Recent Developments in IPO Research

1
Recent Developments in IPO Research

• Jay R. Ritter
• University of Florida
• Oxford EFMA IPO Symposium

2
A few topics

• The international competition for new listings
• Why are IPOs underpriced?
• Do IPOs underperform in the long run?
• How are IPOs sold?

3
The international competition for new listings
4
The vast majority of IPOs list in the issuers
home market

• Every year from 1995-2006, 90-97 of IPOs list in
  the issuers home market
• Why?

5
Demand is local

• Some IPOs list in more than one market
  (cross-listing)
• Mainly very large IPOs, including privatizations
• Cross-listings have been rapidly declining,
  partly due to better clearing mechanisms in the
  EU
• Many Chinese companies list abroad (Singapore,
  London, New York)
• For attracting foreign investors, a Hong Kong
  listing is common

6
Some moderate-size and small firms list abroad
rather than at home

• Hong Kongs market share has been rising
• AIMs market share has been rising
• The market shares of the NYSE and NASDAQ have
  been falling
• Some blame the Sarbanes-Oxley (SOX) Act of 2002

7
Number of Offerings and Average First-day Returns
on US IPOs, 1980-2007
8
(No Transcript)
9
(No Transcript)
10
Singapore has become a major center for small
Chinese companies to go public

• Number of Foreign IPOs in Singapore
• Year All From China and Hong
  Kong
• 2004 14 13
• 2005 29 24
• 2006 32 29
• The median proceeds in 2006 25.6 million

11
Market conditions for IPOs fluctuate between hot
and cold

• French IPO volume has fluctuated between 4 and
  121 IPOs per year
• German IPO volume has fluctuated between zero
  and 175 IPOs per year
• Italian IPO volume has fluctuated between zero
  and 42 IPOs per year
• UK IPO volume has fluctuated between 12 and 264
  IPOs per year

12
Number of Offerings and Average First-day Returns
on French IPOs, 1983-2007
13
Number of Offerings and Average First-day Returns
on German IPOs, 1980-2007
14
Number of Offerings and Average First-day Returns
on Italian IPOs, 1985-2007
15
Number of Offerings and Average First-day Returns
on UK IPOs, 1980-2007
16
The number of U.S. IPOs (not including ADRs) has
fluctuated between 63 to over 600 offerings per
year
17
Investor protection vs. excessive regulation

• Jensen and Meckling (1976 JFE)
• Investors price protect themselves
• Shleifer and Vishny (1997 JF)
• La Porta, Lopez-de-Silanes, Shleifer, and Vishny
  (1998 JPE)
• Laws and their enforcement matter

18
Why are IPOs underpriced?
19
(No Transcript)
20
(No Transcript)
21
Asymmetric Information-based Explanations for
Underpricing   The winners curse or Groucho
Marx theorem (Rock (1986)) If you are
unsure of the fair value of shares being sold,
and there is excess demand, the most
optimistic investors are likely to get
the shares   Thus, conditional on getting
the shares, you find out that you are
probably overoptimisitic   The need to give
institutions an incentive to investigate a
company and buy its shares (Benveniste and
Spindt (1989))
22
These are good explanations if we were seeking
to explain why, on average, underpricing is
5-10   These are not good explanations when we
are trying to explain why underpricing is 15 or
more
23
Underpricing and allocations are related There
are three frameworks for viewing discretion in
allocations The information acquisition
view (Benveniste and Spindt), in which
underwriters favor regular investors who provide
information about demand, resulting in more
accurate pricing The pitchbook view, in
which underwriters seek out buy-and-hold
investors The rent-seeking view, in which
underwriters trade money left on the table for
quid pro quos (commission business) Biased
analyst recommendations appeal to issuing firms
and make them willing to leave money on the
table
24
IPO Underpricing

• Is it equilibrium compensation for risk-bearing
  or providing information?
• Is it excessive, with rent-seeking behavior
  common?

25
Loughran and Ritters (2004 FM) issuer objective
function

• a1IPO Proceeds a2Proceeds from Future Sales
  (1-a1-a2)Side Payments

26
Why do issuers put up with severe
underpricing?   On internet IPOs, underwriters
knew they were overpriced But why did their
analysts put out buy recommendations?   Issuer
stupidity The publicity is worth it
Capital can be raised in a follow-on
offering   Prospect theory When people get
good news about their wealth increasing,
they dont bargain as hard at the pricing
meeting   Analyst lust and spinning
Issuers seek out underwriters where influential
analysts will be bullish Spinning of
hot IPOs to executives
27
Scandals (SLAC)   Spinning Allocating hot
IPOs to the personal brokerage accounts of top
executives in return for company
business   Laddering Requiring the purchase of
additional shares in the aftermarket in return
for IPOs   Analyst conflicts of interest
Giving buy recommendations in return for
underwriting and MA business   Commission
business in return for IPOs Underwriters
allocated IPOs primarily to investors that
generated soft dollar commissions on other
trades
28
Academic Evidence on SLAC Problems
  Spinning Xiaoding Liu and Jay
Ritters Corporate Executive Bribery An
Empirical Analysis   Laddering Grace Haos
2007 JFE paper Laddering in Initial Public
Offerings   Analyst conflicts of interest
Mike Cliff and David Deniss 2004 JF paper and
Liu-Ritter paper Commissions Jon Reuters 2006
JF paper
29
Example of spinning Salomon Smith Barney's
allocations of IPOs to Bernie Ebbers   Ebber's O
ffer Market First-day IPO Date Shares
Price Price Profit McLeod 6/96 200,0
00 20.00 25.13 1,026,000 Tag
Heuer 9/96 5,000 19.55 20.00 2,250 Qwest
Communications 6/97 205,000 22.00 28.00 1,230,0
00 TV Azteca 8/97 1,000 18.25 19.19 900 Box
Hill Systems 9/97 5,000 15.00 20.62 28,100 Next
link Communications 9/97 200,000 17.00 23.25 1,
250,000 China Mobile HK 10/97 2,000 30.50 28.00
-5,000 Metromedia Fiber 10/97 100,000 16.00 21.
38 538,000 Teligent 11/97 30,000 21.50 25.63 1
23,900 Earthshell 3/98 12,500 21.00 23.56 32,00
0 Rhythms Netconnection 4/99 10,000 21.00 69.13
481,300 Juno Online 5/99 10,000 13.00 11.63 -1
3,700 Juniper Networks 6/99 5,000 34.00 98.88 3
24,400 Focal Communications 7/99 5,000 13.00 19.
50 32,500 Williams Communications 10/99 35,000 2
3.00 28.06 177,100 Radio Unica 10/99 4,000 16.0
0 27.44 45,800 Chartered Semiconductor 10/99 5,0
00 20.00 33.19 66,000 UPS 11/99 2,000 50.00 6
7.38 34,800 KPNQwest 11/99 20,000 20.81 29.81
180,000 Tycom Ltd 7/00 7,500 32.00 37.00 37,500
Signalsoft 8/00 5,000 17.00 21.88 24,400  
30
Corporate Executive Bribery An Empirical
Analysis (joint work with Xiaoding Liu)

