The effects of shareholder value-focus on average wage and the effective tax rate of MNEs based in the United Kingdom

(1)

The effects of shareholder value-focus on

average wage and the effective tax rate of

MNEs based in the United Kingdom

Matthijs Slaghekke - S2186160

MSc International Business & Management

University of Groningen

10-08-2017

Supervisor: Drs. A. Visscher

(2)

2 Word count: 14.649 (excluding appendices, figures, tables and references)

Abstract:

In recent years much research has been conducted on how much tax is being avoided and the responsibility firms have for employees in at home and in developing economies. A number of decades ago Milton Friedman advocated that the only responsibility of firms was to make as much profit as possible. In recent years major firms have been developing means, aided by globalization, to evade tax, thereby lowering their effective tax rates and to ship large parts of production to developing economies. Both the decision on tax avoidance and height of wages are theorized to be dependent on the focus on shareholder value present in the firm. As CSR reporting is considered to be an indication of how much a firm takes interests of other individuals into consideration, it is seen as a predictor of shareholder value focus. Therefore, through means of quantitative research, this paper examines whether there is an effect of he shareholder value focus measured by CSR, on ETR and wage relative to other MNEs. No relationship has been found between CSR, ETR and wage.

(3)

3

Abbreviations and acronyms

CSR Corporate social responsibility

CME Coordinated market economy

ETR Effective tax rate

LME Liberal market economy

LW Living wage

MNE Multinational Enterprise

IRS Internal Revenue service

UK United Kingdom

US United States of America

List of figures and tables

Figure 2.1 Conceptual model: The effect of shareholder value focus

Figure 2.2 The spectrum of firm orientations

Figure 4.1 Scatterplot of the relationship between CSR reporting and ETR Figure 4.2 Scatterplot of the relationship between CSR reporting and

average wage

Figure 4.3 Boxplots of the distribution of total assets before and after transformation

Figure 4.4 Scatterplots of all variables

Figure 4.5 Normal Q-Q plots of average wage and average ETR Figure 4.6 Table of Results of the linear regression analysis for CSR

reporting and ETR

Figure 4.7 Table of Results of the linear regression analysis for CSR reporting and wage

(5)

5

1 – Introduction

1.1 – Historical background

In 1776 Adam Smith wrote the wealth of nations in which he argued that when an individual serves his own best interest, he also serves the rest of society. In his view any transaction was positive for both sides, since if either buyer or seller would not value the goods offered by the other party, no deal would be agreed upon. In this way both sides of a transaction get the best out of the deal they make with each other. This theory has been the basis for western economics (Slemrod, 2004). Taxes are an exception on this rule since the payment of taxes is not a voluntary act. In the 18th century Benjamin Franklin made the next remark:

“In this world nothing can be said to be certain, except death and taxes.” - Benjamin Franklin

However, since the 18th century times have changed and in recent years multinational

companies (MNEs) have been shifting their profits all over the globe by making use of elaborate tax avoidance schemes in order to avoid paying taxes (Avi-Yonah, 2000). The Huffington post (2013) reported that the federal US government missed 150 billion dollars in tax revenues on a yearly basis because of these schemes. Multinational firms, facilitated by a globalized world have attempted to bend the rules to their advantage by making use of loopholes, thereby working around mandatory tax legislation. Globalization has also made it possible for MNEs to have products produced in developing economies and gaining competitive advantages by producing as cheap as possible. Sweatshops were among the consequences of the new power distribution in the world (Yu, 2008).

Causes for the development of large scale tax avoidance and the effect on wages have been theorized to be a focus on creating shareholder-value and globalization (Avi-Yonah, 2000) (Yu, 2008) (Alderson & Nielsen, 2002).

Lanis & Richardson (2012) define tax aggressiveness as “The downward management of

taxable income through tax planning activities”. As they point out themselves, this definition

(6)

6

avoidance for the reason that tax evasion will not show in the data sources used in this research. It is neither the focus of this research to uncover tax evasion schemes.

A lot of commotion has arisen about the elaborate tax avoidance schemes which large multinational firms have set up (Foster, 2013). These are firms which are not on the brink of failure, but make millions in profits every year. The public has been calling for a just and balanced tax system in which these firm’s pay their fair share (Christensen & Murphy, 2004). However, politics and tax havens which reap the benefits of the current system are hindering improvement of the tax legislation globally, making progress go very slow (Hines, 2005), (Avi-Yonah, 2000).

Although it is generally considered to be desirable that a firm pays taxes over its profit in order to contribute to society, a number of firms chooses not to do so. According to Friedman (1970) the only social responsibility of firms is to make as much profit as possible while staying inside the boundaries of the law. In his view by making as much profit as possible the whole of society would benefit since this would be the most effective way of production and of creating value. A firm as such could have no other responsibilities because a firm is an artificial entity and could never have the social responsibilities an individual has (Friedman, 1970). Part of this philosophy, when taken to the extreme, is that firms have no social responsibility to pay their fair share in taxes as long as taxes are avoided in a legal way (Avi-Yonah, 2008).

In 1976, 6 years after Friedman has written his famous piece in the New York Times, Jensen & Meckling looked into the issue of how agents could be motivated to act in the best interest of principals. They argued that top management should be given an interest in the firm and in this way an incentive could be provided to enhance firm value as much as possible. The alignment of interests of the agents and the principals should in theory cause agents to make sound business decisions that would benefit the firm. Jensen & Meckling viewed the firms as simply a number of individuals who are connected through the firm, but did not view the firm as a separate identity. This view is hereafter called the “Aggregate view” (Jensen & Meckling, 1976).

(7)

7

interests of principles and their agents, anything that is legal and increases profit is being viewed as a positive outcome. A firm does not face any direct consequences when it evades tax legally and it is therefore a simple way of increasing net profit for top management. Providing top management with incentives to maximize profit can therefore lead to active tax avoidance or even evasion (Desai & Dharmapala, 2006).

Next to the aggregate view of the firm the literature has discerned other views. These are the artificial and the real-world entity views (Avi-Yonah, 2008) or the enlightened shareholder value (Freeman, 2010). In the artificial entity view a firm is a creation of the state and the state facilitates its existence. Therefore a firm should not impose any other burden on the state and has certain responsibilities it should live up to. In the real-world entity view a firm has the same responsibilities as a person has and should act as if it had the same moral and social responsibilities. The enlightened shareholder value of Freeman holds the same standards as the real-world view of Avi-Yonah. These views have in common that firms are allowed and even encouraged to take social responsibility. When firms take their social responsibility they do not focus solely on their own interest (Freeman, 2010), but will rather take the interests of stakeholders other than shareholders into consideration. Both tax expense and wage expense are under less pressure of profit creation.

