pihak relasi

8/17/2019 pihak relasi

1/45

Related Party Transactions and Earnings Management

Elizabeth A. Gordon* Rutgers University

Elaine HenryUniversity of Miami

Current Draft: November 2005

ABSTRACT

In this paper, we investigate whether related party transactions are associated with earnings management.If a firm's executives and/or board members engage in related party transactions to expropriate the firm’sresources, then they have incentives to manage earnings either to justify (or increase) these perquisites or

possibly to mask such expropriation. An alternative view is that related party transactions rationally fulfillother economic demands of a company, for example, by providing access to in-depth skills and expertiseor providing an alternative form of compensation. If this were the case, there would be no incentives tomanage earnings since the related party transaction need not be obscured or offset; consequently, wewould not expect a relation between earnings management and related party transactions. In our paper, wefirst provide a comprehensive empirical description of the types of related party transactions in whichcompanies engage and then investigate whether related party transactions are associated with earningsmanagement. We present a description of the number of, the parties involved with, the types of and whereavailable, the dollar amounts of related party transactions for 331 publicly-traded companies in fiscalyears 2000 and 2001. In our earnings management tests, we present evidence that adjusted absoluteabnormal accruals (our earnings management measure), are positively associated with limited types oftransactions such as fixed-rate financing from related parties. Overall, it appears that concerns aboutrelated party transactions as a factor associated with earnings management are warranted, but for onlycertain related party transactions. The mere presence of related party transactions is not necessarily anindication that a firm is likely to engage in greater earnings management.

* We thank Tom Stober, Ira Weiss, and participants at the 2003 Baruch College Four School AccountingConference, the 2005 annual congress of the European Accounting Association, and the 2005 AmericanAccounting Association annual meeting for helpful comments and suggestions. We are grateful for thesupport provided by the Whitcomb Center for Research in Financial Services at Rutgers Business School.

* Corresponding author: Elizabeth A. Gordon, Boutillier Endowed Faculty Scholar, Assistant Professorof Accounting, Rutgers Business School- Newark and New Brunswick., Department of Accounting andInformation Systems, Rutgers University, Janice H. Levin Building, Room 241, 94 Rockefeller Road,Piscataway New Jersey, 08854, USA; telephone 732-445-5849; email [email protected] .


2/45

1

I. Introduction

Recent corporate scandals have highlighted weaknesses in US corporate reporting and

accountability. One of the recurring areas of concern is related party transactions. 1 These

transactions are diverse and often complex business transactions between a firm and its own

managers, directors, principal owners or affiliates. 2 Users of financial reports view the existence

of related party transactions as an indicator of the increased likelihood of aggressive accounting. 3

In public accounting, related party transactions are considered difficult to audit and a potential

indicator of audit risk (AICPA, 2001; Johnstone and Bedard, 2004). 4 The General Accounting

Office (2003) identified related party transactions as one of the nine major reasons leading

companies to restate financial statements. Regulators, market participants, and other corporate

stakeholders commonly regard these transactions as potential conflicts of interest that can

compromise management's agency responsibility to shareholders or a board of director's

monitoring function. 5 Because they are transactions with firm insiders, RPTs present

1

Concern about related party transactions has increased as a direct result of at least two high profile bankruptcies:Enron, whose collapse has been attributable at least in part to extensive use of undisclosed or “under-disclosed”transactions with special purpose entities for which the CFO was also a director; and Adelphia, which engaged inextensive related party transactions such that the controlling family members’ dealings with the firm have beencharacterized as “using the company as the Rigas family's personal piggy bank, at the expense of public investorsand creditors." (AP newswire 7/5/02, AP staff writer Larry Neumeister). Both Enron and Adelphia includeddisclosure of related party transactions in their financial statements filed prior to bankruptcy.2 A related party can be an executive, a non-executive director, a principal owner or investor, a subsidiary, or a jointventure partner. Alternatively, the party may be a family member of, or a company owned by or affiliated with, anyof these related individuals. FASB Statement No. 57, Related Party Disclosures, as well as SEC reportingrequirements are summarized in Section II.3 Sherman and Young (2001) for example identify six “minefields” where aggressive accounting is most likely tooccur, one of which is “related party transactions, which can allow companies to arbitrarily increase or decreaseearnings.”4 The American Institute of Certified Public Accountants’ publication entitled “Accounting and auditing for related

parties and related party transactions”(2001) gives three reasons why related parties and related party transactionsare difficult to audit: 1.) transactions with related parties are not always easily identifiable, 2.) the auditor relies

primarily upon management and principal owners to identify all related parties and related party transactions and 3.)such transactions may not be easily tracked by a company's internal control. Johnstone and Bedard (2004) includethe existence of related party transactions in their audit risk measure. 5 Section 402 of Sarbanes-Oxley Law (SOX) limits the types of related party transactions in which companies canengage, prohibiting personal loans to executives and directors. Legislators, regulators and standard setters areadditionally considering even more rigorous controls on related party transactions. See "RESTORING TRUST,”Report to The Hon. Jed S. Rakoff, The United States District Court for the Southern District of New York, On


3/45

2

opportunities to expropriate firm resources. If a firm's executives and/or board members engage

in related party transactions to expropriate a firm’s resources, then they have incentives to

manage earnings to justify (or increase) these perquisites or possibly to mask extraction of the

firm's resources. In our paper, we examine whether the existence of and types of related party

transactions (RPTs) are associated with earnings management by the firm.

Theory suggests at least two alternative views of related party transactions, each with a

different implication for earnings management The first view, consistent with the portrayal of

related party transactions in the business press, is that related party transactions encompass

agency issues of the type considered by Berle and Means (1932) and Jensen and Meckling(1976). Executives and/or board members have incentives to manage earnings to justify (or

increase) these perquisites, or possibly to mask such expropriation. An alternative view is that

related party transactions rationally fulfill other economic demands of a company, or are a

mechanism that bonds the party to the company. For instance, related party transactions

potentially fulfill company demands such as the need for in-depth company knowledge and

expertise or the need for alternative forms of compensation. In these cases, there would be no

incentives to manage earnings since the related party transaction need not be obscured or offset;

consequently, we would not expect a relation between earnings management and related party

transactions. As a bonding mechanism, such transactions would tie the related party to the

company, decreasing incentives for the related party to engage in risk-taking behavior such as

earnings management that might jeopardize the company or the related party’s relationship with

the company.

Corporate Governance for the Future of MCI, Corporate Monitor, August 2003, Richard C. Breeden, From the SECwebsite (www.sec.gov.). Also refer to several business press articles that discuss these concerns including “BusinessTies: Many Companies Report Transactions With Top Officers” Wall Street Journal, December 29, 2003 A.1; "EvenGood Insider Deals Raise Doubts ” Wall Street Journal, May 7, 2003 B.6; and "Equity Office Faces Move onRelated-Party Deals" Wall Street Journal, May 14, 2003. C.13.


4/45

3

In this paper, we first provide a comprehensive empirical description of the types of related

party transactions in which companies engage and then investigate whether related party

transactions are associated with earnings management. In our description of related party

transactions, we present evidence on the number of, the parties involved with, the types of and

where available, the dollar amounts of transactions for 331 publicly-traded companies in both

fiscal years 2000 and 2001. By providing this description, our paper contributes to our

understanding of the nature of related party transactions.

Of the several interesting characteristics we document, we find that companies disclose

three related party transactions (at the median) but the number ranges from 0 to 25. Only 16% ofcompanies report none, while approximately 10% report 10 or more. Approximately 48% of all

disclosed transactions are with executives and 41% with non-executive board members. Loans to

related parties for reasons not specified as home or stock purchases are reported by the greatest

number of companies, with about 28% disclosing at least one, and this type of transaction is the

largest group of transactions, representing about 11% of all transactions. Fixed-rate financing

from related parties involves the greatest dollar amount of all transactions with a mean (median)

of $72.6 million ($2.1 million).

In our earnings management tests, we find evidence that adjusted absolute abnormal

accruals are associated with certain types of parties and with certain types of transactions. We

show that transactions involving fixed-rate financing from related parties, both the presence of

and the dollar amount of, are positively associated with adjusted absolute abnormal accruals. In

some cases, these transactions are considered related party transactions because a representative

from a bank or other creditor is a non-executive director of the firm. In other cases, the financing

is provided by officers of the company. (Appendix A provides examples of disclosures related to

fixed-rate financing from related parties.) If companies borrowing from related parties have not


5/45

4

obtained third-party financing, whether because they have been unable to do so or for other

reasons, there may be an incentive to manage earnings in order to obtain third-party financing in

the future and repay these loans.

Additionally, our findings suggest that earnings management is less prevalent in two main

cases: 1.) when companies report a related party transaction with an executive chairman or

executive chairman’s business; and 2.) when management services are provided. We offer two

related interpretations of these results. One explanation is that the transaction ties the related

party to the company, decreasing incentive to engage in risk-taking behavior. Another

explanation is that any benefits accruing directly to the related party as a result of the transactioncould mitigate incentives to manage earnings for the purpose of increasing wealth, for example

through bonuses or other compensation tied to earnings.

