The influence of audit committee with financial expertise on the earnings management in UK companies

(1)

MSc Accountancy & Control, variant Control

Faculty of Economics and Business, University of Amsterdam

Master’s Thesis:

The influence of Audit Committee with Financial

Expertise on the earnings management in UK companies

Final version

Name: Shih-Min, Liu Student Number: 10599460

Program: MsC in Accountancy and Control First Supervisor: Dr. Bo Qin

Second Supervisor: Dr. Mario Schabus Submission Date: 24th of June, 2014

(2)

The influence of Audit Committee with Financial Expertise on the

earnings management in UK companies

Abstract

As the Audit Committee members with financial expertise are expected to strengthen the internal control mechanism and to improve the reliability of the financial statements provided by the firm, this paper intends to explore if the regulation that requires each audit committee has at least one member with financial expertise has an impact on the financial reporting quality of the UK public listed firms. I examine the relation between audit committee member with financial expertise and earnings management through Modified Jones Model (1991) and then ran the regression model to test if my hypotheses are in line with the empirical results. With a British sample of 1,461 annual reports from 2002 to 2009, the results indicate that firms with more accounting-literate professionals in the Audit Committee are less likely to have manipulation in their earnings. The findings indicate that the regulation should further specify the term, financial experience, as to ensure the quality of companies’ earnings.

Keywords: Audit Committee, Financial Expertise, Earnings Management, Motified Jones Model

(3)

1. Introduction

1.1 Background

The Board of Directors has very defined roles and responsibilities within the business

organization and played a very important role in the field of corporate governance. According to the Public Oversight Board (1993), it defines corporate governance as “those oversight activities undertaken by the Board of Directors and Audit Committee to ensure the integrity of the financial reporting process”. The Board of Directors has played an important role of the governance structure of business corporations because the Board of Directors has the ability to hire, expel and compensate the top management team. Besides, it has a monitoring and control function. The board is in charge of the quality of financial reporting and the engagement of the external auditors. Therefore, with the board of directors, it ensures the rights of the shareholders. Audit Committee and Remuneration Committee are the two operating committees under the body of Board of Directors. The responsibilities of Audit Committee include overseeing the process of financial reporting, being in charge of reviewing financial statements and hiring external auditors.

Recent years, there have been many frauds around the business world. In early 2000s, there were huge accounting scandals from American enterprises, such as Enron and WorldCom. As a result, the SEC (U.S. Securities and Exchange Commission) released some rules under Section 407 of the Sarbanes-Oxley (SOX) Act (2002). It requires all the members of the Audit Committee to be independent and at least one member with

financial expertise. The reason behind this decision is under the belief that appointing financial experts to be the member of Audit Committee will improve corporate

(5)

such as Canada and Taiwan. In Europe, UK Corporate Governance Code 20031 (from here on referred to as “the Code”) was released as to change the regulation regarding the composition of the Audit Committee. Under the code, it requires “the board should satisfy itself that at least one member of the Audit Committee has recent and relevant financial experience.”2 However, there is still no specific definition of “financial experience” in the UK Corporate Governance Code. Therefore, I would like to see what type of financial experience actually enhance corporate governance mechanism of a firm.

1.2 Research question

Does Audit Committee with Financial Expertise have influence on earnings management in UK companies?

1.3 Motivation of your study

As stated in the introduction, the change of composition of Audit Committee is to improve corporate governance. There have been many discussions about the relationship between Audit Committee with financial expertise and firm performance and earning quality. One of the studies conducted by Agrawal and Chadha (2005) shows that having directors with financial expertise in the Audit Committees leads to fewer earning restatements. Moreover, ` find that there is a positive association between the Audit Committee with accounting expertise and earnings quality. However, these studies are mainly based on the companies from the United States of America. This is because they changed the regulation regarding Audit Committee composition in the year of 2002 and 2003. In the United Kingdom, UK Corporate Governance Code 2003 changed the regulation regarding the composition of

1

A set of good corporate governance principles aims at public-listed companies on the London Stock Exchange (LSE). The code is overseen by Financial Reporting Council (FRC), the UK’s and the Republic of Ireland’s independent regulator responsible for promoting high quality of corporate governance and reporting standards.

2

The regulation is stated in C.3.1 under The UK Corporate Governance Code July 2003. The most updated UK Corporate Governance Code (September 2012) has the same provision.

(6)

the Audit Committee. The revised Code was published in July 2003 and took effect for companies applying for fiscal years beginning on or after 1 November 2003.

This study contributes to the literature in three ways. To begin with, there is no specific definition of the term “financial experience” in the UK Corporate Governance Code. In my opinion, financial experience can be at least divided into two parts: accounting-related experience and financial-related experience. Although “Guidance on Audit Committee” released by Financial Reporting Council (FRC) states that “It is desirable that the committee member whom the board considers to have recent and relevant financial experience should have a professional qualification from one of the professional accountancy bodies3”, it is used only to assist company boards when implementing the sections of the UK Corporate Governance Code. Secondly, this research contributes to the little existing research that investigates the relationship between Audit Committee with financial experts and earnings quality under principle-based accounting. U.K. and U.S. currently follow two different types of regulations. UK follows International Financial Reporting Standards (IFRS), principle-based standards, as their regulation towards

financial reporting, while US still follow General Accepted Accounting Principle (GAAP), rule-based standards. Nelson et al. (2002) find that it is harder for management to have opportunistic use when principle-based regulations applied because the users cannot know the financial reporting effects of principle-based standards ex ante. Moreover, UK adopts “comply or explain” approach towards Corporate Governance Code, while US requires companies have to comply with the SEC rules in almost every cases. Although

SOX-related regulations use the “comply or explain” method in some cases, for example, in relation to whether a company has “Code of Conducts” or its Audit Committee has a

3

Please refer to the section2.16 of “Guidance on Audit Committee” published by Financial Reporting Council (FRC).

