INTEGRITY IN SUSTAINABILITY REPORTING:
T
HE CASE OF TAXATION
MASTER THESIS ACCOUNTANCY & CONTROLING WRITTEN BY: L.A. OLDENBANNING STUDENTNUMMER: 1909134 VEEMARKSTRAAT 89 9724 GB GRONINGEN L.A.OLDENBANNING@STUDENT.RUG.NL
JUNE 11, 2015
L.A.OLDENBANNING 1 PERSONALIA
Name Laura Alicia Oldenbanning
Date of Birth June 21, 1990
E-mail address l.a.oldenbanning@student.rug.nl
Alternative e-mail address oldenbanning@hotmail.com
Address Veemarktstraat 89, Groningen
Postal code 9724 GB
Phone number +31 (0)6 52 54 51 51
University Rijksuniversiteit Groningen
Faculty Economics & Business
Study program Accountancy & Controlling
Student number 1909134
Internship company PricewaterhouseCoopers
Thesis supervisor University Prof. dr. I.J.J. Burgers
L.A.OLDENBANNING 2 CONTENTS Personalia 1 Abstract 3 Introduction 4 Literature review 8
The role of sustainability reporting 8
Assurance on sustainability reporting 9
Taxation 11
Theoretical framework and hypotheses development 13
Shareholder theory 13 Agency theory 13 Stakeholder theory 14 Hypotheses 15 Research model 17 Data 17 Model 19 Results 24 Descriptive statistics 24
Regression analyses results 25
Results based on model using long-run cash ETR as proxy tax aggressiveness 25
Results based on model using cash ETR as proxy tax aggressiveness 26
Results based on model Transparency Benchmark score and tax aggressiveness 28
Conclusion 30
Further research 32
Appendix 33
1. Variables 33
2. descriptive statistics firm-year observation 34
3. Correlation Matrix 35
L.A.OLDENBANNING 3 ABSTRACT
Encouraged by consumers, governments, institutions and society as a whole, Corporate Social Responsibility (CSR) has become increasingly important to corporations. More often information is disclosed about CSR, and a swelling number of companies voluntarily issue a sustainability report. However, in the meantime a public debate about tax transparency is on the rise. Multinational companies (e.g. Apple, Amazon, Starbucks and Google) face public scrutiny caused by their tax policy. This research investigates the relation between CSR reporting and tax aggressiveness. The challenge in this study is merging several research fields (accounting, CSR, law, and ethics) and the different perspectives researchers have on these topics. This research reveals the relationships between CSR assurance, transparency benchmark score, firm size, leverage and return on assets and tax aggressiveness. Three proxies are used to measure tax aggressive behaviour, respectively: long-run cash effective tax rate, annual cash effective tax rate and book effective tax rate. A significant relation is found between the level of CSR assurance and tax aggressiveness. Implicating companies with assurance on CSR behave less aggressive in regards to income taxes. Findings on the Transparency Benchmark indicate that the TB score in association with tax aggressive behaviour. This paper contributes to merge different research topics in one model and expands research done on CSR reporting, CSR assurance and tax aggressiveness.
L.A.OLDENBANNING 4 INTRODUCTION
Encouraged by consumers, governments, institutions and society as a whole, Corporate Social Responsibility (CSR) has become increasingly important in corporate boardrooms and business activities (e.g., Kitzmueller and Shimshack, 2012; Sweeney and Coughlan, 2012; Veríssimo and Laserda, 2015). An increasing number of corporations go beyond the mandatory requirements of providing an annual report. More often information is disclosed about CSR, and a swelling number of companies voluntarily issue a sustainability report1 or even go
beyond and integrate CSR within their annual report (also known as integrated reporting). Even though measures of sustainability are not yet standardized and thus the audit of environmental and social information is complex (Reilly, 2009), companies nowadays even request an auditor to provide voluntary external verification and assurance of report contents on CSR alongside the assurance on their financial reports (Kolk, 2010).
However, in the meantime a public debate about tax transparency is on the rise. Multinational companies like Apple2, Google3, Amazon and Starbucks made headlines on different occasions about tax-avoidance practices.
As Nouwen (2013) states: “Multinationals are considered to be fiscally over privileged as compared to the man
on the street.” Not only the media, but also governments and institutions are increasingly aware of the powerful
position of multinationals. For example, in the U.K. a Committee of Public Accounts was ordered by the House of Commons to investigate Google and their tax avoidance practices4. A similar situation occurred in America where
Apple was invited for a Senate Hearing by a subcommittee for an explanation of Apple’s tax strategy5. Moreover,
Apple’s tax construction is under investigation by the European Commission to see if the ruling concerning transfer pricing arrangements struck by Ireland’s tax authorities and Apple was unlawful because of the state aid which Apple has received. State aid is defined by the European Union as: “an advantage in any form whatsoever
conferred on a selective basis to undertakings by national public authorities.”6. State aid can only by implemented
after approval by the European Commission and the Commission holds the power to recover incompatible state aid7. Consequently after the announcement of the investigation, Apple warned investors in their annual report
of 2014 of material financial damage from the possible state-aided tax advantages8. To illustrate the size of the
possible consequences the SEC rules an event material usually as 5% of the average of the past three years pre-tax earnings in Apple’s case this could exceed $ 2.5 billion9 and the 10-K form SEC filing by Apple in 2014 shows
a unrecognized deferred-tax liability of $23.3 billion. Meanwhile Amazon choose to let go of their recent tax strategy since May 1, 201510. Amazon channeled their revenue through a holding in Luxembourg to ascertain
low-tax payment11. Even though Amazon denied last year that the corporate structure was motivated by the
favorable tax climate12. The change in reporting and recognizing revenues by the implementation of a different
tax strategy by Amazon (effective since May 1, 2015) can result in a higher tax bill in Europe because revenue realized in for example U.K., Germany, Spain and Italy. The multidisciplinary nature of tax transparency can be
1 Sustainability Reporting is also known CSR Reporting, and Social Responsibility Reporting
2 http://www.forbes.com/sites/beltway/2013/05/21/the-real-story-about-apples-tax-avoidance-how-ordinary-it-is/ 3 http://uk.reuters.com/article/2013/05/22/uk-eu-tax-avoidance-factbox-idUKBRE94L0HE20130522 4 http://www.publications.parliament.uk/pa/cm201314/cmselect/cmpubacc/112/112.pdf 5 http://www.gpo.gov/fdsys/pkg/CHRG-113shrg81657/pdf/CHRG-113shrg81657.pdf 6 http://ec.europa.eu/competition/state_aid/overview/index_en.html 7 http://www.nortonrosefulbright.com/knowledge/publications/120029/tax-rulings-on-transfer-pricing-may-violate-eu-state-aid-rules 8 Apple 10-K annual report, 2014. Page 44, Legal and Other Contingencies
9 http://www.ft.com/cms/s/0/507f19e8-ee77-11e4-98f9-00144feab7de.html#axzz3ZM37Y1gB
http://www.telegraaf.nl/buitenland/24022769/___Boete_Apple_17_miljard_euro___.html
10 http://www.wsj.com/articles/amazon-changes-tax-practices-in-europe-amid-investigations-1432480170 11 http://fortune.com/2015/05/25/amazon-tax-eu-regulators/
L.A.OLDENBANNING 5 difficult because of the diverse disciplines (e.g. shareholders, managements, tax advisors, accountants, governments and society) which are involved and transparency is viewed from different perspectives. From a legal perspective: “Tax transparency demands the tax authority to provide clear tax rules.” (Page 430- Cordes, Ebel & Gravelle, 2005). From an accounting perspective on the other hand, tax research is viewed from an angle to how markets will perceive the tax information (non-)disclosure (Gentry, 2007)13.
