The investment behavior of politically connected firms in Indonesia

Abstract

Purpose

The authors examine the effect of the politically connected supervisory board (PV_SVP) on corporate investment behavior in Indonesia in the period of 2015–2019.

Design/methodology/approach

The authors use Indonesian listing companies as our sample. Ordinary Least Squares regression is applied to investigate this association. Also, the authors address the endogeneity problem by using the generalized method of moments.

Findings

The authors find that firms with political connections through Supervisory Boards (SBs) are negatively significantly associated with corporate investment. Our results are robust to alternative measures and to test for endogeneity.

Research limitations/implications

The authors contribute to prior research by showing empirical findings on the investment behavior of politically connected firms using an emerging economy context, Indonesia, which has a unique political landscape. The authors offer practical implications for practitioners and policymakers, such as improving the corporate governance system and promoting better investment opportunities by establishing a transparent and competitive environment.

Originality/value

Our study differs from other studies due to different corporate governance and political connection settings. While most prior studies examine the investment behavior of politically connected firms using the Chinese context, the authors use Indonesia which has different political and governance landscapes. Indonesia applies a two-tier board system that promotes the strategic role of the political supervisory board.

Citation

Joni, J., Natalia, M. and Leliana, L. (2023), "The investment behavior of politically connected firms in Indonesia", Asian Journal of Accounting Research, Vol. 8 No. 4, pp. 348-355. https://doi.org/10.1108/AJAR-12-2022-0416

1. Introduction

Our study aims to investigate how investment behavior is different between politically connected and nonpolitically connected companies in an emerging economy, Indonesia. Specifically, we examine the effect of political connections through supervisory board members on corporate investment. The research on the effect of political connections on capital market outcomes has been explored around the world, including stock return (Cooper et al., 2010; Faccio and Parsley, 2006), accounting information quality (Chaney et al., 2011) and market valuation (Faccio, 2006). Further, previous studies document that political affiliation is an effective way for companies to obtain easy access to external financing (Claessens et al., 2008; Bliss and Gul, 2012; Boubakri et al., 2012; Houston et al., 2014; Joni et al., 2019; Kamarudin et al., 2021), tax benefits (Abdul Wahab et al., 2017) and audit pricing (Gul, 2006) which can influence their corporate investment behavior (Pan and Tian, 2020). However, limited studies examine the effect of political connections via supervisory board members on investment behavior in a dual board system context. In addition, the empirical results on the relationship between political connections and corporate investment are inconclusive. While several studies argued that politically connected firms experience better investment efficiency than their counterparts because political connections allow them to have extra external funding (e.g. Pan and Tian, 2020; Phan et al., 2020), other studies confirm that firms with government connections may suffer from government intervention to invest in unfavorable projects, but it is concerned politically (e.g. Chen et al., 2011; Chen et al., 2020; Nasih et al., 2020).

Our study is different from prior studies in terms of institutional background. We examine the investment behavior of politically connected firms in Indonesia, where the capital market is imperfect and semi-efficient (Joni et al., 2019). In the context of semi-efficient market, informational asymmetries and agency problems, which have been the most crucial factors in affecting corporate investment efficiency, are more relevant for investment decisions than the efficient market (Xu et al., 2013). Also, most prior studies use China as the setting of their studies, which is very different from Indonesia. While the Chinese government system is communist, the Indonesian government system is a democracy where people have the authority to vote for their legislators and president. Accordingly, the political power in Indonesia is more dispersed compared to China. Unlike the political connection pattern in China, where external resources are controlled by the government and the role of SOEs is important in China (Imamah et al., 2020; Pan and Tian, 2020), political connection in Indonesia depends on various political parties and army power (Joni et al., 2019). Compared to Nasih et al. (2020), our work also promotes another uniqueness of the Indonesian corporate governance system, which implements a two-tier board system. Indonesian dual board system has separated operating and monitoring functions into two different boards called a board of directors and a supervisory board (or board of commissioners). Political connection is mostly developed through elite people appointed to supervisory board, and their roles become very strategic (Joni et al., 2019). Therefore, this institutional context is interesting to investigate whether political connection through the supervisory board effectively influences corporate investment.

