Exploring the determinants of Tax Avoidance: An empirical analysis of
Corporate Social Responsibility, Financial Constraints and Culture
Abstract: This paper examines the relation between corporate social responsibility (CSR) and tax avoidance. Previous literature found contradicting results regarding CSR and tax avoidance, and it is therefore unclear whether the payment of taxes is a component or a substitute of CSR. This paper extends the literature by using a global sample and including two new, moderating variables, namely financial constraints and national culture. Weak evidence is found concerning the negative relation between CSR and tax avoidance. Additionally, this paper cannot find any difference in this relation between CSR and tax avoidance, when it comes to corporations that are financially constraint. Nevertheless, the relationship does seem to differ across culture. Corporations headquartered in countries with higher values on individualism, masculinity and uncertainty avoidance, have a weakened relation between CSR and tax avoidance. This would imply that the negative relation between CSR and tax avoidance is weakened in countries characterized as individualistic, masculine and/or intolerant towards uncertainty.
Student number: S2979071
Name: Stefan Bovenhuis
Study programme: MSc International Financial Management
2
I.
Introduction
The globalisation of business created opportunities for businesses, consumers and countries to experience economic growth. One of those opportunities for businesses was the strategic allocation of costs, to avoid payment of taxes. The avoidance of taxes has been a useful tool to decrease a corporation’s costs while not directly affecting its performance, which would
have been affected if costs were cut in other departments (e.g. decrease in marketing could lead to lower brand recognition and therefore lower performance). Besides, tax avoidance is perfectly legal. However, there is also a downside of globalisation, as international competition got fiercer. The increased competition created an environment where corporations sole aim was to decrease (increase) their costs (profits). Meanwhile, ethical principles got violated as firms paid less attention towards environmental and societal issues. Tax avoidance can be regarded as one of those societal issues. To elaborate on, corporate tax income is one of the strongest indicators for a country’s economic health. Therefore, the avoidance of corporate taxes affects a country’s health directly (Friese et al, 2008). Consequently, some argue that the payment of taxes is an important way for corporations to give back to the society where they are operative (Christensen and Murphy, 2004). Corporations have to make a comparative assessment to see whether they want to maximise profits or whether they want to act as good corporate citizens by paying their fair share of taxes. This comparative assessment leads to intensified attention of stakeholders towards the actions and movements of corporations. Consequently, the importance of corporate social responsibility (CSR) increased. Additionally, the comparative assessment of CSR and tax avoidance leaded to fierce discussion in academic research.
3
firm engages in a more corporate socially responsible manner they will pay their fair share of taxes, or at least they should be, implying that there would be a negative association between CSR and tax avoidance (Park, 2017; Hoi et al., 2013). Moreover, the gross part of the existing literature found their findings in a single-country sample. Taking these facts into consideration, this paper tries to extend the literature by giving a more conclusive answer based on a global sample. Besides, it uses a new proxy for CSR, namely a score created by Thomson Reuters based on a firm’s performance on Environment, Social, and Government, hereafter referred to
as ESG. Additionally, this paper continues the exploration of tax avoidance by researching new moderating variables, namely financial constraints and national culture.
To test the relation between CSR and tax avoidance, and the moderators financial constraint and national culture, this paper employs multiple regressions, using 19,168 firm-year observations gathered from the Thomson Reuters Asset4® database. The sample consists of firms headquartered in 47 different countries, over the period 2009-2017. The performance of CSR is measured by using the ESG score provided by Thomson Reuters’ Database. Corporate Effective Tax Rate, hereafter referred to as CETR, is used as the proxy for the dependent variable tax avoidance, while this paper uses the Kaplan-Zingales index for financial constraints and Hofstede’s cultural dimensions for national culture, respectively.
4
effect of a country’s culture where the firm is headquartered. In line with what is hypothesized, this paper finds individualism, uncertainty avoidance and masculinity to be significant moderators of the relation between CSR and tax avoidance. All three dimensions have a negative moderating effect, implying that these cultural dimensions weaken the negative relationship between CSR and tax avoidance.
This paper is contributing to management and accounting literature in several ways. First, using a global sample to test the association between CSR and tax avoidance has never been done to my best knowledge. This paper finds partial evidence that corroborates the negative relationship between CSR and tax avoidance. These findings strengthen the view of CSR being a core activity for corporations. Besides, it supports the perspective towards CSR as an important tool for a corporation’s tax-positioning strategy. Moreover, it confirms the importance of corporate citizenship, which implies that when a firm gains from the privileges from society, it has the obligation to give something in return and pay its fair share of taxes. Secondly, this paper contributes to the existing literature by exploring new moderating variables. Some of these interactions resulted to be of significant influence. Moreover, evidence is provided for national culture to be a moderator in the relationship between CSR and tax avoidance. These additional identified determinants can be important proxies for policy makers who seek to identify in which circumstances it will be more likely that firms engage in tax aggressive strategies. This paper shows that awareness needs to be raised regarding CSR and national culture as determinants of tax avoidance.
