The Impact of the Malaysian Code on Corporate Governance: Compliance, Institutional Investors and Stock Performance

Effiezal A. Abdul Wahab,# Janice C.Y. How, Peter Verhoeven Department of Accounting and Finance,

The University of Auckland

(JCAE 2007 Forthcoming)

Abstract

In 2001, the Malaysian Code on Corporate Governance (MCCG) became an integral

part of Bursa Malaysia Listing Rules, which requires all listed firms to disclose the extent of

compliance with MCCG. Our panel analysis of 440 firms from 1999 to 2002 finds that the

corporate governance reform in Malaysia has been successful, with a significant

improvement in governance practices. The relationship between ownership by Employees

Provident Fund (EPF) and corporate governance has strengthened in periods subsequent to

the reform, in line with the lead role taken by EPF in establishing the Minority Shareholders

Watchdog Group. The implementation of MCCG has a substantial effect on shareholders’

wealth, increasing stock prices by an average of about 4.8%. Although there is no evidence

that politically connected firms perform better, political connection has a significantly

negative effect on corporate governance, which is mitigated by institutional ownership.

Key words: Corporate Governance; Institutional Investors; Stock Performance; Political

Connection JEL classifications: G21, G22, G23, G32, G34. We would like to thank Agnes Cheng, David Emanuel, Jerry Bowman, an anonymous referee, the associate editor (Bin Srinidhi) and participants at JCAE Symposium and seminar at University of Auckland for helpful comments.

# Corresponding author: Department of Accounting and Finance, The University of Auckland Private Bag 92019 Auckland, New Zealand. Ph: +64 9 373 7599. Email: e.wahab@auckland.ac.nz

1.0 Research Aims

The 1997/98 Asian financial crisis exposed a number of poor corporate

governance practices in Malaysia. Although not unique to this country,

companies suffer from over-leveraging (Fraser et al., 2006); lack of transparency, (financial)

disclosure and accountability (Mitton, 2002); poor legal protection of minority investors

against expropriation by corporate insiders (Claessens et al., 1999); and allegations of

cronyism (Johnson and Mitton, 2003). These matters have been exacerbated by a lack of

corporate takeovers in Malaysia, helped by an extensive network of politically connected

companies (Faccio et al., 2006; Mohamad et al., 2006). Furthermore, acts of shareholder

activism by local institutional shareholders have been rare (Claessens and Fan, 2002).

Although these problems are not easy to solve, the Finance Committee on Corporate

Governance (FCCG) in Malaysia recognizes that improved disclosure practices together with

increased institutional shareholder activism are at the heart of establishing good corporate

governance. In 1999, FCCG made two important recommendations. The first is the

establishment of the Malaysian Code on Corporate Governance (MCCG), which identifies a

framework for best practices in corporate governance. The second is the establishment of the

Minority Shareholders Watchdog Group (MSWG), whose main objective is “to monitor and

combat abuses by insiders against the minority” (FCCG, Chapter 6 paragraph 9.1).

In 2001, the first recommendation of MCCG became an integral part of the (revamped)

Bursa Malaysia Listing Rules. Although compliance with best practices is voluntary, the

listing rules require firms to state in their annual report the extent of their compliance with an

explanation for any departure.1 In August 2000, MSWG was incorporated with five

1 While we note that annual reports of all firms listed on Bursa Malaysia are required to be audited, we cannot

rule out the possibility that the actual extent of compliance with best practices may differ from that disclosed in

1

founding members representing the major institutional investors in Malaysia.2 In carrying

out its role as an independent minority shareholder watchdog group, MSWG is responsible

for encouraging proactive shareholder participation in publicly listed companies. In this

respect, the group aims to harness the strength of large institutional investors to monitor and

institute changes in the companies they invest in. In particular, MCCG (Part 4 paragraphs

4.80 to 4.84) states “institutional shareholders have a responsibility to make considered use

of their votes” and “…should encourage direct contact with companies including

constructive communication with both senior management and board members about

performance, corporate governance and other matters affecting shareholder interest.”

In light of the above events, this study has three major aims. First, we examine the

extent of compliance of Malaysian companies with MCCG. The 2006 Institutional

Shareholders Services (ISS) Global Institutional Investor Study notes that compliance with

regulatory requirements provided the catalyst for increased importance of corporate

governance in recent years, with investors seeing corporate governance as not just an

externally imposed obligation, but an ownership responsibility or the “right thing to do”.

Corporate governance issues and activities have also been increasingly transformed into

competitive and portfolio advantages with investors perceiving corporate governance as a

competitive necessity “just to get in the game”. Consistent with this, we predict corporate

governance of Malaysian firms has improved as a result of MCCG.

Our second objective is to examine how institutional ownership is related to firms’

corporate governance practices, and whether this relationship is strengthened in periods the annual report. We thus assume that the extent of firms’ compliance with best practices is as stated in the

annual report.

2 The five founding members are Employees Provident Fund (EPF), Lembaga Tabung Angkatan Tentera

(LTAT), Permodalan Nasional Berhad (PNB), Lembaga Tabung Haji (LTH) and National Social Security

Organisation of Malaysia (SOCSO).

2

subsequent to the corporate governance reform. The increased awareness of the role of

institutional investors globally (Gillan and Starks, 2003) and the emergence of institutional

investors as fiduciary capitalism (Hawley and Williams, 1997) put pressure on institutional

investors to improve firms’ corporate governance. Since institutional investors hold a

substantial amount of equity, they are expected to play a more participative role in the firm.3

Therefore, we predict that institutional ownership is positively related to firms’ corporate

governance practices, and that this relationship is strengthened in periods subsequent to the

establishment of MSWG.

Finally, we investigate the impact of MCCG on stock performance in Malaysia. The

2002 survey by McKinsey and Company4 shows that 82 percent of Asian institutional

investors perceived corporate governance to be at least equally important as financial issues

in evaluating which companies to invest in. In particular, institutional investors in Malaysia

were willing to pay up to 22 percent premium for well-governed firms. The “codes of best

practices” suggested by MCCG are essentially aimed at improving board independence,

transparency and accountability to the company’s shareholders and other stakeholders, and

the effectiveness of the board in fulfilling both its conformance and performance functions.

Since these help to alleviate the agency problem by monitoring and controlling the

3 The participative role can be divided into (i) internal, where institutional investors take an active role in the

day-to-day management of the firm by having a representative on the board of directors and other committees

(audit, nomination and remuneration); and (ii) external, where institutional investors pressure firms by means of

litigation, media pressure, private negotiations, shareholders proposals and proxy voting.

4 McKinsey and Company’s Global Investor Opinion Survey was undertaken between April and May 2002 in

cooperation with the Global Corporate Governance Forum. Responses were received from over 200

institutional investors in 31 countries in Asia, Europe, Latin America, Middle East, Africa and North America,

which collectively are responsible for some USD 2 trillion of assets under management.

3

opportunistic behaviour of management (Jensen and Meckling, 1976), we expect better

governed firms have better stock performance than poorly governed ones.

Our study is based on a final sample of 440 firms listed on Bursa Malaysia from 1999

to 2002, representing 78 percent of all firms listed on the Main Board in 2002. As expected,

there is a significant improvement in corporate governance after the implementation of

MCCG, irrespective of the method we use to calibrate corporate governance.

Consistent with our prediction, institutional ownership is positively related to corporate

governance. Further analysis shows that this is driven mainly by the ownership of pressure

insensitive investors,5 thus supporting investors’ heterogeneity as an important factor in

determining firms’ corporate governance practices. Of the members of MSWG, it is only for

Employees Provident Fund (EPF) that we find the relationship between institutional

ownership and corporate governance strengthened in periods subsequent to the corporate

governance reform.

Stock performance is positively and significantly related to corporate governance, in

line with past studies (Klapper and Love, 2004; Durnev and Kim, 2005). After the

integration of MCCG into Bursa Malaysia Listing Rules in 2001, stock price performance

increased by an average of about 5%. This suggests that the corporate governance reform has

been successful and well received by the market.

A unique characteristic of Malaysia that we also exploit in this study is her political

economy, which has an effect on how firms are being run externally (based on political

intervention)6 and internally (based on ethnicity). In Malaysia, politically connected firms 5 Pressure insensitive investors are institutional investors that do not have any business relationship with the

firms (Brickley et al., 1988). See later in Section 3.0.

