Tax Aggressiveness, Tunneling and Firm Value: Interactive Effect of Corporate Governance in Business Groups

This study extends the literature on tax aggressiveness (TA) from agency perspectives in the rarely discussed case of group firms. More specifically, the study investigates the relationship between firms TA, tunnelling and value of firm. Moreover, the study also investigates the impact of the moderating role of corporate governance (CG) in counterfeiting the conflicts of interests in group firms. For this purpose, sample data of 160 non-financial Pakistani firms belonging to groups for the period from 2009 to 2018 is analyzed through two Stages Least Square Regression (2SLS) models. The findings reveal that tunneling, group ownership and managerial ownership show direct association with TA. While CG agents, board and audit committee independence and external audit quality exhibits an indirect relationship with TA. Moreover, the tunneling-related TA has a negative effect on the firm value. However, the estimated interaction effects show that CG mitigates this negative relationship between the tunneling-related TA and firm value. Thus, good CG ameliorates the expropriations by the controlling shareholders and TA becomes a value enhancing activity in business groups.


Introduction
In general governments take more than 30% of a firm's pre-tax income and Pakistan is no exception. This gigantic firms' tax cost could influence managers to devise plans to mitigate the cost and enhance profits after tax (Chen et al., 2008). Looking through the lens, managers as stewards, it is not irrational to assume that managers' aggressive tax planning will contribute towards the cherished goal of shareholders wealth maximization (Scholes et al., 2009). However, agency theory predicts that these plans might translate to contribute towards furthering managerial opportunism because it might reduce corporate transparency (Desai & Dharmapala, 2006). Further, if a firm belongs to some business group then it may minimize its tax cost at the group level by shifting profits within a group (Choi, Sami & Zhuo, 2010). To this end, Jung, Kim and Kim (2009) reported that the marginal tax rate of group firms is lower relative to that of non-group firms of the same industry and size due to aggressive tax avoidance. Similarly, Klassen, Lang and Wolfson (1993) in the case of multinational corporations operating in countries with high and low tax rate countries found that these firms, through related party transactions (RPT), transfer their funds from economies with high tax rates to those with low tax rates (Ko, 2000).
Existing studies propose number of diverse factors that are associated with the TA behavior of firms. These factors are size of firm, international operations, leverage, capital intensity, executive compensations and political affiliation of firms (see e.g., Richardson & Lanis, 2007;and Richardson & Lanis, 2007). However, these studies ignore the aggressive tax behaviors that have potential non-tax agency costs such as regulatory or reputational (Scholes, Wolfson, Erickson, Maydew, & Shevlin, 2005).
The agency perspective of corporate TA issues is interlinked with corporate governance (CG). From the agency perspective, cash saved from TA activities is transferred by the firms' controlling shareholders having low cash flow rights to a firm in which these shareholders have high cash flow rights (Desai & Dharmapala, 2006). Kim, Quinn, and Wilson (2016) documented a positive (negative) association between managerial ownership and TA (tax expenses paid in cash). An indirect association between corporate TA and family ownership is also reported (see e.g., and Landry, Deslandes, & Fortin, 2013).
Existing literature suggests that CG mechanisms mitigate agency problems and empirical studies report significant positive effect of CG on value of firm and tax avoidance (see e.g., Desai & Dharmapala, 2009;and Yee, Sapiei, & Abdullah, 2018). Number of studies show that the negative effect of corporate TA on value of firm is mitigated through better CG practices in placed by the firms. Arif and Hashim (2013) found that the value relevance of corporate TA is relatively higher for well-governed firms. Large number of studies report that independent boards with sufficient financial compensations, independent audit committees, separation of the role of chairman and CEO, quality of external audits and close monitoring by creditors and regulatory authorities discipline the expropriating behaviour of controlling shareholders (Cheung et al. 2006;Gao & Kling 2008;Nekhili & Cherif, 2011;and Ryngaert & Thomas, 2012).
Most of the work on this topic is done in developed countries where dispersed ownership with conflict between managers and shareholders or in China where government act as a controlling shareholder (Pilos, 2017;and Kim, Li & Zhang, 2011).
The current study takes this literature a step further and tends to investigate the implication of non-tax costs incurred due to aggressive tax avoidance on value of firm in the case of group firms. This study is conducted in an institutional setting where business groups act as controlling shareholders of firms and are reported to be aggressively involved in tunneling-related activities (see e.g., Ullah & Shah, 2015). The relationships among TA, CG, and firm value are largely unexplored in Pakistani group firms. Earlier, Shabbir, Waheed, and Mahmood (2017) examined the relationships of firm size, growth, debts and inventory changes, ROA, TA and tax avoidance. Akram, Iqbal, and Mughal (2012) followed exploratory research design and considered macroeconomic factors that affect corporate tax. They explored that lack of enforcement, lower penalties, lack of public awareness, and high compliance cost are the factors that discourage TA. Whereas, the current study is entirely different from these aforementioned studies in a way that it follows an agency perspective of corporate TA and the impact of tunneling-related TA on firm value and further analyzes the moderating role of CG in counterfeiting the agency conflict arising from TA in unique institutional settings of business groups.The unique institutional setting is considered the most suitable to conduct this study and to analyse the behaviour of the business groups in managing the tax at group level within the agency framework and determining its effect on value of firm in the presence of CG mechanisms. Thus, following an agency framework, this multi-prong study contributes to the literature and broadens the understanding of tax-sheltering in family-dominated and business groups firms. The study tests that if tax sheltering activities are benefiting the overall shareholders or if the resources saved through such activities are tunneled out by the controlling shareholders. The study also assesses the contribution of the CG agents in constraining the expropriations of saved tax through tax sheltering activities.