II.
Literature review and hypothesis development
A. Corporate Social Responsibility and Tax Avoidance
5
the payment of corporate taxes has consequences for the wider community (Avi-Yonah, 2008). Otherwise, taxes would have been just a regular business transaction and thus a cost for the firm when the firm would not have been minimizing it. Consequently, this paper builds on previous literature arguing and assuming that tax aggressive practices have negative effect on a country’s education, public health care and public transport (Freedman, 2003; Landolf, 2006; Friese et al., 2008), and therefore has an effect on the wider community. Furthermore, this paper does not make any distinction between tax avoidance and tax evasion, following Dyreng et al. (2010). The aim of this paper is not to determine whether it is tax aggressiveness, tax risk or tax evasion, but to investigate the broader concept of tax avoidance, concerning anything that lowers the CETR.
Whilst both CSR and tax avoidance are heavily researched individually, the two have not been researched often together. This paper uses two possible broader theories to forecast the possible association: the corporate citizenship theory and the shareholder theory. The
corporate citizenship theory can be used to describe the negative relation between CSR and tax
avoidance, arguing that when a firm intents to perform the right course of action regarding CSR, it should do so for all CSR related matters, including tax avoidance (Kreps, 1990). This idea is enhanced by Christensen and Murphy (2004). They argue that tax planning is the most tangible and important component of corporate citizenship. To elaborate on, the payment of taxes is the most fundamental way in which corporations engage with broader society, according to Christensen and Murphy. Firms are getting more internationalized and therefore can get more easily around national tax regimes in order to grow corporate profits, at the expense of society. Christensen and Murphy suggest that internationally operating firms need to adopt clear CSR standards, to give a clear oversight of whether a corporation is a “good corporate citizen” or not. The second theory used to predict the association between CSR and tax avoidance, is the
6
responsibility of a firm is to increase its profits. The shareholder theory argues that the firms’ only obligation is to fulfil the interests of the shareholders. Therefore, the shareholder theory suggests that a firm only should engage in CSR if it enhances the shareholder’ value. Moreover, this possibly means that the shareholder value could be increased if a company compensate its irresponsible taxing strategies by engaging in CSR. This would mean that there is a positive relationship between CSR and tax avoidance. The engagement in CSR and tax avoidance solely depends on whether it enhances shareholder value. This could possibly mean that a firm sees tax payments as a component of CSR and therefore there can be a negative relation between CSR and tax avoidance as well. As long as it enhances the value of the shareholders. For this reason, we can conclude that the shareholder theory can reason both in favour and against a negative relation between CSR and tax avoidance.
7
concluding that social services, satisfaction of employees and contributions to economic development were highly negatively related to tax avoidance.
Nonetheless, there is also evidence in favour of a positive relation between CSR and tax avoidance (Davis et al., 2015; Lanis and Richardson, 2012). Davis et al. (2015) finds that companies that engage more in CSR activities have lower five-year cash CETRs. Davis et al. suggest that there is a positive relationship since managers or other influential shareholders do not see the payment of taxes as being a component of CSR. In addition, Davis et al. tested the relationship between CSR and lobbying and identified a significant positive relation, which could be another explanation for the negative relationship between CSR and tax avoidacnce. To elaborate on, a firm could try to strengthen their image concerning socially responsible activities by lobbying. Davis et al. reasons that this is considered by firms as more efficient than paying their fair share of taxes, concluding that CSR and tax strategies act as substitutes, rather than as complements. Another research by Lanis and Richardson (2012), provided another explanation for the positive association between CSR and tax avoidance. They found that Australian publicly listed companies that were accused of tax aggressive activities increased their CSR activities. Lanis and Richardson suggested that firms increased their CSR activities to compensate for their irresponsible taxing policies. For this reason, Lanis and Richardson suggest that firms accused or suspected of tax avoidance will increase their CSR efforts to compensate.
8
this positive relation are the role of CSR as a substitute or/and the opinion on CSR as inefficient. To summarize, the following hypothesis is developed accordingly:
Hypothesis 1:
- Ceteris paribus, corporate social responsibility has a significance influence on a firm’s taxing strategies.
B. Financial constraints and Tax Avoidance
Financial condition or financial constraints is an important factor in decision-making regarding a firm’s investment. Firms can be identified as financial constrained, when it has difficulties to finance its project. A firm that is having trouble with obtaining external financing, must find alternative funds of finance for investments. Often those firms aim for the reduction of costs to generate higher cash flows. There are many cost-cutting techniques focussing on R&D expenses, marketing expenses, staffing etc. However, these techniques often have negative influences on the firm's’ operations. Therefore, firms tend to shift their focus to other techniques concerning tax expenses, as these are less vulnerable for the adversely effects related to the firm’s operations. This would imply that when a firm is financially constrained it would engage in more aggressive tax planning. Kim et al. (1998) argued that cash holdings are valuable when external financing is not sufficient in providing the necessary funds. By using tax aggressive policies, it can save more cash internally to use for their investments.
