Do political connections matter in accessing capital markets? Evidence from China

    Do Political Connections Matter in Accessing Capital Markets? Evidence from China Xiaolu Bao, Sofia Johan, Kenji Kutsuna PII: DOI: Reference:

S1566-0141(16)30057-7 doi: 10.1016/j.ememar.2016.08.009 EMEMAR 467

To appear in:

Emerging Markets Review

Received date: Revised date: Accepted date:

5 May 2015 9 March 2016 25 August 2016

Please cite this article as: Bao, Xiaolu, Johan, Sofia, Kutsuna, Kenji, Do Political Connections Matter in Accessing Capital Markets? Evidence from China, Emerging Markets Review (2016), doi: 10.1016/j.ememar.2016.08.009

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

T

Do Political Connections Matter in Accessing

SC R

IP

Capital Markets? Evidence from China Xiaolu Baoa, Sofia Johanb, Kenji Kutsunaa* Graduate School of Business Administration, Kobe University, Japan b Schulich School of Business York University, Canada

MA

NU

a

D

Abstract We analyze the influence of political connections on firms’ access to capital and the ensuing effect on the cost of raising capital. Using a dataset of 413 IPOs from 2009 to

AC

CE P

TE

2012, we exploit a research setting where government is still highly involved with the process. We find that firms rely on political connections to reduce IPO rejection risk as a firm’s political connections are positively associated with the propensity of obtaining approval for an IPO. We further find political connections are negatively associated with the cost of IPO and connected firms appear to perform better subsequent to their IPOs.

Keywords: IPO; Political connection; Start-ups; China; Emerging markets. JEL classification: G32; G38 ☆ We thank Ting Cao, Douglas Cumming, and participants in the 2015 Emerging Markets Review Special Issue Conference at Shanghai University of International Business and Economics for the helpful suggestions and comments. We also thank comments and suggestions from the referee and two anonymous referees. Sofia Johan thanks the Social Sciences and Humanities Research Council of Canada for financial support. * Corresponding author at: Kobe University, 2-1 Rokkodai, Nada, Kobe, Hyogo, 657-8501, Japan. E-mail address: [email protected]

1

ACCEPTED MANUSCRIPT

1. Introduction

T

Government intervention in business activities has been a common feature of

IP

emerging markets. Through taxation, regulation, policy and so forth, governments

SC R

influence various aspects of business: from output, production processes, to input such as land, energy, infrastructure, and financing (Shleifer and Vishny, 1994; La Porta et al., 1999). As a result of such influence, firms in these emerging markets often find it

NU

difficult to access external capital such as bank loans, which are largely reserved for

MA

state-owned enterprises (SOEs), or are subject to heavy government regulation (Johnson et al., 2000) which result in similar constraints. Competition for external financial

D

resources in these emerging markets thus often involve non-price mechanisms such as

TE

building relationships with bankers, government officers and politicians (Allen et al., 2005, Güçbilmez, 2015).

CE P

A variety of literature has argued that political connections bring value to firms in various ways, ranging from the firms’ terms of borrowing, market valuation, and

AC

long-term performance, to successfully competing for government contracts (Khwaja and Mian, 2005; Faccio, 2006; Fisman, 2001; Fan et al., 2007). The evidence for how political connections affect the process of obtaining arguably the best form of external financing, the public market, however, is still limited. This paper attempts to fill this gap by investigating newly listed firms in China to examine the relation between political connections and the process of going public. Focusing on China, where the government plays a uniquely predominant role in the increasingly capitalist economy, is suitable for conducting this research.1

1

See also Fan et al. (2011), who argue that government plays an extremely important role in China.

2

ACCEPTED MANUSCRIPT

Before a firm is able to access public capital, it must first meet the listing

T

requirements of an exchange. If it believes it is able to do so it will go through the process

IP

of preparing a prospectus, which generally provides the background of the company and

SC R

the terms of the public offering, and that prospectus will be reviewed by the securities regulators of the relevant jurisdiction. For example in the United States, the Securities and Exchange Commission (SEC) requests that a S-1 document will be filed and the SEC will

NU

then negotiate with the firm to either approve, amend or deny the filing. In China, firms

MA

attempting to go public must obtain approval from the China Securities Regulatory Commission (the CSRC),2 otherwise their issues cannot be listed on the stock exchange for trading. We find evidence that suggests that between 2009 and 2012, one in six IPO

TE

D

applications failed to pass the CSRC’s screening. 3 Being rejected by the regulator approving the prospectus is very costly for firms, not only in terms of the foregone

CE P

benefits of the projects the additional capital would have funded, but also the direct cost of preparing the firm for the offering including, but not limited to, the additional audit and

AC

accounting fees, the legal fees, the cost of an organization and board of directors overhaul to meet listing requirements, marketing fees and underwriter fees. This paper focuses on start-up firms because firstly, start-ups are usually more

reliant on external funding than mature firms. As suggested by Huergo and Jaumandreu (2004) and Darby and Zucker (2003), younger firms usually have more growth opportunity than their older peers; obtaining approval from the CSRC and then going public to achieve external funding is thus more valuable for them. However, due to 2

A counterpart to the Securities and Exchanges Commission (SEC) in the United States. This number is calculated by scaling the rejected issues to the total applications. Kutsuna et al. (2009) described that they have incomplete direct information on cancelled offerings but have been advised by industry practitioners that cancellation is rare in Japan. Nomura Securities was able to provide a list of cancellations for the 2001 to 2003 portion of their sample period. During this period, there were 227 completed JASDAQ IPOs. Eleven IPOs were cancelled. 3

3

ACCEPTED MANUSCRIPT

limited performance records, information asymmetry, and greater uncertainty in bad

T

market conditions, it is harder to evaluate a start-ups’ value. Previous studies argue that

IP

investors often use firm specific characteristics, such as the network of the CEOs, to

SC R

assess the potential of an IPO firm (Cosh et al., 2009; Field et al., 2013; Yang et al., 2011). With the use of data from China we are able to identify political connections as a specific characteristic as we believe having a member of the ruling party as a member of

NU

the firm or having a network of contacts with government officials will reduce firm risk

MA

of ideological discrimination, help the firm to tap into key information and therefore to be better prepared for the pre-IPO process (see Li et al., 2008; Chen, 2013).4 We use the CEO and other board director’s political connections and percentage

TE

D

of political directors on the board as the proxies for a firm’s relationship with government. Referring to Li et al. (2006), we define a politically connected director or

CE P

CEO as one who satisfies any one of the following three criteria: (1) a current or former member of the National People’s Congress (NPC); (2) a current or former member of the

AC

Local People’s Congress (LPC); and (3) a current or former member of the Chinese People's Political Consultative Conference (CPPCC). The NPC, LPC and CPPCC are influential quasi-governmental organizations in the Chinese political system, the former being the lawmaker and the latter the sole official advisory body. As the appointments of main government officials must be approved by the PC and CPPCC, and such appointments are often influenced by comments from members of the CPPCC, these two organizations have a significant influence on, and a close relationship with, key government officials.5 4

The Communist Party is the single ruling party in China. Literature has argued that having a membership in Communist Party can help firm to reduce the ideological discrimination, to obtain easier accessing in key information as well as valuable resources. 5 We thank the referee for recommending the use of prior military service or government employment

4

ACCEPTED MANUSCRIPT

We find that 19% of firms have CEOs or other directors who are members of

T

NPC, 15% are members of LPC and 23% are members of CPPCC. After controlling for

IP

other factors that relate to a firms’ riskiness, we find that there exists a significant positive

SC R

association between political connections and the probability of obtaining approval for an IPO from the CSRC. We also find a significant negative association between political connections and direct cost of going public. Regarding the firm’s performance on the

NU

market, there are no statistically significant associations between political connections

MA

and IPO underpricing. However, there is a significant positive association between political connections and long-term performance. In sum, the results of our paper suggest the importance of the role of a CEO’s or

TE

D

other board directors’ political connections in facilitating firms’ access to the public equity capital market. Furthermore, political connections may also afford value to

CE P

connected firms in terms of cost savings, and also future profits. This paper contributes to several strands of literature. First, this paper adds

AC

evidence to previous studies by showing that political connections facilitate a firm’s access to the public capital market, and political connections can help reduce financing costs (Khwaja and Mian, 2005; Faccio, 2006; Chen, 2013). Second, this paper supplements the growing literature (Field et al., 2013; Yang et al., 2011) that argues that the entrepreneurs’ network is valuable for young firms. This paper addresses that as one of the networks, political connection is very valuable for start-ups seeking to raise external capital, especially within the particular institutional background of China. The remainder of the paper proceeds as follows. Section 2 introduces the institutional background of China. Section 3 summarizes the literature and develops as a proxy. Unfortunately, very few of the CEOs and directors are former army or government officials, hence, we did not take this into account.

