Corporate Governance and Corporate Behavior in Japan

SPRINGER BRIEFS IN ECONOMICS
DE VELOPMENT BANK OF JAPAN RESEARCH SERIES
Masaharu Hanazaki
Corporate Governance
and Corporate Behavior
in Japan
The Consequences of
Stock Options and
Corporate Diversification
SpringerBriefs in Economics
Development Bank of Japan Research Series
Series Editor
Keimei Kaizuka
Editorial Board Members

Akiyoshi Horiuchi
Toshihiro Ihori
Eiji Ogawa
Masayuki Otaki
Masaharu Hanazaki
Masaaki Komiya
Jun-ichi Nakamura
Akihiko Mori
This series is characterized by the close academic cohesion of nancial economics,
environmental economics, and accounting, which are the three major elds of
research of the Research Institute of Capital Formation (RICF) at the Development
Bank of Japan (DBJ). Readers can acquaint themselves with how a nancial
intermediary efciently restructuring rms in nancial distress, can contribute to
economic development.
The aforementioned three research elds are closely connected with one another
in the following ways. DBJ has already developed several corporation-rating
methods, including the environmental rating by which DBJ decides whether or not
to make concessions to the candidate rm. To evaluate the relevance of this rating,
research, which deploys not only nancial economics but also environmental
economics, is necessary.
The accounting section intensively studies the structure of IFRS and Integrated
Reporting to predict their effects on Japanese corporate governance. Although the
discipline of accounting is usually isolated from nancial economics, structural and
reliable prediction is never achieved without sufcient and integrated knowledge in
both elds.
Finally, the environmental economics section is linked to the accounting section
in the following manner. To establish green accounting (environmental account-
ing), it is indispensable to explore what the crucial factors for the preservation of
environment (e.g. emission control) are. RICF is well-equipped to address the acute
necessity for discourse among researchers who belong to these three different elds.
More information about this series at http://www.springer.com/series/13542

Masaharu Hanazaki
Corporate Governance
and Corporate Behavior
in Japan
The Consequences of Stock Options
and Corporate Diversication
123
Masaharu Hanazaki
Graduate School of Commerce
and Management
Hitotsubashi University
Tokyo
Japan
and
Research Institute of Capital Formation
Development Bank of Japan
Tokyo
Japan
ISSN 2191-5504 ISSN 2191-5512 (electronic)

SpringerBriefs in Economics
ISSN 2367-0967 ISSN 2367-0975 (electronic)
Development Bank of Japan Research Series
ISBN 978-4-431-56004-3 ISBN 978-4-431-56006-7 (eBook)
DOI 10.1007/978-4-431-56006-7
Library of Congress Control Number: 2016943876
Development Bank of Japan 2016

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microlms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specic statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made.
Printed on acid-free paper
This Springer imprint is published by Springer Nature

The registered company is Springer Japan KK
Preface
Japans corporate governance structure has changed greatly since the second half
of the 1990s. Institutional reform involving enterprise law, such as the introduction
of a stock option and removal of the ban on holding companies, served as back-
ground. In this brief, the author uses microdata on listed companies in an empirical
study of how the new trends in corporate management in Japan, such as the
introduction of a stock option and the evolution of corporate diversication and
division of corporations, influence corporate behaviors and performance. The main
ndings of a stock option analysis show that the introduction of a stock option had
neither a positive impact upon protability nor a negative side effect of promoting
risk-taking behaviors. The results of an empirical analysis of corporate diversi-
cation and division of corporations show their negative impacts upon protability.
However, their negative magnitude is not dependent upon the degree of corporate
diversication or division of corporations. This nding suggests that weak corpo-
rate governance is not the key problem but that Japanese rms managerial strategy
may be unsophisticated or Japanese corporate diversication and division of cor-
porations might aim at attaining goals such as future rm development, ensuring
market share, welfare improvement for rm employees, and job preservation.
Kunitachi, Japan Masaharu Hanazaki
v
Acknowledgments
This brief is an output of the Development Bank of Japan Research Series. Many
people have helped me in this project in various ways. I would like to thank
Akiyoshi Horiuchi, Toshihiro Ihori, and other participants of the DBJ seminar held
on September 25, 2015 for the fruitful discussions. I also thank the editorial
committee for their constructive comments and suggestions. Kanako Matsushita
provided me with valuable statistical assistance for an earlier version. I would also
like to acknowledge with gratitude the editorial assistance from Yuko Hosoda,
Ayako Ozawa, and other members of the Research Institute of Capital Formation of
Development Bank of Japan Inc. I am especially grateful to editorial staff at
Springer Japan KK, who gave me useful suggestions and contributed to the com-
pletion of this brief. I gratefully acknowledge the following nancial support:
JSPS KAKENHI Grant Number 26380389.
vii
Contents
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 Institutional Characteristics and Effects of a Stock Option. . . . . . . . 5
2.1 What Is a Stock Option? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2 The Development of the Stock Option System in Japan . . . . . . . . 6
2.3 Stock Option Studies on Japanese Firms. . . . . . . . . . . . . . . . . . . 11
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3 Empirical Analysis of a Stock Option . . . . . . . . . . . . . . . . . . . . . . . 15
3.1 Estimation of Stock Option Value . . . . . . . . . . . . . . . . . . . . . . . 16
3.1.1 Basic Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.1.2 Estimation Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1.3 Estimation Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2 Has the Protability of Companies Improved
by the Introduction of a Stock Option?. . . . . . . . . . . . . . . ..... 21
3.3 Did the Introduction of a Stock Option Induce Risk-Taking
Firm Behaviors? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 25
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 30
4 Related Studies of Corporate Diversication . . . . . . . . . . . . . . . . . . 31
4.1 Theories of Corporate Diversication . . . . . . . . . . . . . . . . . . . . . 32
4.2 Negative Effects of Corporate Diversication . . . . . . . . . . . . . . . 32
4.3 Reexamination of the Diversication Discount . . . . . . . . . . . . . . 33
4.4 Corporate Governance and Corporate Diversication . . . . . . . . . . 34
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5 Corporate Diversication of Japanese Firms . . . . . . . . . . ........ 37
5.1 How Can I Determine the Actual Conditions
of Corporate Diversication?. . . . . . . . . . . . . . . . . . . ........ 37
5.2 Corporate Diversication and Division of Corporations
in Japanese Firms . . . . . . . . . . . . . . . . . . . . . . . . . . ........ 41
Reference. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........ 45
ix
x Contents
6 Empirical Analyses of Corporate Diversication and Division

of Corporations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 47
6.1 Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 48
6.2 The Empirical Analysis: A Comparison Between
Consolidated Financial Statements and Non-consolidated
Financial Statements for Corporations Which Have
Conducted Corporate Diversication and Division
of Corporations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 49
Firms with Consolidated Subsidiaries and Those Without . . ..... 55
6.4 The Empirical Analysis: Analysis Based on Segmental
Information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 59
6.5 The Empirical Analysis: Analysis of Pure Holding
Companies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 67
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 75
7 Main Results and Their Interpretation . . . . . . . . . . . . . . . . . . . . . . 77
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Appendix Tables: Basic Statistics for Major Parameters

of the Black-Scholes Model . . . . . . . . . . . . . . . . . . . 81
About the Author
Masaharu Hanazaki is Professor at the Graduate School of Commerce and

Management, Hitotsubashi University. He graduated from Waseda University in
1979 and joined the Japan Development Bank. He was Director General and the
Executive Director of the Research Institute of Capital Formation of the
Development Bank of Japan (DBJ) from October 2003 to March 2012. After
retiring from the DBJ in March 2012, he immediately assumed his current position
at Hitotsubashi University. His main elds of research are corporate nance, cor-
porate governance, nancial systems, and the Japanese economy. His major English
articles include:
Corporate Governance and Investment in East Asian FirmsEmpirical
Analysis of Family-Controlled Firms (with Qun Liu), Journal of Asian Economics,
18, 7697, 2007.
A Review of Japans Bank Crisis from the Governance Perspective (with
Akiyoshi Horiuchi), Pacic-Basin Finance Journal, 11, 305325, 2003.
Is Japans Financial System Efcient? (with Akiyoshi Horiuchi) Oxford
Review of Economic Policy, Vol. 16, No. 2, 6173, 2000.
His English book publications include:

Designing Financial Systems in East Asia and Japan, (Joseph P. H. Fan,
Masaharu Hanazaki, and Juro Teranishi, eds.) Routledge Curzon, 2004.
A Vacuum of Governance in Japanese Bank Management, (with Akiyoshi
Horiuchi) in Banking, Capital Markets and Corporate Governance (Hiroshi Osano
and Toshiaki Tachibanaki, eds.), Palgrave, pp. 133180, 2001.
xi
Chapter 1
Introduction
Abstract Many institutional reforms of Japans corporate law system have

occurred since the 1990s, such as the introduction of a stock option (1997), removal
of the ban on pure holding companies (1997), enactment of the Corporate Law
(2005), announcement of Principles for Responsible Institutional Investors (2014),
and Japans Corporate Governance Code (2015). Japanese authorities expect that
these institutional reforms will stimulate Japanese rms and revitalize the Japanese
economy. This book will try to make clear the contents of recent trends in corporate
management, such as the growing importance of stock options, corporate diversi-
cation, and division of corporations and use micro data on listed companies to
analyze quantitatively the influences such corporate management trends have had
on corporate behavior and business performance.

Keywords Corporate governance Stock option Corporate diversication

Division of corporation Institutional reform
Many institutional reforms of Japans corporate law system have occurred since the
1990s, such as the introduction of a stock option (1997), removal of the ban on pure
holding companies (1997), enactment of the Corporate Law (2005), announcement
of Principles for Responsible Institutional Investors (2014), and Japans Corporate
Governance Code (2015). These reforms have introduced into Japan a corporate
governance structure that has already been introduced in advanced countries such as
the US. Japanese authorities expect that these institutional reforms will stimulate
Japanese rms and revitalize the Japanese economy.
In fact, many Japanese rms have tried to use US-style corporate governance
systems such as introducing stock option or advancing corporate diversication and
division of corporations. This represents a new corporate management trend in
Japan, and this book examines what kind of corporate activity and performance
such a corporate trend produces.
From the theoretical point of view, introducing a stock option enhances the effort
of the top manager to whom it is given, corporate value increases, and, nally, all
shareholders prots increase. Since the holder of a stock option can obtain an
Development Bank of Japan 2016 1

M. Hanazaki, Corporate Governance and Corporate Behavior in Japan,
Development Bank of Japan Research Series, DOI 10.1007/978-4-431-56006-7_1
2 1 Introduction
upside return without any restriction and does not need to pay downside risks if the
stock price is less than the exercise price, a top managers behavior with a stock
option should become a high-risk and high-return type.
Corporate diversication and division of corporations could become a leading
channel for realizing the management strategy needed for rms to continue to grow.
By enhancing corporate diversication, the rm can secure a new prot source and
can also expect a synergy effect by pursuing the old business and the new business
together. Moreover, through a division of corporations, it becomes easy to prepare a
different personneladministration strategy for different subsidiaries, and the busi-
ness risk derived from each subsidiary should be intercepted effectively. On the
other hand, rms pursuing corporate diversication and division of corporations
may be producing substantial losses via resource misallocation, such as in
cross-subsidization or overinvestment, compared with rms without an excessive
corporate diversication or division of corporations. Moreover, such corporations
have complicated organizational forms, and it is quite difcult for external investors
to monitor their actual business conditions. In other words, the asymmetric infor-
mation between external investors and corporate managers is a serious problem,
which tends to allow top managers to promote their private interests.
The positive and negative aspects of stock options, corporate diversication, and
division of corporations have been derived from the analysis conducted on
European and American corporations, behind which there are the principalagent
problem resulting from the asymmetric information between shareholders and top
managers.
However, whether similar performance occurs in Japanese rms is not neces-
sarily clear because Japanese corporations feature various kinds of stakeholders,
including not only shareholders but also creditors such as main banks, two types of
keiretsu systems, vertical keiretsu and horizontal keiretsu, trading partnerships
based upon long-term customer relationships, employees working within a lifetime
employment system, and local residents. It is very important for Japanese rms to
maintain favorable relationships with these various stakeholders in addition to
major shareholders. Therefore, the standard hypotheses about stock options, cor-
porate diversication, and division of corporations might not necessarily apply to
Japanese rms.
This book, a revised and integrated version of Hanazaki and Matsushita (2010,
2014), will try to make clear the concrete contents of the recent trends in corporate
management such as the growing importance of stock options, corporate diversi-
cation, and division of corporations and use micro data on listed companies to
analyze quantitatively how such corporate management trends have influenced
corporate behavior and business performance.
In Chaps. 2 and 3, I analyze stock options. From Chaps. 46, I analyze cor-
porate diversication and division of corporations. Finally, Chap. 7 summarizes the
results and offers interpretations and implications.
References 3
References
Hanazaki, M., & Matsushita, K. (2010). Stock options and rm performance: Quantitative analysis
based upon calculated option values. Economics Today, 30(4). (Research Institute of Capital
Formation, Development Bank of Japan. In Japanese).
Hanazaki, M., & Matsushita, K. (2014). Corporate governance and corporate diversication:
Empirical analysis based upon corporate nancial data in Japan. Economics Today, 34(5).
(Research Institute of Capital Formation, Development Bank of Japan. In Japanese).
Chapter 2
Institutional Characteristics and Effects
of a Stock Option
Abstract A stock option is dened as a corporations own share option given as

remuneration to their working staff. In other words, it is a call option that identies
the share of their own corporations as an original asset under the present corpo-
ration law. The use of a stock option used to be prohibited in Japan. In May 1997, a
stock option system was introduced for the rst time, and, in November 2001, the
Commercial Law was revised and a call option system was introduced. The number
of companies that have introduced a stock option has increased steadily from the
beginning of the system in the 1997 scal year. It reached a peak of 798 companies
in the 2008 scal year. After that, affected by the costs-and-expenses treatment of
the stock option, the worldwide economic recession, and a fall in stock prices
accompanying the Lehman shock, the number of companies that have introduced a
stock option declined to 721 companies in the 2010 scal year and 707 companies
in the 2012 scal year. Studies on stock options in Japanese rms multiplied after
the introduction of the system in 1997. Many studies focus upon what kinds of
corporation introduced the stock option. Some studies examine the consequences or
influences of the introduction of the stock option. For example, Nagaoka (J Bus 78
(6):22892315, 2005) shows that rms that are younger, that show more volatile
stock prices, that belong to the high-intensity R&D sector, and that have higher
growth potential tend to have a higher propensity to introduce a stock option. Kato
et al. (J Finan Econ 78(2):435461, 2005) compare between stock option-adopting
rms and non-adoption rms over the three years before and after the adoption of
the stock option and show that management ownership becomes higher for the
adoption rms, that there are no signicant differences in dividends or stock price
volatility between the adopting and non-adopting rms, and that the protability of
the adopting rms measured by return on assets is higher.

Keywords Stock option Call option Commercial law Incentive view

Selection view Managerial remuneration

6 2 Institutional Characteristics and Effects of a Stock Option
2.1 What Is a Stock Option?
In modern corporations with a separation between ownership and control, the

questions of how the different interests between shareholders and corporate man-
agers can be settled and how corporate performance can be improved are important
issues for corporate governance.
One mechanism that gives a pecuniary incentive to a corporate manager is a
stock option. According to accounting standards, a stock option is a corporations
own share option given as remuneration to their working staff. Here, the share
option is the call option that identies the share of their own corporation as an
original asset under the present corporation law.
Put another way, a stock option provides the claim for the stock of its company
to be acquirable at a price determined beforehand (i.e., the exercise price) to cor-
porate directors and employers. As in the original meaning of option, those who
have a claim can freely select whether they will exercise the claim or not. In a stock
market condition in which the stock price is less than the exercise price of the stock
option, an exercised right causes losses. Therefore, a stock option is exercised under
the conditions that the actual stock price is at least higher than the exercise price.
As shown in Fig. 2.1, in a general equity investment, where the stock is the
corporations own stock or other corporations stock, stock holders can obtain
capital gains when the current stock price is higher than its purchasing price, while
they will suffer capital losses when the current stock price falls. The owner of a
stock option can enjoy capital gains when the stock price at the time of its exercise
is higher than its exercise price, but the owner doesnt have to suffer losses even if
the stock price at the time of its exercise is lower than its exercise price because the
owner would not exercise the option. Thus, the owner of a stock option does not
have to pay the downside risk by not exercising a claim.
In other words, the owner of a stock option can increase the expected return if a
high-risk, high-return type of investment project is chosen. Therefore, giving stock
options to top managers may promote risky corporate behavior. In the US, the stock
option has been widely used as part of a top managers reward package since the
1960s. According to Yermack (1995), who investigated the CEO compensation of
792 large US corporations, the share of stock option rewards out of total com-
pensation had reached 20 % in 1984 and 35 % in 1990.
2.2 The Development of the Stock Option System in Japan
Stock options used to be prohibited in Japan because the acquisition of a companys

own stock was restricted due to the belief that it was important for corporations to
improve their equity capital under the old Commercial Law. Moreover, a special
resolution at a stockholders meeting was needed for equity issuances at a price less
2.2 The Development of the Stock Option System in Japan 7
(1) Equity Investment
Expected Profit
Positive Profit
Zero Profit
Negative Profit
Purchasing Price Future Stock Price
(2) Stock Option
Expected Profit
Positive Profit
No Negative
Zero Profit
Exercise Price Future Stock Price
Fig. 2.1 Comparison between equity investment and stock option

