Risk Mgmt in PE-GSAM Jan11

Hitting the Curve Ball: Risk Management in Private Equity

White PaperJanuary 2011

Executive Summary� While private equity has been affected by the challenges of recent years, many investors

are today renewing their focus on these investment strategies, mindful of the considerabledispersion of performance between top managers and average managers, and the possibilityof meaningful outperformance by the strongest general partners (GPs).

� This renewed focus from limited partners (LPs) comes at a time when investors are taking a broader look at their entire portfolio across all asset classes, and challenging themselvesto have a deeper understanding of the risks they are assuming. In private equity, such a riskassessment cannot rely on static checklists and high-level conclusions.

� Recent experience has reminded us that the private equity program cannot be viewed inisolation. It is a means to an end. What really matters is understanding just how the programmight succeed or fail in contributing to broader, portfolio-wide investment objectives in the ways originally intended.

� Recent experience has also revealed that the potential challenges in a private equity portfoliotend to surface at different times and in different ways. In other words, the environmentmatters. A risk framework should distinguish benign regimes from volatile regimes, ideallyringing an alarm bell before the onset of trouble.

� Mindful of these observations, investors should first consider the places from which risks mayemerge in a private equity program. Broadly speaking, these include the unrealized companies(individually and in aggregate, as a portfolio), the remaining unfunded commitments, themanager’s organization, and the investor’s private equity program itself.

� Importantly, the risk assessment should not be guided by the structure and labels of a particularfund—it must focus instead on underlying components and the way they interact with eachother, inside and outside of the private equity portfolio.

� Such a granular focus allows risk managers to identify the key variables, and in turn map tobroader risk factors, permitting more dynamic insights into the behavior of the portfolio underdifferent scenarios. This component-level analysis can also be complemented by a top-downapproach using public-market proxies, appropriately adjusted and tailored. These approachespierce traditional fund classifications, invite interesting hypothesis testing, and offer a mechanismfor integrating with the entire investment program.

� One of the primary reasons for translating the private equity program into factors is toemploy a common language spoken by all of the asset classes, enabling a coordinated analysisand, in turn, a coordinated response.

� A “commit and hope” strategy is not really a risk management strategy. In our view, activemanagement should be a critical part of risk management in private equity. Simply approvingor disapproving new additions to the portfolio is insufficient. Investors should considerchanges that rebalance exposures and re-shape risks, working across all asset classes andavailing themselves of mechanisms like the secondary market.

This is the submitted version of the article: J. Christopher Kojima and Daniel J. Murphy, “Hitting the Curve Ball: Risk Management in Private Equity,”Journal of Private Equity, Spring 2011, Copyright (c) 2011, Institutional Investor Inc.This information discusses general market activity, industry or sector trends, or other broad-based economic, market or political conditions andshould not be construed as research or investment advice. Please see additional disclosures.This material is provided for information purposes only and does not constitute a solicitation or offer to provide any advice or services in anyjurisdiction in which such a solicitation or offer is unlawful or to any person to whom it is unlawful. Please note that Goldman Sachs Asset Management does not maintain any licenses, authorizations or registrations in Asia ex Japan, except that itconducts businesses (subject to applicable local regulations) in and from the following jurisdictions: Hong Kong, Singapore, Malaysia, Korea, and India.

Alternative Investments &Manager Selection (AIMS) (212) 855-0462

2 | Goldman Sachs Asset Management

Risk Management in Private Equity

This information discusses general market activity, industry or sector trends, or other broad-based economic, market or political conditions and should notbe construed as research or investment advice. Please see additional disclosures.

“You’ve gotta be careful if you don’t know where you’re going, otherwise you might not get there.”—Yogi Berra

Investing in top-performing private equity managers can be similar to the challenge of consistentlypicking a winning baseball, football, or soccer team. Investors must choose a team based on a varietyof objective and subjective factors, considering historical track records, the quality and depth of theteam’s bench, the effectiveness of the strategy, the competitiveness of rivals, the passion of the players,and the weather conditions on the field. And like the die-hard sports fan, the private equity limitedpartner (LP) often finds itself invested for the entire game before success truly can be determined. To besure, there are interim scores and directional indications of performance. But unlike the public-marketinvestor who enjoys regular liquidity and reliable valuation-marks, the private equity investor cannotdeclare victory midway through the game, and generally speaking, cannot leave the stadium prematurely.

The commitment can be worth it, however. While the performance of private equity as an asset classis a question complicated by data availability, varied sample sets, and eventful time periods, what isreasonably clear is that the dispersion of performance between top managers and average managers is very wide (generally exceeding 1,200 basis points, as shown in Exhibit 1), and that the top teamsoutperform public equity markets, sometimes by a considerable margin.1 And perhaps even more important,there is meaningful persistence in the relative performance of private equity general partners (GPs),meaning that there is generally something special about the top teams that tends to keep them on top.2

The strongest GPs have the potential to outperform the public markets, often a result of their abilityto enhance company operations, focus on longer-term strategy, inject growth and stabilizing capital,improve corporate governance, and capitalize on unique acquisition and financing dynamics with sellers.While it is no small feat to find these top managers, the promise of consistently strong investment returnshas led an increasingly diverse array of investors to private equity. Indeed, as illustrated in Exhibit 2,over the past decade LPs have committed nearly three trillion dollars to private equity managers.

Exhibit 1 – Private equity exhibits a significant dispersion in managers’ investment performance3

Sources: Thomson Reuters VentureXpert (private data) and Morningstar Principia (public data)

Upper Quartile ManagersMedian ManagersLower Quartile Managers

US FixedIncome

LargeCap

SmallCap

InternationalEquity

AsiaEquity

OpportunisticReal Estate

Private Equity(All VentureEconomics)

Buyouts Venture

35%

30%

25%

20%

15%

10%

5%

0%

-5%

1 Data from Thomson Reuters VentureXpert database indicates the average spread between top and bottom quartiles for leveraged buyout funds raised ineach vintage year from 1996 through 2005 is 17.1%, and the average spread for venture capital funds is 19.7%.

2 See, for example, (Kaplan and Schoar, 2005)3 Private equity fund returns are compared with 4-year annualized average returns of public funds, realized over a period starting three years after the

private equity fund vintage year. This lag and time-averaging is done in order to take into account the private equity investment period and long time-horizonover which private returns are realized. Private equity data is for funds raised between 1989 and 2003 as of June 30, 2010. The public data for public fundreturns over the years 1992 through 2008. Managers are chosen for each market segment described here based on a strategy description from Morningstar.

Exhibit 2 – Fundraising has increased significantly, with considerable cyclicality

Source: Thomson Reuters VentureXpert

I. The Current LandscapeIn contrast to this promising long-term picture, recent years have been hard on just about everyone.An environment in which it was relatively easy to raise, borrow, invest, and return capital turned intoone where every element of the private equity business became more difficult. Some managers had thegood fortune to conclude fundraising just prior to the financial crisis, providing them “dry-powder”for new investments at perhaps better valuations, and additional time to endure some bad weather.Others were less fortunate, running out of both money and time, failing to convince a shrinking numberof LPs who could still handle a long-term illiquid commitment amidst pressures more broadly in theirportfolios. Credit markets retreated, severely limiting not only acquisition financing for new deals, butalso the debt capital that fueled distributions coming from secondary buyouts and special-dividendrecapitalizations. And the abrupt end to a steady stream of distributions disrupted many traditionalcash-flow forecasts, and revealed the extent to which some LPs had perhaps over-committed to theasset class, hoping distributions would forever fund future capital calls. In the face of these difficulties,some GPs struck out and retired from the game entirely, and some LPs decided to sit out the next gamewhile they repaired their broader portfolios.

But today, there are once again some encouraging dynamics. The approach to valuations in privateequity meant that investors never really experienced the severe drawdowns of the public equity marketsin 2008 (or earlier in the decade, for that matter).Those write-downs in net asset value (NAV) thatdid occur throughout the first part of 2009 were considerably reversed before the end of the year, asillustrated in Exhibit 3. Credit markets which seized up in 2008 began their recovery, and in someparts of the market today, could even be characterized as somewhat aggressive. Exit opportunitieshave begun to emerge, as IPO activity has increased significantly year over year.4 Reinforcing thedispersion of returns in the asset class, stronger GPs distinguished themselves from weaker ones, eitherin their ability to conserve cash, negotiate extensions in debt maturities, identify superior managementteams, or swiftly adjust their portfolio companies’ strategies to fit a new operating environment.

(US$ in Billions)

SecondaryFund of fundsReal assetsMezzanineTurnaround/DistressedVenture capitalBuyout

$0

$100

$200

$300

$400

$500

$600

1H1020092008200720062005200420032002200120001999199819971996199519941993199219911990

Goldman Sachs Asset Management | 3


4 “Private Equity, Public Exits.” October 2010. Ernst & Young.