9
decreased availability of external funds. Their findings showed how firms that are financially constrained have a higher likelihood of tax aggressive strategies. Other researchers found that firms that were financially constrained were more likely to be involved in tax evasion activities (Alm et al., 2019). Alm et al. argued that this is caused by the fact that tax evasion helps firms to deal with financial issues that are created by the financial constraints. Additionally, they argued that this association is stronger when it comes to younger, smaller or private ownership-structured firms.
Following the existing literature this paper expects that financially constrained firms engage more in tax aggressive strategies to generate higher internal cash revenues. Moreover, expected is that financial constraints will weaken the association between CSR and tax avoidance. This would imply that firms that appears to be socially responsible and have sufficient funds of financing have higher CETRs, than firms that are financially constrained. This leads to the following hypothesis:
Hypothesis 2:
- Ceteris paribus, financial constraints moderates the relation between corporate social responsibility and tax avoidance.
C. National culture and Tax Avoidance
long-10
term orientation and indulgence were added to Hofstede’s cultural dimensions the years after. The initial dimensions are often acknowledged as the most valuable variables concerning national culture for empirical research. Because of this acknowledgement and inconclusive data on the additional dimensions this paper focuses on the initial dimensions. These dimensions can be shortly summarized as follows (Hofstede, 1980):
- Power distance deals with the hierarchical (in)equality within a country. It describes whether there are large inequalities in society or not.
- Individualism-collectivism can be described as the extent to which a culture focuses on the needs and desires of an individual (individualism) or group (collectivism).
- Masculinity typically can be subscribed as a society where heroism, individual achievement and material rewards are more valuable, while feminist societies enhances quality of life, cooperation and caring for the weak.
- Uncertainty avoidance implies the way that a society tolerates ambiguity and uncertainty. Some societies are less reluctant toward risk than other societies.
11
argumentation this paper expects that a country with a higher degree of power distance has more tax aggressive strategies. This would mean that if there is a negative relation between CSR and tax avoidance, power distance would weaken this relation.
The second cultural dimension addressed is individualism. Individualism concerns the self-concept whether a person in a specific country focuses on its personal goals or the goals of their group. Husted (1999) argues that a more collectivistic country would be more likely to override rules if it will affect the group positively. Collectivistic countries will focus less on ethic matters when there is a stronger group code. In this case collectivistic countries would engage more in tax avoidance. Furthermore, countries with a more individualistic character tend to have stronger economies with better regulations (Hofstede, 2001), and therefore there would be less opportunity to engage in tax aggressive strategies. Moreover, Tsakumis et al. (2007) empirically tested this theoretical reasoning and found evidence that firms in individualistic countries had lower tax avoidance. Richardson (2008) replicated this research using another sample and came to the same conclusion. However, Bame-Aldred et al. (2013) argued that Tsakumis et al. and Richardson lacked the required theoretical justification. Moreover, Bame-Aldred et al. argued the opposite, saying that individualism is positively related to tax avoidance. To elaborate on, Bame-Aldred et al. appealed that collectivism arises from a society where man thinks that members of that society are interdependent (Singelis et al., 1995). Therefore, an entity in a collectivistic country should engage less in tax avoidance since it will affect the society where an entity lives or operates in. Following Bame-Aldred et al., this paper expects individualism to be a weakening moderator, if there is a negative relation between CSR and tax avoidance.
12
material successes in an unjust society. Feminist countries rather put a high value on quality of life and environment. This paper assumes that the payment of a firm’s fair share of taxes, enhances the quality of life and environment. Previous literature corroborated this theoretical argumentation and found evidence that less masculine countries had a lower tendency towards tax avoidance (Richardson, 2008). However, literature is not entirely consistent on the relationship between masculinity and tax avoidance. Tsakumis et al. (2007) found countries with higher masculinity scores to have lower tax evasion. They argued that countries with higher scores on masculinity are “bragging” more about their CSR activities. Following Tsakumis et al. a firm in a masculine society who “brags”, is more conscious of its tax compliance responsibilities, because if it does not comply with it, it will increase the likelihood of scrutiny. Another proposed reason is that masculinity and the national Permissive Index are negatively related (Hofstede, 2001). This would mean that a more masculine society is less permissive when it comes to lawbreakers (e.g. firms that engage in tax evasion). Ensuing previous literature, we see that there is mixed evidence on the relationship between masculinity and tax avoidance. Therefore, this paper expects that masculinity will moderate the relation between CSR and tax avoidance.
13
reduce tax payments (Tsakumis et al., 2007; Richardson, 2008). These members see tax avoidance as a tool to reduce risk since the government has less revenue to “misuse”. Following this reasoning this paper expects that higher uncertainty avoidance leads to more tax avoiding strategies and therefore a weakening moderating effect if the relation between CSR and tax avoidance is negative.
Using the literature, this paper expects each dimension of national culture to affect the relationship between CSR and tax avoidance differently. Therefore, the third hypothesis addressed in this paper will be:
Hypothesis 3
- Ceteris paribus, national culture moderates the relation between corporate social responsibility and tax avoidance.
