Political Connections and CSR Disclosures in Indonesia

Abstract

This research seeks to provide evidence about how political connections, proxied by government ownership and the existence of politically connected board members, affect the extent of corporate social responsibility (CSR) disclosures in Indonesian listed companies. This research uses the legitimacy theory as a basis for explaining management's motivation for disclosing its CSR. The sample consists of 131 firm-year observations from 38 non-financial public companies that published sustainability reports from 2013 to 2017. We measured the CSR disclosures using a disclosure checklist on the sustainability reports. We subsequently processed the data using a random effect (RE) linear regression. The result shows that CSR disclosures were greater in government-owned companies but lower in companies that have politically connected board members. The results support the legitimacy theory that the government intends to demonstrate legitimate national economic and political conditions by showing that government-owned companies are sustainable. However, CSR disclosures seem to have a substitutive relationship with the existence of politically connected board members, since those political connections may protect the company from public pressure and/or the risk of litigation, reducing the need for CSR disclosures. This research provides evidence that different types of political connections may have different impacts on corporate disclosures.

1. Introduction

In this era of increasing business complexity, corporations are always faced with new reporting challenges (Ernst & Young, 2017). Recently, public awareness about the impact of business on social and environmental conditions has increased (CSR Netherlands, 2016); therefore, CSR emerges as an integral part of businesses today (Porter & Kramer, 2006; Widjaja, 2017), especially in large companies (Sahut, Peris-Ortiz, & Teulon, 2019). CSR activities and reporting also attract the attention of the government (Griffin dan Sun, 2013), indicated by the government’s decision to regulate CSR activities and reporting (Porter and Kramer, 2006).

Regulations in Indonesia relating to CSR reporting contain a great deal of ambiguity which causes unsatisfactory reporting, in terms of quantity and quality. Ernst and Young (2017) found that only 32 out of the 100 biggest companies in Indonesia have published sustainability reports as per the Global Reporting Initiatives (GRI), the globally acknowledged sustainability reporting standard. Looking at the reporting quality, only 16% of the companies disclose balanced information, 34% fairly balanced, and 50% unbalanced.

The unsatisfactory quality of CSR reporting motivates researchers to investigate the factors influencing sustainability reporting in Indonesia. The legitimacy theory has become the most widely used theory to explain the motivation for companies disclosing information about their CSR programs (Muttakin, Mihret, & Khan, 2018). The legitimacy theory is based on the concept of a “social contract,” which suggests that society allows companies to operate and expects that they operate in line with the values held by society (Muttakin, Mihret, & Khan, 2018). CSR is a tool that affects the stakeholders’ perceptions that corporate activities are indeed in line with society’s values, so business continuity may exist (Deegan, Rankin, & Tobin, 2002).

In many cases, past studies conducted using the legitimacy theory’s framework ignored the existence of domination amongst stakeholders (Muttakin, Mihret, & Khan, 2018). Most studies assumed a pluralistic condition, where power was widely distributed among the stakeholders (Gray, Kouhy, & Lavers, 1995). This results in a lack of literature discussing the role of certain stakeholders in CSR decision making. Most CSR literature mentions that shareholders, employees, suppliers, society, and investors are the main stakeholders of companies (Gunawan, 2010). Studies found that governments are especially powerful in developing countries (Dieleman & Widjaja, 2018), and pressure from the government can significantly impact corporate disclosure (Faisal, Situmorang, Achmad, & Prastiwi, 2020; Nguyen, Nguyen, Nguyen, Le, & Nguyen, 2019).

Several studies have investigated the relation between political connections and CSR disclosures. Wong and Hooy (2018) found that there are two proxies for political connections that significantly affect the financial performance of companies. The two proxies are state ownership and politically connected board members. Hung, Kim, and Li (2018) stated that those two proxies for political connections have a similar impact on CSR disclosures. However, Wong and Hooy (2018) argued that different types of political connections may result in different findings. The current research attempts to investigate how those two proxies for political connections affect sustainability reporting in Indonesia.