than the market price of the stock; the effect of the special resolution covered a
6-month period because stockholders interests were regarded as the highest
priority.
In May 1997, a partial amendment to the Commercial Law was made, and a
stock option system was introduced for the rst time. Two different modes were
adopted. The rst was called the companys own stock mode. In this mode, an
incorporated company provides a claim whereby the companys own stock held by
the corporation can be purchased at a predetermined price by the directors and/or
employees. The second is the stock subscription rights mode, whereby an
incorporated company provides the claim to the directors and/or employees that
they can purchase new stock at a predetermined price; when the claim is exercised,
new stocks are issued.
An ordinary resolution at a regular shareholders meeting is necessary to provide
the stock option claim in the former case, but, in the latter case, after making
provisions in the articles of association, a special resolution of the shareholders
meeting is needed. In both cases, the exercise-of-right time periods are more than
2 years and less than 10 years from the date of the shareholders meeting resolu-
tion. Using both modes together in the same corporation is not allowed.
In November 2001, the Commercial Law was revised, and a call option system
was introduced. Thereafter, a stock option was regarded as the issuance of a new
share warrant, where the price of the new share has to be less than its market price.
Before this change, the entities for stock option investiture had been restricted to the
directors and employees of the company itself. After the revision, investiture
entities were expanded to include the directors and employees of subsidiary com-
panies, corporate lawyers, and others. Moreover, procedures for making provisions
in the rms articles of association became unnecessary, although a special reso-
lution of the stockholders meeting was still needed. Restrictions on the
exercise-of-right time period were also abolished.1
Accounting standards concerning stock options were released in December 2005
to prevent abuse. This standard required the investiture of a stock option to be
regarded as equity-related compensation costs and expenses for stock options given
after the date of the enforcement of the Company Law in May 2006.
The introduction of the stock option to Japanese TSE-listed companies is shown
in Fig. 2.2. The number of companies that have introduced the stock option has
increased steadily from the beginning of the system in the 1997 scal year, reaching
365 companies in the 2000 scal year. In response to the abovementioned insti-
tutional reform in November 2001, an upsurge in stock option introduction began in
the 2002 scal year; the number of companies using stock options reached 617in
the 2002 scal year, 771 in the 2004 scal year, 779 in the 2006 scal year, and 798
in the 2008 scal year. Afterwards, due to the abovementioned costs-and-expenses
1
However, a preferential tax treatment is applied to stock options exercised from 2 to 10 years
from the date of the decision of the shareholders meeting.
2.2 The Development of the Stock Option System in Japan 9
800
700
600
Number of firms
500
Mothers
Second Section
400
First Section
300
200
100
0
1998 2000 2002 2004 2006 2008 2010 2012
Fiscal Year
Fig. 2.2 Development of the introduction of the stock option for japanese TSE-listed companies.
Note From 1998 to 2004, the authors estimation. From 2006 to 2012, TSEs survey data
treatment of stock options, the worldwide economic recession, and the fall in stock
prices accompanying the Lehman shock, the number of companies that have
introduced stock options declined to 721 in the 2010 scal year and 707 in the 2012
scal year.
The number of companies listed in the First Section of the Tokyo Stock
Exchange was overwhelmingly high during the several years after the beginning of
the system in the 1997 scal year. In the 2000s, however, the number of companies
listed in the Second Section and the Mothers of the Tokyo Stock Exchange
increased. Firms in the Mothers have a higher interest in the introduction of the
stock option, and about 80 % of them have introduced the stock option in recent
years.
Table 2.1 shows the number of companies that introduced the stock option and
the number of listed companies by industry and by Market Section in the Tokyo
Stock Exchange. In manufacturing industries, more than 30 % of companies (e.g.,
electrical machinery, pharmaceutical products, paper and pulp, precision machin-
ery, transportation equipment, chemistry, glass and ceramics, and soil-and-stone
nished products rms) have introduced the stock option. In non-manufacturing
industries, the percentage of companies that have introduced the stock option is
much higher; in industries such as the information and telecommunication, services,
real estate, and retail industries, most companies have introduced the stock option.
Table 2.1 The number of rms that introduced stock option by TSE market and by sector
Unit: The number of
rms
First section Second Mothers TSE all markets
section
SO LC SO LC SO LC SO LC SO/LC
(%)
Manufacturing 256 862 38 250 21 25 315 1,137 27.7
Food 17 73 2 28 2 2 21 103 20.4
Textile 13 46 3 15 0 0 16 61 26.2
Paper and pulp 2 12 3 5 1 1 6 18 33.3
Chemistry 43 121 3 28 0 0 46 149 30.9
Pharmaceutical 10 35 1 1 4 4 15 40 37.5
Petroleum and coal 2 11 0 2 0 0 2 13 15.4
Rubber 5 12 0 5 0 0 5 17 29.4
Glass and ceramics 10 30 2 10 0 0 12 40 30.0
Iron and steel 6 35 0 9 0 0 6 44 13.6
Nonferrous metal 5 25 3 10 0 0 8 35 22.9
Metal products 6 38 0 20 0 0 6 58 10.3
General machinery 26 121 6 40 1 2 33 163 20.2
Electrical machinery 64 166 9 39 8 10 81 215 37.7
Transportation equipment 25 64 1 17 0 1 26 82 31.7
Precision machinery 8 26 2 9 2 2 12 37 32.4
Other manufacturing 14 47 3 12 3 3 20 62 32.3
Non-manufacturing 348 869 72 226 142 164 562 1,259 44.6
Agriculture, forestry and 3 6 1 1 0 0 4 7 57.1
shery
Mining 0 6 0 0 0 0 0 6 0.0
Construction 16 104 6 27 1 2 23 133 17.3
Electricity and gas 0 17 0 6 1 1 1 24 4.2
Land transportation 5 34 2 10 0 0 7 44 15.9
Marine transportation 2 10 1 4 0 0 3 14 21.4
Air transportation 0 4 0 1 1 1 1 6 16.7
Warehouse 3 18 0 9 1 1 4 28 14.3
Information and 63 93 10 20 55 63 128 176 72.7
telecommunication
Wholesale 51 144 14 43 8 9 73 196 37.2
Retail 72 142 17 47 15 19 104 208 50.0
Banking 22 86 0 2 0 0 22 88 25.0
Securities 13 21 1 1 0 0 14 22 63.6
Insurance 5 8 0 0 1 1 6 9 66.7
Other nancial services 14 34 3 4 2 2 19 40 47.5
Real estate 29 54 6 17 11 12 46 83 55.4
(continued)
2.3 Stock Option Studies on Japanese Firms 11
Table 2.1 (continued)

Unit: The number of
rms
First section Second Mothers TSE all markets
section
SO LC SO LC SO LC SO LC SO/LC
(%)
Services 50 88 11 34 46 53 107 175 61.1
Total 604 1,731 110 476 163 189 877 2,396 36.6
SO/LC (%) 34.9 23.1 86.2 36.6
Notes SO represents the number of rms that introduced stock options. LC represents the number
of listed companies. The data are based upon the TSE Corporate Governance Information Service
as of May 25, 2007
2.3 Stock Option Studies on Japanese Firms
Studies on the stock option in Japanese rms multiplied after the introduction of the
system in 1997. Many studies focus on what kinds of corporation have introduced
the stock option. Some studies examine the consequences or influences of its
introduction.
Nagaoka (2005) empirically analyzes the determinants of the rms that intro-
duced the stock option between 1997 and 2000, examining 3,176 Japanese listed
companies. Among these rms, 391 had introduced the stock option. A probit
model set at 1 for rms that had introduced the stock option and 0 otherwise was
estimated. The result showed that rms that were younger, had more volatile stock
prices, that belonged to the high-intensity R&D sector, and that had higher growth
potential tended to have a higher propensity to introduce a stock option. In addition,
when a stock option was given to not only corporate directors but also to
employees, the result for rm age was amplied.
These results can be interpreted as follows. The rms that faced a wide range of
investment opportunities used the stock option to mitigate information asymmetries
with employees who had useful information and nally chose suitable investment
projects. Nagaoka (2005) points out that, in Japan, the stock option system tends to
be utilized not to motivate the managerial efforts of a top manager (incentive view)
but to mitigate information asymmetries when suitable investment opportunities are
being chosen (selection view).
Uchida (2006) also estimates a probit model in which the dependent variable is a
dummy variable set at 1 for stock option-introducing rms and 0 otherwise. The
sample comprises 782 listed rms in the First Section of the Tokyo Stock Exchange
in 2000. Among these, 109 had introduced the stock option. According to the
estimation results, among the independent variables, the coefcients of leverage
ratios showed signicantly negative signs, which tended to become larger for the
rms belonging to keiretsu or that had main banks. Uchida (2006) argues that this is
because Japanese rms tend to nance through debt, and, in order to reduce the
agency costs of debt nancing, highly leveraged rms are reluctant to grant stock
options. The scale factor of the rms had positive coefcients, which Uchida (2006)
interprets to indicate that the larger corporations are better situated to introduce
stock options because these rms have varied staff members, including experts in
law, accounting, and tax systems.2
Aman (2002) examines 117 stock option cases introduced between 1997 and
1999. The dependent variable is the stock option dependency ratios of the sample
rms, dened as the ratio of the number of the stock options already given to the
total number of already issued shares, or the ratio of the value of the stock option to
the market capitalization of the rm.
According to the estimation results of the tobit model, the management own-
ership ratio and nancial institution ownership ratio show statistically signicant
negative coefcients. Aman (2002) concludes that management ownership has a
function similar to that of the stock option and thus that these two methods have an
alternate relationship and that higher and more powerful ownership by nancial
institutions restrains stock options because it increases a kind of uncertainty.
Kato et al. (2005) analyze the determinants of the adoption of the stock option
system and the effects of the stock option. Their sample comprises TSE-listed
companies and covers 19972001. Among them, 344 rms that introduced the
stock option are included. First, a stock price event study of the adoption of the
stock option showed that approximately 2 % of the cumulative abnormal return
(CAR) was observed before and after ve days from the date the information
became public. The CAR tended to be higher when the ratio of the stock option
granted to board members was higher and lower when the ratio of the stock option
granted to employees was higher.
Second, a logit model estimation based upon the panel data showed the deter-
minants of the stock option. The results indicated that the rms with abundant
growth opportunities and facing liquidity constraints tended to introduce the stock
option and that highly leveraged rms tended to be reluctant to adopt it. Kato et al.
(2005) compare between stock option-adopting rms and non-adopting rms
during the three years before and after the adoption of the stock option in terms of
ownership structure, dividend policy, stock price volatility, and protability.
According to the results, management ownership became higher for the adopting
rms. There were no signicant differences in dividends or stock price volatility
between the adopting rms and non-adopting rms, and protability, measured by
the return on assets of the adopting rms, was higher. From these results, Kato et al.
(2005) conclude that the stock option system is functioning positively as an
incentive mechanism for creating higher shareholder value.
2
Moreover, Uchida (2006) shows the positive impact of the rm size upon the introduction of
stock option. He points out that there are many specialists and professionals, such as lawyers,
accountants, and licensed tax accountants, in the big companies. And he interpreted that the
availability of these talented person facilitated the introduction of the stock option which is a new
system for the rms.
2.3 Stock Option Studies on Japanese Firms 13
Kubo and Saito (2008) perform a comprehensive analysis of the managerial

remuneration that includes not only the normal pecuniary compensation such as a
salary and a bonus but also stock-related compensation such as stock options and
management ownership. They analyze the presidents total remunerations for 115
companies extracted from the Nikkei 225. Their estimations show that the stock
option value was an average of 4,860,000 yen per company in 2000, nearly 10 % of
the presidents total compensation (an average of 45,290,000 yen). They also cal-
culate the sensitivities of the presidents total compensation to the change in stock
values and nd that only a 0.85 yen increase in compensation is observed after a
1,000 yen increase in stock value in 1977. Moreover, sensitivities declined, with
only a 0.20 yen increase realized in 1992. Kubo and Saito (2008) argue that this
occurred because the presidents shareholding ratio followed a downward trend.
References
Aman, H. (2002). A stock option and ownership structure. Gendai Finance, 11, 4359. (in
Japanese).
Kato, H. K., Lemmon, M., Luo, M., & Schallheim, J. (2005). An empirical examination of the
costs and benets of executive stock options: Evidence from Japan. Journal of Financial
Economics, 78(2), 435461. doi:10.1016/j.jneco.2004.09.001.
Kubo, K., & Saito, T. (2008). The relationship between nancial incentives for company
presidents and rm performance in Japan. Japanese Economic Review, 59(4), 401418. doi:10.
1111/j.1468-5876.2008.00420.x.
Nagaoka, S. (2005). Determinants of the introduction of stock options by Japanese rms: analysis
from the incentive and selection perspectives. Journal of Business, 78(6), 22892315. doi:10.
1086/497050.
Uchida, K. (2006). Determinants of stock option use by Japanese companies. Review of Financial
Economics, 15(3), 251269. doi:10.1016/j.rfe.2005.08.001.
Yermack, D. (1995). Do corporations award CEO stock options effectively? Journal of Financial
Economics, 39(2), 237269. doi:10.1016/0304-405X(95)00829-4.
Chapter 3
Empirical Analysis of a Stock Option
Abstract I empirically analyze the characteristics of rms that introduce stock

options and the introductions effects on the performances of Japanese rms. First, I
compute the value of the stock options the rms introduced using two methods, the
BlackScholes option price valuation model and the binomial, or tree, model.
Second, I examine whether rm protability improved after the introduction of the
stock option. The result shows that the stock option values had no positive effects
on either ROA or ROE in both the manufacturing and non-manufacturing indus-
tries. For the manufacturing industries, where the dependent variable is ROA, stock
option values show a statistically signicant negative coefcient, suggesting that the
introduction of a stock option would not improve rm protability. Third, I
examine whether the introduction of a stock option induces risk-taking rm
behaviors. I use nancial leverage, capital investment, and research and develop-
ment (R&D) activity (for the manufacturing industry only) as risk-taking measurers.
The empirical analyses of the relationship between stock options and risk-taking
rm behavior show negative effects on nancial leverage as well as positive effects
on R&D activities. Therefore, the general hypothesis that the introduction of a stock
option promotes risk-taking rm activities is not supported empirically.

Keywords Stock option Blackscholes model Binomial model ROA ROE

Risk-taking Financial leverage Capital investment R&D
In the following, I empirically analyze the characteristics of rms that have intro-
duced stock options and the introductions effects on the performance of Japanese
rms. The sample comprises listed companies in the First and Second sections and
the Mothers markets of the Tokyo Stock Exchange as well as nancial rms. The
estimation period covers the 10 years from scal 1997 to scal 2006.

16 3 Empirical Analysis of a Stock Option
3.1 Estimation of Stock Option Value
3.1.1 Basic Model
In this section, I compute the value of the stock options the Japanese rms intro-
duced using two methods, the BlackScholes option price valuation model1 and
the binomial, or tree, model.2
The calculation formula of the option price based on the BlackScholes model is
the following:3
OP eqt S Nd1 ert K Nd2

p
d1 ln S = K r q r2 =2 t = r t 3:1
p
d2 d1 r t
where
OP Stock option value
S Stock price
K Exercise price
t Time to maturity
Volatility of stock price
r Risk-free rate of interest
q Dividend rate (the dividend per share divided by the average stock price for
the year)
N() An accumulated density function of a standard normal distribution
e The exponential function
ln A natural logarithmic function
The binomial model is based on the view that present stock prices are deter-
mined by the relationship of the behaviors of two types of investors: one expects
increases in future stock prices, and the other expects declines. I set the following
notations. p is the probability that the stock price of the company will rise, u is the
degree of the stock price rise, and d is the degree of the stock price decline. The
present stock price S is denoted by the following formula:
1
See Black and Scholes (1973) and Merton (1973). For our analyses, the model includes dividend
rates.
2
Cox et al. (1979) formulate the binomial model for option pricing. In our analyses, the model
includes dividend rates. There are differences between Black Scholes option price valuation
model and a binomial model. The former is a European type, where the exercise date of the option
is limited to the expiration date. The latter is an American type, where the stock option holders can
exercise the option whenever they like before expiration.
3
The basic statistics for the major parameters of the BlackScholes model are shown in the
appendix Tables 3.1, 3.2, 3.3 and 3.4.
3.1 Estimation of Stock Option Value 17
S ert fu S p d S 1 pg 3:2
The option prices determined by the option market are set at a level reflecting the
expectations of the future share price rises or declines. Those with a claim of a stock
option can gain if the stock price outperforms beyond its exercise price. When the
stock price is lower than the exercise price of the stock option, no gains can be
obtained. Therefore, the following formula is obtained:
OP ert fu S K p 0 1 pg 3:3
Here, from formula Eq. (3.2), the probability of the stock price increase, p, can be
written as
ert d
p 3:4
ud
The option price OP is computed by substituting Eq. (3.4) for (3.3).4
3.1.2 Estimation Approach
In this section, the stock option values for TSE-listed companies are calculated
based on the formulas explained in the above section. The main variables are
calculated as follows.
The stock price (S), as an underlying asset, is the average price at the highest and
lowest during the month in which the introduction of the stock option was voted at
the general meeting of shareholders. The volatility of the stock price () is the
historical volatility during the year when the introduction of the stock option was
voted at the general shareholder meeting. Volatility represents the standard devia-
tion of the monthly rate of stock price changes. The risk-free rate of interest (r) is
calculated as the yearly average rate of the 10-year yield on government bonds in
the secondary market from the monthly rates in the year of the date when the stock
option was voted at the general shareholders meeting. For the exercise price (K),
the time to maturity (t), the stock option investiture number of shares, and the
dividend rate (q), information in nancial statements and/or the corporate gover-
nance reports of the rms are utilized. When two or more stock options were voted
4
In our binomial model, the option prices are calculated assuming the case where the remaining
stock option periods are divided into 1,000 periods. Following Cox, Ross, and Rubinstein (1979),
the degrees of stock price increase (u) and decline (d) in Eq. (3.4) are calculated using the
following formula:
q q
r t=n r t=n
ue de n: the number of the divided periods.
in the same general shareholders meeting, the stock option value is calculated by
adding all cases together. Moreover, for rms that performed a stock split, the
necessary reconciliation according to the description in the nancial statements is
performed.5
3.1.3 Estimation Results
Table 3.1 totalizes three kinds of stock option values by scal year, calculated from
the BlackScholes model and the binomial model.
First, the estimated and total amounts of the stock option values newly granted in
each year are seen in the left-hand rows of Table 3.1. The amount for all industries
was 93.5 billion yen as calculated by the BlackScholes model and 92.6 billion yen
for the binomial model at the end of the 2000 scal year, when the stock option
began to be widely used in many rms. Then, the amount increased substantially, to
146.7 billion yen for the BlackScholes model and 145.7 billion yen for the
binomial model, in the 2002 scal year. After that, the amount increased and
decreased until the 2005 scal year. In the 2006 scal year, the amount declined
drastically to 49.7 billion yen for the BlackScholes model and 49.5 billion yen for
the binomial model. This decline reflects the institutional change whereby costs and
expenses have to be included when a stock option is introduced. In addition, the
average annual stock option value per company during the whole period is calcu-
lated at 340 million yen.
Second, the outstanding value of the stock option was computed from the
accumulated values of the newly granted stock options for each year based on the
view that the option values are maintained from the introductory year until the
expiring year of the exercise of the stock option right.6 The estimated outstanding
values are shown in the middle row of Table 3.1. The values increased, as expected,
and reached about 780 billion yen in the 2005 scal year.
Third, the relationship between a stock price and an option value is quantita-
tively examined. As is clearly seen from the formula of the BlackScholes model
5
For example, in cases where one stock is divided into two after the stock option is granted, from
the point of views of pure theory, the stock price is halved and the number of stocks doubles.
However, the stock price data used here are for the month when the introduction of the stock
option was decided. Therefore, the data for the number of stocks also cover before the stock
division.
6
To be precise, the stock option values on which the exercise was carried out should be subtracted
from the outstanding value of the stock option. However, since information on exercised stock
options is not available, we are computing the outstanding value of the stock option assuming that
the stock option is exercised at the exact date of the exercise expiration. In other words, we assume
that the stock option is the European type, where the exercise date is limited to the expiration date.
Moreover, although many parameters such as stock prices and their volatilities, which appear in
the estimation formula for the value, are changing over time, they are computed using data for the
year when the decision was made.
3.1 Estimation of Stock Option Value 19
Table 3.1 Estimated stock option values

Newly granted value Outstanding value (100 Marginally enhanced
(100 million yen) million yen) value (million yen)
BS Binomial BS Binomial BS model Binomial
model model model model model
1997 31.22 31.25 31.22 31.25 58.65 58.69
1998 144.82 144.80 200.75 200.81 260.42 260.86
1999 369.91 367.87 726.87 724.16 625.66 621.98
2000 934.65 925.77 1,738.20 1,726.62 1,884.24 1,849.45
2001 895.38 891.44 2,812.48 2,797.39 2,403.55 2,308.58
2002 1,466.81 1,456.75 4,472.37 4,444.53 3,871.91 3,779.27
2003 1,394.69 1,370.88 5,941.30 5,889.00 5,039.14 5,029.65
2004 1,504.55 1,490.72 6,941.38 6,878.03 7,302.87 7,389.27
2005 1,488.07 1,473.35 7,853.90 7,776.85 11,400.00 11,500.00
2006 497.13 495.01 7,699.87 7,625.79 15,300.00 15,200.00
all 8,727.22 8,647.83
periods
Notes
1. Newly Granted Value represents the sum of the stock option values that were newly introduced
in that year
2. Outstanding Value represents the sum of the stock option values that exist in that year
3. Marginally Enhanced Value represents the marginally enhanced values of the existing stock
option, when the stock price for each company in which the stock option was introduced increases
by 1 %
and the binomial model, stock option values are a function of stock prices. The
incentives to those to whom a stock option is given would heavily depend on how
much the option values would change when stock prices change. The right-hand
row of Table 3.1 shows the increases in option values when stock prices rise by
1 %. That amount increased to about 15 billion yen in 2006, following the
increasing trend in the introduction of stock options.
Next, the stock option values by industry are examined. Table 3.2 shows the
estimated stock option values by industry in 2006. The industries with high stock
option values include the electrical machinery (about 110 billion yen), chemistry
(about 36 billion yen), and transportation equipment (about 32 billion yen) sectors
in the manufacturing industries and the services (about 285 billion yen), retail
(about 72 billion yen), and real estate (about 45 billion yen) sectors in the
non-manufacturing industries.
Regarding the ratio of the stock option amount to the executive compensation,
shown in the parentheses in Table 3.2, the ratio is high for glass and ceramics (more
than nine times) and electrical machinery (more than seven times) rms in the
manufacturing industry and for electricity and gas (around eight times), real estate
Table 3.2 Estimated stock option values and executive compensation

