The role of monitoring mechanisms towards company’s performance: Evidence from politically connected companies in Malaysia

Abstract

Purpose

Past studies show that companies’ connection with the government (or politically connected companies (PCCs)) contributed negatively to their financial performance. The grabbing hand theory suggests that political connection demand companies to serve political and social obligation that exhaust companies’ financial resources. The purpose of this paper is to extend the previous studies by examining the role of monitoring mechanisms, specifically corporate governance mechanism and institutional ownership (IO), whether they weaken or strengthen the financial performance of PCCs in Malaysia.

Design/methodology/approach

The sample consists of all companies listed on the Main Board of Bursa Malaysia (previously known as Kuala Lumpur Stock Exchange) for the year of 2004-2007. The time periods were chosen because there were no significant economic and political events that could possibly distorted the financial and non-financial data.

Findings

The findings show that companies’ political connection (the presence of political figure or government representative as members of board of director) has consistently showing negative relationship with performance. The result is consistent with the grabbing hand theory that argues that companies’ connection with government would actually destroy companies’ value. The monitoring role of corporate governance as measured by the percentage of independent board members does not have any significant effect on firm’s performance. The monitoring role of corporate governance as measured by the composition of independent board members have shown a positive significant effect on the company’s performance. However the second monitoring mechanism, the percentage of institutional investors, have a tendency to weaken the company’s performance.

Originality/value

The findings of this study provide an additional understanding of the consequence of government intervention on companies’ performance. This study also highlights the role of monitoring mechanism (independence board members and IO) in strengthening or weakening the performance. The findings suggest that the proper appointment criteria for board members should be seriously considered to ensure better corporate governance structure. Therefore, the formation of the nomination committee as suggested by the current Malaysian Code of Corporate Governance play an important contribution to ensure candidates nominated as board members have proper credentials and qualifications to carry out responsibilities as board members.

Citation

Jaffar, R. and Abdul-Shukor, Z. (2016), "The role of monitoring mechanisms towards company’s performance: Evidence from politically connected companies in Malaysia", Journal of Accounting in Emerging Economies, Vol. 6 No. 4, pp. 408-428. https://doi.org/10.1108/JAEE-05-2014-0021

1. Introduction

Politically connected companies (PCCs) enjoy substantial economic benefits due to systematic exchange of favor between the companies and politicians (Chaney et al., 2011). PCCs have considerable access to government subsidies and contracts (Agrawal and Knoeber, 2001; Gomez and Jomo, 1999), preference in getting business loans (Johnson and Mitton, 2003; Khwaja and Milan, 2005) and have better competitive positions and greater market power (Faccio, 2006). In Malaysia, during the Asian economic crisis in the late 1990s, PCCs who were facing severe financial distress would be more likely to be bailed out by the government compared to non-PCCs (Gomez and Jomo, 1999; Johnson and Mitton, 2003). Since PCCs are also shown to be paying lower effective tax rates (Adhikari et al., 2006), they should also be enjoying lower operating costs (Boubakri et al., 2012).

The above mentioned advantages enjoyed by PCCs, would ideally lead them to be in better financial position compared to non-PCCs. However, past studies have shown contradicting and inconsistent results. In Singapore, PCCs are considered better performers (Ang and Ding, 2006; Feng et al., 2004; Ramirez and Tan, 2004) and equally efficient as a privately run organization (Feng et al., 2004). In these studies, PCCs are measured based on the percentage of government ownership in the companies. In Malaysia, PCCs are considered to have higher risks of financial misrepresentation (especially during the economic crisis), therefore are charged higher audit fees (Gul, 2006). In China, the findings are a mixture. Fan et al. (2007) find out that non-PCCs performed better than PCCs. The post-initial public offering performance, in terms of stock returns of privatized companies show that non-PCCs perform better than PCCs. Study by Wu et al. (2012) show that listed companies with politically connected managers are financially better compared to companies without any connections. In Indonesia, companies which are close with the ruling parties, relied heavily on domestic financing and not taking advantage of the benefits of available global financing (Leuz and Gee, 2006).

In developed countries, Bertrand et al. (2004) provided evidence that PCCs in France exhibit lower profits than non-PCCs, especially in the election year and in politically contested areas. They conclude that political connections help affiliated politicians extract political benefits at the expense of other stakeholders of the company. On the contrary, Fisman (2001) documented evidence that in the USA, PCCs are shown to have better financial performance compare to non-PCCs.

These inconsistent findings have provided this study with an opportunity to extend the previous works by looking at the role of corporate governance mechanism and ownership structure in the financial performance of PCCs. Malaysia has provided us with a good research setting to extend the study as the government intervention in private business started in 1970s with the formation of New Economic Policy[1] (1970-1990) and the National Development Policy[2] (1991-2000). The two policies were intended to reduce socio-economic imbalance between three main races in Malaysia (mainly Malay, Chinese and Indian) by promoting and encouraging Bumiputra[3] participation in corporate ownership. Thus, Bumiputra companies and companies who are deemed to be compatible with the government policy are likely to be preferred by the government in getting government projects and subsidies (Fraser et al., 2006).

Additionally in 2001, the Malaysia Code of Corporate Governance (MCCG) was introduced to improve the corporate governance mechanism of companies in Malaysia (Abdul Wahab et al., 2008). The MCCG (2001) stressed the role of board of director (BOD) to ensure efforts and resources are geared toward the best interest of the company and its stakeholders. The Code also specified that the minimum level of independent board member should be at least 30 percent. The Code was revised in 2007 and 2012. In latest revision, MCCG 2012 (Securities Commission (SC) Malaysia, 2012) focuses on strengthening board structure and composition of board members. Moreover, MCCG (SC Malaysia, 2012) redefines the definition of independent board members as member whose designation as “independent directors” should cumulatively not exceeding more than nine years. These developments indicate that the issue of “independent board member” is a critical and an ongoing issue that deserves constant investigation of its implication especially toward companies’ performance. Therefore, this study investigate the role of corporate governance mechanism, specifically the percentage of independent board member, to provide further explanations of the performance of PCCs compared to non-PCCs.

