IPO Underpricing and the Evolutions of Regulations: Evidence from Chinese Stock Markets

Presented by: Gang Wang (Steven)

School of Accounting, Economics & FinanceFaculty of Business & LawDeakin University, Australia

Lb4 226, 221 Burwood Highway Burwood, VIC (3125)

Literature review

Category hypothesis LiteraturesAsymmetric information theory

Winner’s curse Rock (1986), Koh & Walter (1989)Information revelation Benveniste & spindt (1989), Loughran & Ritter (2002)

Ex ante uncertainty Beatty and Ritter (1986)Signalling Welch (1989), Allen & Faulhaber (1989)

Principal-agent Baron (1982), Loughran & Ritter (2002)

Institutional theory lawsuit avoidance Lowry & Shu (2002)Price stabilization Ruud (1993)tax advantage Rydqvist (1997)

Control theory Retain control Brennan & Frank (1997)Lessen monitoring cost Stoughton & Zechner (1998)

Behavioural theory Informational cascade Welch (1992),Prospect Kahneman & Tversky, (1979), Loughran & Ritter

(2002)

Summary of underpricing theories

Cross-country underpricing (Ritter, 2006)

Institutional background

Definitive role of China Securities Regulatory Commission (CSRC) Issuance Review Committee (IRC)

High competition and uncertainty in getting approval from CSRC

Frequently varied laws and regulations/directories

Data

Data resource: principally from CSMAR database, some corrections according to SHSE & SZSE.

Number of observations: 1724 (891 listed in SHSE and 833 listed in SZSE).

Sample period: companies listed from 10 December, 1990 to 31 December, 2009. (their issue date has a longer span from 18 November 1984 through 21 December 2009)

IPOs (up to 70) in GEB (Growth Enterprise Board) were excluded from the sample.

Models

Variables in asymmetric information theories Ex ante uncertainty ---industry, firm risk (lngrsprc), pvp (par value to offer

price), statecontrol, region Information revelation---bookbuilding

Variables in control theories—propsold

Variables related to issue policies---lagdays, placement, subperiod

1. Annual Percentage Yield (APY)----Winners Curse APY (Annual percentage yield)=underpricing*ballot*365/lagdays

2. Regression Model

Descriptive statisticsstats underprc lngrsprc pvp lagdays propsold bookbuilding region statecontrol placement

mean 2.37 10.15 0.22 166.03 0.31 0.22 0.28 0.66 0.25

Median 1.2 10.17 0.16 21 0.27 0 0 1 0

Max 49 15.71 1 4046 1 1 1 1 1

Min -0.19 3.69 0.01 0 0 0 0 0 0

Sd 4.27 1.25 0.22 514.45 0.15 0.41 0.45 0.47 0.43

skewness 4.98 0.16 2.71 4.51 2.9 1.38 0.95 -0.69 1.17

kurtosis 34.9 6.54 9.58 24.29 14.93 2.91 1.91 1.48 2.38

N 1724 1724 1724 1724 1724 1724 1724 1724 1724

underprc lngrsprc pvp lagdays propsold bookbuilding region Statecontrol placement

underprc 1

lngrsprc -0.5017* 1

Pvp 0.7601* 0.6207* 1

lagdays 0.6800* 0.4893* 0.7808* 1

propsold 0.1289* -0.0528 0.1194* 0.0911* 1

bookbuilding -0.1358* 0.3240* 0.2528* 0.1551* -0.1937* 1

region 0.0157 0.1110* -0.0491 -0.0634 0.0912* 0.0627 1

statecontrol -0.0200 0.0650 0.0880* 0.0214 0.1455* -0.3181* -0.015 1

placement -0.1094* 0.2695* 0.2217* 0.1315* -0.1533* 0.7750* 0.1361* 0.2687* 1

Yearly underpricing based on listing date

Note: the line is composite index of SHSE

6124

Yearly underpricing based on issuing date

The trend of yearly underpricing is more obvious.

The relation between index and underpricing is more obvious.

6124

19941984

Underpricing of sub-periods

1. 18/11/1984--30/06/1994, 366 Obs (when the issuing of staff shares was banned). 2. 01/07/1994--23/04/2001, 739 Obs (when approval-system took the place of quota system).3. 23/04/2001--30/05/2005, 287 Obs (when nontradable shares reform succeeded)4. 01/06/2005--31/12/2009. 332 Obs (all-tradable era)

6.55401

1.33772.930131

1.28589

02

46

8m

ean

of u

nder

prc

1 2 3 4

Return of uninformed investors (APY)Obs Mean Std. Dev. Min Max

1994 3 0.1304141 0.1324931 0.002541 0.267088

1995 7 0.3392421 0.4051187 0.025748 1.086405

1996 150 0.3471775 0.6739665 -1.97413 4.758532

1997 193 0.5794339 0.8470744 0.001725 6.244348

1998 93 0.0576804 0.0566266 0.002013 0.313152

1999 96 0.0533501 0.0596998 0.001633 0.309538

2000 132 0.0830548 0.0701947 0.003779 0.45672

2001 73 0.0674302 0.0558278 0.006017 0.310252

2002 24 0.0486491 0.0320378 0.015683 0.161266

2003 2 0.0218341 0.0226293 0.005833 0.037835

2004 1 0.1444711 . 0.144471 0.144471

2005 0

2006 65 0.0653317 0.0584036 0 0.328239

2007 123 0.1321216 0.1451768 0.005071 1.157596

2008 76 0.030676 0.0323994 0.001896 0.142991

2009 63 0.089043 0.128347 0.013052 0.994013

overall 1101 0.202397 0.4818681 -1.97413 6.244348

OLS Regression results  (1) (2) (3) (4)

