Abstract
This part of the thesis explores the metrics that private equity firms use to measure and manage the performance of portfolio investments. Following a review of existing literature and an examination of known private equity performance measures, I describe theoretical and practical challenges that private equity firms encounter when attempting to apply metrics from conventional finance theory to private equity investments.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The PEIGG study was conducted as an online survey and distributed to “the membership of the Institutional Limited Partners Association (ILPA), Private Equity CFO Association, National Venture Capital Association’s (NVCA) CFO Task Force, the VCBC network and to various other individuals and entities on an ad-hoc basis” (PEIGG, 2004, p. 2). The sample comprises 55 private equity firms (mostly venture capital), 28 investors, and 7 fund of funds. The study report includes an explicit notice that the results are “not scientific and should not be interpreted as a statistical representation of the private equity industry” (PEIGG, 2004, p. 1)
- 2.
As this part of the thesis is concerned with performance measurement in private equity firms, issues of performance calculation that are particular to external parties (e.g. because they do not have access to detailed cash flow information) are not addressed here (see Kaplan & Schoar, 2005, and Kaserer & Diller, 2004b, on problems of survivorship bias and the accurateness of reported net asset values)
- 3.
Chiampou and Kallett propose the use of alternative methods for evaluating venture capital risk. For example, they compare the percentage distribution of venture capital partnerships across defined return categories with the percentage distribution of securities in an equity capital market across the same return categories, or, they compare the respective percentage of funds and securities in each asset class that were in default at a certain date. These measures are clearly less suitable for measuring the risk of a single portfolio investment
- 4.
The idea that private equity firms may be exposed to idiosyncratic risk was not new. In their study on the determinants of required return in venture capital investments, Manigart et al. (2002) declared that “the existence of huge market imperfections [in the market for venture capital investments] implies that idiosyncratic investment risk and other investment characteristics may be as important as market risk in determining required return (Rea,1989)” (p. 294). Using survey data from 209 venture capital firms in five countries, they found that mean required rates of return varied across investment stages and that the return required for early-stage investments was significantly higher than the return required for later-stage investments. They concluded that “contrary to the expectations of conventional finance theory, diversification by investment stage does not appear a significant risk-reduction strategy of VCCs” (Manigart et al., 2002, p. 292)
- 5.
Fund returns based on actual cash flow data can of course only be calculated after the fund has been liquidated. The use of net asset values (i.e. accounting information) to calculate private equity fund returns is often considered as problematic because net asset values include estimates of the value of unrealised investments (e.g. Ljungqvist & Richardson, 2003a, p. 4; Phalippou & Gottschalg, 2009, p. 2) 63 In a related study, Groh, Baule, and Gottschalg (2008) use a Contingent Claims Analysis to calculate implied idiosyncratic risk of individual leveraged-buyout transactions, but do not assess the aggregate risk of the private equity firm
- 6.
In a related study, Groh, Baule, and Gottschalg (2008) use a Contingent Claims Analysis to calculate implied idiosyncratic risk of individual leveraged-buyout transactions, but do not assess the aggregate risk of the private equity firm
- 7.
Phalippou does point out that such “strategic activity” would also negatively affect a private equity firm’s income from carried interest and that the additional analysis of money multiples can help (external parties) to uncover such distortions (pp. 7–8)
- 8.
Industry practitioners also use other terms, such ascash multiple,cash on cash return, orcash on cash multiple (e.g. 3i Group, 2006; Craigs Investment Partners, 2009; Grabenwarter, 2007; Soulignac, 2002)
- 9.
The definition here is general in that it can be used for individual investments or entire funds. Some authors use other definitions that vary from this one in terminology or scope (e.g. Bader, 1996, p. 314; Grabenwarter & Weidig, 2005, p. 22; Ljungqvist & Richardson, 2003b, p. 21); see also the definitions in the EVCA Reporting Guidelines, 2006, and the Global Investment Performance Standards of the CFA Institute, 2010
- 10.
Required multiples include distributions to paid-in capital (DPI orrealised multiple), residual value to paidin capital (RVPI orunrealised multiple), and the total value to paid-in capital (TVPI orinvestment multiple). Note that the guidelines do not require these ratios to be reported for the underlying investments (theEVCA Reporting Guidelines recommend the inclusion of a market-to-book ratio, i.e.fair value to cost, for individual investments; see CFA Institute, 2010, and EVCA, 2006)
- 11.
In private equity literature Ljungqvist and Richardson (2003b) and Sudarsanam and Nwaghodoh (2005), for example, use annualised multiples or related metrics
- 12.
