Emerging Capital Markets AG907 M.Sc. Investment & Finance - - PowerPoint PPT Presentation

Emerging Capital Markets AG907

M.Sc. Investment & Finance M.Sc. International Banking & Finance Lectures 5 & 6 – Ownership Structure in Emerging Capital Markets

I g n a c i o R e q u e j o – G l a s g o w , 2 0 1 0 / 2 0 1 1

Overview of Lectures

Two main Parts I. Corporate Ownership Worldwide (based on La Porta et al., JF, 1999) II. Ownership Structures in East Asia (based on Claessens et al., JFE, 2000)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

1

Corporate Ownership Worldwide Lectures 5 & 6 – Part I

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

2

• I. Corporate Ownership Worldwide

I.1. Introduction I.2. Data Set & Sample Coverage I.3. Different Types of Control I.4. Ownership vs. Control I.5. Controlling Shareholders Worldwide I.6. Means of Achieving Control I.7. Separation Control–Ownership I.8. The Special Case of Family Control I.9. Main Conclusions

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

3

I.1. Introduction

Evidence on ownership structures of 20 largest publicly traded firms in 27 economies (also smaller firms to keep size constant across countries) Focus on largest firms in richest economies ⇒ For these firms likelihood of dispersed ownership is greatest Principal contribution: Find identities of ultimate owners of capital & voting rights in firms (when shares in a firm are

• wned by another company ⇒ examine ownership of that

company)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

4

I.1. Introduction

Two main questions of study

• 1. How common are widely held firms vs. firms with
• wners with significant voting rights?
• 2. Among firms with significant owners ⇒ Who are they

(families, state, financial institutions, other firms)? Bottom line: Challenge view that dispersed ownership is common pattern around the world

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

5

I.2. Data Set & Sample Coverage

Companies from 27 countries (richest countries based on 1993 per capita income) For each country, two samples of firms i) Top 20 firms ranked by market cap. at end of 1995 (“large firms”) ii) Smallest 10 firms in each country with market cap. of at least $500 million at end of 1995 (“medium firms”)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

6

I.2. Data Set & Sample Coverage

Final sample ⇒ 540 large firms & 691 firms in total To identify ultimate owners ⇒ Shareholders who control

• ver 10% of votes

The 10% cut-off point used because 1) It provides significant threshold of votes 2) Most countries mandate disclosure of 10% ownership stakes

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

7

I.3. Different Types of Control

Definitions of ownership rely on voting rights (rather than CF rights) ⇒ Focus on who controls corporations? Do corporations have shareholders with substantial voting rights (directly or through chain of holdings)? Firms divided into: Widely held & with ultimate owners (5) 1) Family or individual 2) The State 3) Widely held financial institution (bank, insurance co.) 4) Widely held corporation 5) Miscellaneous (cooperative, voting trust, etc.)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

8

I.3. Different Types of Control

State control as separate category ⇒ State uses firms to pursue political objectives but public pays for losses Widely held corps. & widely held financial institutions ⇒ Interest of bank ownership

• Corp. has UO if shareholder’s direct & indirect voting

rights > 20%; Indirect ownership of X% in firm A 1) Shareholder directly controls firm B, which in turn directly controls X% of votes in firm A 2) Shareholder directly controls firm C, which in turn controls firm B (control chain), which directly controls X% of votes in firm A

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

9

I.4. Ownership vs. Control

Definition of ownership (CF rights) & definition of control (voting rights) Separation of CF & voting rights ⇒ Deviations from one- share–one-vote, pyramidal structures, cross-holdings E.g.: Family owns 11% stock of Firm A, which in turn has 21% stock of Firm B ⇒ Family owns 2% CF rights of Firm B (product of ownership stakes in chain) but family controls 11% of Firm B (weakest link in control chain)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

10

I.4. Ownership vs. Control

Example of indirect ownership & control

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

11

Family Firm A Firm B 11% 21% Family links to Firm B: Ownership (CF rights) = 0.11 * 0.21 = 2% Control (voting rights) = Min (0.11 ; 0.21) = 11%

