Does Corporate Governance affect Earnings Quality: Evidence from an Emerging Market
Muhammad Nurul Houqe, Tony van Zijl, Keitha Dunstan, A.K.M. Waresul Karim ABSTRACT This study examines whether firm governance practices have an effect on accounting earnings quality in Bangladesh. Using a sample of 648 firm-year observations for the period of 2001 – 2006, the result suggests that corporate governance mechanisms provide greater monitoring of financial reporting and these mechanisms positively affect firm earnings quality. This knowledge could be used by the capital market regulator, the Securities and Exchange Commission (SEC), to design mechanisms to constrain managers‟ earnings management practices. This, in turn, would contribute towards development of an efficient stock market and thus protect investors‟ interests. Key word: earnings quality, discretionary accruals, corporate governance. JEL classification: J3; K2; M4. 1. INTRODUCTION This study examines whether firm governance practices have an effect on accounting earnings quality in Bangladesh. Extant research suggests that “a better-governed firm reduces control rights which stockholders‟ and creditors confer on managers, increasing the probability that managers invest in shareholders‟ value-creating projects” Shleifer and Vishny (1997:132).
The Term Paper on Corporate Governance and Firm Value
... ownership by blockholders and disclosure quality of firms as corporate governance measures. In regressions (1), ( ... shareholding agreements and the scarcity of mechanisms to control their discretionary power. Thus, ... have a strong incentive to manage earnings during an economic crisis. In contrast, ... G15; G21; G32; G33; G34 Keywords: Corporate governance; Firm value; Chaebol; Controlling shareholder; Main ...
Bangladesh firms has not faced similar to those accounting scandals reported in larger economies such as the US, the UK, the Canada and Australia. Nevertheless, there is concern is poor earnings quality (Kabir et al. 2008, Habib and Islam 2007, and Habib 2005).
During the period 1993 to 2000 six major developments took place in regards to corporate financial reporting in Bangladesh. First, the Securities and Exchange Commission (SEC) was established in 1993, under the SEC Act of 1993. Second, the Companies Act 1913 was replaced by the new Companies Act 1994 that came into force from October 1, 1995. Third, in October 1997 the SEC revised the Securities and Exchange Rules (SER) of 1987. The revised SER require listed companies to prepare half yearly accounts within one month of the end of the first six months of its financial year, and to submit those accounts to the stock exchange (s) on which its securities are listed to both holders of its securities and the SEC. Fourth, the share market debacle of 1996, uncovered wilful misconduct in the management of companies and by stockbrokers of the Dhaka Stock Exchange (DSE) and also the Chittagong Stock Exchange (CSE).
In addition, the Report of the Enquiry Committee (1998) found that there had been lack of enforcement by regulatory bodies and the absence of timely reliable financial information in the market. Fifth, in response to the 1996 stock market debacle, the SEC required public listed companies to hold Annual General Meetings and to publish annual reports on a timely basis. Finally, in January 2000 the SEC further revised the SER to require, among other things, that the accounts of an issuer must be audited within 120 days of year end or within a period as extended by the SEC. The first motivation of this study is the lack of empirical work on earnings quality in emerging economies, particularly Bangladesh. The bulk of the research on earnings management has been conducted on countries with more effective governance mechanisms in place. Few studies have focused on emerging markets where it is relatively easier for management to expropriate minority shareholders interests because of the lack of effective governance structures such as dispersed ownership, independent boards, independent auditors, an active takeover market, and demanding disclosure requirements. Secondly, there is an ongoing discussion and debate about the association between corporate governance practices and earnings quality (Larcker et al. 2007).
The Term Paper on Bangladesh Studies
The main text covers military and guerilla actions and touches upon political as well as other related matters. Comparatively wider coverage of the armed actions does not imply greater importance or precedence of military actions over the long-drawn struggle of the vast majority of the people. The significance of freedom movements cannot be comprehended by studying military actions or guerrilla ...
Using a sample of 648 firmyear observations from 2001-2006, the results of this study suggest that corporate governance mechanisms provide greater monitoring of financial reporting and these mechanisms positively affect firm earnings quality. This study extends the current literature in a number of ways. Firstly, the effect of corporate governance reform in Bangladesh has not been extensively researched. Secondly, this study (as with Peasnell (2006)) examines a full range of governance variables compared to other extant research
1
Electronic copy available at: http://ssrn.com/abstract=1726134
studies which tend to investigate limited governance attributes (Fan and Wong 2005 and Hsu and Koh 2005).