Lanis & Richardson (2012) find evidence for a relationship between corporate social responsibility reporting and the effective tax rate in Australian firms. They conclude that a positive relationship is present between CSR and ETR. The aforementioned aggregate view of the firm is in line with the findings of Lanis & Richardson (2012) that when CSR is relatively low, ETR should also be relatively low. The enlightened shareholder value accounts for the opposite effect when firms score high on CSR reporting and ETR is high.

(8)

8

2013). In a number of situations wages are cut to a bare minimum, forcing employees to make an unrealistic amount of hours or work multiple jobs (Chan, Pun & Selden, 2013). The result can be seen in factories in developing economies in which employees operate in bad working conditions and wages are substandard, or the so called sweatshops (Yu, 2008).

Yu (2008) specifically looked into the effect of codes of conduct on labour. He found a strong relationship between the presence of a code and working conditions. Although codes of conduct can be classified as part of overall CSR, CSR encompasses much more.

This paper aims to extend the current literature in two ways. First, it aims to extend on the research done by Lanis & Richardson (2012) by researching the relationship between CSR reporting and ETR in the United Kingdom. Second, it aims to make an argument for the relationship between the shareholder value focus and the height of average wage. It aims to do this by arguing that an ethic is present in a firm which determines both how much effective tax is being paid and how high the average wage is and that represents itself by means of CSR.

1.2 – Sample selection and methodology

(9)

9

knowledge of the relationship between a focus on shareholder value and wages and effective tax rates.

CSR scores will be taken from CSRHub.com. This site provides a balanced oversight of the CSR reporting of many firms and does not include either effective taxes or wages in its calculation. For the purposes of this research the average wage per employee and the effective tax rate will be taken from the Orbis database. Average wage is considered to be what the firm is spending on wage. The firm only considers the expense it makes for employees which has little to do with the actual monetary benefit the employees receive after taxes and social security payments. The effective tax rate is calculated on the basis of actual tax payments which are divided by net profit and multiplied by 100%.

A relationship will be deducted from the literature in which firms that have a low CSR reporting score are more likely to focus on value creation for shareholders and therefore also pay relatively less wage and have a relatively lower ETR. On the other side are firms which have higher CSR reporting scores and are theorized to have more emphasis on their social responsibility and will therefore also pay more wage and have a higher ETR. Therefore the research question in this paper will be:

What is the relationship between shareholder value focus, the effective tax rate and wages between MNE’s based in the United Kingdom?

The remainder of this thesis will first provide a theoretical background in chapter 2 to substantiate the relationship between the effective tax rate and wages. Thereafter the methodology will be discussed in chapter 3 and the results arising therefrom will be discussed in chapter 4. Chapters 5 and 6 will provide the discussion and conclusion.

(10)

10

2 – Theoretical background

2.1 – Introduction

Lanis & Richardson (2012) define tax aggressiveness as “The downward management of

taxable income through tax planning activities”. Tax planning activities can be executed in

both legal and illegal ways. As was mentioned before, this research will solely focus on legal tax avoidance. Therefore the definition adopted for this research is an adapted version of the definition of Lanis & Richardson (2012), namely:

Tax aggressiveness: The downward management of taxable income through legal tax planning activities.

Stiglitz (1986) discerns three ways in which the taxpayer can avoid taxes. He calls these the “principles of tax avoidance”. These principles are tax postponement, tax arbitrage across individuals facing different tax brackets and tax arbitrage across different income streams (Stiglitz, 1986). Because multinational firms operate in multiple tax jurisdictions it becomes possible to make use of loopholes. These loopholes exist because laws in different jurisdictions do not properly connect, which leads to discrepancies in one of the three areas Stiglitz (1986) summarized. Statutory tax rates also vary between jurisdictions, making it more attractive for firms to let profits be taxed in countries with lower statutory tax rates. Tax aggressiveness will be measured in height of the effective tax rate (ETR) of a firm. The effective tax rate is defined as:

Effective tax rate: The percentage of profit before income taxes which is spent on income taxes.

In different jurisdictions, different components, or different legal entities are taxed. This leads to the current situation which allows multinationals to bring their taxable profit down to almost zero. For example, the Netherlands levies very little taxes on royalties, leading to many large multinational firms shipping their royalties on intellectual property through the Netherlands.

(11)

11

& Meckling (1976). In the aggregate view the firm is seen as merely a set of contracts between individuals which are part of the firm and its owners. In this view the principals require the agents to act in their best interest by making their shares as valuable as possible. Avi-Yonah (2008) and Yu (2008) connects this effect to the practices exercised in the 90’s in which pressure on CFO’s led to minimization of the tax expense account. Next to the aggregate view he studies the artificial and real world-entity view in which a firm takes increasingly more social responsibility. In the artificial entity view of the firm, the firm is created by the state and has been bestowed with different rights and responsibilities. According to Avi-Yonah (2008) the firm must not impose any burdens upon the state which has created it and therefore has a responsibility to pay tax the way it was intended by the state. Taxes provide governments with the funds they need in order to be able to pay for infrastructure, social security and other facilities the government provides for. Therefore paid taxes stand to benefit the whole of society and are not used to purchase a certain specific good. For this reason it is desirable that all firms contribute to government funding through tax and do not avoid their responsibility to contribute to society.

The last view is the real entity view in which firms have the same responsibility as an individual has. Freeman (2010) calls this the enlightened shareholder value. In this case a firm should adhere to a tax ethic (Wenzel, 2005) which requires a firm to pay a fair share in tax. These views will be linked to different theories in the next sections of this chapter.

Krautheim & Eisenlohr (2016) offer another perspective on the matter of the average wage – tax avoidance relationship. They theorize that when MNE’s relocate profits abroad to tax havens, workers have a worse view of the value produced in their section of the firm. This gives the firm a better position to bargain over wages since the firm has an informational advantage. They show that this effect is present in a model with two countries but does not hold when the model is expanded to multiple countries. This research aims to extend on Krautheim & Eisenlohr by empirically testing the relationship in the UK.