With regard to Starbucks; Starbucks is well known and often praised for their CSR practices14. In 2001 Starbucks
published their first sustainability report and has published such a report on an annual basis ever since. However, in respect of taxes the exposure of Starbucks’ tax practices in the UK led to public scrutiny15. As The Economist16
points out, this led to serious concerns for Starbucks. Starbucks therefore relocated their headquarters from Amsterdam, The Netherlands to London, U.K. in order to restore reputation. Additionally, the relocation would administratively yet not substantively change tax payments17, because of the favorable financial jurisdiction in
London18. In the meantime Starbucks is also under investigation by the European Commission to assess if the tax
deal arranged with the Dutch tax authority was lawful. If not it will qualify as state-aid and financial consequences could be expected19. During all the commotion in the recent years Starbucks continued to publish CSR reports
however the topic of taxation, though heavily discussed in society, has not been touched in the corporate social responsibility report. I do not wish to scapegoat Starbucks for tax aggressive behavior drenched in business activities all over the world in this day and age. I applaud the company for being one of the leaders in corporate responsibility. However, a company with such a magnitude and power as Starbucks can lead by example. Initiatives are growing to contribute to the transparency of CSR practices along with initiatives which are pushing for systemic changes in tax behavior. The Organization for Economic Co-operation and Development (OECD) delivered an action plan – Base Erosion and Profit Shifting – in 2013 to the G20 ministers with 15 action plans clearly identified that can result in fundamental changes to international tax rules and end the artificial shifting of profits to jurisdictions to avoid tax payments20. In addition, since 2004, the Dutch Ministry of Economic Affairs
initiated the Transparency Benchmark (TB), to assess the extent to which organizations account their activities in regard to CSR in their annual reports. The Dutch government requires organizations to be transparent about their policies and activities21. The Transparency Benchmark provides transparency in CSR reporting. Furthermore,
the Dutch Association of Investors for Sustainable Development (VBDO) works to create a sustainable capital market, considering not solely financial criteria but also non-financial, social and environmental criteria to increase sustainability awareness among investors and corporations. The public outcry, after widespread media coverage on tax avoidance, did not go unnoticed by VBDO. In the general meeting of shareholders in 2013 the VBDO prioritized tax policy as an item on the agenda for 2013 to contribute to an sustainable tax policy. These
13 This article and research is conducted from an accounting perspective.
14 http://www.usatoday.com/story/money/business/2014/07/06/why-its-hard-to-hate-starbucks/12022699/ http://www.theguardian.com/sustainable-business/blog/best-practices-sustainability-us-corporations-ceres http://fortune.com/2014/10/30/starbucks-innovation-cafe-to-classroom/ 15 http://www.internationaltaxreview.com/Article/3109681/Starbucks-scandal-could-force-tax-onto-CSR-agendas.html, 16 http://www.economist.com/node/21643271/print 17 http://www.theguardian.com/business/2014/apr/16/starbucks-hq-relocation-uk-generate-negligible-tax-revenue 18 http://wolfstreet.com/2014/11/08/as-luxembourg-takes-the-heat-the-worlds-worst-tax-haven-the-city-of-london-remains-hidden-in-plain-sight/
19 Financieel Dagblad, 6-05-2015, pagina 7, Onderzoek naar deal tussen Starbucks en Nederland vertraagd, Rik Winkel 20 http://www.oecd.org/g20/topics/taxation/
L.A.OLDENBANNING 6 initiatives, if proven to be influential, could be a great contribution to perform pressure and enable changes in organizations behavior.
Hoi, Wu & Zhang (2013) showed that companies involved in excessive irresponsible CSR activities are more aggressively engaged in tax avoidance behavior. Further, Lanis and Richardson’ results (2012) indicated that a higher level of CSR disclosure of a corporation, lowered the level of tax aggressiveness. Overall, transparency is essential in corporate sustainability reporting, and findings on CSR and tax transparency show the potential benefits for corporations (Boylan, 2013; Kitzmueller and Shimshack, 2012; Sweeney and Coughlan, 2012). However, corporate social responsibility statements published by multinationals often did not regard tax payment as a part of their CSR agenda (Christensen and Murphy, 2004). I follow the statement made by Christensen and Murphy (2004):
“It is not possible to be ethical in one area of business conduct and to act otherwise in another area, and companies that function in this way reveal a major disconnect in their core organizational values.”
It is interesting that environmentally harmful activities are condemned by public and law (e.g. BP’s Deepwater Horizon oil spill22), abusing human rights leads to boycotts and other forms of irresponsible corporate behavior
seems to be embedded as non-ethical in society. However, the large scale tax avoidance done by corporations (multinationals as well as small and medium enterprises) is still deplorably little condemned by the public as unethical behavior. The argument that tax avoidance is legal and fits within the boundaries of the law raises questions since tax aggressive behavior is against the spirit of the law. When did we lose the spirit of the law as objective out of sight and use the law merely for own personal benefit?
My research will focus on the following key matters:
1. Companies that issue or integrate a sustainability report show lower tax aggressiveness compared to
companies that do not issue such a report.
2. Companies with assurance on CSR contents differ in tax aggressiveness compared to companies without
assurance on CSR contents.
3. The Transparency Benchmark scores ascribed to companies can be used as an indicator for their degree
of tax aggressiveness.
This study contributes to decipher the integrity of a corporation involved in sustainability reporting. More specifically, I will focus on the transparency in taxation based on a corporation’s tax aggressiveness23. Of
particular interest is the alignment of a corporation’s words and actions in the case of integrity in CSR and tax behavior. Prior research focused either on the linkage between tax behavior and ethics or on CSR and ethics. However, little cohesion exists regarding integrity, CSR and taxation. Therefore, this research will fill the gap in current literature on the integrity of an organization in their words about CSR issued in a sustainability report and their action. Besides, this research tries to show the possible influence and relevancy of initiatives to contribute to the transparency of corporate behavior in regard to CSR and taxation.
22 For more information: http://en.wikipedia.org/wiki/Deepwater_Horizon_oil_spill
L.A.OLDENBANNING 7 The paper proceeds as follows. In the next section I briefly review current literature on the topics of sustainability reporting, corporate taxation & tax aggressiveness, and integrity & business ethics. The following section outlines the theoretical framework and hypotheses development. Thereafter, the research method, data and sample are outlined and explained. This section is followed by the findings of the analysis adopted. Finally, I evaluate the findings, consequently resulting in the conclusion of these findings. I conclude this paper with a brief review on the limitations of this research and recommendations for future research.