Note that this characteristic is only true for positive, non-zero IRRs as the monotonic exponential function (1+IRR)t increases with t, only if IRR > 0. In cases where the IRR is zero, the simple sum of all cash flows is equal to zero – hardly an attractive investment proposition. In cases where the IRR is below zero, the IRR equation in fact attributes lower weightings to early cash flows than to later ones
- 13.
Reyes (2003) concedes that the IRR is less suitable for comparing private equity returns to public market returns; he maintains that, in this case, “the IRR becomes problematic and either a TWR or Index Method PME approach may be used” (p. 33)
- 14.
Some authors recommend the use of duration-weighted means (e.g. Phalippou, 2009) or capital-weighted means (e.g. Bader, 1996, p. 324) for approximating pooled IRRs. In the example, the capital-weighted mean of investment IRRs amounts to 15.3%. For some purposes it may also be suitable to calculate thesimultaneous IRR of a portfolio (i.e. the hypothetical pooled IRR, assuming all investment projects having commenced in the same year); in the example simultaneous IRR amounts to 14.5%. See Bader (1996, pp. 318–325) and Kraft (2001, pp. 297–302) for additional information on methods of aggregation
- 15.
Kierulff (2008, pp. 327–328) additionally demonstrates that the MIRR resolves the ranking problem of the IRR for mutually exclusive projects — provided that the projects do not differ in size
- 16.
Busse von Colbe and Laßmann appear to imply the opportunity cost of capital (see Busse von Colbe & Laßmann, 1990, p. 53)
- 17.
A note of caution is required here. As pointed out by Eagle, Kiefer, and Grinder (2008), the incremental principle (and therefore the underlying assumption of independent investments) may be breached if the discount rate reflects a particular investment opportunity
- 18.
This is less of an issue for open-ended or evergreen funds, which is reflected in the fact that they are treated differently in industry reporting guidelines (see, e.g. the Global Investment Performance Standards of the CFA Institute, 2010)
- 19.
See also Footnote 62 on the use of net asset values for calculating private equity fund returns
- 20.
See Reyes (2003) and Long (2008) for a more detailed account of the metric’s early development
- 21.
Long and Nickels provided a detailed description of this methodology in an unpublished working paper presented at the AIMR Conference in 1996 on venture capital investing (see Long, III & Nickels, 1996)
- 22.
Formula of Kaserer and Diller (2004a, p. 405), adjusted for the notation used in this paper
- 23.
More recent extensions to the PME method have tried to address this issue (see, e.g. Peterson, Kasarda, & Grier, in press)
- 24.
The definitions of the profitability index sometimes vary (see Brealey et al., 2011, pp. 143–144)
- 25.
A formal difference is that the public market equivalent sets end values in relation to each other, whereas the profitability index sets present values in relation to each other. As can easily be shown by extending the PME formula with the \( \frac{{\prod\nolimits_{i = 1}^T {{{(1 + {R_i})}^{ - 1}}} }}{{\prod\nolimits_{i = 1}^t {{{(1 + {R_i})}^{ - 1}}} }}, \) the ratio of end values is equivalent to the ratio of present values, and, for practical purposes, is possibly more a matter of preference; perhaps depending upon whether the metric is being used ex ante, for investment decisions, or ex post, for evaluation purposes
- 26.
Let IA … In be contributions to an investment and DA … Dm be distributions from the investment, then the capital gain equates to \( \sum\nolimits_{i = m}^m {{D_i}} - \sum\nolimits_{j = 1}^n {{I_j}} \) and the investment multiple is \( \frac{{\sum\nolimits_{i = m}^m {{D_i}} }}{{\sum\nolimits_{j = 1}^n {{I_j}} }} \)
- 27.
Schefczyk points out that he does not regard the default ratio as a satisfactory measure: “Die Ausfallquote kann aber keinen exakten und finanzwirtschaftlich befriedigenden Risikomaßstab darstellen, da sie letztlich nur eingetretene Misserfolge anstelle des Risikos im eigentlichen Sinne abbildet [the default ratio cannot, however, constitute an exact and financially satisfactory risk measure, because, in the end, it represents only unsuccessful investments rather than risk in the true sense]” (Schefczyk, 1998, p. 169; 2006, p. 156; reiterated by Kraft, 2001, p. 307; and Vater, 2003, p. 38)
- 28.
In the past, illiquidity of private equity partnership interests was also attributed to a lack of vibrant secondary markets (e.g. Ljungqvist & Richardson, 2003b, p. 4). More recent data from practitioner reports suggest that these markets have experienced strong growth over the last years and have now reached a sizeable proportion of the primary private equity market (e.g. Lüchinger & Schnyder, 2009)
- 29.
