The Swiss Federal Debt Brake
Vera Z. Eichenauer and Jan-Egbert Sturm
Preliminary version. Please do not quote without permission
The Swiss federal debt brake pursues two goals: ensure medium- and long-term debt stabilization by avert (chronic) structural imbalances and grant short-term countercyclical budget leeway. The federal debt brake is a constitutional rule that was introduced to correct aspects of the budget process that were perceived to be flawed. Its minimum goal is the business-cycle adjusted equalization of revenues and expenditures. The fiscal rule concerns only the expenditure side; given expected tax revenue subject to a business cycle correction, the budget must be balanced. In principle, the Swiss Confederation can also act on the revenue side but constitutional upper limits on the main tax rates do not allow for micro-management of revenues. The debt brake was first applied in the federal budget 2003 following a steep debt accumulation in the 1990s.
In the decade prior the introduction of the federal debt brake, Switzerland went through a period of recession and stagnation2￼ Between 1990 and 1998, federal debt tripled starting from a debt level of 39 billion and a debt ratio of 10 % (Figure 1). In 1998, it amounted to over 110 billion Swiss francs, or more than 25% of gross domestic product (GDP). During this period of debt accumulation, the existing constitutional rule introduced in 1958 about the compensation of budget deficits proved largely toothless. The debt brake was introduced when other policy measures to curb the growth of the debt ratio were judged insufficient.
The debt brake is anchored in the Federal Constitution, which cannot be changed without a popular vote. The population accepted the introduction of the fiscal rule with an 84.7% Yes-share
- This chapter draws extensively from Sturm et al. (2017).
- An important cause for the rapidly growing debt were the economic stagnation phase (1991-1996) and the subsequent costs of the spin-off and financing of pension funds and federal enterprises (1998-2003). However, deficits were to a considerable degree structural.
1 of 12
and a significant majority in all cantons in 2001. Previously, the proposal passed both legislative chambers with majorities of more than two-third. These strong majorities demonstrate the widely accepted belief that federal public finances need to be tightly and conservatively regulated to ensure long-term fiscal sustainability.
The debt brake treats deficits and surpluses in the budget differently. While the former need to be compensated, the later are automatically used for debt reduction. This modification was made by parliament to the original (symmetric) proposal by the federal government. The debt brake was an immediate success: it stopped debt accumulation and even reduced debt. From 2004-2019, the federal government’s gross debt fell from CHF 124 billion to CHF 88 billion. The main reasons for the absolute reduction in debt were regular budget underruns and the systematic underestimation of revenues. From 2006 to 2019, the federal revenues were larger than the expenditures in almost every year. In combination with economic growth from 2005 to 2019, the federal government’s debt ratio fell from 26% of GDP to 13%. As of 2019, the debt brake had almost reversed the buildup of the debt ratio during the 1990s.3￼
The federal government’s debt ratio has halved between the introduction of the federal debt brake in 2003 and 2019. For international comparisons and for the overall assessment of fiscal policy, the overall public debt ratio is relevant. Switzerland’s public debt ratio has fallen from 46% in 2003 to 26% in 2019. Since 2009, the debt ratio at the cantonal level has slightly increased while the municipal debt ratio has slightly dropped since a peak in the mid-2010s (Figure 1). It is likely that these subsidiary federal levels have partly compensated for the decline in the federal expenditures, but this is difficult to assess rigorously. As all 26 cantons have their own debt brakes in place,4 the small increase in the cantonal debt ratio is not necessarily worrisome.
- In contrast to many cantons, however, the federal government still has no positive equity capital.
- Detailed descriptions can be found on the website of the Conference of the Cantonal Finance
Ministers https://www.fdk-cdf.ch/-/media/FDK_CDF/Dokumente/Themen/Finanzpolitik/Haushaltsregeln/181129_HH_Regeln_ 18_UPDATE.pdf?la=de-CH (18/11/20).
2 of 12
Figure 1: Development of the debt-to-GDP ratio at all political levels and forecasts
Notes: 2020-2022 are projections from 22 October 2020 using recent GDP estimates that exclude value added associated with (capital inflows following major international) sport activities. Source: KOF Swiss Economic Institute (2020)
Figure 2: Development of the gross debt of the Confederation, 1990-2019, in million CHF
Notes: GDP estimates that exclude value added associated with (capital inflows following major international) sport activities. Gross debt and the level of the compensation account in million nominal CHF. The yearly average value of the USD in 2019 was 0,99 CHF.
