University of Southern California Dissertations and Theses

Political connections and access to bond capital: Reputation or collusion?

(USC Thesis Other)

Political connections and access to bond capital: Reputation or collusion?

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

POLITICAL CONNECTIONS AND ACCESS TO BOND CAPITAL:
REPUTATION OR COLLUSION?

by

Fei Du

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BUSINESS ADMINISTRATION)

May 2011

Copyright 2011 Fei Du

ii

Acknowledgements
I gratefully thank my dissertation committee: Mark Young (co-chair), Sarah
Bonner (co-chair), Tatiana Sandino, and Jack McArdle (outside member) for their timely
feedback, insightful advice, and continuous support. They have guided me through my
PhD journey and made sure that I did not get lost on the way. Thank you, Mark, for
believing in my potential and capabilities which gave me the confidence that I badly
needed, for your good cheer, and for your inspiring devotion to quality accounting
research. Thank you, Sarah, for training me and coaching me to design my empirical
work carefully, for holding me to high standards of working ethics, and for being a role
model of a productive and responsible accounting scholar. Thank you, Tatiana, for
always opening doors to answer my myriad questions and for offering frank and
constructive feedback to overcome my weaknesses. Thank you, Jack, for offering your
expertise and insights in statistics when I needed it.
I also appreciate helpful comments made by Charles Lee, Clive Lennox, Claudia
(Zhen) Qi, T. J. Wong, Guliang Tang, and workshop participants at the Hong Kong
University of Science and Technology, the Chinese University of Hong Kong, the
University of Hong Kong, Stanford University, and University of Southern California. A
special thanks to Professor Guliang Tang, who served as my mentor at the outset of my
journey in doing management accounting research in China, for his generosity in sharing
his connections in China with me, for providing access to interview investment bankers
and CFOs during the early stage of my dissertation. Without his help, getting the
necessary institutional details would have been much more difficult.
iii

I gratefully acknowledge financial support from the Leventhal School of
Accounting and Marshall School of Business at University of Southern California. I
thank Deloitte Foundation for a doctoral fellowship in my dissertation stage.
The completion of my dissertation work would have not been possible without the
constant support from my family and my friends. I owe it to my mom, Keke Li, for her
unconditional love and unwavering belief in me. I am grateful to my husband, David
Chohing Mo, not only for his commitment, but more importantly, for his friendship and
for the many, many hours he spent on talking with me trying to understand every piece of
my dissertation. I thank my good friend and colleague Claudia (Zhen) Qi, who provides
enormous intelligent support and emotional support during my dissertation stage. Most
importantly, I thank my lovely son, Davis Tocheuk Mo, for coming into my life and
giving me the main source of strength to complete my dissertation work.
iv

Table of Contents
Acknowledgements ................................................................................................. ii

List of Tables.......................................................................................................... vi

Abstract ................................................................................................................. vii

Chapter 1. Introduction .............................................................................................1

Chapter 2. Institutional Background ..........................................................................7
2.1. Primary Market ......................................................................................9
2.2. Secondary Market ..................................................................................9
2.3. Major Bond Instruments ....................................................................... 10

Chapter 3. Literature Review .................................................................................. 14
3.1. Formation and Measurement of Political Connections. ......................... 14
3.2. Benefits and Costs of Political Connections .......................................... 15
3.3. Channels Through Which Political Connections .................................. 17
Influence Firm Value
3.4. Institutional Features Favoring the Emergence and Survival of ............. 18
Political of Political Connections
3.5. Political Connections and External Financing ....................................... 22

Chapter 4. Theory and Hypotheses ......................................................................... 25
4.1. The Reputation Enhancement Argument .............................................. 25
4.2. The Social Lending Argument .............................................................. 27
4.3. The Collusion Argument ...................................................................... 28
4.4. Political Connections of Bank Underwriters and Issuing Firms’ .......... 30
Access to Bond Capital

Chapter 5. Empirical Design and Analyses ............................................................. 33
5.1. Sample, Constructs, and Measures........................................................ 33
Figure 1. Rating Regime, Symbols, and Definitions .................................... 36
5.2. Regression Models ............................................................................... 39
5.3. Empirical Results ................................................................................. 44
5.4. Additional Analyses ............................................................................. 57
5.5. Robustness Checks ............................................................................... 72

v

Chapter 6. Conclusion and Discussion .................................................................... 74
6.1 Summaries and Conclusions .................................................................. 74
6.2 Contributions ........................................................................................ 76
6.3 Limitations and Future Research ........................................................... 78

Bibliography ........................................................................................................... 81

vi

List of Tables
Table 1: Major Events in Bond Markets Developments in China .............................. 8
Table 2: Trading Volume and Outstanding Balance of Chinese Bond Market ......... 12

Table 3: Descriptive Statistic .................................................................................. 46

Table 4: Associations Between Political Connections of Issuing Firms ................... 47
and Issue Size

Table 5: Associations Between Political Connections of Issuing ............................. 49
Firms and Issuer Ratings

Table 6: Tests of Substitution Effect Between Political Connections ...................... 51
of Issuing Firms and Political Connections of Lead Arranger
Underwriters in Affecting Offering Amounts

Table 7: Tests of Substitution Effect between Political Connections ...................... 53
of Issuing Firms and Political Connections of Lead Arranger
Underwriters in Affecting Entity Ratings

Table 8: Associations Between Political Connections of Issuing Firms and ............ 55
Political Connections of Lead Arranger Underwriters

Table 9: Associations Between Political Connections of Issuing Firms and............. 59
Issue Size in Non-PLCs subsample and PLCs Subsample

Table 10: Associations between Political Connections of Issuing Firms .................. 61
and Issue Size in non-Beijing and Beijing Headquartered Firms

Table 11: Associations Between Political Connections of Issuing Firms and .......... 63
Issue Size in non-SOEs Subsample and SOEs Subsample

Table 12: Interaction Test of Corporate Governance, Organizational ..................... 69
Complexity, and Political Connections of Issuing Firms

Table 13: Associations Between Corporate Governance, Organizational ................ 71
Complexity, Financial Performances of Issuing Firms and Political
Connections of Issuing Firms
vii

Abstract
This study addresses how political connections influence a firm’s access to bond
capital using a database of 1672 new bond issuances in China from 2001 to 2009. In this
context, firm political connections are measured using a score based on the highest
bureaucratic position held by a firm executive, with the highest level of political
connections attributed to those who have held top positions in the central government, an
intermediate level attributed to those in provincial and municipal governments, and the
lowest level attributed to those who have not held any government position. Results
suggest that firm political connections are positively associated with debt offering
amounts and issuer credit ratings, but only in the subsample of firms that have poor
information environments, such as non-publicly listed firms and non-Beijing
headquartered firms, thus lending support to the argument that political connections
contribute to firm reputation. The role of political connections in providing preferential
access to debt is relevant to both state-owned enterprises and privately held firms. In
addition, issuing firm political connections and bank underwriter political connections
serve as alternative mechanisms when explaining the variation of offering amounts and
issuer credit rating.
1

Chapter 1. Introduction
Firms with strong political connections are more likely to get preferential access
to capital, especially in emerging markets and corrupt countries (Johnson and Mitton
2003; Khwaja and Mian 2005; Charumilind et al. 2004; Claessens et al. 2008).
Nevertheless, Chaney et al. (2010) find that politically connected firms have lower
accounting quality. Low accounting quality can result in a number of negative
consequences, including a higher cost of capital (Francis et al. 2005), which begs the
question: what is the mechanism by which political connections facilitate a firm’s effort
to raise capital? Is it because managers’ political connections are one dimension of
management quality that influences the judgments of regulators, ratings analysts, and
market participants? Is it possible that political pressure and intervention on behalf of
connected parties substitute for better quality disclosures? To date there is little empirical
evidence to substantiate either of these arguments, likely because of the lack of data
regarding managers’ political connections. To explore this question, this paper uses a
unique, manually constructed database to study whether and how the political
connections of issuing firms influence offering amounts and issuer credit ratings in the
bond issuance process. This paper also explores the effects of the political connections of
bank underwriters on bond capital access.
At least three mechanisms are possible when explaining the positive relationship
between political connections and a firm’s ability to raise capital. First, the relationship
may be explained using a reputation enhancement argument, which suggests that the
political connections of firm executives serve as an alternative channel for establishing
2

firm reputation when quality disclosure is absent. This helps firms gain easier access to
debt capital. Second, this relationship could be explained through a social lending
argument; that is, the bond issuance process is guided by the government’s efforts to
improve social welfare, and the government assigns bureaucrats to firm executive teams
in order to better monitor firms’ use of financial resources that are granted to perform
these socially beneficial projects. Finally, a collusion argument could explain the relation.
Firms use their connections to engage in activities that will influence the government’s
bond issuance approval decisions, and the government shows partiality to firms whose
executives promise to return politicians’ personal favors. In other words, the bond
issuance process is influenced by the desire of individual politicians to seek rents.
These arguments are not mutually exclusive. This paper attempts to determine
which of these arguments is the most plausible. First, if the reputation enhancement
argument holds, that is, if political connections serve as substitutes for quality disclosure
in establishing firm reputation, then I expect that political connections will affect bond
offering amounts in the subsample of firms operating in a poor information environment,
but not in the subsample of firms that operate in a quality information environment. To
implement this test, I sort firms according to listing status and the location of firm
headquarters. Results show that the effect of political connections is significant only for
firms that operate in a poor information environment, such as non-publicly listed firms
and firms that are not headquartered in Beijing, thus lending support to the reputation
enhancement argument.
3

Second, if the social lending argument holds and the government rewards firms
that involve socially beneficial projects with preferential access to capital, then I expect
to find that political connections influence offering amounts in the state-owned
enterprises (SOEs) subsample, but not in the non-SOEs subsample, since the government
is more likely to engage SOEs in socially beneficial projects but not non-SOEs. Results
suggest that the relationship between political connections and firm access to capital is
robust across the SOEs subsample and the non-SOEs subsample. This finding allows me
to rule out that the idea that the observed data pattern is simply due to the social lending
argument.
Third, if the collusion argument holds and politicians and issuing firms collude
during the bond issuance process, the effect of political connections on offering amounts
with politically influential executives will be stronger in firms with poor corporate
governance mechanisms. This is because managers in firms with poor governance are
better able to return favors from politicians. Results are not consistent with the collusion
argument.
I chose the bond issuance market in China as my research setting for three
reasons. First, the Chinese government employs a range of regulations to oversee the
bond issuance process, including placing restrictions on issuers’ qualifications, a “merit-
based” approval process, issuance quotas, and interest rate controls. Issuing firms have to
meet certain thresholds before issuing bonds; however, meeting these criteria does not
guarantee approval for issuance from government regulators. These restrictions increase
competition for capital among issuing firms. In turn, such competition is expected to
4

change firm behavior when raising capital, making political connections particularly
relevant. Although these are country-specific regulations, the inferences drawn from this
setting are generalizable to highly regulated markets with government intervention.
Second, in the bond issuance process, underwriters work with issuing firms to get
approval from China Securities Regulation Committee (CSRC) to get the “issue quota.”
Thus, in addition to the issuing firms’ political connections, the connections of bank
underwriters may explain firm offering amounts and credit ratings. Third, both publicly
listed companies (PLCs) and non-publicly listed companies (non-PLCs) can issue bonds,
and the variation in the information environment across PLCs and non-PLCs provides a
natural setting for studying whether the political connections of issuing firms
(management quality) serve as potential substitutes for quality financial information
when establishing firm reputation.
The political connections examined in this study derive from the career histories
of SOE executives as officials of the Chinese government. The highest level of political
connection is attributed to those who have held top positions in the central government,
followed by individuals in provincial and municipal governments. I improve upon the
relatively crude measures of political connections used in prior research that treat
political connections simply as a dichotomous variable. I manually collect detailed
resume data for top management team members of issuing firms, and calculate political
connection measures as the sum of the score of the highest career title an individual had
earned in his or her political career before joining the boards of directors of an issuing
firm and the score of the highest rank of government office the person has served.
5

This study also contributes to the literature in several ways. First, prior studies
that examine the relationship of political connections to firm reporting behavior (Chaney
et al. 2010) and to firm financing behavior (Faccio et al. 2006) have established the link
between political connections and preferential capital access without explaining the
mechanism through which political connections contribute to preferential capital access. I
use the Chinese bond issuance market research setting to test three theoretical arguments
that may contribute to this phenomenon: the reputation enhancement argument, the social
lending argument, and the collusion argument. Second, prior studies that discuss factors
that may influence credit ratings have mainly focused on the financial reporting of
information (Jiang 2008) and on corporate governance variables (Ashbaugh-Skaife et al.
2006). According to Moody’s rating report, managerial accounting information,
management quality and organizational structure, are also factors considered by analysts,
when making rating decisions. This study suggests that one important dimension of
management quality– the political connections of executives— can also contribute to
issuer credit ratings.
1
Third, I contribute to the literature on bond underwriter choice by
establishing how the political connections of issuing firms and political connections of
bond underwriters can serve as substitute mechanisms for obtaining government approval
to issue bonds.

1
The corporate political activities literature suggests that directors’ political connections can capture the
breadth of board members’ skills and expertise brought from previous public service. For example, senators
have a wider range of constituents and legislative issues in which they are called to play active decision-
making roles than representatives, who are more specialized in political expertise and geographic influence
(Baker, 1989; Oleszek, 1974). On the other hand, cabinet members possess a great breadth of human and
social capital because they interact with a wide number of constituents and other political decision makers
on both national and international stages (Cohen, 1988; Mann & Smith, 1981).

6

The rest of this dissertation is organized as follows: Chapter 2 provides
institutional details regarding the overall bond market structure and trading instruments in
China. Chapter 3 reviews the political connections literature. Chapter 4 presents the
theoretical framework motivating my investigation, and explains my research hypotheses.
Chapter 5 discusses empirical design models and presents analyses and results. Chapter 6
concludes the dissertation with a summary, discussion, and suggestions for future
research.
7

Chapter 2. Institutional Background
Although corporate bonds represent an important source for stable, long-term
capital, Chinese enterprises still heavily rely on bank loans for capital, which brings in
big risks to the Chinese banking system. Realizing the need to speed up the development
of this market, the Chinese government has enacted regulatory reform such as lifting the
interest cap, establishing a regulated rating system for better risk disclosure, and
introducing listed company bonds and short-term financing bills.
8

Table 1: Major Events in Bond Markets Developments in China
Time Event
1988 Ministry of Finance started to issue treasury bonds in 61 cities in China as an
experiment.

1990
(December)
Establishment of the Shanghai Stock Exchange, which permitted trading of bonds.

1994 Short sell of treasury bonds were permitted, and raised market risk substantially.
T-bond futures contracts were permitted to trade.

1995 Speculations and irregularities in T-bond trading led to the closure of regional T-
bond trading centers, e.g., the Wuhan Trading Center, and T-bond features
contracts were banned.

1995 (August) OTC trading of T-bonds was stopped, and the Shanghai and Shenzhen Stock
Exchanges became the only legal trading platforms.

1996 Book-entry bonds were issued in the Shanghai and Shenzhen Stock Exchange, and
the bond trading system was established with the increasing volume in re-purchase
transactions.

1998 (May) People’s Bank of China started OMOs, which stimulated the development of
interbank bond market of bond/notes trading.

1998
(September)
China Development Bank started to issue financial bonds in the interbank bond
market.

1998 (October) Peoples’ Bank of China approved insurance companies to become members of the
interbank bond market.

1999 (October) A portion of securities firms and all asset management firms became members of
the interbank bond market.

2000
(September)
Financial firms were allowed by People’s Bank of China to become members of
the interbank bond market.

2002 (October) Non-financial institutions were allowed by People’s Bank of China to become
members of the interbank bond market.

2005 Short-term corporate notes were allowed to be issued in the interbank bond
market.

2007 (January) The State Council allowed financial institutions to issue CNY denominated bonds
in Hong Kong.

2007 (March) Corporate pensions funds were allowed to trade in the interbank bond market.
9

2.1. Primary market
The primary market for bond issuance is largely completed through syndication.
Governmental bonds were mostly underwritten by the four state-owned banks, whereas
commercial banks and securities companies played an active role in forming syndicate-
to-market financial and corporate bonds. Starting in late 2002, China loosened regulation
on foreign investors investing in treasury bonds. Launching Qualified Foreign
Institutional Investor (QFII) offered foreign capital an opportunity to invest in the
Chinese bond market. International Finance Corporation (IFC) and Asian Development
Bank (ADB) were allowed to issue CNY-denominated bonds in China in 2005.
New bond issuances require credit rating reports for the issuer and the issue from
one of the five top Chinese rating agencies. All five national rating agencies are linked to
the big international rating agencies.
2

2.2. Secondary market
The market has a multi-layered structure, comprised of the national interbank
market, the exchange market, and the over-the-counter (OTC) market. Foreign
institutional investors can invest in Chinese bonds by seeking regulatory approval for
QFII quota or access to the interbank market.
The interbank market absorbs 95% of trading. “Interbank” is a misnomer, since
the market’s roughly 6,000 registered participants include end-users such as mutual funds,

2
By the end of 2006, Moody’s has gained the majority control right of Zhong Chengxin Global Credit
Rating Company; Fitch has gained the majority control right of China Lianhe Credit Ratings Company;
Xinhua Finance Limited (A U.S. controlled company) has purchased 62% of the common equity of
Shanghai FarEast Global Credit Rating Company; and Standard and Poor’s has started a strategic alliance
with Shanghai Brilliance Credit Rating Company. The only exception is the Dagong Global Credit Rating
Company, which remains to be a domestic credit rating firm.
10

securities companies, insurance companies, and some large non-financial corporations,
along with traditional commercial banks and broker dealers. The interbank market is a
quote-driven OTC market, in which deals are negotiated between two counterparties
through an electronic trading system. It is governed by the PBOC and functions as a
wholesale market for institutional investors. The interbank market relies on the National
Inter-bank Funding Center (i.e., the interbank center) and the Central T-bond Registration
and Settlement Co., Ltd. (i.e., the central registration and Settlement).
2.3. Major bond instruments
At the end of 2008, the total amount of bonds outstanding was RMB 15110.2
billion (1888.75 billion USD), among which government bonds, central bank sterilization
bills, financial bonds and non-financial corporate bonds (including CPs and MTNs)
represented 32.3%, 31.8%, 27.1%, and 8.4% of the total market, respectively.
China’s corporate bond market has been expanding remarkably. The amount of
corporate bonds outstanding has tripled in the past two years, up from 937 billion RMB
(117.12 billion USD) in first quarter of 2008 to 2916 RMB (416.57 billion USD) in first
quarter of 2010. However, by the end of 2008, Chinese corporate bonds still accounted
for less than 10% of the total market.
Six types of nonfinancial corporation bonds are currently regulated in China by
different authorities and are traded in different markets (Table 2). Short-term corporate
financing bills (CPs), medium-term notes (MTNs), and SME Collective Notes are
regulated by PBOC (People’s Bank of China) and NAFMII (National Association of
Financial Market Institutional Investors). They are traded in the interbank market.
11

