Determinants of corporate governance compliance: what matters and what does not?

Abstract

Purpose

The objective of this study is to assess the level of corporate governance (CG) compliance and identify determinants of high compliance in Indonesian publicly listed corporations including family and nonfamily firms. The country uses a voluntary disclosure approach to enforce its regulations; thus, it is important to identify the factors affecting compliance.

Design/methodology/approach

Employing a logistic regression model, this paper analyzes the CG index of high-compliance vs. poor-compliance companies and emphasizes factors that contribute to better governance compliance. The CG index of high-compliant firms is almost twice as high as that of low-compliant firms.

Findings

The study explores factors that contribute to high CG in an emerging market like Indonesian corporations. The study's findings indicate that family-owned businesses predominate in the low-compliance group. High-compliance firms are older and larger with higher financial performance, free float and leverage, as well as a positive influence of the founder's great leadership. The results support theoretical arguments that concentrated ownership and excessive majority shareholder control are key factors in determining the likelihood of good governance practices by firms. Hence, the market and regulators should devise effective strategies to encourage and reward high compliance.

Research limitations/implications

The findings of the research offer several implications for the academic community and policymakers. Improving CG at the firm level is a viable goal, even though the agenda to reform minority investor protection laws and increase judicial quality is challenging and may take a long time to show significant results. Moreover, this study has some limitations that could be addressed in future research. The study focuses on a single-country setting, Indonesia. There are cultural aspects and governance settings that may be unique in the Indonesian context, which may limit the applicability of the findings to other countries with their own cultural settings and institutional legal framework.

Originality/value

The study investigates the factors that influence high governance compliance in specific CG regulations designed for the emerging Indonesian market. The study also discovers evidence that the crisis period has a favorable impact on the firm's decision to comply with governance provisions.

Citation

Tanjung, M. (2023), "Determinants of corporate governance compliance: what matters and what does not?", Journal of Business and Socio-economic Development, Vol. 3 No. 3, pp. 237-252. https://doi.org/10.1108/JBSED-11-2021-0148

1. Introduction

Participants in the private sector, government agencies and scholars are aware that the Asian financial crisis of 1998 altered how companies and boards of directors should be governed. As a result, corporate governance principles have emerged as a critical factor in ensuring a company's survival following the effects of financial distress and crisis. Recently, corporate governance (hereinafter referred to as CG) practices have been integrated into the strategies of worldwide corporations. Since the early 1990s, additional CG provisions have been developed and adopted by many nations, particularly by companies. Since companies have the capacity and discretion to comply with certain CG provisions but not others, adoptions have, by and large, been done voluntarily for at least two decades (Keay, 2014). This concept evolved mostly from the standpoint of stewardship theory (Goto, 2018).

Since the 1998 global financial shock and the 2008 financial crisis, CG principles have become a crucial concern in managing a company's survival in times of financial distress and crisis. These crises have altered how companies and their owners should be governed to minimize the financial difficulty and poor financial performance. However, the literature suggests that CG compliance is not consistently associated with strong financial performance. Why are these results inconclusive? What are some of the potential determinants of effective governance practices? Is there a difference in CG values between family businesses and nonfamily businesses? Furthermore, why is family-businesses compliance lower than that of nonfamily businesses? While this set of questions is intriguing, there are not enough research and empirical findings to address them adequately, particularly in the context of large Indonesian corporations.

As a country in the region that was severely impacted by the crisis, Indonesia urgently needs to properly implement governance provisions. However, the debate regarding CG and its adoption in Indonesia is still in its early stages. The Indonesian government introduced the first version of the country's CG codes in 1999 in response to the Asian foreign exchange crisis, while South East Asian governments such as Malaysia, Thailand, the Philippines and Singapore took parallel moves (Zhuang, 2000). In its early adoption, CG was first intended and dedicated to publicly listed companies traded on the Indonesian Stock Exchange. The Indonesia Financial Service Authority (from now on referred to as OJK) imposes CG for these corporations to govern and monitor the capital market. The OJK enforces CG by drawing on appropriate laws, legislations and generally accepted CG principles, including the Indonesian CG principles.

Owing to data availability and accessibility, Indonesia is relatively under-researched in terms of CG; there are only a few empirical studies that focus on the subject and the concrete implementation of CG within the business sector. Previous studies and reports (e.g. Nam, 2004) concluded that Indonesia has a poorer CG quality than other South East Asian economies, as well as weak legal enforcement and inferior minority investor protection. Given the limited research on this topic in Indonesia, this study aims to identify the determinants of firm-level CG practices in the Indonesian market and thereby contribute to the current debate on how CG mechanisms operate and converge in the unique institutional setting of an emerging market like Indonesia.