6 Faccio et al. (2006) document that from 1997 to 2002, the number of politically connected firms in Malaysia is

81, second to the United Kingdom with 118 politically connected firms. Considering the size of the capital

market, the proportion of politically-connected firms is thus staggeringly high in Malaysia.

4

are not necessarily owned by the state, but they are identified as favoured firms by the ruling

government (Gul, 2006). For example, in documenting the practice of cronyism during the

Asian financial crisis, Johnson and Mitton (2003) note that the capital control restriction

imposed by the Malaysian government in 1998 helped firms that were politically connected

to the then Prime Minister to outperform firms that were politically connected to his ex-

deputy.

Despite the strong and well-documented political connection in Malaysia (Johnson and

Mitton, 2003; Faccio et al., 2006), we find no evidence of a link between political connection

and stock performance. Instead, we find strong evidence that political connection has a

negative impact on corporate governance, consistent with past findings that political

connection is an important determinant of corporate transparency (Ball et al., 2003; Bushman

et al., 2004). It is also consistent with the 2002 McKinsey Emerging Market Policymaker

Opinion Survey on Corporate Governance, which reports politicians as one of the major

obstacles to corporate governance reforms. However, our results show that the negative

association between political connection and corporate governance is mitigated by

institutional ownership.

An important factor that has shaped Malaysia’s capital market is the close identification

between racial and economic functions (Gomez and Jomo, 1999). Ethnicity7 has shaped how

the country and business are run externally, through political means (Mohamad et al., 2006),

and internally, through cultural values (Haniffa and Cooke, 2002). Our results show that

although the proportion of Bumiputera directors on the board does not catalyze stock

7 The three main ethnic groups in Malaysia are Bumiputeras, Chinese and Indians. As at 2004, Malaysia has a

population of about 22.97 million people, of which 65 percent are Bumiputeras, 26 percent Chinese, 8 percent

Indians and 1 percent others (Department of Statistics Malaysia).

5

performance, as previously argued by Tan (2004), it is positively related to corporate

governance.

In the next section, we provide a description of the Malaysian institutional framework

followed by research methods and data in Section 3. Our results are discussed in Section 4

and Section 5 concludes.

2.0 Institutional Background

2.1 Malaysian Code on Corporate Governance (MCCG)

MCCG is largely derived from the recommendations of the Cadbury Report (1992)

and the Hampel Report (1998) in the United Kingdom, although MCCG tends to be more

regulatory driven (Ow-Yong and Guan, 2000). The recommendations set out in MCCG are

prescriptive in nature and fall under four main parts: Part (1) principles; Part (2) best

practices; Part (3) exhortations to other participants; and Part (4) explanatory notes.

Part (1) principles address four main issues: board of directors, directors’ remuneration,

shareholders, and accountability and audit. A narrative statement in the annual report of how

the relevant principles have been applied is perceived sufficient disclosure for investors to

assess the firms. Part (2) best practices provide a set of guidelines or practices relating to the

board of directors and accountability and audit to assist firms in designing their approach to

corporate governance. Compliance is voluntary but firms are required to state in their annual

reports the extent of their compliance with an explanation for any departure. Part (3)

exhortations to other participants are addressed primarily to institutional investors and

auditors to enhance their role in corporate governance. Part (4) explanatory notes provide

further explanation of the three parts mentioned above. However, unlike Part (2) best

practices, Part (4) guidelines on explanatory notes do not require firms to justify departures

from best practices.

6

2.2 Institutional Investors in Malaysia

As at 2003, total institutional shareholdings in Malaysia stood at about 13% of the total

market capitalisation of Bursa Malaysia. Although relatively low compared to those in

developed countries, institutional shareholdings in Malaysia are high compared to most other

nations in the region. This is a primary consequence of the 1970 New Economic Policy

(NEP), which uses Malaysia’s institutional investors as a tool to reduce equity ownership

imbalance between the various ethnic groups through increasing Bumiputera equity

ownership in the capital market (Gomez and Jomo, 1999; Tan, 2004).

The five largest public institutional investors, all members of MSWG, are two pension

funds (Employees Provident Fund (EPF) and Lembaga Tabung Angkatan Tentera (LTAT)),

an investment fund (Permodalan Nasional Berhad (PNB)), a pilgrim fund (Lembaga Tabung

Haji (LTH)) and an insurance company (National Social Security Organization of Malaysia

(SOCSO)). Collectively, their shareholdings represent about 70 percent of total institutional

shareholdings in firms listed on the Bursa Malaysia’s Main Board.

EPF, established in 1951, is the world’s first mandatory national pension fund for

private sector employees. As Malaysia’s largest contractual savings institution, EPF is both a

crucial financial intermediary, providing a key source of long-term investment capital, and a

central pillar of the country’s social policy and social security system. EPF’s investment

portfolio is mandated by the Malaysian law, requiring it to invest 70 percent of its funds in

Malaysian Government Securities, while investment in domestic equity market cannot

exceed 25 percent. As of 2004, EPF managed an investment portfolio of Malaysian equities

worth almost RM 50 billion (www.kwsp.gov.my).

The next major institutional investor is PNB. Established in 1972, PNB is Malaysia’s

first unit trust (“ownership-in-trust”) set up to encourage savings by Bumiputeras. It started

with a single unit trust called Amanah Saham Nasional (ASN) but now has multiple unit

7

trusts that cater for all groups of people such as youths (e.g., Amanah Saham Didik) and non-

Bumiputeras (e.g., Amanah Saham Malaysia). As at the end of 2003, PNB managed over

RM 15 billion worth of public and private equity in Malaysia, representing about two-thirds

of its total investment (www.pnb.com.my).

Established in August 1972, LTAT serves as a superannuation fund8 for the Armed

Forces of Malaysia. Similar to EPF, its objectives are to provide retirement and other

benefits to members of the Armed Forces (compulsory contributors) and to enable officers

and mobilised members of the volunteer forces in the service to participate in a savings

scheme.

LTH was established in 1962 with the aim of encouraging Malaysian Moslems to

save for journey to Mecca for pilgrimage. LTH’s role has evolved over time, from a mere

saving depository to providing Malaysian Moslems some returns on their investment. Like

other major institutional investors, LTH’s investment advisory board includes Islamic

scholars who must make sure that all investments are in accordance with syariah.9

Lastly, SOCSO was set up in 1971 to provide employment injury and invalidity pension

schemes to Malaysian workers. It is a “compulsory” insurance scheme for workers (both

government and private) in Malaysia.

8 Under the superannuation scheme, serving members of the other ranks in the Armed Forces are required to

contribute 10% of their monthly salary to LTAT with the government as employer contributing 15%. For

officers, participation is voluntary and the contributions are a minimum of RM 25 with a maximum of RM 200

monthly (www.ltat.org.my).

9 Syariah refers to the body of Islamic law. It is the legal framework within which public and some private

aspects of life are regulated for those living in a legal system based on Moslem principles of jurisprudence. For

example, Moslems are not permitted to be involved in gambling or any contracts involving future predictions or

uncertainty.

8

The Board and Investment Panel of Malaysia’s major institutional investors are

appointed by and report directly to the Ministry of Finance, with Bumiputeras typically

holding the position of the Chair of the board (Asher, 2001; Norhashim and Abdul Aziz,

2005).10 Coupled with the government’s development goals, this seriously constrains the

investment choices of Malaysia’s public institutional investors (Thillainathan, 2000),11 which

are heavily biased towards Bumiputera-run corporations (Norhashim and Abdul Aziz, 2005).

Constraints in investment choices also limit the exit route for these institutional investors in

relation to underperforming equities.

3.0 Research Methods and Data

3.1 Research Methods

To test the research objectives outlined in Section 1.0, we run the following two basic

regression models, which respectively has corporate governance index and firm performance

as the dependent variable:

CGINDEXit = REFORMit + OWNINSTit + OWNINSTit×REFORMi + control variables (1)

ROR_MADJit = REFORMit + CGINDEXit + ∆CGINDEXit + control variables (2)

10 For EPF, the investment panel comprises a Chairman, a representative of the Ministry of Finance, a

representative of the Central Bank and three individuals with expertise in finance and investment.