Enterprise bonds are regulated by NDRC (National Development and Reform
Commission) and traded in both interbank and exchange markets. Listed company bonds
and convertible bonds are regulated by CSRC and traded in the exchange market. Quite
different from the situation in overseas countries, China’s corporate bond market includes
not only the enterprise bonds or listed company bonds but also nonfinancial corporate
debt financing instruments comprised of six types of nonfinancial corporation bonds. The
interbank corporate bond market has four instruments: CPs, MTNs, enterprise bonds, and
SME collective notes.
Enterprise bonds are the earliest type of corporate bonds issued in China. Because
their issuance was tightly regulated by the National Development & Reform Commission
(NDRC), enterprise bonds have had moderate growth. No more than RMB 700 billion
(87.5 billion USD) was outstanding at the end of 2008. Enterprise bonds are traded in the
interbank market, in which only institutional investors participate. Because they are
usually guaranteed by one of the state-banks, coupon rates of enterprise bonds are lower
than those of listed company bonds with the same maturity. Their maturity tends to be
mid- to long-term.
12

Table 2: Trading Volume and Outstanding Balance of Chinese Bond Market

Panel A. Trading Activities in Chinese Bond Market
Source: Wind. Figures are total value traded in 2008 in billion yuan.
Market Spot Trading Repo
Trading
Total

Shanghai Stock
Exchange

205.87 2430.68 2636.55

Shenzhen Stock
Exchange

21.88 0.00 21.88

Interbank
Market
37090.52 57526.25 94616.78

Total 37318.28 59956.93 97275.21

Panel B. Fixed Income Instruments Outstanding in Chinese Bond Market
Source: Chinabond Data as of December 31, 2008.
Bonds Outstanding
# of Issues
Par Value
(Billion)
Percentage

Regulated by Traded in
Government
Bonds

110 4875.34 32.27%

Central Bank
Bills

135 4812.10 31.85%

Policy Bank
Bonds

222 3672.01 24.30%

Financial
Institution
Bonds

99 424.82 2.81%

Corporate
Bonds

295 680.35 4.50%

NDRC/CSRC the interbank
and
exchange
markets

Corporate
Commercial
Papers

259 420.31 2.78%

PBOC and
NAFMII
the interbank
market

Assets Backed
Securities

49 55.11 0.36%

13

Listed company bonds were introduced in 2007, and are regulated by the China
Securities Regulatory Commission (CSRC). They are issued by listed companies and
traded in exchanges, in which both institutional and individual investors participate.
Listed company bonds are real credit bonds without bank guarantees. They have higher
coupon rates than enterprise bonds with the same maturity. By the end of 2008, 20 listed
company bonds were outstanding, with a total value of RMB 40 billion (5 billion USD).
Corporate CPs and MTNs are issued and traded in the interbank market, and are
regulated by the PBOC. Because the issuers are usually large corporations with good
standing, no bank guarantee is required. Credit rating is required for issuance. CP and
MTN markets have been the fastest growing segments in the past years. At the end of
2008, the size of CPs and MTNs reached RMB 420.3 billion and 167.2 billion,
respectively. They are two of the most actively traded instruments in the Chinese bond
market. In 2008, the turnover was 6.87 times and 4.40 times their outstanding balance,
respectively. Commercial banks and mutual funds are the main investors of CPs and
MTNs.
14

Chapter 3. Literature Review
Economists and sociologists have conducted extensive studies of how political
connections affect a firm’s value. In this section, I briefly review how political
connections are formed and measured; the benefits and costs of political connections; the
moderators that influence the relationship between political connections and firm value,
and the mechanisms through which political connections influence firm value.
Political connections are widespread. Faccio (2008) built a database of political
connections for more than 20,000 listed companies from 47 countries, in which she
identified 541 firms as politically connected, representing 3% of the world’s publicly
listed firms, and almost 8% of the world’s market capitalization. Connections are
particularly common in countries with high levels of corruption, countries that impose
restrictions on foreign investments by their residents, and countries with relatively more
transparent systems (Faccio 2008). Connections are less common in countries with
regulations that set rigorous limits on political conflicts of interest.
3.1. Formation and measurement of political connections.
Political connections can be established in the following ways: hiring former
government officials as corporate directors (Hillman 2005), donating to Political Action
Committees (Aggarwal Meschke and Wang 2008; Feguson and Voth 2006), or having a
firm representative serve in a political capacity (Hillman, Zardkoohi and Bierman 1999).
Accordingly, empirical studies measure political connections in the following ways: 1)
Friendship and social ties between firm executives and government officials (Faccio,
Masulis, and McConnell 2006; Faccio and Parsley 2007; Fisman 2001; Leuz and
15

Obserholzer-Gee 2003; Johnson and Mitton 2003). 2) Businessmen becoming politicians
(Faccio 2006). For example, a business person who is elected prime minister or member
of the parliament can serve as a proxy for political connections. 3) Officers joining
boards (Goldman, Rocholl, So 2008; Fan, Wong, Zhang 2007; Du, Tang, Young 2010).
Firm executives with a career history as a government bureaucrat can serve as a proxy for
political connections. 4) Political contributions and political donations (Snyder 1990;
Grossman and Helpman 1996; Coate 2004; Jayachandran 2006; Aggarwal, Meschke, and
Wang 2008; Roberts 1990; Knight 2006; Shon 2006; Ferguson and Voth 2008;
Claessens, Feijen, and Laeven 2008). 5) Firm executives’ party membership
(Jayachandran 2006; Goldman, Rocholl, and So 2008; Shon 2006; Li, Meng, Wang, and
Zhou 2008). 6) State ownership (Talmud 1992; Talmud and Mesch 1997; Cheung, Jing,
Rau, and Stouraitis 2008; Shleifer 2002; Sapienza 2004; Dinc 2004).
3.2. Benefits and costs of political connections
How firms benefit from political connections has received attention in the
literature in finance and accounting. Most of this literature finds that shareholders in
publicly listed firms benefit from close ties to governments. Firms can gain multiple
sources of values by becoming politically connected (Faccio 2006). As economists have
noted, politically connected firms obtain more favorable regulatory conditions (Stigler
1971; Agrawal and Knoeber 2001), use higher leverage (Cull and Xu 2000; Faccio 2006;
Khwaja and Mian 2005), have prioritized access to bank loans (Bai et al., 2006a; Khwaja
and Mian, 2005; Faccio, 2006; Claessens et al., 2008; Li et al., 2008), have higher market
share (Faccio 2006), enjoy lower taxation (Faccio 2006; De Soto 1989), secure property
16

rights protection (Hellman et al., 2003), are more likely to be bailed out by governments
during financial troubles (Faccio et al. 2004), are preferentially treated in competition for
government contracts, and lobby for stiffer regulatory oversight of their rivals (Faccio
2006). Thus, political connections improve firm performance (e.g., Johnson and Mitton,
2003; Li et al., 2008) and increase firm value (e.g., Roberts, 1990; Fisman, 2001;
Ramalho, 2007; Claessens et al., 2008).
During a regime change, firms betting on the right politician gain value. Recent
work has analyzed the value of political connections using event-study methods in the
context of a regime change, or shift of power, including several event studies use country
specific settings to test the value of betting on Dictator Hitler in Germany in 1930
(Ferguson and Voth 2008), of loyalty towards President Suharto in Indonesia (Fisman
2001), of connections to Mahathir and Anwar in Malaysia (Johnson and Mitton 2003),
and of affiliation with President Collor de Mello in Brazil (Ramalho 2003). In a regional
study of Shanghai firms, Xu and Zhou (2009) reported that firms connected to the
Shanghai government suffered on average 2% cumulative abnormal returns during the
five days that followed disclosure of the dismissal of Chen Liangyu, the mayor of
Shanghai.
The costs of political connections manifest in two ways. First, maintaining
beneficial political connections by providing payments and returning favors to politicians
is costly; Fisman et al. (2006) examined the value of political connections to Vice
President Dick Cheney and found that the value of such connections was zero, a finding
they believe is consistent with the hypothesis that politicians extract rent from the
17

companies they manage (De Soto, 1989; Shleifer and Vishny, 1994; Bertrand, Kramarz,
Schoar, and Thesmar 2006; Morck, Strangeland, and Yeung 2000; Fan, Wong and Zhang
2007), and thus that the costs of connections may offset the benefits. Second, cost occurs
when politically connected parties lose power and political enemies gain power, which
means having the wrong friends at the wrong time. Just as positive ties can lead to
favorable exchanges and other benefits for companies, negative ties can lead companies
to become victims of discrimination, resource exclusion, and even occasional
expropriation and sabotage between rival sociopolitical networks, as shown by Siegel
(2007) in a Korean context and by King (2000) in an Indonesian context.
3.3. Channels through which political connections influence firm value
Political contributions may affect firm value through several pathways. The
political connections held by a firm might afford it a competitive advantage, such as
government contracts and related transactions (Goldman, Rocholl, So 2008; Cheung,
Jing, Rau, and Stouraitis 2008, Li 2004); preferential access to financing (Cole 2004;
Dinc 2005; Sapienza 2004; Charumilind, Kali, and Wiwattanakantang 2004; Claessens,
Feijen, and Laeven 2008; Chiu and Joh 2004; Cull and Xu 2005; Faccio 2003; Johnson
and Mitton 2003; Khwaja and Mian 2005); government bailouts (Facci, Masulis, and
McConnell 2004; Fisman 2001); lower effective tax rates (Adhikari, Derashid, and Zhang
2006); and lower government enforcement risk (Correia 2009).

18

3.4. Institutional features favoring the emergence and survival of political connections
3.4.1. Level of market regulation.
Several institutional features favor the emergence and survival of political
connections. Market regulation means the level of government regulation in which a firm
competes (Baron 1995). A more regulated market means that many business
opportunities are controlled by government bodies that play favorites. Political
connections can be particularly important as they allow firms to control their attainable
opportunities in nonmarket competition. Agrawal and Knoeber (2001) found that
politically experienced managers are more prevalent when exports, lobbying, and sales to
the public sector play an important role, which is consistent with evidence provided by
Goldman, Rocholl, and So (2006).
3.4.2. Corruption.
Corruption is defined as crimes by public officials for personal gain (Rose-
Ackerman 1975). The cost of political connections is more notable in highly corrupt
countries with barriers to foreign investment and weak institutions. A sample of 541
politically connected firms in 47 countries suggests that connected firms underperform
compared to nonconnected companies on an accounting basis, and that the difference
between connected firms and nonconnected firms is more notable in countries with high
levels of corruption and a weak legal system (Faccio 2006). Faccio further shows that the
value of these companies increases when their executives enter politics; however, she
finds that this latter result comes from the subsample of companies in countries with high
levels of corruption. This proposition has been tested by several studies of countries well
19

known for having a high level of corruption, for example, Indonesia (Fisman 2001),
Malaysia (Johnson and Mitton 2003), and Brazil (Ramalho 2003). For example, Fisman
(2001) reported that in Indonesia, to a large extent, political connections— rather than
fundamentals such as productivity— determined the value of firms.
Politically connected firms benefit from having connections in countries with a
higher level of corruption for three main reasons. First, political connections bring about
government protection, which is particularly important in corrupt countries with weak
investor protection and property rights protection. Due to the lingering legacy of the
command economy and the slow development of market-supporting institutions, private
entrepreneurs in corrupt countries face many obstacles in running their businesses. They
are often denied access to bank loans, which are largely reserved for state-owned
enterprises, or are subject to heavy government regulations (red tape) or “extralegal” fees
(Johnson et al., 2000; McMillan and Woodruff, 2002; Guriev, 2004). In addition to the
problems of a weak state and ill-functioning markets, the legal system in such economies
is often too weak to secure property rights and enforce contracts (Hay and Shleifer, 1998;
McMillan and Woodruff, 1999; Frye and Zhuravskaia, 2000). Since corrupt countries are
oftentimes developing economies, which is relationship-based rather than market-based
capitalism, in such an environment, close ties to the government help businesses
overcome these market and state failures and avoid ideological discrimination.
Second, corruption makes countries poor (Mauro 1995), induces capital to flee,
and results in an underdeveloped equity market and debt market. Under these
circumstances, competition for capital among private business is expected to be more
20

intense. If one channel through which political connections may benefit firms is by
allowing them to borrow from state-owned banks on preferential terms, then the benefit
of political connections should be more notable in corrupt countries.
Third, in the absence of large and active private investors and well-functioning
institutions of corporate governance, the direct appointment of political executives is the
most powerful way a government can constrain the abusive behavior of enterprise
insiders and thus reduce agency costs (Qian 2000; Li 2000; Chang and Wong 2004). Xu
et al. (2005) suggest that politically connected managers are more accountable to
politicians and, hence, their ability to abuse their power is possibly curtailed. Qian
(1996), Li (2000), and Chang and Wong (2004) indicate that political executives behave
in accordance with the state’s interests and can mitigate managerial opportunism through
a system of checks and balances instituted by politicians.
3.4.3. Economic or political liberalization.
Business-government ties are likely to be sensitive to political change. In
emerging economies, two or more rival networks often compete for political power. If
one network gains political power, members of its government may use that power not
only to bestow privileged resources on their friends but also to target their enemies—
including members of rival networks— for exclusion and punishment. And, it is unclear
what effect economic or political liberalization has on the value of ties to the state.
Opposing arguments but few empirical tests have examined how economic and
political liberalization influences the value of business-government ties. Some
organizational scholars believe that liberalization reduces the impact of business-
21

government ties (Nee 1989; Guthrie 1990). Others, though, have argued that the value of
business-government ties remains high and even increases after economic liberalization
(Peng 1984).
3.4.4. Regime change.
The impact of business-government ties can change in the face of state policy
change. Talmud (1992, 199) and Talmud and Mesch (1997) have shown in the case of
Israel that being connected to the government through state ownership gave certain Israeli
firms privileged access to resources, but that state-owned defense contractors, such as
Ta’as, suffered when the defense budget was cut after 1985. More importantly, the
impact of business-government ties can change in the face of regime change. Fisman
(2001) showed empirically that political connections to the former Indonesian dictator
Suharto were worth a significant percentage of the politically connected firms’ market
capitalization and that a sizeable portion of their capitalization was erased any time a
legitimate rumor of Suharto’s life-threatening illness or impending death circulated.
3.4.5. New Regulations.
Berkman, Cole, and Fu (2008) reported that firms with weaker governance
experienced significantly larger abnormal returns around announcements of the new
regulations than did firms with stronger governance. Firms with strong ties to the
government did not benefit from the new regulations, suggesting that minority
shareholders did not expect regulators to enforce these new rules on firms in which block
holders have strong political connections.
22

3.5. Political connections and external financing
Although the importance of political connections has been anecdotally accepted
as an endemic feature of emerging economies, empirical work linking these political
connections to preferential finance is limited. This section discusses prior studies of
political connections and external financing from two perspectives: that of bank lenders
and that of firm borrowers.
On one hand, government-owned banks are often subject to capture by politicians.
Sapienza (2003) demonstrates that state-owned banks in Italy are vehicles for supplying
patronage to distressed regions and powerful political parties, showing that Italian state-
owned banks charge lower interest rates than privately owned banks after controlling for
the borrower’s credit-worthiness and other firm characteristics. Dinc (2004) analyzed a
sample of the largest banks in 36 emerging and developed countries, and found that
banks controlled by the government increased their lending during election years relative
to private banks.
On the other hand, Chiu and Joh (2004), Cull and Xu (2005), Faccio (2003),
Johnson and Mitton (2003) show that politically connected (but not publicly traded) firms
have higher leverage ratios than their nonconnected peers. Considering that comparing
the leverage rates between connected firms and nonconnected firms provides only an
indirect test, follow-up empirical studies have used bank loan data to examine whether
access to bank finance is an important channel through which political connections
operate. La Porta et al. (2003) found that companies with direct links to Mexican banks
received better terms and were more likely to default on a sample of 300 loans in Mexico.
23

They also find that related loans were more likely to default, and when they did, have
significantly lower recovery rates than unrelated loans. Charumilind, Kali, and
Wiwattanakantang (2004) showed that Thai firms with connections to banks and
politicians obtained more long-term loans and needed less collateral during the period
preceding the Asian financial crisis of 1997 compared to firms without such connections.
Khwaja and Mian (2005) used loan-level data for Pakistan and found that politically
connected firms (firms with a director participating in an election) borrowed 45% more
and had a 50% higher default rate than control firms, in a sample of 90,000 firms in
Pakistan between 1996 and 2002. Li, Meng, Wang, Zhou (2008) reported that politically
connected private Chinese firms (firms with CEOs who are CCP members) obtained bank
loans and other state institutions more easily than private firms that were not affiliated
with CCP. Consulting a sample of 6,400 Vietnamese firms, Malesky and Taussig (2008)
used government survey data, and concluded that banks placed greater value on
connections than on performance and that firms with greater access to bank loans were no
more profitable than firms without them. Claessens, Feijen, and Laeven (2008) showed
that Brazilian firms that contributed to (elected) federal deputies substantially increased
their bank financing relative to a control group after the 1998 and 2002 elections.
Charumilind, Kali, and Wiwattanakantang (2009) found that firms with connections to
banks and politicians had greater access to long-term debt than firms without such ties, in
a sample of 260 publicly traded Thai firms before the East Asian Crisis of 1997-98.
In addition to debt financing, domestic offerings, cross-listings, and government
bailouts are financing channels in which political connections may play a role. Faccio et
24

al. (2006) showed that politically connected firms are significantly more likely to be
bailed out than similar nonconnected firms, using the bailout data of 450 politically
connected firms from 35 countries from 1997 to 2002. Duchin and Sosyura (2010)
demonstrated that a bank’s political ties and lobbying activity increased the likelihood
and amount of federal TARP (The Troubled Assets Relief Program) investments in
banks. Leuz Obeholzer-Gee (2005) has suggested that firms without political connections
have a greater likelihood of going outside the country to raise capital in a market with
pervasive political intervention, and that firms connected to the losing parties in elections
flee to foreign capital markets to raise funds. Francis, Hasan, and Sun (2009) found that
politically connected firms had a higher offering price and less underpricing in IPOs in
China.
25