Indeed, recent research has observed that businesses backed by private equity sponsors defaulted at less than half the rate of comparable companies,5 and that companies involved in private equitytransactions tend to pursue more influential innovations.6 And while declaring victory in operatingperformance will ultimately require overcoming the lingering headwinds in the broader economy,improving capital market conditions and increasing M&A activity today provide further signs thatthe ingredients for future de-leveraging and distributions are developing.

Exhibit 3 – Net Asset Value in private equity funds has steadily recovered from the lows of early 2009

Source: Goldman Sachs AIMS Group

A similar recovery has been felt by many LPs. Many who feared funding capital calls found that theslower pace of new deal activity in the immediate wake of the financial crisis provided some welcomerelief, and time to examine the health of their entire investment portfolio. Following such a review,mindful of uncertain equity markets and historically low interest rates, many LPs concluded that alternatives(hedge funds, private equity, and real estate) were not options that could be dismissed if they wantedto come close to achieving their actuarial or long-term return assumptions. Today, several investorsurveys suggest that many LPs are either maintaining or increasing their allocation to private equity, as illustrated in Exhibit 4.7

Exhibit 4 – Few LPs plan on reducing private equity allocations

What changes do you expect to make to your portfolio’s allocation to private equity over the next year?

Source: Goldman Sachs 2010 AIMS Diagnostic Survey

Distressed/Special situationsMiddle-market buyoutLarge-market buyoutVenture capital

(NAV Indexed to June 2008, Adjusted for Cash Flows)

60%

70%

80%

90%

100%

Mar-10Dec-09Sep-09Jun-09Mar-09Dec-08Sep-08Jun-08

Increase significantly 6%

Increase moderately 39%No change 46%

Decrease moderately 8%

Decrease significantly 1%

5 (Thomas, 2010)6 (Lerner, Sørensen and Strömberg, 2008)7 For example, the Goldman Sachs 2010 Alternative Investments Diagnostic, a survey of over 400 alternative investment managers and investors, indicated

that 46% of private equity investors planned to maintain their private equity allocation, and an additional 45% planned to increase their allocation.



II. Identifying the Old and New RisksFor many investors, this renewed focus on private equity comes at a time when they are taking abroader look at their entire portfolio and challenging themselves to have a deeper understanding ofthe risks they are assuming. If properly executed with the strongest GPs, a private equity strategy mightvery well contribute returns in excess of the public markets. But what are the risks that accompanysuch an attractive outcome? If we can fully identify these risks, how can we manage them? And forthose risks that we cannot mitigate, are we sufficiently compensated for assuming them?

Some of the risks inherent in any private equity program are straightforward. Illiquidity and leverage,for example, remain familiar considerations. The dispersion of returns in private equity performancealso points to the importance of understanding the risks in manager selection, with real consequencesif mistakes are made. And recent accounts of funds closing their doors, teams in conflict, ownershipstakes being monetized, or combinations with other organizations all further suggest that interestingthings can happen to the manager’s franchise that demand ongoing attention. The LP who has experiencein evaluating managers has these and other items on its formal or informal checklist.

But a deeper risk assessment of private equity cannot rely on a static checklist, however detailed.Recent experience has reminded us of two important points. First, for the LP investor—whether apension fund, financial institution, sovereign wealth fund, endowment, foundation, or individual—the private equity program is a means to an end. It cannot be considered in isolation. What reallymatters is understanding just how the program might fail to contribute to broader, portfolio-wideinvestment objectives in the ways originally intended. In order to do this, private equity investorsneed their risk framework to communicate with the broader portfolio as they detect departures fromoriginal expectations. Leverage, illiquidity, manager risks, and meaningful performance dispersiononly raise the stakes in getting it right.

Second, the potential challenges in a private equity portfolio tend to surface at different times and in different ways. In other words, the environment matters. There will be times when the environmentis accommodating, when credit markets are aggressively lending to private-equity backed companies,when public-market multiples expand, when the broader economy serves as a tailwind—happycircumstances which might easily deliver the GPs’ upside-case scenarios on operational growth or exit proceeds. In contrast, there will be times when everything is difficult, when the credit marketscontract (or even freeze), when the broader economy shrinks along with top-line revenue growth,when raw-input pressures on margins persist, and when acquirers and IPO markets shun what exitingGPs have to offer, closing the door on distributions and discouraging new fundraising. Some challengesremain in either environment; others manifest themselves only selectively. A risk framework should beconsistent but recognize these different regimes, ideally ringing an alarm bell before the onset of trouble.

These two observations—the need for context, and the dynamic nature of these risks—make the riskassessment less about a checklist, and more about identifying the places from which current and newrisks might develop, depending on the environment. These are summarized in Exhibit 5.





Exhibit 5 – Five broad areas where risks originate in many private equity programs

Risks at the Company Level

One obvious place from which risks emerge is the current collection of unsold companies within theexisting investment program. The question here is whether the game is going as planned, and if not,how far off and in what ways is the team ahead or behind. This is an inquiry that happens in everyenvironment, and in accommodating regimes, it might be fairly easy to assess. After all, if your teamis winning, you might be less concerned with precisely how you got there. If the score is in your favor,and particularly if it has remained so for quite some time, LPs may be tempted to dull the sharpnessof the risk inquiry to a simple accounting of which companies are ahead of plan, which are on track,and which are somewhat behind.

Less accommodating regimes, however, teach us that the inquiry must always go deeper. Mindful of itstarget returns, the GP will typically determine an acceptable purchase price for any given company aftermaking some educated forecasts on certain operational factors (future revenue growth, margin improvement,cash-flow generation), balance-sheet factors (debt pay-down, covenant flexibility, maturity schedules),and market-based factors (exit timing, probable buyers and exit proceeds). These are the granular variablesthat require attention. How has the recession impacted top-line growth? How have labor costs changed?Is working capital fatally constrained? Do acquirers or the IPO markets appear as capable of deliveringan exit as originally hoped? What is the amortization schedule of the outstanding debt, and whatcovenant flexibility remains? Have the probabilities and degrees originally ascribed to the upside, base,and downside cases now changed, such that the “shape” of the outcome distribution has been altered?Particularly in the current economic environment, a more fulsome risk assessment must consider howthe operating, financial, and market assumptions have all changed from the original underwriting,quantifying where possible the impact across each granular variable.



Risks at the Portfolio Level

A second perspective from which to observe emerging risks is the portfolio as a whole, consideringhow a given fund’s companies relate to each other and how they relate to the broader pool of unrealizedprivate equity positions in the LP’s entire program. Here, diversification might very well provide somecomfort. Mindful of manager selection risks and the general dispersion of private equity returns, LPs willsensibly diversify across strategies and managers, with the total portfolio ultimately comprised of hundredsof underlying portfolio companies. While some companies may fail to achieve their ambitions, diversificationand some smart manager picks, the argument goes, should cushion most downside scenarios.

Unfortunately, less benign environments may paint a different picture, and reveal more complicatedchallenges. First, as experienced between asset classes broadly in 2008, stressed macroeconomic and market conditions can reveal unusual correlations within the private equity portfolio itself.Consider a credit market crisis that impedes working capital and refinancings in an indiscriminate wayacross all companies, or an equity market collapse that denies an IPO exit to all companies, irrespectiveof industry or GP or region. Second, in environments of heightened competition between GPs—perhapsencouraged by aggressive lending, tight credit spreads, and overall low interest rates—copycat GPsmight be forced to court the third- or fourth-tier business in any given industry (if the already-acquiredindustry leader is unavailable), and these might be weaker businesses and management teams whoseshortcomings only really show in tough times. Third, in difficult periods GPs may need to ration their finite time, operational resources, and negotiating leverage with creditors in the direction of justthe healthiest companies. Finally, syndications between GPs, through which multiple managers maybe invested in the same company, test the benefits of manager diversification in the best of times, but in the worst of times may expose tensions between GPs and paralyzing disagreement on strategiccourse corrections.

Best observed when looking across the entire portfolio, these possibilities really point to an asymmetryof outcomes in each environment. The upside case is a happy outcome that probably rewards selectcompanies considerably, and others more modestly. But the downside case in volatile regimes can hiteveryone dramatically. Even detecting these issues can be problematic. The periodically reported NAV,unable to continually refresh for operational and market factors, will tend to lag the best snapshotestimate of “fair value.”8 In benign environments where asset prices are broadly rising, this time-lagwill generally mean NAV is conservative. In more volatile environments, the opposite may be true, andNAV may range from being merely imperfect, to actually over-stating reality, precisely at the time anaccurate read is most needed.