(12)

12

assumed to have certain values. This makes the claim on the connection between these two statistics very weak according to Clausing. Clausing did research on country level while this research will focus on firm level effects.

Previous research is very scarce and research which has been conducted has been done on other levels like country level. This research seeks to enhance the understanding of this relationship by researching the relationship at the firm level in the United Kingdom.

2.2 – Globalization

Because taxes are mostly perceived as costs to firms, it is therefore believed that taxes need to be enforced in order to get firms to comply, since it is not in their direct self-interest to pay taxes (Wenzel, 2005). Globalization has however caused firms to have subsidiaries all over the world in multiple countries which makes it much easier to avoid taxes and to move labour intensive production processes to developing economies (Avi-Yonah, 2000), (Yu, 2008). Many firms have relocated their headquarters to more favourable tax jurisdictions in order to make it easier for them to evade tax (Laamanen, Simula & Torstila, 2012). This shift in the corporate world has had far stretching consequences for society at large as governmental income revenues declined and much manual labour was transferred overseas (Yu, 2008). Taxes which are not being paid anymore and wage which are paid abroad cause economic development overseas but lead to an increasing amount of dissatisfaction in developed economies.

Only very few firms publicly report their profits on a country-per-country basis. Reported numbers are generally global unless specified otherwise. In some countries it is even forbidden to publicize the height of tax payments. These facts make it hard to gain insight into what profits are taxed in which country or jurisdiction. There have been several initiatives to make tax payments more transparent like the Dodd-Frank act in the US, and the Capital Requirements directive IV in the EU. However, these initiative only target single industries and are mostly not complete due to the aforementioned ban on the publication of tax payments in certain jurisdictions.

(13)

13

often not willing to adapt (Avi-Yonah, 2000). Some countries like the Netherlands have an industry of tax advisors which exist mostly because of the institutional environment the Dutch tax law creates in combinations with its extensive treaty network (Bouwman, 2011). In a research paper by Dharmapala & Hines 2009) countries are found to be more likely to become a tax haven when certain institutions are present. Most tax schemes are constructed by making use of these small states which have unique tax law which makes them exceptionally useful in tax schemes (Avi-Yonah, 2000).

This development is making it much harder for governments to tax firms since these firms have many opportunities to shift profits in the globalized world. In support Lanis & Richardson (2015) find firms with larger foreign activities and larger profits to have lower global ETRs.

Labour has also been affected by the increase in mobility of capital and goods. It has been made relatively easy and cheap in recent decades to ship all kinds of goods to all corners of the world. Outsourcing parts of production to developing economies has been becoming a major part of business practices. Chan, Pun & Selden (2013) did a case study on the political process of Apple outsourcing to China and observed a buyer driven supply chain. In these buyer driven supply chains the firms operating in developed economies are experiencing a continuous pressure to be more efficient in production and to keep costs as low as possible (Chan, 2013). The employees of the factories in which the products are made are the victims of this process. They receive very low wages and have to work many hours per week. This has even resulted in a number of suicides at production plants. According to research (Crane, 2013), (Kramer & Porter, 2011) MNE’s can be held responsible for the firms they outsource to.

(14)

(15)

15 2.3 – The aggregate view and shareholder value focus

The perspective of the firm being purely a profit centre has emerged in the 1970’s with the rise of Milton Friedman and the rise of the aggregate view of the firm introduced by Jensen & Meckling (1976) in the same time period.

In 1970 Friedman wrote an article in the New York Times in which he argues the following:

”There is one and only one social responsibility of business: to use its resources and engage in activities designed to increase its profits so long as it stays within the rules of the game, which is to say, engages in open and free competition without deception or fraud” (Friedman,

1970).

Friedman argues that agents are effectively levying taxes of shareholders when firms are taking more social responsibility than is absolutely needed. The shareholders should, in his opinion, be entitled to the entire profit the firm can make, and nothing less. If the individual shareholders would like to spend their part of the profit on social issues, they can do so privately. It is not up to the firm which social projects should be invested in. In his view social responsibility is the responsibility of the government and the democratic process. Sometimes firms may have to invest into certain resources in their environment, instead of just into themselves, but this should always lead to a better result for shareholders in the long term.

Tax avoidance still lies within the extent of the law and therefore a firm is “sticking to the

rules of the game” as Friedman would state. Lawfully a firm does not do wrong by lowering

its tax bill. An ethical and practical view reveals that this course of action is not desirable (Campbell & Helleloid, 2016). Tax avoidance is however not the only way in which profits can be maximized. Other costs for the corporation should also be kept as low as possible. By imposing this view on a firm, a certain amount of shareholder value focus is presumed to be present in those leading the firm, stimulated by its owners.

(16)

16

new ownership structures motivate top executives to focus on shareholder value and this includes both boosting sales and cutting costs.

The aggregate view of the firm states that all individuals within a firm should be seen as such, since there is no overall responsibility or ethic a firm has besides its bare legal obligations and its goals of profit maximization. All individuals act in their own best interest and do not take interests of other individuals into consideration (Jensen & Meckling, 1976). Transactions typically happen at an arm’s length basis (Hall & Soskice, 2001). Any problems that may arise are presumed to be solved by market mechanisms.

In his research Avi-Yonah (2008) links the aggregate view of the firm of Jensen & Meckling (1976) to Hall & Soskice’s (2001) liberal market economies (LME). In liberal market economies firms and employees tend to behave in a more individualistic way compared to coordinated market economies (CMEs). Taking into consideration the fact that the UK market is an LME, one would expect business to operate at an arm’s length principle and have a focus on creating short term shareholder value. A common institution in LMEs is a flexible labour market in which it is relatively easy to lay off employees, and employees have a broad skillset which is not specific to one firm (Hall & Soskice, 2001). Unionization rates in LMEs typically lie lower than in CMEs which is also an opportunity for firms to more easily vary the wages they pay to employees. The combination of a focus on shareholder-value and the flexible labour market, could amplify the relationship between the focus on shareholder value and average wage. Whittaker & Hurrel (2013) stated that the amount of low paid employees is one of the highest in the UK and the US for all developed economies. Werner & Lim (2016) theorize this may be a consequence of the Reaganomics and Thatcherism policies. In those systems the government would protect the individual employees much less and take a more passive stance to interference in the economy as compared to the decades before.