11 An example is the 2001 gradual takeover of Malaysian Airline System (MAS) from Naluri Berhad by two

main government-run institutional investors, Kumpulan Wang Amanah Pencen and Bank Simpanan Nasional.

Although this may be construed as a pure political bailout, others may see this takeover as an important national

obligation as there were speculations of a foreign takeover of MAS. In 2004, the offer by EPF to take control of

the financial group Rashid Hussain was perceived as another state-arranged banking merger to strengthen banks

ahead of further liberalisation in 2007 when competition from foreign banks is expected to intensify.

9

where for firm i at time t, CGINDEXit and ∆CGINDEXit are respectively the level and change

in the corporate governance index; REFORMit is a dummy that takes a value of one for the

post-2001 (MCCG) period and zero otherwise; OWNINSTit is institutional ownership; and

ROR_MADJit is stock performance.

Since the data are pooled across firms from 1999 to 2002, a panel analysis is used. We

use period seemingly unrelated regressions (SUR) to handle both heteroskedasticity and

contemporaneous correlations in the residuals for a given cross-section. For each test

variable, all observations lying below the first and above the 99th percentiles of the

distribution are censored.

Our first research aim requires us to test whether corporate governance practices of

Malaysian firms have improved as a result of MCCG. In the regression with corporate

governance (CGINDEX) as the dependent variable, we include the period indicator REFORM

and predict it to have a positive sign.

To test our second objective, we include the following two variables in the corporate

governance regression (equation (1)): institutional ownership (OWNINST) and its interaction

with the REFORM dummy (OWNINST×REFORM). A positive coefficient on both variables

would support the prediction that institutional investors utilise the strategy of giving

corporate governance advice to firms and that this strategy has intensified in periods

subsequent to the establishment of MSWG.

To investigate the impact of MCCG on stock performance in Malaysia, which is our

third objective, we include REFORM in equation (2) with stock performance as the

dependent variable. A positive estimate coefficient on REFORM would suggest an

improvement in stock performance subsequent to MCCG. Since stock performance could

change due to many other reasons unrelated to MCCG, we include CGINDEX and

∆CGINDEX. Both are expected to have a positive sign, in line with our prediction that better

10

governed firms and firms with an improvement in their corporate governance practices have

better stock performance.

We also capitalize on the unique political economy of Malaysia by testing whether

political connection (POLITIC) is related to corporate governance and stock performance.

To test whether ethnicity is an important determinant of stock performance and corporate

governance in Malaysia, we include the percentage of Bumiputera directors on the board

(BUMI) in the tests.

In both equations, we include other control variables which past studies have found to

be significant. In the regression with corporate governance as the dependent variable, we

include both the current (ROR_MADJ) and lagged stock performance (ROR_MADJ(-1)) to

test whether there is a lead-lag relationship between corporate governance and stock

performance. We hypothesize that in a properly functioning corporate governance system,

firms respond to poor stock performance by improving their corporate governance. In turn,

this should lead to improved stock performance.

We also control for audit quality (AUDIT) since higher quality auditors are more likely

to ensure greater transparency and to eliminate mistakes in financial statements because they

have a greater reputation to hold (Dye, 1993). Therefore, clients of higher quality auditors

have better corporate governance. Whether a firm has an American Depository Receipt

(ADR) or not is another determinant of corporate governance since ADR firms are associated

with greater disclosure and transparency (Coffee, 1999). So is the proportion of intangible

assets to total assets (INTANG). Because intangible assets are harder to manage and easier to

steal relative to tangible assets (Durnev and Kim, 2005), it is more necessary for firms with

high intangible assets to adopt stricter governance standards.

In the regression with stock performance as the dependent variable, we control for

institutional ownership (OWNINST), which past studies have found to be related to firm

11

performance (Del Guercio, 1996; Woidtke, 2002). Firms’ growth prospects (MTBV) is

another important determinant because if growth prospects are at least partially impounded in

stock prices, growth firms will have a higher market value relative to assets-in-place. This

suggests a positive relationship between growth prospects and stock performance. Demsetz

and Lehn (1985) argue that accounting profits reflect year-to-year fluctuations in underlying

business conditions and are thus an important determinant of share returns. The accounting

rate of return examined in this study is the return on assets (ROA), which measures the capital

intensity of the firm and how many dollars of earnings are derived from each dollar of assets

the firm controls.

Control variables common to both the governance and stock performance regressions

are market risk (MKTRISK); managerial ownership (MANOWN); firm size (ASSETS); and

leverage (DEBT). Industry dummies are also included in the tests to control for industry

effects.

3.2 Sample and Measurement

This study is based on a sample of 440 firms listed on the Main Board of Bursa

Malaysia from 1999 to 2002, giving us a total of 1760 firm-year observations. Data on

institutional ownership and corporate governance variables were hand-collected from annual

reports available on Bursa Malaysia website (www.bursamalaysia.com) and Mergent Online

database. Financial and stock price data were extracted from DataStream.

Unlike other studies which concentrate on just a few corporate governance features

such as duality and board independence, we examine a wider group of governance features

and condense them into one single measure.12 The corporate governance index (CGINDEX)

12 The use of indices in evaluating a firm’s governance structure has become a popular methodological approach

in measuring the overall quality of compliance with a comprehensive set of governance-related

12

is constructed using the 30 provisions of MCCG, which we classify into two groups, as

shown in the appendix. The first group (MCCG_PT2) relates primarily to compliance with

Part 2 of MCCG, best practices. The second group (MCCG_PT4) relates to the disclosure of

governance practices recommended in Part 4 of MCCG, explanatory notes. MCCG_PT2 and

MCCG_PT4 comprise 16 and 14 governance provisions respectively. The approach of

scoring is additive, giving a measure of CGINDEX for firm i based on an equally weighting

scheme used for the two parts:

1002

4_2_×

+= ii

iPTMCCGPTMCCG

CGINDEX

where ∑ ==

16

11612_

j jXPTMCCG and ∑ ==

14

11414_

j jYPTMCCG . Here, Xj and Yj are

equal to 1 if the jth governance provision is adhered to and 0 if it is not so that

0 ≤ CGINDEXi ≤100. We compute the governance measure for each of the four years of our

study period.

We also use MCCG_PT2 and MCCG_PT4 separately as alternative measures of

corporate governance in our tests. Additionally, we compute QUALITY and QUANTITY

measures of corporate governance. The former relates to the quality of governance practices

and consists of provisions that relate mainly to the independence of the board of directors

while the latter focuses on the number of corporate governance provisions that are disclosed

by a firm. Since our measures of corporate governance are highly correlated with each other

(the correlation coefficient is about 0.7 and better), we focus mainly on the composite

measure CGINDEX in subsequent analysis.

recommendations. Examples are the Standard and Poor 500, the CLSA (Credit Lyonnais Securities Asia) and

the ISS Corporate Governance Quotient.

13

Stock performance (ROR_MADJ) is measured by the continuously compounded annual

market-adjusted share returns based on the December (year-end) share prices and Bursa

Malaysia’s Composite Index.

We measure institutional ownership (OWNINST) as the percentage shareholdings by the

top five institutional investors in a firm. We include only the top five institutional investors

since their shareholdings capture reasonably well the level of effectiveness in monitoring a

firm (Cornett et al., 2007). Since institutional investors are not necessarily a homogenous

group, possessing the same investment strategies, monitoring techniques and desired

outcomes from their monitoring activities, we classify them into three separate groups

(Brickley et al., 1988; Bushee, 2001; Cornett et al., 2007). INSENSITIVE institutional

investors have no known business relationship with the firm and are immune to pressure to

comply with management decisions. Examples are pension funds and state-owned

institutional investors. SENSITIVE institutional investors, on the other hand, have a business

relationship with the firm. Afraid of losing business, they will succumb to management

decision and abide by it. Insurance firms, banks and other financial intermediaries fall into

this group. INDETERMINATE institutional investors are those that do not fall into either of

the first two categories.

We also measure institutional ownership as the percentage shareholding by each of the

five members of MSWG (i.e., EPF, PNB, LTAT, LTH and SOCSO) as well as their

cumulative percentage shareholdings (MSWG). These measures allow us to test whether or

not the percentage ownership of MSWG members, either individually or collectively, is

positively related to the corporate governance of sample firms.