Furthermore, the ownership structure such as the level of institutional shareholders could also influence company performance. Institutional investors are investment intermediaries who act on behalf of the beneficiaries (Lang and McNichols, 1997). They make investment decisions to provide a reasonable and expected rate of returns (Jensen and Meckling, 1976). Therefore, institutional investors will play an active role in monitoring the performance of their investment (Abdul Wahab et al., 2008). Past event-type studies have provided evidence to show that institutional investors are effective corporate monitors (Abdul Wahab et al., 2009). Since past studies have shown that the relationship between PCCs and performance is not consistent, this study examines if institutional share ownership can influence the financial performance among companies connected with the government. Specifically this study investigates whether institutional investors (IO) weaken or strengthen financial performance of PCCs.

Our findings show that companies connections with the government does not help them to improve their financial performance. The connections are measured in two ways; the appointment of a politician from a ruling government as a member of the board, and the appointment of government officers as members of the board. Our results indicate that the company’s connections consistently contribute in a negative way to performance. This could be due to the fact that the appointment of members connected with the government might not be due to professional qualifications but due to political considerations. Therefore, the competency and the expected contributions would be the least important criteria in an appointment consideration.

However when the PCCs are moderated by the existence of independent board members, our study found that independent board members consistently show tendency that they contribute in improving companies’ performance among PCCs. This finding provides support on the effort of the government to improve companies’ governance through the MCCG (SC Malaysia, 2007, 2012) requirement for independent directors on the board to be high. On the other hand, the existence of institutional ownership (IO) does not indicate a proper monitoring role when it tends to weaken the financial performance of PCCs. This could be due to the fact that IO in Malaysia highly consists of ownership among entities closely linked to the government. Hence the lack of monitoring role in the context of corporate governance toward companies performance.

The next section of this paper discusses the related past studies and the development of hypotheses. This will be followed by the discussion of the research methodology, findings and conclusion of the study.

2. Literature review and hypotheses development

The extensive network of political connections between private companies and the ruling government is very common in every developing economy (Polsiri and Jiraporn, 2012). The network can be in various forms. First and foremost commonly studied network is based on government ownership (see e.g. Ang and Ding, 2006; Feng et al., 2004; Ramirez and Tan, 2004) and second is based on a government representation at the companies’ highest level of decision-making authority such as representation as member of the board.

There are studies that show government influences in companies especially in the form of share ownership, has not decreased even after the privatization exercises. In Malaysia, for example, the government has continued to own consistently more than 50 percent of ownership in major utility companies such as Tenaga Nasional Berhad and Telekom Malaysia Berhad. These two utility companies also have substantial government representation as board members.

The phenomenon is similar in other developing countries. Boubakri et al. (2012) provide extensive evidence of the degree of government connections and control of private companies after the privatization exercise. On the average, the government ownership across 27 developing countries represented by 221 sample companies is about 33.5 percent with maximum ownership of 100 percent. In fact the data shows that government has ownership control of more than 50 percent of the sample companies.

Agrawal and Knoeber (2001) suggest that the appointment of government officers or representatives as board members would provide companies with the knowledge and procedures to predict government actions or policies. These companies pay lower taxes, acquire government contracts and/or enjoy reduced regulatory requirements (Agrawal and Knoeber, 2001; Faccio, 2006). In fact, compared to non-PCCs (Faccio, 2006), PCCs are almost three times likely to receive government assistance in the form of direct cash payments, purchases of newly issued debt or equity, government-subsidized loans, loan guarantees, and tax relief tied to a bailout and government purchases of company assets. Therefore, PCCs especially in the utility industry tend to enjoy monopolistic position and have greater market power. Boubakri et al. (2012) discuss thoroughly the benefits and costs incurred on companies associated with the government.

The helping hand theory argues that government participants in business entities help to generate positive outcomes on the company’s financial performance. The theory argues for massive state intervention to help market work better in achieving the social good. This help typically takes in the form of taxing those activities that are perceived to be oversupplied and subsidizing those activities that are undersupplied (Chang and Wong, 2004).

While political connections can increase the company’s performance and value, arguments from the corporate governance literature suggest that agency cost and governance issues may plague PCCs, leading to value-decreasing rent-seeking activities. This is in line with arguments advocated by the grabbing hand theory that suggests government participation would demand companies to serve political and social obligation (e.g. to increase employment and corporate social responsibility spending), which in turn could have a negative impact on the company’s financial performance (Chang and Wong, 2004). Moreover, the representation of government representative at the highest decision level of a company (BOD level) also means that the company has to adhere to government requirements, accept members of board that are less competent and employ excess labor input. Therefore, the performance of PCCs might be lower than performance of non-PCCs.

There are many studies that examine the relationship between PCCs and performance. However, the findings are of mixed results. In Singapore, PCCs performed better than non-PCCs (Ang and Ding, 2006; Feng et al., 2004; Ramirez and Tan, 2004). In Malaysia, the performance of PCCs is no different from non-PCCs (Jaffar et al., 2010). In fact, PCCs in Malaysia are considered to have higher risk of financial misrepresentation and pay higher audit fees (Gul, 2006). The empirical findings in China are contradicting. Fan et al. (2007) found that non-PCCs performed better than PCCs, while Wu et al. (2012) found that PCCs are financially better compared to non-PCCs.

PCCs are also less likely to invest and innovate (Desai and Olofsgard, 2008), have lower economic growth (La Porta et al., 2002), poorer quality of accounting information (Chaney et al., 2011) as they practice lenient accounting standards especially in the disclosure practice (Li and Filer, 2007). Furthermore, information disclosed in the annual reports of PCCs tends to be unreliable since the accounting and auditing professionals are less independent (Li and Filer, 2007).

Additionally, the expropriation activities of controlling owners through self-dealing and tunneling are more evident in PCCs compared to non-PCCs as political connections help companies secure bank financing and evade capital market punishment (Qian et al., 2011). Managers frequently use lower quality accounting information to conceal expropriation activities and delay efficient monitoring (El Ghoul et al., 2013). Studies by Chen et al. (2010) have shown that analyst forecasts are less accurate for PCCs than for non-PCCs, implying that political connections aggravate information asymmetry problems. Therefore, this study agrees with the grabbing effect theory that proposes the negative effect of government connection toward companies’ performance. Hence the hypothesis is stated as follows:

H1.