By investigating CG compliance in the Indonesian market, this study uncovers several findings. First, the study reveals that family-owned businesses have lower compliance scores than nonfamily firms. High-compliance firms are older and larger companies with higher financial performance, free float and leverage, as well as a positive influence of the founder's great leadership. Additionally, the study finds that: (1) firm size and performance have a large and significant influence on CG practices, and (2) family-owned businesses comply with governance provisions less frequently. Another intriguing conclusion is the evidence that the 2008 crisis period favorably influences a firm's decision to comply with governance standards. As a result, the study adds to the CG literature in at least three ways: (1) the first family generations have a positive significant influence on compliance, (2) the third generations as board members have a negative impact on CG and (3) the higher the percentage of blockholders and family share ownership, the lower the governance compliance.

The rest of this paper is organized as follows. Section 2 presents a literature review and the proposed hypotheses regarding family firms and performance. Section 3 describes the model development, as well as the datasets and the main variables. Section 4 then presents the empirical findings. Finally, Section 5 summarizes the empirical results and concludes the study.

2. Literature review and hypothesis development

2.1 Corporate governance development in Indonesia

Since the late 1990s, when the crisis began, there has been intense debate on CG in Indonesia. Indonesian firms' poor CG has slowed their recovery from the crisis. Numerous incidents of financial default were triggered by the detrimental intervention of controlling owners, who forced managers to extend excessive credit to affiliated groups and subsidiaries. This intervention by the controlling owners, notably the founders, demonstrates poor risk management and a severe conflict of interest.

As a response to the Asian foreign exchange crisis, the Indonesian government created the first version of the country's CG codes in 1999, while South East Asian governments such as Malaysia, Thailand, the Philippines and Singapore took parallel decisions (Zhuang, 2000). CG was initially established and dedicated to publicly listed companies traded on the Indonesian Stock Exchange (from now on referred to as IDX). The OJK enforces CG for these corporations, to regulate and monitor the capital market. The CG enforced by the OJK is drawn upon relevant laws, legislations, and generally accepted principles of CG including the Indonesian good CG principles.

Companies have encountered several challenges as a result of weak internal governance arrangements, including agency problems. These agency problems occur more frequently during a crisis as a result of weak CG (Kirkpatrick, 2009). For instance, the US government bailed out numerous bankrupt companies, however, these businesses continued to offer their executives and senior managers massive bonuses (Dowd, 2009). Although the provision of incentives and bonuses is appropriate since it complies with the CG structure, it has caused agency problems given that companies were experiencing a downturn. Thus, the US government, as the owner of funds, has decided against continuing with this policy [1]. This problem could arise in other countries, including Indonesia. As a result, the Indonesian FSA issued Rule No. 34/POJK.04/2014 establishing a remuneration committee for publicly listed companies.

Indonesia is relatively under-researched in this topic, with only a few empirical works focusing on the specific subject of CG compliance and practical implementation within the corporate community. Although some studies have detected CG development in Indonesia, the majority of studies and reports have focused mainly on the relationship between CG adoption and its implications on firms' financial performance (Wahyudin and Solikhah, 2017). Another study (Utama et al., 2017) examined the link between CG and the largest shareholders but did not investigate the various generations of corporate ownership. Given the scarcity of empirical research on this topic, the findings of this study should contribute to our understanding of how CG mechanisms operate and converge in the unique institutional setting of Indonesia where family businesses are dominant.

The current study fills the gap in the Indonesian literature by thoroughly investigating the role and influence of business founders and their families on the implementation of CG practices. Furthermore, the study also seeks to determine whether family-owned companies have distinct institutional characteristics in terms of CG compliance quality than widely held firms, as well as whether corporations comply with CG more during a financial crisis. As a result, this study aims to address gaps in the scant CG literature on Indonesia in terms of the characteristics that drive high governance compliance and whether the financial crisis affects CG adoption at the firm level. The key elements of this study are further discussed in the following sub-sections.

2.2 Determinants of CG compliance

Theoretically, CG consists of two mechanisms: external and internal (Young et al., 2008). The external governance mechanisms are associated with various external aspects of a company, which are controlled and directed by the government, debt and capital markets. The internal mechanisms, which typically consist of board structures, ownership and the company's management style, are controlled by shareholders and boards. This mechanism is concerned not only with management's efficacy in achieving corporate objectives but also plays an important role in strategic investment decision-making. Following earlier research (Tanjung, 2020), this study constructs a comprehensive CG index by identifying 15 governance elements or variables that are considered to be significant determinants of high compliance. These variables are grouped into four clusters: (1) firm type and ownership structure, (2) family control, (3) financial characteristics and (4) financial crisis.

