Switzerland

Abstract

Switzerland’s federal system features an internationally unmatched pattern of widespread revenue autonomy and administrative fragmentation on subnational levels of government. In combination with strong participatory institutions, the resulting tax competition among cantons and municipalities widely restricts excessive tax burdens and induces fiscal restraint. A comprehensive institutional framework at the federal level sets boundaries to ensure that tax competition is not at the expense of fiscally weaker cantons. Yet Swiss fiscal federalism is more fragile than it seems. Despite many institutional barriers to centralization, there is a persistent trend to allocate an increasing number of responsibilities with the federal government. In addition, the fiscal equalization scheme sets disincentives for fiscally weaker cantons possibly increasing disparities in the future. Reform proposals should seek to strengthen fiscal equivalence and prevent a system in which cantons are merely executing federal regulation.

1 Introduction

Despite its comparatively small territory, Switzerland has an extensive federal structure.¹ Its subnational jurisdictions, the cantons, enjoy wide-ranging political and fiscal freedom. The high degree of autonomy is extended to the local level and based on the principle of subsidiarity—the idea that tasks should only be centralized if local or subnational authorities cannot perform them effectively. In addition, the Swiss political system grants the most extensive political rights to its citizens worldwide. Initiatives and referenda are institutionalized in all three tiers of government: local, cantonal, and federal.

Comparatively, the fiscal autonomy of lower-tier governments is particularly extensive on the revenue side of the budget. Apart from subnational jurisdictions in Canada, Swiss cantons and municipalities are granted the most autonomy with respect to tax policy (see Fig. 1). However, given the semi-direct-democratic system, tax changes by cantonal governments and parliaments are usually subject to public approval. In addition to participatory institutions, Switzerland has an internationally unmatched degree of administrative (jurisdictional) fragmentation (see Fig. 2).

Fig. 1

(Source OECD revenue statistics)

Tax decentralization OECD countries (2017): subnational tax revenue in % of general government tax revenue (incl. social security revenue)

Fig. 2

(Source Brülhart et al. [2015, Fig. 17.8]. Jurisdictional fragmentation: average number of municipalities per 100,000 inhabitants. ALTS: local tax revenue with real tax autonomy; LTS: all local tax revenuee)

Tax shares at local level and jurisdictional fragmentation

This unique combination has several implications: Fiscal responsibility, referenda, and budget rules induce efficiency in expenditure, high-tax compliance, and satisfaction of regional preferences. Tax competition restricts excess supply of public goods and services. A comprehensive redistributive framework at the federal level sets boundaries to ensure that tax competition is not at the expense of fiscally weaker cantons. This chapter emphasizes the distinctive features of the Swiss fiscal federal system and discusses empirical evidence on its implications as well as recent policy developments.

2 Switzerland: Facts and Figures

This second part of our chapter largely relies on Kirchgässner (2007).

Although Switzerland is small, with a land area of 41,285 km² (15,940 sq. miles) and a population of 8.5 million people, its constituent parts, the 26 cantons, are remarkably different (see Fig. 3).² It has four official national languages along with the corresponding cultures: 63% speak (Swiss) German; 23%, French; 8%, Italian; and 0.5%, Romansh. English, Portuguese and Albanian are the predominant foreign languages.

Fig. 3

(Source Federal Statistics Office [2018])

Switzerland and its cantons

With respect to religion, more than two-thirds of the Swiss population claim affiliation to Christianity with a relative majority of Roman Catholics (37% of the population) over Protestants (25%) and other Christian denominations (6%). The residents who are not members of any religion have been on the rise for several years and now amount to a quarter of the population. Islamic denominations make up for about 5% of the population.

Almost 25% of the population are foreigners. This is a higher percentage than in any other country in Europe (apart from some microstates such as Monaco or Liechtenstein). Given the linguistic, cultural, and religious heterogeneity, Switzerland is a nation shaped by the resolve of its citizens and is well aware of its many diversities.³

The federal roots trace back to the thirteenth century when the three primary cantons—Uri, Schwyz, and Unterwalden—entered into a treaty. In 1848 the modern Swiss Confederation was founded and the current federal structure was institutionalized after a short civil war between the Protestants and Catholic separatists (known as the Sonderbundkrieg). The fact that Switzerland did not split up along its linguistic divisions in the second half of the nineteenth century (when its neighbors, Italy and Germany, created their national states) is presumably owed 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 (see Fig. 3). The average canton has about 325,000 people, but population sizes range from 16,100 in Appenzell Innerrhoden to 1.5 million in Zurich. The average population density is 205 people per km². Compared with some other European countries such as Belgium or the Netherlands this might not seem low. However, Switzerland’s territory is diverse with some densely populated area such as the “Mittelland”, a narrow tract that stretches from Lake Geneva to the Lake of Constance and includes most of the medium-sized cities as well as Zurich and its suburbs. North and west of the Mittelland bordering France are the Jura Mountains, to the south and east are the Alps. Large parts of these mountainous areas are unproductive and, as a result, quite sparsely populated.

Given the volatile economic environment in Europe, recent economic activity in Switzerland has recorded solid growth rates of GDP between 1 and 3%. With the exception of 2009 when owing to the financial crisis, growth was negative, GDP has grown since the end of 1990s with foreign trade playing a key role (FSO 2018). Based on purchasing power, international comparisons show that Switzerland is the fourth richest country in the world with GDP per capita amounting to 68,105 US-Dollars (PPP) lagging behind Luxembourg, Singapore and Ireland only (OECD 2019).

Within Switzerland, however, there are substantial economic discrepancies. In 2016, the average per capita GDP was 78,869 Swiss Francs (median per capita GDP was 68,332 Francs) with the canton Basel-Stadt being 120% above and the canton of Uri being 33% below the national average (FSO 2018). Although the discrepancies used to be larger, they remain a source of public and political controversy.