Studies regarding the relationship between state ownership and CSR disclosures in countries around the globe show varying results. Ghazali (2007) found that state ownership increased the CSR disclosures of Malaysian companies. Makhija and Patton (2004) found that the relationship was insignificant for Czech companies. While in Saudi Arabia (Alotaibi & Hussainey, 2016) and China (Shahab & Ye, 2018) the relationship was found to be negative. There have been no similar studies conducted in Indonesia.

Studies have also used the existence of politically connected board members as a proxy for political connections and these have yielded different results. Marquis and Qian (2014) found that Chinese companies that had at least one of their board members politically connected were more likely to publish sustainability reports. Dicko, Khemakhem, and Zogning (2019) found a positive relationship between the existence of politically connected board members and voluntary Environmental, Social and, Governance (ESG) disclosure in Canada. On the other hand, Muttakin, Mihret, and Khan (2018) found that the existence of politically connected board members decreased the extent of CSR disclosures by Bangladeshi companies.

This research is intended to provide evidence of political connections’ impact on the CSR disclosures of non-financial public companies listed on the Indonesian Stock Exchange from 2013 to 2017, using the legitimacy theory’s framework. This research used a panel data regression and measured CSR’s disclosure using content analysis. The proxies for political connections refer to Wong and Hooy (2018), which are state ownership and the existence of politically connected board members.

There has been no similar research found for Indonesia. The result of this research may differ from those studies conducted in other countries, as the country is the main factor influencing CSR (Zhao, 2012). Economic growth, law enforcement, and culture may lead to different perspectives toward CSR (Cai, Pan, & Statman, 2016).

2. Theoretical Background

2.1. Legitimacy Theory

Legitimacy is the most widely used theory to explain companies’ motivation for disclosing their CSR policies (Muttakin, Mihret, & Khan, 2018). The word legitimacy is defined by Lindblom (1994) as a condition or status when an entity’s value system is consistent with the value system in which the entity operates. Organizations try to align their goals and operations with the social boundaries of the society they operate in, to maintain their positioning in the society, thus forming a social contract between the organization and the society (Deegan & Unerman, 2008).

2.2. Hypotheses

Based on the legitimacy theory, management’s motivation for disclosing its CSR performance is to gain legitimacy from the stakeholders (Deegan, Rankin, & Tobin, 2002). However, power is not distributed evenly among the stakeholders, and certain groups of stakeholders may dominate others (Archel, Husillos, & Spence, 2011). Governments may emerge as the dominant players due to their authority to control resources, especially in developing countries (Dieleman & Widjaja, 2018). Companies try to build legitimacy with their respective governments to get advantages, such as better treatment, less taxation, better chances of winning government contracts, looser supervision, and other advantages (Faccio, 2006).

Several studies find that government ownership has a positive impact on CSR disclosures (Ghazali, 2007; Muttakin and Subramaniam, 2015). State-owned companies have greater political visibility; hence they need to show more disclosure to legitimize their position in society (Ghazali, 2007; Muttakin and Subramaniam, 2015). State ownership can also be defined as public ownership, thus state-owned companies try to increase the extent of their disclosures to fulfill the expectations placed on them (Ghazali, 2007). Based on the above explanation, this research suggests the following hypothesis:

H1: Government ownership has a positive impact on the extent of CSR disclosures

Political connections can also be gained through politically connected members of boards of directors or commissioners. This type of political connection differs from the previous type since private companies may well have this type of political connection. Different types of political connections may lead to different research results (Wong & Hooy, 2018).

Several studies found that the existence of politically connected board members decreases CSR disclosures (Hung, Kim, & Li, 2018; Jahid, Rashid, Hossain, Haryono, & Jatmiko, 2020; Muttakin, Mihret, & Khan, 2018). This is because the legitimacy coming from CSR can be substituted by the advantages a company gets from its politically connected board (Muttakin, Mihret, & Khan, 2018). Political connections facilitate companies by making funding easier to obtain and reducing the risk of litigation, therefore reducing the incentives for public disclosures (Hung, Kim, & Li, 2018). Based on the previous studies, the following hypothesis is developed:

H2: The existence of politically connected board members has a negative impact on the extent of CSR disclosures.

3. Methodology

3.1. Data

This research used secondary data. The sustainability reports were obtained from companies’ official websites, and the financial data were retrieved from the Thomson Reuters Eikon Database.