There is a negative association between the level of government connection and companies’ financial performance.

Past studies show that corporate governance mechanisms such as board financial literacy and diversity, number of board meetings and CEO duality influence companies’ financial performance (Johari et al., 2008). This study proposes that good corporate governance mechanism, as measured by the percentage of independent director, can mitigate the grabbing effect problem associated with PCCs and eventually will improve the company’s performance (McKinsey and Company, 2002). The independent board was chosen as a variable of interest because the issue of board effectiveness that specifically associated with independent board members has received considerable and continuous interest from the community and regulatory bodies. The latest revised version of MCCG 2012, for example, reemphasizes its importance and eventually redefines the definition of board independence. Accordingly, independent board members are members whose designation as “independent directors” should not exceed more than nine years, cumulatively.

In a one-tier board system, boards of directors are the highest governing body that sets company’s policies and makes important strategic decisions. The board represents shareholders and other stakeholders to monitor and influence decisions and policies made by top management which have to be in line with shareholders’ expectations (Dunn and Sainty, 2009). The MCCG (2001) recommends a right balance between independent non-executive directors[4] and executive directors. This is to ensure the establishment of an effective board in the company. Previous studies provide evidence that independent board members contribute to the effectiveness of the board which lead to the improvement of overall company performance. Furthermore, the board is expected to exercise their role and duties effectively (Pearce and Zahra, 2007; McKinsey and Company, 2002).

According to both agency theory and resource dependence theory (Fama and Jensen, 1983) the larger the number of independence non-executive directors on the board, the better they can fulfill their role in monitoring and controlling the opportunistic behavior of the executive directors (Jensen and Meckling, 1976) and over consumption of perquisites (Brickley and James, 1987). In a way, the existence of independent directors are to provide a check and balance mechanism in enhancing the board’s effectiveness as external directors are also considered as decision experts (Fama and Jensen, 1983).

Empirical evidence varies on the association between independent directors and the company performance. Studies have found that having more external independent directors on the board improves performance (Daily and Dalton, 1994; Siagian et al., 2013), while other studies states otherwise (Hermalin and Weisbach, 1991). The point that can be made from these studies is that there are no clear benefits to the company’s performance provided by independent directors. Petra (2005) argues that the mixed results may be reflective of a corporate culture whereas corporate boards are controlled by management and the presence of independent directors has no obvious impact on management decisions. However, other empirical evidence does suggest that independent directors do play an important role of being a shareholders advocate. For example, studies have shown that shareholders benefit more when independent directors have control of the board in tender offers for bidders (Byrd and Hickman, 1992) and in hostile take-over threats (Gibbs, 1993). Furthermore, Beasley (1996) found that independent directors reduce the likelihood of financial statement fraud. These studies indicate that independent directors do monitor and control management and this could lead to better company performance. Therefore, this study propose that the composition of independent board members would mitigate the negative effect of government connection toward companies’ performance. The hypothesis is stated as follows:

H2.

The composition of independent board members reduces the negative association between the level of government connection and companies’ financial performance.

Another variable that can influence the company’s performance is the institutional shareholders. Institutional shareholders hold a substantially ownership in invested company and have less ability to dispose their share quickly without affecting share prices (Gillan and Starks, 2000). Their investment horizon is typically a long-term investment. The institutional investors such as pension funds and insurance companies have an obligation to provide reasonable rate of returns to their beneficiaries.

Jensen and Meckling (1976) argue that substantial shareholdings by institutional investors provide greater incentives for them to monitor top management as their wealth is tied to the company’s financial performance. They further propose that when there is diffusion in ownership, the agency problem is higher due to potential of conflicts between the principal and the agent. Institutional shareholders can mitigate the agency problems through their involvement in monitoring or controlling activities that potentially lead to these problems. Additionally, institutional owners are arguably having an informational advantage over the average investor because of their superior research and analytical skills. Abdul Wahab et al. (2007) has documented evidence that show institutional investors in Malaysia such as Kumpulan Wang Simpanan Pekerja[5], Lembaga Urusan Tabung Haji (LUTH)[6] and Lembaga Tabung Angkatan Tentera[7] often play active monitoring roles. Furthermore, Bethel et al. (1998) found that the performance of a company improves following an acquisition of block of shares by activist investors. Therefore, this study proposes that the level of institutional investor would mitigate the negative relationship between government connection and companies’ performance. The hypothesis is stated as follows:

H3.

The level of institutional shareholders ownership reduces the negative association between the level of government connection and companies’ financial performance.

2.1 Measurement of variables

The data collected comprises of four types of variables: dependent, independent variable of interest, moderating and control variables. Proxy for the dependent variable, which is firm performance (PERFORM) is based on two measurements. First, for the internal performance of the company, we used return on assets (ROA) measured based on net profit after tax divided by total assets. ROA is frequently employed in past studies to measure firm’s financial performance (see e.g. Wu et al., 2012). Another proxy for PERFORM is on the external performance of the company, that is, Tobin’s Q (TQ). TQ is measured based on market value of firm’s equity over book value of firm’s equity (Feng-Li and Tsangyao, 2010).

In this study, the independent variable of interest is the type of company, i.e. whether the company is of PCCs or non-PCCs. To determine whether a company is of PCCs or otherwise, we track the composition of board members over a period of four years by examining the company’s annual report. We consider a company as PCCs if:

1. at least one member of its BODs is or was a politician, that is, a member of state or federal parliament, a minister or any other top appointed-bureaucrat; and

2. at least one member of its BODs is or was a current prominent government officer.

The definition of PCCs companies takes into account the standard definitions used in the literature. For example, Fan et al. (2008) define a Chinese company as being politically connected if the CEO is a current or former officer of the central government, local government or the military. Faccio (2006) defines a company as politically connected if at least one of the company’s largest shareholders or one of its top officers is a member of parliament, a minister, a head of state or closely related to a top government official.

