The effect of related party transactions on earnings management of listed companies in Vietnam

Policies and Sustainable Economic Development | 577 The Effect of Related Party Transactions on Earnings Management of Listed Companies in Vietnam NGUYEN THI THU HA Hue College of Economics, Hue University - thuhanguyen262@gmail.com Abstract Recent presence of related party transactions in scandal accounting frauds has attracted academic researches’ concerns about the role of related party transactions in earnings management There are mixed academic results about the effect of related party on earnings management Hence, the research aims at examining the association between related party transactions and earnings management of listed companies in Vietnam To test this relationship, the match pair analysis is used by matching firms disclosing related party transactions with firms not disclosing related party transactions with similar total assets and the same industry Previous study presents the advantage of match pair analysis in minimizing concerns regarding the role of potential confounding effects The discretionary accruals are measured as the proxy of earnings management The Cross Sectional Modified Jones Model (1991) is used to measure accruals Moreover, the multi regression analysis is applied to investigate the effect of related party transactions on earnings management The findings of the study indicate that the presence of related party transactions may induce higher earnings management Keywords: Fraud; related parties; special purpose entity; earnings management; transparency 578 | Policies and Sustainable Economic Development Introduction A related party transaction is defined as a transfer of resources, services, or obligations between a reporting entity and a related party (IAS 24) Examples of related party transactions include transactions between a company and its subsidiaries, affiliates, principal owners, officers or their families, directors or their families, or entities owned or controlled by its officers or their families (Statement of Financial Accounting Standards No 57, 1982) The presence of related party transactions in fraud scandals such as Adelphia, Enron, Parmalat, or Tyco have increasingly attracted public attention in recent years Two views of related party transactions as efficient transactions and conflicts of interests are considered (Gordon et al., 2004) The first view regards related party transactions as normal transactions to meet the economic need of firms The opposite view indicates that related party transactions are conflicts of interests This view indicates that related party transactions are similar to agency problems (Berle & Means, 1932; Jensen & Meckling, 1976) In this situation, related party transactions can lead to earnings management defined by Healy and Wahlen (1998), in which managers structure related party transactions to alter financial reporting for purposeful intention Indeed, related party transactions are considered as one of red flags of potential fraud in the auditing literature According the ISA 550 - Related parties, there are higher risks of misstatements of related party transactions because the nature and relationship between related parties Reasons for increased fraud of related party transactions are the complicated network of relationships, or related party transactions might not occur as the normal commercial terms such as similar arm’s length transactions Kohlbeck and Mayhew (2005) suggest that the existence of related party transactions is more common when management has incentives such as executive stock option compensation to engage in these related party transactions Empirical evidences have shown that Chinese listed companies have been involved in related party transactions to manage earnings, aiming at avoiding being delisted or during the period of initial public offering (Jian & Wong, 2004) Moreover, Henry et al (2007) investigate 83 cases of SEC enforcement actions from 1983 to 2006 and indicates that related party transactions such as loans to related party transactions, payments to company officers for services or sales of goods or services to related entities were used to misstate financial statements According to Vietnamese Accounting Standard 26 – Related parties, each related party transaction between a parent company and its subsidiaries, controlled companies as well as between subsidiaries of the same company is required to disclose the nature as well as the amount of related party transactions No disclosures of related party transactions are required in consolidated financial statements or in the financial statements of a parent or in the financial statements of a wholly-owned subsidiary if its parent is incorporated in Vietnam and provides consolidated financial statements in Vietnam In addition to this accounting standard, there is no detailed instruction of recording related party transactions for listed companies Hence, litigation of related party transactions in Vietnam