• IPOs in which the executives are being spun are
  underpriced about 18 more than other IPOs,
  holding everything else constant
• Firms that are involved in spinning are much less
  likely to switch underwriters for their next
  public equity offering (5 vs. 31)

31
Analyst coverage
32
Analyst Coverage Is Bundled with Bookbuilding

• Degeorge, Derrien, and Womack (July 2007 RFS)
• Analyst Hype In IPOs Explaining the Popularity
  of Bookbuilding

33
What are the most important qualities of a good
analyst?

• a) Accurate earnings forecasts
• b) Timely buy and sell recommendations
• c) Insightful written reports
• d) Setting up meetings with management
• e) Accessibility/responsiveness of phone calls
• f) Industry knowledge

34
(No Transcript)
35
Example of kickbacks with commissions Credit
Suisse First Boston (CSFB) received commission
business equal to as much as 65 of the profits
that some investors received from certain hot
IPOs, such as the December 9, 1999 IPO of VA
Linux   The VA Linux IPO involved 5.06 million
shares   Offer price 30.00 Closing market
price 239.25 Capital gain 209.25  
Gross spread 2.10   If the investor then
traded shares to generate commissions of
one-half of this profit the total underwriter
compensation per share was 2.10 plus 104.625,
or 106.725
36
According to paragraph 58 of the SECs January
22, 2002 settlement with CSFB, an institutional
customer that had received a 12,500 share
allocation of VA Linux from CSFB paid CSFB at
least 565,000 by engaging in the following
transactions
37
Where were the regulators during the internet
bubble?   At first
38
(No Transcript)
39
But then
40
(No Transcript)
41
And then
42
(No Transcript)
43
Long-run performance of IPOs
44
While IPOs tend to go up on the first day of
trading, in the long run, on average they have
tended to underperform.   But there is a strong
cross-sectional pattern in the U.S. IPOs that
had annual sales of less than 50 million
severely underperform, whereas those that had
achieved annual sales of 50 million dont
underperform. Buy-and-hold stock returns are
skewed there are some big winners, but most
stocks underperform. This is especially true
with young companies, where there is even greater
right skewness.
45
Annual returns in the five years after going
public for U.S. 6,973 IPOs from 1970-2003.
Style-matched firms match on market cap and
book-to-market.
46
(No Transcript)
47
Many of the AIM listings have annual sales below
50 million, a category in which U.S. IPOs have
underperformed in the past.
48
How are IPOs sold?
49
Fixed Price OfferingsBookbuilding Information
acquisition (Benveniste and Spindt
(1989)) Agency problems (Loughran and Ritter
(2002, 2004))AuctionsHybrid Mechanisms
50
Pricing and Allocation

• Is the offer price set before or after
  information about the state of demand is
  acquired?
• Are shares allocated in a discriminatory or
  non-discriminatory manner (favoritism vs. pro
  rata)?

51
August 1992, Shenzhen Before the rioting
started,crowds waited for the new share
subscription forms
52
(No Transcript)
53
Googles IPO

• Andreas Trautens paper this morning
• Why the Google IPO might stay exotic
• an experimental analysis of offering mechanisms
• File price range of 108-135/share
• Auction result 85/share offer price
• Closing first-day price of 100.33/share

54
What went wrong?

• Out of equilibrium disclosure strategy
• First post-IPO earnings announcement on Oct. 21,
  2004, with revenue up 105 from the year-earlier
  quarter price jumped 15.4
• Second post-IPO earnings announcement on Feb. 1,
  2005 , with revenue up 101 from the year-earlier
  quarter price jumped 7.3

55
Summary   Hot and Cold markets will
continue Issuers still put up with
underpricing Long-run underperformance is
restricted to companies going public with less
than 50 million in annual sales In the
U.S., auctions are becoming (slightly) more
common, although bookbuilding has become the
dominant method internationally
56
Summary (continued)

• Regulatory reform has changed the game a little
• Spinning has been nearly eliminated
• Laddering continues to occur
• Analyst lust will continue
• Commission business in return for IPOs is
  still allowed
• Underwriters still have an
  incentive to underprice
• The academic literature still
  focuses too much on
• asymmetric information models
  rather than agency
• models