I.4. Ownership vs. Control

E.g.: Family owns 11% stock of publicly traded Firm A, which in turn has 21% stock of Firm B; & same family

• wns 25% Firm C, which in turn owns 7% Firm B

Looking at control rights ⇒ Family controls 18% of Firm B, (sum of weakest links in chains of voting rights) but family

• wns about 3.5% of CF rights of Firm B (sum of products
• f ownership stakes along two chains)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

12

I.4. Ownership vs. Control

Example of indirect ownership & control

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

13

Family Firm A Firm B 11% 21% Family links to Firm B:

• Own. (CF rights) = (0.11*0.21) + (0.25*0.07) = 3.5%
• Con. (voting rights) = Min(0.11;0.21) + Min(0.25;0.07) = 18%

Firm C 25% 7%

I.5. Controlling Shareholders Worldwide

27 countries divided in two groups

• 1. High

protection ⇒ 12 with better than median shareholder protection

• 2. Low protection ⇒ 15 with median & worse than median

protection Within each country (for given sample & given definition of control) firms classified in one of six types; then compute frequency of each type in each country

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

14

I.5. Controlling Shareholders Worldwide

Sample of large firms & using 20% definition

• 36% widely held, 30% family-controlled, 18% state-

controlled & 15% in residual categories

• Using this definition, all 20 UK firms, 18 out of 20

Japanese firms & 16 out of 20 US firms are widely held

• In Argentina, Greece, Austria, Hong Kong, Portugal,

Israel or Belgium, few widely held firms Exception

• f

dispersed

• wnership

⇒ Important to understand corporate governance in most countries

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

15

I.5. Controlling Shareholders Worldwide

Principal owner types ⇒ Families & state

• High percentage of state-controlled companies (70% in

Austria, 45% in Singapore & 40% in Israel & Italy) ⇒ Largest firms, unfinished privatization in most countries & massive post-war state ownership in the world

• Dominant form of controlling ownership in the world ⇒

By families (not by banks or other corporations)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

16

I.5. Controlling Shareholders Worldwide

Comparison of countries with good & poor shareholder protection ⇒ Widely held firms more common in countries with good protection (48% vs. 27%) Countries with poor shareholder protection ⇒ More of most other types of firms: family-controlled (34% vs. 25%), state-controlled (22% vs. 14%) Dispersion of ownership + good shareholder protection ⇒ Enables shareholders to divest at attractive prices

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

17

I.5. Controlling Shareholders Worldwide

Sample of large firms & using 10% definition

• In rich world as a whole, dispersed ownership is rare
• Many Japanese firms shift into miscellaneous category

⇒ Controlled by groups with no dominant members

• Japanese corporate ownership closer to US & UK than

to Continental European model Bottom line ⇒ Largest firms typically have UOs, particularly in countries with poor shareholder protection

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

18

I.5. Controlling Shareholders Worldwide

Sample of medium-sized firms & using 20% definition

• World average incidence of dispersed ownership =

24% compared to 36% for large firms ⇒ Going down in size = effect as relaxing strictness of control definition

• But in US & UK medium firms remain mostly widely

held ⇒ Attractiveness of selling out

• Percentage of firms controlled by families (dominant
• wnership pattern) rises to world average of 45%

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

19

I.5. Controlling Shareholders Worldwide

Comparison of countries with good & poor shareholder protection ⇒ Widely held firms (38% vs. 13%) Family (39% vs. 50%) & state control (9% vs. 20%) more frequent in poor protection environments Even using 20% definition, medium firms generally have

• wners (especially when shareholder protection is poor)

Using 10% definition, predominant ownership pattern is family control (regardless of shareholder protection)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

20

I.5. Controlling Shareholders Worldwide

Conclusions from empirical evidence

• Among largest firms & using tough definition of control

(20%) ⇒ Dispersed ownership is as common as family

• wnership
• Focusing on medium-sized firms, using more relaxed

definition

• f

control (10%) & paying attention to countries with poor protection ⇒ Widely held firms become exception & family control becomes the rule

