The Future of Private Equity: Evidence and Hypotheses

1
The Future of Private EquityEvidence and
Hypotheses

• Josh Lerner
• Harvard Business School

2
Period of dramatic change

• Growth in fundraising.
• Emergence of diversified mega-funds.
• Entry of new groups.
• Expanding geographic scope.
• Regulatory scrutiny.

3
Scene from a road show
4
Natural questions

• Is this a bubble or a profound shift?
• How will the industry evolve in the next decade?
• What are implications for Abraaj?

5
This talk

• Will seek to answer these questions by looking
  backwards.
• Traditionally, very hard to understand key
  drivers of private equity success.
• In recent years, much more information.
• Drawing on large-sample and case evidence.
• Thoughts about future of private equity more
  generally
• And particularly in new private equity markets.

6
1. Everyone does about the same

• Frequent claim among investors
• Emphasis on balancing portfolio by
• Type of fund.
• Location of fund.
• Vintage year.
• Similar to whats seen in public market
  investing.

7
U.S. private equity returns
8
Recent work

• Has sought to understand how much difference is
• Between fund classes.
• Between funds.
• Seeking to distinguish importance of individual
  performance.

9
Evidence from the Yale endowment
Source Lerner 2003
10
More general patterns
Source Kaplan and Schoar 2005
11
The reality

• The key difference is between different funds
• Unlike public markets.
• Investing in the right categories is not nearly
  as critical as getting into the right companies!

12
2. Regression to the mean

• Frequently heard stories
• Our last two funds were a disappointment, but
  were getting back on track
• I considered investing in the fund, but I
  decided that their success must be a fluke...

13
Recent work

• Has sought to understand nature of performance
• Is there little continuity from
  quarter-to-quarter?
• Many studies of public markets suggest little
  persistence
• Mutual funds.
• Hedge funds.
• Or is the reality different?

14
Persistence of performance

• High likelihood that the next funds of a given
  partnership stays in the same performance bracket
  
• ? Persistence.
• 1 boost in past performance ? 0.77 boost in
  next funds performance.

Source Kaplan and Schoar 2005
15
The reality

• Performance seems to be very sticky
• Good continue to do well.
• Underperformers continue to do so.
• While exceptions, seems to be the basic rule
• Seen in buyouts as well as venture.

16
3. Growth doesnt hurt

• Numerous venture groups have grown dramatically.
• Mid 1980s and late 1990s.
• Recent dramatic growth by buyout funds.
• Have typically argued that can sustain
  performance despite growth.
• But powerful incentives to grow may induce
  skepticism.

17
U.S. private equity fundraising
18
Private equity fundraising outside the U.S.
19
Fund sequence number

• Positive relationship between IRR and fund
  sequence number.
• First time funds perform especially poorly.
• Regression results control for vintage year
  effect, fund category and fund size.

Source
Source Lerner and Schoar 2005
20
Fund size

• Concave relationship between IRR and fund size.
• Fund size is measured as capital committed at
  closing.
• Regression results control for vintage year, fund
  category.

Source Lerner and Schoar 2005
21
Change in fund size

• Negative relationship between change in IRR and
  change in fund size for a given firm.
• Fund size is measured as capital committed at
  closing.
• Regression results control for vintage year
  effect, fund category, and firm fixed effects.

Source Lerner and Schoar 2005
22
Partner to size ratio

• Positive relationship between IRR and the ratio
  of partners to committed capital.
• Regression results control for vintage year
  effect, fund category, and fund size.

Source Lerner and Schoar 2005
23
Partner to total staff ratio

• Positive relationship between IRR and the ratio
  of partners to total staff.
• Total staff includes associates, principals etc,
  excludes purely admin. positions.
• Regression results control for vintage year
  effect, fund category, and fund size.

Source Lerner and Schoar 2005
24
Difference in deal success rate

• Specialist firms are more likely to have
  successful deals.
• I.e., 30 vs. 32.1 vs. 33.1.
• Partners focus especially matters.

Source Gompers, Kovner, Lerner and Scharfstein
2005
25
Returns Disparity between recent past and
historical pattern
26
Returns One past episode
27
One past episode (continued)
28
The reality

• Funds with higher sequence number, i.e.,
  established funds, perform better.
• Larger funds have better performanceto a point.
• Rapid growth in capital under management is
  associated with performance deterioration.
• May be driven by less impact of partners
• Funds with more partners per dollar managed have
  higher returns.
• Funds with higher partner-to-non-partner ratio
  have higher returns.
• Decline in specialization leads to poorer
  performance.

29
4. Anyone can play

• Lately, great deal of interest from new
  investors
• Public pension funds.
• Non-U.S. governmental entities.
• Attracted by high returns that established
  investors have enjoyed.

30
Recent research

• Has sought to understand the differences between
  investors.
• Does everyone do the same?
• Or are there substantial differences?
• Key data
• LP investment decisions.
• Fund returns.
• GP and LP characteristics.

31
Performance summary

• Substantial performance differences
• 13 differential in annual returns between
  endowments and next best.
• Entirely driven by early- and late-stage VC.
• Advisors and banks particularly poor.
• Patterns true when weighted as well.

32
Performance by investor type
Source Lerner, Schoar and Wang 2005
33
Concerns with univariate tests

• Do these reflect other differences
• E.g., endowments early investors and more heavily
  weight VC.
• Examine through regressions
• Regress IRR on fund and LP characteristics.
• Only include lt1999 funds to insure meaningful
  performance numbers.

34
Regression analyses

• Differences persist
• Endowments outperform corporate pensions and
  banks underperform.
• Proximity negatively associated with performance.
• Younger LPs do worse
• At least among advisors, banks, corporate
  pensions, and insurers.