L.A.OLDENBANNING 8 LITERATURE REVIEW
THE ROLE OF SUSTAINA BILITY REPORTING
Following Kitzmueller and Shimshack (2012) CSR can be seen as “the commitment of businesses to behave ethically, [consider culture, history and prevailing institutions], and to contribute to sustainable economic development by working with all relevant stakeholders to improve their lives in ways that are good for business, the sustainable development agenda, and society at large”. Corporate activities which address ethical and social values that go beyond legal requirements succeed as corporate social responsible (Van Aaken, Splitter & Seidl, 2013). As Van Marrewijk (2003) described, reporting about corporate sustainability and CSR activities demonstrates the corporation’s inclusion of social and environmental concerns related to their business operations through interaction with stakeholders. A sustainability report has become a more significant communication source. Initially CSR disclosures took place separately within annual (financial) reports. However, over time an increase arose in disclosing CSR in stand-alone reports; also known as a sustainability report. A sustainability report refers to a stand-alone report where information is disclosed in regard to a company’s environmental, economic, and social activities and impacts (O’Dwyer, 2011). Where annual reports have strict regulation and frameworks, sustainability reporting lacked guidelines. In the course of time several guidelines, frameworks and benchmarks were developed for CSR reporting. For example: Carbon Disclosure Project (CDP) focuses primarily on the environmental impact of an organization; The UN Global Compact is a guideline for principles contemplating human rights, labor, anti-corruption and the environment; Dow Jones Sustainability
Index (DSJI) is a global sustainability benchmark who analyses a company’s economic, environmental and social
performance annually; The Transparency Benchmark, initiated by the Dutch Ministry of Economic Affairs, assesses Dutch organizations on CSR information disclosed in annual reports; AccountAbility AA1000 Principles
standard is a principle-based standard helping organizations to become more accountable, responsible and
sustainable; Global Reporting Initiative (GRI) develops globally applicable Sustainability Reporting Guidelines for organizations. Where GRI is becoming the standard for sustainability reporting (Hess, 2008) a recent shift is being observed to integrate CSR information in the annual report. This practice has become known as integrated reporting. With the International Integrated Reporting Council (IIRC) leading with the standard setting framework. Although the IIRC and GRI share some characteristics there are also essential differences. The GRI focuses on a broader perspective which includes a wider range of stakeholders’ interests, whereas the IIRC increases the focus on the interests and related information needs of financial capital providers (De Villiers & Unerman, 2014). The meaning of sustainability using the IIRC’s framework is linked to efficient use of financial and non-financial resources, aimed at, and contributory to, the longevity of economic value creation (De Villiers & Unerman, 2014). A feature of integrated reporting is interweaving the two types of reporting, financial and CSR, and also combine the short- and long-term thinking, governance and strategy (Van Bommel, 2014). Ultimately, the actual goal of sustainability reporting is to truly change corporate behavior in continuance to aim for a sustainable development and be transparent about their performance – regardless whether the performance is positive or negative. This change can occur through a company’s own self-critical reflection (known as moral development) but also through external pressure forcing changes (Hess, 2008).
L.A.OLDENBANNING 9 ASSURANCE ON SUSTAIN ABILITY REPORTING
Assurance on sustainability reporting promises to provide assurance regarding the completeness and reliability of such a report (O’Dwyer, 2011). More often companies request for external assurance on sustainability reporting or components of CSR information. To some extent due to the increased complexity of sustainability reporting, with much deeper and broader disclosures, assurance on CSR could enhance the credibility of the report (O’Dwyer, 2011). Partially for the reason that since the reporting on CSR skepticism grew on the value of the disclosed CSR information. Some users viewed sustainability reporting as a way of greenwashing and therefore (some) disclosed CSR information is disregarded (Lyon & Maxwell, 2011). Lyon and Maxwell (2011) define greenwashing as: “selective disclosure of positive information about a company’s environmental or social
performance, without full disclosure of negative information on these dimensions, so as to create an overly positive corporate image”. Negative CSR information, however, is viewed as credible without any assurance.
Thus, assurance on sustainability reporting can avoid the risk for companies issuing sustainability reports being perceived as greenwashing (Wong & Millington, 2014). Together with the findings by Casey and Grenier (2015) that assurance on sustainability reporting results in lower analyst forecast errors and dispersion, and lower cost of capital, CSR assurance could lead to benefits for an organization. Assurance on sustainability reporting and components of CSR are provided by professional accounting firms as well as non-accounting firms (O’Dwyer, 2011; Casey & Grenier, 2015). Professional accounting firms concentrate mainly on the accuracy of the data transferred by management into a sustainability report with the providence of assurance statements (O’Dwyer, 2011). In any type of assurance engagement the five elements2425 (see figure 1 on the following page for an
explanation of these elements) need to be present in order for an assurance firm or professional to be able to provide assurance.
The assurance on financial reports is, in most well-developed countries, mandated by law and assurance standards are set. However, assurance on sustainability reporting is voluntary (as is sustainability reporting in most territories). In order for an (accounting-) auditor to give assurance applicable to CSR, the five elements thus need to be present. Therefore CSR assurance standards were developed in order to fulfill the demand for voluntary assurance on CSR. Some examples of frequently used accounting standards for CSR information are:
International Standard on Assurance Engagements (ISAE) 3000 which can only be issued by professional
accountants; the AccountAbility AA1000 Assurance Standard which can be used by different types of assurance providers; the International Federation of Accountants (IFAC) issued the International Standard on Assurance
Engagement (ISAE) 3410 assurance on Greenhouse Gas Statements Standard, and the Netherlands Institute for
Chartered Accountants (NBA) issued specific accounting standard concerning sustainability reports26. Globally,
research has shown that the demand for assurance on CSR reporting differs per country and occurs primarily in the U.K., Australia, Japan and Europe in comparison to North-America (Simnett, Vanstraelen & Chua, 2009; Kolk, 2010). 24 https://www.nba.nl/HRAweb/Richtlijnen/2005/Bundel1%5CEng%5CFramework%20assurance%20engagements%2020-60.htm 25http://www.google.nl/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&cad=rja&uact=8&ved=0CDsQFjAE&url=http%3A%2F%2Fwww.icaew.c om%2F~%2Fmedia%2Fcorporate%2Ffiles%2Ftechnical%2Faudit%2520and%2520assurance%2Fassurance%2Fsourcebook%2520chapter3.ashx &ei=7VVqVdqCFOKE7gaj8IGQBQ&usg=AFQjCNHWphaaOx9n3YXHHJTLiP_o1oYQeg
L.A.OLDENBANNING 10 Figure 1. Explanation the five elements of assurance engagement
L.A.OLDENBANNING 11 TAXATION
The ability to raise sufficient tax revenue is a key element of a government in order to finance education, infrastructure and health and create a sustainable society (Weyzig & Van Dijk, 2009). Organizations contribute indirectly to the financing of public goods and services by a number of tax payments (e.g. value added tax, social security, income tax on employees’ salary and corporate tax based on profit). For tax purposes a country does not look at a multinational as a whole (unitary approach), but it looks only to the parts operating in their jurisdiction. This is known as the separate entity approach (Needham, 2013). Globalization led the road for companies to go abroad and expand their markets and purchase resources from other parts of the world. However, local regulation of companies became more difficult because of the increasing globalization (Barnea, Heinkel & Kraus, 2013). Since multinationals can use their international operations to facilitate the strategic exploitations of the differences in international tax systems to lower overall corporate profit and reducing tax liabilities (Beuselinck, Deloof, & Vanstraelen, 2015).
Tax aggressive behavior is: “Downward manipulation of taxable income through tax planning that may or may not be considered fraudulent tax evasion” (Frank, Lynch & Rego, 2009). Although some literature may use tax evasion and avoidance interchangeably, the difference lies in the legal consequences. Tax avoidance can be defined as: “the arrangement of a taxpayer's affairs that is intended to reduce his tax liability and that although the arrangement could be strictly legal it is usually in contradiction with the intent of the law it purports to follow.”27 Aggressive tax strategies are often used to minimize corporate tax payments by all legal means
possible (Hardeck & Hertl, 2014) and can lead to tax minimization on the long-run (Dyreng, Hanlon & Maydew, 2008) even if the tax practices are in conflict with the spirit of the law. For example, transfer pricing is one of the methods which can be used to minimize taxes. Transfer pricing is a legitimate way for corporations to calculate profits made based on internal sales as long as they abide by the arm’s length principle28. However, the arm’s
length principle is can be easily abused when assigning value to assets which have few comparatives available (like intellectual property). The Organization for Economic Co-Operation and Development (OECD) was addressed by the G20 finance ministers to develop an action plan focusing on addressing Base Erosion and Profit
Shifting and providing countries with domestic and international instruments to align rights to tax better with
the activity (OECD, 2013). In 2013 the OECD published the report: “Addressing Base Erosion and Profit Shifting”29.
