## _sdn1309tn

## Source details

**Canonical URL:** [_sdn1309tn](https://www.imf.org/-/media/websites/imf/imported-publications/external/pubs/ft/sdn/2013/_sdn1309tn.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-publications/external/pubs/ft/sdn/2013/_sdn1309tn.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-publications/external/pubs/ft/sdn/2013/_sdn1309tn.pdf.json)

---

### Fiscal Risk Sharing — Motivation and Channels
- Motivation for risk sharing:
  - Smooths impact of temporary income shocks on national or regional consumption, analogous to household permanent income/consumption-smoothing.
  - Within a currency union, risk sharing is crucial because:
    - Countries cannot use independent monetary policy for country-specific shocks.
    - Wage and price rigidities and limited labor mobility reduce adjustment mechanisms.
    - Domestic fiscal policy may be constrained in crises (for example, credit markets freezing), limiting borrowing as a buffer.
  - Cross-country risk sharing should be limited to country-specific and temporary shocks; common synchronous shocks and permanent shocks are not appropriate targets for mutual insurance.
- Channels for smoothing consumption:
  - Saving and borrowing via international credit markets.
  - Private insurance via international capital markets (diversified asset portfolios or explicit insurance).
  - Fiscal risk sharing across countries (ex ante or ex post).
- Limitations of channels:
  - Credit markets: effective but often impaired during recessions as lending tightens and collateral values fall.
  - Private capital markets: limited by “home bias” and moral hazard/adverse selection for explicit insurance.
  - Fiscal risk sharing: can take the form of automatic stabilizers, centralized provision of public goods/services, and ex post bailout agreements (deposit insurance, common fiscal backstops for bank resolution).

### Empirical Evidence on Fiscal and Overall Risk Sharing
- Fiscal risk sharing estimates (literature):
  - Sachs and Sala-i-Martin (1991): federal taxes and transfers cushion 33-40 percent of a regional income shock (US, nine regions, 1970-80).
  - Von Hagen (1992): federal taxes and transfers insure about 10 percent of regional income risk (48 US states, 1981-86).
  - Broad summary: fiscal risk sharing tends to smooth between 15 and 30 percent of regional shocks; increases with central government expenditure share and in more centralized unitary states.
- Overall risk sharing (fiscal + market + credit):
  - Existing federations: about 80 percent of income shocks are smoothed when all channels are considered (Asdrubali, Sorensen, and Yosha (1996)-based studies).
  - Example (state-level US, Sorensen and Yosha, 1998):
    - Federal tax-transfer and grant system: about 15 percent smoothing.
    - National capital markets (private income and insurance flows): about 45 percent smoothing.
    - Credit markets (saving and borrowing): about 20 percent smoothing.
    - Remaining unsmoothed: about 20 percent.
  - Other federations (Canada, Germany): fiscal risk sharing between 10 and 25 percent; overall smoothing close to 80 percent.

### Risk Sharing in the Euro Area — Key Findings
- Level and composition:
  - Total risk sharing in the EU and the euro area is roughly half that seen in existing federations, with only about 40 percent of income shocks smoothed.
  - Fiscal risk sharing is almost nonexistent in both the EU and the euro area; the EU budget is about 1 percent of area-wide GDP and not designed for risk sharing.
  - Capital markets in the euro area provide less than half the amount of income insurance seen in the US, Canada or Germany, reflecting stronger “home bias”.
  - Aggregate portfolio diversification across euro area countries is far lower than among German regions or U.S. states.
  - Where buffering occurs, it is primarily through saving and borrowing in international credit markets.
- Time and cyclical dynamics:
  - About two third of country-level GDP shocks remains unsmoothed in the euro area.
  - Private saving and borrowing provides the largest amount of smoothing (about 20 percent) in normal times.
  - Degree of smoothing decreased over time:
    - Share of GDP shocks unsmoothed rose from 58 percent over 1979-99 to 66 percent over 1990-2010.
    - Decline largely reflects reduced cross-country smoothing via credit markets after the creation of the euro—credit flows became more pro-cyclical.
    - International capital markets smoothing rose somewhat with financial integration.
  - Risk sharing is particularly ineffective during recessions and sovereign stress:
    - Share of shocks unsmoothed in recessions is significantly larger than during normal times.
    - Europe’s share of unsmoothed shocks in 2008-10 was 76 percent (similar to previous severe downturns at 78 percent).
    - An increase of 100 basis points in the ten-year sovereign yield spread (versus the US ten-year rate) reduces the share of GDP shocks smoothed by about 5 percentage points.