Unit: 100
million yen
Industry Outstanding value Executive
compensation
BS model Binomial model
Manufacturing 2,737.72 (3.54) 2,730.52 (3.53) 774.07
Food 80.93 (1.62) 80.47 (1.61) 50.06
Textile 47.28 (1.63) 47.37 (1.63) 29.08
Paper and pulp 8.01 (0.55) 8.05 (0.55) 14.53
Chemistry 365.35 (2.72) 364.75 (2.72) 134.26
Glass and ceramics 307.67 (9.40) 300.02 (9.17) 32.72
Iron and steel 12.95 (1.31) 12.94 (1.31) 9.86
Nonferrous metal 28.06 (2.05) 28.07 (2.05) 13.67
Metal products 18.59 (1.43) 18.72 (1.44) 12.98
General machinery 215.27 (2.28) 215.90 (2.28) 94.54
Electrical machinery 1,101.65 (7.33) 1,101.19 (7.33) 150.22
Transportation equipment 321.13 (2.18) 324.94 (2.21) 147.24
Precision machinery 48.39 (2.42) 48.21 (2.41) 19.99
Other manufacturing 182.44 (2.81) 179.90 (2.77) 64.92
Non-manufacturing 4,962.15 (5.54) 4,895.27 (5.47) 894.94
Construction 100.48 (2.46) 100.93 (2.47) 40.87
Electricity and gas 38.00 (8.43) 35.51 (7.68) 4.63
Transportation and 361.63 (5.96) 341.45 (5.63) 60.68
telecommunication
Wholesale 408.31 (2.66) 409.18 (2.67) 153.28
Retail 723.14 (4.18) 715.26 (4.14) 172.84
Real estate 460.47 (7.61) 447.09 (7.39) 60.51
Services 2,869.12 (7.13) 2,845.85 (7.08) 402.13
Total 7,699.87 (4.61) 7,625.79 (4.57) 1,669.01
Notes
1. The executive compensation is the sum of the executive compensation for the TSE-listed
companies in the exact year when the stock options were granted
2. The amount of the stock option values has the same values from the time of the grant of the
stock options until the nal year of the exercise of the options
3. The nancial sector companies are excluded
4. The numbers in the parenthesis are the ratio of the amount of the stock option to the executive
compensation
(more than seven times), and services (more than seven times) rms in the
non-manufacturing industries. In contrast, the ratio is low for paper and pulp, iron
and steel, and metal products rms in manufacturing industries and for construction
and wholesale rms in the non-manufacturing industries.
3.2 Has the Protability of Companies Improved 21

by the Introduction of a Stock Option?
As mentioned, the main purpose of introducing a stock option is enhancing the working
incentives of top managers and/or employees and enhancing the business performances
of the company. In this section, I examine whether such a directional effect is realized by
estimating an econometric model ofrm performance. The performance indicators used
are two kinds of typical protability measures: ROA and ROE. The explanatory vari-
ables used are three indicators relevant to stock option values, computed using the
formula of the BlackScholes model and the binomial model used in Sect. 3.1the
outstanding values of the existing stock option, the ratio of the outstanding values of the
existing stock option to the executive compensation, and the marginally enhanced stock
option values of the existing stock option as a result of a 1 % gain in stock prices. The
nancial variables and corporate governance variables of each rm and a
macro-economic variable are also included as explanatory variables.
The main bank dummy is a dummy variable set at 1 for rms whose main bank
relationship is stable from the 1960s to the 1990s and 0 otherwise. Information on
main bank relationships are drawn from Keiretsu no Kenkyu by Keizai Chousa
Kyoukai (Economic Research Association).
The estimated equation is as follows:
PAit const: a1 SOVit a2 SC10it a3 FIit a4 ECit a5 LEVit a6 SALEit
a7 ASSETit a8 MAINi a9 DIt
3:5
where
PA Firm protability (ROA or ROE)
SOV Three indicators relevant to the amount of stock option values
SC10 Top 10 shareholders shareholding ratio
FI Financial institutions shareholding ratio
EC Foreigners shareholding ratio
LEV Debts to total assets ratio
SALE Sales growth rate
ASSET Total assets (natural logarithm)
MAIN The main bank dummy
DI The coincident index of business conditions
The subscripts i and t indicate rms and scal years, respectively. The basic
statistics for the above nancial variables are shown in Appendix 5.
Table 3.3 shows the estimation results of the panel analysis7 using all data
samples, not only rms that introduced the stock option but also rms that did not.
7
Here, I used not the xed effect model but the random effect model for the estimation because the
main bank dummy variable, which is a rm specic variable, is used as one of the independent
variables.
22
Table 3.3 Stock option and rm protability

(1) Manufacturing
ROA ROE
SOV1 0.0288 0.0133
(2.81) (0.41)
SOV2 0.0269* 0.0423
(1.85) (1.02)
SOV3 0.0330** 0.0040
(2.18) (0.08)
SC10 0.0287*** 0.0292*** 0.0287*** 0.0874*** 0.0879*** 0.0874***
(5.39) (5.47) (5.39) (6.75) (6.79) (6.75)
FI 0.0264*** 0.0273*** 0.0265*** 0.0529*** 0.0526*** 0.0525***
(5.07) (5.25) (5.09) (3.81) (3.79) (3.78)
EC 0.0743*** 0.0710*** 0.0736*** 0.1435*** 0.1423*** 0.1447***
(12.55) (12.07) (12.45) (8.29) (8.26) (8.37)
LEV 0.0141*** 0.0140*** 0.0139*** 0.0052 0.0039 0.0048
(3.78) (3.76) (3.72) (0.60) (0.45) (0.56)
SALE 0.0477*** 0.0479*** 0.0475*** 0.0914*** 0.0929*** 0.0917***
(24.22) (24.31) (24.15) (13.09) (13.29) (13.16)
ASSET 0.4569*** 0.4629*** 0.4592*** 0.6031*** 0.5896*** 0.5993***
(4.97) (5.04) (4.99) (3.49) (3.41) (3.46)
MAIN 0.5868** 0.5872** 0.5787** 0.4224 0.4409 0.4264
(2.25) (2.25) (2.22) (1.05) (1.10) (1.06)
DI 0.0223*** 0.0220*** 0.0221*** 0.05667*** 0.0574*** 0.0570***
(17.17) (17.09) (17.08) (11.74) (12.00) (11.86)
(continued)
3 Empirical Analysis of a Stock Option
(1) Manufacturing
ROA ROE
Const. 3.6635*** 3.6536*** 3.6643*** 4.7806*** 4.7909*** 4.7755***
(5.98) (5.95) (5.98) (3.98) (3.98) (3.97)
R2 0.1431 0.1411 0.1434 0.1323 0.1319 0.1325
The number of observations 10,245 10,232 10,245 10,142 10,129 10,142
(2) Non-manufacturing
ROA ROE
SOV1 0.0006 0.0141
(0.06) (0.40)
SOV2 0.0140 0.0146
(1.56) (0.54)
SOV3 0.0272* 0.0518

(1.74) (1.00)
SC10 0.0735*** 0.0754*** 0.0731*** 0.2091*** 0.2130*** 0.2094***
(11.54) (11.88) (11.49) (12.23) (12.46) (12.25)
FI 0.0385*** 0.0375*** 0.0381*** 0.0601*** 0.0584*** 0.0605***
(5.88) (5.74) (5.81) (3.10) (3.02) (3.12)
EC 0.0731*** 0.0719*** 0.0746*** 0.2047*** 0.2116*** 0.2026***
(9.60) (9.35) (9.81) (8.65) (8.85) (8.57)
LEV 0.0109** 0.0104** 0.0113*** 0.0636*** 0.0646*** 0.0637***
(2.47) (2.39) (2.58) (5.25) (5.35) (5.27)
SALE 0.0157*** 0.0158*** 0.0158*** 0.0234*** 0.0233*** 0.0232***
(continued)
23
24
(1) Manufacturing
ROA ROE
(9.00) (9.08) (9.04) (3.93) (3.91) (3.90)
ASSET 0.4417*** 0.4840*** 0.4320*** 1.0480*** 1.0613*** 1.0473***
(4.19) (4.61) (4.10) (4.60) (4.66) (4.60)
MAIN 0.3875 0.2689 0.4119 0.8582 0.8937 0.8652
(0.73) (0.51) (0.77) (0.99) (1.03) (1.00)
DI 0.0043*** 0.0040** 0.0046*** 0.0283*** 0.0293*** 0.0280***
(2.64) (2.54) (2.86) (4.85) (5.10) (4.84)
Const. 5.2235 5.2746 5.2142 20.7206** 21.0598** 20.7325**
(0.98) (0.99) (0.98) (2.49) (2.53) (2.49)
R2 0.2140 0.2123 0.2146 0.1614 0.1601 0.1610
The number of observations 7,995 7,978 7,995 7,858 7,844 7,858
Notes
1. SOV1, SOV2, and SOV3 are the outstanding values of the existing stock option, the ratio of the outstanding values of the existing stock option to the
executive compensations, and the marginally enhanced stock option values of the existing stock option as a result of a 1 % gain in stock prices, respectively
2. The random effects model is applied for the estimation
3. The amount of stock option values are calculated by the BlackScholes model
4. The estimation period is from 1997 to 2006
5. The industrial dummy variables are included as explanatory variables in all estimated models
6. The values for the last year are used for all explanatory variables except for the main bank dummy and the coincident index of business conditions, in which
the current year values are used
7. ***, **, and *represent statistical signicance at the 1, 5, and 10 % levels, respectively. The z values are in parentheses
3.2 Has the Protability of Companies Improved 25
The results show that the stock option values had no positive effects on either ROA
or ROE in both the manufacturing and non-manufacturing industries. For the model
of the manufacturing industries, where the dependent variable is ROA, the stock
option values show a statistically signicant negative coefcient.
The results are similar in the model where the ratio of the outstanding values of
the existing stock option to executive compensation and the marginally enhanced
stock option values of the existing stock option as a result of a 1 % gain in stock
prices are explanatory variables. None of the estimated results shows that these
indicators relevant to the amount of stock option values had signicantly positive
effects upon the protability measures. As in the above case, these indicators had
signicantly negative effects upon ROA.
These estimation results suggest that the introduction of a stock option does not
improve rm protability.
3.3 Did the Introduction of a Stock Option Induce

Risk-Taking Firm Behaviors?
As mentioned, those granted a stock option do not need to pay a downside risk, and
there is no limits to the upside gains; the owner of a stock option can enjoy innite
capital gains. From this point of view, several empirical analyses of US corpora-
tions found that the introduction of the stock option promoted risk-taking activities
(Lekse and Zhao, 2009).8
In this section, I examine whether the constitution of the stock option introduced
in Japanese corporations has induced corporate risk-taking behaviors.
On this matter, the selection of the risk-taking behavior measures is very
important. I examine nancial leverage, capital investment, and R&D activity (only
for the manufacturing industry). The second proposition of the MM (Modigliani
and Miller) theory posits that the higher the nancial leverage (i.e., the higher the
debt ratio), the higher the standard deviation of the stock returns.9 I analyze how a
stock option affects nancial leverage.
Capital investment and R&D might not directly indicate the degree of risk
taking. Plant and machinery investments and R&D expenditures are very important
factors in corporate performance and must be appropriately managed to foster
rms growth strategies. However, as excessive capital investments and superfluous
8
Moriyasu and Uchida (2011) empirically show that the stock option promotes risk-taking
behaviors in Japanese companies in a short-term and long-term event study of stock option
introduction.
9
See the introductory chapter of Hanazaki (2008).
R&D expenditures can also raise corporate management risk, these expenditures are
used as proxies for risk-taking activities.
The following formula is used to examine what kinds of effects stock option
value has had on rms risk-taking behaviors:
RTit const: a1 SOVit a2 SC10it a3 FIit a4 ECit a5 ROAit a6 GSALEit

a7 ASSETit a8 MAINi a9 DIt eit
3:6
RT: rms risk-taking activities (the debttotal asset ratio, the capital expenditure
capital stock ratio, and the R&D expendituretotal assets ratio)
Here, SOVs are three indicators relevant to the stock option values used in
Eq. (3.5). Table 3.4 shows the estimation results of Eq. (3.6). The coefcients of
the outstanding value of the existing stock option for the estimation results
explaining the debt ratio show negative signs at a statistically signicant level for
both the manufacturing industry and the non-manufacturing industry. This means
that the introduction of the stock option has reduced leverage signicantly. On the
other hand, in the estimation results explaining the R&D ratio of the manufacturing
industry, the outstanding value of the existing stock option shows a signicantly
positive sign. Thus, R&D activities are signicantly induced by introducing a stock
option.
Next, how the ratio of the outstanding values of the existing stock option to
executive compensation affects risk-taking activities is examined. The coefcients
show signicantly positive signs for the R&D ratio of the manufacturing industry
and the capital expenditure ratio of the non-manufacturing industry. Statistically
signicant results are not obtained in other cases. Finally, I examine the estimation
results of the marginally enhanced stock option values of the existing stock option
as a result of a 1 % gain in stock prices. The results are similar to those for the
outstanding value of the existing stock option. The marginally enhanced stock
option values have a negative impact on the debt ratio and a positive impact on the
R&D ratio of the manufacturing industry.
The overall empirical analyses on the relationship between stock options and
rms risk-taking behavior show negative effects on nancial leverage as well as
positive effects on R&D activities. However, when I regard R&D activities as a
kind of prior investment for improving growth potential and protability, it is not
clear that they reflect a kind of risk-taking behavior. Bearing these circumstances
fully in mind, I cannot conclude that the general hypothesis that the introduction of
a stock option promotes rms risk-taking activities is supported empirically.
Table 3.4 Stock option and risk-taking activities
(1) Manufacturing
Debt ratio Capital expenditure ratio R&D ratio
SOV1 0.1412*** 0.0328 0.0142***
(5.86) (0.58) (3.37)
SOV2 0.0403 0.1113 0.0105**
(1.17) (1.46) (2.02)
SOV3 0.1541*** 0.1030 0.0186***
(4.38) (1.24) (3.04)
SC10 0.0048 0.0082 0.0051 0.0736** 0.0709** 0.0738** 0.0015 0.0018 0.0015
(0.34) (0.58) (0.36) (2.49) (2.39) (2.50) (0.59) (0.71) (0.61)
FI 0.0405*** 0.0323** 0.0396*** 0.0530* 0.0539* 0.0543* 0.0015 0.0018 0.0015
(3.12) (2.50) (3.05) (1.87) (1.90) (1.92) (0.68) (0.81) (0.68)
EC 0.1933*** 0.2111*** 0.1972*** 0.1482*** 0.1490*** 0.1452*** 0.0023 0.0027 0.0025
(13.65) (15.04) (13.93) (4.53) (4.58) (4.44) (0.87) (1.03) (0.94)
LEV 0.1013*** 0.1016*** 0.1006*** 0.0084*** 0.0089*** 0.0084***
(4.84) (4.85) (4.82) (4.46) (4.76) (4.49)
ROA 0.5834*** 0.5858*** 0.5816*** 0.3177*** 0.3221*** 0.3183*** 0.0013 0.0008 0.0013
(23.67) (23.77) (23.59) (5.37) (5.43) (5.38) (0.30) (0.18) (0.28)
SALE 0.0336*** 0.0325*** 0.0330*** 0.0466*** 0.0453*** 0.0464*** 0.0004 0.0002 0.0003
(7.89) (7.64) (7.76) (4.15) (4.02) (4.13) (0.44) (0.21) (0.37)
ASSET 0.7586** 0.6214** 0.7124** 2.6314*** 2.6809*** 2.6495*** 0.1281** 0.1430*** 0.1303**
(2.48) (2.03) (2.32) (5.58) (5.68) (5.62) (2.33) (2.62) (2.38)
MAIN 11.1143*** 11.2618*** 11.2039*** 0.7026 0.8301 0.7106 0.0462 0.0545 0.0512
(10.05) (10.16) (10.12) (0.55) (0.65) (0.56) (0.25) (0.30) (0.28)
DI 0.0415*** 0.0442*** 0.0427*** 0.0263*** 0.0266*** 0.0267*** 0.0025*** 0.0028*** 0.0026***
(14.01) (15.12) (14.52) (3.61) (3.68) (3.68) (4.78) (5.24) (5.00)
27
(continued)
28
(1) Manufacturing
Const. 37.3617*** 39.4946*** 38.1480*** 59.5888*** 60.5262*** 59.8305*** 0.7857 1.0289 0.8262
(6.58) (6.96) (6.72) (6.97) (7.06) (7.00) (0.75) (0.98) (0.79)
R2 0.2472 0.2458 0.2451 0.0909 0.0921 0.0919 0.1832 0.1850 0.1818
The number of 10,296 10,283 10,296 9,508 9,495 9,508 6,813 6,801 6,813
observations
(2) Non-manufacturing
Debt Ratio Capital Expenditure Ratio
SOV1 0.2292*** 0.1500
(8.12) (1.30)
SOV2 0.0017 0.2242**
(0.08) (2.41)
SOV3 0.2386*** 0.0350
(6.12) (0.22)
SC10 0.0602*** 0.0642*** 0.0621*** 0.1344* 0.1491** 0.1292*
(3.63) (3.86) (3.74) (1.87) (2.07) (1.80)
FI 0.0946*** 0.0859*** 0.0922*** 0.0391 0.0344 0.0354
(5.59) (5.06) (5.44) (0.55) (0.49) (0.50)
EC 0.2811*** 0.3128*** 0.287302*** 0.1512* 0.1813** 0.1593*
(14.55) (15.92) (14.86) (1.82) (2.16) (1.92)
LEV 0.1384*** 0.1348*** 0.1444***
(2.83) (2.79) (2.97)
ROA 0.3165*** 0.3128*** 0.3225*** 0.2433** 0.2184* 0.2442**
(11.56) (11.34) (11.75) (2.09) (1.88) (2.09)
0.0324** 0.0321* 0.0326**
SALE 0.0051 0.0058 0.0050

(1.32) (1.51) (1.29) (2.03) (2.02) (2.04)
(continued)
(1) Manufacturing
ASSET 3.9539*** 3.8232*** 3.8529*** 9.1788*** 9.2778*** 9.0063***
(13.55) (13.08) (13.22) (7.34) (7.44) (7.24)
MAIN 9.3661*** 9.9101*** 9.7043*** 2.1858 2.5382 1.7770
(5.62) (5.96) (5.83) (0.31) (0.36) (0.25)
DI 0.0367*** 0.0430*** 0.0401*** 0.0137 0.0126 0.0102
(9.24) (10.96) (10.21) (0.84) (0.79) (0.63)
Const. 31.4480*** 32.1015***
(3.56) (3.63)
R2 0.3765 0.3816 0.3784 0.1519 0.1540 0.1489
The number of 7,847 7,829 7,847 6,808 6,791 6,808
observations
Notes
1. Dependent variables are as follows
Debt Ratio = total debts/total assets
Capital Expenditure Ratio = capital expenditures/tangible xed assets (the average of the amount of the current year and the last year)
R&D Ratio = research and development expenditures/total assets (the average of the amount of the current year and the last year)
2. SOV1, SOV2, and SOV3 are the outstanding values of the existing stock option, the ratio of the outstanding values of the existing stock option to the
executive compensations, and the marginally enhanced stock option values of the existing stock option as a result of a 1 % gain in stock prices, respectively
3. The random effects model is applied for the estimation
4. The amount of stock option values are calculated by the BlackScholes model
6. The industrial dummy variables are included as explanatory variables in all estimated models
7. The values for the last year are used for all explanatory variables except for the main bank dummy and the coincident index of business conditions, in which
the current year values are used
8. ***, **, and *represent statistical signicance at the 1, 5, and 10 % levels, respectively. The z values are in parentheses
29
References
Black, F., & Scholes, M. (1973). The pricing of options and corporate liabilities. Journal of
Political Economy, 81(3), 637654.
Cox, J. C., Ross, S. A., & Rubinstein, M. (1979). Option pricing: A simplied approach. Journal
of Financial Economics, 7(3), 229263.
Hanazaki, M. (2008). Corporate nance and corporate governance: An informational and
institutional approach. Tokyo: University of Tokyo Press. (in Japanese).
Lekse, W., & Zhao, M. (2009). Executive risk taking and equity compensation in M&A process.
In R. W. Kolb, & D. Schwartz, (Ed.), Corporate boards: Managers of risks, sources of risk.
Wiley-Blackwell.
Merton, R. C. (1973). The theory of rational option pricing. Bell Journal of Economics and
Management Science, 4(1), 141183. doi:10.2307/3003143.
Moriyasu, H., & Uchida, K. (2011). Stock option grants and managerial risk taking: Evidence
from Japanese intraday stock return data. Unpublished manuscript.
Chapter 4
Related Studies of Corporate
Diversication
Abstract Chandler, The visible hand: the managerial revolution in American

business. Belknap Press of Harvard University Press, Cambridge (1977) and Scale
and scope: the dynamics of industrial capitalism. Belknap Press of Harvard
University Press, Cambridge (1990) pointed out that an economy of scope and
corporate diversication could bring about higher protability and/or higher pro-
ductivity. However, many empirical analyses of US corporations have found that
corporate diversication has reduced rm value, in so-called diversication dis-
counts or conglomerate discounts. For example, Wernerfelt and Montgomery,
Am Econ Rev, 78(1):246250, (1988), a pioneering study on diversication dis-
counts, show that corporate performance is inferior in corporations that have
diversied broadly compared with corporations that have not. It is often pointed out
that inefcient corporate diversication is the result of weak corporate governance.
For example, top corporate managers tend to be fonder of large-scale and
wide-ranging businesses than of single and small-scale businesses. This propensity
is called empire building. As the scale and scope of a business expand, top
managers tend to dabble in investment projects in which the discounted net present
values are negative; this is what is called the problem of overinvestment.
Furthermore, in project selection, top managers have a strong tendency to choose
projects that coincide exactly with their own specic skills and experiences instead
of projects that are suitable and manageable for other managers; this is the so-called
problem of entrenching investment. To solve these negative aspects of corporate
diversication resulting from weak corporate governance, mechanisms for moni-
toring and controlling top managers behavior effectively are required.

