Earnings management by real activities manipulations a look at vietnam

ABSTRACT This research thesis tests three hypothesizes: i In Viet Nam, the listed companies that meet earnings target zero earnings and zero earnings growth exhibit the proof of real act

UNIVERSITY OF ECONOMICS ERASMUS UNVERSITY ROTTERDAM

HO CHI MINH CITY INSTITUTE OF SOCIAL STUDIES

VIETNAM THE NETHERLANDS

VIETNAM – THE NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

EARNINGS MANAGEMENT BY REAL ACTIVITIES

MANIPULATION: A LOOK AT VIETNAM

BY

NGUYEN DUY ANH

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

HO CHI MINH CITY, November 2016

UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES

HO CHI MINH CITY THE HAGUE

VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

EARNINGS MANAGEMENT BY REAL ACTIVITIES

MANIPULATIONS: A LOOK AT VIETNAM

A thesis submitted in partial fulfilment of the requirements for the degree of

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

DECLARATION

This is to certify that this thesis entitled “Earnings Management By Real Activities Manipulations:

A Look At VietNam”, which is submitted by me in fulfillment of the requirements for the degree

of Master of Art in Development Economics to Viet Nam – The Netherlands Programme (VNP)

To the best of my knowledge, my thesis does not infringe on anyone’s copyright nor violate any proprietary rights and that any ideas, techniques, quotation, or any other material from the work

of other researchers in my thesis, published or otherwise, are fully acknowledge in accordance with the standard referencing practices

HCMC, November 30th, 2016

Nguyen Duy Anh

I would like to express my gratitude to all lecturers of the Vietnam- Netherlands Program who have provided an interesting lesson to build my economic knowledge during this program Besides, completing this work would have been difficult if it is not supported by my best friends I am indebted to them for their help Moreover, I wish to thank all my friends who are

in VNP 21 who share unforgettable memories in this program

Finally, there are also words of deep gratitude for my family who support and encourage me when I implement my postgraduate studies

ABSTRACT

This research thesis tests three hypothesizes: (i) In Viet Nam, the listed companies that meet earnings target (zero earnings and zero earnings growth) exhibit the proof of real activities manipulation (ii) Hypothesis 2: there is the difference between the extent of using real activities manipulation of the listed firm that meet benchmark and those that meet earning benchmark and have a high value of an asset (iii) there is no relationship between firm using real activities manipulation to just meet earnings benchmark and future performance Our tests were based

on data included 2374 firm-year observation covering 2005 to 2015.We focus on companies that satisfy one of the criteria: zero earnings or zero earnings growth The firms meet the criteria call suspected firms or belong to suspected firms group The rules identify the firms that more likely to using real activities manipulation We examine three types of real earnings management: (1) cutting discretionary expenditures (2) acceleration of timing of sales (sale manipulation) (3) reducing the cost of production To measure real earnings management, we follow the cross-section model developed by Roychowdhury (2006), Gunny (2010); and estimate abnormal production cost, abnormal discretionary spending (sum of SG&A, R&D, and advertising) and abnormal CFO Our finding is that Vietnamese listed firmed apply real activities management to meet earnings benchmark Besides, the degree effect of cutting production cost of the suspected firms with a high value of assets is highest (11.14%) among three types of real activities management (2.54% for sale manipulation and 0.398% for reducing discretionary expenses, which suggest the firm with a good reputation prefer employ cutting production cost to meet companies target Final, the companies which engage in CFO manipulation or cutting discretionary expenses, as real earnings management to just meet earnings benchmarks have no impact on subsequent performance In contrast, the companies which engage in production cost as real earnings management to just meet earnings benchmarks have negative impact on subsequent performance

Keyword: Capital markets; Accounting choice; Earnings manipulation

Table of Contents

DECLARATION i

ACKNOWLEDGEMENT ii

ABSTRACT iii

CHAPTER 1 1

INTRODUCTION 1

1.1 Problem statement 1

1.2 Research objective 3

1.3 Research questions 4

1.4 Structure of study 4

CHAPTER 2 5

LITERATURE REVIEW 5

2.1 Key concept 5

2.1.1 Definition of earnings management 5

2.1.2 Real activities management definition 8

2.1.3 Accruals 8

2.2 Models to detect the use of earnings management 9

2.2.1 Jones model 9

2.3 Incentive to earnings management 11

2.3.1 Debt Covenants 11

2.3.2 Compensation 12

2.3.3 Insider Trading 14

2.3.4 Management buyout 14

2.4 Beating benchmark (zero earnings) 15

2.5 Empirical research on real earnings management 19

2.5.1 Type of real earnings manipulation 21

2.5.2 Impact of real activity management on firm future performance 23

2.6 Hypotheses development 24

CHAPTER 3 26

3.1 Selection of suspect firm-year 26

3.2 Real earnings proxies 26

3.3 Empirical model 28

3.3.1 Empirical model to test hypothesis 1 28

3.3.2 Empirical model to test hypothesis 2 29

3.3.3 Empirical model to test hypothesis 3 30

3.3 Data collection 31

3.3.1 Data collection 31

3.3.2 Variable descriptions 32

CHAPTER 4 35

RESULTS AND DICUSSION 35

4.1 Selected suspected firm-year 35

4.2 Descriptive statistics 36

4.3 Testing the assumptions of panel data regression 40

4.3.1 Multicollinearity 40

4.3.2 Autocorrelation 41

4.4 Empirical and discussion 42

4.4.1 Empirical evidence on the real activities manipulation of Companies listed in the Viet Nam stock market 42

4.6.2 Discussion about real activities manipulation 45

4.6.3 Size effect on real earnings management 49

4.6.4 The accociation between using real activites managmet to meet earnings benchmark and future performance 52

CHAPTER 5 55

CONCLUSION, CONTRIBUTION, AND LIMITATION 55

5.1 Main finding 55

5.2 Contribution 57

5.3 Limitation and further research 57

5 Reference 59

6 Appendix 63

List of Table and Figure

List of Tables

Table 1: Earnings management definition 8

Table 2: Calculating abnormal accruals 11

Table 3: Variable descriptions 32

Table 4: Description statistics for sample firms in 2005-2015 period 37

Table 5: Pearson Correlation matrix 40

Table 6: Variance inflation factor 41

Table 7: Comparison of suspect firm and the rest of sample 45

Table 8: Capture net effect when firms combine one more type of real earnings management 49

Table 9: Multivariate regression analyses of size effect on earnings management in Vietnam listed firms 50

Table: 10 Regression between real activities and future performance 53

List of Figure Figure 1: Framework for understanding the practice of Accounts Manipulation 6

Figure 2: A hypothetical value function 16

Figure 3: The losses have more impact on than an equivalent amount of gain 17

Figure 4: The distribution of changes in net income divided by market value of equity at the beginning of the year 18

Figure 5: histogram of distribution of net income scaled by lagged total assets (Figure 5A) and histogram of distribution of change in net income scaled by lagged total assets (Figure 5B) 35

Figure 6: Comparing mean of the suspect firm group (include observation of firm with zero earnings and firm with zero earnings growth) and men of the rest of sample 39

ABBREVIATION

CHAPTER 1 INTRODUCTION

1.1 Problem statement

With the explosion of the Dot-Com bubble in 2000, the previous stock which used to be bullish now became going down Then, in the end, the awful reality kept up with the firms which were supposed to try to hide the unpleasant truth in their financial reports The beginning of the long list big scandal happened in 2000 when Xerox admitted that over a four year – period, their income had been overstated by US$1 4 billion They boosted their income by reported revenue from the lease

of printers and copier in the long-term too early Unfortunately, it was not an isolated example Enron Corporation uses a special purpose accounting entities which help them hide billion dollars worth of debt away its balance sheet WorldCom employed a simple scheme to change more than US$ 11 billion of cost to assets Tyco International executives were accused of covering million