Chapter 4. Theory and Hypotheses
Before proposing my hypotheses, the theoretical arguments are discussed related
to how political connections may lead to firms’ preferential treatment in issuing bonds.
These three theoretical arguments regarding the relationship between a firm’s political
connections and bond issuance are: (1) The reputation enhancement argument, (2) The
social lending argument, and (3) The collusion argument.
4.1. The reputation enhancement argument
In countries with poor governance mechanisms and weak investor protection,
firms tend to rely on personal relationships for contract enforcement (Allen et al. 2005),
and banks place greater value on connections than on performance in making loan
decisions (Malesky and Taussig 2008). Given the institutional climate
3
in the Chinese
bond market, political connections may be a substitute mechanism for establishing
creditworthiness and reputation, a notion that is in line with the relationship lending
literature.
Individual executives’ political connections contribute to firm reputation because
connected executives have higher personal reputation at stake than non-connected
executives, caused by high expected career mobility between the business and the
politics. Former government officials are likely to join on boards of large companies, and
former businessmen are likely to enter politics and become high rank officials. When the

3
China’s bond market lacks information disclosure to investors due to an inadequate legal framework and
weak investor protection; inadequate accounting and external audit standards; a lack of regulatory emphasis
on proper disclosure by issuers; and a lack of credible credit rating agencies as well as prudent analysis by
investors. (A quote from Xiaochuan Zhou, governor of the People’s Bank of China)

26

Chinese government recruit firm executives as officials, executives whose firms fall into
financial insolvency during their tenures are less likely to be selected, because loan
default and bond default are salient and are subject to media scrutiny. Foreseeing this
potential career advance in the communist party, connected executives are less likely to
have their firms default. On the other hand, there is only a limited pool of connected
executives who can potentially join the firm as an executive. From the companies’ stand
of point, how much value this person’s reputation can bring is weighed against the cost of
hiring this person. From the executives’ stand of point, he considers whether and how the
experience of joining this company helps his future career advances. Having a connected
executive therefore may help to signal a firm’s reputation in not defaulting on bonds.
In bond markets with an opaque information environment
4
and low quality
disclosure, soft information may need to come through channels other than financial
reporting in order to meet the information needs of bond underwriters. In addition,
having politically connected firm executives facilitates information seeking and reduces
information communication costs between firms and bond issuance regulatory agencies
in the bond issuance process, as they share a higher level of in-group identity with
government regulatory agency officials than non-connected executives do. Executives
with a prior career history in the government are more familiar with the bureaucratic
systems, administrative procedures, and the macro-perspectives taken by government
officials when making decisions. Firm executives with previous government experience

4
Morck, Yeung and Yu (2000) measures information opacity as a premium that investors are willing to pay
to obtain transparent and credible information. Empirical data shows that the opacity premium is highest in
China, followed by Indonesia and South Korean among 12 Asian countries.
27

share ideologies and aspirations with government regulatory agency officials, understand
the underlying rules of the bureaucracies, and are more adept at using the language of the
Chinese Communist Party when communicating with regulatory agency officials.
4.2. The social lending argument
Social lending means that the government gives favorable treatment to firms that
involve socially efficient but high-risk projects and firms with politicians on their boards
are more likely to take on such socially efficient projects (Khwaja and Mian 2005). This
argument is in line with Shleifer and Vishny’s (1998) “giving hand model” of
government, which is based on the assumption that government is benevolent, and aims
to maximize social welfare and solve the failures of the free market by encouraging
socially efficient investments (Stiglitz 1989).
In the absence of large, active private investors, directly appointing political
executives is the most powerful way for governments to constrain abusive behavior by
enterprise insiders and thus to reduce agency costs. Xu et al. (2005) have suggested that
politically connected managers are more accountable to politicians and hence their ability
to abuse their power is curtailed. Qian (1996) has indicated that political executives
behave in accordance with the state’s interests and can mitigate managerial opportunism
through the system of checks and balances instituted by politicians. Assuming that
politically connected firm executives serve as agents of the government (Hu and Leung
2008), it is possible to expect them to follow the desires of the government to maximize
economic efficiency and to fulfill social goals. Social responsibility measures are subject
to judgment, hard to quantify, and may differ across industries (Aharoni 1981). The
28

difficulty of quantifying social responsibility may hinder monitors from effectively
determining whether efforts have been exerted to maximize measures, leading to
weakened incentives and potential shirking on the part of firm executives.
4.3. The collusion argument
An important institutional feature of China’s bond market is strong government
intervention. The Chinese government exerts a range of regulations on bond issuance,
including placing restrictions on issuers’ qualifications,
5
and exercising a merit-based
approval process,
6
issuance quotas, and interest rates controls. The administrative
allocation of quotas for issue size and number of issuers was mandated by the central
government for provincial and lower-level governments. The administrative allocation of
quotas was often used as a relief measure for financially distressed enterprises.
Corporate bonds can provide long-term capital without diluting shareholders’
interests; there is no strict limitation imposed on the usage of the funds (as there is for
bank loans); and current regulations of the interest rate caps emphasize that it should be
no more than 40% higher than the prevailing bank lending rate. Thus, it is clear why the
option of issuing corporate bonds is appealing to most Chinese firms, and that issuing
firms are motivated to influence the government regulatory agency’s approval decisions.

5
Only firms with ratings on the highest end of the credit ratings spectrum are allowed to issue bonds in the
Chinese primary bond market. In other words, no junk bonds, or any bonds with a class lower than
investment are allowed. This regulation constrains the size of the whole bond market and the design and
trading of derivative instruments.

6
The official approval procedure for the issuance of corporate bonds is currently a merit-based approval
process, The minister of PBoC, Zhou Xiaochuan, mentioned that the official approval procedure for the
issuance of corporate bonds should be transformed gradually into a verification system, and, ultimately,
into a “registration management” system.

29

Because China’s regulatory agencies adopt a merit-based approval process in bond
issuance, government officials practice a certain level of discretion over the decisions.
Chinese government officials may show favoritism to firms issuing bonds, acting on
personal preference and showing partiality to some firms (Prendergast and Topel 1996),
behavior that is in line with the “grabbing hand model” proposed by Shleifer and Vishny
(1998). According to the grabbing hand model, politicians are self-interested and thus
are more likely to intervene in firm business activities to fulfill their personal agendas
than to intervene to maximize social welfare.
The research setting has several supervising agencies for the bond issuance
process, including the People’s Bank of China (PBoC), the China Securities Regulation
Committee (CSRC), and the National Development and Reform Committee (NDRC).
Considering that officials from these three bureaus may have different political objectives
and personal agendas, officials may treat firms with politically influential executives
favorably during the bond issuance process, hoping for future reciprocity when they need
to fulfill their personal political agendas and advance their careers. In other words, firm
executives and politicians in bond issuance regulatory agencies collude so that firms get
favorable treatment when issuing bonds and politicians gain resources to advance their
careers, such as money, information, and votes. The political connections of firm
executives facilitate their collusion with politicians, thus, the government favors
politically connected firms when granting approval decisions in bond issuance processes.
For example, firms in Brazil were shown to create value by being friends with President
Collor de Mello (Ramalho 2007); in Indonesia by connecting to President Suharto
30

(Fisman 2001); and in the U.S. by contributing to winning campaign candidates (Roberts
1990; Goldman et al. 2009).
Based on these arguments about how political connections affect issuing firms’
access to bond capital, the following hypotheses is offered.
Hypothesis 1: Politically connected firms are more likely than non-
connected firms to issue bonds with larger offer size.
4.4. Political connections of bank underwriters and issuing firms’ access to bond
capital
Before issuing bonds, issuing firms must meet certain thresholds; however,
meeting these criteria does not guarantee approval decisions from the government
regulator. In this process, underwriters work with issuing firms to get approval from
CSRC to get the “issue quota.” The role of underwriters becomes especially important
when the central government intends to tighten the supply of bond capital and assign
limited passes to most connected bank underwriters.
The political connections of bank underwriters may facilitate an issuing firm’s
access to bond capital because these banks can gain trust from the government regulator
by providing capital to underdeveloped areas and politically influential parties, and can
influence the government regulator’s decisions in bond issuance approval. Prior studies
have suggested that politically connected bank underwriters are more likely to be subject
to capture by politicians, and that politically connected banks may in turn obtain trust
from the government regulator by supplying patronage to distressed regions and charging
low interest rates (Sapienza 2003) and to powerful political parties by increasing lending
31

during election years (Dinc 2005). Thus, bank underwriters’ connections may contribute
to bank underwriters’ reputations, which help to certify issue quality to the government
regulator (Myers and Majluf 1984). Underwriters serve as information specialists that can
bridge the information gap between issuing firms and the government regulator
(Peristiani 2006). Investment banks have historically performed the role of information
intermediary in securities underwriting, managing, and selling the offering for issuing
firms. China’s financial system features state-owned banks
7
as the dominant player, with
the banking system as the main source of external corporate financing. Thus, commercial
banks have information advantages resulting from the screening and monitoring of
borrowers.
This idea leads to the second hypothesis.
Hypothesis 2: Issuers with politically connected bank underwriters are
more likely to issue a larger amount of bonds than issuers without
politically connected bank underwriters.
This study also attempts to explore the relationship between the role of firm
political connections and that of underwriter political connections.
On one hand, it is possible that firms with strong political connections are not in
need of connected bank underwriters. They may attach more weight on general
reputations, distribution networks, industry specific experiences, and firm specific
experiences (repeated landing) of individual banks when choosing underwriters.

7
To date, the only two private banks included in my sample are China Minsheng Bank Corp., Ltd. and
China Zheshang Bank Co., Ltd.

32

Foreseeing the fact that political connections facilitate a firm’s effort to raise capital, non-
connected firms may pay premium in underwriting fees to buy political connections from
the bank underwriters. If this is the case, then the political connections of issuing firms
and the political connections of bank underwriters are substitutes when explaining bond
issue size.
On the other hand, if the reputation enhancement holds, then firm executives’
connections signal a lower probability to default. Connected banks may choose to work
with connected firms, relying upon an implicit government guarantee that politically
connected firms will be bailed out should they encounter financial difficulties. It is also
possible that connected bank officers and connected firm executives enjoy working
together since they come from a common background. If this is the case, then the
political connections of issuing firms and the political connections of bank underwriters
are complements when explaining bond issue size.
It remains an empirical question as to what is the relationship between the
political connections of issuing firms and of bank underwriters. This discussion leads to
the third hypothesis.
Hypothesis 3a: The political connections of issuing firms and the political
connections of bank underwriters are substitutes when explaining bond
issue size.
Hypothesis 3b: The political connections of issuing firms and the political
connections of bank underwriters are complements when explaining bond
issue size.
33

Chapter 5. Empirical Design and Analyses
5.1. Sample, constructs, and measures
This study uses the China Bond database prepared by the China Government
Securities Depository Trust & Clearing Co. Ltd (CDC)
8
. I start by collecting bond
issuance statements from 2001 to 2009, including enterprise bonds, listed company
bonds, short-term corporate financing bills, and mid-term notes. I exclude convertible
bonds and bonds issued by financial service firms. My initial sample consists of 1,664
bond issuances and 1,263 firm year observations. Among these 1,263 firm year
observations, I extract the political connections measures of 779 firm year observations
and construct my final sample based on these 779 firm year observations.
I extract the issue size, credit ratings, and political connections variables from
bond issuance statements. First, I obtain information about bond characteristics from
bond issuance statements, including type of bond, issuing amount, maturity terms, and
the collateral terms. Second, I record characteristics of investment banks, rating agencies,
auditors, and law firms involved in the bond underwriting process. I record the names,
ownership types, and sizes of lead arranger banks and participating banks, as well as the
locations of their headquarters. I also calculate the number of participating banks. I
identify whether issuers chose a Big-4 auditor when issuing bonds. Third, I code
characteristics of bond issuing firms, including their financial performance measures,

8
In December 1996, China Government Securities Depository Trust & Clearing Co. Ltd (CDC) was
established for centralized settlement and clearing. In 1997, it was assigned the responsibility for
centralized treasury custodian.

34

organizational structure, board structure, and complete CV records of the top
management teams.
5.1.1. The issue size and the issuer credit ratings.
The issue size serves as a proxy for a firm’s access to bond capital. In the Chinese
bond market, current regulations emphasize that the interest rate caps should be no more
than 40% higher than the prevailing bank lending rate, and only allow firms with ratings
higher than BBB+ to issue bonds. In addition, the issuance market has an overall yearly
quota. These regulations constraint the size of the whole bond market and firms’ access
to bond capital, making the issue size an appropriate measure for representing a firm’s
access to bond capital. Prior studies suggest that the issue size is one of the most
economically important determinants of bond trading volume (Hotchkiss and Jostova
2007) and bond liquidity (Crabbe and Turner 1995). The issue size also reflects a firm’s
overall debt burden, which is generally positively associated with bond yield spread (Shi
2003).
The issuer credit ratings, which are frequently used in bank loan contracts,
represent the rating agencies’ judgment of a firm’s overall financial capacity and can
affect a firm’s future access to bank loans and bonds (Jiang 2008). Downward rating
changes affect both bond and stock prices (Dichev and Piotroski 2001) and can trigger an
accelerated payment of existing debt (Reason 2002). I choose issuer credit ratings over
individual bond ratings to apply in my regression models because the level of analysis in
35

this study is the firm level.
9
Credit ratings symbols and definitions in bond issuances in
China are required to conform to the regulatory document issued by the People’s Bank of
China. The Chinese corporate bond rating system is designed to be almost identical to the
Standard & Poor’s rating criteria, with the lowest possible rating a “C,” and no “+/-”
adjustments can be attached to issuers/issues with ratings lower than or equal to “CCC.”
Ratings letters are coded into ratings numbers in a reverse manner for ease of
interpretation, with higher values of ratings numbers indicating a better rating, according
to the conversion table presented in Figure 1, below.

9
An issuer credit rating is utilized to judge the issuer’s overall financial capacity and does not apply to any
specific financial obligation. A bond credit rating is the current opinion of the issuer’s creditworthiness
with respect to a specific bond. A bond credit rating can be either long term or short term. Short-term
ratings are generally assigned to short-term financing bills with an original maturity of no more than 365
days– including commercial papers. Medium-term notes are assigned long-term ratings (Standard &
Poor’s, 2009).
36

Figure 1. Rating Regime, Symbols, and Definitions
Panel A. Moody’s rating analysis of an industrial company

Panel B. Issuer and Issue Ratings Symbols and Definitions
Long-term/Issuer Short-term Code Definitions
AAA A-1+ 8 Investment
AA+ 7 Investment
AA 6 Investment
AA- 5 Investment
A+ A-1 4 Investment
A 3 Investment
A- A-2 2 Investment
BBB+ 1 Investment
BBB A-3 0 Investment
BBB- N/A Investment
BB+ B N/A Speculative
BB N/A Speculative
BB- N/A Speculative
B+ N/A Speculative
B N/A Speculative
B- N/A Speculative
CCC C N/A Speculative
CC N/A Speculative
C N/A Speculative

Issue Structure
Company Structure
Operating/Financial Position
Management Quality
Industry/Regulatory Trends
Sovereign/Macro-economic Analysis
37

5.1.2. Political connections of issuing firms
The political connections
10
considered in this study derive from career histories of
SOE executives as officials in the Chinese government. I use three connections measures
based on Du et al. (2010), including a CEO’s political connections (CEOConnection), a
CFO’s political connections (CFOConnection), and the political connections of the
management team (MGTConnection). An executive’s connections are coded based on
scores assigned to the highest bureaucratic positions held by CEOs before they joined the
top management teams of their SOEs. MGTConnection is coded as the sum of individual
executives’ connections scores. The highest level of political connection is attributed to
those who have held top positions in the central government, followed by individuals in
provincial and municipal governments. This method enabled measurement of the
intensity of a connection instead of simply measuring for an indication of its presence,
capturing the breadth of the board members’ skills and expertise brought from previous
public service.
In addition, the exploratory factor analysis revealed that CEO connections, CFO
connections, MGT connections, and a dummy indicator of the existence of political

10
Empirical studies measure political connections in the following ways: 1) Friendship and social ties
between firm executives and government officials (Faccio, Masulis, and McConnell 2006; Faccio and
Parsley 2007; Fisman 2001; Leuz and Obserholzer-Gee 2003; Johnson and Mitton 2003); 2) Businessmen
becoming politicians (Faccio 2006). (For example, a businessperson who is elected prime minister or
member of the parliament can serve as a proxy for political connections); 3) Officers joining boards
(Goldman, Rocholl, So 2008; Fan, Wong, Zhang 2007; Du, Tang, Young 2010). Firm executives with a
career history as a government bureaucrat can serve as a proxy for political connections. 4) Political
contributions and political donations (Snyder 1990; Grossman and Helpman 1996; Coate 2004;
Jayachandran 2006; Aggarwal, Meschke, and Wang 2008; Knight 2006; Ferguson and Voth 2008;
Claessens, Feijen, and Laeven 2008). 5) Firm executives’ party membership (Jayachandran 2006;
Goldman, Rocholl, and So 2008; Li, Meng, Wang, and Zhou 2008). 6) State ownership (Sapienza 2004;
Dinc 2005).

38

connections load on one factor. Based on the results from the exploratory factor analysis,
a principal component analysis is implemented to obtain the principal component score of
the political connections of an issuing firm. This factor score is labeled as connectionfac,
which captured 68% of the explained variance of firm political connections. The
coefficient reliability (measured by Cronbach’s alpha) for connectionfac is 0.89,
validating this measure as an overall assessment of a firm’s strength of political
connections.
5.1.3. Political connections of bank underwriters.
The political connections of bank underwriters are measured by a dummy variable
of state-owned lead arrangers (leadSOE) and a dummy variable of big banks (leadBIG). I
argue that bank underwriters with the strongest political connections are most likely to
attract the most lucrative deals.
The first measure is leadSOE, which is coded based on the ownership type of the
lead arrangers. In China, close interaction exists between state-owned banks and state-
owned firms. Take the MTN issuance market in 2009 as an example. The seven issuers
with the largest issuance volume are either large central SOEs such as State Grid,
PetroChina, Sinopec, China Telecom, Shenhua Group, or central governmental agencies
such as the Ministry of Railway. The big seven issued 321 billion RMB MTNs in 2009,
making up 46.52%, or nearly half, of the total MTNs issuance volume. The largest MTN
issuances in 2009 were underwritten by large state-owned commercial banks such as
ICBC (Industrial and Commercial Bank of China), BOC (Bank of China), CCB (China
39

Construction Bank), ABC (Agricultural Bank of China), and BCM (Bank of
Communications).
The second measure is coded based on the market share of an underwriter in the
bond issuance market. Overall, my research sample has more than 200 bank underwriters.
Among these bank underwriters, the big-6 commercial banks and the big-6 investment
banks account for 58% of all issuance deals. If a firm underwriter has a lead arranger at
one of the big-6 commercial banks or the big-6 investment banks, the leadBIG is coded
as 1. Notably, an issuing firm in this research sample may have multiple lead arrangers,
in which case, the political connections of lead arrangers are coded as the sum of
connections of all lead arrangers.
Underwriters are also classified based on their ownership using two dummy
variables, leadPRIVATE and leadFOREIGN, to proxy for bank underwriters who are
privately held and those who are jointly held by foreign institutional investors. Two types
of underwriters are in my research sample: commercial bank underwriters and securities
company underwriters. leadcommercialb and leadinvestmentb are used respectively to
mark these two types of bank underwriters.
5.2. Regression models
To test whether a positive association exists between the variables of interest and
the offering amount of bond issuance, the following regression models are used to assess
the associations between political connections measures and (1) offering amount and (2)
issuer credit ratings.
40

Amount
i,t
= α
0
+ β
1
Issuer/Underwriter Connections
i,t
+ β
2
Term
i,t
+ β
3
Collateral
i,t

+ β
4
AT
i,t
+ β
5
EBIT
i,t
+ β
6
STDEBIT
i,t
+ β
7
Lev
i,t
+ β
8
SGrow
i,t
+ β
9
Margin
i,t
+ β
10
Current

i,t
+ ε
i
(1)
Entityrating
i,t
= α
0
+ β
1
Issuer/Underwriter Connections
i,t
+ β
4
AT
i,t
+ β
5
EBIT
i,t
+
β
6
STDEBIT
i,t
+ β
7
Lev
i,t
+ β
8
SGrow
i,t
+ β
9
Margin
i,t
+ β
10
Current
i,t
+ β
10
Auditorrank
i,t
+
β
10
Boardsize
i,t
+ β
10
DualCEOChair
i,t
+ ε
i
(2)
Where:
Amount
i,t
= Total offering amount, which is the sum of offering amount of bonds
issued by firm i in fiscal year t.
EntityRating
i,t
= Long-term credit ratings assigned to firm i in fiscal year t.