### Illustration: Supranational Rainy-Day Stabilization Fund (Thought Experiment)
- Motivation and design:
  - Banking union and deeper financial integration could help market-based risk sharing but would take time; fiscal risk sharing can play a role in the interim.
  - Fund collects taxes as a share of GNP from each member and pays transfers to members hit by negative output shocks; contributions are saved when no member experiences a negative shock.
  - Transfers are proportional to the size of the shock, the relative size of the economy, and resources available in the fund; GNP used for scaling to control for capital market consumption smoothing.
  - Fund accumulates reserves for intertemporal insurance against common shocks as well as idiosyncratic shocks.
- Estimated contributions and stabilization potential:
  - To raise euro area stabilization close to levels common in federations (e.g., Germany where 80 percent of shocks are smoothed), annual contributions of about 1½ to 2½ percent of GNP could suffice.
  - The estimate depends on distinguishing temporary from permanent shocks; contributions would need to be higher if real-time distinction is not possible.
- Historical-simulation outcomes:
  - Had the fund existed since 1999:
    - Most euro area countries would have been net contributors until 2007 and net recipients during the recent crisis.
    - Smaller countries would have tended to receive larger transfers relative to economy size due to higher volatility.
    - For the entire period, average net contribution by each country would have been close to zero (exception: Luxembourg), indicating the scheme need not entail permanent transfers.
    - Periphery countries would have contributed relatively more than core countries until 2007 but received larger net transfers during the Great Recession.
  - Cross-country and intertemporal insurance:
    - In most years over the past three decades in the simulation, about a fourth to half of countries would have benefited from gross transfers.
    - The fund would have accumulated reserves to deal with rare large common shocks (examples: 1992-93 ERM crisis; dot.com bubble bust; two years after Lehman Brothers failure), at which point positive net transfers would be provided to almost all countries.

### Practical Challenges for Supranational Fiscal Risk Sharing
- Correctly identifying “insurable” shocks is the main challenge: shocks should be unrelated over time (not too persistent) and have zero mean to justify activation.
- Real-time identification is difficult; feasible compromises include operational techniques based on the output gap or deviation from trend growth.
- Risk of misclassification: a shock initially thought temporary could prove permanent, leading to persistent transfers.
- Some risk sharing functions could be institutionalized via existing mechanisms (e.g., centralized unemployment benefits), though these may involve implementation lags and only partially insure overall income risk.