This report presented the an overview of current research and data available in regard to Base Erosion and Profit Shifting (BEPS) and addresses perceived flaws in the international tax rules. The OECD included 15 separate actions points as possible solutions to reduce flaws within international tax rules. As mentioned in the introduction, multinationals (such as Apple, Amazon, Starbucks) often shift profits through different countries to lower the tax burden with the (ab)use of transfer pricing corporations are able to secure minimizing of taxes. Four action plans (Respectively; action 7: prevent the artificial avoidance of Permanent Establishment (PE) status, action 8, 9, and 10 are in regard to assure transfer pricing outcomes are in line with value creation) provided in the BEPS report give a clear overview, but more importantly, also a plan to conquer the abuse of transfer pricing by multinationals.
27 http://www.oecd.org/ctp/glossaryoftaxterms.htm#T
28 The OECD Tax glossary defines the Arm’s length principle as: “The international standard which states that, where conditions between related
enterprises are different from those between independent enterprises, profits which have accrued by reason of those conditions may be included in the profits of that enterprise and taxed accordingly “http://www.oecd.org/ctp/glossaryoftaxterms.htm#T
29 The OECD defines Profit Shifting as: “Allocation of income and expenses between related corporations or branches of the same legal entity
(e.g. by using transfer pricing) in order to reduce the overall tax liability of the group or corporation.” http://www.oecd.org/ctp/glossaryoftaxterms.htm#B
L.A.OLDENBANNING 12 Balakrishnan, Blouin & Guay (2011) find that firms that exhibit aggressive tax planning tend to display increased financial and organizational complexity and decreased information transparency. Conversely, the market recognizes disclosures related to tax transparency as relevant information and this recognition is incorporated into firm value (Eiler & Kutcher, 2014; Boylan, 2013). Whereas an experiment by Boylan (2013) found that investors in markets with greater tax transparency led to higher profits when the investors had access to this information. An important factor to limit tax avoidance behavior is the reputational costs, since research shows that companies could be punished by consumers for aggressive tax-planning with a slight lower willingness to pay for company’s product (Boylan, 2013). Correspondingly, research shows a temporary decline (<30 days) in a company’s stock price after accusations of tax avoidance activities (Hanlon & Slemrod, 2009; Gallemore, Maydew, & Thornock, 2014). Companies which use tax aggressive strategies diminish their corporate success with consumers, whereas Hardeck and Hertl (2014) found that responsible tax behavior enhances corporate success.
L.A.OLDENBANNING 13 THEORETICAL FRAMEWORK AND HYPOTHESES DEVELOPMENT
SHAREHOLDER THEORY
Based on the shareholder theory an organization’s main responsibility is the maximization of shareholders’ value. Therefore, managers will not intentionally behave and/or participate in any activities which could interfere with profit maximization. Reinhardt, Stavins and Vietor (2008) define CSR as “sacrificing profits in the social interest”. Fisher (2014) uses an equivalent definition of CSR: “obligations and inclinations, if any, of corporations organized
for profit, voluntarily to pursue social ends that conflict with the presumptive shareholder desire to maximize profit.” Taking these definitions of CSR into account; evidently organizations will not prioritize CSR activities. Let
alone that companies will sacrifice profit only for the sake of the alignment of their words and actions. From a shareholder’s perspective it can be argued that tax is seen as an expense which should be minimized. This implies that tax minimization is a fiduciary responsibility to shareholders (Dowling, 2014). As Marens & Wicks (1999) state: “This fiduciary duty requires the exercise of care loyalty, and honesty with regard to the financial interest of stockholders”. However, Christensen and Murphy (2004), argue that in the case of tax aggressive strategies this may not be in a shareholders’ broader interest for a firm since it is not compatible with long-term sustainability. Tax aggressiveness has the potential to harm the organization and thus its shareholders by damaging the firm’s reputation (Fisher, 2014).
AGENCY THEORY
In modern organizations, and especially multinationals, shareholder ownership is often widely dispersed. Shareholders (in agency theory terms: the principle) appoint a manager (in agency theory terms: the agent) and delegate decision-making authority on the manager. The manager acts on behalf of the shareholders in business-activities. Building on the shareholder theory which emphasizes the requirement of shareholder value maximization (Donaldson & Davis, 1991). The first problem arises when the desires or goals of the agent and principal are in conflict. The assumption is that human beings are bounded rationale and behave in self-interest. Besides, for the principal it is difficult and/or expensive to verify if the agent acts in the requisite way (Eisenhardt, 1989). In addition, the risk preferences can differ between the agent and principal which could lead that the agent may prefer different actions (Eisenhardt, 1989). The traditional view of tax aggressive behavior implies that the shareholder value directly increases. However, Desai & Dharmapala (2009) show that agency problems characterized by agent-principal relation influence the presumptuous view that tax aggressive behavior positively influences a firm’s value. The agency and information problem can be minimized however this is costly and will therefore decrease the expected positive value of aggressive tax behavior (Desai & Dharmapala, 2009). On the other hand, if an organization obtains a high score on CSR disclosure and/or activities this will lower cost of capital through the reduction of information asymmetry and agency problems (El Ghoul, Guedhami, Kwok & Mishra, 2011; Chen, Chen & Wei, 2009). The disclosure of nonfinancial information, such as CSR, is essential in the reduction of information asymmetry between important stakeholders and management (Cohen et al, 2012). Assurance is integral to agency theory since it is used as a monitoring mechanism (Wong & Millington, 2014). Research shows that obtaining assurance on CSR can lead to a reduction in information asymmetry (Casey & Grenier, 2015). This relates to the firms reduction in management in social and environmental risks, since investors use this information to assess the long-term viability (Casey & Grenier, 2015), because it allows investors to assess key areas of performance better and a broader view of organizations performance that comprehends society at large (Holder-Webb et al. 2009; Cohen et al, 2012).
L.A.OLDENBANNING 14 STAKEHOLDER THEORY
An organization as seen from a stakeholder perspective is embedded in society and corporate behavior will be judged by stakeholders on which the organization depends. Freeman (1984) proposed that stakeholders are ‘any group or individual who can affect or is affected by the achievement of a corporation’s purpose’. From this perspective a company thus relies on their stakeholders. Therefore, the organization needs to identify and interact with the stakeholders to see what the stakeholders’ expectations, goals and views are on the corporation in order to be successful (Burrit & Schaltegger, 2010). As Näsi and Mäkelä (2010) highlight, organizations need to find a balance in managing the demands and rewards from stakeholders. Because every stakeholder has its own demands and expectations of a company and contribute in their own way in the company. Nowadays, stakeholders are expecting more from corporations on demands for an ethical, social or environmental account in addition to the financial account (Adams, 2004). In order to cooperate, in a group, in a sustainable way some actors must comply to the choices of others (Castro, 2014). This means that an organization will receive contradicting demands by stakeholders and therefore needs to be decisive which demands the organization will choose to meet. This so called “zero-sum game” indicates that when an organization allocates its resources to a particular stakeholder group these resources are taken away from another stakeholder (Parmer et al, 2010). From a normative stakeholder perspective “Managers should not try to satisfy the interests of stakeholder groups
in expectation of profitability in the long term, but because it is their fiduciary duty to do so.” stated by Van Aaken,
Splitter & Seidl (2013). However, an argument against the stakeholder approach is known as the stakeholder paradox. The stakeholder paradox can be explained as follows, where the organization is in a position where it has to choose between the fulfillments of fiduciary responsibilities for shareholders and serving other stakeholders’ interests (Marens & Wicks, 1999). Looking at the fiduciary duties paradox from normative standpoint the paradox in the fiduciary duties are irrelevant since the duties must be morally defendable and do not give permission for an organization to violate normative ethics in accordance with other stakeholders (Freeman, 1994; Jones, 1995). Hendry (2001) states that the stakeholder movement seems to have turned people purely into economic stakeholders. This contradicts Kant’s philosophy since stakeholders are now used as purely a mean according to Hendry (2001). However, ethics are not primarily about economics and law. Otherwise tax avoidance would not be a public debate. As Hendry describes it perfectly: “It is primarily about the
social relationships between moral actors, and if we are to establish an ethic of business and of corporate governance we must surely begin with a construction of the firm in terms of these relationships.”
