Cannes, France) - Measuringriskin private markets · 2020. 2. 4. · ¡ Pro-cyclicality ¡...

Measuring risk in private markets

Prof Dr Cyril Demaria

IPEM // Cannes // January 30, 2020

2

20 years of experience in private markets

• Research: Executive director in charge of private markets research, UBS WM CIO

• Investments: venture capital fund manager, former CIO of Tiaré Investment Management AG

• Advisor to the President of CASDEN-Banque Populaire, a Swiss pension fund and a MFO

• Expert for SECA, France Invest, InvestEurope and CAIA

Six books on private equity, among which:

• Introduction to private equity, Wiley, 2020, 3d ed. (trans.: Portuguese, Spanish and Mandarin)

• Asset allocation and private markets, Wiley, forthcoming 2020

• Private equity fund investments, Palgrave, 2015

Prof Dr Cyril DemariaPartner, Head of Private Markets, Wellershoff & Partners

3

AgendaMeasuring risk in private markets

1. Asset price volatility is irrelevant to measure risk

2. What matters: potential losses

3. Strategy- and execution-related risk

4. Selection risk

5. What about... liquidity risk?

6. Conclusion

Asset price volatility is irrelevant to measure risk

Not only in private markets, but also in general

5

Used as a measure of risk, asset price volatility assumes:¡ Frequent and seamless tradability (so-called « liquidity »)¡ Normal distributions of returns (to run simulations such as Monte-Carlo)è Not applicable to private marketsè Instruments used for listed assets are not relevant (nor Sharpe ratios, etc.)

Private market assets are « less liquid » by construction¡ The liquidity horizon is the third dimension of investing (next to risk & return)¡ Liquidity horizons are predictable: the variability around them is the actual risk

è Focus on what matters to investors: losses of capital and variability of results¡ How much capital could I loose? (simplified value-at-risk assessment)¡ How do returns evolve due to macro and other factors? (strategy assessment)¡ What is my selection risk? (skillset assessment)¡ What is the actual evolution versus the plan? (execution & liquidity risk)¡ (Other, yet to be defined: credit risk for leveraged strategies?)

Volatility of asset prices is irrelevant to measure risk

Wellershoff & Partners– C. Demaria – 30/1/2020Material under copyright – Do not reproduce

6

PE: less volatile than the stock exchange

• Since 2007, LBO and Growth capital funds apply mark-to-market rules• VC, distressed debt and turn-around capital keep the historical cost ruleè Volatility is not a relevant measure of risk in PE (illiquidity + fund shield)

Source: Preqin, S&P, Wellershoff & Partners, 2018.

1

324 285 271 349 419 496630

837 743904

1106 12001382

1512 1552 1635 1744

2040

-300

200

700

1200

1700

2200

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

USDbn

Valueofcompaniesinportfolio(unrealizedvalue) S&P500

Compared quarterly evolution of major stock indexes and PE NAV

S&P 500 2008 = 100

Historicalcost FMV


7

Performance distribution is not normal

Sample: 5000 direct US investments. 300 European VC funds and 618 funds of funds.Source: Weidig & Mathonet/Thomson VentureXpert, 2003, Wellershoff & Partners/Cambridge

Associates, 2017.

Dispersion of returns in VC direct (US), funds (Europe) and funds of funds (generalist)

0%

10%

20%

30%

40%

50%

60%

70% 0

0.2

5

0.5

0.7

5

1

1.2

5

1.5

1.7

5

2

2.2

5

2.5

2.7

5

3

3.2

5

3.5

3.7

5

4

4.2

5

4.5

4.7

5

5

Pro

bab

ility

Distributed to Paid-In Directs Funds Funds of funds

Fund-of-funds are highly concentrated also skewed.

Direct highly skewed, 30% of investments are total losses!

Extreme profits for direct.Healthy profits for fund.

Funds are less skewed and widespread.


8Source : Cambridge Associates, Wellershoff & Partners, 2017.

The third dimension of investment: liquidity horizonDispersion of PE strategies based on their performance and duration

• PE: time-to-liquidity is generally of 3.5 to 5.5 years• Possibility to assess the historical and current variation around the reference pointè Portfolio construction can benefit from using this dimension


9

Misleading notions of "illiquidity risk" and "illiquidity premium" to be given up¡ "Illiquidity risk" refers to variable and unpredictable shifts in tradeability of assets è not

applicable to private markets¡ "Illiquidity premium" refers to reward for investors giving up their ability to freely trade

assets (fixed income) è not applicable to private equity/real assets

Sources of risk are to be found elsewhereØ Measurable¡ Historical losses (frequency and amount)¡ Divergence from average: pooled returns, top/bottom performers, vintage, liquidity

Ø More difficult to assess¡ Concentration and contagion risk (diversification)¡ Behavioural biases (home, prospect theory, overconfidence, timing, fashion/fads,

representativeness, aversion to ambiguity, under-reaction to information, noise)¡ Commitment (intrinsic risk)

Consequences

NRI & eFront – C. Demaria – 19/2/2019Material under copyright – Do not reproduce

What matters: potential losses

Assessing frequency and amounts at stake

11

Losses Profits

Distribution of private equity funds-of-funds based on their IRR

Frequency and losses in private equityUsing funds-of-funds as a proxy for private equity index-fund

N = 346. Source: Wellershoff & Partners/Cambridge Associates, 2017. Data as of Q2/2016.