Overall, it appears that concerns about related party transactions as a factor associated

with earnings management are warranted, but only for certain related party transactions such as

fixed-rate financing from related parties. Our results do not show adjusted absolute abnormal

accruals are associated with certain types of (and amounts of) transactions such as loans to,

investment banking, real estate and purchases or direct service.

The remainder of the paper proceeds as follows: We summarize accounting and reporting

requirements for related party transactions in section II. We discuss the theoretical background

and related literature in section III. We describe our sample selection and data collection

procedures in section IV. In section V, we discuss our estimation of adjusted absolute abnormal

accruals and our regression model. We present an extensive discussion and descriptive analyses

of the related party transitions data in section VI. In section VII, we discuss regression results.

We offer conclusions and extensions in section VIII.


6/45

5

II. Accounting and reporting requirements for related party transactions

There are three main sources of accounting and reporting regulations on related party

transactions: FASB disclosure guidelines, SEC requirements, and the Sarbanes-Oxley Act of

2002 provisions. 6 FASB Statement No. 57, Related Party Disclosures, provides guidance for

disclosure of transactions with related parties. The relevant SEC financial statement regulations

are included in Regulation S-X, rules 4-08(k)(1) and (2), and the SEC non-financial statement

disclosure requirements are presented in Regulation S-K (Reg. §229.404. Item 404). The

Sarbanes-Oxley Act of 2002 contains enhanced conflict of interest provisions as well as

amended disclosure requirements. Each of these sources is summarized below.

FASB Statement of Financial Accounting Standards No. 57, Related Party Disclosures.

Related parties are defined to include three basic categories of individuals: board

members, executives, and principal owners (owners of more than 10 percent of voting interests

of the enterprise). In addition, the immediate family members of any of these categories of

individuals as well as any entities controlled by any of these categories of individuals are also

considered to be related parties. Related parties also include affiliates of the enterprise, where an

affiliate is described as “a party that, directly or indirectly through one or more intermediaries,

controls, is controlled by, or is under common control with an enterprise;” (para. 24) and,

specifically, entities for which the enterprise uses the equity method to account for its

investment. Examples include subsidiaries of the enterprise or of the enterprise’s parent

6 In addition, auditing for related party transactions is covered in various Statements on Auditing Standards (SAS)and related interpretations. In audit planning, for example, the existence of related parties and related partytransactions is highlighted as a “condition that may require extension or modification of audit tests.” SAS No. 22,Planning and Supervision, (AICPA, Professional Standards, vol. 1 AU sec 311.03) as referenced in AICPA 2001.Since the focus of this paper is on disclosure for related party transactions, auditing guidelines are not included inthis discussion.


7/45

6

company, and trusts such as pension trusts managed by or under the trusteeship of the

enterprise’s management. Related parties are also defined by SFAS 57 to include “other parties

with which the enterprise may deal if one party controls or can significantly influence the

management or operating policies of the other to an extent that one of the transacting parties

might be prevented from fully pursuing its own separate interests” (para. 24).

Other than compensation and similar arrangements in the ordinary course of business,

transactions with related parties must be disclosed in the financial statements. The disclosure

must include the nature of the relationship, as well as the nature and value of the transaction.

SEC Regulation S-X. 4-08(k) Related party transactions, which affect the financialstatements.

Rule (1) requires related party transactions and amounts to be identified on the financial

statements. Rule (2) requires “in cases where separate financial statements are presented for the

registrant, certain investees, or subsidiaries, separate disclosure shall be made in such statements

of the amounts in the related consolidated financial statements which are (i) eliminated and (ii)

not eliminated. Also any intercompany profits or losses resulting from transactions with related

parties and not eliminated and the effects thereof shall be disclosed.”

SEC Regulation S-K (Reg. §229.404. Item 404.) Non-financial statement disclosurerequirements .

SEC Regulation S-K requires disclosure of the following relationships and related

transactions:

Nature and Amount of transactions exceeding $60,000 with directors (existing andnominees), executives, owners of more than five percent voting interest, and immediatefamily members of any of the foregoing persons.


8/45

7

Nature and Amount of certain business relationships with an entity owned or managed by adirector (or nominee), including: sales to, purchases from, borrowings from, legal orinvestment banking services from the related entity.

Indebtedness of a related party to the enterprise, including amounts owed by: any director orexecutive officer of the registrant, or their immediate families; any corporation or

organization (other than the enterprise or a majority-owned subsidiary of the enterprise)which a director or executive owns more than 10 percent or serves as an executive officer or partner.

Nature and amounts involved in transactions with Promoters.

Note however, that related party indebtedness has been significantly changed by the Sarbannes-

Oxley Act of 2002.

Sarbanes-Oxley Act of 2002

Section 402 of the Sarbanes-Oxley Act of 2002 includes enhanced conflict of interest

provisions, which generally provide that it will be unlawful for an issuer to extend credit to any

director or executive officer. Financial institutions may extend credit for home improvement,

charge cards or securities trading so long as these extensions of credit are made in the ordinary

course of business, of a type generally made available by the institution to the public, and on the

same general terms and conditions available to the general public.

The Act also requires directors, executives and principal owners (of 10 percent) to report

ownership within 10 days of becoming a director, executive or principal owner; and to report

changes of ownership within two business days of the transaction.

III. Theoretical background and related literature

Although commonly portrayed as potential conflicts of interests, theory suggests two

alternative views of related party transactions. The first, consistent with the portrayal of related

party transactions in the business press, is that related party transactions encompass agency


9/45

8

issues of the type considered by Berle and Means (1932) and Jensen and Meckling (1976).

Jensen and Meckling (1976, 90) characterize the agency conflict between a manager and outside

shareholders as the manager's tendency to expropriate the firm's resources for personal

consumption, similar to perquisites. As commonly viewed both in the business press and

accounting standards, related party transactions represent the potential for the expropriation of

the firm's resources.

If a firm's executives and/or board members engage in related party transactions to

expropriate perquisites, then they have incentives to manage earnings to justify (or increase) the

perquisites or possibly to mask such expropriation. This view is consistent with Schipper’s(1989) definition of earnings management as a “purposeful intervention in the external financial

reporting process, with the intent of obtaining some private gain (as opposed to say, merely

facilitating the neutral operation of the process).” An alternative view is that related party

transactions rationally fulfill other economic demands or are a mechanism that bonds the party to

the company. For instance, say a director possesses an in-depth knowledge of firm-specific

activities and an expertise that the company demands such as legal expertise. Then it could be

more effective for the company to engage the related party to provide the service than an

outsider. 7 If this were the case, there are no incentives to manage earnings since the related party

transaction need not be obscured or offset; consequently, we would not expect a relation between

earnings management and related party transactions. Similarly, if related party transactions bind

the related party to the company, there would be less incentive to manage earnings. For instance,

say an executive’s son is employed by the company. Knowing that both his and his son’s

7 Fama (1980) and Fama and Jensen (1983) assert that the optimal board composition should include both inside(executive) and outside (non-executive) board members. The inside board members bring in-depth knowledge andoutside members bring independence and monitoring skills. Bushman et al. (2000) also consider the industry-specific expertise of the outside director as beneficial to the firm and its board. We consider the implications ofengaging a non-executive director to perform other services or provide goods.


10/45

9

livelihoods depend on the company, the executive may engage in less risk-taking behavior such

as earnings management that might jeopardize the company or his (and his son’s) relationship

with the company. Or, say a board member also has a substantial consulting arrangement with

the company. If any risk-taking behavior such as earnings management occurs, both the board

seat and consulting contract are at risk of being terminated. As a bonding mechanism, a related

party transaction would provide limited or decreased incentives for earnings management.

These differing views highlight that the underlying nature of RPTs is complex. A

presumption by standard setters, though, is that transactions involving related parties are not

carried out on an arm's length basis. Some may not have occurred or may have occurred on

different terms with another party (SFAS 57, paragraph 13). Regulators and accounting standard

setters have primary regulated RPTs through disclosure, assuming that:

information about transactions with related parties is useful to users of financialstatements in attempting to compare an enterprise's results of operations andfinancial position with those of prior periods and with those of other enterprises. Ithelps them to detect and explain possible differences. Therefore, informationabout transactions with related parties that would make a difference in decision

making should be disclosed so that users of the financial statements can evaluatetheir significance (paragraph 18).

To address the concern about these transactions not occurring at arm's length, many

companies disclose that their contracts with related parties have been made on terms at least as

favorable as with unrelated parties. If a firm undertakes related party transactions which enhance

-- or at least do not harm -- the economic interests of the shareholders, there should be no

negative impact on the firm and no reason to manage earnings. However, given the existence of

costs other than contracting costs, this disclosure does not necessarily mean the firm has been

unaffected by the related party transaction. There are also several other potential economic costs

to the firm of related party transactions including monitoring costs, opportunity costs and the

costs of reporting complexity.


