AQR Whitepaper 1Q19

Demystifying
Illiquid Assets:
Expected Returns for
Private Equity
Executive Summary
The growing interest in private is based on a discounted cashflow
equity means that allocators must framework similar to what we use for
carefully evaluate its risk and return. public stocks and bonds.
The challenge is that modeling
private equity is not straightforward In particular, we attempt to
due to a lack of good quality data and assess private equity’s realized
artificially smooth returns. We try and estimated expected return
to demystify the subject, considering edges over lower-cost public equity
theoretical arguments, historical counterparts. Our estimates display
average returns, and forward- a decreasing trend over time, which
looking analysis. For institutional does not seem to have slowed the
investors trying to calibrate their institutional demand for private
asset allocation decisions for private equity. We conjecture that this is
equity, we lay out a framework due to investors’ preference for the
for expected returns, albeit one return-smoothing properties of
hampered by data limitations, that illiquid assets in general.

Antti Ilmanen, Ph.D.

We thank Daniel Villalon, Thomas Maloney, Peter Hecht, Scott Richardson,
Principal
Marina Niessner, and Tobias Moskowitz for helpful comments and suggestions,
Portfolio Solutions
and Daniel Rasmussen for data.
Group

Swati Chandra, CFA

Vice President

Nicholas McQuinn
Associate
Contents
Table of Contents 2

Introduction 1

Frameworks for Expected Returns 2

Conclusion 11

Appendix: Assumptions 12

References 17

Disclosures 19

Table of Contents
 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19 1

Introduction

As investors increasingly embrace private We posit that this surprising outcome reflects
equity (PE), they find themselves posing the illusion of the lower risk of illiquid assets
the following questions: How much should or the appeal of their artificially smooth return
they allocate? What are good yardsticks for streams. Due to the absence of mark-to-market
assessing performance? Are the higher fees of accounting, the reported volatility and equity
PE justified by higher expected returns over beta of private assets tend to be understated
public equity counterparts? What is the risk unless one desmooths their returns, which
and diversification potential of PE? may not be a clear-cut exercise. This overstates
their diversification potential or naïvely
The comparison to public equity is not measured alpha.1 Even if one expected PE to
straightforward. In general, illiquid assets provide zero excess return over public equity,
are inherently harder to model, and this is the assumption that PE was less risky, and
exacerbated by a lack of good quality and lowly correlated to public equity, would call for
transparent data. We try to demystify the an increased allocation to PE. Furthermore,
subject of PE risk and return, focusing on the understating the reported risk compared to
medium-term expected return (ER) of PE. We economic risk may in itself appeal to investors.
view the topic from multiple lenses: theoretical The shrinking valuation gap between private
required returns, historical performance, and and public equity, which we show later on, is
finally our favored approach of extending one indication of investor willingness to pay,
our discounted-cashflow-based (DCF) perhaps knowingly overpay, for these return-
methodology for equity and fixed income to smoothing characteristics.
the realm of PE. A common framework helps
highlight how the ER of PE is anchored to that This report is targeted at investors interested
of public equity by similar drivers — say, yield in understanding the relation between private
and growth. While we focus on returns, our and public equity at some depth, and it serves
analysis also touches on the hidden risks and as a background piece for the readers of our
factor exposures of PE, and thereby suggests annual Capital Market Assumptions edition of
potentially better performance benchmarks AQR Alternative Thinking. Our 2019 edition
and comparisons to public equity. will now include a brief section on illiquid
asset classes which may suffice for most
We observe that PE has grown in popularity readers, but those wanting more detail can
despite a decreasing expected and realized refer to this report.2
return edge over public equity counterparts.

1  For example, if one expects PE to deliver a 3% excess return over public equity and the equity risk premium to be 5%, the implied alpha
of PE is 0.5% if PE’s beta to public equity is deemed to be 1.5, and 5.5% if using a beta of 0.5 based on the artificially smooth PE
return series.
2  A companion piece Ilmanen et al. (2019) discusses another illiquid asset class, real estate.
2 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19

Frameworks for Expected Returns

We try to estimate the medium-term real ER for be more useful and informative than the point
private equity, focusing on its edge over public estimates themselves.
equity. Specifically, our estimates are for the
Theory
largest segment of the private equity market,
U.S. buyouts.3 Our expected return estimates Theoretical or risk-based explanations of
are net of fees, as fees can be a substantial asset returns follow the premise that higher
component of returns for illiquid assets.4 return is the compensation investors require
for taking on additional risk. So, if PE has
We approach the topic through three greater exposure than public equity to certain
complementary perspectives as described in risk factors, and if those risk factors have
Ilmanen (2011): theoretical required returns, positive expected returns, one would expect
historical evidence on past average returns, it to have both higher risk and return. Based
and yield-based analysis that considers current on economic intuition and empirical evidence
valuations and market conditions. that we describe later, we expect PE to have the
following factor tilts over public equities: equity
As is the norm with other asset classes, we risk, (il)liquidity premium, size, and value:
present real (inflation-adjusted) compound
rates of return for the asset class as a whole • Equity Risk: The principles of corporate
for a horizon of 5 to 10 years. Over such finance dictate that all else equal,
intermediate horizons, initial market yields companies with greater debt-to-equity
and valuations tend to be the most important (D/E) should have higher volatility and
inputs. For multi-decade forecast horizons, the equity beta, as the required interest
impact of starting yields is diluted, so theory payment to debt holders increases the
and long-term historical average returns (or riskiness of the remaining cashflow to
yields) may matter more in judging expected equity holders. Studies indicate that PE
returns. Our estimates are intended to assist firms take on 100-200% debt for every
investors with their strategic allocation and dollar of equity (down from the 300-400%
planning decisions, and, in particular, with D/E ratios in the 1980s), whereas publicly
setting appropriate medium-term expectations. listed firms, on average, add 50% of debt
They are highly uncertain, and not intended for every dollar of equity.5 This suggests
for market timing. The broad framework may PE’s equity beta is well above 1.6

3  The terminology is mixed, as private equity could either refer to buyouts only, or include other segments like venture capital. We use the
term to refer to buyouts only. Buyouts account for over 60% of the aggregate private equity market, as per market segment estimates
by Døskeland and Strömberg (2018), based on Preqin data on funds raised 2012-17.
4 Gross expected returns could be 5 to 7% higher. See Døskeland and Strömberg (2018).
5 See Døskeland and Strömberg (2018). Further, Axelson et al. (2013) find median debt-to-enterprise value ratios of 70% for LBOs and
35% for public industry-and-region matched companies, implying D/E ratios of 233% for LBOs and 54% for the public match. As of
June 30 2018, Bloomberg estimates of the weighted average D/E ratio for the S&P 500 and Russell 2000 are in-line, at 45% and
49%, respectively.
6 As per Modigliani and Miller (1958), in general, rl = ru + (D/E) (ru – rd) (1-Tc) where rl = required rate of return on levered equity, ru=
unlevered cost of equity (or the return on assets) for an all-equity-financed firm, rd = cost of debt, and Tc is the tax rate. This implies βl
= β u + (D/E) ( β u – β d) (1-Tc). Thus a levered firm (a firm with debt) has a higher required equity return and expected equity beta than an
unlevered firm if D>0 and rd < ru.
 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19 3