Source: Swiss Federal Administration and State Secretariat for Economic Affairs Switzerland.
Despite its obvious success, the debt brake at the federal level has been criticized. The main criticism regards the systematic budget underruns that it appears to have created. A 2017 report on the topic commissioned by the government and headed by Jan-Egbert Sturm, one of the authors of this chapter, concluded that an adjustment of the regulatory framework could undermine trust in the debt brake and thus its effectiveness. As a moral hazard, it could furthermore provoke follow-
3 of 12
up demands. The commission only analyzed the expenditure side of the budget underruns in detail as they considered the measures undertaken to improve the estimation of revenues and the reform of government budgeting as likely to be sufficient to eliminate systematic budget underruns in the future. A 2012 public opinion survey demonstrated that the debt brake remains very popular among the population despite the systematic budget underruns. This strengthens the political commitment to respect the debt brake and thus its effectiveness.
In the remainder of the article, we will describe the background and introduction of the debt brake (section 2) and the design of the Swiss debt brake (section 3). Section 4 discusses recent reforms, discusses the explanations for the budget underruns and other challenges surrounding the federal debt brake. Section 5 concludes.
2 Background for the creation of the debt brake
Switzerland is a strongly federalist country. The competences of the Swiss federal government are delegated from the cantons and are limited compared to most OECD countries. The maximum tax rates of main revenue components are determined by the Constitution which can only be changed by popular vote. The Swiss federation accounts for 30% (2000-2019) of government expenditures in Switzerland. The largest share of government expenditures occurs at the cantonal level which is responsible for basic and higher education, police, health, and social security. On average, 40% of government expenditures are made at the cantonal level while municipal level expenditures account for 20%.5
Switzerland is a fiscally conservative country with a history of fiscal rules. After the two world wars, the federal government’s debt amounted to 9 billion Swiss francs, which corresponded to about 50% of the gross domestic product at that time. In 1959, a deficit rule was introduced into the Swiss constitution.6 High federal surpluses during the 1960s led to a sharp reduction in the debt ratio. Influenced by the worldwide oil crises, this trend was interrupted in the second half of the 1970s. During the 1980s, a further decline was recorded. The constitutional anchoring of debt relief could not stop the increase in debt during the recession and stagnation period of the 1990s. The Swiss National Bank responded to the 1987 financial crisis with expansionary policies but reversed its approach after inflation surpassed 6 percent in 1990. The tight monetary policy increased interest rates, led to an appreciation of the Swiss franc, the burst of a real estate bubble,
- Cantons have far reaching autonomy with respect to personal and corporate income taxes and the exclusive competence for property and inheritance taxes. Considering the (rather loose) restrictions of the Federal Constitution, they are free to set their own tax schedules. Municipalities also have their own revenues and can set their own tax schedules within the cantonal rules. While vertical and horizontal redistribution within and between cantons takes place, its extent is limited and guided by clear rules.
- The following article was inserted into the Federal Constitution by the referendum of 11 May 1958 “Art. 42a. The deficit in the balance sheet of the Confederation shall be deducted. In doing so, the situation of the economy shall be taken into account.”
4 of 12
and to the collapse and acquisitions of several regional banks. In a 1992 referendum, the Swiss narrowly opposed to joining the European Economic Area. As a combination of these factors, the unemployment rate during several years in the mid-1990s was more than four percentage points higher than during the 1980s. This led to high and unprecedented federal expenditures for unemployment benefits.
The expansion of federal debt was perceived to reflect flaws in the budget process. First, there was a spending bias coupled with a pro-cyclical fiscal policy: economic upturns were not used as an opportunity for fiscal consolidation. Second, in order to raise revenues, a constitutional amendment is usually required. Spending increases, on the other hand, require only a simple majority vote in parliament.
The steep rise in the debt ratio during the 1990s was criticized early and acted upon fast. The debt accumulation was first met with several austerity packages in 1992-1994. In 1995, a brake on expenditures was introduced. Subsidy provisions, commitment credits and payment frameworks entailing new one-time expenditures of more than CHF 20 million, or new recurrent expenditures of more than CHF 2 million required the approval of the majority in both chambers of the federal parliament. These majorities are difficult to obtain in the Swiss governmental system with a consensus rather than a majority government.