1. Firm type and ownership structure

The study's framework is based on two CG theories: agency theory and stakeholder theory. Agency theory was developed from the original work of Jensen and Meckling (1976). This theory has been used to explain various issues surrounding the separation of corporate ownership, control and management. In 1932, Berle and Means debated the impact of the separation of ownership and management Stigler and Friedland (1983). The literature states that agency cost is the result of a contract formed by the owners of the firm (the principals) who hire outsiders (the agent) to perform services for the firm on behalf of the owner, a contract that includes a delegation and power-sharing structure in the firm's decision-making. This contract was designed as a result of the owners' intention to promote corporate values through the delegation of authority to managers. Stakeholder theory, on the other hand, states that corporate actors have their duties not only to maximize the profits of the stockholders but also to identify and align all stakeholders' interests. Stakeholder theory highlights the urgency of integrating moral and ethical principles into a firm's economic activity. In accordance with this theory, management should promote the welfare of both internal and external stakeholders (Freeman et al., 2004).

Ownership concentration among main stakeholders plays an important role in achieving a firm's operational and strategic goals. Such owners have better access to company data and the authority to fire underperforming managers and hire new agents to maximize shareholders' wealth. The presence of large shareholders and concentrated ownership has both benefits and drawbacks. Management control is effective when ownership is concentrated, and blockholders may exert influence on managers and promote resource allocation that maximizes value (Shleifer and Vishny, 1986). Previous studies have also found that dominating family ownership, particularly when combined with stronger outside boards and management can help reduce potential principal–agent conflict (Tanjung, 2019). Other countries especially in East Asia, such as Korea, have experienced rapid economic development as a result of financial power and contributions from large family corporations (Kim, 2017; Sung and Kim, 2017). Nonetheless, previous studies reported a negative impact of concentrated ownership on financial performance and agency costs.

The ownership structure is determined by retrieving shareholder information and disclosure in annual reports and financial statements. By using these data, the controlling shareholder(s) of the firm is identified with stakes or voting rights they have. In this study, controlling shareholders are classified as either family or nonfamily firms. High family ownership is expected to be detrimental to the adoption and practice of CG. Family firms are plagued by high agency conflicts (i.e. conflicts between owner-managers and minority shareholders), and the controlling families are reluctant to implement good governance practices, by providing better financial and nonfinancial disclosure.

The expected coefficient signs of family ownership, blockholders and free float ratio are negative, negative and positive, respectively. Build on the above reasons, the following hypotheses are proposed in this study.

H1.

There is a negative association between family share ownership and CG compliance.

H2.

There is a negative association between blockholder ownership and CG compliance.

H3.

There is a positive association between free float ratio and CG compliance.

1. Family control

Following previous governance studies (Isakov and Weisskopf, 2014), family control is defined as a measurement of the active involvement of the controlling family members in the firm's management and boards as the controlling family does not want outsiders to have complete control of the firm. Six dichotomous variables, representing the first, second and third family generations appointed as members of the board of directors or supervisory boards, are used as proxies for active family control.

According to Ricardo et al. (2016), family attachment grows with increasing ownership due to the socio-emotional wealth of family firms. In addition, higher levels of family attachment and control may increase the family's influence over the board of directors. Therefore, as a result of increasing family-related control costs, the adoption of and strong compliance with stricter governance practices may undermine family interests and advantages. The compliance costs for family firms include the appointment of independent directors, CEO duality and increased disclosure (Al Amosh and Khatib, 2021). In light of these arguments, the following hypothesis is proposed.

H4.

There is a negative association between family control and CG compliance.

1. Financial crisis

The Asian financial crisis of 1998 motivated many Asian governments, including the Indonesian government, to introduce governance principles. Two decades have elapsed since the first CG codes were issued in 1999 and a decade has passed since the 2008 global financial crisis. Taking into account the enactment of the codes in Indonesia, this study aims to examine the level of governance during the financial crisis of 2008–2009. The study examines whether firms are more compliant during crisis periods than during noncrisis periods. The author posits that firms are more driven to adhere to CG during times of crisis to sustain business performance and investor confidence. According to Mitton (2002), firms with more accountable accounting disclosures and outside ownership concentration perform better during the 1998 financial shock. According to Bouslah et al. (2018), the financial crisis affects the linkages between corporate risk and social performance. This provides firm-level evidence consistent with the assumption that CG contributes to the understanding of firm performance during a financial crisis. Thus, the study posits the following hypothesis.

H5.

There is a positive association between the 2008 financial crisis and CG compliance.

3. Model development and econometric methodology

3.1 Compliance index data and construction

The study collects ownership data from firms' financial and corporate records. The database of the Indonesia Capital Market Institute (TICMI), corporate annual reports and financial filings and related publicly available reports are the primary sources of data. The ownership structure is identified by retrieving companies' shareholder information and disclosure in annual reports and financial statements. In this study, controlling shareholders are classified as either family or nonfamily firms. The ownership structure is defined as the percentage of shares owned by the family, shareholding and minority shareholders.