Literature Review
The extant literature highlights different terms for the TA such as tax avoidance, tax planning, tax management, and tax sheltering. According to Dyreng, Hanlon & Maydew (2010) the management practices whereby firms pay lower tax per dollar of the pre-tax accounting income is known as TA. Whereas, Frank, Lynch and Rego (2009) stated that TA is tax planning that involve manipulating and reporting reduced accounting income and that may or may not be fraudulent tax evasion. Similarly,  also stated that TA could involve legal and illegal practices. Whereas, Hanlon and Heitzman (2010) describe TA to include activities such as investments in municipal bond, noncompliance and tax evasion. As such TA is multifaceted phenomenon that influences firms' tax payment.

Tax Aggressiveness and Tunneling
The traditional theory of finance assumed that managers undertake TA activities to maximize shareholders wealth (Scholes, et al., 2009). The tax saved may be reinvested or given to the shareholders. However, agency theory predicts that in the case of group firms, cash saved from TA activities might be transferred to other firm(s) by the controlling shareholders to protect self-interests (Desai & Dharmapala, 2006;Desai et al., 2009). There is an opportunity for firms belonging to business group to play down their tax liabilities through RPTs (Choi et al., 2010). Findings of empirical studies show that firms belonging to group shift income and expropriate resources within the group (Ullah & Shah, 2015). Other studies provide more direct evidence of TA by multinational corporations (MNCs); MNCs are reported to shift incomes to business units in other countries and exposed to lower tax rates (Klassen, Lang & Wolfson, 1993;Ko 2000). As revealed in the literature, in economies with concentrated or family ownership of firms the conflict between principals-principals are of major concern (see e.g., La Porta et al., 1999). Further, utility maximizer majority shareholders can influence the management and will expropriate wealth at the expense of minority shareholders through RPTs due to their more cash flow rights (Jensen & Meckling, 1976). Empirical studies document that controlling shareholders with excess power are involved in self-dealing such as TA, excessive compensation, personal loans, transfer pricing, corporate opportunities, outright expropriation of assets and profits (Desai & Dharmapala, 2006;Djankov et al., 2008;Shleifer & Vishny, 1997). Few studies documented value destroying effect of TA because cash saved through TA is expropriated by the controlling shareholders through RPTs at the expense of other shareholders Desai & Dharmapala, 2006;and Ryngaert & Thomas, 2007). Thus, RPTs and tax avoidance will relate positively on the conjecture that the saved tax could be expropriated by the controlling shareholders in group firms. H1: There is positive relationship between RPTs and TA in group firms.