### Fiscal Union: Common Practices in Federal States — Features and Lessons
- Benchmarks and comparators:
  - Seven federations studied as instructive comparators: Australia, Belgium, Brazil, Canada, Germany, Switzerland, U.S.
  - Key dimensions: centralized revenue and expenditure, social safety nets, banking backstops, debt issuance practices, fiscal rules and governance.
- Centralized revenue and spending:
  - Central government revenue shares:
    - "about half of general government revenue in Canada, Switzerland, and the U.S."
    - "two thirds in Germany"
    - "even higher shares in Australia and Belgium."
  - Central control of investment spending in the sample: "10 to 45 percent, and 20 percent on average".
- Social safety nets and banking backstops:
  - All sample federations provide federation-wide social safety nets and (explicit or implicit) federal banking backstops.
  - Unemployment benefits: systems range from fully centralized (Australia) to largely decentralized (United States); U.S. benefit duration ranges from "6 to 30 weeks"; federal emergency programs have extended benefits to "34 weeks and more" in crises.
  - Pension benefits: most non-U.S. countries have unified centrally-funded public pension systems; U.S. offers a federal minimum with state supplements averaging "29 percent" (single pensioners) and "50 percent" (couples) of the maximum federal benefit in some states.
  - Health insurance: funding varies—general taxes (Australia, Brazil), social contributions (Belgium, Germany), individual premiums (Switzerland), large private provision (U.S.).
  - Deposit insurance and bank resolution:
    - Most federations have nation-wide schemes; many have explicit fiscal backstops (Canada, U.S., Australia, Belgium) while Switzerland and Germany rely on private funding with ad hoc fiscal support when needed.
    - Canada’s CDIC: available funds at end-2011 were "0.4 percent of insured deposits" with target increase to "1 percent"; borrowing backstop up to "C$18 bn (1 percent of GDP)".
    - Switzerland’s deposit insurance fund capped at "CHF 6 bn (just over 1 percent of customer savings and deposits)"; no explicit fiscal backstop but federal ad hoc support occurred (UBS in 2008).
- Debt issuance and liability:
  - Public debt is issued separately by each government level; no federation in the sample permits joint and several liability of the central level backed by sub-national revenues.
- Fiscal rules and governance to limit moral hazard:
  - Common sub-national fiscal rules: balanced-budget requirements; many federations complement these with borrowing controls or oversight.
  - Empirical magnitude: in sample federations, about "of every one percent of a country-specific income shock, about 0.15 to 0.2 percent is smoothed through federal fiscal transfers".
  - Higher fiscal risk sharing tends to be paired with stronger fiscal governance and enforcement.
- Modes of enforcement and discipline:
  - Self-imposed (U.S., Canada): market discipline, credible no-bailout commitments, wide fiscal autonomy; rise of rainy-day funds (from 12 in 1982 to 45 in 1995).
  - Center-based (Brazil, Germany): federal oversight and strict rules after bailout episodes (Brazil’s Fiscal Responsibility Law; Germany’s constitutional debt-brake and Stability Council).
  - Cooperative (Australia, Belgium, Switzerland): intergovernmental coordination, peer pressure, direct democratic constraints.
- Lessons for the euro area:
  - Federations combine pooling of fiscal resources, centralization of expenditure (especially social protection), and banking backstops with governance/enforcement mechanisms.
  - The euro area lacks comparable ex ante fiscal risk sharing and has weaker governance.
  - Improvements could involve centralized fiscal capacity, common social safety nets, financial backstops, and governance frameworks reconciling solidarity with discipline.
  - Transition to credible no-bailout regimes is gradual; the Fiscal Compact and the “two-pack” are initial steps toward enhanced transparency and monitoring.

### Sources of Discipline, Enforcement Frameworks, and Implementation Examples
- Enforcement modes observed:
  - Market-sanctioned constraints and self-imposed rules (U.S., Canada).
  - Federal oversight and hierarchical enforcement (Germany, Brazil).
  - Cooperative negotiation and peer-pressure mechanisms (Australia, Belgium, Switzerland).
- Interplay with bailout expectations:
  - Credible no-bailout commitments incentivize sustainable sub-national fiscal policies; lack of credibility can lead to excessive borrowing and expectation of rescue.
  - When bailouts occur, centers often attach conditions and oversight to restore discipline.
- Illustrative institutional mechanisms:
  - Box 1 sequential game: center announces no-bailout; sub-national governments choose spending/borrowing; if bailout requested, center chooses to bail out (switch regime) or allow default (re-establish credibility).
  - Credibility of no-bailout influenced by scope for spillovers, extent of federation-wide risk sharing, and degree of sub-national tax authority.

### Appendix highlights — Banking union, common borrowing, and sub-national fiscal frameworks
- Banking union and deposit insurance features by federation:
  - Examples of arrangements preserved exactly:
    - Canada: CDIC with fiscal backstop; available funds at end-2011 "0.4 percent of insured deposits"; target "1 percent"; borrowing up to "C$18 bn (1 percent of GDP)".
    - Switzerland: deposit insurance fund capped at "CHF 6 bn (just over 1 percent of customer savings and deposits)"; banks required to hold secured domestic assets "125 percent of insured deposits" and liquid assets "50 percent of maximum potential contributions".
  - Euro area: banking union "none"; deposit insurance/bank resolution/supervision "no"; borrowing arrangements "separate"; joint and several "No"; central oversight: SGP "two" and "six" pack; bailout practices: ongoing programs.
- Sub-national fiscal frameworks (selected examples):
  - Australia: Loan Council coordination; balanced-budget rules; enforcement via peer pressure and market discipline.
  - Belgium: negotiated fiscal targets; federal law allows capping regional borrowing for two years.
  - Brazil: Fiscal Responsibility Law with debt ceilings, sanctions, and escape clauses.
  - Germany: Constitutional debt brake; borrowing linked to investment projects until 2020; enforcement via courts and conditional financial assistance.
  - United States: constitutional balanced-budget rules in most states; market discipline; no federal guarantee of sub-national debt.