The issuer Connections take one of the following five specifications.

CEOConnection
i,t

= The highest bureaucracy position a CEO held before
joining a firm issuer as CEO.

The position is manually
coded into two dimensions of categorical measures, the
rank of the bureau in which the CEO had served, and the
rank of the political title the CEO held. The ranking of
the bureau is sorted into three categories: central
government (=3), provincial government (=2), and
municipal government (=1). The ranking of the political
title is also sorted into three levels, ministerial/provincial
level (=3), board/regional level (=2), and unit level (=1).
If the CEO has not served in the Chinese government,
then CEOConnection is coded as 0.
CFOConnection
i,t
= The highest bureaucracy position a CFO held before
joining a firm issuer as CFO. Coded as above for
CEOConnection.
MGTConnection
i,t
= Sum of scores based on the highest bureaucracy position
the executives held before joining the management team
of a firm issuer.
Connection
i,t
= Dummy variable of political connections. If
MGTConnection >0, Connection =1; otherwise,
Connection =0.

41

connectionfac
i,t
= A Principle component score extracted from the
following four measures of firm i in fiscal year t:
CEOConnection, CFOConnection, MGTConnection,
and dummy indicator of political connections.

The underwriter Connections take one of the following four specifications.

LeadBIG
i,t
= If the lead arranger is one of the top-6 commercial banks
or one of the top-6 investment banks in China, LeadBIG
=1; otherwise, LeadBig =0.
LeadSOE
i,t
= If the lead arranger is a state-owned financial institution,
LeadSOE =1; otherwise, LeadSOE =0.
LeadPRIVATE
i,t
= If the lead arranger is a privately held financial
institution, LeadPRIVATE =1; otherwise,
LeadPRIVATE =0.
LeadFOREIGN
i,t
= If the lead arranger is a foreign jointly held financial
institution, LeadFOREIGN =1; otherwise,
LeadFOREIGN =0.

Control variables include:

Term
i,t
= Bond maturity (in the number of years) calculated as the
weighted average bond maturity across bonds issued by
the firm, using offering amounts as weights. Individual
bond maturity is one plus the number of years from bond
offering date to bond maturity date.
Collateral
i,t
= Collateral dummy variable. Bond issues with collaterals
are coded as 1; otherwise, Collateral = 0.
AT
i,t
= Natural log of firm i’s total assets at the end of year.
EBIT
i,t
= Firm i’s earnings before taxes plus interest expenses.
STDEBIT
i,t
= Standard deviation of EBIT over the last 3 years before
the bond issuances.
Lev
i,t
= Firm i’s total debt divided by total assets at the end of
year t.
SGrow
i,t
= The annual percentage change in firm i’s sales from year
t-1 to year t.
Margin
i,t
= (Operational income – operational cost)/operational
income.
Current
i,t
= Firm i’s current assets divided by current liabilities at
the end of year t.
BoardSize
i,t
= Number of board members of firm i.

42

Auditorrank
i,t
= The rank of auditors selected by issuing firms when
issuing bonds. The rank of auditors is sorted into three
categories: Big-4 auditors (=2); domestic top-10 auditors
(=1); local small auditors (=0).
CEODuality
i,t
= Duality of CEO and chair dummy variable. CEODuality
is coded as 1 when the CEO and chair of board is the
same person; otherwise, CEODuality = 0.

Additional control variables include:

#Business Units
i,t
= Number of business units of firm i.
#Provinces
i,t
= Number of provinces that are covered by business units
of firm i.
Beijing
i,t
= City political rank indicator variable. Issuers that are
headquartered in Beijing are coded as 1; otherwise,
Beijing=0.
SOEDummy
i,t
= SOE dummy variable. Issuers that are SOEs are coded
as 1; otherwise, SOEDummy =0.
ListingDummy
i,t
= PLC listing status variable. Issuers that are publicly
listed in Chinese A-stock market are coded as 1;
otherwise, ListingDummy =0.

This study uses Peterson two-way clustered regression models
11
to test whether
my variables of interest led to a larger offering amount. Standard errors are calculated
after adjusting two dimensions of within-cluster correlation, by firm and by year. A
positive coefficient on political connection measures is consistent with my hypothesis
that stronger political connections are associated with a larger issuing amount. To control
for the effects of outliers, observations with absolute standardized residuals greater than
0.01 in all regressions are excluded.

11
I also use the Peterson one-way cluster analysis to obtain results that are robust to within-cluster
correlation in the same firm across years. Note: this method assumes that observations in the same firm but
different years are uncorrelated. I included year dummies to control for the year fixed effect. Results
remain qualitatively consistent.

43

Control variables include log-transformed total assets, EBIT scaled by total assets,
three-year standard deviation of EBIT prior to bond issuance, leverage, sales growth,
sales margin, and current ratio. These variables are included in the regressions to control
for the effect of profitability and financial capacity on issue size. Prior studies have
documented size, asset tangibility, profitability, and sales margin as important
determinants of corporate financing choices (Hovakimian et al. 2004). A positive
correlation between firm size and offering amount is expected because larger firms can
afford to issue bonds of a larger amount (Hovakimian et al. 2004). A positive correlation
between sales growth, sales margin, EBIT, and offering amount is expected because
profitability has been shown in prior studies to be associated with larger issue size (Rajan
and Zingales 1995). High EBIT volatility, high leverage, and low current ratio may signal
higher risks and heavier debt burdens, so they may associate with lower offer amount. To
control for bond features associated with offer size, the bond’s collateral terms and
maturity terms are also included as control variables in the regression model (1).
Ashbaugh-Skaife et al. (2006) suggested that governance variables such as the
ownership concentration, the overall board independence, and the CEO power are
significantly associated with firm credit ratings; hence, in the regression model (2)
controls for corporate governance are added. Corporate governance is measured by rank
of auditors, board size, and CEO/Chair duality.
12

12
I collected firm characteristics data of board independence but was only able to extract 289 firm year
observations that can be merged with my main sample. Neither the number of independent directors nor the
percentage of independent directors on board is significantly associated with issue size or credit ratings in
the univariate analysis and in the multiple regression. Thus, I choose to exclude variables of board
independence as control variables.
44

Additional controls include firm issuer characteristics based on the organizational
complexity, the ownership type, the geographic locations of firm headquarters, and the
listing status. Organizational complexity is measured by the geographic diversification of
firm subsidiaries using the number of subsidiaries and the number of provinces covered
by firm subsidiaries following Bushman et al. (2004). The research sample is classified
into two types of firms based on the ownership type, SOEs and non-SOEs. The research
sample is classified into two types of firms based on the geographic locations of their
firm headquarters: those headquartered in Beijing and those not headquartered in Beijing.
The research sample is classified into two types of firms based on their listing status:
publicly listed companies (PLCs) and non-publicly listed companies (Non-PLCs).
5.3. Empirical results
5.3.1. Summary Statistics
Table 3 reports the summary statistics. Issue sizes of individual firms ranged from
20 million RMB Yuan to 190 billion RMB Yuan (23.75 billion USD). Average maturity
terms of bonds issued by individual firms ranged from three months to 30 years. In my
sample, approximately 47% of firms were politically connected, which is slightly higher
than the Fan, Wong, and Zhang (2007) sample, which shows that almost 27% of CEOs
were politically connected, in a database of CEOs and directors of 790 PLCs (nearly 73%
of all IPOs). That my sample covers 62% of bond issuances and Fan, Wong, and Zhang’s
(2007) covers 73% of all IPOs indicates that bond issuers are more connected than PLCs.
In addition, this statistic suggests that there is more government intervention in the bond
issuance domain than the IPO domain. I also compared the mean and median offering
45

amount between four sets of sum-samples: connected vs. non-connected firms, SOEs vs.
private firms, publicly listed firms vs. non-publicly listed firms, firms headquartered in
Beijing vs. firms headquartered in other cities. Consistent with my expectation, connected
firms are more likely to have higher offering amount and higher credit ratings than non-
connected firms.
5.3.2. Political connections and issue sizes and ratings
Table 4 shows that all four political connections measures significantly and
positively correlate with offering amounts, which is consistent with H1, that is, the
politically connected firms are more likely to issue bonds of a larger amount than non-
connected firms. CEO connections (coeffi. = 0.076), CFO connections (coeffi. = 0.161),
and MGT connections (coeffi. = 0.050) are all positively and significantly associated with
the issue size. Consistent with my expectation, the firm size and the firm profitability is
positively associated with the offering amount and the volatility of firm profitability is
negatively associated with the offering amount. Firms with bonds of longer maturity
terms are associated with a larger amount of offer size, and firms with collaterals are
associated with a smaller amount of offer size.

46

Table 3: Descriptive Statistic
Variable mean Sd min p25 p50 p75 max
Issuesize 25.33 79.98 0.2 6 10 20 1900
REntityrating 6.05 1.49 0.80 5.00 6.00 7.00 8.00
Term 3.66 9.28 0.25 1 1 5 30
Collateral 0.24 0.43 0 0 0 0 1
CityPR 1.08 0.83 0 0 1 2 3
Ownership 1.17 0.56 0 1 1 2 2
listingDummy 0.37 0.48 0 0 0 1 1
Bondtype 0.58 0.87 0 0 0 2 2
SOEDummy 0.92 0.28 0 1 1 1 1
Beijing 0.26 0.44 0 0 0 1 1
CEOConnect~n 0.82 1.45 0 0 0 2 5
CFOConnect~n 0.33 1.04 0 0 0 0 5
MGTConnect~n 2.28 2.9 0 0 0 4 12
DConnection 0.47 0.5 0 0 0 1 1
Auditorrank 0.4 0.68 0 0 0 1 2
BoardSize 8.09 2.85 5 5 8 10 18
DualCEOChair 0.19 0.39 0 0 0 0 1
Numberofbu~s 10.32 15.81 1 2 6 13 314
Numberprov~s 4.16 5.8 1 1 2 5 34
leadBIG 0.66 0.62 0 0 1 1 2
leadSOE 1.08 0.53 0 1 1 1 2
leadPRIVATE 0.05 0.23 0 0 0 0 2
leadFOREIGN 0.03 0.18 0 0 0 0 1
AT 4.53 1.01 3.04 3.91 4.32 4.81 9.87
EBIT1 0.11 0.37 0 0.02 0.05 0.08 3.17
STDEBIT 4183.99 18050.35 1 71.66 253.52 1054.62 137852.5
Lev 2.35 3.33 0.01 0.96 1.63 2.61 30.35
SGrow 0.31 0.57 -0.83 0.08 0.21 0.39 3.84
Margin 17.03 17.03 -0.12 0.4 13.41 25.45 73.37
Current 1.26 1.31 0.16 0.77 1.03 1.28 10.2
CreditMKT 5.22 2.1 0 3.85 5.24 7.67 7.94
GovDecentr~e 6.67 1.29 0 6.15 6.4 7.49 8.37
Legal 6.24 1.46 0 5.32 6.29 7.97 7.97
Table 3 A describes the summary statistics of the main variables. See Table 1 for variable definitions.
Offering amounts are in the units of 0.1 billion. Maturity terms are in the units of 1 year. AT is presented
after log transformation. Summary descriptive are presented before winsorization.

47

Table 4: Associations Between Political Connections of Issuing Firms and Issue Size

Amount
i,t
= α
0
+ β
1
Issuer Connections
i,t
+ β
2
Term
i,t
+ β
3
Collateral
i,t
+ β
4
AT
i,t
+ β
5
EBIT
i,t
+ β
6
STDEBIT
i,t

+ β
7
Lev
i,t
+ β
8
SGrow
i,t
+ β
9
Margin
I,t
+ β
10
Current
i,t
+ ε
i

(1) (2) (3) (4)
Amount Amount Amount Amount
CEOConnection 0.0775
**

(2.79)
CFOConnection 0.161
***

(3.32)
MGTConnection 0.0498
*

(2.43)
DConnection 0.171
(1.92)
Term 0.0709
*
0.0721
*
0.0722
*
0.0729
*

(2.18) (2.25) (2.32) (2.26)
Collateral -0.643
*
-0.643
**
-0.639
**
-0.665
**

(-2.55) (-2.75) (-2.63) (-2.74)
AT 0.340
**
0.333
**
0.335
**
0.344
**

(2.78) (2.73) (2.78) (2.84)
EBIT1 0.254
*
0.252
*
0.250
*
0.261
*

(2.38) (2.35) (2.32) (2.24)
STDEBIT -0.00000207 -0.00000161 -0.00000207 -0.00000214
(-1.44) (-1.35) (-1.32) (-1.30)
Lev -0.0327
**
-0.0304
**
-0.0319
**
-0.0315
**

(-2.65) (-2.60) (-2.61) (-2.66)
SGrow -0.0581 -0.0470 -0.0528 -0.0559
(-1.86) (-1.38) (-1.63) (-1.77)
Margin -0.000777 -0.00152 -0.00119 -0.000573
(-0.54) (-0.95) (-0.91) (-0.35)
current1 -0.0745
*
-0.0777
*
-0.0802
*
-0.0709
(-1.96) (-2.16) (-2.05) (-1.90)
_cons 0.997
*
1.037
*
0.974
*
0.949
*

(2.05) (2.14) (2.07) (2.03)
N 779 779 779 779
adj. R
2
0.189 0.207 0.197 0.184

Table 4 reports the regression results for the effect of CEO/CFO/MGT political connections on offering
amounts. Columns 1, 2, 3, 4 report results when offering amounts are used as dependent variables. Columns
1, 2, 3, 4 report results when CEO connections, CFO connections, MGT connections, and dummy of political
connections are used as independent variables. Columns 1, 2, 3, 4 report results when control variables
include issue maturity terms, dummy of collateral, log transformed total assets, EBIT scaled by total assets,
standard deviation of EBIT, leverage, sales growth, sales margin, and current ratio. To ensure that our
inferences are not an artifact of a few extreme values, all variables are winsorized at the top and bottom 1%.
Results are derived from two way clustered regressions (Peterson 1997), in which observations are clustered
by firm and by year, and standard errors are calculated which account for two dimensions of within cluster
correlation. T-statistics are in parentheses. ***, **, * indicate two-tailed statistical significant of coefficient
estimates at the 0.001, 0.01, 0.05 level.
48

Table 5 shows that all four political connections measures significantly and
positively correlate with entity long-term ratings, which is consistent with the reputation
enhancement argument. Other firm-specific controls are also statistically significant in
regression (2). Consistent with earlier studies, firm size and firm profitability is positively
associated with issuer credit ratings; higher leverage represents higher overall debt
burden and thus is negatively associated with issuer credit ratings. A firm with a higher
quality auditor is more likely to receive higher credit ratings (coeffi. = 0.42) and a firm
with its CEO serving as chair of the board is more likely to receive lower credit ratings
(coeffi. = - 0.62).
49

Table 5: Associations Between Political Connections of Issuing Firms and
Issuer Ratings

Entityrating
i,t
= α
0
+ β
1
Issuer Connections
i,t
+ β
4
AT
i,t
+ β
5
EBIT
i,t
+ β
6
STDEBIT
i,t
+ β
7
Lev
i,t
+ β
8
SGrow
i,t
+
β
9
Margin
i,t
+ β
10
Current
i,t
+ β
10
Auditorrank
i,t
+ β
10
Boardsize
i,t
+ β
10
DualCEOChair
i,t
+ ε
i

(1) (2) (3) (4)
REntityrating REntityrating REntityrating REntityrating
CEOConnection 0.127
***

(4.14)
CFOConnection 0.154
***

(6.32)
MGTConnection 0.0802
***

(9.07)
DConnection 0.406
***

(4.53)
AT 0.341
**
0.345
**
0.336
**
0.345
**

(2.93) (3.07) (3.00) (3.14)
EBIT1 0.729
**
0.743
**
0.742
**
0.756
**

(2.65) (2.83) (2.67) (2.78)
STDEBIT -0.00000692
***
-0.00000691
***
-0.00000713
***
-0.00000741
***

(-5.58) (-6.20) (-5.00) (-4.58)
Lev -0.0626
***
-0.0588
***
-0.0629
***
-0.0620
***

(-9.51) (-13.97) (-8.83) (-12.88)
SGrow -0.0168 -0.0158 -0.000270 -0.00792
(-0.26) (-0.34) (-0.00) (-0.13)
Margin 0.00557 0.00474 0.00471 0.00532
(1.78) (1.39) (1.55) (1.67)
current1 -0.201
***
-0.194
***
-0.207
***
-0.202
***

(-13.52) (-18.14) (-17.47) (-13.38)
Auditorrank 0.424
**
0.380
**
0.396
**
0.414
**

(3.26) (3.08) (3.15) (3.27)
Boardsize -0.244
*
-0.239
*
-0.230 -0.203
(-2.11) (-2.16) (-1.76) (-1.54)
DualCEOChair -0.629
*
-0.518
*
-0.604
*
-0.559
*

(-2.54) (-2.17) (-2.48) (-2.24)
_cons 4.923
***
4.940
***
4.854
***
4.720
***

(5.94) (6.22) (5.64) (5.43)
N 545 545 545 545
adj. R
2
0.209 0.206 0.221 0.212

Table 5 reports the regression results for the effect of CEO/CFO/MGT political connections on long term issuer
entity ratings. Columns 1, 2, 3, 4 report results when issuer entity ratings are used as dependent variables.
Columns 1, 2, 3, 4 report results when CEO connections, CFO connections, MGT connections, and dummy of
political connections are used as independent variables. Columns 1, 2, 3, 4 report results when control variables
include issue maturity terms, dummy of collateral, log transformed total assets, EBIT scaled by total assets,
standard deviation of EBIT, leverage, sales growth, sales margin, current ratio, auditor rank, board size, and
duality of CEO and chair. To ensure that our inferences are not an artifact of a few extreme values, all variables
are winsorized at the top and bottom 1% . Results are derived from two way clustered regressions (Peterson
1997), in which observations are clustered by firm and by year, and standard errors are calculated which account
for two dimensions of within cluster correlation. T-statistics are in parentheses. ***, **, * indicate two-tailed
statistical significant of coefficient estimates at the 0.001, 0.01, 0.05 level.

