Internationale Politikanalyse International Policy Analysis Thomas Rixen, Susanne Uhl Europeanising Company Taxation – Regaining National Tax Policy Autonomy Expertise for the Friedrich-Ebert-Stiftung „ Company tax competition restricts the political capacity of EU member states to design their national tax systems in a fair and efficient manner. Tax competition not only endangers tax revenues, but also leads to a shift of the tax burden from capital to labour and consumption. In addition, it undermines the progressiveness of personal income taxation. „ Clinging on to formal national tax sovereignty will systematically undermine legitimate democratic autonomy over the design of national tax systems. Regaining de facto control over their tax policies will only be possible if member states come to agree on a(partial) Europeanisation of tax policy. „ Therefore, we propose the introduction of a mandatory common consolidated corporate tax base with minimum tax rates. The tax base is to be apportioned to member states according to a dual allocation mechanism. The major part of the tax base should be allocated according to(microeconomic) formula apportionment. Second, there should be a macroeconomic component to compensate low-tax countries for agreeing to the minimum rate. JULY 2007

International Policy Analysis 1 Content I. The Problem – An Overview...................................................... 2 II. The Development of the Single Market and Tax Competition.......................... 3 1. Provisions to Avoid Double Taxation in the Single Market.................................. 3 2. European Court of Justice Judgements................................................ 4 3. The Change in the Method of Integration: Fair and Unfair Tax Competition.................... 4 4. Small Countries – Large Countries: Winners and Losers in Tax Competition.................... 5 III. The Structure of Tax Competition and Its Consequences............................... 6 1. How Does Tax Competition Really Work?.............................................. 6 2. The Effects of Tax Competition on Member States....................................... 7 IV. Corporate Taxation in Europe: A Proposal.......................................... 10 1. Opportunities and Limits of a Common Consolidated Corporation Tax Base.................... 10 2. The Current European Debate on a CCCTB............................................ 12 3. Europeanising Business Taxation: Criteria and Key-points.................................. 13 V. Back to the Political Debate...................................................... 17 Literature........................................................................ 18

2 Rixen/ Uhl Europeanising Company Taxation – Regaining National Tax Policy Autonomy I. The Problem – An Overview The ongoing debate on the draft European Constitution puts questions such as‘What sort of Europe do we actually want?’ back onto the agenda. Other questions include‘How can a common Europe improve the tangible quality of life?’,‘What should a socially just Europe look like?’,‘What common policies and democratic procedures and institutions are required?’, and‘What scope for the independent organisation of their laws should member states retain?’. Answering these questions is also relevant in taxation because the ability to make effective political decisions on tax policies is a key element of a democratic and socially just polity. This autonomy has come under massive pressure in recent years due to the provisions taken to create a Single Market. One result of the freedom to establish subsidiaries and transfer capital is that, at present, neither the individual states nor the European institutions have real sovereignty over taxation; although member states reiterate their deep conviction that they still possess it. In this report, we will explain why national tax sovereignty has become a hollow shell and propose a solution for this troublesome situation. A partial European harmonisation of business taxation would enable member states to regain some political autonomy over the design of their taxation policies. We will proceed in three steps. In part II, we will show how the development of the Single Market has restricted member states’ autonomy over their taxation systems. And we try to provide answers to the question of why member states have not yet reacted to the consequences of tax competition in the Single Market with a common policy to regulate this competition. In part III, we explain what the ramifications of tax competition are; how it works, and how it influences the two main functions of modern tax jurisdictions – decisions on tax rates and the structure of state revenues to finance public expenditure as well as redistribution. Finally, part IV is concerned with the political options in this situation. We take up the current debate on the introduction of a‘Common Consolidated Corporation Tax Base’(CCCTB) and discuss how the present recommendations could be modified and extended in order to accommodate economic, democratic and social demands. Susanne Uhl: Jacobs University Bremen Thomas Rixen: Jacobs University Bremen With thanks to Andreas Kammer, Christian Kellermann and Jana Zitzler for their comments and suggestions. Any remaining mistakes are of course our own.

International Policy Analysis 3 II. The Development of the Single Market and Tax Competition European economic integration began in the 1960s. After two world wars, the European Economic Community was intended to secure the fragile peace. Goods and services, capital, enterprises and workers of the member states should be able to cross frontiers as freely as possible. At that time, tax policy makers focused on the goal of supporting the free mobility of goods because‘the free movement of goods was already much more advanced than the freedom to set up subsidiaries and the free movement of capital’ as the leader of the Taxation and Customs Union Directorate-General at that time, Pieter VerLoren van Themaat(Themaat 1966), summed up the majority opinion. The most important concern of European member states was the standardisation of product based taxes, i. e. value added tax and other consumer taxes. Today, these taxes are harmonised at the European level to a large extent(Uhl 2007). But even then the freedom to move enterprises and capital across frontiers should at least not be hampered by double taxation. European harmonisation provisions strove toward this goal. Judgements of the European Court of Justice since the end of the 1990s have clearly dynamised this objective. Both developments contributed to the impossibility of curbing tax avoidance strategies of multinational enterprises through individual states’ national tax laws. But member states reacted to this with tax competition instead of cooperation. In particular, smaller countries saw advantages for their economies. We want to expand briefly on these developments. 1. Provisions to Avoid Double Taxation in the Single Market By the 1960s, the European Commission was already pressing relatively successfully for member states to conclude bilateral Double Taxation Agreements as foreseen in the Treaty of Rome. 1 The agreements were complemented by European legal guidelines to regulate cross-border tax issues, so that they no longer 1 In order to avoid double taxation, member states sign bilateral treaties under which they reciprocally limit their overlapping tax claims and thus share the tax revenues. The avoidance of double taxation can be achieved either if one of the treaty partners abstains from raising a tax(Exemption Method), or if the tax raised by one is credited in calculating the tax burden in the other state(Credit Method)(Rixen 2007a, Chap. 3; Vogel 1990: Rn 2–3 and Rn 45–47). constricted the freedom of enterprises to move capital or establish subsidiaries in the Single Market. Almost all the European directives in the area of corporate taxation pursued exactly this purpose. The so called parent-subsidiary Directive of 1990 2 was intended to simplify multinational mergers. The Arbitration Convention created a procedure that had to be followed if there was no mutual agreement as to which a member state was entitled to tax under Double Taxation Agreements and European Directives. 3 Since 2003, a common tax regulation for the payment of interest and licence fees between associated companies of different member states has further guaranteed that such income is only taxed once, i. e. where the parent company has its registered office. 4 These tax provisions were part of European efforts to allow free movement of capital within the Single Market. They were constituted in the Treaty of Rome 1957, but only became effective in practice after controls on capital were removed in the 1990s. 5 The provisions to remove limitations on enterprises’ freedom to establish subsidiaries gained practical importance in the 1990s, as international mergers increased. 6 In consequence, multinational enterprises had more opportunities for cross-border transfers of ‘paper profits’ for tax avoidance purposes(a topic we discuss in more detail in Part III.1). The decisions of the European Court of Justice play an important role in this development. Since the late 1990s, it has ever more frequently been asked by national courts to adjudicate on legal proceedings relating to enterprises operating multinationally and the interpretation of European Treaty Regulations. 2 Directive 90/435/EEC. 3 Convention 90/436/EEC. 4 Directive 2003/49/EC. 5 The complete liberalisation of capital movements in the EU was agreed in 1988(Directive 88/361/EEC) with effect from 1990 in most member states. Special transitional periods were agreed for some states. Until the mid-1990s, movement of capital was in practice restricted by the fact that many financial transactions with players in other member states required permission from national authorities under foreign exchange controls. Liberalisation of capital movements followed in the wake of the creation of the Economic and Currency Union and was finally fixed in the Treaty of Maastricht, which came into force in November 1993. 6 According to the European Commission(European Commission 2001a:21), the number of multinational companies in 15 developed countries(EU and non-EU) rose from approx. 7,000 parent companies at the end of the 1960s to about 40,000 by the end of the 1990s.