11/45

10

Little research has been conducted to further our understanding of related party

transactions. Gordon, Henry, and Palia (2004b) find that related party transactions are generally

associated with weak corporate governance mechanisms and lower market returns. In a

concurrent study, Kohlbeck and Maydew (2004) find that certain related party transactions are

associated with executive compensation while other are not. Other concurrent research examines

related party transactions in companies listed in Hong Kong. Institutional features of the Hong

Kong market and firms traded on the market, such as highly concentrated ownership and owners

in mainland China, imply controlling owners could engage in related party transactions

opportunistically. Cheung, Rau and Stouraitis (2004) find that excess returns at theannouncement of such transactions are related to ownership percentage of the controlling

shareholder and to proxies for information disclosure. Ming and Wong (2003) provide evidence

that controlled companies report higher levels of related party sales when they have incentives to

inflate earnings to avoid being delisted or prior to issuing new equity. These inferences about

related party transactions based on a sample of Hong Kong listed companies do not necessarily

apply to other markets. Agency costs in the U.S. are expected to be greater because of dispersed

ownership. Market regulations and oversight also vary. Additionally, aside from controlling

ownership, these papers examine limited other corporate governance mechanisms in place.

Research on board composition uses the presence of related party transactions to classify

non-executive board members as "affiliated" or “grey” directors (Klein, 2002a; Vicknair,

Hickman and Carnes, 1993; Hermalin and Weisbach, 1988). Affiliated directors typically are

viewed as non-independent, outside directors. Klein (2002b) investigates the relation between

earnings management and both audit committee independence and board independence. Based

on the audit committee's role as arbiter between management and auditors to weigh and broker

divergent views of both parties to ultimately produce a balanced, more accurate report, Klein


12/45

11

(2002b) hypothesizes that an independent audit committee is best able to serve as an active

overseer of the financial reporting process. She predicts that audit committee independence will

be negatively related to earnings management. Using a sample of 692 observations of firms with

audit committees listed on the S&P 500 in 1992 and 1993, she finds that both audit committee

and board independence are negatively related to earnings management by the firm and

concludes that boards structured to be more independent are more effective in monitoring the

corporate financial reporting process.

Our study is related to Klein (2002b) in that related party transactions with non-executive

directors are one type of related party transaction we examine. Our scope is broader than andextends Klein (2002b) by also investigating transactions with other key parties in the company

including executives and principal owners. Where Klein (2002) motivates her study based on the

monitoring role of the board, we also investigate the agency conflict between a manager and

outside shareholders. Additionally, our study not only looks the existence of the related party

transactions but also comprehensively examines related party transactions with all parties

involved, by the types of related party transactions and by the dollar amount of the transactions,

where available. Similar to Klein (2002b), our maintained hypothesis is that related party

transactions are associated with earnings management.

IV. Sample selection and data collection procedures

Sample selection

Because we aim to assess related party transactions for a broad group of publicly-traded

firms, we select our sample from the population of companies on COMPUSTAT. We stratify our

sample by size (quintiles of market value) and industry to ensure comprehensive coverage. We


13/45

12

define industries following Fama French (1997). 8 We restrict our sample to those companies

with sales, income, market value and other key data available for the 2000 and 2001 fiscal year. 9

We exclude financial companies due to the nature of regulation and because it is difficult to

define abnormal accrual firms in this industry. We examine two years of RPT disclosures, 2000

and 2001. We select these two years because of increased scrutiny of company's financial reports

and transactions in 2001 following the Enron debacle unfolding in 2000. Our sample contains

331 companies (662 observations) representing approximately 7.5% of eligible COMPUSTAT

companies, for which we present descriptive statistics in tables 1 through 5.

Identification and characterization of related party transactions

We search through a company's proxy statements and 10-Ks for related party disclosures.

RPTs are most commonly found in proxy statements, often being incorporated by reference in

the 10-K. RPTs are described in the 10-K under Item 13, “Related Transactions." We also

examine any footnote(s) to the financial statement titled “Related Party Transactions.” In the

proxy, RPTs are usually described under “Certain Relationships and Related Transactions” or

“Certain Company Transactions with Management” or “Certain Transactions.” Family

relationships are noted in the biographical descriptions of board members and named executives,

which are included in the proxy.

We characterize related party transactions along four main dimensions: the primary party

involved, the secondary party involved (if any), the type of transaction, and the amount of the

transaction, if disclosed. 10 Parties involved are identified first by their relationship with the firm

8 Because we estimate industry-adjusted abnormal accruals, we also exclude small industries with than 50observations.9 In our final sample, we include only those observations with complete data for all variables. We limit our sampleto companies with fiscal year-ends in December or January.10 These categories are similar to those that were used in Gordon, Henry, and Palia (2004a, 2004b).


14/45

13

such as executives, non-executive board members, principal owners, subsidiaries or other.

Because we are interested in potential conflicts of interest on the management side and

impairment of the monitoring function on the board side, we track executives and non-executives

separately and further categorize these groups. Within the executive group, we separate

transactions with an executive chairman, executive board members and non-board executives.

We track executive chairman as a separate group because Jensen (1993) asserts that boards are

less effective when the chief executive officer is also the chairman. Similarly, having a larger

percentage of inside executives on board could impair monitoring. Within the non-executive

group we separately identify non-executive board members and a non-executive chairman.If transactions involve a family member of, or a company owned by or affiliated with,

any of these related individuals, we identify both parties. The party having the most direct or

most senior relationship with the firm is identified as the “primary related party." The family

member, or company owned by or affiliated with the primary related party, is the secondary

related party. We group secondary related parties by executives, executive's business, non-

executives, non-executive's business, principal owners, subsidiaries, and other. When the

transaction is directly between a primary related party and the company (for example, with a loan

from the company to an executive), no secondary party is involved. Within the executive group

and executive's business, we separate the secondary party involved in the transactions by

executive chairman, executive board members and non-board executives. Within the non-

executive and non-executive's business groups we separately identify non-executive board

members and a non-executive chairman.

The third dimension we identify is the type of transaction. We identify seven main types

of transactions: direct service between related parties or the related party and the company,

purchases of goods or contract services acquired from the related party, sales to the related party,


15/45

14

loans to related parties, fixed-rate financing from related parties, investment and other. Within

type, we report the different kinds of transactions disclosed by companies. Examples of direct

service transactions are an executive chairman of the board hiring a relative in a non-executive

position, a non-executive director being directly employed by a principal owner or a relative of

an executive serving on the board. Contracted services acquired from the related party include

management services, legal services, marketing, real estate, accounting, investment banking, and

other. We also identify transactions involving purchases of goods from the related party. We

classify loans into those to the related party for houses and stock purchases, where specified, or

other, where another purpose is given or none is identified. Such loans to related parties are now prohibited under Sarbannes-Oxley. We separately identify fixed-rate financing from the related

party, which can be in the form of loans or preferred stock, sometimes with conversion features

or warrants. The “other” category includes other types of transactions such as shareholder

agreements and shared R&D. We do not include the existence of employment agreements or

indemnification agreements, which some companies disclose as related party transactions, since

these items are unambiguously compensation.

The final dimension we identify is the amount of related party transactions. For loans,

investments and single transactions, we collect principal amounts; we code annual amounts

where the transaction is ongoing or a multiple year involvement such as a contracting service.

We give examples of reported related party transactions disclosures and our classification in

Appendix B.

Variable definition – related party transactions

Because related party transactions are diverse and often complex business transactions,

we investigate different measures of a firm's related party transactions and their complexity.


16/45

15

Unlike audit committee (or board) independence, where a logical measure is the number of

independent directors scaled by the size of the committee (or board), a firm-specific scalar for

related party transactions is less obvious because there is no limit on the number of transactions

of any type and with any related party in which a firm can engage. Therefore, we examine

various measures of a firm's related party transactions including: the total number of different

related party transactions; whether a firm has a related party transaction with a specific type of

primary or secondary related party, or a specific type of transaction; and the amount of disclosed

transactions, where available.

We also measure the overall complexity of a firm's related party transactions. We believethat more parties and more types of transactions indicate extensive and pervasive potential

conflicts of interest and monitoring issues. To do this, we count the number of unique related

parties and types of transactions. For example, say a firm has three different transactions, all with

a single executive: 1.) a relative is also employed by the firm, 2.) the executive received a loan

from the firm, and 3.) the firm leases real estate from a trust owned by the executive. The

number of primary related parties is one. The number of secondary related parties is three. The

number of different types of transactions is three. Each of these is a gauge of the complexity of

transactions. 11

V. Adjusted abnormal accruals and regression model

We use adjusted abnormal accruals as our measure of earnings management, similar to

Klein (2002b). To calculate abnormal accruals, we begin by estimating a cross-sectional version

of the Jones (1991) expected accrual model for all firms j in industry k for year t . The model is:

11 In gauging complexity, our objective is to capture the notion of the firm’s “web” of related party transactions. Asanother measure of complexity, we multiply the number of primary related parties times the number of secondaryrelated parties times the number of different types of transactions. However, we find this measure is highly andsignificantly correlated with the number of related party transactions so do not undertake tests using it.


17/45

16

t jk jk,t t jk t k

jk,t t jk t k jk,t k,t jk,t

e ,1,,

1,,11t jk,

)TA/PPE(

)TA/Δ REV()(1/TAα/TAACCR

++

+=

−

−−−

γ

β ( 1)

Where:ACCR jk,t is the total accruals for firm k in industry j in year t, defined as net income less

operating cash flows.TA jk,t-1 is total assetsΔ REV jk,t is the change in net salesPPE jk,t is gross property, plant and equipment.

