Rent Seeking and Corporate Finance: Evidence from Corruption Cases Joseph P.H. Fan* Oliver M. Rui* Mengxin Zhao** *Chinese University of Hong Kong **Bentley.

Rent Seeking and Corporate Finance: Evidence from

Corruption Cases

Joseph P.H. Fan*

Oliver M. Rui*

Mengxin Zhao**

*Chinese University of Hong Kong

**Bentley College

2

Corporate Financing Patterns in Emerging Markets

Companies in emerging markets rely on debt much more than equity to finance their investment

Moreover, they rely on short-term debt, even when they engage in long-term investment

Banks, not capital markets, are the primary sources of funds for firms in developing countries

3

Cross country pattern of corporate leverage (Fan, Titman, Twite, 2006)

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

Kor

ea, R

ep.

Tha

iland

Indo

nesi

a

Indi

a

Bra

zil

Phili

ppin

es

Chi

na

Aus

tria

Paki

stan

Ital

y

Port

ugal

Japa

n

Finl

and

Den

mar

k

Switz

erla

nd

Bel

gium

Fran

ce

Hon

g K

ong,

Chi

na

Can

ada

Mex

ico

Irel

and

Peru

Tai

wan

New

Zea

land

Spai

n

Chi

le

Sing

apor

e

Mal

aysi

a

Net

herl

ands

Ger

man

y

Swed

en

Nor

way

Uni

ted

Stat

es

Aus

tral

ia

Uni

ted

Kin

gdom Isra

el

Tur

key

Sout

h A

fric

a

Gre

ece

4

Cross country pattern of corporate debt maturity (Fan, Titman, Twite, 2006)

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

Can

ada

Swed

en

New

Zea

land

Uni

ted

Stat

es

Aus

tral

ia

Nor

way

Finl

and

Irel

and

Switz

erla

nd

Mex

ico

Den

mar

k

Indi

a

Net

herl

ands

Sout

h A

fric

a

Chi

le

Port

ugal

Bra

zil

Fran

ce

Bel

gium

Uni

ted

Kin

gdom

Ger

man

y

Phili

ppin

es

Isra

el

Japa

n

Spai

n

Aus

tria

Kor

ea, R

ep.

Hon

g K

ong,

Chi

na

Ital

y

Paki

stan

Peru

Tai

wan

Tur

key

Indo

nesi

a

Mal

aysi

a

Tha

iland

Sing

apor

e

Chi

na

Gre

ece

6

Research Questions in This Study

To investigate the impact of political rent seeking on corporate financing behaviors in China

Research questions How do rent seeking and corruption affect capital

structure, debt maturity, and long-term debt financing of the Chinese firms?

In retrospect, whether political connections provide the firms financing advantages?

Are the financing advantages of the politically connected firms reflected in their stock prices?

Does corruption affect capital allocation efficiency?

7

Why China?Pervasive corruption

One of the world’s highly corrupted countries According to China’s official record, during 1997-2002,

there are totally 861917 corruption cases under investigation 842760 corruption cases concluded 846150 people punished by communist laws, of which

137711 expelled from the communist party Among the punished communist party members,

28996 county （县） level2422 intermediate（厅 ,局） level98 provincial （省 , 部） level or above

8

Corruption in Asian Economies(Source: Transparency International: mean Corruption

Perception Index 1992-2000)

0

1

2

3

4

5

6

7

8

9

Economy

ChinaTaiwanHong KongIndonesiaJapanKorea (South)MalaysiaPhilippinesSingaporeThailandUnited StateUnited Kingdom

9

Why China?Dominant role of bank debt in corporate

finance Equity markets small Banks as the primary external sources of funds

Big-four commercial banks dominate (accounting 63 percent of loans outstanding and 62 percent of deposits in 2001)

10 national and 90 regional commercial banks, 3 policy banks, 3000 urban and 42000 rural credit cooperatives

Limited foreign bank activities The banking system is largely state run, suffering from high

NPL problem and scandals Given the state control of banks and the highly corrupted

system in China, connections with government bureaucrats are likely important in influencing bank loan decisions

17

Empirical Design

Identifying 23 high (provincial) level government officer corruption cases during 1995-2003 Tracking publicly listed companies in China that

are bribers or are connected with the corrupted bureaucrats

Track changes in financial leverage and debt maturities of the event (bribing or connected) firms from 3 years before to 3 years after the corruption event

18

Empirical Design Compared with the prior studies, our single-

economy time-serial empirical design provides advantages Focusing on a specific institutional factor – rent

seeking Providing more directly link between corporate

financing decisions with rent seeking and corruption Mitigating endogeneity issues