has exposed weakness in providing specific guidance on related party transactions Consequently, some Policies and Sustainable Economic Development | 579 companies have involved in earnings management throughout related party transactions (Nguyen, 2015) Moreover, while Vietnam’s stock market has been contributing to the economic development of Vietnam, there is huge room for development such as low transparency of financial information and weak investor protection Therefore, transactions with related parties may be considered as a mechanism to induce managers to manage earnings, targeting at altering financial performance of a firm to mislead users of financial statements Hence, it is believed that the presence of related party transactions of listed companies in Vietnam can lead to higher earnings management Theoretical basis and methodology 2.1 Theoretical basis and analysis framework 2.1.1 Earnings management Agency theory is considered as the major theory for academic literature focusing on explaining incentives for earnings management behavior Indeed, according to the agency theory, the separation of ownership causes agency conflicts between managers and shareholders (Jensen & Meckling, 1976) Agency theory is concerned with conflicts between principal and agents due to information asymmetry between principal and agent Basing on agency theory, there is research about various incentives for earnings management, including capital market, compensation, debt covenant violation, political cost, and regulatory factors (Healy & Wahlen, 1999) a Capital market Capital market known as stock market is an incentive for earnings management Indeed, investors have a tendency to depend on analysts’ views or forecasts to make an investing decision Under capital market pressure, managers might manipulate earnings to meet analyst’ forecasts and allow a firm to improve financial performance (Gramham, 2005) Similarly, Stein (1989) indicates that to meet analysts’ forecasts or avoid negative earnings, managers have more incentives to manage earnings b Management compensation The management compensation hypothesis indicates that managers have incentives for maximizing their compensation throughout selecting accounting procedures (Watts & Zimmerman, 1978) Hence, managers will intend to increase reported earnings that affects their compensation Similarly, Cheng and Warfield (2005) find evidence of a positive association between earnings management and stock-based executive compensation Indeed, management wealth is affected by future stock performance; therefore, management can reserve earnings in the current year to avoid subsequent earnings surprises c Debt contract 580 | Policies and Sustainable Economic Development Debt contract hypothesis indicates that firms with closer accounting-based debt covenants are more likely to choose accounting procedures to shift earnings from future periods to a current period Defond and Jiambalvo (1994) and Klein (2002) find that companies utilize accrual manipulation to decrease debt covenant violation In fact, at firms with high financial constraints, earnings are managed to reduce debt d Political cost Another incentive for earnings management is political cost According to the political cost hypothesis, managers can involve in earnings management to decrease political costs by reducing earnings in the current year Watts and Zimmerman (1978) state that companies facing political risks are expected to have earnings management to lower their risk e Regulatory factors Regulatory factors can be motivations for earnings management Indeed, Fong (2006) presents that in some industries such as banking, due to minimum capital regulation, managers have incentives to implement earnings management techniques to meet these regulatory requirements 2.1.2 Related party transactions Recent presence of related party transactions in scandal accounting frauds has attracted academic researches’ concerns about the role of related party transactions in earnings management However, it is controversial that when a manager has incentives to manage earnings, a firm involves in related party transactions to mask inappropriate wealth transfer (Gordon & Henry, 2005) With the conflicts of interest hypothesis, related party transaction can lead to agent-principal conflicts Therefore, the existence of related party transactions may harm the interest of shareholders Previous studies prove that related party transactions are the contributor of agency costs Indeed, Kohlbeck and Mayhew (2004) suggest that related party transactions are positively associated with executive stock option compensation According to this view, related party transactions are sources of earnings management Indeed, due to dependent transactions among related parties, related party transactions are traded by non-arm’s length transactions Prior researches also prove that related party transactions are related to earnings management Chen et al (2011) indicate related party transactions as a source