(7)

“financial expert”, in most other situations, US Corporate Governance Code relies on the legislations and serious penalties when companies violate SOX requirements. Last, the ownership structure of UK companies is very different from US companies. In the UK companies, individual share ownership is more prevalent than its US counterpart though the individual ownership fell from over 50% of the market in the 1960s to less than one fifth today. When the individual power is concentrated, there might have some

temptations to extract private benefits and even influence the quality of financial reports.

My multivariate test finds a significant and negative relationship between

accounting-literate professional in the Audit Committee and earnings management, but no significant result between the general term of financial expertise (Type B Expert) and earnings management. These study results means that only with accounting professionals in the committee can successfully constrains earnings management, indicating that this regulation helps to enhance the corporate governance mechanism of a firm and that the current regulation under UK Corporate Governance Code is too broad. Furthermore, any future refinements of the current regulations are fully supported by this study. Last, this study also contributes to the current literature as there is still not much discussion in about this regulation change in the UK domain.

The rest of the paper proceeds as follows: First, I review prior literature and develop my hypotheses in more detail. Next, I describe the sample selection procedure and the empirical model. Furthermore, in presenting my main results, I also consider the additional analysis as to support the main findings. Finally, I provide some concluding remarks.

(8)

2. Literature review and hypotheses

Initial works in this type of research are largely from agency theory and view Audit Committee as a potentially helpful tool to reduce agency problem between the shareholders and managers. For the first part of this chapter, I will explain the theory behind this research. Furthermore, I will stress the governance role of Audit Committee under current trend. Later, I will address the characteristics of Audit Committee in relation to earnings quality. Finally, I will propose the hypotheses based on the prior literature mentioned earlier in the chapter.

2.1 Agency Theory

Many research works that have been done in the field of corporate governance were mainly developed through the agency theory (Puat and Devi, 2010; Guner et al, 2008). Agency theory explains the relationship between owners (principals/shareholders) and management team (agents) as a conflict of interests, which happens due to the fact that the people working in the management team are not owners. This theory also assumes that the agents want to pursue self-interest objectives in a way that may conflict with the goals of the shareholders. Moreover, it assumes that the board of directors from outside of the company act independently and therefore monitor effectively for shareholders’ interests. Corporate governance regulator and scholars have maintained a perspective that having members with financial expertise in Audit Committee helps the team to critically analyze accounting policies and financial statements, monitor the implementation process of accounting practices and identify accounting issues as well (Sultana et al,2013).

(9)

2.2 The governance role of Audit Committee

The widespread adoption of Audit Committee as one of the tools for governance reform started in the early 1990. In the earliest days, Audit Committee was regarded as an

important mechanism to improve the statutory audit process. Therefore, the first stream of research aimed at looking at the relationship between Audit Committees and audit quality from different perspectives. For example, the interaction between Audit Committees and external auditor behavior (Chan et al., 2012) and the role of Audit Committee when choosing external auditor for the company (Chen et al., 2005).Moreover, the formation of Audit Committee also helps the internal control process. Cohen et al. (2004) believe that governance capability can be improved by establishing a close and solid relationship between Audit Committee and the internal audit process.

In addition to ensuring audit quality and internal control, Audit Committee also plays an important role when talking about market reaction. Wild (1996) find that the formation of Audit Committee drew a higher market reaction toward earning reports in the stock market through analyzing 260 companies from the US. Furthermore, Davidson III et al (2004) find that appointing Audit Committee member with financial expertise can bring positive stock price reaction from the investors in the United States. It is believed that a firm with lower earnings management may prefer to have an audit, while a firm with higher discretionary accruals may avoid establish an audit committee on a voluntary basis. These findings suggest that investors believe that Audit Committee has played an

important governance role in improving managerial accountability to shareholders and increasing the quality of reported earnings as well.

(10)

2.3 Audit Committee characteristics and earnings management

There are many characteristics need to be noticed for the effective operation of the Audit Committee and the expertise of Audit Committee members is one of them. Several prior literatures have had significant findings regarding the association between the expertise of Audit Committee and its earnings quality after SEC proposed the idea that they would like to require all the public-listed companies to have at least one member with financial expertise. Research conducted by Xie et al (2003) have find that there’s a negative association between the percentage of Audit Committee members with corporate or investment banking backgrounds and the level of earnings management. Moreover, Dhaliwal et al. (2007) have find that there is a positive association between accounting expertise in Audit Committees and accruals quality under US companies from the year of 1995 to 1998. Furthermore, one study finds out that having an Audit Committee with accounting expertise may be essential for preventing distortions in financial reporting by conducting a research aiming at S&P 500 firms (Visvanathan et al., 2008). However, some studies have found different points of views, one study conducted by Van der Zahn and Tower (2004) failed to find a relationship between the extents of earnings management and the Audit Committee’s financial expertise.

2.4 The composition change of Audit Committee

There have been many financial frauds around the world. This has lowered the shareholder confidence in the stock market and companies. Therefore, the SEC (U.S. Securities and Exchange Commission) located in the United States released some rules under Section 407 of the Sarbanes-Oxley (SOX) Act (2002). The new regulation is to require all the members of the Audit Committee to be independent and at least one

member with financial expertise. The reason behind this motivation is under the belief that appointing financial experts to the Audit Committee helps to improve the corporate

(11)

governance. In UK, the government released the new regulation under UK Corporate Governance Code 2003 and required companies listed on London Stock Exchange (LSE) to have at least one member with financial expertise in the Audit Committee.