L.A.OLDENBANNING 15 HYPOTHESES
Sustainability reporting
Buchholz & Rosenthal (2006) and Messner (2009) argued that an organization can be considered as a moral agent and therefore also hold ethical responsibilities. Macro-perspective tax aggressiveness has a negative effect on the national budget (Kirchler, Maciejovsky & Schneider, 2003). As counter-plea for this unsocial behavior one could argue that companies might act in ethical blindness, partially because of the unawareness and unconsciousness of their unethical behavior (Palazzo, Krings & Hoffrage, 2012). However, in issuing a sustainability report or an integrated report a consciousness decision has been made to do so, therefore rejecting the claim of ethical blindness. Hence, a corporation’s words and actions should be align and show their integrity in CSR words expressed in actions. Thus this should lead to lower tax aggressiveness. The complex nature of transactions makes it difficult for stakeholders to fully evaluate firm performance, including the tax implications of these transactions or arrangements (Desai and Dharmapala, 2007). Because the stakeholder theory includes shareholders as well as other stakeholders and is in line with CSR. I therefore state, based on the stakeholder theory, the following hypothesis:
Hypothesis 1. Companies that issue or integrate a sustainability report show lower tax aggressiveness compared to companies that do not issue such a report.
Assurance statement on sustainability reporting
As noted in the introduction, there is a rise in demand for voluntary assurance on CSR contents by organizations. From a stakeholder perspective, the assurance on a sustainability report by an independent professional is considered more reliable (Garcia-Benau, Sierra-Garcia & Zorio, 2013). Many firms issue CSR reports following The Global Reporting Initiative (GRI) guidelines. The latest guideline released by the GRI, the G4 recommends organizations to use external assurance on the CSR report (p.14, GRI G4 guidelines30). However for the time being
the GRI does not require the assurance to be ‘in accordance’ with the G4. Nonetheless, CSR assurance is associated with lower cost of capital and lower analyst forecasts errors and dispersion (Casey & Grenier, 2015). Especially taking into account that unassured CSR reporting could be perceived as greenwashing (Lyon & Maxwell, 2011). As a result of corporate wrongdoings who are inflicted in CSR and the media attention that came along with it growing skepticism emerged because of missing transparency between what organization disclose on CSR information and the actual results whether corporation lives up to their standards and their involvement in CSR activities (Skarmeas, Leonidou & Saridakis, 2014; Lyon & Maxwell, 2011). The sustainability framework issued by IFAC calls auditors to extend their role and pay greater attention to engagement between the organization and its stakeholders in the process of conducting assurance. The possibility that assurance on CSR would help to ensure more transparency and credibility of an organization in relation with all stakeholders (O’Dwyer, 2011). Assurance on CSR influences the assessment conducted by stakeholders on the value of CSR information in a positive way (Wong & Millington, 2014). Thus, from a stakeholder perspective I formulate the following hypothesis:
Hypothesis 2. Companies with assurance on CSR information differ in tax aggressiveness compared to companies without assurance.
L.A.OLDENBANNING 16
Transparency benchmark score
Since 2004 the Dutch Ministry of Economic Affairs (Ministerie van Economische Zaken) carries out an annual Transparency Benchmark (transparentie benchmark). The Transparency Benchmark provides transparency in corporate social reporting. In the Transparency Benchmark (TB) companies are assessed to what extent they report their accountability concerning CSR activities in the annual report31. The aim for TB is to encourage
businesses to be transparent about their policies and achievements in respect of CSR. Furthermore, the TB can facilitate stakeholder dialogue and thereby increase the focus on CSR policy and performance improvement. The organization whose report is ranked as most transparent concerning CSR activities is rewarded with a prize (Kristalprijs). The company’s efforts are assessed in accordance with the transparency benchmark criteria. A self-assessment takes place by the organization and is examined by an appointed executor (currently done by EY32).
The quality of the top twentieth ranked annual reports will be supplementary assessed by a panel of experts. The top 3 annual reports are presented to a jury to determine the winner of the Kristalprijs. The aim is to strive for an annual growth in achieved scores among the organizations. Hence, an increase in score indicates more transparent information is disclosed in the annual and/or CSR report. Therefore I expect that:
Hypothesis 3. The TB scores ascribed to companies can be used as an indicator for their degree of tax aggressiveness
31 Evidently, a published CSR report is interchangeable with an annual report concerning CSR disclosures. 32 http://www.transparantiebenchmark.nl/Over%20Transparantiebenchmark/Beoordelingsproces
L.A.OLDENBANNING 17 RESEARCH MODEL
DATA
This paper is descriptive in nature. The data is gathered in the period from 2009 – 2014 from Dutch organizations. For several reasons I will concentrate merely on Dutch organizations. First of all, a number of large multinationals which lead in CSR reporting are established and domicile in the Netherlands, for example the participation of Dutch organizations (e.g. Aegon, Akzonobel, DSM & Unilever33), in the IIRC Pilot Program. Not only organizations
but also standard setters as the Dutch-based GRI, institutions like VBDO (Vereniging van Beleggers in Duurzame Ontwikkeling) and the Transparency Benchmark introduced by the Dutch Ministry of Economic Affairs play a crucial role in the sustainability practices. On the other hand, concerning tax aggressiveness the Netherlands are often viewed as a tax haven (Van Dijk, Weyzig & Murphy, 2006) because multinationals use the Dutch tax system to minimize and shift profits (For example the “famous” Double Irish with a Dutch Sandwich34 is often used for
tax relocation). The Netherlands is a country known for the concluding tax treaties with a large number of countries35. These tax treaties are concluded to avoid the double taxation of (corporate) income from a
perspective that the Netherlands wants the tax to be charged where the activities take place. Besides large companies often have subsidiaries and often subsidiary companies distribute their profit (in the form of dividend) to parent companies. The Netherlands has a substantial holding exemption ruling that exempts the parent company from paying tax on dividends received from subsidiaries36. When Barack Obama delivered a speech on
tax avoidance in 2009 a press release qualified the Netherlands as a tax haven37. Dutch politicians (i.e. Kees-Jan
de Jager and Wouter Bos) were caught off guard by this statement and demanded rectification and the White House adjusted the press-release deleted the sentence which stated that The Netherlands (as well as Ireland and Bermuda) are low-tax countries 38. The fact that the Netherlands contributes to both fields (i.e. tax avoidance
and sustainability) makes this country very interesting. Also, the leading role in CSR reporting by Dutch organizations implicated a better availability of CSR data.
