leveraged buyouts are likely to be distinctively different from those driving public-to-private buyouts. In the existing academic literature, agency theory of free cash flows (Jensen, 1986, 1989, 1993) has been the important theoretical grounds to understand the motivations and the effects of leveraged buyouts of public firms. Previous studies document that firms with abundant free cash flows and low investment opportunities are more likely to engage in leveraged buyouts (Lehn and Poulsen, 1989, Opler and Titman, 1991, Long and Ravenscraft, 1993, Dittmar and Bharath, 2009), and that target firms reduce capital expenditures and actively
By comparison, during the same period, the number of buyouts of publicly traded firms is 1,398 with a total enterprise value of approximately $1.1 trillion. 2 This undue emphasis on public-to-private buyouts may be ascribed to several high profile deals involving public companies such as RJR Nabisco ($31.1b in 1988), Beatrice ($6.1b in 1985), and, more recently, HCA ($32.7b in 2006), and TXU ($43.8b in 2007) and to the lack of publicly available financial data for privately held targets and gone private companies through leveraged buyouts in the U.S.
1

sell assets or divisions after buyouts (Kaplan, 1988, Smith, 1989, Wiersema et al. 1995, Dennis, 1994, Magowan, 1989, Chevalier, 1995, Aslan and Kumar, 2009). 3 These findings are typically interpreted as being consistent with the view that buyouts reduce free cash flow problems by reversing previously made inefficient investments or acquisitions. However, the agency view cannot explain the leveraged buyouts of private firms (Wright et al. 2000) because it is less likely that private firms suffer from agency problems due to their concentrated ownership structure. One potential explanation for the motivations of the leveraged buyouts of private firms is that private equity firms may be able to improve a targets value by mitigating inefficiencies coming from various investment constraints facing small private firms. In fact, unlike the image reflected in the media as asset-strippers, private equity firms often claim that they take this growth strategy to help the target firms grow and increase firm value. This paper finds evidence supporting this view. I find that privately held targets substantially grow after the buyouts led by private equity firms: assets, sales, capital expenditures, and the number of employees all increase. This finding stands in stark contrast to what the academic literature has documented regarding corporate restructuring process after leveraged buyouts. I investigate a sample of 1,009 buyouts in the U.K. between 1997 and 2006, 887 of which involve privately held targets. The U.K. market provides two advantages: first, the stringent disclosure and financial reporting environment in the U.K. allow observation of
Also managerial compensation is restructured to align the interests of managers with owners, and high leverage and close monitoring by investors reduce inefficient resource wastes (Baker 1992, Baker and Wruck, 1989). As a result, after leveraged buyouts, operating performance and plant productivity improve (Kaplan, 1988, Smith, 1989, Litchenberg and Siegel, 1989, Muscarella and Vetsuypens, 1990, among others).
3

the characteristics of privately held targets and what these companies actually do in the post-buyout period under private ownership. 4 Second, the U.K. leveraged buyout market is, after the U.S., the second largest in the world, making it possible to examine a large sample of buyouts. In the sample of privately held targets, I first document that the average (median) ownership of the largest owner prior to buyout is 87% (100%) and the vast majority (95%) of these owners are also managers of the companies. Therefore, it is unlikely that private targets suffer from the same kinds of agency problems as public companies. I also find that, after buyouts led by private equity firms, privately held targets, unlike publicly traded targets, increase assets, sales, employment, and capital expenditures. For example, from one year prior to the third year after the buyout, the industry adjusted total assets increase by 94% for private targets of private equity firms and the value decreases by 25% for public targets. Similarly, the median value of industry-adjusted capital expenditures to sales ratio increases by 18% for private targets. In contrast, the industryadjusted capital expenditures to sales ratio decreases by 5% for public targets over the same period. Private targets also make substantial acquisitions under private equity ownership: The cash outflows (inflows) associated with acquisitions (disposals) to tangible fixed assets ratio is 0.644 (0.000) in private targets and 0.039 (0.026) in public targets. However, these findings can be driven by selection bias. In other words, private equity firms may be acquiring private firms which could have grown even without private
4

U.K. company laws require all limited liability companies (both private and public) to file periodic reports with the Companies House. See the U.K. Companies House for the Companies Act.

equity led leveraged buyouts. Indeed, I find that targets of private equity sponsored buyouts are more profitable and experience faster growth prior to buyout than industry peer private firms, which suggests that private equity firms do select particular types of private firms. To examine whether a targets growth is more of a selection effect or a treatment effect by private equity firms, I compare the investment and growth of the target firms of private equity with the carefully selected three sets of control sample firms. First set of control firms is private firms which underwent leveraged buyouts without private equity firms involvement. This is a natural set of benchmark because these private firms were actually put up for sale, experienced similar ownership changes, and could have been targets of private equity. Second, I construct non-LBO target private firms which have similar characteristics as private equity led LBO target firms using propensity score matching. Lastly, I compare targets growth with industry median growth, following previous studies (Kaplan, 1989, Smith, 1990). Comparing with industry median will give a general idea about how the target firms are different from other industry peer firms. I find that the targets growth is substantially larger than the growth of other benchmark private firms. Overall, the evidence suggests that even though private equity firms select targets with growth potential, they do help target firms grow and expand post-buyout. To further understand whether private equity firms through leveraged buyouts mitigate investment constraints facing small private targets, I examine the relationship between pre-buyout investment constraints and post-buyout growth. First, I test whether more financially constrained targets in terms of size, age, and leverage experience larger post4

buyout growth. Second, I investigate whether owner-managers risk aversion measured by ownership concentration and age is associated with post-buyout growth. Lastly, I examine whether lack of operational expertise is related to larger post-buyout growth. I assume that if a private firm is managed by owner-managers, the firm does not have access to outside professional management skills. Though I find that more constrained firms tend to experience larger post-buyout growth, the statistically significance of the relationship is not strong, making it harder to draw a strong implication from this analysis. Finally, I examine the operating performance of private targets post-buyout to see how private equity firms growth strategy is associated with post-buyout performance. I find that private targets with private equity sponsors experience an increase in operating performance: industry-adjusted EBITDA increases by 12% during the first three years post-buyout. Not surprisingly, the industry-adjusted ratio of operating income to sales drops by 35%. The reason is because the rate of sales growth exceeds that of EBITDA after buyouts. This pattern implies that buyouts with private equity sponsors result in growth but not improved margins. This deterioration of operating efficiency could be the result of worse investments on the part of private equity firms. However, it could also be that private equity firms are increasing, i.e., optimizing, investments by taking positive NPV but less profitable projects which were not previously exploited prior to buyouts due to investment constraints. Therefore, we observe the decrease in the average profitability of private targets over time after the buyouts. Though the evidence in this paper cannot distinguish between these two hypotheses, it is unlikely that private equity 5

firms are making unprofitable and inefficient investments in light of previous studies. For example, Sheen (2008) finds that private firms which underwent private equity led leveraged buyouts are more likely than public firms to make an efficient investment in response to the expected demand shock in chemical industry. Also, Bargeron et al. (2008) finds that private equity firms tend to pay less acquisition premium than public firms when acquiring a similar target firm, which suggests that private equity firms tend to make investments in a most cost saving way. To my knowledge, this is the first academic study to examine the effects of leveraged buyouts of private firms, transactions which account for a large fraction of the leveraged buyout market. Importantly, I show the importance of private equity sponsors and leveraged buyouts in alleviating the investment constraints facing private firms. With the existing findings on the buyouts of public firms, the overall evidence suggests that private equity firms attempt to reorganize target firms in a way which reduces inherent the targets inefficienciesagency problems in public targets and investment constraints in private ones. A closely related study to this paper is Boucly et al. (2009). From the examination of French leveraged buyouts transactions, they document substantial growth in assets, sales, and employment. They interpret that private equity and leveraged buyouts can provide niche financing for credit-constrained firms in countries with under-developed capital market. Though similar in spirit, the evidence in this paper suggests that, even in the U.K. with relatively well developed financial market, private equity firms take growth strategy to improve firm value by alleviating investment constraints of small private targets. 6

The paper proceeds as follows: Section 1 develops hypothesis. Section 2 describes data sources and sample selection. It also presents summary statistics on the deal characteristics of private and public target firms. Section 3 investigates post-buyout restructuring processes. Section 4 analyzes deal rationales. Section 5 examines whether pre-buyout investment constraints are associated with the post-buyout growth of private targets. Section 6 studies the operating performance of target firms after buyouts, and Section 7 concludes.

1.2. Hypothesis development In private companies, entrepreneurs or owners usually serve as the managers of their company, or the ownership structure is highly concentrated in the hands of a few, such as founders, angel investors, and venture capitalists. These owners perform close monitoring on the management and managerial incentive mechanisms are tightly structured to protect owner wealth from manager expropriation (e.g. Sahlman, 1990, Kaplan and Strmberg, 2003). Therefore, the agency problems associated with incentive misalignment between owners and managers are unlikely to be found in private companies. Consequently, the elimination of agency costs of free cash flows (Jensen, 1986, 1993) is less likely to be an important reason for leveraged buyouts for a private company as it is for a public firm. Private firms, however, tend to have limited or costly access to public resources imposed by concentrated ownership structure and information asymmetry. All else equal, when facing an investment opportunity, managers of private companies are less able or willing to implement the investment. Owners with substantially undiversified wealth tied up in 7

the firm may not want the extra risks associated with the new investment. Further, the existing managers may not possess the intimate knowledge and expertise necessary to execute the new investment, and the firm may not have the financial resources to take advantage of new investment opportunities. Often owners of private firms tend to have different goals, such as preserving wealth, keeping the business stable, maintaining stable income flows, and providing employment opportunities for their descendants. Private equity firms, through leveraged buyouts, can alleviate these investment constraints by providing the owners with a whole or partial exit (thereby lowering the ownership of and reducing the risk exposure to the owners). Private equity firms sponsoring these transactions reduce information uncertainty of target firms through due diligence and their reputation in the capital markets. Private equity firms also import advanced management skills (e.g. operational knowledge and expertise on the corporate control market) and industry and regional networks into the target companies. In addition, private equity firms, being professional investment firms which manage several portfolio companies, are more risk-tolerant than the individual owners and better positioned to take risky investments. In contrast, a public company taken private through a leveraged buyout is less likely to be resource constrained. One of the main reasons that a firm goes public is to tap the public capital market and to exploit current and future investment opportunities (e.g. Kim and Weisbach, 2007). Hence, a public company may pursue a leveraged buyout and go private because it no longer needs public resources. Public status also provides an opportunity to engage in mergers and acquisitions (which is a major investment for a 8

company), either by creating shares that serve as a currency for acquisitions (Brau, Francis, and Kohers, 2005, Brau and Fawcett, 2006) or by establishing a market value for the firm (Zingales, 1995, Mello and Parsons, 2000, Brau and Fawcett, 2006). These theories suggest that a firm without large growth investment opportunities, without a need for large amount of capital, and with no demand for corporate control activities will be more likely to go private. In sum, consistent with the traditional argument in support of leveraged buyouts, publicly held targets will try to increase firm value by eliminating inefficiencies arising from agency problems after a leveraged buyout. In contrast, financial buyers of private targets will try to alleviate investment constraints and capitalize on growth opportunities through a leveraged buyout.

1.3. Data and summary statistics 1.3.1. Data sources and some institutional background Leveraged buyout transactions are collected from Zephyr; 5 deal information (deal date, deal type public-to-private or private-to-private, etc.) is cross-checked using SDC Platinum and Capital IQ. I select completed management buyout, management buy-in, and institutional buyout deals with an acquired stake of at least 50% and target companies that are located in the U.K. The total number of such deals is 4,652. As a comparison,

Zephyr, which is published by Bureau van Dijk, provides information on mergers and acquisitions, IPOs, and private equity deals worldwide since 1997. As of January 2009, it contains information on 703,327 deals. Zephyr does not cover deals involving equity stakes of less than 2 percent, unless the consideration for the stake is greater than GBP 15 million (i.e. where the market capitalization of the target is over GBP 300 million). When the bidder is an investment trust or pension fund, then the threshold is raised to 5 percent. If the purchase is considered to be significant, then it is entered regardless of the deal value.

during the same period, the number of completed leveraged buyout deals where the targets are located in the U.K. and the stake owned after buyout is at least 50% is 4,386 in the SDC database; Capital IQ contains 3,322 such deals from 1997 to June 2007. Therefore, Zephyrs deal coverage appears to be most comprehensive than other databases typically employed in academic research. Table 1 provides summary statistics on leveraged buyout transactions. Panel A presents the distribution of the number of buyouts by transaction year and target status. The buyouts of independent private firms (Private) account for about 40% of the U.K. leveraged buyout market, after divisional buyouts (Divisional, 41%). The buyout of public firms represents only a small fraction, 4.9%. However, the median deal value of private targets is 10 million whereas that of public targets is 74.6 million. Therefore, while public firm buyouts are small in number, they account for 29% of the market in value. Private firm buyouts make up 11.2%, less than half the size of public firm buyouts. However, note that only 37.8% of private buyouts report deal values, whereas all public buyouts do so. If we assume that the size of unreported deals of private buyouts is the same as for reported deals, the total deal value of private firm buyouts is about 69 billion, approximately the same as the total deal value of public firm buyouts. Although I do not attempt to explain the difference in deal pricing across target status in this study, I briefly present the information on deal value multiple. 6 Deal value multiple on EBITDA is highest among secondary buyout deals, 10.61, and lowest among

Officer (2006) and Bergeron et al. (2009) analyze this issue.

10

distressed buyouts, 6.11. Deal value multiples are slightly lower in private buyouts than public buyouts: 7.87 vs. 7.92 for private and public targets, respectively. Panel B reports the distribution of leveraged buyout transactions by transaction type. The most frequent form of buyout (66.5%) is a management buyout (MBO) where the existing managers of a target company purchase a controlling interest from existing owners. The next most popular form of buyout (22%) is an institutional buyout (IBO), in which private equity firms acquire a target company from previous owners and, often, the target's management takes a relatively small stake. In a management buy-out, the incumbent managers, as a bidding group, take an equity stake. In the case of an institutional buy-out, managers receive equity ownership as part of their compensation packages (Renneboog and Simons, 2005). A management buy-in (MBI), where an outside management team takes a majority stake in the target company and often replaces existing managers, accounts for 9.4% of the buyout market. Finally, a buy-in management buyout (BIMBO) indicates that an existing management team, along with outside managers, buys out a company; this type of deal represents about 2% of the market. About half of all buyouts are sponsored by private equity funds. In terms of deal value, institutional buyouts dominate, accounting for 81% of the market. Management buyouts follow at 17%. The median deal value of institutional buyouts is 62 million which is substantially larger than that of transactions led by incumbent or outside management teams. This pattern is not unexpected since large public firms usually complete buyouts with private equity partners.

11

Pre- and post-buyout financial information on target companies is from three sources: Amadeus, Worldscope (for public targets), and from the annual accounts filed with the Companies House. This process is complicated by substantial changes in corporate structure that occur after buyouts. Figure 1 depicts a typical leveraged buyout transaction. Usually one or more acquisition vehicles (e.g. NewCo and TopCo) are established one on top of each other to complete the transaction. After NewCo is incorporated by management and/or private equity firms, it acquires Target and its subsidiaries. TopCo is also concurrently created to acquire NewCo. After the buyout, these acquisition vehicles continue to serve as holding companies of Target and its subsidiaries. 7 In the U.K. when one company becomes a wholly owned subsidiary of another, the subsidiary does not have to prepare consolidated financial statements (in accordance with Section 228 of the Companies Act in 1985). In a leveraged buyout, when a target company has significant subsidiaries and becomes, again, a subsidiary of an acquisition vehicle, the targets financial statements do not usually include its subsidiaries financial information. Therefore, a targets pre-buyout financial statements and its post-buyout financial statements can be significantly different, even though the entity is materially unchanged throughout the process. Hence, for each buyout transaction, I trace and identify the targets parent company. Because their annual accounts consolidate the target and its subsidiaries financial information, I use the parent companys annual account as a comparison against the pre-buyout targets annual account.
There are several reasons for the establishment of multiple acquisition vehicles. When there are several layers of financing, a number of acquisition vehicles are formed into which corresponding financial instruments are invested. Senior lenders take comfort by having subordinated or equity invested into vehicles further away from cash generating target group. Effectively, this structure establishes structural subordination (See Speechly (2008) for further discussion on this issue).
7

12

1.3.2. Sample selection From 4,652 initial deals, I exclude from the analysis target companies without a U.K. registration number, leaving 3,706 deals. I drop divisional buyouts since complete accounting data is rarely available for divisions of larger corporations (1,313 deals). Similarly, I also eliminate distressed buyouts (144 deals): accounting data is not usually available once a firm files bankruptcy. I do not study secondary buyouts in this paper (386 deals), since the motive for these deals is likely to be very different from that of buyouts of independent private firms. I exclude recent deals (those completed in 2007 2009) since I need to examine target firms post-buyout investment and performance (203 deals). This selection process results in a sample of 1,660 buyout deals from 1998 to 2007. Finally, I exclude target firms which do not provide consolidated financial statements (651). 8 The final sample of target firms consists of 886 independent private target firms and 122 public target firms. The distribution of the final 1,009 leveraged buyouts from 1997 to 2006 by calendar year is reported in Table 2. Public-to-private transactions account for 12% and private-toprivate deals make up the rest. The number of transactions is more concentrated towards the end of the sample period, peaking in 2006. The median deal values involving private and public companies are 10.00 and 74.64 million, respectively. The leveraged buyouts transactions are concentrated in manufacturing and services industry. This could be
8

Not all companies provide detailed annual accounts: public companies disclose a substantially greater amount of information than private companies. Small or medium-sized private companies can provide abbreviated accounts which contain only minimal company and financial information (in accordance with Section 248 of the 1985 Companies Act). Also cash flow statements are not filed if the company is a wholly owned subsidiary of another company (Financial Reporting Standards 1).

13

because there are more number of firms in this industry or because firms in these industries tend to have greater amount of fixed assets which can be used as collateral for debt financing.

1.3.3. Construction of control sample To examine the effect of leveraged buyouts led by private equity firms on private targets, I compare the investment and growth of the target firms of private equity with the carefully selected three sets of control sample firms. First, I compare targets growth with industry median growth, following previous studies (Kaplan, 1989, Smith, 1990). The second set of control firms is private firms which underwent leveraged buyouts without private equity firms involvement. This is a natural set of benchmark because these private firms were actually put up for sale, experienced similar ownership changes, and could have been targets of private equity. In addition, to a certain extent, unobservable economic forces leading to leveraged buyout decision can be controlled by using this sample. Second, I construct non-LBO target private firms which have similar characteristics as private equity led LBO target firms using propensity score matching. From all U.K. companies in Amadeus, I first exclude private firms which have engaged in leveraged buyouts. Then I select firms that provide consolidated financial statements. I exclude firms where only unconsolidated statements are available because such statements do not provide a complete view of the business. Among these private firms, I randomly choose 3,000 firms to generate reasonable estimates of the following discrete choice regression 14

model. With 3,000 non-LBO target firms and my sample of private equity led LBO target firms, I estimate a logistic regression to predict the likelihood of being a target. In this regression, I include a list of typical determinants of a firm being a target of leveraged buyout such as firm size, sales growth, profitability, cash holding, and leverage. The predicted value from this logistic regression is called propensity score, which implies the probability of being a target of leveraged buyout. Then for each target firm, I choose a control firm with the closes propensity score. 9 One purpose of the first two control sample is to mitigate sample selection bias. As shown in Table 3, it turns out the targets of leveraged buyouts led by private equity have different characteristics than other industry peer firms. Therefore, to examine whether private equity firms do affect the target firms behavior after the buyouts, I need to carefully construct control sample of firms which could have been targets of private equity but decided not to be acquired by private equity. Comparing the targets postbuyout behavior with the control sample of firms will tell, though not perfect, whether private equity firms are simply selecting particular types of private firms or private equity firms do affect the targets post-buyout restructuring process. Table 3 provides the summary statistics of financial characteristics for private target firms of private equity prior to buyout and compare with three control samples. Compared with industry peer firms, the target firms are smaller, older, more profitable, growing faster, and less leveraged. Relative to private targets which underwent leveraged buyouts without private equity sponsors, the target firms are larger, more profitable, and growing

For a detailed description of the methodology, see Li and Prabhala (2007).

15

faster. There are not much significant differences between the target firms and the matched control firms. In Table 4, I also report the estimates of the logistic regression to predict the likelihood of being a target of private equity led leveraged buyout of the following specification:

The dependent variable is a binary variable equal to 1 for target firms and 0 for control sample firms. Since targets of leveraged buyouts should take substantial leverage, targets profitability, measured by EBITDA, and leverage may be important determinants of leveraged buyouts. The results show that private firms with higher profitability are more likely to targets of private equity led leveraged buyouts. Also, private targets of leveraged buyouts have larger sales growth than peer non-target private firms. In addition, private targets are under-utilizing debt capacity (negative coefficient of total debt to sales), which implies that leveraged buyouts are more attractive for under-leveraged private firms. To the extent that sales growth proxy for growth potential of private firms, the results suggest that private firms with larger growth opportunities are more likely to be targets of private equity led leveraged buyouts.

1.3.4. Ownership structure of private targets In this section, I document the ownership structure of privately held targets to see whether and the degree to which ownership is concentrated. Table 5 details the ownership 16

structure of private targets prior to buyout for the subsample of 721 target firms where ownership information can be identified from the firms annual returns. 10 As expected, ownership is highly concentrated among a few shareholders. The average (median) number of shareholders of private targets is 3.6 (2). Thirty-one percent of private targets are entirely owned by a single owner. The average (median) ownership of the largest owner is 87% (100%). 95% of the largest owners are also directors (chairman or managing director) of their firms. Therefore, the majority of private target firms qualify as zero-agency cost firms of Jensen-Meckling (1976). There is a substantial change in the ownership structure after the buyouts. The average (median) number of shareholders increases from 3.6 (2) to 7.2 (7). The average (median) ownership of the largest owner-managers drops from 86% (100%) to 7% (0%). In an unreported table, I find that 80% of the largest owner-managers leave the management and 71% of owners completely liquidate their ownership through leveraged buyouts. Though I cannot directly observe why the owner-managers would want to exit the business, a modest fraction (23.5%) of the owner-managers is above 65 years of age, which is suggestive of retirement. In the transactions, ownership is transferred to a larger number of shareholders, including existing managers and private equity firms. This change in the ownership structure is distinctively different from the change in public firms undergoing leveraged buyouts, where, not surprisingly, ownership becomes more concentrated.

All companies in the U.K. must submit this annual return form (Form 363a prior to October 2009) to Companies House each year: It provides a snapshot of general information about the company, including the details of key personnel, the registered office, share capital and shareholdings.

10

17

1.4. Post-buyout growth of target firms If private firms face investment constraints which can be alleviated by a leveraged buyout led by private equity firms, we expect to observe increases in firm size and investment post-buyout. Measuring firm size surrounding the time of a leveraged buyout is complicated due to the fair-value adjustment to the book value of assets. Current assets including stocks (inventories) and creditors (account payable) are fair-value adjusted upon completion of the buyout. Also, typically, positive goodwill is generated to reflect the purchase price paid for the targets assets (as a result, intangible fixed asset size increases) and, subsequently, these write-ups are depreciated or amortized over the ensuing years. Therefore, to make a fair comparison between assets in the pre-buyout period and those post-buyout, I adjust the book value of total assets by subtracting write-ups and goodwill generated at the time of the buyout transaction. One limitation of this approach is that the size of the assets after buyout may be understated. Since write-ups and goodwill are depreciated or amortized after buyouts, subtracting write-ups and goodwill generated at the time of the transaction from the book value of assets at each fiscal year-end will lead to an underestimation of the book value of assets. Re-adding depreciation and amortization to the book value of assets in each year will not alleviate this concern, because depreciation and amortization also include assets not associated with write-ups and goodwill due to the buyouts. Hence, to give as fair a view as possible, I also provide other measures of firm size, such as tangible fixed assets (PPE), sales and the number of employees. 18

Before I proceed to examine post-buyout growth of target firms of private equity led leveraged buyouts, I examine post-buyout growth of public targets in Table 6. I find a substantial reduction in firm size among public targets of leveraged buyouts. This finding is consistent with those in previous studies (Kaplan 1989, Smith, 1990, Aslan and Kumar, 2009). For example, Kaplan (1989), with a sample of 48 public targets, documents that the control-adjusted capital expenditures to sales ratio decreases by 16.7%, 16.8%, and 25.6% respectively for years +1, +2, and +3 compared with year -1. He interprets this finding as being consistent with reduced agency costs of free cash flow. Table 7 provides summary statistics for changes in firm growth of the private targets of private equity from the two years preceding the buyout to three years afterwards for private and public targets. All measures of firm growth are benchmarked again three control samples: Industry median, private LBO targets without private equity sponsors, and matched sample using propensity score matching. Private targets substantially increase in total assets, sales, and tangible fixed assets. For example, from year -1 to year +3, the control-adjusted sales and tangible fixed assets of private targets increase by 34% and 23%, respectively. Capital expenditures are the net cash flows from the purchases and sales of fixed assets. Private targets significantly increase capital expenditures, especially during the first year post-buyout. During the first year post-buyout, the control-adjusted capital expenditures of private targets increase by 62%. The control-adjusted capital expenditures to sales ratio, however, does not increase. This is mostly because the growth rate of sales (denominator) exceeds that of capital expenditures (numerator). 19

Many studies examine whether leveraged buyouts benefit investors at the cost of employees (Shleifer and Summers, 1988, Kaplan, 1989, Lichtenberg and Siegel, 1990, Davis et al., 2008). In general, researchers find that employment grows less than other peer firms (Kaplan and Strmberg, 2008). This is partially true in my sample of leveraged buyouts. The industry-adjusted number of employees in public targets at year +3 is 35% lower than it is at year -1. In contrast, private targets undergo substantial increase in employment. The control-adjusted employment grows by 37% from year -1 to year +3. Though prior studies on the effect of leveraged buyouts on employment generate somewhat mixed results, it appears the buyouts substantially increase employment for private targets. However, I find that, in an unreported table, the industry-adjusted wage per employee decreases post-buyout. The difference in post-buyout investment behavior between private and public targets suggests that private firm buyouts occur for very different reasons than those of public firms: leveraged buyouts reduce inefficiencies of private targets arising from investment constraints and those of public targets stemming from free cash flow problems. Finally, to see the extent of the acquisitions and disposals activities of target companies, I compute the sum of all cash outflows (inflows) associated with acquisitions (disposals) from the time of a leveraged buyout to the exit (when the firm exited private equity ownership) or to the most recent fiscal year, and divide this sum by PPE at the end of the fiscal year prior to buyouts. I exclude cash outflows associated with the leveraged buyouts. The intensities of acquisitions and disposals activities are estimated by the following measures: 20

Table 8 presents the results. The median acquisition-related cash outflow to PPE ratio of private targets is 0.66 and that of public targets is 0.043; the difference is statistically significant at the 1 percent level. 11 In addition, the median disposal-related cash inflows to PPE ratios are 0.00 and 0.017 for private and public targets, respectively. Therefore, private targets engage in active add-on acquisition activities after the buyout while public targets do not. The overall evidence in this section shows that private targets grow substantially postbuyout through larger investments and acquisitions. This finding is distinctively different from the post-buyout restructuring and value creation process involving publicly held targets (Kaplan, 1988, Smith, 1989, Wiersema and Liebeckind, 1995, among others), and supports the view that leveraged buyouts alleviate the investment constraints of privately held targets and facilitate the targets growth after the buyouts.

