Swiss Confederation

1.3 million, 1,728.9 square kilometres, 39,319 (Zurich)

14,700, 172.5 square kilometers, 26,873 (Appenzell Inner-Rhodes)

remarkably different. It has four official national languages along with the corresponding cultures: 63.7 percent speak (Swiss) German; 20.4 percent, French; 6.5 percent, Italian; and 0.5 percent, Romansch. The remaining 9 percent speak various other languages.³ Because German and French are the dominant languages, public officials are expected to understand both.

Language is not the only area of diversity. The Swiss population also adheres to different religions: 41.8 percent are Roman Catholics; 35.3 percent are Protestants; 4.3 percent are Muslims; 3.4 percent are adherents of other religions; and 15.4 percent have no religion or did not answer the corresponding question. Moreover, 20 percent of the population are foreigners.⁴ This is a higher percentage than is found in any other country in Europe (except some very small countries such as Monaco). Thus, Switzerland, as a nation, lacks cultural, linguistic, and religious homogeneity; it is a nation shaped by the resolve of its citizens, a “nation of will” (Willensnation), and it is well aware of its many diversities.

This holds for the modern federal system, which was founded in 1848 but has roots going back to the thirteenth century. Although there was a first treaty between the three primary cantons – Schwyz, Uri, and Unterwalden – in 1273, the official founding date of Switzerland is 1291. However, the Swiss nation was strongly affected by Napoleon and came together in 1848 after a short civil war between the Protestants and Catholic separatists, the Sonderbundkrieg, in 1847. That Switzerland has survived and did not split up according to linguistic divisions in the second half of the nineteenth century (when its neighbours, Italy and Germany, created their national states) is presumably due to its rather decentralized federal structure. The other key ingredients that constitute the Swiss nation are its direct democracy and its political neutrality in international affairs.

The cantons vary greatly in size and in population density. The average canton has about 280,000 people, but population sizes range from 14,700 in Appenzell Inner-Rhodes to 1.3 million in Zurich. The average population density is 178 people per square kilometre. Compared with some other European countries, such as Belgium and the Netherlands (as well as with the United Kingdom and Germany), this might not seem to be very high. However, Switzerland’s internal differences are quite large. In Zurich, the population density is 723 persons; in Uri, a canton consisting almost entirely of mountains and valleys, it is thirty-six persons per square kilometer.⁵ Basle-Town, with 190,700 inhabitants in its thirty-seven square kilometres, has a population only a little under that of the geographically largest canton, Grisons, whose 191,200 inhabitants are spread across 7,106 square kilometres and 150 valleys. Thus, some areas are rather densely populated, especially in the “Mittelland,” a relatively narrow tract that runs from Lake Geneva to the Lake of Constance. North and west of the Mittelland are the Jura Mountains, to the south and east are the Pre-Alps and the Alps. Large parts of these mountain areas are unproductive and, correspondingly, quite sparsely populated (e.g., the canton of Uri).

The record of Swiss economic growth during the last fifteen years has been rather bad; Switzerland has had the lowest average growth rate of all oecd countries. Nevertheless, it is still a rather rich country. Based on current exchange rates, Switzerland lags behind Luxembourg and Norway as the third richest country in the world, but it is ahead of the United States, Denmark, and Ireland. Based on purchasing power, its record is not so splendid, as gross domestic product (gdp) per capita is higher in the United States and Ireland, and gross national investment per capita is higher in the United States. Even so, ranking fourth or fifth in the world is still very good.⁶

Inside Switzerland, however, there are strong economic discrepancies. In 2002, gross national product (gnp) per capita was us$49,774 in Zug but us$23,188 in the Jura.⁷ The discrepancies are even stronger with respect to gdp. In 2001, the average Swiss gdp per capita was us$37,456, with the canton Basle-Town being 106 percent above, and the canton Appenzell Inner-Rhodes being 47 percent below, the national average.⁸ Such large discrepancies create tensions in the fiscal system.

the structure of government and the division of fiscal power

The basic constituents of Switzerland are the twenty-six (twenty-five) cantons that founded the federal state in 1848.⁹ Article 3 of the federal Constitution states that they “are sovereign insofar as their sovereignty is not limited by the federal Constitution; they shall exercise all rights which are not transferred to the Confederation.” The Confederation has authority only in those areas in which it is empowered by the federal Constitution (e.g., foreign affairs, defence, customs, and monetary policy). Tasks that do not expressly fall within the scope of the Confederation are handled by the cantons.

Each canton and half-canton has its own constitution, parliament, government, and courts.¹⁰ The canton parliaments have between fifty-eight and two hundred seats, while the canton governments have five, seven, or nine members. All of these are directly elected by the people at the ballot box, with the exception of the canton Appenzell Inner-Rhodes, where they are elected by the canton general assembly (Landsgemeinde), which convenes every year in April.

All the cantons are divided into municipalities, of which there are at present 2,760. Their number is tending to diminish as some municipalities are merging. Around one-fifth of these municipalities have their own parliament; in the other four-fifths, decisions are taken by direct democracy in a local assembly. In addition to the tasks entrusted to them by the Confederation and the canton – such as the population register and civil defence – local authorities also have specific responsibilities of their own with regard to, inter alia, education and social welfare, energy supply, road building, and local planning. To a large extent, these powers are self-regulated. The scope of municipal autonomy is determined by the individual cantons and, therefore, varies widely. However, given the autonomy for local government provided in the canton constitutions, neither the cantons nor the federal authorities have the right to interfere with local decisions. The only exception occurs when the financial situation of a local community deteriorates seriously. When this happens, the local budget has to be approved by the canton government.

The federal Constitution assigns specific tasks to the Confederation; all other matters are canton responsibilities.¹¹ However, there are many shared responsibilities. This applies to situations in which the federal government has legislative responsibility but the cantons execute the legislation (e.g., motorways are built by the cantons as ordered and financed by the federal government). However, joint responsibility applies also to areas such as education. This is a canton task, which means that the cantons are, in theory, responsible for all the universities. However, there are two federal universities: the two Swiss Federal Institutes of Technology in Zurich and Lausanne. Moreover, research is a federal task, which is mainly financed by the Swiss National Science Foundation located in Berne. This foundation also finances research conducted in the canton universities. Further, the federal government gives subsidies to the canton universities, depending on the number of students enrolled. Finally, the federal government tries to develop a consistent national university policy, although the canton ministers of education and, therefore, the canton governments have to agree to it. Nevertheless, due to its large role in financing universities, the federal government has a large impact on decisions that are formally in the domain of the cantons.

University policy reveals another specific trait of the Swiss system. The majority of the cantons are far too small to have their own university. In addition to the two federal institutes, ten cantons have universities, all of which are public, and students’ fees are rather low. To finance the universities, the cantons have agreed among themselves that the canton from which a student comes has to pay a certain amount of money to the canton in which he or she is studying. So part of the financial burden of the universities is spread across the country without the intervention of the federal government. Such inter-canton cooperation is to be enforced by the new system of fiscal equalization described below.