2.5 Hypotheses development

One research finds that Audit Committee with accounting expertise is associated with higher earning quality (Baxter and Cotter, 2009). Moreover, there is a strong association between earnings quality and earnings management. Therefore, I would like to propose hypotheses discussing the relationship between Audit Committee members with

accounting expertise and earnings management. Furthermore, McMullen and

Raghunandan (1996) examines the effectiveness of Audit Committee by distributing surveys to 51 companies in the United States and find that companies with some financial reporting problems were much less likely to have at least one Audit Committee member with CPA certification. Moreover, Agrawal and Chadha (2005) find that having directors with a professional certification, like Certified Public Accountant (CPA) and Chartered Financial Analyst (CFA) leads to lower chances of earnings restatements. Therefore, we can see that these findings suggest that at some certain points, both the expert with financial and accounting background helps companies to strengthen internal control systems and improve corporate governance mechanism, and finally contribute to the quality of financial report. To sum up, I expect a positive relation between Audit Committee with accounting expertise and earnings quality.

As the UK Corporate Governance Code does not specifically define the meaning of financial experience, the first step of this study is to examine the effect of different types of financial experience on earnings quality. It was widely discussed in public about how to define the term “Financial Expertise” in the United States while SEC was proposing the

(12)

new regulation regarding the term, financial expertise. Audit Committees are expected to assess the policies that assess the firm’s financial posture, evaluate judgmental accounting areas and the quality of financial reports, so some people hold the perspective that

financial expertise should be defined as accounting-related experts. However, some people believe that narrowly defining financial expertise as accounting-related expertise was unnecessarily restrictive and drastically limits the pool of qualified directors, for example, the American Association of Bank Directors indicated that even the Alan Greenspan, the former chairman of Federal Reserve, disqualified as the financial expert. There are some studies discussing about the term “financial expertise”. Defond et al (2005) break the term into two subcategories that constitute the broader group: accounting financial experts and non-accounting financial experts, which define as company presidents and CEO. In addition, Qin (2007) also divided the definition of financial expert into two types: Type I expert refers to people who have closely worked with accounting information; Type II expert refers to people who have CEO experience or have worked closely with the financial markets. Therefore, based on prior literature, I would like to break the term, financial experience, into two parts: Type A Expert refers to accounting-related expert while Type B Expert refers to Non-Accounting-related Expert. The detailed descriptions and examples of each category are listed as follows:

(1)Type A Expert: Accounting-related Expert refers to people who have closely worked with accounting information, such as public accountant, auditor, principal/chief financial officer, controller, or principal/chief accounting officer.

(2) Type B: Non-Accounting-related Expert refers to people with more eligible positions and also working closely in the financial markets, such as presidents/CEOs, professional financial analysts, investment bankers and venture capitalists.

(13)

As I would like to see if how the correlation between the different types of experts and earnings management, the hypotheses for this research are listed below:

H1: Firms with Type A Expert are negatively associated with earnings management.

H2: Firms with Type B Expert are negatively associated with earnings management.

Summary of hypotheses

Independent variables Hypotheses

Accounting-related Expert (A) H1 Firms with Audit Committee members with experts (Type A and B) are negatively associated with earnings management. Non-Accounting related Expert

(B)

H2

3. Research methodology

Agency theory is the main theoretical theory for this study. This study has decided to use quantitative research with a focus on answering research question mentioned earlier. For this section, I will explain how I choose the sample and retrieve the data. Later, I will explain the model that will be used for this study.

3.1 Sample Selection and Data collection

Since the new regulation applies to the public-listed companies in the UK, I randomly select companies from the London Stock Exchange (LSE). The data for this research is obtained through two ways: directly download from COMPUSTAT database and collect the background information of each audit committee member from annual reports by hand in order to determine the type of financial expert. COMPUSTAT is a database of financial, statistical and market information on global companies throughout the world. Furthermore,

(14)

as the accessible databases do not provide audit information of the UK public companies, I also manually collect the audit information of the selected companies through annual reports. Therefore, annual reports are regarded as the primary source for this research. The study period is from 2002 to 2009 as to understand how the new regulation influences the earnings quality in the United Kingdom with a comparison between before the change and after the change. From 1,840 observations, I eliminate those without complete background information of audit committee members. Furthermore, I also eliminate observations that are with invalid information when applying values for the variables. The final sample consists of 1,461 observations representing 238 companies across many different industries.

3.2 Research Design

The objective of the hypotheses testing is to investigate the relationship between earning management (EM) and audit committee expertise, which is represented by the coefficient of DFEXP_A and DFEXP_B. EM is the measurement of earning management magnitude.

DFEXP_A and DFEXP_B are dummy variable; DFEXP_A (DFEXP_B) equals 1 if the

Audit Committee has at least one Type A (Type B) expert, otherwise 0. Additional analysis using relative value, the ratio between the number of FEXP_A/ FEXP_B and the size of the audit committee, is provided in Additional Analysis section in this paper.