5

ACCEPTED MANUSCRIPT

hypotheses. Section 4 describes the sample selection method and the characteristics of the

T

data. Section 5 shows the empirical results of regressions. Section 6 reports the robustness

SC R

IP

test by using the propensity score matching method. Section7 concludes the paper.

2. Institutional Background

ChiNext was set up as a junior exchange within the Shenzhen Stock Exchange

NU

(SZSE) in October 2009. Its aim is to list companies in high growth sectors such as

MA

technology and pharmaceuticals. As the SZSE is a multi-tiered capital market comprising the Main Board, SME Board, and the ChiNext market, the listing

D

requirements for each board within the exchange differ although the same regulatory

TE

approvals have to be obtained from the CSRC for all listings. While requiring a far lower level of capital than the Main Board or the SME Board, ChiNext has stricter

CE P

thresholds for other listing requirements (such as for business operations, information disclosure and limitations on stock sales) for transparency and risk management

AC

purposes, which is the norm for many junior exchanges around the world (Cumming and Johan, 2013; Johan, 2010). As with numerous other securities regulators around the world, the CSRC

evaluates the issuers’ filing materials before giving firm issuers the approval to list on ChiNext.6 To engage in an initial public offering, all issuers have to file application documents with the CSRC for approval. An IPO application is first examined by CSRC officials, who then decide within five working days whether or not to accept the application. If the application package meets the basic financial and legislative 6

The described process is according to articles 29~50 of “Interim Measures on the Administration of Initial Public Offerings and Listings of Shares on the ChiNext”, which is disclosed on the website of Shenzhen Stock Exchange as a rule of listing on ChiNext. See http://www.szse.cn/main/en/ListingatSZSE/ListingProcedures/Apply/

6

ACCEPTED MANUSCRIPT

requirements, it is then passed to the Public Offering Review Committee (the

T

Committee), which is a department under the CSRC, for examination and verification.

IP

What is important to note is that during the examination process before the application

SC R

is passed to the Committee, the officials will need to communicate with the provincial government regarding the proposed issue. For Main Board and SME Board issues, the CSRC will need to consult with not only the provincial government, but also the

NU

National Development and Reform Commission and the minister of commerce (for

MA

Sino-foreign joint ventures and foreign enterprises). Before the Committee commences examination, the issuer will disclose the prospectus (submission version) in advance on

D

the website of the CSRC. After their pre-disclosure, the Committee commences review.

TE

In the reviewing process, the Committee will also seek comments on the issuers from the provincial and municipal governments.7 Within 3 months, the CSRC will render its

in figure 1.

CE P

decision on approval or rejection on its website.8 The detailed process is also described

AC



3. Literature Review and Hypotheses Development There is a growing body of research arguing that networks help new firms

access resources (Field et al., 2013; Yang et al., 2011). Hoang and Antoncic (2003) suggest that a well-networked CEO has quicker access to critical information and information on market conditions from a network of contacts (e.g., venture capitalists, underwriters, creditors, and auditors), thus allowing them to tap into key market information to recognize entrepreneurial opportunities, and to prepare for an IPO. Field et al. (2013) argue that the networks of CEOs or directors are more valuable for younger 7 8

Refer to the book “Economic Laws”, by the China CPA Association. Refer to the website of the CSRC: http://www.csrc.gov.cn/pub/zjhpublic/.

7

ACCEPTED MANUSCRIPT

firms when compared to mature firms as younger firms often lack experience and the

T

networks of contacts useful for navigating public markets. They are likely to rely more

IP

heavily on over-boarded directors for their experience and network of contacts. Yang et

SC R

al. (2011) show evidence that a CEO’s network can reduce the firm’s time to IPO. In view of the underdeveloped legal and financial systems in China, ties to the government or political connections play a central role within this network (Allen et al.,

NU

2005). Extant research has established that it is crucial for new firms to establish

MA

relationships with government to gain comparative advantages that enhance firm performance and value (Fan et al., 2008; Fisman, 2001; Goldman et al., 2009; Johnson

D

and Mitton, 2003; Wu et al., 2012a; Wu et al., 2012b; Millar et al., 2005; and Shan et al.,

TE

2015). For the purpose of this paper we refer specifically to Chen et al. (2011) who show that building a relationship with the government becomes essential for firms in order to

CE P

achieve superior economic resources and to avoid discretionary fees, because the local governments in China can either grant preferential treatment to businesses or impose

AC

extra fees and fines on non-SOEs or individual firms with the allocation of economic resources at their discretion. Preferential treatment by government is also suggested by Li et al. (2008), who suggest that having a membership in the communist party can help the non-SOEs or individual firms to gain more confidence in the legal system, and get more easy access to bank loans as well as government loans. Khwaja and Mian (2005) suggest that political connections help firms access resources such as bank loans and Haß et al. (2014) find that political connections has a positive impact on borrowing costs. Faccio (2006) further shows that the value of these companies increases when their executives enter politics. Goldman et al. (2009) also add evidence on political connections helping create firm value. They find that when a connected board member’s 8

ACCEPTED MANUSCRIPT

political party gains control of the presidency, the value generated by the member

T

increases while the value generated by a director connected to the opposing party

IP

decreases.

SC R

There is however another body of research which suggest that political networks may not necessarily be positive for the firm value and performance. Fan et al. (2007) investigate 779 newly partially privatized firms in China, and find that politically

NU

connected boards have a greater propensity to appoint other bureaucrats into their board

MA

instead of professional managers, leading to significantly worse long-term firm performance. Chen et al. (2011) also suggest that appointing bureaucrats to the board

D

reduces the firm’s value and investment efficiency because the bureaucrats extract

TE

resources from listed SOEs to fulfil objectives that are not consistent with value maximization of the firm. With regard to changes in political affiliation, Fisman (2001)

CE P

looks at companies in Indonesia that are connected to President Suharto’s family and shows that these companies lost value following several announcements regarding the

AC

deterioration of his health.

Our concern is to gather evidence on how political connections affect the

process of accessing the public market. As we noted in an earlier section, during the examination process of the listing application, officials of the CSRC do not independently review the applications but have the additional duty to communicate with the provincial government and in certain cases the National Development and Reform Commission and the minister of commerce regarding the proposed issue. Therefore, in addition to referring to literature that find new firms being able to more easily access economic resources and thus enhancing firm value and performance with political connection, we base our hypothesis on Agrawal and Knoeber (2001) who further find 9

ACCEPTED MANUSCRIPT

that political connections help firm secure favourable regulatory conditions. We know

T

that political connections can for example mitigate the incidence of enforcement against

IP

corporate fraud in the US (Correia, 2012; Yu and Yu, 2012) and reduces the likelihood of

SC R

enforcement actions pursuant to regulatory breaches in China (Wu et al., 2014). Firms seeking to access public funding in China’s underdeveloped legal and financial systems rely more heavily on its political network to obtain favourable regulatory conditions, in

NU

this case approval of listing applications. Faccio (2006) suggests that most politically

MA

connected companies are listed in countries with high levels of corruption and a weak legal system and Francis et al. (2009) further show that having a relationship with

D

government can smooth the firms’ initial public offering, by using data from 423 firms

TE

during 1994-1999.

We thus conjecture that within the particular institutional background of

CE P

China, having a relationship with the government can help firms secure regulatory conditions and facilitate a firm’s access to the public equity capital market. Hence, our

AC

first hypothesis is as follows:

H1: Politically connected firms are more likely to obtain approval of their

applications for a public offering from the CSRC.

If politically connected firms are more likely to obtain approval for a public offering, or less likely to have their applications rejected by the CSRC, the politically connected firms will suffer less ex-ante uncertainty, therefore reducing the risk borne by the firm and the underwriter.9 Being rejected by the regulator approving the prospectus

9

Theoretical and empirical research explaining the underwriting costs have mainly concentrated on three areas: the competitive environment of underwriting business, the size of the issue and economies of scale, and the influence of risk parameters (Chen and Ritter, 2000).