III.
Methodology
A. Sample selection and data sources
14
The database starts from 2009 as from that moment on the size increased substantially. It ends in 2017 as it is the last year that the data is available, currently. Moreover, following previous literature (Davis et al., 2015; Watson, 2015; Hoi et al., 2013), financial industries (SIC 6000-6999) and utility industries (SIC 4900-4999) are excluded from the sample. Both industries have different regulatory environments, which possibly causes disruptions in the relationship between CSR and tax avoidance. Hereafter, the data from Data-stream is merged with the data from Hofstede’s Dimension Data Matrix.
B. Model development
This paper examines three regression models. Firstly, the estimation models will be given. Thereafter, the description of the dependent, independent and control variables will be provided.
Model concerning the relationship between CSR and tax avoidance
The first model estimates the association between CSR and tax avoidance. Following hypothesis 1, the formula used for this regression is:
𝐶𝐸𝑇𝑅𝑖,𝑡 = 𝛽0+ 𝛽1𝐶𝑆𝑅𝑖,𝑡+ 𝛴𝛽𝑘𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑠𝑘+ 𝜀𝑖,𝑡
15
Model concerning the moderating effect of financial constraints
The second estimation model focuses on the interaction term between CSR and financial constraints and is formulated as follows:
𝐶𝐸𝑇𝑅𝑖,𝑡 = 𝛽0+ 𝛽1𝐶𝑆𝑅𝑖,𝑡+ 𝛽2𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙𝐶𝑜𝑛𝑠𝑡𝑟𝑎𝑖𝑛𝑡𝑠𝑖,𝑡+ 𝛽3𝐶𝑆𝑅𝑖,𝑡 ∗ 𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙𝐶𝑜𝑛𝑠𝑡𝑟𝑎𝑖𝑛𝑡𝑠𝑖,𝑡 + 𝛴𝛽𝑘𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑠𝑘+ 𝜀𝑖,𝑡
where financial constraint is a dummy variable that indicates whether a firm in a specific year was identified as financially constrained. The value will be 1 if a firm in a specific year belonged to the 50 percent highest scores concerning the Kaplan-Zingales index. A negative interaction term in combination with a positive CSR coefficient would imply that the relationship between CSR and financial constraints is weakened.
Model concerning the moderating effect of national culture
The last model estimates the moderating relationship of national culture on the association between CSR and tax avoidance, by using the following formula:
𝐶𝐸𝑇𝑅𝑖,𝑡 = 𝛽0+ 𝛽1𝐶𝑆𝑅𝑖,𝑡+ 𝛽2𝑁𝑎𝑡𝑖𝑜𝑛𝑎𝑙𝐶𝑢𝑙𝑡𝑢𝑟𝑒𝑖,𝑡+ 𝛽3𝐶𝑆𝑅𝑖,𝑡∗ 𝑁𝑎𝑡𝑖𝑜𝑛𝑎𝑙𝐶𝑢𝑙𝑡𝑢𝑟𝑒𝑖,𝑡 + 𝛴𝛽𝑘𝐶𝑜𝑛𝑡𝑟𝑜𝑙𝑠𝑘+ 𝜀𝑖,𝑡
16 Operationalization of the dependent variable
The dependent variable in this paper is the firm’s tax avoidance. Following previous
literature, I find CETR to be the most useful proxy for tax avoidance. Tax rates are considered as useful since tax-motivated transactions typically reduces CETRs (Rego, 2003). International firms use their foreign operations to shift taxes to jurisdictions where the tax rate is the lowest. This will reduce the overall CETR of a firm. Therefore, the firm’s tax avoidance in this paper is estimated by the CETR in a specified year. When the CETR is rather low for a corporation there is a higher chance of tax aggressive policies. On the other hand, when a CETR is rather high it may be more likely that it is paying its fair share of taxes. In this paper the corporate effective tax is calculated by dividing the tax expense by the taxable income (Rego, 2003). Some firms may have tax refunds (negative CETR) and will be set zero following previous literature (Gupta and Newberry, 1997). Moreover, in line with Gupta and Newberry, all observations with a value higher than 100 percent will be excluded since they are often the result of some exceptional conditions. This can cause model estimation problems (Stickney and McGee, 1982). Therefore, CETR values range between 0 and 1.
Operationalization of the Independent variable
17 Operationalization of the moderators
The firm-level moderator, financial constraint, will be tested using the second model. The degree of financial constraints will be measured by using the Kaplan-Zingales index (1997). The Kaplan-Zingales index can be calculated by the following formula:
Kaplan-Zingales Index = -1.1001909 * Cash Flows from operations / K + 0.2826389 * Q + 3.139193 * Debt / Total Capital + ‘-39.3678 * Dividends / K + ‘-1.314759 * Cash / K
where K is the amount of Property, Plant and Equipment and Q is the Tobin’s Q
calculated by adding equity market value and liability’s market value and dividing this by the amount of equity book value and liability’s book value. A high Kaplan-Zingales index would imply that a firm has higher debts, lower cash and lower dividends. After calculating the Kaplan-Zingales index for each firm each year, the firms are qualified as either financially constrained or not, following Lamont et al. (2001). This paper does so, by creating a dummy variable per year per industry, for each firm that belongs to the top 50 percent of financially constrained firms. By doing so, it does not imply that the top 50 percent are all financially constrained firms and the bottom 50 percent are not. Yet, by following Lamont et al., the claim is made that the top percent as a group is overall more constrained than the bottom 50 percent.