In doing so, this study supports the stance taken by Pergola (2006) that earnings quality is not explained by a single governance variable but by groups of governance variables. Finally, this study extends the work done by Kabir et al. (2007) and Habib and Islam (2007) by undertaking broad based earnings quality research on a sample of Bangladeshi publicly listed companies. In this context, the objective of this paper is to identify corporate governance factors that significantly contribute towards firm earnings quality in Bangladesh. The rest of the paper is organized as follows. The next section describes the corporate governance environment in Bangladesh. Section 3 presents a theoretical framework that outlines the expected determinants of earnings quality and we then develop our hypotheses on the basis of this framework. Section 4 discusses research design issues. Section 5 of the study presents the empirical results and Section 6 presents the conclusions. 2. CORPORATE GOVERNANCE ENVIRONMENT IN BANGLADESH In Bangladesh, the corporate sector is a combination of local, private, foreign-multinational and public sector companies. Only very few have any international operations. The number of public sector companies in Bangladesh is moderately large. In most of the companies, there is no effective structure for internal audit review and reporting on internal control. Moreover, in companies where there is some kind of internal audit, the auditors are not independent, financial reporting is faulty, and there is no audit charter or requirement for internal audit (Iqbal 1997).
The Essay on Corporate Governance Companies Cadbury 1992
"Corporate governance is concerned with holding the balance between economic and social goals and between individual and communal goals. The aim is to align, as nearly as possible, the interest of the individuals, corporations and society" Sir Adrian Cadbury 1999 The UK's economy depends on the drive and efficiency of its companies. The companies must be free to drive forward, but exercise that ...
Table-1: Ownership Structure of Listed Corporations Distribution of Share Percentage Family/Sponsors/Individual 72.5 Bank /financial institution 3.1 Government/Investment Corporation of Bangladesh 7.1 Employees 1.3 Foreigners 16.0 Total 100 Source: Dhaka Stock Exchange Fact Book (Chowdhury and Chowdhury (2004)) In Bangladesh many companies („family companies‟) are controlled individuals or families. These companies are high ownership concentration and strong preference to retain control (Chowdhury and Chowdhury 2004).
Table-1 shows that in Bangladesh 72.5% of the outstanding shares in 2004 were owned by family/sponsors and individuals. Bank and financial institutions held 3.1%, foreign investors 16.0%, and government 7.1%. This ownership structure creates opportunities for management to engage in earnings management to expropriate wealth from minority shareholders (Shleifer and Vishny 1997).
Given the above ownership pattern, agency problems between managers and shareholders (Jensen and Meckling) are more likely to arise between block owners and other shareholders (Shleifer and Vishny 1997).
As a result, there is a potential for expropriation of minority shareholders‟ wealth in the firms by management representing family/sponsors/individual founding families. Nevertheless, other forms of ownership, in particular institutional shareholdings, are able to counterbalance such impacts on earnings quality through their larger stake and ability to monitor managerial opportunistic behaviour. The above ownership pattern means that the agency conflict between minority shareholders and majority shareholder(s) prevails, as suggested by Shleifer and Vishny (1986), Habib (2005), Habib and Islam (2007) and Farooque et al. (2007).
The Essay on Harvard Management Company
Portfolio to correspond to their benchmark, according to the modern portfolio theory (Markowitz, 1952), whose goal is to minimize the variance for a given return. The main advantage of the optimal portfolio allocation lies in its ability to provide weights on how to invest a given amount of money based on a few inputs. Optimal portfolio allocation is easy to implement, yet it faces some issues and ...
Hence, the governance models based on the traditional agency theory are likely to be ineffective. 3. THEORETICAL FRAMEWORK AND HYPOTHESIS DEVELOPMENT In this section we develop we develop hypotheses on the impact of various corporate governance mechanisms on earnings quality. 3.1. Management Shareholding Core et al. (1999) document the prevalent use of stock options in the United States as a form of managerial compensation and alignment of shareholders and managers interests in the late 1990‟s, with
2
Electronic copy available at: http://ssrn.com/abstract=1726134
outstanding options accounting for 7 percent of total outstanding shares, and with executives holding approximately a third of these options. The resulting alignment of interest between the manager and the shareholders has resulted in research findings of a lower level of earnings management by the managers when managerial ownership is present in the company (Weisbach 1998, Warfield and Wild 1995 and Peasnell et al. 2005).
However, a more recent view argues that managerial ownership may result in a higher level of earnings management due to the managers being entrenched. Within the entrenchment range, insiders gain effective control thus making it harder for the owners to monitor the actions of the managers and easier for the managers to overcome the governance mechanisms in place (Pergola 2006).