10

ACCEPTED MANUSCRIPT

is very costly for firms, not only in terms of the foregone benefits of the projects the

T

additional capital would have funded, but also the direct cost of preparing the firm for the

IP

offering including, but not limited to, the additional audit and accounting fees, the legal

SC R

fees, the cost of an organization and board of directors overhaul to meet listing requirements, marketing fees and underwriter fees. Ritter (1987) defines the cost of a firm going public in two parts: direct costs, usually called gross spread, and the indirect

NU

cost, underpricing. The direct cost is compensation for an underwriter’s advisory

MA

function, certification role as well as the risk of selling the offered securities. The underpricing is compensation for investors due to information asymmetry (Rock, 1986;

D

Benveniste and Spindt, 1989; and Chang et al., 2008). Politically connected firms are

TE

less likely to incur higher direct cost as they are less likely to require extensive advice and certification of quality from the underwriters. The underwriters themselves incur

CE P

lower costs in their initiative to collect information to accurately price the IPO due to the political network (Benveniste and Spindt, 1989; Benveniste and Wilhelm, 1990).

AC

The firms’ political network helps provide certification of firm quality, mitigate information asymmetry, thus making it unnecessary for them to pay a premium for more accuracy from their underwriters (Titman and Trueman, 1986). Our hypothesis is as follows: H2a: Firms with political connections incur lower direct costs and therefore lower gross spreads when compared to firms without political connections.

With regard to indirect costs, underpricing, Fan et al. (2007) examine the relationship between politically connected CEOs and first day stock return. They find a negative relationship between the CEO’s political ties and the issuing firms’ initial 11

ACCEPTED MANUSCRIPT

return. They interpret this negative effect as a signal of government intervention.

T

Futhermore, Tian (2011) studies the level of IPO underpricing with the lag between the

IP

issuing and trading, and insider ownership. Although he attributes the extreme level of

SC R

underpricing to government regulation and contends that politically connected individuals and groups benefit most from the high initial returns, he did not directly test how political connections create value in the process of going public. More specifically

NU

he did not address the question of whether some issues may not necessarily be as

MA

extremely underpriced as others due to political relations. Research suggests that there is an economic motivation for underpricing IPOs. Underwrites underprice IPOs to

D

effectively extract information from their investors. Along with our contention that

TE

politically connected firms will pay lower direct fees due to the increased likelihood of the application for listing being approved by the CSRC, we believe that the same firms

CE P

will also incur lower indirect fees or face lower levels of underpricing. Francis et al. (2009) provide evidence that politically connected firms enjoy a higher IPO offering

AC

price in China. In this paper, we argue that a CEO’s or other board directors’ political connections will help to reduce underpricing because members of the NPC, LPC or the CPPCC are usually prominent members of society with a good reputations, which can be identified as a positive signal of firm quality. Tian (2011) also suggests that the extreme underpricing of Chinese issues may be a result of the lock-up risk faced by investors as firms queue for floatation. He finds a time lag between issue approval and floatation of an average of 54 days and median of 23 days may result in the investors being compensated for the risk taken as a result of illiquidity and potential share price discounts. We believe however that this lock up risk may be reduced as politically connected firms potentially obtain preferential treatment within the queue and thus 12

ACCEPTED MANUSCRIPT

reducing underpricing. This can help to reduce the information asymmetry between

T

investors and firms. Hence, we conjecture that the political connections of a firm will

IP

reduce indirect costs, which for the purposes of this paper we refer to underpricing.

SC R

H2b: Firms with political connections incur lower indirect costs and are therefore underpriced less when compared to firms without political connections.

NU

As mentioned above, a variety of studies suggest that political connections will

MA

add to firm value and performance as comparative advantages are obtained more easily (Fan et al., 2008; Fisman, 2001; Goldman et al., 2009; Johnson and Mitton, 2003; Wu et

D

al., 2012a; Wu et al., 2012b, Millar et al., 2005). Both economic resources are more easily

TE

obtained and adverse results mitigated by the discretion of connected government officials (Chen et al., 2011; Khwaja and Mian, 2005; Haß et al., 2014) Preferential

CE P

treatment by government create firm value and enhance firm performance (Faccio, 2006; Goldman et al., 2009; Li et al., 2008). While such studies suggest that political

AC

connections help firms as they seek rents from governments, contrary research has suggested that as conditions under which governments seek rents from firms exist, underperformance may be associated with political connection. We note however that Fan et al. (2007) analyse data from companies listed on the Shanghai Stock Exchange and the Shenzhen Stock Exchange from 1993 to 2000. The data comprised larger, more latter stage companies, companies that were essentially privatized and as such the extent of political oversight with such companies and the companies we analyse in this paper may differ. The companies seeking listing on the junior ChiNext exchange are smaller high growth technology companies that may have received less political attention prior

13

ACCEPTED MANUSCRIPT

to listing and they may be in a better position to seek rents from government to create

T

value. Our final hypothesis is thus as follows:

IP

H3: Firms with political connections perform better than firms without

SC R

political connections.

4. Sample and Descriptive Statistics

NU

4.1. Sample

MA

This section describes the dataset used in this empirical analysis. In this paper, we use issues that filed at ChiNext because ChiNext is the specific stock exchange for

D

start-ups to list. Since the ChiNext was established in October, 2009, and the central

TE

government in China suspended the IPO activity of the ChiNext stock exchange on November, 2012.10 We use the sample period of October, 2009 to October, 2012.

CE P

Since the analyses in this paper require detailed information about the directors in each firm, we searched for the information from the IPO prospectuses of each firm

AC

that listed on ChiNext from October, 2009 to October, 2012. Regarding the rejected issues, we collected their prospectuses from the CSRC’s website.11 For both listed and rejected firms, we obtained a profile of the CEOs (directors) from the “profile of directors and senior managers” section of the company’s prospectus. Since the CSRC requires that prospectuses should provide the details of the personal background of each director, in addition to the CEO’s (director’s) name, the profile typically contains information on age, gender, education, professional background, and employment 10

The central government suspended the IPO activity of the ChiNext stock exchange on November, 2012, and reopened it on January 2014. China’s central government often suspends stock exchange activity for reformation or other purposes to manage market conditions (Wang, 2013). 11 All the firms need to file a preliminary prospectus with the CSRC first, after the CSRC’s evaluation, they will get an approval/rejection announcement. The approval/rejection announcements will be disclosed on the website of the CSRC.

14

ACCEPTED MANUSCRIPT

history. We then traced the CEO’s (director’s) political connections by examining

T

whether he or she is a current or former member of the NPC, LPC or CPPCC.

IP

We recorded details of 355 firms that successfully listed on ChiNext and 70

SC R

firms that were rejected by the CSRC. Among the 70 rejected firms, 2 of them had no available preliminary prospectus, 10 of them re-filed with the CSRC and successfully gained approval for IPO from the CSRC.12 Therefore, we excluded these 12 firms from

NU

our subsample. The final sample consists of 355 successfully listed firms and 58

MA

rejected firms from the period October, 2009 to October, 2012.

D

4.2. Descriptive Statistics

TE

Table 1 provides descriptive statistics on the total sample. The sample firms here are young, at an average age of 11 years. 70% of the CEOs in these firms are founders.

CE P

The average leverage of the sample firms is around 40%. There are an average of 9 directors serving on a board. Almost 40% of the sample firms have politically connected

AC

directors who are current or former members of the NPC, LPC or CPPCC. The first day initial return of our sample firms ranges from -17% to 209%, which indicates that the information asymmetry and uncertainty is significant in this market. The annual breakdown of rejected issues versus approved issues is presented in Panel A of Table 2. The percentage of issues rejected by the CSRC (out of the total filed) is higher in the year the market was launched in 2009 and in the year the market was suspended in 2012; around 22% of total issues applications were rejected by the CSRC in 12

According to the CSRC, each rejected firm cannot re-apply for IPO within 6 months. After 6 months, however, they can re-apply for IPO. In our sample period, 10 firms successfully gained the approval for IPO from CSRC after their re-application. We did not include these 10 firms into the rejected firm sample.

