This is to certify that the
dissertation entitled

THE DISCRETIONARY REPORTING OF
NONCONTROLLING INTERESTS AND ITS ASSOCIATION
WITH THE MARKET ASSESSMENT OF CREDIT RISK

 

presented by

_ =93!

BEI DONG

V -*.

LIBRARY
Mlchlgan State

! ifl-ix. “1‘37"?“

«.—

has been accepted towards fulfillment
of the requirements for the

 

 

 

Doctoral degree in Accounting and Information
System

 

K-K_______\.

 

Major Professor’s Signature

LI'IlgIOX

 

Date

MSU is an affirmative-action, equal-opportunity employer

“'3‘“

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

CII 7. I; 8g.“ 99 9

Add

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 KIProilAccaPreleIRCIDaIeDue.indd

THE DISCRETIONARY REPORTING OF NONCONTROLLING INTERESTS AND
ITS ASSOCIATION WITH THE MARKET ASSESSMENT OF CREDIT RISK

By

Bei Dong

A DISSERTATION
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Accounting and Information Systems

2008

ABSTRACT

THE DISCRETIONARY REPORTING OF NONCONTROLLING INTERESTS AND
ITS ASSOCIATION WITH THE MARKET ASSESSMENT OF CREDIT RISK

By
Bei Dong

This dissertation empirically examines the determinants of firms’ decision to
classify noncontrolling interests in the liability versus mezzanine section of the balance
sheet, and the implication of the classification decision on the credit market’s assessment
of risk. I hypothesize and find that firms with greater exposure to litigation risk from
noncontrolling shareholders and firms with more serious information asymmetry
problems are more likely to classify noncontrolling interests as a liability to mitigate
agency costs. However, consistent with more serious agency conflicts due to controlling
ownership, I predict and find that family firms are more likely to adopt the mezzanine
classification.

After controlling for the cndogeneity of firms’ classification choice and other
factors affecting credit ratings, I find that firms choosing to disclose noncontrolling
interests in the liability section have higher credit ratings (around one notch) compared
with those that disclose noncontrolling interests in the mezzanine section. This finding is
consistent with my hypothesis that creditors are more likely to favorably assess the credit
risk of companies that conservatively classify claims on their balance sheet. Overall, my
results suggest that the classification choice of noncontrolling interests is informative
about the market’s assessment of firms” credit risk, and therefore, limiting accounting
choices could potentially eliminate relevant information from financial statements in

situations where significant information asymmetry and agency costs may be prominent.

To my husband, my mother and father.

iii

ACKNOWLEDGMENTS

I would like to thank members of my dissertation committee for their invaluable
guidance and support: K. Ramesh (Chair), Ana Herrera, Marilyn Johnson, and Ed
Outslay. Without their support, I could not have finished this long journey. I would
especially like to thank Ramesh for his motivation, guidance, inspiration, and insight
throughout the whole doctoral program. My experience to work with him during the past
years has been an extremely valuable and enjoyable one. One thing I learnt through this
journey from him is that you must believe in what you are doing and work confidently in
order to make things happen.

I would also like to thank Donna Coallier of PWC, Chris E. Hogan, John
(Xuefeng) Jiang, Gregory Jonas of Moody’s, Yen-Jung Lee, Edward (Xuejun) Li, Kathy
Petroni, Min Shen, Mike Shields, and Isabel (Yanyan) Wang and workshop participants
at Michigan State University, University of South Florida at Tampa, and Western
Michigan University for helpful comments and suggestions.

Most importantly, I would like to thank my husband, Jing Wang, my parents,

Xiao Bing Yuan and Shou Nian Dong for their love and support.

iv

TABLE OF CONTENTS

 

 

LIST OF TABLES--- - - - _ - - -_ - -- . ------- vii
LIST OF FIGURES - - - - - -- - - ------ - - ---------------------------- viii
CHAPTER 1 INTRODUCTION--- - . ......... - l

 

CHAPTER 2 DETERMINANTS OF THE DECISION TO CLASSIFY

 

NONCONTROLLING INTERESTS AS LIABILITY ........ 7
2.1 Controlling-noncontrolling shareholder agency problem ............... 8
2.1.1 Family firms ................................................................ 10
2.1.2 Exposure to litigation by noncontrolling
shareholders ................................................................ 1 l
2.2 Manager-shareholder agency problem .......................................... 14
2.2.1 Alleviate information asymmetry problems in
growth firms ................................................................ 14
2.2.2 Alleviate manager-shareholder agency problem in
financially-constrained firms ...................................... 15

CHAPTER 3 CREDIT MARKET EFFECTS OF BALANCE-SHEET

 

 

 

CLASSIFICATION OF NONCONTROLLING INTERESTS .......... 17
CHAPTER 4 RESEARCH DESIGN -- - - _ -_ - _ 20
4.1 Determinants of the decision to classify noncontrolling
interests as liabilities ..................................................................... 20
4.2 Credit market effects of the balance-sheet classification of
noncontrolling interests ................................................................. 22
CHAPTER 5 SAMPLE AND UNIVARIATE ANALYSIS ------ -- - - 25
CHAPTER 6 MULTIVARIATE RESULTS - - -- - ----- 27
6.1 Determinants of the reporting decision ......................................... 27
6.2 Credit market analysis of reporting decisions ............................... 29
6.3 Additional analysis ........................................................................ 31
6.3.1 Firm characteristics ..................................................... 31

6.3.2 Earnings quality and liability classification choice 32

 

CHAPTER 7 CONCLUSION ........................................................................................ 35
APPENDICES- -- - -- -- - -- -_ _ _ - ------ 47
Appendix A: Background ................................................................... 47

Appendix B: Credit ratings’ numerical conversions ........................... 51

Appendix C: Sensitivity analysis ........................................................ 52
Appendix D: Balance-sheet classification examples .......................... 57
BIBLIOGRAPHY ........................................................................................................... 62

vi

LIST OF TABLES

Table 1 Summary Statistics ........................................................................................... 38
Table 2 Correlation Matrix ............................................................................................ 40

Table 3 Regression results for Eq. (1): Analysis of the determinants of the decision to
disclose noncontrolling interests in liability sections ....................................... 44

Table 4 Regression results for Eq. (2): Instrumental variable estimation of credit
market effect of disclosing noncontrolling interests as liability ...................... 45

Table 5 Descriptive statistics of 207 S&P 500 firms in fiscal year 2004 across
classification choice for noncontrolling interests ............................................. 46

Table C.1 Descriptive statistics of S&P 500 firms in fiscal year 2004 across classification
choice for noncontrolling interests ................................................................... 52

Table C.2 Rank regression results for Eq. (1): Analysis of the determinants of the
decision to disclose noncontrolling interests in liability sections .................... 53

Table C.3 Regression results for Eq. (1): Analysis of the determinants of the decision to
disclose noncontrolling interests in liability sections ....................................... 54

Table C.4 Regression results for Eq. (2) (using RATING_Klock as dependent variable):
Instrumental variable estimation of credit market effect of disclosing
noncontrolling interests as liability .................................................................. 55

Table C.5 Regression results for Eq. (2) without control: Instrtunental variable estimation
of credit market effect of disclosing noncontrolling interests as liability ........ 56

vii

LIST OF FIGURES

Figure 1 Two types of agency problems: manager-shareholder agency problem and
controlling-noncontrolling shareholder agency problem ................................. 37

viii

CHAPTER 1
INTRODUCTION
I investigate firms’ balance-sheet classification decision of noncontrolling
interests in the pre-F AS 160 era, and the association between the classification choice and
firms’ credit ratings.l The Financial Accounting Standards Board (hereafter, FASB)
defines noncontrolling interests as “the portion of the equity (residual interest) in a
subsidiary attributable to the owners of the subsidiary other than the parent and the
parent’s affiliates.”2 FAS 160 “Noncontrolling Interests in Consolidated Financial
Statements,” which is effective for fiscal periods beginning after 12/ 1 5/2008, views most
noncontrolling interests in subsidiaries as equity of the consolidated group.3 However,
F AS 160 has raised considerable debate among practitioners and in the business

community.4’ 5 Moreover, given that noncontrolling interests fall in a grey area on the

 

I My sample includes S&P 500 firms with noncontrolling interests during the fiscal year 2004 before FAS
160 had taken effect when firms had discretion in classifying noncontrolling as liability, equity, or
mezzanine.

2The FASB believes the term noncontrolling interest is more precise than minority interests, because the
latter implies that the holder has less than 50 percent of the voting shares. However, in the case of a
variable interest entity under Interpretation 46(R), a controlling interest can hold a minority of the voting
shares but still exercise control through other means.

3 FAS 150 requires that mandatorily redeemable securities (e.g., mandatorily redeemable noncontrolling
interests) that “embody an unconditional obligation requiring the issuer’s redemption by transfer of assets
at a specified or determinable date (or dates) or upon an event that is certain to occur" be disclosed in the
liability section. I find that firms in my sample did not change their classification of noncontrolling interests
during the period 2002 — 2004, suggesting that FAS 150 did not have an effect on my sample firms. In
other words, my dissertation focuses on the classification of non-mandatorily redeemable noncontrolling
interests for which firms have discretion in classifying them as liabilities, equity, or mezzanine.

4 Based on forty-nine comment letters to the 2005 Exposure Draft, “Consolidated Financial Statements,
Including Accounting and Reporting Noncontrolling Interests in Subsidiaries,” (hereafter, 2005 ED), the
majority of respondents did not support the classification of noncontrolling interests as equity. Ms.
Seidman, one of the FASB members, disagreed with the equity classification because she believed that
noncontrolling interests is different from the residual interests of parent firms. For example, noncontrolling
interests do not “absorb losses relating to other activities of the parent and thus would be higher in standing
than the residual claim of shareholders of the parent entity (pg. 3, FAS I60)”

5 Appendix A provides background information on accounting rules for the balance-sheet classification of
noncontrolling interests prior to F AS 160, the debate among practitioners, academics. and standard setters
over the proper classification, and the competing theories underlying the debate.

liability-equity spectrum, their balance sheet classification was a key issue in the FASB’s
broader project on liabilities and equity.

Despite the importance of the classification issue, there is no systematic empirical
evidence on corporate reporting of noncontrolling interests. For example, I find that in
year 2004 roughly a quarter of the S&P 500 firms with noncontrolling interests disclose
them in the liability section, with the remainder of the firms reporting in the mezzanine
section (i.e., between liability and equity), suggesting significant cross-sectional variation
in classification practicesé‘ 7 Given the importance of financial leverage to firm value
(e.g., Hamada 1969; Rubinstein 1973; Bowman 1980; Mulford 1985; Choi 1988) and the
significant consequences of liability-equity classification to different stakeholders
(Hopkins 1996; Engel et al. 1999), I argue that the accounting classification choice for
noncontrolling interests is an important financial reporting decision.

Extant literatures argue that accounting choices can be used to alleviate various
agency problems (Jensen and Meckling 1976; Smith and Warner 1979; Watts and
Zimmerman 1986; Fields et al. 2001).8 Further, recent accounting research suggests
that accounting conservatism can mitigate agency costs, and thereby, reduce the cost of

capital (Ahmed et al. 2002; Francis et al. 2004; Zhang 2006). In my setting, I view the

 

6 None of my sample firms disclosed noncontrolling interests in the equity section. In its survey of
consolidated financial statement, the American Institute of Certified Public Accountants reported that only
3 out of 85 companies display noncontrolling interests in the equity section. Further, AICPA survey
“Accounting Trends & Techniques “of 600 public companies during the period 1993 to 2002 also shows
that classification variation in their sample exists only between liability and mezzanine choices. Exhibits 1,
2, and 3 provide examples of liability (United Technologies), mezzanine (Borders Group), and equity (from
Appendix A of F AS 160) classification of noncontrolling interests.

7 Based on my analysis of S&P 500 firms’ lO-K forms during the period 1993 — 2004, I find that 4 out of
263 firms with noncontrolling interests in 2004 changed their classification at least once during this period.
Of those 4 firms, 1 (2) of them changed from mezzanine (liability) classification to liability (mezzanine)
classification, and one firm changed its classification twice.

classification of noncontrolling interests in the liabilities section as a conservative
financial reporting choice as it leads to understatement of net assets. Unlike most prior
research which focuses on owner-manager agency problems, I contend that the
classification choice for noncontrolling interests is also influenced, may be to a greater
extent, by agency problems between controlling and noncontrolling shareholders. With
respect to agency problems arising from control, I hypothesize that: 1) due to better
alignment of the interests of managers and family owners, family firms are more likely to
report noncontrolling interests as lower priority claims (i.e., mezzanine), possibly
reflecting their indifference to protecting the interests of noncontrolling shareholders; and
2) however, firms with higher litigation exposure to noncontrolling shareholders (“deep
pocket”) are more likely to choose the liability classification to signal reduced agency
conflicts. From the standpoint of manager-shareholder agency problems, I examine
whether the managerial decision to classify noncontrolling interests as liability is related
to incentives for: 1) alleviating information asymmetries generally; and 2) limiting
agency conflicts between managers and shareholders when firms are financially
constrained.

My empirical analysis using a sample of 128 S&P 500 firms with noncontrolling
interests in the fiscal year 2004 largely supports of my predictions. Reflecting the agency

conflicts due to control, the likelihood of classifying noncontrolling interests in the

 

8 I consider the classification of noncontrolling interests an accounting choice which “is a decision whose
primary purpose is to influence (either in form or substance) the output of the accounting system in a
particular way” (Fields et al. 2001, page 256).

liability section decreases by 17.9 percent in family firms.9 However, the probability of
choosing the liability classification increases by l 1.3 percent when parent companies
with “deep pockets” face potential litigation risk from the controlling relationship
(proxied by the ratio of consolidated assets to the claims of noncontrolling shareholders).
To mitigate manager-shareholder agency problems, I find that firms with higher
information asymmetry (proxied by the extent of growth opportunities) are 10.9 percent
more likely to choose the conservative approach of classifying noncontrolling interests as
liabilities. I do not, however, find a relationship between financial constrain and the
balance sheet classification of noncontrolling interests. Taken together, while my results
on the role of accounting classification in manager-shareholder agency conflicts is
consistent with prior research, I provide new evidence on how agency conflicts due to
controlling ownership influence accounting choice.