(17)

17

their actions. This could be an adverse effect on the amplification discussed in the previous paragraph, since firms now will have higher CSR reporting scores while there is no immediate effect on ETR or wages. Issues which were in the explicit area of CSR have now become implicit CSR since these issues have become part of regulation. The UK makes for an interesting case to conduct this research because of these different effects. The effect of CSR will be elaborated on in the next section of this chapter.

Avi-Yonah (2008) comes to the conclusion that if the situation Friedman argues for is taken to the extreme governments would fail. This is due to firms not being allowed to exert any social responsibility other than in their own interest. Part of this profit maximizing process will be to lower effective taxes paid to the government, thereby reducing the capital the government has at its disposal to tend to all social matters in society. This is reason to argue taxes are ought to be paid by firms as a part of their social responsibility.

Figure 2.1: Conceptual model: The effect of shareholder value focus

The above theory leads to the first hypotheses.

Hypothesis 1: Firms with relatively high focus on shareholder value will have relatively high tax avoidance.

Hypothesis 2: Firms with relatively high focus on shareholder value will have relatively low wages.

(18)

18

doing so. The effect of shareholder value focus and how to ascertain a viable measure for this concept is discussed in the next section of this chapter.

As multinationals which experience the possibilities offered by a globalized world get the opportunity to increase the value for their shareholders, it is only the ethical considerations of such a choice that can influence both the decision to determine the level of tax aggressiveness and wages.

2.4 – The real world- and artificial entity-view

In the previous paragraph the consequences of a shareholder value driven way of conducting business has been discussed. The conclusion that firms with a focus on creating value for shareholders will result in relatively low ETRs and will have low average wages can also be viewed in a different light. There are many firms which take more social responsibility than is legally required. The generally accepted term for these activities is corporate social

responsibility (CSR). The issue with CSR is however that there is no one “ultimate” definition of the term and that there are even more ways to measure it. The next section will touch on this subject since it is closely related to the relationship researched and the CSR score will be used as a predictor value for the shareholder value focus.

Currently corporate social responsibility has been playing a large role in literature and in practice (Peng & Pleggenkuhle-Miles, 2009). CSR originates from the stakeholder approach (Freeman, 2010) and states that a firm has a responsibility to all stakeholders involved. Dahlsrud (2006) identifies 5 dimensions of CSR: The environmental, social, economic, stakeholder and the voluntariness dimensions. According to many CSR definitions a firm does right by returning part of its resources or profits to its stakeholders (Dahlsrud, 2006). An argument for this statement is that a firm relies on the resources present in its environment and therefore should compensate for the resources used, even if the use of these resources does not formally require compensation. The definition of ‘resources’ Freeman (2010) uses is not limited to natural resources or semi-finished products. It accounts for all the burdens a firm might impose on both the natural environment as well as the social and economic environment.

(19)

19

much CSR they implement in their operations. Kolstad (2007) states that an argument executives often use to justify the implementation of CSR, is that CSR is profitable. He then argues that research has failed to back up this claim. Also, when CSR would be used as a mean to increase profits, it would be simple shareholder value focus in disguise. According to Kolstad (2007) the proper way of approaching CSR is asking yourself what a firm is responsible for and act on these responsibilities, no matter its effect on profitability. Ethical and social consequences of the actions of the firm are then determinants of how much CSR needs to be implemented. Therefore the focus of CSR lies on an ethically opposite side of the shareholder value focus, and as more CSR is implemented less focus will be on shareholder value focus (Kolstad, 2007).

In a growing number of firms codes of conduct are present with the aim of upholding the labour standards for both the firm itself as well as for its suppliers. Yu (2008) looked into the effect of these codes of conduct on actual practices in firms and concludes that the presence of the codes does influence most violations of human rights like child labour and physical punishment. However, the effect on the height of wage is not apparent. Because these codes do not have an apparent effect on wage, it would be easy to state that CSR in its totality has no effect on wage. CSR is however a much broader subject and goes above and beyond these codes. This research aims to expand on the research done by Yu (2008) by looking into the effect of overall shareholder value focus on the wages in the entire firm, where Yu only looked into outsourcing.

Matten and Moon (2008) make a distinction between explicit and implicit CSR. Explicit CSR is the CSR activities that are undertaken by a firm explicitly as a part of the social responsibility the firm actively takes. Implicit CSR on the other hand is “embedded in

broader norms and regulation” (Matten & Moon, 2008). The codes of conduct on labour

practices are part of implicit CSR. It is laid down in law and society that children should not work and physical punishment is not a viable option to maintain order in the workplace. CSR can go further in the area of labour when a firm is also exercising explicit CSR. This can be done by for example providing for healthy food in the canteen or by providing other extras for employees. Therefore CSR goes above and beyond these codes of conduct.

(20)

20

statement into their CSR statement and should refrain from all tax avoidance. Not conforming to the norms that are laid down in law and society could be an indication for the fact that these firms do not care about returning some part of their profits to society (Lanis & Richardson, 2012). This stance would be implemented when a firm holds the aggregate view. The firms that do contribute to society above and beyond what is legally required of them, fall in the real world or enlightened shareholder value categories. In these views firms can implement CSR activities, are less focused on creating value for shareholders and more on being a sustainable part of society. In both of these views it is obligatory to pay a fair share in taxes.

CSR reporting is found to be related to tax avoidance (Lanis & Richardson, 2012) and labour standards (Yu, 2008) in MNE's. The focus of a firm on the shareholder value creation could be determined on the basis of CSR reporting scores. A firm that spends a lot of effort or finances on social issues will care less about the value for shareholders. However, since reporting is not actually measuring how much action a firm actually undertakes, a certain amount of “window dressing” has to be taken into account. Next to this the effective tax rate and the average wage are not part of CSR policies of firms (Yu, 2008) and are not part of research definitions of CSR (Dahlsrud, 2006) which prevents auto correlation in the statistical analysis of chapter 4.

Figure 2.2: The spectrum of firm orientations

CSR Enlightened shareholder value Shareholder value focus Aggregate view

(21)

21

all, firms will have characteristics that can be appointed to both theories. A distinction can however still be made because the level of CSR varies greatly from firm to firm.

The real world and artificial entity should according to Campbell (2007) be more likely to be found in CMEs. Campbell theorizes that the institutions present in a CME are motivation to have more CSR activity. This is partly a result of the long-term focus of CMEs compared to the short-term focus present in LMEs.