The following models are used to test the study’s hypotheses:

Hypothesis 1 testing:

, ( )

Hypothesis 2 testing:

(15)

where

EM Earnings Management. I use discretionary accruals (DACC) obtained through Modified Jones Model (1991) as the level of earnings management. FEXPA Type A Expert, equals 1 if the Audit Committee has at least one Type A

expert, otherwise 0

FEXPB Type B Expert, equals 1 if the Audit Committee has at least one Type B expert, otherwise 0

SIZEF Firm Size, measured by natural log of Net Total Asset

PERMF Firm Performance, calculated by Net Income divided by Total Asset (ROA) BIG4 Big 4 auditors, receives “1” if the company is audited by a Big 4 accounting

firm and “0” otherwise

LEVG Leverage, defined by Total Debt/Net Total Asset

INDUSTRY Industry dummies identified on the basis of Global Industry Classification Group (GIC Group) classification

Some control variables are included in this model, they are: firm size, firm performance, audit firm, leverage and industry. According to prior literature, firm performance and growth are important factors affecting the amount of managed earnings (Lee et al., 2006). Furthermore, large companies may have less incentive to engage in earnings management as they get more scrutiny from the public, including financial analysts and investors (Chen

et al. 2010). In addition, one study shows that the measurement of discretionary accruals

can be problematic for firms with extreme financial performance (Dechow et al., 1995; Kothari et al., 2005). Therefore, firm performance and firm size can be one of the control variables for this study. I use log of net total assets as the market value of the firm to

(16)

define the firm size control variable and use net income divided by total asset (Return on Asset; ROA) to define firm performance control variable. In addition, Becker et al. (1998) conclude that companies with auditors from non-Big 5 accounting firms were more inclined to report discretionary accruals that are, on average, higher than the discretionary accruals of companies audited by Big 5 auditors. Francis et al. (1999) holds the similar perspective, they argue that firms with Big 5 auditors report lower discretionary accruals. Given the evidences from Becker et al. (1998) and Francis et al. (1999), I control for the effect of external audit quality by including a dummy variable (BIG 4), receiving “1” if the company is audited by a Big 4 accounting firm and “0” otherwise. Since the Big Five became the Big Four after the Enron Scandal in late 2001, I use Big Four accounting firms, including Deloitte, Ernst & Young (EY), KPMG and PriceWaterHouseCoopers (PwC), in this research. For the last two control variables, leverage and industry, they are also supported by the prior literature. Beatty and Weber (2003) indicate that leveraged firms were more inclined to engage in earnings management in order to avoid debt covenant default. I use natural log of total debt divided by net total asset to define the level of leverage of each firm. Last, I choose industry as one of the control variables based on the prior literature (Defond and Jiamobalvo 1994; Kuang et al. 2014).

Earning Management Measurement (EM)

Consistent with prior studies, the current study uses the magnitude of discretionary accruals (DACC) as a proxy to examine the extent of earnings management (e.g. Chung and Kallapur, 2003; Klein, 2002a). The reason why discretionary accrual (DACC) as a testable measure of earnings management is that many management teams may use accrual accounting to distort the true financial performance of the firms (Cohen et al., 2007). Furthermore, Dechow et al. (1995) provide evidence that the Modified Jones Model is the most powerful way to detect earnings management among the alternative

(17)

model to measure discretionary accruals.

Discretionary accruals (DACC) are estimated by subtracting the predicted Non-discretionary accruals (NDACC) from total accruals (TACC) (Dechow et al, 1995). The Non-discretionary part of total accruals (NDACC) is predicted by using “The Modified Jones Model (1991):

, , , ∆ , ∆ , , , , , ( )

As to get the discretionary accruals (DACC), I take the following steps: first, I make a regression analysis to the whole samples in order to obtain the value of α0, α1 and α2. Subsequently, those values are put into Non-discretionary part of total accruals (NDACC) to predictive model. Finally, firm-year discretionary part of accounting accruals (DACC) is computed as follows: , , , , ∆ , ∆ , , , , ( ) Variable Definition

TACC_, For each firm i, the total accounting accruals at the end of year t TA_, Total assets at the end of year t-1

∆REV_, The change in sales revenue between year t and year t-1 ∆REC, The change in accounts receivable between year t and year t-1 PPE, Property, Plant and equipment at the end of year t

(18)

4. Research Results

4.1 Data description

TABLE 1 Observation by Industry (n=1,461) (winsorized)

Sector GIC GROUP Percentage Number of Observations

Energy 1010 8.145% 119 Materials 1510 7.871% 115 Industrials 2010-2030 32.170% 470 Consumer Discretionary 2510-2550 24.572% 359 Consumer Staples 3010-3030 8.761% 128 Health Care 3510-3520 4.312% 63 Financials 4030-4040 0.958% 14 Information Technology 4510-4530 8.693% 127 Telecommunication Services 5010 1.506% 22 Utilities 5510 3.012% 44 TOTAL 100% 1,461 Note:

Industrial sector includes Capital Goods Industry, Commercial &Professional Service Industry and Transportation Industry; Consumer Discretionary sector includes Automobiles&

Components Industry, Consumer Durables& Apparel Industry, Hotel Restaurants &Leisure Industry, Media Industry and Retailing Industry; Consumer Staples sector includes Food & Staples Retailing Industry, Food, Beverage & Tobacco Industry and Household & Personal Products Industry.

Industry distribution for the selected sample is shown in TABLE 1. The selected samples are mainly from two different GIC Group sectors: Industrial sector and Consumer Discretionary sector. They represent 54% of the total selected samples of this study.