14

To identify politically connected firms, we use data from three main sources: Mohamad

et al. (2006); Johnson and Mitton (2003); and the Khazanah Berhad13 website

(www.khazanah.com.my). POLITIC takes a value of one for firms that have been identified

to have political connection and zero otherwise. BUMI is measured by the percentage of

Bumiputera directors on the board.

We proxy audit quality (AUDIT) using the traditional Big 5 vs. non-Big 5 dichotomy

(Reed et al., 2000). ADR is a binary variable taking a value of one for firms with ADR

facilities and zero otherwise. INTANG is measured by the proportion of intangible assets to

total assets and growth prospects (MTBV) is measured by market to book value of equity. The

return on assets (ROA) is the ratio of earnings to total assets. Market risk (MKTRISK) is

measured as the beta coefficient obtained from a regression of monthly stock returns on

corresponding monthly market returns from 1995-2002 inclusive. We measure managerial

ownership (MANOWN) by the percentage shareholding by directors. Firm size (ASSETS) is

measured by total assets and leverage (DEBT) by total book value of debt divided by total

equity.

Table 1 describes sample firms’ characteristics. Over the sample period, CGINDEX

averages 38.87 and ranges widely from 0 to 79.91. This wide variation is also echoed in the

alternative measures of corporate governance. Firms score more than twice as high in the

QUALITY than in the QUANTITY aspect of the governance provisions, suggesting that firms

are more likely to comply with provisions that relate to the independence of the board of

directors rather than to the number of corporate governance provisions disclosed. 13 Khazanah Nasional Berhad is the investment holding arm of the Malaysian government to manage its

commercial assets. It is the trustee to the nation’s financial assets. Its objectives are to promote economic

growth and make strategic investments on behalf of the government which would contribute towards nation

building. Khazanah was incorporated in September 1993 and began operations in 1994. It is structured into a

holding company that is a wholly entity of the Ministry of Finance.

15

{Table 1}

The mean (median) percentage shareholding of institutional investors (OWNINST) is

12.58% (5.65%), with the highest proportion held by insensitive investors (mean=6.97% and

median=2.05%). The average (median) shareholding of the five MSWG members (MSWG)

is 8.63% (3.77%), with a maximum of 84.16%.14 Of the members, EPF is the biggest player

in terms of the average proportion of shares held (8.38%), followed by LTAT (4.14%).

The average stock performance (ROR_MADJ) of sample firms is -9.4%. This is not

surprising considering that the sample period is shortly after the 1997/8 Asian financial crisis,

when many firms recorded negative earnings performance.15 The average debt to equity

ratio (DEBT) is 47.7% (median 28.6%), lower than that in many developed countries. About

22 percent of sample firms are politically connected (POLITIC) and an average 46 percent of

board members are Bumiputeras (BUMI). The average (median) percentage shareholding by

directors (MANOWN) is 7.37% (0.41%). About 69 percent of the sample firms are audited

by a Big 5 auditor (AUDIT) while just 2 percent of the firms have American Depository

Receipts (ADR).

4.0 Results

4.1 Univariate

Table 2 reports univariate results of differences in corporate governance, institutional

ownership and firm characteristics in periods before and after the 2001 corporate governance

reform brought about by MCCG. Not surprisingly, there is a significant improvement in the 14 Although the objective of MSWG is to represent minority shareholders by protecting minority interest,

members of the MSWG are not necessarily the minority shareholders themselves.

15 The minimum ROA is -164%, as indicated in Table 1. To ensure that our results are not driven by these

negative earnings observations, we rerun the tests on a sample of firms with non-negative earnings. The results

do not change the thrust of our findings.

16

corporate governance practices of Malaysian firms, irrespective of the method used to

measure corporate governance. The median value of CGINDEX, for example, in the pre-

2001 period almost triples after 2001 (from 19.64 to 53.13). In 2002, the top three

CGINDEX-scoring firms are Asia Pacific Land Berhad, KPJ Healthcare and Kim Hin

Industries with scores of 79.91, 76.63 and 76.63 respectively. Amongst the government-

linked firms, Tenaga Nasional Berhad has the highest CGINDEX score of 69.64, while

Malaysian Airline System is a close second at 66.52 percent. Edaran Otomobil Berhad, a

company set up by the government to enhance the selling of Proton, the national car, and

Telekom, the national telecommunication company, are close behind at 66.07 percent. This

shows that government-linked firms do comply with MCCG and can have above average

corporate governance scores.

{Table 2}

Looking at the alternative measures, we note that compliance with the best practices

(MCCG_P2) has increased two-fold (the median has increased from 31.25 to 62.50) after the

reform, while disclosure of governance practices recommended in explanatory notes

(MCCG_P4) has increased more than 3 times (the median has increased from 14.28 to

50.00). The largest increase is observed for QUANTITY, relating to the number of corporate

governance provisions disclosed, whose median has increased four-fold from 9.52 to 42.86.

In comparison, the median score on QUALITY, relating to the independence of the board of

directors, has improved by about 50%, from 55.56 to 88.89.

Whilst the median institutional shareholding (OWNINST) has remained steady at about

6%, the median cumulative shareholding of the five MSWG members (MSWG) has increased

significantly after 2001, from 2.66% to 3.53%. MSWG members who experience a

significant increase in their shareholdings are EPF, PNB and SOCSO. The rest of the table

17

shows a significant drop in growth potential (MTBV), return on assets (ROA), leverage

(DEBT) and market risk (MKTRISK) after 2001.

Focusing on the 30 individual constituents of CGINDEX in Table 3, Malaysian

companies scored high prior to 2001 on the following six governance provisions: separation

of the Chairman and CEO posts (BOD_001), board independence (BOD_002), independence

of audit and nomination committees (AA_001 and NOM_004), CEO not sitting on the

remuneration committee (REM_003) and disclosure of board appointments (SHA_001B).

This suggests that some best practices were already widely adopted well before the 2001

reform.

{Table 3}

Of the remaining 24 provisions that scored low prior to 2001, 14 showed sizeable

improvements, including disclosures of directors’ remuneration (REM_008), current

appointments of directors (BOD_009) and their relationship with the company or other board

members (BOD_007), directors’ experience and educational background (BOD_010),

directors on remuneration and nomination committees (REM_002 and NOM_002), a

statement of internal control (AA_003), individual members’ attendance at audit committee

meetings (AA_005), frequency of board meetings (BOD_003), whether the CEO sits on the

nomination committee (NOM_003), the existence of a remuneration and nomination

committee (REM_001 and NOM_001) and independence of the remuneration committee

(REM_004).

Post 2001, Malaysian firms’ disclosure practices remained opaque on the following:

disclosure on components and details of directors remuneration scheme (REM_008A and

REM_008B), delegation and separation of duties among directors (BOD_008), methods of

board appointment (NOM_009), recommendations made by the nomination and remuneration

committees (NOM_008 and REM_009), material contracts with major shareholders

18

(SHA_001A), investor relations (SHA_002) and activities by the audit committee (AA_002).

This suggests that sizable improvements on these more sensitive disclosure items remain to

be made.

4.2 Multivariate

The main drawback of univariate tests is that they examine only one variable at a time.

To the extent that the independent variables do interact with each other in affecting the

dependent variable, multivariate tests are more appropriate. This is the focus of this section.

We use the three measures of corporate governance in all the tests. As expected from the

high correlations between them, similar results are found. For brevity, we report only the

results for CGINDEX.

We first examine the determinants of corporate governance, as captured by CGINDEX

in Table 4. Consistent with the univariate results, the significantly positive coefficient on

REFORM confirms a significant improvement in corporate governance (an increase in the

disclosure of about 10 governance provisions) subsequent to the reform in 2001. This result

is robust to the inclusion of control variables.

The results also show a significantly positive relationship between institutional

ownership (OWNINST) and CGINDEX, which disappears once we control for its interaction

with political connection (POLITIC×OWNINST). Consistent with our prediction,

institutional investors with no business relationship with the firm (INSENSITIVE) have

greater incentives and ability to monitor the firm (Brickley et al., 1988; Cornett et al., 2007).

This also supports the notion that INSENSITIVE institutional investors perform a fiduciary

duty to their contributors (Hawley and Williams, 1997). Focusing on the ownership of

MSWG members, the results show that the total shareholding (MSWG) is insignificant. So is

19

the ownership by EPF. Although not reported, we find similar results for the remaining four

MSWG members.