In this study we utilized two moderating variables. The first moderating variable is the interaction between composition of independent directors (IND) being one of the corporate governance characteristics and PCC (IND×PCC). The second moderating variable is the interaction between percentage of institutional investors (IO), being another corporate governance characteristic and PCC (IO×PCC). The first corporate governance characteristic, that is, IND, is calculated based on the percentage of independent directors compared to total number of directors (Abdul Wahab et al., 2007). The second corporate governance characteristic, that is, IO, is calculated based on total percentage of IO having more than 5 percent shareholdings (Eng and Mak, 2003). Our measurement is consistent with several other Malaysian studies (see Saleh et al., 2010; Wong et al., 2009) as well as an Indonesian study (Fauzi et al., 2007).

There are studies that utilized measurement for IO to focus only on the top five institutional investors in a firm (see Abdul Wahab et al., 2007, 2008, 2009). It is well-known that the top five institutional investors would include giant institutions such as the Employees Provident Fund, Lembaga Tabung Haji and Permodalan Nasional Berhad (PNB) (Saleh et al., 2010). However, we utilized a more general measurement so as not to be bias toward the giant institutions only since Malaysian firms are generally owned by multiple types of institutions including insurance companies, pension funds, investment trusts, financial institutions and investment companies (Wong et al., 2009).

In this study, we include seven control variables, namely, leverage (LEV), company size (SIZE), auditor quality (AUDQ), foreign ownership (FOWN), firm’s age (AGE), family firm (FAMILY) and government-owned firms (GOV). Leverage (LEV) is measured based on total long-term debts divided by total assets (Che Haat et al., 2008). It is expected that agency costs are higher for companies that have more debt in their capital structure (Jensen and Meckling, 1976). This is because potential wealth transfers from shareholders and managers to debt holders could rise with the increase in gearing (Meek et al., 1995). Therefore, LEV is expected to have a negative relationship with firm performance (PERFORM).

The company size (SIZE) is calculated based on log of total assets (Che Haat et al., 2008). Larger companies are expected to have more resources that can be utilized to generate more income. This study expects SIZE to have a positive relationship with PERFORM. Audit quality (AUDQ) is measured based on dummy 1 if the firm is audited by Big 4 audit firm and 0 otherwise (Gul et al., 2002). AUDQ is expected to have a positive association with PERFORM. Foreign ownership (FOWN) is measured based on the percentage of company’s share owned by foreign investors. It is expected that foreign investors could be more likely to impose better monitoring mechanism and invest in a company that can give good investment return (Gurbuz and Aybars, 2010). Therefore, FOWN is expected to have a positive relationship with PERFORM. Firm’s age (AGE) is measured based on the number of years since firm’s incorporation and is expected to have a positive association with PERFORM (Tam and Tan, 2007).

It is common among firms in many parts of the world including Malaysia to be controlled by a group of family members (Maury, 2006). Therefore apart from foreign ownership, firms controlled by family members or owned by government could also affect PCC’s performance. Prior studies found mixed results on the association between firm being controlled by a family (FAMILY) and firm performance (Maury, 2006). A firm controlled by a family might not have the same effect on performance in PCC firms compared to non-PCC firms since family firm probably depend less on politics to survive compared to non-family firm. FAMILY is measured based on dummy 1 if the firm is a family firm and 0 otherwise (Jaafar et al., 2012).

Family firms in our sample are based on the information where the shareholders are designated by more than 1 named individual or families. The idea behind this is that family members would probably exert their voting power together (Bureau van Dijk, 2014). Just as we utilized a dummy variable to represent PCC, we also utilized a dummy variable for our measurement of family firm.

Prior studies also found mixed results on the association between government-owned (GOV) firms and firm performance (Feng et al., 2004). The association could be positive if government owners assist firms in securing contracts or provide financial help. The association could be negative if government owners take advantage of firms’ resources for personal wealth. GOV is measured based on percentage of shareholdings in the firms owned by government agencies (Gul, 1999). In this study, we include government ownership (GOV) to act as a control to PCC. Many prior studies consistently found government ownership to have a significant association with firm performance in Malaysia (Lau and Tong, 2008; Taufil-Mohd et al., 2013; Zakaria et al., 2014). Therefore we expect by including GOV could actually allow PCC to represent the actual political influence of the government ownership rather than becoming a proxy on the mere influence of government ownership alone.

2.2 Research model

This study uses the following regression model to test the relationship of independent and moderating variables with the company’s performance:

PERFORM i t = α i t + β 1 PCC i t + β 2 IND i t + β 3 IO i t + β 4 IND × PCC i t + β 5 IO × PCC i t + β 6 LEV i t + β 7 SIZE i t + β 8 AUDQ+ β 9 FOWN i t + β 10 AGE i t + β 11 FAMILY i t + β 12 GOV i t + ε i t

where PERFORM_it: ROA is net profit after tax divided by total assets of firm i at time t, TQ is market value of equity divided by book value of equity of firm i at time t; PCC1_it the number of BOD members who are also members of a political party of the ruling government divided by total BOD members of firm i at time t; PCC2_it the number of BOD members who are also government officers of the ruling government divided by total BOD members of firm i at time t; IND_it the independent directors divided by total BOD members of firm i at time t; IO_it the total percentage of share own by institutional investors who own more than 5 percent shares for firm i at time t; IND×PCC_it the IND multiply by PCC (dummy 1 if PCC, 0 otherwise) of firm i at time t; IO×PCC_it the IO multiply by PCC (dummy 1 if PCC, 0 otherwise) of firm i at time t; LEV_it the total debt over total assets of firm i at time t; SIZE_it the log total assets of firm i at time t; AUDQ_it is the dummy 1 if auditor is Big 4 audit firm, 0 otherwise for firm i at time t; FOWN_it the total percentage of shares own by foreign investors in firm i at time t; AGE_it the number of years since firm’s incorporation for firm i at time t; FAMILY_it the dummy 1 if it is a family firm, 0 otherwise for firm i at time t; GOV_it the total percentage of share own by government and its’ agencies for firm i at time t.