(19)

TABLE 2 Descriptive statistics for Dependent and Independent Variables (winsorized)

TABLE 2 presents the winsorized descriptive statistics about total accruals, discretionary accruals, firm size, firm performance and leverage for the sample of COMPUSTAT Global firm-years covering 2002-2009 meeting the data requirements of this study. I winsorized the extreme observations by setting the value in the bottom and top one percent to the values of 1st and 99th percentiles. Under Modified Jones Model (1991), the average total Variables Total Observations Mean Std. Deviation Minimum (1st) Q1 Median Q3 Maximum (99th) TACC 1,461 −0.0474 0.1005 −0.8438 −0.0831 −0.0461 −0.0109 0.6161 NDACC 1,461 −0.0346 0.0389 −0.3278 −0.0497 −0.0301 −0.0154 0.2854 DACC 1,461 −0.0128 0.1158 −0.8059 −0.0514 −0.0112 0.0242 0.6249 SIZEF 1,461 6.0417 1.7045 −0.1043 4.9381 5.8811 7.0312 11.8013 PERMF 1,461 0.0643 0.0964 −0.7667 0.03048 0.0591 0.0980 0.7899 LEVG 1,461 −1.1192 1.7059 −9.4928 −1.5426 −0.8209 −0.2115 2.9154 FEXPA 1,461 1.0589 0.6749 0 1 1 1 4 FEXPB 1,461 2.3340 1.0106 0 2 2 3 7 FEACA 1,461 0.2979 0.14 0 0.20000 0.3333 0.3333 1.0000 FEACB 1,461 0.6414 0.19 0 0.5000 0.6667 0.7500 1.0000 Additional Information SIZE AC 1,461 3.6632 0.9560 2 3 3 4 8 Notes:

TACC = Total Accruals; NDACC = Non-discretionary accruals; DACC= Discretionary Accruals; SIZEF = Firm Size, control variable, measured by natural log of Net Total Asset; PERMF = Firm Performance, control variable, calculated by Net Income divided by Total Asset (ROA); LEVG = Leverage, control variable, defined by Total Debt/Net Total Asset; FEXPA = Type A Expert, independent variable; FEXPB= Type B Expert, independent variable; SIZE AC = Size of Audit Committee; FEXPA = Proportion of Type A Expert in the Audit Committee; FEXPB = Proportion of Type B Expert in the Audit Committee; Q1 = the first quartile; Q3 = the third quartile

(20)

accruals (TACC) is (−0.0474), similar to the total accruals reported in Ecker et al (2012). Furthermore, the average discretionary accrual (DACC) of the sample companies is around (−0.0128) and the median discretionary accrual is (−0.0112).The average

discretionary accruals is a bit lower than those reported by Puat and Devi (2010). This is most likely due to the difference in the firm sizes.

The mean number of Type A experts among the selected samples is 1.0589, but for the Type B experts is 2.3340. Over the study period, the amount of Type A and Type B accounted for the total amount of Audit Committee is high. The average size of Audit Committee in the sample firms is 3.6632. The Blue Ribbon Committee (1999) in the US recommended that firms maintain Audit Committees consisting of at least three members who were all fully independent. In addition, in section C.3.1 of UK Corporate Governance Code, it suggests that “The board should establish an audit committee of at least three, or in the case of smaller companies two, independent non-executive directors”. Based on my study results, it shows that most of the audit committees are consists of 3 to 4 members. The lower quartile audit committee size of 3 and upper quartile audit committee size of 4 indicate that most of the sample firms maintained audit committee consisting of at least three members. This is around the size recommended by The Blue Ribbon Committee and regulation by UK Corporate Governance Code.

TABLE 2 also reports descriptive statistics for the control variables for the sample firms. The mean of firm performance, net income divided by total assets, is 0.0643, and only a bit larger than its median (0.0591), indicating the absence of strong skewness in the yearly performance of the firms. The mean (6.0417) of firm size is also a bit larger than its median (5.8811).

(21)

TABLE 3 Type A and Type B Expert information by year (winsorized)

This table shows the number of observations and the ratio of experts in the Audit Committee by different subgroups: Type A and Type B Expert

Year

Type A Expert Type B Expert

Total

Number of Observations Number of Observations

Percentage of Total Observations Percentage in the Audit Committee Percentage of Total Observations Percentage in the Audit Committee 2002 95 146 153 62.09% 19.50% 95.42% 53.85% 10.47% 2003 107 148 151 70.86% 23.67% 98.01% 57.96% 10.34% 2004 128 164 166 77.11% 26.77% 98.80% 61.40% 11.36% 2005 153 189 190 80.53% 29.13% 99.47% 65.37% 13.00% 2006 163 184 186 87.63% 32.61% 98.92% 67.75% 12.73% 2007 177 198 201 88.06% 33.68% 98.51% 66.97% 13.76% 2008 194 212 215 90.23% 34.05% 98.60% 67.23% 14.72% 2009 183 196 199 91.96% 34.30% 98.49% 68.31% 13.62% Total 1,200 1,437 1,461 82.14% 98.29% 100%

Note:Number of observations indicates the numbers the samples classified as Type A or Type B expert

each year; Percentage of Total Observations = Type A or Type B Expert per year /the total number of observations; Percentage in the Audit Committee = Type A or Type B Expert/ The total number of members in the Audit Committee

(22)

TABLE 3 exhibits the exact number of Type A and Type B financial expert by each year. Apparently, most of the sample firms already employed either Type A or Type B financial expert in their audit committees before the new regulation was published in the year of 2003. From the table, it is obvious that the growth rate of Type A experts is higher than Type B expert. While I was manually collecting the background information of audit committee members, I found that in the first two years of the study period, companies had a few people with accounting-related expertise, such as Certified Public Accountant or Chief Financial Officers, but after the regulation took effect in late 2003, the number of members with accounting-related background had increased. Furthermore, some

companies did not disclose the audit committee member background in the annual reports in the year of 2003 and 2004, so the observation samples of these two years are lower than the rest of the study years.