Keywords Economy of scope Corporate diversication Diversication dis-

count Conglomerate discount Corporate performance Weak corporate gov-

ernance Empire building Overinvestment Entrenching investment

32 4 Related Studies of Corporate Diversication
4.1 Theories of Corporate Diversication
One of the most popular topics concerning the traditional theory of the rm was
whether rms can improve protability and raise rm value by diversifying their
businesses. Chandler (1977, 1990) pointed out that an economy of scope is iden-
tied when prots and/or cost efciencies are projected in such cases as when the
know-how and/or information of a certain business is applicable to another or when
several frontline divisions can use a common administrative division; in such a
case, corporate diversication will increase protability and/or productivity.
The effects of corporate diversication have often been discussed in the eld of
corporate nance. Lewellen (1971) argues that, due to the conglomeracy of cor-
porations via amalgamation, rms can diversify risk and use more debt and thus
reduce their tax payments. Diamond (1984) analyzed non-nancial companies
multilateralization using a theoretical model and found that there were two kinds of
multilateralization: one that decreases risks for nancial institutions and another that
amplies risks for them. The theoretical model in Stein (1997) shows that efcient
management is realized when the head ofces of diversied companies distribute
nancial resources effectively through their internal capital market.
4.2 Negative Effects of Corporate Diversication
Many empirical analyses of US corporations have found that corporate diversi-

cation has reduced rm value, in so-called diversication discounts or con-
glomerate discounts. Wernerfelt and Montgomery (1988), a pioneering study on
diversication discounts, empirically examine what kinds of elements contributed
to performance changes in corporations. Their performance measure is Tobins q in
1976. Their result for multilateralization shows that corporate performance was
inferior in corporations that diversied broadly compared to the corporations that
did not.
Many studies explore the background of the negative aspects of corporate
diversication. Berger and Ofek (1995), Rajan et al. (2000) study the inefciency
arising from the cross-subsidization between segments. Jensen (1986) investigates
the harmful effects of inefcient and unprotable investment projects, and
Scharfstein and Stein (2000) nd that distortion arose in the internal capital market
because the directors of each segment engaged in rent-seeking behaviors for private
benet.
Various empirical studies on US rms have found that corporations with
diversied businesses tend to have inferior rm value or operational efciency
compared with corporations that run a single business. This inferiority appears to be
caused by factors such as over-investment, the cross-subsidization from protable
segments to unprotable segments, and the inefciency of the internal capital
4.2 Negative Effects of Corporate Diversication 33
market. This suggests that capital expenditures as a result of corporate diversi-

cation tend to be too expansionary and inefcient.1
For the East Asian region, which experienced a serious currency crisis in 1997,
several empirical analyses have shown that corporate diversication has caused
inefciency and has had negative effects upon corporate performance. Among
these, Hanazaki and Liu (2003) examine corporate performance indicators such as
ROA and ROE from 1994 to 2000 for Indonesia, South Korea, Malaysia, the
Philippines, and Thailand. They hypothesize that the performance of corporations
that ran more businesses tended to be worse during the currency crisis. Their
empirical ndings support the hypothesis.2
4.3 Reexamination of the Diversication Discount
Many studies point out the negative aspects of corporate diversication; however, it
is unclear whether diversication is the cause and reduced performance is the effect.
It is also unclear why shareholders accept corporate diversication if it causes
negative impacts upon corporate performance. Investors, who are shareholders, can
lower their risks by diversifying their portfolio. Therefore, shareholders would
reject corporate diversication that does not improve rm value.
The notion that the matter of the diversication discount should be reexamined
emerged in the 2000s. Campa and Kedia (2002) argue that corporate diversication
is not necessarily the cause of the reduced value of diversied rms. Rather, they
argue that rm values fall as a result of the self-selection of the diversied rms.3
According to their empirical study, the diversication discount is substantially
reduced, and a premium may occur after controlling for the endogenous nature of
the issue and the sample selection bias.
Graham et al. (2002) examined corporate diversication through mergers and
acquisitions (M&A) activities and investigated why such M&A would reduce rm
value. They found that many acquiring enterprises had a strong tendency to
1
Many empirical studies show the diversication discount. These include Berger and Ofek (1995),
Lang and Stulz (1994), Comment and Jarrell (1995), Servaes (1996), Denis et al. (1997), Shin and
Stulz (1998), Denis et al. (2002), Mansi and Reeb (2002).
2
The number of business segments is used as the index of corporate diversication.
3
The concrete mechanisms are as follows. First, the companies in nancial distress due to delayed
technological progress or other problems are going to advance diversication to other business
areas, since the opportunity cost for their corporate diversication is presumably low. In that case,
although low corporate value and diversication have occurred simultaneously, the corporate
value did not necessarily fall as a result of diversication. Rather, worsening performance is
connected with the low corporate value. Second, when companies with rm-specic organizational
abilities are searching for business segments suitable for their ability, their corporate values may be
lower than those of single-segment rms. However, the diversications themselves are not the
cause of the low values.
purchase businesses with low rm values and that the corporate diversications
themselves did not necessarily reduce the values.
Moreover, Villalonga (2004) insists that the diversication discount appears
because of the inaccurate nature of the segment data in COMPUSTAT, which are
widely used in the literature. Using company data taken from Business Information
Tracking Series (BITS), a new census database for US rms, instead of
COMPUSTAT, Villalonga (2004) re-estimated the same equation for the same
sample rms used in Lang and Stulz (1994), a typical study in the eld, and found
that the diversied rms generated premiums instead of discounts compared with
the single-business rms.
4.4 Corporate Governance and Corporate Diversication
It is often pointed out that inefcient corporate diversication appears as one side of
the corporate governance issue for modern corporations where ownership and
control are separated. Generally, top corporate managers tend to be fonder of
large-scale and wide-ranging businesses than of single and small-scale ones. This
propensity is called empire building. Moreover, as the scale and scope of the
businesses are expanding, top managers tend to dabble in investment projects in
which the discounted net present values are negative; this is the problem of
overinvestment. Furthermore, in project selection, top managers have a strong
tendency to choose projects that coincide exactly with their own specic skills and
experiences instead of projects that are suitable and manageable for other managers;
this is the so-called problem of entrenching investment.
To solve these negative aspects of corporate diversication resulting from weak
corporate governance, mechanisms for monitoring and controlling top managers
behavior effectively are required. In this regard, rst, major shareholders, who have
powerful controlling rights and a strong interest in rm values, should play an
important role in monitoring the top managers. Second, to mitigate the so-called
free cash flow problem, Jensen (1986) argues that it is most effective to issue
debts for additional nancing or to repurchase the companies own stock instead of
equity nancing.
In this section, I propose hypotheses focusing on the relationship between cor-
porate governance and corporate diversication and empirically test them for
Japanese rms.4
4
Empirical analyses of Japanese corporate diversication behaviors include Yoshiwara et al.
(1981), Shimizu and Miyagawa (2003), Morikawa (1998), Miyajima and Inagaki (2003), Ito et al.
(2003), Kikutani et al. (2007). Among them, few focus on the issue of corporate governance.
References 35
References
Berger, P. G., & Ofek, E. (1995). Diversications effect on rm value. Journal of Financial
Economics, 37(1), 3965. doi:10.1016/0304-405X(94)00798-6.
Campa, J. M., & Kedia, S. (2002). Explaining the diversication discount. Journal of Finance, 57
(4), 17311762. doi:10.1111/1540-6261.00476.
Chandler, A. D, Jr. (1977). The visible hand: The managerial revolution in American business.
Cambridge, Mass: Belknap Press of Harvard University Press.
Chandler, A. D, Jr. (1990). Scale and scope: The dynamics of industrial capitalism. Cambridge,
Mass: Belknap Press of Harvard University Press.
Comment, R., & Jarrell, G. A. (1995). Corporate focus and stock returns. Journal of Financial
Economics, 37(1), 6787. doi:10.1016/0304-405X(94)00777-X.
Denis, D. J., Denis, D. K., & Sarin, A. (1997). Agency problems, equity ownership, and corporate
diversication. Journal of Finance, 52(1), 135160. doi:10.1111/j.1540-6261.1997.tb03811.x.
Denis, D. J., Denis, D. K., & Yost, K. (2002). Global diversication, industrial diversication, and
rm value. Journal of Finance, 57(5), 19511979. doi:10.1111/0022-1082.00485.
Diamond, D. W. (1984). Financial intermediation and delegated monitoring. Review of Economic
Studies, 51(3), 393414. doi:10.2307/2297430.
Graham, J. R., Lemmon, M. L., & Wolf, J. G. (2002). Does corporate diversication destroy
value? Journal of Finance, 57, 695720. doi:10.1111/1540-6261.00439.
Hanazaki, M., & Liu, Q. (2003). The Asian crisis and corporate governance. In M. Hanazaki &
J. Teranishi (Eds.), An economic analysis of corporate governance: Japan and East Asia in a
changing environment (pp. 339368). Tokyo: University of Tokyo Press. (in Japanese).
Itoh, H., Kikutani, T., & Hayashida, O. (2003). Many-sided relationship between the parent
companies and their subsidiaries. In M. Hanazaki & J. Teranishi (Eds.), An economic analysis
of corporate governance: Japan and East Asia in a changing environment (pp. 5280). Tokyo:
University of Tokyo Press. (in Japanese).
Jensen, M. C. (1986). Agency costs of free cash flow, corporate nance and takeovers. American
Economic Review, 76(2), 323329.
Kikutani, T., Itoh, H., & Hayashida, O. (2007). Business portfolio restructuring of Japanese rms
in the 1990s: Entry and exit analysis. In M. Aoki, G. Jackson, & H. Miyajima (Eds.),
Corporate governance in Japan: institutional change and organizational diversity (pp. 227
256). New York: Oxford University Press.
Lang, L. H. P., & Stulz, R. M. (1994). Tobins q, corporate diversication, and rm performance.
Journal of Political Economy, 102(6), 12481280.
Lewellen, W. G. (1971). A pure nancial rationale for the conglomerate merger. Journal of
Finance, 26, 521537. doi:10.1111/j.1540-6261.1971.tb00912.x.
Mansi, S. A., & Reeb, D. M. (2002). Corporate diversication: what gets discounted? Journal of
Finance, 57(5), 21672183. doi:10.1111/0022-1082.00492.
Miyajima, H., & Inagaki, K. (2003). Diversied corporate behaviors and corporate governance.
Policy Research Institute, Ministry of Finance. (in Japanese).
Morikawa, M. (1998). Empirical analysis of new entry to the new business and withdrawal from
the current business. RIETI Discussion Paper Series, 98-DOJ-87. (in Japanese).
Rajan, R., Servaes, H., & Zingales, L. (2000). The cost of diversity: the diversication discount
and inefcient investment. Journal of Finance, 55(1), 3580. doi:10.1111/0022-1082.00200.
Scharfstein, D. S., & Stein, J. C. (2000). The dark side of internal capital markets: Divisional
rent-seeking and inefcient investment. Journal of Finance, 55(6), 25372564. doi:10.1111/
0022-1082.00299.
Servaes, H. (1996). The value of diversication during the conglomerate merger wave. Journal of
Finance, 51(4), 12011225. doi:10.1111/j.1540-6261.1996.tb04067.x.
Shimizu, M., & Miyagawa, K. (2003). New entry, withdrawal, and corporate diversication: the
economic analysis based on the industrial statistic data. Tokyo: Keio University Press. (in
Japanese).
Shin, H. H., & Stulz, R. M. (1998). Are internal capital markets efcient? Quarterly Journal of
Economics, 113(2), 531552.
Stein, J. C. (1997). Internal capital markets and the competition for corporate resources. Journal of
Finance, 52(1), 111133. doi:10.1111/j.1540-6261.1997.tb03810.x.
Villalonga, B. (2004). Diversication discount or premium? New evidence from the business
information tracking series. Journal of Finance, 59(2), 479506. doi:10.1111/j.1540-6261.
2004.00640.x.
Wernerfelt, B., & Montgomery, C. A. (1988). Tobins q and the importance of focus in rm
performance. American Economic Review, 78(1), 246250.
Yoshiwara, H., Sakuma, A., Itami, H., & Kagono, T. (1981). Diversication strategy of Japanese
corporations. Tokyo: Nihon Keizai Shimbun. (in Japanese).
Chapter 5
Corporate Diversication of Japanese
Firms
Abstract Many studies on US rms use information on business segments as the

indicator of corporate diversication. However, this presents problems given the
segmental reporting disclosure system in Japan because around half of corporations
do not disclose segmental information. Therefore, I mainly use an analytical
approach based upon a comparison between non-consolidated nancial statements
and consolidated nancial statements, as parent companies promote corporate
diversication through their consolidated subsidiaries. Comparing among Japanese
companies shows that the ratios of consolidated sales or total assets to the parent
companies sales or total assets are relatively small and that Japanese rms have
many consolidated subsidiaries. These facts would suggest that the division of
corporations often includes the establishment of subsidiaries, which operate the
service and supporting businesses for their parent companies, and may suggest that
corporate diversication and division of corporations in Japanese rms is quite
different from empire building or overinvestment.

Keywords Corporate diversication Business segment Division of corpora-

tion Consolidated nancial statement Non-consolidated nancial statement

Parent company Consolidated subsidiary
5.1 How Can I Determine the Actual Conditions

of Corporate Diversication?
Determining the actual condition of corporate diversication is the most essential

aspect of this study. In many studies of US rms, information on business segments
is used as the indicator of corporate diversication. Business segments are the
components of corporations in which corporate resources are distributed by the
managers, whose sales, costs, and prots are registered and from which their

38 5 Corporate Diversication of Japanese Firms
nancial data and information are available.1 In Japan, the disclosure of segmental
reporting has been required since September 1988. The coverage of the disclosed
information for segments was later expanded and became a target of auditing in
March 1993.
However, there are still problems with the segmental reporting disclosure system
in Japan. Tsumuraya (2010) argues that the current segmental reporting disclosure
standards are ambiguous and arbitrary and may be used for the concealment of
unprotable segments. He also points out that nearly half of all listed rms do not
disclose segmental information.
I examined segmental reporting disclosure based on the nancial statements of
Japanese listed rms and found that around half of corporations are not disclosing
segmental information and that, even where segmental information is available, the
content varies from detailed to vague. I concluded that the current segmental
information is not reliable for empirical analysis.2 In this book, the issue of cor-
porate diversication is approached from another point of viewthe viewpoint of
corporate diversication via subsidiaries. Corporations sometimes acquire rms
and/or establish new rms to enter into new businesses.
As an example of corporate diversication through subsidiaries, Fig. 5.1 shows
segmental information for Toshiba Corp. and Hitachi, Ltd. While parent company
Toshiba has ve segments (energy and infrastructure, community solution,
healthcare system and services, electronic devices and components, and lifestyle
products and services), it has 590 consolidated subsidiaries among the ve seg-
ments, such as digital products, electronic devices, energy and infrastructure, and
home appliances. Hitachi has more varieties of segments based on consolidated
subsidiaries. The parent company runs three segments (information and telecom-
munication systems, power systems, and social infrastructure and industrial sys-
tems). Its 963 consolidated subsidiaries are involved in not only the same segments
as the parent companys but also seven other segments (such as electronic systems
and equipment, construction machinery, high functional materials and components,
automotive systems, smart life and ecofriendly systems, nancial services). These
subsidiaries are utilized for two purposes: splitting up the businesses of the parent
company and launching into businesses entirely different from that of the parent
company.
The corporate behaviors by which parent companies promote corporate diver-
sication through their consolidated subsidiaries can be grasped by analyzing the
differences between non-consolidated nancial statements and consolidated nan-
cial statements. The non-consolidated nancial statements for the parent companies
offer no nancial information on their subsidiaries, while the consolidated nancial
statements are the unied nancial statements for the parent companies and their
1
For details, see the Corporate Accounting Standards Committees Accounting Standards for the
Disclosure of Segment Information (last revised on March 27, 2009).
2
In Japan, there seems to be a kind of sample selection bias in terms of the disclosure of segmental
information because the more detailed segmental information disclosures may be those of com-
panies that achieved successful corporate diversication.
Toshiba Corp.
usiness Segments of a Parent Company
1 Energy & Infrastructure Segment

2 Community Solution Segment
3 Healthcare Systems & Services Segment
4 Electronic Devices & Components Segment
5 Lifestyle Products & Services Segment
ain Consolidated Subsidiaries of Each Business Segment Toshiba
Digital Products Electronic Devices Energy & Infrastructure Home Appliances Others
Toshiba TEC Iwate Toshiba Electronics Toshiba Nuclear Energy Holdings (US Toshiba Consumer Marketing Toshiba America
oshiba Global Commerce Solution NuFlare Technology Toshiba Nuclear Energy Holdings (UK Toshiba Home Appliances Taiwan Toshiba International Procuremen
oshiba America Business Solution Toshiba America Electronic Components Toshiba JSW Power Systems Private Toshiba Carrier Toshiba Trading
Dalian Toshiba Television Kaga Toshiba Electronics Toshiba Solutions Toshiba Lighting & Technology Toshiba Logistics
he number of consolidated subsidiaries : 590
Hitachi, Ltd.
usiness Segments of a Parent Company
1 Information & Telecommunication Systems