3.2. Dependent Variable

CSR disclosure was measured using the content analysis method utilizing a checklist formulated by Gunawan (2010). The checklist comprised 45 disclosure indicators divided into six categories, which are the environment, human resources, community involvement, product, sustainability strategy, and others. A score ranging from 1 to 5 was given manually to each disclosure indicator, based on the criteria below:

a. Score of 1: indicator was disclosed in one or two sentences

b. Score of 2: indicator was disclosed in a paragraph (at least three sentences) and less than half an A4 page

c. Score of 3: indicator was disclosed in more than or equal to half an A4 page and less than a full A4 page

d. Score of 4: indicator was disclosed in a full A4 page.

e. Score of 5: indicator was disclosed in more than a full A4 page.

To arrive at a final CSRD score, the total scores from all the indicators were divided by the maximum score of 225 (total of 45 indicators multiplied by the maximum score of 5).

3.3. Independent Variables

Political connections were measured by the state’s ownership (of the company) and the existence of politically connected board members. State ownership was a dummy variable; a score of one indicated that the company was owned by the Government of Indonesia. This measurement referred to Ghazali (2007).

The existence of politically connected board members was also a dummy variable; a score of one was given when an observation has at least one politically connected board member. This definition of politically connected board members referred to previous studies (Faccio, 2006; Dicko, Khemakhem, & Zogning, 2019) and was adjusted for the Indonesian context (Habib and Muhammadi, 2018). A member of either a board of directors or commissioners was considered to be politically connected when he/she fulfilled one of these criteria: (1) is currently affiliated with a political party; and was or is a: (2) minister; (3) member of parliament (Dewan Perwakilan Rakyat or DPR); (4) cabinet secretary; (5) governor or vice-governor. This definition excluded volunteers for political campaigns because they were more difficult to identify (Hung, Kim, & Li, 2018).

There were a few steps taken to determine whether a board member was politically connected. First, the names of board members were gathered from the companies’ annual reports along with their career history. Positions as a minister, member of parliament, cabinet secretary, or governor are generally stated in the career history. To confirm those positions, an Internet search was conducted. The name of the related board member and his/her position were inserted as keywords in search engines. However, affiliations with political parties are generally not included in the boards’ profile in the annual reports. Henceforth, manual checking was done to identify any potential political party affiliations of the board members. In this case, the inserted keywords were the names of the board members and “partai” (meaning party). When a trustworthy source was found, such as a prominent news portal, which verified the related member’s party affiliation, a mark was made against the board member’s name. If at least one politically connected board member was found in a single observation, a score of one was given to that observation. When there was no trustworthy source that confirmed the political position of all the board’s members, the related observation was given a score of zero.

There were four control variables used in the model. Profitability was proxied by return on assets (ROA) which was calculated by dividing income before interest and taxes to total assets. Leverage was proxied by the debt to equity ratio. Firm size was measured by the natural logarithm of total assets. Firm age showed the natural logarithm of the number of years since the establishment of each company.

Hence, we arrived at the two models, where Model 1 below was to test the first hypothesis:

\(\begin{aligned} \operatorname{CSRD}_{\mathrm{it}}=&\alpha+\beta_{1} \mathrm{GOV}_{\mathrm{it}}+\beta_{2} \mathrm{ROA}_{\mathrm{it}}+\beta_{3} \mathrm{LEV}_{\mathrm{it}}+\beta_{4} \mathrm{SIZE}_{\mathrm{it}} \\ &+\beta_{5} \mathrm{AGE}_{\mathrm{it}}+\varepsilon \end{aligned}\)      (1)

And Model 2 was to test the second hypothesis:

\(\begin{aligned} \mathrm{CSRD}_{\mathrm{it}}=& \alpha+\beta_{1} \mathrm{BOARD}_{\mathrm{it}}+\beta_{2} \mathrm{ROA}_{\mathrm{it}}+\beta_{3} \mathrm{LEV}_{\mathrm{it}} \\ &+\beta_{4} \mathrm{SIZE}_{\mathrm{it}}+\beta_{5} \mathrm{AGE}_{\mathrm{in}}+\varepsilon \end{aligned}\)      (2)

Where:

CSRD : CSR disclosures

GOV : government ownership

BOARD : existence of politically connected board members

ROA : return on assets to measure profitability

LEV : debt to equity to measure leverage

SIZE : natural logarithm of firm size

AGE : natural logarithm of firm age

4. Results and Discussion

4.1. Results

The population of this study was non-financial companies listed on the Indonesian Stock Exchange. Companies from the financial industry were excluded due to different reporting requirements. This research used purposive sampling by only including non-financial public companies that have published stand-alone sustainability reports for the fiscal years from 2013 to 2017. The reports are also accessible from company websites or the GRI database. The sampling process can be observed in Table 1.