Entrenched managers pose a risk to shareholders as such managers not only have control and expertise but also ownership rights making it even more difficult for shareholders to rein in the managers. However, it is difficult to pinpoint a stage where managerial ownership of shares could result in entrenchment, which would be detrimental to the shareholders interests. So, the relevant null hypothesis related to management shareholdings is: H01: There is no relationship between management shareholdings and earnings quality. 3.2. Institutional Shareholding Institutional investors are expected to better monitor the actions of managers given their less dispersed shareholding and better access to resources such as professional advice and analysts‟ reports. Studies in other countries document that institutional ownership levels are a significant proportion of company ownership. Institutional ownership made up 51.6 percent of total market value of US listed companies in 1996 (Gompers and Metrick 2001), an average of 65 percent of all UK listed companies between 1990 and 1995 (Hsu and Koh 2005) and around 49 percent of Australian listed companies in 1997 (Stapleton 1998).
The Research paper on The Earnings Management Issue of WorldCom Case Study Report
... earnings management” and “fraud” based on if the manipulation is intent to affect wealth transfer among company ... Shivakumar, L 2008, ‘Earnings quality at initial public offerings’, ... Earnings Management: Two Main Incentives 1.2.1 The Incentive of Earnings Management for Personal Compensation Plan Daniel and Thomas (2006) states that if CEOs’ potential total ... disclosed in the board meeting was limited, ...
However, extant literature provides mixed views on the role that institutional investors play in corporate governance. One view presents institutional investors as passive actors in corporate governance mechanisms due to their frequent trading and fragmented ownership of companies, which leads to situations where such investors, rather than taking an active role in the operations of their investment, “vote with their feet” and divest of entities that are poor performance or otherwise incongruent with their needs (Porter 2002 and Hsu and Koh 2005).
A contrasting view is provided by Bushee (1998) who presents institutional investors as active participants in governance issues where institutions take a long-term view on the value of their portfolio companies. Hsu and Koh (2005), in a study of the governance role of institutional investors in Australia, found that long-term institutional investors take an active role in governance of companies by constraining earnings management actions by managers. Other factors that contribute towards institutional investors taking on governance roles include the risk of suffering larger losses due to their relatively larger ownership of companies compared to individual ordinary shareholders (ibid).This study extends the current literature by investigating the effect of institutional shareholding on managers‟ opportunistic behaviour to manage earnings for listed companies in Bangladesh and the relevant null hypothesis therefore is: H02: There is no relationship between institutional shareholdings and earnings quality. 3.3. Government Shareholding There are few studies on the impact of government shareholding on earnings quality. Kole and Mulherin (1997) did not find that government shareholding has any significant relationship with earnings‟ quality. Xu and Wang (1999) find that earnings quality is either negatively correlated or not correlated with the proportion of government shareholdings. So, the null hypothesis regarding the government shareholdings is: H03: There is no relationship between government shareholdings and earnings quality. 3.4. Board Size Board size is an important element in board characteristics that may have an impact on earnings‟ management. Goodstein et al. (1994) argued that smaller boards, between four to six members, might be more effective since they are able to make timely strategic decisions, while larger boards are more
The Essay on Public Company Accounting Oversight Board Will It Protect Investors
Public Company Accounting Oversight Board; Will it Protect Investors? The Public Company Accounting Oversight Board (PCAOB) was created by Sarbanes-Oxley Act of 2002. This board was created to oversee the audit of public companies, subject to the securities laws, in order to protect the interests of investors (15 USC 7201, 2002). It was created in wake of the recent financial scandals of Enron, ...
3
capable of monitoring the actions of top management (Zahra and Pearce 1989).
However, large boards having members with varied expertise could increase the synergetic monitoring of top management in reducing the incidence of earnings‟ management. Xie et al. (2003), Peasnell et al. (2006) and Rahman and Ali (2006) found that having a larger board is associated with less earnings management. As such, the null hypothesis relating to board size and earnings quality is as follows: H04: There is a no relationship between earnings quality and the size of boards. 3.5. Board Independence Outside directors can play an important role as independent scrutinisers of managements‟ actions and thus protect shareholders‟ interest. The literature on governance emphasises the role played by outside directors in ameliorating agency problems between the divergent interests of the shareholders and management of the company through monitoring of managerial behaviour (Peasnell et al. 2005).