We calculate the unadjusted abnormal accrual for each firm-year jk,t using the year and

industry specific coefficients estimated in Equation 1:

)]TA/PPE(

)TA/Δ REV()(1/TA[α-/TAACCR AAC

1,,

1,,11,

−

−−−

+

+=

jk,t t jk t k

jk,t t jk t k jk,t k,t jk,t jk,t t jk

γ

β (2)

Where:

AAC jk,t is the abnormal accruals for firm j in industry k in year t .

As Kothari et al (2005) note, a firm's AAC is associated with its earnings process. They

recommend performance adjusting AAC to control for firm performance and earnings process.

We follow Kasnick (1999) and Klein (2002b) and use a matched-portfolio method to adjust the

absolute value of the AAC. Because our study is related to issues of independence and Klein

(2002b) adjusts the absolute abnormal accrual by the median absolute value of abnormal accruals

of industry portfolios based on the standard deviation of past accruals, we also adjust the AAC of

each sample firm by the median absolute value of the AAC for a portfolio of firms matched by

the standard deviation of past total accruals.


18/45

17

To get percentile rankings of past total accruals for each industry, we compute the

standard deviation of total accruals for all COMPUSTAT firms with non-missing data during the

previous 10 years and assign them to percentiles based on their ordered rank within their

industry. We do not restrict the firm to be in the sample each of the ten years. We also compute

the standard deviation of total accruals for our sample firms for the past 10 years. If 10 years of

past accruals are not available, we use the years available. Each sample firm is then matched by

percentile, p, within industry k . The AAAC is computed using Equation 3:

kt,p p jk,t j,t Abs(AAC)MED-)Abs(AACAAAC ,=

( 3)

We use AAAC as the dependent variable in examining the association between related

party transactions and earnings management. If earnings management exists, we expect to find a

positive and significant relation between the adjusted absolute abnormal accruals and related

party transactions.

In the regressions, we include additional variables to control for other factors found to be

associated with earnings management. Not controlling for such factors could result in falsely

rejecting the null hypothesis of no abnormal accruals when the null is true (Klein, 2002b; Bartov

et al, 2000). Prior research finds past profitability is positively associated with earnings

management; therefore we include the absolute value of the change in net income, Abs( Δ NI), as

a measure of changes in profitability. To capture expected growth, we include the market value

of common equity divided by the book value of common equity measured at the beginning of the

year, MV/BVA. We also include operating cash flows, OCF, and an indicator variable for

negative earnings to control for other properties of earnings and accruals. Prior research finds

that leverage is positively associated with earnings management (DeFond and Jiambalvo, 1994).


19/45

18

We include long-term debt divided by beginning of the year total assets, Debt, as a measure of

leverage. Since political costs are negatively associated with earnings management, we include

size, Log(Assets), as a measure of political costs. The general form of the association model we

examine is:

et

+++

+++Δ++=

)Log(AssetsDebt

NI Neg.OCFMV/BVA NI)Abs(*RPTAAAC

65

4321

γ γ

γ γ γ γ β α ( 4)

Where:

AAAC is the absolute value of adjusted abnormal accrual measures as the Abs(AAC) minus

the Median Abs(AAC) for a portfolio of firms matched on the standard deviation of past accruals by industry.

RPT* is the measure or measures of related party transactions.Abs( Δ NI) is the absolute value of the change in net income divided by lagged total assets

between years t and t-1.MV/BVA is the market value of common equity divided by the book value of common

equity measured at the beginning of the year.OCF is the current year's operating cash flows divided by beginning of the year total asset.

Neg. NI is an indicator variable equally one if the firm reports negative net income in thecurrent or prior year and zero otherwise.

Debt is the long-term debt divided by beginning of the year total assets.Log(Assets) is the natural log of total assets.

Firm and time subscripts are omitted. We estimate annual regressions and summarize results.

Our significance tests on the RPT variables are one-sided since we predict a positive association.

We use two-sided tests for other variables. We substitute various measures of related party

transactions as discussed above.

VI. Descriptive analyses of related party transactions

Table 1 describes the industry composition of our sample. Our sample proportionately

represents the population of eligible COMPUSTAT firms. However, because we limit our


20/45

19

sample to larger industries, we have relatively more firms in each industry. Table 2 presents

descriptive statistics on summary related party transaction variables. The mean (median) number

of related party transactions is 4.042 (3). Untabulated analysis indicates that 21% (31%) of

companies decrease (increase) the number or related party transactions they report from 2000 to

2001 by an average of 1.9 (1.8) RPTs. The remaining 48% of companies report no change in the

number of RPTs. However, we do not assess how the composition of their RPTs changes.

To assess the complexity of transactions by the primary related party, we calculate the

number of transactions per individual primary related party by company. The mean (median)

number of transactions per primary related party is 2.82 (2). Another gauge of complexity is thenumber of different primary related party groups (i.e., executives, board members) engaged in

transactions. On average (at the median), companies have related party transactions with 1.79 (2)

different types of primary related parties and the range is from 0 to 6.

We examine the complexity of dealings with secondary related parties in a similar way.

The mean (median) number of transactions per unique secondary related party type is 2.29 (2).

On average (at the median), companies have related party transactions with 2.05 (2) types of

secondary related parties. Finally, we evaluate the complexity by type of transaction, finding that

companies have a mean (median) of 2.19 (2) number of different types of transactions out of the

18 different types we identify.

Table 3 presents additional detail on the number of RPTs per company and primary

related party, and Table 4 presents additional detail on the secondary related party. Table 5

focuses on the types of related party transactions.

Table 3 presents the number of related party transaction per company and by primary

related party. The 331 companies (662 observations over two years) in our sample report a total

of 2,641 different related party transactions (1,299 in 2000 and 1,342 in 2001). Approximately


21/45

20

84% percent of observations report at least one RPT, with number per company ranging from 0

to 25. Untabulated analysis indicates that firms with no RPTs and those with at least one RPT do

not significantly differ (at the mean and median) in their 1) absolute value of the change in net

income (Abs( Δ NI)), 2) market-to-book ratio (MV/BVA), 3) operating cash flows (OCF), 4)

leverage (Debt), or 5) size (Log (assets)). A significantly higher percentage of firms with at least

one RPT report negative net income (neg. NI) compared to those with no RPTs (62% versus

51%). Approximately half of the companies report three or more RPTs. Of the RPTs,

approximately 47.7% are with executives and 40.9% with non-executive board members. The

remaining transactions are with principal owners, subsidiaries or other affiliates.Table 4 describes the number of transactions by primary and secondary party. The first

column gives the percent of companies reporting a transaction with the given secondary related

party. Approximately 81% of companies have at least one transaction where the transaction is

directly between the company and a single related party such as a loan to an executive or

engaging a director to provide management services; these transactions represent approximately

44% of all transactions. Transactions with non-executive directors' businesses are the next

highest group of secondary related parties with 15% of companies reporting at least one,

followed by transactions with principal owners and executive chairmen’s businesses.

Table 5 shows the types of transactions by primary related party. Loans to related parties

where the purpose is not specifically identified as home or stock purchases are the most

prevalent with 28.1% of companies reporting at least one, representing 11.2% of all RPTs. The

mean (median) amount of the loans is $9.2 million ($ 0.263 million). Home loans are also

common, reported by 22.0% of companies, representing 10.6% of all RPTs. Loans specifically

identified for stock purchases are reported by only 5.2% of companies. Transactions involving

fixed-rate financing from related parties are reported by 19.1% of companies and represent 7.4%


22/45

21

of all RPTs. The mean (median) amount of fixed-rate financing from related parties is $72.6

million ($2.1 million). Approximately 25.0% of companies, representing 9.5% of transactions,

report a direct service relation between an executive and another party such as a relative directly

employed by the company or serving on the board. Few companies specifically report the dollar

amounts associated with these transactions, usually remuneration paid to the executive’s relative,

except for directors fees, where applicable.

Contract services include a number of different types of business transactions such as the

company acquiring management services, renting real estate or purchasing goods from the

primary related party. Contract services account for approximately 29% of related partytransactions, with real estate and purchases transactions making up almost half of these.

Purchases from related parties represent the highest dollar amounts with the mean (median) at

$6.046 million ($.504 million). Real estate transactions are the next highest with mean (median)

of $0.929 million ($0.600 million).

Table 6 reports descriptive statistics for accruals variables and correlations between key

variables used in regressions. We find no evidence of a systematic upward or downward bias in

abnormal accruals as the mean of the AAC is not significantly different than zero (untabulated p-

value = 0.19). Therefore, our use of the absolute value of abnormal accruals is appropriate.

Pearson correlations in panel B indicate that the number of related party transactions is

significantly and positively correlated with Debt and Log(Assets), suggesting that more

leveraged and larger companies engage in more of these transactions. As expected our control

variables for the earnings process and size are significantly correlated with adjusted absolute

abnormal accruals, AAAC. However, the correlation between adjusted absolute abnormal

accruals, AAAC, and leverage is positive but not significant.