Why then is PE vaunted for its diversification increased investor demand for PE drives
benefits? The answer lies in the lack of regular, up the purchase multiples of buyout
mark-to-market pricing for illiquid assets in targets, one may expect their future
general. This induces the common practice returns to be lower.
of appraisal-based or self-reported NAVs that
do not reflect the daily fluctuations in public • Size: Buyout targets tend to have smaller
markets, making for artificially smoothed capitalizations and therefore provide
returns that understate risk and correlation to exposure to the size factor.10 This implies
public markets. Naïve regression-based equity a more appropriate benchmark would be a
beta estimates tend to be below 1.0, even if leveraged small-cap index that accounts for
adjusting for illiquidity by using lagged betas. 7
both the higher leverage and small-cap bias.
Many empirical studies conclude that a beta The small-cap tilt is evident from holdings-
estimate of 1.2-1.5 is more realistic, implying based analyses that present the firm
PE has higher volatility and lower risk-adjusted characteristics of typical buyout targets,
returns than naïve reported returns suggest.8 but even this basic fact is hard to confirm
with returns-based regressions, due to the
In short, while PE has low reported risk, it is artificial smoothness of PE returns.11
economically riskier and has higher exposure to
the equity risk premium than public equities, • Value: Over and above a small-cap bias,
a combination that many investors may find buyout targets have tended to trade
appealing. The full risk of PE is most likely to at lower valuation multiples than the
materialize in prolonged bear markets, not in market, though venture capital targets
relatively fast ones like 2008-9. are more likely to be growth companies.
While Stafford (2017) and Chingono and
• (Il)liquidity Premium: In principle, Rasmussen (2015) report a value tilt,
locking up capital for a 5-10 year window broader evidence is mixed. Further, as
warrants a significant illiquidity premium, we show later, the PE industry overall no
as suggested by Ang (2014). However, 9
longer has the valuation discount versus
as we argue later, the data suggests that public equities it used to have. While this
even if such a “fair” illiquidity premium may be partly due to a changing industry
existed, it may in practice be largely offset composition of buyout targets, it is unclear
by investor willingness to overpay for the whether PE’s historical value bias will
return-smoothing described earlier. If persist.

7  Desmoothing illiquid asset returns using a simple auto-regressive AR1 variant may not suffice. Anson (2017) finds that PE lagged
betas are significant up to three quarters back, while Real Estate lagged betas are significant up to four quarters back.
8  See Døskeland and Strömberg (2018) which summarizes betas and risk factor loadings across several papers, datasets and
methodologies. All else equal, higher leverage tends to increase beta, while performance fees dampen net-of-fee returns above the
hurdle rate, thus lowering beta.
9  Ang (2014) discusses a model which suggests investors should require a 4-6% illiquidity premium to lock up their capital for 5-10
years. However, broad evidence on realized illiquidity premia in many asset classes is mixed.
10 Banz (1981) shows that empirically, small-cap stocks have earned higher returns than large-cap stocks. The size premium is much-
debated and may not be as robust as other factor premia, as discussed in Alquist et al. (2018).
11 See L’Her et al. (2016) who find that the average size of LBOs is very small in comparison to listed small-cap equities.
4 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19

Crudely assuming that the fair illiquidity Historical Performance

premium of PE gets fully offset by the
overpayment for smooth returns, investors Now that we’ve outlined the economic
may still require and expect a higher return rationale for the expected excess return of
from PE than public equity due to its higher PE over public equities, what does historical
equity beta and small-cap bias. This could evidence have to say?12 Exhibit 1 compares
thus help inform a public proxy approach PE’s historical performance to various publicly
for a minimum required return for PE. As traded benchmarks as well as to baskets of
a purely illustrative example, if investors stocks that account for PE’s factor exposures.
assumed PE had a 1.2 beta and no net-of-
fee alpha to public small-cap equities, and For the reasons mentioned earlier, comparing
expected small-caps to return 5% excess PE reported returns directly to large-cap
of cash (this includes both the equity risk equities is not a fair measure of alpha or the
premium and the size premium), the implied illiquidity premium. Leveraged, small-cap
PE net-of-fee expected excess return over cash indices are more appropriate as benchmarks.
would be 1.2 times 5%, or 6%. Exhibit 1 shows that over the period 1986

Exhibit 1
Historical Performance of Private Equity: Scant Illiquidity Premium
July 1, 1986 – December 31, 2017

Cambridge Small Value

Cambridge Private Equity 1.2x Russell Stocks
Private Equity (U.S. Buyout) Russell Russell 2000 (Fama
(U.S. Buyout) Desmoothed S&P 500 2000 2000 Value French)

Average Return
9.9% 9.9% 7.5% 7.6% 9.1% 8.5% 11.4%
(Arithmetic)

Excess Return over

2.3% 2.3% 0.7% 1.4% -1.6%
Public (Arithmetic)

Average Return
9.8% 9.2% 6.4% 5.5% 6.0% 6.7% 9.4%
(Geometric)

Excess Return over

3.4% 4.3% 3.7% 3.1% 0.4%
Public (Geometric)

Volatility 9.3% 13.8% 15.8% 20.7% 24.9% 19.4% 21.4%

Source: AQR, Bloomberg, Cambridge Associates (using internal-rate-of-return (IRR)-based raw index returns and an AR(1)-desmoothed
variant), Kenneth French Data Library. PE returns are based on pooled horizon IRRs, net of fees, expenses, and carried interest. Public index
returns are gross of fees and of trading costs. Excess return over public refers to the raw Cambridge PE return in excess of each of the public
market indices to the right. For illustrative purposes only and not representative of any portfolio or strategy that AQR currently manages.