Risks in the Unfunded Commitments

A third source of risk surrounds the remaining time in the current game. When evaluating their existingportfolio, LPs must consider not only the current unrealized companies, but also the remaining uninvestedcapital for which they remain accountable. In many respects, these are really a form of off-balance-sheetliabilities, perhaps easily ignored in favorable environments. After all, when the pace of distributionsexceeds the pace of capital calls, liquidity concerns may be long forgotten and the focus might shift to the difficulty of maintaining (or increasing) a target allocation to private equity. This was the verydynamic facing some LPs in the 2004 to 2008 timeframe, an environment in which concern for the“quality” of the remaining unfunded was perhaps not top of mind for everyone.


8 In September 2006, the US Financial Accounting Standards Board (FASB)—through its Statement of Financial Accounting Standards (SFAS) No. 157—updatedand clarified existing rules on the use of fair market value (FMV) in generally accepted accounting principles. It also provided additional guidance on howto calculate FMV. While it is likely that the SFAS No. 157 changed the valuation practices of some GPs, its impact on the overall industry remains unclear.



In all environments, however, there is risk that the manager might strike out in future innings of thesame game. LPs should always conduct a dispassionate assessment of the return prospects from theremaining unfunded exposure, drawing inferences based on the performance of the already realizedcompanies as well as the health of the remaining unrealized companies. Since every dollar of unfundedexposure is a dollar essentially funded at par (relative to the target return), LPs need to be vigilant thatthe original return expectations remain as achievable as when the commitment papers were signed.

Exhibit 6 – Capital calls and distributions tend to decline during periods of economic stress, with capital callsrecovering earlier than distributions


In less favorable circumstances, this might not be the case. In recent years, a meaningful slowdown indistributions has forced all LPs to take notice of their unfunded obligations, particularly as capital callshave begun to resume (and at a rate faster than the recovery of distributions), as shown in Exhibit 6.What they are discovering in today’s environment is that the scoreboard may indicate that they aretrailing, and that the time remaining to play may be limited. For some GPs with struggling legacyinvestments, the only hope for any performance-based economics (or, less ambitiously, a break-evenresult on the overall fund) may lie in “swinging for the fences,” taking unusual risks in the hopes ofan unusual payoff. Such potential escalations in risky behavior can remind LPs that the GPs’ carriedinterest resembles a call option on the portfolio, with the option’s value enhanced by increasing thevolatility of the underlying assets.9

(% of Commitment)

-2.0%

-1.5%

-1.0%

-0.5%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

Deviation from average contributions (4-quarter rolling average)Deviation from average distributions (4-quarter rolling average)

Jun-10Jun-09Jun-08Jun-07Jun-06Jun-05Jun-04Jun-03Jun-02Jun-01Jun-00Jun-99Jun-98


9 Consider the analogy to an equity call option, which represents the right, but not the obligation, to purchase a stock at a pre-determined price (the “strike price”)before a defined future expiration date. The longer in the future the expiration date is, the more valuable the option, since the stock has a longer periodof opportunity to rise above the strike price. Since the downside of a call option is limited to the initial price paid for the option, an increase in the volatilityof the underlying stock also increases the value of the option, because greater variation in the price of the stock means a higher probability of risingabove the strike price. In the case of private equity carried interest, the analogous expiration date is the termination of the fund, and the strike price isthe preferred return, which increases over time. As an under-water fund comes closer to the end of its investment period, the combination of a nearerexpiration date and higher strike price creates downward pressure on the value of the carried interest call option, which may encourage a manager toattempt to increase that value by investing in riskier deals.



Exhibit 7 – Significant capital commitments are reaching the end of their investment periods

Source: Goldman Sachs AIMS Group. As of September 30, 2010.

For some GPs, time is indeed running out, as noted in Exhibit 7 above and Exhibit 8 below. As theend of the investment period nears, with recent years lost to credit market dislocations or stubbornseller price-expectations, some GPs sitting on expiring uninvested capital might also be tempted tocompromise standards or drift into unfamiliar strategies, rather than return capital to LPs. Investing intechnology in the late 1990s and credit-oriented strategies in 2009 are familiar examples of occasionalstrategy drift. Even for managers less susceptible to such temptations, their traditional investment spacemight find itself suddenly crowded by new entrants, changing the competitive dynamics or elevatingvaluation multiples. The stronger GPs will have sourcing advantages, valuation discipline, and a long-term perspective on their franchise. LPs will need to ensure that they are partnering with suchmanagers to avoid unusual and new unfunded risks.

Exhibit 8 – “Dry powder” is concentrated in 2007 and 2008 vintage-year funds

Source: Thomson Reuters VentureXpert and Goldman Sachs AIMS Group. As of June 30, 2010.

$0

$50

$100

$150

$200

$250

$300

$350

$400

$450

$500

2009200820072006200520042003200220012000

(Capital Commitments, US$ in Billions)

Estimated dry powder

Number of funds

(Total Commitments Expiring, US$ in Billions) (Number of Funds)

Total commitments

$0

$20

$40

$60

$80

$100

$120

$140

$160

Next 36 MonthsNext 30 MonthsNext 24 MonthsNext 18 MonthsNext 12 Months0

5

10

15

20

25

30




Risks in Manager Selection

A fourth source of private equity risk concerns the team itself. Here, the question is whether the LPswant season tickets for next year, or if a switch to a different team is overdue. The risk of choosingpoorly, particularly in private equity where the performance dispersion is so wide and the chances of error so real, has always been considerable. When making a new commitment decision, whether to an existing manager or a new one, investors have always faced a daunting due diligence challenge.Quantitative and qualitative in focus, direct and indirect in approach, the diligence inquiry by LPs hasalways needed to cover a wide range of topics, including the historical track record, the health of thecurrent portfolio, the attractiveness of the investment strategy, the depth and quality of the GP’s financialand operating teams, the alignment of incentives, and the economic and legal arrangements that governwhat might very well be a decade-long relationship.

But the environment can make this already difficult exercise more complicated, for at least three reasons.First, the information needed to make an informed decision is now harder to find. If fueled only by performance-enhancing credit markets, managers’ prior successes may be things of the past, andmay not necessarily reflect any special value-creation capabilities. LPs have to look beyond IRRs and ROIs to identify the top-tier from the average, focusing on how GPs have weathered recentstorms within their portfolio companies, whether in tactically adjusting business strategy, enhancingmanagement, conserving cash, or amending and extending credit packages. To be sure, capitalizingon favorable macroeconomic trends is not unwelcomed. At their most aggressive moments, relaxedlending standards, frothy credit markets, and rising equity markets all benefit the strong and mediocreGP alike. But precisely because these macro factors washed over the entire community of GPs inrecent years, investors need to work that much harder—and the risks of a false positive are that much greater—in identifying the next season’s strongest contenders.

Second, LPs today might not be able to defer the decision. This change in the availability and usefulnessof traditional information comes at an inconvenient time. As observed earlier, there are some seriousquestions to ask about the health of the unrealized portfolio and the ambitions of the unfundedcommitments. But the problem is not just one of heightened LP curiosity. With many investment periodsnearing their end, and perhaps some anxiety building about the sustainability of their franchise, certainGPs may be compelled to force on their LPs a “re-up” decision very soon, perhaps even prematurely.For these LPs, the recent environment creates not only ambiguous signals from the existing companies,but also brings with it fewer distributions, lingering unfunded exposures, and broader portfolio-widechallenges to address. These burdens add weight to an already heavy decision.

Exhibit 9 – Buyout managers have become accustomed to increasing fund sizes… …but recent fundraising has proven to be a challenge

Source: Thomson Reuters VentureXpert and Goldman Sachs AIMS Group. Based on a sample of 1,046 buyout funds. As of June 30, 2010.

(% of Previous Fund Size)

(Fund Sequence Number)

0%

50%

100%

150%

200%

250%

Fund VIIFund VIFund VFund IVFund IIIFund II


% of funds where size increased

(% of Previous Fund Size)

(Vintage Year)

% of previous fund size

0%

50%

100%

150%

200%

250%

300%

200920082007200620052004200320022001200019991998199719961995199419930%10%20%30%40%50%60%70%80%90%100%

(% of Funds Raised that were Larger than Previous Fund Size)



10 For example, if public markets were to suggest a much more difficult environment for future private equity investing, this might argue that future commitments,funded dollar-for-dollar, cannot be assumed to grow at previously-assumed IRRs, translating into a current penalty or haircut to the NAV.


Third, LPs must also consider new stresses on the stability of the GP’s team, pressures which mightonly reveal themselves in an uncertain and volatile environment. How unfamiliar is the burden on the thinly staffed GP, whose efforts were subsidized by a buoyant market in good times, but who intough times must prioritize resources to maintain struggling legacy investments? How committed are the mid-level deal partners—who might be the next generation of leaders, but not yet covered bykey-person provisions—when their carry points are now worth less, or worthless? If successor funds(and streams of management fees) are smaller, how will the GP’s cost structure adjust? As illustrated in Exhibit 9, many GPs have enjoyed years of increasing fund sizes and corresponding revenue streams.For many, facing a new reality in fundraising, a new cost structure may be necessary. Organizationalstability, generational transition, and shifting ownership were risks that industry participants werealready moving to address, even before the financial crisis. Now, informed by the realities of a newenvironment, these legacy issues are amplified and further complicate the manager-risk assessment.