Ownership Structure and Tax Aggressiveness
Firms' structures of ownership and agency problems have association (Shleifer & Vishny, 1997). Majority of the Pakistani firms are family owned, have concentrated ownerships and belong to some groups (Abdullah, Shah, Gohar, & Iqbal, 2011). Family controlled associated firms have concentrated ownership either due to the weaker enforcement of property rights or due to the absences of specialized institutions in the capital market (Fan & Wong, 2005;and La Porta et al., 1997). The possible nexus between family members and the opportunistic could spin the TA activities in the favour of the associated firms. Given the majority and minority shareholders conflict, it cannot be ruled out that the later could discount the stock prices coupled with other tax avoidance cost. However, due to lesser dependence on the capital market and lesser reputational impact in case of penalty imposed, associated family-controlled firms are expected to gain more benefits from TA actions in countries like Pakistan with relatively poor or inefficient law enforcement. Literature reveals that controlling shareholders with the control of the group can shift or share resources within a group, which benefits those firms operating in developing economies (Khanna & Yafeh, 2005;Ullah & Shah, 2015). Moreover, the existing literature documented other benefits associated with business groups such as sharing financial resources, financing needs of newly established firms, TA within group firms (Almeida & Wolfenzon, 2006;and Ryngaert & Thomas, 2007). Therefore, it is hypothesized: H2: Associated ownership has positive association with TA in group firms.
Extant literature document the relationship between TA and family ownership , private equity ownership (Badertscher, Shroff & White, 2012) and dual-class ownership (McGuire et al., 2014). However, there is no study of managerial ownership and TA in the case of group firms. Desai and Dharmpala (2006) suggested that manager execute complex TA transactions that extract rents from shareholders. Thus, the tax saved through TA is used by the opportunistic managers for their private benefits. Kim, Quinn, and Wilson (2016) and Rego and Wilson (2012) found that an increase in managerial ownership increases TA. The possible collusion between majority shareholders and managers also provide the opportunity of sharing the saved tax pie. Hence, it is hypothesised that: H3: Managerial ownership and TA exhibit positive relationship in group firms.

Tunneling, Tax Aggressiveness, and Firm Value
Theoretically, RPTs can have either positive or negative effect on the wealth of minority shareholders and value of firms (Gordon, Henry, Louwers & Reed, 2007). On one hand, the "conflict of interest hypothesis" argues that RPTs negatively affect firms' performance and value due to the existence of conflicts of interests between principals and agents or principals and principals. On the other hand, the "efficient transactions hypothesis" argues that due to inherent expected efficiency, RPTs could improve performance and value of group firms. Wang (2011) investigated the role of corporate transparency in TA and value of firm and concluded that tax sheltering dilute value of firm in opaque firms and positive influence in case of informational transparent firms. The resulting lack of transparency due to the complex structure and complex nature of transactions suggests that that there will be negative association between TA and value of firm in case of group firms.
Empirical studies about the effect of TA on the value of firm have mixed results. For example, Gupta and Newberry (1997) found that firm profitability, liquidity and leverage are positively associated with TA. In the support of a positive effect of tax sheltering activities and firm value, Wilson (2009) reported that market returns were higher for the firms involved in TA practices. Others, such as Arif and Hashim (2013), Moradi, Mohammadi, and Saeedi (2015) also found that TA enhances firm value. However, Cloyd, Mills, and Weaver (2003) found no significant association between TA and firm value in the case of US firms. In contrast, Hanlon and Slemrod (2009) found that the market responds negatively to the firms involved in aggressive TA. Similarly, Desai and Dharmapala (2009) found a significant and negative effect of TA on firm value. Their findings are in line with the agency perspective that information asymmetry could favor controlling shareholders to engage in TA activities Therefore, tax saved from tax sheltering activities may not increase shareholders wealth due to the expropriation by the major shareholders or opportunistic managers at the expense of minority shareholders. In this study, it is hypothesized that: H4: TA is inversely associated with the value of firm in group firms.

