Tunnelling and Company Performance In The Financial Industry In Indonesia

This study aims to find empirical evidence of the impact of related party transactions, particularly tunneling transactions, on the performance of the Indonesian financial industry. The research samples were 105 companies that are members of the financial industry listed on the Indonesia Stock Exchange (IDX) during 2018-2020 with a total of 300 observations. Tunneling was measured using the proportion of related party transactions related to trade receivables, other receivables


INTRODUCTION
The performance of a company is critical because organizations with strong performance can entice investors to invest (Fiyati & Noegroho, 2021). Accounting-based measurements and marketbased measurements can be used to evaluate a company's performance (Al-Matari et al., 2014). According to Supatmi et al. (2019), depending on the type and quantity of related party transactions, they will have an impact on the company's performance. Corporate action data according to the Financial Services Authority (OJK) report shows the proportion of issuers conducting related party transactions for other corporate actions during 2018-2020 was 49.5 percent, 48 percent, and 76.25 percent, respectively. This demonstrates that companies in Indonesia engage in a large number of related party transaction (RPT) to boost the performance of the company.
The impact of RPT to company's performances has two different sides. Related party transactions allow companies to reduce the cost of transactions with third parties or transaction cost efficiency (efficient transaction hypothesis) thereby increasing company performance (Gordon et al., 2004). In contrast, the research of Supatmi et al. (2019) states that tunnelling related party transactions leads to conflicts of interest rather than efficient transactions. This is in line with Gordon & Henry, (2005) who stated that RPT creates a potential conflict of interest in line with the (conflict of interest hypothesis). Villalonga & Amit (2006) stated that countries in Asia, including Indonesia, experience majority and minority shareholders in a type II agency conflict. Related Party Transactions (RPT) are one way for the shareholders of majority to take over the shareholders of minority (Supatmi et al., 2019). Cheung et al. (2009), Yezhen & Wong (2015), Tsai et al. (2015), and Deng et al. (2005) in their research found that tunnelling related party transactions will result in losses for the company and have a negative impact on its performance as measured by return on assets and return on sales. Companies that have high receivables related party transactions also have a high risk of bad debts (Jian & Wong, 2003) thus reducing the company's profit achievement.
Investors, as market participants, have a tendency to avoid companies that engage in tunneling operations. Related Party Transactions (RPT) damage firm value, in line with tunneling activities that create negative market reactions (Cheung et al., 2009). Wong et al. (2015) prove RPT are for the benefit of the majority shareholder and to eliminate the value of the minority shareholder. In line with the research of Tsai et al. (2015) on companies that are classified on the Taiwan Stock Exchange, proved that companies engage in related-party transactions showed tunnelling motivation and caused the company value of affiliated companies to tend to be lower than non-affiliated companies.
The goal of this research is to discover empirical evidence of the impact of related party transactions, especially those that are tunnelling, on the performance of the financial industry in Indonesia. The research hypothesizes that tunnelling-related party transactions have a negative impact on company performance, both in terms of accounting performance and market performance. This research is based on at least three reasons. First, research on related party transactions that focuses on tunnelling has not been widely explored, especially in the context of the financial industry in Indonesia. Tunnelling based on Johnson et al. (2000) represents the transfer of assets and profit coming from the company's subsidiary to the parent company, impacting on the takeover of non-controlling shareholders. This study uses three indicators to measure tunnelling, namely related party transactions related to trade receivables, related party transactions related to other receivables, and related party transactions with assets other than trade receivables (Supatmi et al., 2019). The impact of tunnelling on company performance will be measured by accounting-based performance as proxied by Return On Assets (ROA) and market-based performance as proxied by Tobin's Q. This is because tunnelling related party transactions will not only affect the company's profitability in terms of accounting records but can also affect investor response. Second, previous studies used more research samples in the manufacturing industry or non-financial industry. The financial industry is a highly regulated industry so this research will present a different picture than previous research. The financial industry, with strict regulations, can still carry out related party transactions due to public pressure or public supervision, so there is still a gap to be able to carry out related party transactions in the context of improving company performance. Third, this study uses data from 2020, which is the year of the Covid-19 pandemic as part of the research period, so this study aims to show the effect of the Covid-19 pandemic in the causal relationship of related party transactions on company performance. According to the government's analysis, the present economic crisis brought on by the Covid-19 outbreak will have an influence on numerous companies' profits and financial performance (Devi et al., 2020). Shen et al. (2020) indicated that the pandemic had a significant of negative impact towards the performance of Chinese issuers because of the declining in total revenue and ROA.