While my study focuses on a single balance-sheet accounting classification
choice, I also provide evidence that firms adopting liability classification choice are more
likely to provide higher-quality earnings. This suggests that firms use multiple accounting
choices to satisfy the demand for high-quality financial information to address agency
problems (Fields et a1. 2001).

Considering the prominence of credit-based financing in the capital markets, the
relevance of accounting leverage ratios for credit risk assessment (Shi 2003; Ashbaugh et
al. 2006; J iang 2005; Lee 2005), and the importance of accounting conservatism to

creditors (Ahmed et al. 2002; Beatty et al. 2006; Zhang 2006), I also investigate the

 

9 The marginal effect of continuous variables (dummy variables) is measured as the change in the
probability of adopting liability classification for a change in the independent variable equal to its inter-
quartile range (from zero to one) in the sample, holding all other independent variables at their means.

potential economic significance of the classification choice from the credit market
standpoint. While extant empirical studies have not resolved whether balance sheet
classification choices of hybrid securities are relevant to investors,10 the balance-sheet
classification of noncontrolling interests provides a unique setting for examining this
question because it has no impact on net income, and therefore, the analysis is not
affected by considerations other than the placement of information on the balance sheet. “
Extant literature argues that firms with conservative accounting practices are
rewarded with lower cost of debt (Zhang 2006) and higher credit ratings (Ahmed et al.
2002). Accordingly, I hypothesize that firms would receive higher credit ratings when
they choose to classify noncontrolling interests under liabilities because the classification
choice is likely to signal firms’ desire to use balance sheet conservatism to limit agency
conflicts. Based on the 107 of my sample firms with credit rating data, I find that on
average, firms that classify noncontrolling interests as liabilities have credit ratings that
are around one notch higher than those of firms adopting mezzanine section. The result is

obtained after controlling for the various determinants of credit risk and the cndogeneity

 

'0 Archival studies find that stock market treats securities with both liability and equity characteristics
(e.g., mandatorily redeemable preferred stock) in a fashion that is different from the accounting treatment
the securities receive in financial reports, suggesting that liability-equity classification has no discerning
effect on investor valuation (Kimmel and Warfield 1995; Cheng et al. 2003). However, experimental
research documents that liability-equity classification of hybrid securities matters to equity analysts, an
important group of financial statement users (Hopkins I996).

” Lee at al. (2006) find that insurance companies that have used security gains and losses to manage
earnings and those that have a reputation for poor financial reporting quality choose to provide information
on comprehensive income in a statement of equity rather than a performance statement as recommended by
the FASB. In the context of comprehensive income reporting, firms use different levels of reporting
saliency as a response to differential costs of transparent disclosures. Given the proximity of the locations
for the two classification choices, I do not expect disclosure saliency to be an important issue with respect
to noncontrolling interests. However, one must be careful to note that this dissertation does not provide a
direct test of the comparison between liability and mezzanine classification on the degree of saliency, thus
it is possible that the observed association could be confounded by managers’ concern on and market
reaction to the different saliency of different classification choices.

of the classification choice (e.g., Fisher 1959; Ziebart and Reiter 1992; Sengupta 1998;
Shi 2003; Ashbaugh et al. 2006; Klock et al. 2005).

In summary, this dissertation also has implications for the FASB who recently
curtailed the discretion in the accounting classification of noncontrolling interests. My
credit rating analysis indicates that balance-sheet classification provides useful
information for the market’s assessment of firms’ credit risk. This finding implies that
limiting accounting choices could potentially eliminate relevant information from
financial statements in situations where information asymmetry and agency problems
may be prominent (Fields et al. 2001: p. 261). While eliminating accounting discretion is
arguably driven by the F ASB’s recent focus on decision usefulness, my analysis suggests
that stewardship benefits of accounting discretion may suffer.

The remainder of the dissertation proceeds as follows. Sections 2 and 3 discuss
various determinants of the balance-sheet classification of noncontrolling interests and
the potential impact of the classification on credit markets. Section 4 describes the
research design, and Section 5 discusses sample selection and descriptive statistics.
Section 6 presents empirical findings and the results of robustness tests. Section 7

concludes.

CHAPTER 2
DETERMINANTS OF THE DECISION TO CLASSIFY NONCONTROLLING
INTERESTS AS LIABILITY

According to Dye (2002 page 1125-1126), “financial reporting is, at its roots, a
process of classification: firms are ‘going concerns’ or not; transactions are ‘recognized’
in a firm’s financial statements or not; leases are capital leases or operating leases;
expenditures are assets or expenses; financial claims are liabilities or equity, etc.” The
classification process underlying financial reporting has a real effect on resource
allocation because it helps investors predict firms’ future cash flows, which in turn,
affects their investment decisions (Dye 2002). Consistent with this notion, extensive
research documents that liability-equity classification has significant consequences for
various stakeholders such as buy—side financial analysts (Hopkins 1996) and managers
(Engel et al. 1999).12

Moreover, accounting literature argues that accounting choices are affected by
and can be used to alleviate various types of agency problems (see Fields et al. 2001 and
Watts 2003 for a review). As depicted in Figure 1, there are two types of agency
problems in my setting: 1) between controlling and noncontrolling shareholders, and 2)
between managers and shareholders. Of the two types, I argue that the agency problem
due to control dominates the classification of noncontrolling interests. In a majority-

owned subsidiary, managerial decisions about the operating, investing, and financing

 

'2 Hopkins (I996) provides experimental evidence that buy-side financial analysts price the finn’s
mandatorily redeemable preferred stock as if the firm had issued debt (equity) when the stocks are
classified as liability (equity). Engel et al. (1999) find that firms incur significant costs (i.e., 10 to 43
million) to manage their debt-to-assets ratio by an average of 13 percent through issuing trust preferred

activities of subsidiaries and the ensuing economic benefits are often controlled by parent
company shareholders. 13 The controlling shareholders often have closer ties to both the
parent and subsidiary managers than do noncontrolling shareholders, making
opportunistic alliances between managers and controlling shareholders more likely. My
objective in this section is to identify how these types of agency problems will influence

the noncontrolling interests classification.

2.1 Controlling-noncontrolIing shareholder agency problem

The financial economics literature argues that market forces and/or legal statutes
cannot completely eliminate the controlling-noncontrolling shareholder agency conflicts,
leading to the expropriation of noncontrolling shareholders’ interests by controlling
shareholders (Hazen 1977; Brudney and Clark 1981; Bebchuk 1989; Shleifer and Vishny
1997; La Porta et al. 2000).14 In the case of a majority-owned subsidiary, controlling
shareholders gain if they cause subsidiary assets to be transferred to the parent
corporation without adequate compensation to noncontrolling shareholders. Such wealth
transfer mechanisms could include outright theft, dilution of noncontrolling interests by
issuing additional shares to the controlling shareholders, and asset sales to parent firms at

favorable prices (Shleifer and Vishny 1997; LaPorta et al. 2002).

 

stock. While Hopkins ( I996) examines the behavioral implications of balance sheet classification, Engel et
al. (1999) focus on transaction structuring by managers to achieve a specific accounting classification.

'3 Officers and directors of the parent typically serve in the same capacity for the subsidiary. Schipper and
Smith (1986) document that in 34 of 48 of their sample firms, the President or CEO of the subsidiary is also
a parent manager. In l8 cases, one or more of the secretary, treasurer, and corporate counsel is the same
person for parent and subsidiary. Also, 56 of 57 subsidiary boards have at least one member who is also a
parent director or an officer in the parent firm.

'4 Those studies articulate the agency problem between controlling and noncontrolling shareholders but not
provide empirical evidence.

Direct expropriations related to control are often discussed in the popular press.
For instance, Mr. LeBow, who owned 60-percent of the Brooke Group, apparently
benefited from favorable loans, and from selling assets to his other controlled interests at
non-arms-length prices (Cohen 1993). The Wall Street Journal notes that “a look at Mr.
LeBow’s Securities and Exchange Commission filings for the past three years shows how
easily a dominant shareholder can tap into the cash stream of a company he controls.”

Empirical research also provides evidence on expropriation by controlling
shareholders. For example, Maury and Pajuste (2002) find that the dividend payout made
to noncontrolling shareholders is lower than that to controlling shareholders. Moreover, .
studies linking ownership structure and performance find a “roof-shaped” relationship,
which suggests that as the controlling ownership increases beyond a certain level,
controlling shareholders begin to retain more private benefits that are not shared with the
noncontrolling shareholders (Stulz 1988; Wruck 1989; McConnell and Servaes 1990;
Morck et al. 1998).15 In this section, I will investigate the effect of controlling-
noncontrolling shareholder agency problem on firms’ classification choice of

noncontrolling interests.

 

'5 However, some studies provide competing evidence suggesting no exploitative behavior through
controlling ownership of publicly-traded companies. In particular, Denis and Denis (1994) find that
majority-owned firms are as profitable as diffusely-held firms. Holdemess and Sheehan (1988) find that
majority-owned firms do not invest in overly risky capital projects nor do they provide excessive executive
compensation.

2.1.1 Family firms

Recent research finds that a nontrivial fraction of publicly-traded US. companies
consists of family firms where founders or descendants control senior management
positions (Anderson and Reeb 2003). Even among S&P 500 firms, roughly one third are
family firms in which family members control approximately 19% of the shares
outstanding (Anderson and Reeb 2003). However, extant literature argues that while
family firms in the US. have less severe owner-manager agency problems, they suffer
from more severe agency conflicts between family owners and other shareholders (Gilson
and Gordon 2003; Fan and Wong 2003).’6

The level of financial reporting quality supplied by family firms may depend on
the relative dominance of the manager-shareholder agency problem (the alignment effect)
versus the agency problem due to family control (the entrenchment effect). The
alignment effect predicts that family firms would provide higher-quality financial
information due to: 1) better alignment of interests between shareholders and managers
(or firm value); 2) longer investment horizon of family members (Stein 1998 and 1989;
Chami 1999; James 1999; Anderson and Reeb 2003); and 3) incentives to preserve
family reputation (Anderson et al. 2003). In contrast, the entrenchment effect implies that
given founding family members have incentives and opportunity to extract wealth from
other shareholders, family firms are more likely to manipulate their financial reports for

their private benefits.

 

'6 To my best knowledge, there is no empirical evidence documenting this type of agency problems in US.
family firms. Wang (2006) argue that the results of Fan and Wang (2003) may not be generalized to the
countries like US. with stronger investor protection.

10

In contrast to the competing theories regarding the effect of concentrated family
ownership on the quality of financial reports, empirical studies find that family firms
report better quality earnings (i.e., less positive discretionary accruals, greater earnings
informativeness, higher earnings persistence, and higher earnings predictability) (Wang
2006; Ali et al. 2007), and are more likely to provide bad news warnings in management
earnings forecast (Ali et al. 2007), suggesting the alleviation of the owner-manager
agency problem.

However, I argue that agency problems that arise due to any controlling
ownership would be more serious in family firms when compared to other firms. Because
of better alignment of interests between family members and managers, family-owned
parent companies are more likely to behave opportunistically when dealing with
noncontrolling shareholders of subsidiaries than parent companies that are not controlled
by family members. Consistent with this argument, I predict that, when reporting
noncontrolling interests in the balance sheet, family firms are more likely to resort to a
classification that presents a favorable picture of its financial position from the
perspective of stakeholders other than noncontrolling shareholders. Accordingly, I expect
a negative relation between family membership and the decision to classify
noncontrolling interests as liabilities.

H]: Family firms are less likely to classify noncontrolling interests as liabilities.

2.1.2 Exposure to litigation by noncontrolling shareholders

Expected litigation risk under the federal securities laws is thought to be an
important determinant of public companies’ disclosure decisions (e.g., Francis et al.

1994; Skinner 1994 and 1997; See Healy and Palepu, 2001, for a review). In the setting I

11

examine, firms can face lawsuits from noncontrolling shareholders for any alleged
expropriation by controlling shareholders. Both securities regulations and court decisions
indicate a concern for protecting the rights of noncontrolling shareholders. For example,
the Securities and Exchange Commission (SEC) requires an independent assessment of
the “Fairness of the Transaction’ under Rule l3e-3 of Securities and Exchange Act of
1934 when a controlling owner acquires a noncontrolling interest.17 In a similar vein,
court decisions on conflict of interests in parent-subsidiary transactions (e.g., assets sales)
establish that noncontrolling interests have the right to require that benefits be distributed
on a pro-rata basis. '8

The litigation risk faced by controlling shareholders is highlighted in a recent case
involving Ford Motor Co. of Canada Ltd. in which a court awarded judgment to its
noncontrolling shareholders as the original buyout price paid by the controlling
shareholder, Ford US, was not deemed a fair consideration (Transfer Pricing Report
617, 03/ 1 7/2004). Another case in point is the claim brought by the noncontrolling
shareholders in the management buyout of Quiznos, where a Denver Judge ruled that the
original offer of $8.50 per share substantially underestimated the per-share fair value of at
least $32.50.19

Extant research argues that firms with “deep pockets” are exposed to higher

litigation risk (Francis et al. 1994, Roger and Stocken 2005). For instance, larger firms

 

’7 See http://www.sec.gov/about/forms/ruleI3e-3.pdf (accessed on 1 1/23/2006).

'8 Johnson et al. (2000) argues that, in common law countries, the duty of loyalty principle, or fiduciary
duty principle is specifically employed to deal with the situations of conflict of interests (e.g., controlling
and noncontrolling interests). “In the case of fiduciary duty, the very fact that the interests of a director
are in conflict with those of the company itself constitutes the basis for liability, and if the interests of the
company are prejudiced as a result of such conflict, liability for breach of fiduciary duty arises...”(Shibuya

1972, p. 127).
'9 see http://www.forbes.com/business/global/2004/030l/Ol8.html (accessed on 0l/28/2008)

l2

tend to have better insurance that provides resources for plaintiff attorney fees as well as
paying defense costs, making them an attractive target for lawsuits. The auditing
literature also documents that auditors are more exposed to litigation risk when they work
for larger clients (Carcello and Palmrose 1994). Consistent with these findings, 1 posit
that when parent firms have “deep pockets,” they face a higher risk of litigation from
noncontrolling shareholders.