Risks in the Execution of the Program

A final category of risk looks beyond the uncertainty associated with the underlying companies,funds, or managers. At a higher level, the implementation of the private equity program itself, and its relationship to the broader investment portfolio, present a unique set of risks for the portfolio-wide CIO. Differences in liquidity, data availability, and connectivity to the rest of the portfoliodistinguish private equity from other asset classes. In accommodating environments, these may bemanageable issues. Isolating any strategy—whether private equity, public equity, fixed income, or any other—is surely suboptimal in all circumstances, but in times of plenty, when that stand-alonestrategy is performing nicely and demanding nothing from its adjacent asset classes, such isolationmight be tolerated. Lack of transparency in standard-issue private equity reporting may be just aminor inconvenience when the strategy is outperforming public-market benchmarks. And theoreticalilliquidity and unpredictable capital calls might not disrupt cash-flow models too terribly, if distributionsare well ahead of schedule.

But in more volatile environments, a very different picture emerges. As observed in 2008, diversificationacross asset classes afforded little protection for investors. With volatility spiking, correlationsincreasing, and normally liquid markets suddenly illiquid, all asset classes disappointed at the sametime. These experiences remind us that investors always need to look across traditional asset-classcategories, and that no asset class can be an island. While private equity presents some peculiarfrictions in making such a connection to the broader whole, the temptation to conclude in good timesthat private equity is an uncorrelated source of excess return makes it all the more critical that suchconnections are established. Two particular program-wide correlations are often under-appreciated.

The first is equity-market risk. Is the private equity program truly diversifying, or is it instead amplifying,public equity exposure? Estimating the equity market (beta) risk within a private equity portfolio iscomplicated by the many drivers of performance. But, as summarized in Exhibit 10, certainly things likea welcoming IPO market, multiple expansion in comparable public companies, the acquisitiveness ofstrategic buyers, and top-line revenue growth all meaningfully correlate to the public equity markets.From a return perspective, this connection itself might not trouble investors, particularly if consistentoutperformance to the public markets can be realized. From a risk perspective, however, the correlationscannot be ignored, particularly since the public-market effects can be amplified through private-companyleverage. Periodic movements in underlying funds’ quarterly NAVs might attempt to capture a real-timevaluation, but with their time lag and subjectivity, they almost certainly understate market exposure.This refreshed NAV also entirely ignores equity market risks associated with the portfolio’s ongoingunfunded exposures—whose IRR prospects are also impacted by the public markets—and thereforefails to address whether these fixed unfunded obligations should be welcomed or feared.10



Exhibit 10 – Changes in private equity NAV reflect public market movements


The second portfolio-wide consideration is liquidity risk. All asset classes can suffer moments of illiquidity,and as painfully revealed by recent experience, on occasion do so at the same time. But is this relevantfor the private equity strategy, given the “natural” anticipated illiquidity in the mechanics of such a program? Unfortunately, it is as relevant for the private equity team as it is for the fixed income,hedge fund, real estate, or public equity teams. An economy-wide liquidity and credit problem directlyimpacts underlying portfolio companies, not only in long-term debt refinancings, but immediately inworking capital lines and acquisition revolvers. Fund-level credit facilities can be curtailed by the banks,constraining GPs’ cash management activities and perhaps encouraging pre-emptive capital calls fromunderlying LPs. And if the liquidity issue stems from a broader credit-market problem, LPs mightsuddenly find few distributions fuelled by secondary (GP to GP) buyouts and special dividend recaps.Such LPs might find that their “over-commitment strategy”11 ceases to be clever cash-flow management,and instead crosses the tipping point into another form of unmonitored program-level leverage, de-riskedonly by uncertain distributions and occasionally requiring emergency funding from other asset classes.

These are only some of the many risk factors that cut across the entire portfolio, of which private equityis just a part. Falling long-term interest rates can provide helpful tailwinds to leveraged buyout returnsand fuel a wave of fundraising interest, as experienced initially in the 1980s. Emerging growth marketscan drive end-user demand across numerous private-equity-backed companies, as anticipated by manytoday, and they can also present sovereign debt challenges, as experienced in the 1990s. Fluctuations incommodity prices might impact not just energy-related private equity strategies, but also raw-materialcosts and the likelihood of GP-driven margin improvements. These are all factors that do not fit neatlyinto traditional asset-class classifications.

(NAV Indexed to June 2008, Adjusted for Cash Flows; S&P 500 Total Returns)

Private equity NAV changeS&P 500

50%

60%

70%

80%

90%

100%

Sep-10Jun-10Mar-10Dec-09Sep-09Jun-09Mar-09Dec-08Sep-08Jun-08


11 LPs might sensibly conduct such a strategy, recognizing that capital calls happen unpredictably over time, and that the entire commitment to a fund neednot be funded immediately. For further details on commitment strategies to private equity, see (Goldman Sachs Asset Management, 2007).



III. Taking Action: Mitigating and Managing RisksStepping back, how should investors begin to think about managing these broad sources of risk?Whether originating at the company level, portfolio level, manager level, or program level, risks stemmingfrom these places can be complex, hidden, and dynamic, depending heavily on the environment. And while some of them can be mitigated by helpful GPs, even the most thoughtful GP is exercisingthis thoughtfulness on its particular fund, not on any given LP’s broad portfolio. Simply put, GPs arenot trying (and arguably, should not try) to solve individual LPs’ broader liquidity, diversification, orprogram-wide implementation issues. What this means is that LPs must look to themselves for solutions.

Exhibit 11 – Investors must dive beneath the surface to examine, identify, and mitigate private equity risks

These are not insurmountable challenges, and investors who avoid them may miss an opportunity.The question is whether these risks can be managed. Exhibit 11 above suggests a framework forexploring this question. LPs may find that with the right resources and tools, they are not only able to manage the risks within their private equity program, but also able to enhance the contribution of these strategies to their broader investment objectives.

Step #1: Assessing the Score – Penetrating the Fund Structure

The first step is to quantify the existing risks, reflecting on the broad risk categories discussed earlier. For those investors whose primary means of accessing private equity is through fund selection, this exercise can be difficult, in part because of the disconnect between where many of the risks lie (at the company level) and where much of the LP’s decision-making occurs (at the fund level).Traditional strategy buckets, like “large-market buyouts” or “venture capital” or “emerging markets,”are useful starting points, but of course these are crude labels that are really only invoked to shed somelight on the underlying behavior of managers and companies. More probing analyses—decomposinga particular fund’s past and current performance by industries, by geographies, by vintage years, or by deal-team leaders—are more helpful and certainly necessary, but in most cases also fail to tell the full story of how the game is going. This is particularly the case in volatile environments.


Penetrate the fund structure by drilling down to the individual company level, ignoring the fund wrapper and short-hand classifications.

Map granular exposures to observable and exogenous forces driving portfolio risk, mindful that some factors reveal themselves only in certain environments.

Test hypotheses across multiple scenarios to shed new light on the health of the current exposures, timing of capital calls and distributions, and future funding requirements.

Triangulate the private portfolio by developing a proxy based in the public marketsusing a risk-based approach to calibrate alpha, volatility, and beta assumptions.

Manage the multiple dimensions of private equity risk through portfolio-wideand strategy-specific adjustments, including secondary-market solutions.

LPs looking to deepen their understanding of the remaining portfolio need to drill down to the individualcompany level, ignoring the fund wrapper and short-hand classifications. Only then can creepingconcentrations of risk be identified across the entire pool of unrealized positions, irrespective of how theyare labeled by the GP.

But LPs should not stop there. Ideally, the risk management inquiry should penetrate even deeper tooperational insights at the individual company level, as illustrated in Exhibit 12 below. This is wheremany risk assessments struggle, often due to data or resource limitations in observing company-specificrevenue growth, cash-flow improvement, working capital issues, progress with debt pay-down, or shiftsin exit multiples, all across a multitude of companies. Without looking at the performance of the individualplayers in this way, LPs are forced to make conclusions about the game just by looking at the headlinescore. These operationally driven inquiries not only shed light on the likely outcome, they also educateon how the company is performing relative to the original underwriting case. In turn, this may commenton the sensibility of the current NAV for that particular company.12

Exhibit 12 – A risk assessment should distinguish funded from unfunded exposure, and penetrate to underlying risk factors

Next, LPs must consider the granular risks in the unfunded exposure. Often overlooked in any quantitativeapproach, the operations and behavior of underlying managers must also be incorporated into a riskframework, particularly given the extent to which unfunded exposures and manager stability impactthe risk of the private equity program. How does generational transition of the manager’s leadershipfactor into the analysis? What if the identified key-persons quit the team? What is the likelihood that carried interest will be realized in this particular fund? What fund-level assets or liabilities mustbe considered, beyond the aggregate company valuations? How precisely do the carried interest,preferred return, and clawback mechanics work? Some of these variables will not reveal themselveseasily. These are subjective assessments, but the challenge of ascribing values, weights or rankings to such qualitative fields does not make them any less relevant to the design of a risk scorecard andevaluative framework.