18 Control variables
To enhance the internal validity of this paper, several control variables are included. In doing so, this paper follows the existing literature and use it as a guideline for the methodology. This paper controls for size (SIZE), return on assets (ROA), market-to-book ratio (MKTBK), leverage (LEV), capital intensity (CAPINT), year (YEAR), country (COUNTRY) and industry (INDUSTRY).
19
industries may have a more tolerant approach towards tax avoidance than others. This paper includes YEAR fixed effects since tax disputes may differ over the period 2009-2017 (Richardson, 2015). Lastly, this paper controls for headquarter-country fixed effects as COUNTRY fixed effects can influence the CETR. The country dummy will be excluded in the last regression where the moderating effect of culture is measured. This paper does so, since culture does not vary over time. Moreover, countries with similar cultural characteristics will have comparable estimates on their dummies.
IV.
Empirical findings
A. Descriptive statistics
20
Table Ⅱ provides an overview of the composition of the sample, concerning the dependent and independent variables categorized per industry. Panel A describes the composition of the sample per industry. As depicted we can see that manufacturing industry is most present in this study with 9,076 firm-year observations. As mentioned before, this paper excludes both financial (SIC 6000-6999) and utility (SIC 4900-4999) industries for respective reasons. This can be seen in panel A, where no financial firm-year observations are included.
Table Ⅰ Descriptive statistics Panel A: Regression variables
Full Sample
Variable Mean SD Q1 Median Q3
CETR 0.28 0.13 0.20 0.27 0.34 CSR 58.57 30.00 30.80 66.55 86.10 FINCON 0.50 0.50 0 0 1 PDI 49.75 16.38 39.00 40.00 60.00 IDV 63.93 27.02 46.00 71.00 91.00 MAS 60.69 18.40 52.00 62.00 66.00 UAI 56.55 22.15 46.00 48.00 76.00 SIZE 16.90 2.69 14.89 16.40 18.68 ROA 8.10 6.04 4.01 6.61 10.50 MTBV 3.16 4.02 1.27 2.17 3.73 LEV 0.24 0.17 0.11 0.23 0.34 CAPINT 0.32 0.24 0.12 0.26 0.45 CSR*FINCON 28.33 35.25 0.00 1.35 63.42 CSR*PDI 2819.26 1702.15 1366.80 2870.88 3762.28 CSR*IDV 3939.69 2725.60 1486.28 3523.55 6373.57 CSR*MAS 3485.20 2078.31 1583.48 3509.61 5297.90 CSR*UAI 3264.93 2147.07 1456.64 3054.80 4423.28
All continuous variables are winsorized at the 1st and 99th percentiles.
21
The least present industry in this sample is agriculture with only 97 firm-year observations. The difference in observations per industry causing an unbalanced panel can be explained by the restricted data availability.
From panel B we can obtain the average CSR ratings per industry. The average CSR rating is highest for firms in the Manufacturing industry with an average of 63.44, followed by Mining with an average of 58.61. It is interesting to see that the Mining industry scores second on CSR ratings. This can be explained by the fact that the Mining firm-year observations included in this sample are mostly from the last few years. In these years attention towards ESG policies was raised in the mining industry (Rutten, 2013). Moreover, this raised attention was due to high fines and costly cancelations of projects because of irresponsible ESG activities. The industry with the lowest score is the Services industry with an average score of 51.06.