The austerity efforts and the spending brake were not sufficient to significantly reduce the budget deficit. Moreover, some economists and economic historians argued that the austerity measures at the cantonal and federal level prolonged the economic crisis.
In 1997, the Finance Ministry started planning the debt brake. As a necessary condition for its introduction, a structurally largely balanced budget was required. Against this background, a budget target for 2001 was formulated in 1997: The deficit in the financial accounts was to be reduced gradually from 1999, once the economy had recovered and stabilized, and not exceed 2
- of revenues until the 2001 fiscal year. In order to achieve the budget consolidation aimed for with the 2001 budget target, a new, temporary transitional provision was created in the federal constitution that obligated the Federal Council and Parliament to make savings with a binding deadline if the target was not met. The proposal for the 2001 budget target was accepted by 70.7
- of the people in a 1998 referendum. Subsequently, work began on the design of the federal debt brake which was first applied in the 2003 federal budget.
The government report accompanying the original proposal makes clear that the primary goal of the regulation is to stabilize the federal debt with the secondary goal of granting short-term countercyclical budget leeway.7 To achieve these goals, the government proposed to symmetrically compensate for surpluses or deficits. According to the proposal, the balancing of the budget would have required either an increase of the expenditure ceilings or a reduction of revenues in the following years. The reason for the planned symmetrical management was a
- BBl 2000 4653, 4686
5 of 12
proposed compensation account as an instrument to compensate for estimation errors in revenues. In the absence of systematic forecast errors for revenues, the balance of the compensation account would fluctuate around a stable level. The sanction in the event of a major deficit prevents a systematic overestimation of revenues and thus a constant violation of the debt brake. Conversely, the debt brake would have been overfulfilled in the event of systematic underestimation. The funds in the compensation account could then have been used to increase the expenditure ceiling or reduce taxes.
Parliament did not endorse this proposal and allowed for a more ambitious goal than debt stabilization. There should be the possibility of using surpluses to pay off the debt. In contrast to the original proposal, surpluses cannot be used to increase the expenditure ceiling or to reduce taxes. They thus lead to a reduction in debt. No upper limit was set for the positive balance of the compensation account. The design of the compensation account is thus asymmetric, intended for the case of a negative balance only. A surplus in the compensation account is continued and the obligation to reduce the surplus does not apply. While the constitutional framework permits symmetrical management of the compensation account, the current Financial Budget Act is more restrictive. Parliament thus gave priority to debt reduction as opposed to spending increases or tax relief.
3 The design of the debt brake
The Swiss debt brake has two features that we want to describe in more detail because we believe them to be rather unique in international comparison. First, the current Financial Budget Act (FHG) implements the constitutional provision of the debt brake in an asymmetric way. Deficits and surpluses are not treated in the same way. While the former need to be compensated, the later are automatically used for debt reduction. Second, the Swiss debt brake cyclically adjusts revenues but not expenditures. The aim is to stabilize the expenditure path over time. This means that, in contrast to the revenue side, no short-run economic impulses are intended to come from the expenditure side of the federal budget. Tax revenues are supposed to act as automatic stabilizers. In this context, it is important to note that significant expenditure-side automatic stabilizer that react very strongly to economic fluctuations, such as unemployment insurance (as well as other social insurances) have been recorded separately from federal accounts since 2002. They are therefore not subject to the debt brake rule.8 Finally, it is important to note that constitutional upper limits to federal taxes makes it difficult to adjust tax rates. Changes to tax levels must be accepted by the voters in an optional or even mandatory referendum.
- If the debt level of the insurance increases above a defined threshold, contribution rates are raised as a default measure.
6 of 12
The expenditure ceiling determined by the debt brake corresponds to the cyclically adjusted revenues. During the budget process, the Swiss debt brake is applied twice: first to budget forecasts, then to effective outcomes. It is the second calculation that determines the deviations that must be credited or debited in the compensation account. If the deficit exceeds 6 % of expenditure, the excessive amount must be eliminated within the next three annual budgets by lowering the expenditure ceilings.