### Conclusion and Policy Implications
- Long-term viability of a currency union depends on capacity to deal with country-specific shocks; risk sharing mechanisms for consumption smoothing are essential.
- Empirical studies indicate risk sharing in the euro area is limited via both market-based and fiscal channels and substantially lower than in federations such as Canada, the United States and Germany.
- Existing euro area mechanisms are particularly ineffective during downturns, increasing vulnerability to severe downturns and spillovers.
- A supranational fiscal stabilization mechanism (temporary-transfer rainy-day fund) could materially increase stabilization; a thought experiment suggests contributions of about 1½ – 2½ percent of GNP could raise euro area stabilization to levels similar to Germany, though effectiveness depends on real-time shock identification and political feasibility.
- Policy design must balance insurance with governance to limit moral hazard; political acceptance remains a significant constraint.

*Prepared by a team including Fabian Bornhorst, Esther Perez-Ruiz, John C. Bluedorn, Davide Furceri, Florence Jaumotte, Franziska Ohnsorge, Tigran Poghosyan, and Aleksandra Zdzienicka.*

### 1. Fiscal Risk Sharing ______________________________________________________________________________ 7

### 1. Fiscal Risk Sharing

### A. Motivation for Risk Sharing
- Risk sharing smooths the impact of temporary income shocks on national or regional consumption, analogous to household permanent income/consumption-smoothing.
- Within a currency union, risk sharing is especially important because:
  - Countries cannot use independent monetary policy to respond to country-specific shocks.
  - Wage and price rigidities and limited labor mobility can reduce adjustment mechanisms.
  - Domestic fiscal policy may be constrained in crises (for example, credit markets freezing), limiting borrowing as a buffer.
- Cross-country risk sharing should be limited to country-specific and temporary shocks:
  - Common synchronous shocks offer no scope for mutually beneficial risk sharing; aggregate monetary policy (and coordinated fiscal policy if needed) is preferable.
  - Permanent shocks require adjustment in consumption and would amount to redistribution, not risk sharing.
  - When the nature of shocks is uncertain, partial short-term insurance combined with partial adjustment may be appropriate, implying careful insurance mechanism design.

### B. Channels for Smoothing Consumption
- Three key channels for country-level consumption smoothing within a group of countries:
  - Saving and borrowing via international credit markets.
  - Private insurance via international capital markets (diversified asset portfolios or explicit insurance).
  - Fiscal risk sharing across countries (ex ante or ex post).
- Characteristics and limitations of channels:
  - Credit markets: effective but often impaired during recessions as lending tightens and collateral values fall.
  - Private capital markets: limited by “home bias” in portfolios and by moral hazard/adverse selection for explicit insurance; some sector-specific precedents exist (for example, oil price hedging).
  - Fiscal risk sharing: can include common tax and transfer mechanisms (automatic stabilizers), centralized provision of public goods and services (for example, a universal floor for unemployment insurance or common infrastructure investment), and ex post bailout agreements (deposit insurance, common fiscal backstops for bank resolution).

### C. Fiscal and Overall Risk Sharing: Evidence from the Literature
- Fiscal risk sharing estimates across studies and federations:
  - Early studies (Sachs and Sala-i-Martin, 1991): federal taxes and transfers cushion 33-40 percent of a regional income shock (US, nine regions, 1970-80). Caveat: approach may overestimate fiscal insurance because it does not separate redistributive effects or other channels.
  - Von Hagen (1992): using 48 US states for 1981-86, federal taxes and transfers insure about 10 percent of regional income risk.
  - Broad literature summary: fiscal risk sharing tends to smooth between 15 and 30 percent of regional shocks; it tends to increase with central government expenditure as a share of GDP and is larger within more centralized unitary states.
- Overall risk sharing (simultaneous estimation of fiscal and market channels):
  - Studies building on Asdrubali, Sorensen, and Yosha (1996) find that in existing federations about 80 percent of income shocks are smoothed when all channels (fiscal, capital markets, credit markets) are considered.
  - Example estimates (state-level US, Sorensen and Yosha, 1998): 
    - Federal tax-transfer and grant system smoothes around 15 percent of the income shock.
    - Income smoothing through national capital markets: about 45 percent (private income and insurance flows).
    - Smoothing through credit markets (saving and borrowing): about 20 percent.
    - Remaining unsmoothed: about 20 percent.
  - Other federations in sample (Canada, Germany) show similar patterns: fiscal risk sharing between 10 and 25 percent and overall smoothing close to 80 percent.