3. Results

Table I reports the descriptive information of sample companies[8]. The average ROA of sample companies from 2004 to 2007 is about 3.36 percent. The average board size is about eight members with maximum number of 16. In general, Malaysian companies complied with the Code of Corporate Governance (SC Malaysia, 2007, 2012) requirement for independent directors of 30 percent where the average percentage of independent directors (IND) is about 41 percent. However, there are cases where companies do not fulfill this requirement when the minimum percentage of independent directors is only about 13 percent.

Table I also shows that on the average, companies in Malaysia appoint about three government officers to be board members. This method of government connection with business entity is more common compare to the appointment of politicians as board members (the number is less than 1). IO is quite high, with an average of about 48.50 percent, government ownership on average about 9.50 percent for our sample and foreign ownership is around 2.50 percent.

This study used multiple regression analysis to test the research model. Several assumptions in the regression analysis have been tested to ensure that data fulfill the normality and homoscedasticity assumption. Additionally, based on the Pearson bivariate correlation among independent variables, it shows that data do not have multicollinearity problems. The detail of the correlation information is presented in Table II.

Table II shows that even though IO has correlation with government ownership (GOV), however the correlation is low, at 26.60 percent (i.e. at 0.266) which suggest there is no issue of multicollinearity among the two variables in our sample. Similarly, in the case of family firm (FAMILY) and IO. Even though there is correlation, at 4.3 percent, however the correlation is very low. These findings suggest the link between IO and GOV as well as between IO and FAMILY is not a problem. Table II also shows that PCC1 (BOD members also being members of the political party of the ruling government) and PCC2 (BOD members also being government officers) are only correlated with GOV up to an accepted level (at 3.1 and 39.5 percent, respectively, i.e. less than the multicollinearity level of 80 percent). Absence of multicollinearity between PCC1 and PCC2 with GOV might be because political party members or government officers representing the government on the BOD of firms were not assigned matching to the level of the government ownership (GOV) but based on other considerations.

Table III presents the multiple regression results for our research model. The third column present results on the relationship between PCC1 and moderating variables (IND×PCC1 and IO×PCC1) with PERFORM based on ROA in the presence of control variables. The last column present results on the relationship between PCC1 and moderating variables with PERFORM based on TQ in the presence of control variables. In Table III, PCC1 is measured based on number of BOD members also being members of a political party of the ruling government.

The results show that the existence of political figures as a member of board (PCC1) has consistently show negative relationships with performance. This relationship is significant in both models with PERFORM measured based on ROA (β=−15.852 significant at 1 percent) and TQ (β=−3.428 significant at 5 percent). The result is consistent with the grabbing hand theory that argues that the connection of companies with government would actually destroy their value when they are expected to serve political and social obligations that could exhaust companies valuable resources in the area which are not economically beneficial. Moreover, if the appointments of politicians as board members are due to political considerations, these considerations would not contribute to the company’s performance especially when members are not competent and resourceful in the board discussions (Fan et al., 2007). Table III findings support our first hypothesis.

Table III results also show that the monitoring role of corporate governance mechanism via the appointment of independent directors (IND×PCC1) have a significant effect on the company’s performance especially as illustrated in the regression result of our model with PERFORM measured based on ROA (β=5.434 significant at 5 percent). This finding partially supports our second hypothesis. This finding also supports the argument for the appointment of independent board members to reflect additional contribution that a person can bring into the board discussion, not only to fulfill the requirements set by a regulator. In Malaysia, the Code of Corporate Governance (SC Malaysia, 2012) has set up a requirement of a minimum of 30 percent composition of independent directors.

However since the finding is not consistently significant, it might indicate that companies were rushing to comply with the regulation without really screening the competency and experience of directors. There is also evidence in past studies that show appointments were based on networking (Maurer and Li, 2006). A dominant personality by the chairman/CEO could be another reason of the inconsistent significant finding. The chairman that is also the CEO (have the duality role) would be too powerful in the boardroom discussion; therefore, the presence of independent directors has less obvious impact on management decisions to improve performance.

The second monitoring mechanism of a company, the existence of institutional investors (IO), show positive association with PERFORM. However, the association is not significant. The results indicate that the existence of institutional investors of PCCs companies has potential but not significant to contribute to company’s performance. As highlighted earlier, institutional investors, such as PNB[9] and LUTH in Malaysia, have an obligation to provide acceptable rates of returns to their beneficiaries. Therefore, this gives them an incentive to monitor activities of top management, since they also have better excess of inside information and financial analysis skills. Jensen and Meckling (1976) argue the existence of institutional investors reduces agency costs and increase the company’s performance.

In Table III, all control variables are shown to influence the company’s performance at different degree of significance. Leverage (LEV) show a negative and significant influence toward company performance (PERFORM) based on ROA but positive significant for PERFORM based on TQ. According to the agency theory, companies that have more debt in their capital structure will make the agency costs between shareholders and debt holders higher (Jensen and Meckling, 1976), and this can have an impact to the company’s performance. On the other hand, debt holders can become a monitoring agent and hence reflect a positive action on the part of the company toward investors and other stakeholders as suggested in the association between LEV and TQ (Olokoyo, 2013).

The company size (SIZE) measured based on log of total assets has a positive relationship with ROA but not TQ. Bigger companies are expected to have more resources that can be invested to generate more income, hence the positive association with ROA. However as companies become too large, investors might be skeptical on whether the companies can really manage their performance well, hence the negative association with TQ.

Foreign ownership (FOWN) as measured by total percentage of share owned by foreign investors of company’s share is expected to provide better monitoring and investing mechanisms in a company that can give good investment return. However, the consistent negative relationship between FOWN and PERFORM suggests that in our sample companies, foreign share ownership do not have any significant influence on company performance. This could be due to the small percentages of ownership by foreign investors (about 2.50 percent as reported in Table I) among Malaysia public listed companies.

Table IV present findings on analysis carried out by replacing the independent variable; percentage of politicians as members of board (PCC1), with percentage of government officers as members of board (PCC2). In general, the results for PCC2 (β=−0.028 significant at 5 percent) are quite similar to the results for PCC1 shown in Table III. Company’s connection with the government (PCC2 measured by the appointment of government officers as board members) contributed negatively to the company’s performance. The results reinforce the grabbing hand theory proposition that the government used private business to achieve social goals (such as employment motive) other than economic goals.