23/45

22

VII. Regression results

Main results

Table 7 presents results of regressions of adjusted absolute abnormal accruals on total

number of related party transactions and on transactions by primary related party. We do not find

a significant association between the number of RPTs per firm and adjusted absolute abnormal

accruals. Similarly, we find no significant association between our metrics of RPT complexity

from table 2 and earnings management (untabulated). 12 This lack of significant results suggests

that earnings management is not related to the extent or complexity of a company's RPTs.

Next, we substitute an indicator variable for whether or not a firm has at least onetransaction with a certain type of primary related party (See the "percent of companies with

transaction" reported in table 3). Of the different groups of primary related parties, the

coefficient on RPTs with an executive chairman is significantly related to abnormal accruals.

However, given an assumption that firms with executive chairmen have weaker corporate

governance and thus are more likely to exhibit earnings management, the relation is

unexpectedly negative.

In regressions substituting an indicator variable for the different groups of secondary

related parties, presented in table 8, the estimated coefficients on RPTs with an executive

chairman and an executive chairman's business are significantly negative, again implying less

earnings management. This finding is interesting in light of results in the primary related party

regressions. Taken together, these findings suggest earnings management is less prevalent in

12 We also estimate the regression using different variables to capture the number of related party transactions, alsofinding no significance on these variables. Because of few observations with a large number of related partytransactions, we truncate the number at 12. We also use an indicator variable equaling one if the number of related

party transactions is greater than the median or 3 and zero otherwise.


24/45

23

companies that have RPTs with executive chairmen. 13 One explanation is that the executive chair

is bonded to the firm and has limited incentive to engage in risky behavior such as earnings

management. Another explanation is that if the executive chairmen benefit from the transactions,

they have limited personal incentives to manage earnings to increase wealth through bonuses tied

to earnings or incentive-based stock compensation (Healy, 1985). 14

The estimated coefficient on principal owner as a secondary party is also negative and

significant, suggesting less earnings management. An explanation consistent with less earnings

management is that the principal owner, especially as an involved and related party, has access to

detailed firm-specific information. If the principal owner is involved with the company and hasaccess to management, managers' incentives and ability to manage earnings may be lower,

consistent with the principal owner playing a monitoring role.

Table 9 presents results substituting an indicator variable for the different types of

transactions and table 10 present results for the dollar amounts. 15 We find significantly positive

13

Dechow, Sloan and Sweeney (1996) find that firms manipulating earnings are more likely to have a chiefexecutive officer who is also chairman. However, Dechow, Sloan and Sweeney do not specifically examine otherdealings the chairman has with the companies. In addition to our focus on related party transactions, the difference

between our results and those of Dechow, Sloan and Sweeney could be attributable to sample selection or theearnings management measure used. Dechow, Sloan and Sweeney’s sample is composed of 92 firms (plus controlfirms) subject to enforcement actions related to earnings manipulation by the SEC between 1982 and 1992. Oursample is from the general population of companies in a later time period.14 To provide additional insight into the negative association between earnings management and RPTs with anexecutive chairman, we examine detailed characteristics for a sub sample representing 85% of our sample firms. Weobserve that approximately 52% have an executive chairman and 33% have a RPT with the executive chairman (orhis business). Comparing firms with versus without RPTs with an executive chairman, we note: CEOs hold asignificantly greater percentage of outstanding equity in those firms with such RPTs, mean (median) of 10.7%(5.5%) compared to 4.1% (1.0%) for those firms without; a founder continues to be involved in 38.3% of firms with

such RPTs, compared to 28.0% without, and in approximately 30% of the firms where a founder continues to beinvolved, he is the executive chairman. These observations suggest that characteristics of the executive chairmancould mitigate any potential agency conflict in a related party transaction and thus limit risk taking behavior. Whilesuggestive, further research is warranted as the associations between governance characteristics and firm activity areoften complex and mulitfactoral. Additionally, performance and operating characteristics of those firms with suchRPTs do not significantly differ (at the mean and median) from those without in their 1) absolute value of thechange in net income (Abs( Δ NI)), 2) market-to-book ratio (MV/BVA), 3) operating cash flows (OCF), 4) leverage(Debt), 5) size (Log (assets)), or 6) percent of firms with negative net income (neg. NI).15 Where dollar amounts are available, we use as our measure the log of the total dollar amounts of the type oftransactions. If a company does not have a certain type of transaction, the amount is zero. If a company has atransaction but no dollar amount associated with it, we exclude the observation.


25/45

24

coefficients on the indicator variable and the dollar amounts of fixed-rate financing from related

parties suggesting companies that obtain fixed-rate financing from related parties are more likely

to manage earnings. Companies obtaining funds from related parties have not obtained third-

party financing, either because they were unable to do so or for some other reason, and instead

have relied on related parties. In such cases, there may be incentive to manage earnings in order

to obtain future financing and repay these loans. 16

Results also show significant negative estimated coefficients on the indicator variable and

dollar amounts of management services, implying less earnings management. 17 Additionally, we

find a negative and significant estimated coefficient on other types of transactions (coefficient = -0.033). Because this category includes various types of transactions, it is difficult to assess what

is driving the relation.

In the regressions in tables 7 though 10, the estimated coefficients on Log(Assets) and

OCF are consistently significant as anticipated.

16 To support this explanation, we find that the firms with financing from related parties exhibit weaker performanceindicators than those firms that do not have these transactions. They have (at the mean and median) significantlyhigher leverage (Debt), lower market-to-book ratios (MV/BVA), lower operating cash flows (OCF), and lowerreturn on assets (at the median only). Additionally, a higher percentage reports negative net income (neg. NI).17 To provide additional insight into the negative association between earnings management and management-services RPTs, we examine detailed firm and governance characteristics for a sub sample representing 85% of our

sample firms. Because the majority of management services are provided by board members, we focus on boardcharacteristics. Comparing firms with versus without management-services RPTs, we do not observe significantdifferences in board size, percent of insiders and whether there is an executive chair. Additionally, performance andoperating characteristics of the two groups of firms do not significantly differ. Those firms with a management-services RPT provide significantly less direct director compensation and fewer provide directors options. Theseobservations on director compensation suggest that the management-services RPT could be a substitute form ofcompensation. If providing a form of compensation, the transaction could bind the RP to the firms limitingincentives to engage in risk-taking behavior such as earnings management. While suggestive, further research iswarranted as the associations between governance characteristics and firm activity are often complex andmulitfactoral.


26/45

25

Additional tests

As a sensitivity check to our related party variable, we substitute the number of

transactions with primary related parties, secondary related parties and types of transactions

instead of an indicator variable for the party involved or type of transaction (as reported in tables

7 through 9). Results of the primary related party transaction equation are similar: we find a

negative and significant estimated coefficient on RPTs with executive chairman. In the

secondary related party transaction regression, we find that the coefficient on executive chairman

remains negative and significant while the p-value on that of the executive chairman’s business

drops to 0.12 (untabulated). The estimated coefficient on principal owners is not significant.Results for the types of transaction equation are also similar: we find a negative and significant

coefficient on management services and a positive and significant estimated coefficient on fixed-

rate financing from related parties. The estimated coefficient on other types is not significant.

VIII. Conclusions and extensions

Our paper offers two primary contributions to the literature. First, we provide a

comprehensive description of related party transactions for a broad sample of companies. In our

description of related party transactions, we present evidence on the number of, the parties

involved with, the types of and, where available, the dollar amounts of transactions for 331

publicly-traded companies in fiscal years 2000 and 2001. By providing this description, our

paper enhances the understanding of the nature of related party transactions.

Our second contribution is investigating whether related party transactions are associated

with earnings management. We find evidence that earnings management, as measured by

adjusted absolute abnormal accruals, is associated with certain types of transactions such as those


27/45

26

involving fixed-rate financing from related parties. An explanation for these results is that when

firms have obtained financing from related parties rather than third-parties, either because they

were unable to do so or for some other reason, there may exist incentives to manage earnings in

order to obtain future financing and repay the related party. Additionally, our results indicate a

negative association with adjusted absolute abnormal accruals when a firm reports a related party

transaction with an executive chairman or the executive chairman’s business. One explanation

for these results is that the related party transaction binds the chairman to the company, inducing

less risk-taking behavior such as earnings management. Another related explanation is that any

benefits of related party transactions accruing directly to an executive chairman could limit personal incentives to manage earnings for the purpose of increasing wealth, for example,

through bonuses or other compensation tied to earnings. We also find that the provision of

management services is associated with less earnings management. Again, perhaps those related

parties that also provide management services to a company are bound to the company, thus

limiting incentives for risk-taking behavior such as earnings management.

Perhaps just as interesting as the evidence that aspects of related party transactions are

associated with earnings management, we find no evidence of earnings management for certain

transactions. Our results do not show adjusted absolute abnormal accruals are associated with

certain types of (and amounts of) transactions such as loans, investment banking, real estate and

purchases, or with direct service. Overall, it appears that concerns about related party

transactions are warranted, but for only certain related party transactions. The mere presence of

related party transactions is not necessarily an indication that a firm is likely to engage in greater

earnings management.


28/45

27

References

American Institute of Certified Public Accountants, 2001. Staff of the American Institute ofCertified Public Accountants. Accounting and auditing for related parties and related partytransactions.