### D. Risk Sharing in the Euro Area — Key Findings and Simulations
- New empirical evidence in the source indicates:
  - Cross-country risk sharing in the euro area is limited — roughly half that seen in existing federations.
  - Risk sharing in the euro area falls sharply in severe downturns, just when it is most needed.
  - The level of risk sharing in the euro area is similar to that in the European Union, implying euro area members have not benefited from additional risk sharing despite giving up domestic monetary autonomy.
- Simulations on a euro area rainy-day fund:
  - A rainy-day fund financed by country contributions averaging 1½ -2½ percent of GNP per year could provide significant stabilization against country-specific income shocks.
  - The stabilization provided would be on par with the level of risk sharing seen in Germany and other federally organized countries.
  - Simulations indicate that all countries would be net beneficiaries of such a fund over the long term.

*Prepared by a team including Fabian Bornhorst, Esther Perez-Ruiz, John C. Bluedorn, Davide Furceri, Florence Jaumotte, Franziska Ohnsorge, Tigran Poghosyan, and Aleksandra Zdzienicka.*

### 11.      Risk sharing in the euro area. The estimates from the literature also indicate that the

### 11.      Risk sharing in the euro area. The estimates from the literature also indicate that the

### Key findings on overall risk sharing
- Total risk sharing (through both private and public mechanisms) in the EU and the euro area is roughly half that seen in existing federations studied above, with only about 40 percent of income shocks smoothed.
- The overall level of risk sharing in the euro area is nearly the same as in the EU, indicating euro area members have not benefited from additional risk sharing despite sacrificing domestic monetary autonomy.
- Fiscal risk sharing is almost nonexistent in both the EU and the euro area—unsurprising given that the EU budget is small, at about 1 percent of the area-wide GDP, and not designed for risk sharing.
- Capital markets in the euro area provide less than half the amount of income insurance seen in the US, Canada or Germany (post-unification), reflecting a stronger “home bias” in households’ investment decisions within the euro area.
- Aggregate portfolio diversification across euro area countries is far lower than that common amongst German regions or U.S. states.
- Where consumption is buffered at all against income shocks in the euro area, it occurs primarily through saving and borrowing in international credit markets.

### Risk sharing in the euro area over time and across the business cycle
- About two third of the country-level GDP shocks remains unsmoothed in the euro area; international capital markets (factor income flows) and fiscal transfers have only small effects on income smoothing.
- The main risk sharing channel is credit markets (countries’ saving and borrowing). Differentiating public and private saving and borrowing, private saving and borrowing provides the largest amount of smoothing (about 20 percent) in normal times.
- The degree of income shock smoothing within the panel of 15 European countries has decreased over time:
  - The share of GDP shocks that remains unsmoothed has risen from 58 percent over 1979-99 to 66 percent over 1990-2010.
  - This largely reflects a decline in cross-country smoothing via credit markets after the creation of the euro area—credit flows have been more pro-cyclical than pre-EMU.
  - In parallel, the amount of income shocks smoothed via international capital markets has risen somewhat, reflecting increasing financial integration within the euro area.
- Risk sharing mechanisms are particularly ineffective during recessions and times of sovereign stress:
  - The share of shocks unsmoothed in recessions is significantly larger than during normal times.
  - The deterioration is driven in part by a sharp decline in saving and borrowing in international credit markets, as investors revert to their own borders.
  - The share of unsmoothed shocks in Europe during 2008-10 (76 percent) is similar to previous episodes of severe downturns (78 percent).
  - Higher sovereign stress reduces smoothing: an increase of 100 basis points in the ten-year sovereign yield spread (versus the US ten-year rate) reduces the share of GDP shocks smoothed by about 5 percentage points.