3 The Structure of Government and the Division of Fiscal Power

The Swiss federal system is characterized by the widespread autonomy of its subfederal jurisdictions, the 26 cantons, guaranteed by several articles in the federal constitution (namely Art. 3, 5a, 43, 47 and 48). The latter grants great freedom with respect to the cantons’ political systems. Their constitutions are merely required to be democratic, not contradict federal law and allow for revisions if a majority of the electorate demands. In order to finance expenditure for those tasks, the cantons can levy their own income and property taxes (see V. Public revenue). They are free to decide not only on the tax rates but also on the tax schedule as well as on how progressive these taxes are. The Confederation has authority only in those areas in which it is empowered by the federal Constitution (e.g., foreign affairs, defense, customs, and monetary policy) meaning that tasks that do not explicitly fall within the scope of the Confederation are handled by the cantons. Table 1 provides an overview of the distribution of responsibilities between the respective levels of government.

Table 1 Distribution of responsibilities between the three levels of government

Each canton has its own constitution, parliament, government, and courts. The cantonal parliaments have between 50 and 180 members who are elected based on the system of proportional representation. The cantonal governments have five or seven members. They are directly elected in majority systems by the people at the ballot box except for the government in the canton Appenzell Innerrhoden, where the annual general assembly (Landsgemeinde) elects the government in April.⁴ In all other cantons elections take place every four or five years.

Despite multiple shared responsibilities with the federal government—usually assigned by the federal Constitution—cantons act independently in many policy fields to a large extent. Among other fields, high degree of autonomy is guaranteed in education, health, police, law and courts as well as transport, cultural service, social assistance, energy supply, or waste (water) management. Many of those fields are shaped by local policies on the municipal level. As of 2019, there were 2,212 municipalities. In recent years, a decreasing trend has emerged due to mergers that often are encouraged by financial incentives set by cantonal governments or parliaments. Around one-fifth of municipalities have their own parliament; in the other four-fifths, decisions are taken by direct democracy in a local assembly.

The scope of local autonomy is determined by the respective canton and, therefore, varies considerably. However, given the autonomy for local government provided in the cantonal 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 commune deteriorates seriously. In such a case, the local budget has to be approved by the cantonal government.

3.1 Federal Scope of Responsibilities and Its Evolution

After the foundation of the modern Swiss state in 1848, the federal government’s scope of powers was extended in three waves (e.g. Blöchlinger and Frey 1992). In an early phase, common foreign and military policy were developed and the foundations for a common market were laid. Social policy was enshrined into federal law in several steps during the interwar period and shortly after World War II (Sommer 1978). In a third step, the federal level was endowed with public responsibilities emerging in 1960s and 1970s such as energy, highway, environmental policy and regional development planning.

As a result, the once distinctive division of responsibilities has subsequently been overruled by the mechanisms of intensive cooperation between the three levels of the federal system creating a broad array of shared responsibilities (Linder 2010). In primary education, for example, many cantons grant their municipalities wide-ranging autonomy. At the same time, the federation sets standards for the duration of compulsory school attendance, school entrance age, or the duration of and objectives for different levels of education. While secondary education largely remains a cantonal responsibility, there is a strong federal impact on tertiary education with two federal universities (Swiss Federal Institutes of Technology in Zurich and Lausanne) and research funding being a federal task. In addition, the federal government subsidizes the cantonal universities depending on the number of students enrolled.

Similarly, the federal government has a large impact on health policy that is formally in the domain of the cantons by setting the legal framework for the (private) health insurance (Feld et al., 2017b). Yet the provision of health services such as hospitals is ensured by the cantons and municipalities. Responsibilities for social welfare are split, too. Despite consistent political pressure to harmonize social assistance, it has remained a cantonal and local task. Social insurance such as the pension system or unemployment insurance, however, essentially is a federal task except for the supplementary benefits to old age and disability pensions, which is shared with the cantons. Hence, most cantonal responsibilities are in the provision of public consumption and investment goods rather than social protection or the transfer system although this «principle» is violated in various ways. The federation is more involved in the redistribution of income given its responsibilities with regard to social protection and regional or agricultural policy.

Figure 4 shows the governmental division of expenditure by various functions. Although insightful for public spending analysis, it ignores the high degree of integration between the three levels of government in different policy fields. As mentioned above already, this is particularly striking for health. The federal government’s impact on health policy is significantly larger than its share of expenditure suggests. This is an example of where the central government exhausts or even exceeds its legislative powers at the expense of the subnational jurisdictions that are reduced to executing superior legislation. In order to strengthen their position toward the federal government, the cantons have formed the conference of cantonal governments as well as policy-specific conferences of ministers (e.g. finance, health, education) often resulting in expenditure and revenue sharing agreements with the federal government.

Fig. 4

Expenditure share by level of government (left axis) and total government expenditure in 1,000 CHF (right axis) by function in 2017. Data: Federal Finance Administration

3.2 Issues with Centralization

While the conferences of cantonal governments and ministers are a legitimate measure to represent interests⁵ and co-ordinate cantonal arrangements in policy fields in which they have significant responsibility, increasing revenue and expenditure sharing (further) undermine fiscal equivalence and direct democracy. As the conferences grow more important, they establish an intermediate level of government between the Confederation and cantons resulting in a power shift from cantonal legislature to executives (Rother and Rühli 2017). In a direct-democratic system, this development is problematic because it restricts parliamentary responsibilities and undermines the fact that there is little room for direct statutory interference between the different levels of government.⁶ In addition, their decisions have centralizing character because they imply a national consensus from which it is more difficult to deviate, likely changing the odds of winning a referendum. Similarly, increasingly powerful conferences impair the principle of fiscal equivalence that establishes a multi-level government setting by “perfect mapping” uniting those who fund, benefit from, and determine policy. Although “inequivalence” is prevalent in practice, additional deviation from the principle could threaten accountability or bureaucracy benefits of decentralization.

The tendency of more sharing agreements has two implications: First, the politically painful process of disentangling responsibilities between federal and cantonal level can be postponed further. Second, the procedural provisions that protect the centralization of taxes and tasks such as popular initiative and the referendum are partially impaired.⁷ In particular, the latter is commonly considered a safeguard against centralization.⁸ On the federal level, the signatures of 50,000 citizens are sufficient to call for a petition or veto referendum on any new national legislation. Constitutional changes are subject to a mandatory referendum vote.⁹ The Swiss electorate votes on the resulting policy proposals that appear on the ballot for approval or disapproval three to four times per year.