Wenzel (2005) in addition theorizes that there is a social “tax ethic” in individuals and finds significant correlation to support his theory. Wenzel defines “tax ethic” as:

“One’s belief that there is a moral imperative that one should be honest in one’s tax

dealings”

Wenzel is targeting individual people and individual income tax in his research. Although Wenzel researches the relationship on an individual level, this research is done at the level of the firm.The question remains whether these principles still stand in a corporate environment. According to upper echelon theory the organisational outcomes can be partly predicted by managerial background characteristics (Hambrick & Mason, 1984). If this theory holds for tax dealings of firms, top executives may have a tax ethic which represents itself in their business dealings. If the real world view or enlightened shareholder value of the firm is present, and therefore a firm is considered to have the same responsibilities as an individual, one would expect the same tax ethic to apply here.

(22)

22 The effect of paying low end wages to employees has more consequences. When wages are paid at a sub-LW level, in many countries like the UK, government funded facilities are in place to provide these employees and their families with the basic needs they require in order to live a normal life (Werner & Lim, 2016). These facilities will supplement wages which are perceived to be too low to a level at which the individual employee can sustain him or herself and those who depend on him or her. These government funded facilities need to be financed by the governmental revenues which for the greatest part are generated from taxes. According to Muilenburg & Singh (2007) the taxpayer is paying the cost for firms who keep their wages low. Through this mechanism firms which pay low wages and low effective taxes are effectively shifting the burden of paying the LW to other tax payers (Muilenburg & Singh, 2007). Furthermore, Arnold & Bowie (2003) state that firms have a responsibility to pay tax so that the government can maintain the goods and services its people need in order to accomplish a proper quality so that people can have a fulfilling life.

For the period this research focusses on, the UK was found to have one of the highest rates of low paid employees of the developed economies (Whittaker & Hurrel, 2013). Werner & Lim (2016) connect the promotion of the LW to the Reaganomics and the Thatcherism policies in the 1980’s. These policies are argued to be based on the theory of Milton Friedman, who was also an advisor to Reagan. Under the Reaganomics and Thatcherism the growth of the private sector was emphasized and the government was supposed to be less influential in the economic field and also leading to a lower rate of unionization (Werner & Lim, 2016). As the rate of unionization declines, the low and mid-range wages tend to stagnate compared to the higher range wages (Mishel, 2015). The low rate of unionization and a passive government are typical for an LME and are still part of the overall institutional environment within the UK in the same way it is part of the US environment. The previously named CSR regulation implied in the UK makes it however easier for firms to make a choice towards a more enlightened shareholder value manner of conducting business.

(23)

23

are expected not to shift the economic burden of paying low wages and low taxes to other taxpayers. Therefore the next hypothesis is:

Hypothesis 3: Firms which have a relatively low focus on creating shareholder value will have relatively low tax avoidance.

Hypothesis 4: Firms which have a relatively low focus on creating shareholder value will have relatively high wages.

2.5 – Conclusions

In the aforementioned views of Milton Friedman and CSR two apparent conclusions can be drawn. When a firm chooses to perceive social responsibility purely as a cost to the firm and follow the view of Friedman (2009) and Jensen & Meckling (1976) as in the aggregate or nexus of contracts view, profits will be maximized and costs cut as much as possible. Firms on this side of the spectrum will not take on any corporate social responsibility since only profit matters for these firms. Therefore tax will be evaded as much as is possible, but also wages will be kept low.

On the other side are firms that tend to be more socially responsible and have higher CSR scores. These firms are more prone to have a higher effective tax rate and care more about other stakeholders and not just about shareholders (Lanis & Richardson, 2015). In these firms it will be considered important to pay a fair share in taxes for the better of society. As stated before, Dahlsrud (2006) distinguished five dimensions of CSR of which the stakeholder dimension was one. Employees are a major part of these stakeholders and are therefore theorized to be treated better by firms with higher CSR scores.

(24)

24

in both subsidiaries and in firms which MNE’s outsource to (Yu, 2008). They conclude that MNE’s are responsible for providing employees with a living wage. The living wage goes above and beyond the legal minimum wage and is a level of income that can properly sustain the worker (Arnold & Bowie, 2003).

(25)

25

3 – Methodology

This section will discuss the research methodology that will be used in this study to provide an answer to the research question.

3.1 – Research design

This paper aims at answering the question: What is the relationship between shareholder

value focus, the effective tax rate and wages between MNE’s in the United Kingdom? In order to answer this research question, quantitative research will be conducted. Quantitative research is based on the assumption that the world is viewed objectively by an independent objective researcher and that the social world exists externally (Blumberg, Cooper & Schindler, 2014). This research therefore adheres to a positivist stance which forms the basis for quantitative methods and analysis (Cupchik, 2001).

The data will be collected in a systematic and objective way in order to ensure the reliability of the data. Therefore this is a formal study (Blumberg, Cooper & Schindler, 2014). As the prerequisites for an empirical analysis are present, a statistical analysis will be conducted to answer the research question.

3.2 – Population

3.2.1 – United Kingdom

For this research, the population consists of large and very large publicly listed firms from the United Kingdom. The United Kingdom has a unique position in the world because of its proximity to the EU mainland while still being classified as an LME by Hall & Soskice (2001). Tax evasion is not widely facilitated within the United Kingdom itself, the actual evasion takes place in countries like Bermuda, the Cayman Islands or the Isle of Man.

(26)

26

more of a focus on long-term capital and profit. This long term focus lets firms be more thoughtful about the consequences of their actions in the distant future (Hall & Soskice, 2001). Avi-Yonah (2008) considers this to be a reason for CSR being a much more accepted business practice in the EU mainland than in countries with liberal institutional environments. The UK has some CSR regulation in place which is contrary to what would be expected of a typical LME. The UK therefore has characteristics of both these categories and therefore makes for an interesting case to examine.

The US, in contrast, has much less regulation in this area and therefore less firms report on their CSR activities or related subjects. Firms in the US are more prone to act in a way which is expected from a firm in an LME. Their annual reports state everything they legally have to, but not more. The UK has been leaning more towards the EU mainland and the CMEs vested there. This can also be seen in the annual reports which report on more aspects of the firms than US based firms do. For example, as some preliminary research was done in order to find the proper country to conduct this research for, it was found that US firms do not report on gender ratio in their business where many UK firms do. For this research a country was sought that would have an institutional environment in which firms could most freely choose how to conduct business. This freedom was found to be present in the UK. Since in the UK stimulation is present to perform better in the area of CSR, the UK remains an LME which stimulates the shareholder value focus and US firms do not (or barely) report the required data, the choice has been made to study UK firms.