of earnings management by finding support for positive relationship between related party transactions and Chinese IPOs’ operating performance in pre-IPO period, but after IPO period, this positive association fades Furthermore, Jian and Wong (2010) find that Chinese listed firms from 1998 to 2002 manipulate earnings by using abnormal sales transactions to related parties Moreover, it is found that abnormal related sales are not wholly accrual-based but also cashbased Indeed, there is cash transfer throughout lending transaction to controlling firms after the propping However, there is no evidence for cash transfer through related lending transactions related to earnings management Gordon and Henry (2005) summarize seven different types of related party transactions including 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 Policies and Sustainable Economic Development | 581 related party, loans to related parties, fixed-rate financing from related parties, investment and other They find evidence of relationship between earnings management and related party transactions but only for one certain type of related party transaction such as fixed-rate financing from related parties Basing on 83 cases of SEC enforcement actions from 1983 to 2006, of Henry et al (2007) classify different types of related party transactions and investigate how a firm uses related party transaction to misstate financial statements In detail, fictitious sales to a related party are used to overstate revenues or purchase to related parties are not reported to understate expenses and hence overstate income Moreover, another mechanism is non-recognition of borrowings from related parties, resulting in understated liabilities Nevertheless, some empirical research, i.e Kuan et al (2010) does not find the relationship between related party transactions and earnings management 2.1.3 Related party transactions in Vietnam In an emerging capital market like Vietnam, ownership is significantly concentrated (La porta et al., 1998) Indeed, McGuinness and Ferguson (2005) indicate that when the ownership is belong to only one owner, the agency issue turns into the conflicts of interest between controlling shareholders and minority shareholders Johnson et al (2000) highlight that another form of agency conflict is that controlling shareholders pursue their private benefits at the costs of outsiders Hence, the agency problem shifts from managers and shareholders in developed countries to controlling shareholders and minority shareholders in developing countries While controlling shareholders have a power to make a decision, there is limited power of decision making by minority shareholders Therefore, noncontrolling shareholders’ interests are mainly profit-seeking However, the information asymmetry between controlling shareholders and minority shareholders may lead to agency conflicts when controlling shareholders may exploit the minority shareholders in such as a way to manage earnings for controlling shareholder’s own interests Incentives of controlling shareholders to manage earnings because it is probable that such firms are not attractive and market share values are less than that of market peers Hence, it can lead to the conflicts of interests between large and small shareholders This conflict of interests can include expropriation, i.e., controlling shareholders can manage earnings through using related party transactions such as sales to related parties Especially, in the context of Vietnam, there is the weak corporate governance without an independent board of directors as well as an audit committee; hence, it is probable to have higher opportunities for controlling shareholders to expropriate minority shareholders’ benefits (Johnson et al., 2000) Especially, as noticed in accounting standards, related party transactions expose highly significant risks of the expropriation of the firm’s resources (Gordon & Henry, 2005) It is due to the fact that related party transactions are approved by controlling shareholders as well as implemented by closely dependent relationship between related parties Jiang and Wong (2003) indicate that Chinese business groups manipulate earnings throughout using related party transactions with parent companies Although Vietnamese Accounting Standard 26 – Related party transaction requires listed companies to disclose related party transactions, there is no specific regulation in the accounting standard or Enterprise Law 2005 about prices of related party transactions as well as instruction 582 | Policies and Sustainable Economic Development about recording special purpose entity This induces higher chances for controlling shareholders to involve in related party transactions to manage earnings in listed companies in Vietnam Jing and Wang (2013) introduce some motivations for companies using related party transactions to manage earnings, such as to maintain resources or qualify for refinancing as well as enhance short-term activities of these firms’ controlling