The gathering of data concerning sustainability reporting, level of assurance on CSR, and (long-run) effective tax rate is difficult. Leading databases miss the accurate CSR variables needed in this research. Therefore I manually retrieved the CSR and CSR Assurance data. Likewise, financial data concerning the measurement of the (long-run) ETR were on some occasions incomplete and thus the missing data had to be acquired from the annual report of the company. Organizational information is gathered using Orbis and COMPUSTAT. The organizations are selected based on their publicly listed status in the years ranging from 2009 up to 2014 in Orbis. COMPUSTAT Global gave the possibility to search the database based on country code. The country of Incorporation code identifies the country in which the company is incorporated or legally registered, the country code of the Netherlands is NLD. The companies retrieved from both Orbis and COMPUSTAT were matched in order to gather financial information of the organizations. The financial data was employed from COMPUSTAT. Tax aggressiveness is measured by the effective tax rate. COMPUSTAT is the most suitable database to gather financial data for this study since the measurement of ETR is based on Compustat data items: cash tax paid, pre-tax income and special items. The pre-pre-tax income is the commercial profit of a company. In this research all the
33 http://examples.integratedreporting.org/
34 http://www.nytimes.com/interactive/2012/04/28/business/Double-Irish-With-A-Dutch-Sandwich.html?_r=0 explains, in a simplified
manner, how this tax construction is used by multinationals.
35 http://www.government.nl/issues/tax-treaties/tax-treaties-in-the-netherlands 36 http://www.government.nl/issues/taxation-and-businesses/corporation-tax
37 http://journalistiek.npo.nl/dossiers/belastingparadijzen/30_66903-nederland-belastingparadijs
L.A.OLDENBANNING 18 results and profits are commercial and thus not interchangeable with fiscal results. Depending on the data available per company, there is a range between 2 and 6 years for the calculation of the long-run ETR. In the case of several firms some data is retrieved from annual reports. For the transparency benchmark score I had to gather the data from the TB score archive39. The Dutch statutory corporate tax rate has been 25,5 % in 2009 and from
2010 on the statutory corporate tax rate is 25%. Evidently, tax incentives are existent in Dutch tax legislation. This legislation makes it possible that a company’s average ETR is lower than the statutory tax rate. It is thus important to emphasize that a low ETR not necessarily implies that a company is engaged in anything improper. However, some tax incentives (e.g. leverage) are used for tax aggressive behavior (Wang, Campbell & Johnson, 2014). Since there is no way to measure the intrinsic motivation of a company to use a tax incentive I did not take into account that a company might have a discount on the statutory tax rate.
L.A.OLDENBANNING 19 MODEL
To test the hypotheses, explained in the hypotheses section, I will use a linear regression. The dataset is unbalanced since the number of years to calculate the effective tax rate as well as the number of years for which the TB scores are available can differ per company as a result of missing or incomplete data. Two models are designed to research the proposed hypotheses.
The first model describes the relationship between CSR reporting and CSR assurance in relationship with tax aggressiveness. The following model is used:
𝑇𝐴𝐺 = 𝛼0+ 𝛽1𝑖 𝑆𝑈𝑆 + 𝛽2𝑖 𝐴𝑆𝑆 + 𝛽3𝑖 𝑆𝐼𝑍 + 𝛽4𝑖 𝐿𝐸𝑉 + 𝛽56𝑖 𝑅𝑂𝐴 + 𝜀_𝑖
The second model explores the association of the Transparency Benchmark score with tax aggressiveness. The model is as follows:
𝑇𝐴𝐺 = 𝛼0+ 𝛽1𝑖𝑇𝑅𝐴 + 𝛽2𝑖 𝑆𝐼𝑍 + 𝛽3𝑖 𝐿𝐸𝑉 + 𝛽4𝑖 𝑅𝑂𝐴 + 𝜀_𝑖
For both models the alpha_ 𝛼 is the intercept giving direction to the results whether this is positive or negative, the betas_ 𝛽 are the coefficient with the coherent number, and the epsilon is the error_ 𝜀 term of the regression which shows how much of the result could be explained by conditions external to this model. For an overview of the variables used and explanation see Appendix, Table 9 explanation of the research model.
THE DEPENDENT VARIABLE
In this research tax aggressiveness is measured by the long-run effective tax rate of a corporation. Hanlon and Heitzman (2010) show an overview of measurements used to calculate the effective tax rate (hereafter, ETR). Tax aggressive behavior is predominantly attained by long-term tax planning. The best way to expose tax aggressive behavior is done by looking into the long-term ETR of a firm. If the ETR is measured by only one year an occasional tax liability and/or asset could occur which could lead to a distorted ETR. The benefit of using the long-term ETR as a measure of tax avoidance is the long time period avoids the year-to-year volatility in a company’s annual ETR (Hanlon & Heitzman, 2010). Using cash taxes paid as a basis to calculate the ETR avoids problems associated with using current tax expense as a measure of a company’s tax burden since the tax expense is not equal to the cash flow of taxes. Taking multiple years to measure the long-run cash ETR smoothens out possible accrual management and the estimated tax payments (Frank, Lynch & Rego, 2009). Understandably, measuring the effects of CSR involvement of a company in relation to the tax practices can take some time before changes are reflected in a company’s tax practice. Likewise, a common misunderstanding is that a company’s reported income tax expense is solely cash taxes paid. This is not the case, because of the accrual nature of GAAP. As such, the reported income tax expense in any given year only tax expense incurred during this time period. So, if an organization reports profits in their financial reports, then they will report a corresponding tax expense, regardless of the actual cash taxes paid in that period (Graham, 2012). Enumerated, the ETR measured over several years to unravel a company’s tax behavior gives more value because year-to-year variation is less relevant. Dyreng, Hanlon, and Maydew (2008) introduced a way to measure the Long-run ETR. In my research I will follow their measurement as one of the proxies for tax aggressive behavior. To reduce any distortion, outliers below 0% ETR and above 100% ETR are winsorized to 0 or 1 in this research to prevent biased results. With the
L.A.OLDENBANNING 20 use of COMPUSTAT data items ’17. Special items’, ‘170. Pre-tax Income’, ‘317. cash tax paid’, the long-run ETR is measured40.
𝐿𝑜𝑛𝑔 − 𝑟𝑢𝑛 𝑐𝑎𝑠ℎ 𝑒𝑓𝑓𝑒𝑐𝑡𝑖𝑣𝑒 𝑡𝑎𝑥 𝑟𝑎𝑡𝑒 = ∑ 𝑐𝑎𝑠ℎ 𝑡𝑎𝑥𝑒𝑠 𝑝𝑎𝑖𝑑
∑(𝑃𝑟𝑒 𝑡𝑎𝑥 𝑖𝑛𝑐𝑜𝑚𝑒 − 𝑠𝑝𝑒𝑐𝑖𝑎𝑙 𝑖𝑡𝑒𝑚𝑠)
Since limited data availability causes a relatively small sample size for long-run cash ETR, I will also conduct regression analyses using the cash ETR per firm-year. I recognize that using cash ETR could potentially be noisy on annual basis because the possible volatility of a discrepancy between tax expenses and tax paid. On the other hand, the number of companies used in this data set will be consistent with 72 companies. Using the cash ETR calculation the firm-year observations will increase my total sample size. Because this research only includes companies with a minimum of 2 fiscal-year observations this can evidently lead to minimize any volatility which might occur through deferred tax activities (Dyreng, Hanlon & Maydew, 2010). I find it therefore justified to use annual cash ETR as an extra proxy to measure tax aggressiveness. The ETR calculation is based on COMPUSTAT data items: ‘17. Special items’, ‘170. Pre-tax Income’, ‘317: cash tax paid’.