US dollar debt, which they borrowed from employees with interest – free or very low-interest loans and do not disclose these loans After investigation of the U.S Securities and Exchange Commission, Quest Communication was forced to adjust their profit by US $2.4 billion because Quest Communication reported impressive income from the transaction is booked as revenue without receiving money The series of financial accounting scandal still goes on As a result of the scandals, the collapse of cooperation lead to hundreds of billion dollars in loss for investors, thousand job losses The collapse of WorldCom in May 2002 is the biggest, with approximately

US $180 billion in the loss and also 30.000 lost jobs

Besides financial accounting scandals, it also highlights the failures to audit financial statement correctly For example, Arthur Andersen LLP, which was one member of Big 5 accounting firm in the world, prepared Enron’s financial report Andersen also collapses after Enron scandal According to the Securities and Exchange Commission investigation, firstly, Andersen found out many “trouble” transactions which were highly risky, but Andersen audit firm received a million fees, so Andersen did not give their opinion of these risk Further, Andersen ordered their company’s Houston office destroy a thousand of the document to prevent Securities and Exchange Commission from investigating Enron’s bankruptcy These Arthur Andersen scandals reduce the faith of investor in the integrity of the audit firm After all, United States of America enacted Sarbanes-Oxley Act in July 2002 to improve the business environment and protect investors

The new environment force academics, regulator, and practitioners concern with companies

‘earnings management to protect investment and control capital market efficiency There are two reasons why we should study earnings management Firstly, more research on earnings management, more suggestion for improvement (Ronen, 2002a, 2002b, 2002c) The others try to understand the earnings management phenomenon (Ronen and Yaari, 2007) According to Erickson, Hanlon, and Maydew (2006), if we can understand clearly why earnings management occur, and how it is created, we can effectively prevent futures happening Accounting research shows that not all earnings management is wrong Therefore, each type of earnings management,

we can take an active action to avoid this risk

Schipper (1989) defines: “Earnings management occurs when managers use judgment in financial reporting and structuring transactions to alter financial reports to either mislead some stakeholders about the underlying economic performance of the company or to influence contractual outcomes that depend on reported accounting numbers.” In another word, somebody is acting something which effects negatively on someone else For example, to evaluate firm’s credibility; financial health and viability; creditors apply the numbers reported (Ge, 2010) In turn, among other indicators, investors use earnings to monitor operational performance Nevertheless, their conclusions on a given number reported could be incorrect if they are unable to recognize and adapt

to the impacts of earnings management which is hiding in the financial statements This distortion will become apparent in future results when the firm’s performance does not match their evaluations

Accruals manipulation does not involve by altering operating activities of the firm, but through selecting the relevant accounting system to meet desired goals On the other hand, real earning management is the action of managers to reach their objective by changing real firm operations such as decline research & development cost; decreasing selling, general and administrative outlays; sales promotion These two types of earnings management are tools for managers to increase/decrease revenues; nevertheless, one form has an effect on real activities, and the other form does not involve in operation

There are some possible benefits if managers employ REM, even though the cost of using accruals management could be lower than REM Firstly, based on accruals management, every action on changing financial reports is quickly placed under scrutiny by regulator and auditor, whereas both auditor and regulator cannot prevent financial executives from employing REM Secondly, the flexibility of using accruals management is limited For example, accruals management must occur

at the end of quarter or year, and the efficiency of accruals management is uncertainty, it depends

on the acceptableness of auditor

The convincing proof on firms employs real earnings management is supported by Graham et al (2005) who conducted a survey 401 financial executives to find out core elements which affect decisions involve in presenting income and voluntary disclosure This paper demonstrates 78% of the managers interviewed are willing to make from small to large sacrifices as long as they meet an earnings target In addition to, the decline in research and development (R&D), advertising and maintenance cost are employed by 80 percent of participants, while more than a half of them would delay a new project

In finance and accounting, none of the research topics can be more provocative than earnings management, to the point that a quantity of the investigation on detection and results of earnings management is exponential growth Although, accruals management and real earnings management (REM) are two groups of earnings management (Healy & Wahlen, 1999), most of these current studies do not take into earnings management by manipulating real activities While in the recent year, especially after the biggest accounting scandal of Enron and WorldCom, there has been increasing attention to the identification of various types of real activities management Besides that, whether investors recognize the effects of this manipulation to protect their investment Although in the developed country, these issues got much attention, in Asia in general and Vietnam

in particular, there is little research on these topics Therefore, the aim of this paper is to examine whether the companies in Viet Nam employ real activities manipulation to meet the income target

As far as we know, earnings management area has not studied in Viet Nam Therefore, it is necessary to research that thoroughly investigates the real earnings management in order to bring

an overview of earnings management The results of this research can be applied in various fields such as finance, company governance, studying… The sample is covered last update data in the 2005-2015 period This study utilizes a cross-section estimation for industry-year to measure the proxy of real earnings management and Feasible Generalized Least Squares estimation to estimate the degree of using real earnings management in listed companies in Viet Nam

1.2 Research objective

In overall, the purpose of this thesis research is threefold First, this study examines the companies apply real activities management as a plan for manipulation to meet the firm’s target Then, the second objective investigates whether on the well-known companies use real activities management

to report a good profit Final, we examine the relationship between firm using real activities

manipulation just to meet earnings benchmark and future performance Our tests were based on data included 2374 firm-year observation covering 2005 to 2015.We focus on companies that satisfy one of the criteria: zero earnings or zero earning growth The firms meet the criteria call suspected firms or belong to suspect firms group The rules identify the firm that more likely to using real activities manipulation We examine three types of real earnings management: (1) cutting discretionary expenditures (2) acceleration of timing of sales (sale manipulation) (3) reducing the cost of production To measure real earnings management, we follow the cross-section model developed by Roychowdhury (2006) and Gunny (2010) Then, we will regress the equation with two key variables: real earnings management variable (dependent variable) and suspected firms variable (independent variable) and control variable to meet research object following

1.3 Research questions

To meet the research object, the following research questions need to be answered:

 The first question is: Do the Vietnamese listed companies employed real activities manipulation to meet zero earnings or zero earnings growth?

 Then, is there the difference between the extent of using real activities manipulation of the listed firms that meet zero earnings or zero earnings growth benchmark, and those meet that benchmark with a high value of an asset?

 Finally, is there no relationship between firm using real activities manipulation to just meet earnings benchmark and future performance?

