Political Connection and Ownership Concentration:

1
Political Connection and Ownership
Concentration Evidence from Thailand
2
Introduction
Why Politics?

• Political turmoil in Thailand during 2005

• Early arguments about the problem

• Conflict of interest

• Freedom of media and press

• Nominee and stock price manipulation

Lead to

• The study about stocks that have strong linked
  with political figures

• Also how concentration of ownership determines
  the firm performance

3
Introduction
Historical event for TRT

• Won the election with the majority seats during
  the beginning of 2001

• Regard as the most stable political situation
  after the financial crisis

• Positive sentiment drove the index to over 700
  in just one year

4
Introduction

• The topic on the political connection has been
  around many years.
• Empirical studies are conducted around the world.
  Fisman (2001) focuses in Indonesia, Johnson and
  Mitton (2003) focus in Malaysia. Faccio (2006)
  studies 47 countries including Thailand.
• Imai (2006) concentrates in Thailand,
  investigating firm performance between connected
  and non-connected firms. Wiwattanakantang et al.
  (2006) conduct event study for connected firms.

5
Introduction

• Another interesting topic concerning level of
  firm performance is how firms in developing
  countries structure their shareholding.
• Some evidence from empirical studies suggest
  strong link between ownership concentration and
  countries in East Asia, Claessens et al. (2000).
• Mitton (2002) also focuses in East Asian
  countries, where higher performance relates to
  the presence of concentrated shareholder.

6
Introduction

• Thailand provides an interesting setting for the
  study.
• The period of Thai Rak Thai party attracts many
  business leaders to take the opportunity to
  become involved in politics.
• This paper is similar and different from previous
  literatures in a way describes in the next slide.

7
Introduction

• Similarities
• 1. Measuring firm performance for connected
  firms by using return on assets, return on equity
  and market share.
• 2. Comparing the leverage level between
  connected and non-connected.
• Differences
• 1. Including market measure for firm
  performance, Tobins Q.
• 2. Introducing risk measures for connected
  firms.
• 3. Expanding the scope to include ownership
  concentration effect.

8
Research Question Objectives
Research Question
Whether the political connection and
concentration of ownership have any impacts on
firm performance, leverage, market power and firm
risk
Objectives
To study the impact of political connection and
concentration of ownership on
1. Performance
2. Leverage
3. Market power
4. Risk
9
Scope of Study
Definition of connections
Key government officers
Prime Minister and Cabinet Ministers
Examples, Thaksin Shinawatra, Pracha Maleenont,
Adisai Potharamik
2 Criteria for firm to be connected

• At least 10 shareholder votes (regard as a major
  shareholder)

OR

• At least one of the Board of Directors serves as
  ministers

Examples, CEO, President, Director, Vice-President
10
Some Examples

• Tracing for connection only family name of
  husband and wife of Cabinet Ministers are
  included, further change down the family line
  will be excluded. For example, Shinawatra family

Yaowapa
Somchai
Phayap
Phoruthai
Yinglak
Anusorn
Thaksin
Podjaman
Damaphong
Wongsawat
Jantharaphan
Amonchat
Further intermarried of these families will not
be included.
11
Some Examples
Connection through same family name
12
Some Examples
Connection through in-laws family name
13
Scope of Study

• Firms listed in Stock Exchange of Thailand

• Period of study between year 2001 to 2004

• Firm performance measures by return on assets,
  return on equity, Tobins Q and market power
  (market share)

• Firm leverage

• Firm risk measures by beta, standard deviation
  and firms specific risk (unsystematic risk)

• Ownership concentration uses the number of
  accumulation of at least 5 shareholding for
  every shareholder in the firm

14
Limitations

• Any indirect measure of connection will be
  excluded. For example, connection through
  friendship.
• Foreign nominee firms
• By including only the connection between Cabinet
  Ministers, the result can potentially
  underestimate the value of connection with other
  government officials or Members of Parliament.
• Firm must have a complete information both
  accounting and stock price during the study
  period.

15
Literature Review
Evolution of the topic
Pioneering paper by Anna Krueger
Dated back to 1974
The Political Economy of the Rent Seeking
Society
More recent papers on politically connected firms

• Roberts, 1990
• Agrawal and Knoeber, 2001
• Fisman, 2001
• Johnson and Mitton, 2003
• Dinc, 2005
• Donald, Zhang and Derashid, 2006
• Faccio, 2006

• Cheung, Jing, Raghavendra Rau and Stouraitis,
  2005
• Imai 2006
• Bunkanwanicha and Wiwattanakantang 2006

16
Literature Review
The global view on ownership structure and
concentration

• Claessens, Djankov and Lang, 2000
• Morck, et al., 2000
• Mitton, 2002
• Gursoy and Aydogan, 1999