22
Table Ⅱ Descriptive statistics Panel A: Sample per industry
Four-digit Full Sample
Group SIC code Industry N %
1 100-999 Agriculture 97 0.51 2 1000-1499 Mining 1,409 7.35 3 1500-1799 Construction 1,012 5.28 4 2000-3999 Manufacturing 9,076 47.35 5 4000-4999 Transportation 2,261 11.80 6 5000-5199 Wholesale trade 667 3.48 7 5200-5999 Retail trade 1,800 9.39 8 6000-6799 Finance 0 0 9 7000-8999 Services 2,846 15.27 10 9100-9729 Public administration 0 0 11 9900-9999 Others 0 0 Total 19,168 100
Panel B: CSR ratings per industry
Group Industry Mean SD Min Max
1 Agriculture 51.28 27.27 5.07 94.03 2 Mining 58.61 27.50 3.1 96.36 3 Construction 52.63 31.34 3.02 96.3 4 Manufacturing 63.44 28.98 2.95 97.1 5 Transportation 56.62 30.75 2.79 96.99 6 Wholesale trade 51.1 28.49 3 96.56 7 Retail trade 54.83 31.81 2.74 96.64 8 Finance - - - - 9 Services 51.06 29.85 2.7 97.08 10 Public administration - - - - 11 Others - - - - Average 58.57 30.02 2.7 97.1
Panel C: Effective tax rates per industry
Group Industry Mean SD Min Max
1 Agriculture .27 .1 0 .74 2 Mining .30 .16 0 .97 3 Construction .27 .15 0 .95 4 Manufacturing .27 .13 0 0.99 5 Transportation .27 .14 0 .99 6 Wholesale trade .29 .13 .01 1 7 Retail trade .31 .11 0 .98 8 Finance - - - - 9 Services .28 .12 0 1 10 Public administration - - - - 11 Others - - - - Average .28 .13 0 1 N= number of observations
%= Percentage of each industry in different sample set
23
B. Pearson correlation matrix
25
C. Regression results
This section will tabulate and discuss the corresponding regression results. Furthermore, all the tables show the regression coefficient and the corresponding standard error, which are clustered at the firm level, following Petersen (2009). Additionally, the significance levels are provided in the form of significance stars. Furthermore, this paper multiplies the dependent variable CETR by 100 for the purpose of clarity.
CSR and tax avoidance
Table Ⅳ depicts the results on the relationship between CSR and tax avoidance. In the first column all control variables are excluded to show the main association. Column 2, 3 and 4 describe thereafter the impact of the control variables on the relationship between CSR and tax avoidance. Finally, column 5 decomposes the CSR variable to see where the effect is coming from.
26
Table Ⅳ
Regression results regarding CSR and tax avoidance
CETR CETR CETR CETR CETR
VARIABLES (1) (2) (3) (4) (5) CSR -0.0240*** -0.0165*** -0.0153*** 0.0011 (0.0042) (0.0044) (0.0044) (0.0049) ESGEVN 0.0076 (0.0066) ESGSOC -0.0031 (0.0067) ESGG 0.0126** (0.0063) SIZE -0.0269 0.0909 -0.8805*** -0.9773*** (0.0630) (0.0640) (0.1231) (0.1234) ROA -0.6455*** -0.6469*** -0.6362*** -0.6347*** (0.0179) (0.0179) (0.0177) (0.0177) MTBV 0.0803*** 0.0803*** 0.0746*** 0.0745*** (0.0244) (0.0244) (0.0239) (0.0240) LEV 0.6850 1.0652 2.0495*** 2.0647*** (0.7513) (0.7566) (0.7428) (0.7418) CAPINT -0.0240*** -0.0165*** -0.0153*** 0.0011 -0.8669 (0.0042) (0.0044) (0.0044) (0.0049) (0.5860) Constant 29.0594*** 36.3452*** 34.7553*** 30.8855*** 32.0939*** (0.2825) (1.0988) (2.3789) (6.1912) (6.1851)
Year fixed effects No Yes Yes Yes Yes
Industry fixed effects No No Yes Yes Yes
Country fixed effects No No No Yes Yes
R-squared 0.0020 0.0661 0.0835 0.2133 0.2130
Observations 19,168 19,168 19,168 19,168 19,168
Number of firms 3,905 3,905 3,905 3,905 3,905
Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
27
In column 5, the ESG score (proxy for CSR) is decomposed, and we can find that especially the score regarding governance (ESGG) contributes to a higher rate of taxes (p<0.05). Firms that score one point higher on the governance dimensions pay on average 0.0126 percent more taxes. The governance score comprises of both corporate governance and corporate behaviour. Moreover, corporate behaviour touches upon tax transparency, and corruption and instability. For this reason, it makes sense that there is on this dimension a positive relationship between CSR and tax payments. However, we cannot make any statements regarding a firm’s score on the environmental (ESGEVN) and social (ESGSOC) dimensions. This suggests that firm’s scoring higher on these two dimensions do not necessarily see the payment of taxes as an important component of CSR. Neither see these firms CSR as a tool to compensate for their irresponsible taxing strategies.