𝐶𝑎𝑠ℎ 𝑒𝑓𝑓𝑒𝑐𝑡𝑖𝑣𝑒 𝑡𝑎𝑥 𝑟𝑎𝑡𝑒 = 𝑐𝑎𝑠ℎ 𝑡𝑎𝑥𝑒𝑠 𝑝𝑎𝑖𝑑
𝑃𝑟𝑒 𝑡𝑎𝑥 𝑖𝑛𝑐𝑜𝑚𝑒 − 𝑠𝑝𝑒𝑐𝑖𝑎𝑙 𝑖𝑡𝑒𝑚𝑠
Following McGuire, Omer & Wang (2012) and Minnick and Nofa (2010) I will use the book effective tax rate conducting research on the Transparency Benchmark score in relation to corporate’s tax aggressive behavior. I recognize that there are differences between the measurement of effective tax rate using the CASH ETR and book ETR. The book effective tax rate is measured over a one-year period. Likewise the Transparency Benchmark score obtained by organizations also cover a one-year period. The book ETR is commonly used as a measure of a company’s tax burden (McGuire, Omer & Wang, 2012; Dyreng, Hanlon & Maydew, 2010). The book ETR is less volatile than the cash ETR because it only reflects the tax activities that directly affect a corporation’s net income. Whereas cash ETR also reflects deferred tax activities by taxes paid (Dyreng, Hanlon & Maydew, 2010). To my best knowledge this is the first scientific research as to the Transparency Benchmark. The Transparency Benchmark does not always include the same organizations per year. Therefore, in my opinion I find it best to minimize any volatility and prefer the BOOK ETR over cash ETR. The book ETR is measured by COMPUSTAT data item: ‘170. Pre-tax Income’ and ‘16. total income taxes’.
𝐵𝑜𝑜𝑘 𝑒𝑓𝑓𝑒𝑐𝑡𝑖𝑣𝑒 𝑡𝑎𝑥 𝑟𝑎𝑡𝑒 = 𝑡𝑜𝑡𝑎𝑙 𝑖𝑛𝑐𝑜𝑚𝑒 𝑡𝑎𝑥𝑒𝑠 𝑝𝑟𝑒 𝑡𝑎𝑥 𝑖𝑛𝑐𝑜𝑚𝑒
40 Due to the relative small sample size missing values in item 317: cash tax paid where retrieved from company’s annual report. The value of cash tax paid can often be found in cash-flow statement and/or in tax disclosures in the report.
L.A.OLDENBANNING 21 THE INDEPENDENT VARIABLES
CSR reporting
CSR will be measured by the presence of a sustainability or integrated report of the organization. No distinction was made between sustainability and integrated reporting and therefore a dummy variable was used in order to conduct the linear regression. 0 stands for no CSR report and 1 for the presence of a CSR report. The calculation of the long-term ETR of a company starts from the first year a company issues a CSR with a minimum of at least 2 years to calculate the ETR.
Level of assurance
Assurance on CSR information enhances the creditability. However, different levels of assurance can be given by an auditor. To type of assurance is measured by analyzing the content of the assurance statement in the sustainability report. There are limitations in the degree of assurance provided by an assurance-provider (O’dwyer, 2005). Therefore, the level of assurance will be categorized in 4 categories ascending in level of assurance (see Table 1).
The level of assurance depends on the desires of the responsible party, the intended user and the ability of the practitioner to meet the qualifications for the assurance engagement41 An auditor is not able to give absolute
assurance because there are inherent limitations to an audit (e.g. human error, practical and/or legal limitations to obtain sufficient appropriate evidence). However, an audit provides the highest level of comfort an auditor can give. In an audit or “reasonable assurance engagement” the auditor makes sure to obtain sufficient appropriate evidence to support a company’s reported information. The audit (or reasonable assurance procedures) allow the auditor to express an opinion (or positive form statement) about the reported information. To give an example: After KPMG audited the sustainability statements and sections in the annual (integrated) report concerning social and environmental performance the Independent Auditor’s Assurance Report of 2014 by KPMG Accountants NV and addressed to the Supervisory Board of Koninklijke Philips N.V. stated that: “In our opinion, The Sustainability Information presents fairly, in all material respects, the sustainability performance of Philips in accordance with the reporting criteria as mentioned below.“42
If a company requests limited (sometimes referred to as moderate) assurance on their CSR report the auditor will conduct a review as a basis for the auditor’s conclusion. The auditor gathers sufficient appropriate evidence to conclude that the subject matter (information) is plausible in the circumstances. The auditor reports this in the form of a negative assurance by given an indirect opinion. To give an impression: the directors of Unilever engaged PricewaterhouseCoopers to provide limited assurance on CSR information. The auditor’s conclusion was formulated as: “Based on the procedures we have performed and the evidence we have obtained, nothing has come to our attention that causes us to believe that the Selected Information for the 2014 reporting year has not been prepared, in all material respects, in accordance with the Reporting Criteria.” 43
Organizations sometimes engage auditors to provide limited assurance on certain parts of CSR report/information along with reasonable assurance on other parts of the CSR report. This is categorized as
hybrid assurance where limited as well as reasonable assurance is given in one auditor’s report.
41 https://www.nba.nl/Documents/Publicaties-downloads/Brochure%20het%20assurance%20raamwerk.pdf
42 http://www.2014.annualreport.philips.com/#!/independent-assurance-report/
L.A.OLDENBANNING 22
No assurance means that a company did not subject their CSR disclosure/report under the assessment of an
auditor.
Transparency Benchmark score
The Transparency Benchmark score is released annually. The TB score released in year X is based on the transparency on CSR in regard to the previous year. The score attained in a year is based on the CSR information in a company’s annual (integrated) report or sustainability report. The maximum score to obtain is 200. The scores can be attained in several categories in the transparency benchmark and all the points gathered in the different categories will sum up to total TB score. So for example KPN’s total TB score over 2014 was 193 points. The score is based on KPN’s integrated44 annual report 2013. To link the ETR with the correct TB score I therefore
used t-1, where t is fiscal year. As mentioned before, the dataset is based on the years 2009 up to 2014. For this particular analysis, all the firm observations of the year 2014 have been discarded because the TB score based on the firms’ 2014 CSR disclosure will be released in 2015. I thus used the ETR data of 2009, 2010, 2011, 2012 and 2013 and matched them respectively with TB results of 2010, 2011, 2012, 2013 and 2014.
44 KPN publishes an annual integrated report according to the GRI and IIRC standard since fiscal year 2013.
http://www.duurzaam-ondernemen.nl/een-geintegreerd-jaarverslag-voor-kpn/ Category Level of Assurance
0 No assurance
1 Limited/moderate assurance
2 Hybrid assurance
3 Reasonable assurance
L.A.OLDENBANNING 23 CONTROL VARIABLES
Size
The size of a company is measured by the total assets. To reduce the impact of skewed distribution (heteroskedasticity issues) a standard procedure is to use the logarithm of the total assets in the regression. I will use the average of total assets. This is in line with the calculation of the long-run ETR, since total assets of an organization can vary per year. Simnett, Vanstraelen & Chua (2009) found that the size influences the chance on reporting CSR. Whereas Frank, Lynch & Rego (2009) found a significant relation between tax aggressiveness and size.
log 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑡𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠 =∑ 𝑡𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠 𝑡 Number of years
Return on assets (ROA)
Frank, Lynch and Rego (2009) found that ROA is positively and significantly related with tax aggressiveness. These findings match the research conducted by McGuire, Omer and Wang (2012) where a higher return on assets is associated with tax avoidance. Therefore, the ROA of a company is included in this research as a control variable. The return on assets (ROA) is a performance variable measured by:
ROA = ( mean of earnings before tax and special items average total assets )
Leverage
Previous tax research (McGuire, Omer & Wang, 2012; Frank, Lynch & Rego, 2009; Dyreng, Hanlon & Maydew, 2008; Gupta & Newberry, 1997) shows that a higher leverage has a positive effect on tax aggressiveness. Other research shows that highly leveraged firms are more likely to produce stand-alone sustainability reports (Simnett, Vanstraelen & Chua, 2009).
leverage = ( ∑Long − term debt total assets ) ÷ 𝑡
L.A.OLDENBANNING 24 RESULTS
DESCRIPTIVE STATISTICS
Long-run cash ETR data is included for the years 2009 up to 2014. 323 firm-year observations are examined spread over 72 companies. This leads to an average of 4,49 fiscal years per company to calculate the long-run cash ETR, with a minimum of 2 fiscal years per company and maximum of 6 fiscal years depending on the available and qualified data of the company. The average corporate tax rate paid (based on cash ETR per fiscal year) is 22,05 % (see appendix 2, table 10). The long-run cash ETR over the 72 organizations is 20,85 %. Following Dyreng, Hanlon and Maydew (2008) to make cash ETR more interpretable, values are winsorized at 0 and 1. The dataset with winsorized values has a fairly normal distribution before conducting the linear regression analyses.