12 Beyond telling us something about the conservatism or optimism of the GP’s accounting, this “quality of NAV” assessment is an important thresholdinquiry, since subsequent aggregating analyses will incorporate much of this periodic NAV metric into other analyses (in some ways elevating it to thesame “stature” as prices observed in the public markets).



This level of granularity might seem overwhelming when the many component parts are considered.In a perfect world of unlimited resources, LPs might very well examine each part with equal care.Realistically, most LPs will need to pick their spots. But merely attempting such a micro-level inquirywill almost certainly reveal concentrations in the portfolio, either at the company-level or fund-level,and these concentrations will serve to focus attention. For example, a program considerably moreunfunded than funded will have its outcomes much more driven by assumptions attached to the manager’sbehavior in deploying new capital, and these variables will overwhelm any single revenue, cash-flow,or exit multiple assumption on an individual company in the unrealized portfolio. In aiming finiteresources at this broad analytical exercise, LPs need to look for the most efficient places on which to focus attention, a triage exercise only possible after piercing the traditional fund structure andlooking at the relevant pieces.

Step #2: Mapping to Broader Risk Factors

Armed with these more granular assessments on the score of the game and interim performance ofthe team, LPs can now turn their attention to seeing how the game’s outcome is affected by broaderrisk factors in the economy and markets. Like a well-functioning team on the field, the private equityportfolio consists of many moving parts. To understand better when these individual parts might out-perform or under-perform expectations, LPs need to first identify the outside forces that influence them.

Exhibit 13 – Identifying risk factors at the company level is essential in forming a portfolio-wide view

Some broad risk factors will be easy to spot, and easy to map. Consistently rising oil and gas pricesgenerally will be helpful tailwinds for energy-related funds, and the risks to such funds will beheightened in choppy energy markets. Private equity managers focusing entirely in Spain or Japancould be particularly affected by macroeconomic factors (like Spanish housing prices) and marketfactors (like the Nikkei index) unique to those countries. A company-level understanding remainscritical to ensure the selected risk factors are neither over-inclusive nor under-inclusive. But when the risks stemming from existing portfolio companies, unfunded commitments, and a particularmanager’s franchise all tightly identify with a particular industry or region, observing the volatility in those industries or regions can educate LPs on the related risks in the private equity portfolio.




In contrast, other risk factors will be broader in application, cutting across fund-specific boundariesand revealing themselves only in certain environments. Consider equity market risk as an example. In stable and rising equity markets, it might be easy to conclude a manager’s above-plan realizationsare entirely driven by unusual skill (manager alpha). But closer inspection might reveal that thesedistributions are in fact due to expansion in valuation multiples—not an unwelcomed development,but one correlating quite closely with robust equity markets. And in more challenging environments,volatility in public equity securities—driven not only by market technicals, but also by earnings releasesand balance-sheet events—might remind LPs that the private equity companies are subject to changesin the fundamentals just like comparable public companies. Exhibit 14 illustrates such correlation formedian buyout returns and the public equity markets. (Note that top-quartile managers might verywell reveal less correlation.) Particularly within a given industry, changes in GDP growth, regulatoryuncertainty, consumer demand, or the acquisitiveness of strategic buyers will all impact the equity riskin public and private companies alike.

Exhibit 14 – Median returns at the fund-level exhibit some correlation with cycles in public equities13

Source: Thomson Reuters VentureXpert (private data), Standard & Poor’s. As of June 30, 2010.

Another broad risk factor might be fundraising risk. Less obvious than equity or credit risk, changesto the fundraising environment (particularly negative changes) can drive risks across all dimensions of the portfolio. For existing portfolios, such changes may trigger in GPs a greater propensity to sellexisting companies that are clearly in-the-money, and to generate distributions to impress prospectiveLPs. When coupled with some early historical losses, a challenged fundraising environment might alsoprovoke managers to assume unusual new investment risks. If the prospects for a successor fund arediminished, it might even lead to teams fragmenting, disrupting the stewardship of existing portfoliocompanies (and, in a circular way, adding further burdens to the original fundraising challenge).Unlike more easily measured metrics, this may be harder to quantify. But using objective evidencefrom funds’ capital raises, fundraising extensions, distribution activity, and broad surveys of LPintentions can bring definition to such a risk factor.14

(IRR)

(Fund Vintage Year)

Median buyout IRRS&P 500 (6-year average return)

-5%

0%

5%

10%

15%

20%

25%

200220012000199919981997199619951994199319921991199019891988198719861985


13 S&P 500 returns are the six-year average returns of the index for the time period starting one year after the year indicated, to reflect the time necessaryto invest private equity funds and the holding periods of underlying investments.

14 (Kaplan and Schoar, 2005)



These are just a handful of the broader array of risk factors that LPs might identify as observable and exogenous forces that drive risk in the portfolio. Industry and regional variables, equity markets,credit spreads, interest rates, commodity prices, foreign exchange, and capital raising are among theinfluences that ex ante could reasonably be viewed as impacting the private equity program. Identifyingthe appropriate risk factors is as much art as science, even in public-market contexts where the datachallenges are fewer. But in private equity, where the data is infrequent, inaccessible, or imprecise, the obstacles can be greater. If private equity funds’ NAVs were updated daily by GPs, if there wasless subjectivity in their calculation, if there was some mechanism to incorporate the economic changesin the prospects for unfunded exposures, then traditional econometric techniques might offer someguidance. Unfortunately, this is not the situation enjoyed by LPs.15

Data limitations need not prove fatal, however. Armed with a granular understanding of the company-level, fund-level, manager-level, and program-level risks, LPs can take steps to improve their situation.First, GPs should be enlisted as partners in the exercise, and LPs should consider which operationaldata (really, the only hard data solely within the possession of the GPs) they will require in order torefresh their own granular perspectives. Second, LPs must develop their own systems architecture,cutting through the traditional fund structure and focusing on the underlying companies and othercomponent parts. Such systems must include evolving mechanisms for extracting, synthesizing, storing,and accessing an exceptional amount of information. Finally, LPs will need to use considerable judgment,creativity, and hypothesis testing in determining which risk factors are truly important. Sometimes therelationships might even be counter-intuitive.16 Well-resourced LPs asking the right questions (and aidedconsiderably by accommodating GPs) can ignore traditional asset-class categories, pierce conventionalfund structures, unwind the portfolio into its component parts, and connect these parts to broaderrisk factors. This sets the stage for thoughtful scenario analysis.

Step #3: Testing Hypotheses with Scenario Analysis

Having identified the critical variables and mapped them to various risk factors, LPs can take this webof cause-and-effect relationships and examine how the game might evolve, or at least understand betterthe range of possibilities. As with the exercise of determining individual factors, packaging these relationshipsinto plausible scenarios will require some judgment. LPs will need to tailor for context. The downsidecase—reflecting changes in revenues, margins, capital expenditures, debt retirement, exit multiples, exittiming—for an aircraft-parts manufacturer in the U.S. should differ from the downside case of a healthcareservices business in Germany. The downside probability itself (that is, the chances of a one or twostandard-deviation event occurring) is also likely to differ from company to company, since the outcomesfor every business are unlikely always to be normally distributed. Further, not every permutation andcombination will make sense—the operational downside scenario affecting a counter-cyclical businessis unlikely to occur simultaneously with the downside scenario affecting a pro-cyclical business.

But thoughtfully designed factors and scenarios—going beyond simple fund-level classifications (like “buyouts” or “venture”), and probing deeper than single-variable “what if” hypotheticals—canhelp define a more textured distribution of outcomes for LPs to consider. One analysis might be a simplecomparison to the GPs’ original upside, base, and downside cases, testing each for reasonableness inthe current reality, and pressing GPs for a refreshed view. Mindful of recent experience with convergingcorrelations in distressed environments, additional “shock scenarios” can also be considered. Past events—a 1970s oil shock, a 1980s interest rate spike, a 1990s sovereign debt crisis, a 2002 technology marketcollapse, a 2008 credit crisis—all can be modeled to further stress-test volatilities and correlations inexit proceeds or exit timing.


15 Due to the facts that net asset values are only updated on a quarterly basis, and significant discretion is left to the GP in the calculation of these values,return estimates calculated from capital statements may provide only an approximation of the actual change in economic value over the period. The lackof reliable time series of returns precludes the use of many portfolio analysis techniques, as the fundamental inputs of volatility and correlation cannotbe directly calculated from market data, and regressions cannot easily be run directly on return series.