15

ACCEPTED MANUSCRIPT

these two years. Panel A of Table 2 also reports the annual number and proportion of

T

politically connected firms for both successfully listed issues and rejected issues. The

IP

proportion of political connection is much lower among the rejected sample firms. There

SC R

is also an increasing trend apparent in the proportion of politically connected firms among approved issues, while no particular pattern can be seen in the proportion of politically connected firms among rejected firms on a year-by-year basis. This may suggest that over

NU

time the issuer also recognizes the importance of establishing a relationship with

MA

government and acquiring politically connected persons into their board. The breakdown of the sample of all the issues by industry classification is presented in Panel B of Table 2. We adopt the industry classification system issued by the

TE

D

CSRC, and divide companies into 13 industries. 13 A large proportion of issues on ChiNext are from the Electronics, Information technology (IT), Machinery and

CE P

Pharmaceutical sectors. Among these four main sectors, the rejection rate in the IT sector is the highest, which reached a ratio of 21%, versus 6% in the Electronics sector and 14%

AC

in the Machinery sector. This suggests that the risk factor does have an effect on the decision making of the CSRC when it checks the filing materials of these firms (Braun, 2014).

Panel A of Table 3 shows the difference between the 355 successfully listed firms and the 58 rejected firms. Panel A shows that there are almost no significant differences between the listed issues and rejected issues in terms of their basic attributes such as size, leverage as well as other fundamental attributes. However, the age of the listed firms is on average significantly higher than the rejected issues. Listed firms have

13

Industry classification codes are specified in the preliminary prospectus.

16

ACCEPTED MANUSCRIPT

more founders remaining as CEO. From Panel A of Table 3, we can observe that the

T

completed issues do have a higher proportion of politically connected board members for

IP

each proxy of political connection. Panel B of Table 3 presents the mean difference

SC R

between politically connected firms and non-politically connected firms. From each proxy of political connection, the mean asset and leverage of politically connected firms are statistically significantly larger than their politically unconnected peers, indicating

NU

that the political connections do help firms in accessing financial resources such as loans.

MA



D

5. Empirical Results

TE

5.1. Do Political Connections Increase the Propensity of Passing the Scrutiny of the

CE P

CSRC?

In this section, we perform regression analyses to examine the effects of the firm’s political connections on the probability of going public and also the effects of

AC

different types of connection on a firms’ probability of going public.

5.1.1. Results of Regressions We use a probit model to test whether the political connections of a CEO or director will facilitate a firm’s access to the public equity capital market. The dependent variable is a dummy variable equals to 1 if the firm has successfully listed on ChiNext, and 0 otherwise. Four proxies are used to represent the firms’ political connections. Three proxies represent the politically connected CEOs and other board directors in a firm. Another

17

ACCEPTED MANUSCRIPT

proxy is used to represent firm level political connection, which is defined as the

T

percentage of political connected directors in board. We repeat the four regressions in

IP

Table 4 with separate political connection proxies based on the following: (1) a CEO or

SC R

board director is a former or current member of the NPC, (2) a CEO or board director is a former or current member of the LPC, (3) a CEO or board director is a former or current member of the CPPCC. (4) The percentage of politically connected directors on the board

NU

is considered as a specific empirical proxy for corporate political connection.14

MA

We include characteristics that may be related to the riskiness of a firm as control variables, including a founder CEO dummy, which indicates the extent of the

D

commitment of the CEO, ownership reduction post-IPO, which will be recognized as a

TE

positive signal by outsiders if entrepreneurs retain significant ownership in the start-up after the IPO (Leland and Pyle, 1977), a % Board professional variable, which is defined

CE P

as the to represent the professionalism of a board, and other variables such as firm age, leverage, size, and average growth rate for the three years prior to IPO. In addition, we

AC

also include the venture capital dummy and the reputable underwriter dummy, 15 because these third parties can have a significant effect on the process of a firm going public (Megginson and Weiss, 1990; Cumming and Johan, 2008; Cumming, 2008; and Su and Brookfield, 2013). Table 4 reports the results of the regression. Model (1) shows that a CEO’s or director’s NPC membership has a significant and positive effect on the probability of going public at the 5% level. Model 4 shows that the percentage of politically connected directors on a board is also significant and positively related to the probability of a firm 14

These proxies are employed by Li et al. (2006) and Chen et al. (2011) when they analyze the effect of political connections on firms in China. 15 Following Liu and Ritter (2011), we employ a top 10 underwriter dummy based on the number of IPOs underwritten in the 3 years before the IPO.

18

ACCEPTED MANUSCRIPT

passing through the scrutiny of the CSRC, which indicates that with more politically

T

connected members serving on the board, the higher the propensity that the firm will

IP

obtain approval for IPO from CSRC. Although membership of the PC in the local area

SC R

and membership of the CPPCC do not have any significant effects, where overall the analysis of the specific types of political connection suggests that these measures reasonably capture the CEO and/or director’s ties with government and how it affects the

NU

probability of going public. Even though the measures in this paper do not capture all

MA

possible channels of political connection, we believe our findings support our Hypothesis 1. We posit therefore that when firms attempt to gain access to the financial market,

TE

D

connections with the government are essential, at least in China.

5.2. Do Political Connections Help to Reduce the Cost of IPO?

CE P

In this section, we investigate how political connection will affect IPO cost. First, we examine the relationship between political connection and the direct cost of

AC

IPO, which we call gross spread. We then focus on the indirect cost, which is usually called underpricing.

5.2.1. The Direct Cost-Gross Spread In order to examine whether political connections have an impact on gross spread, we controlled for other factors that we believe may also affect the gross spread. Chen and Ritter (2000) document economies of scale associated with issue size. Therefore, we controlled for the proceeds of the issue. Klein and Leffler (1980) argue that an underwriter’s compensation is positively associated with reputation. Habib and Ljungqvist (2001) explain that the higher the amount of shares sold by initial 19

ACCEPTED MANUSCRIPT

shareholders at the time of IPO, the higher the loss of wealth led by high underpricing.

T

Initial shareholders may try to limit the underpricing by paying higher costs for IPO and

IP

choosing, for instance, a prestigious underwriter.

SC R

The regression results are shown in Table 5. As expected and in support of our Hypothesis 2a, several control factors such as the underwriter reputation, and the proceeds influence the gross spread. The estimated coefficient on each proxy of political

NU

connection is negative, the percentage of politically connected CEOs or other directors

MA

serving on a board is significant at a 5% level, suggesting that more politically connected directors decrease the rejection probability, which in turn decreases the IPO

D

ex-ante uncertainty that is faced by the underwriters.

TE



CE P

5.2.2. The Indirect Cost-Underpricing We conjecture that there is a negative association between political connection

AC

and underpricing as the positive repute of a government or political appointee should be reflected in the moral fibre of the firm, thus mitigating information asymmetry. Table 6 reports the results of multivariate regression on underpricing. The result is as expected, although not statistical significant.16 This result also provides weak support of our Hypothesis 2b suggesting that being a member of the NPC, LPC or CPPCC may not necessarily help to reduce the information asymmetry between investors and firms. It is surprising that the venture capital dummy variable is not statistically significant; this result does not support the certification hypothesis of Megginson and

16

Since the standard deviation of the underpricing is very large in this paper, in order to reduce the bias caused by any outlier, we winsorized underpricing at 1%, 5%, and 20% level. However, even pursuant to doing so, the result is still not statistical significant.

20

ACCEPTED MANUSCRIPT

Weiss (1990). We believe however that this may be explained by Allen et al. (2005),

T

who suggests that China’s venture capital industry is underdeveloped, hence, its role in

IP

supporting the growth of start-up firms is very limited.17 Recent research by Johan and

SC R

Zhang (2015) also suggests that the recent global financial crisis as a market shock decreased the private equity exits probabilities in emerging markets, especially for the strategy to divest by IPOs. In view of our data spanning the years after the financial

NU

crisis, this may further explain the continued underdevelopment of China’s venture

MA

capital support of local firms.