The current study expands on prior research that constructed CG indices (Arora and Bodhanwala, 2018; Iqbal et al., 2019; Tanjung, 2020), by taking into account additional CG factors reflecting the compliance of Indonesian enterprises. The CG index (GOVINDEX) in this study is constructed using 15 Indonesian CG elements or attributes. Each sub-index is a binary variable with a value of 1 if the firm complies with the governance concept or meets the condition and 0 otherwise. The sample consists of 135 publicly listed Indonesian companies and the period spans from 2003 to 2013. The selection of the period is based on periods before and after the 2008 financial crisis. The analysis covers a total of 11 years between 2003 and 2013, assuming that the crisis year dummy is one of the key variables. Because CG data for Indonesian firms are scarce and not readily accessible, data are collected from corporate annual reports and financial statements. After excluding firms with incomplete data to obtain balanced panel data and financial-based companies, the final sample results in 1,485 firm-year observations.

According to the study, the CG elements and index (GOVINDEX) are estimated as follows:

(1)Sub-indexi,t=(Dummy equals to 1 or 0, reflecting the compliance of each governance element)

(2)GOVINDEXi,t =∑i=1N(Sub-Indext) / 15

where sub-index_i,t represents governance dummy variables capturing 15 governance provisions: independent director, independent supervisory board, code of ethics, insider trading prohibition, whistle blower system, ultimate beneficiary shareholders disclosures (BO), sanction, corporate social responsibility (CSR), largest shareholders, publicly-held shares ownerships, an audit by top four largest auditing agencies, employee stock ownership program (ESOP), size of the board of directors (BOD), size of the supervisory board (BOC) and anti-corruption and anti-bribery rules. Table 1 presents the quintile scores of the GOVINDEX, while Table 2 presents the list of the variables included in the model. In addition, Tables 3 and 4 show the construction of the CG index as well as descriptive statistics for the datasets.

Using equation (1), this study estimates the GOVINDEX scores for firm i in period t by summing each score from the 15 CG elements (sub-indices) and dividing it by 15 (total number of sub-indices). Each of the 15 sub-indices is a dichotomous variable with a score of 1 or 0 based on the firm's compliance during each year/observation period (equation [2]). This approach applies to all sub-indices from 2003 to 2013. Tables 2 and 3 define the variables and the CG elements included in GOVINDEX.

3.2 Econometric methodology

In line with previous studies, the logistic regression method is used to assess the possible significant correlations between the governance index and its variables (Aren et al., 2014; Correa-Garcia et al., 2020; Odriozola and Baraibar-Diez, 2017). The logistic model has been extensively used in modern financial management and CG literature, particularly when evaluating binary or dichotomous dependent variables (Hoetker, 2007). Hence, the dependent variable in this study is defined as either firm with a high index score or a low score. The probability of a firm having GOVINDEX or sub-index scores greater than the required threshold (y = 1) is written as follows:

(3)P1 =E(y=1)=eβ′x1+eβ′x

According to the model, the likelihood that a firm has a high GOVINDEX score is as follows.

(4)High GOVINDEX Scoreit=α +β1FirmTypeandOwnershipStructuresit+β2FamilyControlit+β3FinancialCharacteristicsit+β4Financial Crisisit+εit

The objectives of this study are divided into two parts. The first is to examine the factors that contribute to high GOVINDEX scores, and the second objective is to explore the determinants of high sub-index scores in family-controlled firms and nonfamily firms. The criterion for high CG compliance has been carefully selected. More specifically, a high score requires an index score that is equal to or above the fourth quartile of governance index scores (Jo and Kim, 2008), which is 0.462, as shown in Table 1. Table 3 provides a statistical summary of the GOVINDEX and its components (sub-indices).

4. Empirical findings

4.1 Descriptive statistics

Table 4 summarizes the descriptive statistics for the whole firm data sets, including firms with high and low GOVINDEX. High governance is represented by an aggregate score of 0.462 or higher (GOVINDEX ≥0.462). Three crucial findings can be derived from Table 4. First, the average GOVINDEX score of high-compliance firms is nearly twice that of low-compliance firms (0.553 vs. 0.277). Second, family-controlled firms account for the majority of the low-compliance group (approximately 68% of the sample). Third, firms with high GOVINDEX scores (scores >0.462) are older, larger, with better financial performance and higher free float and are highly leveraged.