### Illustration: a supranational rainy-day stabilization fund
- Motivation:
  - Banking union and deeper financial integration could help market-based risk sharing but would take time; fiscal risk sharing has a clear role to smooth shocks in the interim.
  - Cross-country fiscal risk sharing can be ex ante (common taxes/transfers, centralized public goods) or ex post (crisis arrangements); the illustration focuses on a simple rainy-day stabilization fund (temporary transfers).
- Fund design (thought experiment):
  - The fund collects taxes as a share of GNP of each member state and pays transfers to countries negatively hit by output shocks.
  - Each time a member is hit by a negative shock, it receives a transfer proportional to the size of the shock, the relative size of its economy, and resources available in the fund; if no member experiences a negative shock, contributions are saved.
  - GNP is used as scaling to control for consumption smoothing through international capital markets.
  - The fund accumulates over time, allowing it to help offset common income shocks (intertemporal insurance) in addition to idiosyncratic shocks.
- Estimated size of contributions:
  - To achieve a level of stabilization close to what is common in existing federations, a euro area stabilization fund based on temporary transfers (defined as a share of GDP) would need relatively limited contributions, but substantially larger than current EU budget transfers.
  - Counterfactual experiment: a fund put in place in 1999 could have increased overall stabilization (from about 40 percent currently) to the level found within Germany—where 80 percent of income shocks are smoothed— with annual contributions of about 1½ to 2½ percent of GNP.
  - The estimate hinges on the ability to distinguish temporary from permanent shocks; contributions would likely have to be higher if this distinction cannot be made in real time.
- Providers and recipients (historical simulation):
  - Had the fund existed since 1999, most euro area countries would have been net contributors until 2007 and net recipients during the most recent crisis.
  - Smaller countries would have tended to receive larger transfers relative to the size of their economy due to higher volatility.
  - For the entire period, the average net contribution by each country would have been close to zero (with the exception of Luxembourg), indicating this type of risk sharing need not entail permanent transfers.
  - Periphery countries would have contributed relatively more than core countries until 2007 (faster growth) but would have received larger net transfers during the Great Recession.
- Cross-country and intertemporal insurance:
  - Over the past three decades in the simulation, in most years about a fourth to half of the countries in the sample would have benefited from gross transfers (cross-country dimension).
  - The fund would have accumulated reserves to deal with rare large common shocks, at which point positive net transfers would have been provided to almost all countries (intertemporal smoothing). Examples of such common negative output events include the 1992-93 ERM crisis, the burst of the dot.com bubble in the early 2000s, and the two years following the failure of Lehman Brothers.

### Practical considerations and implementation challenges
- Correctly identifying “insurable” shocks is the main challenge: shocks should be unrelated over time (not too persistent) and have zero mean to warrant activation and transfers.
- Real-time identification is difficult; operational techniques based on the output gap or deviation from trend growth may be feasible compromises.
- Risk: a shock initially thought temporary could be permanent, leading to persistent transfers.
- Some risk sharing functions could be institutionalized through mechanisms already established elsewhere (e.g., centralized unemployment benefits), though unemployment insurance may involve implementation lags and only partial insurance for overall income risk.

### Conclusion and policy implications
- The long-term viability of a currency union depends on its capacity to deal with country-specific shocks; risk sharing mechanisms for consumption smoothing are essential.
- Empirical studies suggest risk sharing in the euro area is limited through both market-based and fiscal channels, and considerably lower than in federations such as Canada, the United States and Germany.
- Existing euro area risk sharing mechanisms are particularly ineffective during downturns, increasing vulnerability to severe downturns and attendant spillovers.
- A supranational fiscal stabilization mechanism could provide insurance against shocks; a thought experiment indicates that a temporary-transfer rainy-day fund with contributions of about 1½ – 2½ percent of GNP could have raised stabilization in the euro area to levels similar to Germany.
- Effectiveness and political acceptance of further fiscal integration depend critically on specific design; risk sharing would mean countries doing well support those hit by negative shocks, but not necessarily as permanent recipients—however, political difficulties remain significant.