Corporate Governance, Tax Aggressiveness, and Value of Firm
Shleifer and Vishny (1997) describe CG as "the system of control mechanisms, through which the suppliers of finance to corporations assure themselves of getting a return on their investment." Thus, CG is value maximizing framework that preserves interests of all stakeholders including minority shareholders and mitigates agency costs. Desai and Dharmapala (2009) opined that the influence of TA is value enhancing (destroying) for well-governed (poor governed) firms. Similar results are reported by Arif and Hashim (2013) and Yee et al., (2018). However, the later further added that tax saved is expropriated by the controlling shareholder and the weak CG couldn't benefit the minority shareholders in Malaysian firms. Moreover, Chan, Mo, and Tang (2016) suggested that the value relevance of TA reduces for the firms having high association between TA and tunnelling.
In Pakistani business groups with family dominance and expropriations by the controlling shareholders (Ullah & Shah, 2015), the TA is expected to be used for tunnelling. Tax saved will be beneficial for the controlling shareholders but of no value to the minority shareholders. Hence, TA shall negatively affect value of Pakistani firms. The mitigating effect of CG here is an empirical issue. It might partially or completely control the negative impact of TA on value of firm. Contrary to this view, "efficient transactions hypothesis" will predict that RPTs, not meant to expropriate resource, might increase the performance of Pakistani group firms with family dominance. Thus, tax saved through TA might bear shareholders' wealth maximizing effect. Following the "conflict of interest hypothesis", it is hypothesized that the relationship between TA and value of firm is moderated by the level of CG factors in firms belonging to business groups. The current study considered the following three main pillars of CG.

Board independence, TA and Value of Firm
An effective board of directors composed of independent (neutral) members with ample experience and repute would effectively safeguard the interests of all stakeholders (Khaoula & Moez, 2019). Due to their effective monitoring, the boards in group firms are expected to control expropriation and rules violation-oriented TA activities and shall positively influence value of firms (Adams & Ferreira, 2007;and Masulis & Mobbs, 2011). On the other hand, it can be argued that independent members who are supposed to be more concerned about their repute shall disallow tax planning practices by the firms or controlling shareholders having political powers might negatively influence the independent role of the board members. Hence, an indirect effect of the board independence on the TA and value of firm is also possible. We hypothesize that: H5a: Board independence positively influences the association between TA and value of firm in group firms.

Independent Audit Committee, TA and Value of Firm
Expert, independent member directors in the audit committee, having incentives, will increase monitoring effectiveness and hence quality of the financial information of these firms by sustaining any influence of the firms' management (Beasley, Carcello, Hermanson & Neal, 2010). In fact, the auditors are expected to have the expertise to identify the complex tax avoidance practices in the perplex management structure of group firms. A US based study by Dhaliwal, Naiker, and Navissi (2010) reports that accounting frauds are controlled by fully independent audit committee. Under the premise that TA is a tunneling activity by the controlling shareholders, an independent audit committee will ensure that the saved tax is not expropriated at the expense of the minority shareholders. Therefore, an independent audit committee will positively influence the association between TA and value of firm. However, auditors also provide advices on managing tax planning risk (McGuire, Omer & Wang, 2012). Thus, in the presence of expert and independent audit committee, shareholders might perceive TA as value enhancing practice rather than an opaque action (see Dhaliwal et al., 2010). Here, we hypothesize that: H5b: Independent audit committee will have mitigating effect on the value destroying impact of AT in group firms.

Eexternal Audit from Big4
Experience, expertise, repute, presumed greater independence and accompanied incentives are some of the reasons due to which big audit firms are expected to offer better monitoring of the accounting information of firms. Audit by big firms shall reduce the chances of presumed controlling shareholders driven TA. Moreover, it shall positively influence the TA and value of firm association because the quality of the financial information in the presence of big audit firms could be highly valued by the investors. In addition, the advising role on tax planning also suggests a positive impact of TA on value of firm. We hypothesize that: H5c: In group firms, audit by big4 is expected to positively affect the association of TA and value of firm.