Table 1: Sampling Process

The sample of 131 observations was distributed between several different industries, as shown in Table 2. The industries were based on industry classification data from the Indonesian Stock Exchange. The table shows that the mining industry had the largest number of observations.

Table 2: Sample Classification Based on Industry Groups

Descriptive statistics of the dependent, independent, and control variables can be viewed in Table 3.

Table 3: Descriptive Statistics

Average scores for CSRD based on category, ranked from highest to lowest, were human resources (69%), sustainability energy and others (62%), community involvement (49%), product (45%), and environment (42%).

Table 4 displays the Pearson correlation of the model or the linear relationship between the variables. The dependent variable of CSRD was found to have a significant linear correlation with government ownership. However, CSRD did not seem to have a significant correlation with the existence of politically connected board members.

Table 4: Correlation Analysis

Note: ***, ** and * indicates significant at 1%, 5% and 10% level of significance based on t-statistics.

The hypotheses testing used a random effect (RE) panel data regression. The overall significance for both of the models (global F test) was 0.0012 and 0.0037, respectively. The goodness of fit (overall R square) for both the models was 27.81% and 20.79%, respectively. The results of the hypotheses testing are shown in Table 5.

Table 5: Hypotheses Testing

Note: ***, ** and * indicates significant at 1%, 5% and 10% level of significance based on t-statistics.​​​​​​​

4.2. Discussion

The hypotheses testing, as shown in Table 5, suggested that government ownership (GOV) had a positive and significant impact on CSR disclosures at a 95% confidence level. On the other hand, the existence of politically connected board members had a negative and significant impact on CSR disclosures at a 90% confidence level. Therefore, both hypotheses were accepted.

The result indicated that companies owned by the Government of Indonesia disclosed more about their CSR programs. This finding is consistent with Ghazali (2007) and Muttakin and Subramaniam (2015). The legitimacy theory suggests that government-owned companies try to align their operations with the economic and political agenda of the government (Ghazali, 2007). The current Government of Indonesia has been trying to display its down-to-earth reputation, by organizing social events such as BUMN Berbagi Ramadan, an event where state-owned enterprises conduct charity during the Muslim holy month. That event was shown by the media, with President Jokowi sharing food and necessities with impoverished neighborhoods. Although the event was held by the Government of Indonesia, it was funded by government-owned companies and is part of their CSR activities.

The Government of Indonesia represents the people, so government-owned companies are owned by the public. This results in a great deal of political visibility for those government-owned companies. In other words, people pay more attention to the operations of government-owned companies (Ghazali, 2007). CSR disclosure is used as a legitimizing tool, to give assurances to society that companies live up to the public’s expectations (Ghazali, 2007; Muttakin & Subramaniam, 2015).

Local regulations may also play a role in the positive association between government ownership and CSR disclosure. Government-owned companies in Indonesia are required, by regulation, to conduct a social program called Program Kemitraan dan Bina Lingkungan, or PKBL for short. This program consists of two sub-programs: (1) providing guidance and support to small and medium enterprises, and (2) aiding natural disasters victims, funding education, health, infrastructure development, and nature conservation. The regulation requires government-owned companies to plan a budget, maintain financial records, and report their PKBL activities to the Government of Indonesia. In addition to the government-owned companies, other public companies may also conduct PKBL activities after receiving approval from the annual general meeting of their shareholders.