16 For example, see (Dimson, Marsh and Staunton, 2005) for an analysis of how stock market performance appears to be uncorrelated to GDP.



Similar exercises can be conducted on the future outcomes stemming from the current unfunded exposures.As with the unrealized portfolio, tailoring the inquiry will be equally important. The downside case fortoday’s struggling buyout firm with only months remaining in its investment period will differ from thedownside case for a new firm that just confidently completed its first fundraising. The risk of strategydrift (and unwelcomed capital calls) may be higher for a manager whose limited partnership agreement(LPA) is overly accommodating, and whose traditional investment strategies differ from temptingopportunities in new sectors showing real dislocation. As illustrated in Exhibit 15, LPs should testoutcomes given different return assumptions on remaining unfunded commitments. LPs analyzing anexisting manager with few historical distributions and real challenges in the current portfolio might verywell conclude that the manager’s current fund will not return 100% of the capital called. Depending onthe unfunded assumptions, the viability of any successor funds might also be called into question.

Exhibit 15 – Current status of portfolio may put pressure on unfunded to perform, skewing incentives for GPs

For illustrative purposes only.

These scenario analyses are interesting to observe within the private equity program by itself, butthey are even more relevant when they link up with scenarios being run across the entire portfolio.Seeing how the private equity program behaves along with the fixed income, public equity, hedgefund, real estate, commodities and other asset pools when a particular shock scenario is conductedmay illuminate new insights. Indeed, one of the primary reasons for translating the private equityprogram into factors is to employ a common language spoken by all of the asset classes, enabling acoordinated analysis and, in turn, a coordinated response. The risk of a liquidity crunch, for example,would not truly reveal itself unless the analysis combined observations on probable distributionsfrom unrealized company positions, timing of future capital calls, length of investment period, andavailable liquidity from elsewhere across the entire investment program.

Exhibit 16 – Examination of company-level factors yields portfolio-level insights

0.00x

0.25x

0.50x

0.75x

1.00x

1.25x

1.50x

1.75x

2.00x

Typical Fund Structure Fund A0.5x on Unfunded

to Reach Carry

Fund B2.1x on Unfunded

to Reach Carry

Fund C4.1x on Unfunded

to Reach Carry

UnrealizedNAV

Unfunded Cost

UnfundedProfit

Return of Cost

Receipt of Carry

Realized

UnrealizedNAV

Unfunded Cost

UnfundedProfit

Realized

UnrealizedNAV

Unfunded Cost

UnfundedProfit

Realized

Cost

Preferred

Carry/Profit Split


Penetrate Traditional Labels, Considering Components Granularly

Analyze Aggregate Portfolio, Focusing on Factors

Test Hypotheses Across Multiple Scenarios



Step #4: Triangulating with Proxies in the Public Markets

Such a factor-based approach can help shed new light on the health of the current exposures, timing offuture capital calls and distributions, and future funding requirements for the private equity program asa whole. But it is an undertaking complicated by at least three practical challenges. First, when multipliedacross thousands of companies and hundreds of funds, the exercise demands extensive time, resources,and judgment. Second, mindful of the sheer breadth of expertise that would be required to applysuch judgment across industries and managers, the exercise also demands unbiased consistency by thejudges themselves in assigning scenario probabilities, defining upside or downside cases, or mappingto exogenous factors. Third, while extraordinarily complicated in its permutations and combinations,the exercise is itself ultimately focused on discrete cases, and fails to capture a complete continuum ofpotential outcomes and shifting correlations.17 How, then, can LPs without extensive resources proceedin quantifying private equity risk? And even for those with such capabilities, how should they test theconclusions of their component-level, returns-oriented analyses?

Ideally, the secondary market for private equity would provide some convenient solutions.18 Just likethe data-rich comparable market for public equities, the private equity secondary market reflects thecontinually changing valuations of numerous buyers and sellers. Prices within the market reflect viewsabout underlying companies, unfunded exposures, and managers, while at the same time challengingNAV assumptions, re-computing likely carried interest incentives, considering penalties from unwantedunfunded commitments, and incorporating fund-level assets and liabilities. At first glance, then, whyshouldn’t LPs just conduct traditional regressions on the observed return streams from the secondary market?

Unfortunately, several challenges persist. The information is not well-observed by everyone. While themarket has evolved considerably in breadth and depth in the past decade, it remains essentially a privatemarket between an increasing number of sellers and a modest number of buyers. The clearing priceitself can be heavily influenced by the particular buyer-seller dynamics, and may not necessarily be apurely fundamentally derived valuation. Importantly, transaction frictions—including the need for GPapproval on transfers, as well as certain regulatory limitations on volumes—further limit the depth ofthe market’s liquidity. Directionally, the continuing evolution of this market suggests a promising sourceof future intelligence on private equity risk (as well as potential hedging and other risk managementsolutions). But at present, insights on private equity volatility cannot easily be unlocked from thesecondary market.

If we are willing to approximate, however, risk measures and methodologies from the public marketsmight still offer a partial solution, and helpful triangulation on our component-level conclusions. The great dispersion of returns in private equity, driven by the unique capabilities of individual managers,means that a returns-oriented model that tries to map to the public equity markets will be imprecise,relying on the expected return from some subjective “manager alpha”, plus the expected return derivedfrom the public markets. But if we instead focus on the volatility around the outcomes and material shiftsin risk concentrations, then more convenient parallels to the public markets may be drawn, particularlyas the number of companies grows in the private equity portfolio (reducing the effects of company-specificor manager-specific volatility). For example, it might be the case that the risks impacting the privatecompany—supply and demand pressures, customer behavior, raw material price fluctuations, competitorpricing pressures, among others—are no different than those affecting similarly-situated public companies.


17 While the examination of upside, base, and downside cases illuminates specific potential outcomes and concentrations, the day-to-day examination ofportfolio risk must take into account the low probability of specific events and provide a framework that can extend across an infinite plane of probabilities.This expands the discussion from discrete scenarios, which identify and measure idiosyncratic “tail risks”, to continuous-probability distributions thatmay be better able to represent the “average” or “central moment” risks a portfolio faces. The combined application of these approaches provides aunique window through which risk may be observed and measured.

18 (Kojima, 2005)




Perhaps it is then fair to reason that a portfolio of private companies in the same industry and regionbehave similarly to a portfolio of public companies in the same industry and region. If these are sensibleassumptions, then it might also be sensible to use a collection of these public companies to model therisks of the corresponding private portfolio.

This is not to say that using a comparable pool as a proxy is easy to do. A number of complicationspersist. First, a large private equity portfolio reflects different strategies and different companycharacteristics, and so is likely to be best represented by not just one large proxy, but a few smallerbuckets containing appropriate collections of specific public comparables. Risk managers might considerthe current NAVs of the unrealized companies in these sub-buckets as an estimate for the current weightingof each sub-bucket. Second, the behavior of the pool should be adjusted to reflect the higher leveragein private-equity backed companies, through a beta factor that amplifies the returns of the publiccomparables.19 Third, some thought should be given to estimating how much of the pool’s risk is simplynot well-modeled by this approach. This excess volatility of the private equity portfolio relative to thepublic comparables—the “tracking error” of the volatility model—might be informed by subjectivefactors and some common-sense reflections on how well the comparable pool’s volatility really mapsto the private equity target’s volatility.20

WealthCreation

Time

Performance across private equity managers variessignificantly…

…and is affected by both the volatility of returnsand variability of manager skill.

Risk dueto volatility

Risk due to manager selection

Alpha

Beta

Manager BManager B Risk dueto volatility

Manager AManager A

19 See (Goldman Sachs Asset Management, 2005) for detail on how to estimate leverage beta exposure and an approach to estimating the residual volatilityof a private equity portfolio. An alternative approach for beta estimation would be to estimate beta by taking the returns of a portfolio of realized privatecompanies and finding the beta factor which minimizes the difference between the realized returns of the private companies and the beta-adjustedreturns of the portfolio of public companies. This approach, however, requires a significant amount of historical private company data, which may not beavailable to all investors. An important consideration in calculating beta for highly leveraged companies is the fact that beta increases as equity valuediminishes— while the change in beta for a public company may be relatively small for variations in equity value, for a private company financed with70% debt that has experienced a 20% decline in enterprise value, the change in beta may be significant. In addition, this beta factor might also reflectdifferences in size or growth rates between the private and public businesses, which would require additional analysis for calibration.