The corruption events are likely shocks to connected firms. The connected firms are non-bribers, hence their financing policy changes upon the corruption events are likely caused by lost connections

19

A Corrupted Bureaucrat and His Allies

20

The Corruption ListProvince Name Position Event

Day Sentence Day Sentence Number

of firms in the

province

The Bribing Firms

The Connected Firms

The Non-event Firms

Anhui Wang Huizhong

Vice-Province Governor 20010407 20031229 Death penalty 27 1 2 24

Bank Liu Jinbao Vice-Chairman & CEO of Bank Of China (HK)

20030525 20050812 Suspended death penalty 5 5 0 0

Bank Wang Xuebin CEO of China Construction Bank 20020111 20031210 12 years of imprisonment 5 5 0 0

Beijing Chen Xitong CPC Secretary 19950426 19980731 16 years of imprisonment 11 5 3 3

Central Xu Penghang Vice chairman of national defense technology commission and national economics and trade commission

20001011 20001011 Dismissal from the service 1 1 0 0

Fujian Shi Zhaobin Vice-CPC Secretary 19990818 20010927 Dismissal from the service and CPC 34 5 3 26

Guangxi Xu Binsong Vice-Chairman of Municipality 19980523 19990827 Life imprisonment 6 0 1 5

Guangxi Chen Kejie Chairman of Municipality 20000111 20000731 Death penalty 10 2 1 7

Guangxi Liu Zhibin Vice-Chairman of Municipality 20000319 20020624 15 years of imprisonment 10 2 2 6

Guangxi Wang Qinglu Vice-Chairman of PPCC 20010222 20010222 Dismissal from the service and CPC 13 0 1 12

Guizhou Liu Changgui Vice-Province Governor 20030417 20040430 11 years of imprisonment 12 1 3 8

Guizhou Liu Fangren CPC Secretary & PC Chairman 20030422 20040629 Life imprisonment 12 1 1 10

Hainan Xin Yejiang Vice-PC Chairman 19961227 19980526 5 years of imprisonment 10 2 0 8

Hebei Jiang Dianwu Vice-PC Chairman 19971101 19981207 10 years of imprisonment 13 0 0 13

Hebei Chen Weigao CPC Secretary & PC Chairman 20000301 20030809 Dismissal from CPC 25 1 3 21

Hebei Cong Fukui Vice-Province Governor 20000627 20010518 Dismissal from the service and CPC 24 0 2 22

Hubei Li Daqiang Vice-Province Governor 20000925 20000925 Dismissal from the service and CPC 2 2 0 0

Hubei Meng QingPing

Vice-Province Governor 19980410 19991201 10 years of imprisonment 33 1 4 28

Jiangxi Hu Changqing Vice-Province Governor 19990808 20000215 Death penalty 12 0 2 10

Liaoning Mu TuoXing Vice-Province Governor 20010321 20011113 Death penalty 51 5 6 40

Xingjiang Aman.Haji Vice-Province Governor 20031015 N/A N/A (Still under investigation) 25 1 1 23

Yunnan Li Jiating Vice CPC Secretary & Province Governor 20010620 20030509 Death penalty 17 1 4 12

Zhejiang Xu Yunhong Vice-Province Governor 19990922 20001017 10 years of imprisonment 35 2 3 30

Total 393 43 42 308

24

Figure 1.1 Mean Total Debt/Assets (The event firms and the non-event firms)

26

Figure 2.1 Mean Long Term Debt/Total Debt

(The event firms and the non-event firms)

28

Figure 3.1 Mean Long Term Debt/Assets(The event firms and the non-event

firms)

34

Table 4 Fixed-Effect Regression: Event firms and non-event firms

Dependent Variables

Total Debt/Assets

(Total Debt+AP)/Assets

Long-term Debt/Total

Debt

Long-term Debt/(Total Debt+AP)