Despite these institutional barriers and efforts to disentangle responsibilities in a large reform in 2008 the trend to centralize persists. Out of 159 changes in jurisdictional responsibilities between 2000 and 2016 none—apart from the reform itself—constituted a decentralization (Fässler et al. 2017). While 25% of changes devolved formerly cantonal responsibilities to the federal level, 75% concerned changes from formerly cantonal to shared tasks. The reform whose purpose was to strengthen the subsidiary structure of the Swiss system revealed how difficult the process of disentanglement is. Although a number of responsibilities were allocated to a new, single level of government, many tasks remained integrated due to political compromise.

Centralization is the longer the more accepted as the inevitable outcome of the increasingly globalized and independent world (Rother and Rühli 2017). To many, Switzerland’s high degree of fragmentation seems incompatible with the associated challenges. Yet, centralization and sharing agreements are driven by the mutual interests by many (political) actors. For example, in a more transparent world, acceptance for the trial and error approach of federalism has diminished. If the public perceives the provision of cantonal public goods insufficient, calls for a federal framework follow suit. Furthermore, the erosion of local, traditional media has emphasized the significance of national politics. Therefore, politicians have an incentive to add subnational and local policy issues to the national agenda. Importantly, the cantons are also to blame for this development given their tendency to cede autonomy in return for financial reimbursements by the federal government.

4 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. Its independence is embodied as a principle in the Constitution. According to the National Bank Act, the SNB’s main objective is to ensure price stability. In doing so, it is bound to take into account business cycle developments. After the breakdown of the Bretton Woods system in the 1970s, SNB policy focused on the quantity of money until 1999. However, as this was considered one of the factors contributing to the low growth of the Swiss economy in the second half of the 1990s it changed course. Ever since the SNB’s strategy has been to attempt to keep the rate of inflation between 0 and 2%.

It is generally accepted that the SNB has been effective at achieving its main goal given that Switzerland has one of the most stable currencies in the world. From 1980 (2002) to 2018, the average inflation rate was 1.7 (0.4) percent, compared with 3.1 (2.1) percent in the United States (and percent 1.7 in the Euro area). This is also reflected in the development of the exchange rate. Since 1974, when the Swiss franc began floating against all other currencies, it has appreciated significantly against all major currencies constantly challenging the strongly export-oriented economy.

In the aftermath of the financial and European sovereign debt crisis, the SNB implemented unconventional monetary policy by ceiling the exchange rate to the Euro at 1.20 CHF as a measure against the Franc’s overvaluation. The cap was introduced in September 2011 and, to the surprise of markets, scraped in January 2015. The present monetary policy environment remains challenging. Switzerland continues to be a credible safe haven forcing the SNB to keep interest rates at a record low if it wants to protect the Franc from appreciation. At the same time, high growth rates, continuously increasing investment activity and demand (particularly in parts of the real inheritance market), and the yield requirements in the pension system call for an increase in interest rates.

According to the federal constitution, stabilization policy is a federal responsibility. However, as a small open economy, the scope for fiscal policy in Switzerland is limited. Even large federal deficits, as seen in the 1990s, hardly provide an impulse to the Swiss economy (Schaltegger and Weder 2010). Automatic stabilizers such as the unemployment insurance, “short-time work” compensation, the progressive tax system or the federal debt brake have proven more successful. In addition, cantonal impulses have often been pro-cyclical and tended to cancel out each other.

In order to strengthen the cantons’ fiscal discipline, the conference of the cantonal finance ministers agreed on a model law for cantonal budgeting in 1981 (Burret and Feld 2018). For the current budget the law requires a balanced budget in the medium term and a depreciation of balance sheet deficits by at least 20% p.a. For the investment budget, the law requires the self-financing ratio for net investments to be at least 80% if cantonal net debt exceeds revenue by more than 100%. The two rules are directly linked as depreciation of investments and of balance sheet deficits need to be included in the current budget. Anti-cyclicality is implicitly ensured by the provision to balance the budget in the medium term.

However, the provisions did not prevent cantonal debt from increasing considerably during the 1990s. Low economic growth rates and canton-specific fiscal shocks (e.g. bailouts of public-owned cantonal banks) undermined many cantons’ fiscal position. Consequently, cantonal gross debt increased by 43% (in real terms) from 1990 to 2004, but the development varied depending on the specific canton (FFA 2018). For example, among the six cantons that recorded a real decrease in gross debt, four introduced a debt break during that period or had done so previously. At the other end of the scale, debt in Vaud and Geneva increased twofold or more. While Geneva remains the most indebted canton in per capita terms (36,400 Swiss Francs or 260% above the national average in 2017), Vaud’s debt level has decreased by almost 50% both overall and per capita. Overall, cantonal gross debt is slightly lower than 15 years ago.

Fig. 5

(Source Burret and Feld (2018), Fig. 1)

Cantonal finances in real Swiss Francs per capita and the number of cantonal debt brakes 1980–2011

5 Fiscal Rules and Referenda

Although economic aggregates developed much more favorably in the 2000s, easing many cantons’ debt levels, empirical evidence suggests that fiscal institutions matter (e.g. Burret and Feld 2018). Considerable decreases in debt were primarily observed subsequent to the introduction of debt breaks. Today fiscal constraints can be found in nearly all cantonal constitutions and corresponding budget laws. The fact that per capita debt remains the lowest in the only canton without any fiscal rule (AI) is most likely due to its strong direct-democratic institutions and not an indicator of the lack of effectiveness of fiscal rules.

The effectiveness of fiscal rules largely depends on their credibility. For a rule to be credible one of the following criteria needs to be fulfilled (e.g. Feld and Kirchgässner 2008; Feld et al. 2017a): (a) a link between budget planning and final accounting, or (b) a numeric deficit limit or (c) non-discretionary sanctions in the form of tax or expenditure adjustments. At present, a total of 18 cantons have adopted one or more of these requirements.