The possibilities for the national government to interfere with canton or local policies are quite limited. This creates problems, as there are many areas in which the Confederation is in charge of strategic (and legislative) matters but the cantons are in charge of operational or executive matters. If the cantons do not behave appropriately, the federal government has hardly any means to force cooperation. This is relevant, for example, in environmental policy. By way of illustration, if the Confederation sets limits for the emission of some pollutants, and if these limits are violated in certain areas, there is hardly anything the federal government can do about it. It cannot, for example, force the canton governments by withholding grants or revenues from tax sharing or by any other fiscal means. This situation is known as the “implementation deficit” (Vollzugsdefizit) and is seen as a problem in Swiss politics.

The other side of the coin is that the cantons (and local communities) do not have much say in national politics; the national government and Parliament are free to take decisions – subject always to the will of the people. Whenever the Constitution is to be changed (a question that usually comes up several times a year), the people have to be asked to approve the change. Furthermore, whenever there is a new law or a change in a law, Swiss citizens only have to collect 50,000 signatures to get a referendum.

Table 2 Legislative responsibility and actual provision of services by different orders of government

Function	Legislative Responsibility	Execution
International relations	Federal	Federal
Defence	Federal	Federal
Monetary policy, customs	Federal	Federal
Postal services, telecom, mass media	Federal	Federal
Railway, aviation	Federal	Federal
Atomic energy	Federal	Federal
Water power	Federal	Federal
National roads	Federal	Canton
Trade, industry, labour legislation	Federal	Canton
Agriculture	Federal and canton	Federal and canton
Civil and criminal law	Federal and canton	Federal and canton
Police	Canton	Canton
Churches	Canton	Canton
Education (secondary schools, universities)	Canton	Canton
Taxes	Federal and canton	Federal and canton
Social security	Federal	Canton
Environmental policy	Federal	Canton
Cantonal roads	Canton	Canton
Local roads	Local	Local
Local public transport (in cities)	Local	Local
Local gas, electricity and water supply, waste	Local	Local
Primary schools	Local	Local
Public care	Local	Local

Source: Wolf Linder, Schweizerische Demokratie: Institutionen, Prozesse, Perspektiven (Berne: Haupt, 1999), 140; with reference to Jürg M. Gabriel, Das politische System der Schweiz (Berne: Haupt, 1990), 97.

Table 3 Direct expenditures by function and level of government (percentages)

Function	Federal	Canton	Local	All
Defence	90.8	4.8	4.4	100
Debt servicing	59.6	23.0	17.4	100
General administration	20.7	38.3	41.0	100
Law and order	9.3	65.7	25.0	100
Economic services	41.1	36.1	22.8	100
Social services	20.6	48.3	31.0	100
Health	1.0	57.8	41.2	100
Education	13.7	54.3	32.0	100
Subsidies	41.7	35.4	22.9	100
Total	31.4	41.6	27.0	100
Local public services	5.2	51.5	43.2	100

Switzerland does not have a constitutional court with the power to declare that a law passed by Parliament violates the federal Constitution; the Federal Supreme Court in Lausanne functions as a constitutional court only for the cantons (i.e., it has the power to declare canton laws unconstitutional). There is, of course, the upper chamber of Parliament, which was modelled after the us Senate, and in which each canton has two members. These are representatives of the people of the canton, not of its government or parliament. That is why conferences comprised of the presidents and also of other members of the canton governments (e.g., those who are responsible for education) have been created. Such conferences have two purposes: first, they decide about arrangements between the cantons in areas in which the cantons have sole responsibility but need some coordination; second, they represent the interests of the cantons in political discourse in the national arena.

If, in making a new law, the Confederation interferes too much with the cantons’ interests, the cantons can launch a referendum if at least eight of them demand it. The first such referendum was held in May 2004, when the cantons opposed a tax reform that would have changed the base for the taxation of owner-occupied houses and apartments. The referendum was successful, and the reform was rejected. Since then, the federal finance minister has been much more hesitant to mingle with issues that touch canton interests.

fiscal federalism and macroeconomic management

Monetary policy is a strictly federal issue, although, in practice, the responsibility for this is delegated to the Swiss National Bank (snb), which is located in Zurich and Berne. The snb has been quite independent for some time – a situation that was recognized by the Swiss central bank law, which became effective in 2004. The snb’s main objective is to ensure price stability, but “in so doing, it shall take due account of the development of the economy.”¹² After the breakdown of the Bretton Woods system in the 1970s, snb policy, until 1999, focused on the quantity of money. However, as this was considered to be one of the factors contributing to the low growth of the Swiss economy in the second half of the 1990s, the snb’s strategy since 2000 has been to attempt to keep the rate of inflation between zero and 2 percent.

The snb’s main goal, price stability, is generally conceded to have been very effectively achieved. Switzerland has one of the most stable currencies in the world. From 1980 to 2004, the average inflation rate was 2.29 percent, compared with 2.23 percent in Germany and 2.93 percent in the United States, two other countries whose central banks also have a very high respect for price stability. This is also reflected in the development of the exchange rate. Since 1974, when the Swiss franc began floating against all other currencies, there was an appreciation against the German mark of nearly 40 percent until 1998 and an appreciation against the American dollar of about 140 percent until 2004.

As Switzerland is a small country, the scope for fiscal policy is limited, although it is exclusively a matter for the federal government. Even large federal deficits, as seen in the 1990s, hardly provide an impulse to the Swiss economy. Quite recently, there have been some tentative attempts to create some coordination between the federal government and the can-tons,¹³ but these did not go farther than initial discussions. Indeed, considering the great budgetary independence of the cantons, it is hardly conceivable that there will be effective coordination.

Every canton is responsible for its own fiscal discipline. In 1981, the conference of the canton ministers of finance edited the Handbook of Public Budgeting (Vol. 1),¹⁴ which contains a model for canton budgets. According to Article 2, the principle of a balanced budget has to be observed. This is stated more concretely in Article 4, according to which the current budget has to be balanced in the medium term, and in Article 18, which demands that canton-accumulated debt has to be cut back in the medium term, whereby “medium term” means within about ten years. Today, such rules can be found in nearly all canton constitutions and in the corresponding budget laws. The cantons are obliged to balance their budgets over the business cycle and also to cut down accumulated debt. All this did not prevent canton debt from increasing considerably during the 1990s, partly because of the low economic growth rate. Canton real debt increased by about 106 percent (in real terms) from 1990 to 2002, but the development varied from canton to canton.¹⁵ For example, St Gall and Fribourg showed only modest increases in their public debt over that period; Vaud’s debt increased considerably; and Geneva’s increased dramatically. In 2002, Geneva’s a public debt per capita was us$26,865, which was 418 percent of the average canton debt in the country.

fiscal referenda

A special feature of the canton constitutions, in contrast to the Swiss federal Constitution, is the existence of a fiscal referendum.¹⁶ If it is mandatory and the outlay for it exceeds a certain limit, the canton’s citizens have to be asked whether they agree to the project in question. This is also the case if the referendum is optional and if a given number of signatures are collected. The limit that is specified can be different for nonrecurring expenditures and for recurring expenditures. With the exception of Vaud, all cantons have such a referendum.¹⁷ Because the citizens know that sooner or later they will have to pay for the projects that are carried out by the canton or local community, this acts a restraint on overly ambitious projects.