CSR, tax avoidance and financial constraints
28
Regression results regarding the moderating relation of financial constraints and culture
CETR CETR VARIABLES (1) (2) CSR 0.0001 0.1192*** (0.0059) (0.0346) FINCON -0.1600 (0.4335) CSR*FINCON 0.0018 (0.0064) PDI -0.0240 (0.0257) IDV 0.1337*** (0.0170) MAS 0.1676*** (0.0180) UAI 0.1016*** (0.0139) CSR*PDI 0.0005 (0.0004) CSR*IDV -0.0007*** (0.0002) CSR*MAS -0.0016*** (0.0002) CSR*UAI -0.0005*** (0.0002) SIZE -0.8809*** 0.3990*** (0.1231) (0.0890) ROA -0.6369*** -0.6084*** (0.0179) (0.0177) MTBV 0.0746*** 0.0972*** (0.0240) (0.0241) LEV 2.0738*** 1.1755 (0.7465) (0.7455) CAPINT -0.8543 -1.4553** (0.6207) (0.5918) Constant 30.9690*** 7.6499** (6.1952) (3.5039)
Year fixed effects Yes Yes
Industry fixed effects Yes Yes
Country fixed effects Yes No
R-squared 0.2133 0.1640
Observations 19,168 19,168
Number of firms 3,905 3,905
Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
Table Ⅴ shows the regression regarding the interaction effect of the moderators. The dependent variable is corporate effective tax rate (CETR). The dependent variable in both columns. CSR ratings based on the ESG database. The moderating variable financial constraints (FINCON) in column 1 is a dummy variable based on the Kaplan-Zingales index. Firms that are identified as
financially constrained have a value of 1. In column 2 the moderating variables are indexes on power distance (PDI),
29 CSR, tax avoidance and culture
Table Ⅴ, column 2 reports the regression coefficients regarding the interaction effect of the country-level variable national culture. National culture is measured by four dimensions, namely PDI, IDV, MAS and UAI. As stated in hypothesis 3 the relationship between CSR and tax avoidance is moderated and therefore different across culture. Moreover, column 2 provides evidence that the association between CSR and tax avoidance differs across measures of individualism, masculinity and uncertainty avoidance.
First, we can obtain that the CSR practices of firms negatively affects tax avoidance. If a firm scores one point higher on their CSR practices it pays on average 0.1192 percent more taxes (p < 0.01). The negative association is in line with previous literature (e.g. Lanis and Richardson, 2015; Park, 2017; Hoi et al., 2013). Therefore, it provides evidence that firms see tax strategies as a component of their CSR policy. Besides, it enhances the argumentation that firms that engage more in CSR activities, are expected to value stakeholders higher than firms that engage less in CSR.
30
since it will affect the society where a firm is operative. For this reason, the significant negative coefficient of CSR*IDV (p < 0.01) is line with what is hypothesized. It implies that if a firm is headquartered in a country with one point higher on IDV, it will have a weakening effect on the relation between CSR and tax payments of 0.0007 percent.
Secondly, this paper finds the interaction term regarding masculinity (CSR*MAS) to be significantly negative (p < 0.01). For firms that are headquartered in a country with one point higher on MAS, the relationship between CSR and tax avoidance is weakened by 0.0016 percent. This is in line with Richardson (2008). He argued that the payment of a firm’s fair share of taxes enhances the quality of life and environment. A culture that focuses on the quality of life and environment can be characterized as less masculine or more feministic. Masculine culture focuses more on gaining respect through gathering wealth and materialistic successes.
Thirdly, the interaction term in column 2 regarding uncertainty avoidance (CSR*UAI) shows a significant negative coefficient (p < 0.01). Moreover, a firm that is headquartered in a society where uncertainty is less tolerated the positive relationship between CSR and tax avoidance is weakened: a firm that is headquartered in a country with one point higher on UAI the relationship between CSR and tax payments is weakened by 0.0005 percent.
31
V.
Conclusions
This paper focusses on the relation between CSR and tax avoidance. In addition, the moderating roles of financial constraints and a nation’s culture are examined. Using a new proxy for CSR performance and using two new, moderating variables this paper expands the research on the relation between CSR and tax avoidance. Moreover, this paper draws its conclusions from a global sample, which is different than the country studies in the past literature (Davis et al., 2015; Richardson and Lanis, 2012). Taking these facts into consideration, this paper can be regarded as an important contribution to the existing literature.
32
Nonetheless, this paper finds culture to be a significant moderator in the relation between CSR and tax avoidance. Individualism, masculinity and uncertainty avoidance can be all considered as moderators according to this paper’s findings. The relation between CSR and tax avoidance is weakened in cultures where higher score of individualism, masculinity and uncertainty avoidance can be found. Therefore, this paper concludes that the relationship between CSR and tax avoidance differs across cultures.
Some limitations of this paper should also be addressed. First, this paper does not take into consideration the multicausality problem that arises within the examination of CSR and tax avoidance. This paper acknowledges that this may lead to a decrease of internal validity. Secondly, this paper uses the CETR as a proxy for tax avoidance. Some previous literature questioned the accuracy of financial-statement-based-tax aggressiveness measures (e.g. Plesko, 2003). Therefore, the interpretation of the results and conclusions need a careful approach. Thirdly, this paper only addresses publicly listed firms due to the data availability. These firms are often larger in size and therefore may have different approaches towards CSR and tax avoidance. Lastly, it is important to note that Hofstede’s cultural dimensions were established over two decades ago. Things have changed, and therefore the scores on cultural dimensions may be outdated.
33
VI.
References
Adhikari, A., Derashid, C., & Zhang, H. (2006). Public policy, political connections, and effective tax rates: Longitudinal evidence from Malaysia. Journal of Accounting and
Public policy, 25(5), 574-595.
Alm, J., Liu, Y., & Zhang, K. (2019). Financial constraints and firm tax evasion. International
Tax and Public Finance, 26(1), 71-102.