St. Gallen, for example, with its historically stringent budget rule that was introduced in 1929 fulfills all three requirements. Requirement (a) is satisfied by the provision that a current deficit in the final accounting that cannot be covered by savings is transferred into the budget plan of the year after the next year. The budget law also stipulates a deficit ceiling of 3% of the “simple tax revenue”¹⁰ satisfying requirement (b). Lastly, in order to fulfill requirement (c), if a deficit is expected, the tax multiplier has to be increased up to a level such that the deficit ceiling is not violated. The latest instance when cantonal finances in St. Gallen were constrained by the budget rules was in 2013, forcing the government to adopt spending cuts, liquidating savings worth 110 million Francs, and increase the multiplier by ten percentage points.

5.1 Favorable Evidence on Budget Rules…

The most recent econometric assessment of the effect of cantonal debt breaks on cantonal finances provides conclusive evidence that the fiscal rules are associated with sound cantonal and local finances. Burret and Feld (2018) not only show that cantonal fiscal rules reduce public deficits. As most debt breaks constrain current budget variables, there is an implicit risk of evasion into unconstrained accounts. While there is no evidence of evasion into funds and special financing, debt breaks are partly associated with increased spending in the investment budget. Moreover, the authors reject the common concerns that debt breaks undermine public investments or restrict fiscal measures in response to fiscal shocks. With respect to the third tier of government, the evidence demonstrates that the introduction of cantonal debt breaks improves local finances. A possible explanation, supported by Burret and Feld (2017), lies in the statutory cantonal responsibility for municipal finances that may be taken more seriously subsequent to the introduction of a cantonal debt brake.

The deterioration of federal finances in 1990s—gross debt increased by 160% in real terms between 1990 and 2002—prompted the Confederation to propose a federal debt break. The subsequent vote on the constitutional change was approved by over 85% of the electorate. The basic rule of the debt brake constrains the federal government to balance the financial accounts in the medium term.¹¹ Effectively, the level of expenditure is restricted by the cyclically adjusted revenues ensuring anti-cyclical fiscal policy. Deficits and surplus are credited to the compensation account and, by construction, treated asymmetrically. Deficits, that is the expenditure ceiling is violated, need to be compensated in the subsequent years. Surplus, that is expenditure fall short of revenues, are used to reduce debt.

The federal debt brake has been evaluated favorably. Counterfactual analysis shows that the introduction of the debt brake lead to an annual debt reduction of around 2% until 2010 (Salvi et al. 2020). The positive impact is associated with its precise and cyclically adjusted target, the comprehensive scope to prevent loopholes in the budget, and the strict political enforcement mechanism. In addition, there is no evidence supporting the claim that the debt brake has a negative impact on federal investment spending. The investment share in the federal budget has fluctuated around 12% since the 1990s.

5.2 …and Fiscal Referenda

Besides fiscal rules, Switzerland also has extensive institutional barriers of direct democracy. The major difference in this respect between the federal and cantonal level is the cantonal fiscal referendum.¹² As Table 1 shows, except for Vaud, mandatory and/or petition referendums exist in all cantons.¹³ In the majority of cantons (16), outlays above a certain expenditure threshold require popular approval. In nine cantons, voters can launch a petition for spending programs above a certain expenditure threshold. Twelve cantons have two different thresholds for mandatory and petition referendums (Table 2).

Table 2 Direct democracy in Swiss Cantons. Based on Matsusaka (2018)

Absolute thresholds vary significantly owing to large differences in populations and, subsequently, budgets. Thus, the threshold relative to total cantonal revenues constitutes a more meaningful measure. Thresholds for petition referendums generally are lower than for mandatory requirements (Leuzinger and Kuster 2017). The former range between 0 and 0.5 (0.25) percent of revenues for nonrecurring (reoccurring) spending programs. In Vaud, Neuchatel, and Geneva there is no lower bound for starting a referendum meaning any parliamentary budget decision to be contested. The latter range between 0.1 (0.01) and 5 (1.0) percent of revenue for nonrecurring (reoccurring) spending. Referenda rights are most extensive in Aargau where nonrecurring expenditure amounting to one-tenth or one-hundredth of a percent of total revenue (5 million francs opposite a total median expenditure of 5 billion francs in 2014–2016) can be disputed.

Several studies provide empirical evidence on the constraining impact of the fiscal referendums on cantonal finances (Feld and Matsusaka 2003; Funk and Gathmann 2011, 2013). Depending on the time period studied and the research design, the mandatory referendum is found to reduce both cantonal expenditure and revenue by 12–19%. According to Galletta and Jametti (2015) cantonal fiscal referendum increases local spending for municipalities without fiscal referenda indicating the importance of direct-democratic control on all levels of government. With respect to the effect on debt, evidence is scarce. Feld and Kirchgässner (2001) find an increasing, yet statistically insignificant effect on public debt. Feld et al. (2008) demonstrate that fiscal referenda induce less centralization from the local to the cantonal level working as a safeguard for the interests of municipalities.

6 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. Cantonal tax revenue is mainly financed by personal, corporate income, wealth, and inheritance taxes. The federal government mainly relies on a value-added tax (VAT) and other (indirect) consumption taxes, a highly progressive income tax, and a proportional tax on the profit of legal entities. In addition, it levies a withholding tax on capital income (so-called Verrechnungssteuer), which has a rate of 35%. When capital incomes are declared, the tax is reimbursed. Capital income on private assets (as opposed to business assets) is tax-free. Capital income on real inheritance, however, is taxed by all cantons. Inheritance and wealth taxes are exclusively charged by the cantons as well. While there is no deductibility of personal income taxes, corporate income taxes are partly credited with different tiers of government. Most federal taxes are based on explicit constitutional provisions. The subsequent restricted access to some tax bases is relevant only to a limited extent as other tax receipts are governed either by the respective political area in the Constitution (e.g. motorway tax, heavy vehicle charge) or at statutory level (e.g. CO2 tax, casino tax). Prohibiting cantons to levy a specific tax requires an excluding provision in the federal Constitution such as article 134 that restricts the VAT, other excise taxes, the withholding and stamp tax to the federal level.