shareholders Similarly, Nguyen (2015) presents that some companies in Vietnam such as VPC Saigon created a special purpose entity to manipulate revenues in order to avoid decreasing financial performance Moreover, some Vietnamese listed companies manipulate related party sales transactions (Nguyen, 2015) to improve financial performance Consistent with Jian and Wong (2004), Chinese group companies report high related party sales to increase earnings in order to avoid delisting Thus, with ownership concentration, low transparency of financial information as well as inefficient judicial system, there potentially is higher earnings management and related party transactions Nevertheless, academic research in this field is limited in Vietnam, especially using listed companies to examine Therefore, this study aims at reducing this shortage by investigating relationship between related party transactions and the level of earnings management in Vietnamese listed companies for the fiscal year 2015 2.2 Methods 2.2.1 Sample selection and data collection There have no studies using listed firms in Vietnam to investigate relationship between earnings management and related party transactions Therefore, the sample of listed firms in the latest fiscal years 2015 are selected to examine the relationship between related party transactions and earnings management Moreover, the matched-pair design is applied in this research Indeed, the matched pair design is widely used by previous research such as Dechow et al (1996) and Bartov et al (2000) Bartov et al (2000) present the matched-pair design may alleviate correlated-omitted variables problems that biases results of discretionary accruals Therefore, using matched pair design is believed to minimize concerns related to the effect of potential confounding effects Moreover, Gordon et al (2010) indicate that the most significant advantage of the matched-pair design is to provide more robust findings in a small sample The main criterion for selecting companies in the sample of this research is that each non-disclosing firm is matched with a disclosing firm with the same fiscal year, industry and the closest size (i.e total assets) Moreover, financial firms are excluded from the sample because financial firms have different financial reporting environment and it is difficult to define abnormal accrual for financial companies (Peasnell & Young, 1999) All of listed companies’ financial statements are scanned to classify as a disclosing firms and nondisclosing firms By using key word “related party transactions” in the footnotes of financial statements of listed firms in Vietnam, related party transactions are found Among the total population of 993 companies (Bloomberg Database, 2015), there are 49 non-disclosing firms Hence, all of non-disclosing firms are chosen; furthermore, 49 disclosing firms are matched with 49 nondisclosing firms with the similar total assets, same industry and the given year As for 49 disclosing Policies and Sustainable Economic Development | 583 firms, related party transactions are grouped into relative types of related party transactions to investigate the association with earnings management There are four main types of related party transactions including purchase of goods or contract services, sales to related parties, loan to related parties and other Each type of related party transactions is measured by the amount of related party transactions 2.2.2 Variable definition – Related party transactions To check the effect of related party transactions on earnings management, the related party transaction variable is like a dummy variable (1 for a generic disclosure of related party transaction, otherwise) 2.2.3 Measure of earnings management Past research attempts to use accruals as a measure of earnings management While Healy (1985) and DeAngelo (1986) use total accruals as a measure of earnings management, Jones (1991) and Dechow et al (1995) apply discretionary accruals as a proxy of earnings management Among these models, it is believed that the Jones (1991) model is the most commonly used However, Dechow et al (1995) replace changes in revenues with cash revenue growth They argue that revenue growth is subject to earnings management Guay et al (1996) prove that modified Jones model (1995) is more powerful in detecting earnings management Moreover, Bartov et al (2001) prove that the crosssectional Modified Jones Model is able to detect earnings management throughout examining the association between audit qualification and earnings management Hence, this research uses discretionary accruals as measures of earnings management Discretionary accruals are equal to total accruals minus non-discretionary accruals (DeAnglelo, 1986) In addition, the non-discretionary accruals are estimated, basing on the Modified Jones Model by Dechow et al (1995) for all firms k in industry j at year t As noted that the beginning of total assets are used as a deflator to minimize the scale effects (Barth and Clinch, 