1.5. Analysis of Deal rationales For a subsample of 113 deals which involve privately held targets, I collect deal rationale information though Zephyr and the Lexis-Nexis News Search. These deal rationales are

11

I have not yet examined acquisition and disposals activities of target firms prior to buyouts.

21

comments made by managers of targets, acquirers, or equity sponsors on the completion of transactions. Typically, these comments contain information about the motives for and goals in completing the deals and the future prospects of the target companies. Therefore, from this information we can infer deal participants motivations for the leveraged buyout. Of course, these comments tend to be optimistic about targets future performance since deals would not have otherwise been completed. Also the motivation for the deal and the future plan expressed in deal rationales may not coincide with the actual restructuring process after buyouts. Despite these caveats, it is worthwhile to analyze what deal participants have to say about why they sell and buy target companies and what the buyers intend for the target companies. This analysis will complement the earlier quantitative analysis. The comments on deal rationales are qualitative information where some subjective judgment is called for in interpreting the comments implicit meaning. In 26 cases, deal rationales do not provide clear ideas about the grounds and future plans for target companies. Examples of this kind include: We firmly believe that the company will now benefit from this buy-out, (Rite-Vent Ltd.) It is pleasing to have worked with local professionals to complete the transaction, (MWL Print Group Ltd.) and We believe the deal which has been completed provides an excellent platform for future development (SHG Opportunity Management Ltd.). In 61 cases (61/8770%), managers (of the target company or sponsors) specifically mention that they will expand or grow the target companies through geographical expansion plans (14%), new investments plans (19%), and acquisitions plans (11%). The 22

remainder simply state that managers intend to expand and grow the target businesses. See Appendix I for examples of deal rationales listed according to different future plans. Overall the evidence suggests that the buyers of private targets seem to have a clear intention to grow and expand the business post-buyout.

1.6. Pre-buyout investment constraints and post-buyout growth To further understand whether private equity firms, through leveraged buyouts, mitigate investment constraints facing small private targets, I examine the relationship between pre-buyout investment constraints and post-buyout growth. When facing a similar growth opportunities, a more investment constrained target firm before the buyouts will benefit more from private equity sponsors and leveraged buyouts, and will experience larger growth after the buyouts. I use several proxy variables to measure the degree of investment constraints facing private targets. First, as a measure of financial constraint, I use firm size, firm age, and leverage. Second, to proxy for owner-managers risk aversion, I use the Herfindahl index of ownership concentration and owner-managers age. Lastly, I examine whether a target is managed by a professional manager under the assumption that if a private firm is managed by owner-managers, the firm is less likely to have access to outside professional management skills. Specifically, I estimate the following median regression (4). Median regression, which is a form of quintile regression, estimates the change in the medians of the dependent variables produced by one unit of change in the predictor variables; therefore the estimates are less affected by extreme outliers (Koenker and Hallock, 2001). 23

This specification is also consistent with previous univariate statistical tests where median values are used.

where Growth = Growth rate of firm size (assets, sales, employment, and capital expenditures) from one year prior to buyout to three year post-buyout, SG = Sales growth from year -2 to year -1 (proxy for growth opportunities), FirmSize = The book value of assets at year -1, FirmAge = The number of years from incorporation to the year of leveraged buyouts, Leverage = The total debt to total assets ratio at year -1, Concentration = The Herfindahl index of ownership concentration at year -1, OwnerAge = Owners age, = One if a owner is also the manager and zero otherwise at year -1.

I interact each measure of investment constraints with sales growth to condition on different amount of growth opportunities each target firm faces. Table 9 reports the median regression estimates of targets growth on pre-buyout investment constraints variables. Though I find that more constrained firms tend to experience larger post24

buyout growth, the statistically significance of the relationship is not strong, making it harder to draw a strong implication from this analysis.

1.7. Operating performance after buyouts In this section, I examine the post-buyout operating performance to see whether the different investment behaviors of private and public targets have different implications for operating performance in the post-buyout period. My sample of buyout firms does not suffer from sample selection bias as the performance of targets can be observed irrespective of whether they exit leveraged buyout ownership. The main variable of interest is EBITDA. I exclude the year when the leveraged buyout occurred from the analysis for two reasons: first, buyout-related expenses can understate operating performance in year 0. Second, the first annual account after the buyout provides information on the business from the time of the buyout to the new fiscal yearend, which, typically due to fiscal year-end changes, is shorter than twelve months. This makes a pre- and post-buyout comparison difficult. Table 10 reports the results. Among private targets, the level of EBITDA increases after the buyout. From year -1 to year 3, the control-adjusted EBITDA increases by 12%. I also examine the EBITDA to operating assets (the average of fiscal year-beginning and end current and tangible fixed assets), EBITDA to sales, and EBITDA to the number of employees. The reason I normalize EBITDA by sales or the number of employees is to mitigate the bias due to write-ups in operating assets. As firm size significantly increases

25

post-buyout, these measures of operating efficiencies drop for private targets. Most of the changes of operating efficiency measures are negative. Overall, I find that private targets with private equity sponsors experience an increase in operating performance: industry-adjusted EBITDA increases by 12% during the first three years post-buyout. Not surprisingly, the industry-adjusted ratio of operating income to sales drops by 35%. The reason is because the rate of sales growth exceeds that of EBITDA after buyouts. This pattern implies that buyouts with private equity sponsors result in growth but not improved margins. This deterioration of operating efficiency could be the result of worse investments on the part of private equity firms. However, it could also be that private equity firms are increasing, i.e., optimizing, investments by taking positive NPV but less profitable investment projects which were not previously exploited prior to buyouts due to investment constraints. Though the evidence in this paper cannot distinguish between these two hypotheses, it is unlikely that private equity firms are making unprofitable and inefficient investments in light of previous studies. For example, Sheen (2008) finds that private firms which underwent private equity led leveraged buyouts are more likely than public firms to make an efficient investment in response to the expected demand shock in chemical industry. Also, Bargeron et al. (2008) finds that private equity firms tend to pay less acquisition premium than public firms when acquiring a similar target firm, which suggests that private equity firms tend to make investments in a most cost saving way.

26

1.8. Conclusion I study the effects of leveraged buyouts on private companies. Though the buyouts of private companies account for the majority of deals in the leveraged buyout market, most academic studies base their analysis on a sample of public-to-private buyout transactions. In this respect, our understanding of leveraged buyoutsof why they occur and how firms restructure and perform afterwards is still limited. This paper, by investigating private firm buyout transactions, expands our understanding of leveraged buyout transactions and their effect on target firms. There is a striking difference between private targets and public targets with regard to post-buyout investment policy. Consistent with previous studies based on the sample of public-to-private leveraged buyout (Kaplan, 1988, Smith, 1989), public firms reduce firm size and investment after buyouts. The evidence is consistent with the view that public firms reduce agency costs after leveraged buyouts by reversing previously made inefficient investments. In contrast, private targets significantly increase post-buyout firm size and investment. Owner-managers and incoming managers can complete the transaction either independently or with private equity sponsors. Private companies with larger profitability and growth opportunities tend to become targets of private equity. Post-buyout, rather than improving operational efficiency, these targets of private equity grow in firm size and make greater investments.

27

The evidence supports the view that private equity firms acquire small private firms and try to improve firm value by alleviating various investment constraints facing private firms and by facilitating the target firms growth. In light of the existing findings in the literature, the overall evidence suggests that private equity firms attempt to reorganize target firms in a way which reduces inherent the targets inefficienciesagency problems in public targets and investment constraints in private ones.

28

Chapter 2: Incentives of Private Equity General Partners from Future Fundraising

2.1. Introduction Compensation agreements in private equity (PE) partnerships typically give General Partners (GPs) a management fee that is a percentage (usually 1 to 2%) of the amount of capital committed to the fund, as well as carried interest equal to a percentage of the profits (usually 20%). The carried interest, together with the GPs own equity contribution to the fund, aligns the incentives of the GPs with those of the investors in a private equity fund to a much greater extent than is typical in public corporations. These explicit incentives to make value-maximizing decisions are commonly thought to be an important driver of the success of private equity firms (see, for example, Jensen (1989), Kaplan (1989), Kaplan and Stromberg (2009)). At the same time, Metrick and Yasuda (2010) report that approximately two-thirds of expected revenue to PE partnerships comes from fixed-revenue components that are not sensitive to performance. This has led some to suggest that perhaps PE partnerships incentives to deliver high performance are not as strong as previously thought, or at least not as strong as they should be. 12 Missing from these arguments is the fact that application of the explicit terms of the partnership agreement omits an important source of implicit incentives facing general partners in private equity funds: the effect of current performance on the ability of
12

Its the Fees, not the Profits, The Wall Street Journal, Sept. 13, 2007.

29

partnerships to raise larger funds in the future, and consequently to earn higher fees on those funds. Young partners often give up promising careers in other fields such as investment banking to manage a relatively small fund, their hope being that high returns from such a first fund will enable them to raise larger, more lucrative funds in the future. The quantitative importance of this effect, however, is not known. In other words, for every extra percentage point of returns (or every extra dollar) earned for the current funds investors, how much, in expectation, does the lifetime or future income of the funds general partner change? How large are these implicit incentives relative to the much-discussed explicit ones? Theoretically, what factors ought to affect the size of change in partners lifetime incomes as a function of fund returns? Do these predicted patterns appear true in the data? More generally, how do todays returns affect the ability of partnerships to raise capital subsequently? This paper evaluates the importance of future fund-raising to the incentives of private equity general partners. To do so, we formalize the logic by which good performance today could lead to higher future incomes for GPs. We present a model in which a private equity partnership potentially has an ability to earn abnormal returns for their investors, but this ability is unknown. Given an observation of returns, investors update their assessment of the GPs ability, and, in turn, decide how much capital to allocate to the partners next fund. We derive predictions about the relation between performance of a particular fund and the funds partners abilities to raise capital in the future. Intuitively, the model implies that the more informative the funds performance is about GPs abilities, the more sensitive future fundraising should be to todays performance. In 30

addition, the way in which abilities can be scaled will affect investors willingness to commit higher quantities of capital for a given level of managerial ability. These larger funds will lead, in expectation, to higher compensation for the partners, since compensation agreements almost always change linearly with fund size. Given this setup, we derive an explicit formula calculating the effect of fund performance today on expected future GP compensation. We test these predictions using a sample of 838 partnerships who manage 1,726 buyout, venture capital, and real estate funds. In particular, we evaluate the informativeness criterion, which suggests that performance of later funds (for example, a partnerships third or fourth fund) should be less informative about ability and hence be less strongly related to future inflows of capital than would similar performance in a partnerships first fund. In addition, the ability of managers to translate their skills to larger funds depends on the nature of the production process. Given Metrick and Yasudas (2010) finding that buyout funds are more scalable than venture funds, the model predicts that the future fundraising of buyout funds should be more sensitive to performance than that of venture funds. Our empirical results are consistent with these predictions. For buyout, venture capital, and real estate funds, the estimated relation between the IRR of a partnerships current fund and the expected size of future funds is positive, consistent with Kaplan and Schoar (2005). The magnitude of this relation varies with the scalability of the investments. Buyout funds, which are the most scalable, have the strongest relation between IRR and future fund size, while venture capital funds, which are the least scalable, have the 31

weakest relation. Further consistent with the model, younger partnerships have a stronger relation between future fund sizes and current fund returns than older partnerships. Given these estimates of the sensitivity of future fundraising to current performance, we next turn to calculating the magnitude of general partners incentives. Our model provides an explicit formula for the change in general partners lifetime incomes as a function of the return of the current fund. To perform the calculations, we use this formula, plausible parameters chosen to reflect the characteristics of our sample of private equity funds, and estimates of expected carried interest and management fees taken from Metrick and Yasudas (2010) simulations. We calculate the expected incremental revenue to the general partners from both an additional percentage point of returns (IRR) to limited partners (LPs) in the current fund and an incremental dollar of profits returned to LPs. We break the expected incremental revenue into two parts: the incremental direct compensation (from carried interest on the current fund) and the incremental expected indirect compensation (from carried interest and management fees from future funds, whose size is a function of the performance of the current fund). For an average sized buyout fund in our sample ($853 million), we estimate that for an extra percentage point of return (IRR) to limited partners in the current fund, general partners receive on average an extra $8.7 million in direct fees in the current fund. Estimates of incremental fees on future funds for each additional percentage point of IRR in the current fund vary from $6.6 million to $26.3 million depending on the estimation approach used, with about half of our estimates being approximately twice the

32

incremental direct fees. The level of both direct and indirect fees varies with fund size but their ratio is independent of size. An alternative approach is to calculate the expected future general partner fees per extra dollar returned to limited partners in the current fund. For every extra dollar returned to LPs in the current fund, the GP earns $.25 in carry (assuming that the carry is in the money) 13, while indirect profits are between $.19 and $.76 per dollar of profits in todays fund, for an average sized buyout fund. Of course, expected income from future funds is riskier than direct fees and occurs in the future, so while it is not obvious how to discount these fees appropriately, their true value is substantially less than it their undiscounted expectation. On the other hand, carries are not always in the money, and our econometric methodology is likely to understate the human-capital benefits going to general partners for a number of reasons. Given these estimates of the magnitude of additional expected income through the fundraising channel, it appears that does the indirect benefits to partners of buyout funds are of similar magnitude as the direct fees. We also perform the same calculations for venture capital and real estate funds. Future fundraising is less sensitive to current performance for these types of funds than for buyout funds, with venture capital funds displaying the least sensitivity. For an averagesized venture capital fund, estimates of future profits per dollar of returns in todays fund range from $.05 to $.26, and from $.15 to $.77 for an average-sized real estate fund. This pattern is consistent with the scalability arguments, and also suggests that no matter what

Using a typical carry of 20%, for LPs to receive an extra dollar, the fund must earn an extra $1.25 in profits, with $.25 going to the GPs.

13

33

type of fund one is considering, an important component of general partner incentives is the effect of todays performance on a partners ability to raise future funds. Finally, we consider the cross-sectional pattern of these incentives. Our model suggests that the sensitivity of future fundraising to current performance should depend on the extent to which current performance adds incremental information to the markets assessment of the general partners abilities. We expect that this sensitivity should be greatest for younger partnerships. Empirically, we find that funds of younger vintages have greater sensitivities of fund growth to todays performance, and that this greater sensitivity leads to larger expected future profits per dollar of todays returns for these types of funds. This paper is related to a number of different literatures. It is most directly related to work on the reasons for value improvements in private equity transactions. Kaplan (1989) and Smith (1990) document that operating profitability increases following buyouts, although this pattern appears weaker for more recent buyouts (Guo, Hotchkiss and Song, 2010). Jensen (1989) and Kaplan and Stromberg (2009) attribute these value increases in large part to the incentives facing general partners, although both focus on direct rather than indirect incentives. Kaplan and Schoar (2005), in perhaps the most related analysis to that done here, emphasize the talent of particular partnerships and estimate regressions showing that fundraising is related to historical performance. Still, none of this work attempts to estimate the magnitude of the incentives facing private equity general partners implicit in the effect of todays performance on their future income.

34

Closely related to this work is a large literature on mutual fund inflows and their relation to historical performance. Ippolito (1992), Chevalier and Ellison (1997), Sirri and Tufano (1998), Barclay, Pearson and Weisbach (1998), and Sensoy (2009) all estimate regressions predicting the inflows to mutual funds as a function of a funds historical performance, and find a strongly positive (nonlinear) relation. Berk and Green (2004) explain these patterns in the context of a learning model similar to the one presented below in which mutual fund investors draw inferences about managers abilities from current performance. A key difference between Berk and Greens (2004) analysis and ours is that they focus on the way in which fund inflows dissipate the ability of fund managers to generate abnormal returns, while we focus on how the relation between fundraising and performance affects the incentives of fund managers. More generally, our work adds new empirical evidence to the idea pioneered by Fama (1980) that career concerns can be an important source of incentives inside firms. Chevalier and Ellison (1999) explore the risk-taking of young mutual fund managers compared to old in light of career concerns. Hong, Kubik, and Solomon (2000) and Hong and Kubik (2003) consider how security analysts forecasts may be influenced by career concerns. In contrast to these papers, our focus is not on agency problems, but rather on quantifying the importance of career concerns in providing private equity general partners with incentives to create value for their limited partners. In doing so, our work is in the spirit of Gibbons and Murphy (1992), who emphasize the importance of understanding total, rather than only explicit, incentives. The explicit (and observable) compensation formulas in private equity partnerships together with the empirical relation 35

between fund performance and future fundraising allows for quantification of these incentives, which in other contexts clearly exist but are hard to measure. The remainder of this paper proceeds as follows: Section 2 lays out the model described above. Section 3 describes the database of private equity funds used in the analysis. Section 4 presents regressions estimating the effect of todays fund returns on future fundraising. Section 5 performs calculations that transform the coefficients from these regressions into general partner incentives, using the analysis from the model in Section 2 as a basis for the calculations. Section 6 discusses the implications of this work and concludes.

2.2. Model In this section we construct a learning model in which investors assign cash flows to private equity partnerships based on their perceptions of GPs abilities to earn abnormal profits. Investors observe returns earned by partnerships and allocate their capital to partnerships subsequent funds based on their posterior estimate of their ability. Given that the compensation system in private equity partnerships is almost always a linear function of fund size (Gompers and Lerner (1999)), this capital allocation process leads to a strong relation between performance in a current fund and that funds general partners future compensation.

36

2.2.1. Setup To formalize this idea, we assume that a particular GP manages a series of funds over time. This GP has ability equal to , which is a measure of his ability to earn abnormal returns. We assume that is unobservable and that there is symmetric information, so all agents, including the GPs themselves, have the same estimate of its value. We also assume that is constant over time for a particular partnership, which abstracts away from issues of changing partnership composition, investment environments, or changing ability over time due to health or other considerations. 14 Let i denote the sequence of funds managed by a given GP, ri be the return to LPs for fund i, Ii be the size (committed capital) of fund i, and Ii *k (ri) be the total revenue earned by the GP, where k (r) is an increasing and differentiable function, representing the fraction of the initial size of the fund that is earned by the GP if performance is r. The function k (r) should be thought of as the total profits from running a firm that has a return equal to r, including management fees, carried interest, and other income earned by the fund, such as additional fees earned by funds for managing portfolio companies. We characterize the GP compensation in this manner following Metrick and Yasuda (2010), who provide estimates for k (.) using a simulation approach.

These assumptions greatly simplify the formal analysis but do neglect a number of interesting and potentially important factors. The assumption that there is symmetric information about managers abilities dates to Holmstrom (1982), and has been used in similar learning models by Gibbons and Murphy (1992), Hermalin and Weisbach (1998, 2009), and others. Implicitly, the idea is that anyone who can become a GP is smart, hard-working, well-educated, etc., but the key factor that determining who can earn abnormal returns is an unobservable match between the individual and the tasks associated with earning profits as a general partner.

14

37

We assume that the fund returns are increasing (in expectation) with the GPs ability, , and that ri ~ N (, 1/r). 15 After observing the returns on i funds, the markets updated assessment , i, of i is given by:

for all i (DeGroot, 1970 provides a derivation of this Bayesian updating formula). We assume that investors allocate more funds to managers they believe more able and have higher values of , so the size of fund i, Ii = c (i-1), where c (.) is an increasing, differentiable function. We assume the GP runs a total of N funds over his lifetime. N is exogenously determined (e.g. a function of the GPs initial age). To capture the idea that at some point performance may be so bad, and the updated assessment of so low that the GP is not able to raise additional funds (and so does not actually run N funds), we can think of the c (.) function as producing follow-on funds of size zero for those cases.

2.2.2. Cross-sectional implications This simple learning model characterizes the way that fund returns affect future fundraising and, consequently, the future expected compensation for the funds partners. Future fund size is given by Ii+1 = c (i) = c ( ).

This should be thought of the expected skill of a particular GP conditional on all observable characteristics prior to any returns being observed. Different GPs will therefore have different values of 0 from one another and consequently can raise funds of different sizes.

15

38

2.2.2.1. Informativeness of Returns over a Sequence of funds for a given Partnership The sensitivity of future fundraising to current performance is given by the derivative of I2 with respect to r1, which is equal to:

The sensitivity of the size of the GPs third fund to the performance of his first fund, which equals the sensitivity of the third fund with respect to the second fund is:

Each of these sensitivities given by equations (2) and (3) is the product of c(), multiplied by a weighting factor that reflects the relative infomativeness of the return to the markets perception of the GPs ability. Intuitively, the function c () represents the quantity of funds the market chooses to invest in the fund as a function of their best estimate of . A more steeply sloped c () function means that for a small increase in managerial talent, the fund can profitably invest relatively large increases in funds, i.e., the fund is more scalable. For example, if a manager is shown to be talented at buying out companies and increasing value, he or she can likely buy out larger companies and increase value similarly to what she has done with smaller companies if the market is willing to fund these investments. In contrast, if a manager has demonstrated that he is talented at investing in startup companies, he or she is unlikely to be able to increase fund size much because the size of startup investments is not scalable (and because it is not feasible to 39

simply increase the number of investments given that increasing value is a timeconsuming process). 16 The second term of equations (2) and (3) represents the weight given to returns in forming the markets posterior estimate of . The larger the weighting term, the more informative the signal and the higher the derivative of future fund size to todays returns. As partnerships progress through time, the partnerships becomes known more precisely; so that the optimal updating rule means that subsequent s do not change as much as earlier s for a given return. The overall effect measures the impact on the change in the markets best assessment of for a given return times the amount of capital the market will choose to invest in a particular fund, given this change in their assessment of . Comparing expressions (2) and (3), it seems likely that the sensitivity of future fund sizes to returns is decreasing in fund sequence. The weighting term strictly decreases as the numerator is s in each one, while the denominator increases with the sequence number. If c () is linear or close to being linear, the prediction is unambiguous: the sensitivity of future fund size to current performance is decreasing in the sequence of funds. However, if c () is convex and 2 > 1, or if c () is concave and 2 < 1, the pattern may go the other way. That said, even if c () is highly nonlinear, on average we would not expect 2 to differ much from 1, so it seems likely that the weighting term effect will dominate. Consequently, in the data we expect to observe a decreasing sensitivity of future fund size to current performance as a given partnership manages subsequent funds.
16

Consistent with this logic is the fact that the most successful buyout funds such as KKR and Blackstone have steadily increased the size of their funds to the point where the largest funds are between $15 and $20 billion in committed capital, while the most successful Silicon Valley venture capitalists such as Kleiner Perkins and Sequoia have remained at or under $1 billion in committed capital.

40

2.2.2.2. Informativeness of Returns across Partnerships The model suggests that the effect of one funds return on the capital inflows to a partnership depends on the extent to which the funds return leads the market to update its prior of the partnerships ability. This informativeness of this return depends on how precisely the market knows the partnerships prior to the observation of the return. This idea can be formalized by examining how the sensitivity of fund size with respect to initial returns varies with s, the precision of :

When does this match our intuition that it should be positive? If c () is linear,

() is

equal to zero so the expression is unambiguously positive. If c () is nonlinear, the expression is still likely to be positive because on average should be close to

zero. The intuition for why the expression could be negative is the following: Suppose c () is convex and performance is much worse than expected, so . Then the assessment

of ability adjusts downward, to a point where the slope of c () is smaller (because of the convexity). The more informative the return , the greater the adjustment. Following

similar logic, the expression to be negative if c () is concave.

41

2.2.2.3. Equal Informativeness of Returns within a Sequence One element of the learning model is that, in each assessment of ability, all past returns are equally weighted. This means that the marginal impact of any one historical return is the same as that of any other. In particular, the model implies that:

This implication is potentially testable, though we do not test it here. The extent to which it is true for a particular partnership is likely to depend on the amount that a partnership raises and on how its investment strategy changes over time.

2.2.3. Lifetime compensation of GPs The total revenue earned by the GP over his lifetime is given by:

This formulation assumes that, following practice, GPs are compensated with a combination of management fees, which are a function of committed capital, and carried interest, which is a function of returns times the amount of capital in the fund. We think of the k (.) function as incorporating these two elements, plus other fee income that is likely to be proportional to fund size. The expected total revenue at any point in time would be equal to the realization of k times the current fund size plus the expectation of k times the expected fund sizes for the remainder of the funds (some of which are size zero with some probability). In our empirical implementation, we base our estimates of expected revenue to the GPs from the current fund on the standard 2% management fee 42

plus 20% carried interest fee structure. For expected revenue from future funds, we rely on Metrick and Yasuda (2010), who calculate via simulartion the expected fraction of a funds capital that becomes GPs compensation. 17 We are interested in calculating the incentives of the general partners and decomposing them into the direct and indirect components. In other words, how much do the partnerships expect to keep from incremental revenue, and how much of this additional revenue comes in the form of direct vs. indirect compensation. To perform, this calculation, there are two potential ways to measure incremental revenue: one can calculate the incremental revenue from an additional percentage point in returns or one can calculate the incremental revenue for each extra dollar returned to the funds limited partners.

2.2.3.1. Sensitivity of expected Lifetime GP Compensation to Percentage Return of the GPs first Fund To calculate the additional revenue for each percentage point return, we differentiate the expression for total return (equation (6)) by r1. This calculation leads to:

We will refer to revenue and compensation synonymously throughout the paper. In fact, private equity partnerships do have some (but not many) costs that create a wedge between revenue and partner compensation. However, many of these costs, such as the costs of renting an office and hiring support staff, are more or less fixed and do not affect marginal compensation. In addition, our focus is on the indirect aspects of compensation and its size relative to direct compensation and it seems unlikely that this ratio would be substantially affected by ignoring direct costs in our calculations.