This study contributes additional empirical evidence regarding the effect of related party transactions, especially those that are tunnelling, on the performance of the Indonesian financial industry, which is seen from the accounting and market aspects. Practically, the results of this study are expected to assist investors in considering investment decisions related to tunnelling-related party transactions so that they can determine which companies are more profitable. For the financial industry, the findings of this study are expected to be used to help companies in the financial industry improve their performance. This research is also expected to be useful for Bank Indonesia (BI) and the Financial Services Authority (OJK) in developing policies to manage and supervise related-party transactions in the financial industry.

MATERIALS AND METHODS
This study used the population of the financial industry listed on the Indonesia Stock Exchange (IDX) during 2018-2020 which amounted to 105 companies or 300 research observations. The research data used secondary data from the 2018-2020 financial industry company annual financial statements obtained through the www.idx.co.id website or related company websites. By using the purposive sampling technique, the following samples were obtained: 105 Companies that do not publish annual reports for the year 2018-2020 in a row.
(5) Companies that display financial statements that do not end on December 31 for 2018-2020.

0)
Companies that do not have stock price information.
(0) Number of samples that meet the criteria 100 Source: Processed data, 2021 The financial industry is divided into five categories, namely the banking sector, financial institution sector, securities company sector, insurance sector, and other/general sectors. There were 100 companies that met the criteria for the research sample or which have a total of 300 observations. The test results found several obstacles in the data test where 14 outlier data were found for testing the dependent variable of accounting performance as proxied by ROA and 13 outlier data for testing the dependent variable of market performance as proxied by Tobin's Q. The following table 2 presents the number of observations for each test: Company Performance (KP) as the dependent variable was measured by accounting performance and market performance. Accounting performance is proxied by Return On Assets (ROA) as measured by net income divided by total assets (Jia et al., 2013). While the benchmark for market performance used Tobin's Q (TOBINS) which was obtained from the sum of the market capitalization value and the value of debt divided by total assets (Song, 2016). Tunnelling related party transaction as an independent variable was measured by three measurements (Supatmi, 2020), namely related party transactions related to trade receivables divided by total assets (RPTAR), related party transactions related to other receivables divided by total assets (RPTOR), and related party transactions related to assets other than trade receivables divided by total assets (RPTNAR).
The control variables used in this study are firm size, leverage, and managerial ownership. Company size (UP) is measured as the logarithm of the assets owned by the company (Wijaya et al., 2011), large company size, helps to increase its performances (William & Sanjaya, 2017). Corporate leverage (LEV) is measured by total debt divided by total company assets (Tsai et al., 2015). The positive impact on companies performances is shown from the level of leverage (Bona-Sánchez et al., 2017). Managerial ownership (KM) is the proportion of share ownership by managers to total shares (Supatmi et al., 2019). To accommodate the impact of the Covid-19 (PD) pandemic, this study added a control variable using a dummy variable measured by a value of 1 for the year during the Covid-19 pandemic (2020) and a value of 0 for the year before the Covid-19 pandemic (2018-2019) (Tiwu & Angi, 2021).