Take St Gall as an example. There, fiscal referenda are optional for recurring expenditures of more than us$179,000 and for non-recurring expenditures of more than us$1.79 million. They are mandatory for recurring expenditures of more than us$893,000 and for non-recurring expenditures of more than us$8.93 million. In relation to the budget, which in 2002 was about us$2.06 billion, these limits amount to 0.009 percent or 0.043 percent, respectively, for current expenditures, and 0.09 percent or

0.43 percent, respectively, for non-recurring expenditures. These limits are rather low. In order for an optional referendum to take place, four thousand signatures have to be collected within a thirty-day period. This corresponds to about 1.5 percent of the electorate.¹⁸

However, the existence of the fiscal referendum combined with the regulations for a balanced budget were insufficient to prevent public debt from increasing. Therefore, partly because they had carried debts for longer periods than had the others, eight cantons introduced new instruments within the past ten years in order to limit deficits: St Gall (1994), Fribourg (1994), Solothurn (1995), Appenzell Outer-Rhodes (1995), Grisons (1998), Lucerne (2001), Berne (2002), and, the last one for the time being, Valais (2002).¹⁹ These regulations are, in some cases, in canton constitutions, although they are usually in canton budget laws.

St Gall may again be used as an example.²⁰ The rules require the current budget to be “balanced” – defined as a maximum permissible deficit of 3 percent of the “simple tax revenue,” which at the moment is about 60 percent of total tax revenue.²¹ Whenever a deficit is expected, the tax rate is supposed to be adjusted in order to keep within this limit. Moreover, if there are no savings available, the deficit is transferred to the budget of the year after the next year. In turn, whenever there is a surplus (e.g., because of an economic upswing), the money has to be saved and/or utilized for additional depreciations. Tax rates cannot be reduced if these savings do not amount to seven times the maximum allowable deficit. In addition to considerations regarding the current budget, there are considerations relating to the capital budget, which is used for financing public investment. The rule is that investment projects up to us$2.98 million have to be included in current budgets, while the debt principal of projects between us$2.98 and us$5.95 million has to be paid back within five years, and of projects above us$5.95 million within ten years. These depreciations (as well as the interest payments) have to be included in the current budgets. Thus, such projects cannot lead to a long-run debt increase. It is possible to raise debt in order to buy shares of firms (e.g., of the canton banks), but there must be returns as compensation.

Thus, the citizens of St Gall can – within the boundaries of the federal Constitution – authorize necessary expenditures in relation to the tasks the canton has to perform. With respect to revenue, they decide all constitutional and statutory rules, especially regarding the different taxes as well as the tariff scales (including the progression of the direct taxes) but not about the tax rates. The authority for the latter lies with the canton parliament, although that authority is restricted by the regulations described above, which oblige the canton to build up savings to a certain level before tax rates can be reduced. This rather unusual requirement means that surpluses are built up in “good” years and can be used to cover (to a certain extent) deficits created during “bad” years. This institutionalizes anti-cyclical fiscal policy in the cantons without leading to an increase in public debt.²² This is remarkable, as it is usually assumed that anti-cyclical fiscal policy can only be conducted successfully by the federal government; the medium and lower levels of government are generally thought to be pro-cyclical.²³ Swiss canton experience shows that, with the appropriate culture and institutions, this does not have to be the case.

The combination of direct democratic expenditure restrictions, quasi-automatic revenue adjustment, and the build-up of savings has proved to be successful. In 2002, for instance, St Gall’s public debt per capita was $us2,346; only Schwyz, Zug, Argovia, and the two Appenzells had lower public debt.

As mentioned above, today there are similar rules in seven other cantons. Solothurn and Grisons, for example, also accumulate savings in order to equalize revenue fluctuations over the business cycle. In Appenzell Outer-Rhodes, the rule is that no deficit is allowed in the budget once there is an accumulated deficit that exceeds 5 percent of the canton and local tax revenue budgeted for the current year.²⁴ This rule is intended to force the government to build up reserves in good times and to eliminate structural deficits.²⁵ While Fribourg also strives for a budgetary balance over the business cycle, the regulation is even stricter with respect to balancing the annual budget. The tax rate has to be increased as soon as the deficit in the proposal for the current budget exceeds 3 percent of total revenue.²⁶

The experience of these cantons has also been positive. In Fribourg, debt per capita rose from us$2,069 in 1990 to us$3,165 in 2002 – that is, it rose by only 46 percent (in real terms), far below the average for the Swiss cantons (92 percent). The other cantons that have had debt brakes for more than five years also performed well in this respect.

So much for the evidence of the available case studies.²⁷ An alternative to case evidence is provided by econometric studies that investigate whether cantons and local communities that use such instruments have, ceteris paribus, lower deficits and debts than do those of other cantons and local communities.²⁸ Investigations show that cantons with a fiscal referendum have significantly lower expenditure and revenues than do other cantons. However, because the reduction is stronger for revenue than for expenditure, the deficit is significantly higher. Correspondingly, the public debt is also higher, although the corresponding coefficient is not significantly different from zero. Fiscal constraint leads to somewhat lower expenditure but higher revenue. This leads to a significantly lower deficit and also to significantly lower debt.

Whenever a local community has a mandatory fiscal referendum that requires it to raise additional public debt, this leads to significantly lower expenditure and revenue and, especially, to lower public debt. The estimated impact on the deficit is also negative but is not statistically signif-icant.²⁹ As for the cantons, the debt brake leads to somewhat lower expenditure and somewhat higher revenue, which leads to a significantly lower deficit. The estimated impact on the local public debt is also negative, although not significantly so.