12 We stress that the question of how much excess return over public equities investors require from PE (to compensate them for the
greater illiquidity and risks associated with PE) is distinct from the question of how PE firms can generate those excess returns, over
and above covering their high fees. The various ways PE firms meet this high return hurdle are described in Kaplan and Strömberg
(2009) and Døskeland and Strömberg (2018). PE firms may be able to add significant value through prudent selection of buyout
targets; opportunistic timing; as well as operational, financial, and governance engineering that improve the efficiency and growth
prospects of the companies they hold.
 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19 5

to 2017, PE outperformed large-caps by In Exhibit 1, we show internal-rate-of-return

2.3%, looking at arithmetic means (AM). (IRR)-based returns for PE because they are
But when compared to a 1.2x leveraged commonly used. But we caveat that IRRs for
small-cap index, this falls to just 0.7%, and individual managers are notorious for their
PE actually underperformed a basket of “gameability”.17 A better metric of relative
small-cap value stocks by 1.6%.13 This is performance is the “public market equivalent”
corroborated by Stafford (2017) who finds (PME) that is strongly preferred by academics
that the long-run average excess returns (see Kaplan and Schoar (2005)). The PME
of PE over public equity can be matched approach involves comparing the amount
by a leveraged, small-cap value strategy. of capital generated by a PE strategy to that
Thus, it appears that the PE industry, generated by a public market index (the
on average, has offered scant illiquidity benchmark) over the lifespan of the fund,
premium beyond these typical factor tilts.14 assuming similar amounts were invested with
the same timing.
Nevertheless, for many investors, the bottom
line is that PE firms have delivered clearly Irrespective of whether one uses IRRs or
higher net-of-fee returns than the S&P 500 PMEs, the choice of benchmark is critical. For
over the past 30 years even if those excess example, Harris et al. (2014) find a long-run
returns could be largely accounted for by average PME of roughly 1.2 versus the S&P
using more representative publicly traded 500. A PME of 1.2 implies 20% outperformance
benchmarks. Further, top-quartile managers by PE over the period capital is deployed.
would have served end-investors (LPs) Assuming a typical investment period of
better than the industry average results.15 six years, that implies PE has outperformed
PE managers’ exceptional skill becomes the S&P 500 by 3.1% annually, net-of-fees.
even more evident when we consider their However, L’Her et al. (2016) find that the long-
performance before fees, which are estimated run average PME shrinks to 1 (implying no PE
at around 6% per year. 16
outperformance) when benchmarked against a
leveraged, small-cap index.

13 The 1.2x reflects the market beta of PE versus a small-cap index. Both Driessen et al. (2012) and Franzoni et al. (2012) find a beta of
1.3 against a broad stock index, implying a beta of 1.2 versus a small-cap index. Thus this 1.2x small-cap proxy can also be viewed as
the normative PE required return based on PE’s exposure to the equity premium and the size factor, or the PE expected return using a
public proxy approach.
14 PE would look better if using geometric means (GMs) or public equity returns net of trading costs. In any case, the PE index returns in
Exhibit 1 are IRRs that are not directly comparable to equity total returns.
15 That said, it may be too common for end-investors to assume that they can get top-quartile managers (which may be yet another
reason for the popularity of PE). Such overconfidence may be boosted by the fact that, as described by Harris et al. (2012), about
half of PE funds describe themselves as top quartile. The flexibility that PE funds have in slicing and dicing the data — comparison
universes and time periods — makes this feat possible.
16 It is not straightforward to translate typical PE fund fees of 2% management fee, 20% carry, a hurdle rate and additional portfolio
company fees into a fixed yearly fee. Døskeland and Strömberg (2018) cite a McKinsey (2017) CEM Benchmarking study among large
institutional investors which estimates total fees to be 5.7% p.a. comprising 2.7% in management fees, 1.9% in carried interest
(performance fee), and 1.2% for other fees, including net portfolio company fees.
17 Unlike most index returns, IRRs are not time-weighted and are affected both by the magnitude and timing of cash-flows. Larger
cashflows have a greater effect on IRRs, and IRR calculations embed a non-innocuous assumption that interim cashflows can be
reinvested at the IRR. Thus, PE GPs can time capital calls from LPs as well as deal exits so as to boost IRRs.
6 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19

Exhibit 2 plots PMEs for various vintages, show that PE fund returns tend to be lower
alongside the valuation gap at deal inception after “hot-vintage” years characterized by high
between public equity and PE. Panel A 18 fundraising activity or capital deployment,
plots the EBITDA/EV, or inverted purchase attractive financing conditions, and easy
multiple, for PE alongside the EBITDA/ leverage.19 Skeptics stress that the current
EV for public equity, and thus depicts the environment can be characterized by low
ex-ante return edge PE may have due to its financing rates coupled with increasing
lower valuations. Panel B plots PMEs for each institutional demand for PE, more PE firms,
vintage-year, with and without adjustments record-high dry powder (committed uncalled
for leverage, size and sector, and thus shows capital), and competition from cash-rich public
the future realized return edge of PE over two companies and sovereign wealth funds. Thus,
public equity benchmarks (roughly for the PE faces headwinds that make it less likely to
next five to six years). We see that as PE has deliver the strong returns it has in the past. Of
grown relatively richer and the valuation course, richness versus history is not unique
gap has narrowed, PE’s outperformance over to PE: as described in our Capital Market
public equities has declined, with realized Assumptions editions of AQR Alternative
outperformance for post-2006 vintages Thinking, many other asset classes appear
dropping to virtually zero (PME near 1), even expensive today, perhaps reflecting the easy
before adjusting for size and leverage. global monetary policies of the 2010s.

The key question then is what net returns end- In contrast to our conservative forecasts,
investors can expect in today’s environment of institutional investors widely expect PE to
tighter valuations and greater competition for outperform public equity by 2-3%.20 Despite
deals. How should we weigh the longer 30-year its recent lack of outperformance, investors
history and the more recent 10-year evidence remain optimistic on PE, even as they
when estimating future PE outperformance? increasingly question the value-add of other
forms of active management. Some reasons
Normally we give greater credence to longer- for this may be the lack of transparency on
run evidence, but two disconcerting trends PE returns and fees, slow learning about
point to overweighting the more recent history. performance, and the use of misspecified
First, Exhibit 2 depicts a shrinking valuation benchmarks. PE returns are often presented
gap between PE and public equities, and PE as IRRs, which can be easy to game and which
outperformance ceased at roughly around the evolve slowly. It’s also plausible that investors
same time as the valuation gap closed. This are cognizant of the points we raise and
corroborates our earlier point that increasing knowingly accept a more modest, even zero,
investor demand may have driven up PE net-of-fee outperformance over public equities
valuations. Second, many academic studies because they find the artificially smoothed
returns of private assets desirable.