Recent studies suggest that firms with higher litigation costs are more likely to
engage in economic behavior that would reduce the likelihood of class action lawsuits
under securities laws. From a valuation standpoint, Lowry and Shu (2002) provide
evidence that firms with higher litigation risk substantially underprice their IPOs to lower
expected litigation costs. From a disclosure standpoint, studies show that managers
preemptively disclose bad news to prevent large stock price declines on earnings
announcement dates, and thereby, reduce the potential costs of shareholder lawsuits,
(e.g., Skinner I994; Kasznik and Lev 1995).20

In my setting, this suggests that firms with large litigation exposure from
noncontrolling interests are more likely to follow business practices that would be
construed as respecting the rights of noncontrolling shareholders. Analogously, from an
accounting perspective, I expect these firms will adopt the liability classification for
noncontrolling interests as a signal of fair dealing and protection of the rights of

noncontrolling shareholders. Therefore, my hypothesis is:

 

20 Previous studies (Skinner, I994; Kasznik and Lev, 1995) argue that early voluntary disclosure of large
negative earnings surprise reduces expected legal costs in two ways. First, if the information is disclosed
“voluntarily” prior to the mandated release date, it is difficult for the plaintiff, who does not know for sure
when the manager first received the bad news, to argue that the manager withheld information. Second,
disclosing early limits the period of nondisclosure, thereby reducing the damages that plaintiffs can claim.

13

H2: Firms with “deep pockets” are more likely to classify noncontrolling interests as
liabilities.

While prior research views “deep pockets” purely based on the total economic
resources at a firm’s disposal, I view “deep pockets” in my context as the total economic
resources controlled by parent firms relative to the potential claims by noncontrolling
shareholders. I argue that parent firms become more attractive litigation targets when
they are more capable of covering potential litigation claims from noncontrolling

shareholders.

2.2 Manager-shareholder agency problem

Because of the information asymmetry and the conflict of interest between
managers and shareholders, managers have incentives and opportunities to expropriate
wealth from shareholders, resulting in deadweight losses, and thus, a reduction in firm
value (Berle and Means, 1932; Jensen and Meckling, 1976; Watts 2003a). Accounting
research suggests that financial disclosures can be used to alleviate owner-manager
agency problems, and the consequent reduction in firm value, through providing
verifiable information that is useful in monitoring and evaluating managerial decisions
(Watts and Zimmerman 1986; Bushman and Smith 2001). In this section, I will
investigate the effect of owner-manager agency problem on firms’ classification choice

of noncontrolling interests.

2.2.1 Alleviate information asymmetry problems in growth firms

Previous studies argue that growth firms are more likely to have pronounced
information asymmetry problems that are generated from the nature of the firm’s

investment opportunity or growth Options (Smith and Watts 1992). The increases in

14

information asymmetry provide more opportunities to managers to manipulate financial
reports to transfer wealth to themselves at the expense of uninformed shareholders (e.g.,
excessive compensation of managers). Such behavior reduces resources available for
positive NPV projects, generating deadweight losses due to foregone projects. Therefore,
capital markets tend to charge a higher cost of capital to growth firms to price protect for
expected opportunistic behavior on the part of managers (Jensen and Meckling 1976;
Watts and Zimmerman 1986).

Extant literature argues that accounting conservatism can alleviate information
asymmetry problem between managers and shareholders through: I) constraining
opportunistic managerial behavior in overstating unverifiable net assets by requiring
higher verification standards for gain recognition (Watts 2003); 2) limiting excessive
compensation to managers that is costly to firms (Watts 2003); and 3) assisting in
identifying negative NPV projects by timely recognition of economic losses, thus playing
a monitoring role in firms’ investment policies (Ball 2001). Thus, to alleviate the adverse
effects of information asymmetry on the cost of capital, managers of growth firms are
more likely to adopt conservative accounting practices. Applied to my setting, growth
firms are more likely to disclose noncontrolling interests as liabilities. Accordingly, I
expect a positive relation between firms’ growth prospects and the probability of
classifying noncontrolling interests as liabilities.

H3: Growth firms are more likely to classify noncontrolling interests as liabilities.

2.2.2 Alleviate manager-shareholder agency problem in financially-constrained
firms

Financial economics literature points out that investors will charge a higher cost

of capital to firms with uncontrolled agency problems as compensation for a higher

15

expropriation risk (e.g., F ama and Jensen 1983). Accordingly, firms in this category are
unable to find financing for all of their profitable projects (i.e., face a “financially-
constrained” optimization problem; Lamont et al. 2000), and therefore, are forced to
make suboptimal real decisions, which are reflected in the firms’ stock prices (Sharpe
1964, Lintner 1965, and Merton 1973; Bemanke and Gertler 1989, Gertler and Gilchrist
1994; see Bemanke et al. 1996 for a review).2| Consistent with this argument, recent
financial economics research documents that asset pricing factors that proxy for financial
constrains are positively associated with equity returns (Campello and Chen 2005;
Whited and Wu 2005).

To the extent financial constraints are a reflection of uncontrolled agency
problems, managers would take actions (including accounting choices) to resolve
manager-shareholder agency conflicts. Consistent with my earlier discussions, I predict
that firms facing financial constraints are more likely to classify noncontrolling interests
as liability as part of a conservative financial reporting strategy (Ball 2001, Watts 2003
and 2006, Ahmed and Duellman 2007), which leads the following hypothesis:

H4: Financially-constrained firms are more likely to classify noncontrolling interests as
liabilities.

 

2' Following Lamont et al. (2001), financial constraints are frictions that prevent a firm from firnding all
desired investments. This inability to fund investment might be due to various reasons such as credit
constraints, inability to issue equity, dependence on bank loans, or illiquidity of assets. 1 do not use
“financial constraints” to mean financial distress, although the two constructs are undoubtedly correlated.

l6

CHAPTER 3
CREDIT MARKET EFFECTS OF BALANCE-SHEET CLASSIFICATION OF
NONCONTROLLING INTERESTS
Given the relevance of accounting leverage ratios for credit risk assessment (Shi

2003; Ashbaugh et al. 2006; Jiang 2005; Lee 2005), and the importance of accounting
conservatism to creditors (Ahmed et al. 2002; Beatty et al. 2006; Zhang 2006), I consider
the following research question in this section: does the balance-sheet classification of
noncontrolling interests provide relevant information to the credit market? Most of the
prior research tries to demonstrate whether a claim (e. g., unrecorded capital lease
commitments or unfunded pension liabilities) is viewed by the market as debt- or equity-
like (e.g., Bowman 1980; Dhaliwal 1986; Kimmel and Warfield 1995; Cheng et al. 2003)
through examining how the claim affects the equity market’s assessments of the firm’s
systematic risk.22 However, prior research has largely overlooked the role of accounting
classification for credit market’s assessment of risk. I appeal to the credit market to shed
light on the importance of accounting classification for the following two reasons. First,
creditors are concerned about the firm’s downside risk, which is generally determined by
the ability of firms to pay interest and principal (i.e., creditworthiness). From the
creditors’ perspective, accounting leverage ratios are important indicators that help
“broadly define a company’s position relative to rating categories” (2002 S&P Corporate
Ratings Criteria page 53-55). Thus, creditors appear to consider accounting liability-
equity classification in their assessment of credit risk. Secondly, the credit market plays a

substantial role in satisfying the external financing needs of corporate America. For

17

example, in 2002 (2000), US. corporations raised more than $1.2 trillion in the bond
market compared to $120 ($200) billion in the equity market (Beller 2003).

Given the prominence of credit-based financing in the capital markets, the
question whether the accounting classification of corporate claims is associated with the
credit market’s risk assessment is important. Specifically, I investigate whether firms’
decision on the placement of noncontrolling interests on the balance sheet provides
creditors with relevant information about firms’ desire to alleviate various agency
conflicts.23

Prior studies argue that firms adopting conservative financial reporting have lower
cost of debt, because conservatism helps creditors in terms of: 1) decreasing the downside
risk from limiting inappropriate wealth transfer to equity holders (e.g., Ahmed et al.
2002); and 2) providing early signals about the deterioration of firms’ financial
performance (Zhang 2006). For example, Ahmed et al. (2002) document that firms
employ accounting conservatism to solve the bondholder-shareholder conflicts on
dividend distribution, and firms using more conservative reporting have higher credit
rating. Zhang (2006) also finds that accounting conservatism benefits firms in terms of
lower initial interest rates.

I extend the prior research on accruals-based conservatism to conservatism in

balance sheet classification. In my setting, I view the classification of noncontrolling

 

22 Under the assumption of perfect markets. riskless debt and no taxes, several studies show that debt-like
obligations affect systematic equity risk through financial leverage.

2’ For example, discussions with a senior executive of a major credit-rating agency suggest that the
agency’s analysts do not consider noncontrolling interests as equity in their financial analysis if: I) there is
put option available for noncontrolling shareholders; 2) if there is a plan to sell the subsidiary; or 3) if
noncontrolling interests have a unilateral right to block distributions from subsidiary to parent firm.
However, I argue that the classification choice of noncontrolling interests still matters to creditors in the

18

interests in the liabilities section as a conservative financial reporting choice as it leads to
understatement of net assets. Consequently, I posit that creditors are more likely to
favorably assess the credit risk of companies choosing the liability classification, as firms
have signaled their intent to alleviate various agency problems by adopting a conservative
financial reporting practice. Overall, I expect a positive relationship between firms’
liability classification choice of noncontrolling interests and creditors’ evaluation.

H5: Firms’ decision to classify noncontrolling interests in the liability section of the
balance sheet will be positively associated with the credit market’s assessment of firm
risk.

An alternative prediction is that the accounting liability-equity classification does
not convey relevant information to the market. For example, Kimmel and Warfield
(1995) and Cheng et al. (2003) find that investors view mandatorily-redeemable preferred
stock as equity, and non-redeemable preferred stock as debt, indicating that securities are
treated by the stock market in a fashion which is different from the accounting treatment
the securities receive in financial reports.24 This implies that liability-equity classification

has no effect on investors’ valuation of firms. Both of these papers conclude that liability-

equity classification “may lack representational faithfulness.”

 

sense that it reflects firms’ desire to use balance sheet conservatism to limit various agency conflicts rather
than merely reflect the perceived liability/equity nature of the item.

2" Following ASR268, “Presentation in financial statements of ‘redeemable preferred stocks”, RPF D
should not be classified as equity, suggesting that at least RPFD is regarded as non-equity. Later, SFAS
150, “Accounting for certain financial instruments with characteristics of both liabilities and equity”,
requires that mandatorily redeemable preferred securities be classified as liability.

19

CHAPTER 4

RESEARCH DESIGN

4.1 Determinants of the decision to classify noncontrolling interests as liabilities

I use a probit regression to model the probability of disclosing noncontrolling
interests in the liability versus mezzanine section as a function of the firm characteristics

predicted by the hypotheses in section 3 and various controls:

Prob (LIABILITYi) = F660 + ,8] FAMIL YFIRMI- + fi2 NONCONTROLLING,-
+fi3INV_GROWTHi +fl4C0NSTRAINi +fl5SIZEi (1)
+ fl6 VOLATILITYI- + [97Audit0rl- + 81-),

where

LIABILITYl- = one if firm i disclosed noncontrolling interests in the liability section and

zero if in the mezzanine section,

FAMILYFIRM i = one if firm i is a founding family firm as identified by Business

Week, and zero otherwise,

NONCONTROLLING,- = natural logarithm of the ratio of total assets to noncontrolling
interests of firm i (proxy for litigation risk),

IN V_ GROWTH i = ratio of book value to market value of common equity of firm i,

C ONS TRAIN i = one if firm i had negative average free cash flows calculated over the

preceding five years, and zero other otherwise,25

SIZE,- = natural logarithm of the beginning-of-year total assets of firm i,

 

25 Following Standard and Poor’s Corporate Ratings Criteria (2002 S&P Corporate Ratings Criteria), free
cash flows are defined as cash flows from operating activities (Compustat #308) minus capital expenditures
(Compustat #128).

20

VOLA TILITYi = firm i’s 2-digit SIC decile rank of the standard deviation of

noncontrolling interests scaled by the standard deviation of total liabilities in the
preceding five years,

Auditor, = auditor dummies.

Based on H1, I predict a negative coefficient on FAMIL YFIRM. I use
NONC ONT ROLLING as a proxy for firms’ litigation risk from noncontrolling
shareholders and predict a positive coefficient for it. I use the book-to-market ratio as a
proxy for the inverse of firm growth prospects, and consistent with H3, expect a negative
coefficient on IN V_GROWT H. Consistent with H4, I expect the coefficient of
CONSTRAIN to be positive.

I also include two control variables. Because perceptions of excess volatility may
have negative implications for firm values, firms with noncontrolling interests that are
significantly more volatile than the firms’ existing liabilities may be reluctant to disclose
noncontrolling interests in the liability section. Therefore, I expect a negative coefficient
for the VOLA TILITY variable, consistent with the notion that firms with more volatile
noncontrolling interests than liabilities are more likely to disclose noncontrolling interests
in the mezzanine section. I also include the beginning-of-year total assets as a control for
the effect of firm size on firms’ classification choice. In addition, auditor dummies are
included to control for the possible influence of auditors’ opinion on firms’ classification

choice (Antle and Nalebuff 1991).26

 

26 Literatures argue that published financial statements should be read as “a joint statement from the
auditor and manager” (Antle and Nalebuff I991), indicating that auditors have significant influence on
firms’ financial reporting.

21

4.2 Credit market effects of the balance-sheet classification of noncontrolling
interests

I use credit rating to proxy for the market assessment of firms’ credit risk.
Previous studies find that a firm’s credit rating is positively associated with the
probability that its future cash flows will be sufficient to pay interest and principal in a
timely manner. For example, Altman (1992) documents that more than 90 percent of
defaulting issues are rated by S&P below investment grade (BBB) at least one year prior
to default. Hand et al. (1992) and Liu et al. (1999) find that bond return is positively
related to upgrades in bond rating. Together, these studies suggest that credit ratings are
an appropriate proxy for firms’ credit risk. To test H5, I therefore regress credit rating on
the balance sheet classification of noncontrolling interests (LIABILITY) and other control

variables:

RATINGi = yo + ylLIABILITY+ ngONTROLSi + 1),, (2)

where RATING is the long-term issuer credit rating assigned by Standard & Poor’s six
months after the fiscal year end. Note that the preceding discussion suggests that there are
differences in characteristics between firms that choose to disclose noncontrolling
interests in the liability section versus the mezzanine section. This suggests that
LIABILITY is endogenous and ordinary least-squares (OLS) estimation of Eq. (2) will not
provide consistent parameter estimates given LIABILITY is likely correlated with the
error term t). A common approach to deal with cndogeneity is to use an instrumental
variables estimator. Accordingly, I employ fitted probabilities from Eq. (1) as an

instrument for LIABILITY (Wooldridge, 2002, pp. 623—625).