TE

5.3. Post-IPO Performance

D



To determine the relation between political connection and firm performance,

CE P

we use Change in ROA as the performance measure. As referred to by Fan et al. (2007), the Change in ROA variable is measured as the difference between the average annual

AC

ROA of the two years after the IPO and that of two years before the IPO.18 As extant literature suggests, political connections bring value to firms, therefore we expect a positive effect as suggested by Hypothesis 3. Table 7 presents the results of OLS regressions that analyzes the effects of politically connected CEOs on post-IPO accounting performance changes. The independent variable includes the dummy variable for politically connected CEOs or other connected directors, the natural log of total assets, the leverage, and the

17

We thank our anonymous referee for suggesting a differential effect between domestic and foreign VCs. Pursuant to the referee’s suggestion, in our analysis we separated the VCs into foreign and domestic VC, and we found that the results did not change. 18 Since the sample period is from 2009 to 2012, we only can conduct a two year period performance change. For the ROA in year 2012, we use the accounting data on September, 2012.

21

ACCEPTED MANUSCRIPT

market-to-book-ratio. In support of our Hypothesis 3, the regression results show that

T

firms with politically connected CEOs or directors experience better accounting

SC R

IP

performance subsequent to their IPOs at a 1% significance level.



NU

6. Robustness Test

MA

To better control the other factors that may be related, in this section, we employ the propensity score matching method to further test the effect of political

D

connections on the propensity of passing the screening of CSRC. In regards to whether

TE

firms with good potential tend to attract more politically-connected directors, we tried to figure out when the CEO or board directors entered their political careers, and when the

CE P

politically connected CEO or board directors entered the firm, do they enter the firm first then they became the politician or they enter the politician first then they are

AC

invited into the firm to together prepare for IPO. It is difficult to collect these data. Therefore we conduct the propensity score matching to further check the effect of political connection. We first attend p-score to those firms with NPC-member on their board, then we find the firms have the similar p-score but do not have NPC-member on their board, which means they have similar firm quality except for one with NPC-member on board and one not. After this matching, we use this sample to run the regression, and finally we find the political connections still has significant effect even the firms have the similar quality.

22

ACCEPTED MANUSCRIPT

For the first step, we use the firms that have CEOs or board members that are

T

members of NPC (NPC dummy=1) as the treatment observations. Next, we attach a

IP

propensity score to each observation. The propensity score, defined as e() is defined as

SC R

e (Xi) = Pr (NPC dummy=1| Xi),

where Xi is a vector of the covariates in the probit estimation.

For each treatment observation, we identify the matched observations from the

NU

sample of firms that do not have a NPC member serving on a board. We use the nearest

MA

neighbour matches to match control group with treatment group. The matched observations demonstrate the “closest” propensity scores to a particular treatment

D

observation and are labeled as the control observations. After completing the propensity

TE

score matching, we conduct the baseline probit estimation. The results are showed in

CE P

Table 8.



Table 8 shows that the positive significant association between NPC

AC

membership and the propensity of passing the screening from CSRC remains unchanged. It is still significant at 10% level.

7. Conclusions Research has shown that political connections impact firm value in various ways, ranging from the firms’ process of privatization, financing strategies, borrowing cost, to long-run performance. However, how political connections relate to the IPO, one of the most important events in the life of a firm, has not been formally addressed. This paper tries to fill this gap by examining the role of political connection in a firms’ access to the public equity capital markets. More significantly, we examine the cost of 23

ACCEPTED MANUSCRIPT

IPO; we look at the direct cost of gross spread as well as the indirect cost of

T

underpricing.

IP

This paper indicates that political connections are valuable to firms during the

SC R

process of going public, especially within the particular institutional background of China. This implies that firms that lack political connections may have to accept less preferable treatment or seek other costly options in order to relax capital constraints.

NU

This paper provides support for the findings that firms in emerging market are more

MA

likely to establish political connections in order to achieve superior resources (Allen et al., 2005; Faccio, 2006; Fisman, 2001). This paper also contributes to the growing

D

literature (Field et al., 2013; Yang et al., 2011) that argues that the entrepreneurs’

TE

network is valuable for young firms and addresses that as one of the networks, political

AC

CE P

connections, is very valuable for start-ups in emerging market like China.

24

ACCEPTED MANUSCRIPT

References

IP

T

Agrawal, A., Knoeber, C.K. (2001). Do some outside directors play a political role? Journal of Law and Economics, 44(1), 179-198.

SC R

Allen, F., Qian, J., Qian, M.J. (2005). Law, finance, and economic growth in China. Journal of Financial Economics, 77(1), 57-116.

NU

Benveniste, L.M., Spindt, P.A. (1989). How investment bankers determine the offer price and allocation of new issues. Journal of Financial Economics, 24(2), 343-361.

MA

Benveniste, L.M., Wilhelm, W.J. (1990). A comparative analysis of IPO proceeds under alternative regulatory regimes. Journal of Financial Economics, 28(1-2), 173-207.

TE

D

Braun, M. (2014). The structure of equity markets across countries: Scarcity and stock valuations. Finance Research Letters, 11(4), 385-397.

CE P

Chang, E., Chen, C., Chi, J. and Young, M. (2008). IPO underpricing in China: New evidence from the primary and secondary markets. Emerging Markets Review, 9(1), pp.1-16.

AC

Chen, C.J.P., Li, Z., Su, X., Sun, Z. (2011). Rent-seeking incentives, corporate political connections, and the control structure of private firms: Chinese evidence. Journal of Corporate Finance, 17(2), 229-243. Chen, H.C., Ritter, J.R. (2000). The seven percent solution. The Journal of Finance, 55(3), 1105-1131. Chen, M.X. (2013) The matching of heterogeneous firms and politicians. Economic Inquiry, 51(2), 1502-1522. Chen, S., Sun, Z., Tang, S., Wu, D. (2011). Government intervention and investment efficiency: evidence from China. Journal of Corporate Finance, 17(2), 259-271. Correia, M. (2012). Political connections, SEC enforcement and accounting quality. Stanford University, Working Paper. 25

ACCEPTED MANUSCRIPT

IP

T

Cosh, A., Cumming, D.J., Hughes, A. (2009). Outside entrepreneurial capital. The Economic Journal, 119(October), 1494-1533

SC R

Cumming, D.J. (2008). Contracts and exits in venture capital finance. The Review of Financial Studies, 21(5), 1947-1982. Cumming, D.J., Johan, S.A. (2008). Information asymmetries, agency costs and venture

NU

capital exit outcomes. Venture Capital, An International Journal of Entrepreneurial Finance, 10(3), 197-231.

MA

Cumming, D.J., Johan, S. (2013). Listing standards and fraud. Managerial and Decision Economics, 34(7-8), 451-470.

D

Darby, M., Zucker, L. (2003). Growing by leaps and inchies: creative destruction, real cost reduction, and inching up. Economic Inquiry, 41(1), 1-19.

CE P

TE

Faccio, M. (2006). Politically connected firms. American Economic Review, 96(1), 369-386.

AC

Fan, J.P.H., Wei, K.C.J., X. Xu (2011). Corporate finance and governance in emerging markets: A selective review and an agenda for future research. Journal of Corporate Finance, 17(2), 207-214. Fan, J.P.H., Wong, T.J., Zhang, T. (2007). Politically connected CEOs, corporate governance, and post-IPO performance of China's newly partially privatized firms. Journal of Financial Economics, 84(2), 330-357. Fan, J., Rui, O.M., Zhao, M. (2008). Public governance and corporate finance: evidence from corruption cases. Journal of Comparative Economics, 36(3), 343-364. Field, L., Lowry, M., Mkrtchyan, A. (2013). Are busy board detrimental? Journal of Financial Economics, 109(1), 63-82. Fisman, R. (2001). Estimating the value of political connections. American Economic Review, 91(4), 1095-1102.

26

ACCEPTED MANUSCRIPT

IP

T

Francis, B.B., Hasan, I., Sun, X. (2009). Political connections and the process of going public: evidence from China. Journal of International Money and Finance, 28(4), 696-719.

SC R

Goldman, E., Rocholl, J., So, J. (2009). Do politically connected boards affect firm value? The Review of Financial Studies, 22(6), 2331-2360.

NU

Güçbilmez, Ufuk. (2015) IPO waves in China and Hong Kong. International Review of Financial Analysis 40, 14-26.

MA

Habib, M., Ljungqvist, A. (2001). Underpricing and entrepreneurial wealth losses in IPOs: theory and evidence. The Review of Financial Studies, 14(2), 433-458.