Figure 6 shows how tax revenue is split between three levels of government. The subnational (cantonal) ratio of total tax revenue amounts to a high value of 53 (33) percent. Income and wealth tax revenue amounts to 84 (51) percent of subnational (cantonal) tax revenue illustrating how the federal government is much less dependent on income tax revenue. For 2017, 82% of the federal tax revenue can be allocated to excise taxes (with two thirds of revenue from VAT) and income taxes (with equal shares from personal and corporate income), respectively. An institutional idiosyncrasy is the fact that income taxes are levied by the cantons. The federal government compensates this effort by reimbursing 21% of its income tax revenue to the cantons based on the cantonal tax base. The federal personal income tax is piecewise linear with constant marginal tax rates by brackets. Marginal tax rates range from roughly 1 to 11.5%. Owing to a basic tax exemption, the highest 5% of income taxpayers (taxable income above CHF 150,000) pay about two-thirds of the revenue of the federal income tax (figures based on 2016 federal tax statistics).

Fig. 6

(Source Federal Finance Administration [2019]Footnote

As Fig. 5 does not include social security revenue the numbers in Figs. 1 and 5 deviate.

)

Tax revenue by level of government 2017 in billion Swiss Francs and in % of combined government revenue

6.1 Cantons Exploit Tax Autonomy

The federal constitution guarantees cantons the autonomy to determine their tax rates, tax surcharges (multiplier), and exemptions (Feld et al. 2017b).¹⁵ Until the adoption of the federal tax harmonization law in the 1990s, cantons enjoyed wide-ranging flexibility in tax design matters. Today restrictions apply with respect to tax liability, tax base, tax calculation, and procedural provisions. Although those provisions are of formal nature, court rulings on constitutional tax principles have limited cantonal flexibility. Particularly, the federal supreme court’s decision to rule the regressive tax schedule in Obwalden unconstitutional in 2007 can be considered an interference with cantonal tax autonomy as there are no statutory restrictions with respect to tax rates. The court ruled against Obwalden’s tax reform because the tax schedule was in violation of the constitutional rule to tax households according to their economic capacity.

Municipalities’ fiscal autonomy depends on the extent to which they are explicitly authorized to levy taxes by the respective canton. In practice, they determine their level of taxation by adding multipliers to the canton-specific tax rates. Tax competition among municipalities differs substantially to tax competition among cantons because the former is limited to the scope of taxation that is needed to finance public expenditure, whereas cantons are free to determine their tax burden to a large extent as well as the design of their tax legislation.

A comprehensive cantonal comparison of tax burdens is shown in Fig. 7. Tax exhaustion quantifies the share of a canton’s economic capacity that, on average, can be fiscally absorbed.¹⁶ The countrywide average fiscal burden is 25% (or 100 on the tax exhaustion index) with substantial differences among cantons. While taxpayers in Geneva cannot resort to 34% of their income, profits, or accretion, Schwyz and Zug absorb 11% of their taxpayers’ resources. From an international perspective, the Swiss tax burden is low compared with most other locations in Europe and the United States. This holds even for the higher-tax Swiss cantons. However, this perspective fails to include other compulsory charges such as for mandatory health or accident insurance or the occupational pension scheme.¹⁷

Fig. 7

Tax exhaustion as a percentage of economic capacity (left axis) and as an index (right axis). Data Federal Finance Administration

These differences imply that cantons take advantage of their autonomy in tax matters which can lead to large differences in liabilities. Let’s consider, for example, the 2018 tax burden for two identical taxpayers in the city of Zurich and the municipality Wollerau in the canton of Schwyz, which is about a 30-min drive or train ride from the city. Taxes due for a single person with a gross income of CHF 50,000 amount to 6.5% in Zurich and 4% in Wollerau, respectively (FFA 2019). For a married couple with a gross income of CHF 500,000, two kids and a workload split of 70–30%, Zurich charges 17.4%, whereas the cantonal tax burden Wollerau is 7.3%. Note that both couples pay an additional 7.7% to the federal government. As mentioned, the federal income tax is highly progressive meaning, in this specific case, the cantonal tax burden of the individuals with a gross income of CHF 50,000 increases by merely 0.4%.

The progressivity of the tax system is further put into perspective if the fact is accounted for that there is spatial sorting of taxpayers one effect being that high-income households are more likely to live in low-tax jurisdictions. Quantifying the effective level and progressivity of income taxation in Switzerland, Schmidheiny, and Roller (2016) show that high-income households face significantly lower average and marginal tax rates. Very high-income households without children even face regressive taxes. Half of the reduction in the tax burden on top incomes between 1975 and 2009 is due to reductions in statutory tax rates and about half to stronger income sorting of the population.

6.2 Largely Benevolent Tax Competition

In recent years, there has been an increasing number of scientific contributions providing robust evidence on fiscal interdependencies among cantons and municipalities. They confirm both the presence of strong tax competition as well as subsequent behavioral responses to differences in tax burdens. Investigations confirm the intuitive conjecture that location (canton or municipality) choices are tax-induced. The proportion of high-income earners in a canton is significantly influenced by the canton’s tax rate (e.g. Schmidheiny 2006). Among local communities, the effects of tax competition are also prevalent. This can be shown by exploiting an institutional feature that keeps foreign nationals with an annual income below CHF 120,000 subject to a special tax scheme (Quellenbesteuerung) for at least the first five years of residency (Schmidheiny and Slotwinsky [2018]). The authors observe two behavioral effects: First, foreign households located in a high-tax municipality are likely to move away once they are no longer subject to the scheme. Second, the same households systematically seek to keep their income below the threshold, whereas foreign household in low-tax municipality attempt to exceed the threshold. In a different attempt to detect the presence of tax competition, Eugster and Parchet (2019) provide evidence that the decentralization of fiscal responsibilities limits local governments from setting their culturally preferred tax rates. This finding is based on the empirically established fact that differences in municipal tax rates narrow as proximity to a (intra-cantonal) language border increases despite differences in preferences for public expenditures between French and German-speaking municipalities (with the former exhibiting preferences for more public expenditure).