Thus, for the cantons as well as for the local communities, the combination of the fiscal referendum and the debt brake has a stabilizing effect on public finances. This does not necessarily lead to a lower tax ratio, but it does lead to a lower deficit and lower public debt. To that extent, the St Gall model can be seen as an example of institutionalized arrangements resulting in sound public finances, as described by the finance minister of this canton.³⁰

public revenue

A special feature of the Swiss fiscal constitution is the substantial autonomy of the cantons not only on the expenditure side but also on the revenue side of the budget.³¹ The main progressive taxes on personal and corporate income are state and local taxes. The cantons have the basic power to tax income, wealth, and capital. The municipalities can levy a surcharge on canton taxes. The federal government relies mainly on indirect (proportional) taxes, a value-added tax, and specific consumption taxes, such as the mineral oil tax. There is, however, a small but highly progressive federal income tax, which amounted to 29 percent of total federal tax revenue in 2002, while the cantons and municipalities rely on income and property taxes for about 46 percent of their total current revenue and 90 percent of their tax revenue. The federal income tax has a maximum marginal rate of 13.2 percent and an average rate of 11.5 percent. Owing to a basic tax exemption, the highest 3 percent of income taxpayers pay about 50 percent of the revenue of the federal income tax. Thirty percent of this tax is paid back to the cantons.³² The federal government can also rely on a source tax on income from interest, the so-called Verrechnungssteuer, which has a rate of 35 percent. While there is no federal or canton deductibility of personal income taxes paid to the cantons or localities, there is a tax deductibility in the case of corporate income taxes.

The tax burden varies considerably between the cantons and, in some cantons, also between the municipalities. In the index of the burden of personal taxes for the year 2003,³³ there were huge discrepancies between “rich” cantons, like Zug and Schwyz, and “poor” ones, like Obwalden, Uri, and Jura. It seems obvious that such discrepancies demand a fiscal equalization system. The current system is apparently not able to sufficiently equalize the situations. This is the reason a new system is to be implemented, which – it is hoped – will be effective from 2008 onward.

Strong tax competition takes place in Switzerland. Investigations show that the proportion of rich people in a canton is significantly influenced by the canton’s tax rate.³⁴ Choice of canton for residence often depends, for high-income earners, on the amount of income tax they would have to pay. Social transfers are less influential for the choice of residence. Thus, fiscal competition consists of tax competition rather than of transfer competition, and tax competition is stronger for self-employed than it is for dependent employees and retirees. Among local communities, the effects of tax competition are even stronger than they are for cantons.

From an international perspective, the Swiss tax burden is low compared with other locations in Europe and the United States. This holds even for the higher-tax Swiss cantons. In terms of the index of the effective average tax rate for people with an annual income of us$170,000,³⁵ the burden is considerably lower in the low-tax cantons.

Table 4 Tax assignments for various orders of government

Shares in revenue (percentages)

Determination Determination Collection and of base of rate administration Federal Canton Local

Federal

Direct federal income tax Federal Federal Withholding tax Federal Federal Capital transfer tax Federal Federal Value added taxes Federal Federal Excise taxes Federal Federal Import duties Federal Federal Traffic duties Federal Federal Agriculture duties Federal Federal Steering taxes Federal Federal Gambling house taxes Federal Federal Patents and concessions Federal Federal Fees for legal acts Federal Federal Hospital fees Federal Federal User fees, services Federal Federal Other Federal Federal Canton/ local 70.0 30.0 0.0 Canton/ local 90.5 9.5 0.0 Canton/ local 100.0 0.0 0.0 Canton/ local 100.0 0.0 0.0 Canton/ local 100.0 0.0 0.0 Canton/ local 100.0 0.0 0.0 Canton/ local 100.0 0.0 0.0 Canton/ local 100.0 0.0 0.0 Canton/ local 100.0 0.0 0.0 Canton/ local 100.0 0.0 0.0 Canton/ local 100.0 0.0 0.0 Canton/ local 100.0 0.0 0.0 Canton/ local 100.0 0.0 0.0 Canton/ local 100.0 0.0 0.0 Canton/ local 100.0 0.0 0.0

Canton

Personal income taxes Canton Canton/ local Canton/ local 0.0 54.9 45.1 Property taxes Canton Canton/ local Canton/ local 0.0 55.3 44.7 Corporate income taxes Canton Canton/ local Canton/ local 0.0 55.6 44.4 Capital taxes Canton Canton/ local Canton/ local 0.0 60.6 39.4 Real estate taxes Canton Canton/ local Canton/ local 0.0 26.5 73.5 Property gain taxes Canton Canton/ local Canton/ local 0.0 53.8 46.2 Property transfer taxes Canton Canton/ local Canton/ local 0.0 72.2 27.8 Heritage and gift taxes Canton Canton/ local Canton/ local 0.0 91.5 8.5 Motor vehicle taxes Canton Canton Canton 0.0 100.0 0.0 Amusement taxes Canton Canton/ local Canton/ local 0.0 31.8 68.2

Table 4 Tax assignments for various orders of government (Continued)

Shares in revenue (percentages)

Determination Determination Collection and of base of rate administration Federal Canton Local

Dog taxes

Other property and expense taxes Patents and concessions Fees for legal acts Hospital fees User fees, services Other

Table 5 Vertical fiscal gaps

Canton	Canton/ local Canton/ local	0.0	25.6	74.4
Canton	Canton/ local Canton/ local	0.0	76.0	24.0
Canton	Canton/ local Canton/ local	0.0	84.5	15.5
Canton	Canton/ local Canton/ local	0.0	76.4	23.6
Canton	Canton/ local Canton/ local	0.0	44.4	55.6
Canton	Canton/ local Canton/ local	0.0	26.5	73.5
Canton	Canton/ local Canton/ local	0.0	49.2	50.8

	Total revenue collected	Total revenue available	Expenditures
National	44,512,121	35,951,574	38,806,603
Subnational	74,278,429	82,838,976	84,893,416
Canton	45,729,651	49,771,186	51,447,276
Local	28,548,778	33,067,790	33,446,140
All orders	118,790,550	118,790,550	123,700,019

Corporate income taxes in Switzerland also vary considerably among the cantons. From anecdotal evidence, it is known that two tax havens are in or close to Switzerland – the small country of Liechtenstein, which forms an economic union with Switzerland, and Zug. Taking the value of the (weighted) average for Switzerland as 100, the index of the tax burden of corporate income and capital taxes varied in 2003 from 49.0 in Schwyz to 141.7 in Grisons.³⁶

Again, from an international perspective and compared with other oecd countries, Swiss taxes are comparatively low. In terms of the effective average tax rates,³⁷ only Ireland has average tax rates as low as Zug. Thus, with respect to personal as well as company taxation, at least some Swiss cantons have a very strong position in international tax competition.

By contrast, the Swiss system of corporate income taxation appears to be quite complicated, partly because of the canton and local competencies. All in all, capital may bear seven different taxes: the corporate income tax on profits, the capital tax, the federal source-tax on interest and dividend income, an emission charge, the property tax, the church tax, and – in some cantons – a minimum tax if revenue from corporate income tax does not reach a certain amount. Estimates place the taxation of profits and cap-ital-induced administrative costs at about us$8,300 per firm per year.³⁸ This amount is about 40 percent of the average administrative costs that small and medium-size firms bear due to public regulation and accounts for about 3 percent of their investment in equipment.

Three characteristics are fundamental to the taxation of corporate income:

1 In many cantons, the tax on profits follows a progressive tax schedule according to the rate of return on capital. The federal government has levied a proportional tax on corporate profits since the tax reform act of 1998. Seven cantons already had a proportional tax rate before that reform, while Geneva introduced it after the reform.