2 Power Systems
3 Social Infrastructure & Industrial Systems
ain Consolidated Subsidiaries of Each Business Segment

Hitachi
Information & Telecommunication Systems Power Systems Social Infrastructure & Industrial Systems Electronic Systems & Equipment Construction Machinery High Functional Materials & Components Automotive Systems Smart Life & Ecofriendly Systems Financial Services Others
tachi Information & Telecommunication
Horizon Nuclear Power Hitachi Industrial Equipment Systems Hitachi High-Technologies Corporation Hitachi Construction Machinery Hitachi Chemical Clarion Hitachi Appliances Hitachi Capital Corporation Hitachi America
ystems Global Holding Corporation
Hitachi Data Systems Hitachi Power Europe Hitachi Industry and Control Solutions Hitachi Koki Hitachi Metals Hitachi Automotive Systems Hitachi Consumer Products(Thailand) Hitachi (China)
Hitachi Solutions Hitachi-GE Nuclear Energy Hitachi Building Systems Hitachi Kokusai Electric Hitachi Automotive Systems America Hitachi Consumer Marketing Hitachi Europe
Hitachi Systems Hitachi Power Solutions Hitachi Elevator (China) Hitachi Medical Corporation Hitachi Transport System
he number of consolidated subsidiaries : 963
Fig. 5.1 The diversication of Toshiba and Hitachi. Source Financial Statements of Each Company
5.1 How Can I Determine the Actual Conditions of Corporate Diversication?
39
consolidated subsidiaries. Comparing the two claries the corporate diversication

behaviors through the consolidated subsidiaries.3
There are, however, limitations to trying to clarify corporate diversication
behaviors through the parentsubsidiary relationship. First, corporate activities in
which the subsidiaries take charge of a part of the main business or the existing
business do not constitute corporate diversication but the division of corporations.
Second, the purpose of establishing subsidiaries is not only corporate diversication
or division of corporations but also activities such as IT-related system development
and the maintenance used at the parent companies, the maintenance of their own
building, and the operation of company cafeterias and welfare facilities. These are
regarded as service businesses for parent companies and are quite different from
corporate diversication. However, it is very difcult to distinguish these two cases
from pure corporate diversication cases in the current conditions, with the
insufcient and imperfect segmental reporting disclosure system.
Despite these limitations, an analytical approach based upon a comparison
between non-consolidated nancial statements and consolidated nancial state-
ments make senses because, rst, subsidiaries that take charge of service businesses
for parent companies are not reflected in the consolidated nancial statements, and,
second, regarding the division of corporations, consolidated subsidiaries abroad are
established for overseas production. This trend includes corporate diversication
because these overseas subsidiaries often mainly produce differentiated products
suitable for local needs.
In addition, the current disclosure system for segmental information is unsatis-
factory in the sense that any analysis based upon the information is accompanied by
the serious problem of sample selection bias. An analysis based upon a comparison
between non-consolidated nancial statements and consolidated nancial state-
ments is helped by the fact that the information of all corporations with consoli-
dated subsidiaries is disclosed and available.
For these reasons, corporate diversication behaviors are mainly analyzed in this
book based on a comparison between the nancial information in consolidated
statements and that in non-consolidated statements. The development of corporate
diversication in Japan discussed in Sect. 5.2 and the empirical analyses in
Sects. 6.2 and 6.3 are based upon this information.
Additionally, I also perform two other analyses. One is based upon the segmental
information discussed in Sect. 6.4. Despite the problems with and limitations of the
information, it is interesting to examine the results and compare them with those in
Sect. 6.2. The other is an analysis of pure holding companies. The characteristics of
pure holding companies, which do not undertake any business and are involved in only
equity holdings, are quite different from those of other rms. Therefore, pure holding
3
In Japan, the non-consolidated nancial statements have traditionally been thought important. In
April 1977, however, the consolidated nancial statement was introduced, and the consolidated
nancial statement began to be included in nancial reports in April 1991. Finally, beginning in
the accounting period starting in April 1999, consolidated statements became the main part of the
disclosed accounting information.
5.1 How Can I Determine the Actual Conditions of Corporate Diversication? 41
companies are completely excluded from the empirical analyses in Sects. 6.26.4.
The prohibition on pure holding companies was removed in 1997 in Japan, and the
appearance of pure holding companies is steadily increasing. That is why I focus upon
and examine pure holding companies in Sect. 6.5.
5.2 Corporate Diversication and Division

of Corporations in Japanese Firms
The data I use are nancial and corporate data for Japanese listed companies,
including companies for emerging stock markets, recorded on the Financial
Databank for Listed Companies compiled by the Development Bank of Japan.
Financial sector rms are excluded. The data period covers the 20 years from the
1990 scal year, when the consolidated accounting system was introduced. Pure
holding companies, which began to operate in 1997 because of deregulation, and
companies close to pure holding companies are excluded from the data sample.4
In this section, I attempt to grasp the developments of corporate diversication
and the division of corporations via the barometer of the number of consolidated
subsidiaries and the ratios of consolidated sales or total assets to the parent com-
panies sales or total assets. It is necessary to pay attention to cases featuring great
divergences between the number of consolidated subsidiaries and the ratio of
consolidated sales or total assets to the parent companies sales or total assets,
which would indicate that the wholly owned subsidiaries are concentrating upon the
businesses and services for the patent companies. For example, suppose cleanup
services for the head ofce building previously performed internally were provided
by a newly established wholly owned subsidiary. In that case, although the number
of consolidated subsidiaries increases by one company, the ratio of consolidated
sales or total assets to the parent companies sales or total assets would not change,
as the new wholly owned subsidiary provides services only for the parent company.
Figure 5.2 shows the distribution of the number of consolidated subsidiaries
over time. In 1990, the rst year, companies without consolidated subsidiaries total
43 % of all listed companies, and the percentages of companies with one, two, and
three consolidated subsidiaries were 9, 9, and 6 %, respectively. As the years pass,
the percentage of the corporations with no consolidated subsidiaries falls gradually,
while the number of companies with many consolidated subsidiaries increases. In
4
Holding companies are companies in which the value of the shares of consolidated subsidiaries
out of total assets exceeds 50 %. Companies that do not undertake any kind of businesses except
for the equity holding are called pure holding companies, and those involved in business are
called business holding companies. In 1997, the Anti-Monopoly Law was amended, and the
system of pure holding companies was institutionalized. In this book, we consider that the business
holding companies whose ratios of consolidated sales to the parent companies sales exceed 10 are
quite similar to pure holding companies; hence, they were excluded from the data sample in the
empirical analysis.
1990 100 and over 1995

50-99 100 and over
50-99
10-49
10-49 0
0
5-9
4
5-9
1
3
4 2
2 1 3
The number of firms : 2,345 The number of firms : 2,899
The average number of subsidiaries : 6.7 The average number of subsidiaries : 10.8
2000 2005
100 and over
50-99 100 and over 50-99
0 0
10-49
10-49
1 1
2 2
5-9 5-9
3 3
4 4
2010 100 and over

2011 100 and over
50-99
0 50-99
0
10-49 10-49
1
1
2 2
5-9 3
5-9 3
4 4
Fig. 5.2 The number of consolidated subsidiaries
2011, while only 18 % of corporations have no consolidated subsidiaries, 24 %

have from 10 to 49.
Firms have an average of 6.7 consolidated subsidiaries in 1990. In 1995, the
gure exceeds 10 companies for the rst time, and the gure reaches 14.8 com-
panies in 2011.
Next, the ratios of consolidated sales or total assets to the parent companies
sales or total assets are examined in order to determine the situation of corporate
diversication and the division of corporations in Japanese rms. First, Fig. 5.3
shows the development over time of the ratio of consolidated sales to the parent
companies sales. The number of companies whose ratio is equal to unity, meaning
that no consolidated subsidiaries are held by the parent companies, declined sub-
stantially from 50.2 % in 1990 to 31.3 % in 2011. By contrast, the number of
relatively high-ratio companies gradually increased. The average ratio among all
data samples was 1.1 in 1990 and 1.4 in recent years. Second, the ratio of con-
solidated total assets to the parent companies total assets is shown in Fig. 5.4.
5.2 Corporate Diversication and Division of Corporations in Japanese Firms 43
100%
90%
80%
70%
1
60%
1.0 1.1
1.1 1.2
50% 1.2 1.4
1.4 1.6
1.6 2.0
40% 2.0 3.0
over 3.0
30%
20%
10%
0%
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Fig. 5.3 The development of the ratio of consolidated sales to the parent companies sales
Although the ratio has been rising gradually, the average ratio among all data
samples is within a range from 1.10 to 1.16. This is fairly low compared to the ratio
of consolidated sales to the parent companies sales.
Comparing the two indicates that the ratios of consolidated sales or total assets to
the parent companies sales or total assets are relatively small and that Japanese
rms have many consolidated subsidiaries; this would suggest that the division of
corporations often includes the establishment of subsidiaries that operate services
and supporting businesses for their parent companies. This situation might suggest
that corporate diversication and division of corporations in Japanese rms is quite
different from empire building or overinvestment. The empirical analysis in the
following section is developed based on these results.
44
100%
90%
80%
70%
1
1.0 1.1
60% 1.1 1.2
1.2 1.3
50% 1.3 1.4
1.4 1.6
1.6 1.8
40%
1.8 2.0
2.0 3.0
30%
over 3.0
20%
10%
0%
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Fig. 5.4 The development of the ratio of consolidated total assets to the patent companies total assets
5 Corporate Diversication of Japanese Firms
Reference 45
Reference
Tsumuraya, S. (2010). The survey of disclosure system in Japan from the point of views of the
segmental information. Pronexus Financial Disclosure Institute Report, 4, 312. (in Japanese).
Chapter 6
Empirical Analyses of Corporate
Diversication and Division
of Corporations
Abstract I formulate several hypotheses on the relationship among rm charac-

teristics for corporate governance, the degree of corporate diversication and
division of corporations, and corporate performances and test them empirically.
There are some cases in which the protability of the consolidated nancial
statements bases are better than that of the non-consolidated nancial statements
bases. However, this result does not necessarily mean that corporate diversication
and division of corporations themselves improved earnings because the accounting
gures of the non-consolidated nancial statements are part of the outcomes of
corporations with consolidated subsidiaries. On the other hand, I argue that the
impacts of corporate diversication and division of corporations are better grasped
through a comparative analysis between rms with consolidated subsidiaries and
rms without. I nd that the protability of the rms with consolidated subsidiaries
is signicantly lower than is that of the rms without. I also nd that the negative
impacts of corporate diversication and division of corporations are not dependent
upon the degree of corporate diversication and division of corporations and are
quite similar between the moderately diversied and aggressively diversied rms.
This would suggest that weak corporate governance is not the key problem. The
main propose of Japanese corporate diversication and division of corporations
might not be prot maximization but rather rm development, ensuring market
share, welfare improvement for rm employees, or job preservation.

Keywords Corporate diversication Division of corporation Corporate per-

formance Concentrated corporate ownership Foreign ownership Free cash

flow Consolidated nancial statement Non-consolidated nancial statement

Consolidated subsidiary Segment information Pure holding company

48 6 Empirical Analyses of Corporate Diversication
6.1 Hypotheses
In this section, I focus on the causal relationship between corporate governance and
the behaviors relating to corporate diversication and division of corporations. As
mentioned, theoretically, corporate diversication produces several kinds of bene-
ts, such as economies of scope and risk diversication. If these are realized, they
positively affect corporate value.
However, amid serious agency problems between shareholders and top man-
agers, managers may promote excessive corporate diversication and division of
corporations, wasting corporate resources. In that case, corporate performance will
suffer. Therefore, in order to avoid or mitigate such agency problems, mechanisms
by which shareholders can monitor top managers are needed. Examining how such
mechanisms can be built effectively is the starting point of the corporate governance
approach.
In this section, I focus on the typical mechanisms of corporate governance and
develop hypotheses about the relationship among rm characteristics, degree of
corporate diversication and division of corporations, and corporate performance.
Finally, these hypotheses are examined empirically.
The rst hypothesis is related to large shareholders corporate governance. Many
empirical studies on Japanese rms have discussed the positive roles played by
large shareholder governance. For example, Kaplan and Minton (1994), Kang and
Shivdasani (1995) show that top manager turnover takes place more frequently in
corporations with large shareholders than in corporations without when rm per-
formance worsens. Moreover, Yafeh and Yosha (2003) argue that large share-
holders play an important role in controlling corporate expenditures due to the
arbitrary nature of top managers, such as advertising expenditures, R&D expen-
ditures, and entertainment expenses. Prowse (1992), Berglof and Perotti (1994) also
point out that the ownership structure of Japanese rms is not concentrated com-
pared with the countries of Continental Europe and East Asia. However, they also
argue that corporate management is monitored effectively in Japan, since the large
shareholders are banks in many rms. This research suggests that high ownership
concentration may prevent inefcient and excessive corporate diversication.
Therefore, the following hypothesis on the relationship between ownership con-
centration and corporate diversication is formulated:
Hypothesis 1 More concentrated corporate ownership leads to a more moderate
and efcient degree of corporate diversication and division of corporations. This
corporate diversication and division of corporations improve corporate
performance.
Some in Japan argue that the corporate governance system should be converted
from the traditional Japanese system into the Anglo-Saxon system, which empha-
sizes the importance of shareholder values. The general view is that the corporate
governance system of foreign developed countries such as the US is better than that
of Japan and that shareholders in those foreign countries have better monitoring
ability over corporate management than do Japanese shareholders. If such views are
6.1 Hypotheses 49
accurate, high foreign ownership may prevent inefcient and excessive corporate
diversication and exert positive effects on the operational efciency of corpora-
tions. A hypothesis based on such a view is thus proposed:
Hypothesis 2 Higher foreign ownership leads to a more moderate and efcient
degree of corporate diversication and division of corporations. This corporate
diversication and division of corporations improves corporate performance.
Jensen (1986) argues that debt contracts exert disciplinary effects on corporate
management. Thus, in corporations with plenty of free cash flows, these free cash
flows tend to be used inefciently and discretionarily for top managers private
benets. In order to avoid such a problem, efcient operation is realized by making
the corporations concerned issue a certain amount of debt. Concerning this free cash
flow problem, the following hypothesis is proposed:
Hypothesis 3 For corporations with plenty of free cash flows, higher debt ratios
lead to more disciplinary effects on corporate management. As a result, the degree
of corporate diversication and division of corporations becomes more moderate
and efcient. Therefore, corporate diversication and division of corporations
improve corporate performance.
When the agency problem between shareholders and top managers is not
resolved and when top managers pursue corporate management for their
self-interest, then corporate diversication and division of corporations induce
negative influences upon corporate value. The following is thus proposed:
Hypothesis 4 Corporate diversication and the division of corporations lead to
overinvestment and worsen corporate protability.

Consolidated Financial Statements
and Non-consolidated Financial Statements
for Corporations Which Have Conducted Corporate
Diversication and Division of Corporations
In this section, the above hypotheses concerning corporate diversication and

division of corporations are examined. The focus of our attention is the differences
between consolidated nancial statements and non-consolidated nancial state-
ments for corporations which have conducted corporate diversication and division
of corporations. I use panel data from the 1990 scal year to the 2009 scal year
created based on the Development Bank of Japans Corporate Financial Data Bank.
The estimation method is the xed effects model for panel data analysis. As proxy
variables for the degree of corporate diversication and division of corporations,
three indicators are used for all hypothesesthe ratios of consolidated sales to the
parent companies sales, the ratios of consolidated total assets to the parent
companies total assets, and the number of consolidated subsidiaries. To examine

whether corporate diversication and division of corporations demonstrate positive
effects upon corporate performance, I use two protability indexes, ROA and ROE,
and assess the effects by taking the differences of the gures, one calculated from
the consolidated nancial statements and another calculated from the
non-consolidated nancial statements.
The estimation results for each hypothesis are presented below in order to
examine the hypothetical validities. The proxy variable showing the degree of
ownership concentration for Hypothesis 1 is the top 10 shareholders shareholding
ratio (SC10), the key variable for Hypothesis 2 is the foreign ownership ratio (EC),
and the key variable for Hypothesis 3 is the leverage ratio (LEV).
In the rst stage, the equation was estimated, whose dependent variables are the
ratios of consolidated sales to the parent companies sales, the ratios of consolidated
total assets to the parent companies total assets, and the number of consolidated sub-
sidiaries, and whose main explanatory variables are the top 10 shareholders share-
holding ratio, the foreign ownership ratio, and the leverage ratio (LEV), and whose
controlling variables are rm size, protability, and the time trend. In the second stage,
equations were estimated. The explained variables (PA) are the differences of ROA or
ROE calculated from the consolidated nancial statements and non-consolidated
nancial statements. The main explanatory variables are the estimated values for the
three proxies for corporate diversication and division of corporations used in the rst
stageratios of consolidated sales to the parent companies sales, ratios of consolidated
total assets to the parent companies total assets, and the number of consolidated sub-
sidiaries. The controlling variables are proxies for rm size, rm growth, and rm
soundness together with a variable for macroeconomic conditions.
The basic equation for the rst stage is as follows:
DIVit const: a1 SC10it a2 ECit a3 LEVit a4 ASSETit a5 ROAit a6 TTit eit

6:1
where
DIV three measures of the degree of corporate diversication and division of
corporations, namely the ratios of consolidated sales to the parent
companies sales (DIV1), the ratios of consolidated total assets to the
parent companies total assets (DIV2), and the number of consolidated
subsidiaries (DIV3)
SC10 top 10 shareholders shareholding ratio
EC foreigners ownership ratio
LEV debt to total assets ratio
ASSET total assets (a natural logarithm)
ROA ROA (consolidated accounting base)
TT time trend
i rm
t year
6.2 The Empirical Analysis: A Comparison Between Consolidated Financial 51
Second, the basic equation for the second stage is as follows:
PAit const: a1 EDIVit a2 GSALEit a3 EMPit a4 CRit a5 CIt eit 6:2
where
PA performance variables for corporate diversication and division of
corporations (the differences of ROA or ROE calculated from the
consolidated nancial statements and non-consolidated nancial
statements)
EDIV the degree of corporate diversication and division of corporations (the
estimated values for the three kinds of proxies for corporate diversication
and division of corporations at the rst stage equation, namely the
estimated ratios of consolidated sales to the parent companies sales
(EDIV1), the estimated ratios of consolidated total assets to the parent
companies total assets (EDIV2), and the estimated number of consoli-
dated subsidiaries (EDIV3)
GSALE sales growth rates
EMP the number of employees (a natural logarithm)
CR current ratio (liquidity assets/short-term liabilities)
CI a composite index of economic indicators (a coincidence index)
Figures for the consolidated accounting base are used as nancial data for the
corporations, except for the ownership structure gures, for which only
non-consolidated accounting bases are available. The basic statistics for the above
nancial variables are shown in Appendix 6.
The estimation results are presented in Table 6.1.1 The rst-stage regressions
show that the higher the top 10 shareholders shareholding ratio and the higher the
debt ratio, the lower the ratios of consolidated sales to parent companies sales.
These results support the hypotheses. Moreover, the higher the foreign ownership
and the higher the debt ratio, the lower the ratios of consolidated total assets to the
parent companies total assets. These results also support the hypotheses. By
contrast, the higher the top 10 shareholders shareholding ratio, the higher the ratios
of consolidated total assets to parent companies total assets, and the higher the
foreign ownership, the higher the debt ratio and the greater the consolidated sub-
sidiaries. These results apparently reject the hypotheses. Therefore, the estimation
results for the rst-stage regressions are fairly mixed.
Next, the second-stage regressions examine the effects of the estimated values of
the three proxies for corporate diversication and division of corporations upon
protability. The estimated number of consolidated subsidiaries has statistically
signicant positive effects upon improvements in ROA and ROE. The estimated
ratio of consolidated total assets to the parent companies total assets also has a
1
In this table, only the results for the xed effects model are shown because the results for the
random effects model are similar to those. This also applies to Tables 6.2, 6.3, and 6.4.
Table 6.1 Corporate governance and corporate diversication