Data Collection and Sample Size
This study used a sample of 160 firms belonging to different business groups and listed in Pakistan Stock Exchange for a period from 2009 to 2018. Data of TA measures, CG, ownership, and RPTs and other variables is extracted from the companies' published annual reports available on their respective websites. The sample size is primarily determined by the availability of data on the variables of interest for the sample period.

Regression Models and Estimation
The study uses panel method to estimate the models used in the study of Yee et al., (2018). The study used 2 Stage Least Square (2SLS) regressions wherein the 1 st Stage equation the dependent variable is DETR and the independent variables include CG factors and firm-specific variables. It is least expected that all firms have same level of TA e.g., intangible assets intensive firms may engage in more TA, hence endogeneity cannot be completely ignored. The 2SLS is expected to control the issue. Two instrumental variables, INST (intangible assets scaled by total assets) and ATR (applicable tax rate) are added to eq -2.

Measures of Tunneling
This study examines that controlling shareholders of business groups use RPTs to tunnel the saved tax. Therefore, tunneling is measured through related party receivables and related party loans both scaled by total asset (RPR) (Jiang, Lee & Yue, 2010;Ullah and Shah, 2015).

Measures of Tax Aggressiveness
The study follows Amiram, Bauer, and Frank (2013) and Inger (2014) to compute TA as the difference of effective tax rate (ETR) and applicable tax rate (ATR). ETR is current income tax expense to current pre-tax income. The other two measures of TA that are used for robustness check are DETR2 and DETR3; where DETR2 is provisions of income tax scaled by pre-tax income (Gupta & Newberry, 1997;Hanlon & Heitzman, 2010), and DETR3 is income tax divided by cash flows from operations (Annuar, Salihu & Obid, 2013).

Measure of Firm Value
Value of firm is measured through the frequently used Tobin's Q (Wang, 2011;and Inger, 2014).

Measures of Corporate Governance and Ownership Structure
CG elements include board independence, independent audit committee and external audit from Big4 1 . Board independence is determined through the numbers of outside directors divided by the total directors (Tandean & Winnie, 2016). The independence of the audit committee is the ratio of non-executive members to total members of the audit committee (Tandean & Winnie, 2016). It is expected that an increase in the numbers of outside member will increase the independence of board and audit committee. The higher independence of these two bodies could lower the likelihood of diverting the saved tax from TA activities and are expected to positively affect value of the firm. An external audit from big4 is valued as 1 if the firm auditor is from big4 otherwise it is coded as 0 (Ullah & Shah, 2015). The engagement of Big4 would signal that value enhancing tax sheltering decisions would be made and diversion of the saved tax by the controlling shareholders would be refrained. Managerial ownership is the percentage of shares held by the directors and their family members out of total shares and associated firm ownership is the shares held by other firms or affiliated firms (Abdullah et al., 2011).

Control variables
Following the literature, this study includes various control variables to account for their effects on the TA and firm value. Firm size and growth of firm is computed as log of total assets and changes in fixed assets respectively. Leverage is represented by debts to equity ratio and ROA is computed as the ratio of net income to total assets (Sari, Utama & Rossieta, 2017;and Yee et al., 2018).

Results and Discussions
This section includes discussions on the results of descriptive analysis, correlation, and panel regression analysis.