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

21

I.6. Means of Achieving Control

I.6.1. Different types of control-enhancing mechanisms Firms in EMs typically affiliated with a business group ⇒ Controlled by owner through complex web of ownership (pyramidal structures, cross-holdings, dual class shares) Some control-enhancing mechanisms frequently used by controlling owners are the following

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

22

I.6. Means of Achieving Control

• 1. Pyramidal structures

Control structures where controlling owner holds shares in the firm through one or more intermediate entities (e.g., trusts, funds, foundations, limited partnerships, holdings, any other form of corporation) of which he owns < 100% Pyramidal ownership structures can be beneficial or detrimental from point of view of whole organization ⇒ “tunnelling” vs. “propping up”

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

23

I.6. Means of Achieving Control

• 1. Pyramidal structures

“Tunnelling” ⇒ Transfer of value from one firm to another in a pyramid (e.g., via transfer pricing, provision of capital at artificial prices, inflated payments for intangibles) whereby wealth transfer benefits one firm at expense of the other “Propping up” ⇒ Controlling owners support distressed firms within pyramid to help them survive & prosper (e.g., help to attract additional funds); Pyramidal structures as substitutes for lack of external capital markets

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

24

I.6. Means of Achieving Control

• 2. Cross-holdings

Control structures in which firm

• wns

shares in a corporation that belongs to largest owner’s chain of control

• 3. Multiple share classes / Dual class shares

Voting structures such that firm has issued two or more classes of stock with different voting rights

• 4. Voting agreements

Pacts among shareholders that result in largest owner’s voting power over more shares than what he owns

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

25

I.6. Means of Achieving Control

• 5. Controlling owner alone

Ownership structure in which, apart from controlling

• wner, there is no other large shareholder in company
• 6. Management control

Control structures in which top managers (i.e., CEO, honorary chairman, chairman or vice-chairman) come from largest shareholder’s family or are affiliated with controlling owner

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

26

I.6. Means of Achieving Control

I.6.2. Some examples of complex ownership structures

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

27

Nordström family group (Sweden)

I.6. Means of Achieving Control

I.6.2. Some examples of complex ownership structures

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

28

Nordström family group (Sweden)

I.6. Means of Achieving Control

I.6.2. Some examples of complex ownership structures

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

29

Unicem (Italy)

I.6. Means of Achieving Control

I.6.3. Main objective of control-enhancing mechanisms

• Complex ownership structures used by controlling
• wners to enhance control beyond ownership level
• Common

belief ⇒ Control-enhancing mechanisms designed to expropriate minority shareholders’ wealth

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

30

I.6. Means of Achieving Control

I.6.3. Main objective of control-enhancing mechanisms

• But some structures may have potential financing

benefits ⇒ Internal capital markets

• Overall,

control-enhancing mechanisms lead to separation between ownership & control ⇒ Largest

• wner’s voting rights > CF rights (Effects: value

discounts & corporate policy distortions)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

31

I.7. Separation Control–Ownership

Empirical evidence from around the world

• 1. Multiple classes of shares

Deviations from one-share–one-vote tend to be small ⇒ 18.6% of capital to control 20% of votes Differences across legal protection environments ⇒ 17.7% (poor) vs. 19.7% (good) Some countries (Scandinavia) have more significant deviations but overall deviations are small Conclusion ⇒ Multiple classes of shares are not main mechanism of separating ownership & control

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

32

I.7. Separation Control–Ownership

• 2. Pyramidal structures

26% of firms controlled through pyramids Different predominance across countries ⇒ 18% (good)

• vs. 31% (poor)

Pyramidal ownership as more important mechanism to separate CF ownership from control Bottom line ⇒ Through pyramids (more than through high voting rights shares) controlling shareholders acquire disproportionate power