1.4 Structure of study

The remainder of this study is organized as follows Chapter 2 will present the fundamental concept and discussion of earnings management as well reviews the literature on real activities manipulation Chapter 3 presents our empirical methodology, the method used for regressing, a conceptual framework, and sample selection construction Next, data collection and data description are presented in the same chapter Chapter 4 demonstrates and analyzes the empirical results while the last chapter shows research’s conclusion and gives a suggestion for further research

CHAPTER 2 LITERATURE REVIEW

This chapter presents basic background information, and the literature review includes (theoretical and empirical review) about real earnings management The first section of this chapter provides some fundamental concept in our area: earnings management, real earnings management, and accruals After that, we introduce Jones model which is a basic model together with the important idea to explore earnings management in general The reason why companies applied real earnings management is presented in the next section In the fourth section, we present beating benchmark event which helps find out a firm that is more likely to apply earnings management The final part,

it presents empirical research on real earnings management

2.1 Key concept

2.1.1 Definition of earnings management

Earnings management is one forms of accounts manipulation - other types remain big bath accounting, creative accounting, window dressing, accounting income smoothing Stolowy and Breton (2004) defined accounts manipulation as managers use their judgment to build transactions

or make accountant system choices which have an impact on the probability of affluence move between the firm and fund providers (cost of capital), society (political costs) or managers (compensation plans) The diagram below presents the framework of studying earnings management And, it also shows where earnings management is

Few empirical researches

Many empirical researches

Few real empirical researches

No empirical researches Professional opinion

Type of

research

Few real empirical researches

Many empirical researches

Main research

streams

real earnings management

accruals management

Variance of EPS

EPS Debt/equity ratio

Reduction of current EPS

to increase future EPS

Earnings management

Main research

streams

Income smoothing

Big bath accounting

Academic perspective

Professional prespective

Potent Interpretation

ACCOUNTS MANIPULATION

Outside the limits of Potential Wealth Transfer

Within the limits of laws and standards

Earnings per share (EPS)

Structural risk: Debt/equity ratio

Figure 1: Framework for understanding the practice of Accounts Manipulation Adapted from

“Accounts manipulation: A literature review and proposed conceptual framework”, by, H Stolowy

and, G Breton, 2004 Review of Accounting and Finance, 3(1), 5-92

Healy and Wahlen (1999) are the first authors set standard setter shapers for earning management definition: “Earning management occurs when managers use judgment in financial reporting and structuring transaction to alter financial reports to either mislead some stakeholders about the underlying economic performance of the company or to influence contractual outcome that depends

on reported accounting numbers”

The goodness of this definition is that it covers the costly contracting approach (earnings management is applied to affect contractual results) and informational approach (earnings management is used to deceive stakeholders) Besides, it demonstrates administration as the party responsible for employing those judgments involved the exercise of earnings management The connotation of definition also means opportunistic earnings management

Nevertheless, this description has two weaknesses First, not all earnings management lead to misinterpretation For instance, investors would like to distinguish earnings persistence from one times shocks The company, which manages earnings let investors readily recognize the elements

do not influence on earnings Besides, this action increases the informational benefit of a company announces earnings Second, this definition does not create a distinct frontier between normal operations whose production is earnings and earnings management Dharan (2003) shows that it is

a complicated issue for people who use company’s financial report is how to separate clearly between earnings manipulation which related to fraud and reasonable strategy of managers to meet the firm’s target

The table 1 below summarizes the key point of definition of earnings management, categorizing them as beneficial, neutral, harmful Beneficial earnings management improves report transparency; while misrepresentation and fraud are related to the harmful entirely The neutral is the manipulation of the financial statement within the limits of conformity with obvious law rule

of standing This could be either efficiency improvement or opportunistic

Table 1: Earnings management definition

Earnings management is using

the benefit of choosing freely

in accounting regulation to

transfer executive’s messenger

on future cash flows

Earnings management is a decision on accounting regulation which optimizes a benefit of management only or maximizing firm value

Earnings management is directorate intervention to mislead or decline in the transparency of the financial statement

Ronen and Sadan

“Earnings management is a collection of managerial decisions that result in not reporting the true short term, value-maximizing earnings as known to management Earnings management can be: Beneficial: it signals long-term value; Pernicious: it conceals short- or long-term value; Neutral: it reveals the true short-term performance The managed earnings result from taking action on business activities before income are recognized, or making the decision on accounting choices that affect the earnings numbers and their interpretation after the true earnings are realized.”

2.1.2 Real activities management definition

Roychowdhury (2006) defines real activities management as “Real activities manipulation, on the other hand, entails departures from normal operational practices, motivated by a managers’ desire

to mislead at least some stakeholders into believing certain financial reporting goals have been met

in the normal course of operations.”

Gunny (2010), defines REM as: “… when managers undertake actions that change the timing or structuring of an operation, investment and financing transaction in an effort to influence the output

of the accounting system.”

2.1.3 Accruals

- Accrual basis of accounting and cash basis of accounting

According to International Accounting Standards (IAS), under the accrual basis of accounting, income and costs are recognized when they meet the recognized standard for those components in the IAS, whether or not cash is received, cash is paid

- Accruals in earnings management

Accruals happen when the timing of accounting records of transactions and timing of cash flows is different

2.1.3.2 Non-discretionary accruals and discretionary accruals

Identification between normal accruals and accruals caused managed earnings is the major issue which researchers attempts to distinguish

Ronen and Yaari (2008) definite that:

- Non-discretionary accruals – so called normal accruals - are accruals which start from firm’s activities happened at present that is usual for the company given firm’s strategy, performance level and other factors (effect of the industry, macroeconomic…)

- Discretionary accruals – also known as abnormal accruals - are accruals that begin by firm’s activities created or accounting regulation selected to manage earnings

2.2 Models to detect the use of earnings management

Study of Ronen and Sadan (1981), Healy (1985), DeAngelo (1986, 1988b), Dechow and Sloan (1991) and Jones (1991) seems to be milestones in the study which try to measure abnormal accruals However, nowadays, the idea of the study of Jones (1991) – call Jones model- applies widely in examining earnings management Therefore, in this section, this study introduces a method detect earnings management through abnormal accruals

an adjustment to a financial statement, whether they think the regulators are confused about making

a decision Jones (1991) studies earnings management by the United States of America Firm in Import Relief investigation by United States International Trade Commission (ITC) Jones

examined hypothesis: executives’ U.S companies, which gain the advantage of import protection, intervene income statement which decline recorded profit at examination time compared to non-examination time

Jone model:

Jones’s assumption is before the event; companies do not exercise earnings management Therefore, the progress of the company’s earning can be divided into two sub-period, accruals expectation stage, where abnormal accruals = 0, and the event period

TA𝑖𝑡 : total accruals in year t,

∆REV𝑖𝑡: Revenue in year t subtract revenue in year t-1, PPE𝑖𝑡: gross plant, property and equipment in year t,

𝐴𝑖𝑡−1: total assets in year t -1, 𝜀: error term,

𝑖: indicate to firm, i = 1,2,3…n

The intercept does not include in this equation because of the inversion of the lagged assets All variables in Jones model are divided by lagged assets to overcome heteroscedasticity Managers use revenues as the benchmark to measure companies’ performance before they decide manipulations Besides, revenues are not totally exogenous Therefore, they are included in the equation to control the economic environment of the firm While the ratio between total accruals and normal accruals depreciation expense is controlled by Gross property, plant, and equipment variable

2 Stage 2:

After obtaining 𝛽̂ ,𝛽𝑖 ̂ , 𝛽1𝑖 ̂ of 𝛽2𝑖 𝑖,𝛽1𝑖,𝛽2𝑖 from the first stage, respectively, the parameters 𝛽̂ ,𝛽𝑖 ̂ , 𝛽1𝑖 ̂ 2𝑖

is put into the equation The Jones model assumes that there is stationary in relation between normal accruals and independent variables The error term of the equation is abnormal accruals (discretionary accruals)

Table 2: Calculating abnormal accruals

at time t

(discretionary

accruals)

𝜇𝑖𝑡 = TA𝑖𝑡− TA ̂ 𝑖𝑡

2.3 Incentive to earnings management

There are many factors that encourage executive manage earnings However, debt covenants, executive compensation contracts, insider trading and capital market transactions (equity offerings) are four common elements