17

43

Our goal is to provide empirical estimates of this derivative. We obtain estimates of the terms using regression analysis.

1) One approach is to regress

on r1. The from that regression is an estimate of

So:

2) Because fund growth rates tend to be skewed due to a few exceptionally popular funds, it is possible that these equations could fit the data better if we use a logarithmic transformation of fund growth rates. To use this method, we regress on r1 (the

+1 is to avoid taking the log of zero). We have:

So:

44

Regressions substituting later fund sizes for I2, and later returns for r1, provide estimates of the other terms in the same way as 1) and 2) above.

2.2.3.2. Sensitivity of expected Lifetime GP Compensation to Dollar Return of the GPs first Fund An alternative way of measuring incentives is to compute the change in general partners compensation for an extra dollar returned to the limited partners, which is the metric emphasized by Jensen and Murphy (1990). It is fairly straightforward to adjust the estimates we obtain following the previous subsection to be in terms of dollar return rather than percent return. If the return measure used in the previous section is the total return for the fund, then we just multiply by initial fund size. However, the usual return measure (and the one typically quoted in the private equity databases, including the Preqin one used below), is the internal rate of return (IRR) of the fund. The IRR is an annualized return measure which is subject to well-known problems such as the implicit assumption that intermediate distributions are reinvested at the IRR. To make things as simple as possible, and because the data used below are not sufficient to make more accurate calculations, we can assume that all capital is called at once and all distributions are made at once. That is, we can assume that each fund has a single capital call and a single 45

distribution, and the time (denoted T) between the two matches the average length of time in the data between the call of a dollar and the return of the profits associated with investing that dollar. For a private fund, this length is typically between 3 and 6 years. Metrick and Yasuda (2010) use an expected time to exit of 5 years for all buyout and venture investment, following Metricks (2007) evidence that 5 years is the median holding period for a first-round VC investment. Under these assumptions, the total dollar return to limited partners in the first fund, D, is given by: (12)

where r1 is the IRR of the first fund. Note that because IRR is a net-of-fee measure, D represents the total dollars to limited partners, not the total dollars earned by the fund (some of which go to the GP in the form of management fees and carried interest). To obtain the sensitivity of the lifetime GP compensation to dollar return of the GPs first fund, we can use the estimate of from the previous subsection together with the

identity:

Which holds given that D is invertible:

46

So:

Therefore:

Note that returning an extra dollar implies different marginal percentage returns depending on the baseline return. To get expectations, we will need to make assumptions about the baseline return.

2.3. Data To estimate the relation between fund performance and capital raising, we rely on fundlevel data provided by Preqin. We consider the largest three types of funds: buyout, real estate, and venture capital. The total number of buyout, real estate, and venture capital funds is 9,523 in Preqin as of June 2009. Preqin claims to cover about 70% of all capital ever raised in the private equity industry. In addition, in private communication Preqin informs us that about 85% of their data is collected via Freedom of Information Act requests and thereby is not subject to self-reporting biases. While we cannot directly 47

verify these claims, our data appear similar on key dimensions (notably performance) to that used in prior work. In all of our analysis, we exclude funds without vintage year data (64) or without fund size data (1,137). We also exclude 78 funds which are still being raised. To construct our sample of preceding (i.e., current) funds, we require performance (IRR) data. Because very small funds can grow at extremely high rates, we follow Kaplan and Schoar (2005) and drop funds with less than $5m (in 1990 dollars) in committed capital. In addition, in order to allow for sufficient time to elapse between a fund and its follow-on (should the latter ultimately be raised), we drop funds raised after 2005. Finally, when a private equity firm raises multiple funds in a given year, we aggregate funds in that year and compute the fund size weighted IRR. The exception to this is a few cases in which the same partnership manages, say, both buyout and real estate funds. In those cases, we treat the partnership as two separate partnerships, one each for buyout and real estate funds. We do so to avoid allowing, for example, a real estate fund to be a follow-on fund to a buyout fund. These sample screens leave us with a final sample of 1,726 preceding funds. The sample consists of 848 (49%) venture capital funds, 640 (37%) buyout funds, and 238 (14%) real estate funds. For each preceding fund, we ask whether we observe a follow-on fund in the database. We define a follow-on fund as the next fund raised by the same partnership for which we have information on fund size. Thus each preceding fund is allowed to have at most one follow-on fund. If we observe a follow-on fund, we record the size of the follow-on fund and compute the growth rate in fund size from the preceding fund to the 48

follow-on fund. If we do not observe a follow-on fund in the data, or if the data indicate follow-on funds but do not provide size information, we treat this as if the partnership did not raise a follow-on fund. The working assumption we use throughout the paper is that the absence of a follow-on fund with size information in the data means the partnership was unable to raise one, and thus we assume that the partnership raised a fund of size zero in these cases. 18 Tables 1 through 4 report the descriptive statistics for the variables which enter the multivariate regressions, i.e. preceding fund size, preceding fund performance, follow-on fund size, and fund growth. Since the focus is on growth rates between funds of a given partnerships, we present statistics by fund sequence, for both preceding and followon funds, as well as with and without zero-sized funds. Table 1 presents basic statistics about the funds in our sample. Since the database is organized by fund pairs to focus on the effect of one funds performance on the subsequent funds size, we present statistics for preceding funds and follow-on funds. Panel A shows the number of partnerships, broken down by whether each was buyout, venture capital, or real estate. It also presents statistics the number of preceding funds partnership for each type. Note that these figures are particular to our study, as to be considered a preceding fund, it is necessary to be founded by 2005 and have available data on fund IRR, so the numbers in Panel A of Table 1 understate the true number of funds per partnership. The distribution is clearly skewed, with many firms
18

This assumption has the effect of downward-biasing our estimates. Undoubtedly some partnerships do raise follow-on funds that are missing from the data because the data are incomplete. Additionally, in practice partnerships may dissolve for some reason even though the market would have been willing to provide capital for a follow-on fund had the partnership desired one.

49

raising just one or two funds and a few substantially more (the maximum in the sample is 12 funds). Panel B contains information about fund size and contains statistics for preceding funds and follow-on funds, with the latter presented both for the whole sample, which includes zeros for the cases where the partnership did not raise a subsequent fund, and for the subsample in which they did raise a follow-on fund. Even if we average in zeros for the cases where there was no follow-on fund, funds grew substantially, and of course the rates are even higher if we restrict the sample to those cases where there was a follow-on fund. Table 2 breaks down the sizes by type of fund by sequence number. Panel A contains information for preceding funds, Panel B for follow-on funds including zero-sized funds for the cases when partnerships did not raise a subsequent fund and Panel C for follow-on funds not including these cases. There are substantial differences in size across types of funds, with buyout funds being the largest, venture capital the smallest, and real estate in between. In addition, higher sequence number funds are substantially larger than lower sequence funds, both because they represent successful partnerships with a substantial alpha, and also because they tend to be located later in time when funds were larger. Finally, the numbers in Panel B tend to be larger than those in Panel A, indicating that funds grow over time even when one averages in zeros for the partnerships who do not raise subsequent funds. The numbers in Panel C are of course even larger because they do not average in the zero funds. Table 3 reports statistics on fund growth rates (the proportional difference between the

50

preceding fund and the follow-on fund). Once again we break down the growth rates by type of fund and present the data with follow-on funds of size zero (with growth set equal to -1) included in Panel A, and without those funds in Panel B. Similar to the numbers on fund size, buyout funds have grown faster than venture capital funds, and the growth rates are (by construction) much higher when the growth rates of -1 are not included in Panel B. Table 4 presents fund-level returns (Panel A) as well as statistics on the time between fundraisings (Panel B). The mean annual return over the entire sample is 16% for buyout funds, 14% for venture capital funds and 15% for real estate funds, with somewhat lower median returns. These numbers are similar to those in Kaplan and Schoar (2005), who report average returns of 19% for buyout funds and 17% for venture capital funds (p. 1798). The similarity with Kaplan and Schoar (2005), who use a different data source (Venture Economics) and a different time period (their sample ends in 2003), is reassurance that our data do not suffer from important biases missing from data used in prior work. The time between fundraising averages 3.3 years for the entire sample, with somewhat longer periods for buyout funds (3.76 years) and shorter periods for real estate funds (2.36 years). As funds become more established, they tend to raise funds more quickly, with an average time period of 3.95 years between the first and second fundraisings, and averages of less than 3 years for all fundraisings after the fourth.

51

2.4. The Empirical Relation between todays Returns and Future Fundraising 2.4.1 Calculating indirect incentives for different types of funds We next estimate regressions that predict the growth in the capital committed to the subsequent fund of a particular partnership as a function of the return in the partnerships prior fund. In doing so, we have two goals: first, the models main empirical predictions concern the relation between one funds performance and future fundraising, so these equations allow for testing of these implications. Second, we can use these estimates to calculate the implicit incentives for general partners arising from the possibility of future fund raising. Table 5 contains the results of regressions predicting the growth rate between a partnerships current fund and its next one as a function of the IRR of the current fund. Panel A uses the raw growth rate, while Panel B reports results using the natural logarithm of the growth rate as the dependent variable, similar to the specification used in Table 10 of Kaplan and Schoar (2005). We report results separately for buyout, venture, and real estate funds, as well as results pooling all types (with type dummies included). Each cell reports three specifications: the first uses IRR by itself as the independent variable, the second contains vintage year fixed effects, while the third uses IRR adjusted by the benchmark return provided by Preqin (the median IRR of all other funds of the same type with the same vintage year and geographic focus). Because the dependent variable is bounded below at a growth rate of -1 (reflecting partnerships that do not raise

52

subsequent funds), we estimate all regressions using Tobit. 19 In all specifications, we cluster standard errors at the partnership level. The results in Table 5 suggest that regardless of the specification used, a higher IRR in the current fund leads to a higher fund size in the future. The coefficients on IRR are all positive and statistically significantly different from zero. This pattern is true for both the specifications using raw growth and the log of growth as a dependent variable, although the R2 values are substantially higher for the logarithmic specifications, indicating that these specifications fit the data better. However, there are substantial differences across types of funds in the estimated sensitivity of future fund size to current fund IRR. The coefficients on buyout funds are noticeably larger than those on real estate funds, which are themselves larger than the coefficients for venture capital funds. These differences suggest that buyout funds are the most aggressive at increasing fund size in response to good performance while venture capital funds the least aggressive. This pattern is very persistent and is consistent with the idea that buyout funds are the most scalable; in the context of the model presented above, the results suggest that the buyout funds have the highest ().

Several measurement issues are of concern with these regressions and potentially cloud their interpretation. First of all, the measure of performance used, the funds IRR, is calculated at the end of the funds life, and will not be known precisely by market participants at the time the subsequent fund is raised, although they are likely to have a

The pattern is similar but the coefficients on IRR are much larger if we eliminate these observations and estimate the equations using ordinary least squares.

19

53

reasonably good idea of the funds performance through the exits of its first few deals. This problem is likely to be most severe when the follow-on fund is raised very quickly following the initial one. In addition, some large fund companies operate multiple types of funds simultaneously, such as one focusing on American buyouts and another focusing on European ones. The performance of the American funds is likely to be reflective of the ability of the American partners and relatively uninformative about their European counterparts. In this case, there are likely to be multiple fundraisings shortly after one another but the reason for one being able to raise a large fund is likely to have little relation to the performance of some of the partnerships other funds performance. While a perfect solution to these issues is impossible because we do not have access to the precise information available to market participants at the time of the fundraising and we do not know the identity of the actual partners involved with particular deals, it seems likely that both of these issues are particularly problematic when funds are raised shortly after one another. Consequently, we repeat the analysis conducted in Panels A and B, eliminating all preceding-follow-on fund pairs that are raised less than three years apart. 20 We present the results including only funds with a three-year gap in fundraising in Panels C and D of Table 5. This restriction causes us to lose about a third of our observations for buyout and venture funds and well over half of our real estate funds since real estate

Suppose funds were raised in years 1994, 1996, 1997, and 1998 (situations like this are not common in our data). For our main analysis, we code the 1996, 1997, and 1998 funds as follow-on funds to the 1994, 1996, and 1997 funds, respectively. For this robustness analysis, we drop all three pairs because no pair involves at least a three-year gap. An alternative approach would be to treat the 1994 fund as the preceding fund for the other three funds, and use the performance of the 1994 fund to predict the fund size of the following three funds. Another alternative would be to count the 1997 fund as the only follow-on fund to the 1994 fund. Both of these alternative screening methods yield qualitatively similar results to those presented in Panels C and D of Table 5.

20

54

funds typically raise new funds relatively quickly. Nonetheless, the coefficients on the IRR of the initial fund are all still positive, mostly still statistically significant, and of approximately the same size as those in Panels A and B. Consequently, it appears that the issues of not knowing performance perfectly at the time of the fundraising and of unrelated funds being grouped in the same partnership are not serious issues in interpreting the coefficients from our analysis.

2.4.2. Indirect incentives of older and younger partnerships One prediction of the model presented above is that the sensitivity of fundraising to performance decreases over time, as a partnerships ability becomes more well-known to investors. To examine whether this prediction holds in the data, in Table 6 we add to the regressions in Table 5 variables for the funds sequence number as well as the sequence number interacted with IRR. The results in Table 6 indicate that the effect of sequence is two-fold: First, higher sequence funds tend to grow the fastest as the coefficient on Sequence number is positive and statistically significantly different from zero. Second, the interaction of sequence number with IRR is negative and significantly different from zero in most specifications. This pattern of coefficients is consistent with the model. Higher sequence numbers are associated with funds that have done well historically and hence have high _s, so they grow the fastest, but _ is estimated more precisely over time, so the marginal impact of current returns on future fundraising grows smaller over time.

55

2.4.3. Indirect incentives and fund size In Table 7 we present analogous regressions to those in Table 6 for fund size instead of sequence, entering the natural logarithm of the preceding fund size and the interaction of this variable with preceding fund IRR. If fund size is positively related to the precision with which the market estimates , we would expect that larger funds would have a lower sensitivity of future fundraising to current IRR than smaller ones, because current performance would not lead to as much updating of . The results in Table 7 are not particularly supportive of this idea. The coefficients on size-interacted IRR tend to be negative, but not significantly different from zero. These results suggest that fund size is more reflective of the markets assessment of but that the precision of that assessment does not vary strongly with firm size.

2.5. General Partner Incentives Implied by the Regression Estimates 2.5.1. Basic results The positive relation between a funds performance and subsequent fundraising indicates that fund managers do receive some indirect, market-based incentives from the possibility of raising funds in the future. But how large are these incentives, both in absolute terms and relative to the direct incentives offered by the carried interest in the current fund? What factors determine the magnitude of these incentives across funds? To answer these questions, one must convert the coefficients from Tables 5 and 6 to a more useful metric for making comparisons. We consider two such metrics, the expected change in dollar GP revenue for each additional percentage point of IRR in the initial 56

fund, and the expected change in dollar GP revenue for each dollar returned to the investors. Equations (9) and (11) from the model presented above provide a method of converting the coefficients in this manner. In this section we do this conversion and calculate the magnitude of implied indirect incentives to general partners. As a benchmark, we first calculate the change in the direct income the general partners receive from an additional percentage point of IRR. We do this calculation assuming that the fund is of mean size for each type of fund, earns the mean IRR and General Partners carried interest is in the money, so that the partners receive the standard 20% of the gross profits, (which equals 25% of the amount of incremental income returned to the limited partners, who receive 80% of the gross profits). Because it assumes the carried interest is in the money, this calculation is therefore an overestimate of the expected change in general partners compensation for an addition dollar returned to the limited partners; for a fund whose capital is not yet returned to investors or who has not met the hurdle (typically 8%), the incremental effect of an percentage point return on general partners compensation is zero. These estimates of the direct effect of an incremental percentage point return are presented in Panel B of Table 8 for the base-case equation taken from Table 5. For an average buyout fund with $853m in capital, an additional percentage point in IRR means an additional $8.7m in carried interest for the general partners. For an average-sized

57

($215m) venture capital firm the effect is smaller, equal to $2.1m for each percentage point return. 21 The calculation of the indirect effect for our base case regressions is the product of four terms: INk. First, for the regressions with raw fund growth (log fund growth) as the dependent variable, we use the coefficients associated with lag(IRR) from Table 5, Panel A (Panel B) for each type of fund using specification (1). Second, we multiply this regression coefficient by the mean size of preceding funds (when the dependent variable is log of fund growth, we also add the mean size of follow-on funds, as a proxy for expected follow-on size). The third term in this product is N, the number of future funds raised. Finally, we multiply by k, the fraction earned by the GP. There are a number of factors that affect the calculation. First, and perhaps most important, is the type of fund. As previously discussed, more scalable funds are more able to utilize increased capital commitments profitably, leading to a stronger empirical relation between fund performance and subsequent growth in capital. We present calculations of these indirect incentives for each type of fund, and also using both the raw growth rate and logarithmic specifications. A key component of this calculation is the k parameter, defined as the fraction of fund value that will be paid out to the general partners in expectation as a combination of management fees and carried interest. The appropriate value of k is not obvious as it depends on the fee structure as well as the entire distribution of returns, which matters because it affects the likelihood of the carried interest being in the money and the
The large difference between the two comes simply from differences in fund sizes and serves to emphasize the importance of fund size in determining general partners compensation.
21

58

amount that it will be worth conditional on being in the money. Our analysis relies on the work of Metrick and Yasuda (2010), who perform simulations using details of the compensation structure in the partnership agreements of venture capital and buyout partnerships as well as data on the distribution of fund returns. Metrick and Yasuda provide estimates of the distribution of k for venture capital and buyout funds, and we use similar values for real estate funds (not considered by Metrick and Yasuda) and the overall sample of funds. In addition, the number of funds a partnership will manage over its lifetime clearly has an important impact on the importance of future fundraising. This variable is unobservable; most successful partnerships are still active albeit sometimes with different individuals as partners. We calculate indirect incentives for varying values of N, the parameter that represents the number of future funds the partnership represents. Finally, the indirect incentives depend on the size of the fund, as the dollar incentives for a given percentage point return are magnified when a fund is larger. Panel A of Table 8 calculates the magnitude of the indirect effect, in terms of dollars in compensation to general partners per percentage point of return, and also per dollar returned to the investors. A convenient way of characterizing this effect is to take the ratio of these indirect incentives to the direct incentives from the carried interest in the current fund. We present these ratios in Panel C of Table 8. It is evident from Panel C that the ratios are large. Recall that the direct incentives are substantial and are commonly credited as an important source of value in leveraged buyouts. Yet, the ratios reported in Panel C of Table 8 are sizeable; most values are 59

greater than 1 for buyout and real estate funds, with many greater than 2. Clearly, the incentives implicit in the ability to raise capital in the future are large, and should be much more emphasized in discussions of the drivers of private equity value creation. The observed ratios are (by construction) increasing in both N, the number of future funds to be raised, and k, the fraction of fund value that goes to the general partners. While it is impossible to know which is the correct value of each, the findings do suggest, not surprisingly, that the indirect incentives are most important for younger partnerships (who have more years to reap the benefits of a reputation for ) and partnerships with higher fee structures. In addition, the ratios are substantially higher for buyout partnerships than for venture capital partnerships, with real estate partnerships in the middle. This effect comes directly from the higher regression coefficients for buyout than for venture capital from Table 5, and is likely due to the higher scalability of buyout investments than venture capital ones.

2.5.2. Indirect incentives over the partnerships life The model presented above suggests that incentives from future fundraising are most important, not only for younger partnerships, who have the most time left in their careers, but also for newer partnerships, who have not had time to establish a very precise . To consider the quantitative importance of this effect, we present calculations of indirect incentives for partnerships by fund sequence, using the estimated equations presented in Table 6. Table 9 contains these calculations for a fund with Sequence equal to 1, 3 and 5, using the estimated equations for All Funds from Table 6. 60

The results in Table 9 suggest that the incentives from future fundraising are substantial for new partnerships and decline rapidly. For funds with Sequence = 1, the incentives from fundraising are substantial, with ratios of about 1 relative to direct incentives. The estimates clearly indicate that the career-oriented incentives are substantial, especially for newer partnerships made up of younger individuals who have the potential to raise many funds in the future. The fundraising incentives decline dramatically over time. By the fifth fund in a partnership, they are close to zero, as indicated by the low ratios in Panel C of Table 9. This pattern is consistent with the model presented above, in which alpha is known fairly precisely after several observations of returns, so each additional return has little incremental information. This pattern is illustrated in Figure 1, which contains a graph of the estimated ratio of the indirect to direct incentives as a function of sequence number for varying N. For the first fund in the partnership (Sequence = 1), the ratio is highest for higher Ns. It declines with the Sequence number for each value of N, becoming negative for each around fund 5. While theoretically, this ratio should never be negative (as that would imply negative incentives from future fundraising), this pattern is likely due to the imposition of a linear functional form in sequence number. In future work we hope to explore this issue more carefully using nonlinear specifications.

61

2.5. Discussion and Conclusion The analysis presented above suggested a mechanism by which the possibility of future fundraising can provide incentives to general partners in private equity funds. Partnerships establish a reputation for being able to generate returns early in their careers, which can be lucrative later on as it allows partners to raise larger funds subsequently. We provide a formal model of this process and empirical work consistent with the predictions of the model. An important innovation is that the model provides explicit formulas for expected lifetime compensation for private equity partners that allow calculation of the magnitudes of the incentives implicit in the development of a partnerships reputation. These incentives are of the same order of magnitude as the direct incentives from carried interest that have been commonly credited with much of the value creation in private equity. They are particularly important for newer partnerships who have yet to establish a reputation and for younger partnerships, who have the potential to reap the benefits of the reputation for a long period of time. Clearly there are a number of measurement issues that affect the interpretation of the results. The numbers discussed throughout the paper are undiscounted, which overstates their value since the payoffs to a better reputation are both risky and in the future. On the other hand, it seems likely that aside from the discounting issue, the ratios presented in Tables 8 and 9 understate the true career-oriented incentives in private equity. The calculation of the direct effect is likely an overestimate, as it assumes that general partners keep 20% of each incremental dollar while in fact they get less than that in expectation. In contrast, the calculation of the indirect incentive effect ignores the 62

possibility that individual partners can use their personal reputations to raise new funds on their own, or join other existing firms for lucrative salaries. These possibilities, which do occur fairly regularly, are examples of valuable reputational capital acquired by general partners through earning high returns that the formal analysis in this paper does not consider. This paper contributes to the debate about the incentives of private equity managers and their effect on value creation. Metrick and Yasuda (2010) find that roughly two-thirds of the compensation in private equity partnerships comes from fixed rather than variable components of compensation. Our results suggest that their calculations understate the total incentive compensation that general partners have, and that incentive-based compensation in private equity partnerships is larger than previously thought. The analysis in this paper could be easily applied to other forms of organization. Perhaps the most straightforward would be to other parts of the money management industry, because the explicit fee structures in this industry allow for straightforward calculation of the returns to managing a larger quantity of funds. Hedge funds have a somewhat different institutional structure than private equity funds with their infinite lives and their compensation system based on the high-water mark. Nonetheless, it should be possible to do similar calculations for hedge funds as was done in Section 5 of this paper for private equity funds. In addition, mutual funds and private management of large institutional money, although with a different fee structure, have the same implicit incentives operating through the channel of inflows chasing returns. Calculating the

63

incentives implicit in this inflow-return relation would be an important addition to our understanding of these industries. Most generally, the paper provides some empirical content to the idea started by Fama (1980) and Holmstrom (1982, 1999) that career concerns can be an important source of incentives inside firms. Holmstrom in particular argues that in an intertemporal setting, agents will take actions to maximize peoples perception of their abilities, which can but do not necessarily coincide with increasing a firms profitability. The advantage of focusing on a private equity setting as we do here is that it is possible to quantify the long-term pecuniary benefits to agents from these perceptions. Private equity is nonetheless an industry where incentives, both direct and indirect, are particularly important. The extent to which indirect, market-based incentives are important in other industries both in absolute terms and relative to direct incentives, is likely to be an important topic of future research.

64

Chapter 3: Performance Persistence in Private Equity Funds

3.1. Introduction Does past performance contain information about future performance in private equity funds? How much information is contained? How strong and persistent is the relationship between current fund and future fund performance? What economic factors explain the relationship? What are the implications of this relationship for investors in private equity? In this paper, we provide empirical answers to these questions. While many studies have examined whether performance persists in the mutual fund and hedge fund industry which are important asset classes of institutional investors, few studies have investigated this issue in the context of private equity funds. Recently, Kaplan and Schoar (2005) document persistence of private equity fund performance. They ascribe the persistence to differential and proprietary skills of general partners (GPs), leaving a puzzling question of why GPs do not appropriate more rents from investors, for example, by charging higher fees. Several subsequent studies such as Glode and Green (2008) and Hotchberg, Ljungqvist, and Vissing-Jorgensen (2009) 22 have attempted to rationalize this phenomenon. Obviously, whether, how, and how much performance persists are important questions for investors in private equity funds. The answer to these questions will affect limited

22

Both studies assume that incumbent GPs have information (over outside potential GPs) which can be used to lever their bargaining power against LPs.

65

partners (LPs) investment strategies, their relationship with GPs, and the terms of contract between the two parties. Given the importance of the subject from both practical and academic perspectives, we believe a more thorough study is needed to shed light on aforementioned unanswered questions: We do not yet know the form, the strength, and the economic magnitude of performance persistence in the private equity industry. We examine performance correlation between current fund and the next follow-on funds as well as the second and third follow-on funds using methodologies typically found in the mutual fund literature: Contingency table tests, Pearson and Spearman correlation, and cross sectional regression of future fund performance on current performance. To quantify the economic magnitude of performance persistence, we also measure the performance of quartile portfolios ranked on past performance. We find that, consistent with some previous studies, 23 current fund performance is significantly and positively related to the next fund performance. However, current fund performance is not or weakly correlated with the second follow-on funds and after. Currently best performing and worst performing funds perform similarly in their second and third follow-on funds. Therefore, performance persistence is short-living, and, more surprisingly, performance converges across funds over time. What are the implications of this finding for investors in the private equity industry? First, the short-living performance persistence makes one doubt whether GPs have differential skills. Second, LPs should be careful in using information contained in the

For example, Kaplan and Schoar (2005), Phalippou and Gottschalg (2009), and Axelson, Jenkinson, Stromberg, and Weisbach (2010) document a statistically significant association between current and the first follow-on fund performance.