18 We caveat that the PME data for most recent vintages may not be fully representative of those vintages, as it is likely based only on the
subset of deals that have been exited (the so-called J-curve effect).
19 Kaplan and Strömberg (2009), Axelson et al. (2013), Harris et al. (2014), Robinson and Sensoy (2015), and L’Her et al. (2016) find that
private equity returns are inversely related to the amount of money flowing into the PE industry as well as GP access to cheap financing.
20 Andonov and Rauh (2018) find that institutional investors extrapolate past performance when setting return expectations, and in
recent years have expected the PE industry to outperform public equity by 2.5% (arithmetic) and 1.5% (geometric) over the long run.
 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19 7

Exhibit 2
The Valuation Gap and Performance Gap between PE and Public Equities
January 1, 1998 – September 30, 2018

Panel A: Ex Ante Valuation Gap

16%

14%

Wide
12% Valuation Gap
between Private
EBITDA / EV

and Public
10%

8%
Narrow
Valuation Gap
6% between Private
and Public

4%
1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

S&P 500 EBITDA/EV PE EBITDA/EV at Purchase

Panel B: Ex Post Performance Gap

1.6

1.4
Public Market Equivalent (PME)

1.2 PE Outperforms at PME > 1.0

1.0

0.8

PE Underperforms at PME < 1.0

0.6

0.4
1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

PME vs. S&P 500 PME vs. Leverage, Size, Sector Adjusted S&P 600

Source: PMEs from L’Her et al. (2016). Vintage years are assigned based on the year of the first investment by a fund. EBITDA/EVs from
2008 to 2018 are calendar-year averages of the median EBITDA/EV from Pitchbook and the average EBITDA/EV from Bain & Company. PE
EBITDA/EV from 1998 to 2008 are a proprietary dataset from Dan Rasmussen, based on data from Cambridge Associates and CapitalIQ.
S&P 500 EBITDA/EV is from Bloomberg.
8 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19

Yield-Based Approach selection; higher earnings growth rates through

operational improvements; multiple expansion
Our third approach to PE ER estimation is through opportunistic timing of entries/exits;
yield-based. Here we apply the discounted and financial leverage. We should expect yields
cashflow (DCF) framework we use to forecast and growth rates for PE to be at least loosely
5-10 year expected returns of public equities anchored to those for public equities.
and bonds in our Capital Market Assumptions
editions of AQR Alternative Thinking. Exhibit 3 illustrates our framework for PE ER.
Admittedly, each of our inputs is debatable First, we estimate unlevered ER ru using the
as data limitations on PE necessitate many DDM: ru ≈ yu+ gu, where yu = dividend yield
simplifying assumptions. Still, the broad and gu = real earnings-per-share growth rate.
framework remains relevant, as it explains Then, we estimate the theoretical required
the mechanism of how PE firms can generate levered return to equity rl by plugging in
higher returns than public equity. PE firms leverage D/E and the cost of debt kd, to which
can employ multiple levers to boost returns: we finally add expected multiple expansion m
namely, higher starting yields through deal to arrive at gross PE ER rg.21

Exhibit 3
Building Blocks for U.S. Private Equity Expected Real Returns

Financial
Unlevered Levered
Leverage

r| = ru +
ru = yu (D/E) * rg = r|
yu gu + gu D/E kd (ru - kd) m +m

Real Multiple Net Exp.

Earnings Growth Real Debt to Real Cost Real Expansion Gross Real
Yield Rate Return Equity of Debt Return (Ann.) Real ER Fees Return

Current 2.1% + 3.0% = 5.1% 109% 1.2% 9.3% + 0.3% = 9.6% - 5.7% = 3.9%

Historical
Average
3.1% + 3.0% = 6.1% 181% 2.3% 12.9% + 1.0% = 13.8% - 5.7% = 8.1%
(1993–
2018)

Source: AQR, Pitchbook, Bain & Company, Bloomberg, CEM Benchmarking, Consensus Economics. Current estimate as of September 30,
2018, and subject to change. Historical averages cover period January 1, 1993 to September 30, 2018. Please see the Appendix for further
detail. For illustrative purposes only and not representative of any portfolio or strategy that AQR currently manages. There is no guarantee,
express or implied, that long-term return targets will be achieved. Realized returns may come in higher or lower than expected.

21 Strictly speaking, we should lever up arithmetic mean (AM) estimates of the unlevered expected return, in the equation described in
footnote 6 and numerically estimated in Exhibit 3. Assuming that our unlevered return is more like a geometric mean (GM) and we want
to ultimately derive GM-like ERs, a more precise estimate would involve a “roundtrip” of converting unlevered ER into AM, then applying
leverage and adjusting for multiple expansion and fees, and then converting back to GM. Under the current conditions depicted in
Exhibit 3, we estimate the net impact of this roundtrip from GM to AM and back to GM as -0.3%.to +0.3% on PE ER, depending on
the assumed PE volatility level (10-25%) and leverage conditions. Our approximate approach in Exhibit 3 ignores this roundtrip for
simplicity, given the small magnitude of error.
 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19 9

We elaborate on our framework and our • Multiple Expansion: We estimate the

assumptions for PE in the Appendix. To return from multiple expansion as the
summarize: annualized return if PE multiples converge
partly (20% of the way) toward the initial
• Yield: We assume PE’s payout yield to be public market multiple, over the lifetime of
half of its unlevered earnings yield (EBIT- the deal (assumed to be six years).
to-EV), and somewhat crudely estimate
EBIT-to-EV as half of the EBITDA/EV • Fees: We assume PE fees of 5.7%, as per
at time of purchase, based on historical the Døskeland and Strömberg (2018) and
averages observed for public equities. CEM Benchmarking survey estimate
average.
• Growth Rate: We assume an unlevered
real growth rate of 3%, which is
more than double what we assume Putting it together, we estimate a real ER for
for public equities. This is further U.S. PE of 9.6% gross and 3.9% net-of-fee.22
amplified through financial leverage. In comparison, our U.S. public equity real
Hypothetical
return estimate is 3.1%, net of a 10bps fee
• Leverage: For the post-2008 period, for passive investing.23 We thus expect PE to
we interpolate annual D/E ratios from have a roughly 80bps higher net-of-fee ER.24
Pitchbook. Pre-2008, we assume an As mentioned earlier, we do not interpret
aggregate D/E ratio that tapers from 300% this outperformance of PE as an illiquidity
in the 1990s to the 150% D/E reported premium, but the warranted risk premium
in 2008, to capture the documented given the higher equity risk of PE.
downtrend in PE leverage.

• Cost of Debt: We estimate PE’s cost of

debt as real LIBOR plus a spread proxied
by 33% of the High Yield index OAS over
duration-matched Treasuries.

There is no guarantee, express or implied, that long-term return targets will be achieved. Realized returns may come in higher or lower than
expected.