3.2.2 – Outsourcing and offshoring

Many modern day MNEs outsource many of their activities or incorporate foreign subsidiaries. Nike is a striking example of a firm that lets all the actual production be done by third parties (Leavy, 2005), but Nike is just an example of a much larger group of MNEs which are using this method to distribute activities over a larger set of legal entities. As a firm enters a foreign market, it can do business in two ways. By buying or by making the products it requires. When a firm chooses to produce its own products, it creates subsidiaries. When buying the products it will simply outsource and buy them from an independent firm in the foreign market.

(27)

27

subsidiaries into consideration by using data on the average wage and ETR for the global company. The aggregate financial statements found in annual reports and in Orbis already takes all these subsidiaries into account.

Next to subsidiaries an MNE may also choose to outsource part of its production process. These business partners are many, and to find out what firms are partners to the firms in the sample would be an immense task. Next to this it would be hard to find out and incorporate all these different effective tax rates and average wages so that these will accurately and correctly be represented in the final formula. In order to obtain good measurement parameters on this subject would require wholly different study. Next to these practical arguments, management of the firms in the sample have no direct authority over the wages and tax payments of the firm it outsources to. As was stated before, the firms to which the outsourcing is done are not a part of the structure of legal entities of the rest of the firm. An MNE can be held responsible for worker’s wage in firms which it outsources to on ethical grounds (Crane, 2013), (Kramer & Porter, 2011), but often can exert little actual influence on wage and tax expenses (Yu, 2008). For these reasons the choice has been made to not incorporate outsourcing in this research.

The CSR reporting score will be compared to the average wage and the effective tax rate between all firms in order to find out whether or not there is a relationship between the two statistics. CSR, wage and ETR will be averaged over three years for the entire firm including subsidiaries abroad. Depending on the availability of data, research will be done over 3 years in order to increase external validity.

3.3 – Sample criteria and data collection

(28)

28

losses from previous years left to offset the profit in 2013, 2014 or 2015, positive taxes had to be paid in 2012. Profits had to be made in all of the three years for which the research was done in order to exclude firms in financial distress. Positive taxes had to be paid for the same period as well to ensure that tax treatment for all firms would be equal. Because of their unique tax treatment, insurance and financial firms were excluded from the sample by selecting all other industries to be included. Furthermore, only firms that reported on the gender ratio of all employees working for the firm in the years 2013 to 2015 were selected. These data were taken from annual reports, CSR reports and official websites of the specific firms. All sources have been listed in appendix A. These criteria yielded a list of 48 firms for the sample.

3.4 – Variables

In this section, the variables as represented in the conceptual model in figure 2.1 will be elaborated on, and control variables that can potentially influence the relationship will be discussed.

3.4.1 – Independent variable

The independent variable in this analysis will be the focus on creating shareholder value which will be measured by using the individual average CSR score of the firms in the sample as a predictor. CSR is a concept which is defined in different ways in the literature, however no unified definition has been agreed upon (Dahlsrud, 2006). The lack of a unified definition has also led to a multitude of different ways to measure CSR. The difficulty in measuring CSR is mainly because of its many components which cannot be easily quantified and compared. To arrive at a CSR score which encompasses the many different aspects of how a firm deals with other stakeholders and its environment, a great many factors have to be taken into consideration. Examples are donations to charities, waste reduction, upholding labour standards or countering climate change.

(29)

29 different categories of CSR that CSRHub discerns. The four areas are community, employees, environment and governance. It is worth looking at the employees category since auto-correlation could be present if the wage of the employees is taken into account in this CSR score. However, the employees category is solely concerned with non-financial data on health, safety and other working conditions (CSRHub, 2012a). Although a correlation between the ETR and CSR scores is to be expected (Lanis & Richardson, 2012), no auto correlation is present since this CSR score is not based on any tax dealings.

The scores of the four categories are merged into one overall CSR score per firm by CSRHub, which are normalized and given a score between 0 and 100. 100 being a good CSR score and at 0 not doing any CSR at all (CSRHub, 2012b). These are the scores which will be used in the moderation analysis.

CSRHub uses a lot of different data sources and takes reliability and bias into account and therefore makes for a reliable source of CSR scores. However, it still remains a non-academic dataset and should be handled with proper caution. The reason that this data source has been selected, is that in this source most firms for which financial data was present were also given a CSR score that was based on the same set of sources, handled the same criteria and were subject to the same arithmetic so that the scores can be properly compared between firms. 3.4.2 – Dependent variables 1: wage

The average wage is the next statistic which needs to be obtained. In financial statements as well as in databases like Orbis the number of employees and the costs of wages during a year is public knowledge. Also in financial statements the pay and bonuses top management receive are named which makes it easy to gain insight into these numbers. As was stated in the theory, the wage expense account of the firm is taken as the dependent variable. The firm only cares about its own expense account while the monetary benefit the employee receives is of less importance to the firm.

3.4.3 – Dependent variables 2: tax avoidance

(30)

30

accessible through the Orbis database. Since access to tax returns is very limited, these will not be used.

In order to arrive at the effective tax rate the total income tax expense of a firm is divided by the profit before taxes and multiplied by one hundred. The resulting number is the percentage of the profit that has been really paid to governments or will be paid in the near future.

𝐸𝑇𝑅 = 𝑇𝑎𝑥 𝑒𝑥𝑝𝑒𝑛𝑠𝑒

𝑁𝑒𝑡 𝑖𝑛𝑐𝑜𝑚𝑒 𝑥100%

By adopting this method, deferred taxes are not taken into account. Deferred taxes state the amount of taxes expected to be paid in the future, but which has not become payable due to the difference in fiscal and economic profit. Deferred taxes are not taken into consideration because it is hard to give a proper estimation of its value, since these taxes will be paid at a later date. Some deferred taxes may take a very long time to actually get payable. In extreme cases these amounts may be deferred indefinitely. Firms do not report on the nature and sources of the deferred taxes existing within the firm. The situation of these deferred taxes is therefore unknown and tax is temporarily avoided. The deferred taxes are therefore considered to be part of the tax avoidance, even though it may only be temporarily. Tax actually paid is therefore the only tax taken into consideration in the effective tax rate.