Descriptive Statistics
On the average firms have 18.9% related party receivables out of the total assets, 7.6% difference of receivables to payables scaled by total assets, and 4% related party loans out of total assets. These statistics show that a significant amount of assets are transacted through RPTs by the group firms. The DETR(s) represents the difference of the tax amount that group firms were required to pay and what actually these firms have paid. The positive average value of DETR(s) suggests that the group firms have practiced TA. The average value of the DETR1 is 0.138, DETR2 is 0.183 and DETR3 is 0.204. The ownership structure variables show that on average managerial ownership is 25.5% and associated ownership 34.5% out of total shares of firms. On the average 40% of the group firms take services from Big4 audit firms. About 66% of the directors are non-executive directors where audit committee includes two non-executive directors. Further, on the average firm assets show 5.7% growth, the average size of firms is 15, whereas, the debt to equity ratio has an average value of 0.65 which suggests that 0.65 out of 1 rupee is used to finance assets. The average ROA is 6.9% and the average value of Tobin's ratio is 1.791 which shows that group firms have high market value relative to their book value. Table 4.2 reports no case of high correlation among the variables and rules out the issue of multicollinearity. The measures of TA (DETR1, DETR2, DETR3) and RPTs (RPR, RPloan) exhibit positive correlations between them. This association of RPTs with TA suggests that the former is used to save tax. Furthermore, managerial and associated ownership are positively related to the different measures of RPTs and TA. As such it can be inferred that managerial ownership in Pakistani group firms might be a mean to collude with the controlling shareholders. As expected, the three CG variables are inversely linked to both RPTs and TA. Similarly, growths, ROA and debts to equity exhibit an inverse relationship with the measures of RPTs and TA. Finally, RPTs is inversely related to Tobin's Q but TA show evidence of a positive correlation with the value of firm. Table 4.3 shows results of the 2SLS models with TA and Tobin's Q as the explained variables. The results of the 1 st stage model of TA show a positive effect of the RPRA on TA. An increase in RPRA motivates the group firms to aggressively engage in tax planning (Chan et al., 2016;Jiang et al., 2010). This indirectly implies that the saved tax could be diverted by the controlling shareholder to other group firms through RPTs. In fact, the positive association between RPTs and TA is indicative of the incidents of the income diversion from a firm with probably a high tax rate in the group to a firm with probably a low tax rate and hence the findings are in line with Jung et al., (2009) who reported lower pre-tax return on assets for group firms with more tax burden.

Tax Aggressiveness, Tunneling and Firm Value
As expected, managerial and associated ownership has direct significant effect on TA (Ryngaert & Thomas, 2007). Thus, firms with relatively more managerial and associated ownership are most likely to engage in TA activities (Kim et al., 2016). Desai and Dharmapala (2006) suggested that TA facilitates managerial rent extraction and could strengthen its opportunistic behavior. The CG variables such as board independence, independence of audit committee and audit from Big4 show statistically significant inverse relationship with TA in all the estimated models. These findings are similar to those documented by Jiang et al., (2010) andLi et al., (2016). Thus, increase in the outside members in board and audit committee would improve governance and reduce the aggressive tax opportunism that would risk the minority shareholders wealth. Among the control variables growth, leverage and ROA have a negative influence on the TA activities, whereas, size of firm shows a direct and significant influence on TA activities in the sample firms. Thus, it is argued that to preserve reputation and image growing firms and firms with higher level of leverage would have lower level of TA practices (Jiang et al. 2010). However, relatively large firms are more engaged in aggressive tax sheltering due to their large volume and sophisticated transactions.
The results of the Tobin's Q models show that all the three measures of TA i.e. DETR1, DETR2, and DETR3 have an inverse and significant relationship with the value of firm. As expected, the results imply that TA is conflicting in nature in the Pakistani group firms and perceived as value reducing activity by investors. These results supports the view that TA increase after-tax cash flows, however its effect on firm value is diluted, particularly in group firms, due to the possible expropriation of saved tax by the controlling shareholders (Desai & Dharmapala, 2006. Likewise, perceived as a mean of expropriation by the investors, RPRA shows an indirect influence on the Tobin's Q (Gao & Kling, 2012;Ullah & Shah, 2015). In line with the entrenchment hypothesis, both, managerial and associated ownership are inversely related to the Tobin's Q. Therefore, following the results of Ullah and Shah (2015), it is argued that increase in associated ownership or managerial ownership enhance controlling power in other group firms and the controlling shareholder, in self-interest, thus tunnel out resources through RPTs that eventually decrease the firm value (Gao & Kling, 2012;and Ullah & Shah, 2015). Consisting with the prior studies, firm growth, leverage and applicable tax rate have negative effect on Tobin's Q, and firm size and intangible assets have positive and significant effect on Tobin's Q (Chan et al., 2016;Yee et al., 2018).
The moderating effect of the CG variables between the relationship of TA and value of firm is determined through interaction term of the governance and TA variables. 2 As reported in Table 4.4, the interaction of CG variables with TA has a positive and significant effect on the firm value. These findings demonstrate that in the presence of more independent members on the board, independence of audit committee, and Big4 auditor, the effect of the TA on the firm value becomes positive. Even in the case of group firms, these measures of CG constrain expropriation of the saved tax through tax sheltering activities, improve the overall governance, and hence improve the firm value (Chan et al., 2016;Jiang et al., 2010;Li et al., 2016). These results are similar to the findings documented by Arif and Hashim (2013), Yee et al., (2018), and Desai and Dharmapala (2009). They reported that case of well-governed firm, there is a significant positive influence of TA on firm value; the results of all other variables remain consistent with the results of the baseline model. From these findings we conclude that presence of good CG mechanisms is perceived by investors to reduce the probability of expropriations and shifting resources from a group firm to another by the controlling shareholders.

