I. INTRODUCTION
Tax avoidance has been around for a very long time, but public outcry against it is firmly on the rise. The fact that companies like Google, Starbucks and Amazon use creative ways to lower the amount of taxes they pay is well known, but the last few years protests and other sorts of public scrutiny have increased (BBC, 2013). Also pressure from the European Union against countries with corporate-friendly tax rules has increased. For instance Ireland, Luxembourg and the Netherlands have faced a European commission inquiry about the tax practices in their countries (Finacial Times, 2013).
A lot of prior research has tried to explain different kinds of tax avoidance activities of firms and what its determinants are. These studies have focused on different firm characteristics and their effect on the tax avoidance activities of firms. This paper will look at the effect of the top management team size on the tax planning aggressiveness of a firm.
This research is relevant because the relation between the top management team size and tax planning has not been studied yet. Previous research about this subject has focused on the board size (Zemzem & Fhouti, 2013; Ribeiro, Cerqueira, & Brandão, 2015).
This paper seeks to answer the question: What is the influence of the top management team size within a firm of the tax planning aggressiveness? This study focusses on the fact whether there is a negative relation between the two, meaning that when the top management team size increases the tax planning aggressiveness decreases.
Results from an regression analysis done on 138 observations of US companies in the retail sector, acquired from the Orbis database, shows no significant relation between the top management team size and tax planning aggressiveness. Studies that looked at the effect of the board size on corporate tax avoidance have found mixed results, so the results of this study can also vary under different circumstances.
The structure of this paper will consist of a literature review and hypothesis development in section 2. Section 3 will discuss the method that is used and section 4 the sample selection. Section 5 will contain the results of which the conclusion can be found in section 6.
II. LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT
Different studies have been done in the past to look at management characteristics and their effects on tax avoidance activities within companies. For example different studies found there was not a significant relation between Ceo duality and tax aggressiveness (Minnick & Noga, 2010; Zemzem & Fhouti, 2013). Research about whether political orientation of managers helps explains the level of tax avoidance found that on average, companies with top executives who lean toward the republican party are less aggressive in their tax planning (Christensen, Dhaliwal, Boivie & Graffin, 2014). There has not yet been any prior research done whether the top management team size has an effect on the tax avoidance activities undertaken by firms.
Multiple previous studies on decision making within groups find that it takes more of an effort for groups that are larger in size to reach consensus. The result of this is that the decisions made by larger groups are less extreme than those made by smaller groups (Kogan & Wallach, 1966; Moscovici & Zavalloni, 1969; Sah & Stiglitz, 1986, 1991). Sah and Stiglitz make the suggestion that group decisions are a compromise that reflects the different individual opinions, resulting in the fact that larger groups are more likely to reject risky projects. Cheng (2008) shows that because of this, a larger board size resulted in a lower variability in corporate performance because the decisions of larger boards are less extreme. Adams and Ferreira (2010) also find that when betting, larger groups are less extreme. This can be linked the fact that team size is positively correlated with cognitive conflict within top management teams (Amason & Sapienza, 1997).
Dyreng, Hanlond and Mayhew (2010) find that individual executives have a significant effect on the quantity of tax avoidance activities firms undertake and thus how aggressive the tax planning activities of a firm are. This is important for this study because if just a firms CEO would have an impact and not the rest of the executive team, the effect of the number of executives on tax planning is more likely to not exist or be very small.
Rego and Wilson (2012) show that more equity risk incentive for executives led to more aggressive tax planning. In this paper they argue that tax planning activities involve risk for not only the firm, but also for the executives themselves. Meaning that tax planning activities can involve cost for both the firm and the executives. Based on the fact that larger groups are less likely to make extreme and risky decisions, it is argued that the size of the management team and therefore the number of executives has an impact on tax planning aggressiveness of a firm.
In a paper by Rego and Wilson (2008) they find that CEO and CFO compensation has a positive relation with tax planning aggressiveness. When companies pursue more aggressive tax planning activities, executives receive a higher compensation. On the other hand, Desai and Dharmapala (2006) argue that a rise in high power incentives leads to a reduction in tax sheltering activities in better-governed firms.
It is possible for executives to entrench themselves in the firm, making it more costly for the shareholders to replace them (Shleifer and Vishny, 1989). They argue that their model can be applied to anyone who values their job. This mean that companies with more executives have more problems resulting from entrenchment, because there are more mangers who can choose for this strategy. Companies with more entrenched management are less likely to manage tax and engage in tax avoidance activities (Minnick & Noga, 2010).