An inherent issue of public policy debates is whether tax competition is in the interest of society. One recent study tries to answer this question in the setting of Swiss municipalities (Brülhart and Jametti 2019). The authors find that tax (multiplier) competition is beneficial when the intensity of direct-democratic control in local tax matters is low. The likelihood of an excessive tax burden is significantly higher without fiscal competition and insufficient control by direct-democratic institutions such as town-hall meetings or compulsory referendum. In this case, policymakers are less constrained to increase taxes for the purpose of preferred projects. In addition, the findings are valid outside of the Swiss case as the scope of vertical fiscal externalities and revenue maximization is likely larger in other federations. The former can be explained by Swiss municipalities’ high fiscal share in international comparison which indicates increased scope as the prevalence of vertical externalities is inversely related to the size of local fiscal shares. And the latter is based on the fact that direct-democratic rights additionally constraining revenue maximization are less extensive in other nations.

The emphasis on tax-induced location choices may also be overstated in the political process as indicated by an investigation that studied the mobility of elderly taxpayers in response to a continuous decrease in cantonal inheritance taxes that—with a few exceptions—have been significantly reduced (Brülhart and Parchet 2014). The authors provide evidence that there was no systematic effect of cantonal inheritance taxes on the tax base from retired taxpayers and that the widespread exemptions had a negative effect on cantonal inheritance tax revenue after a tax cut.

Switzerland, relative to the size of its public sector, has the highest level of annual wealth taxation in the developed world. Revenue amounts to around 9% of total subnational tax revenue. The tax is raised on all types of wealth, e.g., cash, financial assets, real inheritance, and luxury durable goods. Exemptions apply for standard durable household goods, compulsory pension assets, and a limited amount of voluntary pension. Exemption levels and tax rates vary by canton with current top marginal wealth tax rates between 0.13 and 1.01%. An investigation into the elasticity of taxable wealth shows strong behavioral responses (Brülhart et al. 2019): A 1 percentage point decrease in the wealth tax rate increases reported wealth by 43% after six years owing to taxpayer migration, capitalization into housing prices, and changes in taxable financial assets of taxpayers suggesting changes in evasion behavior. In contrast to popular belief, the results suggest that the extent of tax avoidance is remarkably constant along with the wealth distribution.

6.3 Corporate Tax System Under Change

Corporate income taxes in Switzerland also vary considerably among cantons. The lowest tax rates can be found in central Switzerland where rates are 40–50% lower than in the higher-tax cantons.¹⁸ While two out of three companies subject to corporate income taxation did not pay any taxes in 2015, almost 90% of federal revenue was born by 2.8% of liable companies (KPMG 2019).

Due to peculiarities of the corporate income tax system, statutory and effective tax rates can differ. Half of the Swiss cantons have an effective average tax rate below 15%. Economic centers such as Zurich, Basel, or Geneva position themselves favorably in international tax competition although differences between cantons and international economic hubs have declined in recent years.¹⁹ As Table 3 shows, low effective tax rates are predominantly located in central and eastern Switzerland. The figures refer to maximum effective tax rate that a company located in a cantonal capital faces and that does not enjoy any tax exemptions, that is it includes federal, cantonal, and local corporate tax liabilities.

Table 3 Key parameters of cantonal corporate tax reforms

In the past decade there has been a downward trend in cantonal corporate tax rates. On average the statutory tax rate has decreased by 3.7% since 2007. Cantons with a reduction of more than five percentage points (GR, BS, VD, SH, LU, NE and AR) can be found throughout the country. This trend is likely to continue with further changes announced (subject to parliamentary and/or voter approval) in the wake of recent federal and cantonal tax policy reforms.

The most recent changes in federal corporate tax legislation were approved by voters in May 2019 and brought Switzerland into line with internationally accepted tax rules. The reform has come a long way and was rooted in strong opposition by the EU and OECD against Switzerland’s treatment of so-called holding, mixed, and domiciliary companies. It was mostly multinationals that benefited from preferential rates on foreign revenue opposite domestic income. As a result of the abolishment of the privileged tax regimes, affected companies would have faced a significant increase in their tax burden. In order to prevent an erosion of the tax base, the reform introduced new, internationally accepted tax instruments (e.g. patent box or discounts for R&D spending) and most cantons cut corporate tax rates. Table 3 offers an overview of cantonal changes in corporate tax rates and the respective allowances for different tax instruments.

With increased mobility of many firms in the wake of globalization and the subsequent effects on (international) tax competition, the interest in tax sensitivity of corporate income and its mobility across jurisdictions has increased too. The only study assessing the tax elasticity of corporate income in Switzerland finds a semi-elasticity of −2.2, i.e., an increase in the local corporate tax rate by 1 percentage point leads to a decrease in reported corporate income by 2.2% (Staubli 2018). In addition, tax sensitivity depends on the level of the corporate tax rate: The semi-elasticity is higher for municipalities with relatively low and relatively high-tax rates, whereas municipalities with medium tax rates record lower semi-elasticities.

7 Redistribution Framework

Fiscal equalization constitutes another essential element of Swiss fiscal federalism. Given the profound differences in tax burdens and fiscal capacity, the institutional framework can be seen as the counterpart to tax competition ensuring redistribution through various features. The highly progressive federal income tax and the withholding tax on capital incomes have been mentioned previously in this article.

In addition, there is substantial redistribution through the (first) pay-as-you-go pillar of the pension system. Contributions are proportional to labor income and of pure tax character above CHF 85,000 (i.e. there is no upper limit in contrast to many other countries) as this is the threshold that defines the maximum insurance which, as of now, is set CHF 2,370 (3,525) for a single individual (couple) per month. The minimum insurance is CHF 1,185 per month. Almost 60% of recipients receive the maximum pension. The fourth element in the framework is the fiscal equalization system aimed at reducing regional disparities.

The current equalization scheme redistributes around 5.2 billion Swiss francs or 0.75% of Swiss GDP (2018) between the Confederation, high-income cantons, and low-income cantons. The endowment is determined by parliament every four years. The mechanism’s primary objective is to ensure that every (low-income) canton reaches 85% of the country’s average tax potential (the threshold has been increased to 86.5% starting in 2020).