Based on 1,523 firm-year observations of Indonesian listed companies from 2015 to 2019, we find that firms with the politically connected supervisory board have significantly lower investment expenditure than their counterparts. The results are robust after addressing the endogeneity issue using the Generalized Method of Moments (GMM) model and alternative measure of investment expenditure. Our paper contributes to the literature by providing empirical results on the role of the politically connected supervisory boards in determining corporate investment behavior in the context of Indonesian dual board system. Also, we contribute to the policymakers in terms of the improvement of corporate governance system which can offer better investment opportunities in Indonesia.

The remainder of our paper is reported as follows: in Section 2, we provide theoretical background and hypothesis development, followed by research design in Section 3. Next, research findings and discussion are explained in Section 4. The final section is the conclusion.

2. Theoretical background and hypothesis development

The association between political connections and investment behavior has been examined and the results are inconclusive and need more explanation. On the one hand, Resource Dependency Theory explains that organization is an open access and uses political connection to deal with external uncertainty and its dependency on government bureaucracy in conducting business. A firm with political connection obtains more external funding and lower cost of finance than nonpolitically connected firms (Joni et al., 2020; Houston et al., 2014; Boubakri et al., 2012), that cause a better investment efficiency. Xu et al. (2013) test whether Chinese family firms can mitigate their underinvestment problem when they have connections with the government. Using the Shanghai and Shenzhen listed firms in the period of 2000–2007, they show that family firms have difficulty accessing external funding and suffer from underinvestment problem. However, political connections help family firms to mitigate the underinvestment problem. The effect is more substantial in financially constrained companies. Pan and Tian (2020) set up a natural experiment using the recent anti-corruption based on listed firms in China and examined the association between political affiliation and corporate investment from 2003 to 2014. They find that politically connected firms experience lower investment after the ousting of the corrupt government officers compared to nonpolitically connected firms. Also, politically connected SOEs obtain more efficient investment than nonpolitically connected SOEs after the corrupt government officer’s ousting. Since the recent anti-corruption campaign, the investment efficiency for politically affiliated firms in China becomes more significant. Imamah et al. (2020) explore the influence of the anti-corruption campaign on the investment behavior of politically affiliated companies in China from 2007 to 2016. Their finding indicates that all types of politically connected companies (political connections of non-SOEs/central SOEs/local SOEs) reduced their investment expenditures after the anti-corruption campaign compared to the period before the campaign. Also, political connections between non-SOEs and local SOEs improve their investment efficiency after the anti-corruption campaign. Phan et al. (2020) examine the relationship between political connections and corporate investment in Malaysia from 2002 to 2016. Based on 631 listed firms, they document a positive influence of political connections on corporate investment. Firm with political connections with top political elites has 2.1–2.9% more investment than nonpolitical firms in the context of a stable political environment, such as Malaysia.

On the other hand, the agency perspective argues that information asymmetry between management and shareholders distracts the company from making an efficient investment decision. And political connection worsens it because firms with political connections should invest in unfavorable projects due to political concerns. Using a Chinese sample, Chen et al. (2011) indicate that government intervention in SOEs prevents them from making optimal investments and even lower investment efficiency. Another study using the Chinese context by Chen et al. (2020) also found that regional political favoritism is related to less efficient corporate investment. Next, Nasih et al. (2020) examine the association between political connections and a company’s overinvestment in Indonesia. They show a negative association between those variables. It might happen because of mixed economy applied in Indonesia, where both public and private sectors are essential.

In line with the theoretical background and the prediction of the previous empirical studies, the hypothesis is proposed as follows:

H1.

Politically affiliated firm through supervisory board connections is associated with corporate investment expenditure, ceteris paribus.

3. Research design

3.1 Data and sample

We use all listed companies in Indonesia during 2015–2019 as our sample, except financial sectors because they are regulated under different mechanisms. The chosen period is considered as a stable political regime. In our research, we collected all information from several different sources. We first use corporate governance information, including supervisory board connections, which is manually collected from company’s annual report, its website and Google search engine. Next, our financial data is downloaded from the Datastream database.

3.2 Models and variables

We use the following regression models to examine the effect of politically connected firms through supervisory boards on corporate investment expenditure in Hypothesis 1.

INVassetit=α1PC_SPVBit+α2SIZEit+α3DERit +α4GROWthit+α5INDUSTRYit+α6YEARit+εit

A detailed definition of variables and references are outlined as follows (see Table 1).