Robustness Check
To check that these findings are robust and remain consistent for another type of RPTs; this study also considered related party loan scaled by total assets and examines its effect on TA and firm value (RPloan). Table 4.5 presents the results of 2SLS models. The only changed proxy is RPloan in the model. The results of RPloan are not different from RPRA, the earlier proxy of tunneling. More specifically, an increase in RPloan increases TA in the Pakistani group firms because controlling shareholders might pressurize managers to aggressively engage in the tax sheltering activities (Jiang et al. 2010;Sari, Utama & Rossieta, 2017). Chan et al., (2016) also reported that RPloan is used to divert saved tax and further showed its detrimental effect for the minority shareholders. Frank et al., (2009) suggested that information risk from TA deepens the uncertainty of firms which confers on the controlling shareholders greater flexibility in expropriating resources. Therefore, following the results of Sari et al., (2017) we also concluded that increase in the related party loan would increase the demand for aggressive tax sheltering activities in group firms in Pakistan and this prediction is remained consistent with the baseline model.
Similar to the results of the base line regression, the ownership variables, managerial and associated ownership, are directly and statistically significantly influencing the various measures of TA; whereas, CG variables, board independence, independent audit committee, and audit from Big4 exhibit the same negative and significant effect on TA. The results of other control variables also exhibit no inconsistency such that firm growth, leverage, and ROA have negative influence and size of firm shows a direct effect on TA.

Conclusion
To the extent aggressive tax planning could reduce tax costs, the shareholders would prefer it. However, this argument overlooks the prospective non-tax costs associated with TA, say for example, agency related costs. Firms owned or controlled by business group or family are exposed to conflict of interests between majority and minority shareholders. This study analyzed data of 160 Pakistani group firms through 2SLS estimation to find impact of RPTs on TA and the later impact on the value of firm. In addition, the moderating role of measures of CG on the relationship between RPTs and TA is analyzed. It is found that RPTs, associated firm ownership and managerial ownership have direct influence on the TA, whereas, board and audit committee independence and audit from Big4 have negative effect on the TA. Moreover, the TA and RPTs show indirect impact on the value of firm while controlling for ownership and CG variables. The interaction models show that CG variables positively moderate the TA and firm value relationship. We conjecture that good CG practices are evenly important in the case of group firms. In these firms, as perceived by investors, the indirect effect of TA on the value of firm is significantly mitigated by employing good CG practices. Related party lending's measured through related party receivables scaled by total asset and related party loan scaled by total assets. (DTER1) is the difference of effective tax rate (ETR) and applicable tax rate (ATR), whereas, effective tax rate is the ratio of current income tax expense to current pre-tax income. DETR2 is the effective tax rate is computed as provisions of income tax divided by pre-tax income, DETR3 is the effective tax rate is computed as the ratio of income tax to operating cash flows. Tobin's Q is computed as the market value of equity plus book value of liabilities divided by total assets. Board independence (NED) is computed as the ratio of outside directors to total directors. The independence of the audit committee (IAC) is computed by the ratio of non-executive members to total members of the audit committee. An external audit from big4 is coded as 1 if the firm auditor is from big4 otherwise it is coded as 0. Firm size is measured by the log of total assets, firm growth is measured by changes in fixed assets, DE represents leverage which is measured by the ratio of debts to equity and ROA is computed as net income divided by total assets.