{Table 4}

Looking at the interaction between the various measures of institutional ownership and

the regulatory reform dummy, only EPF×REFORM is significant. Therefore, the positive

relationship between the shareholdings of EPF and corporate governance is increased

significantly after 2001, with the establishment of MSWG.16 This is consistent with the lead

role that EPF has taken in establishing MSWG.

The results show that only ROR_MADJ(-1) is significant, suggesting that better past

stock performance is related to better corporate governance. This suggests that firms may

decide to adopt good governance after a period of good performance. As expected, firms

with high intangible assets (INTANG) have better corporate governance to prevent misuse

and misappropriation of these assets. So have firms with higher market risk (MKTRISK).

Interestingly, CGINDEX is negatively and significantly related to POLITIC. This

finding is consistent with Ball et al. (2003) and Bushman et al. (2004), who respectively

argue that politically connected firms have lower financial reporting quality and

transparency. It also suggests that politically connected firms are more secretive than their

non-politically counterparts. Perhaps they can well afford to do as politically connected

firms are favoured by the government (Johnson and Mitton, 2003; Gul, 2006) and can use

this means to secure economic dealings or loans from financial institutions. Given their less

reliance on external equity funds, politically connected firms may pay less attention to

improving their corporate governance practices.

16 To date, EPF is the only member of the MSWG that has set up its own in-house unit to research and supervise

the corporate governance of its investments (http://www.mswg.org.my).

20

The coefficient on the interaction variable POLITIC×OWNINST is positive and

statistically significant, suggesting that the positive relationship between institutional

ownership and corporate governance is stronger for politically connected firms, possibly

because of their larger shareholdings.17 In particular, we note that the interaction variable

with the share ownership by MSWG members (POLITIC×MSWG) is significant. This result

suggests that the top 5 institutional investors and MSWG members are more effective in

improving corporate governance of politically connected firms than in non-connected firms,

thus mitigating the negative effect of political connection on corporate governance.

BUMI is positive and mostly significant, implying that firms with a higher proportion of

Bumiputera directors on the board have a higher CGINDEX. This finding is contrary to the

Hofstede-Gray framework of Haniffa and Cooke (2002), which suggests that Bumiputeras

are relatively more secretive, compared to the Chinese. Since high secrecy implies lower

disclosure, our results suggest that Bumiputera directors are less secretive and are more

willing to inform investors about firms’ governance structure.

Next, we examine the determinants of stock performance in Table 5. The results show

that stock performance is positively and significantly related to both the integration of

MCCG into Bursa Malaysia Listing Rules in 2001 (REFORM) and the level and change in

CGINDEX.18 These results remain intact when various control variables are introduced into

the model. Therefore, firms with good corporate governance practices perform significantly

better than poorly governed firms, consistent with our prediction and past studies (Klapper

and Love, 2004; Durnev and Kim, 2005). Firms that have improved their corporate 17 The average (median) institutional ownership for politically connected firms is 16.79% (7.29%), which is

significantly higher than the 11.40% (5.44%) for politically independent firms. Significant differences between

politically connected and unconnected firms are also observed for the ownership of MSWG members.

18 We do not include both CGINDEX and REFORM together in the same equation as they are highly correlated

with each other.

21

governance provisions also perform better, as indicated by ∆CGINDEX. In regression 4, the

estimated coefficient on ∆CGINDEX is 0.155. Based on an average increase in CGINDEX of

30.99 subsequent to the reform (see Table 2), the direct impact of MCCG on stock

performance is thus around 4.8% (=30.99×0.155).19

{Table 5}

Table 5 shows no relationship between institutional ownership and stock performance,

even after accounting for heterogeneity in institutional investors. We find similar results for

the percentage ownership by MSWG members, either collectively or individually. For

efficiency, we report only the results for EPF.

As expected, ROA is positive and mostly significant, connoting that a higher

accounting rate of return is associated with better stock performance. The estimated

coefficient on ASSETS is positive and generally significant, suggesting that larger firms,

which are more able to diversify their risk, perform better than smaller firms (Ghosh, 2001).

There is also some evidence that firms with higher leverage (DEBT) perform poorer.

However, no evidence is found to support market risk (MKT_RISK) and managerial

ownership (MANOWN) as determinants of stock performance.

We also do not find support for Tan’s (2004) argument that Bumiputera directors

(BUMI) enhance stock performance by means of their connections and ability to secure

contracts with the government. Contrary to Mohammad et al. (2006) and Johnson and

Mitton (2003), our measure of political connection (POLITIC) is also insignificant, perhaps

because the effect of political connection on stock return is fully anticipated.

19 This estimate depends on the validity of the model and the assumption that all the change in CGINDEX is due

to MCCG alone.

22

5.0 Conclusion

In 2001, Malaysia experienced two important events which have significant

implications for firms’ corporate governance in that country. The first is the integration of

MCCG as part of Bursa Malaysia Listing Rules and the second is the establishment of

MSWG, which emphasizes an increased monitoring role of institutional investors. Based on

a sample of 440 firms listed on Bursa Malaysia over a period that encompasses the regulatory

change in corporate governance (1999-2003), we find a significant improvement in corporate

governance subsequent to the governance reform. The implementation of MCCG has been

well received by the market, increasing stock price performance by an average of about

4.8%.

Of the MSWG members, it is only for EPF that we find the relationship between

institutional ownership and corporate governance strengthened in periods subsequent to

MCCG. This is consistent with the lead role that EPF has taken in establishing MSWG. We

also find the relationship between institutional ownership and corporate governance is

sensitive to the heterogeneity in institutional investors, with the ownership of only insensitive

investors significant in explaining corporate governance.

A unique characteristic of Malaysia, which we also exploit in this study, is her large

number of politically connected firms. Politically connected firms are found to have weaker

corporate governance in place than politically independent firms, in line with Ball et al.

(2003) and Bushman et al. (2004). The negative effect of political connection on corporate

governance is however mitigated by institutional ownership. Contrary to past studies

(Johnson and Mitton, 2003; Mohamad et al., 2006), we do not find any relationship between

political connection and stock performance.

We also examine the unique ethnicity background of Malaysia in relation to corporate

governance and stock performance. The results show that the proportion of Bumiputera

23

directors on the board is positively associated with corporate governance but unrelated to

stock performance. The former suggests a shift in Bumiputera directors from being secretive

(Haniffa and Cooke, 2002) to becoming more open in informing investors about firms’

governance structure. Our results are robust to various variable specifications and sampling

issues.

24

References

Asher, M.G., 2001, “Pension reform, capital markets and corporate governance in Malaysia”,

Journal of Financial REFORM and Compliance 9, 30-37.

Ball, R., A. Robin and J.S. Wu, 2003, “Incentives versus standards: Properties of accounting

income in four East Asian countries”, Journal of Accounting and Economics 36, 235-

270.

Brickley, J.A., R.C. Lease, and C.W. Smith, 1988, “Ownership structure and voting on

antitakeover amendments”, Journal of Financial Economics 20, 267-291.

Bushee, B., 2001, “Do institutional investors prefer near-term earnings over long-run

value?”, Contemporary Accounting Research 182, 207-246.

Bushman, R., J. Piotroski and A. Smith, 2004, “What determines corporate transparency?”,

Journal of Accounting Research 42, 207-252.

Claessens, S., S. Djankov, J.P.H. Fan and L.H.P. Lang, 1999, “Expropriation of minority

shareholders: Evidence from East Asia”, The World Bank Working Paper Series.

Claessens, S. and J.P.H. Fan, 2002, “Corporate governance in Asia: A survey”, International

Review of Finance 3, 77-103.

Coffee, J.C., 1999, “The future as history: The prospects for global convergence in corporate

governance and its implications”, Northwestern University Law Review 93, 641-707.

Cornett, M.M., A.J. Marcus, A. Saunders and H. Tehranian, 2007, “The impact of

institutional ownership on corporate operating performance”, Journal of Banking and

Finance (In Press, Corrected Proof).

Del Guercio, D., 1996, “The distorting effect of the prudent-man laws on institutional equity

Investments”, Journal of Financial Economics 40, 31-62.

Demsetz, H. and K. Lehn, 1985, “The structure of corporate ownership: Causes and

consequences, Journal of Political Economy 93, 115-1177.