2 In addition to taxation of profits according to the rate of return on capital, capital is taxed separately by all cantons. In most cases, a proportional rate is used. The federal government abolished its tax on firms’ capital in 1998.

3 The Swiss corporate income tax is similar to the corporate income tax in the United States. Profits are taxed at the corporate level and again at the shareholder level as dividend income.

A particular feature of corporate income taxation in Switzerland is the fact that holding companies are taxed at lower rates or, in some cantons, not at all. This is to avoid double taxation of the profits of parent and affiliate companies. However, generous tax exemptions for holding companies provide incentives for profit shifting on the part of firms. Zug is supposed to owe its economic wealth to such a policy. In addition, nearly all cantons, with the notable exceptions of Zug and Argovia, have special tax holidays for “newly founded” firms. These holidays are restricted by the federal tax-harmonization law of 1993 to ten years at the most. “Newly founded” may mean anything from the construction of new firms or affiliates to the relocation of companies that, for years, have been located in other cantons.

Given the large differences in tax burdens among Swiss cantons, double-taxation agreements between cantons and profit allocation rules for firms with plants in different cantons play a non-negligible role. Between Swiss cantons, an exemption system is used exclusively. If, for example, a firm resides in Zurich and has a subsidiary in Zug, Zurich exempts the profits earned in Zug from taxation in Zurich. Moreover, profit allocation between both cantons is regulated by a kind of formula apportionment. There is no unique, harmonized formula apportionment rule for all cantons, and this leaves room for a large variety of such rules between cantons. Payroll, capital, or sales are used as a basis for calculating profit shares. For example, the profits of retail firms are usually allocated according to sales, while the profits of manufacturing firms are allocated according to capital and payroll (capitalized by 10 percent). Given that capital is taxed in addition to profits and that profits are taxed on the basis of rates of return on capital, it is not only profit allocation rules that are used but also capital allocation rules.

Due to the small size of the country and its subfederal units, corporate taxpayers can easily move to places with low tax burdens and respond to canton tax differentials accordingly. The exemption system basically provides an incentive for tax-induced relocation, while profit-sharing rules, in the sense of a kind of formula apportionment, reduce incentives for profit shifting. Moreover, the differences in canton legal and accounting systems are not so substantial as to render the relocation of firms difficult. All in all, firms may have sufficient fiscal incentives to relocate between cantons. In addition, formula apportionment might not work well enough to completely prevent the occurrence of profit shifting. Thus, tax competition for mobile capital in Switzerland may take place either through relocation of real capital, leading to subsequent changes in economic activity, or through profit shifting among cantons.

What are the effects of tax competition on the location of firms and on employment? Corporate and, in particular, personal income taxes have a stronger impact on the canton distribution of firms than they do on em-ployment.³⁹ This provides some indirect evidence for the existence of profit shifting among cantons and/or for smart tax management on the part of firms located in one canton only. This holds despite several incentives provided by inter-canton tax laws, such as the tax exemption system and formula apportionment.

fiscal equity and the equalization system

As explained above, fiscal competition in Switzerland leads to huge differences in the tax burden, as well as in the economic potential, between the different cantons (and, in some cantons, between the different local communities). Nevertheless, redistributive (progressive) personal income taxes are canton and local taxes first and federal taxes second. There is considerable redistribution within cantons and local communities.⁴⁰ This clearly contradicts the usual textbook wisdom, which says that, in a federal polity, redistribution should be undertaken by the federal government.

The Swiss post-tax income distribution is somewhat more unequal than is such distribution in other European countries, especially the Scandinavian and Benelux ones; however, it is in the same range as that of the southern European countries and the United Kingdom and Ireland.⁴¹ Thus, the special design of the Swiss federal system does not inhibit income redistribution comparable to that of other European countries. The main reason for this is the existence of an institutional framework that ensures that high-income people also have to contribute. Further:

1 The federal income tax is highly progressive, and 30 percent of the revenue is paid back to the cantons, one part of it directly and the rest via the fiscal equalization system.

2 There is a federal source tax of 35 percent on interest and dividend income.

3 The federal government upholds the first pillar of the old-age pension system, which is financed on a pay-as-you-go basis and is highly redistributive. Contributions are proportional to labour income (without any limit), but the maximum pension is about us$1,200 for singles and about us$1,800 for couples. Today, about 60 percent of all senior citizens receive the maximum pensions, and the share receiving these is increasing.

4 There is a fiscal equalization system.

Besides these arrangements, it is possible that direct democracy also helps to secure the system. Whenever the people themselves decide public issues, especially issues in relation to the tax burden, they are more prepared to accept the decisions and to contribute their share. There is clear evidence that tax evasion is lower when the people have more direct political rights.⁴²

As mentioned in the introduction, there are large discrepancies between the Swiss cantons: we have some small and very rich cantons, like Zug, Nidwalden, and Schwyz, and we have others that are relatively poor. The rich cantons spend more money per capita than do the poor cantons, but they are nevertheless able to have – ceteris paribus – lower tax rates. Moreover, the discrepancies have increased in the recent past. In order to keep the country together, a system of fiscal equalization necessary.

The existing fiscal equalization system is highly inefficient and, therefore, is currently being reformed. The reform, which is expected to be effective beginning in 2008, consists of four elements:⁴³

1 Some tasks and financial responsibilities that are, at present, the joint responsibility of the Confederation and the cantons will be separated. However, some tasks will still be common responsibilities.

2 New kinds of collaboration between the federal and canton governments and new kinds of financing will be introduced. The traditional system of matching grants will be replaced by a system in which the cantons get all the financial means that are necessary for those tasks for which the Confederation controls strategic issues and the cantons control implementation. Objectives for these tasks will be fixed and described in an intergovernmental contract.

3 There will be new forms of collaboration between the cantons, with cost compensation. If some cantons agree to collaborate on tasks they cannot perform by themselves (e.g., because they are too small), and if their activities have benefits for other cantons, the federal government can require reluctant cantons to participate if at least half of the cooperating cantons ask for such an intervention. The idea is to prevent free-rider behaviour.

4 The fiscal equalization system (in the narrow sense) will be reformed. The new system consists of three building blocks. First, there will be revenue equalization. Its objective is to provide at least 85 percent of the average financial means for all cantons. About 70 percent of the us$1,445 million the “poor” cantons will get will be provided by the Confederation, and the remaining 30 percent will be provided by the “rich” cantons. Second, there will be a cost equalization scheme, financed by the Confederation: about us$165 million for geo-topographic burdens and another us$165 million for socio-demographic burdens. Third, there will be a “cohesion fund.” The idea behind this fund is that “no canton with a weak financial capacity which today benefits from equalization, should suffer from worse conditions in the new scheme.”⁴⁴ Two-thirds of this fund’s money will come from the Confederation and one-third from the cantons. There will be full payment for the first eight years, and then a decrease of 5 percent each year for twenty years. Thus, this transition fund will exist for twenty-eight years.