23

66

second or prior funds because the information is very likely staled. Typically, a private equity firm raises a following fund three to five year after a preceding fundraising. In other words, at the time of a fundraising, investors cannot fully evaluate the performance of the most recently raised fund, although the interim performance may have valuable information about the final outcome at the end of a funds life. Therefore, investors are likely to base their investment decision on the performance information contained in several preceding funds. However, our finding suggests that the second or prior funds performance can be a misleading indicator for future performance. Next, we investigate why performance persists transiently. We examine two potential explanations. First, we examine the effect of common market environment on persistence. Since a private equity funds life is about ten years, and a follow-on fund is usually raised three to five years after a preceding fund raising, successive funds share several years of overlapping investment period during which the common economic condition or shocks can influence the performance of preceding and following funds simultaneously. Therefore, persistence can be affected by the length of overlapping investment period of successive funds. Second, we examine the effect of capital inflows on persistence. As Berk and Green (2004) argue in fact, assume if investors of private equity direct capital to the managers with higher past performance and the inflow of capital is associated with a decline in future performance, performance persistence will disappear. We show that better performing funds raise larger follow-on funds than worse performing counterparts, confirming the findings of Kaplan and Schoar (2005) and Chung, Berk, 67

Lea, and Weisbach (2010). However, funds which have grown more significantly underperform subsequently. The return-chasing-capital phenomenon is slightly more pronounced for buyout funds, and the diminishing returns to capital inflows is only found among venture capital funds. These results suggest that the resources and skills necessary for managing venture capital funds are not readily scalable compared to buyout funds. Importantly, capital inflows conditional on performance reduce performance persistence. One standard deviation increase in fund growth reduces performance persistence by about 0.17 (or 44%). A 2.25 standard deviation increase in fund growth which occur with a probability of about 3.1% completely eliminates persistence again only for venture capital funds. Next, we find that the longer the time gap (the number of years elapsed) between two consecutive fund raisings, the less the performance persistence is. In other words, as the duration of overlapping investment periods becomes shorter, there is less performance correlation between current funds and follow-on funds. On average, one year increase in the number of years between two successive funds leads to 0.10 (or 27%) decrease in performance persistence. However, we find this evidence only for buyout funds. While we find that capital inflows and overlapping investment period affect performance persistence, the former is significant only for venture capital funds and the latter is more important for buyout funds. This asymmetry appears to be consistent with the view that venture capital industry is labor-incentive and buyout industry is capital-intensive. In managing portfolio companies of a venture capital firm, fund managers, often themselves, provide not only capital but also various kinds of resources such as industry 68

network and management skills. An increase in fund size which will in turn increase either target size or the number of investments will require greater amount of management care. To the extent that the resources of a venture capital firm are not quickly or readily scalable, capital inflows will likely lead to a decline in performance. Overall, venture capital funds show a similar pattern as mutual funds (Sirri and Tufano (1998), Lynch and Musto (2003), Chen, Hong, Huang, and Kubik (2004)) or hedge funds (Fung, Hsieh, Niak, and Ramadorai (2008)) in that LPs direct more capital toward funds with superior performance and funds face decreasing returns to capital inflows. The performance of buyout funds, on the other hand, largely depends on macroeconomic condition such as credit market. Since each buyout transaction incurs large amount of leverage, pricing and liquidity of credit directly affect buyout performance. 24 Funds raised close to each other in time, therefore, are likely to be affected by common macroeconomic conditions and to exhibit higher performance correlation. We find more direct evidence that the similarity of macroeconomic conditions increases performance correlations of successive funds. Specifically, we find that similar IPO market condition, stock market performance, GDP growth and credit market conditions over two successive funds lives increase performance persistence. The overall evidence suggests that performance of private equity funds persists only for short-run and even this short-run persistence appears to be largely explained by common market conditions facing neighboring funds. This result does not support the view that

For example, Axelson, Jenkins, Stromberg, and Weisbach (2010) show that the economy-wide cost of credit drives both the quantity and the capital structure of buyout transactions, although they do not directly examine the relationship between credit market condition and performance.

24

69

GPs have differential and proprietary skills. However, the short-run persistence is also explained by excessive capital flows conditional on past fund performance. According to the Berk and Green (2004) model, the return-chasing capital and diminishing returns to capital flows do not necessarily confront the view that GPs have differential ability. Therefore, the findings in this paper are not definitively conclusive about whether GPs have unique skills or not. But the findings have important implications for investors in the private equity industry. The next section briefly describes the data. Section 3 tests performance persistence and examine the dynamics of it. Section 4 investigates the causes of performance persistence in the short-run and Section 5 concludes.

3.2. Data We use fund-level data such as vintage year, funds IRRs, and fund size provided by Preqin. 25 We consider buyout and venture capital funds. The total number of buyout (venture capital) funds is 2,250 (4,588), out of which 888 (1,157) funds report IRRs. See Table 1 for the distribution of funds by fund type and vintage year. In later analyses we estimate the relation between current fund performance and follow-on fund performance. In doing this we aggregate funds in a given year and compute the fund size weighted IRR when a private equity firm raises multiple funds in a given year. Usually, large private equity partnerships tend to raise multiple funds targeting different investors or geographical regions. For each preceding fund, we ask whether we observe a follow-on

25

See Chung et al. (2010) for more detailed description of Preqin.

70

fund in the database. We define a follow-on fund as the next fund raised by the same partnership. Thus each preceding fund is allowed to have at most one follow-on fund. Table 1 reports the descriptive statistics (1st quartile, mean, median, 3rd quartile, and standard deviation) on fund performance and fund size by fund type (buyout or venture capital) and vintage year. Fund size information is available in Preqin for most funds, about 90% of funds reporting fund size data. However, performance data (IRR or multiples) is missing for the majority of the funds in Preqin. Figure 1 plots fund performance and committed capital by vintage year. The time trend of committed capital is largely consistent with other studies (e.g. Acharya et al. (2009), Stromberg, (2009)) using a different database (e.g. Venture Economics or Capital IQ). There are peaks in fund raisings in 2000 and 2007. After the financial crisis in 2007-8, there is a large drop in fund raising activities in the private equity industry. We see substantial underperformance of venture capital funds since the late 1990s in Table 1. This is not solely due to the fact that many of the recent funds are not yet completely liquidated (i.e. not completely realized). 26 Even when we include only liquidated (realized) funds post-2000, we find a similar trend, i.e. venture capital performance is substantially lower post-2000 than previous years. Comparing time series of venture capital and buyout fund performance, we observe that buyout funds tend to comove together in terms of returns. Also the standard deviation of buyout funds is almost half of that of venture capital funds: 23.4 versus 50.1. This
26

Industry practitioners often show concerns about poor performance of venture capital given its level of risk taking in recent years. For example, see an article by Ray Maxwell (http://altassets.net/private-equityfeatures/article/nz18642.html) or by Claire Miller (http://www.nytimes.com/2009/07/07/technology/startups/07venture.html).

71

suggests that buyout funds tend to be more affected simultaneously by some common factors than venture capital funds. We examine this issue further in Section 4.3.

3.3. Testing Performance Persistence We examine whether GPs of better (worse) performing funds tend to continue to outperform (underperform) others in subsequent funds. 27 To this end, first, we compute the conditional probability that a partnerships subsequent funds will either stay in the same performance quartile as the current funds, or move into one of the other three quartiles (Section 3.1). If funds in one portfolio tend to stay in the same portfolio, this will suggest that there is persistence in performance. Second, we test performance persistence by examining Pearson and Spearman correlation between current fund performance and follow-on fund performance in Section 3.2. Third, we examine performance persistence in a multivariate regression framework in Section 3.3 where we will be able to see whether performance persists after controlling for relevant factors. Lastly, we form quartile performance portfolios by ranking current fund performance and trace subsequent performance of initial performance quartile portfolios in Section 3.4. By

Unlike investors in mutual funds, investors in PE funds cannot freely move capital from one to another fund in response to an interim performance result. They are basically locked up with a PE partnership for ten to thirteen years by contract (although secondary markets exist where claims on PE partnerships can be traded at a deep discount prior to fund liquidation). This means that even if a fund is performing well (bad), an immediate cash inflow (outflow) does not follow. The only decision that they can make is whether to invest in a follow-on fund by the same GPs which is typically raised three to five years after the preceding fund raising. Therefore, investors have to make this decision without fully observing the final outcome of the preceding on-going fund. However, surprisingly, the interim IRR at the end of third year of a fund is a good proxy for the final IRR. Even though the IRR in early years of the fund life is typically negative due to cash outflows associated with new investments, Spearman correlation between interim IRRs and final IRRs is over 90%.

27

72

observing subsequent performance will enable us to measure the economic magnitude of performance persistence. In the following analyses, we use unadjusted IRRs, as opposed to risk or style adjusted IRRs. Obviously, it is difficult to know what kind of risks and how much risks private equity investments are exposed to. We do not know what kinds of investments are made by a fund, nor the risk characteristics of those investments. Therefore, we do not attempt to adjust for risks. Alternatively, we may adjust investment returns by benchmarking with a funds investment characteristics (similar to Daniel, Grinblatt, Titman, and Wermers (1997) for mutual funds). For example, private equity investment returns can be adjusted by the returns of portfolio of other private equity funds with same industry focus, geographic focus and so on. This style adjustment is difficult to implement, however, because, unlike mutual funds, there are substantially small number of funds in the market, which limits the number of style dimensions that we can employ. Second, it is difficult to observe detailed investment (i.e. portfolio company) characteristics of a private equity fund. In addition, it is not clear why one needs to adjust for such risk or style. First, private equity industry is inherently risky, the reason why only accredited investors or qualified investors are allowed to invest. Second, unlike mutual funds, private equity funds have absolute target returns, not benchmarked target returns. Taken together, it is less important for private equity partners and investors how much risk they take or what kind of style of investments they make. For these reasons, we focus on the examination of raw investment returns. But we also provide results using benchmark adjusted IRRs in 73

Appendices. The styles used to form benchmark portfolio are vintage year, geographic focus, and stage/investment type (early stage, later stage, expansion, or buyout). The benchmark adjusted IRRs are raw IRRs minus the median of IRRs of benchmark portfolio.

3.3.1. Transitional Probabilities This is the first test to examine whether there is any association between current fund and future fund performance. Table 2 reports transitional probabilities, i.e. the conditional probability that a partnerships subsequent funds will either stay in the same performance quartile as the current funds, or move into one of the other three quartiles. If funds tend to stay in the same performance quartile over time more than expected, this will suggest that performance tend to persist. The first rows of each panel (the raw headings with "Expected") report the expected fraction of funds which should belong to one of the performance quartiles under the assumption that the classification into one of the follow-on performance quartiles is purely random. We compute transitional probabilities from current fund performance quartile into the first through third follow-on fund performance quartile to see how long performance would persist. We use unadjusted IRR as a measure of performance (reported in Table 2) and the results using benchmark adjusted IRRs are reported in Appendix A. The probabilities that current funds in the top performing portfolio (1st quartile in the second row) stay in top performing portfolio in their follow-on funds (1st quartile in the 74

column heading) are 37.08% and 35.46% for buyout and venture capital funds, respectively. These probabilities are substantially greater than the expected probabilities which are 21.14% and 26.59% for buyout and venture capital funds, respectively. The bottom performing portfolio (4th quartile) also tends to continue to underperform subsequently. The conditional probability that current funds in the 4th quartile portfolio stay in the same quartile is 45.28% and 43.68% for buyout and venture capital funds, respectively, which are almost twice as large as the expected probabilities. The Chisquare tests also reject the null hypothesis of no association between current and followon fund performance. Therefore, there is a strong persistence from current fund performance to the first follow-on fund performance. We also examine transitional probabilities from current fund to the second and third follow-on funds in the second and third sub-panels of Panel A and B. Here we do not find strong evidence that funds in one performance quartile tend to stay in the same performance quartile later. For buyout funds, it appears that funds tend to stay in the same quartile in their second follow-on funds as suggested by marginally significant Chisquare statistics. However, the Chi-square test does not reject that the transition from current fund quartile to the third follow-on fund quartile portfolio is random. For venture capital funds, all the transitional probabilities do not show a statistically significant nonrandomness. The results are almost identical when using benchmark adjusted IRRs instead of unadjusted IRRs (See Appendix A). In sum, even though there is some performance persistence from the current funds to the right next follow-on funds, performance persistence does not last after the first follow-on 75

funds except buyout funds transition from current to the second follow-on funds. We confirm this result using different methodologies below.

3.3.2. Correlation between current fund performance and follow-on fund performance We test performance persistence using Pearson and Spearman correlation between current funds performance and the first, second, and third follow on funds performance. The results are presented in Table 3. The corresponding results using benchmark adjusted IRR as opposed to raw IRRs are reported in Appendix B. Panel A presents Pearson correlation and Panel B reports Spearman correlation. First row of each sub-panel use all funds which have IRR data for any of their following funds. The second through fourth row of each sub-panel requires that a fund raise first through third follow-on funds and report IRR data for those funds. For example, the row heading with F~F+2 requires that funds have IRR data for their first and second follow-on funds. Pearson correlation between current funds IRRs and the first follow-on funds IRRs is 35% and 16% for buyout and venture capital funds, respectively, which are statistically significant at the 1% level. Pearson as well as Spearman correlations between current funds and the second and third follow-on funds are not statistically significant for venture capital funds. But for buyout funds Pearson and Spearman correlations between current funds and the second follow-on funds are 12% and 17%, respectively, and statistically significant at the 10% and 5% level, respectively, when we use all funds with IRR data in any of the following funds. However, note that the linear relationship between current funds and second follow-on funds drop almost by half: from 35% to 12% for buyout and 76

from 29% to 17% for venture capital funds. When we require IRR data to be available for all follow-on funds, there is no association between current fund performance and all follow-on funds performance. Essentially, the results are qualitatively identical to those from the test using transitional probabilities. There is some performance persistence from the current funds to the right next follow-on funds, but performance persistence does not tend to last after the first follow-on funds.

3.3.3. Multivariate regression Next, we show the association (persistence) of current and future fund performance in a multivariate regression framework. We regress current fund performance on the first, second, and third previous fund performance, each of which corresponds to the first, second, and third panel in Table 4. Specifically we estimate the following regression equation: (IRR) i,t = + (IRR)i,t- + (Fund Size)i,t- + i,t, where is 1, 2, or 3. (1)

(IRR)i,t- and (Fund Size)i,t- are the -th previous fund IRR and fund size, respectively. If the coefficient estimate of is positive and significant, this will suggest that past performance contains information about future fund performance. In Table 4, columns (1) and (2) report the estimates for buyout and columns (3) and (4) for venture capital funds. In columns (1) and (3), unadjusted IRRs (logarithmized) are used for the estimation and vintage year fixed effects are included. We drop funds which 77

were raised after 2005 to eliminate the potential bias by using interim IRRs of unrealized (unliquidated) funds of recent years. Using an earlier cut such as 2000 does not alter the results both qualitatively and quantitatively. In column (2) and (4), we use benchmark adjusted IRRs (again logarithmized) instead of unadjusted IRRs. Benchmark adjusted IRRs are the differences between unadjusted IRRs and benchmark IRRs which are the median IRRs of portfolios of funds with same vintage year, geographic focus, and investment type (early stage, later stage, expansion, or buyout). Standard errors are clustered at the PE firm level. Current fund performance is not strongly associated with the second and third previous funds for both buyout and venture capital funds. For buyout funds, in the specification (1) where we use unadjusted IRRs current performance is marginally statistically significantly related to second previous fund performance. Other than that, all the other coefficients are statistically insignificant, generally being consistent with the results from tests using transition probabilities and correlations. By and large, performance persistence lasts for only one "period" for both buyout and venture capital funds. In other words, only performances of two consecutively raised funds are significantly related. However, as we will see this statistical significance somewhat overstates economic significance. The estimates are quantitatively similar regardless of different specifications. So, lets focus on column (1) and (3) where we use unadjusted IRRs to estimate the persistence of performance. The coefficients on the first previous fund's IRR are 0.444 for buyout and 0.234 for venture capital funds. Since both independent variables and dependent variable are of logarithmic, we can interpret the coefficients as the elasticity of current funds IRR 78

on the preceding fund IRR. Therefore, one percentage increase in the first previous funds IRR lead to a 0.444 (0.234) percentage increase in current funds IRR for buyout fund (venture capital funds). Put differently, since the average and standard deviation of logarithmic IRRs of the first previous funds are 0.783 and 0.086 for buyout and 0.781 and 0.182 for venture capital funds, one standard deviation increase in the first previous fund performance will increase the average current fund performance by 3.5% for buyout funds and 3.8% for venture capital funds. Our result that performance persistence is short-living may seem different from that of Kaplan and Schoar (2005). Table VII of Kaplan and Schoar (pp. 1806) reports that there is a statistically significant relation between current fund performance and second previous fund performance. Note first that they do not find such a significant relationship when they estimate the regressions for buyout and venture capital funds separately. The discrepancy may come from using different measures of private equity performance. Kaplan and Schoar use, what they term, public market equivalent which is the ratio of discounted cash inflows and cash outflows where contemporaneous S&P 500 returns are used for discounting. Our measure of performance is internal rate of return reported by private equity partnerships. Another source of discrepancy may be different sample period. Kaplan and Schoar study the 1980-2001 period, whereas we examine funds from as early as 1969 to 2005. In fact, when we use performance data for funds which were raise prior to 2001, we find more statistically significant association between current fund performance and the second previous fund performance but only for buyout funds. We still do not find a positive and significant association between current fund and the third 79

preceding funds. This may suggest that recently raise funds show less performance persistence.

3.3.4. Subsequent performance of initial performance quartile Lastly, we examine subsequent performance by performance quartile portfolio. In each year, we rank funds by their performance (based on IRRs) and form quartile portfolios, 1 being the best performing portfolio and 4 being the worst performing portfolio. Then we compute the average and median IRRs of the first through third follow-on funds of each initial performance quartile portfolio. By doing so, we can examine the actual economic magnitude of performance persistence. Table 5 reports the results using unadjusted IRRs. As we see in Figure 1, performance of private equity funds varies substantially by vintage year. Therefore, it seems reasonable here to adjust raw IRRs with vintage year benchmark IRRs or use IRR rankings instead of the actual IRRs to mitigate this "vintage effect." The results are qualitatively identical whichever performance measure we employ. See Appendix C for the results using benchmark adjusted IRRs. In Table 5, the column heading F reports the mean and median IRRs of current funds which are used to form quartile portfolio. And the column headings F+t where t=1, 2, or 3 report the mean and median IRRs of the t-th follow-on funds of each initial quartile portfolio. Panel A reports the results for buyout funds, and Panel B for venture capital funds. In the first sub-panel (first five rows), the mean and median IRRs are computed using all funds with performance data available in Preqin. Since performance reporting is voluntary, often private equity partnerships intermittently report performance information 80

and there are many missing performance data. To see whether fund survivorship into the database affects the results, we also require funds to have the first through third follow-on fund performance data and report the results in the sub-panel B, C, and D using this survivorship free sample. For example, sub-panel D requires that a fund raise the first, second, and third follow-on funds and report IRR data for all those follow-on funds. The average (median) IRR of currently top performing portfolio (Portfolio 1) is 39.5% (33.0%) (See column F) and that of bottom performing portfolio (Portfolio 4) is -4.4% (2.3%) for buyout funds (Panel A). The differences in means and medians between two extreme portfolios are statistically significant. Let's examine the first follow-on fund performance (Column F+1). The mean (median) of the initially top performing portfolio drops to 24.4% (19.7%) and that of the initially bottom performing portfolio increases to 7.7% (6.2%). Again, the difference of performance between top and bottom portfolios is statistically significant. However, the economic magnitude of the difference substantially declines as one moves from one fund to the right next follow-on fund. The differences between first and fourth portfolio are not statistically significant for the second and third follow-on funds. This result is also true for venture capital funds. In fact, when we require funds to raise the first through third follow-on funds (Sub-panel D in Panel B), even the difference between top and bottom portfolios for the first follow-on funds is not statistically significant. By the time funds raise their third follow-on funds, initially best and worst performing fund portfolios are indistinguishable in terms of performance. Graph A and B

81

of Figure 2 plot the median of current and subsequent funds' IRRs by initial quartile portfolios. We can clearly observe performance convergence across portfolios over time. When we require funds to survive (i.e. report performance data) through their third follow-on funds, we also find a similar result. Studies on mutual fund performance find that spurious persistence patterns may arise when fund survival depends on historical performance (Baquero, ter Horst, and Verbeek, 2005; Brown, Goetzmann, Ibbotson, and Ross, 1992; Carpenter and Lynch, 1999; ter Horst, Nijman, and Verbeek, 2001). It seems, however, the high attrition rate does not spuriously increase performance persistence in private equity funds. If better performing funds are more likely to raise follow-on funds or they are more likely to voluntarily disclose performance, we expect to see increases in performance as we require longer survival, i.e. as we move from sub-panel A (first five rows) to D (last five rows) in Table 5. This seems to be true. The averages of IRR in Table 5 are 19.7, 24.1, and 30.2% when requiring funds to have the first, second, and third follow-on funds, respectively, for buyout funds (Panel A) and 23.2, 29.8, and 36.8 %, respectively, for venture capital funds (Panel B). However, we still observe the same dynamic over time performance convergence across quartile portfolios. It may be that among funds in portfolio 4 (worst performing fund portfolio) only those which investors believe to perform better in the future are able to raise follow-on funds. If investors in private equity have ability to distinguish funds which were simply unlucky and those which lack skills, then we may observe improvement in performance among funds in portfolio 4 which succeed raising subsequent funds which is in fact what we see. 82

However, the same survivorship bias cannot explain performance impairment in top performing portfolios. Alternatively, it may be possible that GPs of successful PE firms those with skills leave the PE firms and start their own PE firms. This may explain why better performing funds cannot sustain its performance in the long run. Unfortunately, the data on individual GP is not available, making it difficult to test this alternative hypothesis.

3.3.5. Robustness of the results We repeat the analyses using different methodologies and sub-samples, and find the qualitatively same results. First, we redo the analyses by excluding top and bottom performing portfolios because, as we see in Figure 2, the long-run performance convergence appears to occur only for top and bottom quartile portfolios. Second, we repeat the analyses by using IRR rankings instead of IRRs to further eliminate vintage year effect, i.e. the phenomenon that private equity performance is sensitive to when a fund was raised. Third, we use quintile or decile performance portfolios instead of quartile portfolios to see the arbitrarily formed portfolios affect the results. Lastly, we drop funds which are artificially combined. As mention above, we consolidate multiple funds which are raised in the same year by a private equity firm. This could falsely generate our findings, so we repeat the study by excluding these combined funds.

83

3.4. Why (Not) Performance Persists? In this section, we try to understand why performance persists in the short-run and persistence disappears in the long-run. We first examine the interaction between fund performance and fund flows to see whether fund flows affect performance persistence. Next, we turn to the question of whether and how economic commonality influences performance persistence in private equity funds.

3.4.1. Fund flows and fund performance Previous studies find that capital tend to chase returns for mutual funds (e.g. Sirri and Tufano, 1997) and hedge funds (Fung, Hsieh, Naik, and Ramadorai, 2008), and there is a decreasing returns to scale in mutual funds (Chen, Hong, Huang, and Kubik, 2004) and hedge funds (Fung et al. 2008). Similarly, Chung, Lea, Sensoy, and Weisbach (2010) also document that superior performance leads to greater fund inflows in the private equity partnership. And Lopez de-Silane, Phalippou, and Gottschalg (2010) find a negative relation between fund scale and performance among buyout funds. Based on these recent findings, it seems natural to suspect that return-chasing capital and decreasing returns to capital flows may erode persistence of private equity fund performance. In other words, if a PE fund performs well, then it is likely that the PE partnership is able to raise a larger follow-on fund. However, if there is diminishing returns to scale, GPs who now manage larger funds will not be able to perform as well as before. Therefore, we see performance persistence declines over time.

84

Before we show whether this argument holds, we first examine the relationship between fund flows and fund performance. In Table 6, we replicate Table 5 of Chung et al. (2010) to see how current fund performance affects future fund raising. And in Table 7, we investigate how current fund inflows influences future fund performance. Specifically, Table 6 and 7 report the estimates of the following equations, respectively: Table 6: (Fund Growth)i,t = + (IRR)i,t-1 + (Fund Size)i,t-1 + i,t, Table 7: (IRR)i,t = + (Fund growth)i, t-1 + (IRR)i, t-1 + (Fund Size)i, t + i,t, (2) (3)

where fund growth, IRRs, and fund size are all logarithmized and vintage fixed effects are included in the equations. Standard errors clustered at the PE firm level. In the last two columns ((5) and (6)) of Table 6 and 7, we also include the interaction term between buyout dummy variable and (IRR) t-1 (in Table 5) and between buyout dummy variable and (Fund growth) t-1 (in Table 6) to see whether the effect of past performance or fund growth is different between buyout and venture capital funds. Columns (1), (3), and (5) in each table use unadjusted IRR as an independent and dependent variable, and columns (2), (4), and (6) use benchmark adjusted. As noted above, the purpose of using benchmarked IRR is, among others, to control for the effect of substantial variability of fund performances by vintage year. First two columns ((1) and (2)) report the estimates for buyout funds and the second two columns ((3) and (4)) are for venture capital funds. In all regressions, we include the logarithm of fund size of preceding funds as a control variable because it may be difficult for larger funds to grow at the same rate as smaller funds. 85

Consistent with Chung et al. (2010), we find positive and statistically significant coefficients on preceding funds IRRs. Past fund performance has a strong impact on follow-on fund raisings. This effect is especially stronger for buyout funds. When we estimate a regression including the interaction of buyout fund dummy variable (1 if a fund is a buyout fund and 0 if a venture capital fund) and preceding fund IRR, the coefficient for the interaction term is positive, but not statistically significant. After excluding recently raised funds (2001 as a cut), the estimate of the coefficient on the interaction term is statistically significant (not reported). This difference suggests that buyout funds are more scalable than venture capital funds. However, the scalability of buyout funds seems diminishing in recent years because the statistical significance of the interaction term disappear as we include more recently raised funds in the regression estimations. The results when using benchmarked IRRs are similar. Table 7 reports the relationship between fund flows and follow-on fund performance using unadjusted IRRs as well as benchmark adjusted IRRs. It appears that capital flows are negatively related to follow-on fund performance. However, the negative relationship is statistically significant only for venture capital funds (Column (3) and (4)). None of the coefficients of fund growth is statistically significant for buyout funds (Column (1) and (2)). When we estimate the regression including the interaction term between buyout fund dummy variable and preceding fund growth, the coefficient on the interaction term is positive (Column (5) through (6)). Therefore, a venture capital firm performance tends to decrease more when capital inflows are greater compared with a buyout firm. This result is consistent with the evidence in Table 6: The reason that buyout funds are more 86

scalable than venture capital funds may be because buyout funds suffer less from diminishing returns to capital inflows. Taken together, we find that capital tends to chase returns in both buyout and venture capital funds, but the effect seems to be slightly greater for buyout funds. In addition, venture capital fund performance is decreasing in capital inflows, but buyout fund performance is not. These findings suggest that, conditional on fund growth, buyout funds will show stronger performance persistence than venture capital funds. We investigate this point in the next section.

3.4.2. The effect of fund flows on performance persistence We examine whether fund growth affects performance persistence. Specifically we estimate the following ordinary least square equation and report the results in Table 8: (IRR)i,t = + (IRR)i,t-1 + (Fund Growth)i,t-1 + (Fund Growth)i,t-1 (IRR)i,t-1 + i,t = + [ + (Fund Growth)i,t-1](IRR)i,t-1 + (Fund Growth)i,t-1 + i,t, (4)

where (IRR)i,t-1 and (IRR)i,t are current fund and follow-on fund performance and (Fund Growth)i,t-1 is fund growth from current to the next fund. The IRR and fund growth variables are demeaned to reduce possible multicollearity problem and to ease the interpretation of the interaction term (McClelland and Judd (1993)). Vintage year fixed effects are included and standard errors are clustered at the PE firm level. The estimate of will tell us how much excessive capital inflows conditional on past performance would affect future fund performance. Also, we can measure how capital 87

inflows can affect the relationship (i.e. persistence) between current fund performance and follow-on fund performance. If the estimate of is negative, this will suggest that performance persistence decrease as a fund has grown greater as evident from the second equation. The estimate results are reported in column (1), (4), and (7) in Table 8. The coefficient estimate for the interaction term is negative and statistically significant for venture capital funds (Column (4)). This is consistent with the results in Table 7 where we find that only venture capital funds performance deteriorates as fund grows greater. The coefficient estimate is -0.392. The standard deviation of the log of fund growth is 0.435 and the mean is zero by definition; therefore a 2.25 increase in the standard deviation will completely eliminate performance persistence between two successive venture capital funds. If the log of fund growth follows standard normal distribution, the probability that a fund would grow more than a 2.25 standard deviation from its mean is about 0.012. From the empirical distribution of the log of fund growth, the probability is about 0.031. The probability that performance persistence would cut into half is approximately 0.087. In column (7), we interact buyout dummy variable with the interaction of IRR and fund growth to see whether the effect of fund growth on persistence is different between buyout and venture capital funds. The coefficient estimate is positive and statistically significant, suggesting that indeed the negative effect of fund growth on persistence is stronger for venture capital funds. In sum, one potential reason that private equity fund performance persists only in the short-run is because investors move capital in response to performance and fund 88

managers performance is negatively associated with capital inflows, especially for venture capital funds.

3.4.3. The effect of time gap on performance persistence In this section, we investigate whether common economic conditions can explain performance persistence. Typically, private equity partnerships raise follow-on funds three to five years after preceding fund raising. Since a private equity partnership usually last ten to thirteen years, two neighboring funds share five to ten years of overlap in their investment periods. These features of private equity industry imply that successive funds are likely to be exposed to common economic conditions, therefore the performance of successive funds may be highly correlated due to this economic commonality. One prediction is that the longer the investment period that successive funds share, the greater the performance correlation will be. 28 If two funds are longer apart in time, they should be more dissimilar in terms of performance. To test this, we estimate the following ordinary least square equation: (IRR)i,t = + (IRR)i,t-1 + (Time Gap)i,t-1 + (Time Gap)i,t-1 (IRR)i,t-1 + i,t = + [ + (Time Gap)i,t-1](IRR)i,t-1 + (Time Gap)i,t-1 + i,t, (5)

where (Time Gap)i,t-1 is the log of the number of years between two successive fund raisings. The time gap and IRR variables are centered on the respective means. Vintage

28

Though they do not report the results, Kaplan and Schoar (2005) also estimate a similar regression specification using the time gap variable and find a positive but statistically not significant coefficient on the interaction term of time gap and previous funds performance.

89

year fixed effects are included and standard errors are clustered at the PE firm level. If the estimate of is negative, this will suggest that performance persistence decrease as a time gap grows conditional on the previous funds performance. Column (2), (5), and (8) in Table 8 report the estimation results. The coefficient estimate on the interaction term is negative, but only for buyout funds. The estimate of the coefficient, , is -0.419 in column (2) and the standard deviation of the log of time gap is .410. Therefore one standard deviation increases in the log of time gap results in the decrease in persistence by 0.171 (or 44%). Put differently, if a time gap increases by one year from its mean (3.6 years), performance persistence will decrease by 0.102 (or 27%). In addition, performance persistence will disappear if the log of time gap increases by a 2.25 standard deviation from its mean. The probability that this would happen under the empirical distribution of the log of time gap is about 0.016. Similarly, the probability that performance persistence would reduce in half is approximately 0.129. In column (8) we interact buyout dummy variable with the interaction of IRR and time gap to see whether the effect of time gap on persistence is different between buyout and venture capital funds. The coefficient on this triple interaction term is negative and statistically significant, suggesting that buyout fund performance persistence may be affected more by common economic factors than venture capital funds. This result is consistent with what Figure 1 alludes a strong co-movement of buyout funds performance compared to venture capital funds. This implies that buyout funds tend to be more affected simultaneously by some common factors compared to venture capital funds. 90

In column (9), we regress current funds IRR on the first previous funds IRR, fund growth, time gap, and all interaction terms. Previous results consistently hold true in this specification. Fund growth and time gap between two successive fund raisings reduce performance persistence. The results are consistent with the hypothesis that decreasing returns to capital flows reduces performance persistence and market commonality increases performance persistence in private equity funds. In addition, there are some subtle difference between buyout funds and venture capital funds. The effect of fund growth on persistence is stronger for venture capital funds and the effect of the time gap on persistence is stronger for buyout funds. This difference suggests that management skills and technology managing venture capital funds are not readily scalable, while they seem more scalable for buyout funds. On the other hand, for buyout funds economic conditions seem more important for their performance. Especially, the effect of credit market condition is more important for buyout funds than for venture capital funds because buyout funds typically lever up their investment with large amount of debt, whereas venture capitals usually don't. We believe the results largely are consistent with the view that venture capital industry is more of labor (management) intensive one while buyout industry is more of capital intensive one.

3.4.4. Common market conditions In the previous section, we find suggestive evidence that common economic conditions underlying the successive funds with overlapping investment period may affect performance persistence. What are the relevant economic factors influencing private 91

equity fund performance and its persistence? We first briefly describe the investment cycles of private equity funds to gain insights about what macro economic conditions may matter to private equity performance. Fund raising usually takes from six months to two years depending on market conditions and various factors such as past performance and a PEs reputation. Only accredited investors can invest in private equity funds, and institutions such as pension funds, university endowments, insurance companies, foundations, and family offices and trusts are major investors in this market. Fund of funds which invest in other private equity funds also play an increasingly important role in the private equity industry. These investors commit a certain amount of capital to a partnership, which means that they do not invest the committed capital up front but over the course of a funds life capital is drawn-down when needed up to a committed amount. As noted above, a typical private equity fund is managed for ten years with two to three year extension conditional on an agreement between GPs and LPs. During first several years GPs focus on investments (acquisition of target firms by buyout funds and provision of financing to portfolio companies by venture capital funds) and later part of a funds life GPs focus on divesting their investments before the expiration of the partnership contract. Figure 3 plots typical cash contribution by investors and cash distribution to investors from a private equity partnership over the funds life. The pattern is usually called J-curve (Graph C) in that lots of drawn-downs and investments are made early in the funds life and cash inflows start kicking in as a fund liquidates (divests) its early investments through trade sales and IPOs. The duration of investment 92

period on a portfolio company varies substantially. But on average buyout funds are known to hold a portfolio company shorter than venture capital funds: on average, 3 years versus 5 years. Buyout funds almost always lever up each investment (buyout) with large amount of debt, whereas venture capital funds usually do not do so. Targets of buyout funds have stable cash flows, long track record, high level of cash, and low leverage, whereas targets of venture capital funds are young without much track record or profits and face substantial risks associated with product development and marketing. There are basically three sources of returns in private equity funds: earnings multiples, earnings, and leverage. 29 Private equity funds can improve returns by increasing (or reducing) multiples at exit (multiples at acquisitions), improving performance (earnings or cash flows), or using an appropriate level of leverage. Therefore, various market conditions at the inception of funds, at the time of investments, and at the time of divestments directly affect funds performance. For example, pricing and liquidity of capital are important determinants of private equity performance, especially for buyout funds. Exit market condition such as IPO market and corporate control market directly affect a funds performance. Relatedly, the market-wide price (multiples) at which business are purchased and sold as well as public stock market multiples are also important.

29

The description in this paragraph is based on Fraser-Sampson (2010)

93

Ultimately, what we want to show is to what extent the correlation of economic conditions explains the correlation of fund performance. To test this, we first construct a market similarity measure (MSM): MSM i,t = | (Market Condition) i,t+1 / (Market Condition) i,t -1 |, (6)

which is the absolute value of the ratio of market condition during a follow-on funds life to that during a current funds life. A larger MSM value implies that the market conditions of current and the next fund life are more dissimilar. We examine the following non-exhaustive market condition variables: 1) IPO volume from fifth year to tenth year of a funds life, 2) GDP growth during a funds life, 3) S&P 500 stock returns over a funds life, 4) three month Treasury bill yield during first five years of a funds life, and 5) the ratio of the average S&P 500 price earnings ratio during first five years to that from fifth to tenth year of a funds life. After constructing MSM in equation (6), we estimate a similar regression equation as Equations (4) and (5): (IRR)i,t = + (IRR)i,t-1 + (MSM)i,t-1 + (MSM)i,t-1 (IRR)i,t-1 + i,t = + [ + (MSM)i,t-1](IRR)i,t-1 + (MSM)i,t-1 + i,t, (7)

Standard errors are clustered at vintage year level since funds which are raised in the same year tend to have, but not necessarily, a similar level of MSM. The negative coefficient, , on MSM will imply that as the market conditions under which two funds are managed become dissimilar, the correlation (persistence) between current fund and previous fund performance becomes smaller. Table 9 reports the estimation results. 94

All the coefficients on MSM are negative except that on S&P500 price earnings ratio. In column (1), for example, the negative coefficient on MSM suggests that as the dissimilarity of IPO market condition of current fund and follow-on fund gets larger, performance persistence becomes weaker. The MSM variable is normalized with mean zero and standard deviation one. Therefore, the coefficient of the interaction term between IRR and MSM, -0.219, suggests that one standard deviation increase in the dissimilarity measure, MSM, leads to a -0.219 decrease in performance persistence. The overall evidence suggests that as the common economic conditions under which the successive funds are managed become more similar, there is more performance persistence in private equity funds.

3.5. Conclusion This study examines performance persistence in private equity funds and uncovers several novel findings. First, we show that even though performance seems to persist, it is at best short-living. In the long-run performance tends to converge across funds. Second, capital inflows into a fund after controlling for past performance reduce performance persistence. Third, similar market conditions facing successive funds increase performance persistence. Contrary to Kaplan and Schoar (2005), the evidence is skeptical about whether fund managers in private equity have differential and proprietary skills. However, the shortliving persistence does not necessarily imply that fund managers do not have differential skills if one believes the implication of the Berk and Green (2004) model. 95

Notwithstanding, the evidence in this paper has important implications for investors in the private equity industry. Investors need active monitoring and rebalancing of their portfolios. The second or prior fund performance can be a misleading indicator of future performance. To further see how useful the performance information of current fund for future funds performance is, we examine the average performance of the second and third follow-on funds between top and bottom quartile portfolio ranked based on current fund performance. 30 In an unreported table, we show that the difference between these two average returns is not statistically significant. This result suggests that holding onto a current winner is not likely to generate superior returns to investors over the long run. Using the average of the second and third previous funds performance to rank portfolios does not improve future performance, either. The difference in the performance of top and bottom quartile portfolio based on two previous funds performance is not statistically significant. We also find that the effect of capital flows on persistence is significant only for venture capital funds. And the effect of common market condition on persistence is more important for buyout funds. This finding implies different nature of the two types of funds in that venture capital is more labor-intensive while buyout is more capitalintensive.

Note that we do not require a fund survive through it third follow-on fund because investors do not know whether the fund would be able to raise several follow-on funds.

30

96

References

Chapter 1 References Aslan, H. and Kumar, P., 2009, Lemons or Cherries? Growth Opportunities and Market Temptations in Going Public and Private, Journal of Financial and Quantitative Analysis, forthcoming.

Asquith, P. and Wizman T.A., 1990, Event risk, covenants, and bondholder returns in leveraged buyouts, Journal of Financial Economics, 27, 195-213.

Baker, G. P., 1992, Beatrice: A study in the creation and destruction of value, Journal of Finance 47, 1081-1119.

Baker, G. P. and Wruck K. H., 1989, Organizational changes and value creation in leveraged buyouts: The case of the O.M. Scott & Sons Company, Journal of Financial Economics 25, 163-190.

Bargeron, L., Schlingemann, F., Stulz, R., and Zutter, C., 2008, Why Do Private Acquirer Pay so Little Compared to Public Acquirers?, Journal of Financial Economics, 89, 375390. 97

Boucly Q., Sraer D., and Thesmar D., 2009, Growth LBOs, HEC Paris working paper.

Brau, J.C. and Francis, B. and Kohers, N., 2003, The Choice of IPO versus Takeover: Empirical Evidence Journal of Business 76, 583-612.

Brau, J.C. and Fawcett, S.E., 2006, Initial Public Offerings: An Analysis of Theory and Practice, Journal of Finance 61, 399-436.

Brav, O., 2009, Access to Capital, Capital Structure, and the Funding of the Firm, Journal of Finance 64, 263-308.

Chevalier, J., 1995, Do LBO Supermarkets Charge More? An Empirical Analysis of the Effects of LBOs on Supermarket Pricing, Journal of Finance, 50, 1095-1112.

Dittmar, A. and Bharath, S., 2009, Why do firms use private equity to opt out of public markets, Review of Financial Studies forthcoming.

Fama, E. and Jensen, M., 1985, Organizational forms and investment decisions, Journal of Financial Economics 14, 101-120.

98

Jensen M., 1986, Agency Costs of Free Cash Flow, Corporate Finance and Takeovers, American Economic Review 76, 323-329.

Jensen M., 1989, Eclipse of the public corporation, Harvard Business Review, Sep-Oct, 61-74.

Jensen M., 1993, The Modern Industrial Revolution, Exit, and the Failure of Internal Control Systems, Journal of Finance 48, 831-880.

Kaplan S., 1989, The Effects of Management Buyouts on Operating Performance and Value, Journal of Financial Economics 24, 217-254.

Kaplan, S.N. and Strmberg P., 2003, Financial contracting theory meets the real world: an empirical analysis of venture capital contract, Review of Economic Studies 70, 281315.

Koenker, R. and Hallock K., 2001, Quantile Regression, Journal of Economic Perspectives, 15, 143-156.

99

Kim, W. and Weisbach, M.S., 2008, Motivations for public equity offers: An international perspective, Journal of Financial Economics 87, 281-307.

Lehn, K. and Poulsen, A., 1989, Free cash flow and stockholder gains in going private transactions, Journal of Finance 44, 771-787.

Levine J.S., 2006, Structuring Venture Capital, Private Equity, and Entrepreneurial Transactions, Aspen Publishers.

Li, K. and Prabhala, N., 2007, Self-Selection Models in Corporate Finance, Chapter 2 in: Handbook of Corporate Finance: Empirical Corporate Finance, Handbooks in Finance Series, Elsevier/North Holland.

Lichtenberg F. and Siegel D., 1990, The effects of leveraged buyouts on productivity and related aspects of firm behavior, Journal of Financial Economics 27, 165-194.

Long, W.F. and Ravenscraft, D.J., 1993, LBOs, debt and R\&D intensity, Strategic Management Journal 14, 119-135.

100

Muscarella, C.J. and Vetsuypens, M.R., 1990, Efficiency and organizational structure: a study of reverse LBOs, Journal of Finance 45, 1389-1413.

Opler, T. and Titman, S., 1993, The determinants of leveraged buyout activity: Free cash flow vs. financial distress costs, Journal of Finance 48, 1985-1999.

Sahlman, W.A., 1990, The structure and governance of venture-capital organizations, Journal of Financial Economics 27, 473521.

Sheen, A., 2009, Do Public and Private Firms Behave Differently? An Examination of Investment in the Chemical Industry, UCLA working paper.

Shleifer A. and Summers L. H., 1988, Breach of Trust in Hostile Takeovers, In Alan J. Auerbach, editor, Corporate Takeovers: Causes and Consequences. Chicago: University of Chicago Press.

Smith A., 1990, Corporate Ownership Structure and Performance: The Case of Management Buyouts, Journal of Financial Economics 27, 143-64.

Speechley T., 2008, Acquisition Finance, Tottel Publishing 101

Strmberg P. 2007, The New Demography of Private Equity, working paper, SIFR, in The Global Economic Impact of Private Equity Report 2008, World Economic Forum, January 2008.

Stein J. 2003, Agency, Information and Corporate Investment , in Handbook of the Economics of Finance, edited by George Constantinides, Milt Harris and Rene Stulz. Amsterdam: North Holland, 2003.

Weir C., Jones P., and Wright M., 2007, Public to Private Transactions, Private Equity and Performance in the UK: An Empirical Analysis of the Impact of Going Private, working paper.

Wiersema M. F. and Liebeckind J. P.,1995, The Effects of Leveraged Buyouts on Corporate Growth and Diversification in Large Firms, Strategic Management Journal 16, 447-460.

Zingales, L., 1995, Insider ownership and the decision to go public, Review of Economic Studies 62, 425-448.

102

Chapter 2 References Barclay, M. J., N. D. Pearson, and M. S. Weisbach, 1998, Open-End Mutual Funds and Capital Gains Taxes, Journal of Financial Economics 49, 3-43.

Chevalier, J., Ellison, G., 1997, Risk-taking by mutual funds as a response to incentives, Journal of Political Economy 105, 1167-1200.

Chevalier, J., Ellison, G., 1999, Career Concerns of Mutual Fund Managers, Quarterly Journal of Economics 114, 389-432.

DeGroot, M. H., 1970, Optimal Statistical Decisions, New York, NY, MacMillan. Fama, E. F., 1980, Agency Problems and the Theory of the Firm, Journal of Political Economy 88, 288-307.

Gibbons, R., and Murphy K. J., 1992, Optimal Incentive Contracts in the Presence of Career Concerns: Theory and Evidence, Journal of Political Economy 100, 468-505.

Gompers, P. and J. Lerner., 1999, An Analysis of Compensation in the U.S. Venture Capital Partnership, Journal of Financial Economics 51, 3-44.

103

Guo, S., E. S. Hotchkiss, and W. Song., 2010, Do Buyouts (Still) Create Value? Journal of Finance, forthcoming.

Hermalin, B. E., and M. S. Weisbach., 1998, Endogenously Chosen Boards of Directors and Their Monitoring of Management, American Economic Review 88, 96-118.

Hermalin, B. E., and M. S. Weisbach., 2009, Information Disclosure and Corporate Governance, Working Paper.

Holmstrom, B., 1999, Managerial Incentive Problems: A Dynamic Perspective, Review of Economic Studies 66, 169-182.

Hong, H., Kubik, J., 2003, Analyzing the Analysts: Career Concerns and Biased Earnings Forecasts, Journal of Finance 58, 313-351.

Hong, H., Kubik, J., Solomon, A. 2000, Security Analysts Career Concerns and Herding of Earnings Forecasts, RAND Journal of Economics 31, 121-144.

Ippolito, R., 1992, Consumer reaction to measures of poor quality: Evidence from the mutual fund industry, Journal of Law and Economics 35, 45-70. 104

Jensen, M. C., 1989, Eclipse of the Public Corporation, Harvard Business Review 67, 6174.

Jensen, M. C., and K. J. Murphy, 1990, Performance Pay and Top-Management Incentives, Journal of Political Economy 98, 225-64.

Kaplan, S. N., 1989, The effects of management buyouts on operations and value, Journal of Financial Economics 24, 217-54.

Kaplan, S. N., and A. Schoar, 2005, Private equity performance: Returns, persistence, and capital flows, Journal of Finance 60, 1791-823.

Kaplan S. N., and P. Strmberg, 2009, Leveraged Buyouts and Private Equity, Journal of Economic Perspectives 23, 121-46.

Metrick, A., 2007, Venture Capital and the Finance of Innovation, Wiley.

Metrick, A., and A. Yasuda, 2010, The Economics of Private Equity Funds, Review of Financial Studies, forthcoming.

105

Sensoy, B., 2009, Performance Evaluation and Self-Designated Benchmark Indexes in the Mutual Fund Industry, Journal of Financial Economics 92, 25-39.

Sirri, E., Tufano, P., 1998, Costly search and mutual fund flows, Journal of Finance 53, 1589-1622.

Smith, A. J., 1990, Corporate ownership structure and performance: the case of management buyouts, Journal of Financial Economics 27, 143-64.

Chapter 3 References Acharya V., J. Franks, and H. Servaes, 2009, Private Equity: Boom and Bust?, Journal of Applied Corporate Finance 19, 44-53.

Axelson U., T. Jenkinson, P. Stromberg, and M. Weisbach, 2010, Borrow Cheap, Buy High? The Determinants of Leverage and Pricing in Buyouts, NBER working paper.

Baquero G., J. ter Horst, and M. Verbeek, 2005, Survival, Look-Ahead Bias, and Persistence in Hedge Fund Performance, Journal of Financial and Quantitative Analysis 40, 493 -517.

106

Berk, J. and R. Green, 2004, Mutual fund flows and performance in rational markets, Journal of Political Economy, 112, 1269-1295.

Brown S., W. Goetzmann, R. Ibbotson, and S. Ross, 1992, Survivorship Bias in Performance Studies, Review of Financial Studies, 5, 553-580.

Carpenter J. and A. Lynch, 1999, Survivorship bias and attrition effects in measures of performance persistence, Journal of Financial Economics 54, 337-374.

Chen J, H. Hong, M. Huang, and J. Kubik, 2004, Does Fund Size Erode Mutual Fund Performance? The Role of Liquidity and Organization, American Economic Review 94, 1276-1302.

Chung J., B. Sensoy, L. Stern, and M. Weisbach, 2010, Incentives of Private Equity General Partners from Future Fundraising, Ohio State University working paper.

Daniel K., M. Grinblatt, S. Titman, and R. Wermers, 1997, Measuring Mutual Fund Performance with Characteristic-Based Benchmarks, Journal of Finance 52, 1035-1058.

Fraser-Sampson G., 2010, Private Equity as an Asset Class, John Wiley & Sons, Ltd. 107

Fung W., D. Hsieh, N. Niak, and T. Ramadorai, 2008, Hedge Funds: Performance, Risk, and Capital Formation, Journal of Finance 62, 1777-1803.

Glode V. and R. Green, 2008, Information Spill-Overs and Performance Persistence in Private Equity Partnerships, working paper.

Hotchberg Y., A. Ljungqvist, and A. Vissing-Jorgensen, 2009, Informational Hold-up and Performance Persistence in Venture Capital, working paper.

Kaplan S. and A. Schoar, 2005, Private equity performance: Returns, persistence, and capital flows, Journal of Finance 60, 1791-1823.

Lopez de-Silane F., L. Phalippou, and O. Gottschalg, 2010, Giants at the Gate: Diseconomies of Scale in Private Equity, working paper.

Lynch A. and D.Musto, 2003, How Investors Interpret Past Fund Returns, Journal of Finance 58, 2033-2058.

McClelland G. and C. Judd, 1993, Statistical Difficulties of Detecting Interactions and Moderator Effects, Psychological Bulletin 114, 376-390. 108

Phalippou L., 2010, Venture Capital Funds: Performance Persistence and FlowPerformance relation, Journal of Banking and Finance, forthcoming.

Sirri E. and P. Tufano, 1998, Costly Search and Mutual Fund Flows, Journal of Finance, 53, 1589-1622.

ter Horst J., T.Nijman, and M.Verbeek, 2001, Eliminating Look-Ahead Bias in Evaluating Persistence in Mutual Fund Performance, Journal of Empirical Finance 8, 345-373.

109

Appendix A: Tables for Chapter 1

110

Table A.1. Distribution of leveraged buyout transactions in the U.K

Panel A reports the number of leveraged buyout transactions and deal values by transaction year and by target status. Private indicates independent private targets, Public public targets, Divisional divisions of corporations, SBO a portfolio company owned by a private equity firm, and Distressed firms in bankruptcy (in receivership). % Sponsor is the proportion of deals which are sponsored by private equity funds. Total deal value is the sum of all deal values. Median deal value is the median deal value per deal. Median deal multiple EBITDA is the median of deal value to EBITDA ratio. % Available deal value is the proportion of deals where deal values are available in a given target status. Panel B reports the number of leveraged buyout transactions and deal values by transaction type and target status. Transaction type is classified based on who is leading the transaction. MBO is management buy-out, IBO is institutional buy-out, MBI is management buy-in, and BIMBO is buy-in management buy-out. Panel A: Distribution of leveraged buyout transaction by target status Target status LBO year Total Private Public Divisional SBO Distressed 1997 70 3 136 9 7 225 1998 62 20 144 11 6 243 1999 110 32 207 35 6 390 2000 85 29 196 32 7 349 2001 112 27 213 17 23 392 2002 123 7 172 17 36 355 2003 190 27 193 24 26 460 2004 233 14 148 47 24 466 2005 202 15 143 38 26 424 2006 237 22 122 61 14 456 2007 248 21 145 78 15 507 2008 208 12 105 41 19 385 Total 1,880 229 1,924 410 209 4,652 % 40.41% 4.92% 41.36% 8.81% 4.49% % Sponsor 48.14% 74.24% 48.28% 100.00% 22.97% Total deal value ( million) 26,364.98 69,845.06 85,773.87 52,153.82 787.73 234,925.46 Median deal value ( million) 10.00 74.64 10.00 62.63 4.50 15.00 Median deal multiple on EBITDA 7.87 7.92 7.76 10.61 6.11 8.51 % Available deal value 37.77% 100.00% 57.59% 82.93% 20.10% 52.21% (Continuing)

111

111

Table A.1 - (Continued)

Panel B: Distribution of leveraged buyout transaction by transaction type and target status Transaction type Target status MBO IBO MBI Private 1278 254 275 67.98% 13.51% 14.63% Public 82 145 2 35.81% 63.32% 0.87% Divisional 1380 394 132 71.73% 20.48% 6.86% SBO 186 208 12 45.37% 50.73% 2.93% Distressed 168 24 17 80.38% 11.48% 8.13% Total 3094 1025 438 66.51% 22.03% 9.42% % Sponsor 38.43% 100.00% 46.80% Total deal value ( million) 40,324.60 190,624.24 3,357.91 Median deal value ( million) 7.50 62.00 6.57 Median deal multiple on EBITDA 11.63 11.78 10.43 % Available deal value 46.61% 75.51% 40.64%

BIMBO 73 3.88% 0 0.00% 18 0.94% 4 0.98% 0 0.00% 95 2.04% 57.89% 618.72 12.50 8.87 36.84%

Total 1880 229 1924 410 209 4652 53% 234,925.46 15.00 8.64 52.21%

112

112

Table A.2. Distribution of final sample of leveraged buyouts

This table presents the distribution of leveraged buyout transactions by transaction completion year and by target status for the final sample. Panel A. Panel B reports the distribution of leveraged buyout transaction by industry. % Sponsor is the proportion of deals which were sponsored by private equity funds. Median deal value is the median deal value. Deal values are the total consideration paid for the actual stake acquired and are in millions. Panel A: Yearly distribution of leveraged buyouts Target status LBO year Private 1997 44 1998 39 1999 69 2000 53 2001 70 2002 77 2003 118 2004 145 2005 126 2006 148 Total 887 % Sponsor 48.14% Median deal value ( million) 10.00

Public 2 12 20 18 17 4 17 9 9 14 122 74.24% 74.64

Total 45 51 88 71 87 81 135 154 135 161 1,009

Panel B: Industry distribution of leveraged buyouts Industry (by SIC classification) Agriculture, Forestry, And Fishing Construction Finance, Insurance, And Real Estate Manufacturing Retail Trade Services Transportation, Communications, Electric, Gas, And Sanitary Services Wholesale Trade Total Private 2 0.25% 98 11.03% 33 3.76% 331 37.34% 80 9.02% 213 24.06% 51 5.76% 78 8.77% 887 77.03% Public 0 0% 7 5.88% 8 6.72% 39 31.93% 17 14.29% 31 25.21% 10 8.40% 9 7.56% 122 22.97% Total 2 0.25% 105 16.91% 42 10.48% 370 69.27% 97 23.31% 244 49.27% 61 14.16% 87 0.1633 0 100%

113

Table A.3. Financial characteristics of target firms prior to leveraged buyouts

This table reports the summary statistics of financial characteristics for private target firms and control firms prior to buyout. Financial information is at the end of the fiscal year preceding the year of the leveraged buyout. Targets of PE led LBO is the sample of the target firms of interest, i.e. private firms which engaged in leveraged buyouts with private equity sponsors. Industry Median is the median value of the corresponding accounting variable. Targets of Non-PE led LBO is private targets which underwent leveraged buyouts without private equity sponsors. Lastly, Matched Firms is the control firms matched with the target firms using propensity score matching. The Wilcoxon signed rank sum test is performed for the difference between targets and control firms. The significance level is reported with asterisks: *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.

Targets of PE led LBO Book value of total assets ( million) Sales ( million) PPE ( million) Employees Firm age EBITDA / Sales Cash / Sales Sales growth Total debt / Assets 5.21 10.42 1.02 94 17 0.12 0.14 0.13 0.47

Industry Median 12.14 18.84 2.23 159 15 0.04 0.05 0.06 0.57
*** ** ** ** * *** *** *** **

Targets of Non-PE led LBO 3.56 9.42 0.70 92 17 0.05 0.09 0.02 0.44
*** * *** ***

Matched Firms 6.12 11.23 2.11 124 13 0.11 0.13 0.12 0.48
**

114

114

Table A.4. Logistic regression to predict the likelihood of being a leveraged buyout target
This table reports the estimate of a logistic regression to predict the likelihood of being a target of private equity, based on firm characteristics. The dependent variable is a binary variable equal to 1 for target firms and 0 for non-target private firms. *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively. Coefficient Variables log(Total assets) Sales growth EBITDA / Sales Cash / Sales Total debt / Sales Intercept Year fixed effects Industry fixed effects Obs. of target firms Obs. of non-target firms Adjusted R2 Estimates -1.95** (-2.12) 3.92** (2.32) 8.23** (2.41) 0.62 (0.69) -0.83* (-1.82) -4.32*** (-2.87) Yes Yes 412 3,000 0.36

115

Table A.5. Ownership characteristics of privately held targets

The table reports the ownership structure of private targets for the subsample of 721 private firms before and after the year of leveraged buyout transactions. Ownership information (the number of shareholders, the ownership of the largest owners, and owner age) is collected from Form 363a (annual return). All companies in the U.K. must submit this annual return form to Companies House each year: It provides a snapshot of general information about the company, including the details of key personnel, the registered office, share capital and shareholdings. Variable # of shareholders Before buyouts After buyouts Before buyouts After buyouts Departing Incoming 1st Quartile 1 4 72.12% 0.00% 51 41 Mean 3.60 7.20 87.20% 7.47% 56.92 46.17 Median 2 7 100.00% 0.00% 56 45 3rd Quartile 4 9 100.00% 6.87% 63 50

Ownership of the largest owner(s) Owner-managers age

116

116

Table A.6. Changes in firm growth after leveraged buyouts of public targets
The median (control-adjusted) percentage changes of total assets, sales, property, plant, and equipment (PPE), the number of employees, and capital expenditures are presented from year -2 to year +3 relative to the year of the leveraged buyout completion. Numbers in brackets are accounting value at one year prior to leveraged buyouts. *, **, and *** indicate that the median change is significantly different from zero at the 10%, 5%, and 1% levels, respectively, as measured by the two-tailed Wilcoxon signed rank sum statistics. Year relative to buyout Year relative to buyout Total assets ( million) Percentage change Control adjusted change PPE ( million) Percentage change Control adjusted change Sales ( million) Percentage change Control adjusted change Employees Percentage change Control adjusted change CAPEX ( million) Percentage change Control adjusted change CAPEX / Sales Percentage change Control adjusted change -2 to -1 [87.50] 0.01 -0.02 [29.34] -0.01 0.00 [85.52] 0.02 -0.04 [754] -0.03 -0.06 [1.41] -0.21 -0.18 [0.03] -0.28 -0.25
** * * ** * ** *

-1 to 1 0.05 -0.10 -0.17 -0.20 0.02 -0.18 -0.08 -0.14 -0.34 -0.42 -0.16 -0.07
** * **

-1 to 2 -0.01 -0.12 -0.18 -0.25 0.01 -0.22 -0.18 -0.23 -0.22 -0.33 -0.06 -0.03
*** **

-1 to 3 0.07 -0.25 -0.24 -0.31 -0.05 -0.32 -0.22 -0.35 -0.20 -0.12 -0.04 -0.05
*** ***

*** ***

*** ***

*** ***

*** ***

*** ***

*** ***

** ***

** ***

**

117

Table A.7. Changes in firm growth after leveraged buyouts of private targets with private equity
The median percentage changes of total assets, sales, property, plant, and equipment (PPE), the number of employees, and capital expenditures are presented from year -2 to year +3 relative to the year of the leveraged buyout completion. Numbers in brackets are accounting value at one year prior to leveraged buyouts. All measures of firm growth are benchmarked against three control samples: Industry median (Industry adjusted), private LBO targets without private equity sponsors (Control adjusted (1)), and matched sample using propensity score matching (Control adjusted (2)). *, **, and *** indicate that the median change is significantly different from zero at the 10%, 5%, and 1% levels, respectively, as measured by the two-tailed Wilcoxon signed rank sum statistics.
Year relative to buyout Year relative to buyout Total assets ( million) Percentage change Industry adjusted change Control adjusted change (1) Control adjusted change (2) PPE ( million) Percentage change Industry adjusted change Control adjusted change (1) Control adjusted change (2) Sales ( million) Percentage change Industry adjusted change Control adjusted change (1) Control adjusted change (2) Employees Percentage change Industry adjusted change Control adjusted change (1) Control adjusted change (2) CAPEX ( million) Percentage change Industry adjusted change Control adjusted change (1) Control adjusted change (2) CAPEX / Sales Percentage change Industry adjusted change Control adjusted change (1) Control adjusted change (2) -2 to -1 [5.21] 0.16 0.15 0.11 0.06 [1.02] 0.00 0.06 0.00 0.00 [10.42] 0.17 0.14 0.15 0.01 [94] 0.09 0.09 0.07 0.02 [1.40] 0.01 0.04 -0.02 0.00 [0.01] -0.11 -0.06 -0.62 -0.10
*** * * *** *** *** * *** *** *** *** *** *** ***

-1 to 1 0.81 0.78 0.58 0.52 0.23 0.32 0.30 0.21 0.32 0.25 0.18 0.12 0.21 0.27 0.17 0.18 0.68 0.61 0.58 0.62 0.40 0.31 0.03 0.01
*** *** *** ***

-1 to 2 0.91 0.94 0.66 0.53 0.46 0.44 0.62 0.43 0.50 0.42 0.30 0.24 0.33 0.44 0.31 0.29 0.76 0.72 0.23 0.68 0.18 0.12 -0.02 -0.03
*** *** *** ***

-1 to 3 0.94 1.15 0.68 0.53 0.26 0.50 0.27 0.23 0.71 0.45 0.42 0.34 0.41 0.42 0.43 0.37 0.76 0.81 0.42 0.69 0.18 0.12 -0.04 -0.05
*** *** *** ***

*** *** *** **

*** *** *** ***

*** *** *** **

*** *** *** **

*** *** *** **

*** *** *** **

*** *** *** **

*** *** *** ***

*** *** *** ***

*** *** *** ***

** ** ** **

** ** *** **

*** **

** *

* *

118

Table A.8. Acquisitions and disposals of businesses and operations after leveraged buyouts
This table presents the total cash outflows and inflows associated with acquisitions and disposals of operations and divisions during leveraged buyout ownership. Specifically, I compute the sum of all cash outflows (inflows) associated with acquisitions (disposals) from the time of a leveraged buyout to the exit (when the firm exited leveraged buyout ownership) or otherwise to the present, and I divide this sum by tangible fixed assets at the end of fiscal year-end prior to buyout. I exclude cash outflows associated with the leveraged buyouts. I also report the cash flows of control sample firms. I match the sample firms with control firms with the same two-digit SIC code whose sales most closely matched the sample firm's sales at the end of the fiscal year preceding the year of the going-private transaction. *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively, for the Wilcoxon signed rank sum test. Lower Quartile Private Acquisitions Control sample Disposals Control sample Public Acquisitions Control sample Disposals Control sample Median (Z-stat) Difference (public - private) Acquisitions 3.543
***

Median 0.660 0.000 0.000 0.000 0.043 0.002 0.017 0.000

Upper Quartile 2.170 0.234 0.000 0.000 0.302 0.012 0.446 0.098

0.309 0.000 0.000 0.000 0.000 0.000 0.000 0.000

Disposals -2.666
***

119

Table A.9. Pre-buyout investment constraints and post-buyout growth

This table presents the estimates of median regression of firm growth rates on pre-buyout investment constraints among private target firms of private equity led leveraged buyouts. The regression estimates the change in the median of firm growth rates produced by one unit of change in the predictor variable. The dependent variable is the change in firm growth measures from year -1 to year +3. SG is sales growth from year -2 to year -1 (proxy for growth opportunities), FirmSize is the book value of assets at year -1, FirmAge is the number of years from incorporation to the year of leveraged buyouts, Leverage is the total debt to total assets ratio at year -1, Concentration is the Herfindahl index of ownership concentration at year -1, OwnerAge is owners age, and OwnerManager is a dummy variable taking one if a owner is also the manager and zero otherwise at year -1. Industry growth is the median growth rate of each growth measure of industry-peer private firms. *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively. Numbers in brackets are t-values. Assets growth FirmSize*SG FirmAge*SG Leverage*SG Concentration*SG OwnerAge*SG OwnerManager*SG SG Industry growth Year fixed effect Number of observations Pseudo R2 -0.19* [1.72] -0.03* [1.92] 0.01 [1.32] 0.02* [1.82] -0.01 [0.48] 0.02 [1.40] 0.18** [1.93] 0.01 [0.96] Yes 368 0.09 Sales growth -0.12* [1.69] -0.02 [1.52] 0.01 [1.23] 0.01 [1.45] 0.02 [0.68] 0.03 [0.98] 0.21*** [2.23] 0.03* [1.73] Yes 322 0.11 Employment growth -0.03 [1.21] -0.04* [1.90] -0.01 [0.32] 0.02* [1.68] -0.02 [0.49] 0.02 [1.12] 0.08* [1.75] 0.01** [2.21] Yes 368 0.12 CAPEX growth -0.08 [1.34] -0.03 [1.24] 0.00 [0.23] 0.01 [0.79] 0.01 [0.36] 0.01 [0.89] 0.10** [2.01] 0.02** [2.17] Yes 368 0.08

120

Table A.10. Operating performance after a buyout: Private equity sponsored targets
This table presents the median changes in operating performance surrounding the year of leveraged buyouts for private targets with private equity sponsors. Operating performance is measured in several ways. 1) Percentage changes in EBITDA, 2) Percentage changes in the EBITDA to operating assets (the average of current assets and tangible fixed assets) ratio, 3) Percentage changes in the EBITDA to sales ratio, and 4) Percentage changes in the EBITDA to the number of employees ratio. Control adjustment is made by the median value of the changes in each variable among control firms matched by propensity score matching. That is, (Control-adjusted change in X)=Median value of [(Change in X from year -1 to year Y of a target firm)-(Change in X from year -1 to year Y of a control firm)], where X is the variable of interest and Y is 2, 1, 2, or 3 relative to year -1 (one year before buyouts). *, **, and *** denote statistical significance at the 10%, 5%, and 1% levels, respectively, for the Wilcoxon signed rank sum test. Numbers in brackets are profitability measures at year -1. Year relative to buyout Year relative to buyout EBITDA ( million) Percentage change Control-adjusted change EBITDA / Operating assets Percentage change Control-adjusted change EBITDA / Sales Percentage change Control-adjusted change EBITDA / Employees ('000) Percentage change Control-adjusted change 0.10 0.07
** **

-2 to -1 0.27 0.22 0.03 0.07 0.06 0.02

*** ***

-1 to 1 [1.31] 0.21 0.17 [0.27]

** * ** **

-1 to 2 0.23 0.16 -0.16 -0.31 -0.12

* *** *

-1 to 3 0.20 0.12 -0.28 -0.36 -0.26

* ** ** *

-0.19 -0.21 [0.12] 0.02 0.00 [11.32] -0.12 -0.30

**

**

-0.15 -0.22 -0.45

-0.35 -0.21 -0.36

**

* **

121

Table A.11. Examples of deal rationales For a subsample of 113 leveraged buyout transactions, I collect deal rationale information though Zephyr and Lexis-Nexis news search. Deal rationales are comments made by managers of targets, acquirers, or equity sponsors on the completion of transactions; they contain information on the motive behind and goals for leveraged buyouts. Below are examples of deal rationales. 1. Geographical expansion Fontygary Parks Ltd, (11/21/2003), a caravan park operator: We would look to expand into other places or new builds even in Spain or France. Quadrate Ltd, (3/12/2006), an ERP systems software developer: The MBO will allow the company to expand into the U.S. 2. Expansion through new investments CFM City Financial Mailing (8/16/1999), a mailing annual financial reports and shareholder circulars: "We will be using our specialist knowledge of the direct mail market to expand the business of CFM and to capitalise on the growth opportunities in the direct mail sector." West Cornwall Pasty Co. Ltd. (10/11/2007): "Partnering with Gresham will provide us with a strong platform from which to grow the business in the future. The new funding will enable us to implement our roll-out plans and we are very excited by the opportunities that lie ahead of us." 3. Acquisitions Maybin Support Services Ltd, (3/5/2006), a cleaning and security service firm: "It is our intention to build upon this first-rate business platform throughout Ireland and, when appropriate, to expand our services farther afield through both organic growth and through further acquisitions," said Mr. Terry Brannigan. Metal & Waste Recycling Ltd, (2/14/2006), a metal recycling service: The acquisition by Barclays will allow the company to expand by making further acquisitions.

122

122
Figure A.1. Typical corporate structure after a buyout
This figure depicts the change in corporate structure after a leveraged buyout. Typically one or more acquisition vehicles (TopCo and NewCo in this figure) are created for the purpose of the transaction. After NewCo is incorporated by management and/or private equity firms, it acquires Target and its subsidiaries. Subsequently, TopCo is established and acquires NewCo. Usually TopCo issues equity capital (private equity) and NewCo finances the transaction with debt.

123

Appendix B: Tables for Chapter 2

124

Table B.1. Descriptive Statistics

The table reports the number of funds per partnership by fund type in Panel A. These statistics are computed for funds whose fund size information is available, preceding fund size is greater than or equal to $5M in 1990 dollars, and preceding fund performance (IRR) information is available. If a fund is raised in 2005 or before and does not have a follow-on fund, then we assume that the fund failed to raise a follow-on fund and, therefore, the followon fund size is set to zero. Panel B reports descriptive statistics for fund size (in $M), growth, performance (IRR), and the time between successive funds.

125 125

Table B.2. Committed Capital by Type of Fund and Fund Sequence

Fund size data is provided by Preqin. Preceding funds smaller than $5 M in 1990 dollars, or raised after 2005, are excluded from the analysis. If a fund is raised in 2005 or before and does not have a follow-on fund, then we assume that the fund failed to raise a follow-on fund and, therefore, the followon fund size is set to zero. Panel A reports the descriptive statistics for fund size of all preceding funds. Panels B reports descriptive statistics for fund size of all follow-on funds, and Panel C reports the same statistics conditional on raising a fund.

(Continuing)

126 126

Table B.2. (Continued)

127 127

Table B.3. Fund Growth

Fund size data is provided by Preqin. Preceding funds smaller than $5 M in 1990 dollars, or raised after 2005, are excluded from the analysis. If a fund is raised in 2005 or before and does not have a follow-on fund, then we assume that the fund failed to raise a follow-on fund and, therefore, the followon fund size is set to zero. Panel A reports descriptive statistics for fund growth, defined as (It/It-1)-1, where It-1 is preceding fund size and It is followon fund size. Panel B reports the same statistics conditional on raising a fund.

128 128

Table B.4. Fund Performance and Time between Successive Funds

Fund performance data is provided by Preqin. Preceding funds smaller than $5 M in 1990 dollars, or raised after 2005, are excluded from the analysis. If a fund is raised in 2005 or before and does not have a follow-on fund, then we assume that the fund failed to raise a follow-on fund and, therefore, the follow-on fund size is set to zero. Panel A reports the descriptive statistics for fund performance of all preceding funds, as measured by their IRR. Panel B reports descriptive statistics, by sequence number, for the time elapsed before raising a follow-on fund.

129 129

The table presents Tobit regression estimates of the following specifications: (Fund Growth)t = t + t (IRR)t-1 + t. For Panels A and C, the dependent variable is fund growth, defined as (It /It-1)-1, and ln(It/It-1 +1) for Panels B and D, where It-1 is preceding fund size and It is follow-on fund size. Model (1) uses raw fund IRR (Preceding fund IRR), (2) includes vintage year fixed effects, and (3) uses raw IRR minus the benchmark IRR. The benchmark IRR is the median IRR of other funds with same vintage year, fund type, and geographical focus. In All Funds regressions, fund type fixed effects are included. Also, IRRs are demeaned by fund type. In all regressions, we exclude preceding funds smaller than $5 M in 1990 dollars, or raised after 2005. If a fund is raised in 2005 or before and does not have a follow-on fund, then we assume that the fund failed to raise a follow-on fund and, therefore, the follow-on fund size is set to zero, i.e., fund growth is -1. In Panels C and D, we require there be at least a three year gap between two consecutive fund raisings. Standard errors are clustered at the PE firm level. *, **, and *** indicate statistical significance at the 10, 5, and 1% level, respectively.

Table B.5. Future Fundraising and Current Performance

(Continuing)

130 130

Table B.5. (Continued)

131 131

The table presents Tobit regression estimates of the following specifications: (Fund Growth)t = t + t (IRR)t-1 + t (Fund Sequence)t + t (Fund Sequence)t-1 (IRR)t-1 + t. The dependent variable is fund growth, defined as (It/It-1)-1 for Panel A and ln(It/It-1 +1) for Panel B, where It-1 is preceding fund size and It is follow-on fund size. Model (1) uses raw fund IRR (Preceding fund IRR), (2) includes vintage year fixed effects, and (3) uses raw IRR minus the benchmark IRR. The benchmark IRR is the median IRR of other funds with same vintage year, fund type, and geographical focus. In All Funds regressions, fund type fixed effects are included. Also, IRRs are demeaned by fund type. In all regressions, we exclude preceding funds smaller than $5 M in 1990 dollars, or raised after 2005. If a fund is raised in 2005 or before and does not have a follow-on fund, then we assume that the fund failed to raise a follow-on fund and, therefore, the follow-on fund size is set to zero, i.e., fund growth is -1. Standard errors are clustered at the PE firm level. *, **, and *** indicate statistical significance at the 10, 5, and 1% level, respectively.

Table B.6. Future Fundraising, Current Performance, and Fund Sequence

132 132

The table presents Tobit regression estimates of the following specifications: (Fund Growth)t = t + t (IRR)t-1 + t (Fund Size)t + t (Fund Size)t-1 (IRR)t-1 + t. The dependent variable is fund growth, defined as (It/It-1)-1 for Panel A and ln(It/It-1 +1) for Panel B, where It-1 is preceding fund size and It is follow-on fund size. Model (1) uses raw fund IRR (Preceding fund IRR), (2) includes vintage year fixed effects, and (3) uses raw IRR minus the benchmark IRR. The benchmark IRR is the median IRR of other funds with same vintage year, fund type, and geographical focus. In All Funds regressions, fund type fixed effects are included. Also, IRRs and fund sizes are demeaned by fund type. In all regressions, we exclude preceding funds smaller than $5 M in 1990 dollars, or raised after 2005. If a fund is raised in 2005 or before and does not have a follow-on fund, then we assume that the fund failed to raise a follow-on fund and, therefore, the follow-on fund size is set to zero, i.e., fund growth is -1. Standard errors are clustered at the PE firm level. *, **, and *** indicate statistical significance at the 10, 5, and 1% level, respectively.

Table B.7. Future Fundraising, Current Performance, and Fund Size

133 133

Table B.8. Future Revenue and Current Performance

This table presents the estimates of the impact of the current fund's performance on the expected change in total revenue for the GP. Panel A reports the calculation, by fund type and for all funds, of the indirect effect (from subsequent funds) on total revenue of an extra 1% in IRR. Two model specifications are used. In specification (1), the dependent variable is fund growth, whereas in specification (2), we use the log of fund growth. We also report the sensitivity of GP total revenue to an extra dollar returned in the current fund (TR/D). We use the regression coefficients from (1) in Table 5. For specifications (1) and (2), the indirect TR/IRR are equal to NkIi, and Nk(Ii +Ii+1) respectively, where is the regression coefficient with respect to (IRR)t-1, N is the number of subsequent funds raised by the GP, I i is the size of current fund i (in $M), and k is the revenue share from future funds. For buyout and venture funds, we used the distribution of k as estimated in Metrick and Yasuda (2010). Panel B reports the calculation, by fund type and for all funds, of the direct effect on total revenue (from the current fund) of an extra 1% in IRR. We assume in this calculation that the fund is "in the money" and that the fund has a unique cashflow in, and a unique cashflow out. We use the mean IRR of our sample as the baseline IRR. The direct effect is therefore computed as follows: Iik{[1+IRRi+1%]^(Ti)-1] - [(1+IRRi)^(Ti)-1]}, where k is the revenue share from the current fund to the GP (in this case equal to 25%), I i is the mean size of the current fund (in $M), IRR i is the mean IRR, and T i is the expected lifetime of the current fund. Panel C reports the ratio of the indirect effect over the direct effect for the mean fund size: TR/IRR (Indirect) / TR/IRR (Direct). (Continuing)

134 134

Table B.5. (Continued)

135
(Continuing)

135

Table B.5. (Continued)

136

Table B.9. Future Revenue, Current Performance and Fund Sequence

Panel A reports the calculation, for all funds, of the indirect effect (from subsequent funds) on total revenue of an extra 1% in IRR, controlling for the sequence number of the fund. Two model specifications are used. In specification (1), the dependent variable is fund growth, whereas in specification (2), we use the log of fund growth. We also report the sensitivity of GP total revenue to an extra dollar returned in the current fund (TR/D). We use the regression coefficients from (1) in Table 6. For specifications (1) and (2), the indirect TR/IRR are equal to k{(1 + (i+1)3)Ii + (1 + (i+2)3)Ii+1 + (1 + (i+3)3)Ii+2} and k{(1 + (i+1)3)(Ii + Ii+1) + (1 + (i+2)3)(Ii+1 + Ii+2) + (1 + (i+3)3)(Ii+2 + Ii+3)} respectively, where 1 is the regression coefficient with respect to IRR and 3 the regression coefficient with respect to IRRSequence#. I i is the size of current fund i (in $M), and k is the revenue share from future funds. N is the number of subsequent funds raised by the GP. For buyout and venture funds, we used the distribution of k as estimated in Metrick and Yasuda (2010). Panel B reports the calculation, for all funds, of the direct effect on total revenue for the GP, of an extra 1% in the IRR of the current fund. To see how sensitive this estimate is to the sequence number of the fund, we successively consider funds with sequence number 1, 3 and 5 to be the current fund. We assume that the fund is "in the money" and that it has a unique cashflow in, and a unique cashflow out. We use the mean IRR of our sample as the baseline IRR. The direct effect is therefore computed as follows: Iik{[1+IRRi+1%]^(Ti)-1] - [(1+IRRi)^(Ti)-1]}, where k is the revenue share from the current fund to the GP (in this case equal to 25%), I i is the mean size of the current fund (in $M), IRR i is the mean IRR of the current fund, and T i is the expected lifetime of the current fund. Panel C reports the ratio of the indirect effect over the direct effect for the mean fund size: TR/IRR (Indirect) / TR/IRR (Direct).

137

(Continuing)

137

Table B.9. (Continued)

138
(Continuing)

138

Table B.9 (Continued)

139

Figure B.1. Importance of incentives from future fundraising over the partnership's life
The figure illustrates the future fundraising incentives dynamic for all funds. Values from specification (2) Panel C in Table 9 are used to plot the evolution of the ratio of indirect incentives from future fundraising over direct incentives as a function of the sequence number of the fund (i ). N is the parameter representing the number of future funds to be raised. The fraction of fund value going to the GP is assumed to be 18% (k =18%).

140

Appendix C: Tables for Chapter 3

141

Table C.1. Private equity fund performance and fund raising by vintage year
The table reports the distribution of buyout funds (Panel A) and venture capital funds (Panel B) performance and fund size by vintage year. We use internal rate of returns (IRR) reported by private equity partnerships and collected by Preqin as performance measure. Fund size is in 1990 dollars. Q1 is first quantile and Q3 is third quantile.
Panel A. Buyout Funds Fund Size ($mm in 1990 dollars) IRR # Funds Median Q3 Std. w/Size Mean Median Sum . . 0 . . 35.50 35.50 . 1 66.86 66.86 66.86 19.40 19.40 . 1 59.95 59.95 59.95 25.80 31.10 8.64 4 193.39 78.52 773.57 . 0 . . . 39.20 39.20 . 1 444.25 444.25 444.25 8.40 8.40 . 2 55.11 55.11 110.23 18.40 28.90 30.66 8 213.47 66.04 1,707.74 10.70 20.35 9.77 3 267.66 216.21 802.98 28.90 39.00 22.02 11 160.19 91.51 1,762.06 20.50 30.45 9.82 13 758.73 143.82 9,863.48 13.10 20.00 7.45 12 654.98 263.50 7,859.75 30.00 33.40 15.78 19 353.33 274.29 6,713.23 16.80 27.00 18.59 24 180.77 100.00 4,338.43 23.75 30.25 15.60 10 277.12 127.94 2,771.23 21.20 36.90 26.03 22 247.43 97.20 5,443.50 19.80 27.30 15.39 29 269.39 198.99 7,812.37 19.75 33.00 19.58 54 404.40 234.24 21,837.38 9.90 25.40 20.79 49 311.69 128.64 15,272.86 9.30 22.00 30.93 63 289.97 163.86 18,267.92 7.70 15.00 15.50 86 484.91 188.93 41,702.52 7.90 15.30 18.81 103 520.90 257.39 53,652.49 9.95 19.20 13.61 116 429.97 214.75 49,876.98 16.95 23.50 11.18 146 559.91 189.75 81,747.44 23.85 33.70 17.99 103 332.47 159.99 34,244.72 18.30 30.50 16.63 134 338.38 110.60 45,342.65 16.50 37.80 30.26 94 377.70 147.77 35,503.71 7.30 23.20 24.01 121 372.20 172.98 45,036.44 6.30 14.40 23.74 197 553.43 226.50 109,025.32 -9.90 0.10 21.40 211 731.40 225.02 154,325.19 -12.30 -4.80 19.28 246 688.38 211.24 169,342.22 . . 159 768.61 273.53 122,208.20 . 58 767.13 307.67 44,493.42 . . . 5 346.58 244.51 1,732.92 9.95 22.50 23.41 2105 519.78 194.49 1,094,141.11

Vintage # Funds # Funds w/IRR 1976 1 0 1977 1 1 1979 1 1 1980 4 3 1981 2 0 1982 1 1 1983 3 1 1984 11 7 1985 6 4 1986 14 9 1987 13 8 1988 19 11 1989 20 13 1990 30 18 1991 15 8 1992 27 21 1993 32 19 1994 61 38 1995 58 29 1996 67 33 1997 95 51 1998 108 61 1999 121 56 2000 159 78 2001 112 40 2002 138 46 2003 98 39 2004 130 43 2005 202 83 2006 216 83 2007 251 83 2008 167 0 2009 62 0 2010 5 0 Total 2250 888

Q1 . 35.50 19.40 14.20 39.20 8.40 12.40 6.40 15.70 11.70 9.80 20.40 7.00 19.80 10.30 11.40 11.00 2.70 0.20 0.20 -0.60 1.35 9.20 12.45 11.10 5.70 0.70 -2.20 -20.00 -25.60 . . . -2.20 . . . .

Mean . 35.50 19.40 23.70 39.20 8.40 28.37 13.38 31.28 20.43 14.59 31.14 20.85 25.23 19.51 21.22 23.47 16.55 15.91 8.90 4.53 9.38 16.19 25.09 21.36 20.15 13.68 9.37 -9.06 -14.74 . . . 10.59

142

Table C.1. (Continued)

Panel B. Venture Capital Funds IRR # Funds Fund Size ($mm in 1990 dollars) Median Q3 Std. w/Size Mean Median Sum 8.70 8.70 . 2 277.78 277.78 555.56 . 1 16.78 16.78 16.78 28.25 35.00 9.55 2 17.67 17.67 35.33 . 1 24.29 24.29 24.29 . 0 . . 48.55 57.10 12.09 2 39.09 39.09 78.18 18.50 18.50 . 1 61.21 61.21 61.21 13.95 19.10 18.64 5 59.01 44.41 295.03 11.30 26.10 22.79 7 43.88 36.67 307.13 9.30 14.90 17.51 8 55.63 31.51 445.00 9.90 15.30 13.53 18 66.58 37.00 1,198.44 12.00 13.75 8.78 24 74.55 59.04 1,789.20 13.00 18.70 8.30 24 54.34 39.77 1,304.04 9.50 15.00 8.32 14 62.63 47.34 876.88 14.80 22.00 10.05 34 87.71 48.90 2,981.97 22.50 31.95 15.82 23 77.07 46.40 1,772.52 14.70 27.50 36.24 39 57.08 42.16 2,226.28 18.15 23.45 23.91 32 74.74 50.00 2,391.74 25.10 39.70 75.94 23 100.07 62.23 2,301.50 17.50 29.80 27.99 35 75.89 55.90 2,656.12 31.70 44.80 34.44 58 55.42 42.96 3,214.53 23.80 54.80 34.85 60 69.65 42.77 4,178.94 20.00 75.10 88.66 67 71.85 43.74 4,814.22 14.60 63.70 50.58 102 84.32 54.53 8,600.23 24.10 75.90 64.75 173 72.01 50.65 12,458.25 5.75 15.10 150.42 200 93.55 52.89 18,710.00 -5.30 5.90 13.47 267 137.89 78.45 36,815.56 -2.20 6.30 13.08 412 156.67 75.90 64,546.39 -0.70 6.60 17.73 321 119.36 55.35 38,315.84 3.70 11.35 16.17 268 67.02 36.33 17,961.90 2.00 10.20 16.85 187 69.11 35.52 12,923.94 -4.00 4.80 17.48 243 82.09 42.52 19,948.36 -3.00 5.20 19.85 307 107.77 66.92 33,085.08 -9.45 0.80 13.92 331 123.37 71.31 40,834.48 -11.15 -3.55 26.69 340 133.08 74.33 45,246.18 . . . 236 139.72 76.49 32,974.86 . . . 128 131.38 60.92 16,816.77 . . . 13 305.73 240.90 3,974.52 4.30 16.00 50.10 4008 108.97 58.20 436,735.73

Vintage # Funds # Funds w/IRR 1969 2 1 1971 1 0 1972 2 2 1975 1 0 1977 1 0 1978 2 2 1979 1 1 1980 7 6 1981 12 7 1982 15 11 1983 20 13 1984 30 20 1985 47 21 1986 28 19 1987 46 22 1988 37 24 1989 59 33 1990 42 20 1991 29 19 1992 48 28 1993 66 33 1994 77 31 1995 90 35 1996 116 35 1997 193 59 1998 232 66 1999 308 87 2000 471 121 2001 355 85 2002 303 64 2003 221 42 2004 276 43 2005 332 69 2006 354 74 2007 368 64 2008 249 0 . 2009 134 0 . 2010 13 0 . Total 4588 1157

Q1 8.70 21.50

Mean 8.70 28.25

40.00 18.50 5.50 6.10 7.00 6.40 7.60 8.50 5.30 7.20 9.60 10.00 6.15 10.60 5.20 3.40 6.70 3.40 5.20 1.30 -8.60 -11.00 -8.00 -6.80 -5.90 -6.90 -9.90 -7.30 -17.70 -20.80 . . . -6.60

48.55 18.50 16.47 18.93 13.82 13.05 12.33 14.55 10.63 13.33 21.65 21.45 15.90 40.69 22.40 33.08 29.32 50.94 36.58 45.80 36.10 -4.14 -0.67 0.20 2.20 2.80 -3.73 2.18 -8.04 -8.57

11.72

143

Table C.2. Transition probabilities from current funds performance quartiles to follow-on funds performance quartiles
We sort all funds for which we have follow-on funds into performance quartiles and calculate the conditional probability that a partnerships 1st through 3rd follow-on funds will either stay in the same performance quartile as current funds, or move into one of the other three quartiles. The row headings with Expected are the expected probabilities that a follow-on fund will be in one of the four quartiles (column headings 1 through 4) under the assumption that the classification into one of the follow-on performance quartiles is purely random. The column heading with Chi-square reports Chi-square test statistics testing the null hypothesis of no association between current and follow-on funds performance. First, second, and third sub-panels of each panel report transition probabilities from current fund to 1st, 2nd, and 3rd followon funds, respectively. Panel A and B report transition probabilities for buyout and venture capital funds, respectively. Statistical significance at the 1%, 5%, and 10% levels is denoted by ***, **, *, respectively.
Panel A. Buyout Funds 2 3 4 28.08% 26.81% 23.97% 33.71 16.85 12.36 27.27 32.95 19.32 31.03 33.33 27.59 15.09 22.64 45.28 25.54% 36.21 17.86 22.92 22.73 28.30% 25.71 25.00 40.91 23.08 28.26% 31.03 23.21 37.50 13.64 29.25% 37.14 16.67 36.36 30.77 21.74% 10.34 28.57 20.83 36.36 23.58% 22.86 33.33 9.09 23.08

Current to 1st follow-on

Expected 1 2 3 4 Expected 1 2 3 4 Expected 1 2 3 4

1 21.14% 37.08 20.45 8.05 16.98 24.46% 22.41 30.36 18.75 27.27 18.87% 14.29 25.00 13.64 23.08

Total Chi-sqaure 317 45.25 *** 89 88 87 53 184 58 56 48 22 106 35 36 22 13 16.31 *

Current to 2nd follow-on

Current to 3rd follow-on

9.64

Current to 1st follow-on

Expected 1 2 3 4 Expected 1 2 3 4 Expected 1 2 3 4

1 26.59% 35.46 36.44 20.91 9.20 26.15% 31.34 28.05 24.24 17.78 22.02% 31.25 23.64 16.22 10.71

Panel B. Venture Capital Funds 2 3 4 27.29% 24.94% 21.18% 25.46 30.00 9.09 25.42 22.03 16.10 32.73 25.46 20.91 25.29 21.84 43.68 25.77% 25.37 21.95 27.27 31.11 27.38% 27.08 20.00 29.73 39.29 26.54% 25.37 25.61 28.79 26.67 29.76% 25.00 32.73 24.32 39.29 21.54% 17.91 24.39 19.70 24.44 20.83% 16.67 23.64 29.73 10.71

Total Chi-sqaure 425 52.10 *** 110 118 110 87 260 67 82 66 45 168 48 55 37 28 4.31

Current to 2nd follow-on

Current to 3rd follow-on

11.87

144

Table C.3. Pearson and Spearman correlations between current fund performance and follow-on fund performance
This table reports Pearson (Panel A) and Spearman (Panel B) correlation between current fund performance and follow-on fund performance. First four rows and second four rows of each panel are for buyout and venture capital funds, respectively. First row of each sub-panel uses all funds which have IRR data for any of their following funds to compute correlation coefficients. Second through fourth row of each sub-panel requires that a fund raise first through third follow-on funds. The row headings with F~F+t (t=1, 2, or 3) represents that a fund has IRR data from current through t-th follow-on funds. First three columns report correlations; second three columns report p-values, and the last three columns reports the number of funds used to compute correlations.
Panel A. Pearson correlation p -value F+1 F+2 F+3 0.00 0.10 0.39 0.00 0.00 0.16 0.00 0.64 0.28 0.00 0.79 0.30 0.00 0.00 0.92 0.02 0.61 0.29 Panel B. Spearman correlation p -value F+1 F+2 F+3 0.00 0.02 0.40 0.00 0.00 0.02 0.01 0.62 0.22 0.00 0.12 0.21 0.00 0.00 0.15 0.00 0.53 0.33

Buyout

Venture

All F ~ F+1 F ~ F+2 F ~ F+3 All F ~ F+1 F ~ F+2 F ~ F+3

F+1 0.35 0.35 0.26 0.36 0.16 0.16 0.22 0.20

IRR F+2 0.12 0.11 0.05 0.02 0.01 -0.04

F+3 0.09

F+1 317 317 166 82 425 425 228 135

N F+2 184 166 82 260 228 135

F+3 106

0.12 -0.08

82 168

145

-0.09

135

Buyout

Venture

All F ~ F+1 F ~ F+2 F ~ F+3 All F ~ F+1 F ~ F+2 F ~ F+3

F+1 0.29 0.29 0.24 0.29 0.42 0.42 0.34 0.28

IRR F+2 0.17 0.18 0.06 0.10 0.10 -0.05

F+3 0.08

F+1 317 317 166 82 425 425 228 135

N F+2 184 166 82 260 228 135

F+3 106

0.14 -0.10

82 168

-0.09

135

145

Table C.4. Cross sectional regression of current performance on past performance

The table reports the estimates of the following regression: (IRR)t = t + t (IRR) t- + t (Fund Size) t- + t, where is 1, 2, or 3 and t- represents -th previous funds. Funds raised after 2005 are excluded from the estimations. The dependent variable is the logarithms of IRRs of current funds. The independent variables are the logarithm of IRRs and the logarithm of fund size of first, second, and third previous funds in the first, second, and third panels, respectively. In columns (1) and (3), unadjusted IRRs are used as independent and dependent variables, and vintage year fixed effects are included. In columns (2) and (4), benchmark adjusted IRRs are used as independent and dependent variables, where benchmark IRRs are the median IRRs of portfolios of funds with same vintage year, geographic focus, and investment type (early stage, venture capital, or buyout). The standard errors are clustered at the PE firm level. Columns (1) and (2) estimate the equation for buyout funds and columns (3) and (4) for venture capital funds. The numbers in the parentheses are t-statistics. Statistical significance at the 1%, 5%, and 10% levels is denoted by ***, **, *, respectively.
Buyout Adjusted IRR (2) 0.409*** (5.720) -0.012** (-2.512) 0.480*** (7.762) 317 0.133 0.150 (1.618) -0.020*** (-2.684) 0.714*** (7.437) 184 0.063 0.111 (1.060) -0.012 (-1.313) 0.696*** (5.872) 106 0.011 Venture Capital IRR Adjusted IRR (3) (4) 0.234*** 0.278*** (4.541) (4.704) -0.009 -0.025** (-0.875) (-2.547) 0.630*** 0.634*** (9.316) (8.600) 425 425 0.220 0.101 0.094 (1.329) -0.010 (-0.618) 0.782*** (8.903) 260 0.153 0.079 (1.195) -0.031 (-1.530) 0.941*** (9.125) 168 0.180 0.107 (1.455) -0.033** (-1.978) 0.813*** (8.504) 260 0.026 -0.072 (-1.116) -0.045** (-2.090) 1.034*** (8.735) 168 0.050

IRR(t-1) Fund Size (t) Constant Obs. Adjusted R2 IRR(t-2) Fund Size (t) Constant Obs. Adjusted R2 IRR(t-3) Fund Size (t) Constant Obs. Adjusted R2

IRR (1) 0.444*** (6.155) -0.009** (-2.000) 0.489*** (7.533) 317 0.217 0.178* (1.809) -0.018** (-2.476) 0.772*** (7.516) 184 0.072 0.109 (1.009) -0.005 (-0.585) 0.740*** (6.621) 106 0.029

146

Table C.5. Subsequent fund performance (unadjusted IRRs) by quartile portfolios based on current fund performance
Each fund is sorted into quartile rank portfolios in each vintage year based on its IRR. Next, mean and median IRRs are computed for follow-on funds in each quartile. The column headings with F+t where t takes 1, 2, or 3 represent t-th follow-on fund. F is current funds used to form the initial portfolios. Panel A is for buyout funds and Panel B for venture capital funds. Sub-panel A includes all funds. Sub-panels B through D require that a fund has 1st, 2nd, and 3rd following funds, respectively. The last four columns report the number of funds included in the computation of the mean and median values. The last row of each sub-panel reports t-statistics (for the tests of difference in mean IRRs between portfolio 1 and 4) and Wilcoxon rank sum Z-statistics (for the tests of difference in median IRRs between portfolio 1 and 4). Statistical significance at the 1%, 5%, and 10% levels is denoted by ***, **, *, respectively.

(Continuing)

147 147

Table C.5. (Continued)

Panel A. Buyout Mean Quartile F F+1 F+2 F+3 F A. All funds All 15.93 15.45 15.43 15.08 13.85 1 39.52 24.40 18.49 12.51 33.00 2 19.53 15.73 12.80 16.11 19.70 3 10.35 10.71 15.00 16.85 9.30 4 -4.44 7.73 14.98 16.15 -2.30 Diff(1-4) 19.67*** 3.59*** 0.5 -0.27 14.38*** B. Funds with 1st follow-on funds' IRR available All 19.77 15.45 17.90 1 38.47 24.40 32.83 2 20.71 15.73 20.95 3 12.05 10.71 11.60 4 -0.49 7.73 -0.60 Diff(1-4) 12.1*** 3.59*** -9.22*** C. Funds with 1st and 2nd follow-on funds' IRR available All 24.09 18.75 14.65 21.30 1 40.85 25.00 17.47 36.45 2 24.11 18.26 11.58 22.75 3 13.19 14.57 14.19 13.80 4 3.89 12.66 15.17 7.55 Diff(1-4) 9.73*** 2.38** 0.56 -6.1*** D. Funds with 1st, 2nd, and 3rd follow-on funds' IRR available All 30.17 21.65 17.56 14.99 25.75 1 45.83 28.09 20.11 14.89 47.30 2 26.27 20.08 15.08 14.21 25.70 3 18.28 17.74 16.25 15.69 18.00 4 7.03 9.40 18.69 16.31 9.90 Diff(1-4) 5.11*** 4.62*** 0.27 -0.21 -3.9*** Median F+1 12.90 19.70 13.10 9.40 6.20 -3.82*** 12.90 19.70 13.10 9.40 6.20 -3.82*** 16.21 21.60 19.10 12.80 7.90 -2.63** 19.10 24.70 19.00 15.10 8.40 -3.05*** 13.25 16.10 13.70 10.15 14.40 -0.95 16.61 17.50 15.20 16.10 15.10 -0.46 15.00 15.00 15.20 13.70 12.90 0 F+2 13.50 16.90 13.70 11.05 14.40 -0.73 F+3 13.50 12.90 15.20 14.40 12.90 0 F 668 157 173 173 165 F+1 317 89 88 87 53 N F+2 184 58 56 48 22 F+3 106 35 36 22 13

317 89 88 87 53

317 89 88 87 53

148

166 54 46 46 20

166 54 46 46 20

166 54 46 46 20

82 31 25 19 7

82 31 25 19 7

82 31 25 19 7

82 31 25 19 7

(Continuing)

148

Table C.5. (Continued)

Panel B. Venture Capital Mean Quartile F F+1 F+2 F+3 F A. All funds All 14.01 16.83 21.10 24.88 4.84 1 54.35 28.33 31.44 24.28 23.55 2 13.51 26.07 17.52 21.91 6.60 3 2.27 9.93 24.85 38.10 -1.40 4 -11.80 -1.53 6.74 14.28 -11.20 Diff(1-4) 8.72*** 3.71*** 1.51 0.89 16.05*** B. Funds with 1st follow-on funds' IRR available All 23.23 16.83 10.00 1 71.62 28.33 39.80 2 18.69 26.07 15.15 3 4.37 9.93 4.69 4 -7.91 -1.53 -8.80 Diff(1-4) 5.65*** 3.71*** -10.19*** C. Funds with 1st and 2nd follow-on funds' IRR available All 29.84 24.70 21.53 15.10 1 77.30 38.25 28.29 63.70 2 24.76 36.20 19.74 23.65 3 7.89 9.27 25.49 8.70 4 -2.92 5.69 7.84 -1.42 Diff(1-4) 8.96*** 2.36** 1.2 -6.52*** D. Funds with 1st, 2nd, and 3rd follow-on funds' IRR available All 36.84 29.06 30.92 28.79 22.10 1 81.10 44.13 34.07 27.20 63.80 2 27.58 31.26 21.72 23.69 24.10 3 11.40 12.76 48.75 45.44 10.70 4 2.58 16.15 19.16 19.16 1.91 Diff(1-4) 7.82*** 1.17 -0.11 0.42 -4.3*** Median F+1 4.30 8.40 8.10 2.90 -4.00 -4.7*** 4.30 8.40 8.10 2.90 -4.00 -4.7*** 7.95 13.00 11.25 3.70 0.40 -2.04** 16.50 26.20 19.35 10.30 8.10 -0.23 3.55 4.00 4.00 2.00 2.70 -0.31 7.00 6.85 4.75 10.00 7.60 0.82 2.10 3.30 1.65 0.90 3.36 0.4 F+2 3.40 4.20 2.26 2.90 1.10 -0.89 F+3 2.55 3.80 2.30 0.90 4.02 -0.08 F 884 210 226 228 220 F+1 425 110 118 110 87 N F+2 260 67 82 66 45 F+3 168 48 55 37 28

425 110 118 110 87

425 110 118 110 87

149

228 61 70 59 38

228 61 70 59 38

228 61 70 59 38

135 41 46 29 19

135 41 46 29 19

135 41 46 29 19

135 41 46 29 19

149

Table C.6. Current fund performance and follow-on fund growth

The table presents ordinary least square regression estimates of the following specifications: (Fund Growth)t = t + t (IRR) t-1 + t (Fund Size) t-1 + t. Fund growth (from current to 1st follow-on fund), IRRs, and fund size are logarithmized. In columns (5) and (6), we also include buyout fund dummy variable (taking 1 if a fund is a buyout fund and 0 if a venture capital fund) and the interaction term between the buyout dummy variable and IRRs. Columns (1), (3), and (5) show the estimates of the regression of fund growth on unadjusted IRRs, and columns (2), (4), and (6) for the regression of fund growth on benchmark adjusted IRRs. Benchmark IRRs are the median IRRs of portfolio of funds with same vintage year, geographic focus, and investment type (early stage, venture capital, and buyout). In all regression estimations, we exclude funds raised after 2005. In columns (1), (3), and (5), we include vintage fixed year effects. The numbers in the parentheses are t-statistics. Standard errors are clustered at the PE firm level. *, **, and *** indicate statistical significance at the 10, 5, and 1% level, respectively.
Buyout Funds IRR Adjusted IRR (1) (2) 0.596** 0.583** (2.125) (2.319) Venture Capital Funds IRR Adjusted IRR (3) (4) 0.379*** 0.432*** (2.794) (3.465) All Funds IRR Adjusted IRR (5) (6) 0.409*** 0.428*** (3.245) (3.442) -0.097 0.154 (-0.409) (0.767) 0.491 0.176 (1.580) (0.613) -0.161*** -0.140*** (-9.339) (-9.699) 1.226*** 1.220*** (9.551) (10.547) 912 912 0.221 0.179

IRR(t-1) Buyout=1 Buyout*IRR Fund size(t-1) Constant Number of observations Adjusted R2

150

-0.166*** (-6.259) 1.396*** (5.156) 383 0.232

-0.153*** (-6.730) 1.465*** (6.779) 383 0.196

-0.155*** (-6.730) 1.226*** (8.682) 529 0.208

-0.128*** (-7.235) 1.162*** (9.498) 529 0.148

150

Table C.7. Fund growth and follow-on fund performance

The table reports ordinary least square regression estimates of the following specifications: (IRR)t = t + t (Fund growth) t-1 + t (IRR) t-1 + t (Fund Size) t + t. Fund growth, IRRs, and fund size are logarithmized. In columns (5) and (6), we also include buyout fund dummy variable (taking 1 if a fund is a buyout fund and 0 if a venture capital fund) and the interaction term between the buyout dummy variable and fund growth. Columns (1), (3), and (5) show the estimates of the regression of fund growth on unadjusted IRRs, and columns (2), (4), and (6) for the regression of fund growth on benchmark adjusted IRRs. Benchmark IRRs are the median IRRs of portfolio of funds with same vintage year, geographic focus, and investment type (early stage, venture capital, and buyout). In all regression estimations, we exclude funds raised after 2005. In columns (1), (3), and (5), we include vintage fixed year effects. The numbers in the parentheses are t-statistics. Standard errors are clustered at the PE firm level. *, **, and *** indicate statistical significance at the 10, 5, and 1% level, respectively.
Buyout Funds IRR Adjusted IRR (1) (2) 0.001 -0.009 (0.092) (-0.568) 0.443*** 0.417*** (5.727) (5.585) Venture Capital Funds IRR Adjusted IRR (3) (4) -0.064*** -0.069*** (-3.555) (-4.662) 0.250*** 0.257*** (5.569) (4.810) All Funds IRR Adjusted IRR (5) (6) -0.068*** -0.069*** (-3.735) (-4.132) 0.289*** 0.281*** (6.682) (5.852) -0.036 -0.067*** (-1.523) (-2.814) 0.056** 0.065*** (2.485) (2.890) -0.004 -0.008 (-0.845) (-1.569) 0.637*** 0.622*** (11.449) (11.145) 742 742 0.186 0.099

Fund Growth (t-1) to (t) IRR (t-1) Buyout=1 Buyout*Fund Growth Fund size (t) Constant Number of observations Adjusted R2

151

-0.009* (-1.900) 0.489*** (7.520) 317 0.214

-0.010* (-1.894) 0.471*** (8.048) 317 0.127

-0.001 (-0.061) 0.644*** (9.601) 425 0.237

-0.006 (-0.725) 0.630*** (9.398) 425 0.086

151

Table C.8. The effects of fund growth and time gap on performance persistence
The table reports ordinary least square regression estimates of the following specifications: (IRR)t = t + t (IRR) t-1 + 1t (Fund growth) t-1 + 2t (IRR*Fund Growth) + 3t (IRR*Fund Growth*Buyout) + 4t (Time Gap) + 5t (IRR* Time Gap) + 6t (IRR* Time Gap *Buyout) + 7t (Buyout) + 8(Fund size) t + t. IRRs, Fund growth, and Time Gap are logarithmized and demeaned. Fund growth is the growth rate from current to the follow-on fund. Buyout is a dummy variable taking 1 if a fund is a buyout fund and 0 otherwise. Time Gap is the difference between current funds vintage year and the follow-on funds vintage year. The column headings (1), (2), (3) are for buyout funds, and (4), (5), (6) are for venture capital funds. The last three columns include all funds with buyout dummy variables and various interaction terms. In all regression estimations, vintage fixed year effects are included. Standard errors are clustered at the PE firm level. *, **, and *** indicate statistical significance at the 10, 5, and 1% level, respectively.
Buyout Funds IRR (1) (2) (3) 0.441*** 0.387*** 0.382*** (6.472) (5.613) (5.386) -0.000 -0.001 (-0.002) (-0.080) 0.024 0.040 (0.188) (0.312) Venture Capital Funds IRR (4) (5) (6) 0.383*** 0.235*** 0.382*** (7.449) (5.022) (7.421) -0.053*** -0.054*** (-2.739) (-2.770) -0.392*** -0.389*** (-5.338) (-5.252) All Funds IRR (8) 0.257*** (7.078)

IRR (t-1) Fund Growth (t-1) to (t)

IRR*Fund Growth IRR*Fund Growth*Buyout Time Gap (t-1) to (t) IRR*Time Gap IRR*Time Gap*Buyout Buyout Fund size (t) Constant Number of observations Adjusted R2

(7) 0.397*** (10.093) -0.041*** (-3.267) -0.403*** (-6.490) 0.548*** (3.158)

0.001 (0.049) -0.419*** (-2.776)

0.001 (0.064) -0.421*** (-2.779)

-0.006 (-0.284) 0.093 (0.696)

-0.009 (-0.434) 0.038 (0.299)

-0.009* (-1.820) 0.828*** (21.595) 317 0.212

-0.008* (-1.858) 0.825*** (21.933) 317 0.233

-0.008* (-1.818) 0.825*** (21.685) 317 0.228

-0.004 (-0.591) 0.782*** (19.650) 425 0.286

-0.009 (-1.254) 0.806*** (19.912) 425 0.217

-0.005 (-0.607) 0.784*** (19.580) 425 0.283

0.019 (1.555) -0.007 (-1.489) 0.795*** (29.154) 742 0.225

0.004 (0.312) 0.131 (1.152) -0.679*** (-2.816) 0.028** (2.278) -0.010** (-2.069) 0.807*** (29.052) 742 0.174

(9) 0.385*** (9.721) -0.041*** (-3.313) -0.389*** (-6.224) 0.530*** (3.059) -0.001 (-0.074) 0.075 (0.685) -0.528** (-2.252) 0.020 (1.636) -0.007 (-1.447) 0.793*** (29.085) 742 0.228

152

152

Table C.9. The effects of similar market conditions on performance persistence

The table reports ordinary least square regression estimates of the following specifications: (IRR)t = t + t (IRR) t-1 + 1t (MSM) t-1 + 2t (IRR*MSM) + 3t 3t(Fund size) t + t. MSM is market similarity measure defined as the absolute value of ((Market Condition) i,t+1 / (Market Condition) i,t -1). Market condition variables are 1) IPO volume from fifth year to tenth year of a funds life (the column heading with IPO volume), 2) GDP growth during a funds life (GDP growth), 3) S&P 500 stock returns over a funds life (S&P500 returns), 4) three month Treasury bill yield during first five years of a funds life (T-bill), and 5) the ratio of the average S&P 500 price earnings ratio during first five years to that from fifth to tenth year of a funds life (P/E). IRRs and MSM are logarithmized and normalized with mean zero and standard deviation one. The first five columns are the estimates for buyout funds, and the last five columns are for venture capital funds. Standard errors are clustered at vintage year level. *, **, and *** indicate statistical significance at the 10, 5, and 1% level, respectively.
Buyout Funds S&P500 T-bill returns (3) (4) 0.341*** 0.346*** (5.325) (6.990) 0.184*** 0.023 (6.966) (0.332) -0.032** -0.160*** (-2.135) (-3.787) -0.110** -0.119** (-2.336) (-2.366) 0.675** 0.707** (2.393) (2.334) 317 317 0.162 0.163 Venture Capital Funds GDP S&P500 T-bill growth returns (7) (8) (9) 0.279*** 0.319*** 0.411*** (4.589) (5.783) (4.530) -0.228*** -0.265** -0.109** (-4.365) (-2.517) (-2.076) -0.091*** -0.400*** -0.222** (-3.171) (-2.656) (-2.563) -0.146** -0.136** -0.107* (-2.030) (-2.239) (-1.779) 0.710** 0.632** 0.548* (2.055) (2.004) (1.743) 363 425 423 0.150 0.171 0.146

MSM: IRR (t-1)

MSM (t) IRR*MSM (t) Fund size (t) Constant Number of observations Adjusted R2

IPO volume (1) 0.250*** (3.569) 0.203** (2.541) -0.219** (-2.080) -0.070 (-1.451) 0.482 (1.583) 251 0.168

GDP growth (2) 0.270*** (3.915) 0.211*** (2.744) -0.249*** (-2.651) -0.075 (-1.428) 0.488 (1.453) 251 0.178

P/E (5) 0.297*** (4.853) -0.055** (-2.250) 0.057*** (2.889) -0.067 (-1.486) 0.455* (1.667) 269 0.083

IPO volume (6) 0.283*** (3.761) -0.080** (-2.272) -0.038** (-2.054) -0.144* (-1.960) 0.713* (1.936) 363 0.098

P/E (10) 0.299*** (3.460) 0.103*** (3.827) 0.127*** (2.688) -0.125* (-1.775) 0.605* (1.654) 382 0.115

153

153

Table C.10. Transition probabilities from current fund performance quartiles to follow-on funds performance quartiles (using benchmarked IRRs)
We sort all funds for which we have follow-on funds into performance quartiles (by benchmarked IRRs) and calculate the conditional probability that a partnerships 1st through 3rd follow-on funds will either stay in the same performance quartile, or move into one of the other three quartiles. The row headings with Expected are the expected probabilities that follow-on fund will be in one of the four quartiles (column) under the assumption that the classification into one of the follow-on performance quartiles is purely random. The column heading with Chi-square reports Chi-square test statistics testing the null hypothesis of no association between current and follow-on funds performance. First, second, and third five rows of each panel report transition probabilities from current fund to 1st, 2nd, and 3rd follow-on funds. Panel A and B report transition probabilities for buyout and venture capital funds, respectively. Statistical significance at the 1%, 5%, and 10% levels is denoted by ***, **, *, respectively.

Current to 1st follow-on

Expected 1 2 3 4 Expected 1 2 3 4 Expected 1 2 3 4

1 21.14% 34.83 19.32 13.64 13.46 24.46% 27.59 25.00 17.02 30.43 20.75% 17.14 30.56 13.04 16.67

Panel A. Buyout Funds 2 3 4 27.13% 29.65% 22.08% 29.21 25.84 10.11 30.68 34.09 15.91 27.27 31.82 27.27 17.31 25.00 44.23 24.46% 25.86 25.00 25.53 17.39 24.53% 34.29 19.44 21.74 16.67 28.26% 36.21 19.64 34.04 17.39 30.19% 28.57 22.22 43.48 33.33 22.83% 10.34 30.36 23.40 34.78 24.53% 20.00 27.78 21.74 33.33

Total Chi-sqaure 317 35.69 *** 89 88 88 52 184 58 56 47 23 106 35 36 23 12 13.27

Current to 2nd follow-on

Current to 3rd follow-on

7.83

Current to 1st follow-on

Expected 1 2 3 4 Expected 1 2 3 4 Expected 1 2 3 4

1 25.41% 37.27 30.51 16.82 14.44 26.15% 34.78 28.05 23.81 13.04 22.62% 36.74 19.64 10.26 20.83

Panel B. Venture Capital Funds 2 3 4 27.06% 25.41% 22.12% 27.27 24.55 10.91 28.81 25.42 15.25 30.84 25.23 27.10 20.00 26.67 38.89 25.77% 24.64 25.61 23.81 30.44 26.19% 22.45 19.64 30.77 41.67 25.00% 20.29 24.39 26.98 30.44 29.17% 28.57 26.79 35.90 25.00 23.08% 20.29 21.95 25.40 26.09 22.02% 12.24 33.93 23.08 12.50

Total Chi-sqaure 425 38.52 *** 110 118 107 90 260 69 82 63 46 168 49 56 39 24 7.62

Current to 2nd follow-on

Current to 3rd follow-on

18.49 **

154

Table C.11. Pearson and Spearman correlations between current fund performance and follow-on fund performance using benchmark adjusted IRR
This table reports Pearson (Panel A) and Spearman (Panel B) correlation between current fund performance and follow-on fund performance (benchmark adjusted IRR). The benchmark adjusted IRRs are raw IRRs minus the median IRRs of portfolios of funds with same vintage year, geographic focus, and investment type (early stage, venture capital, or buyout). First four rows and second four rows of each panel are for buyout and venture capital funds, respectively. Second through fourth row of each sub-panel requires that a fund raise first through third follow-on funds. The row headings with F~F+t (t=1, 2, or 3) represents that a fund has benchmark adjusted IRR data from current through t-th follow-on funds. First three columns report correlations; second three columns report p-values, and the last three columns reports the number of funds used to compute correlations.

Buyout

Venture

All F ~ F+1 F ~ F+2 F ~ F+3 All F ~ F+1 F ~ F+2 F ~ F+3

F+1 0.34 0.34 0.24 0.35 0.13 0.13 0.20 0.20

IRR F+2 0.11 0.10 0.09 0.01 0.00 -0.04

F+3 0.09

0.12 -0.06

-0.08

Panel A. Pearson correlation p -value F+1 F+2 F+3 0.00 0.13 0.36 0.00 0.00 0.22 0.00 0.42 0.28 0.01 0.89 0.43 0.01 0.00 0.95 0.02 0.67 0.39 Panel B. Spearman correlation p -value F+1 F+2 F+3 0.00 0.04 0.29 0.00 0.00 0.06 0.00 0.20 0.12 0.00 0.14 0.99 0.00 0.00 0.41 0.00 0.66 0.82

F+1 317 317 166 82 425 425 228 135

N F+2 184 166 82 260 228 135

F+3 106

155
Buyout Venture

82 168

135

All F ~ F+1 F ~ F+2 F ~ F+3 All F ~ F+1 F ~ F+2 F ~ F+3

F+1 0.30 0.30 0.25 0.33 0.28 0.28 0.26 0.30

IRR F+2 0.15 0.15 0.14 0.09 0.06 0.04

F+3 0.10

F+1 317 317 166 82 425 425 228 135

N F+2 184 166 82 260 228 135

F+3 106

0.17 0.00

82 168

0.02

135

155

Table C.12. Subsequent fund performance (benchmark adjusted IRRs) by quartile portfolios based on current fund performance
Each fund is sorted into quartile rank portfolios in each vintage year based on its benchmark adjusted IRR. Benchmark adjusted IRRs are raw IRRs minus median IRRs of portfolio of funds with same vintage year, geographic focus, and investment type (early stage, venture capital, and buyout). Next, mean and median benchmark adjusted IRRs are computed for follow-on funds in each quartile. The column headings with F+t where t takes 1, 2, or 3 represent t-th follow-on fund. F is current funds used to form the initial portfolios. Panel A is for buyout funds and Panel B for venture capital funds. Sub-panel A includes all funds. Sub-panels B through D require that a fund has 1st, 2nd, and 3rd following funds, respectively. The last four columns report the number of funds included in the computation of the mean and median values. The last row of each sub-panel reports t-statistics (for the tests of difference in mean IRRs between portfolio 1 and 4) and Wilcoxon rank sum Z-statistics (for the tests of difference in median IRRs between portfolio 1 and 4). Statistical significance at the 1%, 5%, and 10% levels is denoted by ***, **, *, respectively.

(Continuing)

156 156

Table C.12. (Continued)

Panel A. Buyout Mean Quartile F F+1 F+2 F+3 A. All funds All 2.63 2.98 2.96 2.87 1 26.23 11.39 5.17 0.74 2 6.07 3.15 0.59 4.12 3 -2.98 -0.69 3.38 5.01 4 -17.66 -5.50 2.33 1.24 Diff(1-4) 20.87*** 3.51*** 0.35 -0.42 B. Funds with 1st follow-on funds' IRR available All 5.52 2.98 1 24.51 11.39 2 6.32 3.15 3 -2.51 -0.69 4 -14.75 -5.50 Diff(1-4) 13.27*** 3.51*** C. Funds with 1st and 2nd follow-on funds' IRR available All 8.69 5.37 2.22 1 26.22 11.72 4.20 2 8.22 4.48 -0.78 3 -2.15 1.41 2.95 4 -11.67 -0.40 2.21 Diff(1-4) 13.01*** 2.41** 0.56 D. Funds with 1st, 2nd, and 3rd follow-on funds' IRR available All 12.98 7.82 3.68 2.27 1 29.29 15.04 6.25 2.82 2 9.06 6.27 1.35 0.86 3 0.14 3.54 1.41 2.39 4 -9.04 -5.67 6.12 4.22 Diff(1-4) 8.41*** 6.02*** 0.08 -0.26 Median F 0.63 20.45 5.25 -2.93 -14.43 14.55*** 2.60 17.85 5.23 -2.60 -12.05 -9.35*** 6.10 20.80 6.70 -2.60 -11.15 -6.37*** 8.15 29.05 6.75 -1.20 -7.75 -4.18*** F+1 0.55 5.25 1.03 -2.03 -5.03 -4.03*** 0.55 5.25 1.03 -2.03 -5.03 -4.03*** 2.20 7.40 2.15 -0.60 -4.70 -3.09*** 5.65 10.00 2.10 2.83 -5.53 -3.59*** 1.18 1.80 1.03 -0.15 2.75 -0.95 2.15 2.10 2.20 1.73 4.23 -0.19 0.88 2.95 -0.15 -1.48 1.33 0.02 F+2 1.63 1.98 2.38 -0.85 4.25 -0.81 F+3 0.88 0.95 0.62 2.10 -5.55 0.3 F 668 157 173 174 164 F+1 317 89 88 88 52 N F+2 184 58 56 47 23 F+3 106 35 36 23 12

317 89 88 88 52

90 31 28 21 10

157

166 54 46 44 22

166 54 46 44 22

82 31 25 18 8

82 31 25 18 8

82 31 25 18 8

82 31 25 18 8

82 31 25 18 8

(Continuing)

157

Table C.12. (Continued)

Panel B. Venture Capital Mean Quartile F F+1 F+2 F+3 A. All funds All 8.07 11.41 16.62 20.90 1 47.76 22.93 27.85 31.54 2 7.35 20.05 12.08 8.64 3 -3.41 1.34 20.47 2.82 4 -17.33 -2.04 2.60 57.19 Diff(1-4) 8.85*** 3.12*** 1.56 -0.71 B. Funds with 1st follow-on funds' IRR available All 14.70 11.41 1 62.10 22.93 2 9.62 20.05 3 -3.57 1.34 4 -14.87 -2.04 Diff(1-4) 5.52*** 3.12*** C. Funds with 1st and 2nd follow-on funds' IRR available All 17.31 16.88 17.12 1 61.30 29.34 25.08 2 12.29 28.82 13.55 3 -3.45 -0.22 21.37 4 -15.22 -0.08 4.10 Diff(1-4) 9.7*** 2.28** 1.18 D. Funds with 1st, 2nd, and 3rd follow-on funds' IRR available All 22.13 17.07 24.51 24.79 1 61.61 31.22 30.32 35.48 2 12.88 20.72 13.38 8.25 3 -2.13 -2.44 38.92 4.12 4 -11.95 5.11 13.83 82.36 Diff(1-4) 8.27*** 1.74* 0.05 -0.84 Median F 0.07 20.50 4.07 -3.17 -13.80 16.51*** 1.60 29.40 5.95 -3.60 -13.65 -10.75*** 3.73 42.05 8.08 -3.65 -12.90 -7.01*** 8.10 42.05 8.85 -1.69 -7.43 -4.52*** F+1 0.50 4.05 3.70 0.00 -3.28 -3.96*** 0.50 4.05 3.70 0.00 -3.28 -3.96*** 1.68 6.10 5.38 -0.35 -1.20 -1.83* 5.35 12.90 8.70 -1.18 4.18 -0.43 0.18 2.40 -0.02 -0.79 0.53 0.01 1.30 2.40 -0.02 0.20 0.50 0.87 0.00 3.40 -2.25 0.08 -0.35 1.53 F+2 0.18 3.40 -0.15 -0.60 -0.91 -0.92 F+3 0.36 3.30 -1.48 0.00 4.00 1.14 F 884 211 224 229 220 F+1 425 110 118 107 90 N F+2 260 69 82 63 46 F+3 168 49 56 39 24

425 110 118 107 90

149 44 51 35 19

158

228 63 70 57 38

228 63 70 57 38

135 43 46 30 16

135 43 46 30 16

135 43 46 30 16

135 43 46 30 16

135 43 46 30 16

158

Table C.13. The effects of fund growth and time gap on performance persistence using benchmark adjusted IRRs The table reports ordinary least square regression estimates of the following specifications: (IRR)t = t + t (IRR) t-1 + 1t (Fund growth) t-1 + 2t (IRR*Fund Growth) + 3t (IRR*Fund Growth*Buyout) + 4t (Time Gap) + 5t (IRR* Time Gap) + 6t (IRR* Time Gap *Buyout) + 7t (Buyout) + 8(Fund size) t + t, where IRRs are benchmark adjusted. Benchmark adjusted IRRs are raw IRRs minus median IRRs of portfolio of funds with same vintage year, geographic focus, and investment type (early stage, venture capital, and buyout). Benchmark adjusted IRRs, Fund growth, and Time Gap are logarithmized and demeaned. Fund growth is the growth rate from current to the follow-on fund. Buyout is a dummy variable taking 1 if a fund is a buyout fund and 0 otherwise. Time Gap is the difference between current funds vintage year and the follow-on funds vintage year. The column headings (1), (2), (3) are for buyout funds, and (4), (5), (6) are for venture capital funds. The last three columns include all funds with buyout dummy variables and various interaction terms. In all regression estimations, vintage fixed year effects are included. Standard errors are clustered at the PE firm level. *, **, and *** indicate statistical significance at the 10, 5, and 1% level, respectively.
Buyout Funds Adjusted IRR (1) (2) (3) 0.432*** 0.357*** 0.385*** (6.610) (5.445) (5.499) -0.002 -0.007 (-0.177) (-0.496) -0.143 -0.093 (-1.066) (-0.699) Venture Capital Funds Adjusted IRR (4) (5) (6) 0.370*** 0.232*** 0.384*** (7.339) (5.140) (7.956) -0.060*** -0.067*** (-3.164) (-3.498) -0.351*** -0.354*** (-4.626) (-4.721) All Funds Adjusted IRR (8) 0.260*** (7.529)

IRR (t-1)

Fund Growth (t-1) to (t) IRR*Fund Growth IRR*Fund Growth*Buyout Time Gap (t-1) to (t) IRR*Time Gap IRR*Time Gap*Buyout Buyout Fund size (t) Constant Number of observations Adjusted R2

(7) 0.390*** (10.352) -0.040*** (-3.335) -0.378*** (-6.058) 0.449** (2.516)

0.011 (0.793) -0.389** (-2.445)

0.005 (0.395) -0.374** (-2.336)

0.020 (0.996) 0.136 (0.983)

0.003 (0.166) 0.079 (0.593)

-0.010** (-2.147) 0.757*** (26.520) 317 0.128

-0.011** (-2.366) 0.762*** (27.325) 317 0.140

-0.009* (-1.872) 0.749*** (26.053) 317 0.137

-0.008 (-1.117) 0.760*** (21.925) 425 0.128

-0.013* (-1.826) 0.785*** (22.661) 425 0.061

-0.004 (-0.500) 0.735*** (21.045) 425 0.144

-0.016 (-1.346) -0.007 (-1.538) 0.751*** (34.182) 742 0.144

0.012 (0.947) 0.158 (1.353) -0.628** (-2.544) -0.007 (-0.567) -0.011** (-2.468) 0.771*** (35.179) 742 0.093

(9) 0.380*** (9.975) -0.040*** (-3.289) -0.366*** (-5.817) 0.438** (2.453) 0.004 (0.329) 0.081 (0.710) -0.473** (-1.960) -0.015 (-1.305) -0.007 (-1.545) 0.751*** (34.100) 742 0.145

159

159

Table C.14. The effects of similar market conditions on performance persistence (using benchmark adjusted IRRs)
The table reports ordinary least square regression estimates of the following specifications: (IRR)t = t + t (IRR) t-1 + 1t (MSM) t-1 + 2t (IRR*MSM) + 3t 3t(Fund size) t + t, IRRs are benchmark adjusted. Benchmark adjusted IRRs are raw IRRs minus median IRRs of portfolio of funds with same vintage year, geographic focus, and investment type (early stage, venture capital, and buyout). MSM is market similarity measure defined as the absolute value of ((Market Condition) i,t+1 / (Market Condition) i,t -1). Market condition variables are 1) IPO volume from fifth year to tenth year of a funds life (the column heading with IPO volume), 2) GDP growth during a funds life (GDP growth), 3) S&P 500 stock returns over a funds life (S&P500 returns), 4) three month Treasury bill yield during first five years of a funds life (T-bill), and 5) the ratio of the average S&P 500 price earnings ratio during first five years to that from fifth to tenth year of a funds life (P/E). IRRs and MSM are logarithmized and normalized with mean zero and standard deviation one. The first five columns are the estimates for buyout funds, and the last five columns are for venture capital funds. Standard errors are clustered at vintage year level. *, **, and *** indicate statistical significance at the 10, 5, and 1% level, respectively.
Buyout Funds S&P500 T-bill returns (3) (4) 0.334*** 0.343*** (4.844) (7.621) 0.199*** -0.068 (7.576) (-1.307) -0.027** -0.140*** (-2.397) (-4.238) -0.094* -0.105** (-1.830) (-1.997) 0.572* 0.640* (1.767) (1.916) 317 317 0.156 0.153 Venture Capital Funds GDP S&P500 T-bill growth returns (7) (8) (9) 0.267*** 0.284*** 0.323*** (4.647) (5.801) (3.554) -0.160*** -0.172** -0.057 (-3.222) (-2.460) (-1.288) -0.086*** -0.350*** -0.134 (-3.323) (-3.152) (-1.552) -0.095 -0.095* -0.077 (-1.386) (-1.672) (-1.365) 0.459 0.452 0.379 (1.339) (1.510) (1.264) 363 425 423 0.091 0.099 0.071

MSM:

Adjusted IRR (t-1) MSM (t) IRR*MSM (t) Fund size (t) Constant Number of observations Adjusted R2

IPO volume (1) 0.233*** (2.880) 0.140* (1.666) -0.223* (-1.867) -0.059 (-1.056) 0.356 (0.999) 251 0.120

GDP growth (2) 0.240*** (3.372) 0.143* (1.876) -0.256** (-2.405) -0.062 (-1.062) 0.368 (0.977) 251 0.127

P/E (5) 0.287*** (4.031) 0.047** (2.200) 0.050** (2.315) -0.054 (-1.025) 0.327 (0.996) 269 0.070

IPO volume (6) 0.255*** (3.687) -0.081* (-1.867) -0.042*** (-3.933) -0.095 (-1.351) 0.460 (1.283) 363 0.064

P/E (10) 0.257*** (3.092) 0.167*** (8.452) 0.076* (1.784) -0.072 (-1.097) 0.350 (1.021) 382 0.080

160

160

A. IRRs of Buyout Funds

50.00 40.00 30.00 20.00 40.00 10.00 20.00 0.00 -10.00 -20.00 -30.00 Q1 Mean Median Q3 0.00 -20.00 -40.00 100.00 80.00 60.00

B. IRRs of VC Funds

Q1

Mean

Median

Q3

C. Committed Capital (Buyout funds)

180,000.00 160,000.00 140,000.00
50,000.00 70,000.00 60,000.00

D. Committed Capital (VC funds)

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

161

120,000.00 100,000.00 80,000.00 60,000.00 40,000.00 20,000.00 0.00

10,000.00 0.00 40,000.00 30,000.00 20,000.00

Fund Size ($mm in 1990 dollars)

Fund Size ($mm in 1990 dollars)

Figure C.1. Private equity fund performance and fund raising by vintage year
The figures plot fund performance and committed capital by vintage year for buyout funds (Graph A and C) and venture capital funds (Graph B and D). We use internal rate of returns (IRR) reported by private equity partnerships and collected by Preqin as performance measure. Fund size is in 1990 dollars.

161

A. Median IRRs of subsequent funds (Buyout)

35.00 30.00 25.00 20.00 15.00 10.00 5.00 0.00 F -5.00 F+1 F+2 F+3

1 (best) 2 3 4 (worst)

B. Median IRRs of subsequent funds (Venture Capital)

30.00 25.00 20.00 15.00 10.00 5.00 0.00 -5.00 -10.00 -15.00 F F+1 F+2 F+3

1 (best) 2 3 4 (worst)

Figure C.2. Performance of quartile portfolios ranked on current fund performance

Funds are sorted into four rank portfolios based on current funds IRR. Next, median IRRs are computed for follow-on funds in each quartile and are plotted. F+t where t takes 1, 2, or 3 represents t-th follow-on fund. F is current funds used to form portfolios. Graph A is a plot for buyout funds and Graph B is for venture capital funds.

162

A. Cumulative Contribution
12,000,000 10,000,000 8,000,000 6,000,000 4,000,000 2,000,000 -

B. Cumulative Distribution
18,000,000 16,000,000 14,000,000 12,000,000 10,000,000 8,000,000 6,000,000 4,000,000 2,000,000 -

C. Net Cash Flows

8,000,000 6,000,000 4,000,000 2,000,000 -2,000,000 -4,000,000 -6,000,000 -8,000,000 -10,000,000 -12,000,000

Figure C.3. Cash flows of a private equity fund over its life
This graph plots cumulative contribution (Graph A), cumulative distribution (Graph B), and net cash flows (Graph C) of a private equity fund over its funds life. The fund was raised in 1993 and its size is $473 m. Cumulative contribution is the sum of all drawn down capital to that date and cumulative distribution is the sum of all profits returned to investors. Net cash flow is cumulative distribution minus cumulative contribution at a given date.

163