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

33

I.7. Separation Control–Ownership

• 3. Cross-holdings

With the exception of few countries (Sweden, Germany) cross-shareholdings are rare Cross-shareholdings are restricted by law in only six sample countries (Belgium, France, Germany, Italy, Korea, Spain) & appear more common in those countries

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

34

I.8. The Special Case of Family Control

Empirical evidence on family-controlled firms

• Ultimate family owners control on average 25% of

value of top 20 firms

• On average, owner families control 1.33 of top 20 firms

(2.5 in countries like Israel & Sweden)

• Significant control of productive resources by largest

shareholding families

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

35

I.8. The Special Case of Family Control

Separation of ownership & control in family firms

• Family

management (CEO, Chairman, Honorary Chairman or Vice-Chairman) ⇒ 69% cases

• Differences across regions with different levels of

protection ⇒ 75% (good) vs. 64% (poor)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

36

I.9. Main Conclusions

• Dispersed ownership is common only among largest

companies in richest common law countries (US)

• Outside US (countries with poor investor protection) ⇒

Even largest firms have controlling owners

• Most frequent types of controlling shareholders ⇒

Family & state

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

37

I.9. Main Conclusions

• Controlling shareholders have control in excess of CF

rights ⇒ Pyramidal structures, family management

• Problem of ownership–control separation ⇒ Power &

interest to expropriate minority shareholders

• Existing corporate structures as “equilibrium response”

to legal environments ⇒ Controlling shareholders do not support legal reform

• Surprising! ⇒ Increase in value of dividend rights vs.

benefits from expropriation (value of control)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

38

I.9. Main Conclusions

• Private benefits of control ⇒ Only accrue to controlling
• wner; can be defined as “psychic” value attributed to

being in control (i.e., pure pleasure of command)

• Improvement of minority investors protection ⇒ Wealth

transfer from controlling to minority shareholders

• Other potential lobbying agent for CG reform ⇒

Entrepreneurs interested in issuing equity

• Unlikely convergence towards dispersed ownership

patterns among corporations worldwide

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

39

Ownership Structures in East Asia Lectures 5 & 6 – Part II

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

40

• II. Ownership Structures in East Asia

II.1. Data Set & Sample Coverage II.2. Use of Complex Ownership Structures II.3. Separation Ownership–Control across Types of Large Shareholders II.4. Relevance of Family Control in East Asia II.5. Some Examples of Control Structures II.6. Main Conclusions

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

41

II.1. Data Set & Sample Coverage

• 2,980 publicly traded corporations in HK, INDO, JAP,

SK, MAL, PHI, SIN, TAIW & THAI (9 countries)

• About 3/4 of market value of assets covered even

though sometimes only 1/2 of listed firms ⇒ Largest 100 firms in market cap.

• Analysis of CF & control rights of firms studying all

shareholders with > 5% of votes ⇒ In most cases principal shareholders ≠ ultimate owners

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

42

II.2. Use of Complex Ownership Structures

Mechanisms that separate ownership & control

• 1. Multiple classes of shares

Deviations from one-share–one-vote very small ⇒ 19.76%

• wnership to get 20% control; no superior voting shares in

some countries (JAP, SK & SIN)

• 2. Pyramidal structures

In 38.7% of companies pyramidal structures are used ⇒ High (INDO, SIN) & low (THAI, HK) prevalence across countries

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

43

II.2. Use of Complex Ownership Structures

• 3. Cross-holdings

Not widely widespread ⇒ Exceptions: MAL (14.9%) & SIN (15.7%); least predominance in THAI (0.8%)

• 4. Controlling owner alone

About 2/3 companies with single UOs ⇒ Difference across countries (JAP, SK vs. PHI, SIN, THAI) Interesting results for THAI ⇒ Less prevalence

• f

pyramidal structures & cross-holdings (informal alliances & interfamily business cooperation)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

44

• 5. Participation in management

Manager is employee of controlling financial institution/ corporation or member of controlling family High overlap control–management (average = 60%) ⇒ Differences: 4/5 companies in INDO, SK, MAL, TAIW but less than 1/2 in JAP (professional management) & PHI (interlocking directorates & management boards) Higher incidence of owners–managers than worldwide ⇒ Family members with ≠ last names & smaller companies