50

5.3.3. Test of effects of political connections of lead arrangers
Table 6 reports that the political connections of bank underwriters significantly
and positively influence the issue size. Results provide evidence supporting H2, that is,
issuers with politically connected bank underwriters are more likely to issue a larger
amount of bonds than issuers without politically connected bank underwriters. Hiring one
of the top 12 bank underwriters is significantly positively associated with the issue size
(coeffi. = 0.176), and this result remains robust after controlling for the issuing firm’s
political connections summary measure, connectionfac. Hiring a state-owned bank
underwriter is significantly and positively associated with the issue size (coeffi. = 0.416),
and this result is robust when connectionfac is controlled for. In addition, how the
political connections of issuing firms and political connections of bank underwriters
jointly influence offering amounts is investigated. Results from model 6, Table 6 shows
that when both leadBIG and leadSOE is included in the full regression, the effect of
leadBIG on the issue size becomes insignificant; the coefficient of leadSOE on the issue
size remains to be significant (coeffi. = 0.367, t=3.78). This evidence supports that the
political connections of issuing firms and the political connections of bank underwriters
are alternative mechanisms when explaining the bond issue size.

51

Table 6: Tests of Substitution Effect between Political Connections of Issuing Firms
and Political Connections of Lead Arranger Underwriters in Affecting Offering
Amounts

Amount
i,t
= α
0
+ β
1
connectionfac
i,t
+ β
2
LeadBIG
i,t
+ β
3
LeadSOE
i,t
+β
4
Term
i,t
+ β
5
Collateral
i,t
+ β
6
AT
i,t
+
β
7
EBIT
i,t
+ β
8
STDEBIT
i,t
+ β
9
Lev
i,t
+ β
10
SGrow
i,t
+ β
11
Margin
i,t
+ β
12
Current
i,t
+ ε
i

(1) (2) (3) (4) (5) (6)
Amount Amount Amount Amount Amount Amount
connectionfac 0.151
*
0.159
**
0.137
*
0.140
**

(2.57) (2.82) (2.51) (2.62)
Term 0.0718
*
0.0690
*
0.0677
*
0.0680
*
0.0674
*
0.0664
*

(2.31) (2.14) (2.32) (2.27) (2.43) (2.41)
Collateral -0.635
**
-0.619
**
-0.568
*
-0.526
**
-0.494
**
-0.483
**

(-2.63) (-2.71) (-2.57) (-2.90) (-2.78) (-2.71)
AT 0.333
**
0.343
**
0.323
**
0.337
**
0.322
**
0.319
**

(2.76) (2.86) (2.74) (2.78) (2.67) (2.67)
EBIT1 0.255
*
0.238
*
0.233
**
0.229
*
0.228
*
0.223
**

(2.34) (2.56) (2.65) (2.48) (2.57) (2.65)
STDEBIT -0.00000211 -0.00000182 -0.00000202 -
0.00000252
**

-0.00000267
*
-
0.00000260
*

(-1.32) (-1.42) (-1.31) (-2.78) (-2.37) (-2.27)
Lev -0.0322
**
-0.0302
**
-0.0314
**
-0.0279
*
-0.0291
*
-0.0291
**

(-2.67) (-2.67) (-2.77) (-2.47) (-2.57) (-2.61)
SGrow -0.0553 -0.0631 -0.0660 -0.0315 -0.0341 -0.0392
(-1.71) (-1.92) (-1.95) (-1.03) (-1.11) (-1.30)
Margin -0.00126 -0.000434 -0.00156 0.000166 -0.000798 -0.000931
(-0.91) (-0.25) (-1.16) (0.08) (-0.46) (-0.56)
current1 -0.0822
*
-0.0534 -0.0744 -0.0564 -0.0754 -0.0734
(-2.09) (-1.38) (-1.73) (-1.58) (-1.88) (-1.79)
leadBIG 0.176
***
0.194
***
0.0632
(4.00) (4.24) (1.57)
leadSOE 0.416
***
0.399
***
0.367
***

(4.96) (4.50) (3.78)
_cons 1.101
*
0.898 1.017
*
0.540 0.669 0.675
(2.18) (1.82) (2.03) (1.22) (1.43) (1.45)
N 779 779 779 779 779 779
adj. R
2
0.199 0.188 0.212 0.223 0.241 0.241

Table 6 reports the regression results for the effect of firm issuer connection factor scores and lead arranger
underwriter connections on offering amounts. Columns 1, 2, 3, 4, 5, 6 report results when offering amounts are
used as dependent variables. Column 1 reports results when the factor score of firm issuer connections is used as
the independent variable. Column 2 reports results when the dummy variable of leadBIG is used as the
independent variable. Column 3 reports results when the factor score of firm issuer connections and the dummy
variable of leadBIG are used as independent variables. Column 4 reports results when the dummy variable of
leadSOE is used as the independent variable. Column 5 reports results when the factor score of firm issuer
connections and the dummy variable of leadSOE is used as independent variables. Column 6 reports results the
factor score of firm issuer connections, the dummy variable of leadSOE, and the dummy variable of leadSOE are
used as independent variables. Columns 1, 2, 3, 4. 5 report results when control variables include issue maturity
terms, dummy of collateral, log transformed total assets, EBIT scaled by total assets, standard deviation of EBIT,
leverage, sales growth, sales margin, and current ratio. To ensure that our inferences are not an artifact of a few
extreme values, all variables are winsorized at the top and bottom 1% . Results are derived from two way
clustered regressions (Peterson 1997), in which observations are clustered by firm and by year, and standard
errors are calculated which account for two dimensions of within cluster correlation. T-statistics are in
parentheses. ***, **, * indicate two-tailed statistical significant of coefficient estimates at the 0.001, 0.01, 0.05
level.

52

Table 7 reports that political connections of bank underwriters significantly affect
issuer credit ratings, but only for the measure of leadSOE. It appears that hiring a state-
owned bank underwriter is significantly and positively associated with the issuing firm’s
credit ratings (coeffi. = 0.517), and this result is robust after controlling for both the
issuing firm’s connection factor scores and the top-12 bank underwriter indicator
variable. The fact that the coefficient of leadBIG is insignificant and the coefficient of
leadSOE is significantly positive indicates that hiring a reputable underwriter does not
necessarily increase the issuing firm’s credit ratings, but that hiring a politically
connected underwriter increases the issuing firm’s credit ratings. This evidence is
consistent with my second hypothesis—the political connections of bank underwriters,
rather than the general reputation of bank underwriters affects the issuing firm’s ability to
raise bond capital in China.

53
Table 7: Tests of Substitution Effect Between Political Connections of Issuing Firms and Political Connections of Lead Arranger
Underwriters in Affecting Entity Ratings

Rentityrating
i,t
= α
0
+ β
1
connectionfac
i,t
+ β
2
LeadBIG
i,t
+ β
3
LeadSOE
i,t
+ β
4
Term
i,t
+ β
5
Collateral
i,t
+ β
6
AT
i,t
+ β
7
EBIT
i,t
+ β
8
STDEBIT
i,t
+ β
9
Lev
i,t
+ β
10
SGrow
i,t
+ β
11
Margin

i,t
+ β
12
Current
i,t
+ ε
i

(1) (2) (3) (4) (5) (6)
REntityrating REntityrating REntityrating REntityrating REntityrating REntityrating
connectionfac 0.199
***
0.250
***
0.223
***
0.229
***

(3.61) (9.77) (5.88) (7.39)
Term 0.144
***
0.155
***
0.153
***

(5.23) (5.92) (5.08)
Collateral -1.222
***
-1.251
***
-1.158
***

(-7.70) (-8.18) (-11.62)
AT 0.321
**
0.332
**
0.324
**
0.337
**
0.333
**
0.328
**

(2.96) (3.11) (3.04) (3.05) (2.95) (3.01)
EBIT1 0.630 0.618 0.725
*
0.621 0.723
**
0.720
*

(1.61) (1.57) (2.46) (1.65) (2.62) (2.52)
STDEBIT -0.00000500 -0.00000471 0.00000686
***
-0.00000564
*
0.00000784
***
0.00000766
***

(-1.69) (-1.76) (-4.08) (-2.35) (-5.87) (-5.14)
Lev -0.0590
***
-0.0545
***
-0.0600
***
-0.0510
***
-0.0564
***
-0.0559
***

(-10.45) (-7.33) (-11.07) (-11.15) (-8.74) (-9.59)
SGrow 0.0368 0.00560 -0.0311 0.0460 0.0197 0.00514
(0.86) (0.14) (-0.50) (1.79) (0.55) (0.13)
Margin 0.00496 0.00597
*
0.00449 0.00673
*
0.00578 0.00558
(1.74) (2.03) (1.59) (1.99) (1.79) (1.79)
current1 -0.181
***
-0.153
***
-0.198
***
-0.150
***
-0.187
***
-0.185
***

(-9.64) (-5.16) (-9.05) (-5.86) (-9.38) (-7.51)
Auditorrank 0.334
**
0.337
*
0.364
**
0.281
*
0.311
*
0.307
*

(2.76) (2.48) (2.70) (2.12) (2.32) (2.24)
Boardsize -0.189 -0.287
**
-0.286
*
-0.215 -0.210 -0.244
(-1.87) (-2.98) (-2.20) (-1.94) (-1.64) (-1.95)
DualCEOChair -0.519 -0.475 -0.568
*
-0.455 -0.546
*
-0.541
*

(-1.83) (-1.67) (-2.36) (-1.87) (-2.55) (-2.50)
leadBIG 0.224 0.287 0.131
(1.67) (1.78) (0.98)

54
Table 7, Continued

leadSOE 0.517
***
0.537
***
0.474
***

(6.64) (4.51) (6.06)
_cons 4.776
***
4.699
***
5.020
***
4.068
***
4.373
***
4.451
***

(6.46) (6.18) (5.88) (5.10) (4.72) (4.97)
N 542 542 545 542 545 545
adj. R
2
0.259 0.248 0.234 0.271 0.255 0.256

Table 7 reports the regression results for the effect of firm issuer connection factor scores and lead arranger underwriter connections on offering amounts. Columns 1, 2, 3, 4, 5, 6
report results when entity ratings are used as dependent variables. Column 1 reports results when the factor score of firm issuer connections is used as the independent variable.
Column 2 reports results when the dummy variable of leadBIG is used as the independent variable. Column 3 reports results when the factor score of firm issuer connections and
the dummy variable of leadBIG are used as independent variables. Column 4 reports results when the dummy variable of leadSOE is used as the independent variable. Column 5
reports results when the factor score of firm issuer connections and the dummy variable of leadSOE is used as independent variables. Column 6 reports results the factor score of
firm issuer connections, the dummy variable of leadSOE, and the dummy variable of leadSOE are used as independent variables. Columns 1, 2, 3, 4. 5 report results when control
variables include log transformed total assets, EBIT scaled by total assets, standard deviation of EBIT, leverage, sales growth, sales margin, current ratio, auditor rank, board size,
and duality of CEO and chair . To ensure that our inferences are not an artifact of a few extreme values, all variables are winsorized at the top and bottom 1% . Results are derived
from two way clustered regressions (Peterson 1997), in which observations are clustered by firm and by year, and standard errors are calculated which account for two dimensions
of within cluster correlation. T-statistics are in parentheses. ***, **, * indicate two-tailed statistical significant of coefficient estimates at the 0.001, 0.01, 0.05 level.

55
Table 8: Associations Between Political Connections of Issuing Firms and Political Connections of Lead Arranger Underwriters

Lead Arranger Connections
i,t
= α
0
+ β
1
connectionfac
i,t
+ β
2
Term
i,t
+ β
3
Collateral
i,t
+ β
4
AT
i,t
+ β
5
EBIT
i,t
+ β
6
STDEBIT
i,t
+ β
7
Lev
i,t
+ β
8
SGrow
i,t
+
β
9
Margin
i,t
+ β
10
Current
i,t
+ ε
i

(1) (2) (3) (4) (5) (6)
leadBIG leadSOE leadPRIVATE leadFOREIGN leadcommercialb leadinvestmentb
connectionfac -0.0150
*
0.0636
**
-0.0161
***
0.00362 0.0305
*
0.00557
(-2.20) (3.24) (-5.57) (0.30) (2.28) (0.24)
Term 0.0214
*
0.0111 -0.00488
**
0.00944
***
-0.0223
**
0.0322
**

(2.38) (0.96) (-2.75) (3.58) (-3.00) (3.06)
Collateral -0.349
***
-0.350
*
0.0570
*
0.000401 -0.841
***
0.701
***

(-3.36) (-2.18) (2.23) (0.02) (-18.95) (11.23)
AT 0.0494
***
0.0256 0.00150 0.00145 0.0466
**
0.0120
(3.35) (1.12) (0.38) (0.21) (3.15) (1.22)
EBIT1 0.111 0.0660 -0.0108 0.0460 0.0362 0.0341
(0.90) (1.11) (-1.42) (1.44) (0.77) (1.23)
STDEBIT -0.000000481 0.00000159 -0.000000477 0.000000198 0.000000489 -3.79e-08
(-0.80) (1.61) (-1.78) (0.80) (1.32) (-0.10)
Lev -0.00437 -0.00729 -0.000331 -0.000449 -0.00547 -0.00276
(-0.63) (-1.25) (-0.21) (-0.30) (-1.26) (-1.54)
SGrow 0.0545
*
-0.0503
**
0.0239 0.00302 -0.0192 0.0148
(2.20) (-2.92) (1.14) (0.34) (-0.46) (0.72)
Margin 0.00156 -0.00143 0.000457 0.0000632 -0.00224
*
0.000211
(0.85) (-0.94) (0.96) (0.24) (-2.23) (0.32)
current1 -0.0403
*
-0.0187
***
0.00394 0.000381 -0.0248
*
0.00740
(-2.29) (-3.37) (0.52) (0.06) (-2.40) (0.48)
_cons 0.469
***
1.079
***
0.0418
***
-0.0173 1.061
***
-0.0611
(6.06) (10.62) (4.22) (-0.55) (10.58) (-0.97)
N 779 779 779 779 779 779
adj. R
2
0.040 0.081 0.004 0.049 0.547 0.671

56
Table 8, Continued

Table 8 reports the regression results for the effect of factor scores of firm issuer connections on the political connections of lead arranger underwriters.
Column 1reports results when the independent variable is a dummy indicator variable of whether the lead arranger is a national top 6 commercial banks
or a national top 6 securities houses. Column 2 reports results when the dependent variable is a dummy indicator of whether the lead arranger is state
owned. Column 3 reports results when the dependent variable is a dummy indicator of whether the lead arranger is private held. Column 4 reports
results when the dependent variable is a dummy indicator of whether the lead arranger is foreign jointly held. Column 5 reports results when the
dependent variable is a dummy indicator of whether the lead arranger is a commercial bank. Column 6 reports results when the dependent variable is a
dummy indicator of whether the lead arranger is a securities house/investor bank. In columns 1, 2, 3, 4, 5, 6, CEO connections, CFO connections, MGT
connections, and dummy of political connections are used as independent variables; control variables include issue maturity terms, dummy of collateral,
log transformed total assets, EBIT scaled by total assets, standard deviation of EBIT, leverage, sales growth, sales margin, and current ratio. To ensure
that our inferences are not an artifact of a few extreme values, all variables are winsorized at the top and bottom 1% . Results are derived from two way
clustered regressions (Peterson 1997), in which observations are clustered by firm and by year, and standard errors are calculated which account for two
dimensions of within cluster correlation. T-statistics are in parentheses. ***, **, * indicate two-tailed statistical significant of coefficient estimates at the
0.001, 0.01, 0.05 level.

57
The choice of issuing firms in selecting lead arrangers when issuing bonds is
tested in Table 6, which indicates that connected firms are less likely to choose one of the
national top-12 bank underwriters (coeffi. = -0.015, t=-2.20), and connected firms are
more likely to choose state-owned bank underwriters (coeffi. = 0.064, t= 3.24), providing
mixed evidence in testing H3, the substitutes versus complements hypothesis. Politically
connected issuing firms are less likely to hire privately held bank underwriters (coeffi. = -
0.016, t= -5.57). These connected issuing firms are more likely to hire a commercial bank
to underwrite their bond issuances (coeffi. = 0.031, t=2.28).
5.4. Additional Analyses
This section offers additional analyses to further explore the following two issues:
First, the mechanism through which political connections influence firm access to bond
capital in China; and second, antecedents of firm political connections.
The sample is partitioned according to ownership type, information environment,
and geographic proximity in order to test which of the following arguments explains why
political connections matter in bond issuances: the reputation argument, the social
lending argument, or the collusion argument.
5.4.1. Test of the reputation enhancement argument
The reputation enhancement argument is tested by partitioning the research
sample into issuing firms that are publicly listed (PLCs) and issuing firms that are not
publicly listed (non-PLCs). Results from Table 9 show that political connection measures
correlate significantly and positively with offering amounts in the subsample of non-
PLCs, but not in the subsample of PLCs. Given that publicly listed issuing firms are

58
required to disclose more information than non-publicly listed issuing firms, PLCs
operate in a better information environment than for non-PLCs. Thus, PLC issuing firms
can use their disclosure advantage to establish their reputations and to distinguish
themselves from less reputable issuing firms, weakening the role of political connections
in the investors’ decision making process. On the other hand, non-PLC issuing firms lack
credible disclosure and are thus more likely to fall back on the political connections of
their executives to establish reputation. As shown in Table 9, the effect of political
connections is significant only in the non-PLCs subsample, but not in the PLCs
subsample. Results lend support to the reputation enhancement argument.