A total of 42 companies publishes sustainability reports (or an integrated report), leaving the rest of the organizations, 30 companies lacking a stand-alone CSR report. Overall CSR reporting was conducted in 182 out of a total of 323 firm-years of all 72 companies, leaving 141 firm-years without CSR reporting.
The level of assurance is based on 182 firm-year observations, these firm-years are evidently from the 72 corporations who issue CSR reports. The mean of these observations is 0,8297 with a minimum of 0,00 and a maximum of 3,00. The mean of the level of assurance per company is measured by the sum of the level of CSR assurance divided by the number of assurance years per company. The average is 0,4759 over 72 companies with a minimum of 0,00 and the maximum average per company is 2,6667.
The size (based on the logarithm of total assets) lies between 1,266 – 6,061. The mean of the sample of 72 companies is 3,187. The average based on the 323 firm-year observations is 3,1621, with the highest being 6,1050 and the lowest being 1,1028.
The leverage in the sample size based on companies lies between 0,00 and maximum of 0,505, with the mean of an average of 0,151 within the 72 companies. The highest observation is a 55,19 % leverage per firm-year. The mean of the 323 firm-year sample is 14,7 % and the lowest observation is 0% per firm-year.
The average return on assets for the 72 companies is 0,083 when the proxy long-run cash ETR is used to measure tax aggressiveness. Using annual cash ETR the mean of ROA is 0,09 with the sample of 323 firm-years. The minimum average return one of the seventy-two company performed is 0,004 while the maximum average return was 0,403.
Availability csr report Per firm year Per company Frequency Percentage Frequency percentage
No CSR report 141 43,7 30 41,7
CSR report 182 56,3 42 58,3
Total 323 100,0 72 100,0
L.A.OLDENBANNING 25 REGRESSION ANALYSES R ESULTS
RESULTS BASED ON MOD EL USING LONG-RUN CASH ETR AS PROXY TAX AGGRESSIVENESS
A linear regression analyses is used to examine the association of CSR reporting and Assurance on CSR with tax aggressiveness. For the first regression to test the significance of the model the variable Long-run cash ETR is used as a proxy for tax aggressiveness. In table 11 (See appendix 3) the correlations of the model is explained. For the independent variables: CSR reporting and Assurance on CSR correlated respectively 0.084 (significance level. P=<.485) and 0.039 (significance level p=<.745). For the control variables size, leverage and ROA the correlations are respectively -0.009 (p=<.939), 0.101 (p=<.400), -0.168 (p=<.150). Indicating that it is highly unlikely to find any significance in the regression analysis. In table 3 (see below) a significance level of 0,364 on sustainability reporting, Assurance on CSR has a significance 0,760. The firm size shows a significance of 0,223, the leverage of a company shows 0,343 significance, and return on assets indicates a significance of 0,116. In this research a significance level is of 0,10 or lower indicates that the variable has a significant influence on tax aggressiveness. As noted above, all variables (Sustainability reporting, assurance on CSR, size, leverage and ROA) show higher score. Therefore, the conclusion can be made that no significant indication can be found that leads to believe that companies with CSR reports are less tax aggressive. The same conclusion can be made in regard to companies with assurance on CSR, no suggesting leads to believe that assurance on CSR differ in tax aggressive behavior. These conclusions are made based on the proxy long-run cash ETR as a measure for tax aggressiveness. Although, Lanis and Richardson (2012) found that the more CSR disclosure led to less tax aggressive behavior the previous results do not argue the same findings.
Model Unstandardized
Coefficients
Standardized Coefficients
t Sig. Collinearity Statistics
B Std. Error Beta VIF
1 (Constant) ,304 ,087 3,501 ,001 sustainability report ,045 ,049 ,140 ,914 ,364 1,659 Assurance CSR ,010 ,033 ,048 ,307 ,760 1,713 Size -,037 ,030 -,213 -1,229 ,223 2,102 Leverage ,158 ,165 ,123 ,955 ,343 1,161 Return on Assets -,410 ,258 -,197 -1,593 ,116 1,079
Dependent Variable: Long-run cash ETR table 3. Results of linear regression using long-run cash ETR
L.A.OLDENBANNING 26
RESULTS BASED ON MOD EL USING CASH ETR AS PROXY TAX AGGRESSIVE NESS
Measuring tax aggressiveness based on annual cash ETR increased the sample to 323 firm-year observations. Increasing the firm-year observations could perhaps lead to a more robustness to the model tested with a linear regression analysis. The linear regression analysis conduct on the model including annual cash ETR as proxy for tax aggressiveness is shown in table 4. The independent variable sustainability reporting shows a significance level of 0,434. Assurance on CSR has a significance of 0,407. The control variables: size, leverage and return on assets show significance with the level of 0,684, 0,492, and 0,000. The results indicate that the variables: CSR reporting, level of assurance, size, and leverage do not expose any influence on tax aggressive behavior. Return on Assets is with is significant in relation to tax aggressiveness (when using a significance level of p=0,10 in order to accept the significant influence). The beta (shown in table 4 under unstandardized coefficients B) of Return on Assets show a negative coefficient of 0,469. Indicating that a higher Return on Assets by a firm shows a lower cash ETR and thus implies more aggressive tax behavior. This finding is in line with previous research (Frank, Lynch and Rego, 2009; McGuire, Omer and Wang, 2012) conducted on tax aggressiveness.
Model Unstandardized
Coefficients
Standardized Coefficients
t Sig. Collinearity Statistics
B Std. Error Beta VIF
1 (Constant) ,303 ,046 6,581 ,000 sustainability report -,022 ,028 -,057 -,783 ,434 1,741 Assurance CSR ,014 ,017 ,058 ,830 ,407 1,598 Size -,007 ,016 -,033 -,408 ,684 2,137 Leverage -,062 ,090 -,041 -,688 ,492 1,184 Return on Assets -,469 ,124 -,213 -3,780 ,000 1,059
Dependent Variable: annual cash ETR table 4. Results of linear regression using annual cash ETR
All in all, the previous results on tax aggressiveness using long-run and annual cash ETR show no association whatsoever. The possibility of an increase in sample size could lead more robustness do not stand with the findings. Based on the model used to conduct the research it can be implicated that the independent variable CSR reporting do not have a significant relation on tax aggressiveness. No difference is found between the tax aggressive behavior with companies who do issue a CSR report and firms which choose to and CSR assurance based on these results show no difference in using the proxies annual cash ETR and long-run cash ETR. Furthermore, companies with assurance on CSR do not show any difference in tax aggressive behavior according to these findings, than firms without any assurance on CSR. Concluding that, hypothesis 1 (Companies that issue
or integrate a sustainability report show lower tax aggressiveness compared to companies that do not issue such a report.) and hypothesis 2 (Companies with assurance on CSR information differ in tax aggressiveness compared to companies without assurance.) are not statistically supported in this research.