Moreover, Fama (1980) argues that independent directors have an incentive to protect shareholders‟ wealth in their role on the board of directors in order to protect the value of their reputational capital. Peasnell et al. (2006) and Ebrahim (2007) find that companies with a higher proportion of independent directors on the board tend to have a lower magnitude of abnormal accruals. Liu and Lu (2002) find that the earnings management endeavours of managers in China are constrained to a certain extent if the firms are dominated by outside directors and where firms have their shares traded by foreign investors. Hence, previous empirical findings seem to suggest that boards which are structured to be more independent of the management are effective in monitoring the corporate financial accounting process, thus leading us to the following null hypothesis: H05: There is no relationship between board independence and earnings quality. 3.6. audit quality External auditors play an important role in monitoring managerial behaviour and ensuring that information being provided to the shareholders by management provides a “true and fair” view of the company‟s affairs. Fan and Wong (2005) document that companies that have significant agency problems are more likely to employ a Big 5 auditor (as the study was conducted prior to Arthur Andersons collapse as an auditing firm).
International brand-name auditors („Big auditor‟) are commonly believed to more independent and provide better quality audits compared to other auditors (Kabir et al. 2008).
Becker et al. (1998) and Chen et al. (2006) report that Big audit firms allow less incomeincreasing discretionary accruals choices compared to other auditors. Caramanis and Lennox (2008), and Lin et al. (2006) test the greater audit effort required to reduce the extent to which managers report aggressively high earnings. They suggest that low audit effort increases the extent to which managers are able to report aggressively high earnings. Chia et al. (2007) and Johl et al. (2007) examine the effect of the choice of auditors in constraining earnings management within a rule-based reporting framework during the Asian financial crisis and suggest that service-oriented companies engaged in income decreasing earnings management during the crisis period. More importantly, the results indicate that only the Big audit firms are able to significantly constrain the earnings management of managers of such companies. The null hypothesis is: H06: There is no relationship between audit quality and earnings quality. 3.7. Family Control Family companies produce better information than other companies and prevent managerial opportunism (Shleifer and Vishny 1997).
Recent studies such as Jiraporn and DaDalt (2007), state that family companies are less likely to manage earnings than other companies. The corporate sector in Bangladesh is mainly controlled and owned by groups of families, founder families, or government or foreign owners. According to the Companies Act 1994, a sponsor can retain a maximum of 50% of the total issued capital of a firm making an initial public offering. In Bangladesh, the five largest shareholders hold more than 50% of listed firms share (Habib 2005).
In most cases they are families, major individual owners and domestic or foreign firms. The relevant null hypothesis is: H07: There is no relationship between family control of companies and earnings quality
4
4. RESEARCH DESIGN This study uses financial statement data extracted from the OSIRIS database and also hand collected data from annual reports for those variables that are not included in the OSIRIS database. The initial sample covered all Bangladesh companies listed on the Dhaka Stock Exchange for the period 20012006, but it was subsequently trimmed down to 648 companies given the data requirements in computing discretionary accruals (DACCR).
This study utilises the cross-sectional discretionary accruals model (Jones 1991) in order to derive a measure of discretionary accruals to proxy for earnings‟ quality. The variables and their measurements used in this study are summarized in Table 2. To calculate DACCR, the following OLS regression was estimated for all companies in the sample: TA/At-1 = α0 + α1 (1/At-1) + α2 (ΔREVt /At-1) + α3 (PPEt /At-1) + ε ……………………………….……. (1) and then DACCRt = TA/At-1 – [ά0 + ά1 (1/At-1) + ά2 (ΔREVt /At-1) + ά3 (PPEt /At-1)]
where At-1 DACCRt TAt ΔREVt PPEt = beginning of year total assets. = discretionary accruals of firm for the period t, scaled by beginning year of total assets, = total accruals of firm for the period of t, scaled by beginning year of total assets, = the change in revenue of firm for the period t-1 to t, scaled by beginning year of total assets, = gross property, plant, and equipment of firm for the period t, scaled by beginning year of total assets.
The absolute value of discretionary accruals was then used to proxy for earnings quality then as the dependent variable in the OLS regression to test for association between governance (dependent) variables and earnings quality. ABSDACCR = β0 + β1Size + β2BM + β3ROE + β4CAP + β5LEV + β6MGT + β7INS + β8GOVT + β9BIND + β10 AUDQUA + β11BSIZE + β12FC…………..…………….…………….. (2) SIZE and LEV are included as control variables as Klein (2002) documents that discretionary accruals are negatively associated with company SIZE and positively associated with LEV. The BM ratio is included as a control variable in order to take into account company growth (Lee et al. 2007).
Similarly, Young (1999) reports a negative association between capital intensity and the level of discretionary accruals. We proxy for capital intensity related incentives with the ratio of non-current (fixed) assets to the total assets, labeled CAP. Finally ROE and CFO are included as a control variable in order to control for the performance of the company (Lee et al. 2007).