2.3.1 Debt Covenants

Financial statements cover a primary source of information for creditors Debtors can evaluate company performance through this information to set up debt contract terms Mulford and Comiskey (2002) explained that debt covenants are designed as a condition in debt obligations in order to supervise company performance For example, a creditor could require a maintain a specific value of an accounting ratio or investing and financing within the limits of some activities

If the debt covenant is violated, the borrower might levy a heavy cost of debt as the creditor could increase the rate of interest, requiring additional assets as collateral or require immediate repayment Therefore, Beneish (2001) believe that debt covenants may provide an incentive for directors to adjust income either to avoid the limitation of accounting system based on constraints

in debt obligations or to reduce the cost of agreements violations Healy and Palepu (1990) provide evidence that if the cost of violating the constraints in debt agreement is heavy, the company can circumvent the restrictions of a debt obligation by adjusting income through accounting decision Watts and Zimmerman, 1986 offered the debt covenant hypothesis which forecast the company would make a decision on the choice of accounting systems to reduce the probability of debt covenant violation In particular, Beatty and Weber (2003) studied whether the provision of a company’ debt agreement affects the firm’s accounting choices Their results provided evidence that the companies whose the debt agreement permit changes in accounting method have an impact

on contract calculations are more incentive to manage income-increasing compared to their covenant counterparts

non-DeFond and Jiambalvo (1994) use time-series and cross-sectional model for analyzing abnormal to exanimate whether the firms which are closer to debt covenant is more likely to increase income

by making accounting choice The research applies time-series and cross-sectional model to analysis abnormal accruals One year before the violation, income and abnormal accruals increase together In the year of abuse, in spite of none of the detection of earnings management, these results imply some firm in this study did not become successful in manipulation due to monitor of auditor and debtor

Dichev and Skinner (2002) collect a massive database of the loan agreement to exanimate debt covenant hypothesis (firms have more motivations to make earnings reporting decisions that decrease the probability that accounting-based constraint in their companies’ debt obligation would

be violated) The study indicates an exceptionally a few companies just under contract thresholds and a huge quantity of companies just reach or beat agreement thresholds These result provide that management intervenes accounting choice to prevent debt covenant violations

Bharath, Sunder, and Sunder (2008) study borrower accounting quality measured by company's abnormal accruals influences on debt contracting The data included 7334 private bank loans to 3,082 firms obtained from the DealScan database in the 1988–2001 period, this research points to

a U - shaped relationship between the cost of debt and accounting quality They consider their measure of accounting quality under accruals as an indicator of harmful earnings management, which increases the risk for debtors This study demonstrates that companies with higher abnormal accruals deal with complicated issues such as shorter maturity, higher interest rate or greater probability of collateral The interest rate is 8 percent greater than mean interest rate in the sample, about one-month shorter at maturity of loan and 7.7 percentage point increases in the likelihood of giving collateral

2.3.2 Compensation

The manager’s remuneration is the most apparent reason explaining why executives manipulate earnings In most of the business, the executive will be received a reward such as an increase in salary, cash and stock bonuses, grants of stock option… if they meet targets (accounting returns, sales revenue, net income…)

The former literature present a lot of the relationship between the reported earnings and structure

of executive compensation Healy (1985) is an early paper connect CEO bonus plan and earnings management This study indicates that a CEO bonus program encourages managers choose accounting treatment and affect accruals to maximize their bonus award By using changes in

procedures and total accruals, this study finds that (i) in the first condition, if earning after using earnings manipulation is not meet the lower bound of the bonus plan, the manager will pick the minimum abnormal accruals option since even if he try to maximize earnings, reported income does not reach the lower point and manager will not receive a bonus This strategy allows the manager

to optimize their future bonus (ii) In the second case, if earnings before manipulating in the lower and upper area, the manager will increase earnings by raising accruals until earnings reach the upper bound The manager chooses maximum accruals because the bonus is constant when income goes beyond the upper bound However, the manager will maximize earnings if the bonus plan does not have the upper bound (iii) Therefore, in the final case, when earnings before abnormal exceeding the upper bound, the manager is more like to choose income-decreasing accruals and reported earnings at the upper point By delaying income beyond the upper bound, the managers increase their future bonus Murphy (2000) compares bonus payments of companies with external performance standard to those with the internal performance standard The results indicate that firms which adopt internal standard are more probable to manage earnings than firms using external standard More detail, companies with external performance standard depend largely on previous year performance in planning performance standard for next year Therefore, there are two scenes Firstly, when earnings does not meet the norm, the manager has more incentive to manipulate earnings to meet targets While the managers have more incentive to decline in their attempt when the target standard is reached

In recently, researchers are more focus on option and stock award, which are the major type of bonus package For example, Bergstresser and Philippon (2006) who show that the use of abnormal accruals for managing earnings is more active at companies where the award of executives is nearly connected to the value of option and stock One of the manipulations methods is an exercise option

in the high-accruals period and sell a huge number of shares These results are similar to Burns and Kedia (2005)’s finding: comparison with other types of compensation, such as an increase in salary, incentive payouts, equity… a stock option is powerful incentives for executes to earnings restatement Finally, because of the critical role of chief financial officer (CFO) in preparing a financial statement, a CFO who awarded by equity should have a more powerful effect than those

of chief executive officers (CEO) in manipulating earnings Jiang, Petroni and Wang (2010) show that the effect of CFO equity incentive on the probability of hitting analyst prediction and the level

of accruals is stronger than those of the CEO

2.3.3 Insider Trading

Insider trading is one interesting object in economics, accounting or law By using nonpublic information, insider trading increases the likelihood to manage earnings harmfully Some of the studies indicate apparently evidence of a relationship between insider trading and coming deception

or insolvency (Summers and Sweeney, 1998; Beneish, 1999a; Agrawal and Cooper, 2007) Bones (1999a) finds that: the executive of companies with income overstatement which break GAAP are more probable to sell their shares and to receive stock appreciation right during the time earnings are overstated Banish is based on previous papers to expand his hypotheses about insider trading and earnings management incentives This study proposes that manager's action like informed traders, selling or buying before decline (increase) in stock price, and sees the benefit from trading

as compensation for supplying their individual information to investors This research also explains

in detail about manager insider trading The managers will use their information about income overstatement to buy (sell) for their gain By overstating income, managers provide market participant optimistic company’s prospects or hide the deterioration of business performance In case supplying good signal about business performance, the manager can raise the value of their equity In another case, the manager will sell their shares with a limitation volume to avoid an attention of the SEC

However, insider trading does not always mean harm Boyer, Ciccone, and Zhang (2004) investigate whether earnings reporting is beneficial or pernicious earnings management The paper arranges company stock – year into quintiles based on abnormal accruals (the company with highest quintiles stand for the largest income-increasing company with lowest quintiles stand for smallest income-decreasing abnormal accruals) The results provide evidence correct with opportunism hypothesis: Manager’s purchasing volume of enterprises with optimistic prospects (manipulate earnings upward) is relatively higher than the other directors, while the manager’s selling volume

of the company with losses and optimistic market forecast (manipulation earnings downward) is significantly smaller than the other managers