The ownership structure and corporate control in
Thailand

• Wiwattanakatang, 2001
• Wiwattanakatang, Kali and Charumilind, 2002
• Dhnadirek and Tang, 2003

17
Literature Review
Political connection, benefit or expropriation to
firm

• Y. Wiwattanakantang, P. Bunkanwanicha (2006)
• The difference in firms market share before and
  after when tycoon took position in the government
  increases substantially.
• M. Imai (2006)
• Higher in return on assets (ROA) for connected
  firms, the result is greater if firm is connected
  with Cabinet Ministers.
• Cabinet Ministers exert stronger value of
  connection than among political officials.

18
Literature Review
Ownership concentration and firm performance and
risk
Mitton (2002) found higher ownership concentrated
can lower the agency cost

• Positive relationship between shareholding and
  performance

• Gursoy and Aydogan (1999) illustrate how
  ownership concentration affects level of firm
  risk differently.
• Higher concentrated ownership results in higher
  total risk or standard deviation.
• Moreover the presence of different type of
  shareholders show different level of risk.
  Government owned firm exhibits higher level of
  market risk and standard deviation due to low
  level of corporate governance and management
  skills.

19
Methodology
Data Sources

• For public companies sources

Example, www.set.or.th, www.setsmart.com,
University subscriptions to financial sources,
Datastream, various newspaper issues

• For private companies sources

Example, BOL website, various newspaper issues

• For background and political information sources

Example, Thai Business Groups published by
Brooker group, www.mof.go.th, www.thaigov.go.th,
www.parliament.go.th or wikipedia website
20
Methodology
Sample description

• The sample in this study uses firms listed in
  Stock Exchange of Thailand.
• The study period is from year 2001 to 2004
• Each firm needs to have a complete information
  in order to arrange data into a balanced panel
  structure.

21
Methodology
Hypothesis
Hypothesis 1 Connected firms have higher ROA
than non-connected firms
Hypothesis 2 Connected firms have higher ROE
than non-connected firms
Hypothesis 3 Connected firms have higher Tobins
Q than non-connected firms
Hypothesis 4 Connected firms have higher
leverage than non-connected firm
22
Methodology
Hypothesis
Hypothesis 5 Connected firms have higher market
share than non-connected firm
Hypothesis 6-8 Connected firms have lower risk
than non-connected firm
23
Methodology
Performance measures
Return on Asset (ROA)
Earning before interest and tax (EBIT)/Total
assets
Return on Equity (ROE)
Net profit/Total equity
Tobins Q (Q)
(Market value of equity book value of
liabilities)/book value of assets
Leverage measures
Leverage (LEV)
Long-term debts/(book value of liabilities
market value of equity)
Market power measures
Market Power (MKTP)
Firm sales/Total market sales
24
Methodology
Risk measures
Beta (BETA)
The ratio of covariance between weekly stock
return and market premium divide by the variance
of market premium
Standard deviation (STDEV)
The standard deviation of weekly stock return
Unsystematic risk (UNSYS)
The residual variance
25
Methodology
Explanatory Variables
Political connection (PCON)
Connection through a major shareholder
(accumulation of at least 10 percent
shareholding) or through firms Board of
Directors
Concentration of ownership (CONC)
Accumulation of each shareholder who owns more
than 5 percent in the firm
Interaction term (POLCON)
Multiply dummy variable PCON with CONC
26
Methodology
Control Variables
Log of firms total assets is a proxy for firm
size. Firm size is widely used to control for the
firms market power and efficiency. This control
variable should reflect a positive relationship
with firm performance.
Size (SIZE)
Number of years since incorporation. Well
established firms may have superior performance
as a result of experience and reputation. This
control variable should reflect a positive
relationship with firm performance
Age (AGE)
Sales to asset (STA)
The ratio of firms sales to total assets. This
variable is a proxy for firms efficiency. It
should be positively related with firm
performance.
27
Methodology
Control Variables
As a dummy variable for firm with government
owned more than 10 percent. It should be
negatively related to firm profitability
Government (GOV)
State owned enterprise (SOE)
This is a dummy variable for firm with state
enterprise owned more than 10 percent. This
should be negatively related with firm
performance.
Foreign (FOREIGN)
This is a dummy variable for firm with foreign
owned more than 10 percent. This should be
positively related with performance.
28
Methodology
Control Variables
A dummy variable to remove variation from
industry effects
Industry (IND)
Regression method

• The regression in this study uses panel least
  square procedure with fixed period effect.