Bartov, E., Gul, F.A., Tsui, J.S.L., 2000. Discretionary-accruals models and audit qualifications.Journal of Accounting and Economics 30, 421–452.

Berle, A.A. and G.C. Means, 1932. The Modern Corporation and Private Property. New York,Commerce Clearing House, Loose leaf service division of the Corporation Trust Company,.

Bushman, R.M., Q. Chen, Q., E. Engel, and A.J. Smith, 2004. Financial accounting information,organizational complexity and corporate governance systems. Journal of Accounting andEconomics, 37 (2), 167-201.

Cheng, Y., P.R. Rau and A. Stouraitis, 2004. Tunneling, propping and expropriation: Evidencefrom connected party transactions in Hong Kong. Journal of Financial Economics, Forthcoming.

Dechow, P. M, R.G. Sloan, R. G. and A.P.Sweeney, 1996. Causes and consequences of earningsmanipulations: An analysis of firms subject to enforcement actions by the SEC. ContemporaryAccounting Research 13, 1-36.

DeFond, M. and J. Jiambalvo, 1994. Debt covenant violation and manipulation of accruals.Journal of Accounting and Economics 17, 145-176.

Fama, E., 1980. Agency problems and the theory of the firm. Journal of Political Economy 88,

288-307.

Fama, E. and K. French, 1997. Industry costs of equity. Journal of Financial Economics 43, 153-93.

Fama, E. and M. Jensen, 1983. Separation of ownership and control. Journal of Law andEconomics 26, 301-25.

Financial Accounting Standards Board. Statement of Financial Accounting Standard No. 57.1982. Related Party Disclosures.

Gordon, E. A., E. Henry and D. Palia. Related party transactions in corporate governance. Advances in Financial Economics , Volume 9 (2004a): 1-27.

Gordon, E. A., E. Henry and D. Palia. Related party transactions: associations with corporategovernance and Firm Value. Working paper, Rutgers University, 2004b .http://ssrn.com/abstract=558983.

Healy, P., 1985. The effect of bonus schemes on accounting decisions. Journal of Accountingand Economics 7, 85-113.


29/45

28

Hermalin, B. and M. Weisbach, 1988. The determinants of board composition. Rand Journal ofEconomics 19, 589-606.

Jensen, M. and W. Meckling, 1976. Theory of the firm: managerial behavior, agency costs and

ownership structure, Journal of Financial Economics 3, 305-360.

Johnstone, K. and J. Bedard. 2004. Audit firm portfolio management decisions. Journal ofAccounting Research 42, 659- 690.

Klein, A., 2002a. The economic determinants of audit committee independence, The AccountingReview, 77, 435-452.

Klein, A., 2002b. Audit committee, board of director characteristics, and earnings managementJournal of Accounting & Economics; 33, 375-400

Kohlbeck, M.J. and B.W. Mayhew. Related party transactions. Working paper, University ofWisconsin, 2004. http://ssrn.com/abstract=591285

Kothari, S.P., Leone, A.J., Wasley, C.E., 2005. Performance matched discretionary accruals.Journal of Accounting and Economics, 39, 163-197..

Ming, J.J and T.J. Wong, T.J. Earnings management and tunneling through related partytransactions: Evidence from Chinese corporate groups. EFA 2003 Annual Conference Paper No.549. http://ssrn.com/abstract=424888

Schipper, K., 1989. Commentary on earnings management. Accounting Horizons, 3 (4), 91.

Sherman, H. D. and S. D. Young, 2001. Tread lightly through these accounting minefields.Harvard Business Review: July-August.

Vicknair, D., K. Hickman, K.C. Carnes, 1993. A note on audit committee independence:Evidence from the NYSE on "grey" area directors, Accounting Horizons, 7, 53-57.


30/45

29

Table 1: Sample Industry Composition andSummary of Related Party Transactions by Industry a

Related Party Transactions

IndustrySample

Companies b Sample

CompositionCompustatPopulation

Mean Median Min Max

Entertainment 6 1.8% 1.9% 4 3 0 10Consumer Goods 10 3.0% 1.9% 3.1 3 0 8

Apparel 7 2.1% 1.3% 2.57 2 0 7Healthcare 14 4.2% 1.9% 3.79 3 1 10Medical Equipment 18 5.4% 3.1% 2.17 1.5 0 8Pharmaceutical Products 34 10.3% 7.0% 3.97 4 0 12Chemicals 7 2.1% 1.9% 4.14 2 0 14Construction Materials 10 3.0% 1.8% 3.6 2.5 0 12Steel Works, etc 8 2.4% 1.6% 1.75 1 0 6Machinery 13 3.9% 3.1% 1.77 1 0 10Petroleum and Natural Gas 24 7.2% 4.6% 3.63 3 0 10Telecommunications 19 5.7% 3.4% 8.37 8 0 25Business Services 49 14.9% 19.0% 3.43 3 0 16Computers 19 5.7% 4.9% 3.16 2 0 14

Electronic Equipment 23 6.9% 5.5% 3.52 2 0 19Measuring and Control Equip 8 2.4% 1.9% 3.38 3 0 8Transportation 10 3.0% 2.7% 6.4 5.5 0 19Wholesale 17 5.1% 3.4% 4.71 4 0 10Retail 26 7.9% 5.3% 6.77 6 0 20Restaurants, Hotel, Motel 9 2.7% 2.0% 4.33 3 2 10 Total 331

a We define industries following Fama French (1997). b We collect data on related party transactions from 331 companies for fiscal years 2000 and 2001.

Table 2: Descriptive Statistics of Related Party Transaction Summary Variables

Mean Std. Dev. Median

Related Party Transaction (RPT) Summary Variables: (n=662)

Overall - Number of RPTs per company 4.042 4.289 3

Primary Related Parties -

Number of RPTs per individual primary related party 2.817 2.492 2 Number of different types of primary related parties per company 1.794 1.274 2

Secondary Related Parties - Number of RPTs per individual secondary related party 2.290 2.642 2 Number of different types of secondary related parties per company 2.051 1.172 2

Types of Transactions - Number of different types of transactions per company 2.194 1.712 2


31/45

30

Table 3: Summary of Related Party Transactions by Company and Primary Related Party

Percent of Transactions with: b

Number ofRelated Party Number of Executives c Non-Executives c Transactionsper Company

Numberof Obs. a

Percentof Obs.

Related PartyTransactions a

Percentof RPTs Chairman

BoardMember

Non-Board Director Chairman

PrincOw

0 103 15.6% 0 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%1 103 15.6% 103 3.9% 0.8% 1.0% 0.1% 1.3% 0.2% 0.4% 2 100 15.1% 200 7.6% 1.0% 1.5% 0.5% 3.4% 0.5% 0.6% 3 79 11.9% 237 9.0% 1.5% 1.8% 1.0% 2.9% 1.0% 0.6% 4 49 7.4% 196 7.4% 1.7% 1.1% 1.1% 2.2% 0.8% 0.3% 5 51 7.7% 255 9.7% 1.6% 2.0% 1.2% 2.9% 0.8% 0.3% 6 42 6.3% 252 9.5% 1.4% 2.1% 0.7% 3.4% 0.4% 1.0% 7 38 5.7% 266 10.1% 1.4% 1.7% 1.4% 2.8% 1.2% 1.2% 8 24 3.6% 192 7.3% 1.5% 1.4% 0.4% 2.2% 0.8% 1.0% 9 8 1.2% 72 2.7% 0.0% 0.2% 0.2% 1.4% 0.2% 0.7%

10 23 3.5% 230 8.7% 1.8% 1.9% 1.3% 2.3% 0.8% 0.3% 11 10 1.5% 110 4.2% 0.9% 0.6% 0.7% 1.2% 0.2% 0.1%

12 - 25 32 4.8% 528 20.0% 4.0% 3.3% 2.9% 7.3% 1.0% 0.5% 662 100.0% 2,641 100.0% 17.6% 18.5% 11.6% 33.1% 7.8% 6.8%

Percent of Companies with Transaction 37.4% 40.6% 26.3% 63.8% 17.9% 17.4%

Notes:a Related party transactions data is presented for 331 companies for two years (fiscal years 2000 and 2001). Of the number of relaare reported in 2000 and 1,342 in 2001.

b The 'Percent of Transactions with:' columns report the percent out of 100% by primary party and number of transactions.c

Includes both current and former executives and non-executives. Approximately 2% (1%) of transactions with executives are wiexecutives).d Also includes and transactions with joint venture partners.