22

I allow a six-month lag between the close of the fiscal year and the production of
credit ratings to ensure that rating agencies have sufficient time to incorporate the
information disclosed in firms’ annual reports into their credit rating decisions. The
ratings range from AAA (highest rating) to D (lowest rating — payment in default),
reflecting S&P’s opinion of an issuer's overall creditworthiness. Following Ashbaugh et
al. (2006), I recode the credit rating data (Compustat Quarter data SPDRC) to remove
unassigned codes and collapse the ratings into seven categories, with a higher value of
RATING indicating a better credit rating.27 Ordered probit and logit models are used to
estimate Eq. (2) because I cannot assume a constant difference in credit risk between
adjacent rating categories.

Consistent with H5, I expect a positive coefficient for LIABILITY, suggesting that
creditors assign high ratings to firms that classify noncontrolling interests as liabilities.
Following prior literature (e.g., Fisher 1959; Ziebart and Reiter 1992; Sengupta 1998; Shi
2003; Ashbaugh et al. 2006), I identify the following control variables that are associated
with credit ratings:

FAMILYFIRM as defined above, which is expected to be positively correlated with
credit ratings (Anderson et al., 2003),

LEV = ratio of total debt to total assets. The higher this ratio, the higher the firm’s default
risk and the lower will be its credit rating,

SIZE = natural logarithm of beginning-of-the year total assets. Larger firms are expected
to have lower default risk, and hence, higher rating,

 

27 Appendix A demonstrates the recoding methods employed by Ashbaugh et al. (2006) and Klock et al.
(2005). As a sensitivity check, I also recode the credit ratings following Klock et al. (2005) to allow credit
ratings to range from I (D ratings) to 22 (AAA ratings). Table C5 of Appendix C shows that the results
are qualitatively similar to those presented in Table 4.

23

CAP_INT EN = property, plant and equipment scaled by total assets. Capital intensity is
expected to be positively correlated with credit ratings,

INT_COV = operating income before depreciation divided by interest expense. This ratio
is expected to be positively associated with credit ratings,

ROA = income before extraordinary items divided by total assets. ROA is expected to be
positively correlated with credit ratings.

24

CHAPTER 5
SAMPLE AND UNIVARIATE ANALYSIS

For my determinants analysis, my sample includes 128 S&P 500 firms during
fiscal year 2004 with noncontrolling interests that have COMPUSTAT data necessary for
my analysis.28 Firms’ balance-sheet classification choice of noncontrolling interests is
manually collected from Form lO-K. Exhibit 1 (Exhibit 2) contains an illustrative
example of a firm that discloses noncontrolling interest in the liability (mezzanine)
section. Following the 2003 Business Week family firm classification, I identify 48
(37.5%) family firms in my sample.29 Of these 128 firms, 3] (24%) reported
noncontrolling interests in the liability section, while 97 (76 %) reported noncontrolling
interests in the mezzanine section.

In my analysis on the association between credit rating and the classification
decision, the sample includes 106 of the 128 firms with long-term domestic issuer credit
rating data (COMPUSTAT Quarterly data SPDRC). Comparable to the full sample, I find
that 26 of the 106 firms (25%) reported noncontrolling interests in the liability section
while 80 (75%) reported noncontrolling interests in the mezzanine section.

Panel A of Table 1 presents descriptive statistics on the two groups of firms as
well as p-values from tests of whether the means and medians of the two groups differ.

Although I use log-transformed values for total assets in my primary analysis, I report

 

2’ I do not model the decision of firms to have controlling interests in other companies. However, I did
compare S&P 500 firms with and without noncontrolling interests in terms of various firm characteristics
(i.e., FAMIL YFIRM, CONSTRAIN, IN V_GROW TH, SIZE, and ROA). Table CI of Appendix C shows that
these two groups of firms are not significantly different from each other in terms of the firm characteristics
I examine.

29 In their November 10, 2003 issue, Business Week identifies 177 (35.4%) S&P 500 firms as “family
firms”. Business Week classification is based on Anderson and Reeb’s (2003) methodology. Given the

25

raw values in this table for ease of interpretation. In univariate comparisons of the two
groups of firms, I find that FAMIL YFIRM, NONCONTROLLING, IN V_GROWT H, and
total assets are significantly different. The evidence suggests that family firms are more
likely to disclose noncontrolling interests in the mezzanine section, while firms with
higher litigation exposure to noncontrolling shareholders, larger firms, and growth firms
are more likely to disclose in the liability section.

Panel B of Table 1 presents descriptive statistics on the mean and median credit
ratings of the two groups of firms. Univariate tests provide some evidence that firms
classifying noncontrolling interests in the liability section have higher credit ratings than

firms classifying noncontrolling interests in the mezzanine section.

 

stickiness in family ownership (Ali et al. 2007), I feel that family ownership in 2003 would be a good proxy
for that in 2004.

26

CHAPTER 6

MULTIVARIATE RESULTS

6.1 Determinants of the reporting decision

In Panel A of Table 2, I report the correlations among the variables in Eq. (1).
Pearson (Spearman) correlations are in the upper right (lower left). Consistent with my
univariate descriptive statistics, LIABILITY is negatively correlated with FAMIL YFIRM,
and positively correlated with NONC ONT ROLLING and SIZE.

The results of the probit, logit, and OLS estimations are in Table 3.30 The
negative coefficient on FAMILYFIRM suggests that family firms are more likely to adopt
mezzanine classification. This is consistent with the argument that founding family
members are more likely to adopt classification that presents a favorable picture of its
financial position from the perspective of stakeholders other than noncontrolling
shareholders. The coefficient on NONCONTROLLING is positive and significant,
suggesting that firms with more economic resources controlled by parent firms relative to
potential claims by noncontrolling shareholders (i.e., “deep pocket”) are more likely to
disclose in the liability section. This is consistent with my argument that such firms face
higher litigation risk from noncontrolling shareholders, and thus are more likely to adopt
the liability classification to signal fair dealing and protection of the rights of
noncontrolling shareholders.

The coefficient on IN V_GROWTH is negative and significant, demonstrating that

growth (value) firms are more likely to include noncontrolling interests in the liability

 

3° 1 estimate Eq. ( I) in both raw values as well as ranked values for the independent variables. Table C2 in
Appendix C shows that the results are qualitatively similar.

27

(mezzanine) section. This result supports H3 that growth firms are more likely to adopt a
conservative accounting classification as part of a strategy to alleviate information
asymmetry problems between managers and shareholders, and thus lower their cost of
capital. The insignificant coefficient on CONSTRAIN does not support H4. With respect
to control variables, the coefficient on SIZE is positive and significant, indicating that
larger firms are likely to adOpt conservative accounting treatment. I do not find a
significant coefficient on VOLA T ILITY .

I use coefficient estimates to calculate the marginal effect of each independent
variables on LIABILITY to evaluate its economic significance (Wooldridge 2002, pp.
458—9). The marginal effect of FAMIL YFIRM indicates that family firms are 17.9
percentage points less likely to classify noncontrolling interests as liabilities, holding all
other variables at their mean values. In addition, the probability of liability classification
increases by 11.3 percent and decreases by 10.3 percent when moving from the 25th
percentile to the 75‘h percentile of NONCONTROLLING and IN V_GROW T H in the
sample. Overall, the marginal effects of family control, litigation risk, and information
risk from growth opportunities appear to be economically important.

To sum up, the empirical evidence in this section supports the hypotheses that
firms with greater exposure to litigation risk from noncontrolling shareholders and
growth firms are more likely to classify noncontrolling interests as liability. However,
family firms are more likely to adopt the mezzanine classification, reflecting more

serious controlling-noncontrolling shareholder agency conflicts due to family control.

28

6.2 Credit market analysis of reporting decisions

In this section, I examine whether the balance sheet classification of
noncontrolling interests is associated with credit ratings after controlling for the
endogeneity of firms’ classification choices and other factors affecting credit ratings. The
empirical analyses in this subsection are based on the instrumental variable methods
described earlier.

In Panel B of Table 2, I report the correlations among the variables in Eq. (2).
Pearson (Spearman) correlations are in the upper right (lower left). Consistent with H5 ,
LIABILITY is positively correlated with RATING.

Table 4 presents formal tests of H5. The coefficient on LIABILITY is significant
and positive, suggesting that firms classifying noncontrolling interests in the liability
section receive a higher credit rating. This finding is consistent with my hypothesis that
creditors are more likely to favorably assess the credit risk of companies that
conservatively classify claims on their balance sheet. This is because conservatism in
balance sheet classification may help alleviate downside risks through signaling restraint
on the part of controlling shareholders as well as providing early signs of deterioration in
firm performance. As expected, credit ratings are positively associated with firms’ size,
ability to cover interest (IN T_COV), and profitability (ROA). I do not find FAMIL YFIRM,
LEV, and CAP_IN TEN have significant impact on RATING in my sample.

Table 4 shows the results obtained from running ordered probit and ordered logit
regressions using fitted values from the probit and logit estimation, respectively, of Eq.
(1) as an instrument. Two-sided p-values are calculated based on bootstrap standard

errors using 1000 replications and are adjusted for heteroskedasticity (White 1980). By

29

assuming Eq.(l) and Eq.(2) are both linear, I also employ a ZSLS to estimate Eq. (2) as a
sensitivity check. One common reservation about using linear probability models in bond
rating studies is that the approach assumes the average credit rating difference between
any two adjacent ratings is the same. Several studies that compare the performance of the
linear probability model and the ordered probit model (e. g., Kaplan and Urwitz 1979;
Noreen 198 8) find that the conceptually-superior ordered probit model performs no better
than the simplistic linear probability model in terms of rating prediction. In addition,
Angrist and Krueger (2001) argue that 2SLS can typically capture an average economic
effect even for dummy endogenous regressors. Table 4 shows that the tenor of my results
does not change across the different estimation methods.31

While previous literatures argue that liability-equity classification has no effect on
stockholders’ valuation of firms (Kimmel and Warfield 1995; Cheng et al. 2003), I find
that firms’ classification choice matters to creditors. I think this is because prior studies
focus on whether the classification provides stock market with information about the
liability/equity nature of a corporate claim. However, my study focuses on the value
relevance of classification choice to creditors in the sense that it reflects firms’ desire to

use balance sheet conservatism to limit various agency conflicts.

 

3 ' To ensure that my results are not sensitive to the choice of control variables included in the credit rating
model, I also perform the estimation with no control variables in the credit rating model. Table C6 in
Appendix C shows that the results are similar.

30

6.3 Additional analysis

6.3.1 Firm characteristics

Consistent with prior research on conservatism (Givoly et al. 2007), I find that the
balance-sheet placement of noncontrolling interests is highly sticky over time (see
footnote 7).32 However, the measurement date of my independent variables so far is
driven by the calendar time (i.e., 2004) that I chose to collect information on my
dependent variable. To examine the sensitivity of my results to the measurement period
of the independent variables, I re-measure NONC ON TROLLING, IN V_GROWT H and
CONSTRAIN to proxy for firms’ litigation risk, growth prospectus, and financial
constrain status in a longer time horizon. Specifically,

NONC ON T ROLLING = natural logarithm of the average ratio of total assets to
noncontrolling interests calculated over the preceding five years for firm i,

IN V__GROWTH= one if firm i’s ratio of book value to market value of common equity
calculated over the preceding five years is lower than industry mean, and zero other
otherwise.

CONSTRAIN = one if firm i’s average free cash flows calculated over the preceding five
years is lower than industry mean, and zero other otherwise,

Table C3 of Appendix C shows that the tenor of my results is unaffected,
indicating that firms’ such consistent classification choice is not sensitive to the
measurement period of the independent variables. More important, this suggests that the

determinants I investigate represent certain stable firm characteristics that affect firms’

31

“sticky” classification choice. Such results are also consistent with the argument that the
degree of firms’ conservatism shouldn’t vary significantly from period to period because
it is determined by firm characteristics that are fairly stable over time (Givoly et al.

2007)

6.3.2 Earnings quality and liability classification choice

To the best of my knowledge, this is the first paper documenting that firms
employ liability classification of hybrid securities as part of their conservative financial
reporting strategies. Fields et al. (2001) argue that management will adopt a multi-
dimensional financial reporting strategy for the firm by jointly choosing a set of
accounting choices (Hagerman and Zmijewski 1979; Zmijewski and Hagerman 1981).33
In other words, given a single accounting choice is just one element of the optimal set of
financial reporting choices, I expect different accounting choices to be highly correlated.
In this section, I investigate relationship between firms’ classification choice of
noncontrolling interests and earnings quality.34

I predict that firms adopting liability classification are more likely to provide
higher-quality earnings. Given liability classification of noncontrolling interests is

considered as one of firms’ conservative financial choices, firms’ doing so may reflect a

 

32 Givoly et al. (2007) argues that the firm’s degree of reporting conservatism is stable because it is mostly
affected by firm characteristics that would not fluctuate significant from period to period.

33 Most of the extant literatures focus on one accounting choice at a time (Fields et al. 2001). Exceptions
are Hagerman and Zmijewski (1979), and Zmijewski and Hagerman (1981). Both papers simultaneously
consider four specific accounting choices (LIFO vs. FIFO, straight-line vs. accelerated depreciation, the
amortization period of past service pension costs, and flow-through vs. deferral method for the investment
tax credit) and analyze underlying managers’ incentives. However, they don’t provide any evidence on the
relationship among those accounting choices and only focus on income-statement accounting choices.

3" My measure of earnings quality is based on firms’ discretionary accrual which is considered to be a
summary measure of timing differences of all accounting choices (Watts and Zimmerman, 1990, p. I38).
Thus, I am investigating the relationship between firms’ balance-sheet classification choice and the
accounting choices affecting income statement.