Active tax avoidance will lead to a substantially lower effective tax rate compared to the economic profit, than the statutory tax rate installed by the government of the country of residence. By comparing the effective tax rates between different firms a conclusion can be drawn that the company with a lower effective tax rate is better at avoiding tax (Rego, 2003), (Lanis & Richardson, 2007). The effective tax rate is a widely used statistic used in other research while it proves to be an effective measurement tool to conduct research on the tax policy of firms (Rego, 2003), (Lanis & Richardson, 2007).

(31)

31

avoidance. It would be very hard to obtain insight into the tax returns of firms since these are not public and it is not clear how taxes are built up from financial statements alone. Therefore it is impossible to determine the actual difference between the intended facilities governments have put in place, and active tax avoidance. These facilities are however not substantial enough to have an impact as large as an actual tax avoidance scheme and any significant differences will therefore be assigned to active tax avoidance (Avi-Yonah, 2000).

The purpose of this research is to investigate the relationship between shareholder value focus and how much tax is legally avoided. By doing this the firms which make use of tax evasion schemes are not taken into account. It is not known, and neither the intention of this research to uncover the amount of tax evasion done by MNEs. This is due to the illegal nature of a tax evasion scheme, which makes it hard to conclude whether a firm evades tax. Profits which are made but have not been reported to the IRS will not show up in financial statements firms publish, nor will they be available anywhere else. A detailed audit and fraud detection tools for every company in the sample or a detailed study of documents like the “Panama Papers” would be needed. This is however far beyond the scope of this research.

3.4.4 – Control variables

Controls are done for scale by total amount of assets, presence of debt by the gearing ratio and gender differences by the ratio of women to total amount of employees. Not all data for which should be controlled is present in the databases used in this research, other publicly available databases or annual reports. The firms which are present in the sample have all reported the statistic of gender ratio in their annual reports or have otherwise published these numbers. This may also lead to a bias resulting in a sample of firms which score better in CSR reporting and therefore having higher ETRs (Lanis & Richardson, 2015).

Rego (2003) and Lanis & Richardson (2007) find evidence for economies of scale and scope. They find firm size to have an effect on ETR. The political costs theory states that a larger firm will be less prone to evade tax since the firm is more visible to the outside world and will face more costs if it visibly evades tax (Rego, 2003). Another view is that under the political power theory which states:

(32)

32

Lanis & Richardson 2007

Lanis & Richardson (2007) find a negative relationship for firm size and ETR in Australian firms, while Rego (2003) finds a positive relationship for the same variables in a USA sample. Therefore there will be a control for company size in the sample based on total asset value. What direction the connection between firm size and its ETR will have is hard to determine since the relationship seems to be country specific (Lanis & Richardson, 2007). Rego (2003) also finds firms with higher profits to have lower ETRs. He theorizes that firms which are more profitable are also more likely to have more incentive and more resources available to avoid taxes. His results substantiate his theory.

Another effect is found by Lanis & Richardson (2007) on ETR which is the ratio of debt present in a firm. Tax law observes interest and dividends in a different way and taxes these different kinds of income in different ways, which leads to very different ETRs. Interest is in most tax jurisdictions deductible from the taxable income. Dividends are not. A firm with relatively high debt-to-equity is therefore expected to have a lower ETR than firms with a relatively low debt-to-equity ratio. Therefore the gearing statistic is used to determine the ratio between debt and equity.

𝐺𝑒𝑎𝑟𝑖𝑛𝑔 =𝑇𝑜𝑡𝑎𝑙 𝑙𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠 𝑇𝑜𝑡𝑎𝑙 𝑎𝑠𝑠𝑒𝑡𝑠

The above formula shows the calculation by which the gearing ratio is obtained.

Phillips, Wamhoff & Smith (2014) and Rego (2003) both find the scale of foreign operations to have an influence on ETR. Therefore only multinational firms will be taken into account in this research. The difference in statutory tax rates of the different countries will not be taken into account, since it is impossible to discern how much profit is made in which country because only very few firms sporadically publish these numbers. Only firms incorporated under UK law will be taken into account in the sample, since firms which have been incorporated in foreign legislation may be treated differently under UK tax law.

(33)

33

firms will be taken into account which have been profitable for the duration of the study and have paid positive taxes in the year 2012.

Furthermore the effect of board composition on tax aggressiveness has been researched by Lanis & Richardson (2011). They find a negative relationship between the proportion of outside directors present in boards and tax aggressiveness.

Slemrod (2004) theorizes that after-tax incentives for management would increase tendency to evade tax. There have been several studies into the subject of CEO bonuses related to tax avoidance practices. This study found no relationship between bonuses based on after tax profit and tax avoidance when these bonuses were paid to top-management, but did find an effect when bonuses were offered to unit-level managers (Phillips, 2003). Jensen & Meckling (1976) and Desai & Dharmapala (2006) have argued that paying top management in shares their interests and those of principals can be aligned. The ratio of executive owned shares should be a control variable. For both the board composition and the after-tax incentives the author recognizes that these variables should be accounted for by controlling for them in the formula. However due to the lack of data on these subjects the ability to account for them is not present. This may distort the outcome of this research and readers should interpret results with caution.

There are other factors to take into account when measuring wage. It has been shown in numerous previous researches that women receive less pay compared to men for the same kind of work (Petersen & Morgan, 1995). Therefore a control will be done by taking the ratio of women to the total workforce.

Oi (1999) theorized and has found a relationship between firm size and average wage. Larger firms are theorized to have more need for employees with more scarce skills, experience and education. Being a larger part of the workforce, these employees lift the average wage up compared to smaller firms.

(34)

34 data on these statistics, however some data is available in annual reports and financial data can be collected from the Orbis database. Age, experience and education levels are controls that cannot be taken into account although the author recognizes that they influence the relationship between CSR and average wage.

3.5 – Analysis

Two linear regression analyses will be performed with the CSR reporting scores as the independent variable. The dependent variable for the first test will be ETR and for the second test average wage. Controls for the test for the relation between CSR reporting and ETR will be firm size and gearing of the firms in the sample. The second test, which examines the relationship between CSR reporting and average wage, controls for firm size and gender ratio.

If a relationship would be found, this would be an inclination towards establishing the theory that firms with a shareholder focus will cut costs on both ETR and wage.