More formally, the expenditure ceiling required by the debt rule is Gt = kt Rt where k is the so-called economic factor that adjusts for economic movements on the revenue side. The economic factor k corresponds to the quotient of the estimated real GDP according to the long-term smoothed trend and the expected real GDP in the forecast year: kt = Yt/Yt*. For smoothing the GDP series, a modified HP filter is applied. This method has advantages with respect to alternative measures (e.g. production functions) in terms of transparency and symmetry. In addition, it requires only a small number of hypotheses about the future development of output or production factors. The filter method has the further advantage of yielding a symmetrical value of the output gap over time. A disadvantage of the HP filter is that it is known to suffer from a lack of smoothing properties at the end of a series that is to be smoothed. This problem can be handled either by using forecasts, so that the trend-calculation does not occur at the end of the series or by a modification of weights within the filter. The latter approach was preferred for the Swiss debt brake, as the use of forecasts actually magnified the problem. The modified HP filter interprets a (forecasted) change in GDP during the budget year to be around 80 % cyclical and 20 % structural. The forecasting error regarding the output gap will translate into corresponding errors in both revenue and forecasts of the cyclical adjustment factor k. However, these errors will have opposite signs and generally cancel each other out within the debt brake equation.
In extraordinary circumstances, e.g. severe recessions or natural disasters, a debt-financed increase in the expenditure ceiling is possible. Apart from legal restrictions, extraordinary expenditures must be accepted as such by a qualified majority in both houses of parliament and therefore require a larger consensus than ordinary expenditure items. Such expenditures are debited to a special “amortization account” (not to be confused with the compensation account) and generally must be compensated for during the following six years.
The implementation of the debt brake led to several unexpected side-effects that were dealt by introducing new accounts and improving the calculations of revenue forecasts.
4 Reforms and challenges
4.1 Amortization account: Extraordinary budget receipts and expenditures
7 of 12
Since 2010, the extraordinary budget is subject to the debt brake too. The extended debt brake rule requires deficits in the extraordinary budget to be offset via the ordinary budget in the medium term. Since 2010, extraordinary receipts (without earmarking) and expenditure shown in the Federal accounts will be credited or debited to an amortization account held outside the Federal accounts. Shortfalls in the amortization account must normally be compensated within six financial years by means of surpluses in the ordinary budget. In the event of foreseeable deficits in the amortization account (and provided that the compensation account does not have a negative balance that needs to be corrected first), the maximum amount of ordinary expenditure may be reduced as a precautionary measure.
4.2 Compensation account: Repeated budget underruns
The current implementation of the debt brake led to an increasing balance of the compensation account and thus to the nominal decrease of Swiss federal debt (Figure 2).9 The compensation account is credited (debited) if the actually incurred expenditures are below (above) the maximum permissible expenditure. The expenditure ceiling is recalculated based on the closing accounts for the receipts achieved and the revised economic outlook. Therefore, the balance of the compensation account is also influenced by forecasting errors regarding receipts and economic growth, as these lead the estimated expenditure ceiling relevant for the budgetary planning phase to be too high or too low.
Since 2006, the revenue growth was under- and the expenditure growth overestimated. During the time period 2004-2019, forecast errors for revenues explained 52% of the inflows into the compensation account (whereof 44% were due to ‘Anticipatory Taxes’), forecast errors for expenditures 40% and the remaining 8% are due to requirements of the debt brake and estimation errors related to it. According to the expert report on the debt brake (Sturm et al. 2017), civil servants expected the positive bias in the estimation of ‘ordinary revenues’ and especially the revenues from the ‘withholding tax’ to disappear after 2012 when an improved forecasting model was implemented.10The claim was not evaluated in the 2017 report on budget underruns or at any point thereafter.￼ In this chapter, we examine for the first time the evolution of revenue forecasts since the introduction of the debt brake.
We provide a systematic evaluation of the differences between the estimates and the realized budget values for the expenditure side (in Table 1) and for revenues (Table 2). We present the difference between estimates and the realized values both Swiss Francs and as share of the estimates for two different periods. The period 2004-2019 includes all budget years after the first application of the
- Note that Federal Finance Administration reports that “[t]he compensation account was set to zero in 2006 (reduction path), reduced by 1 billion in 2010 (introduction of the supplementary rule), and reduced by around 4.4 billion and 1.9 billion in 2016 and 2018, respectively, due to accounting changes.” (translation by the authors from Table Notes).