32

demand for higher-quality accounting information as a whole for alleviating various
types of agency problem concerns. Consistent with this notion, empirical studies
document that different attributes of firms’ financial accounting information are highly
correlated. For example, Francis et al. (2004) investigates seven attributes of earnings
(i.e., accrual quality, persistence, predictability, smoothness, value relevance, timeliness,
and conservatism) and find that correlations among those attributes are positive and
significant.3'5

To examine the association between firms’ classification choice of noncontrolling
interests and earnings quality, I consider the two measures of earnings quality widely
used in the previous literature. Following Ashbaugh et al. (2006) and Kothari et al.
(2005), I alternatively measure Earnings Quality as the standard deviation of finn-
specific performance-matched discretionary accruals from the prior five years, where
discretionary accruals are calculated from a cross-sectional estimation of Dechow and
Dichev’s (2002) model modified by McNichols (2002), or from a cross-sectional
estimation of the Dechow et al. (1995) model. I estimate the regression model using the
population of COMPUSTAT firms at the 2-digit SIC code level for each year.

Table 5 presents descriptive statistics on the two groups of firms as well as p-
values from tests of whether the means and medians between them differ. In the
univariate comparisons, I find that Earnings Quality is significantly different, suggesting

that firms adopting liability classification have earnings of higher quality. This indicates

 

35 In stead of examining specific accounting choices, Francis et al. (2004) focus on summary measures of
different attributes of earnings. However, Fields et al. (2001) argue that both accounting choices and
earning management (or earnings quality) are used to influence the outcome of accounting information
system to achieve certain goals (e.g., alleviate agency problems).

33

that firms use multiple accounting choices to satisfy the demand for hi gher—quality

financial information to address agency problems (Fields et al. 2001).

34

CHAPTER 7
CONCLUSION

In this dissertation, I examine firms’ balance-sheet classification decision of
noncontrolling interests, and the economic significance of the classification choice from
the credit market standpoint. I find that firms with greater exposure to litigation risk from
noncontrolling shareholders (i.e., firms with “deep pockets”) and firms with more serious
information asymmetry problem (proxied by the extent of growth opportunities) are more
likely to classify noncontrolling interests as liabilities. However, consistent with more
serious agency conflicts due to controlling ownership, I find that family firms are more
likely to adopt the mezzanine classification. Contrary to expectations, my analyses did
not find that financially—constrained firms are more likely to adopt liability classification.
Moreover, I demonstrate that firms adopting liability classification choice have higher-
quality earnings, suggesting that it is employed together with other accounting choices to
satisfy the demand for higher-quality financial information.

After controlling for the endogeneity of the classification choice and various
determinants of credit risk, I find that on average, firms that classify noncontrolling
interests as liabilities have credit ratings that are around one notch higher than those of
firms that include them in the mezzanine section. This is consistent with my hypothesis
that creditors are more likely to favorably assess the credit risk of companies adopting
conservative financial reporting.

My results should be of interest to both standard setters and academia. First, my
study addresses the current debate about the classification of noncontrolling interests. To

my best knowledge, this is the first paper providing consistent evidence on the

35

determinants and informativeness of firms’ balance sheet classification choices if
noncontrolling interests. Second, my dissertation provides evidence for the debate over
whether the balance-sheet placement of information is relevant to investors. My credit
rating analysis indicates that accounting balance-sheet classification provides information
about the market’s assessment of firms’ credit risk. Thus, limiting accounting choices
could potentially eliminate relevant information from financial statements in situations in
which significant information asymmetry and agency costs may be present. Third, this
study contributes to the accounting choice literature. Fields et al. (2001) note that some
methodological limitations in previous work have hampered progress, for example,
failing to correct for endogeneity problems and failure to model multiple incentives
driving a reporting choice. My analyses and results provide support for their assertion
that resolving these issues can yield value to researchers studying firms’ accounting

choices and the consequences of those choices.

36

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

     

     

 

 
   

  

 

o o

o o

O

4 .

I EoEommSNE o e o o o o o o -

EoEowmcd—z «xmiommfizohaoox o o o o o o o o o o o

NoLEoO 8th SHE Evan "

Ea shaman .

p n

O

O

.5395 Duane. n

hoe—eaouaamfiowaaa—Z o

O

O

O

O

.5395 Gnome. "

mam:oh:3:oZ-u=:-ob=o0 meoEonoumnm 358m 0

whogonoumzm .. fa. .. .. . , ,. . o o o o o o O
. _ . I . .. Gaming—Em, w£=obzoOV.

.wE obzoocoz. . _
,. 3: _ ._ : meow—.onosamwhgmfiosm

 

8038a xoaowm 822085 we:Bungaocmfizobcoo ES :8an Gnome uoEononefiéomeSwE “mEoan twosome mo 39$ 03%
aeaa

37

 

$3.8 03.0

38.8 R _ .o
Owe. 83 x92 coxoozav

“we: 0388-93 Sm o:_m>-m BEES/H

 

 

 

83: :52 80.: Re:
83 :3 83. 8:.
382 as: 5E: as:
Aownzv a Gmuzvii i i i
RENEE: TEENS:

 

Goom are x85:
«8E DEE
seem a s @333
ems-3:. BEE

moEmE>

 

$8.6 ooHnZv 385:: mam—35:85:
How @2050 coceoEmmEo $085 voom So» Roma 5. mafia com gym 2: mo moumuflm oZEtomoQ ”GEES mcsfi H683 m 3.8m

 

98.8 58 mood 89¢ 53 3.3 ES: 39“
93.8 2:... an: $98 08.2 8&8 25:25 .aoa<
$2.8 Ed c Sec 0 mad EwEmwSo
82.8 ms... 82 0:3 22 $3 SEES E
92.8 5:. mm? 33. 83. 5% ezedofizoozoz
55.8 E; o 32 o 63 55: SEE
9mg gm an GOXOOZBV CE o 50 Cd— 0 ~80 mm Nam
58; 2983625 pom 2.37% $66-9: b 2 E b 2 2 3 . >
I I - Imonzv I, IAEWEI I
enigma: 33”:qu

 

 

 

 

AWE wmfiuzv $332: man—82828: 8m

853:5 bafifiam — finch.

86:0 cosmofimmflo mmouoa voom 80% Roam 3 menu oom mfim wfi mo moumufim 0351809 ”638% aommmoou coteofimmflov < 353

38

.Awmi 2339808 3:56:35 1338 SEE Awoma Hmwmzafioov 3:333
message Eofi $5: 5.8 3 amazon 0.8 $5: 58 BE .Amcofiu mwccmm 888.50 macm moomv £5ch mwcsmm 38380 {com 28 Banzai wEBozom a“

 

.Amoomv .3 Ho xoog mo 225:8 565250 2: no 3me 022.43” 8 BEE/coo mm Omamm .AOMQmma

Samsafioov can 89» 30mm 05 Stu 25:08 xmm mmoom ow pudendum 3 wocwmmmm worse :35 53mm atofiwaé H «oSMIDZEEN
.688 a s SEER .6 2328 sausage as 8 Ban BEE 3 causes a 95% .Gmfima

833808 use :8» Roma 2: Ste 262: 5m mzoom a. nausea 3 cocmmmmm marsh :38 53mm EoYmGE u awzeeiwIOZEE
.38» 26 @6805 05 E 83:52 :39

mo com§>ow Havana 05 E 838. memouowfi wfizobcooco: mo count/2o 3.858% 05 mo x52 26% Dam SEEN mm SE "ESE VINO;
..~ SE Sm .338 38 83¢?waafiwon 06. mo :3an2 Rea-m: "MNNW

4 SE no.“ 333 .8888 .«o 0:?» 98.38 9 029» xoon mo 0:8 ”ESQfivIE/Q

3.339850 5:8 83 use £80» 02% wcmwoooa 05 ~26 wean—3:8 950$ :98 RE owfiocé 33me me: .~ 8.6 b 28 HEEEWZQU
.N SE Sm Ewen—BE mam—35:85: 9 Somme :33 we 2:2 05 mo 85:..wa ESE: HOZNNNQKNZQUZQZ

.ommEofio 83 use .939: ummfiwzm 3 REESE mm SE 333% wcmcgom a mm .N SE a use ”EEK; SVEQ

.couoom vague 05 E .2 23 can couoom bane: 2: E 385:: mes—868:8 tome—“En .~ 85 a use “\ENNEV:

355.52. 033:;

~ 935—.

39

 

$8.8 $88 88.8 288$ :88 88.8
83 88 28.8- mod- 82.. ES- 88.8 83:33
£88 £88 888 92.88 3:88 88.8
m :8- 83 m2... 2:. mm. .o- a _ .8. 8:. 28%
63.8 288$ 28.3 68.8 88.8 82.8
088- a8... 88. _ 8a... 8:. 83- 38. 2302281822
388 38.8 2888 $8.8 $8.8 $2.8
M28- 888 82 83 8 8 38.8- 88- 282828
288$ $8.8 $8.8 $8.8 $8.8 88.8
82.. N88- :2 88 83 88.8 m8... 02382228202
9:8 €88 $8.8 $8.8 $8.8 €88
N88- 88. 82.- «8.8. 88.8 88. _ 82.. 3.2223382
88.8 :8... 88.8 9.28 68.8 98.8
:88 8:. 88.8- 8 8- m8... 88- 83 22283

2.2333 2N8 5885 822 2828209 958328202 5833i 3883

 

 

a $56 «NTZV mozafia> 8098885 Ea 802308 05 mo anE coca—880 A2958 586$ 838$mean < Edam
5.5a: noun—«tau N oEaH

40

 

$8.8 $8.8 83.8 $38 23.3

32.- 32.- 83.. Sod 8m... $62
$8.3 $2.8 $3.8 $2.8 285$ u
8.....- 5.? mod :3 83. \59 ES
$8.8 $8.8 $3.8 $2.8 E88 I
22. 85.8 E; $3- 83. 55: $0
:8.ch $8.8 $38 82.8 $223

2m... 2 3 83 $2.- $2.. :3
9?.8 $8.3 $8.8 638 3:8

ego- $2- 85¢. :38 ”8.8 ESESE
2: _.8 EH _ .8 $8.8 $3.8 83.8

82 £3- £3- E; 8:. mma.
38:3 $8.8 $3.8 $2.8 :8.ch I
mm»... SN... 92...- 33. En..- FESG ES

$5.8 $8.8 $8.8 28:3

83 SE SE. 8.2.. $3- 335.29g
35.8 $3.8 $8.8 $2.8

~25 83 35¢ 83 ~28- 82335209202
$58 $2.8 $8.8 638

m3? 838 83 SS- 83- ESE SEE
$2.8 $8.8 $2.8 $2.88

32.. 8.3 $2.. 83 3:. Egg:
38:2 $3.8 $8.8 $5.8

En..- $3- 8 S- E... 83 fizufiawawéwz

 

2~§5w2©9 UZSNQKNZQUZQZ 353m EEVR \ENSMSQ «wzcfiww DZNE

 

a $85 oSHZV 835E? 8095385 98 #89698 05 80 x59: noun—8.80 A0383 mass“ “688 m 35$
$888 N 28?

41

 

28:3 238 2......3 88.8 28.8 28:3

8.... at... .8... mm...- 8...... SN...- amv...- .8...
28:3 888 28....v. 23.8 88.8 28:3

E... 8.... 8.....- m:...- 2...... $2.- 83...- Agni-.2.
$3.8 $38 2.3.... $8.8 88.8 238 u
8...... 8.....- 8.... 8:... 5.... $8... 8...... 2882. m5
$8.8 28:3 82.8 $3.8 28.8 23.8

2.2.- 8.2.- 23 8.... .8...- owo... OS... 83
$8.8 $3.8 $8.8 .838 23.8 23.8

8...... .8... E... 8.....- 8.... 3......- E....- 8-8-5308
2. .8 88.8 23.8 5...... 22.8 83.8

8:.- ~.~...- 2...... E... 5.....- 8.... «fl... 8N8
2.....3 28.8 at... 28.8 2.3.... 88.8 I
8.....- ma...- .8... 08:. Q8...- :~... 8.... 583%. ES
$....8 23.8 88.... 28:3 23.8 2:8 28:3

88...- ..n~.... «.3... 22. NB...- mm; 3.... 2.883208
2.....3 $3.8 8%.... $2.8 28:3 88.8 28.8

«3.... 2 2.- 88:.- ..m... 8.2.- ..m....- 2:... 823388209262
2:8 :38 2%.... .58 $88 288 2.8.8

3.....- ....:. m3... m8...- ow.....- 2:.- mb....- 88:33:...
22.8 228 2.8.8 2. .8 $8.8 $3.8 2....8

8;... m8... 8:...- mm....- 2.... N3... 8.....- .3338:
2......3 2......3 2.88 2......3 $8.8 $8.8 28:3

3.... N3... 8......- Sm...- 8.....- m2... 83....- Eaéfiwnbéwx

.68 \BD .82. 252. $6 83 ES: 59. MN;- 58.8.6 82.

 

 

$.88. N 2...;

42

.. 0.8.08 8. 00800.. 0.0 003080., 808.0 ..<

.6... 80.088800. 808.0 .80. .3 000...... Q E. 888800. 080.. 808.800....0 0.0.0.. 08008. H 80%

Am... .80888000 080880 80.0.8. .3 .0003... Am 3... 88.88000 82.800800 0.0.0.. 08008. 880.080 H AQ0IE<~
.3... 80-88800. 8080 .80. .2. 00.08 S... 88.8800. 80888.00 .80 .80... £80808. .0 Z M 2m 8 2.1880

.6... 8088800. 80000 .80. 0. .00.. 888800 + 9.. 888800. 500 .80. -.0 0.8.8 M 808

"080.850.. 0.8.0.80.»

80000 .80. 8008-08.00-w8.88.w0n
.0 8:...0w0. .8308 0.... 0. 08m A0280»... .0>0. o . .o 0... .0 800....8w.m -.. 00.0.00 0. 83.0.0800 0088.80.08 8. 0.0 38.0.78 000.003.

8.3 880.0508 82.0.0800 8.80-0.80. 808.00% 80. .030. 0... 080 82.0.0800 80808808008. 80800.. 8.080. Em... .0888 08-.- 0

90.800. N 030,—.

43

3.3. 0.8.3. 0...0..m0000.wo.0.00 .0. 00.08.00 8080 08.0880 80 00000 00.0.8200 0.0 88.9-.. 000.0030 0

.8...0> 8008 ..0... .0 8.00..0> 80080800... .080 :0 0.8.0.0.. 0.8800 08 8.