D

Haß, L, Johan, S., Schweizer, D. (2014). Is corporate governance in China related to performance persistence? Journal of Business Ethics, online.

CE P

TE

Hoang, H., Antoncic, B. (2003). Network-based research in entrepreneurship: a critical review. Journal of Business Venturing, 18(2), 165-187. Huergo, E., Jaumandreu, J. (2004). Firm's age, process innovation and productivity growth. International Journal of Industrial Organization, 22(4), 541-559.

AC

Johan, S. (2010). Listing standards as a signal of IPO preparedness and quality. International Review of Law and Economics, 30(2), 128-144. Johan, S., Zhang, M. (2015) Private Equity Exits in Emerging Markets. Working Paper. Available at SSRN: http://ssrn.com/abstract=2590128 or http://dx.doi.org/10.2139/ssrn.2590128 Johnson, S., Kaufmann, D., McMillan, J., Woodruff, C. (2000). Why do firms hide? Bribes and unofficial activity after communism. Journal of Public Economics, 76(3), 495-520. Johnson, S., Mitton, T. (2003). Cronyism and capital controls: evidence from Malaysia. Journal of Financial Economics, 67(2), 351-382.

27

ACCEPTED MANUSCRIPT

IP

T

Khwaja, A.I., Mian, A. (2005). Do lenders favor political connected firms? rent provision in emerging financial maket. The Quarterly Journal of Economics, 120(4), 1371-1411.

SC R

Klein, B., Leffler, K.B. (1980). The role of market forces in assuring contractual perfermance. Journal of Political Economy, 89(4), 615-641.

NU

Kutsuna, K., Smith, J.K., Smith, R. (2009). Public information, IPO price formation, and long-run returns: Japanese evidence. The Journal of Finance, 64(1), 505-546.

MA

La Porta, R., Lopez-de-Selanes, F., Shleifer, A., Vishny, R. (1999). The quality of government. Journal of Law, Economics and Organization, 15(1), 222-279.

D

Leland, H.E., Pyle, D.H. (1977). Informational asymmetries, financial structure and financial intermediation. The Journal of Finance, 32(2), 371-387.

CE P

TE

Li, H., Meng, L., Wang, Q., Zhou, L. (2008). Political connections, financing and firm performance: evidence from Chinese private firms. Journal of Development Economics, 87(2), 283-299.

AC

Li, H., Meng, L., Zhang, J. (2006). Why do entrepreneurs enter politics? evidence from China. Economic Inquiry, 44(3), 559-578. Liu, X., Ritter, J.R., (2011). Local underwriter oligopolies and IPO underpricing. Journal of Financial Economics, 102(3), 579-601. Megginson, W.L., Weiss, K.A. (1990). Venture capitalist certification in initial public offerings. The Journal of Finance, 46(3), 879-903. Millar, C.CJM., Eldomiaty, T.I., Choi, C.J, Hilton, B. (2005). Corporate governance and institutional transparency in emerging markets. Journal of Business Ethics, 59(1-2), 163-174. Ritter, J. (1987). The costs of going public. Journal of Financial Economics, 19(2), 269-281.

28

ACCEPTED MANUSCRIPT

T

Shan, Y.G. (2015). Value relevance, earnings management and corporate governance in China. Emerging Markets Review, 23, 186-207.

SC R

IP

Shleifer, A., Vishny, R. (1994). Politicians and firms. The Quarterly Journal of Economics, 109(4), 995-1025. Su, C. and Brookfield, D. (2013). An evaluation of the impact of stock market reforms on

NU

IPO under-pricing in China: The certification role of underwriters. International Review of Financial Analysis, 28, 20-33.

MA

Tian, L. (2011). Regulatory underpricing: determinants of Chinese extreme IPO returns. Journal of Empirical Finance, 18(1), 78-90.

D

Titman, S., Trueman, B. (1986). Information quality and the valuation of new issues. Journal of Accounting and Economics, 8(2), 159-172.

CE P

TE

Wang, Y. (2013). Determinants of IPO gross spreads: evidence from China. working paper. Wu, W., Wu, C., Rui, O. M. (2012a). Ownership and the value of political connections: evidence from China. European Financial Management, 18(4), 695-729.

AC

Wu, W., Wu, C., Zhou, C., Wu, J., (2012b). Political connections, tax benefits and firm performance: evidence from China. Journal of Accounting and Public Policy, 31(3), 277-300. Wu, W., Johan, S.A., Rui, O.M. (2014). Institutional investors, political connections, and the incidence of regulatory enforcement against corporate fraud. Journal of Business Ethics, online. Yang, Q., Zimmerman, M., Jiang, C. (2011). An emprical study of the impact of CEO characteristics on new firms' time to IPO. Journal of Small Business Management, 49(2), 163-184. Yu, F., Yu, X. (2012). Corporate lobbying and fraud detection. Journal of Financial and Quantitative Analysis, 46(6), 1865-1891. 29

ACCEPTED MANUSCRIPT

Table 1. Summary statistics of variables Table1 summarizes information of sample firms. Among them, 355 firms are successfully listed

T

on ChiNext during the sample period of October, 2009-October, 2012. The NPC dummy, LPC

Variable

Obs

Mean

Median

Min

Max

NPC dummy

413

0.19

0.00

0.39

0

1

LPC dummy

413

0.15

0.00

0.35

0

1

CPPCC dummy

413

0.23

0.00

0.42

0

1

% Board politician

413

0.07

0.00

0.49

0

0.40

% Board professional

413

0.29

0.29

0.13

0

0.78

Firm age

413

11.49

11.00

3.69

4

25.00

Size (in million yuan)

413

413.76

244

709

67

12354

Leverage

413

0.39

0.39

0.16

0.03

0.97

Growth rate

413

0.84

0.54

0.98

0.05

10.21

Founder CEO Dummy

413

0.74

1.00

0.44

0

1

CEO age

413

47.92

47.00

6.63

30

76

Ownership reduction

413

11.53

10.92

4.94

0.34

48

413

0.52

1.00

0.50

0

1

413

0.75

1.00

0.43

0

1

355

7.33

7.00

2.32

2.09

14.21

355 355

10.92 34.42

10.86 25.24

0.54 37.31

9.75 -16.68

12.45 209.73

VC dummy Gross spread

NU

MA D

AC

Proceeds Underpricing

TE

Top10 U dummy

SC R

S.D.

CE P

study. The definition of each variable is described in Appendix I.

IP

dummy, CPPCC dummy and % Board politician are proxies for political connections in this

30

ACCEPTED MANUSCRIPT

TE

Panel B: By industry Manufacturing Electronics Machinery Metals & Non-metals Pharmaceutical

D

MA

NU

SC R

IP

T

Table 2. Distribution of successfully listed issuances and rejected issuances Panel A shows the distribution of successfully listed issuances and rejected issuances that filed with the CSRC over the period from 2009-2012. The N. of listed issues (rejected issues) and N. of political connected represents the number of listed issues and political issues. The distribution is divided into year basis in Panel A. The breakdown of the sample of all the Issues by industry classification is presented in Panel B. The industry classification is published by CSRC in 2009. Rejected Issues Total sample Successful Issues N. of N. of % of N. of % of N. of % of Number listed politically politically rejected being political politically issues connected connected issues rejected connected connected Panel A: By year 2009 45 36 26 72% 9 20% 4 44% 2010 131 117 83 71% 14 11% 4 29% 2011 144 128 99 77% 16 11% 7 44% 2012 93 74 28 38% 19 20% 4 21%

80 74 15 25

31 31 7 10

39% 42% 47% 40%

5 12 2 3

6% 14% 12% 11%

3 5 0 1

60% 42% 0% 33%

Petrochemicals Other manufacturing Agriculture Construction

40 11 10 8

37 8 7 6

20 3 4 3

3 3 3 2

6

5

3

IT Media Research & Development Others

89 7 6 20

70 6 6 16

11 2 1 16

19 1 0 4

8% 27% 30% 25% 17% 21% 14% 0% 20%

2 1 1 1

Environmental Protection

54% 38% 57% 50% 60% 16% 33% 17% 100%

3 0 0 1

67% 33% 33% 50% 100% 16% 0% 0% 25%

Total

413

355

142

40%

58

14%

19

33%

AC

CE P

85 86 17 28

31

1

1

ACCEPTED MANUSCRIPT

levels respectively.