TABLE 4 Correlation Matrix for all variables in the sample

DACC FEXPA FEXPB SIZEF PERMF BIG4 LEVG

DACC 1.0000 FEXPA −0.0558 1.0000 FEXPB −0.0295* 0.3488 1.0000 SIZEF 0.0347* 0.0268* 0.2468 1.0000 PERMF 0.0803 −0.0097** −0.0144* −0.0700 1.0000 BIG4 −0.0038** 0.0338* 0.0714 0.1189 −0.0236* 1.0000 LEVG −0.0182* 0.1006 0.0619 0.1211 −0.1894 0.0918 1.0000 Notes:

* and ** denote significance at the 0.05 and 0.01 levels, respectively. DACC= Discretionary Accruals; FEXPA = Type A Expert, independent variable; FEXPB= Type B Expert, independent variable; SIZEF = Firm Size, control variable, measured by natural log of Net Total Asset; PERMF = Firm Performance, control variable, calculated by Net Income divided by Total Asset (ROA); BIG4 = Big 4 auditors, receiving “1” if the company is audited by a Big 4 accounting firm and “0” ; LEVG = Leverage, control variable, defined by Total Debt/Net Total Asset.

(23)

The correlation matrix for the variables is reported in TABLE 4. The table explains the interrelationship among all the variables included in the study. Correlations are important in this type of study as they not only highlight the explanatory variables but also identify the significant correlations among the independent variables used in the research. As shown in the table, the variations in FEXPA and FEXPB are negatively correlated with the variations in DACC. The negative relation conforms to the statement in Hypothesis 1 that firms with Type A experts are negatively associated with earnings management and also conforms to the statement in Hypothesis 2 that firms with Type B experts are negatively associated with earnings management.

The positive relationship between DACC and firm size as well as firm performance proves the assumption that the firms are more involve in discretionary accruals when the companies have better performances and bigger company sizes. It is worth noting that ROA, the indicator for firm performance, shows the most significant correlation (0.0803) with discretionary accruals (DACC) among all the control variables. This result enhances the argument that it is important to consider firm performance when measuring earnings management. Furthermore, according to the correlation results (−0.0182) in TABLE 4, higher leveraged firms will employ less earnings management, which is consistent with a few prior studies, such as Jelinek (2007). This, at some certain point, implies that high debt ratio constrains irrational behavior.

In TABLE 4, there is a positive correlation between firm size and BIG 4 auditors (0.1189). This result is consistent with those from the research conducted by Zhou J., Elder R. (2001), suggesting that larger firms may have higher possibility of hiring Big 4 auditors for their financial reports. In addition, it is worth noting that there is a negative relation between discretionary accrual (DACC) and the control variable called BIG4. The actual

(24)

coefficient is around (−0.0038), indicating that discretionary accruals are negatively related to the use of Big 4 auditor, suggesting that Big 4 clients engage in less earnings management. Furthermore, this result is consistent with the result of the prior study, which stated that Big 4 auditors were associated with lower level of discretionary accruals (Becker et al. 1998; Francis et al. 1999). Last, only 53 out of total observation numbers (1,461) hire non-Big 4 auditors during the study period, pointing out that most of the public listed firms in London Stock Exchange hiring Big 4 auditors for their financial statement audit.

4.2 Multiple regression test

Before I give the results coefficients, I would like to report the model fit. The result of F-Statistics shown in TABLE 5 is 2.9128 with a p-value less than 0.01 and 2.6099 also with a p-value less than 0.01, indicating that there is a significant relation between the independent variables and dependent variables. Moreover, the number of R-squared shows how well a statistic mode fits a set of observations. From TABLE 5, it is around 0.03 for Type A and Type B Expert. The results are similar to the one from Baxter and Cotter (2009).

(25)

TABLE 5 Regression Analysis with discretionary accruals

Using the magnitude of discretionary accruals (DACC) from Modified Jones Model (1991) as a proxy to examine the relationship between variables and earnings management for London Stock Exchange (LSE) listed companies between 2002 and 2009.

Independent Variables

Dependent variable Discretionary Accruals (DACC)

Type A Expert Type B Expert

Coefficients P-Value Coefficients P-Value

EXPA/ EXPB −0.0169* 0.0351 0.0133 0.5770

SIZEF 0.0014 0.4643 0.0012 0.5397

PERMF 0.0753* 0.0200 0.0734* 0.0236

BIG4 −0.0030 0.8519 −0.0032 0.8448

LEVG −0.0015 0.4070 −0.0019 0.3133

INDUSTRY YES YES

Number of

observations 1,461 1,461

R-squared 0.0274 0.0246

F-statistics 2.9128** 2.6099**

Notes:

* and ** denote significance at the 0.05 and 0.01 levels, respectively. FEXPA = Type A Expert, equals 1 if the Audit Committee has at least one Type A expert, otherwise 0; FEXPB = Type B Expert, equals 1 if the Audit Committee has at least one Type B expert, otherwise 0; SIZEF = Firm Size, control variable, measured by natural log of Net Total Asset; PERMF = Firm Performance, control variable, calculated by Net Income divided by Total Asset (ROA); BIG4 = Big 4 auditors, receiving “1” if the company is audited by a Big 4 accounting firm and “0” otherwise; LEVG = Leverage, control variable, defined by Total Debt/Net Total Asset; INDUSTRY, control variable, industry dummies identified on the basis of Global Industry Classification Group (GIC Group) classification.