22 PE returns and ERs can be measured for a given vintage or across many vintages for a given period. Our approach uses the most
recent purchase multiple and tries to loosely estimate the ER for the current vintage year for the next 5-10 year period. (In practice,
vintage year data may provide better transparency on prevailing PE valuations even if it only reflects just-deployed capital. We
implicitly assume that if the capital deployed in previous vintages were properly marked-to-market at the same point in time, and was
deployed to a similar mix of industries, under similar financing conditions, it would have similar valuations and expected returns. In this
light, our estimates apply to the whole buyout market, although a value-weighted purchase multiple may better represent the entire
market than the median purchase multiple we use).
23 See Alternative Thinking Q1 2017. Averaging our two methods for public equities would lead to a higher ER for U.S. public equity
(the S&P 500), as that includes net buybacks. However, here we estimate public equity real ER using only the earnings yield based
methodology, ignoring net buybacks, as that is closer to our PE framework.
24 It may seem misleading to compare PE fees to equity index fund fees, when a more natural comparison would be against active public
equity funds. Since our focus is on asset class expected returns, we use public equity index funds as the implicit PE benchmark, but
we note that the PE edge over public equity would be higher against a benchmark of active equity funds if the latter are collectively
assumed to underperform passive funds due to their higher fees.
 10 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19

Our current estimate of PE outperformance the 2010s, driven by richening PE multiples

is undoubtedly low compared to history. (resulting in both lower yields and lower
Exhibit 4 charts net-of-fee PE ER and public multiple expansion) and a gradual decline
equity ER through time (we caveat that in leverage. The early 1990s and 2002-5 were
limited data especially in the earlier part of halcyon years when both PE valuations and
the sample necessitates the use of simplifying the cost of debt were low; it is no wonder then
assumptions and imperfect proxies which we that those vintages delivered high subsequent
describe in the Appendix). The gyrations in returns. Our current outlook is far more
the PE ER line are driven most by fluctuations modest reflecting PE’s rich valuations and
in the cost of debt, as it is the only input based low leverage. The Appendix includes a visual
on mark-to-market data. We clearly discern 25
decomposition of the PE expected return edge
a downtrend in PE ER from the 1990s to over public equity.

Exhibit 4
Net-of-fee Expected Returns for Private Equity and Public Equity
January 1, 1993 – September 30, 2018

25%

20%
Real Net Expected Return

15%

10%

5%

0%
1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017

Public Real Net ER Private Real Net ER

Source: AQR, Pitchbook, Bain & Company, Shiller, Consensus Economics, Bloomberg. For public equity real net ER, see Alternative Thinking
Q1 2017 and footnote 23. Private equity real net ER described in the Appendix in further detail. There is no guarantee, express or implied,
that long-term return targets will be achieved. Realized returns may come in higher or lower than expected. For illustrative purposes only and
not representative of any strategy that AQR currently manages.

25 One limitation of our PE framework may be its sensitivity to the cost of debt, and our imperfect proxy for the cost of debt. During
periods like 2000 when real cash rates were high, or 2008-9 when credit spreads spiked, our cost of debt may be overstated and
our methodology can give misleadingly low estimates of broad PE market ER. Under such conditions, new leveraged buyouts become
uneconomical and primary PE markets slow down as the cost of debt is too high to warrant more leverage. On these rare occasions,
the secondary market may be more active and provide a better estimate of actual transaction prices and thus the broad PE market
ER. During the 2008-9 turmoil, the secondary market pointed to a very high ER, which seems more intuitive than the low ER seen in
Exhibit 4.
 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19 11

Conclusion

Our analysis suggests that private equity does Recent surveys suggest that investors still
not seem to offer as attractive a net-of-fee have high expectations of prospective PE
return edge over public market counterparts returns. This may be due to the inherent
as it did 15-20 years ago, from either a difficulty of modeling illiquid assets, and lack
historical or forward-looking perspective. of transparency on fees and performance.
Institutional interest in private equity has In this article, we present more comparable
increased despite its mediocre performance in benchmarks or suitable adjustments for
the past decade versus corresponding public evaluating past performance, and a yield-
markets, and weak evidence on the existence based framework to estimate future returns.
of an illiquidity premium. While this While some specific assumptions are
demand may reflect a (possibly misplaced) debatable, our framework helps to illustrate
conviction in the illiquidity premium, it may the basic arithmetic or the ‘moving parts’
also be due to the appeal of the smoothed underlying expected returns for private equity.
returns of illiquid assets in general. One We humbly admit that return estimates for
possibility is that investor overpayment any asset class come with a great deal of
for the smoothing characteristics offsets a uncertainty, and our framework is a work
large part of the fair illiquidity premium. in progress that we may fine-tune in the
future. We hope it is a first step toward a more
intuitive and transparent comparison between
public and private equity.
12 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19

Appendix: Assumptions

Here we expand on the assumptions for expected returns that were summarized in the main body:

• Yield: We average EV/EBITDA purchase multiples from several sources as data on PE

purchase multiples often covers only a small subset of the deal universe and can thus be noisy.
We define PE’s unlevered earnings yield as its EBIT-t0-EV, which we approximate as half of
its EBITDA-to-EV, based on historical averages of public equities.26 While PE does not have
regular payouts the way public equities pay dividends, we then estimate the payout yield of PE
as half of its EBIT-to-EV, along the lines of our methodology for public equities.27 As seen in
Exhibit 2, until the mid-2000s, PE yields were almost always higher than public equity yields.

• Growth Rate: We assume that, even in the absence of financial leverage, PE firms may achieve
higher earnings growth rates through operational improvements resulting in higher margins,
and by being overweight sectors with higher growth rates.28 Leverage may further amplify
this effect if operating income exceeds the interest expense. Our 3% unlevered real growth
assumption factors in both the initial higher growth rate as the GPs improve operations, as
well as the later, lower steady-state growth rate after the company goes public. The latter will
be closer to our 1.5% earnings-per-share real growth assumption for public equities.

• Leverage: In principle, leverage (higher D/E) should boost equity returns if operating income
is greater than the interest expense on the debt. In reality, however, Axelson et al. (2013)
show that low funding costs and high leverage tend to coincide with hot, overpriced vintages
that have lower future returns. Our usage of PE yield and multiple expansion partly captures
this effect. While historical simulations often assume a constant D/E ratio of, say, 200% over
time, PE leverage levels have been trending lower despite lower funding costs.29 For the post-
2008 period, we use D/E ratios from Pitchbook. Pre-2008, we assume an aggregate D/E ratio
that tapers from 300% in the 1990s to the 150% D/E reported in 2008. This is roughly in-line
with the D/E trends depicted in Axelson et al. (2013).