In Table IV, the monitoring role of corporate governance variable (independence of director and IO) also reinforce Table III findings on independent directors (IND) but highlight possibility of a different issue in the case of IO. When we moderate independent directors with PCC2 (IND×PCC2), Table IV (see the third column) shows that there is tendency independent directors are doing their job in monitoring PCC performance (β=5.417 significant at 1 percent). This finding strengthen our conclusion from findings in Table III, when we moderate independent directors with PCC1 (IND×PCC1).

As discussed in our theoretical framework section, both agency theory and resource dependence theory (Fama and Jensen, 1983) suggest that the existence of more independence non-executive directors on the board could enhance the board role in monitoring and controlling the opportunistic behavior of executive directors (Jensen and Meckling, 1976). At the same time more independent directors could monitor management opportunistic behavior on the issue of wealth transfer (Brickley and James, 1987) hence providing a check and balance mechanism to enhance board’s effectiveness (Fama and Jensen, 1983).

However, the second monitoring mechanism, which is the existence of institutional investors (IO), show tendency of a different nature, not strengthening but weakening the company’s performance. When we moderate institutional investors (IO) with PCC2 (IO×PCC2), Table IV shows that there is tendency institutional investors (IO) in the presence of many government officers on the board are weak in monitoring PCC performance (β=−0.037 significant at 1 percent). This finding does not support H3. This could be due to the possibility that these government officers are in fact individuals representing the IO entities themselves. Hence the presence of institutional investors (IO) in such PCCs is not a good monitoring mechanism for the PCCs performance but in reality reflecting a conflict of interest issue with regards to PCC performance.

4. Conclusion

There are two different perspectives on the effects of government connections toward companies’ performance. The helping hand theory proposes that government participant in business entities help to generate positive financial outcomes. On the other hand, the grabbing hand theory suggests that government participation would demand companies to serve political and social obligation, which in turn could have a negative impact on the company’s financial performance.

The findings of this study show that the companies’ connections with the government contribute negatively to performance, indicating that connections is generally detrimental to companies’ value and shareholders’ wealth maximization goal might not always be the aim of the management. However, this study found that companies monitoring mechanism measured by the independence of board members moderate the negative effect of government involvement and subsequently strengthen companies’ financial performance. This indicates that by having independent directors on the board, the presence of politicians on the board can be monitored, so as to increase possible professional contribution as well as reduce possible political considerations among the politicians on board, eventually enhancing company’s strategic direction and performance.

For the sample in this study, we find that IO is not a really effective in their monitoring role to influence companies’ financial performance among politically PCCs, especially in the case of many government officers on the board. Even though institutional investors can be serious investors, their presence would not be effective, if the board comprises of individuals who are representing the ruling government or politicians. In this case, the grabbing hand theory seems to come into play more seriously than the helping hand theory.

The findings of this study provide an additional understanding of the consequence of government intervention on the company’s performance in Malaysia especially during the time period of our study. This study also highlights the role of monitoring mechanism (independence board members, and IO) in strengthening or weakening company performance. Our findings suggest a possibility that by having at least 30 percent independent board members composition might provide a starting point to improve monitoring and consequently improve financial performance of companies in Malaysia.

In Malaysia, companies rely heavily on business and social network in the appointment of top management and awarding of business contracts (Maurer and Li, 2006). Therefore proper appointment criteria for board members should be considered to ensure better corporate governance structure. In this line, the formation of the nomination committee suggested by the current MCCG (SC Malaysia, 2007, 2012) play an important contribution to ensure candidates nominated to be board members have proper credentials and qualifications to carry out responsibilities as board members. Our findings are also beneficial to the Minority Shareholder Watchdog Group who are seriously monitoring the activities of listed companies to ensure the benefits of majority and minority shareholders in politically PCCs are not jeopardized.

Table I

Descriptive statistics

General	Mean	Median	Min.	Max.	SD
BOD	7.69	8.00	3.00	16.00	1.91
No. IND	3.14	3.00	1.00	8.00	0.99
% IND	41.40	40.00	12.50	83.33	10.68
PCC1 (no.)	0.19	0.00	0.00	3.00	0.48
PCC1 (%)	3.00	0.00	0.00	50.00	0.03
PCC2 (no.)	2.51	2.00	0.00	11.00	1.85
PCC2 (%)	32.00	28.57	0.00	100.00	0.21
Variables
ROA	3.36	3.68	−29.75	18.84	7.54
TQ	1.23	0.84	−35.65	34.05	2.12
PCC1	3.00	0.00	0.00	50.00	0.07
PCC2	32.00	28.57	0.00	100.00	0.21
IND	0.41	0.40	0.13	0.83	0.11
IO	48.52	50.83	0.00	99.16	18.86
LEV	0.44	0.42	0.03	1.08	0.23
SIZE	13.27	13.09	7.26	19.36	1.38
AUDQ	0.71	1.00	0.00	1.00	0.45
FOWN	2.55	1.18	0.00	9.93	2.88
AGE	25.68	23.00	1.00	102.00	18.86
FAMILY	0.04	0.00	0.00	1.00	0.19
GOV	9.42	3.57	0.00	93.93	15.90

Notes: General information: BOD, actual number of members on board of directors; No. IND, actual number of independent members of BOD; % IND, number of independent members of BOD divided by total members of BOD; PCC1 (no.), actual number of BOD members who at the same time members of a political party of the ruling government; PCC1 (%), number of BOD members who at the same time members of a political party of the ruling government divided by total members of BOD; PCC2 (no.), actual number of BOD members who at the same time government officers; PCC2 (%), number of BOD members who at the same time government officers divided by total BOD members. Variables: ROA, net income after tax divided by total assets; TQ (Tobin’s Q), market value of equity over book value of equity; PCC1, number of BOD members who at the same time members of a political party of the ruling government divided by total BOD members; PCC2, number of BOD members who at the same time government officers divided by total BOD members; IND, number of independent board members divided by total BOD members; IO, share ownership by institutional investors; LEV, total debt divided by total assets; SIZE, logarithm of total assets; AUDQ, dummy 1 if auditor is Big 4 audit firm, 0 otherwise; FOWN, share ownership by foreign investors; AGE, number of years since firm’s incorporation; FAMILY, dummy 1 if it is a family firm, 0 otherwise; GOV, share ownership by government entities