.080 0. 0.00 80... 0.00..0> 08 .0 0.880808 8mm 0... 0. 0.8.00.2. 8mm 08 80.. 0.00..0> 80080800... 08 8. 08000 0 .0. 800000.880
30.00.. 0.8.8.000 .0 b...0000... 08 8. 08000 08 00 008.808 0. 90.0088. >888... 8.00..0> 8088.800 .0 800.0 .08....08. 00F

 

 

 

 

 

 

 

 

0
.. 01.0.; 8. 008.00 0.0 8.00..0> ..< a
0mm. .0 0..
00m... $8.... NM 00800..
mam... NS..- vmmd 03.0 3.0... mam... 03... N00... - 8.3.5.308
N3... $0... .30... 000... «um... M00... .3... N0 . .0 + MNE
R . .o .30- mom... own..- mum. .- 0mm... 3.0- M00... + 2.5.0.0200
03... 0mm..- 8.... 2.....- mbN.N- .3... 2.....- 0mm..- - 08.80.5182.
02.... 5...... 8...... m3... mum... 02.... 0...... m5... + 023008.200202
as... 8.....- .00... me... «an..- 03... 05...- 000...- - 3.5.... 0.... S.
0020?... 80.0.0000 U08.0?8. .MMVMMM“. 80.0....000 008.0?8. .MMMHWA 8060.000
80 :03 .38.
0080.). 80808.80”. 4 89m 8.0080 >
. . _ 00.0.08. 0 .
as... 8."...

080.800 30.00.. 8. 0.8.0.8. 8808800808 08.8.0 0. 808.000 08 .0 8808.88.00 0... .0 0.08.08... .9. ..0m. .0. 8.88. 80.0883.
m 0.00..

44

.§§ .353: bsuméemeefi
8: :83 Em 08 ES 326232 8:: mafia 28.5 @3983. 92580: no :83 3818—8 08 $33-: 3:663: o

.38: 58 2. 3 2a 3 E 3833 B s: 2 3%
AN: .:m :0 message sz 35:8 2.: E 52%waqu 8.: 6083ng 5 mm com: mm C: .:m :0 coumfismo :wE 80¢ 829’ 32.:
AN: .3. :0 559550 :38: 3on 05 E \EEEV: .8: :8ng ca mm. :8: mm 2: .cm :0 552580 :38: EB: 32.? BEL

.N 29¢ Ea ~ 2.5 a 8&2. as 85%? =<

 

 

 

 

45

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2:6 mm
:mmm: 33: -~ 353.—
vm:.: Gmé vm:.: 3.92 :m:.: 23.5 + ".62
b::.: 3:: 9:: :m:.: «3.: 3:: + kQUIKZN
m2: wow: .3:: m::.: wmw: Q ~.: + ZED/filmy»:
3:: «MN: m::.: mg: N::.: :3: + MN?
wmm: 0:2- Sm: :mmN- cam: m3.7 - Am:
2m: mmm: NS: 2:: :2: Eq.: + EEKNNNEVK
mm:.: :NNA ~::.: ~ 3.: ~::.: and + xhflm 3%
o “SEEN 3205000 cos—gé. E20580 ”029$ E20580
c mqmm 0 two: :08qu a :5on @2030
8502 Sumecmm =me . m93.2.5
:33:on
$8.6 8T2:

.3253 mm 336:: wfizobcooao: wfimofimmv :0 Soto gov—.88 :38 Mo 53850 28%? 3583me Am: .Um go: 3152 :Bmmouwom
v «Bah.

.325 58 an S 55 E .55 55:5me 3 a: 5 58mm

.652 338 865538532
8: 35:85 58 :8 58:82:58 83 :85: 88.5 :8355 @8580: 8 355: 3535—55 58 5525?: 325-03.: 5

 

 

 

 

 

 

 

 

 

 

 

46

 

 

 

 

 

 

 

 

 

 

 

 

AN: .:m .8 8:58:55 awn: 35:8 55 8 KEN‘NEV: 8: 858858 :5 mm 35.: mm 2: :m .8 8:58:55 2:2 88: 5589/ 38: N
Am: 6m .8 8:58:55 :88 35:8 56 8 8;:me 8: :58:me :5 mm :55: mm C: .:m .8 8:58:55 :88 80.: 553? 3:5 2
.m 535:. ::5 ~ 535:. 8 38.5: 55 55358; =< a
m? _ .o N:
:mmm: 33: NM 8:35.:
3:: Gm: 3:: $.82 :m:.: :3: + VQK
3:: 3:: mg: :m:: 2:: as: + :9: ES
m2: vow: «mo: 8:: wmw: Q g: + ZMRZNIKVU
2:: «MN: m::.: 3:: N::.: :mm: + MNE.
wmm: :3. T Sm: :mmd- :3: ~62- - Am:
2m: mmm: NC: 2:: :2: :3: + EEKxSEVK
mm:.: :34 3:: 3m: 5:: mg: + x5335:
5 5235‘ 8555500 55238. 855580 55:_5>-m 8555500
5 mqmm 5 two: 3580 .158: 35:8:
3.3.2 838:5 $6 a 553583,
3:335
588.: :oTZ:

523:: 55 $5558: was—0885.8: 85258: .8 55:3 5538 :38 .8 8:58:55 5388; 5:58:88: Am: :m 8.: 52:55: 855585”:
m 53.;

APPENDICES

Appendix A: Background

Pre—F AS 160 GAAP

FAS 160 requires that noncontrolling interests in subsidiaries be presented in the
consolidated balance sheet within the equity section, but separate from the parent
shareholders’ equity. As stated in F AS 160, the objective is to “improve the
comparability by eliminating that diversity” of noncontrolling interests classification in
the practice.

Prior to F AS 160, there is little official guidance on how to classify
noncontrolling interests: “ARB No. 51, Consolidated Financial Statements, and F ASB
Statement No. 94... are the prevailing authoritative GAAP rules on accounting and
reporting standards for consolidated financial statements.” (FASB, 1991, par. 14).
However, neither pronouncement offers a definition of noncontrolling interests nor
prescribes how to classify noncontrolling interest in financial statements. In addition to
those accounting guidelines, the SEC and FASB have promulgated standards (i.e., ASR
268 and FAS 150) that shed some light on the disclosure of mandatorily redeemable
noncontrolling interests.

In the pre-F AS 1 50 period, noncontrolling interests that are subject to mandatory
redemption or whose redemption is outside the control of the issuer were required to be
disclosed outside of permanent equity. In the post-FAS 150 period, noncontrolling
interests that “embody an unconditional obligation requiring the issuer’s redemption by
transfer of assets at a specified or determinable date (or dates) or upon an event that is

certain to occur” are required to be disclosed in the liability section (F AS 150). However,

47

noncontrolling interests that are outside the purview of FAS 1 50 could be classified in the

liability, mezzanine or equity section.
Debate on the balance sheet classification of noncontrolling interests

The FASB received forty-nine comment letters in response to the 2005 ED. The
majority of respondents favored the parent company approach to financial reporting over
the economic unit approach, and therefore, did not support the classification of
noncontrolling interests as equity. Specifically, they argued that since noncontrolling
interests do not have an ownership interest in the equity of the parent company, they
should not be classified as equity in the consolidated balance sheet. They also argued
that classifying noncontrolling interests in the consolidated equity section would affect
the calculation of financial and performance ratios, which could confuse users.

In contrast, the few respondents who agreed with the equity-classification of
noncontrolling interests (i.e., academics, accounting societies, certain preparers, and one
accounting firm) supported the economic unit concept as more appropriate than the
parent company concept. They pointed out that there is no support in the conceptual
framework for classifying noncontrolling interests as liabilities or mezzanine items and
that the holders of noncontrolling interests are owners of a residual interest in a

component of the consolidated entity.
Theories underlying classification of noncontrolling interests

The lack of agreement on a theory of consolidation and a clear definition of
liabilities and owners’ equity has led to variations in the reporting of noncontrolling

interests. The debate as to whether the appropriate basis of accounting should rely upon

48

the economic entity concept or the parent company concept has been flourishing for a
long time.” 38

At one end of the spectrum, the economic entity theory views the subsidiary as
part of the consolidated firm, and given that the investors representing the noncontrolling
interests are equity holders of the subsidiary, they are also deemed equity holders of the
consolidated firm (Paton 1922; Moonitz 1942 and 1960; Sapienza 1960; Hendriksen
1970). At the other end, the parent company theory evolved from the representative
viewpoint proposed by Husband (1938 and 1954), who argues that for the consolidated
entity, corporate entrepreneurs rather than the holders of noncontrolling interests are the
parent company’s equity claimants (Kohler 1938; AACS 1955). Hence, a noncontrolling
interest is an outside interest and should not be reported as an element of stockholders’
equity. Proponents have used this theory to argue that noncontrolling interest is a
liability, or should be presented in its own special category separate from the parent’s
equity.

Standard setters’ views on the adoption of the economic entity concept versus the

parent company concept have been mixed. ARB 43 (1953) provides support for the

economic entity theory. Chapter 7 of ARB 43 states that “the income of the corporat[ion]

 

37 The earliest reference is a presentation made by William M. Lybrand at the annual meeting of the
American Association of Public Accountants in October 1908 which was published in two parts in the
Journal of Accountancy in November 1908 and December 1908. Lybrand depicts “common Stock of
Subsidiary Companies Not Owned by the Holding Corp.” under a general heading of “Liabilities,”
following “Common Stock of the Holding Corp.” [November 1908, p. 40]. In part 11, Lybrand states that
“under capital stocks will be included the stock issues of the holding company and separately stated, such
part of the stocks of the subsidiary companies as are not owned by the holding company” [December 1908,
. 120].
E38 Differences of opinion were evident from the start. Newlowe (1948) examined 150 journal articles and
books from 1908 through 1945. Four authors preferred that noncontrolling interest be placed among
liabilities, and 28 preferred to classify noncontrolling interest as an element of stockholders’ equity. The
other 34 sources cited did not address the nature of noncontrolling interests.

49

is determined as that of a separate entity without regard to the equity of the respective
shareholders in such income.” (Section B, par. 6). However, ARB 51 states that “. . .the
purpose of consolidated statements is to present, primarily for the benefit of the
shareholders and creditors of the parent company, the results of operations and the
financial position of a parent company and its subsidiaries essentially as if the group were
a single company with one or more branches or divisions.” Even though ARB51 does not
discuss where to place noncontrolling interests on the balance sheet, the above statement
provides support for the parent company theory of equity. While the FASB’s stated
preference in the F AS 160 for classification under the equity section reflects its leaning
toward the economic entity theory, it also proposes that EPS calculations be based on
parent firm information, consistent with the parent company theory. Furthermore, the

F ASB’s Statement of Financial Accounting Concepts No. 6 does not give clear guidance
on the nature of noncontrolling interests. In sum, there is no clear theoretical consensus

that has emerged on the balance sheet classification of noncontrolling interests.

50

Appendix B: Credit ratings’ numerical conversions

 

 

S&P Letter Compustat Quarter Conversion Conversion
Rating Data SPDRC Number (Ashbaugh Number
et al. 2006) (Klock et a1. 2005)
AAA 2 7 22
AA+ 4 6 21
AA 5 6 20
AA- 6 6 19
A+ 7 5 18
A 8 5 17
A- 9 5 16
BBB+ 10 4 15
BBB 1 1 4 14
BBB- 12 4 13
BB+ 13 3 12
BB 14 3 11
BB- 15 3 10
B+ 16 2 9
B 17 2 8
B- 18 2 7
CCC+ 19 1 6
CCC 20 l 5
CCC- 21 1 4
CC 23 1 3
C 24 1 2
D 27 1 1

 

51

._ 29$ 5 Eamon 2“ 85%? E. _.

 

 

 

 

 

 

 

smug €75

32.8 83 33 83 mod ~86 wow
8?qu ca TzC

Game mm; 485 a? 83.0 $3 mma
samué a TB

88.8 32 c on; o 88 >23:sz
Ammmnzv @Sué I

938 8.8 $2 $8 335 82c >355 Es
Gmmuzv A3 2qu

its $3 0 83 o 93 SKEESE

93“ 8% VHS canoe—$6
H83 0383 5602 :82 SE52 =32 «mo—nmgxw
-95 com o=_m>-m quméBH
385:: wadzobcoocoq 32E? 335:: wcfiobcoonoc 5:5

 

 

 

 

$852: wanbaooaoq 8m 88:0 consommmmfio mmobm voom so» Roma E wee 8m mfim mo mosmcfim ozatomom
~ .0 035—.

mania—5 bgmmuom "U fleuoand‘

52

.Swa 823v gosmmwoxmotgo: com tom: .8“ Echo wawcfim no woman Bum—:38 8m moigd 32993. n
._ mam/2. E pounce 2a 8323:? a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8%... N:
£3... 38... N: 355
med de 30o vwod ovfio owed - \ENNEEQ;
wmmd mmod homd 3N6 ode Efio + MNE.
:86 Sad- om fl .o N _ h. T cmcd mead- + ZNSCWZIQU
v2.0 Rod- mad mmfid- wmfio wood- - IEQMU ES
mac... baa... $9: $5.: mweé baa... + UZ~<NQ~EZQDZQZ
ewe... v26- 53.: a: . 7 $6.: $9... - 3.3K» SEEK
037w>um HGOMOWMDOU PMS—grim “COMOWMOOU map—grim «$205000
30 z. 3 505
$6
Bea: Smaacmm 328$ smsfifi>
anew wm Tzv

228% base: E 385:: was—838:0: 328% 8 commmooc 2t .3 mgqmcmcbfiov 65 mo £9392 A: .cm How 3168 ~868ch 018%
N .U 03.3.

53

Aowo. 8.23. 5638:3858: Ho. 865.? £88 28ch co comm: 3.2523 08 $39»-.. 3.66-03... n

.. 2%... a. 88:0: 2a mozmflg .550 =<
3.3550 $50

83 can .538 @535. SE. 832 m. memo» 02.. winoooa 2.. .26 38.3.3 v.30... fine 02.. owfiog m... at: e: 28 H ZNEEMZQD

6.0.3328 88o Sow can .538 .5265 SE. 532
m. may. 02.. wfiwoooa on. 85 38.523 3.3.6 .8888 .o 02$ 86.38 3 033 xoo: mo 2.8 m... 8.5 .3 one HEEQNmuIAZN
A 82.. 8.. may. 02..