(26)

TABLE 5 illustrates the regression results using regression model in Equation 1 and 2. Hypothesis 1 predicts that firms with Type A Expert are less likely to engage in earnings management. As shown in the table, when the other explanations are controlled, the significantly negative coefficient between discretionary accruals (DACC) and Type A expert is (−0.0169), indicating that firms with more Type A Expert in the Audit

Committee are negatively associated with earnings management. This result is consistent with the literature (Puat and Devi, 2010).

Hypothesis 2 predicts that firms with Type B Expert are negatively associated with earnings management. As shown in TABLE 5, the coefficient between discretionary accruals (DACC) and Type B Expert is 0.0133. However, the result is not at a significant level (P=0.5570).

The results for control variables shown in the table indicate significant association

between discretionary accruals (DACC) and firm performance (PERMF). This result is in line with the one from Puat and Devi (2010). As for the rest of control variables, no significant relations are found.

Given the test results above, it is obvious that more members with accounting expertise (Type A Expert) in the Audit Committee constrain earnings management, indicating that Type A Experts are more capable of applying their professional judgment towards earnings management problems. This finding is consistent with the Qin (2006) that documents firms with accounting professionals in the Audit Committee are more likely to have high quality of reported earnings than firms without this type of experts.

(27)

4.3 Additional Analysis

Based on the Empirical results, it indicates that there is a significantly negative association between Type A Expert and discretionary accruals (DACC). Therefore, I would like to step further to see how the size of Type A Expert in the Audit Committee influence the earnings management.

TABLE 6 Regression of Independent Variables and Dependent Variables: The size of Type A Expert

Coefficients Standard Error t Stat P-value

Intercept −0.0403 0.0344 −1.1731 0.2409 FEACA −0.0315 0.0160 −1.9691 0.0491 SIZEF 0.0010 0.0020 0.5197 0.6033 PERMF 0.0778 0.0324 2.3999 0.0165 BIG 4 −0.0049 0.0163 −0.2985 0.7653 LEVG −0.0018 0.0019 −0.9601 0.3372 Notes:

FEACA = Proportion of Type A Expert in the Audit Committee; SIZEF = Firm Size, control variable, measured by natural log of Net Total Asset; PERMF = Firm Performance, control variable, calculated by Net Income divided by Total Asset (ROA); BIG4 = Big 4 auditors, receiving “1” if the company is audited by a Big 4 accounting firm and “0” otherwise; LEVG = Leverage, control variable, defined by Total Debt/Net Total Asset.

I use the same sample in my primary tests to collect the ratio of the Type A Expert in Audit Committees. TABLE 3 also illustrates the percentage of Type A Experts in Audit Committees from 2002 to 2009. The ratio was growing as the year went by. This explains that the new regulation propels the growth in the Type A Expert and more firms notice the importance of accounting-related professionals in Audit Committees.

As shown in TABLE 6, the significantly negative relation for the discretionary accrual (DACC) and the size of Type A Expert in the Audit Committee is consistent with the

(28)

result for discretionary accruals (DACC) and Type A Expert shown in TABLE 4. Both of the test results show that hiring accounting-related professionals lead to less earnings management.

5. Conclusion and Discussion

This study examines whether Audit Committee with financial expertise have an impact on earnings management by applying Modified Jones Model (1991). The selected samples are all public-listed companies in London Stock Exchange from the year of 2002 to 2009. The motivation behind this study is to define what kind of financial expertise actually help to strength the role of Audit Committee and lead to less earnings management problems. Though there are already much discussion about the definitions of term, financial

expertise, in the United States, the UK has different standards towards accounting standards and different ownership structure when compared to the US. Therefore, this study focuses on the British public-listed firms.

The initial idea of implementing at least one member with financial expertise was to increase the effectiveness of Audit Committee in monitoring the financial reporting

process, but there is still no specific definition of the term, financial experience, in the UK. For this study, our samples consist of 1,461 observations representing 238 companies across 10 different types of industries under Global Industry Classification Standard (known as GIC Groups). Through this study, a negative relation is found between Type A Expert and earnings management in the selected samples from UK after applying

Modified Jones Model (1991). However, there is no significant result for the association between Type B Expert and earnings management.

(29)

The regulators, legislators and stakeholders are all very concerned with the issue of earnings management, so it is important to have regulations that can constrain earnings management and ensure public confidence in the reporting of accounting information. One implication of the finding that accounting-literate members are negatively associated with discretionary accruals (DACC) is to recommend the appointment of more members with professional accounting background. The presence of accounting professionals in the committee helps in constraining earnings management. Another implication is that the appointment of accounting-literate professionals on the Audit Committee will enhance the confidence of the investors regarding the accounting information and further improve firms’ corporate governance mechanism. According to a research conducted by Defond et

al. (2004), it finds a positive market reaction to the appointment of accounting-related

financial expert to the Audit Committee, but only when the firms have relatively strong corporate governance.

For this study, I manually collected the background information of each Audit Committee member through annual reports and then see if those members meet the standard of Type A and Type B Expert. However, there is a limitation when classifying audit members into two different types of experts. I am dependent on companies’ public disclosure in their annual reports for the data collection and analysis. Some firms did not have the

background of the board of directors in their annual reports in the first two years of the study period. In addition, some firms did not disclose the financial or accounting-related in detail before the regulation was released. Therefore, the number of Type A and Type B expert might be underestimated in the 2002 and 2003. This paper still leaves a question for the future research. It has proven that accounting professionals in Audit Committee constrain earnings management, but is it worth it to hire accounting professionals as Audit

(30)

Committee member? This is a very interesting topic for this field of study.