32/45

31

Table 4: Summary of Related Party Transactions by Secondary and Primary Party

Percent ofCompanies

Percent ofRelated

Primary PartyPercent of Transactions w

withTransaction

PartyTransactions

Executives c Non-Executives

Secondary Party(N = 662

Obs.) a (N=2,641Trans.) a Chairman

BoardMember

Non-Board Director

Chai(not C

Executives* - Chairman 10.6% 3.7% 3.7% 0.0% 0.0% 0.0% 0.0%Board Member 5.9% 2.2% 0.0% 1.9% 0.1% 0.2% 0.0%

Non-Board 7.9% 2.5% 0.0% 0.2% 1.0% 0.7% 0.7%Executive's business - Chairman 14.7% 7.0% 7.0% 0.0% 0.0% 0.0% 0.0%

Board Member 10.7% 3.7% 0.0% 3.7% 0.0% 0.0% 0.0% Non-Board 8.4% 4.1% 0.0% 0.0% 3.2% 0.7% 0.1%

Non-Executives - Director 11.3% 3.1% 2.1% 0.1% 0.0% 0.3% 0.7%Chairman 9.7% 4.3% 0.0% 0.1% 0.2% 15.1% 0.0%

Non-Executive's bus - Director 32.9% 15.4% 0.1% 0.1% 0.2% 0.1% 3.9%Principal Owner 14.8% 7.8% 0.3% 0.5% 0.2% 5.6% 0.4%Subsidiary 1.4% 0.7% 0.1% 0.2% 0.1% 0.2% 0.0%

Other 2.1% 1.2% 0.0% 0.0% 0.4% 0.4% 0.0%Company direct 81.0% 44.3% 0.0% 0.2% 0.0% 0.4% 0.1%Total 100.0% 4.5% 11.6% 6.4% 9.4% 2.0%

Notes:a Related party transactions data is presented for 331 companies for two years (fiscal years 2000 and 2001). Of the number of relaare reported in 2000 and 1,342 in 2001.

b The 'Percent of Transactions with:' columns report the percent out of 100% by primary party and number of transactions.c Includes both current and former executives and non-executives. Approximately 2% (1%) of transactions with executives are wiexecutives).

d Also includes and transactions with joint venture partners.


33/45


34/45

33

Table 6: Descriptive Statistics and Correlations of Accruals Variables and Control Variables

Panel A: Descriptive Statistics

Variable (n=662) Mean Std. Dev. Median

Accruals Variables:

Total accruals -0.074 0.183 -0.057Abs(total accruals) 0.131 0.148 0.085AAC, Abnormal accruals (p=0.19) 0.016 0.215 0.009Abs(AAC), Absolute value of abnormal accruals -0.009 0.197 -0.000AAAC, Adjusted absolute value of abnormal accruals 0.087 0.151 0.034

Control Variables: Abs( ∆ NI) 0.211 0.516 0.064 MV/BVA 2.663 3.989 1.315OCF 0.007 0.220 0.056

Neg. NI 0.602 0.490 1 Debt 0.218 0.301 0.103

Log(Assets) 5.269 2.064 5.105

Panel B: Correlations Between Main Variables Used in Regression Analysis(Pearson correlations with two-sided probability values)

(N=662) AAAC Abs( ∆ NI) MV/BVA OCF Neg. NI Debt Log(Assets)

Number of related partytransaction per company -0.036 0.016 -0.031 0.016 0.023 0.129 0.139

0.366 0.688 0.434 0.288 0.553 0.009 0.000

AAAC 0.017 -0.054 -0.108 0.131 0.042 -0.2090.661 0.162 0.006 0.001 0.288 0.000

Abs( ∆ NI) 0.402 -0.224 0.246 0.001 -0.1690.000 0.000 0.000 0.980 0.000

MV/BVA -0.214 0.139 0.012 -0.0370.000 0.000 0.763 0.344

OCF -0.503 0.092 0.4040.000 0.019 0.000

Neg. NI -0.032 -0.4010.417 0.000

Debt 0.269

0.000 Notes:Variable definitions: RPT is related party transaction. Total accruals are defined as net income less operating cash flows accrualsfor firm and year; AAC is the abnormal accruals for firm by industry and year; Adjusted absolute abnormal accrual (AAAC) ismeasured as the absolute abnormal accrual Abs(AAC) minus the Median Abs(AAC) for a portfolio of firms matched on the standarddeviation of past accruals by industry; Abs( Δ NI ) is the absolute value of the change in net income divided by lagged total assets

between years t and t-1; MV/BVA is the market value of common equity divided by the book value of common equity measured atthe beginning of the year; OCF is the current year's operating cash flows divided by beginning of the year total assets; Neg. NI isan indicator variable equally one if the firm reported negative net income in the current or prior year and zero otherwise; Debt islong-term debt divided by beginning of the year total assets; Log(Assets) is the natural log of total assets.


35/45


36/45

35

Table 8: Regressions of Adjusted Absolute Abnormal Accruals on Related Party Transactions by Secondary Rela

(n=662) INT RPT* Abs( ∆ NI) MV/BVA OCF Neg. NI Debt Log(

RPT* = Indicator Variable for Transactions with Secondary Party: Executives:*

- Chairman 0.159*** -0.041 * -0.002 -0.004 -0.005* 0.013 0.049 - Board Member 0.152*** 0.037 -0.002 -0.003 -0.010* 0.013 0.051

- Non-Board 0.157*** -0.021 -0.002 -0.004 -0.008* 0.013 0.050 Executive's business:- Chairman 0.162*** -0.039 * -0.001 -0.004 -0.010* 0.012 0.051

- Board Member 0.153*** 0.005 -0.002 -0.003 -0.008* 0.014 0.049 - Non-Board 0.154*** 0.002 -0.002 -0.003 -0.009* 0.013 0.049 Non-Executives:- Director 0.154*** 0.012 -0.002 -0.003 -0.008* 0.014 0.049

- Chairman (not CEO) 0.155*** -0.010 -0.002 -0.004 -0.008* 0.014 0.051

Non-Executive's business:- Director 0.153*** 0.010 -0.002 -0.003 -0.008* 0.014 0.049

Principal Owner 0.159*** -0.035 * 0.000 -0.004 -0.006* 0.016 0.053 Subsidiary 0.154*** -0.017 -0.001 -0.003 -0.008* 0.014 0.050 Other 0.156*** 0.034 -0.001 -0.004 -0.006* 0.014 0.051 Company 0.142*** 0.002 0.013 -0.003 -0.011* 0.016 0.052 See table 6 for variable definitions.***, **, * denote significance at the 0.01, 0.05 and 01.10 levels, respectively. p-values on related party variables are one-sided, others


37/45

36

Table 9: Regressions of Adjusted Absolute Abnormal Accruals on Related Party Transactions by Type of Trans

(n=662) INT RPT* Abs( ∆ NI) MV/BVA OCF Neg. NI Deb

RPT* = Indicator Variable for Type of Transaction:

Direct Service 0.158*** -0.037 -0.002 -0.004 -0.009* 0.013 0.0Direct Service - Executive 0.159*** -0.023 -0.002 -0.004 -0.007* 0.013 0.0Direct Service - Non-Executive 0.155*** -0.002 -0.002 -0.003 -0.009* 0.013 0.0

Contracted Services:- Management Services 0.161*** -0.032 * -0.002 -0.004 -0.003* 0.015 0.05

- Legal Services 0.155*** -0.009 -0.002 -0.003 -0.008* 0.013 0.05 - Marketing 0.157*** -0.058 -0.001 -0.004 -0.006* 0.014 0.04

- Real Estate 0.154*** -0.003 -0.001 -0.003 -0.008* 0.014 0.05

- Investment Banking 0.153*** 0.025 -0.002 -0.003 -0.008* 0.014 0.05

- Other 0.154*** 0.010 -0.001 -0.003 -0.008* 0.013 0.0

- Purchases from 0.156*** -0.021 -0.002 -0.003 -0.011* 0.013 0.0Sales to 0.157*** -0.015 -0.002 -0.004 -0.006* 0.013 0.0Loans to - Other 0.153*** 0.010 -0.002 -0.004 -0.009* 0.013 0.0Loans to - Home 0.157*** -0.025 -0.002 -0.003 -0.017* 0.012 0.0Loans to - Stock 0.155*** 0.021 -0.002 -0.003 -0.008* 0.013 0.0Fixed-rate financing from 0.136*** 0.049 * 0.000 -0.003 -0.008* 0.012 0.0Investment 0.155*** -0.017 -0.002 -0.003 -0.008* 0.014 0.0

Other 0.158*** -0.033* -0.001 -0.004 -0.010* 0.014 0.

See table 6 for variable definitions.***, **, * denote significance at the 0.01, 0.05 and 01.10 levels, respectively. p-values on related party variables are one-sided, others


38/45

37

Table 10: Regressions of Adjusted Absolute Abnormal Accrualson the Log of the Dollar Amount of the Transaction by Type of Transaction

(N=662) INT RPT* Abs( ∆ NI) MV/BVA OCF Neg. NI

RPT* = Log of Dollar Amount by Type of Transaction: Direct Service 0.155*** -0.002 -0.001 -0.004 -0.009* 0.014

Contracted Services:- Management Services 0.160*** -0.006 * -0.001 -0.004 -0.003* 0.014

- Legal Services 0.156*** -0.004 -0.002 -0.003 -0.008* 0.013

- Marketing 0.157*** -0.009 -0.001 -0.003 -0.006* 0.014 - Real Estate 0.155*** -0.001 -0.001 -0.004 -0.008* 0.013

- Investment Banking 0.153*** 0.003 -0.001 -0.003 -0.008* 0.013

- Other 0.155*** -0.003 -0.002 -0.003 -0.010* 0.013

- Purchases from 0.157*** -0.003 -0.002 -0.004 -0.005* 0.013Sales to 0.155*** -0.001 -0.002 -0.003 -0.008* 0.014Loans to - Other 0.154*** 0.001 -0.002 -0.003 -0.009* 0.013Loans to - Home 0.156*** -0.003 -0.002 -0.003 -0.010* 0.014Loans to - Stock 0.155*** 0.003 -0.003 -0.003 -0.007* 0.014Fixed-rate financing from 0.140*** 0.006 ** 0.000 -0.003 -0.003* 0.012Investment 0.155*** -0.002 -0.002 -0.003 -0.009* 0.014

Other 0.156*** -0.003 -0.001 -0.004 -0.009* 0.013

See table 6 for variable definitions.***, **, * denote significance at the 0.01, 0.05 and 01.10 levels, respectively. p-values on related party variables are one-sided, other# Amounts not disclosed for these transactions.