4.2 Logistic model results

Table 5 presents the logistic regression results, showing variables that have a strong correlation with a high GOVINDEX. At a 1% significant level, firm size, Tobin's Q, equity, blockholders, family share ownership, the first generation as director and the crisis are significantly affecting high governance compliance. While equity, family director and the financial crisis are positively influencing firms' compliance with the Indonesian CG codes, blockholders and family ownership have a detrimental impact on the firm's compliance with the regulations. Table 5 (model 2) also demonstrates that the third generation of family founders harms CG compliance, consistent with previous literature (Luan et al., 2018) which includes the concept of transgenerational entrepreneurship as stated by (Habbershon and Pistrui, 2002). According to Deloitte's (2019) report, only around 13% of family enterprises survived into the third generation. Based on this premise, there are challenges with successful intergenerational governance and succession, including issues with third-generation leadership and competency in business activity (Ramadani et al., 2020).

Furthermore, the positive role of crisis documented in this study is consistent with previous studies showing that the crisis increases the likelihood of the firm adopting CG mechanisms (Catuogno et al., 2018). However, this finding differs, at least in part, from the findings of Karaibrahimoglu (2010) who discovered that companies choose to cut their spending on CSR activities during the financial crisis. This study's finding on crisis–CG relationships also confirms Souto's (2009) study who documented the long-term benefit of CSR in times of financial trouble.

4.3 Logistic model results for family and nonfamily firms

Table 6 presents the logistic regression results for family and nonfamily firms (Models 1 and 2). It summarizes the key drivers or variables that are significantly correlated with higher CG practices as measured by GOVINDEX. Results show that family firms are more likely to comply if they have higher equity (more profitable and more capitalized). Meanwhile, the number of shares owned by the largest shareholders and family share ownerships are negatively associated with high compliance in the third generation as directors. The level of equity and financial crisis are determinants of high compliance for nonfamily firms, with strong evidence that financial shock positively impacts a firm's decision to comply with the CG codes.

In conclusion, family and nonfamily businesses differ significantly in several key characteristics. Compared to nonfamily firms, family businesses are more concerned with market (equity) values. They are more likely to adhere to market regulations and policies when their equity and Tobin's Q (market value proxies) are high. These firms are compelled to follow governance codes to send a positive signal to outside investors. Hence, the market or investors are willing to pay higher or premium prices for companies' stocks and assets.

By contrast, third-family generation and blockholders may act as deterrents to family firms' compliance with CG provisions. The greater the voting rights of families (who also serve as the largest shareholders), the lesser the probability of better compliance will be. According to the data collected for this study, the largest shareholders are also the controlling families, and they have excessive control over and private interest in the firms.

These findings are in line with those of Gomez-mejia and Nunez-nickel (2001) and Cai et al. (2012), who found increased family entrenchment in family firms where tighter ties between the principal and the manager exist. By contrast, our results differ from previous studies conducted in Taiwan (Kao et al., 2019), Malaysia (Shahrier et al., 2020) and Turkey (Ciftci et al., 2019).

CG regulations impose significant costs on family firms. Compliance with additional regulations, particularly governance codes, has resulted in dwindling family influence and control over the board of directors (Ricardo et al., 2016). Consequently, family shareholders or firms are less likely to comply with governance codes, which might reduce their supremacy and control over the assets and other benefits of the company. Other published studies, as discussed in (Krenn, 2015), highlighted the finding that firms may assess compliance costs and noncompliance costs, including the power of outside stakeholders to observe the actual implementation of CG codes. Firms with relatively high noncompliance and outsider monitoring costs are more inclined to have strategic decoupling, in which they may adopt codes without actively executing them in actual operation.

This study reveals challenges with CG enforcement in Indonesia. Firstly, most Indonesian companies are highly closely held as the founders normally own 70% of the business. This concentration damaged operational performance and market valuation. Secondly, family firms are run by their founding families and descendants. The result is an increase in agency conflict between the controlling shareholders and the minority shareholders, putting minorities at a disadvantage. Thirdly, there is a complex corporate structure and a lack of transparency in the ultimate beneficiary shareholders' disclosures. The market system fosters large conglomerates, yet there is no sufficient enforcement to prevent and deter conglomerate builders' abuses. The monopolistic power and lack of incentives to improve are the fourth and final issues.

The Indonesian economy is concentrated in a few conglomerates with monopoly power. Monopoly power promotes business inefficiency and opacity in companies, tolerates poor management and ensures that profits are sufficient to satisfy investors. There is less motivation to improve corporate transparency because business acumen is less important to corporate success than connections and favoritism. Finally, the country continues to struggle with lax regulatory and legal enforcement. Previous studies (e.g. Keay, 2014) classified voluntary CG provisions as soft law or nonbinding regulations, and compliance with such laws is largely left in the hands of the key actors, notably the board and management. This study demonstrates that compliance has remained relatively low over the last decade. In light of this, it is crucial to ensure that the burden and initiatives are not just placed on companies but also on other parties who have the right to regulate and monitor the actual execution of such compliance requirements.