26 For the Russell 2000, on average, EBITDA is roughly twice EBIT over the period January 1, 1995 to September 30, 2018. We perform
this additional step of estimating EBIT-to-EV from EBITDA-to-EV, as EBIT-to-EV does not include depreciation and amortization and
is thus more comparable to the earnings yield we use for public equities, that is based on net income and thus, net of depreciation and
amortization too.
27 See AQR Alternative Thinking Q1 2017: Capital Market Assumptions. Historically, the dividend payout ratio for public equities (the
S&P 500) has averaged roughly 50% over the period January 1, 1900 to December 31, 2016.
28 Acharya et al. (2013) find that PE ownership causes the operating margin (EBITDA/Sales) to increase by around 4% on average
relative to the pre-acquisition phase, while Guo et al. (2011) report an even higher 12% increase in net cash flow to sales. On the other
hand, Cohn et al. (2014) suggest that operating improvements are way more modest and emanate from a natural mean-reversion in
operating efficiency, not the changes introduced by PE GPs. Døskeland and Strömberg (2018) find that PE tends to be overweight
Technology and underweight Financials, even if excluding the venture capital and growth equity segments of PE.
29 Both Axelson et al. (2013) and L’Her et al. (2016) find a decreasing trend in D/E for LBOs. This likely reflects evolving risk preferences
by GPs (bigger firms protecting their brands) and their LP clients (pension funds may be less risk tolerant than family offices and
endowments).
 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19 13

• Cost of Debt: Studies indicate PE firms have superior access to credit and borrow more when
credit is cheap. A high proportion of PE debt is secured bank debt financed at floating rates
plus a spread. In the absence of historically accurate bank loan data for PE, we proxy the
PE bank loan spread as two-thirds of the OAS of the High Yield (HY) index over duration-
matched Treasuries. 30 As the entire spread overstates the cost of debt ultimately borne by the
firm, we estimate the actual PE cost of debt as half of this, at real LIBOR plus one-third of
the HY OAS.31 As the purchase multiples we use are one-year averages, we also use one-year
averages of the cost of debt.

The tax deductibility of interest expense decreases a company’s de facto cost of debt, giving
rise to a debt tax shield that is touted as a value-add of PE. However, as tax laws vary by
jurisdiction, we do not account for taxes here. Our assumption effectively increases the cost
of debt for PE, but this is offset by the aforementioned haircut we apply to the cost of debt.

• Multiple Expansion: We assume no multiple expansion in our yield-based frameworks for

passively managed public equities and fixed income. However, we make an exception for PE
due to its active ownership and some evidence that PE GPs can time deal entries and exits.32
The principle of mean-reversion suggests that PE multiple expansion is more likely if it has
an initial discount versus the market. Hence, as described in the main body, we estimate
the return from multiple expansion as the annualized return if PE multiples converged
partly, say 20% of the way, towards the initial public market multiple, over the lifetime of
the deal, assumed to be six years. We floor this return at zero; that is, we do not allow for
multiple compression, as there is evidence that PE GPs delay exits so as to sell at higher
multiples. Given the arbitrary nature of our estimate, our general skepticism around multiple
expansion for any asset class, and the noise in data on PE purchase multiples, we choose to
apply only a conservative, partial convergence towards the market multiple.33

• Fees: PE fees are partly based on performance. As our assumption of 5.7% is based on
historical averages and we expect future PE returns to be lower than in the past, at least
performance fees may be lower going forward. We stick with the historical average of 5.7% as
our best estimate, as PE fees can vary vastly based on deal terms.

30 Axelson et al. (2013) suggest that a large proportion of LBO debt is secured bank debt that is generally financed at lower rates than
HY bonds. Demiroglu and James (2010) find that over the period 1997 to 2007, PE firms incurred a spread over LIBOR that averaged
around 70% of the HY OAS and was less sensitive to credit conditions than HY spreads. Our use of a constant proportion of the HY
OAS may thus overstate the PE cost of debt when HY spreads spike, as in 2008-9.
31 As explained in AQR Alternative Thinking Q1 2016, investors do not earn the entire credit spread as they incur default losses.
Giesecke et al. (2011) find that over the long-term, the average credit risk premium that investors realize is roughly half the average
credit spread. The flip side of this is that the issuing firm’s actual cost of debt ends up being lower, roughly by half the credit spread, as
the lenders take on part of the default losses.
32 Jenkinson et al. (2018) present some evidence of PE managers' market timing skills related to entries and exits, while Kinlaw et al.
(2015) point to sector timing abilities.
33 The multiple expansion we assume here does not involve assuming that the whole PE asset class richens but rather that even amidst
an unchanged capital market environment, at the time of deal exit, individual deals are able to justify higher multiples than their
purchase multiples, thanks to GPs’ skills in entry/exit timing or to operational improvements that boost expected growth beyond the
going-public date. This is analogous to the rolldown gains bonds can earn when we assume an unchanged yield curve. The market
timing skills of GPs may not help LPs if they fund their PE allocations from public equity.
 14 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19

To illustrate the effect of these inputs on our estimates of PE ER, we can decompose the expected
PE net-of-fee return edge over public equity into its different moving parts. Exhibit A1 presents
this decomposition and reveals that, under our yield-based framework, the declining expected
return differential of PE over public equity has been driven first by the relative richening of PE
mentioned earlier, and second by the decrease in PE leverage which is reflected in both the
declining levered growth differential and partly in the decreasing levered yield differential.34 As
PE leverage has declined from around 300% D/E in the 1990s to 100%-150% D/E more recently, it
has had less of an amplifying effect on PE ER.

Exhibit A1
Decomposition of the Net-of-Fee Expected Return Differential of Private over
Public Equity
January 1, 1993 – September 30, 2018

20%

15%
Return Differential

10%

5%

0%

-5%

-10%

-15%
1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017

Levered Yield Differential Levered Growth Differential

PE Cost of Leverage Multiple Expansion Differential

Fee Differential Total Differential (Net)

Source: AQR, Pitchbook, Bain & Company, CEM Benchmarking, Shiller, Consensus Economics, Bloomberg. For public equity real net ER, see
Alternative Thinking Q1 2017 and footnote 23. Private equity real net ER described in the Appendix.