"mafia—Eon. 2:525»

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3m . .: Lfl
$2.: mew... wk 359mm
:9... 3...: mg... 0.5.: vow... mmod - NEE... .30;
m5: 5:: 3:: vum: 3:: m:..: + MNR.
:5: God- :3... EN... 3:... 2.....- + Z~§RM>SU
:8... $2 3:. RS- 23. :2 - :SQmwnsE
3:: gm: 3:: 3m. mm:.: 3:: + UZEQQKNZQUZQZ
3:: am _ .:- m3: 2:. T 3.: :wmé- - 35%. $3. 3.
002?-.. £06580 son—gd Sumo—too“. 0029.: E06500“.
mAO z. 0.. .3on
8:82 5385mm. :w_m $3.253
. . @8068: a .

 

 

 

 

GEE 3H7: Soto 8.63 8.. mam—05:8

30.6% .3233. a. $35.5 958.880: owe—ommv o. “SE8: 2.. .o 3.85.580: on. :o mix—«:4 H... .
m .U 03¢...

cm 8.. 9.38 commmohwom

wfiwooofi 2.. H26 wows—:28 385.5 mafiobcoococ 9 38mm .83 :0 0:8 aways on. :o gimme. .858: H UZNQQQMRZQDZOZ

w

54

3:2 853:
3653:5856: Ho: 3533 03 can mcoumozmou 0:3 wEms Bot» @5983 92308. no :83 683528 2m 8:15-: 3:66er o

.38: £8 2“ a: Ea 5 am mesa: B S: 2 3% u

AN: .5: mo :oflmfiumo :wS @8090 2: S. KEENMEN 5m EoEHEmE cm mm com: fl AC .:m :0 :oumfiumo :wS 80¢ 82$ BE“: 0
AN: .3: mo noumEcmo :5on @2238 on: 5 >583ng .8: :58:me an 3 :3: mm C: .3: we sausage :9on 8on 33g :35: a
.N DES Ea _ 22¢ a 8&2. em mosfig 3. a

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0mm H .o N:
33.: 82.: mg 359?:
3:: 3.0m: : S: :5: :3: ::m: + VO%
3:: :v:.: 33: 3:: m3: 3:: + :09 ES
1mm: mm: mm:.: 3?: wmm: :: m: + 29;: KN:
N::.: 3»: ~::.:v v9: ~:::v mg: + NEW
:8: m2}- :3: mSN- m2: wmvg- - ANN
SN: 3:: v3: 3:: vm:: mm: + 35.5: SEN:
:3: 3:: 3:: :v:.m ~::.:v :3: + x5853:
oofigd Eamofiooo cos—ST: E06580 eon—g: 3205000
o mamm u awe: 8.530 a :5on @2090
@0502 cocmaumm $5 a 333:;
:82on
$85.: 372:

3253: 3 9.8.5:: wE=obqooaoa mammofioflc mo
“coho 6me8 565 :0 832530 033:; 358:me ”Au—numb; E0253: 3 «89:196ka mama: AN: em 8.: $152 mommmouwom
v .U 935.

55

Sm comics 2m 98 228838 83 mi? Echo waccfim @5308 no woman 85:28 8m weigh onméz/w
.38: 58 as S Ea E E macaw“ 3 S: 2 3%
.ANV .wm mo aoumfismo :wB @0590 05 E \bbflmw: Ho.“ :58:me nu 8 com: mm 2v .wm mo :ozmficwo «ES 8on 338, ESE
.va .Um mo :oflaESmo :9on @8090 05 E NENEQVE Sm “58ng 5 mm wow: mm A: .v@ mo cocflbflmo H503 Boa 82g REE
A 035 a 8539 a ozacg oi

.SwS 8:53 bfismacoxmobg

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

chobcooao: mafia—8% mo Soto H858 :88 .«o sausage oEuflg 3585me ”65:8 8923 AS .cm 8% 33%.: nommmohwom

bass: 8 335%:

m .0 933—.

N:
823 98.0 885 «a £525
33. E: 3.2V £3 3...? 33. + $333:
333:; Eucaoc 3 «8% b2: $3
386 NM
£85 28o N: 25:.—
33. E2 SSV 23 33V 3% + his:
AOEEE “5225““ a ESE? 32E 3:
0 263-9 E20580 fl “weigh 2865000 _ Doggd _ E30580
c 3% _ 0 a3 8520 _ 3 BE 8520
8502 8:355 gym a 83m?»
BEBE
Gee 8T6

0'5

:3 .D

56

momnv

mcowmwzno 50:02 #5883058 98 :ommcom 835m

 

 

 

5th Bot SEVwcoA
nvmamfi 85:531— thFU Each
ow 26 32:53 3% govwcoq
~23 35:55: 3803‘
oatm Bambwa 3583‘.
ommJ mwcrsobon 9:8-tonm
bEwm whogoefim 98 82—534

m Show 3354 EB .H
2&6 38mm 550
EQN $0me Bflwcfifi
: 2: =E€oc
fimmnm mwomma taxi
VN: mavens 53 088E 235m
coo; mwommm wfiocacm 58830
mmflfl 333‘ 2.2.50 .33.
m3 33mm 3830 $50
3%; wwwocon 53 2:02: 835
wood 38on E 385:8 98 3:8235

:3 Ea $3 06 35.88 25:8 E
m ~ m6 85352—5 wo 85 oEmZooB 3558<
mwmflm Eco—«>33 :23 was :96

Bomm<
voom

80am 852mm 3328980 900 mflwo—ocnooh BED

5.53m bang: 2: 5 38.3.5 mum—3.5.395:— w:_mc_om_c ma «En—Sn...—

muEEaxu Eats—Emma? nooaméoaa—am "G flan—25¢.

57

 

 

 

 

 

 

mmodv N3.0.0... 300300035 .50 00.3...005 .30...
.50... 5.5.0m 30:30ouaam .30 H
5.03. 0. mowcwnu
Amm .. .v Eogooamsméo Z 0050 030.0058». .30 ...
mm. 0050 .
8%. 0 0:52 8.82. 8:832
SN 00.32833 .0503 $.20...
”3.0.00 :. mowcmno 3003003500: .850 0030.=§00<
6va $5.... 0000 8:380
$0: 3:05.. 8593.
€50. 500 E 885 006800 033.”. 0:0 NNm.mm . 0.005 >803...
02... 023.0
20.03 .00.N 80600 000mm. 303% ooodoqm 63.00530. m020> an. .m J03m 008800
m:.0033:0 00 000mm. 0007. $085 ooo.omm-00N..~0€0< .00.? .3. .m £005 02.00.05
0.85 .330
5.3m. 300300.80m
nmodm 8.3...005 .30...
0.0 3.0.2.800 50.0.33 0. 38030. 5.80.3.
Am. .000 v 83on 30:30.. “cognac“. .000 3025.880“.
$3. 30:53 €3-32 550

3.2.00. 00.33. 3:30.. 2.» a. 38.53:. an...0.=:00:0: 3:00.00... .0 0.0833

58

 

gamma—

EEEE 3.;

S: 8020 mflocomctcoo van 3898:8800

 

 

2..ch 33:52— ::2-w8o_ 580
3.? Bow 8.83-80;
coda fl . H 85.5.:— 82.89 .83.
8.2 $53 0888 wobomom
om“: _ $83 @888 menus—08 .8me
$.ch 8::sz 580 can Evian 388<
o_.m G 0333 $8508 068,—.
5% 8.8.82 mo
cm. s: .8an 8085 28 $830.88 88-825
”SE—Ba: 88.80
bgam .muoEoAv—oem 88 «3.88— bflcE—z £25531—
ow.wmo,m 30mm.“ .33.
8.3— 53880
3.3 853 @888 @28on
33% $0me 850
86$ 6: 808853 28 35895
Simon; 30mm.“ 82.89 .85.
onw: 38mm 8885 $50 EB 035382 mausooo<
3.8m; 380838 8888802
3.3 $828ng
owévm 3:0?353 8me can :80
”gamma 82.80
383‘
«com

Hoonm 8.83m BEE—8800 .2: £580 828cm

85on 08.8808 95 8 $8835 mam—.2888:— wfimfiummc a: 29883.—

59

 

 

 

 

 

 

owdmoflm m 3.5.00 0001.05.00: .80 000.58. N0.0088 .00....50: .80H
00.30“. 3850 80205.83 .30h
8.0mm $8800 0080.3.
3.3 08008 03080808800 0080 00.088306V
82: 8030800800 008003.
20% 230838 .38 “mm 093 0% 88 .2 hang a
.0 8.088080 0:0 .0030. 0082..m .thng .80 5065.? 60889.80 0089.0 ooodoodom a0.00.0 808800
238.00 800.08.008m
00.0mm. 30.58. margin .80 00.5.00: .30...
Ex. .0208. bugs».

 

2:800. 80.0000 08:35.8 0... 8 300808. w:...0.::00:0: $800.00... .0 0.980mm

60

Example of disclosing noncontrolling interests in the equity section (from Appendix

Assets

Liabilities:

Equity:

A of FAS 160)

ABC Co.

Consolidated Statement of Financial Position

As of December 31

Cash
Accounts receivable
Available-for-sale securities

Plant and equipment
Total assets

Total Liabilities

ABC Co. shareholders' equity:

Common stock, $1 par

Paid-in capital

Retained earnings

Accumulated other comprehensive
income

Total ABC Co. shareholders' equity

Noncontrolling interests

Total equity

Total Liabilities and equity

61

 

 

 

 

 

 

20x3 20X2
$ $
570,000 475,000
125,000 1 10,000
125,000 120,000
220,000 235,000
$ 1,040,000 940,?)00
$ $
555,000 459,000
200,000 200,000
42,000 50,000
194,500 167,000
22,500 16,000
459,000 433,000
26,000 48,000
485,000 481,000
$ 1,040,000 940,%00

 

BIBLIOGRAPHY

Accountants International Study Group. 1972. Consolidated financial statements. Current
Recommended Practices in Canada, the United Kingdom and the United States.

Ahmed, A., B. Billings, R. Morton, and M. Harris. 2002. The role of accounting
conservatism in mitigating bondholder-shareholder conflict over dividend policy
and in reducing debt cost. The Accounting Review 77: 867-890.

Ali, Ashiq, T.Y. Chen, and S. Radhakrishnan. 2007. Corporate disclosures by family
firms. Journal of Financial Economics (forthcoming).

Almeida, H., and M. Campello. 2002. Financial constraints and investment-cash flow
sensitivities: new research directions. New York University and University of
Illinois, Working Paper.

Altman, E. 1992. Revising the high-yield bond market. Financial Management 21 (2):
89-92.

American Accounting Association. 1957. “Accounting and reporting standards for
corporate financial statements: 1957 Revision”. The Accounting Review
(October): 536-546.

-------- . Committee on Concepts and Standards. 1955. “Consolidated financial statements,
supplementary statement No. 7”. The Accounting Review (April): 194-197.

-------- . 1964 Concepts and Standards Research Study Committee — The Business Entity
Concept. 1965. “The entity concept”. The Accounting Review (April): 358-367.

-------- . Financial Accounting Standards Committee. 1999. Liabilities and equity.
Accounting Horizons 13:305-307.

------- . 2001. Evaluation of the FASB’s proposed accounting for financial instruments
with characteristics of liabilities, equity, or both. Accounting Horizons 15:387-
400.

American Institute of Certified Public Accountants. 1956. “A survey of consolidated
financial statement practices”. P. 18.

62

------ . 1959. “Consolidated financial statements”. Accounting Research Bulletin No. 51.
-------- . 1994-2003. Accounting Trends & Techniques. New York, NY: AICPA.

-------- . 1953. “Reinstatement and revision of accounting research bulletins”. Accounting
Research Bulletin No. 43

Anderson, R.C., S. A. Mansi, and D. M. Reeb. 2003. Founding family ownership and the
agency cost of debt. Journal of Financial Economics 68 (2003): 263-85.

-------- and D.M. Reeb. 2003. Founding-family ownership and firm performance:
evidence from the S&P 500. The Journal of Finance LVIII (3): 1301 — 1328.

Angrist, J .D. and A. B. Krueger. 2001. Instrumental variables and the search for
identification: From supply and demand to natural experiments. The Journal of
Economic Perspectives (15) 4: 69-85.

Ashbaugh, H., D. Collins, and R. LaFond. 2006. The effects of corporate governance on
firms’ credit ratings. Journal of A ccounting and Economics 42: 203 - 243.

Aslan, H., D. Easley, S. Hvidkjaer, and M. O’Hara. 2007. Firm characteristics and
informed trading: Implications for Asset Pricing. Working Paper. University
Houston, Cornell University, and University of Maryland.

Ball, R. 2001. Infrastructure requirements for an economically efficient system of public
financial reporting and disclosure. Brookings-Wharton Papers on Financial
Services: 127-182.

Barclay, MJ. and CG. Holdemess. 1989. Private benefits from control of public
corporations. Journal of Financial Economics 25 (2): 371-395.

Barry, CB. and S]. Brown. 1984. Differential information and the small firm effect,
Journal of Financial Economics 90, 629-650.

Basu, S. 1997. The conservatism principle and the asymmetric timeliness of earnings.
Journal of Accounting and Economics 24: 3-3 7.

Beatty, Anne, J Weber, and J. Yu. 2006. Conservatism and Debt. Working Paper. The
Ohio State University, and Massachusetts Institute of Technology.

Bebchuk, L. 1989. Limiting contractual freedom in corporate law: the desirable
constraints on charter amendments. Harvard Law Review 102: 1820-1860.

63

Bemanke, B. and M. Gertler. 1989. Agency costs, net worth, and business fluctuations.
The American Economic Review 79 (1): 14-31.

, , and S. Gilchrist. 1996. The financial accelerator and flight to quality.
Review of Economics and Statistics 78:1-15

 

Botosan, C. A. and M. A. Plurnlee. 2005. Assessing alternative proxies for the expected
risk premium. The Accounting Review (January): 21-53.

Brudney, V. and R. Clark. 1981. A new look at corporate opportunities. Harvard Law
Review 94: 997-1062.

BusinessWeek, 2003. Family, Inc. November 10.

Campello, M. and L. Chen. 2005. Are financial constraints priced? Evidence from firm
fundamentals, stocks, and bonds. University of Illinois and Michigan State
University, Working Paper.

-------- . Chen and L. Zhang. 2005. Expected returns, yield spreads, and asset pricing tests.
University of Rochester, Working Paper.

Carlstrom, C. T. and T. S. Fuerst. 1997. Agency costs, net worth, and business
fluctuations: A computable general equilibrium analysis. The American Economic
Review 87 (5): 893-910.