Table II

Pearson correlation for data involving PCC1 and PCC2

	1	2	3	4	5	6	7	8	9	10	11	12	13
Panel A: Pearson correlation for data involving PCC1 (BOD members also political person of ruling government)
1. ROA
2. TQ	0.202***
3. PCC1	−0.059***	−0.016
4. IND	−0.091***	0.029	0.019
5. IO	0.200***	0.066***	−0.048**	−0.055***
6. IND×PCC1	−0.060***	0.002	0.889***	0.130***	−0.054**
7. IO×PCC1	−0.004	0.011	0.819***	0.001	0.111***	0.872***
8. LEV	−0.370***	0.024	0.054**	0.090***	−0.138***	0.073***	−0.005
9. SIZE	0.040*	−0.001	−0.001	0.102***	0.050**	0.007	−0.003	0.361***
10. AUDQ	0.023	0.056***	−0.053**	0.020	0.130***	−0.039*	−0.037	0.008	0.193***
11. FOWN	−0.057***	−0.053**	−0.036	−0.028	0.000	−0.047**	−0.021	−0.027	−0.179***	−0.024
12. AGE	−0.136***	−0.016	0.034	0.144***	−0.055***	0.072***	−0.011	0.003	0.177***	0.050**	−0.095***
13. FAMILY	0.012	−0.034	0.024	−0.074***	0.043**	0.009	0.050**	−0.027	−0.025	0.018	0.044**	−0.066***
14. GOV	0.037*	0.047**	0.031	−0.001	0.266***	0.008	0.080***	0.047**	0.362***	0.145***	0.002	−0.033	−0.007
Panel B: Pearson correlation for data involving PCC2 (BOD members also government officers)
1. ROA
2. TQ	0.202***
3. PCC2	−0.070***	0.033
4. IND	−0.091***	0.029	0.173***
5. IO	0.200***	0.066***	0.157***	−0.055***
6. IND×PCC2	−0.085***	0.054**	0.771***	0.503***	0.103***
7. IO×PCC2	0.022	0.050**	0.742***	0.096***	0.553***	0.828***
8. LEV	−0.370***	0.024	0.148***	0.090***	−0.138***	0.168***	0.049**
9. SIZE	0.040*	−0.001	0.270***	0.102***	0.050**	0.246***	0.260***	0.361***
10. AUDQ	0.023	0.056***	0.065***	0.020	0.130***	0.070***	0.109***	0.008	0.193***
11. FOWN	−0.057***	−0.053**	−0.005	−0.028	0.000	−0.019	−0.003	−0.027	−0.179***	−0.024
12. AGE	−0.136***	−0.016	0.112***	0.144***	−0.055***	0.155***	0.054**	0.003	0.177***	0.050**	−0.095***
13. FAMILY	0.012	−0.034	−0.051**	−0.074***	0.043**	−0.071***	−0.031	−0.027	−0.025	0.018	0.044**	−0.066***
14. GOV	0.037*	0.047**	0.395***	−0.001	0.266***	0.318***	0.491***	0.047**	0.362***	0.145***	0.002	−0.033	−0.007

Notes: ROA, net income after tax divided by total assets; TQ (Tobin’s Q), market value of equity divided by book value of equity; PCC1, number of BOD members who at the same time members of a political party of the ruling government divided by total BOD members; IND, number of independent board members divided by total BOD members; IO, share ownership by institutional investors; IND×PCC1, IND multiply by PCC1 (where dummy 1 if PCC, 0 otherwise); IO×PCC1, IO multiply by PCC1 (where dummy 1 if PCC, 0 otherwise); LEV, total debt divided by total assets; SIZE, logarithm of total assets; AUDQ, dummy 1 if auditor is Big 4 audit firm, 0 otherwise; FOWN, share ownership by foreign investors; AGE, number of years since firm’s incorporation; FAMILY, dummy 1 if it is a family firm, 0 otherwise; GOV, share ownership by government entities; PCC2, number of BOD members who at the same time government officers divided by total BOD members; IND×PCC2, IND multiply by PCC2 (where dummy 1 if PCC, 0 otherwise; IO×PCC2, IO multiply by PCC2 (where dummy 1 if PCC, 0 otherwise). *,**,***Significant at 10, 5, 1 percent levels, respectively

Table III

Multiple regression results: PCC1 based on BOD members also political person

	Exp. sign	PERFORM (ROA)	PERFORM (TQ)
Intercept		−7.539 (−4.095)***	1.458 (2.629)***
Independent variables
PCC1	−	−15.852 (−2.955)***	−3.428 (−2.123)**
IND	+	−2.911 (−1.872)*	0.974 (2.077)**
IO	+	0.047 (5.049)***	0.005 (1.618)
Moderating variables
IND×PCC1	+	5.434 (1.985)**	0.820 (0.996)
IO×PCC1	+	0.006 (0.319)	0.005 (0.932)
Control variables
LEV	−	−13.797 (−17.910)***	0.575 (2.470)**
SIZE	+	1.381 (9.904)***	−0.098 (−2.320)**
FOWN	+	−0.101 (−1.789)*	−0.042 (−2.467)**
FAMILY	+/−	−0.169 (−0.198)	−0.382 (−1.489)
GOV	+/−	−0.041 (−3.669)***	0.006 (1.908)*
AGE	+	−0.068 (−7.779)***	−0.0003 (−0.113)
AUDQ	+	−0.088 (−0.244)	0.386 (3.560)***
n		1,820	1,819
Adjusted R2		20.08%	1.86%
F-statistics		39.094	3.874
Sig.		0.000	0.000