In a recent study, Franzoni, Nowak, and Phalippou (2011) provide evidence for the pricing of liquidity risk in private equity returns
- 30.
Bodie, Kane, and Marcus (2008, pp. 317–322) also discuss extensions of the standard capital asset pricing model that incorporate liquidity costs and liquidity risk. These extensions are rather complex and require data that are typically not available for private equity investments
- 31.
In his discussion of criteria for assessing the quality of measures, Hannula (1999, pp. 69–79) refers to a 1979 dissertation by Vehmanen on accounting measures. Unfortunately, a copy of Vehmanen’s work was not available to the author of this paper
- 32.
Sink (1985) put forward a list of quality criteria very similar to that of Emory. Sink’s criteria included among othersvalidity andreliability as well ascomprehensibility andcost effectiveness
- 33.
More precisely, this description of validity is what Emory refers to asinternal validity, which he sets in contrast to external validity, defined as the “ability [of a measure] to be generalized across persons, settings, and times” (Emory, 1985, p. 94)
- 34.
In later publications, Hannula labels his quality criteria for productivity measurement more generally as a “framework for the criteria of sound performance measurement” (Lönnqvist & Hannula, 2000, p. 1)
- 35.
The implicit assumption here is that the interests of private equity firms and their managers are highly aligned with each other so that the “value and usefulness” of measures for the “users” (i.e. investment managers of private equity firms) are principally determined by their ability to measure the attainment of “organisational goals and strategy”
- 36.
- 37.
This is not a circular argument because private equity firms compete with each other for investor capital and because fund managers have incentives to attract investors by offering purposeful information — thus lowering information asymmetries and agency conflicts (Kemmerer & Weidig, 2005, pp. 5, 9, and 23)
- 38.
Example: Let carried interest be 20% of all distributions exceeding a threshold of 108% of capital contributions. In rough terms, an investment will contribute (proportionately) to the achievement of carried interest by the private equity firm if its investment multiple exceeds the level of 1.08. The proportionate amount of carried interest roughly equates 20% of the positive difference between the investment multiple and 1.08 (disregarding possible catch-up provisions, repayments of management fees, and taxes etc.)
- 39.
Moreover, top performing private equity firms can also hardly be considered to be capital constrained (Gompers & Lerner, 2001, p. 152)
- 40.
As a matter of convenience the rule is here formulated as a rule of acceptance, rather than as a rule of rejection
- 41.
In this context, (Brealey et al., 2011, pp. 141–142) point out another source of error: Computation of the opportunity cost of capital (i.e. the benchmark for the IRR) can be challenging if the opportunity cost of capital is not the same for all cash flows, for example, when short-term interest rates are different from long-term rates. This is indeed a concern, especially for investments with sizeable interim cash flows. But the issue should not be a problem for investments that consist of only two cash flows (i.e. investment and exit cash flow)
- 42.
Indeed, this should be a concern for investors who rank private equity firms and funds using IRRs and multiples
- 43.
Contrary to this expectation, Harris, Jenkinson, and Kaplan (2012) present results suggesting that the investment multiple is a more reliable indicator of (fund-level) PME than is the IRR
- 44.
On the latter point, Phalippou (2009) points out that comparing multiples and IRRs can help investors to identify “distorting” IRR effects that are potentially a result of “strategic” behaviour of private equity firms (i.e. private equity firms seeking early exits or dividends from investments in order to boost reported IRR, although additional returns above opportunity cost of capital could have been achieved by waiting longer), because investment multiples are not affected by timing of cash flows and because they are lower if the investment is exited at a lower price than might have been achieved, for example, by waiting longer
- 45.
The relationship between the investment multiple (m) and IRR for an investment with only two cash flows C0 < 0 and CT > 0 can be described as follows: \( \frac{{ - {c_0}}}{{{{(i + IRR)}^0}}} + \frac{{{c_T}}}{{{{(1 + IRR)}^T}}} = 0;m = \frac{{{c_T}}}{{{c_0}}} \Leftrightarrow m = {(1 + IRR)^T} \)
- 46.
This should be a particularly sensitive issue for the purposes of external reporting, as investors and private equity firms may well have different views on the appropriate level of discount rate
- 47.
Practitioners sometimes refer to the difference between expected return and minimum return requirement (the hurdle rate) as headroom. The headroom gives a feel for the extent that the constituent amounts can change adversely before the investment criterion is breached
- 48.