Desai and Dharmapala (2006) argue that were there is an asymmetry between managers and shareholders in relation to tax planning, it can help managers acting in their self-interest resulting in a negative relation between tax planning activities and the value of a firm. A survey done by Henderson Global Investors (2005) shows that there was indeed a reluctance by managers to share tax related risk management information to shareholders. This is because there is a difficulty in the measurement of actual effect of tax avoidance activities, which can cover up rent extraction by managers (Ribeiro, Cerqueira, & Brandão, 2015). In this paper they also find that companies with larger board sizes, have a higher effective tax rate. They explain this because for larger boards it is more difficult to reach consensus, thus the ability to monitor the management in weakened. This indicates that companies under better corporate governance follow more aggressive tax planning strategies. It is more difficult to monitor a management team when they consist of more executives. Because of the fact that companies with boards that have less ability to monitor management, the number of executives in a firm could have an effect on its tax planning aggressiveness.
Gallemore and Labro (2014) find that effective tax rates for firms with higher internal information quality are substantially lower than for firms with lower internal information quality. An increase in the size of the top management team will probably lead to a reduction in information quality because the information has to be shared amongst more people. This is because larger groups have more problems with coordination, communication and decision making (Lipton & Lorsch, 1992; Jensen, 1993; Yermack 1996).
Based on the above information the hypothesis of this paper will be:
H1: Firms with a higher top management team size are less aggressiveness in their tax planning activities than firms with a lower top management team size.
Having argued the relation between the top management team size and tax avoidance, a measurement for tax avoidance is required. Because of the difficulty of gathering an exact number for tax avoidance, proxy measures to estimate tax avoidance are frequently used. Often employed proxies are the effective tax rate (ETR) and the book-tax differences (Desai & Dharmapala, 2006). A this time opinions are still dispersed as to which is the best measurement for corporate tax avoidance . In this paper tax avoidance will be measured by the ETR. This is because it the most frequently used proxy when investigating tax avoidance activities of firms in researches like this one.
III. METHOD
In this study, one regression model will be used to test the previously stated hypothesis about the relation between tax avoidance aggressiveness and top management team size within a firm.
Model: TA = β0 + β1TMT + β2SIZE + β3ROA + β4LEVER + β5RDI + ε
Dependent variable: The dependent variable in this is study is corporate tax avoidance (TA). The measurement of this variable will be the measured GAAP ETR. This will be obtained by dividing the Income tax expense by the pre-tax income. Because we are measuring tax avoidance activities, the higher the effective tax turns out to be, the lower actual tax avoidance is. Thus, this the number will be multiplied by minus one for ease of interpretation. The data for this variable is retrieved from the Orbis database.
Independent variable: In this research the explanatory variable is the top management team size (TMT). The data for this variable is hand collected from the Orbis database.
Control Variables: Multiple control variables are taken into account in this paper that are consistent with prior research. These variables are namely size, return on equity (ROA), leverage (LEVER) and R&D intensity (RDI). Firm size is measured by a natural logarithm of a firm’s total assets and controls for the size effects of a firm on the effective tax rate. Multiple studies have found a positive relation between firm size and the effective tax rate (Zimmerman,1983; Omer, Molloy & Ziebart, 1993; Rego, 2003) which means that larger firms have a higher effective tax rate and therefore have less possibilities to avoid taxes. In contrast there have also been papers that find a significant negative relation between the two (Siegfried, 1972; Stickney & McGee,1982; Porcano,1986). Because of these mixed results in prior literature, no predictions regarding this relation will be made. Return on assets (obtained by dividing a firms pre-tax income by the total assets) is used as a measure for firm profitability and is used to control for the effect on the effective tax rate. For this relation mixed results have also been found in past research. Wilkie (1988) and Wilkie and Limberg (1993) both find a positive relation between pre-tax income and the effective tax rate. On the other hand both Manzon and Plesko (2002) and Rego (2003) find a negative relation between the two. For this reason we similarly will not make any predictions for this variable. Finally the last two variables this paper control for are the Leverage (Lever) and the R&D intensity (RDI). Leverage (long-term debt divided by total assets) has a positive effect on firms’ tax avoidance activities, resulting in a lower ETR (Stickney & McGee, 1982; Gupta & Newberry, 1997). The paper of Gupta and Newberry also notes this relation between the R&D intensity (R&D expenditures divided by total assets) and the effective tax rates, meaning that firms with a higher R&D intensity have a lower effective tax rate. ε is the error term of the equation.
IV. SAMPLE SELECTION
The sample consist of US companies that operate in the retail sector with more than 1000 employees with data from the year 2015. The sample is selected from within one specific industry to control for the fact that top management team size can be related to firm complexity. The reason this study looks at the retail sector is because Graham, Hanlon, Shevlin and Schroff (2013) find that executives believe that tax avoidance can have reputational damage as a consequence. Because retail companies are very much in the public limelight, their reputation is a very important factor for them and will be valued highly by their executives. Excluded from the sample were companies who recorded a net loss in 2015. Also not included were firms with a negative effective tax rate based on tax refunds or with a effective tax rate higher than 1. Lastly companies for which not all the data is available were eliminated from the final sample. The size of the final sample is 138 observations.