The main difference between the old and new systems is twofold. First, there will be more division of power between the different spheres of government, and there will be a stricter correspondence between the tasks the cantons have to perform and their financial means. Second, the incentives for the cantons to care about their own tax base by, for example, attracting new firms will be strengthened. Thus, there is hope for efficiency gains that might eventually lead to a reduction of the tax burden.

The Swiss people accepted the constitutional changes necessary for this reform in November 2004.⁴⁵ In the meantime, the corresponding law passed the parliament and will be effective in 2008.

There are also fiscal equalization systems within the cantons. They are, as usual in Switzerland, very different between the cantons. Some are rather strict, while others, like the one in Schwyz, allow huge discrepancies within the canton.

financing capital investment

Some cantons have nearly zero debt, while others have considerable debt. This is also true for local communities. The main rationale for public debt is capital investment. However, for some cantons there are special reasons for their recent debt problems: insolvency of the canton bank due to risky investments (as in Berne and Solothurn), and the challenge posed by undercapitalization of public pension funds (e.g., Vaud and Geneva).

There are two main ways in which the cantons can borrow: (1) from the public by issuing bonds and (2) from commercial banks by obtaining loans. The municipalities have a third source: the Emission Centre of the Swiss local communities. This is a cooperative that currently has 928 local communities as members. It was founded in 1971, a time when it was difficult for local communities to get loans from Swiss commercial banks. Cantons as well as local communities can also get loans from foreign banks, and some of them do so. They are, however, not allowed to issue bonds outside of Switzerland.

What happens if a canton or local jurisdiction violates fiscal discipline and incurs excessive debt? How realistic is it for that canton to hope that it will be bailed out by the upper federal or canton government? How credible is the statement that such a bailout would not take place? It is difficult to believe that a canton or a local community could go bankrupt. Moreover, Switzerland (like every other country) does not have explicit bankruptcy rules or laws for such situations.

The federal Constitution provides each of the administrative levels in the country with an adequate financial basis. Cantons retain tax autonomy with respect to personal as well as corporate income and property taxes. In theory, there is no reason why a canton should ever find itself in a financial crisis. The cantons can increase tax revenue, should this be necessary. In fact, since 1848, there has never been a situation in which the federal government has needed, or been asked, to intervene financially. It must, however, be pointed out that the alternatives to raising tax revenue, and the expectations regarding what they have to contribute to national tasks, varies by canton. The problems that arise from this situation have to be addressed by the fiscal equalization system described above. This will prevent the country’s being divided into much richer and much poorer communities (provided all parties continue to address problems with reasonableness) while maintaining the incentives that encourage the cantons to take care of their own tax base. If this objective can be met by the fiscal equalization system, there is no reason why different cantons should not take different approaches to debt. And their varying indebtedness will be reflected in their different ratings in the capital market.

The picture is somewhat different for the local communities. In principle, they also have a sufficient tax base to perform their tasks. However, if a local community is highly indebted or actually goes bankrupt, as happened in the case of the community of Leukerbad, first of all the private banks (and individuals who hold the relevant bonds) have to at least partially depreciate their credits. However, the canton is responsible for supervising the situation. In the case of Leukerbad, the banks blamed Valais for having violated this duty and took the canton to court. However, the Federal Supreme Court in Lausanne decided that the canton was not responsible,⁴⁶ so there was no bailout. On the other hand, the bankruptcy of Leukerbad signalled the importance that ratings have in this market, and this induced higher interest rates for cantons and local communities with less sound finances.

In most cases, however, at least if a financial crisis is foreseeable, cantons intervene long before attempts to reach a settlement are necessary. For example, if the financial situation of a local community in St Gall deteriorates to the point where it has to be included in the canton fiscal equalization system, it loses its sovereignty to some degree. While the system allows the canton to prevent the local community from going bankrupt by simply pooling resources, the local communities have a strong interest in their sovereignty. This being the case, they do what they can to avoid such a situation.

Of course, in federal countries, local community irresponsibility in fiscal policy can never be totally excluded. The Swiss example shows, however, that, with appropriate institutional rules, the bailout problem can be solved satisfactorily and that federalism does not have to encourage irresponsible behaviour on the part of subfederal communities.

fiscal federalism dimensions of the public management framework

In Switzerland, agencies of the federal government are never involved in canton or local appointments, and there are no federal government restraints on subnational hiring and firing. Nor is there an elite federal service that is appointed and rotated through subnational positions in the cantons. Consequently, canton and local governments have full autonomy in hiring and firing personnel. Within the restraints of the federal Constitution, the cantons and local communities have full autonomy and flexibility in the exercise of executive powers. There are no avenues open to the federal government for undermining local autonomy.

Since the 1990s, New Public Management (npm) instruments and structures have been developed and implemented by all three orders of government. It started with pilot projects: the first was launched in 1993, while the bulk of them were started between 1995 and 1998. In recent years, evaluations of these projects have shown that they provide incentives for the administration to reduce costs and to take the interests of the clients more into account. Despite some ongoing discussions about its effectiveness, npm is currently fairly widespread among Swiss cantons and local communities.⁴⁷

There is, of course, some corruption in Switzerland, but it plays an extremely marginal role. According to the Corruption Perception Index of Transparency International, in 2004 Switzerland ranked seventh out of 146 countries.⁴⁸ So there is neither great concern about corruption in Switzerland nor any special policy in relation to it.

the way forward

As mentioned in the introduction, Switzerland is a rather small federal country, which nevertheless has strong tax competition between the cantons. This causes problems as it creates strong divergences between cantons as well as between local communities. There are relatively poor cantons with a high tax burden and relatively rich cantons with a low tax burden. To keep the country together while preserving canton tax sovereignty, a new fiscal equalization system is being implemented. In a few years we will see whether this will meet expectations or whether there will be a need for additional measures to reduce the discrepancies in Switzerland.

However, another development could cause problems in the future. In 2004, Schaffhausen introduced regressive personal income taxes for high incomes; the marginal tax rate is considerably lower for incomes above us$650,000 than it is, for example, for incomes of us$150,000. Consequently, at a certain point, the average tax rate also declines. In December 2005 Obwalden introduced a similar scheme. So we see increasing competition for high-income earners. The Federal Court decided that the new tax law of Appenzell Outer-Rhodes was against the Federal Constitution; thus, it had to be abolished once more.

Yet, even for very high-income earners, the tax load in Schaffhausen will still be higher (and in Obwalden hardly lower) than it is in Zug and Schwyz, which have a progressive tax schedule but, up until now, have been the cantons with the lowest tax burden. Thus, it is doubtful whether Schaffhausen and Obwalden (and perhaps also other small cantons that follow their strategy) will really be able to attract enough high-income people from other cantons to balance losses from reductions in the tax rate. It may be a zero-sum game between these cantons, with few spillovers to other cantons and, therefore, few effects on Switzerland as a whole;⁴⁹ instead, regressive income tax schedules might increase the incentives for tax evasion for those who have lower incomes but have to pay relatively more taxes. This could, in the long run, deteriorate the fiscal situation of the Swiss cantons, which – at least until today – have not suffered from a race to the bottom with respect to income and property taxes.