4 Rixen/ Uhl Europeanising Company Taxation – Regaining National Tax Policy Autonomy 2. European Court of Justice Judgements The European Court of Justice(ECJ) has reached about 70 judgements relating to tax matters and about 60 of these in the last ten years. At the start of 2006, more than 50 legal cases relating to direct taxation were pending, of which most would have budgetary consequences in member states(Wathelet 2006). The Court has frequently emphasised in its judgements that‘although direct taxation does not as such fall within the purview of the Community, the powers retained by the member states must nevertheless be exercised consistently with Community law’. 7 A result of the Court’s judgement is that national provisions against legal tax avoidance by companies have become increasingly impossible. Two recent cases illustrate this. The first case relates to corporate tax avoidance by transferring profits(Cadbury Schweppes 8 ), the second case to the transfer of losses(Marks& Spencer 9 ) within the Community. Both precedents affected not only the case which the ECJ was concerned with but similar rules can be found in many member states and will have to be altered accordingly. In Cadbury Schweppes the judgement means that setting up a subsidiary with the explicit aim of avoiding tax and enjoying advantageous statutory provisions is not an abuse of the freedom of establishment. Companies will merely have to show that the establishment is not a mere mailbox. Proof of(some) personnel or office inventory will do. They should then have no difficulties with shifting profits in the future. The Judgement of the Court in the Marks& Spencer case will have as a consequence that‘loss transfer’ within the Community will not be restricted but could increase under the new ruling 10 , if no contrary Community regulation can be passed(Uhl 2007, Chapter 1.4). 7 This is the standard formulation of the ECJ in its Judgement of 14 February 1995, Case C-279/93 Schmacker; most recently: Judgement of 7 September, Case C319/02, Manninen margin number 9. It refers to the right to free movement under Article 18, freedom of movement for workers under Article 39, freedom of establishment under Article 43, freedom to provide services under Article 49, and free movement of capital under Article 56 of the EU Treaty. 8 Judgement of the ECJ 12 September 2006, Case C-196/04,Cadbury Schweppes. 9 Judgement of the ECJ 13 December 2005, Case C-446/03, Marks& Spencer. 10 The core of the judgement in brief: In so far as a parent company in a country can prove to the local tax authorities that they have exhausted all possibilities of offsetting the losses of a subsidiary in its own country and that no other possibility exists to take the loss into account there, it would infringe the freedom of establishment if they were not allowed to offset the losses of a subsidiary in other countries against their own taxable profits. All in all, on the basis of the Court’s judgement, the member states must successively abolish their national anti-avoidance legislation, such as Controlled Foreign Corporation(CFC) rules. Consequently, this also means that the tax systems of member states can no longer be viewed in isolation from one another and that member states’ tax autonomy is limited in practice. If member states cannot agree on harmonisation, multinational companies will bring actions for discrimination-free taxation to the Court. To summarise: increased mobility of capital, changes in company structure – supported by tax provisions at a European level – as well as the precedents of the European Court of Justice have reduced the effectiveness of national regulations since the 1990s, with the effect that member states can – de facto – no longer maintain purely national regulations today. They must decide either to agree on common regulations or to relinquish the power to design their tax systems and sacrifice it to the imperatives of the Single Market. But two aspects have made agreement on common regulations very difficult so far: a change in the method of integration and the idea of fostering economic development by means of tax policy that is pursued especially by the small and new member states. 3. The Change in the Method of Integration: Fair and Unfair Tax Competition That the member states have as yet hardly been able to agree on any measures of harmonisation is a result of the change in the method of European integration from the dominance of institutional convergence to mutual recognition(competition). Whereas the Commission’s recommendations on the harmonisation of corporate taxation were originally based on the concept of mandatory tax harmonisation, Brussels changed its policy on corporate taxation in 1990 (European Commission 1990). In this communication, it emphasised the subsidiarity principle for the area of direct taxes. Accordingly, the national systems of direct corporate taxes should remain as they were and regulations at a Community level should only be restricted to those provisions that are essential to the accomplishment of the Single Market. After that the Commission began to appreciate the positive aspects of tax competition. It argued that tax competition could, ex post and in a decentralised manner, bring about an equalisation of the tax burden (European Commission 1990, 32). It could strengthen the budgetary discipline of member states and thus lead to a healthy reduction of the overall tax burden

International Policy Analysis 5 (European Commission 1990, 32). Today, there is hardly a European document that does not mention the positive side of tax competition and identifies subsidiarity – in taxation matters i. e. the responsibility of member states for various income taxes – as the prerequisite – or starting point – for tax competition. The European Code of Conduct for Corporate Taxation 11 in which the member states agreed among themselves to avoid the use of so called‘unfair’ tax practices was no exception. Even if the passing of the Code of Conduct was hailed as a comparatively successful attempt to regulate corporate taxation, 12 competition as a method of integration was thereby strengthened. Regulations which counted as‘fair’ tax competition were those which treated all those liable to tax in a territory equally, however low or narrow their tax base was. So if the previously selective‘unfair’ tax advantages were generalised, they would automatically become‘fair’. As a result of the Code some countries did exactly that and implemented a low tax for all taxpayers. This demonstrates a central problem of‘fair’ tax competition between member states. Not all member states see themselves as victims of tax competition. 4. Small Countries – Large Countries: Winners and Losers in Tax Competition Some countries profit from so called‘asymmetric tax competition’. Small countries benefit from reducing tax because the resulting tax deficit on‘home’ capital can be over-compensated by the attraction of foreign capital. From the perspective of small countries, reducing the tax rate leads to the inflow of foreign 11 The Code of Conduct, which is not legally binding, and which was passed by the Council in 1997 but did not take effect until 2003, qualifies tax incentives by member countries which reduce the tax burden selectively for especially tax sensitive activities such as financial, insurance and consultative services to below the usual national rate as unfair or damaging. The member countries should not introduce any new damaging tax incentives(‘stand still’) and to evaluate and if necessary change their existing statutory provisions and practices(‘roll back’). 12 Certain tendency in this direction is observable – as the Chief Advisor of the Commission’s Taxation and Customs Union Directorate-General, Matthias Mors(Mors 2006: 20), carefully formulated. Today, member states design preferential tax regimes so that they are formally within the letter of the Code. Despite these difficulties, the Code was a step forward as the member states were ready for the first time to debate their respective tax systems and their effects on other countries. capital, especially from large countries, and leads to an income and welfare gain for them. In a situation of tax competition, the welfare of small states rises while that of large states falls. Overall, however, the welfare loss of large countries is greater than the gain experienced by small countries(Bucovetsky 1991; Wilson 1991; Dehejia/Genschel 1999). Perhaps the best known case of a successful small country in tax competition is Ireland. Low taxes in Ireland attracted considerable foreign investment and thus contributed to the rapid economic modernisation of the country and the long 1990s boom(Genschel 2002). The new East European accession countries tried to copy this success and thus attracted resentment from old member states. Germany and France were particularly critical of the East European low tax strategy. The conflict is ambivalent. On the one hand, large member states’ complaints are understandable because the low tax strategy of the small countries is openly aimed at capturing their capital and productive businesses. On the other hand, it is not clear with what right the old and large member states may challenge the freedom of the new and small member states to align their tax policies to the goal of economic development, which they were indirectly promised by their entry into the EU. Incidentally, the old members were hardly in a position to break or at least alleviate tax competition between themselves even before the extension to the East occurred(Genschel/Uhl 2006). If, however, harmonisation decisions can only be made unanimously, as in the case of taxation policy, then each of the 27 member states has a veto. Thus, it is possible for the small and new countries as well as the UK, as another prominent opponent of tax harmonisation, to obstruct every attempt to coordination. 13 We will show below what this means for the community of European states and what effects will accompany tax competition. 13 It is not surprising that the European Economic and Social Committee, in connection with the implementation of the harmonisation provisions that were stipulated in the Treaty of Nice, welcomed the fact that a group of member states were enabled to perform pioneer work regarding the Community regulations(Economic and Social Committee 2002: 9). According to Articles 43 to 45 of the Treaty, it is possible for at least 8 member states to set regulations in the name of flexible integration; of course only under various conditions that promote the aims of the Community and under the pre-condition of their ultima-ratio character.

6 Rixen/ Uhl Europeanising Company Taxation – Regaining National Tax Policy Autonomy III. The Structure of Tax Competition and Its Consequences In this part we show that tax competition between states limits their ability to organise their taxation policies. First, we explain how tax competition works – its structure and mechanisms(for an overview see Rixen 2007b). Then we portray the consequences of this competition. Even if tax competition has not yet led to a large loss of revenue, it has caused meaningful changes in the structure of tax systems which restrict the possibilities of redistribution. 1. How Does Tax Competition Really Work? Tax competition occurs when countries adapt their tax policies strategically in order to make themselves attractive to new enterprises or to keep themselves attractive for existing ones. Governmental tax strategies concern all aspects of a national tax system. International tax differentials are thus reflected in types of taxation, the basis of assessment, tax rates and the enforcement of tax laws. The fact that companies wish to minimise their tax payments reinforces the competition between governments. From the point of view of European member states, there are three connected aspects to tax competition, which we want to highlight separately: first, a company’s real choice of location, second, direct investments and third, transfer of‘paper profits’. We show that all three play a different role in tax competition. Choice of Location by Companies The first aspect concerning choice of location by companies is a well known argument in many member states. An apparently very popular public opinion is that if a state has a higher corporate tax rate than others, then for tax reasons large companies will move their production and jobs to low taxation countries. This argument is not convincing. After all, taxes are only one factor among many that influence investment decisions. There is almost no dispute in the economics literature that factors such as access to markets, infrastructure, labour costs and levels of training have a stronger influence on the choice of location. A company does not relocate solely because of tax burdens(European Commission 2001a). However, this does not apply to all industries. Surveys show that companies choosing a location for a financial services centre clearly focus their attention on tax factors (Ruding Report 1992, 114). Direct Investment Doubts as to the importance of tax policy seem to be confirmed when one investigates the relationship between the level of foreign direct investment 14 and the tax burden. Foreign direct investments are in no way concentrated on countries with low taxes. On the contrary. The available data for Germany show that despite the high taxes on corporate profits in the 1990s, it dramatically caught up in the amount of direct investment it received. The amount of foreign capital in Germany increased almost six-fold between 1990 and 2002 and that occurred before tax rates were significantly reduced in 2001(Fuest/Becker 2006, 1). Large surveys confirm this. In a meta-analysis of empirical surveys, Ruud de Mooij and Sjef Ederveen(2003) come to the conclusion that, as a general rule, raising tax burdens decreases the inflow of foreign direct investment. But the direction and strength of the correlation is strongly affected by the method of measurement used(see also de Mooij/ Ederveen 2006; Schwarz 2007). Thus, it is valid to say that there is hardly any proof of a direct relationship between the amount of direct investment and the rate of tax. Shifting Profits Nevertheless, tax policy makers do not have reason to be relieved. The problem is that companies do not have to move their production abroad to save tax. They are spared the expenses of relocating their businesses and retraining specialist workers if they appoint a good tax consultant. An important reason for the stiff competitive pressure in corporate taxation is that multinationally integrated companies can perform‘tax arbitrage’. They can avoid taxes by transferring‘paper profits’. In this way they can benefit from the good infrastructure and other locational advantages in high tax countries and the tax advantages offered in low tax countries or tax havens. The current structure of international tax law allows them many opportunities to save tax by shifting profits from high 14 International investments made by a company in one economic location to acquire a long-term participation in a company in another economic location. Long-term participation means an enduring relationship between the direct investor and its acquisition exists and the investor exercises significant influence on business policies.