This study used panel data regression analysis techniques using the EViews 9 data processing program for hypothesis testing with the following mathematical equations: = α 0 +α 1 RPTAR + α 2 RPTOR + α 3 RPTNAR + α 4 UP + α 5 LEV + α 6 KM + α 7 PD + ε  Table 3 results of descriptive statistics show that the performance of accounting-based companies in the financial industry sector as represented by ROA during 2018-2020 on average obtained a net loss of 0.1% of their total assets. In terms of market performance (TOBINS), the average Tobin's Q shows that the financial industry has succeeded in developing market value and increasing investment with an average Tobin's Q value of more than 1 (Al-Matari et al., 2014). The highest accounting performance (ROA) was Magna Investama Mandiri Tbk (MGNA), which was 0.83 in the 2020 period as well as the lowest accounting performance (-1.37) for the 2019 period. The Syariah Life Insurance Company Jasa Mitra Abadi Tbk (JMAS) is the company with the highest market performance during the research period, which is 19.97 in the 2018 period and Bank Net Indonesia Syariah Tbk (BANK) is the company with the lowest accounting performance, which is 0.06 for the period 2020.

RESULTS AND DISCUSSION
Three measures were used to evaluate tunnelling, namely related party transactions related to trade receivables (RPTAR), related party transactions related to other receivables (RPTOR), and related party transactions related to assets other than trade receivables (RPTNAR) on average low value (0.1% to 2.3% of total assets). This indicates that related party transactions involving tunnelling in the financial industry in Indonesia were low during the study period. The low tunnelling in the finance industry was due in part to the fact that 13 financial companies did not perform any related party tunnelling transactions during this time period. On the other hand, Bank Rakyat Indonesia Agroniaga Tbk (AGRO) was found to have made transactions with related parties related to trade receivables (RPTAR) up to 69% of total assets, which means that almost 70% of its total assets traded receivables transactions with related parties.
Company size (UP) in the sample of financial industry companies during the study period had an average asset of Rp 29.62 trillion. In addition, the average level of leverage (LEV) in the sample of financial industry companies during the study period is 66%, which indicates that more than 60% of the company's existing assets are derived from third-party loans, indicating a high debt risk. Furthermore, managerial ownership (KM) in the financial industry has a low average of 3% which indicates that the directors and the board of directors of the financial industry had a low proportion of shares compared to the total shares outstanding in the market.
The results of the classical assumption test which includes normality, heteroscedasticity, multicollinearity, and autocorrelation tests, showed that the research data passed the classical assumption test, except for the normality test as shown in Table 4. The normality test using the Jarque-Bera test shows that the research data was not normally distributed because the probability is less than 0.05. However, this study covered 95.24% of the total population. The sample mean that is close to the average of the entire population can be considered data that is estimated to be normally distributed (Islam, 2018) because the sample is large or almost covers the entire population. This is in line with the Central Limit Theorem (CLT), i.e. data that are not normally distributed will not be a problem if the number of observations is more than 30 so that hypothesis testing can be carried out.
A panel data regression estimation test was performed prior to the panel data regression test, which included the common effect (CE), fixed effect (FE), and random effect (RE) models. According to the panel data regression estimation test, the suitable model for the dependent variable ROA and Tobin's Q in this study is a random effect model, as shown in Table 5.  Table 6 presents a summary of the findings of hypothesis testing using panel data regression with the random effect model.  Table 3. Source: Data processed, 2021 The value of Adjusted R 2 in Table 6 shows that tunnelling proxied by related party transactions related to trade receivables, other receivables, and other assets of trade receivables, along with company size, leverage, managerial ownership, and the Covid-19 pandemic situation can explain the proportion of variance of accounting performance (ROA) and market performance (TOBINS) respectively by 18 percent and 5 percent and other variables outside of this study account for the rest.
Taken together, all of these independent variables have been shown to significantly affect the performance of the financial industry. Based on this, the regression model for hypothesis testing statistically meets the goodness of fit, allowing this research model to be utilized to forecast a company's success.