SC R

IP

T

Table 3. Mean difference of firm traits The sample compared the basic attributes of the 355 successfully listed firms and 58 rejected firms that filed with the CSRC from 2009 to 2012. The columns named “Difference in mean” report the t-values of the t-test of equality of means (Wilcoxon rank-sum test for equality of medians) between successfully listed start-ups and rejected start-ups. Asterisks indicate whether the differences between listed firms and rejected firms are significant: ***, **, and * designate significance at the 1%, 5%, and 10% Panel A: Differences between successfully listed firms and rejected firms

NU

Mean 0.19 0.15 0.23 0.07 11.49 19.54 0.39 0.84 0.74 47.9 8.37

D

Number 413 413 413 413 413 413 413 413 413 413 413

TE

NPC dummy LPC dummy CPPCC dummy % Board politician Firm age Size Leverage growth rate Founder CEO dummy CEO age Board size

Listed Issues (a) N Mean 355 0.21 355 0.15 355 0.24 355 0.07 355 11.63 355 19.54 355 0.39 355 0.85 355 0.75 355 47.92 355 8.37

MA

Full Sample

Rejected Issues (b) Mean N 0.05 58 0.14 58 0.21 58 0.05 58 10.67 58 19.54 58 0.37 58 0.79 58 0.64 58 47.90 58 8.36 58

Difference in mean (a)-(b) 0.16 *** 0.01 0.03 0.02 0.95 * 0.00 0.01 0.46 0.11 * 0.02 0.00

NPC Non-NPC Difference Mean Mean in mean 12.35 11.29 1.05 ** 19.74 19.49 0.25 *** 0.43 0.37 0.06 *** 0.79 0.85 -0.06 0.81 0.72 0.09 49.24 47.61 1.64 ** 8.67 8.30 0.37 ** 7.43 6.95 -0.49 34.81 34.31 0.49

AC

Firm age Size Leverage growth rate Founder CEO dummy CEO age Board size Gross spread Underpricing

CE P

Panel B: Differences between political connected firms and unconnected firms LPC Mean 12.08 19.82 0.43 0.86 0.75 49.15 8.58 6.98 30.72

32

Non-LPC Difference in Mean mean 11.39 0.69 19.49 0.33 *** 0.38 0.05 ** 0.84 0.02 0.73 0.02 47.71 1.44 8.33 0.25 7.39 -0.41 35.05 -4.34

CPPCC Non_CPPCC Difference in Mean Mean mean 11.84 11.39 0.45 19.76 19.47 0.29 *** 0.42 0.37 0.05 *** 0.87 0.83 0.04 0.79 0.72 0.08 48.62 47.70 0.92 8.40 8.35 0.05 7.03 7.43 -0.40 29.83 35.86 -6.03

ACCEPTED MANUSCRIPT

Table 4. The effect of political connections on the propensity of passing the screening from CSRC

T

This table reports the probit estimation of listing probability on different type of political

IP

connections. The dependent variable is a dummy variable equals to 1 if the firm has successfully listed on ChiNext, and 0 otherwise. % Board politician is measured on the

SC R

percentage of political directors on the board of directors at the time of IPO, % Board professional is defined as CEO or directors with accounting, law, or finance backgrounds. Top10 U dummy refers the top 10 underwriters calculated by their market share 3 years prior to the IPO year. Definitions of other variables are provided in Appendix I. ***, **, and * denote

NU

significance at the 1%, 5%, and 10% level, respectively. (1) 0.73 (2.97) **

MA

NPC dummy LPC dummy CPPCC dummy

Size Leverage

AC

Growth rate

CE P

Firm age

Founder CEO dummy CEO age

2

CEO age

Ownership reduction Top10 U dummy VC dummy Year dummies Industry dummy Constant Observations Pesudo R2

TE

% Board professional

D

% Board politician

(2)

(3)

0.02 (0.11) 0.05 (0.33)

1.27 (2.75) ** 0.04 (1.49) -0.07 (0.4) 0.17 (0.21) 0.07 (0.55) 0.22 (1.35) -0.18 (2.31) * 0.0018 (2.4) * -0.003 (0.26) -0.13 (0.91) 0.02 (0.14) YES YES 5.63 (1.38) 413

1.31 (3.15) ** 0.04 (1.71) -0.05 (0.3) 0.24 (0.29) 0.06 (0.48) 0.24 (1.51) -0.18 (2.04) * 0.0018 (2.14) * -0.007 (0.54) -0.12 (0.73) 0.01 (0.09) YES YES 5.07 (1.33) 413

1.30 (3.15) ** 0.04 (1.72) -0.05 (0.32) 0.23 (0.28) 0.06 (0.48) 0.24 (1.5) -0.18 (2.00) * 0.0018 (2.10) * -0.007 (0.55) -0.12 (0.74) 0.01 (0.05) YES YES 5.11 (1.32) 413

0.08

0.06

0.06

33

(4)

1.15 (2.15) 1.27 (3.04) 0.04 (1.69) -0.08 (0.46) 0.23 (0.28) 0.05 (0.43) 0.23 (1.45) -0.18 (2.10) 0.0018 (2.16) -0.009 (0.71) -0.11 (0.73) 0.00 (0.00) YES YES 5.74 (1.48) 413 0.06

* **

* *

ACCEPTED MANUSCRIPT

Table 5. The effect of political connections on Gross Spread This table reports the regression estimates of gross spread on different type of political

T

connections. The dependent variable is Gross spread, calculated by the total underwriting cost

IP

divided by total proceeds. % Board politician is measured on the percentage of political directors on the board of directors at the time of IPO, top10 U dummy refers the top 10

SC R

underwriters calculated by their market share 3 years prior to the IPO year. Definitions of other variables are provided in Appendix I. ***, **, and * denote significance at the 1%, 5%, and 10% level, respectively. (1)

NPC dummy

-0.05 (0.39)

MA

LPC dummy

(2)

CPPCC dummy

D

% Board politician

Ownership reduction

VC dummy Proceeds

AC

Year dummies Industry dummy Constant

CE P

Top10 U dummy

Observations R-squared

-0.01 (0.66) -0.69 (1.04) 0.39 (3.78) ** 0.20 (1.44) -2.74 (26.85) *** YES YES 35.88 (28.68) *** 355 0.517

(4)

-0.18 (1.27)

-0.01 (0.6) -0.66 (1.08) 0.40 (3.86) ** 0.20 (1.44) -2.75 (28.37) *** YES YES 35.92 (30.89) *** 355 0.518

TE

Firm age

(3)

NU

Gross spread

34

-0.15 (2.81) *

-0.01 (0.62) -0.80 (1.02) 0.40 (3.92) ** 0.21 (1.48) -2.73 (27.66) *** YES YES 35.83 (29.11) *** 355 0.518

-0.83 (3.21) ** -0.01 (0.55) -0.63 (0.76) 0.39 (3.9) ** 0.21 (1.49) -2.73 (27.78) *** YES YES 35.78 (29.12) *** 355 0.519

ACCEPTED MANUSCRIPT

Table 6. The effect of political connections on Underpricing This table reports the regression estimates of underpricing on different type of political

T

connections. The dependent variable is underpricing, calculated by the closing price in first day

IP

of trading minus the offer price, then divided by offer price. % Board politician is measured on

SC R

the percentage of political directors on the board of directors at the time of IPO, top10 U dummy refers the top 10 underwriters calculated by their market share 3 years prior to the IPO year. Prior 30 market return is the average market stock return 30 days prior to firm's IPO date. Definitions of other variables are provided in Appendix I. ***, **, and * denote significance at

Underpricing NPC dummy

(1) -0.33 (0.08)

(2)

CPPCC dummy

-0.06 (0.15) -4.51 (1.67) 7.37 (0.68) 3.31 (4.34) *** -6.76 (1.74) 0.19 (0.99) -7.06 (2.93) ** 4.94 (1.42) 41.67 (10.36) *** YES YES 151.00 (3.31) ** 355 0.462

-0.05 (0.11) -4.12 (1.52) 7.03 (0.67) 3.33 (4.40) *** -6.66 (1.74) 0.22 (1.11) -7.06 (2.90) ** 5.01 (1.44) 41.13 (9.50) *** YES YES 143.67 (3.06) ** 355 0.463

TE

Firm age

D

% Board politician

Gross spread

CE P

Size Leverage

AC

Founder CEO dummy Ownership reduction Top10 U dummy VC dummy Prior 30 market return Year dummies Industry dummy Constant Observations R-squared

(3)

(4)

-3.50 (1.18)

MA

LPC dummy

NU

the 1%, 5%, and 10% level, respectively.