References

Agrawal, A., and Chadha, S. 2005. Corporate governance and accounting scandals. Journal of

Law and Economics 48:371-406.

Becker, C., M. DeFond, J. Jiambalvo, and K. R. Subramanyam. 1998. The effect of audit quality on earnings management. Contemporary Accounting Research 15(1): 1-24. Beatty, A., and Weber, J. 2003. The effects of debt contracting on voluntary accounting

method changes. The Accounting Review 76:27-59.

Qin, B. 2005. The influence of audit committee financial expertise on earnings quality: US evidence. The IUP Journal of Accounting Research and Audit Practice IV(3):7-28. Brown, G., R. Collins, and D. Thornton. 1999. Professional Judgment and accounting

Standards. Accounting, Organizations and Society 18(4):279-289.

Burak Guner, Ulrike Malmendier, and Geoffrey Tate. 2008. Financial expertise of directors.

Journal of Financial Economics 88:323-354.

Chan, A. M. Y., Liu, G., and Sun, J. 2013. Independent audit committee members’ board tenure and audit fees. Accounting and Finance 53(4):1129-1147.

Chen, K. Y., and Liu, J. 2010. Earnings management, CEO domination and growth opportunities-Evidence from Taiwan. International Journal of Public Information

Systems 2010(1):43-69.

Chen, Y. M., Moroney, R. and Houghton, K. 2005. Audit Committee composition and the use of an industry specialist audit firm. Accounting and Finance 45:217-239.

Cohen, J., Krishnamoorthy, G. and Wright, A. 2004. The corporate governance mosaic and financial reporting quality. Journal of Accounting Literature 23:87-152.

Cohen, J., Gaynor, L.M., Krishnamoorthy, G. and Wright, A.M. 2007. Audit communication with the audit committee and the board of directors: policy recommendations and opportunities for future research. Accounting Horizon 21(2):165-187.

(31)

Dechow, P., R. Sloan, and A. Sweeney. 1995. Detecting earnings management. The

Accounting Review 70 (April): 193-225.

DeFond, M. L., and J. Jiambalvo. 1994. Debt convenant violation and manipulation of

accruals: Accounting choice in troubled companies. Journal of Accounting & Economics, 17 (102):145-176.

Defond, M., R. Hann and X. Hu. 2005. Does the market value financial expertise on audit committee of boards of directors?. Journal of Accounting Research 43(2):153-193. Dhaliwal, D., V. Naiker, and F. Navissi. 2007. Audit committee financial expertise, corporate

governance and accruals quality: An empirical analysis. Working paper. Available at SSRN: http: paper.ssrn.co/sol3/papers.cfm?abstract_id=906690. University of Arizona and Monash University.

Dimitrios V. Kousenidis, Anestis C. Ladas, and Christos I. Negakis. 2013. The effect of the European debt crisis on earnings quality. International Review of Financial Analysis 30:351-362.

Ecker F., Francis J., Olsson P., and Schipper K. 2012. Estimation sample selection for discretionary accruals models. Journal of Accounting and Economics.56(2-3):190-211. Francis, J., E. Maydew and H. Sparks. 1999. The role of Big 6 auditors in the credible

reporting of accruals. Auditing: A Journal of Practice and Theory 18 (Fall):17-34. Jelinek, K. 2007. The effect of leverage increases on earnings management. Journal of

Business and Economic Studies.13:24-46.

Jerry W. Lin, June F. Li, and Joon S. Yang. 2006. The effect of audit committee performance on earnings quality. Managerial Auditing Journal 21(9):921-933.

Krishnan, G. V. and G. Visvanathan. 2008. Does the SOX definition of an accounting expert matter? The association between audit committee directors’ accounting expertise and accounting conservatism. Contemporary Accounting Research Journal 25(3):827-858. Lee, C., Li, and L., Heng, Y. 2006. Performance, growth and earnings management. Review of

(32)

Kuang, Y., Qin, Bo and Wielhouwer, Jacco L. 2014. CEO origin and Accrual-Based Earnings Management. Accounting Horizons, forthcoming.

McMullen, D., and Raghunandan, K. 1996. Enhancing audit committee effectiveness. Journal

of Accountancy 182(2):79-81.

Nelson, M, J. Elliott, and R. Tarpley. 2002. Evidence from auditors about managers’ and auditors’ earnings management decisions. The Accounting Review 77:175-202. Nigar Sultana, and L. W. Mitchell Van der Zahn. 2013. Earnings conservatism and audit

committee financial expertise. Accounting and Finance Journal :1-32

P. Baxter, and J. Cotter. 2009. Audit committee and earnings quality. Accounting and Finance

Peer Review Journal 49(2):267-290.

S. Puat Nelson and S. Devi. 2013. Audit committee experts and earnings quality. Corporate

Governance Journal 3(4):335-351.

Wallace N. Davidson III, Biao Xie, and Weihong Xu. 2004. Market reaction to voluntary announcements of audit committee: The effect of financial expertise. Journal of

Accounting and Public Policy 23:279-293.

Wild, J. 1994. Managerial accountability to shareholders: audit committees and the explanatory power of earnings for returns. British Accounting Review 26:353-374. Xie, B., W. Davidson, and P. DaDalt. 2003. Earnings management and corporate governance:

the role of the board and audit committee. Journal of Corporate Finance 9:295-316. Zhou J., and Elder R. 2001. Audit Firm Size, Industry Specialization and Earnings

Management by Initial Public Offering Firms. Working paper. Available at SSRN: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=321041. Syracuse University and University of Hawaii at Manoa.