The t-test results show that tunnelling proxied by related party transactions related to trade receivables and related party transactions related to other receivables had no effect on accounting performance (ROA) and market performance (Tobin's Q). Meanwhile, tunnelling proxied by related party transactions related to assets other than trade receivables was proven to have a negative effect on accounting performance. This means that an increasing number of financial institutions are engaging in related party transactions involving assets other than accounts receivable, such as related party investments, prepaid rents to related parties, loans to related parties, premiums and reinsurance, financing leases and consumer financing, and other assets of related parties will lower the company's ability to earn profits by using its assets. As a result, the hypothesis that tunnelling, as measured by related party transactions involving assets other than trade receivables, has a negative impact on a company's accounting performance, is supported.
The more companies that engage in related party transactions involving assets other than trade receivables, the higher the risk of bad debts (Jian & Wong, 2003) thus lowering the company's profit margin. The results of this study also support the agency theory and conflict of interest hypothesis (Gordon et al., 2004) that related party transactions can have a negative impact on the company since they generate a potential conflict of interest hypothesis that could compromise the management agency's responsibilities to shareholders or the board of directors' supervisory function. Related party transactions, such as opportunistic management income, tunnelling, and wealth expropriation from shareholders, are used as a strategy for managerial opportunism  so that expropriation solely benefits the controlling company's interests (Johnson et al., 2000). The results of this study are in line with Cheung et al. (2009), Yezhen & Wong (2015, and Supatmi (2020) who proved that tunnelling has an impact on reducing the company's accounting performance, especially profitability.
On the other hand, tunnelling proxied by related party transactions related to assets other than trade receivables has proven to have a positive effect on market performance. The more financial institutions engage in related party transactions involving assets other than trade receivables, the better their market performance. This research shows that when the company performs related party transactions involving assets other than trade receivables, investors who mirror the market have a positive reaction.
The positive regression coefficient shows that there is an indication that related party transactions related to assets other than trade receivables have a positive impact on the performance of market-based companies. Related party transactions related to assets other than trade receivables, such as investments and other receivables, are associated with higher interest income which implies higher net income and better profitability. As a result, the market reacted positively, boosting market performance by increasing the market price of the company's shares. This finding shows that related party transactions related to assets other than trade receivables are more likely to be propping than tunnelling, or are more supportive of efficient transaction costs from related parties or the efficient transaction hypothesis (Gordon et al., 2004). The findings of this study support the findings of  and Supatmi et al. (2019), which proved that related party transactions involving assets other than trade receivables have an impact on the company's profitability and market performance.
Table 6 also shows that the company's financial performance and market performance are influenced by other variables. Company size (UP) has a positive effect on the company's accounting performance (ROA) and has a negative effect on the company's market performance (Tobin's Q). The bigger the company, the stronger its potential to achieve company performance because management has a lot of funds to run its firm. However, larger companies may also reflect greater company risk and therefore have a negative impact on their market performance. The level of leverage (LEV) has been shown to have a negative impact on the company's accounting and market performance, reducing the company's capacity to meet its obligations with current assets. Managerial ownership and dummy factors during the year of the Covid-19 epidemic, on the other hand, have not been demonstrated to affect financial industry company performance.

CONCLUSIONS AND SUGGESTION
The results of the study found that tunnelling through related party transactions related to assets other than trade receivables caused the financial industry's accounting performance (ROA) to decline but caused its market performance (Tobins' Q) to increase. This proves that related party transactions do not necessarily raise agency conflict and have a negative effect on the financial industry, but they can also increase a company's market performance. Despite the fact that this study focuses on the financial business, the dominance of the banking sector is deemed to have influenced the research outcomes. Furthermore, this study disregards economic conditions that has the potential to affect the company's performance, such as interest rates and inflation. Future research can take into account the limitations of this study with a specific area to focus on and incorporate the concept of tunnelling from a different perspective, such as related party transactions related to expenses.