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

45

II.2. Use of Complex Ownership Structures

II.3. Separation Ownership–Control across Types of Large Shareholders

Separation ownership–control across different firm sizes (largest 20, median 50, smallest 50) ⇒ More variation in control structures across median & small-firm groups? Family firms have most deviation

• wnership–control;

except in JAP (WH financial firms) & SIN (state-controlled) Less clear pattern across firm size (focusing on FFs) Overall ⇒ Family firms most likely to have separation

• wnership–control & small firms most likely to have

deviation CF–control rights (regardless of ownership type)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

46

II.4. Relevance of Family Control in East Asia

Concentration of control in hands of family groups ⇒ Market entry, access to financing, government policy

• No. of firms controlled by single family ⇒ Largest no. in

INDO (> 4) & fewest in JAP (≈ 1) Control of listed corporate assets in hands of few families ⇒ Especially in some countries (INDO, PHI, HK)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

47

II.4. Relevance of Family Control in East Asia

Additional evidence ⇒ Value of total assets controlled by largest family group in each country Largest 5 families in PHI control 52.2% of market cap. in sample & sample firms = 82% total market cap. ⇒ 42.8% (52.2% * 82%) control of largest 5 families Extreme ⇒ 16.6% (57.7) & 17.1% (52.5) total market cap. controlled by single (top 10) family in INDO & PHI High concentration of control in families in THAI, SK & HK but family control is insignificant in JAP (top 15 – 2.8%)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

48

II.4. Relevance of Family Control in East Asia

Corporate assets held by top 15 families as % of GDP ⇒ High concentration of control in HK (84.2%), MAL (76.2%) & importance of family control ↓ in INDO, SK, THAI In JAP ⇒ Top 15 families control corporate assets = 2.1% GDP in 1996 For comparison ⇒ Wealth of 15 richest American families in 1998 was 2.9% of GDP Conclusion ⇒ Relatively few families effectively control most East Asian economies

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

49

II.4. Relevance of Family Control in East Asia

Have these families strong effect on economic policy of governments? ⇒ Some examples: Suharto family (INDO), Marcos family (PHI) Evidence of crony capitalism in East Asia ⇒ Incentives & abilities to “lobby” governments for preferential treatment Concentration of control ⇒ Detriment to evolution of countries’ legal systems (e.g., minority investors’ rights) Participation by governments in control ⇒ Possibility of conflicts between public & private interests

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

50

II.5. Some Examples of Control Structures

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

51

The Ayala group (The Philippines)

II.5. Some Examples of Control Structures

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

52

The Li Ka-shing group (Hong Kong)

II.6. Main Conclusions

• In 9 East Asian countries, control is enhanced through

pyramidal structures & cross-holdings

• Voting rights exceed CF rights ⇒ Especially in INDO,

JAP & SIN

• More than 2/3 of firms are controlled by single owner &

separation ownership–management is rare

• Implications

for ability & incentives

• f

controlling shareholders to expropriate from minority investors

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

53

II.6. Main Conclusions

• Family control in > 1/2 East Asian corporations but with

differences ⇒ WH firm in JAP & FFs in INDO & THAI

• Significant state control in INDO, SK, MAL, SIN & THAI
• Higher separation ownership–control in family firms
• Older & smaller firms are more likely family-controlled
• Control concentration ↓ with economic development
• Significant share of corporate assets rests in hands of

few families (except JAP)

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

54

Recommended Readings

• The Economist (2011): ‘Emerging-market giants. Tata

sauce’.

• Financial Times. (2010): ‘India’s family groups groom

future leaders’.

• The Economist (2010): ‘South Korea’s industrial giants.

The chaebol conundrum’.

• The Economist (2010): ‘A special report on Latin
• America. Efficiency drive’.

E m e r g i n g C a p i t a l M a r k e t s – I g n a c i o R e q u e j o

55