35

-0.98 (0.43)

-0.06 (0.15) -4.44 (1.65) 7.52 (0.69) 3.30 (4.30) *** -6.68 (1.72) 0.19 (1.02) -7.03 (2.91) ** 4.98 (1.43) 41.67 (10.59) *** YES YES 149.62 (3.26) ** 355 0.463

-9.02 (0.71) -0.05 (0.12) -4.25 (1.58) 7.74 (0.71) 3.30 (4.33) *** -6.58 (1.71) 0.21 (1.08) -7.05 (2.91) ** 5.01 (1.44) 41.61 (10.62) *** YES YES 145.96 (3.14) ** 355 0.463

ACCEPTED MANUSCRIPT

Table 7. The effects of Political connections on the long-term operating performance

T

The dependent variable in this table is the Change in ROA, which is measured as the difference

IP

between the average annual ROA of the two years after the IPO and that of the two years before the IPO. ROA is defined as net profits divided by total asset of each year. The independent

SC R

variables, measured upon the IPO year, include a dummy variable equals to 1 if the CEO is a NPC (LPC, CPPCC), and 0 otherwise, the natural log of total assets, the leverage and the market to book ratio. ***, **, and * denote significance at the 1%, 5%, and 10% level, respectively.

(1) 0.009 (2.82) *

TE

% Board politician

CE P

Size

Market to book value

AC

Year dummies Industry dummy Constant Observations R-Squared

(4)

0.008 (2.26) *

D

CPPCC dummy

Leverage

(3)

0.017 (4.63) **

MA

LPC dummy

(2)

NU

Change in ROA NPC dummy

0.012 (1.36) 0.191 (6.59) *** 0.010 (0.83) YES YES -0.45 (2.62) * 355 0.25

0.012 (1.28) 0.192 (6.64) *** 0.010 (0.9) YES YES -0.441 (2.56) * 355 0.26

36

0.013 (1.47) 0.191 (6.6) *** 0.011 (0.85) YES YES -0.453 (2.85) * 355 0.25

0.053 (5.65) *** 0.012 (1.36) 0.189 (6.69) *** 0.010 (0.89) YES YES -0.439 (2.71) * 355 0.25

ACCEPTED MANUSCRIPT

Table 8. Robustness test

SC R

IP

T

This table reports the probit estimation of listing probability on NPC dummy after conducting the propensity score matching. The dependent variable is a dummy variable equals to 1 if the firm has successfully listed on ChiNext, and 0 otherwise. NPC dummy is a dummy variable equals to 1 if the CEO or board director is a member of national people congress. % Board professional is defined as CEO or directors with accounting, law, or finance backgrounds. Top10 U dummy refers the top 10 underwriters calculated

NU

by their market share 3 years prior to the IPO year. Definitions of other variables are provided in Appendix I. ***, **, and * denote significance at the 1%, 5%, and 10% level, respectively.

MA

NPC dummy % Board professional

TE

D

Firm age

Leverage Growth rate

CE P

Size

AC

Founder CEO dummy CEO age

CEO age2 Ownership reduction Top10 U dummy VC dummy Constant Observations Pesudo R2 37

(1) 0.94 (1.92) * 1.13 (1.07) 0.01 (0.23) -0.25 (0.94) -0.34 (0.32) 0.06 (0.5) -0.62 (2.21) ** -0.42 (1.72) 0.0037 (1.51) 0.053 (1.97) ** -0.12 (0.43) -0.14 (0.47) 15.92 (2.12) 112 0.17

ACCEPTED MANUSCRIPT

Appendix I. Variable definitions Equal to 1 if the CEO or other board member is a member of National People’s Congress and 0 otherwise.

LPC dummy

Equal to 1 if the CEO or other board member is a member of Local People’s Congress and 0 otherwise.

CPPCC dummy

Equal to 1 if the CEO or other board member is a member of Chinese People’s Political Consultative Conference and 0 otherwise.

% Board politician

The percentage of board memebrs with a political background at the time of IPO. The percentage of directors who have accounting, law, finance or academic background serving on the board.

NU

% Board professional

SC R

IP

T

NPC dummy

Age of the company in the year of IPO.

Size (in million yuan)

Firm asset showed in million yuan.

Size Leverage

The natural logarithm of the assets in the year prior to IPO. It is calculated based on the Size (in million yuan). The debt ratio (total liabilities/assets) in the year of IPO.

Growth rate

Average growth rate of profitablity over three years prior to IPO .

Founder CEO dummy

Equal to 1 if the Founder remain as the CEO of the firm in the IPO year.

CEO age

The age of the CEO of the firm.

Ownership reduction Top10 U dummy

The ownership selled by owner during the IPO. Equal to 1 if the underwriter's market share is belong to top 10 and 0 otherwise, the reputation ranking is according to their 3 years average market share prior to the IPO year. Equal to 1 if the issuing firm is backed by Venture Capital.

AC

Gross spread

D

TE

CE P

VC dummy

MA

Firm age

Proceeds Underpricing

Market to book value Prior 30 market return

Calculated by the total underwriting cost devided by total proceeds The natural logarithm of Proceeds of the issuing company. Calculated by the closing price in first day of trading minus the offer price, then divided by offer price A ratio calculated by the market value of the firm divided by the book value of the firm. The average market stock return 30 days prior to firm's IPO date.

38

ACCEPTED MANUSCRIPT

Appendix II. Correlation coefficients of variables

T

1.000 0.113 0.032 0.009 0.075 -0.101 -0.063 -0.040 -0.078

1.000 0.017 -0.039 -0.114 -0.067 0.022 -0.186 0.057

NU

1.000 0.523 0.059 -0.028 0.122 -0.038 0.063 0.066 -0.312 -0.232

MA

1.000 0.094 0.105 -0.079 0.074 0.136 -0.005 0.000 0.006 0.001 0.019

SC R

IP

Firm age

Founder Growth CEO CEO Ownership Top10U VC Gross Size Leverage rate dummy age reduction dummy dummy Spread Underpricing

AC

CE P

TE

D

% NPC LPC CPPCC Board dummy dummy dummy politician NPC dummy 1.000 LPC dummy 0.274 1.000 CPPCC dummy 0.502 0.178 1.000 % Board politician 0.786 0.586 0.782 1.000 Firm age 0.109 0.087 0.053 0.102 Size 0.178 0.161 0.207 0.214 Leverage 0.165 0.094 0.157 0.183 Growth rate -0.026 -0.051 -0.025 -0.063 Founder CEO dummy 0.073 0.035 0.093 0.100 CEO age 0.100 0.064 0.022 0.090 Ownership reduction -0.029 0.119 0.049 0.105 Top10U dummy 0.008 0.021 0.058 0.019 VC dummy 0.022 0.014 0.059 0.034 GrossSpread -0.086 -0.062 -0.074 -0.093 Underpricing 0.005 -0.041 -0.069 -0.059

39

1.000 -0.117 -0.001 0.015 0.007 0.096 -0.140

1.000 -0.016 -0.026 -0.034 -0.035 0.010

1.000 0.068 1.000 -0.016 0.126 1.000 0.047 -0.002 0.062 1.000 -0.008 -0.080 0.049 0.031 1.000

ACCEPTED MANUSCRIPT

Figure 1: Description of the IPO process

IP

T

1.Due Deligience by main underwriter (1~3years)

SC R

2. Prepare application ducuments together with underwriters. Sponsors under the main underwriter will submit the application ducuments to CSRC.

NU

2.Within 5 working days, CSRC decide whether to receivethe application or not.

D

MA

3.After receipt of the application documents, issuers make predisclosure of their preliminary prospectus on CSRC’s website.

CE P

TE

4.Then the Public Offering Review Committee commences review.

5.Within 3 months, CSRC will render its decision on whether the issuer can go public and make relevant announcement on its website.

AC

6. After receipt of this approval, the firm starts road show and prepares for listing on ChiNext. (One approval valid for 6 months)

40

To be rejected