V. RESULTS
In Table 1 the descriptive statistics are shown of the different variables in this paper. The key variables are the tax avoidance (TA) which is the effective tax rate times minus one and the top management team size (TMT). The tax avoidance has a mean of -,3578 which shows that in the sample the companies have an average effective tax rate of 35,78 percent. In the sample the highest effective tax rate was 82,98 percent and the lowest was 0,3 percent. The average top management team size was 8,83, with a minimum of 2 and a maximum of 50.
Table 1: Descriptive Statistics
N Mean Minimum Median Maximum Standard deviation
TA 138 -,3578 -,8298 -0,3673 -0,0030 0,1113
TMT 138 8,83 2 7,50 50 6,614
SIZE 138 21,7764 18,71 21,6307 40,998 1,3950
ROA 138 0,1154 0,0048 0,0968 0,4100 0,0798
LEVER 138 0,2149 0,00 0,1741 1,3567 0,2047
RDI 138 0,0038 0,00 0,00 0,1916 0,0200
Table 2 shows the Pearson correlation matrix. The dependent (the ETR minus one) and independent variable TMT show a small and positive correlation, but it is insignificant. A high correlation (0,605) can be found between the firm size and the top management team size. This correlation is also significant at a 0,01 significance level. Another significant correlation is between the firm size and the firm leverage (0,263)
No correlation of over 0,8 has been found, which indicates that there are no multicollinearity problems in this model.
Table 2: Correlation Matrix
TA TMT SIZE ROA LEVER RDI
TA 1
TMT 0,043 1
SIZE 0,145 0,605** 1
ROA 0,039 -0,040 -0,061 1
LEVER 0,100 0,070 0,263** -0,092 1
RDI -0,102 -0,041 0,131 -0,073 -0,010 1
** correlation is significant at 0,01 level
As an extra test for multicollinearity the VIFs were calculated. None of these values exceeded 10, so there is again no sign of a multicollinearity problem.
Table 3 shows the results of the regression analysis. The results show a small negative relation between the top management team size and the tax avoidance 0f -0,001 (meaning that for every extra executive in the top management team the ETR increased by 0,1 percent in this model). However this number is insignificant (p = 0,423). Because of this there is no evidence supporting H1.
For the control variables only SIZE shows a significant result for a significance level of 0,1 (β = 0,016, p = 0,423). Because the relation is positive larger corporations engage in more tax avoidance than smaller corporations. All other control variables prove insignificant, so nothing can be said about the effects of these variables on corporate tax avoidance. The relation between ROA and tax avoidance is also positive, but as stated before proved insignificant. In contrast to previous literature the there is a negative relationship between R&D expense and tax avoidance.
Table 3: Regression Results
Variables Prediction Coefficient p-value
Constant -0,708 0,000
TMT – -0,001 0,423
SIZE ? 0,016 0,076
ROA ? 0,060 0,615
LEVER + 0,030 0,535
RDI + -0,715 0,143
N = 138
R2 = 0,046
VI. CONCLUSION
The aim of this study is to investigate whether firms with a higher top management team size are less aggressive in their tax planning than firms with a smaller top management team size. Past research about this subject is narrow, so this study will contribute to the growing literature trying to explain to determinants of corporate tax avoidance activities.
The expectation of this study is that companies with a bigger top management team are less aggressive in their tax planning. However, this paper finds no evidence supporting a negative relation between top management team size and corporate tax avoidance. This is because the effect of the top management team size on tax avoidance is insignificant. So it cannot be concluded that top management team size has any effect on a firm’s tax planning aggressiveness.
From the control variables in this study only size shows a significant relations with tax avoidance with a significance of 0,1. The relation is positive meaning that a larger firms size leads to more aggressive tax planning. The other control variables Return on Assets, Leverage and R&D expenditures all proved insignificant in this study, so no relation between these variables and corporate tax avoidance can be concluded.
Some limitations of this research must be noted so these can be used for further research. One limitation of this study is the small sample size. The results of larger sample will be more reliable and the use of a bigger sample with more observations might lead to different results. Another constraint of this study is that it only looks at one industry. Comparing multiple industries might lead to different results for the different sectors instead of just looking at the retail sector as in this study. Another drawback is the reliability of the data, which is mostly the case for the top management team size. For this the data was hand collected from the Orbis database. This database is a large independent database which is often used for research, but of course there is no guarantee that the data is completely accurate, mostly with the non-financial data. Finally the last limitation is the fact that this study just looked at one measurement for corporate tax avoidance, the effective tax rate. Looking a different measures for the dependent variable might lead to different results in future research.
...(download the rest of the essay above)