First stage: the effects of corporate governance on corporate diversication
DIV1 DIV2 DIV3
SC10 0.0008** 0.0003* 0.0032
(2.28) (1.68) (0.22)
EC 0.0006 0.0008*** 0.3458***
(1.58) (4.34) (21.76)
LEV 0.0011*** 0.0010*** 0.0567***
(4.79) (9.11) (5.82)
ASSET 0.2256*** 0.1966*** 8.6417***
(30.38) (54.00) (27.39)
ROA 0.0017*** 0.00002 0.0495***
(3.95) (0.11) (2.76)
TT 0.0183*** 0.0036*** 0.8587***
(41.56) (16.57) (45.82)
Const. 2.8250*** 2.3241*** 152.3429***
(20.81) (34.93) (26.42)
R2 0.0067 0.0572 0.2279
The number of observations 40,439 41,143 41,143
Notes
1. The xed effects model is applied for the estimation
3. ***, **, and * represent statistical signicance at the 1, 5, and 10 % levels, respectively. The t
values are in parentheses
Second stage: the degree of corporate diversication and corporate performance
ROA ROE
EDIV1 0.0068 0.9694
(0.02) (1.12)
EDIV2 1.9155*** 2.2209
(3.47) (1.26)
EDIV3 0.0196*** 0.0747***
(3.62) (4.41)
GSALE 0.0579*** 0.0584*** 0.0578*** 0.1153*** 0.1151*** 0.1132***
(35.84) (36.22) (35.74) (22.10) (22.24) (21.81)
EMP 0.3000*** 0.3702*** 0.2728*** 1.2518*** 1.3305*** 1.1245***
(3.21) (3.90) (2.91) (4.19) (4.42) (3.78)
CR 0.0004 0.0003 0.0004 0.0030*** 0.0031*** 0.0031***
(1.33) (1.18) (1.32) (3.36) (3.42) (3.47)
CI 0.0851*** 0.0891*** 0.0916*** 0.2082*** 0.2037*** 0.2244***
(16.55) (17.88) (17.70) (12.82) (13.01) (13.81)
Const. 10.6891*** 9.2742*** 10.7811*** 28.6939*** 27.4039*** 29.2998***
(12.54) (10.11) (13.21) (10.56) (9.37) (11.34)
R2 0.0377 0.0255 0.0301 0.031 0.0299 0.0197
The number of 39,101 39,729 39,729 37,987 38,590 38,590
observations
Notes
3. ROA = the consolidated based ROAthe non-consolidated based ROA
4. ROE = the consolidated based ROEthe non-consolidated based ROE
5. ***, **, and * represent statistical signicance at the 1, 5, and 10 % levels, respectively. The t values are
in parentheses
6.2 The Empirical Analysis: A Comparison Between Consolidated Financial 53
positive effect upon ROA but not on ROE. The estimated ratio of consolidated sales
to the parent companies sales has no statistically signicant effects upon
protability.
Next, the rst-stage equation for Hypothesis 4 is as follows:
I=K it const: a1 DIVit a2 ROAit a3 INTit a4 LEVit a5 ASSETit a6 GSALEit eit

6:3
where
I/K corporate investment ratio (capital expenditures over tangible xed assets)
INT average interest rate paid (interest paid over interest-bearing liabilities)
The above equation is estimated to clarify whether corporate diversication and
division of corporations affect decision making for corporate investment. The
estimation results shown in Table 6.2 indicate that the I/K ratio was not affected by
any of the proxy variables for corporate diversication and division of corporations.
The result that would have supported the hypothesis that capital expenditures are
highly expanded by corporate diversication and division of corporations is thus
not obtained.
However, ex-post protability should be examined to judge whether capital
expenditures are excessive. I dene overinvestment as a case in which corporations
fall into negative prot or negative net worth as a result of the investment. Then,
based on a logit model set at 1 for rms fallen into negative prot or negative net
worth and 0 otherwise, the effects of corporate diversication and division of
corporations are analyzed. The estimated equation is as follows:

PDPAit 1
LogitPDPAit 1 log
1 PDPAit 1
const: a1 DIVit a2 EI=Kit eit 6:4
where
DPA two types of rm dummy variables concerning undesirable nancial
conditions, DPA1 set at 1 for rms fallen into negative prot and 0
otherwise, and DPA2 set at 1 for rms fallen into negative net worth and 0
otherwise
E(I/K) the estimated corporate investment ratio from Eq. (6.3)
The second stage of Table 6.2 shows the estimation results of the logit model. In
the results in which the dependent variable is the dummy for negative prots, the
coefcient of the number of consolidated subsidiaries is negative at a statistically
signicant level. This means that the development of corporate diversication and
division of corporations, measured by the number of consolidated subsidiaries,
prevents rms from falling into negative prot. In the model where the negative net
worth dummy is the dependent variable, on the other hand, the coefcients for
variables reflecting corporate diversication and division of corporations are
insignicant in all cases.
Table 6.2 Corporate diversication, corporate investment, and nancial difculty

First stage: corporate diversication and corporate investment
I/K
DIV1 0.8463*
(1.78)
DIV2 0.7006
(0.66)
DIV3 0.0083
(0.81)
ROA 0.1459*** 0.1574*** 0.1566***
(3.81) (4.16) (4.14)
INT 0.0176 0.0249 0.0295
(0.38) (0.54) (0.63)
LEV 0.0188 0.0195 0.0196
(0.92) (0.96) (0.96)
ASSET 1.6888** 2.0995*** 1.8650***
(2.49) (3.06) (2.77)
GSALE 0.0145 0.0128 0.0127
(1.61) (1.43) (1.42)
Const. 48.5914*** 53.8760*** 50.6844***
(4.05) (4.50) (4.24)
R2 0.0216 0.0324 0.0288
Notes
3. ***, **, and * represent statistical signicance at the 1, 5, and 10 % levels, respectively.
The t values are in parentheses
Second stage: corporate diversication and nancial difculty
(1) Negative prots
DPA1
DIV1 0.0080
(0.17)
DIV2 0.2024
(1.62)
DIV3 0.0084***
(5.94)
E(I/K) 0.0486*** 0.0560*** 0.0851***
(6.24) (7.16) (9.32)
Const. 2.8552*** 2.4435*** 2.0327***
(11.08) (8.30) (7.65)
(2) Liabilities in excess of assets
DPA2
DIV1 0.1763
(1.37)
(continued)
6.3 The Empirical Analysis: A Comparison 55

Second stage: corporate diversication and nancial difculty
(1) Negative prots
DPA1
DIV2 0.0158
(0.05)
DIV3 0.0015
(0.35)
E(I/K) 0.1035*** 0.0887*** 0.1060***
(4.02) (3.73) (4.03)
Const. 14.8755*** 13.9533*** 14.3935***
(8.62) (8.33) (8.59)
Notes
1. The logit model is applied for the estimation
3. Industry dummy variables are introduced as controlling variables
4. ***, **, and * represent statistical signicance at the 1, 5, and 10 % levels, respectively. The t values
are in parentheses

Firms with Consolidated Subsidiaries and Those
Without
The second approach to the empirical analysis of corporate diversication and

division of corporations is the comparative analysis between rms with consoli-
dated subsidiaries and rms without. It may be inappropriate to deal equally with
rms with one or two consolidated subsidiaries and rms with 10 or more from the
viewpoint of corporate diversication and division of corporations. Therefore, in
this section, I classify rms with consolidated subsidiaries into two groups, one
comprising rms with one to four consolidated subsidiaries and another comprising
rms with ve or more. I compare between rms without consolidated subsidiaries
and the two groups of rms with them. Needless to say, the rms with one to four
consolidated subsidiaries are considered to have moderate corporate diversication
and division of corporations companies, and rms with ve or more are considered
to have aggressive corporate diversication and division of corporations companies.
I use ROA and ROE as performance measures in the comparative analysis. In the
estimated equation, the dependent variables are ROA and ROE, and the explanatory
variables are ownership structure, nancial structure, rm size, rm growth,
macroeconomic conditions, and the dummy variable (DUMA) for rms with one to
four consolidated subsidiaries and the dummy variable (DUMB) for rms with ve
or more. The estimated coefcients for the two dummy variables DUMA and
DUMB are thus understood to indicate whether each degree of corporate diversi-
cation and division of corporations has positively or negatively affected the two
protability measures. The estimated equation is as follows:2
PAit const: a1 DUMAit a2 DUMBit a3 SC10it a4 FIit a5 ECit

6:5
a6 LEVit a7 ASSETit a8 GSALEit a9 CIt eit
Moreover, I use cross terms for variables reflecting ownership structure and
nancial structure and two dummy variables for the degree of corporate diversi-
cation and division of corporations. By examining the estimated coefcients of
these cross terms, I can determine how the coefcients shift according to the degree
of corporate diversication and division of corporations. The estimated equation is
as follows:3
PAit const: a1 SC10it a2 SC10DUMAit a3 SC10DUMBit a4 FIit a5 FIDUMAit

a6 FIDUMBit a7 ECit a8 ECDUMAit a9 ECDUMBit a10 LEVit a11 LEVDUMAit
a12 LEVDUMBit a13 ASSETit a14 GSALEit a15 CIt eit
6:6
The estimated results for Eq. (6.5) are shown in Table 6.3. The estimated coef-
cients of the two dummy variables for the degree of corporate diversication and
division of corporations show statistically signicant negative values. This means
that the protability of rms that are diversifying businesses and splitting up cor-
porations is lower than that of rms that are not. Next, the negative magnitudes of
the coefcients for DUMA and DUMB are compared; the two coefcients are fairly
equal, and the coefcient of DUMA is somewhat greater than that of DUMB. This
result is very important from the point of view of corporate governance. If corporate
diversication and division of corporations produce inefciency because of weak
corporate governance, there will be a greater negative influence on the protability
of rms involved in aggressive corporate diversication and division of corporations
than on that of rms with moderate corporate diversication and division of cor-
porations. However, the estimation results in this section show the contrary.
Next, the estimation results for model (12) are shown in Table 6.4. The esti-
mated coefcients concerning ownership structure, such as the top 10 shareholders
concentration ratio, the shareholding ratio of nancial institutions, and the
2
In Sect. 6.2, two-stage estimation methods are used because the effect of the estimated degree of
corporate diversication and division of corporations upon protability is calculated. In this
section, however, simple panel analysis is used because the effect of the actual degree of the
number of consolidated subsidiaries upon protability is calculated.
3
Because of the strong correlations between DUMA and DUMB on the one hand and the inter-
section terms of DUMA and DUMB on the other, we estimate the two equations separately.
Table 6.3 The comparison ROA ROE

between rms with
consolidated subsidiaries and DUMA 0.9194*** 1.5168***
without them: the basic cases (10.62) (6.53)
DUMB 0.8648*** 1.5084***
(8.32) (5.42)
SC10 0.0222*** 0.0432***
(5.79) (4.19)
FI 0.0326*** 0.0331***
(8.12) (3.08)
EC 0.0044 0.0039
(1.03) (0.34)
LEV 0.0071*** 0.0291***
(2.85) (4.26)
ASSET 0.3854*** 0.4687**
(4.62) (2.08)
GSALE 0.0303*** 0.0526***
(28.76) (18.83)
CI 0.0191*** 0.0442***
(5.38) (4.65)
Const. 9.3960*** 7.3053*
(6.36) (1.84)
R2 0.0593 0.0330
The number of observations 57,295 57,060
Notes
1. DUMA: dummy variable set at 1 for rms with one to four
consolidated subsidiaries and 0 otherwise
DUMB: dummy variable set at 1 for rms with ve or more
4. ***, **, and * represent statistical signicance at the 1, 5, and
10 % levels, respectively. The t values are in parentheses
shareholding ratio of foreigners, all demonstrate statistically signicant positive

impacts upon rm protability. On the other hand, the cross terms between these
variables for ownership structure and the variables concerning the degree of cor-
porate diversication and division of corporations, such as DUMA and DUMB, all
show negative coefcients at a highly signicant level. This means that these
ownership structure variables have positive impacts upon the protability of rms
that are not diversifying businesses and splitting up corporations. However, these
positive impacts decline sharply, and in some cases diminish completely, for rms
that are diversifying businesses and splitting up corporations. On the other hand,
although the nancial leverage ratio has negative impacts upon the protability of
Table 6.4 The comparison ROA ROE

between rms with
consolidated subsidiaries and SC10 0.0799*** 0.1006***
without them: the cases (18.03) (8.29)
including cross terms SC10DUMA 0.0749*** 0.0694***
(23.86) (8.10)
SC10DUMB 0.0712*** 0.0732***
(19.19) (7.24)
FI 0.0800*** 0.0674***
(14.24) (4.39)
FIDUMA 0.0595*** 0.0401***
(10.78) (2.65)
FIDUMB 0.0573*** 0.0417***
(10.13) (2.69)
EC 0.0570*** 0.0545**
(6.56) (2.27)
ECDUMA 0.0797*** 0.0923***
(8.51) (3.64)
ECDUMB 0.0641*** 0.0546**
(6.62) (2.08)
LEV 0.0694*** 0.0766***
(20.40) (7.90)
LEVDUMA 0.0855*** 0.0600***
(26.02) (6.55)
LEVDUMB 0.0780*** 0.0592***
(21.86) (5.97)
ASSET 0.2203*** 0.2828
(2.64) (1.25)
GSALE 0.0296*** 0.0520***
(28.28) (18.61)
CI 0.0216*** 0.0463***
(6.13) (4.87)
Const. 5.3138*** 2.4744
(3.56) (0.61)
R2 0.1098 0.0466
Notes
1. DUMA: dummy variable set at 1 for rms with one to four
DUMB: dummy variable set at 1 for rms with ve or more
4. ***, **, and * represent statistical signicance at the 1, 5, and
10 % levels, respectively. The t values are in parentheses
rms that are not diversifying businesses and splitting up corporations, a positive
effect appears for rms that are pursuing corporate diversication and division of
corporations, to the extent that those negative impacts are nearly offset. Moreover, it
is interesting that the magnitudes of the coefcients of cross terms for DUMA and
DUMB are generally similar; in some cases, the coefcients of the cross terms for
DUMA have exceeded those for DUMB. These results, together with the results
shown in Table 6.3, suggest that there is little difference in the negative impacts of
corporate diversication and division of corporations between rms with moderate
corporate diversication and division of corporations and rms with aggressive
corporate diversication and division of corporations.
Overall, the comparative analysis between rms with consolidated subsidiaries
and rms without shows that corporate diversication and division of corporations
do not have positive impacts upon rm protability but, rather, negative impacts.
However, these negative impacts are quite similar for rms that pursue corporate
diversication and division of corporations aggressively and for those that pursue
them moderately.
6.4 The Empirical Analysis: Analysis Based on Segmental

Information
In this section, I empirically analyze the hypotheses concerning the relationship

between corporate governance and corporate diversication using the number of
segments, a direct indicator of degree of corporate diversication.
Although the disclosure of segmental information is still restricted (as men-
tioned), the analysis based on the restricted segmental information determines
whether the results are similar to those in the analysis in Sect. 6.2 using the ratios of
consolidated sales or assets to the parent companies sales or assets and the number
of consolidated subsidiaries. The following hypothesis is proposed:
Hypothesis 5 More concentrated corporate ownership leads to more moderate and
more efcient corporate diversication. Therefore, these numbers of segments
The rst-stage estimation model is as follows:
SEGit const: a1 SC10it a2 ASSETit a3 ROAit a4 TTt eit 6:7
The variable notations are the same as those in Sect. 6.2. Model (13) estimates
how the top 10 shareholders shareholding ratio affects the number of segments
(SEG), the variable reflecting the degree of corporate diversication. In the second
stage, how the estimated value of the number of segments (ESEG) from the rst
stage affects ROA and ROE is estimated. The formula is as follows:
PAit const b1 ESEGit a1 b2 GSALEit b3 EMPit b4 CRit b5 CIt eit

6:8
where
PA corporate performance variables, ROA or ROE
ESEG the estimated value of the number of segments in Eq. (6.7)
The estimation results are shown in Table 6.5. The rst-stage estimation shows
that a higher ownership concentration leads to fewer segments. This result supports
the hypothesis. The estimation result of the second-stage equation, however, shows
that the estimated number of segments in the rst-stage equation has no statistically
signicant positive influence on rm protability.
Next, I propose the following:
Hypothesis 6 Greater ownership by nancial institutions leads to more moderate
and efcient corporate diversication. Therefore, these numbers of segments
According to the traditional main bank theory in Japan (Aoki and Patrick 1994),
the main bank plays an important role as a monitor for client corporations. If this
theory is correct, greater ownership by the banking sector will lead to more cor-
porate diversication and thus improve corporate performance.
The structure of the equation used for the estimation of Hypothesis 6.2 is the
same as that in Eqs. (6.7) and (6.8), but with the nancial institutions ownership
ratio (BK) instead of the top 10 shareholders shareholding ratio.
Table 6.6 shows the estimation results. In the rst-stage model, the nancial
institutions ownership ratio has a signicantly negative impact upon the number of
segments. This could be consistent with the main bank theory because the banking
sector might impose restrictions on the degree of corporate diversication.
However, in the second-stage model, the estimated number of segments in the
rst-stage equation has not signicantly improved rm protability.
Hypothesis 7 Greater foreign ownership leads to more moderate and efcient
corporate diversication. Therefore, these numbers of segments improve corporate
performance.
This hypothesis is based upon the view that foreign investors have better
monitoring abilities. The key variable for this hypothesis is foreign ownership ratio
(EC).
Table 6.7 indicates that foreign ownership has no signicant impact upon the
number of segments. Moreover, the estimated number of segments has no signif-
icant relationship with corporate protability. These results completely reject
Hypothesis 3.
6.4 The Empirical Analysis: Analysis Based 61
Table 6.5 Ownership concentration and the number of segments

First Stage: the effects of ownership concentration on the number of segments
The xed effect model The random effect model
The number of segments
SC10 0.0028*** 0.0032***
(4.15) (5.27)
ASSET 0.2370*** 0.2095***
(15.79) (21.17)
ROA 0.0007 0.0005
(0.79) (0.58)
TT 0.0201*** 0.0199***
(18.14) (18.20)
Const. 1.1218*** 0.8956***
(4.11) (4.25)
R2 0.0648 0.1372
Notes
2. Industry dummy variables are introduced as controlling variables for the random effects model
values for the xed effects model and the z values for the random effects model are in the
parentheses
Second stage: the number of segments and corporate performance
ROA ROE ROA ROE
ESEG 9.4604 1.8304 18.10 23.90
(0.42) (0.03) (1.10) (0.59)
GSALE 0.0043*** 0.0062 0.0046*** 0.0044
(2.79) (1.34) (3.07) (1.03)
EMP 0.2054** 0.1247 0.0142 0.0499
(1.96) (0.39) (0.23) (0.39)
CR 0.0003 0.0013 0.0004* 0.0006
(1.25) (1.57) (1.85) (0.94)
CI 0.0142** 0.0055 0.0142*** 0.0075
(2.50) (0.32) (2.57) (0.46)
Const. 5.2007*** 1.9434 6.2819*** 2.2776
(5.51) (0.68) (7.28) (1.11)
R2 0.0005 0.0002 0.0059 0.0076
The number of observations 20,647 20,334 20,647 20,334
Notes
for the xed effects model and the z values for the random effects model are in the parentheses
Table 6.6 Bank ownership and the number of segments

First stage: the effects of bank ownership on the number of segments
BK 0.0016** 0.0013*
(1.97) (1.65)
ASSET 0.2515*** 0.2263***
(16.59) (21.87)
ROA 0.0009 0.0007
(1.11) (0.88)
TT 0.0191*** 0.0192***
(15.80) (16.17)
Const. 1.4725*** 1.3084***
(5.53) (6.38)
R2 0.0638 0.1327
Notes
2. Industry dummy variables are introduced as controlling variables for the random effects
model
parentheses
ROA ROE ROA ROE
ESEG 16.2086 3.1360 45.87 60.57
(0.42) (0.03) (1.10) (0.59)
GSALE 0.0043*** 0.0062 0.0046*** 0.0044
(2.79) (1.34) (3.07) (1.03)
EMP 0.2054** 0.1247 0.0142 0.0499
(1.96) (0.39) (0.23) (0.39)
CR 0.0003 0.0013 0.0004* 0.0006
(1.25) (1.57) (1.85) (0.94)
CI 0.0142** 0.0055 0.0142*** 0.0075
(2.50) (0.32) (2.57) (0.46)
Const. 5.2007*** 1.9434 6.2819*** 2.2776
(5.51) (0.68) (7.28) (1.11)
R2 0.0005 0.0002 0.0059 0.0076
Notes