Durnev, A. and E.H. Kim, 2005, “To steal or not to steal: Firm attributes, legal environment

and valuation”, Journal of Finance 603, 1461-1493.

Dye, R., 1993, “Auditing standards, legal liability and auditing wealth”, Journal of Political

Economy 101, 887-914.

Faccio, M., R. Masulis and J.J. McConnell, 2006, “Political connections and corporate

bailouts”, Journal of Finance 61, 2597-2635.

Fraser, D.R., H. Zhang and C. Derashid, 2006, “Capital structure and political patronage: The

case of Malaysia”, Journal of Banking and Finance 30, 1291-1308.

25

Gillan, S.L. and L.T. Starks, 2003, “Corporate governance, corporate ownership and the role

of institutional investors: A global perspective”, Journal of Applied Finance 13, 4-22.

Gomez, E.T. and K.S. Jomo, 1999, Malaysia’s Political Economy: Politics, Patronage and

Profits (Cambridge University Press).

Gul, F.A., 2006, “Auditors’ response to political connections and cronyism in Malaysia”,

Journal of Accounting Research 44, 931-963.

Hawley, J.P. and A.T. Williams, 1997, “The emergence of fiduciary capitalism”, Corporate

Governance: An International Review 5, 206-213.

Haniffa, R.M. and T.E. Cooke, 2002, “Culture, corporate governance and disclosure in

Malaysian corporations, Abacus 383, 317-349.

Jensen, M.C. and W.H. Meckling, 1976, “Theory of the firm: Managerial behaviour, agency

costs and ownership structure”, Journal of Financial Economics 34, 305-360.

Johnson, S. and T. Mitton, 2003, “Cronyism and capital controls: Evidence from Malaysia”,

Journal of Financial Economics 67, 351-382.

Klapper, L.F. and I. Love, 2004, “Corporate governance, investor protection, and

performance in emerging markets”, Journal of Corporate Finance 105, 703-728.

Mitton, T., 2002, “A cross-firm analysis of the impact of corporate governance on the East

Asian financial crisis”, Journal of Financial Economics 642, 215-241.

Mohamad, S., T. Hassan, and C.M. Chen, 2006, “Impact of political-business relationship:

ownership patterns and corporate performance in Malaysia”, Paper presented at the

AFA/FMA 2006 Meeting, Auckland, New Zealand.

Norhashim, M. and K. Abdul Aziz, 2005, “Smart partnership or cronyism? A Malaysian

perspective”, The International Journal of Sociology and Social Policy 258, 31-48.

Ow-Yong, K. and C. Kooi Guan, 2000, “Corporate governance codes: A comparison

between Malaysia and the U.K.”, Corporate Governance: An International Review 82,

125-132.

Reed, B.J., M.A. Trombley and D.S. Dhaliwal, 2000, “Demand for audit quality: The case of

Laventhol and Horwath’s auditees”, Journal of Accounting, Auditing and Finance 152,

183-198.

Tan, M.G, 2004, “Ownership, corporate control, corporate governance and firm performance

in Malaysia”, Unpublished PhD, Monash University.

Thillainathan, R., 2000, “The Employees Provident Fund of Malaysia: Asset allocation,

investment strategy and governance issues revisited”, Paper presented at the Pension

Fund Management, Kuala Lumpur, Malaysia, 14-15 August.

26

Woidtke, T., 2002, “Agents watching agents?: evidence from pension fund ownership and

firm value”, Journal of Financial Economics 63, 99-131.

27

Appendix Constituents of corporate governance index (CGINDEX)

MCCG_PT2 (based on Part 2 of MCCG) References BOD_001 Does the company split the Chairman and CEO/Managing Director posts? Sect AA Par II BOD_002

Does the company comply with MCCG recommendation on the proportion of independent directors on the board?

Sect AA Par III

BOD_003 Is the frequency of board of directors’ meetings disclosed? Sect AA, Par XIV NOM_001 Does the company have a nomination committee? Sect AA, Par VIII NOM_004 Are the majority of directors on the nomination committee independent? Sect AA Par VIII NOM_003 Does the CEO not sit on the nomination committee? Sect AA Par VIII NOM_008 Does the company disclose recommendations made by the nomination committee? Sect AA Par VIII NOM_009 Does the company disclose methods of board appointments? Sect AA Par X REM_001 Does the company have a remuneration committee? Sect AA Par XXIV REM_002 Is the list of remuneration committee members disclosed? Sect AA Par XXIV REM_003 Does the CEO not sit on the remuneration committee? Sect AA Par XXIV REM_004 Are the majority of directors on the remuneration committee independent? Sect AA Par XXIV REM_009 Does the company disclose recommendations made by the remuneration committee? Sect AA Par XXIV AA_001 Are the majority of directors on the audit committee independent? Sect BB Par I AA_002 Does the company disclose activities carried out by the audit committee? Sect BB Par II AA_003 Does the company disclose a statement on internal control? Sect BB Par VII MCCG_PT4 (based on Part 4 of MCCG) BOD_007

Does the company disclose relationships that directors have with the company or other board members?

Sect AA Par 4.23

BOD_008 Does the company disclose delegation and separation of duties among directors? Sect AA Par 4.19-4.20 BOD_009 Does the company disclose current appointments of directors? Sect AA Par 4.42 BOD_010 Does the company disclose directors’ experience and education background? Sect AA Par 4.22 NOM_002 Is the list of the nomination committee members disclosed? Sect AA Par 4.34 NOM_005 Is the frequency of nomination committee meetings disclosed? Sect AA Par 4.34 REM_008 Does the company disclose directors’ remuneration? Sect B Par 4.6 REM_008A Does the company disclose components of the remuneration scheme of directors? Sect B Par 4.8 REM_008B Does the company disclose details of individual remuneration scheme of directors? Sect B Par 4.10 SHA_001 Does the company disclose affiliations with major shareholders? Sect CC Par 4.70 SHA_001A Does the company disclose material contracts with major shareholders? Sect CC Par 4.69-4.78 SHA_001B Does the company disclose board appointments? Sect CC Par 4.70 SHA_002 Does the company disclose investor relations? Sect BB, Par 4.80 AA_005 Does the company disclose individual members’ attendance at audit committee meetings? Sect BB Par 4.64

28

Table 1 Descriptive statistics of 440 sample Malaysian firms, 1999-2002

CGINDEX is a composite measure of the Malaysian Code on Corporate Governance (MCCG). MCCG_PT2 is a composite measure of corporate governance based on Part 2 of MCCG, which requires firms to explain and provide alternative practices adopted when departing from best practices. MCCG_PT4 is a composite measure of corporate governance based on Part 4 of MCCG, which provides explanatory notes to the principles and best practises. QUALITY and QUANTITY are respectively the quality and quantity of governance provisions based on MCCG. OWNINST is the percentage shareholdings by the top 5 institutional investors. INSENSITIVE, SENSITIVE and INDETERMINATE are the percentage shareholdings by respectively pressure insensitive, sensitive and indeterminate institutional investors. MSWG is the total percentage shareholdings by the five MSWG members; the shareholding of each of the members is denoted by the abbreviations of their name (i.e., EPF, PNB, LTAT, LTH and SOCSO). ROR_MADJ is the continuously compounded annual market-adjusted returns. MTBV is market to book value of equity. ROA is earnings divided by total assets. ADR takes the value of 1 if the firm has American Depository Receipts and zero otherwise. AUDIT takes the value of 1 for Big 5 auditors and zero otherwise. INTANG is total intangible assets over total assets (ASSETS). DEBT is total debt over total equity. MANOWN is total directors’ shareholdings. MKTRISK is systematic risk (beta). POLITIC takes the value of 1 for politically connected firms and zero otherwise. BUMI is the percentage of Bumiputera directors on the board.