International Policy Analysis 7 tax countries to low tax countries. This‘profit shifting’ happens through various techniques such as the(legal) manipulation of internal transfer pricing 15 for preliminary or intermediate products or the skillful choice of financial structures, especially debt rather than equity financing. In this way multinational companies can book the profits in favourable tax countries and their losses in high taxation countries, without changing their location of real production. Many empirical studies have investigated whether and how strongly tax differences between countries influence decisions on where companies transfer their paper profits. Despite different approaches, all the studies come to the same conclusion: the transfer of taxable profits is very sensitive to taxation. 16 Acknowledging that profit-shifting plays an important role in tax competition could offer an explanation for the apparently inconsistent findings on the effects of tax policy on direct investment and choice of location. This becomes clear if one considers that the level of direct investment in a country measures both decisions on new foreign establishments(discrete investment decisions) and the upgrading of investments (decision on the size of investments). Companies establish subsidiaries in low tax countries(discrete investment) but are not prevented from upgrading investments in high tax countries because they can transfer the profits from those to low tax countries. That would explain why for financial service centres – essential for intra-group transfer of profits and losses – the choice of location is highly tax sensitive. The existing opportunities to transfer profits reduce the overall sensitivity of direct investment to taxation. But the sensitivity of discrete investment choices, which often have the purpose of creating a multinational structure to make use of profit-shifting, increases(cf. Devereux 2006). To summarise: in Europe corporate tax competition is a reality. It is driven by the possibilities of companies to shift profits. Because it has not yet been possible to realise a common European regulation, governments feel themselves compelled to reduce their tax rates in competition with other countries. 17 15 See part IV.1 for more on this and other techniques of tax planning. See also Rixen(2007a, Chapter 3 and 6). 16 See Devereux(2006, 28-40) for an overview of the many empirical studies and the various approaches to measuring the importance of profit-shifting. 17 At the discursive level, politicians do of course not justify the need to reduce taxation only with the existing opportunities of profit-shifting, but rather with the risk of real economic consequences in the form of loss of direct investment and the emigration of production sites. 2. The Effects of Tax Competition on Member States What have the consequences of tax competition been so far? What are the facts concerning tax revenues, tax structure and the achievement of income redistribution? Direct Effects: The Level and Structure of Taxation In the past, governments reacted to international tax arbitrage by reducing their tax rates and at the same time broadening the tax base in order to offset the revenue consequences. Figure 1 shows that on the one hand the nominal tax rate has fallen noticeably since the 1980s, whereas on the other hand tax revenues as measured by gross domestic product have not fallen. 18 This development has taken place in almost every country in the past two decades. Precisely because of the broadening of the tax base, reductions of tax rates have not led to appreciable losses of revenue or a reduction of the effective corporate tax burden(e. g. Garrett 1998, 85 ff.; Stewart/Webb 2006). Empirical investigations into the sequence and timing of nominal tax rate reductions show that states react to tax reductions in other countries by reducing their own rates. States are conscious that they are in a situation of strategic interdependence and perceive competitive pressures on their nominal tax rates(see e. g. Devereux/Lockwood/Redoano 2004). The general trend to a reduction of nominal corporate tax rates in the EU has not yet weakened, but has instead continued in 2007, as Table 1 shows. Besides Spain reducing its corporation tax from 35% to 32.5% in 2007 and the Netherlands reducing it from 29.6% to 25.5%, there were reductions in Bulgaria (from 15% to 10%), Greece(from 29% to 25%) and Slovakia(from 25% to 23%). In Slovakia the tax rate will be reduced in several steps to 20% by 2010. Estonia raises no tax at all on retained profits. The Estonian tax rate on distributed profits was reduced in 2007 from 23% to 22%; by 2009 it will be reduced to 20% and then will also be assessed on retained profits. Germany will reduce its corporation tax from 25% to 15% in 2008. 19 18 We want to emphasise, however, that these are average figures and that the situation can be quite different in individual countries. 19 In Germany companies pay corporation tax as well as‘Gewerbesteuer’ and the‘Solidaritätszuschlag’(the average total tax rate of 38.6% in 2007 should fall to29.8% in 2008). In other countries, the following surcharges to corporation

8 Rixen/ Uhl Europeanising Company Taxation – Regaining National Tax Policy Autonomy Figure 1: Average Corporate Tax Rates and Tax Revenues of the 15 EU States, 1980-2006 Table 1: Corporation Tax on Retained Profits of Corporations(without any additional local taxes) in% in 2006/7 Source: adapted from Ganghof and Genschel(2007) These developments show that tax competition between member states is by no means at an end. One may also assume that the tax reduction announced in Germany will provoke strategic reactions in the form of further tax reductions among other member states. The importance of the nominal tax rate becomes clear if one considers that for multinational companies this is the decisive factor in the decision to transfer ‘paper profits’ to low tax countries. Only profits which are taxable because they cannot be offset against depreciation and other tax benefits are considered for transfer. For this taxable part, the nominal rate is also the effective tax rate. The trend toward low tax rates and broad tax bases is an attempt to defend against the outflow of profits and at the same time compensate for this by broadening their tax base(see Ganghof 2005; 2006a; Haufler/Schjelderup 2000). But the‘tax cut cum base broadening’ policy results in a shift of the tax burden within the business sector: highly profitable multinational companies benefit, while nationally active, less profitable enterprises are more heavily burdened(see Ganghof 2006b). Thus, especially small and medium sized companies are more heavily burdened and affiliated groups are relieved(see Devereux/Griffith/Klemm 2002, 483). 20 tax are raised: Belgium(3%), France(3.3%) and Portugal (10%). In Hungary and Italy a value added tax of 2.3% and 4.25% respectively is raised. Further, in Luxembourg there is a local tax on corporations of 6.8% on profits(Luxembourg City) and Spain imposes a yield tax which depends on the industry of a maximum of 7.5% of profits(cf. Status: Recht 26 January 2007, Issue 2, p. 40). 20 Incidentally, the policy of broadening the tax base while reducing the rate of tax can have a negative effect on growth. Belgium Denmark Germany Finland France Greece United Kingdom Ireland Italy Luxembourg Netherlands Austria Portugal Sweden Spain EU-15 Average Bulgaria Estonia(1) Latvia Lithuania Malta Romania Slovakia Slovenia Czech Republic Poland Hungary Cyprus NMS-12 Average EU-27 Average 2006 33 28 25 26 33.3 29 30 12,5 33 22 29.6 25 25 28 35 27.6 15 0 15 15 35 16 19 25 24 19 16 10 17.4 23.3 2007 33 28 25 26 33.3 25 30 12,5 33 22 25.5 25 25 28 32.5 26.9 10 0 15 15 35 16 19 23 24 19 16 10 16.8 22.6 Source: Status: Recht, 27 January 2007, own research. Tax competition also causes other changes in the structure of national tax systems. First there is emPrecisely those young innovative enterprises which are not yet profitable are disadvantaged in contrast to‘old’ capital that has already benefited from previously favourable depreciation opportunities and now enjoys the lower tax rate. Because the growth potential of national economies depends to a high degree on the readiness to invest in new undeveloped business areas, this is counter-productive. This is indeed an important reason why representatives of finance ministries indicate that the policy of rate reduction with simultaneous base broadening is slowly but surely reaching its limits. In current discussions on the German business tax reform, which will extend the tax base to elements that are independent of profit, similar arguments are surfacing(e. g. see Tagespiegel of 14.03.07).