2.3.4 Management buyout

A management buyout (MBO) is a form of business acquisition where the directors of a firm buy the enterprise from the current owners or parent company When management buyout exercises, there is a decline in earnings management This case brings the manager into a conflict: the executive who buys the company prefers paying as low a price as possible, while stakeholder would like to sell their equity at the best price Therefore, the manager has more likely to reported downward earnings before the buyout event The empirical research either support this hypothesis

or at least not deny it The evidence of Perry and Williams (1994) indicates that in the year before

a management buyout, there is a reduction in reported earnings in their data of 175 buyouts in

1981-1988 Estimation of Wu (1997) shows that on average, the expenses of trading are decreased about

50 million US dollars by earnings management Marquardt and Wiedman (2004a) investigate 100 management buyout during 1995-99 Firstly, the paper measures total discretionary accruals which are a proxy for earnings management Then, the authors determine single accruals which might be used by managers to meet their income target: accounts receivable, inventory, accounts payable, accrued liabilities, depreciation, special items Their finding reveals that companies apply income downward (exhibited by unexpected accounts receivable accruals, UAR) before management buyouts Especially, positive UAR of the management buyout is 38%, while this is 45% for the control group

2.4 Beating benchmark (zero earnings)

Beating a benchmark has been identified as one of the events which companies is more likely to accomplish earnings management to influence the market price Accounting number will be useless

if not some benchmark being compared Therefore, companies have reasons to manage earnings to meet criterion such as zero earning

The comprehensive survey of Graham et at (2005) reveals income is commonly employed as an important indicator to measure the level of business success And it is one of the most aims which managers concern In particular, the earning per share (EPS) is considered as the proper index for market participants Since investors need simple indicator which easy to understand, sum up firm performance and compare across companies, which reduce the cost of accessing the large number

of available financial information

Figure 2: A hypothetical value function Retrieved from

http://4.bp.blogspot.com/-LuPWeGpTRUY/Td7yB-OGH1I/AAAAAAAAAkE/TYGKEOSdXAw/s1600/prospect%20theory.png

Then, the question is why market participants rely on zero benchmarks There are two explanations why zero earnings is an important point Firstly, Prospect theory which developed in the field of psychology is suggested by Kahneman and Tversky (1979) From psychology approach, Prospect theory supposes that decision maker feels gains or losses from the neutral reference point, rather than from the absolute level of welfare Moreover, the characteristics of the person’s value function are usually convex for losses and commonly concave for gain (Figure 2) This catches the view that losses make a more uncomfortable feeling compared to the level of pleasure given by an equivalent gain (Figure 3) As a result, people obtain the highest benefit when welfare goes from a loss to a gain relative to the point of reference The investors used commonly three natural reference points (a) zero earnings; (b) the predicted value of analyst and (c) earnings of the previous year According

to these features, Koonce and Mercer (2005) stated that ceteris paribus, companies which announce

a series of small profit will be preferable than enterprises that reported volatile earnings (some huge profit and some minor loss)

Figure 3: The losses have more impact on than an equivalent amount of gain Retrieved from http://i1.wp.com/exclr8.co.za/wp-content/uploads/2015/02/graph.jpg

Kothari (2001) reviews empirical research on the link between the financial report and capital markets Evidence from this paper indicates the reaction of the market is positive with good earnings reported Durtschi and Easton (2005) reported that the median price share of companies which presented one- cent profit is higher ($1.31) than those reported a one-cent loss ($0.25) In addition, the rate increasing in price share of companies which reported profit is higher than whose enterprises that announced a loss For instance, for loss companies, the median price is $1.25 for companies reported ten-cent loss and $3 for companies announced fifty-cents loss

Xue (2003) suggests an alternative explanation The author argues that there are many costs for exceeding benchmark, since, in the current period, accruals are boosted to reach the benchmark, which causes reducing future earnings As a result, the flexibility in exceeding benchmark is harder

in the future Therefore, when information in judgment companies’ future performance is an asymmetry between the enterprises and the market, the focal benchmark will be a tacit agreement for communication between the market and the companies Talented performers meet the benchmark while poor performers miss the target The evidence from this paper give strong support that capital market recognizes the signaling hypothesis, rewards the companies which slightly hit the benchmark while disciplining that slightly loss target Brown (2001) indicates that income distribution has moved from small negative (miss estimation by a small quantity) to zero income (reach evaluation) and then to low income (hit the estimation by a small amount) during 1984 – 1999.Figure 4 displays a visualization of the strange distribution function of the earnings management around the threshold

Figure 4: The distribution of changes in net income divided by market value of equity at the beginning of the year Reprinted from “Earnings management and accounting income aggregation Journal of Accounting and Economics”, by Jacob and Jorgensen, 2007- The sample includes 22,015 distinct companies with 920,926 quarterly observations from 1981 to 2001

Figure 4 is typically in the literature, which shows an unusual small quantity of firms which fail to hit the benchmark (column to the left of the zero) and unusually large quantity of firms which hit the benchmark (column to the right of zero) Graphically, column diagram has a strange Burgstahler and Dichev (1997a) suggest an explanation for this strange is the incentive to avoid

losses make firms manage earnings to shift from companies with small losses to a small profit This paper also points out cash flow from operations and changes in working capital which have been used to manipulate earnings

2.5 Empirical research on real earnings management

Even though accrual-based management has been commonly examined as management real earnings management, finding of Graham et al (2005) survey indicate that for manage earnings, managers prefer applying real earning management to accrual-based management

Managers prefer REM to accruals management for at least three reasons Firstly, the auditors and regulators are less attention on real activities about production and price Another benefit of REM

is flexibility Operating decision can take place during the particular period, while managers must wait until the quarter, half or year – end to use accruals to manipulation earnings However, according to Barton and Simko (2002), the managers may face the risk that the quantity of earnings which requires for meeting targets is higher than allowing accruals as the manipulation on accruals

is controlled by GAAP Therefore, there is a limitation manager’s ability when they use accruals earnings compared to REM Finally, average investors are difficult to understand REM

After the study of Roychowdhury (2006) in the real earnings management, this issue becomes receiving more attention and there are an enormous number of papers investigate the relationship between real earnings management activities with many factors in firms

Roychowdhury (2006) developed a model to capture real earnings management activities He collected annual data and focused on earnings thresholds (the zero earnings thresholds) The paper provides the evidence consistent with companies trying to avoid announcing loss in a three-way Firstly, firm boosts sales throughout the year by proposing price reduction or more lenient credit terms Next, company employs overproducing strategy to increase in production level, more overhead spread over a vast quantity of product lead to reduce the cost of gold sold As a result, company can improve operating performance Finally, manager decreases discretionary expenditures such as advertising, Expenditure of research and development expense or Selling, general and administrative expenses (SG&A) to enhance margin

Zang (2012) examines whether managers use real activities manipulation or based-accrual management for managing earnings The paper applies Roychowdhury (2006)’s model to measure real earnings management with a larger sample over 1987-2008 period The results suggest manager based on the costs of two strategies to decide whether real or accrual-based manipulation And, the probability of use real activities manipulation is higher than accrual-based management Besides,

the relationship between real manipulation and the cost of accrual based management is positive, which mean manager will apply real manipulation if the cost of accrual-based management is high Also, real manipulation is negatively correlated with based-accrual manipulation This result suggests managers use two strategies as substitutes

While Roychowdhury (2006) classifies suspect - firm into real earnings management group based

on beating benchmark (earnings intervals into the range -0.075 to +0.075- net income scaled by total assets nearly or equal zero), Cohen (2010) apply two stage model using the Heckman (1979) model In the first stage, the paper run selection model to detect whether companies manage earnings or not Next step, the authors analyze the element measuring the preference of real manipulation compared to accrual-base manipulation The second stage will classify whether companies using real earnings management or accrual-based management Evidence from Cohen (2010)’s study shows that companies apply both real and accrual-based manipulation activities At the time of the seasoned equity offerings (SEOs) Then, there is a trade-off between real and accrual- based management, and manager choose management strategy based on the comparison the cost of two strategies And, a significant finding is that firms use real earnings management activities to decline the company's’ operating performance before SEOs event