Long term Debt/Assets

Short-term Debt/Assets

AP/Assets (Short-term Debt+AP)/Assets

Intercept -0.442*** -0.544*** -0.785*** -0.734*** -0.432*** -0.026 -0.108* -0.082 (-2.46) (-2.87) (-2.70) (-2.99) (-4.48) (-0.18) (-1.84) (-0.49) Corrupt 0.015 0.016 0.042 0.019 0.003 0.013 0.001 0.014 (0.97) (0.97) (1.36) (0.70) (0.33) (1.14) (0.20) (1.01) Post 0.037 -0.283 -1.073*** -0.664*** -0.233* 0.376* -0.372*** -0.170 (0.13) (-0.88) (-3.01) (-2.26) (-1.85) (1.66) (-3.54) (-0.55) Corrupt*Post -0.082*** -0.093*** -0.043* -0.029* -0.034*** -0.053*** -0.005 -0.041 (-3.13) (-3.10) (-1.84) (-1.66) (-3.07) (-2.50) (-0.49) (-1.33) Lass 0.033*** 0.042*** 0.040*** 0.037*** 0.022*** 0.009 0.009*** 0.018*** (3.67) (4.48) (2.75) (3.02) (4.54) (1.22) (3.13) (2.23) Lass*Post 0.005 0.022 0.051*** 0.030*** 0.011* -0.011 0.020*** 0.019 (0.32) (1.40) (2.91) (2.11) (1.82) (-1.00) (3.83) (1.27) Tangible 0.029 0.028 0.529*** 0.442*** 0.121*** -0.089*** -0.057*** -0.147*** (0.75) (0.70) (6.04) (5.30) (4.26) (-3.16) (-3.49) (-4.22) Tangible*Post -0.019 -0.041 -0.050 -0.010 0.035 -0.116*** -0.027 -0.174*** (-0.30) (-0.59) (-0.52) (-0.12) (1.08) (-2.58) (-1.06) (-2.71) Sales Growth 0.003 0.001 0.004 0.008* 0.002 0.001 -0.002* -0.002 (0.36) (0.05) (0.85) (1.68) (0.77) (0.06) (-1.95) (-0.37) Sales Growth *Post

0.002 (0.22)

0.007 (0.90)

0.004 (0.88)

-0.002 (-0.34)

0.002 (0.75)

-0.001 (-0.25)

0.005*** (3.81)

0.003 (0.60)

Profit -0.791*** -0.870*** 0.166 -0.019 -0.117*** -0.658*** -0.083*** -0.741*** (-5.71) (-5.65) (0.93) (-0.12) (-2.21) (-4.97) (-2.82) (-5.39) Profit*Post 0.776*** 0.859*** -0.166 -0.015 0.112*** 0.650*** 0.086*** 0.737*** (5.05) (5.56) (-0.93) (-0.10) (2.11) (4.89) (2.91) (5.34) N 1933 1933 1917 1917 1933 1933 1933 1933 Adj R-squared

21.05% 24.13% 22.44% 22.07% 20.27% 19.41% 17.47% 17.38%

35

Table 4 Fixed-Effect Regression: Event firms and matching firms

Dependent Variables

Total Debt/Assets

(Total Debt+AP)/Assets

Long-term Debt/Total

Debt

Long-term Debt/(Total Debt+AP)

Long term Debt/Assets

Short-term Debt/Assets

AP/Assets (Short-term Debt+AP)/Assets

Intercept -0.137 -0.262 -1.178*** -0.658*** -0.392*** 0.319 -0.117 0.202 (-0.63) (-1.07) (-3.02) (-2.35) (-4.42) (1.58) (-1.38) (0.89) Corrupt 0.028* 0.026 0.062* 0.050* 0.027*** 0.001 -0.002 -0.001 (1.75) (1.34) (1.83) (1.77) (3.61) (0.07) (-0.33) (-0.08) Post -0.270 -0.713 0.243 0.101 0.082 -0.280 -0.628*** -0.950** (-0.65) (-1.56) (0.40) (0.21) (0.47) (-0.85) (-2.62) (-2.10) Corrupt*Post -0.076*** -0.059 -0.119*** -0.115*** -0.053*** -0.002 -0.006 -0.020 (-2.17) (-1.64) (-2.50) (-2.86) (-4.05) (-0.06) (-0.28) (-0.36) Lass 0.019* 0.030*** 0.060*** 0.032*** 0.019*** -0.003 0.010*** 0.007 (1.80) (2.46) (3.11) (2.40) (4.63) (-0.30) (2.41) (0.62) Lass*Post 0.017 0.038* -0.014 -0.005 -0.004 0.018 0.032*** 0.055*** (0.84) (1.74) (-0.47) (-0.22) (-0.54) (1.10) (2.61) (2.45) Tangible -0.031 -0.095 0.452*** 0.430*** 0.083*** -0.129*** -0.065*** -0.194*** (-0.62) (-1.57) (3.96) (4.13) (2.85) (-3.22) (-2.69) (-3.56) Tangible*Post 0.070 0.067 0.102 0.031 0.094*** -0.104 -0.058 -0.290** (0.68) (0.56) (0.67) (0.23) (2.12) (-1.29) (-1.06) (-2.06) Sales Growth 0.051* 0.053* -0.031 -0.018 0.012 0.037 0.004 0.041 (1.83) (1.74) (-0.54) (-1.40) (0.74) (1.63) (0.34) (1.53) Sales Growth *Post