International Policy Analysis 9 pirical evidence that the tax burden is shifted from mobile to immobile factors. The burden on labour rises while that on capital falls(e. g. Winner 2005; Schwarz 2007). Besides that, there is a tendency to shift from direct to indirect taxation(Bach 2001). Admittedly, the tax rate increases on value added taxes are modest in comparison with the tax rate reductions on corporate and personal income taxes(Schratzenstaller 2006). The result of all these shifts is that the distribution of the tax burden becomes less fair. Capital income, especially of highly profitable investments, is relieved while consumption and earned income is more heavily burdened. In addition, the policy of reducing tax rates and broadening the base as a reaction to tax competition has an important effect on the progressiveness of personal income taxation, as we shall now show. Indirect Effects: The Progressiveness of Personal Income Taxes Tax rate reductions on corporate taxes spread out to income taxes and can lead to a reduction of the top personal income tax rate and a flattening of the progressiveness of personal income taxation. This can – since progressive income taxes are one of the most important and effective instruments of the redistribution of wealth – endanger the redistributive capacity of the entire tax system(Uhl 2006). The mechanism through which pressure on progressive personal income taxation is caused can be briefly described as follows: for tax purposes, the only difference between individuals and corporations lies in the legal form. If corporation tax is lower than personal tax, then it is worthwhile for private individuals to‘hide’ their income by incorporating and thus re-labelling their income. In order to make such tax avoidance unattractive, tax policy makers usually attempt to align the corporate tax rate on retained profits and the top rate of personal income tax. In order to maintain the integrity of personal income tax systems, it is necessary to rely on the‘backstop function’ of corporate tax for income tax(see Mintz 1995; Ganghof 2007). If corporation tax has to be reduced because of tax competition, this has an effect on the integrity of the personal income tax system. A government can decide to reduce top personal income tax rates as well and thus keep income shifting between the corporate and non-corporate sectors to a minimum. In this case, the progressiveness of income taxes and thus the redistribution within the tax system is decreased. A government could, however, also decide to allow a gap between corporation and top personal income tax rates and thus open up exactly the kind of tax arbitrage opportunity described above. 21 In this case, tax progression, at least for personal income tax, is maintained on paper but in fact the redistributive objective is undermined because high-income taxpayers can use the tax arbitrage opportunities of incorporation. One can see this close relationship between the level of corporation taxes and top personal income tax rates in the developments of the last few years. An obvious symptom is that progressive income taxes have come under pressure in the last few years through the introduction of a‘flat tax’ in some countries, particularly in Eastern Europe. A‘flat tax’ in its purest form has only one tax rate regardless of the level of income. Such a proportional tax is indeed also set according to the ability of taxpayers to pay – low earners pay less than income millionaires in absolute terms. But only if wealthy people pay a greater percentage of their income than low earners is there also a relative redistribution of wealth through the tax system. Other countries, such as the Scandinavian countries in the 1990s, chose the option of allowing a large gap between the two rates. They introduced a dual income tax under which enterprises and capital are taxed at a lower rate than progressively taxed personal income. Of course, there exist incentives to ‘hide’ private earned income and these were exploited – which the states then attempted to prevent through complex regulations(see Sørensen 2005). It is possible to argue that a dual income tax system is problematic from a redistributive point of view because the redistribution only takes place between those who earn their income on the labour market, whereas large capital income is only proportionally taxed at a much lower rate. But redistribution under such a system is still more effective than under a flat tax where not even earned income is progressively taxed. 22 To summarise: the negative influence of tax competition on the progression of income tax and the shift in the structure of revenues is a very real threat to policies aiming at distributive fairness and social justice. 21 Steffen Ganghof(2006a, Chapter 1) shows that OECD Nations were successful in narrowing the spread of rates from 1975 to 1989. After 1989, as tax competition intensified, they were forced to allow the gap to widen perceptibly. It is possible that inland tax arbitrage between personal income tax and corporation tax offers an explanation as to why corporation tax revenue has not fallen. 22 For this and a number of other good reasons some authors consider a dual income tax to be a reasonable reaction to the challenge of tax competition, which socialist parties should adopt(see e. g. Ganghof 2006a). For a comparison between the dual income tax and the flat tax, see Advisory Council to the German Federal Ministry of Finance(2004).

10 Rixen/ Uhl Europeanising Company Taxation – Regaining National Tax Policy Autonomy IV. Corporate Taxation in Europe: A Proposal What might be a sensible response to the imperatives of tax competition? In the following, we argue that a solution to the tax competition problem can be successful only if there is at least partial harmonisation at the European level. We proceed in three steps. First, we show how the introduction of a‘Common Consolidated Corporation Tax Base’(CCCTB) could restrict the possibilities of shifting profits and losses in comparison to the present system. We show that tax competition is redirected to other factors and thus merely restructured if the CCCTB is not accompanied by a minimum tax rate. And finally, we want to briefly describe the current status of the discussion on a common tax base at the European level. After that, we present those criteria and key-points which, in our opinion, must be contained in a European agreement on corporate taxation. For the member states, such a European solution would let them regain the tax policy autonomy which they have lost through tax competition. Entering into a process of Europeanising business tax offers an opportunity for a pluralistic and thus controversial democratic debate over tax policy questions. 1. Opportunities and Limits of a Common Consolidated Corporation Tax Base What possibilities would the introduction of a CCCTB offer in relation to the current European taxation system? And what are the problems which a CCCTB introduced in isolation would not be able to answer? To make the advantages which a CCCTB would offer the European member states comprehensible, we must first briefly present some principles of existing international tax law. For one, tax is due in the country of residence. The duty of a natural person or a legal entity to pay tax is tied to their place of residence or registration respectively(residence principle). In addition, basically all countries in the world also tax business activity or wealth in their own territory if it belongs to non-residents(source principle). If a company maintains a place of business in two tax territories, both countries could charge tax on its activities and thus there would be double taxation. In order to avoid such distortions, the company’s profit should only be taxed once – neither twice nor not at all. This leads to the question of which method should be used to separate the profits of the various company parts? There are basically two methods: one is separate entity accounting, in which the profits of a place of business are calculated on the basis of independent bookkeeping, as is done with autonomous companies. The other is unitary taxation with formula apportionment, which takes the total profit of the multinational company and splits it between the parent and its subsidiaries in the various countries according to certain factors(see e. g. McLure/Weiner 2000). The first method is used in current international tax law, as embodied in double taxation agreements(see above) and used in the European Union. The second is used in some federal states like the USA, Canada and in German law for the apportionment of business taxes among local authorities. It is also the point of reference for a European CCCTB. Problems of Separate Entity Accounting Each part of a multinational enterprise is treated as a separate entity for tax purposes under the system of separate entity accounting and the profits assigned to each part are taxed according to the national rules in the respective location(Li 2003, Chapter 3). In order to ensure that the parts of a multinational company can be taxed in a nationally separated way, each part of the company must, for the calculation of its taxable profits, set prices for transactions of goods and services which are internal to the enterprise as a whole but which cross borders. Such so-called internal transfer prices should be the same as if the parts of the company were independently operating market participants(arm’s length principle). These internal transfer prices can be manipulated for a variety of reasons, so that profits occur in low tax countries and losses in high tax countries(Bartelsman/Beetsma 2003). 23 The structure of separate national taxation creates versatile opportunities for tax arbitrage on the part of companies. 23 Such‘manipulation’ is in no way illegal. On the contrary, there is great flexibility in setting internal transfer prices for many intermediate products which are‘traded’ within a company. For instance, knowledge and expertise are very difficult to price correctly at arm’s length. The irony is that according to current economic thinking multinationals exist precisely because the market mechanism for these transactions does not work. Because of this the appropriateness of the principle of arm’s length pricing can be put into question. In practice, the accepted range within which internal transfer prices may lie is very wide(Holzleitner 2005). Of course companies or their trusted tax consultants exploit this manoeuvring room to optimise their tax payments.

International Policy Analysis 11 In addition to this, international tax law as a general rule is based on the legal form and not on the economic substance of a subsidiary or other place of business. Little more than a mailbox in a low tax country is required in order to assign profits to the country or route them through in a tax efficient way. Empirical evidence suggests that companies make use of these opportunities to a considerable extent(Desai/Fritz/ Hines 2006). In theory countries could react to this by controlled foreign corporation(CFC) legislation. As we showed before with the examples of the UK and the Cadbury Schweppes case before the ECJ, such provisions of national tax law are not, according to the European Court of Justice, compatible with the freedom to establish subsidiaries within Europe. Precisely the principles of nationally separate taxation, which preserve countries’ formal sovereignty to design their own tax systems, lead to a restriction of their actual capacity to achieve the objectives of distributive fairness and efficiency. Precisely the fact of their formally unlimited tax sovereignty leads to a limitation of their de facto policy autonomy(Rixen 2006). Opportunities and Limits of Consolidated Profit Calculation with Formula Apportionment As already described, under a common consolidated tax base with formula apportionment one assigns the total profit of the enterprise to its various parts using factors other than profit. This means that at first, independently of the locations of parts of the enterprise, the total taxable profit of all the individual companies of a consolidated group are brought together. 24 The consolidated profit is then attributed to the individual parts of the company using a predetermined formula. They can then be taxed according to the current national tax rate. The advantage of this method is that, if the profits(and losses) of a company are not separated nationally, and an adequate apportionment system is chosen, then the most important possibilities for shifting profits disappear. Internal transfer prices do not have to be set for the calculation of taxable profits because the profit is calculated for the entire enterprise. 25 Furthermore, mailbox companies lose their 24 It is important that a standardised and consolidated assessment basis be used. Standardisation means that the profit is calculated in all countries in the same way. Consolidation means that, in addition, profits and losses are offset across borders. 25 Admittedly if a CCCTB is introduced in Europe, the question arises as to how international tax rules for non-European significance in so far as the attribution of the total profit is based on‘real’ economic values, such as total labour costs, turnover and capital invested – these are the criteria that are typically used in North America. Under these circumstances, a mailbox company would only be assigned a very small or no part of the enterprise’s profit(McIntyre 1993, 319). But at this point we must refer to new problems. With a common consolidated tax base plus formula apportionment, tax competition is no longer only about shifting profits in the books, but rather – in so far as no minimum tax rate is set – real investments. Companies and countries can compete on the factors which are part of the apportionment formula. It is conceivable that, because a CCCTB limits the possibilities of shifting paper profits, it reinforces the competition for real investment. The significance of this competition for member states depends, among others, on the weighing of mobile and immobile factors in the formula. The more mobile the factors in the formula, the more susceptible the system is to tax competition between member states. Examples of mobile factors are capital(real investments) or turnover at the place of origin. 26 Labour costs, numbers of employees or turnover at the final point of destination are more immobile factors. 27 countries should be designed. Because they do not participate in the common tax base, the traditional system of separate entity accounting will continue to be used for these nations. That would, however, mean that companies that are active in an area greater than Europe must undertake a consolidation of their profits as well as continue to be taxed on their non-European subsidiaries on the basis of separate entity accounting. Special regulations are necessary to prevent this parallelism being used for shifting‘paper’ profits to third countries(see Advisory Council to the German Federal Ministry of Finance 2007, 34 ff.). 26 Basically, the use of the factor‘turnover at the place of origin’ means that the allocation of the tax base goes to that location from which an enterprise delivers its products. Turnover at the place of origin can easily be used to shift profits because the place of consignment can be controlled by the enterprise (European Commission 2006c). Irrespective of the fact that the factors of capital and turnover according to place of origin can be seen as more or less mobile and therefore vulnerable to tax competition, these factors allow other opportunities for manipulation. For instance, capital investments can be ‘hidden’ in the capital of foundations for fiscal reasons. 27 The use of turnover at the final point of destination as a factor results in the assignment of the tax base to that country where the product is finally sold or put into use. This assignment, at least, is not usually under the company’s control, but throws up further problems, such as those which Ana Agùndez-Garcia(2006, 52 f.) elaborates under the heading ‘nowhere sales’: i. e. as soon as sales are made in a member state where the company has no presence, the throw-back rule is applied whereby such sales are assigned to the place of origin with all the problems of the mobility and entrepreneurial flexibility of this turnover.