Razzaque, Ali, and Mather (2015) investigate another factor related to real manipulation is family ownership in Bangladesh The author applies Roychowdhury (2006)’s model with the industry year fixed effect and robust with a Fama-Macbeth regression Furthermore, in order to correct for endogeneity problem, they follow the Jaggi (2009) suggestion: the square of the natural logarithm

of total assets, the natural logarithm of total assets and volatility of stock return as an instrumental variable (IV) After conducting exogeneity tests for the strength of this IV by Sargan–Hansen test, they reject volatility of stock return as IV and suggest the percentage of equity of general investor variable Their result provides evidence that in family firms in Bangladesh apply more real manipulation than non-family firms through cutting expenses, boost sale and decline in cash flow from operation in the current period Second, the relationship between real earnings management and the family company is inverted U-shaped Finally, the future operating performance of companies that exercise REMs is lower than those companies that do not use

Li (2016) studies knowledge and experience of top management team (TMT) have an influence on real earnings management (REM) They applied some robust test for their results such as suspect company analysis, using lagged independent variable as the instrument variable (Sovey & Green ,2011) By collecting a set of data over 4000 listed firm-year in 2006-2010 in Taiwanese, the results reveal that the professional level of manager and managing a core function area is negatively

correlated with REM However, the relationship between the percentage of manager at the CPA level and REM is positive Also, as a result of enhancing company performance by the decision of the high-level educated manager, the probability of manager engaging in REM is declining Finally, the longer firms exist, the smaller effect of top managers on real earnings management (company age variable)

Yuliana, Anshori and Alim (2015) provide evidence support the theory that there is the unusual distribution of companies around zero earnings The authors interpret this event that the companies have a strong incentive for using real earnings management since firms need to hide reporting slight loss by shifting into small profit and these companies are known as suspect firms This study uses data from the listed businesses in The Indonesian Capital Market in the 2007-2012 period And, the companies satisfy criterion: net income deflated by the total assets beginning of the year in the interval 0 to 0.005 Mover, this result also shows the suspect companies exercise real earnings manipulation is higher than non-suspect firms through declining discretionary expenses, boosting sales and production

Kouaib and Jarboui (2016) investigate whether the level of real earnings manipulation has an impact

of a feature of managers include behavior biases (overconfidence) and visible demographic traits Data are collected in 2000-2014 cover 190 companies in European and over 450 top managers and applied the model to panel data This research used Roychowdhury (2006) model as a proxy measure real earnings manger

They also create one more variable to measure earnings manager called real earnings management index to capture the total effect of real manipulation activities Next, the paper applied Feasible Generalized Least Squares estimation (FGLS) to correct heteroscedasticity and first-order autocorrelation problem in panel data The result obtains indicated that manager with long experience; overconfidence; high cognitive ability is positive with REM This mean manager has incentive to manage upward earnings through decreasing cost of goods and sold; cutting discretionary costs, accelerating timing of sale

2.5.1 Type of real earning manipulation

There are several types of real earning manipulation Operating decision can be classified into four main categories below: (Roychowdhury (2006) and Gunny (2010) recognizes that)

- Sales manipulation: the effort of the financial executive for a short time to boost sales throughout the year by proposing price reduction or more lenient credit terms Roychowdhury (2006) indicate

that overproduction is one of real earnings management, which uses to avoid presenting unexpected income

- Decrease in discretionary expenditures: firms incurred some cost such as research and development; selling, general and administrative and marketing communication in the same period Therefore, companies can cut these expenditures to improve earnings, which is more probability of happening when these outlays do not make a profit immediately The evidence in study of Baber et

al (1991) shows that R&D expenditure is significant reduction when these costs threatens the ability to present positive or increasing earnings in the current period Dechow and Sloan (1991) provide evidence that the growth of marketing and R&D spending is lessened in the executive’s final years (two years before the CEOs retirement) Bens et al (2002) demonstrate when earnings per share dilution issue appears, the executive will trade off between real investment capital and cost for using employee stock options (ESOs), which means CEOs spend less on R&D expenditure

to focus cash on using ESOs

- Overproduction: to meet earnings target, goods can be manufactured more than essential to reach expected demand, which means more products is produced lead to a decrease in fixed cost per unit Besides, total costs per unit decrease as long as increasing in marginal cost per unit are not bigger than a drop in fixed cost per unit As a result, the cost of goods sold is small and firm performance

is better Thomas and Zhang (2002) find out evidence that company used overproduction strategy (increase in inventory) to reduce the cost of goods sold, which enhances the level of profitability

- Gaining from timing the sale of fixed assets: managers can choose the time to sell assets and underlying accounting, income came from this sale is presented in the period of sale (the difference between the market value of an asset and net book value of it) This follows the timing of asset sales could be considered as a smart strategy for manipulating earnings The finding of Bartov (1993) provides that managers manipulate earnings through the schedule of fixed assets to get a smoother degree of income or satisfying debt covenant Herrmann, Inoue and Thomas (2003) explore that Japanese managers use the sale of the property as a way to manage earnings The authors show that when current income does not meet forecast of earning, the executives will apply the sale of fixed assets and marketable securities to boost (or decrease) income

There are other types of operating decision: debt-equity swaps (Hand, 1989); derivative hedging (Barton, 2001; Pincus & Rajgopal, 2002); share repurchase (Hribar et al., 2006); selling profitable assets (Herrmann et al., 2003; Bartov, 1993); sales price reductions (Jackson & Wilcox, 2000); securitization (Dechow & Shakespeare, 2009); advertising cost reduction (Cohen et al., 2010); putting off or reducing the travel budget and maintenance cost; delaying or removing capital

investments (to avoid depreciation charges); asset securitizations and controlling the funding of pension plans (Graham et al., 2004)

2.5.2 Impact of real activity management on firm future performance

There are two schools though about effect of real activity management on firm future performance: The relationship between real activity management and firm future performance could be positive

or negative

Firstly, REM negatively impacts future firm performance because the manager is willing to sacrifice future cash flows for current period income Stein (1989) shows that managers employ real activities manipulation to pump up current earning in an effort to mislead participants about firm’s value The reason why managers can fool stock market participants is myopic behavior even though they are rational In addition to, results of Jensen and Murphy (1990) reveal that in average, per 1000$ decline in shareholder wealth, CEO’s wealth grow 3$, especially 1.85$ per 1000$ for large enterprises and 8.05$ for small enterprises Their results imply that the CEO will scarify the benefit of shareholders through earning management to meet the requirement target for investors and gain private benefit

Yu (2008) provides some evidence that the operating performance of firm employed real earnings management is lower than those not involved in real activities Similarly, Gunny (2010) demonstrates that firm future performance (future performance measured by ROA) is negatively affected significantly by real activities management In particular, the return on asset of enterprises used four types of real earning manipulation1 is lower non-REM companies with ceteris paribus (variable control: industry, performance, size, accruals deciles)

Secondly, there is positive relationship between REM and future operating performance The firm performance could be better in the future as a result of advantage of REM For example, Trueman (1988) indicates that cooperate manager wants to use REM for smoothing earning to lower the creditor's assessment of the probability of firm bankruptcy In turn, the company can borrow at low interest Bartov et al (2002) show that increasing stock price, creating management’s reputation and avoiding litigation are advantages of beating analysts’ estimation expectations