29
Classification of connection by type
This table presents the sample firms across
industries and classified firms with connection
between shareholders and management
30
Explanation

• The sample consists of 278 firms.
• The politically connected firms are present in
  12 industries.
• Most concentrated in communication sector.
• The sample covers approximately 80 percent of
  total market capitalization.
• The connected firms represent over 30 percent of
  total sample capitalization.
• 14 firms connected through management, 2 firms
  connected through shareholders and 7 firms are
  connected through both management and
  shareholders.

31
Descriptive Statistics
32
Explanation

• On average connected firms show higher level of
  total assets, total liabilities and market cap.
• Non-connected firms are more concentrated in
  ownership with 51.68 percent when compare with
  sample average of 51.44 percent.
• Connected firms operating years are longer than
  sample average and bigger in size.
• Connected firms show lowest ROA. But ROE,
  Tobins Q and market power are highest. On
  average connected firms carry higher leverage
  than non-connected firms.
• On average, beta is higher for connected firms
  when compare with sample average. While SD and
  specific risk show no difference between the
  groups.

33
Regression results with statistically significant
level
34
Regression results with statistically significant
level
35
Regression results with statistically significant
level
36
Explanation

• ROE
• The coefficient on political connection is
  statistically significant at 95 percent confident
  level with positive relationship. This is in line
  with the hypothesis where connection provides a
  rewarding benefit to the shareholders and clearly
  this is an incentive to become connected.
• The ownership concentration is not statistically
  significant but has a positive relationship with
  ROE.
• The interaction term is positively related but
  not statistically significant, meaning firm with
  concentrated ownership and politically connected
  has shown no prove on improving ROE.

37
Explanation

• Tobins Q
• The result shows positive relationship between
  connection and Tobins Q at 95 percent level.
• The ownership concentration is also positive and
  statistically significant at 99 percent level.
  This is in line with Wiwattanakantang (2001)
  where she illustrates firm with large
  shareholding can become more competitive and
  resulting in low agency problem.
• The interaction term is positively related and
  statistically significant. This result further
  proves firm with high concentration and
  politically connected increases firm performance.
• Foreign controlled firm shows a positive
  relationship with Tobins Q at 90 percent level.
  This indicates how foreign partnership equips
  firm with advanced technology and highly skilled
  management.

38
Explanation

• Market Power
• The coefficient on political connection shows a
  positive relationship with statistically
  significant level. This is in line with the
  hypothesis where connected firms enjoy higher
  market share through government policies or state
  licensing.
• Faccio (2006) shows a similar result where
  connected firms maintain higher market share by
  receiving privileges from the government.
• The positive coefficient on ownership
  concentration with significant level determines
  how this variable becomes one of the success
  factor in order to increase level of market
  share. Level of agency problem seems to be
  diminished as ownership becomes more concentrated.

39
Explanation

• Leverage
• The result shows the coefficient is negatively
  related for ownership concentration at 95 percent
  level, indicating lower level of leverage as firm
  becomes more concentrated.
• Beta
• The result shows political connection to be
  statistically insignificant.
• The ownership concentration indicates a negative
  relationship with market risk at a statistically
  significant level. This is in line with Gursoy et
  al. (1999).

40
Regression results with no statistically
significant level
41
Regression results with no statistically
significant level
42
Explanation

• ROA
• From the table, the result shows relationship to
  be statistically insignificant for political
  connection.
• Ownership concentration is also showing sign of
  positive relationship but insignificant.
• The connection itself is providing value, but
  due to the amount of firm resources are being
  devoted to get connection which offsetting the
  benefit firm receives.
• The interaction term, POLCON, shows positive
  relationship but statistically insignificant.

43
Explanation

• Standard Deviation
• Both political connection and ownership
  concentration variables show insignificant level
  with standard deviation (SD). This is
  contradicting to the hypothesis, where I expect
  connected firms to show lower SD compare with
  non-connected firms. The result proves level of
  total risk has no impact on connection and
  ownership concentration.
• Unsystematic risk
• Political connection has shown an insignificant
  level with unsystematic risk or firms specific
  risk. This is different from what I expect
  connected firm to show lower unsystematic risk
  than non-connected firm. The result proves
  political connection has no impact on level of
  unsystematic risk.

44
Conclusion

• The consequences of being politically connected
  through either management or shareholders has
  been explored and these eventually result in
  higher market shares, better performance in term
  of return on equity and Tobins Q. Firms with
  connection and high ownership concentration
  result in higher market share and Tobins Q.
• Lower agency problem can be seen in highly
  concentrated firm.
• Hence, political connection is a favorable
  factor for firm to increase its competitive level
  and stay dominant in the industry.
• However, political connection proves to have no
  impact on level of firm risk. Only ownership
  concentration that associates with beta.

45
Thank You