VARIABLES underprc underprc underprc underprc         

lngrsprc -0.352*** -0.404*** -0.587*** -0.422***

(-5.78) (-5.02) (-7.60) (-5.18)

pvp 11.390*** 11.219*** 11.145*** 11.453***

(11.29) (10.91) (9.96) (10.79)

lagdays 0.002*** 0.002*** 0.002*** 0.002***

(3.62) (3.64) (3.81) (3.62)

propsold 1.394** 1.481** 0.685 1.691**

(2.12) (2.29) (1.04) (2.55)

bookbuilding 0.548*** 0.511*** -0.238 0.033

(2.92) (2.74) (-1.17) (0.19)

region 0.740*** 0.719*** 0.730*** 0.776***

(4.86) (4.75) (4.13) (4.79)

statecontrol -0.422** -0.413** -0.417** -0.395**

(-2.45) (-2.41) (-2.43) (-2.23)

placement 0.135 0.166 0.177 0.040

(0.68) (0.83) (0.70) (0.17)

Constant 2.645*** 4.268*** 5.641*** 4.089***

(3.79) (2.73) (4.04) (2.59)

industry dummy no yes yes yes

year dummy no no yes no

sub-period dummy no no no yes

Observations 1,713 1,713 1,713 1,713

R-squared 0.641 0.644 0.678 0.645

Robust t-statistics in parentheses

*** p<0.01, ** p<0.05, * p<0.1

OLS Regression results by subperiod　 [1] [2] [3] [4]

VARIABLES underprc underprc underprc underprc

lngrsprc -0.525 -0.372*** -0.432*** -0.337***

(-1.64) (-6.76) (-7.73) (-7.31)

pvp 14.689*** 0.463 2.275*** 1.081

-9.77 -0.52 -4 -1.34

lagdays 0.001** 0.004 0.023** 0.009***

-2.44 -1.27 -2.57 -2.62

propsold 0.907 1.382*** 0.144 -1.649*

-0.84 -3.43 -0.26 (-1.90)

bookbuilding 1.074** -0.680*** 0.408*** -0.402

-2.14 (-3.05) -3.73 (-0.61)

region 3.066*** 0.243*** 0.117 0.166

-4.48 -2.89 -1.38 -1.39

statecontrol -0.63 0.029 0.248*** 0.114

(-0.83) -0.39 -2.85 -0.84

placement -0.243 0.275** -0.471*** 0.046

(-0.44) -2.38 (-3.41) -0.21

Constant 2.186 4.447*** 4.498*** 5.410***

-0.7 -7.22 -6.65 -6.17

Observations 355 739 287 332

R-squared 0.583 0.152 0.32 0.177

Robust t-statistics in parentheses

*** p<0.01, ** p<0.05, * p<0.1

1. The extremely high underpricing (655.40%) in first sub period comes from rulers and policies. For example, more than 50% of IPOs were issued at par value (around net asset per share), the par value to offer price in first sub-period is 0.495, while less than 0.2 in other 3 sub periods. More than 20% of IPOs issued shares to related people which need high underpricing to guarantee their profits.

2. The winner’s curse doesn’t apply in Chinese stock markets. The high yield of subscription results in huge demands and low ballot.

3. The private-owned companies have a much higher underpricing, 272.84%, than that of state-controlled companies, 216.55%. The fact reveals that the regulators and investors have less confidence on private-owned companies who have to pay more to buy entrance tickets.

Main findings

4. The underpricing of companies in information-advantaged regions, 247.35%, is higher than that of companies in information-disadvantaged regions, 232.50%. It can be ascribed to home bias.

5. There is a positive relation between the proportion of sold share and underpricing, which seems to be consistent with control theory. However, It can be only interpreted by information asymmetric theory that the higher the proportion granted to issue, the better quality the issuer is.

6. The overall cash dividend is 0. 0794043 yuan per share per year, while the average issuing price is 8.845818. For issuers, the cash-out for possess capital from equity market is 0.8976% per year. It can be easily understood that many companies are willing to go public and leave a large amount of money on the table. The issuers have behavioural bias that they are like to intake money and not to let it out.

Main findings

Conclusion & Further study

The issuing policies in the earlier period are the principal determinant of extremely high underpricing.

The critical revolutions of policies result in obvious variation of IPO underpricing.

The underpricing is influenced by the prosperity of secondary market over the long haul.

Except for the first sub-period (the highest underpricing), the underpricing of other three sub-periods are around 100%, which is still highest over the world. Traditional theories can’t explain it. The further study will focus on the secondary market and the co-action between primary market and secondary market.

Thank you for listening!