Another explanation is signal managerial competence Subramanyam (1996) reports that firms use earnings management to signal firm value Burgstahler and Dichev (1997) indicate that firms’ image is enhanced with stakeholders such as lenders, suppliers, buyer and valuable workers when

companies meet earning benchmark For example, customers pay a higher price for goods and services, supplier and creditors suggest better term and valuable employees is more motivation for staying Graham et al (2005) provide evidence that in order to meet short-term earning benchmark, CEO is willing to postpone maintenance and advertising cost and even ignore positive NPV project, which helps to convey future growth prospects, keep or increase the stock price, creates trust with creditors and enhance management team’s reputation

2.6 Hypotheses development

The research investigates whether the companies apply real activities management as a plan for manipulation to meet the firm’s target In particularly The research pays attention to a sample of companies which is more likely to implement manipulation According to Graham et al (2005)’s survey, He shows concrete proof that the executive use generally real economic actions in order to keep company’s performance For example, 55.3% of survey managers report that they are a willingness to postpone beginning a new project to reach the profit target Nearly 80% state they would decide decline on R&D spending, maintenance, and advertising to get income target Gunny (2010) and Zang (2012) suggest that an unusual event around zero earning and zero earnings growth and explain that it is the proof of companies which employ real earnings management to beat the slightly earnings benchmark Therefore, I believe that:

Hypothesis 1: In Viet Nam, the listed companies that meet earnings target (zero earnings and zero earnings growth) exhibit the proof of real activities manipulation

The research develops the second hypothesis based on the real big scandal in Viet Nam and in around the word For example, Lehman Brothers Scandal (2008): Lehman brothers try to hide US$50 billion in the loan by sale toxic asset to another the bank with the promise that they will buy bank this sale toxic As a result, instead of report loss 50$ billion, they had more 50$ billion cash and less bad asset than it did Next, In September -2015, Volkswagen admitted that for reducing the cost of cars, they had installed the program on million cars in order to pass emission testers This cheat makes the Environmental Protection Agency thought those cars were more environment-friendlier than they were, which boosted Volkswagen sales volume Next, Toshiba‘s accounting scandal, electronics conglomerate Toshiba revealed overstated income by more than US$ 1.2 billion over seven years In Viet Nam, the Vietnam Shipbuilding Industry Group (Vinashin) is one

of largest cases; the estimated losses are about US$40.4 million, and their debt is run up to US$4.5 billion And, there is long list of corporate financial scandals around the world The question is raised whether the well-known companies use real activities management to report profit and maintain business reputation In addition, the reputation of companies is positively correlated with

their size (the paper measure size of the company as a nature log of total assets) Therefore, the second hypothesis is:

Hypothesis 2: there is the difference between extent of using real activities manipulation of the listed firm that meet benchmark and those meet earnings benchmark and have a high value of an asset

As already mentioned in “impact of real activity management on firm future performance” section, the relationship between firms meet benchmarks by applying real activities manipulation and future company performance could be positive or negative This leads to following hypothesis:

Hypothesis 3: There is no relationship between firm using real activities manipulation to just meet earnings benchmark and future performance

CHAPTER 3 METHODOLOGY AND DATA DESCRIPTION

This chapter presents the methodology and data description used to investigate the extent of real activities manipulation The first part is to introduce how to classify firm into the suspect firm group Proxy capture real earnings management and model for measurement this proxy is presented

in the next chapter Following section, it is a presentation of the empirical model Final section present data is used in this research

3.1 Selection of suspect firm-year

In order to test H1: Firstly, this paper selects suspect firm-year involved in REM Based on Roychowdhury (2006) and Zang (2012), there are two benchmarks:

Zero earnings: net income divided by lagged total assets, between 0 and 0.01

Zero earnings growth: (a variation in earnings): a variation in earnings before interest and taxes, after deflation by lagged total assets, that falls between 0 and 1%

Secondly, identification of manipulation through real activities This paper focus on three types of real earning management: sales manipulation, Decrease in discretionary expenditures and overproduction

3.2 Real earnings proxies

Following Roychowdhury (2006)’s work, we use three variable abnormal cash flow from operating, abnormal discretionary expenses and abnormal production cost as a variable to measure the magnitude of three types of real activities manipulation First all of, the normal level of cash flow from operating, abnormal discretionary expenses and abnormal production cost are estimated, which applies the model is created by Dechow, Kothari and Watts (1998) as developed by Roychowdhury (2006) The next step is calculating abnormal of these three variables (abnormal cash flow from operating, abnormal discretionary expenses and abnormal production cost) The abnormal level of each type of real earning management for every firm-year is measured as true observed value minus the estimated value obtained by applying the models

The paper concentrates on following three real activities management based on idea of Roychowdhury (2006), but we relax some assumptions:

The first real earnings management is an acceleration of timing of sales (sale manipulation) We define sales manipulation as executive attempts to enhance temporarily sale by any approach While

Roychowdhury (2006) limits his definition by sale manipulation through raising price discounts or easy lenient credit terms The manager can apply various way for sale manipulation, all of them lead to affect the current – period CFO (or affect abnormal CFO)

The next, it is production cost manipulation We define production cost manipulation as any action

of manager to reduce production cost instead of the lower cost of goods sold through increased production as Roychowdhury (2006)’s description Because a company can combine many temporal solutions to reduce production cost

Final real earnings management is a decline in discretionary expenditures (Advertising, Research and development spending (R&D), and Selling, General and Administrative (SG&A) expenses)

First, the normal level of discretionary expenses (DISEXP) as a function of lagged sale and is calculated using the following model (Dechow et al (1998) and Roychowdhury (2006))

𝐴𝑖,𝑡−1 = 𝛼0+ 𝛼1 1

𝐴𝑖,𝑡−1+ 𝛽1𝑆𝑖,𝑡−1

𝐴𝑖,𝑡−1+ 𝜀𝑡 (Model 1) Where:

DISEXPit represents discretionary expenses for firm i in year t: sum of selling, general and

administrative expenses (SG&A); R&D spending and advertising spending Provided that SG&A is available, R&D and advertising are put to zero if they are not available,

The regression residual 𝜀𝑡 is related to the magnitude of manipulation by cutting discretionary expenses (Ab_DISEXP)

Abnormal discretionary expenses are actual discretionary expenses minus the normal level of discretionary expenses (fitted values) calculated using the estimated from model 1

Secondly, normal levels of production as a function of the current and lagged change in sales and

is estimated using the following model (presented by Dechow et al (1998)) with production costs are measured as the sum of the cost of goods sold (COGS) and change in inventory during the year

𝑃𝑅𝑂𝐷𝑖,𝑡 is sum of the cost of COGS and change in inventory during period t,

∆𝑆𝑖,𝑡 change in sale,

ΔSi,t−1: lagged change in sales,

S is sales for firm i in period t

Abnormal production cost is the actual production cost minus the normal level of production (fitted values) calculated using the estimated from model 2

Finally, normal level cash flow levels as a function of sales and change in sales and is estimated using the following model (Roychowdhury (2006) and Gunny (2005)):

𝐶𝐹𝑂𝑖,𝑡 is cash flow from operation,

𝐴𝑖,𝑡−1 is lagged of total assets for firm i,

S is sales for firm i in period t,

∆𝑆𝑖,𝑡 change in sale for firm i

Abnormal operating cash flows are conducted as the difference between the actual values and the normal levels predicted from model 3