Hypothesis 8 A higher debt ratio leads to more moderate and efcient corporate
diversication. Therefore, these numbers of segments improve corporate
performance.
This hypothesis is based on debts disciplinary mechanism of controlling the free
cash flows rms generate. I use the debt to total asset ratio (LEV) as a key variable.
As Table 6.8 shows, contrary to the hypothesis, a higher debt ratio brings a
signicantly greater number of segments. The estimated number of segments has no
statistically signicant influence on rm protability.
Table 6.7 Foreign ownership and the number of segments

First stage: the effects of bank ownership on the number of segments
EC 0.0002 0.0008
(0.20) (1.03)
ASSET 0.2387*** 0.2190***
(15.35) (21.59)
ROA 0.0005 0.0003
(0.57) (0.38)
TT 0.0202*** 0.0203***
(16.51) (16.83)
Const. 1.3005*** 1.2092***
(4.69) (5.79)
R2 0.063 0.1333
Notes
model
3. ***, **, and * represent statistical signicance at the 1, 5, and 10 % levels, respectively.
The t values for the xed effects model and the z values for the random effects model are in
the parentheses
ROA ROE ROA ROE
ESEG 176.6548 34.1788 76.50 101.01
(0.42) (0.03) (1.10) (0.59)
GSALE 0.0043*** 0.0062 0.0046*** 0.0044
(2.79) (1.34) (3.07) (1.03)
(continued)

ROA ROE ROA ROE
EMP 0.2054** 0.1247 0.0142 0.0499
(1.96) (0.39) (0.23) (0.39)
CR 0.0003 0.0013 0.0004* 0.0006
(1.25) (1.57) (1.85) (0.94)
CI 0.0142** 0.0055 0.0142*** 0.0075
(2.50) (0.32) (2.57) (0.46)
Const. 5.2007*** 1.9434 6.2819*** 2.2776
(5.51) (0.68) (7.28) (1.11)
R2 0.0005 0.0002 0.0059 0.0076
Notes
parentheses
We next propose the following:

Hypothesis 9 A greater number of segments causes overinvestment and worsens
corporate protability.
The estimation model is quite similar to Eqs. (6.3) and (6.4) in Sect. 6.2, except
it uses the number of segments (SEG) instead of three measures for the degree of
corporate diversication and division of corporations.
The estimation results are shown in Table 6.9. A result consistent with
Hypothesis 5, that an increased number of segments seems to promote capital
investment, is obtained from the rst-stage result. Moreover, a result supporting the
hypothesis that capital investment that considers the number of segments induces
negative prots is derived from the second-stage logit model estimation. However,
for the second-stage model, in which the negative net worth dummy is the
dependent variable, the result does not support the hypothesis.
Thus, the results of the analysis based upon the number of segments do not
robustly support the standard hypotheses on the relationship between corporate
governance and corporate diversication.
Table 6.8 Debt ratio and the number of segments

First stage: the effects of debt ratio on the number of segments
LEV 0.0029*** 0.0031***
(6.39) (7.73)
ASSET 0.2375*** 0.2156***
(16.80) (22.88)
ROA 0.0009 0.0008
(1.07) (0.96)
TT 0.0215*** 0.0216***
(19.24) (19.57)
Const. 1.4508*** 1.3464***
(5.75) (6.85)
R2 0.0699 0.1405
Notes
model
parentheses
ROA ROE ROA ROE
ESEG 9.3472 1.8085 18.61 24.57
(0.42) (0.03) (1.10) (0.59)
GSALE 0.0043*** 0.0062 0.0046*** 0.0044
(2.79) (1.34) (3.07) (1.03)
EMP 0.2054** 0.1247 0.0142 0.0499
(1.96) (0.39) (0.23) (0.39)
CR 0.0003 0.0013 0.0004* 0.0006
(1.25) (1.57) (1.85) (0.94)
CI 0.0142** 0.0055 0.0142*** 0.0075
(2.50) (0.32) (2.57) (0.46)
Const. 5.2007*** 1.9434 6.2819*** 2.2776
(5.51) (0.68) (7.28) (1.11)
R2 0.0005 0.0002 0.0059 0.0076
Notes
Table 6.9 The number of segments, corporate investment, and nancial difculty
First stage: corporate investment and the number of segments
I/K
SEG 1.1692** 1.0313**
(2.23) (2.30)
ROA 0.0441 0.0397
(0.72) (0.69)
INT 0.0840 0.1168
(0.75) (1.09)
LEV 0.0008 0.0159
(0.02) (0.59)
ASSET 3.8472*** 4.5816***
(3.48) (8.60)
GSALE 0.0289** 0.0257**
(2.40) (2.22)
Const. 84.4075*** 98.6756***
(4.31) (9.46)
R2 0.0418 0.0667
Notes
model
parentheses
Second stage: corporate investment and nancial difculty (Logit Model)
(1) Negative prot
Without industry dummies With industry dummies
E(I/K) 0.0041*** 0.0043***
(10.46) (9.78)
Const. 3.2737*** 3.6431***
(68.34) (19.85)
(2) Liabilities in excess of assets
Without industry dummies With industry dummies
E(I/K) 0.0053 0.0065*
(1.56) (1.67)
Const. 9.7546*** 12.2796***
(39.22) (7.87)
Notes
are in the parentheses
6.5 The Empirical Analysis: Analysis of Pure Holding Companies 67
6.5 The Empirical Analysis: Analysis of Pure Holding

Companies
Generally speaking, a company designed to own and govern another company is

called a holding company. Such companies have ratios of subsidiary stock value
to total assets exceeding 50 %.
Among holding companies, those that do not undertake any business and are
involved only in equity holdings are called pure holding companies while compa-
nies that do undertake business are business holding companies. A pure holding
company controls and governs a group of companies by owning their stocks.
Traditionally, pure holding companies were prohibited by the Anti-Monopoly
Law. However, the law was revised in 1997 to remove the ban except for cases
where the business concentration of the business group is too high.
There are several benets to shifting to a pure holding company system. First,
reorganizing the business is possible without paying any merger costs for the
integration of the internal divisions, such as the personnel division. Second, the
delegation of authority to each subsidiary company speeds up decision making.
Third, parent companies can concentrate on planning and coordinating their man-
agement strategy, and on supervising and auditing their group companies. On the
other hand, there are also drawbacks. For example, it is quite difcult to change the
organization because the independence of each company in the group is quite high.
Moreover, many mutually independent and inefcient businesses can coexist in a
group. These situations could lead to a conglomerate discount.
Figure 6.1 shows the changes in the number of pure holding companies. In
2000, there are only two pure holding companies in the manufacturing and
non-manufacturing sectors. Afterwards, the number increases steadily. In 2011, the
total number of pure holding companies reaches 264. In the non-manufacturing
sector, there are 178 pure holding companies in 2011, 4.8 % of the listed companies
in the sector. In the manufacturing sector, there are 86 pure holding companies in
2011, 2.8 % of the listed companies in the sector.
I carry out an empirical analysis of the relationship among the ownership
structure, degree of corporate diversication, and corporate performance of pure
holding companies. As proxy variables for diversication degree (common to all
hypotheses), I use the number of consolidated subsidiaries and the number of
segments. I use the difference between the consolidated-based ROA or ROE and the
non-consolidated-based ROA or ROE as well as the ROA and ROE alone as
performance variables to assess corporate diversication. I explain the hypotheses
and their estimation results below.
I propose the following:
Hypothesis 10 More concentrated corporate ownership leads to a more moderate
and efcient corporate diversication and division of corporations. Therefore, this
corporate diversication and division of corporations improve corporate
performance.
300
250
200
number of firms
150
100
50
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
manufacturing non-manufacturing
Fig. 6.1 The development of pure holding companies
Table 6.10 shows the estimation results for the hypothesis. In the rst-stage
estimation, the top 10 ownership concentration ratio has negative impacts upon the
degree of corporate diversication. However, the coefcient is statistically signi-
cant only for the xed effect model, in which the number of segments is the
dependent variable. In the second-stage estimation, checking whether the estimated
degree of corporate diversication enhances corporate protability, while the esti-
mated results in which the estimated number of consolidated subsidiaries is an
explanatory variable support the hypothesis, the results for the estimated number of
segments reject it. Therefore, the results are mixed.
Hypothesis 11 Greater ownership by nancial institutions leads to more moderate
and efcient corporate diversication. Therefore, this corporate diversication and
division of corporations improve corporate performance.
According to the estimation result shown in Table 6.11, no negative impacts of
the nancial institutions ownership ratio upon the degree of corporate diversi-
cation are observed in the rst-step model. In the second-stage estimation, the
estimated degree of corporate diversication had negative impacts upon corporate
protability, and the estimated number of segments has no signicant effects upon
protability. These results completely reject the hypothesis that the banking sector
is a good monitor.
Next, we propose the following:
Hypothesis 12 Greater foreign ownership leads to more moderate and efcient
corporate diversication and division of corporations. Therefore, this corporate
diversication and division of corporations improve corporate performance.
As shown in Table 6.12, foreign ownership leads to a signicantly smaller
number of segments in the rst-stage model. This is consistent with the hypothesis.
Table 6.10 Corporate diversication and ownership concentration: the case for pure holding companies
First stage: the effects of ownership concentration on corporate diversication
DIV3 SEG DIV3 SEG
SC10 0.0278 0.0128** 0.1025 0.0055
(0.19) (2.50) (1.03) (1.51)
ASSET 11.0913*** 0.4039*** 14.9450*** 0.2011***
(3.61) (4.15) (14.37) (5.37)
ROA 0.2266 0.0113* 0.0928 0.0078
(1.15) (1.71) (0.57) (1.40)
TT 2.5692*** 0.0310** 2.2135*** 0.0357**
(5.71) (2.05) (5.16) (2.45)
Const. 208.2939*** 3.3333* 270.6462*** 0.1427
(3.66) (1.83) (12.05) (0.18)
R2 0.3737 0.0770 0.3885 0.0784
The number of observations 963 757 963 757
Notes
6.5 The Empirical Analysis: Analysis of Pure Holding Companies
3. ***, **, and * represent statistical signicance at the 1, 5, and 10 % levels, respectively. The t values for the xed effects model and the z
values for the random effects model are in the parentheses
Second stage: corporate diversication and corporate performance
ROA ROA ROE ROE ROA ROA ROE ROE
EDIV3 0.0713* 0.3247*** 0.1371** 0.5465***
(1.77) (2.56) (2.14) (3.30)
ESEG 9.4604 1.8304 18.10 23.90
69
(0.42) (0.03) (1.10) (0.59)

(continued)
70

GSALE 0.0591*** 0.0043*** 0.1157*** 0.0062 0.0594*** 0.0046*** 0.1162*** 0.0044
(36.93) (2.79) (22.63) (1.34) (37.95) (3.07) (23.96) (1.03)
EMP 0.3487*** 0.2054** 1.0568*** 0.1247 0.5747*** 0.0142 1.4325*** 0.0499
(3.75) (1.96) (3.58) (0.39) (8.69) (0.23) (9.24) (0.39)
CR 0.0004 0.0003 0.0029*** 0.0013 0.0003 0.0004* 0.0020*** 0.0006
(1.32) (1.25) (3.25) (1.57) (1.22) (1.85) (2.73) (0.94)
CI 0.0823*** 0.0142** 0.1867*** 0.0055 0.0843*** 0.0142*** 0.1919*** 0.0075
(17.22) (2.50) (12.49) (0.32) (17.86) (2.57) (13.23) (0.46)
Const. 10.7853*** 5.2007*** 26.3627*** 1.9434 12.0308*** 6.2819*** 28.1294*** 2.2776
(13.35) (5.51) (10.32) (0.68) (13.79) (7.28) (13.57) (1.11)
R2 0.0419 0.0005 0.0336 0.0002 0.0587 0.0059 0.0537 0.0076
The number of observations 41,602 20,647 40,395 20,334 41,602 20,647 40,395 20,334
Notes
2. ROA = the consolidated based ROA the nonconsolidated based ROA
5. ***, **, and * represent statistical signicance at the 1, 5, and 10 % levels, respectively. The t values for the xed effects model and the z values for the
random effects model are in the parentheses
6 Empirical Analyses of Corporate Diversication
Table 6.11 Corporate diversication and bank ownership: the case for pure holding companies
First stage: the effects of bank ownership on corporate diversication
DIV3 SEG DIV3 SEG
BK 0.4596** 0.0024 0.0964 0.0001
(1.99) (0.31) (0.65) (0.03)
ASSET 10.1954*** 0.4117*** 15.9992*** 0.2174***
(3.19) (4.11) (13.00) (5.00)
ROA 0.2413 0.0102 0.0951 0.0075
(1.22) (1.56) (0.59) (1.36)
TT 2.8352*** 0.0405*** 2.2104*** 0.0420***
(6.08) (2.57) (5.04) (2.84)
Const. 207.2170*** 4.3275*** 293.0880*** 0.8210
(3.63) (2.40) (13.47) (1.08)
R2 0.3389 0.0811 0.3941 0.0820
Notes
EDIV3 0.0127* 0.0576*** 0.0143** 0.0570***
(1.77) (2.56) (2.14) (3.30)
ESEG 16.21 3.1360 45.87 60.57
71
(0.42) (0.03) (1.10) (0.59)

(continued)
72

GSALE 0.0591*** 0.0043*** 0.1157*** 0.0062 0.0594*** 0.0046*** 0.1162*** 0.0044
(36.93) (2.79) (22.63) (1.34) (37.95) (3.07) (23.96) (1.03)
EMP 0.3487*** 0.2054** 1.0568*** 0.1247 0.5747*** 0.0142 1.4325*** 0.0499
(3.75) (1.96) (3.58) (0.39) (8.69) (0.23) (9.24) (0.39)
CR 0.0004 0.0003 0.0029*** 0.0013 0.0003 0.0004* 0.0020*** 0.0006
(1.32) (1.25) (3.25) (1.57) (1.22) (1.85) (2.73) (0.94)
CI 0.0823*** 0.0142** 0.1867*** 0.0055 0.0843*** 0.0142*** 0.1919*** 0.0075
(17.22) (2.50) (12.49) (0.32) (17.86) (2.57) (13.23) (0.46)
Const. 10.7853*** 5.2007*** 26.3627*** 1.9434 12.0308*** 6.2819*** 28.1294*** 2.2776
(13.35) (5.51) (10.32) (0.68) (13.79) (7.28) (13.57) (1.11)
R2 0.0419 0.0005 0.0336 0.0002 0.0587 0.0059 0.0537 0.0076
Notes
Table 6.12 Corporate diversication and foreign ownership: the case for pure holding companies
First stage: the effects of foreign ownership on corporate diversication
DIV3 SEG DIV3 SEG
EC 0.4222** 0.0146** 0.1748 0.0117***
(2.49) (2.42) (1.49) (2.64)
ASSET 9.4856*** 0.4461*** 14.8450*** 0.2479***
(2.98) (4.41) (13.69) (6.40)
ROA 0.2466 0.0110* 0.1001 0.0085
(1.24) (1.68) (0.61) (1.55)
TT 2.5659*** 0.0402*** 2.2670*** 0.0415***
(5.67) (2.67) (5.24) (2.88)
Const. 185.5010*** 4.7262*** 277.0856*** 1.2352*
(3.23) (2.57) (13.30) (1.68)
R2 0.3492 0.0866 0.3864 0.0881
Notes
EDIV3 0.0095* 0.0434*** 0.0102** 0.0406***
(1.77) (2.56) (2.14) (3.30)
(continued)
73
74

ESEG 176.65 34.18 76.50 101.01
(0.42) (0.03) (1.10) (0.59)
GSALE 0.0591*** 0.0043*** 0.1157*** 0.0062 0.0594*** 0.0046*** 0.1162*** 0.0044
(36.93) (2.79) (22.63) (1.34) (37.95) (3.07) (23.96) (1.03)
EMP 0.3487*** 0.2054** 1.056*** 0.1247 0.5747*** 0.0142 1.4325*** 0.0499
(3.75) (1.96) (3.58) (0.39) (8.69) (0.23) (9.24) (0.39)
CR 0.0004 0.0003 0.0029*** 0.0013 0.0003 0.0004* 0.0020*** 0.0006
(1.32) (1.25) (3.25) (1.57) (1.22) (1.85) (2.73) (0.94)
CI 0.0823*** 0.0142** 0.1867*** 0.0055 0.0843*** 0.0142*** 0.1919*** 0.0075
(17.22) (2.50) (12.49) (0.32) (17.86) (2.57) (13.23) (0.46)
Const. 10.7853*** 5.2007*** 26.3627*** 1.9434 12.0308*** 6.2819*** 28.1294*** 2.2776
(13.35) (5.51) (10.32) (0.68) (13.79) (7.28) (13.57) (1.11)
R2 0.0419 0.0005 0.0336 0.0002 0.0587 0.0059 0.0537 0.0076
Notes
6.5 The Empirical Analysis: Analysis of Pure Holding Companies 75
However, when the number of consolidated subsidiaries is the dependent variable,

the contrary result is produced. In the second-stage estimation, while the estimated
number of consolidated subsidiaries improves corporate protability, the estimated
number of segments has no signicant impacts upon protability.
These results suggest that no consistent or plausible relationship supports the
standard hypothesis concerning ownership structure, corporate diversication, and
rm protability.
References
Aoki, M., & Patrick, H. (Eds.). (1994). The Japanese main bank system: Its relevance for
developing and transforming economies. New York: Oxford University Press.
Berglof, E., & Perotti, E. (1994). The governance structure of the Japanese nancial keiretsu.
Journal of Financial Economics, 36(2), 259284. doi:10.1016/0304-405X(94)90026-4.
Jensen, M. C. (1986). Agency costs of free cash flow, corporate nance and takeovers. American
Economic Review, 76(2), 323329.
Kang, J. K., & Shivdasani, A. (1995). Firm performance, corporate governance, and top executive
turnover in Japan. Journal of Financial Economics, 38(1), 2958. doi:10.1016/0304-405X(94)
00807-D.
Kaplan, S. N., & Minton, B. A. (1994). Appointments of outsiders to Japanese boards:
Determinants and implications for managers. Journal of Financial Economics, 36(2), 225258.
doi:10.1016/0304-405X(94)90025-6.
Prowse, S. D. (1992). The structure of corporate ownership in Japan. Journal of Finance, 47(3),
11211140. doi:10.1111/j.1540-6261.1992.tb04007.x.
Yafeh, Y., & Yosha, O. (2003). Large shareholders and banks: Who monitors and how? Economic
Journal, 113(484), 128146. doi:10.1111/1468-0297.00087.
Chapter 7
Main Results and Their Interpretation
Abstract In this book, many results that are quite different from the standard
hypotheses on corporate governance were obtained from the empirical study using
data on Japanese rms. Potentially powerful explanations for this difference include
the differences in corporate governance structures between Japan and the US. In the
latter, the primary focus of corporate governance is on the relationship between
shareholders and top managers. In Japanese rms, however, there are traditionally
several kinds of influential stakeholders aside from shareholders, such as main
banks, other creditors, member companies for both vertical keiretsu and horizontal
keiretsu, business acquaintances utilizing trade credit, employees working within a
long-term employment system, and local residents in company towns. It is essential
that Japanese rms build credible long-term relationships with these stakeholders
and maintain a good reputation. In these business circumstances, it is not neces-
sarily rational to give high priority to promoting the benets and interests of
shareholders and conducting management to promote the self-interest of top
managers. In order to reorganize the regional economy, deal with global environ-
mental issues, maintain long-run sustainable development, and reconstruct after the
Great East Japan Earthquake, an institutional arrangement for the corporate gov-
ernance system that attaches a great deal of importance to a variety of stakeholder
interests is required.
Keywords Corporate governance

Corporate behavior
Stock option

Risk-taking Corporate diversication Division of corporation Stakeholder
Main bank Keiretsu
Long-term employment system
Local residence
Long-term relationship
Regional economy
Global environmental issue
Long-run sustainable development
This book has conducted empirical studies on the influences of the new trend of
corporate management in Japan, such as the introduction of stock options and the
evolution of corporate diversication and division of corporations, upon corporate
behaviors and corporate performance.