• Low loss ratio, and limited losses• PE = less risky than listed assets!


Strategy- and execution-related risks

The variation of pooled returns and the divergence of individual vintage years

13

Strategy-related risk: a macro aspect

Source: Cambridge Associates, Wellershoff & Partners, 2018.

Evolution of the pooled average TVPI of global small and mid-size LBO (fully realized)

• Significant variation of performance over time (macro-related risk)• No loss on a pooled-average basis (but CA data is skewed upwards)è Important for portfolio construction and scenario modelling

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

Multiple (TVPI)

Strategy- related risk


14Source: eFront Insight, Wellershoff & Partners, 2018.

Evolution of the TVPI of active LBO funds compared to historical average

Strategy-related risk: individual vintage years

• Tracks active vintage years è benchmarks VY and individual funds• Usable to assess opportunities on the secondary market as wellè Use for tactical purpose and active portfolio management

Historical average

2009

2010

20112012

2013

2014

20152016

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39

Mutiple (TVPI)

Quarters


Selection risk

Extreme and most frequent selection risk

16Source: Cambridge Associates, Wellershoff & Partners, 2018.

Evolution of the top/bottom 5% & 25% TVPI of small and mid-size LBO (fully realized)

• Selection risk is not symmetrical, nor constant• Even worse performers can do well… and are more stable è less « risky »?è Higher performers exhibit a higher dispersion

Assessing selection risk

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

Multiple (TVPI)

Top 5% Top 25% Bottom 25% Bottom 5%

Extreme selection risk

Most frequent selection risk



Correlation of TVPI and TVPI spread (5%) of small and mid-size LBO (1986-2005)

• There is a direct link between TVPI and extreme selection risk• Risk is related to the maturity of funds è limits the assessment of recent fundsè Possible to calculate a simplified/adjusted equivalent of Sharpe/Sortino ratios?

The acid test: risk and return correlation (1/2)

R² = 0.36367 R² = 0.42291

R² = 0.87918

R² = 0.49537

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0

0 0.5 1 1.5 2 2.5 3 3.5 4

TVPI

TVPI spread (5%)

1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 All Linear (All) Linear (93-99) Linear (00-05) Linear (87-92)



Correlation of TVPI and TVPI spread (25%) of small and mid-size LBO (1986-2005)

• There is a direct link between TVPI and most frequent selection risk• As time goes by and the industry matures, the relationship increases

The acid test: risk and return correlation (2/2)

R² = 0.18168 R² = 0.53206

R² = 0.92178 R² = 0.06857

1.00

1.20

1.40

1.60

1.80

2.00

2.20

2.40

2.60

2.80

3.00

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

TVPI

TVPI spread (25%)

1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Linear (All) Linear (93-99) Linear (00-05) Linear (87-92)


What about... liquidity risk?

Liquidity is a dimension of investment: risk can be assessed


Correlation of TVPI and time-to-liquidity of small and mid-size LBO (1986-2005)

• No relationship between performance and time-to-liquidityè Relationship independent of the period considered

Liquidity: independent from returns

R² = 0.00635

R² = 0.41819 R² = 0.32093

R² = 0.0079

1.00

1.20

1.40

1.60

1.80

2.00

2.20

2.40

2.60

2.80

3.00

0 1 2 3 4 5 6 7

TVPI

Time-to-liquidity 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Linear (All) Linear (93-99) Linear (00-05) Linear (87-92)



Correlation of TVPI spread and time-to-liquidity of small/mid-size LBO (1986-2005)

• No relationship between risk and time-to-liquidityè Liquidity is not a risk in itself, but there is a liquidity risk. How to assess it?

Liquidity: independent from risks

R² = 0.01022

R² = 0.16458

R² = 0.07254

R² = 0.02721

1

1.2

1.4

1.6

1.8

2

2.2

2.4

2.6

2.8

3

0 1 2 3 4 5 6 7

TVPI spread (5%)

Time-to-liquidity 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Linear (All) Linear (93-99) Linear (00-05) Linear (87-92)


22

0.00

0.50

1.00

1.50

2.00

2.50

3.00

0 1 2 3 4 5 6 7

TVPI

Time-to-liquidity (years)

Min Average Max

VY 1993 3.23 years

Source: Cambridge Associates, Wellershoff & Partners, 2018.

Relationship TVPI and time-to-liquidity of small/mid-size LBO (1986-2005)

• Liquidity risk can be understood as a continuum: 3.23 to 6.48 years• Average is 4.24 years, the pivot for portfolio constructionè Lower and higher liquidity thresholds are the risk to consider

Liquidity: measuring risks

VY 19876.48 years

4.24 years

Risk: shorter durationConsequence: opportunity cost

Risk: longer durationConsequence: payout shortfall


Conclusion

24

Conclusion

Positive features of these instruments¡ They can be applied to listed assets¡ They are compatible with mid- to long-term investment horizons¡ They support the set-up of ratios (Sortino equivalent) with bespoke thresholds (0, minimum

return)

More pratical measures could be identified¡ Concentration risk¡ Pro-cyclicality¡ Under-reaction¡ Vintage auto-correlations

Note: this is a work in progress, product of a book under way

Thank you for your attention!


Cannes, France) - Measuringriskin private markets · 2020. 2. 4. · ¡ Pro-cyclicality ¡...

Documents

Transcript of Cannes, France) - Measuringriskin private markets · 2020. 2. 4. · ¡ Pro-cyclicality ¡...