Model 1,2, 3 is estimated for every industry every year Two-digit SIC version 2 code are applied

to classify industries There are no estimated for industry-year with lower than 5 companies (Following Bartov et al (2001) and Zang (2012))

3.3 Empirical model

3.3.1 Empirical model to test hypothesis 1

Finally, to test the first hypothesis: the relationship between companies just meet benchmarks (zero earnings and zero earnings growth) and real activities manipulation, we estimate flowing model (Roychowdhury (2006) and Gunny (2010)):

𝑌𝑖,𝑡 = 𝛼0+ 𝛽1𝑆𝐼𝑍𝐸𝑖,𝑡+ 𝛽2𝑀𝑇𝐵𝑖,𝑡+ 𝛽3𝑅𝑂𝐴𝑖,𝑡+ 𝛽4𝑆𝑢𝑠𝑝𝑒𝑐𝑡𝑖,𝑡+ 𝜀𝑖𝑡 (4) Where 𝑌𝑡 refers to one of the forms of manipulation through real activities (Ab_DISEXP, Ab_PROD and Ab_CFO), MTB refers to the market-to-book ratio, and SIZE is the natural logarithm of the company’s market value Suspect is a binary variable that takes the value of 1 when the firm-year observation meets one of the benchmarks indicated in section 3.1

𝛽4 will indicate the relationship between companies just meet benchmarks and real activities manipulation In other words, if 𝛽4 is statistically significant, it shows the firms that beat benchmarks using real activities manipulation, absolute value of 𝛽4 presents the extent of real activities manipulation

Equation (4) is estimated using three determine abnormal of real activities management as the depend variable: abnormal CFO, abnormal discretionary expenses, and abnormal production cost Natural log of total assets (size) is used as a control variable to control for company size The effect

of company performance, firm growth opportunities are controlled through the use of net income scaled by lagged of assets (ROA), market to book value (MTB) Model 4 in a particular form as: Ab_DISEXP𝑖,𝑡 = 𝛼0 + 𝛽1𝑆𝐼𝑍𝐸𝑖,𝑡+ 𝛽2𝑀𝑇𝐵𝑖,𝑡 + 𝛽3𝑅𝑂𝐴𝑖,𝑡+ 𝛽4𝑆𝑢𝑠𝑝𝑒𝑐𝑡𝑖,𝑡+ 𝜀𝑖𝑡 (4-1)

Ab_PROD𝑖,𝑡 = 𝛼0+ 𝛽1𝑆𝐼𝑍𝐸𝑖,𝑡+ 𝛽2𝑀𝑇𝐵𝑖,𝑡+ 𝛽3𝑅𝑂𝐴𝑖,𝑡+ 𝛽4𝑆𝑢𝑠𝑝𝑒𝑐𝑡𝑖,𝑡 + 𝜀𝑖𝑡 (4-2) Ab_CFO𝑖,𝑡 = 𝛼0+ 𝛽1𝑆𝐼𝑍𝐸𝑖,𝑡+ 𝛽2𝑀𝑇𝐵𝑖,𝑡+ 𝛽3𝑅𝑂𝐴𝑖,𝑡 + 𝛽4𝑆𝑢𝑠𝑝𝑒𝑐𝑡𝑖,𝑡+ 𝜀𝑖𝑡 (4-3) 3.3.2 Empirical model to test hypothesis 2

According to Siregar and Utama (2008), he suggests that firm size is seemed to be a proxy for information asymmetry in reporting information environments Because the lager firms are more easily placed under scrutiny by regulator and auditor than smaller firms, the small firms are able to manage earnings more successful than small firms Following Siregar and Utama (2008) , we examine that there is the difference between extent of using real activities manipulation of the listed firm that meet benchmark and those meet earnings benchmark and have a high value of an asset In another word, we examine whether the big or well-known companies manage earnings by real activities manipulation We add to equation 4 one dummy variable: Dsize DSize variable assumes the value of 1 if total asset of firm above the 75th percentile of data in year t, and 0 otherwise In instance, DSize for a particular company in the year 2009 equal one if that companies’ total assets

is above 75th percentile for the year 2009

𝑌𝑖,𝑡= 𝛼0+ 𝛽1𝑆𝐼𝑍𝐸𝑖,𝑡+ 𝛽2𝑀𝑇𝐵𝑖,𝑡+ 𝛽3𝑅𝑂𝐴𝑖,𝑡+ 𝛽4𝑆𝑢𝑠𝑝𝑒𝑐𝑡𝑖,𝑡+ 𝛽5𝐷𝑆𝑖𝑧𝑒𝑖,𝑡+ 𝛽6𝐷𝑆𝑖𝑧𝑒𝑖,𝑡∗ 𝑆𝑢𝑠𝑝𝑒𝑐𝑡𝑖,𝑡+

𝜀𝑡 (model 5)

𝑌𝑖,𝑡 : the dependent variable, set equal to abnormal CFO, abnormal production cost, abnormal discretionary expenses

3.3.3 Empirical model to test hypothesis 3

To exanimate whether there is relationship between firm using real activities manipulation to just meet earnings benchmark and future performance, we follow Gunny (2010)’s model as expressed below

ROA𝑖,𝑡+1 = 𝛼0+ 𝛽1𝐵𝐸𝐴𝑇𝑖,𝑡+ 𝛽2𝐽𝑈𝑆𝑇𝑀𝐼𝑆𝑆𝑖,𝑡+ 𝛽3𝑆𝑢𝑠𝑝𝑒𝑐𝑡1𝑖,𝑡+ 𝛽4𝑅𝑀𝑖,𝑡+ 𝛽5𝑆𝑢𝑠𝑝𝑒𝑐𝑡1𝑖,𝑡∗ 𝑅𝑀𝑖,𝑡+

𝛽6𝑅𝑂𝐴𝑖,𝑡+ 𝛽7𝑆𝐼𝑍𝐸𝑖,𝑡+ 𝛽8𝑀𝑇𝐵𝑖,𝑡 (model 6)

Where

 ROA: Income before extraordinary items divided by lagged total asset,

 BEAT: Set equal 1 if net income divided by lagged total assets equal or greater than 0.01, zero otherwise,

 JUSTMISS: set equal 1 if net income divided by lagged total assets is in range -0.01 and 0

 Suspect1: set equal 1 if net income divided by lagged total assets, between 0 and 0.01,

RM_CFO: set equal 1 if the abnormal CFO value of company from model 3 is

in the highest quintile, zero otherwise,

 SIZE: Natural log of total assets,

 MTB: market to book value

We assume that the companies which suspected using cutting discretionary expenditures to manage earnings are companies in the lowest quintile of abnormal discretionary expenditures (model 1) Next, the companies which suspected using production cost manipulation to manage earnings are companies in the lowest quintile of abnormal production cost (model 2) Finally, the companies which suspected using CFO to manage earnings are companies in the highest quintile of abnormal CFO (model 3) (we follow Gunny (2010)’s assumption)

3.3 Data collection

3.3.1 Data collection

Data collected on Orbis website The sample includes firms listed in Viet Nam from 2005 to 2015 except firms in banking, financial institutions, and real states And this paper needs at least 5 observations each industry-year Each company’s industry is defined by two-digit SIC code The sample have 2374 firms-year observations which have full available variable after we winsorize at the bottom 1% and the top 99% percent of their distribution We winsorize to avoid

an effect of outlier