78 7 Main Results and Their Interpretation
The main ndings of the empirical analysis on stock options and corporate
diversication are as follows. First, the impacts of stock options upon rm prof-
itability are examined using three kinds of indicators relevant to the amount of stock
option values computed by the formula of the BlackScholes model (Black and
Scholes 1973): the outstanding values of the existing stock option, the ratio of the
outstanding values of the existing stock option to the executive compensations, and
the marginally enhanced stock option values of the existing stock option as a result
of a 1 % gain in stock prices. The results show no signicantly positive impacts.
Second, from the theoretical point of views, risk-taking rm behaviors may be
induced by stock options. I empirically examine the effects of a stock option upon
rms risk-taking activities. The proxies for risk-taking activities are the debttotal
asset ratio, the capital expenditurecapital stock ratio, and the R&D expenditure
total assets ratio. The ndings do not indicate that stock options induce risk-taking
rm behavior except for R&D ratio in the manufacturing sector.
The introduction of a stock option has neither a positive impact upon prof-
itability nor the negative side effect of promoting risk-taking behaviors. Therefore, a
result of neither harmful nor good impacts was observed.
Third, the empirical analysis of corporate diversication and division of cor-
porations based upon a comparison between consolidated nancial statements and
non-consolidated nancial statements shows that the protability of the consoli-
dated nancial statements bases is sometimes better than that of the
non-consolidated nancial statements bases because of the promoted corporate
diversication and division of corporations. However, this result does not neces-
sarily mean that corporate diversication and division of corporations improved
earnings on their own because the accounting gures in the non-consolidated
nancial statements are part of the outcomes of corporations that have consolidated
subsidiaries; not only the accounting gures in the consolidated nancial statements
but also those in the non-consolidated nancial statements are affected in many
ways by the corporate behaviors of corporate diversication and division of
corporations.
On the other hand, I argue that the impacts of corporate diversication and
division of corporations are better grasped from the comparative analysis between
rms with consolidated subsidiaries and rms without. The negative impacts
derived from our empirical studies on corporate diversication and division of
corporations can be interpreted as follows. Japanese rm behaviors concerning
corporate diversication and division of corporations may include a certain unso-
phisticated element, such as a failure to formulate an effective managerial strategy
due to a lack of sufcient experience. Moreover, the main purpose of Japanese
corporate diversication and division of corporations might not be prot maxi-
mization but rather rm development, ensuring market share, welfare improvement
for employees, or job preservation.
Finally, the ndings that the negative impacts of corporate diversication and
division of corporations are not dependent upon the degree of corporate diversi-
cation and division of corporations and that they are quite similar between
7 Main Results and Their Interpretation 79
moderately diversied rms and aggressively diversied ones suggest that weak
corporate governance is not the key problem.
In this book, many results that are quite different from the standard hypotheses
on corporate governance were obtained from the empirical study using data on
Japanese rms. How can I interpret them? Potentially powerful explanations for this
difference include the differences in corporate governance structures between Japan
and the US. In the latter, the primary focus of corporate governance is on the
relationship between shareholders and top managers, and top managers are
expected to be faithful agents of shareholders. Many of the hypotheses in this book
are premised on the US-style corporate governance model.
For example, as surveyed in Chap. 2, the introduction of a stock option is
regarded as an institutional mechanism for solving or mitigating the agency
problem between shareholders and top managers. In other words, a stock option can
increase shareholders payoffs by increasing rms stock prices and expanding
corporate value. When the agency problem between shareholders and top managers
is serious, the negative aspects of corporate behaviors relating to corporate diver-
sication and division of corporations, such as empire building and overinvestment,
will become more noticeable and severe.
As argued by Hanazaki (2008), however, Japanese rms traditionally have
several kinds of influential stakeholders apart from shareholders, such as main
banks, other creditors, member companies for both vertical keiretsu and horizontal
keiretsu, business acquaintances utilizing trade credit, employees working within a
long-term employment system, and local residents in company towns. It is essential
for Japanese rms long-term protability that they build credible long-term rela-
tionships with such stakeholders and maintain a good reputation. In these business
circumstances, it is not necessarily rational to give high priority to promoting the
benets and interests of shareholders and conducting management to promote the
self-interest of top managers.
Needless to say, shareholders are among the most important stakeholders and are
becoming increasingly so. In Japanese rms, however, the controlling rights of
various stakeholders are complexly entangled with each other. Therefore, the
straightforward hypotheses tested in this book that assume a simple agency problem
between shareholders and top managers are difcult to support.
The institutional reforms to corporate law systems such as the introduction of
stock options (1997), the removal of the ban on pure holding companies (1997), the
inauguration of the committee-system company (2002), the enactment of the
Corporate Law (2005), the announcement of Principles for Responsible
Institutional Investors (2014), and Japans Corporate Governance Code (2015)
represent attempts to introduce US- and European-style corporate governance
systems into Japan. These trends might be important and meaningful given that
alternative measures of the corporate management of top managers have increased
and they could extend the degree of freedom for their corporate management.
However, in view of the substantial differences in economic and social structures
and in the historical evolution of corporate governance structures between Japan
80 7 Main Results and Their Interpretation
and the US and Europe, the empirical analyses in this book suggest that the
expected effects and problems may not occur.
In order to reorganize the regional economy, deal with global environmental
issues, maintain long-run sustainable development, and reconstruct after the Great
East Japan Earthquake, the formation of a corporate governance that stresses the
relationship between shareholders and corporate managers would be not only
obviously unsuitable but also erroneous. Instead, an institutional arrangement for
the corporate governance system that attaches a great deal of importance to a variety
of stakeholder interests is needed. As a substitute for the introduction of the
Anglo-Saxon-type corporate governance system, examining how surplus values for
various stakeholders can be maximized and distributed appropriately among them
will become more important for corporate management. A design for institutional
arrangements that would meet this objective is urgently required.
References
Black, F., & Scholes, M. (1973). The pricing of options and corporate liabilities. Journal of
Political Economy, 81(3), 637654.
Hanazaki, M. (2008). Corporate nance and corporate governance: An informational and
institutional approach. Tokyo: University of Tokyo Press. (in Japanese).
Appendix Tables
Basic Statistics for Major Parameters
of the Black-Scholes Model
Appendix A
All Listed Firms
Time to maturity
Unit: years
Manufacturing Non-manufacturing
Mean Median Maximum Minimum Mean Median Maximum Minimum
1997 4.83 4.00 9.00 2.75 5.04 4.99 10.00 3.00
1998 4.16 4.00 8.00 2.00 3.71 3.00 9.00 1.00
1999 4.09 4.00 8.00 1.00 4.29 3.58 8.00 0.92
2000 3.94 3.75 10.00 1.00 4.19 3.71 10.00 2.00
2001 3.95 3.79 10.00 1.00 4.22 3.75 10.00 1.00
2002 4.16 4.00 10.00 1.00 4.47 4.00 10.00 0.92
2003 4.13 4.00 10.00 1.00 4.67 4.00 10.00 0.75
2004 4.50 4.00 20.00 1.00 5.10 4.82 19.75 1.00
2005 6.44 4.00 30.00 1.00 5.70 5.00 30.00 0.17
2006 10.11 5.00 30.00 2.00 6.73 5.00 30.00 0.08
Time for exercise
Unit: years
1997 0.80 1.00 2.00 0.00 0.98 1.00 3.00 0.00
1998 1.38 1.76 2.00 0.00 1.57 2.00 3.00 0.00
1999 1.74 2.00 5.00 0.00 1.74 2.00 4.00 0.00
2000 1.72 2.00 4.00 0.00 1.64 2.00 4.00 0.00
2001 1.77 2.00 3.00 0.00 1.75 2.00 3.00 0.00
(continued)

Development Bank of Japan Research Series, DOI 10.1007/978-4-431-56006-7
82 Appendix Tables: Basic Statistics for Major Parameters of the Black-Scholes Model
(continued)
Unit: years
2002 1.80 2.00 3.00 0.00 1.88 2.00 4.00 0.00
2003 1.87 2.00 4.00 0.00 1.82 2.00 4.00 0.00
2004 1.79 2.00 4.00 0.00 1.82 2.00 4.00 0.00
2005 1.64 2.00 7.00 0.00 1.80 2.00 7.00 0.00
2006 1.50 1.76 3.00 0.00 1.70 2.00 3.00 0.00
Stock price over exercise price (S/K)
1997 1.09 0.96 1.56 0.89 1.71 1.14 4.68 0.55
1998 1.28 0.92 9.43 0.51 1.97 0.96 17.81 0.26
1999 0.99 0.94 2.84 0.50 1.64 0.95 11.22 0.20
2000 9.96 0.94 904.76 0.27 1.79 0.96 38.02 0.16
2001 1.06 0.98 4.79 0.64 4.44 1.00 218.18 0.00
2002 1.72 0.99 44.99 0.01 11.91 1.10 932.50 0.08
2003 1.07 0.93 7.16 0.10 2.98 0.98 52.41 0.03
2004 12.32 0.96 1278.50 0.63 2139.79 0.99 207500.00 0.33
2005 183.29 0.95 5545.00 0.35 1035.06 0.95 188500.00 0.10
2006 419.55 0.95 7620.00 0.48 247.79 0.96 9800.00 0.01
Volatility of stock price
1997 0.42 0.34 0.82 0.23 0.73 0.73 1.12 0.42
1998 0.52 0.47 1.52 0.23 0.70 0.65 1.50 0.19
1999 0.62 0.55 1.49 0.16 1.00 0.80 3.95 0.26
2000 0.54 0.47 1.55 0.12 0.77 0.59 3.96 0.14
2001 0.59 0.53 2.06 0.17 0.75 0.64 2.40 0.11
2002 0.51 0.37 9.93 0.10 0.70 0.54 3.33 0.12
2003 0.50 0.41 3.30 0.07 0.92 0.63 9.41 0.11
2004 0.47 0.37 3.30 0.06 0.81 0.58 4.24 0.08
2005 0.47 0.36 4.08 0.08 0.74 0.59 3.77 0.11
2006 0.41 0.39 1.18 0.14 0.54 0.44 2.35 0.11
Note The volatility is demonstrated by the standard deviation of the rate of change of the monthly
stock prices within the year in which the decision to introduce the stock option is made at the
shareholders meeting
Appendix Tables: Basic Statistics for Major Parameters of the Black-Scholes Model 83
Distribution of exercise price
Unit: yen
Exercise price 1 1100 1011000 100110,000 over 10,000
Number of samples 73 17 747 856 434
Appendix B
Listed Firms of the First Section of the Tokyo
Stock Exchange
Time to maturity
Unit: years
All industries
Mean Median Maximum Minimum
1997 4.76 4.67 9.00 2.75
1998 3.96 3.17 9.00 1.00
1999 4.25 4.00 8.00 0.92
2000 4.02 3.50 10.00 1.75
2001 4.03 3.75 10.00 1.00
2002 4.16 3.79 10.00 0.92
2003 4.06 3.96 10.00 0.75
2004 4.58 4.00 20.00 1.00
2005 5.89 4.00 30.00 0.17
2006 8.91 5.00 30.00 0.08
Time for exercise
Unit: years
All industries
1997 0.98 1.00 3.00 0.00
1998 1.50 2.00 3.00 0.00
1999 1.76 2.00 5.00 0.00
2000 1.69 2.00 4.00 0.00
2001 1.77 2.00 3.00 0.00
2002 1.88 2.00 4.00 0.00
2003 1.89 2.00 4.00 0.00
2004 1.82 2.00 4.00 0.00
(continued)
(continued)
Unit: years
All industries
2005 1.69 2.00 7.00 0.00
2006 1.53 1.76 3.00 0.00
All industries
1997 1.73 1.13 4.68 0.89
1998 1.82 0.94 17.81 0.48
1999 1.17 0.94 7.67 0.26
2000 6.17 0.94 904.76 0.46
2001 3.10 0.98 218.18 0.08
2002 8.03 1.01 932.50 0.01
2003 1.42 0.94 36.62 0.25
2004 818.07 0.97 201500.00 0.34
2005 160.53 0.95 5545.00 0.23
2006 416.46 0.95 9800.00 0.35
Volatility for stock price
All industries
1997 0.61 0.71 1.12 0.23
1998 0.59 0.54 1.50 0.19
1999 0.81 0.65 3.95 0.16
2000 0.62 0.50 3.96 0.12
2001 0.63 0.53 2.40 0.11
2002 0.57 0.44 9.93 0.10
2003 0.61 0.45 4.41 0.07
2004 0.53 0.40 4.24 0.06
2005 0.50 0.38 3.77 0.08
2006 0.41 0.38 2.11 0.14
Appendix C
Listed Firms of the Second Section of the Tokyo Stock Exchange
Time to maturity
Unit: years
All industries
1997 5.92 4.75 10.00 3.00
1998 3.48 3.00 5.00 2.00
1999 3.56 3.00 5.00 2.00
2000 3.96 3.00 10.00 1.00
2001 3.96 3.00 10.00 1.00
2002 4.26 3.75 10.00 1.00
2003 3.71 3.00 9.00 1.92
2004 3.97 3.00 9.00 2.00
2005 6.27 4.82 30.00 2.00
2006 8.81 4.91 30.00 2.00
Time for exercise
unit: years
All industries
1997 0.67 1.00 1.00 0.00
1998 1.43 1.00 3.00 0.00
1999 1.33 2.00 2.00 0.00
2000 1.52 2.00 4.00 0.00
2001 1.61 2.00 3.00 0.00
2002 1.52 2.00 3.00 0.00
2003 1.73 2.00 3.00 0.00
2004 1.57 2.00 3.00 0.00
2005 1.47 2.00 3.00 0.00
2006 1.57 2.00 2.00 0.00
All industries
1997 0.83 0.79 1.14 0.55
1998 0.77 0.81 1.03 0.26
(continued)
(continued)
All industries
1999 1.57 0.94 7.83 0.61
2000 1.17 0.96 7.83 0.26
2001 0.91 0.96 1.23 0.27
2002 1.63 0.96 12.72 0.74
2003 3.02 0.98 25.57 0.73
2004 1.42 0.96 10.65 0.53
2005 7569.59 0.92 188500.00 0.14
2006 228.17 0.98 2995.00 0.01
All industries
1997 0.73 0.73 1.00 0.46
1998 0.85 0.72 1.52 0.44
1999 0.74 0.69 1.40 0.26
2000 0.71 0.58 2.93 0.23
2001 0.75 0.55 1.89 0.24
2002 0.54 0.40 1.89 0.15
2003 0.66 0.51 2.86 0.16
2004 0.74 0.59 2.86 0.26
2005 0.79 0.57 4.08 0.18
2006 0.47 0.43 1.20 0.11
Appendix D
Listed Firms of the Mothers Section of the Tokyo
Stock Exchange
Time to maturity
Unit: years
All industries
1999 5.15 5.00 8.00 1.00
(continued)
(continued)
Unit: years
All industries
2000 5.26 5.00 8.00 3.00
2001 5.20 5.00 8.00 2.00
2002 5.99 6.38 9.50 2.00
2003 6.56 7.00 10.00 2.00
2004 6.36 7.00 10.00 1.08
2005 6.13 6.67 10.00 1.00
2006 6.70 7.92 10.00 2.00
Time for exercise
Unit: years
All industries
1999 2.40 2.00 3.00 2.00
2000 1.73 2.00 3.00 0.00
2001 1.81 2.00 3.00 1.00
2002 1.81 2.00 3.00 0.00
2003 1.66 2.00 3.00 0.00
2004 1.87 2.00 3.00 0.00
2005 1.92 2.00 4.00 0.00
2006 1.92 2.00 3.00 0.00
All industries
1999 4.27 2.84 11.22 0.20
2000 4.59 4.75 9.64 0.16
2001 1.52 1.01 5.55 0.00
2002 3.41 1.15 15.38 0.27
2003 5.81 2.29 52.41 0.03
2004 3481.13 1.18 207500.00 0.33
2005 1.65 0.94 18.69 0.10
2006 1.32 0.95 7.12 0.32
All industries
1999 1.53 1.49 2.28 0.78
2000 1.38 1.22 2.56 0.78
2001 1.22 1.27 2.06 0.34
2002 1.11 1.16 2.19 0.34
2003 1.41 1.15 9.41 0.34
2004 1.16 0.88 3.65 0.33
2005 0.94 0.85 2.46 0.33
2006 0.82 0.69 2.35 0.25

Corporate Governance and Corporate Behavior in Japan

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Corporate Governance and Corporate Behavior in Japan

Hochgeladen von

Copyright:

Verfügbare Formate

SPRINGER BRIEFS IN ECONOMICS

DE VELOPMENT BANK OF JAPAN RESEARCH SERIES

Development Bank of Japan Research Series

Editorial Board Members

More information about this series at http://www.springer.com/series/13542

ISSN 2191-5504 ISSN 2191-5512 (electronic)

Library of Congress Control Number: 2016943876

Development Bank of Japan 2016

Printed on acid-free paper

This Springer imprint is published by Springer Nature

Kunitachi, Japan Masaharu Hanazaki

6 Empirical Analyses of Corporate Diversication and Division

Appendix Tables: Basic Statistics for Major Parameters

Masaharu Hanazaki is Professor at the Graduate School of Commerce and

His English book publications include:

Abstract Many institutional reforms of Japans corporate law system have

Development Bank of Japan 2016 1

Abstract A stock option is dened as a corporations own share option given as

Development Bank of Japan 2016 5

2.1 What Is a Stock Option?

In modern corporations with a separation between ownership and control, the

2.2 The Development of the Stock Option System in Japan

Stock options used to be prohibited in Japan because the acquisition of a companys

(1) Equity Investment

Purchasing Price Future Stock Price

(2) Stock Option

Exercise Price Future Stock Price

Fig. 2.1 Comparison between equity investment and stock option

Table 2.1 (continued)

2.3 Stock Option Studies on Japanese Firms

Kubo and Saito (2008) perform a comprehensive analysis of the managerial

Abstract I empirically analyze the characteristics of rms that introduce stock

Development Bank of Japan 2016 15

3.1 Estimation of Stock Option Value

3.1.1 Basic Model

OP eqt S Nd1 ert K Nd2

The option price OP is computed by substituting Eq. (3.4) for (3.3).4

3.1.2 Estimation Approach

3.1.3 Estimation Results

Table 3.1 Estimated stock option values

Table 3.2 Estimated stock option values and executive compensation

3.2 Has the Protability of Companies Improved

Table 3.3 Stock option and rm protability

SOV3 0.0272* 0.0518

3.3 Did the Introduction of a Stock Option Induce

RTit const: a1 SOVit a2 SC10it a3 FIit a4 ECit a5 ROAit a6 GSALEit

SALE 0.0051 0.0058 0.0050

Abstract Chandler, The visible hand: the managerial revolution in American

Development Bank of Japan 2016 31

4.1 Theories of Corporate Diversication

4.2 Negative Effects of Corporate Diversication

Many empirical analyses of US corporations have found that corporate diversi-

market. This suggests that capital expenditures as a result of corporate diversi-

4.3 Reexamination of the Diversication Discount

4.4 Corporate Governance and Corporate Diversication

Abstract Many studies on US rms use information on business segments as the

5.1 How Can I Determine the Actual Conditions

Determining the actual condition of corporate diversication is the most essential

Development Bank of Japan 2016 37

1 Energy & Infrastructure Segment

ain Consolidated Subsidiaries of Each Business Segment Toshiba

he number of consolidated subsidiaries : 590

usiness Segments of a Parent Company

1 Information & Telecommunication Systems

PAit const b1 ESEGit a1 b2 GSALEit b3 EMPit b4 CRit b5 CIt eit