2009) The model is as below: NDAjk,t = αj,t[(1/Ajk,t-1)] + βj,t[(∆REVjk,t- ∆RECjk,t)/ Ajk,t-1] + γj,t[(PPEjk,t)/ Ajk,t-1] (1) where: NDAjk,t: non-discretionary accruals for firm k in industry j in year t Ajk,t-1: Total assets for firm k in industry j in the year t-1 ∆REVjk,t: Changes in revenue for firm k in industry j in year t ∆RECjk,t: change in net receivables for firm k in industry j in year t PPEjk,t: Gross property, plant and equipment for firm k in industry j in year t As noted that the coefficient α, β, γ from the Equation (1) acquired from original Jones Model (1991) not from Modified Jones Model (Bartov et al., 2000) 584 | Policies and Sustainable Economic Development Similar to previous study with discretionary accrual estimation (Healy, 1985; Jones, 19991, and Dechow et al (1995), the total accruals are estimated by the following formula: TAk,t = (∆CAk,t - ∆Clk,t - ∆CASHk,t + ∆STDk,t – Depk,t)/(Ak,t-1) (2) where: TAk,t: total accruals for firm k in the year t ∆CAk,t: changes in current assets for firm k in the year t ∆Clk,t: changes in current liabilities for firm k in the year t ∆CASHi,t: changes in cash and cash equivalent for firm k in the year t ∆STDk,t: changes in short term debts for firm k in the year t Depk,t: depreciation and amortization for firm k in the year t Ak,t-1: total assets for firm k in the year t-1 Therefore, the abnormal accrual defined as: DAC=TAk,t - {αj,t [(1/Ak,t-1)] + βj,t [(∆REVk,t- ∆RECk,t)/ Ajk,t-1] + γj,t [(PPEk,t)/ Ak,t-1]} where α, β, γ are the fitted coefficients from Equation (1) 2.2.4 Control variables In the regression analysis, other variables are included in the model to control for other factors associated with earnings management Therefore, to examine the relationship between related party transactions and earnings management, other variables having relationship with earnings management must be controlled, which includes cash flow, leverage, firm size, return on asset and Big-4 company Bartov et al (2000) and Kothari et al (2001) indicate that no control of these factors leads to type I error that results wrongly reject null-hypothesis when it is true a Operating cash flow (OCF) Chung et al (2005) prove that low growth companies with high free cash flows have motivations to use discretionary accruals to downgrade their earnings levels Moreover, Dechow (1994) indicates that accruals are negatively associated with operating cash flows b Leverage (LEV) While positive accounting theory indicates that managers have incentives for managing earnings to decrease the costly debt covenants, recent empirical research that investigates a manger’s accounting choices when a firm is closer to debt covenant constraints has mixed results (Healy & Palepu, 1990; Defond & Jiambalvo, 1994; DeAngelo et al., 1994) Moreover, leverage is considered as significant incentives for earnings management (Christie, 1990) Sweeney (1994) examines 130 firms reporting violations of their financial statements and finds that managers of firms reaching defaults select income-increasing accounting changes Policies and Sustainable Economic Development | 585 c Firm size (Log(Assset)) There are various views regarding the association between firm size and earnings management While Warfield et al (1995) illustrate that large firms with stronger internal control systems have less earnings management, Barton and Simko (2002) reveal that large firms have incentives to management earnings in order to meet the expectation of analysts Similarly, Myers and Skinner (2007) indicate that large firms did not show their real earnings Furthermore, Watts and Zimmerman (1987) indicate that large firms have more incentives to manipulate earnings rather than small firms due to a compensation policy d Return on asset (ROA) The return on asset is included to measure the firm’s earnings performance Some prior empirical research suggests that the performance of a firm is associated with earnings management Indeed, Lee et al (2005) find that the firm’s earnings performance is positive associated with earnings management (Lee et al., 2005) Moreover, Dechow et al (1995) prove that firms with negative reported earnings have negative discretionary accruals Similarly, McNichols (2000) proves that discretionary accruals are positively associated with the firm’s performance e Big-4 audit firm (BIG-4) De Angelo (1981) indicates that large audit firms such as Big Four probably provide higher quality audit than small audits firms; therefore, Bauwhede et al (2003) prove that audit quality is associated with earnings management Moreover, Becker et al (1998) prove that firms with a Big 4’s auditors have lower amount of earnings management 2.2.5 Empirical model Equation To examine the association between related party transactions and earnings management as well as further investigate which type of related party transactions related to earnings management, the regression analysis is used The discretionary accrual is used as the proxy for earnings management and the related party