12 Rixen/ Uhl Europeanising Company Taxation – Regaining National Tax Policy Autonomy On the other hand, focussing only on immobile factors also carries risks. An example: if the profit of the entire enterprise were allocated on the basis of labour costs paid by a part of the enterprise in a member state, then the higher the comparative rate of tax in this member state the more the company would either try to arrange that the labour costs occur in a country where lower rates are payable or attempt to decrease the costs themselves. That means that the higher the weighting of labour costs in the formula the greater the(extra) pressure on wage agreements in countries with comparatively high labour costs and comparatively high tax rates. While we demonstrated in Part III that currently differences in the tax burden of member states empirically play a minor role in the relocation of entire facilities, the situation could change if corporate profit taxation were turned into an implicit tax on labour costs. Competition over investment would then be based on wage agreements (McLure 1980; Agúndez-García 2006). This example shows that formula apportionment does not offer a complete solution to the problem of tax competition. The introduction of a CCCTB with formula apportionment does not lead to the containment of tax competition, but to a shift of competition to the factors included in the formula. This is true until there is no alignment of tax rates – or at least a minimum tax rate that is valid for all countries. If the assignment of the tax base to the respective member states followed a formula and was taxed at unified rates, then companies’ incentive to restructure their tax affairs on the basis of the factors contained in the formula would be removed. If tax rates were completely equalised, they would pay the same amount of tax everywhere. A minimum tax rate achieves this purpose, at least partially, as well. Because empirically it is absolutely plausible to assume that a limited spread between national tax rates, in so far as the rates are above the minimum rate and that pure profit shifting is barred, would not lead to dramatic relocation of enterprises or distorted incentives to relocate. The proposal below follows this consideration. We advocate the introduction of a minimum rate of tax. But first, we briefly sketch the present status of the European process in the development of CCCTB. corporate taxation(European Commission 2001c; 2001b). The Commission pointed to significant tax obstacles for companies with cross-border activities within the EU. Nationally diverging regulations for tax assessment were causing high compliance costs and the risk of double taxation was still present. One of the provisions that was supposed to improve the situation for companies was the introduction of a CCCTB for cross-border entrepreneurial activities with the aim of making the corporate taxation system of the member states more efficient, effective, transparent and simple. A Working Group, whose interim report was sent to the Commission in April 2006, was set up. 28 The timetable for the work on the CCCTB foresees the presentation of comprehensive recommendations for a common directive at the end of 2008. The interim report(European Commission 2006a; see also Czakert 2006) states that the tax base should be consolidated, so that the taxable gains and losses of individual companies of an enterprise group are combined according to unified rules. The consolidated company results should finally be assigned to the individual member states, according to an as yet not determined allocation key, and then be taxed at the respective country’s current rate of tax. Two other concepts were introduced in the working papers of the CCCTB Working Group(2006c) on the system of allocation of consolidated profits to the member countries besides the formula apportionment system we have already described: the macroeconomic method of distribution, whereby the consolidated tax base is shared on the basis of aggregated factors, such as GDP, or on the basis of added value(the Value Added Method). At present(March 2007) the Working Group has not arrived at a decision on which criteria should be used for apportionment, even if the Commission has made its preference for the formula apportionment method clear. In its interim report, the Commission advised that CCCTB should only be used for enterprises that are taxed as corporations and not unincorporated firms. In addition, it stated that the common tax base could be optional, so enterprises would have a choice of whether to use the CCCTB or traditional separate entity accounting. 2. The Current European Debate on a CCCTB After discussions on the convergence of corporate taxation more or less ground to a halt in the 1990s, the Commission returned to the issue in 2001 with the publication of its Company Tax Study(European Commission 2001a) as well as other memoranda on 28 Experts from ministries from all 27 member states and personnel of the Commission took part in this Working Group. The member states were not obligated to anything from the start. The Working Group was set up for a period of three years and around four meetings a year are planned. Documentation on their work can be found at: http://ec.europa. eu/taxation_customs/taxation/company_tax/common_tax_ base/index_en.htm(last visited 21 March 2007).

International Policy Analysis 13 All three aspects – the allocation system(thus the weighting of factors in the formula), the optionality of the system, and its applicability to incorporated businesses only – are relevant to the structure of tax competition. We have already explained the effects of weighting certain factors differently. In addition, the susceptibility of states to give way to pressure from companies considering relocation would be increased by the restricted group of participants and the optionality of the system. The optionality could have an influence on the design of the formula. On their way to a CCCTB member countries will have to agree on such a formula. For a CCCTB plus formula apportionment to be interesting for an enterprise, it must at least be more attractive than the current system of separate entity accounting. It is therefore quite likely that in the European decision process on the apportionment formula mobile factors will be preferred to immobile ones because states will worry about their ability to compete. Added to that, the more attractive the European solution appears to companies, the greater the pressure on member countries to reduce tax rates. In individual member countries, three different tax systems would be available for multinational companies in the future: CCCTB with appropriately adjusted national tax rates, which could easily be different to other corporation tax rates; secondly the old corporation tax; and thirdly, personal income tax which many member countries use for unincorporated firms. If companies have the choice between these three, the competitive pressure on tax rates for all systems would accordingly not be weakened but may be increased. As early as 2002, the Economic and Social Committee(Economic and Social Committee, 8 f.) discussed some of the problems of a CCCTB that was not neutral with respect to the legal structure of companies and allowed freedom of choice between the old and the new system. They found it unacceptable to create a variety of regulations for companies. Also the Committee repudiated the idea that it be left exclusively to member countries to set their tax rates because the transparency of actual tax rate levels which would be associated with a CCCTB would enable enterprises to exercise strong pressure on national tax authorities to reduce tax rates. Precisely this transparency without a compulsory alignment of rates of taxation, which the Economic and Social Committee criticised, is, however, an aspect that the Commission staff uses to argue in favour of the CCCTB. It argues that the creation of a CCCTB would make tax competition more transparent and fair. Member states do not have to fear interference with their tax sovereignty because, according to the Committee, they would remain free to set their own rates of tax(European Commission 2006b, 22). While a CCCTB would lead to a more transparent – and thus arguably fairer – structure of tax competition, it would not eliminate competition, but may even intensify it. It is thus very likely that member states will not be able to make use of their sovereignty to set tax rates independently. Rather, they will continue to be pressured by tax competition to lower their tax rates. In this respect, the tax sovereignty referred to by the Commission would be a mere formality. Thus, the adoption of a CCCTB without a minimum tax rate would hardly improve the situation. As we have made clear, member countries have already lost the actual capacity to design their tax systems as they wish. The next section deals with how they could regain at least some part of it. 3. Europeanising Business Taxation: Criteria and Key-points The key-points which we recommend have already been implicitly discussed and will briefly be explicated in this section. We do not deal with all the technical aspects and details of a CCCTB in this report. 29 Four Criteria for a European Solution a. Protection of the autonomy of member countries: member states’ room for manoeuvre should be at least partially recovered. We have shown above how the tax sovereignty of nations with respect to the composition of tax revenues, the generality of income taxation and redistribution has been restricted in the past. Our recommendation 29 Accordingly, there will be no recommendations concerning the following very relevant questions: Which accounting standards are appropriate to determine the CCCTB? What is the right way to deal with the depreciation of capital goods; with capital gains, reserves, debts and losses? How should corporate mergers be handled? Who is part of the group of entities whose profits and losses are consolidated? These are just some of the questions which tax experts from the European member countries must deal with within the CCCTB Working Group. Realising common standards on these issues requires detailed knowledge of the various tax laws of the member countries. For an overview of the discussion on this we direct you to the website of the CCCTB Working Group:(http://ec.europa.eu/taxation_customs/taxation/company_tax/common_tax_base/index_en.htm), as well as to the Report of the Advisory Council to the German Federal Ministry of Finance(2007, 38 ff.).