0.049 (0.62)

0.054 (0.75)

0.131* (1.68)

0.108* (1.68)

0.029 (1.18)

-0.049 (-0.97)

0.082 (1.36)

0.089 (0.83)

Profit -1.596*** -1.748*** 0.805*** 0.452** -0.081 -1.514*** -0.157*** -1.671*** (-9.30) (-8.00) (2.94) (2.01) (-1.43) (-8.99) (-2.20) (-7.75) Profit*Post 1.279*** 1.616*** -0.959*** -0.717** -0.036 1.287*** 0.474*** 2.029*** (4.55) (4.84) (-2.39) (-2.03) (-0.34) (4.94) (3.90) (4.98) N 792 792 792 792 792 792 792 792 Adj R-squared

21.79% 23.09% 15.18% 14.33% 15.73% 25.38% 12.74% 19.43%

41

Table 7 Regressions of Long-term Changes in Performance on Changes in Financial Policy around

the Corruption Scandals (Panel A Full Sample)

Change in ROS Change in Market-to-Book ratio Intercept 0.036 0.032 0.029 -0.123*** -0.109 0.163 0.038 0.393*** (0.68) (0.68) (0.55) (-3.08) (-0.79) (1.20) (0.26) (2.17) Intercept 0.080 1.650*** (1.20) (9.44) Change in total leverage 0.394*** 3.595*** (2.20) (7.08) Change in long-term debt 0.129 1.741*** (1.44) (6.89) Change in short-term debt 0.074 1.108*** (1.20) (3.93) Change in debt maturity -0.124*** -0.113*** -0.108* -0.034 -0.054 -0.062 -0.108 -0.202 (-2.12) (-2.36) (-1.86) (-0.74) (-0.35) (-0.46) (-0.66) (-0.97) Corrupt 0.049 0.505 (0.30) (1.15) Corrupt* Change in total leverage

0.398 (0.86)

1.200 (0.91)

Corrupt* Change in long-term debt

-0.042 (-0.20)

0.417 (0.71)

Corrupt* Change in short-term debt

0.268*** (2.18)

0.608 (1.09)

N 334 334 334 334 334 334 334 334 Adj R-squared 4.54% 6.34% 4.56% 5.51% 29.70% 19.83% 19.47% 12.89%

47

Does rent seeking facilitate capital allocation?

Examining efficiency (performance) around the corruption scandals

The endogeneity hypothesis The event firms should outperform the non-event firms

before the scandals, and should not underperform the non-event firms after the scandals

The social injustice hypothesis Event firms should underperform non-event firms after

the scandals The randomness hypothesis

No difference in performance between event and non-even firms. Non-even firms outperform event firms after the scandals

48

Table 8 Mean and Median Differences in Performance between the Event Firms and the Non-

event Firms before and after the Corruption Scandals (Panel A Differences Between the Event

Firms and the Non-event Firms)

Three years before the scandal Three years after the scandal

Mean Median Mean Median ROA 0.012

(1.06) 0.042 (1.08)

-0.021*** (-2.39)

-0.007* (-1.95)

ROE 0.024 (0.98)

0.293 (1.02)

-0.041*** (-2.26)

-0.018* (-1.74)

Operating Margin 0.042 (1.27)

0.216 (1.15)

-0.091*** (-3.06)

-0.037*** (-2.44)

ROS 0.049 (1.36)

0.168 (1.09)

-0.077*** (-2.66)

-0.014 (-1.55)

Market-to-Book 0.244*** (2.88)

0.230*** (2.43)

0.046 (0.34)

0.140 (0.16)

50

Summary of findings An overall increasing trend of financial leverage, but

decreasing trend of debt maturity Significant lowered leverage, debt maturity, and the use

of long-term debt of the bribing firms and the connected firms, relative to control (unconnected or matching) firms

The results hold even if we focus on the connected firms that are not involved in the corruption cases

The weakened debt financing ability in the sample is not just due to the corruption, but also due to loss of political connections

Change in stock value and change in leverage are significantly positively related around the corruption events

We find little evidence from the sample suggesting that rent seeking facilitates capital allocation in China.

51

Conclusions

Corporate financing policies are importantly affected by rent seeking and corruption activities

Connections with government bureaucrats provide firm financing advantages, in particular access to long-term bank debt

Compared with the prior studies, this paper provides more direct links between rent seeking and corporate finance, and empirical design less subject to endogeneity issues

The overall evidence is consistent with recent cross-country studies’ findings that country-level institutional factors matter to corporate financing decisions