Notes: PERFORM (ROA), net income after tax divided by total assets; PERFORM (TQ), market value of equity divided by book value of equity; PCC1, number of BOD members also members of political party of the ruling government divided by number of BOD members; IND, number of independent board members divided by total BOD members; IO, share ownership by institutional investors; IND×PCC1, IND multiply by PCC1 (where PCC is dummy 1 if PCC, 0 otherwise); IO×PCC1, IO multiply by PCC1 (where PCC is dummy 1 if PCC, 0 otherwise); LEV, total debt divided by total assets; FOWN, share ownership by foreign investors; SIZE, log of total assets; FAMILY, dummy 1 if family firm, 0 otherwise; GOV, share ownership by government entities; AGE, number of years since incorporation; AUDQ, dummy 1 if audited by Big 4, 0 otherwise. *,**,***Significant at 10, 5, 1 percent levels, respectively

Table IV

Multiple regression results: PCC2 based on BOD members also government officer

	Exp. sign	PERFORM (ROA)	PERFORM (TQ)
Intercept		−6.904 (−3.987)***	1.457 (2.825)***
Independent variables
PCC2	−	−0.028 (−2.438)**	−0.005 (−1.573)
IND	+	−4.543 (−2.632)***	0.294 (0.573)
IO	+	0.068 (6.838)***	0.006 (2.155)**
Moderating variables
IND×PCC2	+	5.417 (3.179)***	0.937 (1.847)*
IO×PCC2	+	−0.037 (−2.849)***	−0.001 (−0.355)
Control variables
LEV	−	−13.596 (−19.137)***	0.573 (2.702)***
SIZE	+	1.350 (10.396)***	−0.081 (−2.095)**
FOWN	+	−0.067 (−1.322)	−0.037 (−2.448)**
FAMILY	+/−	−0.632 (−0.823)	−0.375 (−1.643)
GOV	+/−	−0.020 (−1.874)*	0.007 (2.240)**
AGE	+	−0.067 (−8.522)***	−0.0007 (−0.316)
AUDQ	+	−0.089 (−0.273)	0.362 (3.738)***
n		2,086	2,085
Adjusted R2		20.37%	2.07%
F-statistics		45.443	4.668
Sig.		0.000	0.000

Notes: PERFORM (ROA), net income after tax divided by total assets; PERFORM (TQ), market value of equity divided by book value of equity; PCC2, number of BOD members also government officers divided by number of BOD members; IND, number of independent board members divided by total BOD members; IO, share ownership by institutional investors; IND×PCC2, IND multiply by PCC2 (where PCC is dummy 1 if PCC, 0 otherwise); IO×PCC2, IO multiply by PCC2 (where PCC is dummy 1 if PCC, 0 otherwise); LEV, total debt divided by total assets; FOWN, share ownership by foreign investors; SIZE, log of total assets; FAMILY, dummy 1 if family firm, 0 otherwise; GOV, share ownership by government entities; AGE, number of years since incorporation; AUDQ, dummy 1 if audited by Big 4, 0 otherwise. *,**,***Significant at 10, 5, 1 percent levels, respectviley

Notes

1. National Economic Policy (NEP) was a program implemented by the government following the racial riot in 1969. The purpose of the program is to correct economic imbalance between three main races in the late 1960s. The Bumiputera, who initially owned only 2.4 percent share of the economy, were targeted (under the NEP) to control 30 percent of the economy. Therefore, Bumiputera were giving business privileges including owning state assets, getting government business contracts and opportunity to buy privatized state assets.

2. The National Development Policy replaced the New Economic Policy in 1990 but continued to pursue most of NEP policies of affirmative action is to correct economic imbalance between races in Malaysia. The Bumiputera share of the economy, though substantially larger, was not near the 30 percent target according to government figures. In its review of the NEP, the government found that although income inequality had been reduced, some important targets related to overall Malay corporate ownership had not been met.

3. Bumiputra is a Malaysian term to describe the Malay race and other indigenous peoples of Malaysia.

4. Independent directors refer to directors who are not officers of the company, neither related to officers nor represent concentrated or family holding of its shares. These independent directors in view of company’s board represent the interest of shareholders and are free of any relationship that would interfere with the exercise of independent judgment (Bursa Malaysia Listing Requirements, 2001).

5. Kumpulan Wang Simpanan Pekerja (Malay) also known as Employees’ Provident Fund (EPF) is a Malaysian government agency under the Ministry of Finance. It manages the compulsory savings plan and retirement planning for private and public sector workers in Malaysia. Membership of the EPF is mandatory for Malaysian citizens and voluntary for non-Malaysian citizens.

6. Lembaga Urusan Tabung Haji (Malay, LUTH) was established in 1963 and is the premiere Islamic Financial Institution in Malaysia. LUTH manages more than RM41 billion Islamic funds and invests both domestically and internationally by venturing into several investment sectors in accordance with Islamic principles such as plantation, property development and construction, Islamic finance, information technology, oil and gas, travel services and halal food. Today, LUTH has more than eight million depositors and a network of 119 branches with more than 6,000 touch-points nationwide. LUTH also makes its presence globally by operating an office in Jeddah, Kingdom of Saudi Arabia.

7. Lembaga Tabung Angkatan Tentera (Malay, LTAT) was established in August 1972 by an Act of Parliament (Act 101/1973). The aim is to provide retirement and other benefits to members of the Armed Forces (compulsory contributors) and to enable officers and mobilized members of the volunteer forces in the service to participate in a saving scheme. The fund managed by LTAT pays an annual dividend to its contributors.

8. The data are free from outliers. Outlier data were determined using Mahalanobis distance statistics and were treated by replacing larger (smaller) value data with the next value in the sequence (Tabachnick and Fidell, 2001).

9. Permodalan Nasional Berhad (Malay, PNB) is the biggest fund management company in Malaysia. It was incorporated on March 17, 1978, under the Malaysia Companies Act, 1965, as a wholly owned subsidiary of Yayasan Pelaburan Bumiputra (YPB) (Bumiputra Investment Foundation). The role of PNB is to implement the objectives and policies of the YPB, whose primary function is to receive and administer funds voted by Parliament and from other sources for the purpose of promoting the ownership of share capital by the Bumiputra community in the corporate sector in Malaysia. The YPB is governed by a Board of Trustees Chaired by the Prime Minister of Malaysia. As the operating arm of the YPB, the primary objective of PNB is to evaluate, select and acquire a sound portfolio of shares in companies in Malaysia to promote the ownership of share capital by the Bumiputra community in the corporate sector in Malaysia.