14 Rixen/ Uhl Europeanising Company Taxation – Regaining National Tax Policy Autonomy would give member countries back their political autonomy to design their tax systems. First, they would actually be able to establish their own corporate and personal tax rates in a way that is most suitable for their individual revenue requirements, their public duties and responsibilities, such as the supply of services and financing their welfare systems. Second, member countries should regain the possibility to design a progressive system of direct taxation and thus be able to engage in redistribution. b. European Solidarity: creating additional opportunities for solidarity with the aim of creating equivalent living conditions in Europe. Instead of engaging in tax competition, tax policy should contribute to the realisation of the European welfare promise. We propose an allocation system of the CCCTB that accords more weight to disadvantaged regions so that they will not continue to depend on the low tax regimes some of them have implemented for regional economic development reasons (‘catching-up’). rather than home investment. There is political agreement among the EU-27 that such tax regimes are unfair. Multinational companies particularly try to reduce their taxes by shifting profits and losses across borders. The anti-avoidance legislations of individual member states, which are supposed to protect the member countries from transactions that are obviously implemented for taxation reasons alone, are no longer possible because of the freedom to establish subsidiaries as interpreted by the ECJ. Therefore, it is necessary to find a way to restrict preferential tax regimes as well as profit shifting by companies in a way that is in line with European law. Proposal for a Mandatory Common Consolidated Corporate Tax Base with a Minimum Tax Rate These criteria could be realised by the following proposal. a. Consolidation and Allocation c. Economic Efficiency: reduction of tax obstacles and distortions in the Common Market. There is no question that the plurality of corporate tax systems creates difficulties for cross-border activities of companies within the EU. These difficulties can only be smoothed out by standardised corporate taxation that is mandatory for everyone and, if possible, includes all legal forms. While companies may welcome distortions caused by differing tax bases or tax rates because they facilitate tax avoidance, such distortions are not efficient in the sense of creating a level playing field for all companies in a common market. Accordingly, all kinds of distortions and tax obstacles should be removed. This would also facilitate the achievement of the goals of the Lisbon Process where the EU set itself the target of becoming the most competitive economic region in the world. d Fairness: curbing unfair tax competition by member states and limiting companies’ tax avoidance strategies. We have shown that in order to gain locational advantages and entice investments from other states, member countries designed preferential tax regimes – that is, those that give preference to foreign capital Our proposal takes up the idea of a CCCTB as recommended by the Commission, the European Parliament, and the European Economic and Social Committee. As to the allocation of the tax base to the individual member countries, we advocate a dual allocation system consisting of a micro and a macro component. The microeconomic allocation component, according to which the major portion of the tax base should be allocated, should consist of a system of formula apportionment. Formula apportionment allocates the tax base according to real economic activity, i. e. only those countries in which a company actually engages in economic activity are included in the allocation. This is complemented by a macroeconomic component taking the idea of compensating small and new member countries – i. e. the criterion of solidarity within Europe – into account. 30 This allocation is independent of whether a multinational company actually engages in business activity in the particular country. Every member state would be included in the macroeconomic component of the allocation system. 30 This is not the economic promotion targeted within the framework of the European Structural Funds but instead a general mechanism of solidarity between the member countries, which is also operated in many federal systems by use of financial compensation. It is questionable how much political preparedness there is in Europe for such compensation, but this does not alter the fact that it is a good idea.

International Policy Analysis 15 The microeconomic component of the allocation system should be based on an apportionment formula in which immobile and mobile factors are included: labour costs, turnover in the country of final destination and invested capital. That means every member state receives its part of the CCCTB in proportion to the labour costs, turnover and capital that the company generates in its territory. The weighting of these three factors in the apportionment formula is left open here. One possibility could be the‘Massachusetts’ formula, in which each factor has an equal value, thus one third. But it might be sensible for different industries to have differently weighted formulas. For instance, in Canada solutions that were matched to the demands of various industries were developed over a long process(see e. g. Hellerstein/ McLure 2004; Weiner 2005). In addition, the exact composition of a formula will be thoroughly discussed between governments because, obviously, the formula decides about the proportion of the tax base that will be assigned to a member state. In other words, there are many aspects to consider in the decision as to the weighting of the factors in the formula. We cannot discuss all the details here. But it is important for us that in applying the specified factors there is a close relationship between real economic activity at a location and the share of the tax base assigned to the respective country. The microeconomic factors in the formula have to portray real business activity so that we end up with a fairer allocation of the tax base than under the current principle of separate entity accounting, which reflects the legal form and not economic substance. Allocation by microeconomic factors should be complemented by a macroeconomic component according to which a smaller share of the CCCTB should be allocated among member states. We recommend an allocation according to the regional GDP per inhabitant of the EU-27 in Purchasing Power Standards (PPS). 31 Gross Domestic Product is a measure of total 31 A comparison between the gross domestic product of different countries or large regions which have differently sized economies is possible if one divides it by the number of inhabitants and then expresses this in euros. The rate of exchange does not reflect the different level of prices between countries. Even in individual countries in the euro zone, the purchasing power of the euro varies according to the respective national level of prices. In order to standardise this, one converts this regional GDP using a conversion factor, so-called purchasing power parity, into artificial purchasing power standards, which make the purchasing power of different currencies comparable. Purchasing power standards portray the price relationship for the same goods in different countries in their own currency. For the European Union, the purchasing power standards of the EU-27 are used as an artificial common reference value for the calculaeconomic activity and can thus be used to compare the economic development of regions. The largest portion of the tax base should go to the country with regions which have the lowest GDP per capita in PPS. This regional compensation provision would explicitly help to achieve the aim of regional compensation within Europe and attempt to reconcile those member countries which currently base their‘catching-up’ economic development on a low tax regime with a unified European minimum tax rate. Furthermore, this macroeconomic part of the allocation system would give no incentive to relocate for tax avoidance reasons (see also European Commission 2006c, 5). Our recommendation is not orientated on the average Gross Domestic Product per capita of an entire member country, as for example the European Cohesion Fund is, but instead takes the regions as its starting point because countries which are‘rich’ on average can have‘poor’ regions and there can be large regional spreads. 32 Consider the example of the United Kingdom. In 2004, Inner London was the region with the highest GDP per inhabitant in the EU with 303% of the average GDP in the EU-27. Cornwall, however, was clearly under the average with 79.2%(Eurostat 2007). The aim of our recommendation is to have ‘rich’ member countries accept the macroeconomic component in so far as they have‘poor’ regions. Regions in the respective country would then also benefit from the system. Allocation using a macroeconomic component interferes with the equivalence between the location of real business activities of a company and the location where tax is payable. If GDP was the only factor in the allocation system, then the company would have no influence at all on the state where its taxes were payable. That would be a very far reaching modification to generally accepted principles of taxation. The principle of equivalence as it is known in public finance requires a connection between the taxpayer and the state, whereby a direct equivalence between the state’s willingness to provide infrastructure and other public goods and their use by the company is required. This reasoning is one of the most important arguments for taxing corporations at all(see e. g. Weichenrieder 2005) and is thus not questioned by us. Instead, we put forward the macroeconomic component as an explicitly redistributive complement to microeconomic formula apportionment. 33 tion of purchasing power parity(see Thalheimer 2005; Eurostat 2007). 32 The nomenclature of territorial units for statistics(NUTS) which was agreed in Europe 2003 counts 268 regions in the 27 countries of Europe. 33 The proportion of the complete basis of assessment that should be allocated according to the macroeconomic method

16 Rixen/ Uhl Europeanising Company Taxation – Regaining National Tax Policy Autonomy b. Minimum Rate of Taxation We have already explained why formula apportionment which is primarily based on microeconomic factors will not curb locational competition, even if it leads to a more transparent form of competition in comparison with the status quo of tax competition based on shifting paper profits. That is why the CCCTB with formula apportionment must be accompanied by a minimum tax rate if tax competition is not merely to be pushed onto other factors. 34 As previously shown, rates of corporate taxation have been in a downward spiral across Europe for some years. The‘old’ EU-15 countries still raise corporate tax at an average rate of 26.9% on retained profit in 2007 – as shown in Table 1. The 12 new member countries(NMS-12) raise just 16.8% in 2007. Admittedly, if the Estonian plans to raise their tax rate on retained profit from the current 0% to 20% in 2009 is taken into account, then the average tax rate in the NMS-12 increases to 18.5%. The average corporate tax rate of all 27 member countries is 22.6% in 2007. In 1992, when the Ruding Committee Report(Ruding Report 1992, 114) on tax distortions in the Internal Market was presented to the European Council, only Ireland had a corporate tax rate of less than 30%. Now only four of the EU-15 countries have a rate above 30%(excluding local taxes). Even today, with a tax rate of 12.5%, Ireland is the maverick at the bottom of the range – and thus beats many of the new member countries, of which only Bulgaria, Estonia(until 2009) and Cyprus have a rate below that. The spread of rates within the European Union indicates that any agreement on a common minimum tax rate will not be a simple endeavour. Given this variance of rates, it is very difficult to propose a rate that will satisfy the large high-tax countries as well as the low-tax countries. But since in our proposal we foresee a macroeconomic(re-)distribution component that is to the advantage of the current‘low-tax’ countries, the minimum tax rate could be set closer to the preferred rates of the‘high-tax’ countries. It is not easy to propose a concrete number for the minimum tax rate. Which tax rates are appropriate depends very much on how exactly the CCCTB is defined. In so far as a narrow tax base is chosen the tax rate must be higher. The broader the tax base, the cannot be determined without exact data, which is not available to us. 34 The Advisory Council to the German Federal Ministry of Finance shares this view(2007, 76). lower the tax rate can be. 35 Our proposal for an exemplary minimum tax rate is based on the assumption that the common European tax base is as broad as the current average national tax bases. This rate, inclusive of local business taxes, should be 30%. 36 c. Mandatory Application and General Validity We have described above how and why an optional application of the CCCTB – that is, companies would be free to choose the new or the old system – could lead to a significant intensification of tax competition. Consequently, it is important that a common consolidated basis of assessment would apply to every company so that there will not be any new tax distortions in the Single Market. Altogether, member countries could regain some of their tax autonomy with a mandatory common consolidated corporate tax base with a minimum tax rate. 35 Precisely this consideration of the connection between the tax base and the rate of tax shows why the Commission’s argument that the question of standardising the tax base must be separated from the discussion on the tax rate is not convincing. On the differences in the effective tax burden in the case of a CCCTB, see Jacobs et al.(2005). Basically, the effects of a CCCTB with formula apportionment on the tax revenue of the member countries can hardly be estimated due to lack of appropriate data at present(Advisory Council to the German Federal Ministry of Finance 2007, 65 f.). 36 Regarding the question of how a European-wide minimum tax rate would affect tax competition between EU countries and non-EU countries: the European-wide implementation of a minimum tax rate could lead to investors choosing locations outside the EU. This issue has played a major role in discussions over the directive on taxation of interest payments(Genschel 2003). But direct investment is less mobile than portfolio investments which are caught by taxation on interest payments, so that this problem is not as acute in the taxation of corporations. In addition, the traditional unilateral anti-avoidance provisions which are intended to restrict the tax avoidance strategies of multinational enterprises in non-EU countries could be Europeanised.