20 This residual volatility, or tracking error, would be greater if the public pool’s risks were a poor proxy for the private portfolio’s risks. It is also influenced by things like diversification and leverage factors. Importantly, this residual is not the same thing as the risk of incorrectly choosing the right manager. It is instead a measure of the extent to which a given manager’s (or group of managers’) behavior differs from the behavior captured by the market proxy.The risk of choosing poorly (that is, the manager-selection risk) is really outside of this volatility framework, since the volatility inquiry really begins after the manager selection happens. One can think of the public-private framework as a model that generates a stream of returns (with correspondingvolatility) for a selected set of investments. Alternative manager-selection scenarios are really akin to generating different return streams for differentinvestments (each with their own corresponding volatility), thus taking a different vector along another dimension. Volatility is the risk of generatingreturns measured over time, while manager selection incorporates risks measured over alternative scenarios, with different collections of managers.This is illustrated in the figure below:




Exhibit 17 – With company-level detail, private equity portfolios may be approximated by public market proxies

For illustrative purposes only.

With these ingredients in place, however, investors can bring together the pieces and develop a roughestimate of historical volatility, trends in risk concentrations, and potential implications through scenarioanalysis and stress tests. With each sub-bucket appropriately defined and its beta to various marketfactors estimated, an abstracted expression for the portfolio’s exposure to market factors can be derived,as illustrated above in Exhibit 17. Sensitivity analysis around estimated beta factors and residualvolatility will provide additional color on the impact of private equity on the broader portfolio.

To be sure, this entire exercise is as much art as science, and certain challenges remain. For example,too broad a grouping of strategies may result in the public comparables serving as ineffective proxies,and too narrow a grouping may result in too little diversification within sub-buckets, causing possiblecompany-specific distortions. The analysis also fails to consider the considerable risks in the remainingunfunded commitments. Indeed, making the assumption that these commitments should be funded (at par)is implicitly assuming the return and volatility on these future commitments is no different than originalexpectations. Additionally, many of the critical variables in the modeling reflect considerable subjectivity(the comparable pools, or the beta factors) or have acknowledged accuracy issues (the NAV weightings).But a model is only meant to be a simplified representation. Whether derived from a component-level,micro-oriented assessment (Steps #1, #2 and #3 above) or from this proxy-based macro-oriented assessment,this exercise of identifying and mapping the drivers of risk and return aims to deepen the investor’sunderstanding of the stand-alone private equity portfolio.

It may also offer a further benefit and provide a mechanism for the private equity analysis to communicatewith similar analyses happening elsewhere. Of course, this may be easier said than done. A unifieddashboard across all asset categories is required for CIOs to assimilate, integrate, and interpret variouscommon factors.21 But with the right technology and tools, investors can imagine a common frameworkthat examines new questions. In what ways does the private equity portfolio increase or decrease overallexposure to equity markets, long-term interest rates, credit spreads, currencies, or commodity prices?

private equity public equity additional variation

21 (Global Portfolio Solutions, 2010)

Additional risk not captured by markets

Unobservable private equity returns

Observable public proxy risk-adjusted returns, combined with beta factors for leverage and size

Private equity shares risk factors with public equity

Proxy built with public equity chosen to behave similarly with private companies

Does the portfolio protect against inflation? How does its illiquidity impact the dynamics with otherasset classes? Does the private equity program complement or conflict with broad macro-perspectives?Can private equity be incorporated within broader tactical tilts? These questions are exceptionallydifficult to address if each investment activity is unconnected, isolated within its own vertical asset classsilo. But broader portfolio-wide concentrations, correlations, and risks reveal themselves when such acommon analytical framework is established and traditional asset classes are penetrated. To be sure,any precise quantification of factor loadings is perhaps unrealistic. The broader goal is to identifyexposures that could materially impact portfolio value. While there will be some false positives that mayrequire further investigation, this approach should help develop a better understanding of cross-assetclass portfolio risk, validate the implementation of specific views, prompt analysis of new and differentfactors, and suggest occasional changes to the portfolio.

Step #5: Stepping Toward Active Management

What more can the LP do, beyond ensuring that its intelligence is as accurate, its understanding as deep,its appreciation for outcomes as broad as possible? Certain LPs may take affirmative steps to actuallychange their risk exposure. At a minimum, all material new decisions associated with the private equityprogram—new commitments, asset-allocation adjustments, co-investments—should be evaluatedthrough the prism of a factor-based multi-disciplinary approach, with the incremental impact of thesedecisions clearly understood. In some cases, LPs might discover that they already have plenty of whatthey are looking for—increasing credit exposure, for example, might at first seem necessary, but afterlooking past conventional labels, the existing portfolio might turn out to have credit instruments inmiddle-market buyout funds, pools of bank loans in macro hedge funds, and a variety of other exposuresthat already correlate with the credit-oriented thesis. In other cases, LPs might discover that change isalready happening in their private equity portfolios. Since ongoing capital calls and distributions willcontinuously change the playing field, a regularly refreshed factor-based analysis will be required inany event. The risk management optimist might even conclude that with both markets and managerseffecting changes, further action might be unnecessary.

Exhibit 18 – Risk factors must be analyzed not only within the private equity allocation, but also across all assetsand obligations of the portfolio







But this is doubtful. As noted earlier, GPs are managing their funds’ risks, not their LPs’ program-wideconcerns, and any comfortable equilibrium arising from the markets’ and managers’ behavior is likelyto be a happy coincidence. A “commit and hope” strategy is not really a risk management strategy,and simply approving or disapproving new additions to the portfolio—however well-reasoned andwell-informed—is likely an awfully crude risk management approach. Even if effective over the longterm, such a passive approach to private equity management may deny the LPs the tactical flexibility totake advantage of unique moments when a short-term shift in asset allocation is advised, when capitalcan be better deployed elsewhere, or when safety is worth more than the status quo.

Private equity LPs should see their role as extending far beyond reactive responses. Active managementmay take several forms. Armed with a factor-based approach and penetrating perspective into theportfolio, CIOs and others looking across the entire playing field can decide where best to correctimbalances that naturally evolve, or those resulting from sudden shocks. It might very well involvechanges to the private equity program. But such changes might also be more efficient—in time,transaction costs, or effectiveness—if they occur in more liquid places elsewhere in the portfolio. For example, a discovery that recent commitments to energy-related private equity funds haveunexpectedly elevated exposure to commodity prices might best be mitigated by adjusting similarlycorrelated public market exposures, by reducing direct commodity exposure, or by implementing hedgesor other risk-overlays. Meaningful shifts in dollar-euro rates may give rise to imbalances in existingcompany exposures as well as change the funding plans for future unfunded commitments; addressingthese imbalances might best be solved by currency hedges or direct adjustments to euro-denominatedpublic equity or fixed income investments.

Other active management steps might be more process-oriented. For example, they may involve revisitingcustomary terms in the LPA, refining key-person provisions, tightening opportunistic investing buckets,or better defining other places of unusual GP flexibility. Or they may be as basic as refining the approachto manager diligence, adding a layer of operational diligence on the trading and clearing functions of a distressed-oriented manager whose credit strategies more closely resemble a hedge fund than aprivate equity fund.22 They may also take the form of a more active engagement with the GP, not onlyarticulating views when solicited in the advisory committee, but also expressing these opinions morestrongly as pre-conditions to future re-commitments.23 An explicit pre-commitment to a successor fund,made early in the fundraising process and with customized economic arrangements, might be sensiblein furthering the LP’s near-term and long-term agenda. Such LPs might consider a more concentratedstrategy, actively managing their unfunded and manager risks by focusing on fewer managers withlarger commitments, and mitigating any corresponding increase in company exposures with the enhancedcompany-level analytical tools discussed earlier.

22 See (Goldman Sachs Asset Management, 2010)23 Equally interesting can be an assessment of the stability of other current LPs. While not subject to the redemption (or “run on the fund”) risks experienced

by hedge funds, private equity LPs must still make affirmative ongoing decisions, whether in funding capital calls, approving fund extensions, authorizingvaluations, conflicts, and other advisory board matters. The capital-call compliance and voting records of informed LPs may provide further evidence ofmanager stability, and can also contribute to the risk assessment.



Even more activist and impactful are those initiatives that immediately re-define or re-shape theportfolio’s risks. For the over-allocated LP, a secondary-market transaction that streamlines non-coremanagers or trims unwanted concentrations may be a sensible alternative. For the under-allocated LP,selective acquisitions in the secondary market—at the right price—might deliver a more immediate andefficient solution. Such active management decisions are nothing new, and are not without complication(including assessing a fair ask price, securing approval from the GP to transfer the interest, andcomplying with important tax and other regulatory considerations). But the secondary market hasevolved considerably over the last decade. Today, it represents an effective mechanism for immediateportfolio rebalancing, when accessed properly. In recent years, the market has also originated morecreative solutions—splitting funded and unfunded exposures to eliminate unwanted manager risk,swaps to convert uncertain future distributions into more bond-like income streams, synthetic structuresthat share upside outcomes while truncating left-tail outcomes. Of course, the greater the ambition,the greater the complexity. But LPs and CIOs need to be aware of the possibilities as they weigh thebenefits of a private equity program alongside the risks.