Mean Median Maximum Minimum Std. Dev. Panel A: Corporate governance CGINDEX 38.870 36.607 79.911 0.000 18.379 MCCG_PT2 44.812 31.250 87.500 0.000 19.457 MCCG_PT4 32.982 35.714 78.571 0.000 19.344 QUALITY 65.846 55.556 100.000 0.000 22.170 QUANTITY 27.935 28.571 71.429 0.000 18.572 Panel B: Institutional ownership OWNINST 12.579 5.654 90.553 0.000 18.159 INSENSITIVE 6.972 2.047 78.566 0.000 13.364 SENSITIVE 1.191 0.000 74.254 0.000 3.963 INDETERMINATE 4.416 0.000 75.269 0.000 10.825 MSWG 8.634 3.765 84.156 0.000 13.973 EPF 8.379 3.032 63.025 0.000 4.751 PNB 2.378 0.000 75.269 0.000 10.628 LTAT 4.136 0.000 74.545 0.000 5.738 LTH 0.939 0.000 29.713 0.000 3.123 SOCSO 0.866 0.000 2.944 0.000 0.221 Panel C: Firm characteristics ROR_MADJ -9.395 -7.655 164.524 -149.39 34.979 MTBV 1.378 0.980 18.830 -15.650 1.903 ROA 4.870 4.427 140.162 -164.020 12.500 ADR 0.022 0.000 1.000 0.000 0.145 AUDIT 0.693 1.000 1.000 0.000 0.461 INTANG (million) 494 0.094 2,700 0.000 202 ASSETS (million) 2,620 580 91,185 3.626 805 DEBT 0.477 0.286 8.718 -6.338 0.845 MANOWN 7.366 0.413 78.256 0.000 14.151 MKTRISK 1.126 1.128 3.045 0.126 0.355 POLITIC 0.219 0.000 1.000 0.000 0.413 BUMI 0.459 0.400 1.000 0.000 0.275

Table 2 Univariate analysis of differences in corporate governance, institutional ownership and firm

characteristics in the pre- and post-MCCG periods for 440 Malaysian firms, 1999-2002 CGINDEX is a composite measure of the Malaysian Code on Corporate Governance (MCCG). MCCG_PT2 is a composite measure of corporate governance based on Part 2 of MCCG, which requires firms to explain and provide alternative practices adopted when departing from best practices. MCCG_PT4 is a composite measure of corporate governance based on Part 4 of MCCG, which provides explanatory notes to the principles and best practises. QUALITY and QUANTITY are respectively the quality and quantity of governance provisions based on MCCG. OWNINST is the percentage shareholdings by the top 5 institutional investors. INSENSITIVE, SENSITIVE and INDETERMINATE are the percentage shareholdings by respectively pressure insensitive, sensitive and indeterminate institutional investors. MSWG is the total percentage shareholdings by the five MSWG members; the shareholding of each of the members is denoted by the abbreviations of their name (i.e., EPF, PNB, LTAT, LTH and SOCSO). ROR_MADJ is the continuously compounded annual market-adjusted returns. MTBV is market to book value of equity. ROA is earnings divided by total assets. ADR takes the value of 1 if the firm has American Depository Receipts and zero otherwise. AUDIT takes the value of 1 for Big 5 auditors and zero otherwise. INTANG is total intangible assets over total assets (ASSETS). DEBT is total debt over total equity. MANOWN is total directors’ shareholdings. MKTRISK is systematic risk (beta). POLITIC takes the value of 1 for politically connected firms and zero otherwise. BUMI is the percentage of Bumiputera directors on the board. Significant p-values are in bolds.

1999-2000 2001-2002 p-value (Mann

Whitney) Mean Median Mean Median p-value

(t-test)

Panel A: Corporate governance CGINDEX 19.675 19.643 50.668 53.125 0.000 0.000 MCCG_P2 28.645 31.250 57.083 62.500 0.000 0.000 MCCG_P4 13.879 14.286 47.488 50.000 0.000 0.000 QUALITY 49.991 55.556 77.842 88.889 0.000 0.000 QUANTITY 9.676 9.524 41.792 42.857 0.000 0.000 Panel B: Institutional ownership OWNINST 12.653 5.622 12.653 5.684 0.857 0.431 INSENSITIVE 6.934 2.073 7.001 2.026 0.923 0.158 SENSITIVE 1.127 0.000 1.241 0.000 0.578 0.521 INDETERMINATE 4.423 0.000 4.411 0.000 0.983 0.422 MSWG 8.047 2.657 8.652 3.531 0.393 0.006 EPF 2.194 0.000 2.530 0.414 0.162 0.004 PNB 3.937 0.000 4.301 0.000 0.499 0.030 LTAT 1.007 0.000 0.882 0.000 0.667 0.536 LTH 0.847 0.000 0.882 0.000 0.829 0.101 SOCSO 0.038 0.000 0.057 0.000 0.090 0.009 Panel C: Firm characteristics ROR_MADJ -15.138 -15.304 -4.809 -3.771 0.000 0.000 MTBV 1.654 1.200 1.149 0.840 0.000 0.000 ROA 5.747 5.035 4.144 4.094 0.011 0.005 ADR 0.022 0.000 0.021 0.000 0.836 0.836 AUDIT 0.711 1.000 0.679 1.000 0.185 0.185 INTANG(million) 49.620 0.239 49.250 0.000 0.972 0.631 ASSETS (million) 3030 622 2890 513 0.793 0.018 DEBT 0.584 0.313 0.388 0.259 0.000 0.001 MANOWN 6.795 0.280 7.840 0.594 0.145 0.000 MKTRISK 1.144 1.141 1.111 1.115 0.061 0.070 POLITIC 0.236 0.000 0.204 0.000 0.121 0.121 BUMI 0.462 0.400 0.457 0.380 0.718 0.698

30

Table 3 Univariate analysis of differences in constituents of CGINDEX in the pre- and post-MCCG

periods for 440 Malaysian firms, 1999-2002 CGINDEX is a composite measure of the Malaysian Code on Corporate Governance (MCCG). MCCG_PT2 is a composite measure of corporate governance based on Part 2 of MCCG, which requires firms to explain and provide alternative practices adopted when departing from best practices. MCCG_PT4 is a composite measure of corporate governance based on Part 4 of MCCG, which provides further best practices and explanatory notes to the principles and best practises. BOD is board of director, NOM is nomination committee, REM is remuneration committee, AA is audit committee, and SHA is shareholders communication. Significant p-values are in bolds.

1999-2000 2001-2002

Mean Median Mean Median p-value (t-test)

p-value (Mann Whitney)

Panel A: MCCG_PT2 BOD_001 0.707 1.000 0.733 1.000 0.240 0.240 BOD_002 0.860 1.000 0.918 1.000 0.000 0.000 BOD_003 0.015 0.000 0.739 1.000 0.000 0.000 NOM_001 0.017 0.000 0.685 1.000 0.000 0.000 NOM_004 0.904 1.000 0.917 1.000 0.275 0.274 NOM_003 0.014 0.000 0.644 1.000 0.000 0.000 NOM_008 0.000 0.000 0.005 0.000 0.003 0.003 NOM_009 0.000 0.000 0.009 0.000 0.001 0.001 REM_001 0.020 0.000 0.689 1.000 0.000 0.000 REM_002 0.018 0.000 0.652 1.000 0.000 0.000 REM_003 0.977 1.000 0.866 1.000 0.000 0.000 REM_004 0.020 0.000 0.576 1.000 0.000 0.000 REM_009 0.000 0.000 0.002 0.000 0.034 0.034 AA_001 0.980 1.000 0.978 1.000 0.183 0.183 AA_002 0.044 0.000 0.278 0.000 0.000 0.000 AA_003 0.009 0.000 0.441 0.000 0.000 0.000 Panel B: MCCG_PT4 BOD_007 0.046 0.000 0.898 1.000 0.000 0.000 BOD_008 0.027 0.000 0.068 0.000 0.000 0.000 BOD_009 0.213 0.000 0.950 1.000 0.000 0.000 BOD_010 0.224 0.000 0.955 1.000 0.000 0.000 NOM_002 0.012 0.000 0.662 1.000 0.000 0.000 NOM_005 0.002 0.000 0.050 0.000 0.000 0.000 REM_008 0.117 0.000 0.728 1.000 0.000 0.000 REM_008A 0.000 0.000 0.023 0.000 0.000 0.000 REM_008B 0.003 0.000 0.066 0.000 0.000 0.000 SHA_001 0.402 0.000 0.560 1.000 0.000 0.000 SHA_001A 0.011 0.000 0.064 0.000 0.000 0.000 SHA_001B 0.869 1.000 0.899 1.000 0.180 0.180 SHA_002 0.000 0.000 0.022 0.000 0.000 0.000 AA_005 0.017 0.000 0.705 1.000 0.000 0.000