7.1 The Widely Successful Reform of the Equalization Scheme

It has been in force since 2008 when it replaced an older system that was widely criticized for its poor incentive structure and small equalization effect relative to the overall transfer volume. In addition to restructuring the scope and volume of transfers, the reform also included the disentanglement of tasks that had been carried out by both the federal and cantonal governments often reducing cantons to pure enforcers of federal regulation. With respect to transfers, the reform included the following elements (Bessard 2013):

• The main redistribution mechanism, fiscal capacity or resources equalization, is now based on cantonal tax potential, defined as standardized (opposed to effective) tax revenues over a six-year period from income, wealth, and profits in relation to taxable income and wealth. The barely straightforward indicator was introduced because, under the old system, cantons could directly manipulate their fiscal capacity through tax policy, for example. Resources equalization amounts to about 80% of total transfers. Within the pot, the federal government transfers about 2.5 billion francs to financially “weak” cantons (19) to increase their fiscal capacity, whereas the seven financially “strong” cantons contribute about 1.7 billion francs (2019 figures).

• Charges equalization, which was not separated from resources equalization prior to the reform, considers financing needs for various specific factors that can be pooled in three categories: equalization for (1) geographic factors such as mountainous, unproductive areas, socio-economic factors faced by (2) urban agglomeration or (3) large city centers (by Swiss criteria). The division of funds between geographic-topographic and socio-economic categories (1:1) is the result of political bargaining. Given the comfortable majority of cantons benefiting from transfers equalizing geographic factors, this is unlikely to change despite a study indicating that socio-economic factors constitute burden cantonal and local expenditure significantly higher (Ecoplan 2013). Charges equalization is entirely financed by the Confederation and amounts to about 14% of total transfers.

• In contrast to the old equalization scheme, the federal government can no longer earmark transfers meaning cantons are free with respect to the disposition of transfers. Before 2008, about three quarters of transfers were matching grants. As a result, grants were often independent of cantons’ fiscal performance.

• For political reasons, the reform included a “hardship” fund that reimburses cantons that were adversely affected by the introduction of the new mechanisms. As of 2019, six cantons remain recipients of this fund which will be terminated no later than 2035. It is financed by the other cantons (1/3) and the Confederation (2/3).

Figure 8 shows the average per capita equalization payments for each canton over the last five years. The wealthy and lower-tax cantons of Schwyz, Nidwalden, and Zug as well as the economic centers Zurich, Basel, and Geneva have the highest per capita contribution. The highest net recipients (Uri, Valais, Jura) are all situated in the periphery (by Swiss standards) and receive more than 2000 Swiss francs per capita. Considering the disparities in fiscal capacity, it is no surprise, that there are large differences in the recipient cantons’ dependency on equalization transfers too. The highest net recipients receive around 20% of their revenue directly from the equalization grants, whereas up to 45% of the contributing cantons’ revenue is redistributed.

Fig. 8

(Source Federal Finance Administration [2019])

Per capita equalization payments by fund 2015–2019

There is little doubt whether the reform improved the fiscal equalization scheme. Thanks to the elimination of matching grants and the new tax potential indicator, redistribution of resources work leading to a reduction in disparities. Since 2012 all cantons have reached 85 or more points on the tax potential index thus exceeding the mechanism’s main objective. The main beneficiaries of the redistribution effects were those 11 cantons whose tax potential was below the key threshold. The mechanism additionally redistributes funds to seven more cantons although their potential was higher than the threshold. This is due to the progressive allocation of funds at the lower end of the resource index as can be observed by the significant leveling at the lower end of the resource index in Fig. 9—in contrast to the linear absorption of resources from the contributing cantons at the top (Leisibach und Schaltegger 2019). As a result, contributing cantons have faced increasing liabilities to the system for years. In order to address this source of increasing tension the tax potential threshold (86.5) is now a statutory variable and no longer subject to parliamentary bargaining.

Fig. 9

(Source Leisibach and Schaltegger (2019), Fig. 4)

Resource equalization 2019

7.2 Disincentives and Flaws of Redistribution Mechanism

The construction of the redistribution mechanisms has a direct impact on cantons’ and (municipalities’) incentives in determining their tax policy (Brülhart and Schmidheiny 2015). It substantially restricts cantons’ fiscal benefits from attracting additional tax revenue. One way to quantify those restrictions is by calculating so-called marginal absorption rates, that is how much transfer payments change as a result of changes in fiscal capacity (e.g. as a result of changes in tax revenue). For fiscally weaker (stronger) cantons it is defined as the share by which transfers decrease (increase) for additional tax revenue.²⁰

Figure 10 shows the marginal absorption rates for each canton in 2019 (Leisibach und Schaltegger 2019). For example, an increase in resource potential by 100 francs in the canton of Schwyz, increases its contribution by 5 francs regardless of whether or how much the canton exploits the newly gained potential. Most fiscally weak cantons face higher absorption rates owing to the progressivity of the mechanism explained above. On average, the (weighted) average absorption rates for those cantons amount to (13,4) 16%. With the exception of Obwalden, that adopted a bold tax reform in 2008 introducing a flat-rate tax on personal incomes and reducing corporate tax rates by half, marginal absorption rates have remained stable since the introduction of the new fiscal equalization scheme over a decade ago.

Fig. 10

(Source Leisibach and Schaltegger [2019], Fig. 5)

Marginal absorption rates for new resources in 2019 in percent for contributing (red) and receiving (blue) cantons

The debate on these disincentives is ongoing. They certainly have not undermined tax competition which has been relatively intense since the last reform. There is also good economic reasoning why high marginal absorption rates are justified—namely if cantons attract corporate profits from other cantons instead of from abroad. And while under the current mechanism disincentives could only be avoided if corporate profits would be excluded from resource potential, the implicit tax rates on new resources do reduce the incentives to maintain revenue growth considerably with unknown long-term consequences. The fact that fiscally weak cantons are affected stronger relative to fiscally strong cantons may increase disparities in the future and indirectly strengthen support for uniform corporate taxes. This problem is amplified by a recent decrease in taxes on corporate profits in almost all cantons as a reaction to the recent abolition of privileged tax regimes (see V. Public revenue). A fundamental reform of the resource equalization and its mechanism could address disincentives in the interest of the entire federal system.