34 For simplicity, the decomposition shows the difference in levered yields and levered growth rates. Thus, it does not disentangle the
effect of time-variation in PE leverage from the effect of time-variation in the unlevered yield.
 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19 15

We describe the decomposition below:

As per AQR Alternative Thinking Q1 2018 and the dividend discount model (DDM),

Net-of-fee Public Equity ER = ypub + gpub + mpub – fpub 1)

where

ypub = dividend yield

gpub = long-term expected growth rate (assumed to be a constant 1.5%)

mpub = multiple expansion (assumed to be zero)

fpub = management fee (assumed to be 10 bps) for public equities

As per Exhibit 3 and the Modigliani-Miller equation,

Net-of-fee PE ER = ru + (D/E)*(ru – kd) + mpvt – fpvt 2)

where

ru = unlevered PE ER

D/E = debt-to-equity

kd = PE cost of debt

mpvt = PE multiple expansion

fpvt = PE fees (assumed to be a constant 5.7%) for PE

As ru = yu + gu, Equation 2 can be re-written as

Net-of-fee PE ER = ypvt + gpvt – dpvt + mpvt – fpvt 3)

where

ypvt = yu * (1 + D/E), i.e., the levered yield of PE35

gpvt = gu * (1+D/E), i.e., the levered growth rate of PE

dpvt = kd * (D/E), i.e., the interest expense or payout to debtholders

35  Penman et al. (2018) show the relation between unlevered and levered earnings yield as per Modigliani and Miller (1958).
16 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19

Thus, subtracting 1) from 3), the PE net-of-fee return edge over public equity can be attributed to
5 components:

• Levered Yield Differential: PE levered yield minus that of public equities.

• Levered Growth Differential: As we make the simplifying assumption of constant

unlevered growth rates for both private and public equity, the difference is driven entirely
by the time-varying leverage of PE. The near-steady decline we see is due to the declining
trend in PE leverage.

• Multiple Expansion Differential: As we assume zero multiple expansion for public equities,
this equals PE expected multiple expansion. As seen in Exhibit A1, this is just a small
component of the return differential.

• Fee Differential: As we assume constant fee for PE (5.7%) and public equities (10bp), this is a
constant -5.6%.

• PE Payout to Debtholders: Listed firms also pay interest expense to their debtholders. But
as our method for public equity ER starts from dividend yields, it is already net of interest
expense, and thus already accounted for.
 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19 17

References

Acharya, V.V., Gottschalg, O.F., Hahn, M., Kehoe, C. (2013). Corporate Governance and Value
Creation: Evidence from Private Equity, Review of Financial Studies 26 (2), 368-402.

Alquist, R., Israel, R., Moskowitz, T. (2018). Fact, Fiction, and the Size Effect. Journal of Portfolio
Management 45 (1), 3-30.

Andonov, A., Rauh, J.D. (2018). The Return Expectations of Institutional Investors, Stanford
University GSB research paper.

Ang, A. (2014). Asset Management: A Systematic Approach to Factor Investing, Oxford

University Press.

Anson, M. (2017). Measuring Liquidity Premiums for Illiquid Assets, Journal of Alternative
Investments 20 (2): 39-50.

AQR (2016, 2017, 2018, 2019). Alternative Thinking Q1: Capital Market Assumptions.

Axelson, U., Jenkinson, T., Strömberg, P., Weisbach, M.S. (2013). Borrow Cheap, Buy High? The
Determinants of Leverage and Pricing in Buyouts, Journal of Finance 68 (6), 2223-2267.

Bain and Company (2018). Global Private Equity Report 2018.

Banz, R., (1981). The Relationship Between Return and Market Value of Common Stocks.
Journal of Financial Economics 9 (1): 3–18.

Chingono, B., Rasmussen, D. (2015). Leveraged Small Value Equities, SSRN working paper.

Cohn, J.B., Mills, L.F., Towery, E.M. (2014). The Evolution of Capital Structure and Operating
Performance after Leveraged Buyouts: Evidence from U.S. Corporate Tax Returns, Journal of
Financial Economics 111 (2014), 469-494.

Demiroglu, C., James, C.M. (2010). The Role of Private Equity Group Reputation in LBO
Financing, Journal of Financial Economics 96 (2), 306-330.

Døskeland, T.M., Strömberg, P. (2018). Evaluating Investments in Unlisted Equity for the
Norwegian Government Pension Fund Global (GPFG), Report for the Norwegian Ministry
of Finance.

Driessen, J., Lin T., Phalippou, L. (2012). A New Method to Estimate Risk and Return of
Nontraded Assets from Cash Flows: The Case of Private Equity Funds, Journal of Finance
and Quantitative Analysis 47 (3), 511-535.

Franzoni, F., Nowak, E., Phalippou, L. (2012). Private Equity Performance and Liquidity Risk,
Journal of Finance 67 (6), 2341-2373.
18 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19

Giesecke, K., Longstaff, F.A., Shaefer, S., Strebulaev, I. (2011). Corporate Bond Default Risk: A
150-Year Perspective, Journal of Financial Economics 102 (2): 233-250.

Guo, S., Hotchkiss, E.S., Song, W. (2011). Do Buyouts (Still) Create Value?, Journal of Finance 66
(2), 479-517.

Harris, R., Jenkinson, T., Stucke, R. (2012). Are Too Many Private Equity Funds Top Quartile?,
Journal of Applied Corporate Finance 24 (4), 77-89.

Harris, R., Jenkinson, T., Kaplan, S. (2014). Private Equity Performance: What Do We Know?,
Journal of Finance 69 (5), 1851-1882.

Ilmanen, A. (2011). Expected Returns: An Investor's Guide to Harvesting Market Rewards,

Wiley.

Ilmanen, A., Chandra, S., McQuinn, N. (2019), Demystifying Illiquid Assets: Expected Returns
for Real Estate, AQR white paper.

Jenkinson, T., Morkoetter, S., Wetzer, T. (2018). Buy Low and Sell High? Do Private Equity Fund
Managers Have Market Timing Abilities? University of St. Gallen working paper.

Kaplan, S.N., Schoar, A. (2005). Private Equity Performance: Returns, Persistence, and Capital
Flows, Journal of Finance 60 (4), 1791-1823.

Kaplan, S.N., Strömberg, P. (2009). Leveraged Buyouts and Private Equity, Journal of Economic
Perspectives 23 (1), 121-46.

Kinlaw, W., Kritzman, W., Mao, J. (2015). The Components of Private Equity Performance:
Implications for Portfolio Choice, Journal of Alternative Investments 18 (2) 25-38.

L’Her, J., Stoyanova, R., Shaw, K., Scott, W., Lai, C. (2016). A Bottom-Up Approach to the Risk-
Adjusted Performance of the Buyout Fund Market, Financial Analysts Journal 72 (4), 36-48.

McKinsey&Company (2017). Equity Investments in Unlisted Companies, Report for the

Norwegian Ministry of Finance.

Modigliani, F., Miller, M.H. (1958). The Cost of Capital, Corporation Finance, and the Theory of
Investment, American Economic Review 48 (3), 261-297.

Penman, S.H., Reggiani, F., Richardson, S.A., Tuna, A.I. (2018) A Framework for Identifying
Accounting Characteristics for Asset Pricing Models, with an Evaluation of Book-to-Price,
European Financial Management 24 (4), 488-520.