However, there is a clear race to the bottom with respect to inheritance taxes. In the last fifteen years, several cantons completely abolished these taxes for direct descendents. This is problematic in so far as these taxes are

– from an economic point of view – less harmful than are, for example, labour or capital income taxes.

Finally, from time to time, there are proposals to reduce the number of cantons in order to have larger cantons with less divergent structures. This would, of course, reduce many problems. For example, there would be no need for fiscal equalization between the cantons as the necessity for such a system arises mainly from the asymmetries between cantons. However, all attempts to merge various cantons have failed because voters have rejected them.⁵⁰ As Swiss citizens have a strong commitment to direct political rights, and as the cantons are deeply rooted in the consciousness of the population, the chance for any merger between them is extremely low.

no t e s

1 For a description of the Swiss Federal System, see also Nicolas Schmitt, “Swiss Confederation,” in Constitutional Origins, Structure, and Change in Federal Countries, ed. John Kincaid and Alan Tarr, 347–81 (Montreal and Kingston: McGill-Queen‘s University Press, 2005); and Thomas Fleiner, “Swiss Confederation,” in Distribution of Powers and Responsibilities in Federal Countries, ed. Akhtar Majeed, Ronald Watts, and Douglas M. Brown, 265–94 (Montreal and Kingston: McGill-Queen’s University Press, 2006).

2 Switzerland consists of twenty “full cantons” and six “half cantons.” With respect to fiscal matters, there are twenty-six cantons with twenty-six different fiscal constitutions.

3 These data refer to the year 2000. See Bundesamt für Statistik, ed., Statistisches Jahrbuch der Schweiz (Zurich: Verlag Neue Zürcher Zeitung, 2005), table 1.5.1.1,

p. 111. 4 Ibid. 5 The most densely populated canton is Basle-Town, with 5,045 persons per square

kilometre.

6 The data are for 2004. Source of the data: oecd, , viewed 2 October 2006; and World Bank , viewed 2 October 2006.

7 See Bundesamt für Statistik, ed., Statistisches Jahrbuch der Schweiz (Zurich: Verlag Neue Zürcher Zeitung, 2005), table 4.5.1, p. 249. 8 No official gdp data are available for the cantons. The data provided are estimates made by Basel Economics.

340

9 In 1848, Switzerland consisted of nineteen “full cantons” and six “half cantons.” The last canton, the Jura, was founded in 1979 (earlier, it had been part of the canton Berne).

10 The six half-cantons are Basle-Town and Basle-Country, Appenzell-Inner-Rhodes and Appenzell Outer-Rhodes, and Obwalden and Nidwalden. Basle, Appenzell, and Unterwalden were split into the six half-cantons for historical reasons. However, the only difference between them and other cantons is that they have only one member in the upper chamber of the Swiss Parliament and have only half the weight in constitutional referenda and initiatives.

11 For the assignment of powers among the different levels of governmental, see Tables 2 and 3. See also Fleiner, “Swiss Confederation.”

12 Art. 5 of the Federal Act on the Swiss National Bank.

13 See Yves Ammann, “Quelques réflexions à propos des règles de politique budgétaire,” Staatssekretariat für Wirtschaft, Wirtschaftspolitische Grundlagen (wp), Diskussionspapier No. 13, May 2002.

14 See Thomas Stauffer, Instrumente des Haushaltsausgleichs: Ökonomische Analyse und Rechtliche Umsetzung (Basel: Helbing und Lichtenhahn, 2001), 83ff.

15 Source of the data: Eidgenössische Finanzverwaltung, 18 Öffentliche Finanzen der Schweiz (Bern/Neuchâtel: Bundesamt für Statistik, 1990), 52; ibid., 2002, 74.

16 At this point there is no fiscal referendum at the federal level, but the Constitution limits the maximum rates of the federal income tax and the value-added tax; any change needs a constitutional referendum. Moreover, up until 2004, with respect to those two taxes, the taxing power of the federation was only given to it based on a clear constitutional time limit.

17 See the overview in Georg Lutz and Dirk Strohmann, Wahl – und Abstimmungsrecht in den Kantonen (Bern: Haupt, 1998), 151.

18 See Alexander Trechsel and Uwe Serdült, Kaleidoskop Volksrechte: Die Institutionen der direkten Demokratie in den schweizerischen Kantonen 1970 – 1996 (Basel: Helbing und Lichtenhahn, 1999), 330ff. To calculate these figures in us dollars and to make them independent from short-run exchange rate fluctuations, average purchasing power parity-values for the period from 2000 to 2004 are used.

19 Actually, in 1994 St Gall merely codified a practice that had been operative since 1929. For a detailed description of the different regulations, see Stauffer, Instrumente des Haushaltsausgleichs, 85ff.; and Verfassungsrat des Kantons Basel-Stadt, “2. Zwischenbericht der Verfassungskommission Finanzverfassung: Einführung einer Schuldenbremse”(Basel, 5 March 2002 [B/Nr. 503]). Attempts to introduce similar regulations are currently under way in other cantons as well (e.g., Berne and Zurich). Whether they will be successful remains an open question. Not all such attempts have been successful in the past. In the canton of Vaud, for example, such a proposal was rejected in 1998.

20 See Art. 82 of the canton constitutions and, especially, Arts. 61 and 64 of the “Staatsverwaltungsgesetz.” A detailed description of these institutions in the canton St Gall is given in Peter Schönenberger, “Institutionelle Massnahmen zur Verschuldungsbegrenzung im Kanton St Gallen”(Vortragsmanuskript, Dornach,

31 März 1995). See also Stauffer, Instrumente des Haushaltsausgleichs, 86ff.

21 The “simple tax revenue” is the basis for the income and property tax revenue; actual revenue is given by the simple tax revenue times a multiplier in the sense of a surcharge (called “tax foot”), which is currently 115 percent.

22 James M. Buchanan and Richard E. Wagner, Democracy in Deficit: The Political Legacy of Lord Keynes (New York: Academic Press, 1977); and James M. Buchanan and Richard E. Wagner, “The Political Biases of Keynesian Economics,” in Fiscal Responsibility in Constitutional Democracy, ed. James M. Buchanan and Richard E. Wagner, 79–100 (Leiden/Boston: Martinus Nijhoff, 1978). Buchanan and Wagner stated that the transition from balanced budgets to a counter-cyclical policy that allows for deficits during recessions in order to dampen economic fluctuations leads to a huge increase in public debt.