transactions as well as control variables are the dependent variables If earnings management exits, there is the relationship between discretionary accruals and related party transactions The model below is applied to investigate whether a manager has incentives for earnings management, related party transactions are involved The model is used in the regression analysis is as follow: DAC = α +βRPT + γ1CFO + γ2LEV + γ3ROA + γ4Firm size + γ5Big4 + e Table below summarises the definition of variables applied in this study 586 | Policies and Sustainable Economic Development Table Variables definition and description Variable title Dependent variable DAC Independent variable RPTs Control variables Operating cash flow (OCF) LEV ROA Firm size Big4 Variable description Discretionary accruals measured by the Modified Jones model (1991) of listed companies Dummy variable with for a generic disclosure of related party transactions, otherwise Current year's operating cash flows divided by beginning of the year total asset Current year's long term debt divided by beginning of the year total assets Ratio of net income to book value of the current year’s total assets Natural logarithm of total book value of the current year’s total assets Big4 firms (PWC, EY, KPMG, Deloitte) Results 3.1 Descriptive analyses of related party transactions Table below shows the summary of the sample selection for the fiscal year 2015 The total of firms with non-disclosing related party transactions is 49, occupied of 5% of total listed firms and the total number of disclosing firms is 935, made up 95% of total population By using the matched pair tests, 49 non-disclosing firms are matched with 49 disclosing firms with the closest total assets and industry in the same latest year 2015 Moreover, Table describes the industry distribution of non-disclosing firms and disclosing firms, classified by the International Standard Industrial Classification of All Economic Activities (ISIC), which is taken from the Bloomberg database Among different industries, the Capital Goods and Material Industry occupy the highest percentage with 26% and 24%, respectively Moreover, the smallest sample is the Commercial and Professional Services as well as Utilities Industry Table Summary of the sample selection Number of firms Percentage Total listed companies in the year 2015 993 Less banks (9) Valid observations, firm-years 984 Total firms with non-disclosing related party transactions 49 5% Total firms with disclosing related party transactions 935 95% Disclosers (firm-years) 49 50% Non-disclosers (firm-years) 49 50% Sample size Policies and Sustainable Economic Development | 587 Total Sample (firm-years) 98 100% Table Sample industry composition Classified industry by ISIC Sample Companies Sample Composition Capital Goods 26 27% Commercial & Professional Services 2% Consumer Durables & Apparel 8% Energy 4% Food Beverage & Tobacco 6% Health Care Equipment & Services 2% Materials 24 24% Media 4% Pharmaceuticals, Biotechnology & Life Sciences 4% Real Estate 6% Retailing 4% Technology Hardware & Equipment 2% Transportation 6% Utilities 2% Total 98 100% Table as follow presents the descriptive statistics of disclosing firms and non-disclosing firms As noted, there is little difference regarding to total assets, net incomes, operating cash flows and long-term debt between disclosing firms and non-disclosing firms Hence, the matched pair design might help alleviate the concerns related to different firms’ characteristics to total accruals Table Descriptive statistics Disclosing firms Thousand VND N Mean Median Min Max Total Assets 49 493,185,855 326,858,633 15,281,489 2,224,012,209 Net income 49 27,082,881 10,526,052 (51,897,878) 260,158,328 Operating cash flows 49 66,320,148 15,176,380 (83,255,311) 542,598,937 Long-term debt 49 32,991,778 0 417,576,842 Non-disclosing firms Thousand VND N Mean Median Min Max Total Assets 49 464,779,251 252,075,258 15,586,706 2,434,420,979 Net income 49 28,274,756 12,074,317 (47,999,809) 282,339,976 Operating cash flows 49 33,262,318 18,601,445 (62,979,773) 558,849,684 Long-term debt 49 67,349,177 2,289,222 544,942,017 588 | Policies and Sustainable Economic Development 3.2 Earnings management As shown in Table 5, the average discretionary of disclosing firm is higher than non-disclosing firm, which is 0.02 and (0.02), respectively This mean DAC is quite similar to other countries such as China or Singapore (Rusmin, 2006; Cheng & Chen, 2007) However, it is not consistent with Leuz et al (2003) indicate that developing countries like Vietnam have higher levels of earnings management However, the distribution of discretionary accruals for disclosing firms and nondisclosing firm are the same with 0.14 Furthermore, the percentage of positive and negative DAC occupies approximately 50% In addition, Table indicates the paired t-test for average DAC of disclosing firms and non-disclosing firms Indeed, the mean DAC of disclosing firm is statistically higher than that of non-disclosing firm (p