59
Table 9: Associations between Political Connections of Issuing firms and Issue Size in non-PLCs subsample and
PLCs Subsample

Amount
i,t
= α
0
+ β
1
Issuer Connections
i,t
+ β
2
Term
i,t
+ β
3
Collateral
i,t
+ β
4
AT
i,t
+ β
5
EBIT
i,t
+ β
6
STDEBIT
i,t
+ β
7
Lev
i,t
+ β
8
SGrow
i,t
+ β
9
Margin
i,t
+ β
10
Current
i,t
+ ε
i

(1) (2) (3) (4) (5) (6) (7) (8)
Amount Amount Amount Amount Amount Amount Amount Amount
CEOConnection 0.0755
*
0.0610
(2.53) (0.97)
CFOConnection 0.139
***
0.185
(3.97) (1.79)
MGTConnection 0.0458
***
0.0522
(3.42) (1.27)
DConnection 0.158 0.178
(1.96) (1.41)
Term 0.0567
*
0.0573
*
0.0578
*
0.0585
*
0.160
**
0.163
**
0.159
**
0.160
**

(2.09) (2.14) (2.23) (2.18) (2.80) (2.83) (3.01) (2.89)
Collateral -0.519
*
-0.513
**
-0.515
**
-0.544
**
-1.273
**
-1.300
**
-1.249
***
-1.250
***

(-2.47) (-2.72) (-2.65) (-2.82) (-3.28) (-3.16) (-3.70) (-3.42)
AT 0.217
*
0.218
*
0.213
*
0.219
*
0.629
*
0.597
*
0.620
*
0.636
*

(2.08) (2.20) (2.09) (2.10) (2.29) (2.01) (2.23) (2.32)
EBIT1 0.0850 0.109 0.0896 0.0982 0.581 0.526 0.575 0.584
(0.48) (0.66) (0.50) (0.54) (1.73) (1.62) (1.69) (1.65)
STDEBIT -0.000000116 -9.45e-08 -5.98e-08 -0.000000265 -0.00000831
*
-0.00000677
*
-0.00000860
*
-
0.00000842
*

(-0.09) (-0.08) (-0.06) (-0.28) (-2.31) (-2.17) (-2.25) (-2.14)
Lev -0.0238
**
-0.0215
**
-0.0231
**
-0.0224
**
-0.0417 -0.0394 -0.0414 -0.0414
(-2.90) (-3.17) (-2.74) (-2.81) (-1.93) (-1.89) (-1.97) (-1.97)
SGrow -0.0370 -0.0328 -0.0337 -0.0364 -0.0844 -0.0555 -0.0789 -0.0795
(-1.20) (-1.08) (-1.09) (-1.24) (-0.80) (-0.50) (-0.71) (-0.73)
Margin 0.000147 -0.000597 -0.000210 0.000188 -0.00178 -0.00258 -0.00251 -0.00150
(0.09) (-0.31) (-0.13) (0.11) (-0.60) (-0.92) (-0.99) (-0.50)
current1 -0.0864
**
-0.0891
***
-0.0922
**
-0.0818
**
0.135 0.151 0.143 0.138
(-2.77) (-3.46) (-3.24) (-2.80) (0.88) (1.04) (0.96) (0.88)

60
Table 9, Continued
_cons 1.539
***
1.551
***
1.518
***
1.511
***
-0.539 -0.426 -0.565 -0.617
(3.64) (3.78) (3.70) (3.70) (-0.50) (-0.38) (-0.53) (-0.58)
N 538 538 538 538 241 241 241 241
adj. R
2
0.126 0.141 0.133 0.119 0.362 0.381 0.374 0.362

Table 9 reports the regression results for the effect of CEO/CFO/MGT political connections on offering amounts. Columns 1, 2, 3, 4 report results when issuing firms are
not publicly listed companies, and columns 5, 6, 7, 8 report results when issuing firms are publicly listed companies. In 1-8 columns, offering amounts are used as
dependent variables; CEO connections, CFO connections, MGT connections, and dummy of political connections are used as independent variables; control variables
include issue maturity terms, dummy of collateral, log transformed total assets, EBIT scaled by total assets, standard deviation of EBIT, leverage, sales growth, sales
margin, and current ratio. To ensure that our inferences are not an artifact of a few extreme values, all variables are winsorized at the top and bottom 1% . Results are
derived from two way clustered regressions (Peterson 1997), in which observations are clustered by firm and by year, and standard errors are calculated which account for
two dimensions of within cluster correlation. T-statistics are in parentheses. ***, **, * indicate two-tailed statistical significant of coefficient estimates at the 0.001, 0.01,
0.05 level.

61
Table 10: Associations Between Political Connections of Issuing firms and Issue Size in Non-Beijing and Beijing Headquartered
Firms

Amount
i,t
= α
0
+ β
1
Issuer Connections
i,t
+ β
2
Term
i,t
+ β
3
Collateral
i,t
+ β
4
AT
i,t
+ β
5
EBIT
i,t
+ β
6
STDEBIT
i,t
+ β
7
Lev
i,t
+ β
8
SGrow
i,t
+ β
9
Margin
i,t
+ β
10
Current
i,t
+ ε
i

(1) (2) (3) (4) (5) (6) (7) (8)
Amount Amount Amount Amount Amount Amount Amount Amount
CEOConnection 0.0679 0.0234
(1.91) (0.54)
CFOConnection 0.137
**
0.0673
(2.90) (1.07)
MGTConnection 0.0397
*
0.0202
(2.45) (0.62)
DConnection 0.202
**
-0.0323
(2.64) (-0.27)
Term 0.0784
*
0.0780
*
0.0778
*
0.0792
*
0.0255 0.0293 0.0276 0.0260
(2.36) (2.41) (2.45) (2.52) (0.59) (0.66) (0.65) (0.59)
Collateral -0.518
*
-0.514
*
-0.508
*
-0.523
*
-0.607 -0.639 -0.614 -0.635
(-2.06) (-2.24) (-2.15) (-2.24) (-1.37) (-1.51) (-1.46) (-1.50)
AT 0.180
*
0.186
*
0.181
*
0.179
*
0.735
*
0.714
*
0.724
*
0.740
*

(1.98) (2.10) (2.03) (2.06) (2.20) (2.07) (2.13) (2.20)
EBIT1 0.151 0.170 0.158 0.163 0.477 0.462
*
0.460 0.480
(1.24) (1.35) (1.33) (1.30) (1.93) (1.99) (1.86) (1.92)
STDEBIT 0.00000179 0.00000166 0.00000174 0.00000169 -0.0000143
***
-0.0000134
***
-0.0000142
***
-0.0000141
***

(1.93) (1.78) (1.91) (1.74) (-3.94) (-3.54) (-4.34) (-4.35)
Lev -0.0332
***
-0.0320
***
-0.0334
***
-0.0327
***
-0.0117 -0.0112 -0.0106 -0.0110
(-5.17) (-4.72) (-5.11) (-5.47) (-0.32) (-0.30) (-0.29) (-0.30)
SGrow -0.00181 0.00992 0.00422 0.00486 -0.188 -0.200 -0.191 -0.195
(-0.08) (0.37) (0.17) (0.19) (-0.63) (-0.67) (-0.66) (-0.68)
Margin 0.00141 0.00112 0.00134 0.00161 -0.00141 -0.00266 -0.00209 -0.000754
(1.05) (0.76) (0.97) (1.20) (-0.22) (-0.46) (-0.36) (-0.11)
current1 -0.0741
**
-0.0771
**
-0.0782
**
-0.0745
**
-0.0118 -0.0163 -0.0149 -0.00514
(-2.59) (-3.16) (-2.93) (-2.82) (-0.10) (-0.13) (-0.11) (-0.04)
_cons 1.460
***
1.451
***
1.424
***
1.411
***
-0.398 -0.303 -0.373 -0.397
(4.07) (4.10) (4.14) (4.25) (-0.27) (-0.20) (-0.25) (-0.27)
N 571 571 571 571 208 208 208 208
adj. R
2
0.128 0.133 0.133 0.132 0.294 0.301 0.296 0.293

62

Table 10, Continued

Table 10 reports the regression results for the effect of CEO/CFO/MGT political connections on offering amounts. Columns 1, 2, 3, 4 report results when issuing firms are
headquartered in cities other than Beijing, and columns 5, 6, 7, 8 report results when issuing firms are headquartered in Beijing. In 1-8 columns, offering amounts are used as
dependent variables; CEO connections, CFO connections, MGT connections, and dummy of political connections are used as independent variables; control variables include issue
maturity terms, dummy of collateral, log transformed total assets, EBIT scaled by total assets, standard deviation of EBIT, leverage, sales growth, sales margin, and current ratio.
To ensure that our inferences are not an artifact of a few extreme values, all variables are winsorized at the top and bottom 1% . Results are derived from two way clustered
regressions (Peterson 1997), in which observations are clustered by firm and by year, and standard errors are calculated which account for two dimensions of within cluster
correlation. T-statistics are in parentheses. ***, **, * indicate two-tailed statistical significant of coefficient estimates at the 0.001, 0.01, 0.05 level.

63
Table 11: Associations Between Political Connections of Issuing Firms and Issue Size in Non-SOEs Subsample and SOEs
Subsample

Amount
i,t
= α
0
+ β
1
Issuer Connections
i,t
+ β
2
Term
i,t
+ β
3
Collateral
i,t
+ β
4
AT
i,t
+ β
5
EBIT
i,t
+ β
6
STDEBIT
i,t
+ β
7
Lev
i,t
+ β
8
SGrow
i,t
+ β
9
Margin
i,t
+ β
10
Current
i,t
+ ε
i

(1) (2) (3) (4) (5) (6) (7) (8)
Amount Amount Amount Amount Amount Amount Amount Amount
CEOConnection 0.199
*
0.0568
*

(2.13) (2.19)
CFOConnection 0.297
*
0.139
**

(2.44) (3.17)
MGTConnection 0.0896
**
0.0395
*

(3.29) (2.04)
DConnection 0.412
*
0.107
(2.48) (1.48)
Term 0.135 0.130 0.125 0.114 0.0660
*
0.0669
*
0.0669
*
0.0673
*

(1.03) (1.02) (0.99) (0.93) (2.17) (2.25) (2.30) (2.23)
Collateral -0.783 -0.776 -0.837 -0.797 -0.689
**
-0.684
**
-0.680
**
-0.705
**

(-0.87) (-0.89) (-1.13) (-1.22) (-2.71) (-2.90) (-2.73) (-2.85)
AT 0.0172 0.0461 0.0341 0.0327 0.354
***
0.346
**
0.348
***
0.358
***

(0.04) (0.11) (0.09) (0.08) (3.34) (3.29) (3.34) (3.37)
EBIT1 -0.196 -0.110 -0.170 -0.198 0.365
*
0.356
*
0.357
*
0.372
*

(-1.15) (-0.92) (-1.20) (-1.37) (2.53) (2.38) (2.40) (2.38)
STDEBIT 0.0000175
**
0.0000155
*
0.0000175
**
0.0000178
**
-0.00000376 -0.00000329 -0.00000379 -0.00000379
(3.29) (2.67) (3.48) (3.30) (-1.84) (-1.87) (-1.73) (-1.70)
Lev -0.0426
*
-0.0356 -0.0431
***
-0.0273
*
-0.0352
**
-0.0331
*
-0.0342
*
-0.0348
**

(-2.64) (-1.79) (-3.79) (-2.18) (-2.61) (-2.56) (-2.56) (-2.65)
SGrow -0.610
**
-0.625
**
-0.643
***
-0.665
***
-0.0426 -0.0332 -0.0380 -0.0403
(-2.77) (-2.74) (-3.51) (-4.16) (-1.26) (-0.86) (-1.10) (-1.14)
Margin -0.0143 -0.0140 -0.0148 -0.0139 0.000420 -0.000336 0.0000623 0.000613
(-0.79) (-0.74) (-0.83) (-0.77) (0.31) (-0.25) (0.05) (0.41)
current1 -0.265
***
-0.280
***
-0.263
***
-0.264
***
-0.0427
*
-0.0477
*
-0.0494
*
-0.0390
*

(-4.69) (-4.21) (-5.55) (-4.89) (-2.30) (-2.46) (-2.35) (-2.19)
_cons 2.258 2.183 2.155 2.145 0.965
*
1.007
*
0.954
*
0.935
*

(1.59) (1.45) (1.57) (1.54) (2.36) (2.51) (2.38) (2.32)
N 58 58 58 58 721 721 721 721
adj. R
2
0.201 0.186 0.215 0.202 0.197 0.215 0.204 0.193

64
Table 11, Continued

Table 11 reports the regression results for the effect of CEO/CFO/MGT political connections on offering amounts. Columns 1, 2, 3, 4 report results when
issuing firms are not state owned enterprises, and columns 5, 6, 7, 8 report results when issuing firms are state owned enterprises. In 1-8 columns, offering
amounts are used as dependent variables; CEO connections, CFO connections, MGT connections, and dummy of political connections are used as
independent variables; control variables include issue maturity terms, dummy of collateral, log transformed total assets, EBIT scaled by total assets,
standard deviation of EBIT, leverage, sales growth, sales margin, and current ratio. To ensure that our inferences are not an artifact of a few extreme
values, all variables are winsorized at the top and bottom 1% . Results are derived from two way clustered regressions (Peterson 1997), in which
observations are clustered by firm and by year, and standard errors are calculated which account for two dimensions of within cluster correlation. T-
statistics are in parentheses. ***, **, * indicate two-tailed statistical significant of coefficient estimates at the 0.001, 0.01, 0.05 level.

65
In addition, the effect of political connections is significant only in the subsample
of firms that is not headquartered in Beijing. Because greater geographic distance leads to
higher communication and information costs, bank underwriters and government
regulators are more likely to use political connections to make decisions if issuing firms
are located in provinces at a greater distance. The research sample is then partitioned into
two subsamples according to their geographic proximity to Beijing. The “home bias”
literature implies that lower communication costs exist between parties that are more
geographically proximate to each other, and that geographic proximity may influence a
set of economic behaviors. For example, retail investors tend to maintain a higher
fraction of local stocks in investment portfolios (Grinblatt and Keloharju 2001) due to
familiarity with those stocks and better information regarding those stocks (Ivkovich and
Weisbennar 2005). Mutual fund investors earn higher returns on their investments in
local firms, according to Coval and Moskowitz (2001). Analysts have been found to
make more accurate forecasts about firms located in closer proximity to their own
brokerage firms (Malloy 2005). Uysal, Kedia, and Panchapagesan (2005) reported that
acquirer firms earned higher acquirer returns for acquisitions within closer geographic
proximity. Geographic proximity also provides an information advantage in the context
of IPO syndicates, as noted by Schultz (2003). In this research setting of the Chinese
primary bond market, greater geographic distance leads to higher communication and
information costs. It is more likely that bank underwriters and government regulators will
fall back on political connections to make decisions if issuing firms are located in
provinces at a greater geographic distance. Thus, the political connections of firms

66
located farther from Beijing were expected to be positively associated with offering
amounts, but not necessarily so for firms headquartered in Beijing. The results, as
indicated in Table 8, supported this argument. The effect of political connections on
offering amounts were only significantly positive in the subsample of firms that was not
headquartered in Beijing, not in the subsample of firms that was.
5.4.2. Test of the social lending argument
The whole sample is partitioned into SOEs subsample and non-SOEs subsample
to test the social lending argument; the results are shown in Table 9. According to the
social lending argument, the Chinese government subsidizes firms that assume more
social responsibilities, such as advancing national technologies, providing more job
opportunities, and taking socially beneficial but individually risky projects. An anecdotal
example would be the “Three Gorges Bond” issuance, with a total amount of 3 billion
RMB Yuan and no bank guarantees or collateral terms. The social lending argument may
account for the relationship between stronger political connections and higher offering
amounts, because if the Chinese government wanted a specific firm to assume more
social responsibility and to take socially beneficial projects, and this firm were to receive
easier access to financing as a reward for doing so, it is rational for the Chinese
government to assign its representatives to firm management teams to ensure the
allocation of financial resources to these socially beneficial projects. Given the monopoly
status of SOEs in Chinese national strategic industries, I assume that the Chinese
government is more likely to select SOEs than non-SOE if socially beneficial projects
were available. Thus, if the social lending argument holds, I expect to find that the effect

67
of political connections on offering amounts was significantly positive in the SOEs
subsample, but not in the non-SOEs subsample. However, as shown in Table 11, the
effects of political connections are not only significantly positive in the SOEs subsample,
but also significantly positive in the non-SOEs subsample. Thus, the analysis does not
support the social lending argument.
5.4.3. Test of the collusion argument
The collusion argument suggests that during the bond issuance process
government regulators favor firms with politically influential executives, hoping for
future reciprocity when they need to fulfill their personal political agendas and advance
their careers. To test this argument, I use a mean-centered interaction test to examine
whether the effect of political connections on offering amounts was larger in poorly
governed firms. In well-governed firms, in which minority investors’ rights are well
protected and boards of directors effectively monitor corporate management, it is more
difficult for the executives to appropriate corporate resources to fulfill their personal
needs. Thus, when corporate governance functions well, managers are less likely to use
corporate resources to reciprocate politicians’ favors in granting approvals in bond
issuances. If the collusion argument held, officials of bond issuance regulating agencies
would favor firms with strong political connections and poor corporate governance in
making bond issuance approval decisions. The collusion argument predicts that the
difference in the offering amount between politically connected firms and non-connected
firms would be larger in firms with poor governance.

68
A negative coefficient on the interaction term between political connections and
corporate governance measures suggests that the difference in access to bond capital
between politically connected firms and non-connected firms is larger in firms with
poorer governance. Table 12 shows that the interaction term is not significant; thus the
analysis does not lend support to the collusion argument.