- The FFA (2019) investigates only the evoluation of budget underruns on the expenditure side up until 2018.
8 of 12
debt brake in 2003.11 For the expenditure side, we show average differences for 2007-2019 period. In the context of the expert report, civil servants reported their concerns about the estimation of the budget sub-item ‘own expenditures’ which has been available only since 2007.
Regarding the forecast errors related to expenditure, Table 1 shows that expenditures incurred during the financial year are consistently lower than the expenditure approved by the parliament in the budgetary process. Although some budget items are increased during the financial year, the unused budget lines are more significant. The table shows that the expenditures for interest payments are significantly overestimated for both of the periods analyzed.
Table 1: Difference between the realized federal budget and the budget estimate for ordinary expenditures
|million CHF||% of the budget estimate|
|Ø 2004-2019||Ø 2007-2019||Ø 2004-2019||Ø 2007-2019|
|Transfers’ to third parties12||-194||-199||-2.5%||-2.6%|
|Other ordinary expenditures||956||988||1.9%||1.9%|
|thereof own expenditures13||440||4.2%|
Notes: A positive (negative) value in Table 1 indicates an underrun (overrun) of the expenditure side of the budget. Standard errors of the respective realized budget deviations in brackets. Source: Swiss Federal Finance Administration.
Table 2 displays the differences between the estimated revenues and the actual receipts for the first time. Our main interest is the evolution of the forecasting error of the ordinary revenues and in particular of the withholding tax following the application of an improved estimation method applied since the budget year 2012. We find that the average forecasting error remains positive in the budget years 2012-2019 but was reduced compared to 2004-2011. The relative reduction in the budget deviation is most pronounced for the withholding tax which drops from 43% to 17%. Looking at the absolute values estimated and realized receipts, receipts from the withholding tax are still underestimated by almost 1 billion in the 2012-2019, although this is an improvement compared to the average 1.3 billion in the 2004-2012 period. Our analysis thus suggests that the
- In 2003, the compensation account was not credited with any of the surpluses i.e. we assume that the implementation of the debt brake was not fully realized in this first year.
- The largest share of federal revenues to third parties are distributed to cantons.
- Own expenditures include personnel, the defense budget, amortization, material, and other operating expenses.
9 of 12
expectations of civil servants in the Federal Finance Department that the forecasting errors for the revenue side and the withholding tax would become more symmetric were partially realized.
Finally, a comparison of the standard deviations for the differences between estimations and realized values for expenditures and revenues reveals that forecast errors for revenues were more severe but were reduced substantially reduced in the more recent time period.
Table 2: Difference between the realized federal budget and the estimate for ordinary revenues
|million CHF||% of the budget estimate|
|Ø 2004-2011||Ø 2012-2019||Ø 2004-2011||Ø 2012-2019|
Notes: A positive (negative) value in Table 1 indicates a surplus (deficit) of the revenue side of the budget. Standard errors of the respective realized budget deviations in brackets. Source: Swiss Federal Finance Administration.
Regarding the explanations for budget underruns, the government-commissioned report (Sturm et al. 2017) notes four possible reasons: two technical explanations and two hypotheses about the behavior of the administration.
First, the realized costs for the fulfillment of government tasks or the quantitative demand for federal services could be lower than expected. Over the medium- to long-run, such forecast errors should balance out. The last decade was characterized by economic events with consequences for the federal public finances that are difficult to forecast. Since the financial crisis in 2008/09, the forecasts for economic growth as well as interest and inflations rates deviated consistently from their realizations. Interest rates on government bonds were lower than expected and the compensation for inflation in the federal financial planning was set too high. Although the compensation for inflation was corrected downward multiple times, unplanned expenditures increases of seven percent in real terms were realized from 2009-2016.
Second, expenditures might be postponed to the next fiscal year, e.g. for technical reasons. The government may carry over budgeted and supplementary credits, delayed disbursements for planned investments and certain other types of expenditures to the next fiscal year. However, the possibility to carry-over the budget is associated with administrative costs and may therefore not be fully exploited, leaving surpluses in the running year.