23 A classical reference is the following: “It remains to note that responsibility for stabilisation policy has to be at the national (central) level. Lower levels of government cannot successfully carry on stabilisation policy on their own for a number of reasons. This is obviously the case for the unitary state, where fiscal decentralisation is limited to the provision of local public goods. But it also holds for the federation.” See Richard A. Musgrave and Peggy B. Musgrave, Public Finance in Theory and Practice, 4th ed. (New York: McGraw-Hill, 1984), 515.

24 Art. 9, Finanzhaushaltsgesetz des Kantons Appenzell Ausserrhoden of 30 April 1995.

25 See Ernst Buschor, Klaus Vallender, and Thomas Stauffer, Kommentierter Entwurf für ein Finanzhaushaltsgesetz des Kantons Appenzell Ausserrhoden, ausgearbeitet für die Finanzdirektion des Kantons Appenzell Ausserrhoden (St Gallen: Institut für Finanzwirtschaft und Finanzrecht an der Hochschule St Gallen, 1993), 12ff.

26 Art. 38 (3), Gesetz vom 25. November 1994 über den Finanzhaushalt des Staates, Kanton Freiburg. This rule goes back to the Finanzhaushaltsgesetz des Kantons Freiburg of 1960. This law had similar regulations in Art. 5. As in St Gall, the law from 1994 did not create a really new situation. See Stauffer, Instrumente des Haushaltsausgleichs, 93.

27 Such instruments are also used in many local communities. We restrict the discussion, however, to the canton case studies.

28 See Lars P. Feld and Gebhard Kirchgässner, “The Political Economy of Direct Legislation: Direct Democracy and Local Decision Making,” Economic Policy 33 (2001): 329–67; and Lars P. Feld and Gebhard Kirchgässner, “Does Direct Democracy Reduce Public Debt? Evidence from Swiss Municipalities,” Public Choice 109 (2001): 347–70. See also the corresponding results in Christoph A. Schaltegger, “Budgetregeln und ihre Wirkung auf die öffentlichen Haushalte: Empirische Ergebnisse aus den us-Bundesstaaten und den Schweizer Kantonen,” Schmollers Jahrbuch 122 (2002): 369–413.

29 For the impact of the fiscal referendum on local debt, see also Feld and Kirchgässner, “Does Direct Democracy Reduce Public Debt?”

342

30 Peter Schönenberger, “Institutionelle Massnahmen zur Verschuldungsbegrenzung im Kanton StGallen,” Vortragsmanuskript, Dornach 31 (März 1995): 1.

31 For the assignment of taxes among the different government levels, see Table 4.

32 This is the main reason for the fiscal gaps between the federal and subfederal levels shown in Table 5.

33 See Bundesamt für Statistik, ed., Statistisches Jahrbuch der Schweiz (Zurich: Verlag Neue Zürcher Zeitung, 2005), table T18.2.2.3.1, p. 773.

34 Gebhard Kirchgässner and Werner W. Pommerehne, “Tax Harmonization and Tax Competition in the European Union: Lessons from Switzerland,” Journal of Public Economics 60 (1996): 351–71; and Feld and Kirchgässner, “Does Direct Democracy Reduce Public Debt?”

35 See Christina Elschner and Robert Schwager, The Effective Tax Burdens on Highly Qualified Employees (Heidelberg: Physica, 2005), 5.

36 See Bundesamt für Statistik, ed., Statistisches Jahrbuch der Schweiz (Zurich: Verlag Neue Zürcher Zeitung 2005), table T18.2.2.3.1, p. 773.

37 See Elschner and Schwager, Effective Tax Burdens, 5.

38 Milad Zarin-Nejadan, Die Besteuerung der kmu in der Schweiz Die Volkswirtschaft 70 (February 1997): 56–59.

39 Lars P. Feld and Gebhard Kirchgässner, “The Impact of Corporate and Personal Income Taxes on the Location of Firms and on Employment: Some Panel Evidence for the Swiss Cantons,” Journal of Public Economics 87 (2003): 129–55.

40 According to Lars P. Feld, “Tax Competition and Income Redistribution: An Empirical Analysis for Switzerland,” Public Choice 105 (2000): 125–64, excluding the redistributional impact of the Swiss pension system, two-thirds of the redistribution occurs at the subfederal levels. See also Lars P. Feld, Steuerwettbewerb und seine Auswirkungen auf Allokation und Distribution: Ein Überblick und eine empirische Analyse für die Schweiz (Tübingen: Mohr [Siebeck], 2000).

41 See Anthony B. Atkinson, “Income Distribution in Europe and the United States,” Oxford Review of Economic Policy 12 (1996): 15–28.

42 See Werner W. Pommerehne and Hannelore Weck-Hannemann, “Tax Rates, Tax Administration and Income Tax Evasion in Switzerland,” Public Choice 88 (1996): 161–70; and Lars P. Feld and Bruno S. Frey, “Trust Breeds Trust: How Taxpayers Are Treated,” Economics of Governance 2 (2001): 87–99.

43 For a more detailed description of the new system, see Bernard Dafflon, “Federal-Cantonal Equalisation in Switzerland: An Overview of the Present System and Reform in Progress,” University of Fribourg: benefri Centre for Studies in Public Sector Economics, Working Paper 356, updated version, May 2004.

44 P. Siegenthaler and P. Wettstein, “Finanzausgleich – alle können Gewinner sein: Entgegnung an den Finanzdirektor des Kantons Zürich,” Neue Zürcher Zeitung 182 (2001): 13. Cited by Dafflon, “Federal-Cantonal Equalisation,” 45.

45 Even the people of the canton Zurich accepted the new system, despite the fact that Zurich has to pay more into it in the future. Zürich, on the other hand, largely benefits from the new fund to compensate socio-demographic burdens.

46 Decisions 2C.1/2001, 2C.4/1999, 2C.4/2000, and 2C.5/1999, 3 July 2003. See Charles B. Blankart and Achim Kleiber, “Wer soll für die Schulden von Gebietskörperschaften haften?” in Perspektiven der Wirtschaftspolitik, ed. Christoph A. Schaltegger and Stefan C. Schaltegger, 137–50 (Zurich: vdf, 2004), 137–50.

47 For the development and effects of npm, see Stefan Rieder and Luzia Lehmann, “Evaluation of New Public Management Reforms in Switzerland: Empirical Results and Reflections on Methodology,” International Public Management Review 3 (2002): 25–42; Kuno Schedler, “… and Politics? Public Management Developments in the Light of Two Rationalities,” Public Management Review 5 (2003): 533–50.

48 See , viewed 21 November 2005. There is, moreover, no indication that some cantons are more corrupt than others.

49 Due to the construction of the new system of fiscal equalization described above, such a policy cannot be used to get more out of (or to pay less into) this system.

50 This is contrary to mergers of local communities. There have been quite a lot in recent years, and this has reduced the number of local communities from 2,915 in 1990 to 2,760 today. However, in many of these cases, the cantons provided financial incentives in order to induce the citizens to accept the merger.