3.3 Statistical analyses

The objective of this study is to examine the impact of political connections through the supervisory board on investment expenditure, using an Ordinary Least Squares (OLS) regression model. We begin by presenting descriptive statistics to better understand the sample, followed by a pairwise correlation analysis to check the correlation of the key variables. To address the potential issue of multicollinearity, we conduct a variance inflation factor (VIF) test before proceeding with the regression analysis.

3.4 Sensitivity analyses and alternative measure

Endogeneity is a critical issue in corporate governance literature, particularly when investigating the association between political connections and investment efficiency. The possibility of firms with efficient investment connecting with government officials introduces the potential for an endogeneity problem. The independent variable, political connections, may be endogenously determined and influence the dependent variable, investment expenditure. To address this problem, the study employs the GMM to estimate the model. GMM is a widely accepted econometric technique that addresses endogeneity. These methods are consistent with previous research (Phan et al., 2020; Pan and Tian, 2020) and aim to ensure the validity and reliability of our findings. Moreover, the study re-estimates the model using a different measure of investment expenditure, a five-year fixed asset, which represents the long-term investment commitment made by the firm. This alternative measure can serve as a proxy for investment expenditure, offering a more accurate assessment of the relationship between political connections and investment efficiency.

4. Research findings and discussion

4.1 Descriptive statistics

Table 2 reports descriptive statistics for the total sample of 1,523 firm-year observations to examine the relationship between the politically affiliated firm and corporate investment. We winsorized at the 1st and 99th percentiles for continuous variables, except dummy variables. The average of several main variables is presented in Table 2. The average of INV_asset is 6.680 and PC_SPV is 15.6%. We find that these values are consistent with prior literature (e.g. Pan and Tian, 2020; Joni et al., 2019). The average firm size (SIZE) value is 22.179 with a maximum of 32.921 and a minimum of 2.607. Next, the value of DER and GROWth are considered reasonable compared to prior literature (e.g. Pan and Tian, 2020; Joni et al., 2019).

Next, Pairwise Pearson Correlation is applied to check correlation among key variables in the model, industry and year effects are excluded. In Table 3, it reports that the highest correlation is between firm size and leverage (r = 0.131). Then, we run the multicollinearity test using VIF in Table 4 and the result shows that the model is free from the multicollinearity problem.

4.2 Politically connected supervisory board and corporate investment

Table 4 reports the OLS estimate to examine the effect of the politically connected supervisory board on corporate investment. The result of the model shows that the coefficient of political connection via the supervisory board is negative and significant at the 5% level (the coefficient = −1.116 and t = 1.95), suggesting that the presence of political connections through supervisory board reduces corporate investment expenditure. The result is consistent with Agency Theory which explains the reason why politically affiliated companies tend to have lower investment efficiency and avoid optimal investment. It argues that information asymmetry between agent (management) and principal (shareholders) distracts the company from making an efficient investment decision. The leverage is positively and significantly associated with corporate investment. However, firm size and company growth are not statistically significant. Mostly, the results are consistent with prior studies, such as Nasih et al. (2020).

4.3 Sensitivity analysis

It is clearly argued that firms with politically connected SBs are negatively associated with corporate investment. Table 5 reports an alternative analysis using the GMM model to check for endogeneity issues in testing the relationship between politically connected SBs and Corporate Investment. Overall, the result remains similar after applying the GMM model.

4.4 Alternative measure

To improve the reliability of our main result, we use a five-year fixed asset to measure capital expenditure as an alternative measure and rerun the regression. As shown in Table 6, the result shows a negative association between politically connected firms through the supervisory board and corporate investment. The result in Table 6 is consistent with the result in Table 5.

5. Conclusions

Our paper contributes to the literature by examining the relationship between the politically affiliated supervisory board and corporate investment behavior in an emerging market, Indonesia. After controlling several main variables, we find a negative association between political connections through the supervisory board and investment expenditure. Consistent with agency theory, our finding suggests that firms with political connections via supervisory boards make inefficient investment decision-making. Politically connected firms through supervisory board tend to invest in unfavorable projects due to political concerns. Additionally, we conduct robustness tests by addressing endogeneity problem and using an alternative measure of corporate investment, the results remain similar.