Robinson, D.T., Sensoy, B.A. (2015). Cyclicality, Performance Measurement, and Cash Flow
Liquidity in Private Equity, Journal of Financial Economics.

Stafford, E. (2017). Replicating Private Equity with Value Investing, Homemade Leverage, and
Hold-to-Maturity Accounting, HBS working paper.
 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19 19

Disclosures
This document has been provided to you solely for information purposes and does not constitute an offer or solicitation of an offer or
any advice or recommendation to purchase any securities or other financial instruments and may not be construed as such. The factual
information set forth herein has been obtained or derived from sources believed by the author and AQR Capital Management, LLC (“AQR”)
to be reliable but it is not necessarily all-inclusive and is not guaranteed as to its accuracy and is not to be regarded as a representation or
warranty, express or implied, as to the information’s accuracy or completeness, nor should the attached information serve as the basis of
any investment decision. This document is intended exclusively for the use of the person to whom it has been delivered by AQR, and it is not
to be reproduced or redistributed to any other person. The information set forth herein has been provided to you as secondary information
and should not be the primary source for any investment or allocation decision.

Past performance is not a guarantee of future performance.

This presentation is not research and should not be treated as research. This presentation does not represent valuation judgments with
respect to any financial instrument, issuer, security or sector that may be described or referenced herein and does not represent a formal
or official view of AQR.

The views expressed reflect the current views as of the date hereof and neither the author nor AQR undertakes to advise you of any changes
in the views expressed herein. It should not be assumed that the author or AQR will make investment recommendations in the future that
are consistent with the views expressed herein, or use any or all of the techniques or methods of analysis described herein in managing
client accounts. AQR and its affiliates may have positions (long or short) or engage in securities transactions that are not consistent with the
information and views expressed in this presentation.

This analysis is for illustrative purposes only. This material is intended for informational purposes only and should not be construed as legal or
tax advice, nor is it intended to replace the advice of a qualified attorney or tax advisor. The recipient should conduct his or her own analysis
and consult with professional advisors prior to making any investment decisions.

The information contained herein is only as current as of the date indicated, and may be superseded by subsequent market events or for
other reasons. Charts and graphs provided herein are for illustrative purposes only. The information in this presentation has been developed
internally and/or obtained from sources believed to be reliable; however, neither AQR nor the author guarantees the accuracy, adequacy
or completeness of such information. Nothing contained herein constitutes investment, legal, tax or other advice nor is it to be relied on in
making an investment or other decision.

There can be no assurance that an investment strategy will be successful. Historic market trends are not reliable indicators of actual future
market behavior or future performance of any particular investment which may differ materially, and should not be relied upon as such.
Target allocations contained herein are subject to change. There is no assurance that the target allocations will be achieved, and actual
allocations may be significantly different than that shown here. This presentation should not be viewed as a current or past recommendation
or a solicitation of an offer to buy or sell any securities or to adopt any investment strategy.

The information in this presentation may contain projections or other forward-looking statements regarding future events, targets,
forecasts or expectations regarding the strategies described herein, and is only current as of the date indicated. There is no assurance
that such events or targets will be achieved, and may be significantly different from that shown here. The information in this presentation,
including statements concerning financial market trends, is based on current market conditions, which will fluctuate and may be superseded
by subsequent market events or for other reasons. Performance of all cited indices is calculated on a total return basis with dividends
reinvested.

Diversification does not eliminate the risk of experiencing investment losses. Broad-based securities indices are unmanaged and are not
subject to fees and expenses typically associated with managed accounts or investment funds. Investments cannot be made directly in an
index.

The investment strategy and themes discussed herein may be unsuitable for investors depending on their specific investment objectives
and financial situation. Please note that changes in the rate of exchange of a currency may affect the value, price or income of an investment
adversely.

Neither AQR nor the author assumes any duty to, nor undertakes to update forward looking statements. No representation or warranty,
express or implied, is made or given by or on behalf of AQR, the author or any other person as to the accuracy and completeness or fairness
of the information contained in this presentation, and no responsibility or liability is accepted for any such information. By accepting this
presentation in its entirety, the recipient acknowledges its understanding and acceptance of the foregoing statement.
20 Demystifying Illiquid Assets: Expected Returns for Private Equity | 1Q19

The data and analysis contained herein are based on theoretical and model portfolios and are not representative of the performance of funds
or portfolios that AQR currently manages. The information generated by the above analysis are hypothetical in nature, do not reflect actual
investment results, and are not guarantees of future results. The analyses provided may include certain statements, estimates and targets
prepared with respect to, among other things, historical and anticipated performance of certain assets. Such statements, estimates,
and targets reflect various assumptions by AQR concerning anticipated results that are inherently subject to significant economic,
competitive, and other uncertainties and contingencies and have been included solely for illustrative purposes. The results shown represent
a hypothetical illustration. The hypothetical or simulated performance results are compiled with the benefit of hindsight. No representation
is being made that any account will or is likely to achieve profits or losses similar to those shown. Changes in the assumptions may have a
material impact on the model presented. Other periods may have different results, including losses. There can be no assurance that the
analysis will achieve profits or avoid incurring substantial losses. AQR did not manage or recommend this allocation to clients during periods
shown, and no clients invested money There is a risk of substantial loss associated with trading commodities, futures, options, derivatives
and other financial instruments. Before trading, investors should carefully consider their financial position and risk tolerance to determine
if the proposed trading style is appropriate. Investors should realize that when trading futures, commodities, options, derivatives and other
financial instruments one could lose the full balance of their account. It is also possible to lose more than the initial deposit when trading
derivatives or using leverage. All funds committed to such a trading strategy should be purely risk capital.

This material is intended for informational purposes only and should not be construed as legal or tax advice, nor is it intended to replace the
advice of a qualified attorney or tax advisor. The recipient should conduct his or her own analysis and consult with professional advisors prior
to making any investment decision.

The Cambridge Associates Private Equity U.S. Buyout Index is a pooled horizon internal rate of return (IRR)-based index compiled from the
Cambridge Associates database of U.S. private equity buyout funds. The index returns are reported net of fees, expenses, and carried
interest. The Russell 2000 Index is a market capitalization weighted index designed to represent the performance of the 2,000 smallest
companies in the Russell 3000 Index. The Russell 2000 Value Index is designed to represent the performance of small cap companies that
exhibit value characteristics. The S&P 500 Index is a market capitalization weighted index designed to represent the performance of the
500 largest companies in the U.S. stock market.
 www.aqr.com

AQR Capital Management, LLC Two Greenwich Plaza, Greenwich, CT 06830 P +1.203.742.3600 F +1.203.742.3100