69
Table 12: Interaction Test of Corporate Governance, Organizational Complexity, and
Political Connections of Issuing Firms

Amounts
i,t
= α
0
+ β
1
CityPR
i,t
+ β
2
Ownership
i,t
+ β
3
Auditorrank
i,t
+ β
4
Boardsize
i,t
+ β
5
DualCEOChair
i,t

+β
6
NumownedBU
i,t
+β
7
Numprovinces
i,t
+ β
8
AT
i,t
+ β
9
EBIT
i,t
+ β
10
STDEBIT
i,t
+ β
11
Lev
i,t
+ β
12
SGrow
i,t
+
β
12
Margin
i,t
+ β
14
Current
i,t
+ ε
i

(1) (2) (3) (4) (5)
Amount Amount Amount Amount Amount
connectionfac 0.103
**
-0.154 0.180
***
0.184
*
0.171
***

(2.63) (-0.72) (4.65) (2.20) (3.47)
Auditorrank 0.257
***

(5.44)
pcauditor 0.0652
(1.56)
Term 0.0660
***
0.105
***
0.105
***
0.0619
***
0.0867
***

(5.45) (5.61) (6.55) (4.44) (5.36)
Collateral -0.567
***
-0.934
***
-0.924
***
-0.590
***
-0.635
***

(-4.59) (-5.76) (-6.39) (-4.23) (-4.19)
AT 0.316
***
0.338
***
0.328
***
0.306
***
0.305
***

(9.51) (9.09) (9.41) (7.98) (8.82)
EBIT1 0.240
*
0.159 0.272
*
0.314
*
0.336
**

(2.13) (1.14) (2.32) (2.47) (2.71)
STDEBIT -0.00000213 -0.00000215 -0.00000177 -0.00000245 -0.00000272
(-1.01) (-0.90) (-0.81) (-1.03) (-1.22)
Lev -0.0371
***
-0.0414
***
-0.0396
***
-0.0307
**
-0.0450
***

(-3.89) (-3.77) (-3.96) (-2.65) (-4.62)
SGrow -0.0330 -0.0907 -0.0310 -0.0826 -0.0297
(-0.60) (-1.31) (-0.52) (-1.26) (-0.51)
Margin -0.00170 -0.00261 -0.00154 0.000271 -0.000651
(-0.89) (-1.14) (-0.73) (0.11) (-0.29)
current1 -0.0674
**
-0.0876
**
-0.0699
*
-0.0339 -0.0933
**

(-2.60) (-2.91) (-2.56) (-1.12) (-3.22)
Boardsize 0.00735
(0.07)
pcboardsize 0.151
(1.45)
DualCEOChair -0.0829
(-0.97)
pcdual -0.0910
(-1.20)
NumownedBU 0.182
***

(4.50)
pcnumBU -0.0308
(-0.87)
Numprovinces 0.165
***

(4.42)
pcnumprovince -0.0369
(-1.23)
_cons 1.069
***
1.082
***
1.081
***
0.837
***
1.041
***

(6.78) (3.65) (6.48) (4.25) (6.05)
N 776 635 704 541 614
adj. R
2
0.235 0.221 0.219 0.225 0.244

70
Table 12, Continued

Table 12 reports the regression results for the interaction effect of corporate governance measures and
firm issuer political connections, and the interaction effect of organizational complexity measures and
firm issuer political connections. Columns 1, 2, 3, 4, 5 report results when offering amounts are used as
dependent variables. Columns 1, 2, and 3 test the interaction effect of corporate governance measures and
firm issuer political connections, with corporate governance proxied by auditor rank, board size, and
duality of CEO and chair. Columns 4, 5 report the interaction effect of organizational complexity
measures and firm issuer political connections, with organizataional complexity proxied by number of
business units owned and number of provinces covered. In columns 1, 2, 3, 4, and 5, control variables
include log transformed total assets, EBIT scaled by total assets, standard deviation of EBIT, leverage,
sales growth, sales margin, and current ratio. To ensure that our inferences are not an artifact of a few
extreme values, all variables are winsorized at the top and bottom 1% . Results are derived from two way
clustered regressions (Peterson 1997), in which observations are clustered by firm and by year, and
standard errors are calculated which account for two dimensions of within cluster correlation. T-statistics
are in parentheses. ***, **, * indicate two-tailed statistical significant of coefficient estimates at the 0.001,
0.01, 0.05 level.

5.4.4. Determinants of firm political connections
Table 13 presents how firm political connections are influenced by firm
characteristics, governance variables, and financial performance measures. Firms with
headquarters located in national capital cities are more likely to have connected
executives; likewise, state-owned firms are more likely to have connected executives.
Firms with larger assets, higher current ratios, and higher sales margins are more likely to
have connected executives. No significant relationship exists between political
connections measures and profitability measures (ROE), which is in line with Malesky
and Taussig’s (2008) findings in a sample of 6,400 Vietnamese firms that firms with
greater access to bank loans were no more profitable than firms without them.

71
Table 13: Associations Between Corporate Governance, Organizational Complexity,
Financial Performances of Issuing Firms and Political Connections of Issuing Firms

Issuer Connections
i,t
= α
0
+ β
1
CityPR
i,t
+ β
2
Ownership
i,t
+ β
3
Auditorrank
i,t
+ β
4
Boardsize
i,t
+ β
5
DualCEOChair
i,t

+β
6
NumownedBU
i,t
+β
7
Numprovinces
i,t
+ β
8
AT
i,t
+ β
9
EBIT
i,t
+ β
10
STDEBIT
i,t
+ β
11
Lev
i,t
+ β
12
SGrow
i,t
+
β
12
Margin
i,t
+ β
14
Current
i,t
+ ε
i

(1) (2) (3) (4)
CEOConnection CFOConnection MGTConnection DConnection
CityPR 0.319
***
0.0995 0.774
***
0.125
***

(3.42) (1.59) (3.90) (3.55)
Ownership 0.360
*
0.353
**
0.400 0.00464
(2.39) (2.86) (1.22) (0.08)
Auditorrank -0.115 0.214 0.268 0.0287
(-0.80) (1.80) (0.89) (0.68)
Boardsize 0.0473 -0.176 0.0658 -0.0906
(0.18) (-0.88) (0.13) (-1.19)
DualCEOChair 0.351 -0.321
**
0.292 -0.00649
(1.75) (-2.64) (0.71) (-0.10)
NumownedBU 0.0275 -0.0598 0.123 0.0243
(0.33) (-0.88) (0.77) (0.90)
Numprovinces -0.0751 0.144
*
-0.115 -0.0691
*

(-0.85) (2.13) (-0.60) (-2.26)
AT 0.143
*
0.106 0.330
*
0.0422
(2.21) (1.82) (2.39) (1.76)
EBIT1 0.367 0.157 0.131 -0.0895
(1.28) (0.84) (0.18) (-0.66)
STDEBIT -0.00000629 -0.00000447 -0.00000308 0.000000978
(-1.86) (-1.76) (-0.39) (0.67)
Lev -0.0219 -0.0179 -0.0457 -0.000422
(-1.54) (-1.74) (-1.48) (-0.06)
SGrow 0.00445 -0.0587 -0.211 -0.0191
(0.04) (-0.80) (-1.06) (-0.52)
Margin 0.00778 0.0114
*
0.0263
*
0.00304
(1.51) (2.40) (2.35) (1.95)
current1 0.162
***
0.0889
*
0.379
***
0.0615
***

(5.05) (2.12) (4.50) (6.40)
_cons -0.928 -0.539 -1.503 0.235
(-1.29) (-0.86) (-0.99) (1.07)
N 407 407 407 407
adj. R
2
0.067 0.133 0.102 0.083

Table 13 reports the regression results for the effect of corporate governance measures, organizational
complexity measures, and financial performance measures on firm issuer political connections. Columns
1, 2, 3, 4 report results when CEO connections, CFO connections, MGT connections, and dummy
indicator of connections are used as dependent variables. Columns 1, 2, 3, 4 report results when
independent variables include political rank of firm headquarter located cities, firm ownership type,
auditor rank, board size, duality of CEO and chair, number of owned business units, number of provinces
covered, log transformed total assets, EBIT scaled by total assets, standard deviation of EBIT, leverage,
sales growth, sales margin, and current ratio. To ensure that our inferences are not an artifact of a few
extreme values, all variables are winsorized at the top and bottom 1% . Results are derived from two way
clustered regressions (Peterson 1997), in which observations are clustered by firm and by year, and
standard errors are calculated which account for two dimensions of within cluster correlation. T-statistics
are in parentheses. ***, **, * indicate two-tailed statistical significant of coefficient estimates at the 0.001,
0.01, 0.05 level.

72
5.5. Robustness Checks
The following robustness checks are performed to increase the statistical validity
of my empirical analyses. First, Firm’s domestic listing status serves as a proxy for the
information environment in the main tests. It is possible that the disclosure requirement is
equally high for Chinese firms listed in foreign capital markets. Thus, as a robustness
check, a dummy variable is coded as 1 if a firm is domestically listed, internationally
listed, or cross-listed. Results remain robust, that is, results indicate that the role of
political connections was significant in non-PLCs but not in PLCs. However, if the
dummy variable is coded in the manner that equals 1 if a firm was listed or its affiliated
companies were listed, and 0 otherwise, the role of political connections was significant
in both the PLCs subsample and the non-PLCs subsample, suggesting that the listing
status of firms’ affiliated companies is not a good measure to proxy for the information
environment of the firm itself.
Second, both bond level of analysis and firm level of analysis are conducted as
robustness checks, results remained qualitatively the same. The firm level of analysis is
more appropriate for addressing the research question of whether having political
connections facilitate a firm’s ability in raising bond capital; the firm level of analysis is
reported in the main results section.
Third, an alternative explanation is that the government makes bond approval
decisions based on provincial development levels. To control for the different levels of
provincial development, three province level indices are included: credit market index,
government decentralization index, and higher legal environment index. The higher credit

73
market index scores refer to more developed credit markets, the higher government
decentralization index scores refer to less government intervention, and higher legal
environment index scores refer to more developed legal institutions. Results are robust
after controlling for organizational complexity measures, corporate governance measures,
and province indices measures. Furthermore, results remain robust after adding listing
status dummy indicator, SOE dummy indicator, and Beijing headquarter dummy
indicator as additional controls in both the offering amount regression model and the
credit ratings regression model.
Fourth, to test whether the geographic proximity influences the effect of the
issuing firms’ political connections and issue size, the geographic distance from issuing
firms’ headquarters to the government regulators in Beijing are used. Another relevant
geographic proximity measure is the distance from issuing firms’ headquarters to the lead
arrangers’ headquarters. According to this criterion, the research sample is partitioned
into two subsamples based on whether issuing firms’ headquarters and the lead arrangers’
headquarters are located in the same city, and the regression model (1) and (2) are applied
respectively in these two subsamples. Results remain qualitatively consistent. Political
connections of issuing firms that are located in the same city as the lead arrangers are not
significantly associated with offering amounts, and political connections of issuing firms
that are located not in the same city as the lead arrangers are significantly associated with
offering amounts.

74
Chapter 6. Conclusion and Discussion
6.1 Summaries and Conclusions
This study suggests that political connections facilitate Chinese firms interested in
raising bond capital. An issuing firm’s political connections are significantly associated
with the offer size even after controlling for financial performance measures (size,
profitability, leverage, current ratio, volatility of cash flows) and debt contracting
variables (bond maturity terms, bond collateral terms). An issuing firm’s political
connections are also significantly associated with the firm’s credit ratings after
controlling for financial performance measures and corporate governance measures
(board size, rank of auditor, and duality of CEO and chair). In addition, the political
connections of issuing firms and the political connections of bank underwriters are
alternative mechanisms that can explain a firm’s ability to raise the bond capital. Results
suggest that by having someone with government experience directly allied to the firm, a
firm may gain the benefits of favorable governmental action.
Three theoretical arguments, the reputation enhancement argument, the social
lending argument, and the collusion argument, may explain this relationship. This study
uses a sample-partitioning design to explore these arguments separately. Results support
the reputation enhancement argument. Political connections are associated with larger
offering amounts, but only in the subsample of firms in a poor information environment,
such as non-publicly listed companies or firms that are not headquartered in Beijing. The
evidence indicates that when quality disclosure is absent, political connections serve as
an alternative channel that facilitates information communication between firms and bond

75
issuance regulatory agencies in the bond issuance process. Further, this study provides
evidence that political connections of firm executives are associated with higher credit
ratings, suggesting that political connections help signal a firm’s reputation in not
defaulting on bonds.
This study partitions the sample into state-owned enterprises and private
enterprises to test the social lending argument. Considering the dominant role of state-
owned enterprises in the Chinese economy, when the government selects firms for
socially beneficial projects, it is more likely to choose state-owned enterprises. This
indicates that the relationship between political connections and bond issuance amount (a
proxy for government subsidizing capital to involve these firms in socially beneficial
projects) should only maintain in the subsample of state-owned enterprises. Because this
relationship holds in both the subsample of state-owned enterprises and private
enterprises, it is unlikely that the social lending argument is descriptive of the empirical
data.
To test the collusion argument, that is, that politicians and executives collude
during the bond issuances, this study investigates whether the role of political
connections is more important in the poor-governed firms. The collusion argument
predicts that it is easier for executives with political connections to return favors to
politicians using firm resources only in poor-governed firms, and results from the
interaction test are not consistent with the collusion argument.

76
6.2 Contributions
This study addresses an important broad issue: In emerging markets where firms
thirst for capital, what can facilitate a firm’s efforts in raising capital? Some prior studies
have established the link between political connections and preferential capital access
without explaining the mechanism through which political connections contribute to
preferential capital access. As noted above, this study uses the Chinese bond issuance
market research setting to test three theoretical arguments that may contribute to this
phenomenon: the reputation enhancement argument, the social lending argument, and the
collusion argument. This study is also the first attempt to explore the role of political
connections of bank underwriters in bond issuance process in emerging markets. The
evidence contributes to the literature on bond underwriter choice by establishing how the
political connections of issuing firms and political connections of bond underwriters can
serve as substitute mechanisms for obtaining government approval to issue bonds.
This dissertation explores an important channel through which political
connections operate - career concerns of connected firm executives facilitate a firm’s
reputation in not defaulting on bonds, and establishes some preliminary evidence for the
reputation enhancement argument. Prior literature that investigates the benefit of political
connections focuses on the regulatory treatment (better property protection, cheaper
capital, lower tax, and more stipend and government contracts) from the government’s
perspective. This study is the first paper that discusses how executives’ career concerns
brought by political connections influence firm financing behavior. The major reward of
maintaining a good reputation is career advancement, that is the career mobility between

77
politics and business motivated connected executives to remain an impeccable history to
prepare them to run for politics. The punishment of not keeping a good reputation is more
severe for connected executives than for non-connected executives. Since connected
executives are mostly communist party members, connected executives in SOEs may get
deductions in their performance evaluation scores assigned by the SASAC; connected
executives in private firms may be asked to be investigated at designated time & place
(“Double Regulations”) and face immediate career termination. Considering the
institutional background in China that there is great career mobility for business elites to
run for politics, connected executives want to remain an impeccable history to the
government and the media, which helps to signal a firm’s reputation in not defaulting on
bonds.
This study suggests that connected firm executives face two types of incentives,
the incentive to collude with politicians and the incentive for their own career advances in
politics. Political connections are not necessarily always beneficial for firm value,
because career concerns of politically connected executives may be aligned with the
company value in some cases, but may not be the case in other settings. It remains an
empirical question whether and when these incentives are in conflicts, and which
incentive dominates under what circumstances.
This dissertation also has implications for practitioners. Chinese firms raised over
50% more from bonds than from equities during the year 2010, and the bond financing
amounts to 1585.08 billion RMB (226.43 billion USD) in 2010. Although the
government has promised to reduce direct intervention of the state economy gradually,

78
this study suggests that the percentage of firms that are politically connected in the debt
issuance sample has steadily increased from 25% in the year 2001 to 54.3% in the year
2009. This study also provides evidence on whether having a politically connected
executive helps a firm’s ability in raising debt capital. The answer is yes but only in
certain subsamples. Politically connections are associated with a higher level of bond
issuance amount, but only in the subsamples of non-listed firms and non-Beijing
headquartered firms. The result indicates that boundary conditions should also be taken
into considerations in future political connections studies.
6.3 Limitations and Future Research
There are some limitations to this study that need to be addressed. Disclosing
these limitations also suggests other opportunities for future research. First, this study
uses the career history of firm executives as a measure of political connections. More
broadly, there are two types of political connections. The first type of connection relates
to the political objectives of the Chinese government (for example, firms that are owned
by central government, firms that compete in strategic industries, and firms that are
relevant to the nation’s technique advances and national defense). The second type of
connection is the private agenda and personal interests of politicians. This study uses the
concept of executive identity and thus does not completely disentangle these two types of
political connections.
Second, this dissertation uses sample partition techniques to test the reputation
enhancement argument. If the reputation argument holds, then executives that become
politicians in the future, or executives who run to become politicians in the future, there

79
is a higher desire for them to remain an impeccable history and to keep a good reputation.
They are more likely to try to make sure they don’t default on bond. A more direct test is
to keep track of these executives and divide the sample to 2001-2005 and 2005-2009 and
test this possibility. If results suggest that someone who is run in the future is more likely
to care about their reputation, then this would be stronger evidence to support the
reputation enhancement argument.
Third, an alternative explanation for my result is the government bailout
hypothesis. La Porta et al. (2003) found that companies with direct links to Mexican
banks received better terms and were more likely to default on a sample of 300 loans in
Mexico. They also found that related loans were more likely to default, and when they
did, have significantly lower recovery rates than unrelated loans. It is possible that higher
credit ratings reflect the belief held by analysts and investors that the government is more
likely to give politically connected firms bailout money when these firms become
distressed. The evidence that political connections of bank underwriters matter can help
rule out the government bail-out hypothesis. If the government bailout story holds, then it
should only be the political connections of firm issuers that matter, but not the political
connections of lead arrangers matter. Once the lead arrangers help firms issue debt, they
are out of the picture and market participants cannot count on the political connections of
lead arrangers to influence the government’s decision in giving bailout money. The
market participants can only count on the political connections of firm issuers to get more
bailout money once they are in financial stress. A more direct test of this alternative
explanation would be to follow up on these bond issuances and enrich this hand-collected

80
database by adding the credit ratings granted after the initial bond issuance to directly test
the government bailout hypothesis against the reputation enhancement hypothesis.
Fourth, results suggest that politically connected firms are more likely to hire
state-owned bank underwriters and that hiring a state-owned bank underwriter is
associated with larger offer size and higher credit ratings. A likely follow-up question is
whether these state-owned bank underwriters charge premium underwriting fees to
extract rents from their political connections. For firms with political connections, why
do they need bank underwriters to be connected at all? Why don’t they need higher
general reputation, more extensive distribution network, more experience in the industry?
For firms without political connections, are they buying political connections from the
bank underwriters? Do they pay premium in underwriting fees to buy political
connections from the bank underwriters? Is this a complementary story or a substitute
story? Additional data on underwriting fees charged by bank underwriters in the Chinese
bond market may help addressing this question. These and other potential limitations
suggest a number of fruitful areas for future research.