Consider a private equity firm that needs to decide between investment projects A and B. Both projects require immediate investment of 100. Project A is expected to generate proceeds of 200 at the end of the first year, whereas project B is expected to generate proceeds of 300 at the end of the second year (no interim cash flow in either project). As sole decision criterion, the multiple would lead to a preference of project B (MultipleB = 300/100 = 3.0 is larger than MultipleA = 200/100 = 2.0). The IRR as sole decision criterion would lead to a preference of project A (IRRA = 200/100 – 1 = 1 is larger than IRRB = [300/100]1/2 – 1 = .73). Net present values would yield a preference of project B for any discount rate below 50% and a preference of project A for any discount rate above 50%
- 49.
This is particularly evident in private equity reporting: Private equity reporting standards and guidelines typically do not require private equity firms to report measures of risk or risk-return to investors; in fact, the word risk occurs only twice in a 32 page document on reporting guidelines issued by the European Private Equity & Venture Capital Association (EVCA, 2006) and not at all in the respective 22-page document issued by the Private Equity Industry Guidelines Group (PEIGG, 2005). The Global Investment Performance Standards of the CFA Institute (2010) have only recently included provisions that relate to risk
- 50.
A glossary of success measure definitions from the survey package is included in Appendix B
- 51.
One buyout respondent specified “profit increase and amount of deleveraging”. One venture capital respondent specified “internal key performance indicators” as important for the firm
- 52.
The results of the paired samples tests are available from the author upon request
- 53.
A Wilcoxon signed ranks test with all respondents who had assigned differing (paired) importance ratings to the IRR and multiple also yielded a significant preference for the multiple (z = 2.25, p = .024, two-tailed, n - ties = 75)
- 54.
The five exclusive number one ranks that were not assigned to the multiple or IRR (see Table 27), are distributed among NPV, DCF Or APV, profitability index, and PME
- 55.
It is also conceivable that private equity firms use different success measures for different geographic areas, investment stages or perhaps investment types — this would make comparing or aggregating performance very cumbersome
- 56.
Consistent with respondents’ evaluations of individual success measures (see Chapter 4.5.1), the multiple was generally attributed a little more importance than the IRR
- 57.
Additional statistical tests of the relationship between rejection frequencies and IRR importance (both as success measure and as decision criterion) did not yield significant results (alpha = .05, two-tailed)
- 58.
Detailed information on these statistical tests are available from the author upon request
- 59.
This ambiguity of interpretation is unfortunately the result of an ill-defined survey question (see Appendix B for the exact wording), making it impossible to distinguish between the importance that respondents attribute to each decision type and the importance they attribute to the respective metrics for these decisions
- 60.
For example, a high importance rating of the factor inflation rate would suggest that the respective private equity firm is accustomed to undertaking investments that are exposed to varying inflation rates
- 61.
The results for (time-insensitive) Capital Gain point in the opposite direction; however, both the level of significance and the strength of correlation are low
- 62.
These results stand in some contrast to findings of Kemmerer and Weidig (2005) regarding the effect of fund experience and focus on reporting characteristics
- 63.
The influence of the variable CEF on respondents’ evaluations of success measures is possibly caused by a link between the variables CEF and Outside Inv: χ2 = 16.9, Φ = .36, df = 1, p < .0005, two-tailed, n = 133 (one cell in the contingency table had an expected cell count lower than 5; p is the value of Fisher’s exact test)
- 64.
Because the question was formulated as an open question it is not possible to discern whether respondents who had left the response field blank had been unwilling to respond to this item on the questionnaire, or had meant to imply that their firm did not use any measure of risk or risk-return (see Appendix B for the exact wording of the question)
- 65.
In Chapter 4.4 a number of conflicting opinions on the relative importance of the IRR and the multiple for industry practice were presented. The empirical results of the survey offer support for the views of Bader (1996, p. 310), Kraft (2001, pp. 290–308), and Weidig and Mathonet (2004, p. 6), who had suggested that the IRR and multiple are both typical metrics in private equity
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer Fachmedien Wiesbaden
About this chapter
Cite this chapter
Broere, M. (2014). Success Measures and Decision Metrics. In: Decision-Making in Private Equity Firms. Springer Gabler, Wiesbaden. https://doi.org/10.1007/978-3-658-03780-2_4
Download citation
DOI: https://doi.org/10.1007/978-3-658-03780-2_4
Publisher Name: Springer Gabler, Wiesbaden
Print ISBN: 978-3-658-03779-6
Online ISBN: 978-3-658-03780-2
eBook Packages: Business and EconomicsEconomics and Finance (R0)