Third, the precautionary principle may dominate in the public administration. Budgetary hurdles or social and cultural customs may create an asymmetry in the costs occurred for budget overruns
- Withholding tax (Verrechnungssteuer) refers to the anticipatory tax on the payment of interest and dividends.
10 of 12
than for underruns. In principle, the credits defined in the budgeting process must not be exceeded. According to Swiss federal budgeting, supplementary credit can only be requested due to an unforeseen event. Since future needs cannot be forecast accurately and budget overruns are more costly in terms of administrative expenses and reputation than budget underruns, over-budgeting tends to be assumed in the literature as a precautionary motive. Administrative units might implicitly reserve potential savings to avoid supplementary credits, for future additional expenditure, or for future budget cuts. These asymmetries in the budgeting phase can lead to a tendency for administrative units to adopt (over)cautious budgeting practices but also create the intended incentive to use funds economically). Budget residuals resulting from precautionary considerations may well be an expression of an efficient budgeting process, as forecasting expenditures tends to involve uncertainty about costs or demands. In order to reduce budget underruns due to the precautionary motive, the “costs” of the actors involved in the budget process for exceeding and falling short of the budget would have to be symmetrical.
Fourth, budget underruns can be due to asymmetries in the institutional rules that address the problem of the fiscal commons in the budgeting process and during budget execution. Control mechanisms for budget execution tend to be stricter for practical reasons. To diminish this asymmetry, the role of the finance department during the budgeting process could be further strengthened. Another possibility would be to sanction budget underruns, but this would diminish incentives for an economical budget management and might increase the “December fever” phenomenon of heightened spending at the end of the fiscal year.
5 Concluding remarks
At the beginning of the SARS-CoV-2 pandemic, Swiss federal public finances were sold. This allowed the federal government to respond rapidly and, at during the first infection wave in spring 2020, generously to the economic supply and especially demand shocks triggered by the restrictive health regulations in Switzerland and in many relevant markets abroad. Overall, the fiscal deficit for 2020 is reckoned to be 20 billion (as of 28 October 2020) but considerable uncertainty remains. This is less than the extraordinary expenditure envisaged in the addenda (30.9 billion).
Despite a strong second wave of COVID-19 infections, the (conservative) finance minister Ueli Maurer and parts of the economic elite have argued that Switzerland cannot and should not provide similar financial support to the economy as during the first wave in spring 2020. The importance of healthy federal finances is also exemplified by the early start of the discussion about respect of the debt brake during and after the pandemic.
Nevertheless, debt will increase substantially in 2020. While the law provides for a period of six years for the reduction of this debt, this is hardly realistic. The parliament can decide to extend the repayment period. Prominent public finance experts in the Swiss-German part of the country and the Director of the Federal Finance Administration argue for a reduction of expenditures over the next decade. Temporary tax increases, which would require a constitutional amendment subject to a (voluntary) popular referendum, or higher debt levels are not discussed although the yearly average interest rate on short-term Swiss government bonds was –0.5% in 2019.
11 of 12
Proposals to modify the federal debt brake, e.g., to a symmetric management of deficits and surpluses, or any other proposals for debt accumulation or tax increases are highly unlikely to gain majority approval in the federal executive, the parliament or from the people. This is is an almost unanimous opinion that the federal government must expect a sharp increase in spending in the coming decades because of the commitments it has made and due to the high level of implicit debt without structural legislative revisions (FFA 2016).
- FFA (2019). Bericht zur Entwicklung der Budgetunterschreitungen (no English version available), Federal Department of Finance. Available at: https://www.efv.admin.ch/efv/de/home/themen/finanzpolitik_grundlagen/schuldenbremse.h tml
- FFA (2016). Langfristperspektiven der öffentlichen Finanzen in der Schweiz 2016 (no English version available), Federal Department of Finance (FFA), Bern. Available at: https://www.efv.admin.ch/efv/de/home/themen/publikationen/oeko_grundlagenarb.html
- Sturm, Jan Egbert Sturm, Brülhart, Marius, Funk, Siegenthaler, Peter (2017). Gutachten zur Ergänzung Schuldenbremse (no English version available). ethz.ch/handle/20.500.11850/235655
Patricia, Schaltegger, Christoph A., der Schuldenbremse, Expertengruppe Available at: https://www.research-
12 of 12