5. Discussion

This study conducts two tests to investigate the determinants of firm-level CG compliance: (1) determinants of the GOVINDEX and (2) determinants of CG compliance in two firm institutional settings, namely, family and nonfamily firms. This study's CG index is constructed using 15 CG provisions. The logistic regression model is employed given the dichotomous or binary nature of the investigated dependent variables. The first test is used to identify factors that contribute to high GOVINDEX scores, while the second test aims to determine the variables that influence strong governance in publicly traded and family-owned firms.

The study concludes that financial performance, firm size, market crisis and the firm's founders are the most important components of high governance. Bigger firms as measured by equity and sales are more likely to comply with the regulations than smaller firms. Profitable and healthy firms also comply more, as shown by the ROA coefficient. This study argues that poor financial performance may reduce a firm's likelihood of complying with market requirements. However, the study demonstrates a negative relationship between firm size and a high index. This study also identifies the detrimental impacts of family share ownership and blockholders. Because family firms have lower compliance rates than nonfamily firms, this poor performance is affected by the governing families being shareholders and board members.

This study has the following implications. The occurrence of corporate failures and economic crises in the past decades possibly may have been the primary factor in the emergence of CG practices in the corporate world. Scholars and corporations have regarded CG as an effective mechanism for regaining market and stakeholders' confidence and trust. This study will be beneficial for various parties, including policymakers, market regulators and corporations involved in strategic decision-making.

6. Conclusion

The empirical findings of this study reveal that firm size and performance have a sizeable and significant influence on CG practices, and family firms are less compliant than nonfamily firms. These findings imply that family-controlled firms should pay greater attention to CG issues. Thus, adopting CG practices should go beyond checking a box. Family firms must make substantial efforts to be more consistent in their governance practice, as the trend toward CG compliance is projected to continue in the near future, due to demands for good governance and transparency. Meanwhile, larger and more widely held firms may learn from the corporate financial scandals of Enron, Parmalat and Lehman Brothers (Di Miceli da Silveira, 2011) that poor management can lead to financial fraud, moral hazard, corporate crisis and organizational failures

The results of this study offer some implications for both academicians and policymakers. Enhancing CG at the firm level is a viable goal, even though the agenda or goal of reforming minority investor protection laws and improving judicial quality is challenging and may take a long time to show significant results. Policymakers and regulators in Indonesia should establish a short-term national agenda; they should design policies or specific CG for small firms, as well as stricter and more thorough regulations to empower minority shareholders' roles and rights. Firm-level governance issues will become even more critical in the future; thus, it will be advantageous in the long run to establish a dedicated body or task force to monitor and assess CG practices and develop specific rewards for high-compliance firms. To conclude, rigorous regulatory enforcement would be a crucial driver in ensuring compliance with the current laws and codes.

The results of the study corroborate theoretical arguments that concentrated ownership and excessive control by majority shareholders are critical in determining the likelihood of good governance practices by firms. Furthermore, the market and regulators must devise effective strategies to encourage and reward high levels of compliance. Moreover, some limitations could be addressed in future research. The current study focuses on a single-country setting, namely, Indonesia; there may be cultural elements unique to the Indonesian context that may limit the transfer of the results to other countries with their own cultural settings and institutional legal environments. The GOVINDEX used in this study consists of only 15 CG elements covering the period between 2008 and 2013. Thus, future research could carefully incorporate more CG elements and extend the period by including more recent years. Finally, this study's findings and implications will contribute to a better understanding of CG development and adoption in Asia, a region with the quickest CG adoption.

Table 1

Quintile scores of the samples

Quintile	Observation	% of sample	Quantile index	Quartile group share of total GOVINDEX
1st	328	22.09	0.231	20.69
2nd	293	19.73	0.308	20.99
3rd	454	30.57	0.462	32.86
4th	410	27.61	≥ 0.462	25.46
Total	1,485	100		100
Mean of total samples (GOVINDEX)			0.353