International Policy Analysis 17 V. Back to the Political Debate Summarising the findings of this report, it is quite clear that EU member states must give up their formal tax sovereignty to regain effective autonomy over the design of their tax systems. Only then will national parliaments regain the scope for decision making on a major political question again, i. e. how high and progressive the respective tax rates should be. As we have shown, the implications of the European Single Market – especially the freedoms of capital movement and establishment – have created very strong interdependencies among the taxation systems of member countries. As yet, member states have not reacted with more tax cooperation, but rather with more intense tax competition against one another. Multinational enterprises take advantage of the resultant differences in tax levels. But this rarely involves the relocation of entire production sites, and mostly consists of shifting paper profits. This hardly requires more than the establishment of a simple mailbox company. Protective measures(e. g. controlled foreign company legislation), which are still allowed in the international tax regime, are hardly possible within Europe because of the very far-reaching interpretation of the freedom of establishment by the ECJ. Both these factors – Single Market and tax competition – mean that the member countries have practically robbed themselves of their competence to organise their own tax policies effectively, without at the same time assigning this competence to the European level. Thus, at present, in significant areas of tax policy, there is a political vacuum and, since governments are forced to react more to the policies of other nations than the desires of their own constituents, little democratic legitimation. Our proposal of a mandatory common consolidated corporate tax base with a minimum tax rate would change that. We want to reinstate the capacity to make political decisions over European tax systems back to democratically legitimised bodies. In order to achieve this, we have to begin with a farewell – a farewell to the myth of member states’ sovereignty over tax matters. It is quite simply not possible to have both the advantages of a Single Market and de facto sovereignty over tax matters at the same time. One cannot have the one, a Single Market without tax obstacles and at the same time keep the other, i. e. complete and effective national control over the type, level, and form of taxation. Tax competition among member countries helps us to ignore this truth because it implies formal sovereignty. After all, governments can decide to reduce taxation unilaterally. This freedom, however, is very lopsided since it clearly gives preference to those who demand a general reduction of corporate and income taxes for material or ideological reasons. Everyone else has allegedly no other political option(Genschel/Rixen/Uhl 2007). Our main objective in this report was to demonstrate that this type of argument is false because a European solution would open a new path that would allow a more democratic organisation of taxation policy again. So, the farewell to the myth of sovereignty over tax matters should not be too difficult because, in fact, member states lost their real sovereignity a long time ago. The responsibility for the fact that this development has not been perceived as dramatic as it is lies with the various actors in the political arena who have continued to allude to the idea of national tax sovereignty being still intact and the notion that tax competition is something that is forced upon governments without them having a choice. Against this, it is necessary to point out that the developments of the past few years have not just appeared from nowhere and that we can only like them or hate them according to our political disposition, but that instead the developments are firstly a result of policy and secondly are also politically controllable in the future. This political control can be regained in and for member countries only if in a first step formal sovereignty is handed up to the European level. We see the European debate on a common consolidated corporate tax base as evidence that some member countries are conscious of the notion that Europe – and with it the single member states – has a significant tax policy problem that cannot be adequately solved by simple laissez-faire. Any national and European debate on tax policy is certainly positive. But this is not enough. In addition, it is necessary to make visible various alternatives in the political arena so that a democratic search for the best solution can begin. The politicisation of the topic of tax competition is required. Our proposal is intended to contribute to this. Tax policy is not politically neutral. On the contrary, tax policy is central to questions of distribution and equity and thus basic democratic and social standards.

18 Rixen/ Uhl Europeanising Company Taxation – Regaining National Tax Policy Autonomy Literature Advisory Council to the German Federal Ministry of Finance (2004): Flat Tax oder Duale Einkommensteuer? Zwei Entwürfe zur Reform der deutschen Einkommensbesteuerung. http:// www.bundesfinanzministerium.de/lang_de/DE/Service/Down loads/Downloads__4/25706__0,templateId= raw,property= publicationFile.pdf(last visited on 28 March 2007). Advisory Council to the German Federal Ministry of Finance (2007): Einheitliche Bemessungsgrundlage der Körperschaftsteuer in der Europäischen Union. http://www.bundesfinanzministerium.de/lang_de/DE/Service/Downloads/Abt__I/0703 231a3003,templateId= raw,property= publicationFile.pdf (last visited on 28 March 2007) Agúndez-García, Ana(2006): The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-jurisdictional Corporate Income Taxation: a Review of Issues and Options. European Commission – Taxation Papers. Brussels. Alesina, Alberto/Spolaore, Enrico(2003): The Size of Nations. Cambridge: MIT Press. Bach, Stefan(2001): Internationale Entwicklungstendenzen nationaler Steuersysteme – von der direkten zur indirekten Besteuerung? Berlin: Duncker& Humblot. Bartelsman, Eric J./Beetsma, Roel M.W.J.(2003): Why Pay More? Corporate Tax Avoidance through Transfer Pricing in OECD Countries. Journal of Public Economics 87: 2225-2252. Bucovetsky, Sam(1991): Asymmetric Tax Competition. Journal of Urban Economics 30: 167-181. Czakert, Ernst(2006): Der Stand der Arbeiten an einer gemeinsamen konsolidierten Körperschaftssteuer-Bemessungsgurndlage in der Europäischen Union. Internationales Steuerrecht 15(16): 561-566. de Mooij, Ruud A./Ederveen, Sjef(2006): What a Difference does it Make? Understanding the Empirical Literature on Taxation and International Capital Flows. General Directorate Economic and Financial Affairs Workshop on Corporate Tax Competition and Coordination in Europe. Brussels, 25 September 2006. Dehejia, Vivek H./Genschel, Philipp(1999): Tax Competition in the European Union. Politics& Society 27(3): 403-430. Desai, Mihir A./Fritz, Foley C./Hines, James R.(2006): The Demand for Tax Haven Operations. Journal of Public Economics 90: 513-531. Devereux, Michael P./Griffith, Rachel/Klemm, Alexander(2002): Corporate Income Tax Reforms and International Tax Competition. Economic Policy 17: 451-495. Devereux, Michael P./Lockwood, Ben/Redoano, Michela(2004): Do Countries Compete over Corporate Tax Rates? University of Warwick. http://www2.warwick.ac.uk/fac/soc/economics/ staff/faculty/devereux/wp/dlrrev1.pdf(last visited on 02.01.2006). Devereux, Michael P.(2006): The Impact of Taxation on the Location of Capital, Firms and Profit: A Survey of Empirical Evidence(with Data Appendix by Giorgia Maffini). Working Paper, University of Warwick. Economic and Social Committee(2002): Direct company taxation. Opinion of the Economic and Social Committee(2002/C 241/14). Brussels. European Commission(2001a): European Commission. 2001. Company Taxation in the Internal Market, SEC(2001)1681. Brussels: Commission Staff Working Paper. European Commission(2001b): Tax Policy in the European Union – Priorities for the Years Ahead. COM(2001) 260 final. Brussels. European Commission(2001c): Towards an Internal Market without Tax Obstacles. A Strategy for Providing Companies with a Consolidated Corporate Tax Base for their EU-wide Activities. COM(2001) 582 final. Brussels. European Commission(2006a): Implementing the Community Lisbon Programme: Progress to date and next steps towards a Common Consolidated Corporate Tax Base(CCCTB). Communication From the Commission to the Council, the European Parliament and the European Economic and Social Committee COM(2006) 157 final. Brussels. European Commission(2006b): Progress to Date and Future Plans for the CCCTB. Working Document of the Working Group CCCTB. CCCTB/WP/046/doc/en. 20 November 2006. Brussels. European Commission(2006c): The Mechanism for Sharing the CCCTB. Working Document of the Working Group CCCTB CCCTB/WP/047/doc/en. 17 November 2006. Brussels European Commission(1990): Commission Communication to Parliament and the Council. Guidelines on Company Taxation (SEC(90) 601). Brussels. Eurostat(2007): Regionales BIP je Einwohner in der EU27. Pressemitteilung vom 19. Februar 2007. Luxemburg. Fuest, Clemens/Becker, Johannes(2006): Kapital und Arbeit unterwegs? Empirische Befunde zu Steuerarbitrage und Unternehmensverlagerungen.“ Harmonisierung – Kooperation – Wettbewerb. Steuerpolitische Vorschläge in und für Europa”, Konferenz der Friedrich-Ebert-Stiftung, 4-5 December 2006, Berlin. Ganghof, Steffen(2005): Konditionale Konvergenz. Ideen, Institutionen und Standortwettbewerb in der Steuerpolitik von EU- und OECD-Ländern. Zeitschrift für Internationale Beziehungen 12(1): 7-40. Ganghof, Steffen(2006a): The Politics of Income Taxation. A Comparative Analysis of Advanced Industrial Countries. Colchester: ECPR Press. Ganghof, Steffen(2006b): Wie schränkt der europäische Steuerwettbewerb nationale Handlungsspielräume in der Einkommensbesteuerung ein? Das Problem der Besteuerung übernormaler Gewinne.“ Harmonisierung – Kooperation – Wettbewerb. Steuerpolitische Vorschläge in und für Europa”, Konferenz der Friedrich-Ebert-Stiftung 4-5 December 2006, Berlin. Ganghof, Steffen/Genschel, Philipp(2007): Taxation and Democracy in the EU. Köln, Max-Planck-Institut für Gesellschaftsforschung, Diskussionspapier 07/2, www.mpifg.de. Garrett, Geoffrey(1998): Partisan Politics in the Global Economy. Cambridge: Cambridge University Press. Genschel, Philipp(2002): Steuerharmonisierung und Steuerwettbewerb in der Europäischen Union. Frankfurt a. M.: Campus. Genschel, Philipp(2003): Kontexteffekte in der europäischen Integration: Das Beispiel der Zinsbesteuerung. Politische Integration. Hrsg. von Thomas Plümper. Opladen: Westdeutscher Verlag: 28-52. Genschel, Philipp/Rixen, Thomas/Uhl, Susanne(2007): Die Ursachen des europäischen Steuerwettbewerbs, Politische Vierteljahresschrift, Sonderheft(i. E.)

International Policy Analysis 19 Genschel, Philipp/Uhl, Susanne(2006): Der Steuerstaat und die Globalisierung. Transformationen des Staates. Hrsg. von Stephan Leibfried und Michael Zürn. Frankfurt am Main: Suhrkamp. Haufler, Andreas/Schjelderup, Guttorm(2000): Corporate Tax Systems and Cross Country Profit Shifting. Oxford Economic Papers 52: 306-325. Hellerstein, Walter/McLure, Charles E.(2004): The European Commission’s Report on Company income Taxation: What the EU can learn from the Experience of the US States. International Tax and Public Finance 11: 199-220. Holzleitner, Christian(2005): Transfer Pricing – Tax Consulting Practice and Economics Research. Presentation, SFB“ International Tax Coordination”, Wirtschaftsuniversität Wien, 4 March 2005. Jacobs, Otto H./Spengel, Christoph/Stetter, Thorsten/Wendt, Carsten(2005): EU Company Taxation in Case of a Common Tax Base: A Computer-based Calculation and Comparison Using the Enhanced Model of the European Tax Analyzer. Zentrum für Europäische Wirtschaftsforschung(ZEW) Mannheim, Discussion Paper No. 05-37, ftp://ftp.zew.de/pub/zewdocs/dp/dp0537.pdf.(last visited on 21 March 2007). Li, Jinyan(2003): International Taxation in the Age of Electronic Commerce: A Comparative Study. Toronto: Canadian Tax Foundation. McIntyre, Michael J.(1993): Guidelines for Taxing International Capital Flows: The Legal Perspective. National Tax Journal 46(3): 315-321. McLure, Charles E.(1980): The States Corporate Income Tax: Lambs in Wolves´ Clothing. The Economics of Taxation. Hrsg. von H.J. Aaron und M.J. Boskin. Washington: Brookings Institution. McLure, Charles E./Weiner, Joann M.(2000): Deciding whether the European Union Should Adopt Formula Apportionment of Company Income. Taxing Capital Income in the European Union: Issues and Options for Reform. Hrsg. von Sijbren Cnossen. Oxford: Oxford University Press: 243-292. Mintz, Jack(1995): The Corporation Tax: A Survey. Fiscal Studies 16(4): 23-68. Mors, Matthias(2006): Der Steuerwettbewerb in der EU: Erfahrungen mit dem Verhaltenskodex zur Unternehmensbesteuerung.“ Harmonisierung – Kooperation – Wettbewerb. Steuerpolitische Vorschläge in und für Europa”, Konferenz der Friedrich-Ebert-Stiftung, 4-5 December 2006, Berlin. Rixen, Thomas(2006): Die institutionellen Grundlagen des globalen Steuerwettbewerbs.“ Harmonisierung – Kooperation – Wettbewerb. Steuerpolitische Vorschläge in und für Europa”, Konferenz der Friedrich-Ebert-Stiftun g, 4-5 December 2006, Berlin. Rixen, Thomas(2007a): The Political Economy of Global Tax Governance. Institutional Choice and Development in International Taxation. Dissertation, Jacobs University Bremen. Rixen, Thomas(2007b): Taxation and Cooperation: International Action Against Harmful Tax Competition. Globalization. State of the Art and Perspectives, edited by Stefan A. Schirm. London: Routledge. Ruding Report(1992): Report of the Committee of Independent Experts on Company Taxation. Luxemburg. Schratzenstaller, Margit(2006): Steuerwettbewerb: Spürbare Folgen?“ Harmonisierung – Kooperation – Wettbewerb. Steuerpolitische Vorschläge in und für Europa”, Konferenz der Friedrich-Ebert-Stiftung, 4-5 December 2006, Berlin. Schwarz, Peter(2007): Does Capital Mobility Reduce the Corporate-Labor Tax Ratio? Public Choice(im Erscheinen). Sørensen, Peter B.(2005): Dual Income Taxation: Why and How? FinanzArchiv 61(4): 559-586. Stewart, Kenneth/Webb, Michael C.(2006): International competition in corporate taxation: Evidence from the OECD time series. Economic Policy: 153-201. Thalheimer, Frank(2005): Stuttgart und Karlsruhe im Spitzenfeld der wirtschaftsstärksten Regionen Europas. Statistisches Landesamt Baden-Württemberg http://www.statistik.badenwuerttemberg.de(last visited on 20 February 2007). Themaat, Pieter Verloren van(1966): Die Steuerharmonisierung in der EWG. Der Betrieb 19: 834-840. Uhl, Susanne(2006): Die Gebührengesellschaft. Kehrt die Steuerpolitik zu ihren Anfängen zurück? vorgänge 45(2): 15-22. Uhl, Susanne(2007): Steuerstaatlichkeit in Europa. Über die Transformation der Steuersysteme europäischer Mitgliedstaaten, deren Bedeutung für die nationale Steuerautonomie und die steuerpolitische Verfasstheit Europas. Dissertation, Jacobs University Bremen. Vogel, Klaus(1990): Doppelbesteuerungsabkommen der Bundesrepublik Deutschland auf dem Gebiet der Steuern vom Einkommen und Vermögen: Kommentar auf der Grundlage der Musterabkommen. München: C.H. Beck. Wathelet, Melchior(2006): Direct Taxation and European Law (The Recent Case Law of the ECJ). CFE Forum. Brussels. Weichenrieder, Alfons J.(2005):(Why) Do We Need Corporate Taxation? CESifo Working Paper No. 1495, www.CESifo.de (last visited on 05.07.2006). Weiner, Joann Martens(2005): Formulary Apportionment and Group Taxation In the European Union: Insights From the United States and Canada. European Commission – Taxation Papers. Brussels. Wilson, John D.(1991): Tax Competition with Interregional Differences in Factor Endowments. Regional Science and Urban Economics 21: 423-451. Winner, Hannes(2005): Has Tax Competition Emerged in OECD Countries? Evidence from Panel Data. International Tax and Public Finance 12: 667-687.

A selection of publications by the“International Policy Analysis” Uwe Wissenbach The EU’s effective multilateralism – but with whom? Functional multilateralism and the rise of China May 2007 Roger Hällhag New Sweden: Crushing or Confirming a Social Democratic Model? FES-Country-Report April 2007 Jana Zitzler Plea for a European Minimum Wage Policy April 2007 Michael Sommer: A Social Europe Needs Workers’ Consultation and Participation[only available online] April 2007 James K. Galbraith Maastricht 2042 and the Fate of Europe. Toward Convergence and Full Employment March 2007 Stefanie Flechtner European Security and Defense Policy: between “Offensive Defense” and“Human Security” January 2007 Sven Biscop The International Security Engagement of the European Union – Courage and Capabilities for a“More Active” EU. Report from the 1st European Strategic Forum, Warsaw 2006. January 2007 Michael Dauderstädt, Arne Schildberg(eds) Dead Ends of Transition – Rentier Economies and Protectorates September 2006 Marius Busemeyer, Christian Kellermann, Alexander Petring, Andrej Stuchlik Political Positions on the European Economic and Social Model – a Map of Interests [also available in German] August 2006 Working Group“European Integration” The EU needs a new economic policy! Launching the debate[also available in German] January 2006 Alexander Petring, Christian Kellermann New Options for a European Economic and Social Policy October 2005 Find these and further publications online: www.fes.de/internationalepolitik orders to: Friedrich-Ebert-Stiftung Internationale Politikanalyse Ursula Müller D–53170 Bonn E-Mail: info.ipa@fes.de Tel.:+49(228) 883-212 Fax:+49(228) 883-625

Imprint Friedrich-Ebert-Stiftung International Policy Analysis International Dialogue Department D-10785 Berlin www.fes.de/internationalepolitik E-Mail: info.ipa@fes.de ISBN 978-3-89892-687-4 Orders Friedrich-Ebert-Stiftung International Policy Analysis z. Hd. Ursula Müller D-53170 Bonn E-Mail: info.ipa@fes.de Fax:+49(228) 883-625 All texts are available online: www.fes.de/internationalepolitik The views expressed in this publication are not necessarily those of the Friedrich-Ebert-Stiftung or of the organization for which the author works.