7.3 The Effect of Direct Democracy and Fiscal Federalism on Income Inequality

Spatial sorting is often said not only to affect regional disparities but also the distribution of income among citizens. The fact that in a federal system tax bases are mobile lead exponents of the traditional theory of federalism to favor the assignment of distributional responsibilities at the federal level (e.g. Musgrave 1959; Oates 1972, etc.). This ideal, however, is a long way from the reality of fiscal structures—in Switzerland and elsewhere. Many responsibilities on the cantonal level have redistributive character.

It seems that this has not been to Switzerland’s disadvantage as an international comparison of inequality outcomes show. Inequality in market incomes is low compared to most other industrialized countries reducing the need to redistribute and the associated efficiency losses (Frey et al. 2019). Inequality in disposable incomes after redistribution is equivalent to the OECD median.²¹ A similar pattern is observable for poverty rates.

Recent evidence shows that fiscal federalism and direct democracy play a systematic role in those favorable outcomes as they largely determine how fundamental economic trends translate into income inequality. Fiscal federalism actually reduces income concentration conditional on jurisdictional fragmentation, i.e., the inequality decreasing effect of decentralization becomes weaker the smaller municipalities are (Feld et al. 2021). Similarly, the voter initiative is associated with a significant decrease in top incomes to the benefit of the upper middle class, whereas referendums seem to have an opposing effect (Frey and Schaltegger 2019). These effects are observed in market outcomes already meaning the results are not driven by shifts in redistribution via income taxes.

The public health and economic response to COVID-19 in Switzerland

Switzerland’s first COVID-19 case was discovered on February 25, 2020 in Ticino, the Swiss canton closest to some of the Italian regions that were most significantly affected by the virus. By May 15, Switzerland had confirmed more than 30,400 positive cases and 1879 deaths due to COVID-19 (see www.corona-data.ch for the latest figures). There are striking spatial differences with more than 600 positive cases per 100,000 residents in Ticino, Vaud, or Genera opposite 150 to 300 positive cases in most German-speaking cantons.

The Swiss public health response has largely been based on the Law on Epidemics adopted in 2012 that grants the federal government—in consultation with the cantons—the authority for the interference into public life in case of an unusual or extraordinary epidemiological situation. A few cantons made use of the statutory provision that allows them to extend federal measures, most notably Ticino that declared a state of emergency five days prior to the federal response on March 16. The Swiss Federal Council adopted measures similar to those in other countries including the ban of all private and public events and gatherings of more than five people. With the exception of grocery stores, banks, pharmacies and post offices, stores were closed for six weeks. By mid-April, a three-step exit strategy was announced to be employed until the beginning of June. Ticino was the only canton to extend the state of emergency until the beginning of May.

To tackle what is likely to develop into the worst economic crisis since World War II, the federal government has implemented a variety of measures concerning liquidity bottlenecks, monetary and macro-financial stability, employment, compensation for self-employed as well as the culture or sports sector. In total, it has set aside around 65 billion Swiss francs in support of the economy, an amount almost equivalent to its annual budget. Many cantons have adopted additional stabilization measures in favor of areas not covered by federal efforts. In light of the many uncertainties regarding the virus It is unclear to what extent the commitments will exert pressure on the existing landscape of fiscal institutions on different levels of government.

8 Conclusion

Fiscal federalism and direct democracy are key parts of the Swiss political DNA that have contributed to political stability and economic success. They are associated with several fiscal benefits such as lower debt levels, lower government expenditure, or the prevention of an excess supply of government services. The widespread autonomy of subnational jurisdictions with respect to fiscal power as well as participatory institutions creates a framework that has proved conducive to increased citizen impact on political outcomes.

Recent empirical evidence has significantly advanced the understanding of the effects of decentralized tax autonomy and its interdependencies. These largely favorable outcomes place Switzerland at top positions by international comparison and provide a universal economic rationale for a fiscal hierarchy and subsequent tax competition. Yet the underlying equilibrium between competitive and cooperative elements of Swiss fiscal federalism is more fragile than it seems.

This is exemplified by the recent corporate income tax reform and subsequent cantonal changes in tax rates. The current mechanism of the fiscal equalization scheme partly impedes a level playing field for cantonal tax competition as fiscally weaker cantons are disproportionally affected by disincentives rooted in the mechanism design. While not all disincentives are unjustified from an economic point of view, their long-term effects on the system are uncertain. Specifically, they could undermine the public’s trust in the system if disparities increase and lead contributing cantons to question the scheme’s solidarity as the funds to redistribute continue growing. For now, a bold reform is politically unrealistic given the receiving cantons solid majority in both councils in the federal assembly. In the medium term, however, this moral hazard is likely to weaken the effectiveness of the equalization scheme at best.

There is also increasing pressure on the system with respect to the division of responsibilities between different levels of government. Some call for another, more profound disentanglement to correct for what was missed by the 2008 reform. Others see most policy solutions at the federal level, possibly because the federal system’s potential benefits are of dynamic nature (e.g. the taming effect on the Leviathan state, the competition of ideas), whereas its potential drawbacks (e.g. coordination costs, economies of scale not fully exploited) are static and therefore more easily observable (Rother and Rühli 2017).

In order to preserve the principle of subsidiarity and not adopt a system in which cantons are reduced to executing federal regulation, advocating another reform is necessary. Importantly, such an institutional revision should prevent the cantons from exchanging responsibilities for reimbursements or grants. Instead, a decentral shift in responsibilities should be combined with a decentral shift in tax autonomy ensuring that cantons have enough revenue to finance their expenditure.

Integrated tasks will exist beyond a reform reflecting the complex reality of fiscal structures. Ideally, federal legislation sets a regulating framework without interfering with specific cantonal autonomy. For example, health insurance providers need to be regulated on the federal market to guarantee access to the domestic market but standardizing the financing of hospitals leads to competition for subsidies instead of innovative financing solutions. Furthermore, cantonal conferences should refrain from legislative action in order to uphold the principle of fiscal equivalence and direct democracy and ensure the proper working of procedural provisions. Instead, they could adopt an “IMF”-like role: support cantons in the provision of public goods and services by assisting them with knowledge and developing benchmarks.