Table 2

Definition of variables

Variables	Explanation	Expected sign
GOVINDEX	A dummy variable t 1 if firm has CG index score ≥ required threshold; 0 otherwise
FAMFIRM	Dummy variable that equals one if the family has at least 10% (or 30%) voting rights and has the founder and any member of the family acting as director and/or commissioner; zero otherwise	–
DIR1	Dummy variable that equals one if the founding family held the position as director; zero otherwise	–
DIR2	Dummy variable that equals one if founding family descendant (second generation) held the position as director; zero otherwise	–
DIR3	Dummy variable that equals one if founding family descendant (third generation) held the position either director; zero otherwise	–
COM1	Dummy variable that equals one if the founding family held the position as commissioner; zero otherwise	–
COM2	Dummy variable that equals one if founding family descendant (second generation) held the position as commissioner; zero otherwise	–
COM3	Dummy variable that equals one if founding family descendant (third generation) held the position either commissioner; zero otherwise	–
AGE	The observation period minus the date of the firm's establishment	+
SIZE	Book value of assets (IDRbillion)	+
TQ	MV of equity plus the BV of total assets minus equity and deferred taxes divided by BV of total assets	+
EQUITY	Total equity (IDR 1 trillion)	+
DER	Total debt/equity	+
BLOCKSHARE	The percentage of shares in the hand of the largest shareholders	–
FAMSHARE	Ratio of shares held by a controlling family	–
PUBLIC	Ratio of shares held by the minority shareholders	–
CRISIS	Dummy variable that equals one for year 2008 and 2009; zero otherwise	+

Table 3

Corporate governance index construction and statistics summary

Index	Definition	Mean	Std. Dev
GOVINDEX	Aggregate index that consists of 15 CG elements	0.35	0.15

Elements of GOVINDEX	Definition	Mean	Std. Dev
1. Code of ethics	Equals 1 if firm has anti-corruption and anti-bribery policy, 0 otherwise	0.32	0.47
2. Insider trading rules	Equals 1 if firm has policy against insider trading, 0 otherwise	0.03	0.18
3. Anti-corruption and bribery	Equals 1 if firm has anti-corruption and anti-bribery policy, 0 otherwise	0.05	0.23
4. Whistleblower system	Equals 1 if firm has whistleblower protection mechanism, 0 otherwise	0.09	0.28
5. Sanction	Equals 1 if firm does not breach any stock market regulations or rules, 0 otherwise	0.22	0.42
6. Audit	Equals 1 if firm hires one of the biggest 4 international auditing company, 0 otherwise	0.53	0.49
7. Largest shareholders	Equals 1 if firm's largest shareholder owns less than 50% voting rights, 0 otherwise	0.48	0.49
8. Free float ratio	Equals 1 if minority shareholders own >7.5% of total voting rights, 0 otherwise	0.89	0.3
9. Ultimate BO	Equals 1 if firm discloses ultimate beneficial owner, 0 otherwise	0.17	0.37
10. Independent supervisory board	Equals 1 if independent supervisory boards are more than 30% of the total board, 0 otherwise	0.77	0.41
11. Independent director	Equals 1 if the firm has ≥ 1 independent director, 0 otherwise	0.01	0.14
12. BOC size	Equals 1 if the firm has 4 to 8 supervisory boards in the board, 0 otherwise	0.47	0.49
13. BOD size	Equals 1 if the firm has 5 to 9 directors in the board, 0 otherwise	0.45	0.49
14. ESOP	Equals 1 if firm issues employee stock option program, 0 otherwise	0.09	0.29
15. CSR	Equals 1 if firm discloses CSR programs in the annual report, 0 otherwise	0.66	0.47

Table 4

Descriptive statistics

Observation	(1) All samples N = 1,485		(2) Firms with high index (GOVINDEX ≥0.462)		(3) Firms with low index (GOVINDEX <0.462)
	Mean	Std. Dev.	Mean	Std. Dev.	Mean	Std. Dev.
	N = 1,485		N = 410		N = 1,075
DEPENDENT VARIABLE
GOVINDEX	0.353	0.154	0.553	0.097	0.277	0.091
INDEPENDENT VARIABLES Firm type
1. Family	0.595	0.491	0.462	0.499	0.683	0.465
2. Nonfamily	0.405	0.332	0.535	0.499	0.316	0.465
Financial characteristics
3. Age	33.985	19.455	39.641	25.707	31.828	15.951
4. Size	5.569	14.949	13.634	24.588	2.493	6.652
5. TQ	1.338	0.893	1.818	1.093	1.155	0.724
6. Equity	2.394	6.984	6.146	11.914	0.963	2.432
7. DER	1.616	2.531	1.404	1.969	1.696	2.711
Family control
8. DIR1	0.277	0.448	0.193	0.395	0.309	0.462
9. DIR2	0.290	0.454	0.217	0.413	0.317	0.466
10. DIR3	0.036	0.193	0.010	0.098	0.046	0.217
11. COM1	0.362	0.481	0.222	0.416	0.415	0.493
12. COM2	0.155	0.362	0.112	0.316	0.171	0.377
13. COM3	0.013	0.124	0.024	0.183	0.008	0.091
CG mechanism
14. Blockholders	0.499	0.218	0.502	0.224	0.498	0.216
15. Famshare	0.303	0.312	0.183	0.265	0.348	0.317
16. Publicshare	0.28	0.182	0.316	0.198	0.276	0.175