81
Bibliography
Adelson, M. 2009. “Understanding Standard & Poor’s Rating Definitions.” Ratings
Direct. Retrieved at: www.standardpoors.com/ratingsdirect.
Adhikari, A., C. Derashid, and H. Zhang. 2006. “Public policy, political connections, and
effective tax rates: Longitudinal evidence from Malaysia.” Journal of Accounting and
Public Policy 25 (5) (Sep-Oct): 574-95.
Aggarwal, R. K., F., Meschke, and T. Y., Wang. 2009. “Corporate political contributions:
Investment or agency?” EFA 2008 Athens Meetings Paper, available at SSRN:
http://ssrn.com/abstract=972670.
Agrawal, A., and C. R. Knoeber. 2001. “Do some outside directors play a political role?”
Journal of Law and Economics 44 (1) (Apr): 179-98.
——–, and ——–. 1996. “Firm performance and mechanism to control agency problems
between managers and shareholders.” Journal of Financial and Quantitative Analysis
31: 377-97.
Aharoni, Y. 1981. “Performance evaluation of state-owned enterprises: A process
perspective.” Management Science 27, (11) (Nov): 1340-47.
Allen, F., J. Qian, and M. Qian. 2005. “Law, finance, and economic growth in China.”
Journal of Financial Economics 77 (1): 56-116.
Ashbaugh-Skaife, H., D. W. Collins, and R. LaFond. 2006. “The effects of corporate
governance on firms’ credit ratings.” Journal of Accounting and Economics 42 (1-2):
203-43.
Baron, D. P. 1988. “Regulation and legislative choice.” The Rand Journal of Economics
19 (3) (Autumn): 467-77.
Barth, M., and W. Landsman, 2003. Cost of capital and the quality of financial statement
information. Working Paper, Stanford University and the University of North
Carolina.
Bharath, S.T., J. Sunder, and S.V. Sunder, 2008. Accounting quality and debt contracting.
The Accounting Review: 83 (1): 1-28.
Bhattacharya, N., H. Desai, and K. Venkataraman, 2009a. Earnings quality and
information asymmetry: evidence from trading cost. Working paper, Southern
Methodist University.

82
Berger, A. N., and G. F. Udell. 2002. “Small business credit availability and relationship
lending: The importance of bank organizational structure.” The Economic Journal
112 (477): 32-53.
Berkman, H., R. A. Cole, and L. J. Fu. 2010. “Political connections and minority-
shareholder protection: Evidence from Securities Market Regulation in China.”
Journal of Financial and Quantitative Analysis, forthcoming.
Bessembinder, H. and W. Maxwell, 2008. Transparency and the Corporate Bond Market.
Journal of Economic Perspectives, Spring.
Brick, I.E. and S.A. Ravid, 1985. On the relevance of debt maturity structure. The Journal
of Finance, Vol. XL, No. 5, 1423-1437.
Chaney, P. K., M. Faccio, and D. Parsley. 2010. “The quality of accounting information
in politically connected firms.” Journal of Accounting and Economics:
doi:10.1016/j.jacceco.2010.07.003.
Charumilind, C., R. Kali, and Y. Wiwattanakantang. 2006. Connected lending: Thailand
before the financial crisis. Journal of Business 79 (1): 181-218.
Cheung, Y., P. R. Rau, and A. Stouraitis. 2006. “Tunneling, propping, and expropriation:
Evidence from connected party transactions in Hong Kong.” Journal of Financial
Economics 82 (2) (Nov): 343-86.
Claessens, S., E. Feijen, and L. Laeven. 2008. “Political connections and preferential
access to finance: The role of campaign contributions.” Journal of Financial
Economics 88 (3) (Jun): 554-80.
Coate, S. 2004. “Political competition with campaign contributions and informative
advertising.” Journal of the European Economic Association 2 (5): 772-804.
Correia, M. M. 2009. “Political connections, SEC enforcement and accounting quality.”
Rock Center for Corporate Governance at Stanford University Working Paper,
available at SSRN: http://ssrn.com/abstract=1458478.
Coval, J. D., and T. J. Moskowitz. 2001. “The geography of investment: Informed trading
and asset prices.” The Journal of Political Economy 109, (4) (Aug): 811-41.
———. 1999. “Home bias at home: Local equity preference in domestic portfolios.” The
Journal of Finance 54, (6) (Dec): 2045-73.

83
Cull, R., and L. C. Xu. 2000. “Bureaucrats, state banks, and the efficiency of credit
allocation: The experience of Chinese state-owned enterprises.” Journal of
Comparative Economics 28 (1) (Mar): 1-31.
Crabbe, L., and C. Turner. 1995. “Does the liquidity of a debt issue increase with its size?
Evidence from the corporate bond and medium-term note markets.” Journal of
Finance 50: 1719-34.
Crouhy, M., D. Galai, and R. Mark. 2001. “Prototype risk rating system.” Journal of
Banking & Finance 25: 47-95.
Cull, R., and L. C. Xu. 2000. “Bureaucrats, state banks, and the efficiency of credit
allocation: The experience of Chinese state-owned enterprises.” Journal of
Comparative Economics 28, (1) (Mar): 1-31.
———, and ———. 2005. “Institutions, ownership and finance: The determinants of
profit reinvestment among Chinese firms.” Journal of Financial Economics 77: 117-
46.
Diamond, D.W. 1985. Optimal releases of information by firms. Journal of Finance 40,
1071-94.
———, and R. Verrecchia, 1991. Disclosure, liquidity and the cost of capital, Journal of
Finance 46, 1325-59.
Dichev, I., and J. Piotroski. 2001. The long-run stock returns following bond ratings
changes. Journal of Finance 56: 173-203.
Dinc, I. S. 2005. “Politicians and banks: Political influences on government-owned banks
in emerging markets.” Journal of Financial Economics 77 (2) (Aug): 453-79.
Dixit, A., G. M. Grossman, and E. Helpman. 1997. “Common agency and coordination:
General theory and application to government policy making.” The Journal of
Political Economy 105 (4) (Aug): 752-69.
Du, F., S. M. Young, and G. Tang, 2010. Dysfunctional behavior and subjective
performance evaluation: Evidence from Chinese state-owned enterprises. University
of Southern California Working Paper.
Duchin, R., and D. Sosyura. 2010. “TARP Investments: Financials and politics.” Ross
School of Business Working Paper, available at SSRN:
http://ssrn.com/abstract=1426219.

84
Easley, D., and M. O’Hara, 2004. Information and the cost of capital. Journal of Finance
59:1553-83.
Farh, J., A. S. Tsui, K. Xin, and B. Cheng. 1998. “The influence of relational
demography and guanxi: The Chinese case.” Organization Science 9, (4) (Jul/Aug):
471-88.
Faccio, M. 2006. “Politically connected firms.” The American Economic Review 96 (1)
(Mar): 369-86.
——–, R. W. Masulis, and J. J. Mcconnell. 2006. “Political connections and corporate
bailouts.” The Journal of Finance 61 (6) (Dec): 2597-635.
——–, and D. C. Parsley. 2009. Sudden deaths: Taking stock of geographic ties. Journal
of Financial and Quantitative Analysis 44 (3): 683-718.
Fan, J. P. H., T. J. Wong, and T. Zhang. 2007. “Politically connected CEOs, corporate
governance, and post-IPO performance of China's newly partially privatized firms.”
Journal of Financial Economics 84 (2) (May): 330-57.
Ferguson, and T., H. Voth. 2008. “Betting on Hitler – The value of political connections
in Nazi Germany.” The Quarterly Journal of Economics 123 (1) (Feb): 101-37.
Fisman, R. 2001. “Estimating the value of political connections.” The American
Economic Review 91 (4) (Sep): 1095-102.
Fisman, D., R. Fisman, J. Galef, and R. Khurana. 2006. “Estimating the value of
connections to vice-president Cheney.” Unpublished working paper, available at
http://fairmodel.econ.yale.edu/ec483/ffgk.pdf.
Francis, J., R. LaFond, P. Olsson, K. Schipper. 2005. “The market pricing of accruals
quality.” Journal of Accounting and Economics 39: 295–327.
Garzarelli, F., S. Lawson, M. Vaknin, Z. Sheng, and T. Xu. 2006. “Bonding the BRICs:
The ascent of China’s debt capital market.” Goldman Sachs, Global Economics Paper
149, November.
Goldman, E., J. Rocholl, and J. So. 2009. “Do politically connected boards affect firm
value?” The Review of Financial Studies 22 (6) (Jun): 2331-60.
Guriev, S. 2004. “Red tape and corruption.” Journal of Development Economics 73 (2):
489-504.

85
Guthrie, D. 1998. “The declining significance of Guanxi in China’s economic transition.”
China Quarterly 154: 254-82.
Graham, R. C., and R. D. Kingy. 2000. “Accounting practices and the market valuation
of accounting numbers: Evidence from Indonesia, Korea, Malaysia, the Philippines,
Taiwan, and Thailand.” The International Journal of Accounting 35 (4): 445-70.
Grinblatt, M., and M. Keloharju. 2001. “How distance, language, and culture influence
stockholdings and trades.” The Journal of Finance 56, (3) (Jun): 1053-73.
Hay, J. R., and A. Shleifer. 1998. “Private enforcement of public laws: A theory of legal
reform.” The American Economic Review 88 (2): 398-403.
Hellman, J. S., G. Jones, and D. Kaufmann. 2000. “Seize the State, seize the Day: State
Capture, Corruption, and Influence in Transition.” Unpublished paper, World Bank,
available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=240555.
Hillman, A. J. 2005. “Politicians on the board of directors: Do connections affect the
bottom line?” Journal of Management 31 (3) (Jun): 464-81.
——–, A. Zardkoohi, and L. Bierman. 1999. “Corporate political strategies and firm
performance: indications of firm-specfic benefits from personal service in the U. S.
government.” Strategic Management Journal 20 (1) (Jan): 67-81.
Hotchkiss, E. S., and J. Gergana. 2007. “Determinants of corporate bond trading: A
comprehensive analysis.” Available at SSRN: http://ssrn.com/abstract=1001459.
Hovakimian, A., G. Hovakimian, and H. Tehranian. 2004. “Determinants of target capital
structure: The case of dual debt and equity issues.” Journal of Financial Economics
71 (3): 514-40.
Hu, F., and S. Leung. 2009. “Appointment of political top executives and subsequent
performance and corporate governance: Evidence from China’s listed SOEs.”
Working Paper, available at SSRN: http://ssrn.com/abstract=1361617.
Huang, H., and N. Zhu. 2009. “China’s emerging financial markets.” The Milken
Institute Series on Financial Innovation and Economic Growth 8 (4): 523-46.
Hung, M., T. J. Wong, and T. Zhang. “Political Relations and Overseas Stock Exchange
Listing: Evidence from Chinese State-owned Enterprises.” Working paper, University
of Southern California, The Chinese University of Hong Kong, City University of
Hong Kong, 2008. Available at SSRN: http://ssrn.com/abstract=1224162.

86
Ivkovic, Z., and S. Weisbenner. 2005. “Local does as local is: Information content of the
geography of individual investors' common stock investments.” The Journal of
Finance 60, (1) (Feb): 267-306.
Jayachandran, S. 2006. “The Jeffords effect.” Journal of Law and Economics 49 (2): 397-
425.
Johnson, S., and T. Mitton. 2003. “Cronyism and capital controls: evidence from
Malaysia.” Journal of Financial Economics 67 (2) (Feb): 351-82.
Johnson, S., J. McMillan, and C. Woodruff. 2002. “Property rights and finance.” The
American Economic Review 92 (5): 1335-56.
Ji, R., and N. Wu. 2010. “The growth of China’s inter-bank corporate bond market.”
China Securities Working Paper, available at
www.afdc.org.cn/afdc/down_research.asp?id=37.
Jiang, J. 2008. “Beating earnings benchmarks and the cost of debt.” Accounting Review
83 (2): 377-416.
Khwaja, A. I., and A. Mian. 2005. “Do lenders favor politically connected firms? Rent
provision in an emerging financial market.” The Quarterly Journal of Economics 120
(4): 1371-411.
Knight, B. 2006. “Are policy platforms capitalized into equity prices? Evidence from the
Bush/Gore 2000 presidential election.” Journal of Public Economics 90 (4-5) (May):
751-73.
La Porta, R., F. Lopez-de-Silanes, and G. Zamarripa. 2003. “Related lending.” The
Quarterly Journal of Economics 118 (1): 231-68.
La Porta, R., F. Lopez-de-Silanes, and A. Shleifer. 2002. “Government ownership of
banks.” The Journal of Finance 57, (1) (Feb): 265-301.
——–, and F. Oberholzer-Gee. 2006. “Political relationships, global financing, and
corporate transparency: Evidence from Indonesia.” Journal of Financial Economics
81 (2) (Aug): 411-39.
Li, H., L. Meng, Q. Wang, and L. Zhou. 2008. “Political connections, financing and firm
performance: Evidence from Chinese private firms.” Journal of Development
Economics 87 (2) (Oct): 283-99.

87
Malesky, E., and M. Taussig. 2009. “Out of the gray: The impact of provincial
institutions on business formalization in Vietnam.” Journal of East Asian Studies 9
(2): 249-90.
Malloy, C. J. 2005. “The geography of equity analysis.” The Journal of Finance 60, (2)
(Apr): 719-55.
Mauro, P. 1995. “Corruption and growth.” The Quarterly Journal of Economics 110 (3)
(Aug): 681-712.
McMillan, J., and C. Woodruff. 2002. “The central role of entrepreneurs in transition
economies.” The Journal of Economic Perspectives 16 (3): 153-70.
Morck, R., B. Yeung, and W. Yu. 2000. “The information content of stock markets: why
do emerging markets have synchronous stock price movements?” Journal of
Financial Economics 58 (1-2): 215-260.
Nee, V. 1989. “A theory of market transition: From redistribution to markets in state
socialism.” American Sociological Review 54: 663-81.
Peng, M. W., and Y. Luo. 2000. “Managerial ties and firm performance in a transition
economy: The nature of a micro macro link.” Academy of Management Journal 43:
486-501.
Peng, Y. 2004. “Kingship networks and entrepreneurs in China’s transitional economy.”
American Journal of Sociology 109: 1045-74.
Peristiani, S. 2006. “Corporate bond underwritten by commercial and investment banks:
Analyzing the default risk.” Available at SSRN: http://ssrn.com/abstract=879806.
Prendergast, C., and R. H. Topel. 1996. “Favoritism in organizations.” The Journal of
Political Economy 104, (5) (Oct): 958-78.
Qian, Y. 1996. “Enterprise reform in China: Agency problems and political control.”
Economy Transition 4 (2), 427-448.
Ramalho, R. 2007. “The persistence of corruption: Evidence from the 1992 presidential
impeachment in Brazil.” Unpublished paper, World Bank, available at
http://www.enterprisesurveys.org/documents/collorpaper_jan_16.pdf.
Rajan, R., and L. Zingales. 1995. “What do we know about capital structure? Some
evidence from international data.” Journal of Finance 50: 1421–60.

88
Reason, T. 2002. Not trigger happy. CFO Magazine. February 21, Available from
www.cfo.com.
Roberts, B. E. 1990. “A dead senator tells no lies: Seniority and the distribution of federal
benefits.” American Journal of Political Science 34, (1) (Feb): 31-58.
Rose-Ackerman, S. 1975. “The economics of corruption.” Journal of Public Economics 4
(2) (Feb):187-203.
Sapienza, P. 2004. “The effects of government ownership on bank lending.” Journal of
Financial Economics 72 (2) (May): 357-84.
Schultz, P. 2003. “Who makes markets.” Journal of Financial Markets 6, (1) (1): 49-72.
Shleifer, A., and R. W. Vishny. 1998. The Grabbing Hand: Government Pathologies and
Their Curses. Cambridge: Harvard University Press.
Shi, C. 2003. “On the trade-off between the future benefits and riskiness of R&D: A
bondholders’ perspective.” Journal of Accounting and Economics 35 (2): 227-54.
Siegel, J. 2007. “Contingent political capital and international alliances: Evidence from
South Korea.” Administrative Science Quarterly 52: 621-66.
Snyder, J. M. 1990. “Campaign contributions as investments: The U. S. House of
Representatives, 1980-1986.” The Journal of Political Economy 98 (6) (Dec): 1195-
227.
Stigler, G. J. 1971. “The theory of economic regulation.” The Bell Journal of Economics
and Management Science 2 (1) (Spring): 3-21.
Stiglitz, J. E. “On the Economic Role of the State.” In A. Heertje, ed., The Economic Role
of the State. Oxford: Blackwell, 1989.
Talmud, I., and G. S. Mesch. 1997. “Market embeddedness and corporate instability: The
ecology of inter-industrial networks.” Social Science Research 26: 419-41.
Uysal, V. B., S. Kedia, and V. Panchapagesan. 2008. “Geography and acquirer returns.”
Journal of Financial Intermediation 17, (2) (Apr): 256-75.
Xu, H., and J. Zhou. 2008. “The value of political connections: Chinese evidence.”
Working Paper, available at ssrn: http://ssrn.com/abstract=1267472.

89
Xu, L. C., T. Zhu., and Y. Lin. 2005. “Political control, agency problems and ownership
reform: Evidence from China.” Economics of Transition 13 (1), 1-24.
Zeng, L. 2009. “Chinese bond markets – An introduction.” Standard & Poor’s Research
Paper, available at www.standardpoors.com.
Zhou, X. 2006. “China’s corporate bond market development: Lessons learned.” Bank of
International Settlement Research Papers No. 26. Retrieved at:
http://www.ksri.org/bbs/files/research02/bispap26.pdf#page=17.

Abstract (if available)

Abstract This study addresses how political connections influence a firm’s access to bond capital using a database of 1672 new bond issuances in China from 2001 to 2009. In this context, firm political connections are measured using a score based on the highest bureaucratic position held by a firm executive, with the highest level of political connections attributed to those who have held top positions in the central government, an intermediate level attributed to those in provincial and municipal governments, and the lowest level attributed to those who have not held any government position. Results suggest that firm political connections are positively associated with debt offering amounts and issuer credit ratings, but only in the subsample of firms that have poor information environments, such as non-publicly listed firms and non-Beijing headquartered firms, thus lending support to the argument that political connections contribute to firm reputation. The role of political connections in providing preferential access to debt is relevant to both state-owned enterprises and privately held firms. In addition, issuing firm political connections and bank underwriter political connections serve as alternative mechanisms when explaining the variation of offering amounts and issuer credit rating.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

CEO reputation: who benefits -- the firm and the CEO?

PDF

Essays on improving human interactions with humans, algorithms, and technologies for better healthcare outcomes

PDF

Toward counteralgorithms: the contestation of interpretability in machine learning

Asset Metadata

Creator Du, Fei (author)

Core Title Political connections and access to bond capital: Reputation or collusion?

School Marshall School of Business

Degree Doctor of Philosophy

Degree Program Business Administration

Degree Conferral Date 2011-05

Publication Date 04/29/2011

Defense Date 03/30/2011

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag bond issuance,bond rating,China,collusion,OAI-PMH Harvest,offer size,Political connections,Reputation

Place Name China (countries)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Bonner, Sarah (committee chair), Young, S. Mark (committee chair), McArdle, Jack (committee member), Sandino, Tatiana (committee member)

Creator Email dufeiusc@gmail.com,fdu@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3813

Unique identifier UC1474314

Identifier etd-Du-4565 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-462086 (legacy record id),usctheses-m3813 (legacy record id)

Legacy Identifier etd-Du-4565.pdf

Dmrecord 462086

Document Type Dissertation

Rights Du, Fei

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu