MPIfG Working Paper 01/1, May 2001

Globalization, Tax Competition, and the Fiscal Viability of the Welfare State

by Philipp Genschel

Philipp Genschel is a research fellow at the Max Planck Institute for the Study of Societies in Cologne.

Abstract

Does globalization undermine the fiscal basis of the welfare state? The conventional wisdom believes so: open borders cause tax competition, which in turn leads to a race to the bottom in capital taxation. However, the data show that revenues from capital taxation are fairly stable in OECD countries. Some observers conclude from this that globalization does not pose much of a challenge to the welfare state. This conclusion is unwarranted because it overlooks that tax competition was not the only challenge facing welfare states during the 1980s and 1990s. There was also slow growth, rampant unemployment, and high levels of precommitted spending. These problems exerted countervailing pressures that prevented a race to the bottom in taxation. Yet, this does not mean that national autonomy has not been diminished. The welfare state is trapped in between external pressures to reduce the tax burden on capital and internal pressures to maintain revenue levels and relieve the tax burden on labor.

Contents

1 The conventional wisdom and its critics

2 Did tax competition change the structure of taxation?

3 Would total taxation be higher without tax competition?

4 Would capital tax revenues be higher without tax competition?

4.1 Corporate taxes

4.2 Taxes on financial income

5 Would taxes on labor and consumption be lower without tax competition?

5.1 Labor taxes

5.2 Consumption taxes

5.3 Green taxes

6 Taxing dilemmas

Literature

1 The conventional wisdom and its critics

During the 1990s, there was a lot of concern in Europe that globalization would undermine the fiscal basis of the welfare state. Newspapers were full of dire warnings from policy makers who saw the coming of "cut throat tax competition" (Maystadt 1994: 2) and bemoaned the loss of "billions of Euro" to unfair tax poaching (Lafontaine and Strauss-Kahn 1999: 18). G-7 summits, the European Commission, and the OECD issued alarmist reports on "harmful tax competition" (OECD 1998c; European Commission 1996a). Scholars warned of "beggar-thy-neighbor policies" (Tanzi und Bovenberg 1990: 187) and an impending "race to the bottom" that would force "down the share of total government revenue generated by taxes on business and capital income" (Scharpf 1997: 531). "In equilibrium, the tax rate on capital in each state will be driven to zero" (Frey 1990: 89) with potentially serious consequences for public goods provision and distributional equality (CEPR 1993; Sinn 1997). "The end of redistribution" (Steinmo 1994) seemed near. Doomsday for the welfare state.

Recently, a group of scholars has challenged this view. They maintain that there is little evidence for a close relationship between economic integration and national tax policy and no indication that tax competition is eroding the mobile capital tax base or depressing tax revenues. Even sophisticated empirical analyses fail to hint of a race to the bottom in capital taxation. The "conventional wisdom," so it seems, "is too simple and considerably overdrawn" (Garrett 1995: 682; see also Garrett 1998a; Garrett 1998b; Kirchgässner and Pommerehne 1996; Quinn 1997; Swank 1998).

Who is wrong? Don't policy makers know what they are talking about? Do they suffer from 'false consciousness' concerning tax competition, or are Geoffrey Garrett et al. looking at the wrong data or drawing the wrong conclusions? Probably both. As I will show in the next section of this paper, the critics are right in claiming that the evidence on tax competition does not fit the conventional 'race-to-the-bottom' scenario. In recent years, the average OECD country has neither suffered a dramatic decrease in total tax revenue nor experienced a clear shift of the tax burden from mobile to immobile bases (section 2). However, the critics are wrong in concluding from this that tax competition is not a serious constraint on national tax policy. As I will show, tax levels and tax ratios have remained stable not because tax competition was impotent, but because there have been countervailing pressures that neutralized its impact. Globalization was not the only challenge facing the welfare state during the 1980s and 1990s. There was also slow growth, high unemployment, high levels of precommitted spending, and mounting public debt. The handling of these problems, which were not directly linked to globalization, compromised the adjustment to tax competition: The reduction of tax revenue was not an option given high levels of spending and debt (section 3). Governments had to maintain tax revenue even though tax competition increased the difficulties of taxing capital (section 4). At the same time, high levels of unemployment limited how far the tax burden could be shifted to immobile tax bases such as labor and consumption. Labor and consumption may not normally flee abroad to avoid high taxes. But they 'evade' them through unemployment or the shadow economy (section 5). In conclusion, the conventional wisdom is correct: Tax competition is a constraint on national tax policy. But this constraint makes itself felt differently than the conventional wisdom assumes. It does not force the welfare state into a race to the bottom, but traps it in between external pressures to reduce the tax burden on capital and internal pressures to maintain revenue levels and relieve the tax burden on labor. The result is more austerity, more deficit finance, and a less employment-friendly tax mix than would have prevailed in a world without tax competition (section 6).

2 Did tax competition change the structure of taxation?

The national tax systems of OECD countries are products and symptoms of economic closure.[1] Their main components were conceived when national borders were relatively closed to economic transactions. The progressive income tax made its breakthrough during the huge fiscal expansion of the First World War. Turnover taxation was introduced in most countries during the interwar years when first inflation and then depression cut into income tax revenues. Corporate taxes and social security contributions also made their debut during this period. After World War II, social security contributions were expanded massively to finance the build-up of the welfare state. All this occurred in a context of separated national markets. Trade barriers and capital controls restricted tax-base mobility to national markets and prevented any international spillover effects of national tax-policy choices. Taxation was a purely national affair. The governments' fiscal sovereignty went unchallenged.

As the fences separating national markets were coming down during the 1980s and 1990s, many observers feared that this sovereignty might drain away. The elimination of trade barriers and capital controls made exit a viable option for mobile factors, such as human, physical, or financial capital. Governments would have to compete for these factors and could no longer turn a fiscal profit on them. As a consequence, it was feared that the welfare state would shrink in scale and its power to redistribute would be diminished. The scale would shrink because high tax levels were believed to be unsustainable if tax levels were significantly lower elsewhere. Redistribution would decrease because capital and other mobile tax bases would no longer pay high taxes - if they paid any tax at all (CEPR 1993: 67). In short, two consequences were predicted:

• the level of total taxation would decline (see e.g. European Commission 1996a; Steinmo 1996: 3; Tanzi 1998: 19; Hagen et al. 1998: 161), and

• the tax burden would shift from mobile tax bases, most importantly capital, to immobile bases, such as labor, consumption, and real estate (see e.g. Sinn 1990: 501; CEPR 1993: 82; Steinmo 1996: 2; Hagen et al. 1998: 193-194; Schulze and Ursprung 1999: 17-18).

These predictions have become the conventional wisdom on tax competition and have dominated the public debate on this subject (see Radaelli 1998). But are they also true? Are they sustained by the data? There is some reason for doubt.

Figure 1 shows how total tax receipts, averaged across sixteen OECD countries,[2] have developed since 1970 (figure 1). So far there is no evidence of a decline in tax revenues. In fact, the share of tax revenues in GDP has risen by eight percentage points from roughly 32 percent in 1970 to about 40 percent in 1998.

Are changes in the composition of the tax burden more in line with conventional wisdom? Figure 2 presents the revenues of various types of taxes as percentages of the total tax revenue (tax ratios) and shows how these percentages have changed in the average OECD-16 country since 1965. Interestingly, the most obvious changes occurred before 1975, when national borders were still fairly closed. After the mid-1970s, tax ratios changed remarkably little. The changes that did occur make little sense in terms of the conventional wisdom. Revenue from taxes on property and consumption has decreased rather than increased since the mid-1970s. Corporate tax revenues have gone up rather than down. Only the rising percentage of social security contributions suggests that the tax burden on immobile tax bases - wage earners - has increased.[3]

Closer inspection of figure 2 hints that there may be more shifting of the tax burden taking place than is immediately apparent (see Ganghof 1999: 34-35). The nearly constant percentage of the personal income tax in total tax revenue may mask a shift of the tax burden from mobile to immobile sources of personal income, i.e. from capital to labor income. The constancy of the tax ratios may conceal changes in the size of the underlying tax bases. If unchanging proportions of tax revenue derive from tax bases of changing proportion, then clearly a redistribution of the tax burden has occurred. National account data suggest that the relative weight of the major macroeconomic tax bases has changed during recent years (Kramer 1998). Michael Webb claims, for example, that the share of corporate profits in GDP has increased significantly since the early 1980s. Hence, the slight increase of the corporate tax ratio reported in figure 2 may actually conceal a decrease of the effective tax burden per unit of corporate profit (Webb 1998: 12).

In order to get a better picture of potential shifts in the tax burden, it is necessary to break down personal income tax revenues according to source and to control for tax-base effects. Personal income tax receipts have to be classified according to whether they derive from capital or labor income to control for shifts between these two categories of personal income, and changes in tax revenues have to be related to changes in the underlying tax base in order to see if the effective tax burden per unit of tax base has changed. Currently, there are two data sources that satisfy these requirements, the statistics on average effective tax rates (table 1) calculated by Mendoza and colleagues (Mendoza et al. 1994; Mendoza et al. 1997) and the statistics on implicit tax rates (table 2) provided by Eurostat, the statistical office of the EU (Eurostat 1997: 14). The underlying concepts are very similar: Tax revenues are classified according to the tax base from which they derive and then expressed as a share of this tax base (Mendoza et al. 1994: 299; Eurostat 1997: 14). What results is the effective/implicit tax burden per unit of tax base. Unfortunately, these two data sets operate with different country samples and different tax-base definitions. Mendoza et al. use the trisection of consumption, labor, and capital to classify tax bases, while Eurostat looks at consumption, labor, and other factors of production. Capital clearly is the most important of these other factors, but the category also includes energy, environmental resources, and self-employed labor. While Eurostat focuses exclusively on EU member states, Mendoza et al. look at a sample of 14 OECD countries. Nevertheless, tables 1 and 2 show interesting similarities.

Both tables are fairly consistent with respect to consumption taxation. They show that the effective/implicit tax burden on goods and services is low relative to labor or capital income, and that the level of taxation has increased very little since the 1970s. The picture of labor taxation is also similar. Both tables show that the effective/implicit tax burden on labor has grown faster than the tax burden on either capital income or consumption, i.e. part of the tax burden has been shifted to labor. The tables diverge with respect to taxes on capital and other factors of production. While the implicit tax rate on other factors of production hardly increased between the 1970s and the 1990s according to table 2, table 1 records a substantial increase of nine percentage points in the average effective tax rate on capital.

In conclusion, there is scant empirical evidence to support the conventional wisdom. To be sure, the effective/implicit tax burden on labor has grown relative to capital. Yet in the following respects, the empirical evidence fails completely to uphold the conventional wisdom:

• Total tax revenues have not been declining since the 1970s, but continue to increase.

• The capital income tax base has not been eroded visibly. On the contrary, the tax ratio of corporate income has increased slightly, and the average effective tax rate on capital has increased greatly. It is true that, contrary to this finding, the stagnation of the implicit tax rate on other factors of production suggests that the tax burden on capital has not increased. But even that is "a long way from predictions of a free fall in capital taxation" (Garrett 1998a: 814).

The critics seem to have a point when they argue that the conventional rhetoric on a "policy race to the neoliberal bottom"is at variance with the available data (Garrett 1998a: 823; see also Swank 1998: 690; Quinn 1997: 539; Kirchgässner and Pommerehne 1996: 366). Are they also right to conclude that tax competition does not seriously constrain national policy autonomy? That governments "wishing to expand the public economy for political reasons may do so (including increasing taxes on capital to pay for new spending)" (Garrett 1998a: 823)? That they "have room to pursue their preferred policy goals" (Swank 1998: 691) and may even find their governmental capacity enhanced (Quinn 1997: 541)? This is, perhaps, more difficult to believe. In any event, such conclusions cannot be drawn from the lack of evidence for a race-to-the-bottom in capital taxation.

To argue that tax competition does not constrain national tax policy implies that the observed tax policy choices and outcomes would have been the same in the absence of tax competition. In other words, it assumes that the slow increase of total tax revenues since the 1980s, the constancy of tax ratios, and the increase of the effective/implicit tax burden on labor would have occurred in any event because they reflected government preferences rather than the structural constraints imposed by economic integration. As I am going to show next, this is not a very plausible assumption. There is good reason to believe that in a counterfactual world without tax competition the level of total taxation would be higher, capital taxation would generate more revenue, and the tax burden on labor income and consumption would be lower.

3 Would total taxation be higher without tax competition?

The level of total taxation would indeed be higher in a world without tax competition because the growth of public expenditure is hard to contain and even harder to reverse, and because large deficits are not a long-term solution for bridging the gap between stagnating revenues and increasing expenditures. While total tax revenues have risen since the 1970s, public spending has risen even further, leaving most OECD countries with a mounting burden of public debt (see figure 1).

The high levels of public expenditure during the 1980s and 1990s were rooted in decisions made in the 1960s and early 1970s to create and expand welfare state programs in old-age pensions, health care, unemployment, and other areas of social protection. These entitlement programs created expenditure obligations that proved hard to control. Once legislators had stipulated who was to be eligible for transfer payments of what size, then the number of households receiving such payments and, consequently, the amount of transfers paid, depended on factors largely beyond legislative control: demographic trends, health status of the population, labor market developments, growth rates, etc. (Cordes 1996: 98; Pierson 1996).

Most of the new or expanded welfare programs were based on the assumption that the high growth rates and low unemployment levels of the 1960s would continue into the future (Kawai and Onitsuka 1996: 188). Yet, after the first oil crisis in 1973, growth rates slumped, and unemployment began to rise. Initially, governments reactioned to these developments in Keynesian fashion: benefit levels were maintained, the range of beneficiaries was increased, new welfare programs were established, and the level of public expenditure was raised in an effort to shorten the recession, dampen its social impact, and fight unemployment (Stephens et al. 1999: 172). The revenue shortfall was met through large fiscal deficits. Deficit finance was an inexpensive option, since long-term real interest rates were low. Despite sizeable deficits, the public debt grew only slowly during the 1970s (figure 1). Also, deficits seemed to be a temporary expedient only. It was thought that, once the recession was over and OECD economies returned to the high growth rates and high employment levels of the period preceding the oil crisis, governments would reduce fiscal imbalances. As it turned out, however, the economic difficulties represented more than just a cyclical downturn. Slow growth continued, and unemployment rose even further in the wake of the second oil crisis. Long-term interest rates began to rise, and the level of public debt began to escalate (figure 1). The net debt service doubled between 1980 and 1985 from 1.7 percent of GDP to 3.4 percent (table 3). Higher interest payments meant more precommitted spending and less budgetary flexibility; they thus contributed to the fiscal predicament of the welfare state.

When it slowly dawned on policy makers and public opinion alike that the logic of the economic game had changed in a fundamental way, the expansive fiscal policy of the 1970s came to be regarded as a mistake. Leftist governments were replaced by conservative, supply-side-minded ones in many countries, and fiscal consolidation and expenditure restraint became top priorities throughout the OECD. Most attention and political energy focused on welfare retrenchment. But cuts in entitlement programs actually contributed little to the consolidation process (Pierson 1996; Stephens et al. 1999). Social expenditures continued to rise and were only partly compensated by cuts in public investment (table 3). An economic boom helped to improve appearances during the second half of the 1980s. Deficits were reduced, and some countries managed to balance their budgets, including Denmark, Germany, the UK, and Sweden. Yet the debt level remained high. Interest payments continued to tie up 3 percent of GDP in the average OECD country and more than 5 percent in EC member states. The structural discrepancy between spending requirements and tax revenues continued.

When economic conditions worsened in the 1990s, levels of public expenditure immediately shot up again. Large deficits reoccurred and pulled upward the level of gross public debt and the cost of debt service. Governments were in for another round of expenditure containment. Again, the focus was on welfare retrenchment. As a result, the welfare state has undoubtedly become meaner and less generous (Clayton and Pontusson 1998). Still, social expenditures continued to rise. In part, this was due to the character of the reforms, which were often incremental rather than radical and reduced future rather than current spending. Yet other factors also played a role in this rise of social expenditure: rising levels of unemployment that sustained high spending even as social rights and benefits were curtailed, (4) growing health care costs related to an aging population, and increased spending on public pensions due to early retirement schemes and population aging (Pierson 1996; Steinmo and Swank 1999: 30). As before, the only real savings came from cuts in public investment (OECD 1998a: 152). By the mid-1990s, public investment had reached the "lowest level since the beginning of the century" (Tanzi and Schuknecht 2000: 47).

As we enter the new century, the fiscal situation of most OECD countries has markedly improved. Deficits have fallen again. Some countries run budget surpluses and have started to discharge their debt. However, most of these improvements are due to the cyclical upswing - higher growth rates and better employment figures - and not to contractionary spending policies. Expenditure containment has reached its limits. The continual cuts in capital spending have raised concerns that public investment may be too low and may jeopardize long-term growth (OECD 1998a: 152; Sturm 1998). The process of social security reform is becoming increasingly painful. All the 'easy' changes have already been implemented. The reform process has reached the core of the system and has become correspondingly controversial (Hemerijck and Schludi 2000: 224).

During the past 20 years, austerity has not been enough to achieve fiscal consolidation [5] in advanced welfare states (Kawai and Onitsuka 1996: 184). This goal also required tax increases (Messere 1997: 313). A survey of fifteen episodes of significant fiscal consolidation in OECD countries shows that all cases but one were based on an increase of total taxation, while only ten cases involved a cut in expenditure. "Overall, more than half of the consolidation episodes under review relied more on revenue increases than on expenditure cuts" (OECD 1996: 36). Evidence from the consolidation efforts in the EU during the period leading up to the establishment of the monetary union confirms this result (OECD 1998a: 152). Sound public finances were predicated on higher taxes.

In fact, many countries made conscious efforts to raise government revenue (Kawai and Onitsuka 1996: 182). During the 1980s, these attempts were still modest. The most visible feature of the supply-side-inspired reforms of this decade was the reduction of headline tax rates. Almost everywhere, governments cut, at times significantly, the top rates on personal and corporate income (Hallerberg and Basinger 1998). Yet simultaneously, they broadened the tax base. They curtailed interest deductions, cut savings and investment incentives, eliminated tax shelters, tightened capital gains taxation, and intensified the enforcement of tax laws. As a consequence, the reforms were usually at least revenue neutral (Steinmo and Swank 1999). The tax policy of the 1990s was more overtly "budget driven" (Sunley and Stotsky 1998: 426). In the Euro-area, in particular, it was sometimes difficult to see any rationale behind policy initiatives "apart from getting more revenue in the short term" (Castellucci 1998: 159). The base-broadening continued. Social security contributions, VAT rates, and - in some instances - excises increased. A few governments, including the Netherlands, Denmark, and Sweden, introduced new 'green taxes' on energy and the consumption of other natural resources. Germany, France, Italy, and Sweden levied temporary surcharges on gross personal and corporate income.

The increase in total tax revenue (figure 1) and the rise of average/implicit tax rates on labor and capital (tables 1 and 2) show that these efforts were not without effect. However, revenue gains would have been larger in the absence of tax competition. As I will demonstrate in the following, tax competition prevented governments from tapping into important sources of capital tax revenue and forced them to rely more on labor taxation to meet revenue targets. This had detrimental repercussions on employment and growth, and eventually on tax revenues.

4 Would capital tax revenues be higher without tax competition?

The critics of the conventional wisdom doubt that capital tax revenues would be higher in a world without tax competition. According to Garrett, the observed increase in average effective tax rates on capital (table 1) "should give ... pause" to those who assume that tax competition bids down capital taxation (Garrett 1998c: 87; see also Swank 1998: 672). Yet, the reasoning is not entirely clear. If capital owners shift capital out of high-tax jurisdictions, this does not necessarily reduce the effective tax burden on the remaining capital. On the contrary, governments may be forced to increase the effective tax burden in order to maintain the same revenue from an eroding tax base (see Hagen et al. 1998: 166). Given tax-base mobility, an increase in effective tax rates may indicate intense tax competition and revenue losses rather than the reverse. Hence, the highly aggregated data on effective/implicit tax rates, tax ratios, and total tax revenues are a poor guide for assessing if and how deeply tax competition cuts into the mobile capital tax base. To get a better picture, it is necessary to look at the behavior of taxpayers at the microlevel. Do individual and corporate capital owners make efforts to avoid or evade taxation, and how widespread and effective are these efforts? Obviously, the answer is hard to quantify: "if we could measure it, we could tax it" (Perez-Navarro 1999: 18). But still, some instructive evidence exists.

4.1 Corporate taxes

The most obvious way in which corporations can reduce their corporate tax bill is by setting up shop in a low-tax country. Econometric evidence confirms that, by and large, foreign direct investment reacts negatively to high effective tax rates (Leibfritz 1997: 31; Sørensen 1992; Hines 1999: 308). However, the estimated effects are often weak in terms of size and statistical significance, suggesting that non-tax factors, such as access to new technologies, large markets, or a well-educated work force, are more important determinants in the choice of business location (e.g. Garrett 1998a: 800). This does not mean, however, that companies do not avoid taxes. They just take another route to avoidance. Instead of relocating real activities to low-tax jurisdictions, multinational companies manipulate commercial and financial exchanges within the company to shift paper profits out of high-tax environments and into low-tax jurisdictions. For example, to reduce the taxable profits of a subsidiary in a high-tax country, affiliates in less tax-heavy locations will charge inflated prices for deliveries to this subsidiary and pay deflated prices for deliveries they receive from it (transfer pricing). To increase deductible expenses in the high-tax country, affiliates will finance the subsidiary through intracompany loans rather than equity, because interest payments are tax-deductible while dividend payments are not (thin capitalization). To reduce the tax burden on the resulting interest income, they will collect the interest income through a holding company in a low-tax environment, where it is taxed lightly, if at all (base company) (Ruding Report 1992: 121-141; Hines and Rice 1994: 156-157; Gammie and Radaelli 2000: 15-18).

As long as multinational companies can move profits artificially between high- and low-tax jurisdictions, there is less need to also move the underlying profit-generating activities. They can place location-specific business functions in countries featuring the best local conditions, irrespective of tax, if the tax factor is taken into special consideration when they decide where to place business functions that are not location-specific. In other words, multinational companies cannot shift profits out of high-tax jurisdictions without maintaining a presence in a low-tax jurisdiction. Therefore, while foreign direct investment in general is not very sensitive to tax differentials, the investment in location-unspecific activities, such as intragroup finance, the management of intangible assets, headquarter administration, and other overhead services, tends to be extremely sensitive to tax differentials. Because these functions provide the strategic nodes for successful international tax planning, it matters a lot that they are placed in a favorable tax environment. Survey data confirm that taxation is a much more important factor in determining the location of a financial service center than that of a production plant or sales outlet (Ruding Report 1992: 114). This, in turn, makes it attractive for governments to specifically target investment in tax-sensitive business activities through preferential tax regimes.

This is not a new phenomenon. Luxembourg passed its first law on holding companies in the late 1920s. But the importance of preferential tax regimes has increased in step with the multinationalization of production. During the 1980s, a large number of new regimes were set up in the EU. Ireland, for example, established the so-called International Financial Services Center in Dublin - a regime that awards a special, reduced corporate tax rate of only 10 percent to companies providing financial services to non-residents. Belgium has a similar tax regime for trade in intangibles. So-called coordination centers that sell finance, consultancy, marketing, R&D, and other services to affiliated companies are allowed to submit only a fraction of their profits to corporate tax. Other preferential tax regimes have been established in the Netherlands (favorable tax treatment for financial service companies), Luxembourg (new rules for holding companies), Italy (Trieste off-shore financial and insurance services center), and France (preferential tax treatment of headquarter services) (Valenduc 1994; Pinto 1999).

Multinationals use preferential tax regimes as a platform for international tax planning. A coordination center in Belgium, for example, may help to reduce the corporate tax burden of a German subsidiary belonging to the same multinational company. To this end, the coordination center sells, say, a brand name to the subsidiary, which also receives a loan from the coordination center to pay for the brand name. This transaction reduces taxable profits in Germany in two ways: The - probably inflated - transfer price for the brand name cuts into the subsidiary's operating surplus, while the interest payments on the loan create deductible expenses. The corresponding profits are collected by the coordination center, where they remain largely tax-exempt. As a result, the tax bill of the overarching multinational company is reduced (Ruding Report 1992: 121-141; Owens 1993; Valenduc 1994; Schreiber 1998).

Preferential tax regimes have become popular because they boost the financial service industry, provide high-quality employment for professionals, and generate additional tax revenue. The problem is that they do so by undermining the ability of foreign governments to collect corporate taxes. Research on large German companies shows that the share of foreign income in total company income has increased greatly since the 1980s. In part, this reflects the surge of foreign acquisitions in the wake of the initiative for a Single European Market. It also reflects increased incentives for profit shifting. Since a sizeable share of the new acquisitions was in countries with preferential tax regimes - Ireland, Belgium, the Netherlands -international tax planning has become easier. As a consequence, the corporate tax burden on large multinational companies in Germany has fallen considerably, compared with that on nationally based, small- and medium-sized companies that do not enjoy the same international options to avoid taxes (Jacobs and Spengel 1997; Jacobs and Spengel 1998; see also Weichenrieder 1996; Wuntsch 1998). Allegedly, some large and profitable companies, including Daimler Chrysler and BMW, hardly pay any tax in Germany. It is impossible to give an exact number for the size of the attendant revenue loss. But experts suggest that it may be significant (OECD 1995: 73; Grigat 1997: 404; Deutsche Bundesbank 1997: 93).

Similar evidence is available for the United States. A House of Representatives study claimed in 1990 that half of a sample consisting of almost forty foreign companies had paid virtually no US taxes over a ten-year period. The revenue lost through transfer-price manipulations was estimated at US $35 billion (Dicken 1998: 247). Another study discovered a significant relationship between the income reported by US subsidiaries abroad and the tax rate in the host country. Reported profits were found to be lower in high-tax countries, confirming that multinational enterprises shift profits to low-tax countries (Hines and Rice 1994: 162; Hines 1999). [6] This is also consistent with the observation that foreign-controlled companies in the United States report significantly less taxable income than their domestically owned counterparts, which is at least partly attributable to international tax planning (Grubert et al. 1993). In general, market-based methods of transfer pricing seem to be less common in international transactions within US multinationals than in domestic transactions, for which tax avoidance incentives for price manipulation do not exist (Benvignati 1985). Considering this evidence, it seems fair to conclude that, even though corporate tax revenue has slightly increased as a share of total tax revenue in recent years (recall figure 2), the increase would have been larger in the absence of tax competition.

4.2 Taxes on financial income

Despite the attention that political scientists pay to them, corporate taxes have not been a major revenue source in recent decades for "funding the welfare state" (Swank 1998). The personal income tax has always been much more important (see figure 2). Is there any evidence that tax competition weakens the grip of the personal income tax on capital income?

Two types of capital income are taxed under the personal income tax: profits from unincorporated business and private financial income. Small, unincorporated businesses typically lack the channels for international tax planning that multinational companies have. Hence, the taxation of their profits is not threatened much by international tax avoidance. Financial income, by contrast, is very vulnerable. Financial assets such as bank accounts, bonds, or equity are highly mobile and easy to relocate. This often allows taxpayers to reduce their tax bill by simply moving their assets across the border. Just to get an idea of the dimension of the potential revenue loss, Kramer has added the interest payments on government bonds and bank deposits in the fifteen EU member states. Assuming an effective taxation of 20 percent,[7] the resulting revenue would, according to Kramer, be almost double the corporate tax revenue (Kramer 1998: 61). Tax competition has largely prevented governments from tapping into this revenue source.

In the past, almost all OECD countries taxed financial income under the global personal income tax. Nevertheless, the tax burden on financial income was significantly lower than on other sources of personal income. First, financial income enjoyed large tax privileges. Throughout the OECD, governments granted generous tax breaks for specific types of investment in order to stimulate long-term savings and to channel it into specific savings instruments such as government bonds. Second, financial income was easy to hide. Unlike wage income, from which tax was withheld at source, financial income was taxed on the basis of taxpayer self-assessment. If taxpayers did not declare their financial income, it usually went untaxed. Given high levels of top marginal tax rates and inflation rates that sometimes ran into double digits during the 1970s, the incentive to exploit this opportunity was large. It was sometimes the only way to prevent interest income from being confiscated by a combination of personal income tax and inflation (Vanistendael 1988: 160; Robson 1996).[8] Since governments understood this problem and did not want to choke off private savings, they often turned a blind eye to this type of evasion. As a result, a large share of the tax base for interest income went untapped. It is difficult to measure this share, but the estimates reported in table 4 suggest that it was rather large, reaching 60 percent and more in many countries.

During the 1980s, governments became more assertive in taxing financial income. Inflation rates were coming down, and personal income tax rates were cut, so that there was less reason for leniency with tax evasion. More importantly, increased taxation of financial income appeared to be an elegant way to ease the budgetary strain. Since, formally, this type of income was already taxed, no new tax was necessary. It was sufficient to enforce and extend the tax that was already in place (e.g. Schlesinger 1985: 238; Tanzi and Shome 1993: 807-808). Finally, this move also promised to improve distributional equity, presumably because the lax treatment of financial earnings favored (wealthy) capital owners over (poor) wage earners (e.g. Ishi 1993: 147).

To tighten the tax grip on financial income, governments reduced or eliminated tax privileges and improved the collection of taxes that were theoretically due but often went unpaid. They introduced withholding taxes in order to make sure that at least a fraction of the tax due was collected at source. [9] Some countries also set up automatic reporting systems obliging banks to inform the tax administration about taxpayers' interest earnings. With the sole exception of Luxembourg, all OECD countries now use at least one such protective measure against domestic tax evasion (table 5).

To the chagrin of national treasuries, the introduction of these protections generated little additional revenue. Many investors reacted to withholding taxes or reporting systems not by paying more taxes but by switching from domestic to international tax evasion.

• When Belgium introduced a withholding tax of 25 percent in 1983, many investors moved their assets to withholding-tax-free Luxembourg and the Netherlands. As a consequence, the competitiveness of the Belgian financial services industry declined. Interest rates rose. In 1990, the government reduced the withholding tax rate to once again attract financial assets and financial intermediation to Belgium (Brean et al. 1991: 32).

• Austria went through much of the same experience in 1984, when it introduced a withholding tax on financial income: Savings flew to Germany, where no such withholding tax existed, and the volume of the securities market contracted significantly (Schuster 1999: 29-30).

• In the Netherlands, the introduction of an automatic reporting system in 1988 induced a drop in domestic savings deposits and prompted a major flight of capital. Short-term capital exports increased by 1.4 percent of GDP in the period following the announcement of the reporting system in July 1987 (Gardner 1992: 68).

• After an aborted attempt in 1989, Germany introduced a withholding tax of 30 percent in 1993. The result was a massive outflow of funds and a banking boom in Luxembourg. The loss for the German treasury was considerable. Instead of the projected receipts of DM 24 billion, the new measure generated only DM 11 billion in 1993. The yield did not significantly improve in later years (Deutsche Bundesbank 1994: 49-55; Deutsche Bundesbank 1997: 94).

To be sure, governments could stop international tax evasion through international cooperation. Since the late 1980s, the European Commission has continually attempted to bring the member states to agree on a European reporting system through which each government would inform the others about the foreign investments of their citizens, or, alternatively, to introduce a harmonized withholding tax on financial income. Both measures would substantially reduce the incentives for international tax evasion within the Single European Market. The problem is, however, that small countries often have no interest in stopping evasion. They have little domestic tax base to lose but a lot of foreign tax base to win from tax competition. For them, tax competition may actually be a very attractive proposition. Luxembourg is an excellent example. Hence, the negotiations on tax cooperation are progressing rather slowly, if at all (Dehejia and Genschel 1999; Bernauer 2000: chap. 5). Meanwhile, the member states compete for foreign tax evaders. Some countries, including Austria and Luxembourg, introduced or strengthened bank secrecy laws to provide foreign evaders better protection against investigations by authorities from their home countries. Others have reduced or eliminated withholding taxes on non-resident financial income to increase the advantages of evasion (OECD 1994: 175-176). As table 6 shows, only a minority of countries still levy such taxes. In those countries which do, the enforcement is often lax, and the tax systems offer multiple exemptions and reduced rates. In effect, unless the investor is scrupulously honest, interest receipts from abroad go tax free (Owens 1993: 33-34).

To reduce their vulnerability to international tax evasion, many countries have lowered their tax claims on resident financial income. Top marginal income tax rates were cut practically everywhere, and some countries even decided to exempt financial income completely from the global income tax. Norway, Sweden, and Finland turned to so-called dual income tax systems, in which capital income is taxed at low proportional rates while labor income continues to be taxed at progressive rates (Ganghof 1999: 26-27). Others only exempted financial income from progressive taxation by turning the withholding tax on this income into the final tax. Table 5 indicates the trend. While in 1985, only two out of eighteen OECD countries taxed financial income outside the general income tax, by 1998 the number had risen to eight (see table 5). evidence suggests that the exemption from progressive taxation has occasionally been successful at attracting financial investment back home that had fled abroad (see e.g. Gardner 1992: 68; Schuster 1999: 50,80). This required, however, that governments scale down their tax claims on financial income.

The willingness to take financial income out of the scope of the global income tax represents a remarkable break with past traditions. Until recently, a schedular approach to income taxation, in which each source of income is taxed separately at flat rates, was considered to be a second-best choice for countries that lacked the administrative skills and capacity to operate a global income tax. Global income taxation was believed to be superior because it promised to tax income from all sources equally without discrimination (horizontal equity), and because it could tax individuals at progressive rates according to their individual ability to pay (vertical equity) (Tanzi 1995). By moving back in the direction of schedular taxation, governments implicitly admit that they no longer have the power to enforce a global income tax in a globalized economy (Slemrod 1995: 145). Schedular taxation sacrifices both horizontal and vertical equity. But it seems to be governments' best bet to collect at least some revenue from financial income.

5 Would taxes on labor and consumption be lower without tax competition?

In contrast to corporate and financial capital, labor and consumption are fairly immobile tax bases. In border regions, some people may work, live, and shop across the border for tax reasons. But these are exceptions. Usually people don't move to another country just because income taxes and social security contributions are lower. Cross-border shopping is more common. Still its impact on national tax policy is very limited (Hagen et al. 1998: 171). Even in the EU, there is no indication of any tax competition in the area of consumption taxes (Ratzinger 1997: 469).

Nevertheless, the tax burden on consumption has increased very little over the past twenty years (recall figure 2, and tables 1 and 2). While many governments raised general consumption taxes, especially VAT, the receipts from special excises fell. Operating on specific tax rates, excises are liable to negative fiscal drag. Their real value declines with inflation unless governments adjust the rates, which they rarely did during the 1970s and 1980s (Messere 1997: 306). The effective tax burden on labor, by contrast, has increased sharply, higher social security contributions being the main driving force (figure 2). As tables 1 and 2 show, the effective/implicit tax rate on labor has gone up in recent decades almost continually in nearly every country. Now workers frequently receive only half of the total wage bill to the employer; the rest is paid to the government (OECD 1995: 96). Would that be different in a world without tax competition? Would governments use additional revenues from capital taxation to reduce the tax burden on labor?

5.1 Labor taxes

There is widespread agreement in Europe that the tax burden on labor is too high and should be reduced for three reasons: the shadow economy, distributive equity, and, most importantly, unemployment (see e.g. Tullio 1987; Europäische Kommission 1994; OECD 1995; Piekkola 1998; Daveri and Tabellini 2000).

High taxes on labor tend to increase unemployment if trade unions manage to shift the tax burden forward onto employers via higher gross wages. The employers may try to pass the cost increase on to consumers. But in a globalized economy, the scope for price increases is limited. Therefore, they reduce the demand for labor instead. Unemployment increases. The evidence "is, with some exceptions, reasonably convincing" that taxes on labor increase pre-tax wages, and, hence, unemployment (OECD 1995: 68; see also Leibfritz et al. 1997: 44-46). Daveri and Tabellini find that the association between high labor taxes and high unemployment is particularly strong in continental Europe. They estimate that the rise of ten percentage points in the average effective tax rate on labor since the 1965-1975 period (table 1) may account for as much as four percentage points of the increase in European unemployment. The relation is less pronounced in the USA, Japan, and Scandinavia (Daveri and Tabellini 2000; see also Tullio 1987: 756). [10]

High tax rates on labor are also considered a major factor behind the shadoweconomy (European Commission 1996a: 2; Schneider and Enste 2000a: 82). If an average production worker's marginal after-tax wage equals only 50 percent of his marginal pre-tax wage, as is often the case in OECD countries (OECD 1995: 96), he could offer his labor at half the price and still earn the same amount in a marginal hour of work in the untaxed shadow economy. This leaves a fairly wide ranges for mutually beneficial agreements between sellers and buyers of labor at the expense of the tax authorities. High tax wedges make it particularly hard for labor-intensive industries such as agriculture, construction, or hotels and restaurants to stay clear of shadow employment practices (Scharpf 2000: 205). Faced with price-elastic demand, they often hire labor off-the-books to reduce costs. High labor taxes also tend to crowd out markets for certain household services, encourage unreported economic activity such as babysitting, gardening, house repair, and private lessons, and cause in general "excessive self-production" and do-it-yourself activity (Rosen 1996: 736). Obviously, it is difficult to measure the size of the shadow economy. However, estimates by Friedrich Schneider and Dominik Enste suggest that it has grown in step with increases in the taxation of labor (table 7).

Finally, high taxes on labor create equity problems, especially if a large part of these taxes consists of social security contributions. Based on proportional rates and often with income caps, social security contributions tend to be proportional or regressive with regard to their distributive effect.[11] Growing reliance on them has led to a redistribution of the tax burden downwards. According to OECD data, the effective marginal tax rate on families with low-labor income rose by more than seven percentage points in OECD countries between the late-1970s and the mid-1990s, while the effective tax rates for high-labor income families rose only modestly or fell in most cases (Leibfritz et al. 1997: 43; OECD 1998b: 161). It is difficult to gain popular acceptance for high taxes on labor, especially if financial income is taxed outside the personal income tax through final withholding taxes or the dual income tax (Sørensen 1998: 23). Why should poor laborers pay more taxes than wealthy capital owners? Maintaining a certain level of perceived fairness is important (Bird et al. 1998: 87), and most people still assess fairness in terms of tax progressivity (Steinmo 1994: 10,13).

In short, high labor taxes tend to increase inequality, discourage employment growth, and spur the growth of the shadow economy. However, low participation rates and high levels of shadow activity need not lead to tax cuts. On the contrary, governments faced with an erosion of the labor tax base may be forced to raise the taxes on labor even further to maintain revenue. The continental European countries in particular are at risk of getting locked into a mutually perpetuating trap in which high labor taxes erode the labor tax base, and tax base erosion prompts the raising of taxes: "welfare states without work" (Esping-Andersen 1995). To avoid this pathology, the European Commission suggested in 1994 (Europäische Kommission 1994), and the European Council agreed the same year at its summit in Essen, that the tax burden on labor should be reduced by at least one or two percentage points of GDP. But how should the revenue loss be compensated? The European Commission advocates higher taxes on capital (see e.g. European Commission 1996b). However, this requires coordinated action to eliminate tax competition, which so far has not been forthcoming. An alternative option is to raise consumption taxes (Messere 1997: 305). Since consumption taxes affect not only labor income but unearned income from capital and transfers as well, they may help reduce the tax burden on labor. No international cooperation would be required. Unfortunately, consumption taxes suffer from the same deficiencies as labor taxes: They tend to increase unemployment, encourage the growth of the shadow economy, and create equity problems.

5.2 Consumption taxes

Consumption taxes may increase unemployment in two ways. First, by reducing the consumption that workers can finance out of their earnings, they induce trade unions to increase their wage claims with similar results as in labor taxation: higher gross wages, lower labor demand, increased unemployment (OECD 1995: 65). Second, by reducing the demand for goods and services, consumption taxes reduce the demand for the factors that go into the production of these goods and services. If the goods are labor-intensive, as in the case of non-tradable services - hairdressing, cleaning, house repair, child care, etc. -this implies a reduction of labor demand. Given a price-elastic demand for such services, the burden of adjustment will almost entirely fall on labor (Scharpf 2000: 205): Workers, often unskilled, are laid off. A reduction of consumption taxes on non-traded services has therefore been suggested as a strategy to raise low-skill employment (OECD 1995: 65). Sweden, for example, introduced reduced VAT tax rates for restaurant services, contracting work, and tourist services in the early 1990s.[12] It has been argued that these reductions had a positive impact on the labor market (Andersson and Mutén 1998: 340,347). But they also limited the scope for consumption taxes as an alternative source of revenue.

A second problem with consumption taxes is that they stimulate the shadow economy. They drive a tax wedge between gross-consumer and net-producer prices and thereby create opportunity for mutually beneficial deals: The consumer pays cash and does not receive a regular invoice, the producer does not report the income, and the tax authorities lose consumption tax receipts. Italy is notorious for such illegal dealings (Castellucci 1998: 192), but they are also commonplace in less suspicious countries such as Canada. Much of the alleged recent growth in the Canadian shadow economy is attributed to the introduction of the Goods and Services Tax in 1991. The new tax covers a wide range of services previously untouched by consumption taxes. "Within a short time, for example, it became common practice in the home renovation business for many firms to quote a lower 'no GST' price for cash transactions - and if they did not do so initially, their customers often asked them to do so" (Bird et al. 1998: 70). Reductions in tax rates are sometimes suggested to lure illegal business back into the formal economy (OECD 1995: 84).

Finally, a switch to consumption taxes may create equity problems because poor people tend to consume a higher share of their income than rich people. It is possible to attenuate regressivity by excluding daily necessities from consumption taxation (Garrett 1998b: 154). But again, this reduces the revenue potential of consumption taxes and, in turn, their capacity to compensate for lower labor tax revenues. Note also that a move towards consumption taxation reduces the tax burden on labor only if those outside the labor force are not compensated for the increase in prices (OECD 1995: 95). As many in this group live on transfer payments, this can be socially and politically tricky: Taxing poor pensioners and social assistance recipients in order to lessen the tax burden on labor may not strike everybody as fair. For this reason, raising taxes on consumption can be politically costly. This was experienced most dramatically by governments that attempted to introduce new consumption taxes. In Canada, for example, the governing Conservative party was virtually wiped out in the election following the introduction of the Goods and Services Tax (Bird et al. 1998: 76). The defeat of the Australian Liberals in the 1993 elections has also been attributed to their pledge to introduce the VAT (Sandford 1993: 106).

5.3 Green taxes

Some governments in Scandinavia and continental Europe have recently become interested in 'green taxes' on energy, pollution, and natural inputs to production as an alternative source of revenue. The European Commission also advocates green taxes as a means to reduce the tax burden on labor (see e.g. Europäische Kommission 1994; European Commission 1996a). Yet, the tax base of green taxes is narrow. According to Eurostat, energy tax receipts from excises on mineral oils and other fuels accounted for 2.2 percent of GDP in EU member states in 1995, while environmental taxes narrowly defined - everything from duties on tap water to charges on aircraft noise - raised only 0.7 percent of GDP. Taxes on labor, by contrast, accounted for 21.4 percent of GDP (Eurostat 1997: 99). In other words, reducing the effective taxation of labor by two percentage points of GDP, as suggested by the European Commission, requires energy and environmental taxes to be almost doubled. No government has attempted anything like that. Even in the most environmentally minded countries, increases of green taxes have been modest with marginal effects on employment, if any.[13] Steeper increases face numerous difficulties.

Basically, green taxes pose problems similar to those found in consumption taxation. They may not substantially relieve labor because much of the tax burden is passed on in higher prices. In unionized labor markets, higher prices may translate into higher wage claims, higher gross wages, and, eventually, lower labor demand (OECD 1995: 95). In addition, green taxes are likely to create distributive inequities because, as many studies show, energy consumption is higher among low-income households (Piekkola 1998: 90). Finally, to the extent that green taxes affect the competitiveness of domestic producers, they may be vulnerable to tax competition. There is concern that the uncoordinated introduction of charges on the environment may result in the closure or relocation of production facilities and in labor shedding (OECD 1995: 78-79) - a horrifying scenario which has induced many governments to go slow on green taxes and wait for international coordination (see e.g. Europäische Kommission 1994: 166).

Given the difficulties of shifting the tax burden to consumption or the environment, it appears plausible that, in the absence of tax competition, governments would have used additional tax receipts from corporate and financial income to reduce labor taxation. Indeed, taxing capital more heavily has often been promoted as a way of correcting a perceived bias against labor in the tax system. This was, for example, a theme of the corporate tax reforms in the UK in 1984-86, and the introduction of the Generalized Social Contribution in France in 1991, which extends social charges to non-labor income including interest, dividends, and capital gains (Blotnicki and Heckly 1998: 93). It was one of the reasons why the Danish Tax Reform Committee of 1992 suggested progressive rather than proportional taxation of capital income (Sorensen 1998: 23). It was discussed in Belgium in connection with a report on the financing of the social security system, and it was a core element of the tax reform plan of Germany's former minister of finance, Oskar Lafontaine (OECD 1995: 72; Lafontaine 1998: 5). Finally, it has been used by the European Commission as an argument to push for tax harmonization. In order to increase the taxation of capital, tax competition first must be stopped (Europäische Kommission 1989; European Commission 1996a).

6 Taxing dilemmas

The contest between the conventional wisdom and its critics ends in a draw: The critics have the better data, but the conventional wisdom has the better intuition. The critics are right in pointing out that there is no evidence for a race to the bottom in capital taxation or a meltdown of total tax revenues. But they are wrong in concluding that tax competition does not constrain national tax policy. Even Garrett admits that "income-based tax evasion is a significant problem in the global economy" (Garrett 1998b: 155) and acknowledges "a strong tax-based constraint on the future of the welfare state" (Garrett 1998b: 153). Unfortunately, he does not explain how this fits into his "much more optimistic ... picture of the prospects for national autonomy in the global economy" (Garrett 1998c: 94). The conventional wisdom is correct in assuming that globalization constrains national tax policy. Indeed, it is hard to see how it should not. How should the elimination of barriers to cross-border movements not lead to more international tax arbitrage and evasion, and how should this not increase the difficulties of taxing mobile tax bases? The conventional wisdom is wrong, however, when it assumes that arbitrage pressures will automatically lead to a race to the bottom in taxation.

The race to the bottom has not taken place because tax competition was not the only problem that the welfare state faced during the 1980s and 1990s. There was also slow growth, unemployment, high levels of precommitted budget expenditure and a mounting public debt. These problems, which were not directly related to globalization, exerted countervailing pressures on tax policy that eclipsed the effects of tax competition (Ganghof 1999; Steinmo and Swank 1999). The welfare state is not trapped in a race to the bottom, but boxed in between external pressures to reduce the tax burden on capital, on the one hand, and internal pressures to maintain revenue levels and relieve the tax burden on labor, on the other. There is no policy that can offer an easy way out. Governments can try to reduce their exposure to competition by cutting taxes on capital and by relying more on taxes on labor and consumption. But this may depress employment levels, encourage the growth of the shadow economy, and create equity problems. Alternatively, they can try to stimulate employment by reducing the tax burden on labor and consumption. But this implies higher taxes on capital and thus threatens to accelerate tax flight to other countries. Governments can take temporary refuge from this dilemma by running a larger deficit. Yet this merely postpones the problem. The taxes that are not raised today will lead to higher interest payments tomorrow. Less pain now causes more pain later. Whatever solution is chosen, it may backfire. If taxes on labor and consumption are raised too much, the welfare state may end up in what Gösta Esping-Andersen described as the welfare-state-without-work trap. If taxes on mobile capital are too high, they may drive investors and investment abroad. If budget deficits are too high, the ensuing interest payments will drive up spending requirements and make it ever more difficult to balance the budget.

Caught between high spending requirements and the risk of eroding the tax base of both capital and labor, many European states suffered from a sense of permanent fiscal failure during much of the 1990s. Something always seemed too high: the debt, the deficit, the taxes on capital, the taxes on labor. Yet it was impossible to reduce any of these factors without at the same time increasing one of the other factors. Tax reforms seemed urgent, but given high spending levels, there was little room for reform. Problems were shifted from one end of the tax system to another, but hardly solved (Ganghof 2000). Many reforms were inconsistent and unstable over time. France, for example, reduced the tax burden on corporations during the late 1980s to stimulate investment, only to introduce a surcharge on corporate income when budget consolidation became pressing during preparation for the monetary union. Germany reduced the personal income tax burden while at the same time increasing social security contributions (Manow and Seils 2000). Sweden reduced the progressivity of its income taxes during the 1991 tax reform, only to reintroduce progressivity during the mid-1990s (Steinmo 2000). There were a lot of tax reforms, but they made little difference in terms of total taxation and tax mix.

At the start of the new decade, fiscal prospects look much better. Higher growth rates and receding unemployment have reduced spending requirements and improved tax revenues. Countries as diverse as the United States, Ireland, Sweden or Denmark run budget surpluses, and even Germany and France slowly approach a balanced budget. The fiscal policy dilemmas and trade-offs of the 1990s are still around. But they seem to have lost much of their bite. Will the welfare state simply grow out of its fiscal predicament? That is possible if high growth persists and expenditure containment continues. The recent slowdown of the US economy does not augur well for the first of these requirements, and the prospects of demographic aging raise doubts about the second. When the first baby boomers reach retirement age around 2015, even tight expenditure controls may not be enough to prevent substantial increases in social spending. Some experts already predict a new fiscal crisis of the welfare state (Ferguson and Kotlikoff 2000). Then, at the latest, the concern about tax competition will grow again.

Literature

Andersson, Drister, and Leif Mutén, 1998: Sweden. In: Ken Messere (Ed.), The tax system in industrialized countries. Oxford: Oxford University Press, 327-353.

Benvignati, Anita M., 1985: An Empirical Investigation of International Transfer Pricing by US Manufacturing Firms. In: Alan M. Rugman and Lorraine Eden (Eds.), Multinationals and Transfer Pricing. London/ Sydney: Croom Helm, 193-211.

Bernauer, Thomas, 2000: Staaten im Weltmarkt. Zur Handlungsfähigkeit von Staaten trotz wirtschaftlicher Globalisierung. Opladen: Leske & Budrich.

Bird, Richard M., David B. Perry and Thomas A. Wilson, 1998: Canada. In: Ken Messere (Ed.), The tax system in industrialized countries. Oxford: Oxford University Press, 39-92.

Blotnicki, Laurence, and Christophe Heckly, 1998: France. In: Ken Messere (Ed.), The tax system in industrialized countries. Oxford: Oxford University Press, 93-127.

Brean, Donald J.S., Richard M. Bird and Melvyn Krauss, 1991: Taxation of international portfolio investment. Ottawa: Centre for Trade Policy and Law.

Castellucci, Laura, 1998: Italy. In: Ken Messere (Ed.), The tax system in industrialized countries. Oxford: Oxford University Press, 159-213.

Caves, Richard E., 1996: Multinational enterprise and economic analysis. Cambridge: Cambridge University Press.

CEPR, 1993: Making sense of subsidiarity: How much centralization for Europe? Brussels: CEPR.

Clayton, Richard and Jonas Pontusson, 1998: Welfare state retrenchment revisited. Entitlement cuts, public sector restructuring, and inegalitarian trends in advanced capitalist societies. In: World Politics 51, 67-98.

Cordes, Joseph J., 1996: How yesterday's decisions affect today's budget and fiscal options. In: C. Eugene Steuerle and Masahiro Kawai (Eds), The new world fiscal order. Implications for industrialized nations. Washington, D.C.: The Urban Institute Press, 95-116.

Daveri, Francesco, and Guido Tabellini, 2000: Unemployment, growth and taxation in industrial countries. In: Economic Policy 30, 47-104.

Dehejia, Vivek H., and Philipp Genschel, 1999: Tax competition in the European Union. In: Politics & Society 27, 403-430.

Deutsche Bundesbank, 1994: Aufkommen und ökonomische Auswirkungen des steuerlichen Zinsabschlags. In: Deutsche Bundesbank, Monatsbericht 46, 45-58.

Deutsche Bundesbank, 1997: Neuere Entwicklung der Steuereinnahmen. In: Monatsbericht 49, 83-104.

Dicken, Peter, 1998: Global shift. Transforming the world economy. London: Sage.

Esping-Andersen, Gösta, 1995: Welfare states without work: The impasse of labor shedding and familialism in Continental European social policy. Madrid: Centro de Estudios Avanzados en Ciencias Sociales, Working Paper 1995/71.

Europäische Kommission, 1989: Mitteilung der Kommission an den Rat: Steuerliche Maßnahmen, die die Gemeinschaft im Zusammenhang mit der Liberalisierung des Kapitalverkehrs zu treffen hat, Kom(89) 60/3/Revision endg.

Europäische Kommission, 1994: Wachstum, Wettbewerbsfähigkeit, Beschäftigung. Herausforderungen der Gegenwart und Wege ins 21. Jahrhundert, Weißbuch. Luxembourg: Office for Official Publications of the European Communities.

European Commission, 1996a: Taxation in the European Union. Brussels: European Commission.

European Commission, 1996b: Taxation in the European Union: Report on the Development of Tax Systems. Brussels: European Commission.

Eurostat, 1997: Structures of the taxation systems in the European Union, 1970-95. Luxembourg: Amt für amtliche Veröffentlichungen der Europäischen Gemeinschaften.

Ferguson, Niall, and Laurence J. Kotlikoff, 2000: The degeneration of EMU. In: Foreign Affairs 79, 110-121.

Frey, Bruno S., 1990: Intergovernmental tax competition. In, Influence of tax differentials on international competitiveness. Proceedings of the VIIth Munich Symposium on international taxation. Deventer: Kluwer, 89-98.

Gammie, Malcolm, and Claudio Radaelli, 2000: The future of tax policy in the EU. From 'harmful' tax competition to EU corporate tax reform. Brussels: CEPS.

Ganghof, Steffen, 1999: Adjusting national tax policy to economic internationalisation: strategies and outcomes. Cologne: Max Planck Institute for the Study of Societies Discussion Paper 99/6.

Ganghof, Steffen, 2000: Corporate tax competition, budget constraints, and the new trade-offs in domestic tax policy. Germany and Italy in comparative perspective. Cologne: Max Planck Institute for the Study of Societies, mimeo.

Gardner, Edward H., 1992: Taxes on capital income. A survey. In: George Kopits (Ed.), Tax harmonization in the European Community: Policy issues and analysis. Washington: IMF, 52-71.

Garrett, Geoffrey, 1995: Capital mobility, trade, and the domestic politics of economic policy. In: International Organization 49, 657-687.

Garrett, Geoffrey, 1998a: Global markets and national politics: Collision course or virtuous circle? In: International Organization 52, 787-824.

Garrett, Geoffrey, 1998b: Partisan politics in the global economy. Cambridge: Cambridge University Press.

Garrett, Geoffrey, 1998c: Shrinking states? Globalization and national autonomy in the OECD. In: Oxford Development Studies 26, 71-97.

Grigat, Hans-Goerg, 1997: Verlagerung von Unternehmensgewinnen in das Ausland und Steuerdumping. In: WSI Mitteilungen 50, 404-414.

Grubert, Harry, Timothy Goodspeed and Deborah Swenson, 1993: Explaining the Low Taxable Income of Foreign-Controlled Companies in the United States. In: Alberto Giovanni et al. (Eds), Studies in International Taxation. Chicago: University of Chicago Press, 237-275.

Hagen, Kare P., Erik Norrman and Peter B. Sorensen, 1998: Financing the Nordic welfare states in an integrating Europe. In: Peter B. Sorensen (Ed.), Tax policy in the Nordic countries. Houndmills: Macmillan, 138-203.

Hallerberg, Mark, und Scott Basinger, 1998: Internationalization and changes in tax policy in OECD countries. The importance of domestic veto players. In: Comparative Political Studies 31, 321-352.

Hemerijck, Anton C., and Martin Schludi, 2000: Sequences of policy failures and effective policy responses. In: Fritz W. Scharpf and Vivien A. Schmidt (Eds), Welfare and work in the open economy. Vol. 1: From vulnerability to competitiveness. Oxford: Oxford University Press, 125-228.

Hines, James R., 1999: Lessons from behavioral responses to international taxation. In: National Tax Journal 52, 305-322.

Hines, James R., and Eric M. Rice, 1994: Fiscal paradise: Foreign tax havens and American business. In: Quarterly Journal of Economics 109, 149-183.

Ishi, Hiromitsu, 1993: The Japanese tax system. Oxford: Oxford University Press.

Jacobs, Otto H., and Christoph Spengel, 1997: Entwicklung der Steuerbelastung der DAX-Unternehmen von 1989-1994. Mannheim: Universität Mannheim, mimeo.

Jacobs, Otto H., and Christoph Spengel, 1998: Steuerbelastung in Deutschland nach Unternehmensgröße und im internationalen Vergleich - eine empirische Analyse. Mannheim: Universität Mannheim, mimeo.

Kawai, Masahiro, and Yusuke Onitsuka, 1996: Fiscal policy, global saving and investment, and economic growth. In: C. Eugene Steuerle and Masahiro Kawai (Eds), The new world fiscal order. Implications for industrialized nations. Washington, D.C.: The Urban Institute Press, 171-222.

Kirchgässner, Gebhard, and Werner W. Pommerehne, 1996: Tax harmonization and tax competition in the European Union: Lessons from Switzerland. In: Journal of Public Economics 60, 351-371.

Kramer, Helmut, 1998: Economic aspects of tax co-ordination in the EU. In: Austrian Federal Ministry of Finance (Ed.), Tax competition and co-ordination of tax policy in the European Union. Vienna: Austrian Federal Ministry of Finance, 51-62.

Lafontaine, Oskar, 1998: Globalization and international cooperation - social democratic policy in the age of globalization. In: Dieter Dettke (Ed.), The challenge of globalization for Germany's social democracy. New York: Berghahn, 1-10.

Lafontaine, Oskar, and Dominique Strauss-Kahn, 1999: Europa - sozial und stark. In: Die Zeit 54, 17-18.

Leibfritz, Willi , John Thornton and Alexandra Bibbee, 1997: Taxation and economic performance. Paris: OECD Economics Department Working Paper No. 176.

Manow, Philip, and Eric Seils, 2000: Adjusting badly: The German welfare state, structural change, and the open economy. In: Fritz W. Scharpf and Vivien A. Schmidt (Eds), Welfare and work in the open economy. Vol. II: Diverse responses to common challenges. Oxford: Oxford University Press, 264-307.

Maystadt, Philippe, 1994: EMU: The Tax Challenges Ahead. In: EC Tax Review 2, 2-3.

Mendoza, Enrique G. , Giancarlo Milesi-Ferreti and P. Asea, 1997: On the ineffectiveness of tax policy in altering long-run growth: Harberger's superneutrality conjecture. In: Journal of Public Economics 66, 99-126.

Mendoza, Enrique G., Assaf Razin and Linda L. Tesar, 1994: Effective tax rates in macroeconomics. Cross-country estimates of tax rates on factor incomes and consumption. In: Journal of Monetary Economics 34, 297-323.

Messere, Ken C., 1993: Tax policy in OECD countries. Choices and conflicts. Amsterdam: IBFD Publications.

Messere, Ken C., 1997: OECD tax developments in the 1990s. In: IBFD Bulletin, 298-313.

Neumark, Fritz, 1948: Internationale Gemeinsamkeiten und nationale Eigenheiten der Finanzpolitik. In: KYKLOS 2, 317-348.

OECD, 1994: Taxation and household saving. Paris: OECD.

OECD, 1995: Taxation, employment and unemployment. Paris: OECD.

OECD, 1996: Economic outlook 59. Paris: OECD.

OECD, 1998a: Economic outlook 64. Paris: OECD.

OECD, 1998b: Economic outlook 63. Paris: OECD.

OECD, 1998c: Harmful tax competition. An emerging global issue. Paris: OECD.

Owens, Jeffrey, 1993: Globalisation: the implications for tax policies. In: Fiscal Studies 14, 21-44.

Perez-Navarro, Grace L., 1999: Overcoming fiscal borders in the global economy. In: OECD Observer 215, 18-21.

Piekkola, Hannu, 1998: Reducing the tax burden on labor and how to finance it. In: Austrian Federal Ministry of Finance (Ed.), Tax competition and co-ordination of tax policy in the European Union. Vienna: Austrian Federal Ministry of Finance, 77-95.

Pierson, Paul, 1996: The new politics of the welfare state. In: World Politics 48, 143-179.

Pinto, Carlo, 1999: (Harmful) tax competition within the European Union: concept and overview of certain tax regimes in selected Member States. http://www.lof.nl/artikel/.

Quinn, Dennis, 1997: The correlates of change in international financial regulation. In: American Political Science Review 91, 531-551.

Radaelli, Claudio, 1999: EU business taxation: fighting for competitiveness or combating tax competition? In: Thomas C. Lawton (Ed.), European industrial policy and competitiveness. Houndsmill: Macmillan, 114-135.

Radaelli, Claudio M., 1998: Policy narratives in the European Union: The case of harmful tax competition. Florence: European University Institute Working Paper RSC No. 98/34.

Ratzinger, Jürgen, 1997: Perspektiven für die Umsatzbesteuerung in der Europäischen Union. In: Wirtschaftsdienst 77, 464-471.

Robson, Mark, 1996: Deficits, tax policy, and saving. In: C. Eugene Steuerle und Masahiro Kawai (Eds), The new world fiscal order. Implications for industrialized nations. Washington, D.C.: Urban Institute Press, 197-222.

Rosen, Sherwin, 1996: Public Employment and the Welfare State in Sweden. In: Journal of Economic Literature 34, 729-740.

Ruding Report, 1992: Report of the committee of independent experts on company taxation. Luxembourg: Office for Official Publications of the European Communities.

Sandford, Cedric, 1993: Successful Tax Reform: Lessons from an Analysis of Tax Reform in Six Countries. Bath: Fiscal Publishers.

Scharpf, Fritz W., 1997: Introduction: the problem-solving capacity of multi-level governance. In: Journal of European Public Policy 4, 520-538.

Scharpf, Fritz W., 2000: The viability of advanced welfare states in the international economy: Vulnerabilities and options. In: Journal of European Public Policy 7, 190-228.

Schlegelmilch, Kai (Ed.), 1999: Green budget reform in Europe. Countries at the forefront. Berlin: Springer.

Schlesinger, Helmut, 1985: Zur Besteuerung von Kapitalerträgen. In: Dieter Cansier and Dietmar Kath (Eds), Öffentliche Finanzen Kredit und Kapital. Berlin: Duncker & Humblot, 237-253.

Schneider, Friedrich, and Dominik H. Enste, 2000a: Shadow economies: Size, causes, and consequences. In: Journal of Economic Literature 38, 77-114.

Schneider, Friedrich, and Dominik H. Enste 2000b: Schattenwirtschaft und Schwarzarbeit. Umfang, Ursachen, Wirkungen und wirtschaftspolitische Empfehlungen. Munich: Oldenbourg.

Schreiber, Ulrich, 1998: Internationale Mobilität und Steuerbasen - Ist nationale Steuerpolitik möglich? In: Gerold Krause-Junk (Ed.), Steuersysteme der Zukunft. Berlin: Duncker & Humblodt, 29-88.

Schulze, Günther G., and Heinrich W. Ursprung, 1999: Globalization of the Economy and the Nation State. In: The World Economy 21.

Schuster, Helmut, 1999: Die österreichische Abgeltungssteuer - ein Modell für Deutschland? Cologne: Bundesverband deutscher Banken.

Sinn, Hans-Werner, 1990: Tax harmonization and tax competition in Europe. In: European Economic Review 34, 489-504.

Sinn, Hans-Werner, 1997: The selection principle and market failure in systems competition. In: Journal of Public Economics 66, 247-274.

Slemrod, Joel, 1995: Comments. In: Vito Tanzi, Taxation in an integrating world. Washington, D.C.: Brookings, 141-146.

Sørensen, Peter Birch, 1992: The effects of taxation on international investment and economic efficiency. In: Ruding Report, Report of the committee of independent experts on company taxation. Luxembourg: Office for Official Publications of the European Communities, 313-359.

Sørensen, Peter Birch, 1998: Recent innovations in Nordic tax policy: From the global income tax to the dual income tax. In: Peter Birch Sorensen (Ed.), Tax policy in Nordic countries. Houndmills: Macmillan, 1-27.

Steinmo, Sven, 1993: Taxation and democracy: Swedish, British and American approaches to financing the modern state. New Haven: Yale University Press.

Steinmo, Sven, 1994: The end of redistribution? International pressures and domestic tax policy choices. In: Challenge 37, 9-17.

Steinmo, Sven, 1996: The new political economy of taxation: international pressures and domestic policy choices. Berkeley: University of California, Center for Western European Studies Working Paper 19.

Steinmo, Sven, 2000: Bucking the trend? Social Democracy in a global economy: The Swedish case up close. Boulder: University of Colorado, mimeo.

Steinmo, Sven, and Duane Swank, 1999: The new political economy of taxation. Boulder: University of Colorado, mimeo.

Stephens, John D., Evelyne Huber and Leonard Ray, 1999: The welfare state in hard times. In: Herbert Kitschelt, Peter Lange, Gary Marks and John D. Stephens (Eds), Continuity and change in contemporary capitalism. Cambridge: Cambridge University Press, 164-193.

Sturm, Jan-Egbert, 1998: Public capital expenditure in OECD countries. The causes and impact of the decline in public capital spending. Cheltenham: Edward Elgar.

Sunley, Emil M. , and Janet G. Stotsky, 1998: The United States. In: Ken Messere (Ed.), The tax system in industrialized countries. Oxford: Oxford University Press, 377-430.

Swank, Duane, 1998: Funding the welfare state: globalization and the taxation of business in advanced market economies. In: Political Studies 46, 671-692.

Tanzi, Vito, 1995: Taxation in an integrating world. Washington, D.C.: Brookings.

Tanzi, Vito, 1998: Globalization, tax competition and the future of tax systems. In: Gerold Krause-Junck (Ed.), Steuersysteme der Zukunft. Berlin: Duncker & Humblodt, 13-27.

Tanzi, Vito, and A. Lans Bovenberg, 1990: Is there a need for harmonizing capital income taxes within EC countries? In: Horst Siebert (Ed.), Reforming capital income taxation. Tübingen: Mohr, 171-197.

Tanzi, Vito, and Ludger Schuknecht, 2000: Public spending in the 20th century. A global perspective. Cambridge: Cambridge University Press.

Tanzi, Vito, and Parthasarathi Shome, 1993: A primer on tax evasion. In: IMF Staff Papers 40, 807-828.

Tullio, Giuseppe, 1987: Long run implications of the increase in taxation and public debt for employment and economic growth in Europe. In: European Economic Review 31, 741-780.

Valenduc, Christian, 1994: Tax havens and fiscal degredation in the European Community. In: EC Tax Review 1, 20-25.

Vanistendael, Frans J., 1988: Trends of tax reform in Europe. In: Australian Tax Forum 5, 133-179.

Wagschal, Uwe, 2001: Deutschlands Steuerstaat und die vier Welten der Besteuerung. In: Manfred G. Schmidt (Ed.), Wohlfahrtsstaatliche Politik: Institutionen - Prozesse - Leistungsprofil. Opladen: Leske & Budrich.

Webb, Michael C., 1998: Global markets and state power: Explaining the limited impact of international tax competition. Victoria, B.C.: University of Victoria, mimeo. 

Webber, Carolyn , and Aaron Wildavsky, 1986: A history of taxation and expenditure in the Western World. New York: Simon and Schuster.

Weichenrieder, Alfons, 1996: Fighting International Tax Avoidance: The Case of Germany. In: Fiscal Studies 17, 37-58.

Wuntsch, Michael von, 1998: Internationale Steuerplanung im Zeitalter der Globalisisierung. In: WSI Mitteilungen 51, 735-746.

Endnotes

1 On the evolution of the modern tax system see, for example, Neumark 1948; Webber 1986; Steinmo 1993.

2 The 16 OECD countries are Austria, Australia, Belgium, Canada, Germany, Denmark, France, Finland, Italy, Ireland, Japan, Norway, the Netherlands, Sweden, United Kingdom, and the United States of America.

3 For a thorough analysis of tax structure developments in OECD countries, see Messere 1993; Wagschal 2001.

4 OECD simulations suggest that the effect of unemployment on the fiscal balance is very strong. In some countries, including Italy, Belgium, and Sweden, an increase in unemployment by one percentage point is estimated to deteriorate the fiscal balance by as much as roughly one percentage point of GDP (OECD 1998b).

5 Significant fiscal consolidation is defined as an improvement in the general government structural (i.e. cyclically-adjusted) fiscal balance that is equivalent to at least three percentage points of GDP, which takes place continuously over consecutive years (two years minimum).

6 On the problem of profit shifting, see also: Caves 1996; Radaelli 1999: 121.

7 Note that an effective tax rate of 20 percent is rather low if compared with the average effective/implicit tax rates that tables 1 and 2 report for capital taxation in general.

8 As Tanzi shows, even a combination of moderate inflation rates and moderate personal income tax rates can translate into an effective marginal tax rate on real interest income close to or in excess of 100 percent (Tanzi 1995: 123-130).

9 Technically, withholding taxes are prepayments on the income tax that are withheld at the source of the income. In case of financial income, this is often the bank, which pays out interest or dividends.

10 Daveri and Tabellini's finding is controversial, however (Bertolila in Daveri and Tabellini 2000: 89-92).

11 If workers perceived a direct link between their social security contributions and future benefits, the contributions would have the character of insurance premiums rather than taxes. This might induce workers to accept real-wage reductions in exchange for increased contribution payments because they feel that they get something in return (Esping-Andersen 1995: 8). However, this is unlikely to be the case in many OECD countries. Usually, the link between contributions at the margin and future benefit levels is unclear, and contributions are regarded as income taxes rather than insurance premiums (Leibfritz et al. 1997: 34).

12 Technically, Sweden did not cut the tax rate as such, but reduced the VAT tax base to which the tax rate is applied – by 30 percent and 80 percent, respectively.

13 See contributions in Schlegelmilch (1999).

Copyright © 2001 Philipp Genschel

No part of this publication may be reproduced or transmitted without permission in writing from the author. 
Jegliche Vervielfältigung und Verbreitung, auch auszugsweise, bedarf der Zustimmung des Autors.

MPI für Gesellschaftsforschung, Paulstr. 3, 50676 Köln, Germany