This research suggests that the existence of politically connected board members causes a company to disclose less CSR information. This finding is consistent with the previous studies (Hung, Kim, & Li, 2018; Jahid, Rashid, Hossain, Haryono, & Jatmiko, 2020; Muttakin, Mihret, & Khan, 2018). Political connections provide advantages to companies, such as making it easier for them to raise funds, reducing their litigation risk (Hung, Kim, & Li, 2018), and reducing public pressure on them (Muttakin, Mihret, & Khan, 2018). Therefore, if these connections exist, they may undermine the need to disclose those companies’ CSR plans.

This research finds that government ownership and politically connected board members work as opposite forces in influencing CSR disclosures. This finding contrasts with that of Hung, Kim, and Li (2018) who found that those two forces had a similar impact on corporate voluntary disclosures. This research, however, supports Wong and Hooy (2018) that different types of political connections may impact differently on companies’ decisions.

The result suggests that in Indonesia, direct government influence through ownership tends to increase CSR disclosures, as an indication that government-owned companies are sustainable in their operations. However, the existence of politically connected board members in a company, such as former ministers, members of parliament, or governors, surprisingly has the opposite effect. It may be because those politically connected companies use their connections to protect themselves from potential litigation and public pressure.

From the four control variables used in the model, only two variables had a significant influence on the dependent variable, which were leverage and size. Leverage was found to have a negative impact on CSR disclosures, which is consistent with Muttakin and Khan (2014). Companies with a high debt ratio do not disclose much CSR information, to avoid excessive attention from creditors (Gantyowati & Agustine, 2017). Company size had a positive impact on CSR disclosures, which was consistent with most previous findings (Ghazali, 2007; Muttakin & Khan, 2014; Muttakin, Mihret, & Khan 2018). Bigger companies have greater political visibility, and hence need to disclose more about their CSR programs (Ghazali, 2007).

Firm age did not seem to significantly influence CSR disclosure, which was inconsistent with the previous findings (Muttakin & Khan, 2014; Muttakin, Mihret & Khan, 2018). ROA was not found to have a positive impact on CSR disclosures either. This finding differs from some previous findings (Gamerschlag et al., 2011; Muttakin & Khan, 2014; Muttakin, Mihret & Khan, 2018) while it was consistent with Siregar and Bachtiar (2018). The political visibility of companies is mostly affected by their size instead of their profitability.

4.3. Sensitivity Analysis

Two sensitivity analyses were also conducted in this study, to corroborate the findings. The first was to check whether the results would differ when the two independent variables were tested together in one model. This sensitivity analysis showed a similar result to those from the main testing. The second sensitivity analysis was to check whether the percentage of shares owned by the Government of Indonesia mattered. We created a subsample consisting of 48 firm-year observations from 13 companies owned by the Government of Indonesia. The result indicated that the percentage of shares did not have a significant impact on CSR disclosures. It meant that all the government-owned companies obeyed the government and complied with the same regulations, regardless of the percentage of shares owned by the Government of Indonesia.

5. Conclusions

This research aimed at providing evidence of how political connections, proxied by government ownership and the existence of politically connected board members, influence the extent of CSR disclosures in companies’ sustainability reports. The theoretical framework was built within legitimacy notions which see CSR as a tool for companies to legitimize their existence in society. The research sample consisted of 131 firm-year observations from 38 non-financial companies, listed on the Indonesian Stock Exchange, which published stand-alone sustainability reports for the fiscal years from 2013 to 2017.

We analyzed the sustainability reports using the content analysis method. Two independent variables, government ownership and the existence of politically connected board members were tested in separate models. A board member is politically connected if he/she: (1) is affiliated with a political party; and is or was a: (2) minister, (3) member of parliament; (4) cabinet secretary; or (5) governor or vicegovernor. Four control variables are used: return on assets, leverage, company size, and company age.

This research processed the data using a random effect panel data regression. The result indicated that CSR disclosures are more extensive in government-owned companies, but less extensive in companies with politically connected board members. The legitimacy theory suggests that government-owned companies need more CSR disclosure as a legitimizing tool for society, due to their politically sensitive nature. Companies having politically connected board members may have less incentive to disclose information about their CSR programs, since their political connections may protect them from any risk of litigation and public attention.

Sustainability reporting is relatively rare, so any future studies are suggested to also consider using annual and integrated reports. The quality of the reporting may also be called into focus by any future studies. Content analysis may also be subject to the researchers’ subjectivity.