39/45

1

APPENDIX AEXAMPLES OF TRANSACTIONS INVOLVING FIXED-RATE FINANCING FROM

RELATED PARTIES

EXAMPLE I

From: Alexander & Baldwin, Form DEF 14A, filed with SEC on 3/12/2001

CERTAI N RELATI ONSHI PS AND TRANSACTI ONS. Wal t er A. Dods, J r . , adi r ect or of A&B, i s Chai r man of t he Boar d and Chi ef Execut i ve Of f i cerof BancWest Cor por at i on, and Chai r man of t he Boar d and Chi ef Execut i veOf f i cer of i t s banki ng subsi di ar y, Fi r st Hawai i an Bank.

Fi r st Hawai i an Bank ( i ) has a 32 per cent par t i ci pat i on i n and i sagent f or a $140, 000, 000 r evol vi ng cr edi t and t erm l oan agr eement wi t hA&B, under whi ch $113, 500, 000 was out st andi ng at Febr uar y 15, 2001,( i i ) has a r evol vi ng cr edi t agr eement wi t h A&B under whi ch t he amountout st andi ng ( $6, 000, 000 at Februar y 15, 2001) , when combi ned wi t h Fi r stHawai i an Bank' s shar e of amount s dr awn under t he previ ousl y descr i bed$140, 000, 000 r evol vi ng cr edi t and t erm l oan agr eement , may not exceed$70, 000, 000, ( i i i ) has a $25, 000, 000 r evol vi ng credi t f aci l i t y wi t hMat son t o suppor t t he i ssuance of commer ci al paper , under whi ch noamount was out st andi ng at Febr uar y 15, 2001, ( i v) has i ssued l et t er s ofcr edi t t ot al i ng $13, 226, 000 on behal f of Mat son f or i nsur ance secur i t ypur poses, and ( v) has i ssued l et t er s of cr edi t t otal i ng $6, 882, 576 onbehal f of cer t ai n r eal est at e subsi di ar i es t o secur e obl i gat i ons t ogover nment al agenci es i n connect i on wi t h r eal est at e devel opment s.

EXAMPLE II

From: Alpharma Inc. Form DEF 14A, filed with SEC on 4/10/2001

Excerpt from “CERTAIN RELATIONSHIPS AND RELATED TRANSACTIONS” (page 22)Emphasis added.

Dur i ng 2000, t he Company t hrough Al phar ma Osl o and i t ssubsi di ar i es sol d $2, 002, 000 of pr oduct s, pr i mar i l y mul t i vi t ami ns andadhesi ve pr oducts , t o a subsi di ar y of A. L. I ndust r i er f or di s t r i but i onof t hese pr oduct s t o ret ai l f ood st or es. I n addi t i on, dur i ng 2000Al phar ma Osl o pur chased $8, 000 of pr oduct f r om a subsi di ar y of A. L.I ndus t r i er.

At December 31, 2000 A.L. Industrier held a convertiblesubordinated note issued by the Company (the "Industrier Note") in the

principal amount of $67,850,000, whi ch was i ssued by t he Companyconcur r ent l y wi t h $125, 000, 000 of Conver t i bl e Subordi nat ed Not es s ol dt o unaf f i l i at ed t hi r d par t i es ( t he "Not es") . The I ndust r i er Not e hassubst ant i al l y t he same t er ms ( i ncl udi ng i nt er est r at e, conver si on pr i ceand mat ur i t y dat e) as, and wi l l r ank par i passu wi t h, t he Not es, exceptt hat t he I ndust r i er Not e i s conver t i bl e i nt o Cl ass B Common St ocki nst ead of Cl ass A Common St ock. Subj ect t o cer t ai n adj ust ment s t heI ndust r i er Not e i s conver t i bl e i nt o 2, 372, 896 shares of Cl ass B CommonSt ock. The I ndust r i er Not e, so l ong as i t i s hel d by I ndust r i er or any


40/45

2

ot her af f i l i at e of t he Company, i s not conver t i bl e at t he hol der ' sdi scr et i on but i nst ead wi l l aut omat i cal l y be conver t ed i nt o Cl ass BCommon St ock on or af t er Apr i l 6, 2001, i f at l east 75% of t he Not esar e conver t ed i nt o Cl ass A Common St ock by t he hol der s t her eof . TheI ndust r i er Not e i s exchangeabl e, i n whol e or i n par t , i nt o Not es at anyt i me af t er Oct ober 31, 1999, sol el y f or t he pur pose of enabl i ng t hehol der t o sel l t he Not es r ecei ved i n such exchange to an unaf f i l i at edt r ansf er ee. The Company pai d A. L. I ndust r i er $3, 901, 000 i n i nt er est i n1999 and $3, 901, 587 i n 2000 on t he Not e.

Al l t r ansact i ons wi t h A. L. I ndust r i er are subj ect t o r evi ew by,and i n cer t ai n ci r cumst ances pr i or appr oval of , t he Audi t Commi t t ee.See "Board of Di r ect or s and Commi t t ees - - Commi t t ees of t he Boar d"above.

Excerpt from footnotes to “SECURITY OWNERSHIP OF CERTAIN BENEFICIAL OWNERSAND MANAGEMENT” (page 7)

Mr. Sissener is Chairman of the Board of A.L. Industrier and t oget herwi t h A/ S Swekk ( Mr . Si ssener ' s f ami l y- cont r ol l ed pr i vat e hol di ngcompany) ( " Swekk") and cer t ai n of hi s r el at i ves benef i ci al l y owns 52. 2%of A. L. I ndust r i er ' s out st andi ng or di nar y shar es ent i t l ed t o vot e and,accor di ngl y, may be deemed a cont r ol l i ng per son of A. L. I ndust r i er .

Exerpt from “NOMINEES FOR DIRECTORS” (page 9)

Einar W. Sissener.72 Chairman of the Company si nce 1975. Chi efExecut i ve Of f i cer f r om J une 1994 t o J une 1999. Member of t he Of f i ce oft he Chi ef Execut i ve of t he Company J ul y 1991 t o J une 1994. Chai r man oft he Of f i ce of t he Chi ef Execut i ve J une 1999 t o December 1999.Pr esi dent , Al phar ma AS si nce Oct ober 1994. Pr esi dent , Apot heker nes AS( now AL I ndust r i er AS) 1972 t o 1994. Chai r man of A. L. I ndust r i er ASsi nce November 1994.


41/45

1

APPENDIX BEXAMPLES OF CODING VARIOUS RELATED PARTY TRANSACTION

DISCLOSURES

EXAMPLE IThe sections below are from the Proxy, form 14A, of Looksmart Ltd. filed as of April 26, 2002and obtained from the SEC website: www.sec.gov. All indented items in black have been cut and

pasted from the SEC website and so are direct quotes. The grey circles and the remarks in greyitalics are comments added to indicate how the relevant related party transaction information wscoded for this study. There are three RPTs.

DEF 14APUBLI C DOCUMENT COUNT: 3CONFORMED PERI OD OF REPORT: 20020605FI LED AS OF DATE: 20020426

FI LER:COMPANY DATA:

COMPANY CONFORMED NAME:LOOKSMART LTD

PROPOSAL ONE — ELECTION OF DIRECTORS

Our board of directors consists of seven directors, three of whom are standingfor election: Evan Thornley, Tracey Ellery and Edward West. In addition to thethree directors standing for election, we have two incumbent directors with termsexpiring in 2003 and two incumbent directors with terms expiring in 2004. Our

bylaws provide that the board of directors be divided into three classes, with eachclass to be as nearly equal in number as possible. There is no difference in thevoting rights of the members of each class of directors. Each class of directorsserves a term of office of three years, with the term of one class expiring at theannual meeting of stockholders in each successive year. There are no familyrelationships among any directors or executive officers of the Company, exceptthat Evan Thornley, Chairman and Chief Executive Officer, is married to TraceyEllery, a member of the board of directors.

Nominees for Election to the Board of Directors

The nominees for election to the board of directors are Evan Thornley, TraceyEllery and Edward West. The board of directors unanimously recommends thatyou vote FOR election of all nominees as directors.

Evan Thornley co-founded LookSmart and has served as its Chairman andChief Executive Officer since July 1996. From July 1996 to June 1999, Mr.Thornley also served as President. From 1991 to 1996, Mr. Thornley wa

Documents

pihak relasi