35
Regression analyses (continued)

• Market inflows
• Negative in general.
• Especially for advisors, corporate pensions, and
  insurers.
• Hot markets appear to lead to more herding by
  these investors.
• Robust to other controls
• Also excess IRR, all funds, and median
  regressions.

36
Reinvestment decisions

• Reinvestment decision should be made with better
  information and without access constraints.
• Look at follow-on funds in our sample
• Only look at same classes of funds.

37
Statistics on reinvestment

• Reinvestment rates differ
• Public pensions, insurers higher.
• Higher in VC than buyouts.
• More likely to reinvest when high IRR.
• Next fund has higher IRR when reinvest.

38
Reinvestment (continued)

• Pension funds and advisors tend to invest when
  current returns are high.
• But much more dramatic difference in future
  returns from endowments.
• Also smaller funds.
• Substantial differences in ability to identify or
  act on inside information.

39
Reinvestment and current returns
Source Lerner, Schoar and Wang 2005
40
Reinvestment and future returns
Source Lerner, Schoar and Wang 2005
41
Is access an explanation?

• Do endowments do well because they were there
  first?
• Other way to look at
• Funds that were undersubscribed.
• Funds which took a long time to raise.
• Same patterns appear!

42
The reality

• Huge disparities in performance.
• Superior performance has been largely confined to
  endowments.
• Raise substantial questions about ability of new
  entrants to succeed.

43
Summary

• Funds with higher sequence number, i.e.,
  established funds, have performed better.
• ?Lesson Being early is critical.
• Rapid growth in capital under management was
  associated with performance deterioration.
• May be driven by organizational challenges
• Funds with more partners per dollar managed have
  higher returns.
• Funds with higher partner-to-non-partner ratio
  have higher returns.
• Decline in specialization leads to poorer
  performance.
• ?Lesson Managing growth is major challenge.
• Investors have had wildly uneven returns.
• ?Lesson Having right investment partners matters.

44
The special challenges of new private equity
markets

• Lessons from Celtel, Skype and Shanda
• HBS field cases written on all three
• Represent Africa, Europe and China
• The checks have cleared!

45
1. All markets are global

• Skypes business model depended on consumers
  calling across borders.
• Celtels pan-African strategy attracted
  international telecoms equipment companies to
  become second largest source of financing.
• Shandas revenues were more than 80 dependent on
  a game written and owned by a Korean company.

46
2. Expect deals to be massively more work
intensive

• Skypes code was written in Estonia, the
  management team was spread throughout Europe, the
  customers were all over, and the founders could
  not travel to the US.
• Celtel needed to raise over 400mm from 2001 to
  2004 during the meltdown of the telecommunication
  investing marketsall of it to be spent in
  Africa.
• Shanda was threatened by a lawsuit from the
  Korean vendor whose game accounted for most of
  their revenue.

47
3. Back to basics who are you in business with
and are your interests aligned?

• Skypes founders had control over a sale.
• Celtel had minority partners in all 15 of its
  operations (i.e. 15 different groups in 15
  different African countries). It also had an all
  common stock equity capitalization.
• Shanda management was furious that SAIF sold some
  stock after the IPO.

48
4. The real value added is transparency

• Skype owned its technology as a result of the VC
  investment
• Celtel had a prestigious board who insisted on
  transparency and openly refused licenses that had
  the taint of corruption
• Shandas settlement with its key vendor was
  negotiated by SAIF

49
5. Luck still counts

• (this space intentionally left blank)

50
Advice Top tier firms must be global

• LPs will prefer to go global with people they
  know, but will be skeptical about executionso
  they will pursue a mixed model
• The top tier firms will need to be global to
  maintain top tier statusthere is simply too much
  information they would otherwise miss, and the
  overseas growth and return rates will be higher
  than US return rates
• Global top tier firms will outperform local top
  tier firms over longer periods of time
• Brands will likely cross borders but are no
  guarantee of success.

51
Advice Tourist VCs will not be successful, you
must be on the ground

• Act global but think local.
• Local, permanent, day in, day out presence is an
  absolute must.

52
Advice The industry will not be replicate that
in the U.S.

• Less developed PE markets require much more
  resource on each deal
• Investment strategy may vary from location to
  location
• Less of a premium for early stage investing in
  less developed markets

53
Advice There are no settled models for running a
global private equity firm

• Lots of models to choose from
• Large PE firms provide useful models of satellite
  offices
• There are interesting models of building an
  affiliate firm with different GPs and LPs (Accel,
  Benchmark)
• There are many examples of investment in
  independent firms (Chengwei, Argnor)
• But no set answers yet.
• Keys to success communication among investing
  partners, expectation setting between offices,
  portfolio management globally.

54
Advice You cant do this on the cheap

• Must be approached with the same intensity and
  vigor and commitment as your most important
  initiative.

55
Advice The key company value builder is
transparency

• Exits are through one global market, whether they
  are MA or public floats (NY/London).
• This is the lesson of the 60s and 70s all over
  again dont treat portfolio companies as small
  companies, treat them as large companies who
  happen to be small right now.
• PE firms will need to stockpile management talent
  and keep overseas offices staffed well enough for
  frequent, persistent oversight of portfolio
  companies.

56
Five Easy Pieces (of Advice)

• Top tier firms must be global
• Global top tier firms will outperform local top
  tier firms over longer periods of time
• Tourist VCs will not be successful, you must be
  on the ground
• Overseas industry will not be replicas of U.S.
• There are no settled models for running a global
  private equity firm
• PE Firms can not go global on the cheap
• Must be approached with intensity, vigor and
  commitment.
• The key company value builder is transparency
• Teaching the culture of minority equity ownership
  may be the lasting legacy of US venture capital.

57
Thank you