Avi-Yonah, R. S. (2008). Corporate social responsibility and strategic tax behavior. In Tax and
Corporate Governance, 183-198.
Bame-Aldred, C. W., Cullen, J. B., Martin, K. D., & Parboteeah, K. P. (2013). National culture and firm-level tax evasion. Journal of Business Research, 66(3), 390-396.
Christensen, J., & Murphy, R. (2004). The social irresponsibility of corporate tax avoidance: Taking CSR to the bottom line. Development, 47(3), 37-44.
Davis, A. K., Guenther, D. A., Krull, L. K., & Williams, B. M. (2015). Do socially responsible firms pay more taxes? The Accounting Review, 91(1), 47-68.
Dyreng, S. D., Hanlon, M., & Maydew, E. L. (2010). The effects of executives on corporate tax avoidance. The Accounting Review, 85(4), 1163-1189.
Edwards, A., Schwab, C., & Shevlin, T. (2015). Financial constraints and cash tax savings. The
Accounting Review, 91(3), 859-881.
Freedman, J. (2003). Tax and corporate responsibility. Tax Journal, 695(2), 1-4.
Friedman, M. (1970). A theoretical framework for monetary analysis. Journal of Political
34
Friese, A., Link, S., & Mayer, S. (2008). Taxation and corporate governance—The state of the art. In Tax and Corporate Governance, 357-425.
Gupta, S., & Newberry, K. (1997). Determinants of the variability in corporate effective tax rates: Evidence from longitudinal data. Journal of Accounting and Public Policy, 16(1), 1-34.
Hite, P. A., & Roberts, M. L. (1992). An analysis of tax reform based on taxpayers’ perceptions of fairness and self-interest. Advances in Taxation, 4(1), 115-137.
Hofstede, G. (1980). Culture and organizations. International Studies of Management &
Organization, 10(4), 15-41.
Hofstede, G. (2001). Culture's consequences: Comparing values, behaviors, institutions and organizations across nations. Sage publications.
Hoi, C. K., Wu, Q., & Zhang, H. (2013). Is corporate social responsibility (CSR) associated with tax avoidance? Evidence from irresponsible CSR activities. The Accounting
Review, 88(6), 2025-2059.
Husted, B. W. (1999). Wealth, culture, and corruption. Journal of International Business
Studies, 30(2), 339-359.
Kim, C. S., Mauer, D. C., & Sherman, A. E. (1998). The determinants of corporate liquidity: Theory and evidence. Journal of Financial and Quantitative Analysis, 33(3), 335-359.
Kreps, D. M., Alt, J. E., & Shepsle, K. A. (1996). Corporate culture and economic theory.
Firms, Organizations and Contracts, Oxford University Press, 221-275.
Lamont, O., Polk, C., & Saaá-Requejo, J. (2001). Financial constraints and stock returns. The
35
Landolf, U. (2006). Tax and corporate responsibility. International Tax Review, 29(July), 6-9.
Lanis, R., & Richardson, G. (2012). Corporate social responsibility and tax aggressiveness: An empirical analysis. Journal of Accounting and Public Policy, 31(1), 86-108.
Lanis, R., & Richardson, G. (2015). Is corporate social responsibility performance associated with tax avoidance? Journal of Business Ethics, 127(2), 439-457.
Milliron, V., & Toy, D. (1988). Tax compliance: an investigation of key features. The Journal
of the American Taxation Association, 9(1), 84-104.
Park, S. (2017). Corporate Social Responsibility And Tax Avoidance: Evidence From Korean Firms. Journal of Applied Business Research, 33(6), 1059.
Plesko, G. A. (2003). An evaluation of alternative measures of corporate tax rates. Journal of
Accounting and Economics, 35(2), 201-226.
Rego, S. O. (2003). Tax‐avoidance activities of US multinational corporations. Contemporary
Accounting Research, 20(4), 805-833.
Richardson, G. (2008). The relationship between culture and tax evasion across countries: Additional evidence and extensions. Journal of International Accounting, Auditing and
Taxation, 17(2), 67-78.
Rutten, A (2013). Materiality of ESG risks in the mining sector. Retrieved from www.vigeoeiris.com/wpcontent/uploads/2013/11/Sustainable_Focus_Materiality_of_ ESG_risks_in_the_Mining_sector_November_2013.pdf?x62552.
36
Stata Corporation. (1997). Stata User's Guide: Release 5. Stata Press.
Stickney, C. P., & McGee, V. E. (1982). Effective corporate tax rates the effect of size, capital intensity, leverage, and other factors. Journal of Accounting and Public Policy, 1(2), 125-152.
Tsakumis, G. T., Curatola, A. P., & Porcano, T. M. (2007). The relation between national cultural dimensions and tax evasion. Journal of International Accounting, Auditing and
Taxation, 16(2), 131-147.
Watson, L. (2015). Corporate social responsibility, tax avoidance, and earnings performance.
The Journal of the American Taxation Association, 37(2), 1-21.