Chami, R. 1999. What’s different about family business? Working paper, University of
Notre Dame and the International Monetary Fund.

Cheng, Q., P. Frischmann, and T. Warfield. 2003. the market perception of corporate
claims. Research in Accounting Regulation 16: 3-28.

Choi, D.S., Levich, R.M., 1990, “The capital market effects of international accounting
diversity”, Irwin Professional Pub

Cohen, L. 1993. Head of Brooke group draws on its coffers to tune of millions. The Wall
Street Journal (July 30): p. A — 2.

Cooper, 1. and S. Davydenko. 2004. Using yield spreads to estimate expected returns on
debt and equity. London Business School, Working Paper.

Cooper, W. W., and Y. Ijiri, eds. 1983. Kohler ’s Dictionary for Accountants. Sixth
edition. Englewood Cliffs, NH: Prentice Hall, Inc.

64

DeAngelo, H., L. DeAngelo and E. M. Rice. 1984. Going private: minority freezeouts
and stockholder wealth. Journal of Law and Economics 27 (2): 367-401.

DeFond, M. L. and C. W. Park. 2001. The reversal of abnormal accruals and the market
valuation of earnings surprises. The Accounting Review 76 (3): 375-404.

Denis, D. and D. Denis. 1994. Majority owner-managers and organizational efficiency.
Journal of Corporate Finance 1 (1): 91-118.

Dhaliwal, D. 1986. Measurement of financial leverage in the presence of unfunded
pension obligations. The Accounting Review 61: 651-661.

Dye, R., 2002. Classifications Manipulation and Nash Accounting Standards. Journal of
Accounting Research 40 (2002): 1125—62

Duke, J ., and G. Hunt. 1990. An empirical examination of debt covenant restrictions and
accounting-related debt proxies. Journal of Accounting and Economics 12: 45-63

Elton, E. J. 1999. Expected return, realized return, and asset pricing tests. Journal of
Finance 54: 1199-1220.

-------- , J. J. Gruber, D. Agrawal and C. Mann. 2001. Explaining the rate spread on
corporate bonds. Journal of Finance 56: 247-277.

Engel, E., M. Erickson and E. Maydew. 1999. Debt-equity hybrid securities. Journal of
Accounting Research 37: 249-274.

Fama, E. and MC. Jensen. 1983. Separation of ownership and control. Journal of Law
and Economics 26: 301-326.

Fazzari, S. M., R. G. Hubbard, B. C. Petersen, A. S. Blinder and J. M. Poterba. 1988.
Financing constraints and corporate investment. Brookings Papers on Economic
Activity 1: 141-206.

Fields, T.D., T.Z. Lys, and L. Vincent. 2001. Empirical research on accounting choice.
Journal of Accounting and Economics 31: 255-307.

Financial Accounting Standards Board (FASB). 1985. “Elements of Financial
Statements Concepts Statement No. 6. Stamford, CTzFASB.

-------- . 1987. “Consolidation of all majority-owned subsidiaries”. Statement of Financial
Accounting Standards No. 94 .

65

-------- . 1990. “Distinguishing between liability and equity instruments and accounting
for instruments with characteristics of both”. Discussion Memorandum.

-------- . 1991. “An analysis of issues related to consolidation policy and procedures”.
Discussion Memorandum.

-------- . 2003. “Accounting for certain financial instruments with characteristics of both
liabilities and equity”. Norwalk, CT: FASB.

-------- . 2004. FASB Project Update: Liabilities and Equity. www.fasb.org

-------- . 2005. “Consolidated financial statements, including accounting and reporting of
noncontrolling interests in subsidiaries Exposure Draft. Norwalk, CT: F ASB.

-------- . 2007. “Noncontrolling interests in consolidated financial statements Statement
of Financial Accounting Standards No. 160

Fisher, L., 1959. Determinants of risk premiums on corporate bonds. The Journal of
Political Economy 67: 217 — 237.

Francis, J ., R. LaFond, P. M. Olsson, and K. Schipper. 2004. Cost of equity and earnings
attributes. The Accounting Review 79 (4): 967-1010.

Gilson, .R. J. and J. Gordon. 2003. Controlling controlling shareholders. Working Paper
#228. Columbia law School, The Center for Law and Economic Studies, New
York.

Givoly, Dan, C. Hayn and A. Natarajan. 2007. Measuring reporting conservatism. The
Accounting Review 82 (1): 65-106

-------- , and --------. 2000. The changing time-series properties of earnings, cash flows
and accruals: Has financial reporting become more conservative? Journal of
Accounting and Economics 29: 287-320.

Gomes, J .F., A. Yaron, and L. Zhang. 2004. Asset pricing implications of firms’
financing constraints. University of Pennsylvania, Working Paper.

Graham, RC. and CE. Lefanowicz. 1999. Majority and minority ownership of publicly-
traded firms: a test of the value of control using market multiples. Journal of
Business Finance and Accounting 26 (1) & (2): 171-198.

Greene, W., 1998. LIMDEP, Version 7.0: User’s Manual. Econometric Software Inc,
Plainview, NY.

66

-------- . 2000. Econometric analysis, 4th Edition. Prentice Hall, Saddle River, NJ.

Hamada, R. 1969. Portfolio analysis, market equilibrium, and corporate finance. The
Journal of finance 24: 13-31

------- . 1972. The effect of the firm’s capital structure on the systematic risk of common
stocks. The Journal of Finance (May): 435-452.

Hazen, Thomas, 1977, Transfer of corporate control and duties of controlling
shareholders — common law , tender offers, investment companies-and a proposal
for reform, University of Pennsylvania Law Reuiew 125, 1023-1067.

Hand, J. R. M., R. Holthausen, and R. Leftwich. 1992. The effect of bond rating agency
announcement on bond and stock prices. Journal of Finance 47 (2): 733-753.

Healy, P., and K. Palepu. 1993. The effect of firms’ financial disclosure strategies on
stock prices. Accounting Horizons (March): 1-11.

Hendriksen, E. S. 1970. “Accounting theory”. Homewood: Richard D. Irwin.

Holdemess, C. D. and D. P. Sheehan. 1988. The role of majority shareholders in
publicly-held corporations: an exploratory analysis. Journal of Financial
Economics 20 (1/2): 317-340.

Husband, G. R., 1938, “The corporate-entity fiction and accounting theory”. The
Accounting Review: 241-253.

-------- . 1954. “The entity concept of accounting”. The Accounting Review (October):
552-563.

Jaffe, T. 1996. New issues in reverse. Forbes (June 17): 158-160.

Jensen, M.C., and W.H. Meckling. 1976. Theory of the firm: Managerial behavior,
agency costs and ownership structure. Journal of Financial Economics 3: 305-
360.

Johnson, M. F ., R. Kasznik, and K. K. Nelson. 2001. The impact of securities litigation
reform on the disclosure of forward-looking information by high technology
firms. Journal of Accounting Research 39 (2): 297-327.

Johnson, S, R LaPorta, F. Lopez-de-Silanes, A. Shleifer. 2000. Tunnelling. Discussion
Paper. Harvard University.

67

Kaplan, R., and G. Urwitz. 1979. Statistical models of bond ratings: a methodological
inquiry. Journal of Business 52: 231-261.

Kasznik. R., B. Lev , 1995, "To warn or not to warn: management disclosures in the
face of an earnings surprise," The Accounting Review, 113-134.

Kellogg, R.L. 1984. Accounting activities, security prices, and class action lawsuits.
Journal of Accounting and Economics 6: 185-204.

Kennedy, P., 1998. A guide to econometrics, 4th Edition. MIT Press, Cambridge, MA.

Kimmel, P. and T. Warfield. 1995. The usefulness of hybrid security classifications:
Evidence from redeemable preferred stock. The Accounting Review: 151-167

Klock, M., S. Mansi, and W. Maxwell. 2005. Does corporate governance matter to
bondholders. Journal of Financial and Quantitative Analysis (40) 4: 693-719

Kohler, E. L. 1938. “Some tentative propositions underlying consolidated reports”. The
Accounting review (March): 63-77.

Kothari, S.P., Lys, T., Smith, C.W., Watts, R.L., 1989. Auditor liability and information
disclosure. Journal of A ccounting, Auditing and Finance 4, 307-339.

LaFond, R. and R. L. Watts. 2006. “The information role of conservative financial
statements”. Working Paper. Massachusetts Institute of Technology.

Lamont, 0., C. Polk and J. Saa-Requejo. 2001. Financial constraint and stock returns.
Review of Financial Studies 14:529-554.

La Porta, R., F. Lopez-de-silanes, A. Shleifer and R. W. Vishny. 2000a. Agency
problems and dividend policies around the world. Journal of Finance 55: 1-33.

-------- . 2000b. Investor protection and corporate governance. Journal of Financial
Economics 58: 3-27.

Lee, Y.J. 2005. The market perception of employee stock options. Working Paper.
Georgia State University.

Linsmeier, T., C. Shakespeare, and T. Sougiannis. 2004. Liability/equity classifications
and shareholder valuation. Working Paper. Michigan State University.

68

Liu, P., F. Seyyed, and S. Smith. 1999. The independent impact of credit rating changes:
The case of Moody’s rating refinement on yield premiums. Journal of Business
Finance & Accounting 26 (3-4): 337-363.

Loss. L., J. Seligman, 1991, “Securities Regulations”, Aspen Law & Business

Maddala, G., 1986. Limited dependent and qualitative variables in econometrics.
Econometric Society Monographs.

Marquardt, C. and C. Wiedman. 2003. Earnings management through transaction
structuring: contingent convertible debt and diluted EPS. New York University,
Working Paper.

Maury, CB. and A. Pajuste. 2002. Controlling shareholders, agency problems, and
dividend policy in Finland. The Finnish Journal of Business Economics 1: 15-45.

Merton, R. C. 1973. An intertemporal capital asset pricing model. Econometrica 41: 867-
887.

-------- , 1987, “A simple model of capital market equilibrium with incomplete
information”, The Journal of Finance 42, 483-510.

Moonitz, M. 1960. “The changing concept of liabilities”. The Accounting Review (May):
41-46.

-------- . 1942. “The entity approach to consolidated statements”. The Accounting Review
(July):236-242.

Morck, R. A. Shleifer, and R. Vishny. 1998. Management ownership and market
valuation: An empirical analysis. Journal of Financial Economics 20: 293-315.

Murphy, R. 1956. Corporate divisions vs. subsidiaries. Harvard Business Review 34: 83-
92.

Myers, S. C. and N. S. Majluf. 1984. Corporate financing and investment decisions when
firms have information that investors do not have. Journal of Financial
Economics 13 (2): 187-221.

Newlowe, G. H. 1948. “Consolidated statements: including mergers and acquisitions”.
Boston: D. C. Health and co.

Noreen, E., 1988. An empirical comparison of probit and OLS regression hypothesis
tests. Journal of Accounting Research 26: 119 — 133.

69

Pacter, P. 1991. “Consolidation Policy and Procedures”. Discussion memorandum.
Norwalk, CT: FASB.

Paton, W. A. 1922. “Accounting theory”. New York: The Ronald Press.

Qiang, Xinrong. 2007. The effects of contracting, litigation, regulation, and tax costs on
conditional and unconditional conservatism: Cross-sectional Evidence at the firm
level. The Accounting Review 82 (3): 759-796.

Rogers, J. L. and P. C. Stocken. 2002. Credibility of management forecasts. Working
Paper. University of Pennsylvania.

Rubin, S. 1988. “Applications in accounting”. Journal of Accountancy 166: 123-126.

Sapienza, S. R. 1960. “The divided house of consolidations”. The Accounting Review
(July): 750-763.

Schipper, K. and A. Smith. 1996. A comparison of equity carve-outs and seasoned equity
offerings. Journal of Financial Economics 15 (1/2): 153-186.

Securities and Exchange Commission. 1979. “Presentation in financial statements of
‘redeemable preferred stocks”. Rule 5-02.28 of Regulation S-Xl

Sengupta, P. 1998. Corporate disclosure quality and the cost of debt. The Accounting
Review 73: 459 - 474.

Sharpe, W. F. 1964. Capital asset prices: A theory of market equilibrium under
conditions of risk. Journal of Finance 19: 425-442.

Shleifer, A. and R.W. Vishny. 1997. A survey on corporate governance. Journal of
Finance 52 (2): 737-783.

Skinner, D., 1994, “Why firms voluntarily disclose bad news”, Journal of Accounting
Research 32, 38-61.

Slovin, MB. and ME Sushka. 1997. The implications of equity issuance decisions
within a parent-subsidiary governance structure. The Journal of Finance LII (2):
841-857.

Stein, J. 1988. Takeover threats and managerial myopia. Journal of Political Economy
96: 61-80. -

70

Stein, J. 1989. Efficient capital markets, inefficient firms: A model of myopic corporate
behavior. Quarterly Journal of Economics 103: 655-669.

Stulz, R. 1988. managerial control of voting rights. Journal of Financial Economics 20:
25-59.

Verrecchia, R. 1983. Discretionary disclosure. Journal of Accounting and Economics
mecember): 179 — 194.

Wang, D. 2006. Founding family ownership and earnings quality? Journal of Financial
Economics 44 (3): 619-655.

Watts, R.L., 1993 A proposal for research on conservatism. Working Paper. University of
Rochester.

-------- . 2003a. Conservatism in accounting Part I: explanations and implications.
Accounting Horizons 17: 207-221.

-------- . 2003b. Conservatism in accounting Part II: evidence and research opportunities.
Accounting Horizons 17: 287-301.

-------- , and J. Zimmerman. 1986. Positive accounting theory. Chapter 8. Prentice Hall,
Englewood cliffs, NJ.

-------- . 1990. Positive accounting theory: a ten year perspective. The Accounting Review
65: 131-156.

Whited, T. M. and G. Wu. 2005. Financial constraints risk. University of Wisconsin and
University of Michigan. Working Paper

Wooldridge J. 2002. Econometric analysis of cross section and panel data. Cambridge,
Massachusetts: The MIT Press.

Zhang, J. Y. 2006. Efficiency gains from accounting conservatism: Benefits to lenders
and borrowers. Working paper. University of South California.

Ziebart, D., and S. Reiter. 1992. Bond rating, bond yields and financial information.
Contemporary Accounting Research 9: 252 -282.

71

..............

 

utjmtjijjjjj«ErrI