Table 2: Summary of the operationalisation
Variables Dependent variable Absolute Discretionary accruals (ABSDACCR) Explanatory variables Management shareholdings (MGT) Institutional shareholdings(INST) Government shareholdings(GOVT) Board Size(BSIZE) Board independence(BIND) Auditor quality(AUDQUA) Family company dummy(FC) Measurement Obtained using Jones (1991) model
Number of shares held by management/ total no of ordinary shares Number of shares held by financial institution / total no of ordinary shares Number of shares held by government / total no of ordinary shares No of directors on the board No of non executive directors in the board/ total no of directors in the board Indicator variable with the value of 1 if the firm audited by one of the BIG 4 and otherwise 0 Indicator variable with the value of 1 if the firm is controlled by founding family and otherwise 0.
Control variables SIZE Leverage (LEV) Book to market ratio (BM) Return on equity (ROE) Capital intensity (CAP)
Log of firm total assts Total long-term debt/Total assets Book value of equity/Market value of equity Net income/Total equity Non-current (fixed) assets/ Total assets
5
5. FINDINGS 5.1. DESCRIPTIVE STATISTICS Table 3: Descriptive statistics of ABSDACCR the control variables, and corporate governance.
Variables ABSDACCR SIZE LEV BM ROE CAP MGT INST GOVT BSIZE BIND AUDQUA FC Mean 0.0391 2.5776 0.5782 1.1573 0.1767 0.5109 0.5032 0.1853 0.0491 8.20 0.7049 37.21 0.71 Median 0.0241 2.5580 0.5833 0.7873 0.1089 0.5471 0.5000 0.1736 0.0000 6.00 0.8000 0.0000 1.00 S.D. 0.0511 0.67690 0.62355 2.97216 0.88165 0.25321 0.15830 0.11710 0.16872 6.164 0.25444 0.48359 0.454 Min. 0.0101 0.88 0.01 -14.61 -5.55 0.00 0.00 0.00 0.00 3 0.00 0.00 0 Max. 0.0491 4.70 5.36 15.65 6.69 1.00 1.00 0.58 1.00 37 0.97 1.00 1
Note: ABSDACCR is the absolute value of discretionary accruals for firm i in year t. SIZE is the natural logarithm of total assets in thousand (Tk) for firm i in year t. LEV is the total long-term debt to total assets for firm i in year t. BM is the book value of equity to market value of equity for firm i in year t. ROE is the net income to total equity firm i in year t. CAP is the non-current (fixed) assets to total assets for firm i in year t. MGT is the number of shares held by management to total no of ordinary shares for firm i in year t. INST is the number of shares held by financial institution to total no of ordinary shares for firm for firm i in year t. GOVT is the number of shares held by government to total no of ordinary shares for firm i in year t. BSIZE is the no. of directors on the board for firm i in year t. BIND is the No of non executive directors in the board to total no of directors in the board for firm i in year t. AUDQUA takes the value of 1 if the firm audited by one of the BIG 4 and otherwise 0, for firm i in year t. FC takes the value of 1 if the firm control by founding family and otherwise 0, for firm i in year t.
The mean (median) measure of absolute value of discretionary accruals in this study is 0.0391(0.0241).
The mean (median) value of management shareholding is 50.32 (50.00) percent. The mean (median) value of institutional shareholding is 18.53 (17.36) percent. The institutional ownership value in this study is comparable to that reported in previous studies such as Farooque et al. (2007).
The value is also consistent with Peasnell et al. (2005) (UK listed companies) of 21.60 percent. The mean (median) proportion of outside directors of 70.49 (80.00) percent is comparable to that reported in Lee et al. (2007) of 68 percent but is very different from that reported in Peasnell et al. (2005) of 43 percent. Of the sample firms, 37.21 percent employ a Big audit firm. This result is consistent with Karim and van Zijl (2008) who report 33 percent for their sample of Bangladesh firms; Habib (2007), and Habib and Islam (2007) of 30%. Companies size shows considerable variation with a mean of 2.58 and standard deviation of 0.6769. The sample companies have a mean (median) debt ratio of 57.82 (58.33) percent, mean (median) book to market ratio 1.1573 (0.7873), mean (median) return on equity 17.67(10.89) percent and mean (median) capital intensity ratio is 0.5109 (0.5471).
These findings are consistent with the study by Farooque et al. (2007).
Table 4 reports the pair wise correlation coefficients. Tests (not reported) show that, there is no predictor variable that produces a variance inflation factor greater than 5, confirming that multicolinearity is not a problem for estimation of regression equation (2).
6
Table 4: Correlation analysis of variables
ABSDACCR ABSDACCR SIZE LEV BM ROE CAP MGT INST GOVT BSIZE BIND AUDQUA FC 1 -.075 (.014) .315 (