This study offers implications for scholars on the cost of having political connections through supervisory boards. We also provide implications for practitioners and regulators to strengthen the corporate governance system in Indonesia. Further, policymakers should establish a transparent and competitive environment to promote better investment opportunities. However, our findings should be interpreted by considering several limitations which can be potential future research. First, we consider firm size, growth and leverage as control variables in the model, but other control variables (such as free cash flow, firm’s age, return on asset, state-owned enterprises, etc.) are not included. Second, the effect of political connections on corporate investment behavior is examined using a quantitative approach. It will be interesting to address this issue using a qualitative approach to obtain a deep and comprehensive understanding of how and why the presence of political connections reduces corporate investment. Third, a firm can establish a connection with the government through several different channels, such as family members, friends, etc. Our paper needs to address these issues, and it raises the need for future studies.

Table 1

Variables’ definition

Variable	Definition
INVassetit	Capital expenditure of the company in year t scaled by five-year total assets of the company in year t−1 (Phan et al., 2020)
PC_SPV_it	Dummy variable is set to 1 if the company has political connections through the supervisory board in year t and 0 otherwise. We consider our sample as politically connected firms if one of the top management (supervisory board) was a former government officer, such as a minister, a member of parliament, a president, etc. (Faccio, 2006)
Control variables – firm characteristics
SIZEit	The natural log of the company’s total assets in year t (Phan et al., 2020; Pan and Tian, 2020)
DERit	Total debts scaled by total assets in year t (Phan et al., 2020)
GROWthit	Sales growth (Phan et al., 2020)
Control variables – fixed effects
INDUSTRYit	A vector of industry indicator variables categorized using two-digit GICS (Global Industry Classification Standard)
YEARit	A vector of year indicator variables: 2015; 2016; 2017; 2018; 2019

Source(s): Authors’ own work

Table 2

Summary statistics of main variables

Variable	N	Mean	Std. dev.	Min	Max
INVasset	1,523	6.680	6.653	0.070	34.950
PC_SPV	1,523	0.156	0.290	0.000	1.000
SIZE	1,523	22.179	9.362	2.607	32.921
DER	1,523	0.734	0.371	0.000	0.960
GROWth	1,523	0.116	0.614	−0.831	5.267

Note(s): The table reports the summary statistics of the key variables. The sample includes 1,523 firm-year observations for the period 2015–2019. We use all variables based on calendar year. The variable definitions are available in Table 1

Source(s): Authors’ own work

Table 3

Correlation matrix

	(1)	(2)	(3)	(4)	(5)
(1) INVasset	1.000
(2) PC_SPV	−0.046c	1.000
(3) FSIZE	0.046c	0.009	1.000
(4) LEV	0.100a	0.058b	0.131a	1.000
(5) GROWth	0.002	−0.020	−0.026	0.052a	1.000

Note(s): The table reports the Pairwise Pearson Correlation matrix for the full sample (1,523 firm-year observations). Please see Table 1 for definitions of variables. The superscripts a–c indicate two-sided significance at the 1%, 5% and 10% levels

Source(s): Authors’ own work

Table 4

Politically connected SBs and corporate investment – pooled OLS

	Model
INTERCEPT	5.125a (3.59)
PC_SPV	−1.116b (1.95)
SIZE	0.069 (1.38)
DER	1.777a (3.57)
GROWth	0.137 (0.51)
INDUSTRY	Included
YEAR	Included
Average VIF	2.46
Adj. R2	0.083
F	12.54
Prob > F	0.000a
N	1,523

Note(s): The table reports OLS coefficient estimates, and indicator variables are included in the regression to control for year and industry-fixed effects. Please see Table 1 for definitions of variables. The superscripts a–c indicate two-sided significance at the 1%, 5% and 10% levels, respectively

Source(s): Authors’ own work

Table 5

Politically connected SBs and corporate investment – GMM model

	Model 1
INTERCEPT	−1.976 (1.28)
PC_SPV	−0.011a (5.42)
SIZE	0.000 (0.35)
DER	0.018a (9.43)
GROWth	0.001 (0.75)
INDUSTRY	Included
YEAR	Included
N	1,523

Source(s): Authors’ own work

Table 6

Politically connected SBs and corporate investment – alternative measure

	Model
INTERCEPT	24.474a (2.00)
PC_SPV	−16.373a (3.76)
SIZE	0.381 (0.92)
DER	−7.159a (−1.55)
GROWth	3.810 (1.55)
INDUSTRY	Included
YEAR	Included
Average VIF	2.71
Adj. R2	0.022
F	2.76
Prob > F	0.001a
N	845

Source(s): Authors’ own work