The model for the first linear regression analysis is:

ETRna = α0 + β1 Gearingna + β2 Assetsna + β3 CSR Scorena + ε

The model for the second linear regression analysis is:

Average wagena = α0 + β1 Assetsna + β2 Gender rationa + β3 CSR Scorena + ε

In the model the ETRna depicts the average wage for firm n, and with the average a taken over

(35)

35

4 – Results

Statistical analysis was done with the SPSS software. The two linear regression analyses which were mentioned in the methodology chapter have been done. For the first test CSR Reporting was the independent, ETR the dependent and firm size, R&D and gearing the control variables.

Figure 4.1: Scatterplot of the relationship between CSR reporting and average ETR.

(36)

36

For the second test the independent variable was CSR reporting, the dependent variable was average wage and controls were firm size and gender ratio.

4.1 – Reliability analysis of the data

Before a linear regression analysis can be conducted, the data needs to conform to certain standards and requirements in order to be able to draw a sound conclusion. Descriptive statistics has been used to test the data on these assumptions. The data has been scanned for outliers which can distort the sample. All data is continuous, therefore the data has been checked for linearity, multicollinearity, homoscedacity, auto correlation and normality. The size of the sample for both tests is 48 firms for which all data were available.

4.1.1 – Outliers

One outlier was evident from the rest of the data as it had an ETR of 95%. This was due to the fact that this firm paid more than twice its net income in taxes in 2014. The author suspects this to be a mistake in the database since the annual reports of the firm mention different numbers in 2014. To avoid the influence of this bad data on the analysis, the firm was

removed from the sample before any other tests of reliability were done. The sample therefore consists of 47 firms.

(37)

37 Figure 4.3: Boxplots of the distribution of total assets before and after transformation.

The boxplot on total assets (left) shows outliers as asterisks. After normalization (right) these have disappeared.

4.1.2 – Linearity

(38)

38

Figure 4.4: Scatterplots of all variables

4.1.3 – Normality

(39)

39 Figure 4.5: Normal Q-Q plots of average wage and average ETR.

When the Kolmogorov-Smirnov test is taken into account the conclusion remains the same. The data is normally distributed. The Shapiro-Wilk test is more suitable for smaller samples and is commonly referred to as a reliable way of testing for normality of the data (Lilliefors, 1967).

4.1.4 – Multicollinearity & homoscedacity

(40)

40

4.2 – Statistical outcomes

After all assumptions were found to be met, the two linear regression analyses were performed on the data.

4.2.1 – The relation between CSR reporting and ETR

The first test yielded the results seen in figure 4.4.

As can be seen in the figure above, two models are present. The first containing the control variables, the second containing the CSR score. The results show there is an insignificant effect between CSR reporting and ETR, B=-,003 P=.361. The negative relationship is the opposite of the result which was expected. The arguments in the theory stated that a positive relationship was to be expected between CSR reporting and ETR. All these statistics are however non-significant and therefor seem to have no explanatory effect on the DV. The R2 of the control variables is 0,012. However, adding CSR in the equation only adds .031 to the explained variance of the dependent variable (P=.361). The control variables show no

indication that there is any relationship between firm size, gearing and the ETR of the firms in the sample.

Figure 4.6: Table of Results of the linear regression analysis for CSR reporting and ETR

Model 1 Model 2

Step and variables B SE B SE

(41)

41 4.2.2 – The relation between CSR reporting and ETR

The second test yielded the results seen in figure 4.5.

Again two models are present in the above figure. The controls, gender ratio and firm size are part of the first model, which can explain for 14,2% of variance in average wage. Adding CSR reporting to the model in model 2 only increases this number by .003. There is no significant evidence that CSR reporting has an influence on the average wage for the firms in the sample (P=.678). There is however a significance found in the first model in which the firm size is found to have a significant positive effect on wages. Previous research can explain the effect of firm size on the average wage which is present. Oi (1999) confirms the effect of firm size on wages and finds a similar result in his research. The found relationship between gender ratio and average wage (B=-29.123) seems to nudge towards a negative effect, though the model shows no significance (P=.231).

A correlational analysis (see figure 4.5) shows a positive correlation of (.238) which is not significant (P=.052) between CSR reporting and wages. A negative correlation is found to be

Figure 4.7: Table of Results of the linear regression analysis for CSR reporting and wage

Model 1 Model 2

Step and variables B SE B SE

(42)

42

present when CSR reporting is compared to the ETR. This correlation is however again not significant (P=.143)

Figure 4.8: Table of correlations

The correlation between ETR and firm size being -0,079 and non-significant (P=0,297) indicates no relationship between the two variables. The implications of this result are unclear since Rego (2003) and Lanis & Richardson (2007) find opposite results of each other

implying that this relationship is country specific. Apparently in the UK, in contrary to the USA and Australia, firm size and tax avoidance have little to do with each other.

(43)

43

4.3 – Hypotheses

The four hypotheses stated in the theoretical background, have not been empirically proven. The first hypothesis stated that a shareholder value focus would negatively impact ETR. The second hypothesis stated that a firm with relatively high shareholder value focus would also have relatively low wages. Although the correlation matrix implied the predicted relationship, the regression analysis has shown no support for both the tests that were conducted.

(44)

44

5 – Discussion and directions for further research

5.1 – Limitations

This research set out to extend the current literature in two ways. Therefore it has argued that an ethic and focus on shareholder value is a determinant of ETR and wage. It has however been a lot harder to isolate other effects from the equation. Some effects that have been found to have an influence on either taxes or wage by other research has been difficult to reproduce. The data needed for an accurate representation of the relationship is often private or requires an extensive amount of labour to find. This has resulted in a small sample of firms which does not show empirical evidence for the relationship sought after.

The sample size would be 303 firms based on data availability in Orbis. However, many firms did not report on gender ratios which brought the sample size back to 50 firms, which was thereafter reduced to 48 firms because of CSR data availability issues.

The fact that some sample selection criteria are in place could have led to a bias. Firms which do not take pride in how much they pay their employees may be less prone to publicize these numbers or make them harder to find and therefore the sample may tend to hold more firms which pay higher wages. Furthermore the range of the research stretching back to 2012 has not granted the opportunity to young firms to be part of the sample. A similar situation occurs in the case of CSR reporting scores, which were not available for some firms due to a lack of data.

The effects of shareholder value-focus on average wage and the effective tax rate of MNEs based in the United Kingdom