Table 5

Logistic model for determinants of CG compliance

Variable	Model 1			Model 2
	Dependent variable = 1 if a firm has GOVINDEX ≥0.353, “0” otherwise			Dependent variable = if a firm has GOVINDEX ≥0.462, “0” otherwise
	Coeff.	Std. Err.	Sig.	Coeff.	Std. Err.	Sig.
Family Firm	−1.3353	0.23	***	−1.0885	0.2434	***
Age	0.0140	0.01	**	−0.0009	0.0064
Size	0.0622	0.02	***	0.0400	0.0184	**
TQ	0.5981	0.11	***	0.4076	0.1028	***
Equity	0.0775	0.05		0.1388	0.0454	***
DER	0.0038	0.03		0.0192	0.0475
DIR1	0.6422	0.17		0.7824	0.1815	***
DIR2	0.1080	0.11	***	0.2057	0.1165	*
DIR3	−3.0674	1.03		−3.9539	1.7718	**
COM1	0.2259	0.15	*	0.1930	0.1775
COM2	0.4431	0.23		0.1147	0.2998
COM3	−0.0291	0.60	***	0.8254	0.7314
Blockholders	−2.1463	0.45	***	−1.5653	0.4930	***
Famshare	−2.2857	0.40	**	−2.6507	0.4156	***
Publicshare	−1.1929	0.52	***	−0.4781	0.5633
Crisis	2.2279	0.38	***	2.3856	0.4743	***
Year Dummy	Yes			Yes
Industry Dummy	Yes			Yes
Predicted 1s that were actual 1s (%)	0.8060			0.7649
Predicted 0s that were actual 0s (%)	0.8192			0.8535
Prediction Power	0.8140			0.8340
Log Likelihood function	−627.8628			−525.7663
Prob (χ2>value)	0.0000			0.0000
Pseudo r-squared	0.3615			0.3916
Chi-square	326.3200			352.0100
Number of obs	626			410

Note(s): The table presents results of the logistic regression of the likelihood of firm to comply with the corporate governance codes. Dependent variable is the dichotomous variable of “1” if the firm has high GINDEX score ( ≥ 0.462), or “0” otherwise. GINDEX is the aggregate corporate governance index consists of 15 governance elements. Dependent variables include: Tobin's q, ROA, size of the firm, debt to equity ratio, operating expense ratio, ownerships structures (largest, family and public shares ownerships) and family generations' representative in the boards. Asterisks denote significance at the 1% (***), 5% (**) or 10% (*)

Table 6

Logistic model for determinants of CG compliance between family and non-family firms

Variable	Model (3) Family firm			Model (4) Non family firm
	Dependent variable = 1 if a family firm has GOVINDEX ≥0.462, “0” otherwise			Dependent variable = 1 if a nonfamily firm has GOVINDEX ≥0.462, “0” otherwise
	Coeff	Std Err.	Sig.	Coeff	Std Err.	Sig.
Age	−0.0080	0.0140	**	−0.0287	0.0164	*
Size	0.0310	0.0234		0.0471	0.0288
TQ	0.6182	0.1828		0.7753	0.2381	***
Equity	0.1324	0.0545	**	0.1226	0.0761
DER	−0.1152	0.0107		−0.0053	0.0670
DIR1	0.6945	0.2927
DIR2	0.4232	0.3431
DIR3	−3.6538	1.3180	**
COM1	0.1074	0.2945	*
COM2	0.0651	0.3538
COM3	0.3331	1.0512
Blockholders	−2.2881	0.6606	***	−1.1170	1.0752
Famshare	−2.4702	0.5651	**
Publicshare	−0.0614	0.7072		−0.2892	1.2342
Crisis	2.5516	0.7551	***	2.6024	0.6655	***
Year Dummy	Yes			Yes
Industry Dummy	Yes			Yes
Predicted 1s that were actual 1s (%)	0.7486			0.8125
Predicted 0s that were actual 0s (%)	0.8814			0.8645
Prediction Power	0.8561			0.8434
Log Likelihood function	−303.8098			−159.2274
Prob (χ2>value)	0.0000			0.0000
Pseudo r-squared	0.4237			0.4240
Chi-square	253.6800			156.0100

Note(s): The table presents results of the logistic regression for the determinants of governance compliance between family-owned business and nonfamily ones. Dependent variable is the dichotomous variable of “1” if the firm has high GOVINDEX score ( ≥ 0.462), or “0” otherwise. Asterisks denote significance at the 1% (***), 5% (**) or 10% (*)

Note

1. See https://www.nytimes.com/2009/02/04/business/04pay.html

Funding: This study received no funding.

Competing interests: The author declares that there are no competing interests.