31

Table 4 Determinants of corporate governance for 440 Malaysian firms, 1999-2002

The dependent variable is CGINDEX, which is a composite measure of the Malaysian Code on Corporate Governance (MCCG). OWNINST is the percentage shareholdings by the top 5 institutional investors. INSENSITIVE, SENSITIVE and INDETERMINATE are the percentage shareholdings by respectively pressure insensitive, sensitive and indeterminate institutional investors. MSWG is the total percentage shareholdings by the five MSWG members. EPF is the percentage shareholding by Employees Provident Fund. REFORM takes a value of one for the post-2001 period and zero otherwise. ROR_MADJ and ROR_MADJ(-1) are respectively the continuously compounded annual market-adjusted returns. MTBV is market to book value of equity. ROA is earnings divided by total assets. ADR takes the value of 1 if the firm has American Depository Receipts and zero otherwise. AUDIT takes the value of 1 for Big 5 auditors and zero otherwise. INTANG is total intangible assets over total assets (ASSETS). DEBT is total debt over total equity. MANOWN is total directors’ shareholdings. MKTRISK is systematic risk (beta). POLITIC takes the value of 1 for politically connected firms and zero otherwise. BUMI is the percentage of Bumiputera directors on the board. Industry dummies are included but not reported. t-statistics are italicized. *, ** and *** denote significance at the 10%, 5% and 1% level respectively.

Regressions 1 2 3 4 5 6 REFORM 30.578 30.553 29.915 29.618 30.43 29.94 48.006*** 48.026*** 43.631*** 46.769*** 49.230*** 50.930*** OWNINST 0.057 0.028 0.028 3.188*** 1.292 1.090 OWNINST×REFORM 0.004 0.002 -0.002 0.130 0.078 -0.078 INSENSITIVE 0.070 1.982** SENSITIVE -0.040 -0.464 INDETERMINATE -0.022 -0.568 MSWG -0.001 -0.027 EPF 0.014 0.212 INSENSITIVE×REFORM 0.041 1.003 MSWG×REFORM 0.022 0.565 EPF×REFORM 0.270 2.565** ROR_MADJ -0.008 -0.008 0.000 0.000 -0.008 -0.008 -1.026 -1.101 0.022 -0.036 -1.100 -1.164 ROR_MADJ(-1) 0.021 0.021 0.021 0.022 2.244** 2.315** 2.286** 2.372** AUDIT 0.171 0.181 0.252 0.179 0.297 0.149 0.244 0.260 0.317 0.225 0.423 0.212 ADR 3.253 2.724 2.982 3.624 3.428 3.930 1.471 1.233 1.189 1.455 1.552 1.783* INTANG 9.287 10.592 11.078 10.941 9.863 10.151 1.781* 2.029** 1.955* 1.936* 1.882* 1.945* Ln(ASSETS) 0.029 -0.022 -0.086 -0.102 0.005 -0.047 0.125 -0.097 -0.322 -0.382 0.020 -0.203 DEBT -0.115 -0.084 0.084 0.211 -0.086 -0.002 -0.300 -0.220 0.167 0.413 -0.223 -0.006 MKTRISK 1.789 1.797 1.968 1.674 1.664 1.544 1.886* 1.904* 1.804* 1.532 1.753* 1.630

32

Table 4 continued MANOWN 0.017 0.016 0.017 0.019 0.012 0.016 0.765 0.729 0.656 0.724 0.536 0.726 BUMI 1.839 1.927 2.800 2.698 2.650 2.486 1.412 1.486 1.885* 1.819* 2.109** 1.982** POLITIC -1.261 -2.584 -2.664 -2.093 -2.198 -2.188 -1.530 -2.636*** -2.295*** -1.877* -2.311*** -2.218*** POLITIC×OWNINST 0.094 0.094 2.415*** 2.018** POLITIC×INSENSITIVE 0.028 0.477 POLITIC×MSWG 0.108 2.063** POLITIC×EPF 0.202 1.284 CONSTANT 16.970 18.314 20.268 21.1400 17.671 19.150 3.275*** 3.529*** 3.372*** 3.521*** 3.387*** 3.668*** Adjusted R2 0.683 0.686 0.613 0.621 0.683 0.686 Cross sections included 431 431 427 427 431 431 Total panel observations 1412 1412 1164 1164 1412 1412

33

Table 5 Regression results for stock performance for 440 Malaysian firms, 1999-2002

The dependent variable is ROR_MADJ, the continuously compounded annual percentage market-adjusted return. CGINDEX is a composite measure of the Malaysian Code on Corporate Governance (MCCG) and ∆CGINDEX is the annual change in CGINDEX. REFORM takes a value of one for the post-2001 period and zero otherwise. OWNINST is the percentage shareholdings by the top 5 institutional investors. INSENSITIVE, SENSITIVE and INDETERMINATE are the percentage shareholdings by respectively pressure insensitive, sensitive and indeterminate institutional investors. MSWG is the total percentage shareholdings by the five MSWG members. EPF is the percentage shareholding by Employees Provident Fund. MTBV is market to book value of equity. ROA is earnings divided by total assets. ADR takes the value of 1 if the firm has American Depository Receipts and zero otherwise. AUDIT takes the value of 1 for Big 5 auditors and zero otherwise. INTANG is total intangible assets over total assets (ASSETS). DEBT is total debt over total equity. MANOWN is total directors’ shareholdings. MKTRISK is systematic risk (beta). POLITIC takes the value of 1 for politically connected firms and zero otherwise. BUMI is the percentage of Bumiputera directors on the board. Industry dummies are included but not reported. t-statistics are italicized. *, ** and *** denote significance at the 10%, 5% and 1% level respectively.

Regressions 1 2 3 4 5 6 7 REFORM 11.121 19.220 5.945*** 6.868*** CGINDEX 0.213 0.289 0.214 0.213 0.212 4.302*** 4.079*** 4.316*** 4.298*** 4.265*** ∆CGINDEX 0.238 0.155 3.084*** 2.106** OWNINST -0.018 -0.010 -0.037 -0.021 -0.245 -0.137 -0.448 -0.260 INSENSITIVE 0.063 0.626 SENSITIVE -0.274

-1.061-0.030-0.315

-0.004

INDETERMINATE MSWG -0.041 EPF 0.054 0.227 MTBV -0.332 -0.160 0.215 0.733 -0.323 -0.321 -0.346 -0.611 -0.296 0.333 1.135 -0.596 -0.590 -0.634 ROA 0.147 0.176 0.130 0.159 0.151 0.147 0.151 1.707* 2.045** 1.319 1.627 1.752* 1.704* 1.751*

34

Table 5 continued Ln(ASSETS) 1.167 1.255 0.858 1.008 1.179 1.132 1.121 1.671* 1.803* 1.057 1.258 1.693* 1.624 1.598 DEBT -1.670 -1.374 -3.393 -3.012 -1.442 -1.694 -1.515 -1.235 -1.017 -1.861* -1.673* -1.065 -1.252 -1.117 MKTRISK 1.142 1.934 -3.600 -2.202 0.721 1.331 1.097 0.390 0.663 -1.069 -0.661 0.245 0.454 0.373 MANOWN 0.015 0.014 -0.115 -0.115 0.016 0.018 0.020 0.212 0.208 -1.432 -1.463 0.235 0.260 0.291 BUMI -3.978 -3.590 -4.931 -4.564 -4.055 -4.362 -4.556 -1.021 -0.926 -1.109 -1.041 -1.041 -1.157 -1.206 POLITIC -3.093 -3.406 0.459 -0.433 -2.432 -3.294 -3.406 -1.031 -1.142 0.131 -0.125 -0.845 -1.143 -1.123 POLITIC×OWNINST 0.070 0.080 0.042 0.065 0.614 0.711 0.325 0.510 POLITIC×INSENSITIVE -0.009 -0.058 POLITIC×MSWG 0.123 0.835 POLITIC×EPF 0.330 0.682 CONSTANT -47.580 -48.830 -44.170 -49.791 -47.191 -46.880 -46.350 -2.978*** -3.081*** -2.378*** -2.725*** -2.960*** -2.943*** -2.892*** Adjusted R2 0.013 0.024 0.048 0.077 0.012 0.013 0.012 Cross sections 435 434 431 431 434 434 434 Total observations 1476 1468 1098 1098 1465 1465 1465

35