IV. ConclusionNot unlike the challenge of predicting the top contenders for the next World Series, Super Bowl, orWorld Cup, risk management in private equity is a complicated undertaking. Operational and marketuncertainties add to company-specific risks, which in turn aggregate to fund-level risks, which in turncontribute to manager-level risks, which all factor into the program-wide risks of any private equityportfolio. And these risks themselves are all evolving, accentuated by the current economic environment.The consequences in getting them wrong are surely increasing.

But so are the opportunities from successfully managing them. Ultimately, no risk management systemwill be effective without the right organizational leadership to support it. Effective reporting lines,multi-dimensional internal incentives, procedural checks-and-balances, a focus on transparency, and a commitment to continual evolution are all necessary ingredients, emanating from the highest levelsof the organization. Within the right risk management culture—and armed with the proper tools, resourcesand infrastructure—LPs may be equipped not only to better understand and manage their existingprivate equity exposures, but also to collaborate with GPs in delivering better solutions.

This invites a number of new questions. What if the LP prefers the manager’s healthcare-industrycapabilities, but is less interested in its energy-industry capabilities? Armed with an ability to managecompany-level exposures, such an LP could perhaps partner with versatile GPs to design a bespokedelivery mechanism, far superior to a traditional one-size-fits-all fund structure. How should the LP looking to tailor its particular sensitivities—whether j-curve mitigation, preference for specific sub-sectors, cost concerns, innovative governance, or alignment of incentives—distinguish the long-established GP raising its next multi-billion dollar fund from a newly formed GP starting its first?Insights gained in assessing fund-level and manager-level risks can help shape the LPA negotiation,cutting through precedent, and focusing on first principles and case-specifics. Can the private equityprogram itself be made more nimble and effect tactical changes to exposures, even after initial commitmentshave been made? The secondary market, perhaps in collaboration with the GPs, can evolve further to empower such dynamic risk management.

What all of this really suggests is a broader possibility: can a defense-oriented risk management effortin private equity also serve as an offense-oriented alpha generator? These are exciting possibilities to contemplate. The promise of private equity—material, and perhaps consistent, outperformance tothe public markets—is not without its risks. And even the most robust analytical framework cannotoperate without experience and judgment. But if these risks can be identified, managed, and perhapseven harnessed, investors in private equity might not just know better where they are going—they mightend up getting there faster (and safer) than originally planned.





The Alternative Investments & Manager Selection (AIMS) Group of Goldman Sachs provides investorswith diversified and customized portfolio solutions. With over 200 professionals around the worlddrawing on Goldman Sachs’ global sourcing network, due diligence capabilities, risk managementexpertise, and extensive manager relationships, the AIMS Group sources, evaluates, and invests inmanagers across traditional long-only, hedge fund, and private equity strategies. To date, the investmentstrategies of the AIMS Group represent more than $70 billion of capital commitments across long-only strategies, hedge fund portfolios, private equity fund-of-funds, secondary-market investmentsand co-investments.24

24 As of September 2010.




References“A Practical Guide to Managing Private Equity Commitments.” Goldman Sachs Asset Management.

June 2007.

“Active Risk Budgeting in Action: Assessing Risk and Return in Private Equity.” Goldman SachsAsset Management. August 2005.

“An Evolving Landscape: Operational Due Diligence and Investing in Hedge Funds.” Goldman SachsAsset Management. October 2010.

Dimson, Elroy, Paul Marsh & Mike Staunton. “Global Investment Returns Yearbook 2005.” ABNAMRO/London Business School. February 2005.

“Houston, We (May) Have a Problem.” Global Portfolio Solutions Perspectives. Goldman SachsAsset Management. Fourth Quarter 2010.

Kaplan, Steven and Antoinette Schoar. “Private Equity Performance: Returns, Persistence and CapitalFlows.” Journal of Finance 60 (August 2005).

Kojima, J. Christopher. “Liquidity Solutions and the Secondary Market in Private Equity.” CFA Institute Conference Proceedings Quarterly. March 2005, Vol. 2005, No. 2, pp. 39-45.

Lerner, Josh, Morten Sørensen, Per Strömberg. “Private Equity and Long-Run Investment: The Caseof Innovation.” NBER Working Papers 14623. National Bureau of Economic Research, Inc. 2008.

“Private Equity, Public Exits.” Ernst & Young. October 2010.

Thomas, Jason M. “The Credit Performance of Private Equity-Backed Companies in the ‘Great Recession’of 2008-2009.” Private Equity Council. 2010


DisclosuresThis material is provided for educational purposes only and should not be construed as investment advice or an offer or solicitation to buy or sell securities.There may be conflicts of interest relating to the Alternative Investment and its service providers, including Goldman Sachs and its affiliates, who are engaged in businesses and have interestsother than that of managing, distributing and otherwise providing services to the Alternative Investment. These activities and interests include potential multiple advisory, transactional andfinancial and other interests in securities and instruments that may be purchased or sold by the Alternative Investment, or in other investment vehicles that may purchase or sell such securitiesand instruments. These are considerations of which investors in the Alternative Investment should be aware. Additional information relating to these conflicts is set forth in the offering materialsfor the Alternative Investment.This material has been prepared by GSAM and is not a product of the Goldman Sachs Global Investment Research (GIR) Department. The views and opinions expressed may differ from thoseof the GIR Department or other departments or divisions of Goldman Sachs and its affiliates. Investors are urged to consult with their financial advisors before buying or selling any securities.This information may not be current and GSAM has no obligation to provide any updates or changes.Alternative Investments such as private equity funds are subject to less regulation than other types of pooled investment vehicles such as mutual funds, may make speculative investments,may be illiquid and can involve a significant use of leverage, making them substantially riskier than the other investments. An Alternative Investment Fund may incur high fees and expenseswhich would offset trading profits. Alternative Investment Funds are not required to provide periodic pricing or valuation information to investors. The Manager of an Alternative InvestmentFund has total investment discretion over the investments of the Fund and the use of a single advisor applying generally similar trading programs could mean a lack of diversification, andconsequentially, higher risk. Investors may have limited rights with respect to their investments, including limited voting rights and participation in the management of the Fund.Alternative Investments by their nature, involve a substantial degree of risk, including the risk of total loss of an investor’s capital. Fund performance can be volatile. There may be conflictsof interest between the Alternative Investment Fund and other service providers, including the investment manager and sponsor of the Alternative Investment. Similarly, interests in anAlternative Investment are highly illiquid and generally are not transferable without the consent of the sponsor, and applicable securities and tax laws will limit transfers.Private Equity investments are speculative, involve a high degree of risk and have high fees and expenses that could reduce returns; they are, therefore, intended for long-term investors whocan accept such risks. The ability of the underlying fund to achieve its targets depends upon a variety of factors, not the least of which are political, public market and economic conditions.The trading market for the securities of any portfolio investment of the underlying funds may not be sufficiently liquid to enable such funds to sell such securities when it believes it is mostadvantageous to do so, or without adversely affecting the stock price. In addition, such portfolio companies may be highly leveraged, which leverage could have significant adverse consequencesto these companies and the funds offered by the GS Private Equity Group (“PEG funds”). Furthermore, restrictions on transferring interests in the PEG funds may exist. Volatility in political,market or economic conditions, including an outbreak or escalation of major hostilities, terrorist actions or other significant national or international calamities could have a material adverseeffect upon the PEG fund and its portfolio investments. The possibility of partial or total loss of PEG fund capital exists, and prospective investors should not subscribe unless they can readilybear the consequences of such loss.Although certain information has been obtained from sources believed to be reliable, we do not guarantee its accuracy, completeness or fairness. We have relied upon and assumed withoutindependent verification, the accuracy and completeness of all information available from public sources.The portfolio risk management process includes an effort to monitor and manage risk, but does not imply low risk.Past performance is not indicative of future results, which may vary. The value of investments and the income derived from investments can go down as well as up. Future returns are notguaranteed, and a loss of principal may occur.Opinions expressed are current opinions as of the date appearing in this material only. No part of this material may be (i) copied, photocopied or duplicated in any form, by any means, or (ii) distributed to any person that is not an employee, officer, director, or authorized agent of the recipient, without GSAM’s prior written consent.This material has been communicated in the United Kingdom by Goldman Sachs Asset Management International which is authorized and regulated by the Financial Services Authority (FSA).This material has been issued or approved for use in or from Hong Kong by Goldman Sachs (Asia) L.L.C.This material has been issued or approved for use in or from Singapore by Goldman Sachs (Singapore) Pte. (Company Number: 198602165W).

Copyright © 2011, Goldman, Sachs & Co. All Rights Reserved. 43453.OTHER, 44266.OTHER, 44267.OTHER USIHCB/01-11

Risk Mgmt in PE-GSAM Jan11

Documents