*Source: IMF background paper section on “Risk Sharing in the Euro Area”*

### References

### _sdn1309tn - References

### Bibliographic references (selected)
- Afonso, A. and D. Furceri (2008). “EMU Enlargement Stabilization Costs and Insurance Mechanisms”, Journal of International Money and Finance 27, 169-187.
- Andersson, L. (2004). “Regional Risk Sharing Provided by the Fiscal System: Empirical Evidence from Sweden,” Regional Studies, 38, no. 3, 269-280.
- Asdrubali, P., B. Sorensen, and O. Yosha (1996). “Channels of Interstate Risk Sharing: United States 1963-90”, Quarterly Journal of Economics, 111, 1081-1110.
- Balli, F., S. Basher, and J. Rosmy (2011). “Channels of Risk Sharing Among Canadian Provinces: 1961-2006,” MPRA Paper No. 30876.
- Bayoumi, T. and P. Masson (1995). “Fiscal Flows in the United States and Canada: Lessons for Monetary Union in Europe,” European Economic Review, 39, no. 2, 253-274.
- Cecchetti, S., R. McCauley, and P. McGuire (2012). “Interpreting TARGET2 Balances,” BIS Working Papers No. 393.
- Decressin, J. (2002). “Regional income redistribution and risk sharing: how does Italy compare in Europe?”, Journal of Public Economics, 86, no. 2, 287-306.
- Friedman, M. (1957). A Theory of the Consumption Function, Princeton University Press.
- Furceri, D. and A. Zdzienicka (2013). “The euro area crisis: need for a supranational fiscal risk sharing mechanism?”, IMF Working Paper (forthcoming).
- Goodhart, C. E. A. and S. Smith (1993). “Stabilization,” in The Economics of Community Public Finance, European Economy Reports and Studies 5, 417-455, European Commission.
- Gros, D. and E. Jones (1994). “Fiscal Stabilizers in the US Monetary Union: Measurement Errors and the Role of National Fiscal Policy,” CEPS Working Document 83.
- Harding D. and A. Pagan (2002). “Dissecting the cycle: a methodological investigation,” Journal of Monetary Economics, 49, no. 2, 365-381.
- Hepp, R. and J. von Hagen (2012). “Interstate risk sharing in Germany: 1970-2006”, Oxford Economic Papers, (forthcoming).
- Laeven, L. and F. Valencia (2008, 2010). “Systemic Banking Crises: A New Database,” IMF Working Paper 08/224.
- Makipaa, A. and J. von Hagen (2005). “Regional Insurance and Redistribution in Germany,” Center for European Integration Studies.
- Masson, P. and M. Taylor (1992). “Common Currency Areas and Currency Unions: an Analysis of the Issues,” Center for Economic Policy Research Discussion Paper No. 617.
- Mélitz, J. and F. Zumer (2002). “Regional Redistribution and Stabilization by the Center in Canada, France, the United Kingdom and the United States: new estimates based on panel data econometrics,” CEPR Discussion Paper 1829.
- Obstfeld, M. and G. Peri (1998). “Regional non-adjustment and fiscal policy,” Economic Policy, 26, 205-259.
- Pisani-Ferry, J., A. Italianer, and R. Lescure (1993). “Stabilization Properties of Budgetary Systems: A Simulation Analysis,” in The Economics of Community Public Finance, European Economy Reports and Studies 5, European Commission.
- Sachs, J. and X. Sala-i-Martin (1991). “Fiscal Federalism and Optimum Currency Areas: Evidence for Europe from the United States,” in Establishing a Central Bank: Issues in Europe and Lessons from the U.S., ed. Matthew Canzoneri, Vittorio Grilli, and Paul Masson, Cambridge University Press.
- Sorensen, B. and O. Yosha (1998). “International Risk sharing and European Monetary Unification”, Journal of International Economics, 45, 211-238.
- von Hagen, J. (1992). “Fiscal Arrangements in a Monetary Union: Some Evidence from the U.S.,” in Fiscal Policy, Taxes, and the Financial System in an Increasingly Integrated Europe, ed. Don Fair and Christian de Boissieux, 337-360, Kluwer Academic Publishers.

### Appendix: Technical note on the Asdrubali, Sorensen, and Yosha approach
- Objective: Measure degree of risk sharing between sub-national or regional entities by decomposing variability in output and relating it to variability in consumption using national income and product accounts.
- Identity and notation (as presented):
  - indexes countries,  indexes time (years),
  -  is gross domestic product (output),
  -  is gross national product (income),
  -  is net national income,
  -  is net national disposable income,
  -  is private consumption,
  -  is public consumption (government consumption expenditures).
- Identified risk sharing channels (as listed in the source):
  -  = international income flows (factor income flows)
  -  = capital depreciation
  -  = net international tax and transfer flows (fiscal risk sharing)
  -  = saving
- Aggregation and channel labels:
  - International income flows and capital depreciation aggregated as a capital markets channel (bulk of factor income flows are capital income).
  - Saving channel referred to as the credit market channel (works through fixed income securities and deposits).
- Estimation strategy:
  - Evaluate how much each component moves with output to derive the share of income shock smoothing provided by each channel.
  - System of log-difference equations estimated (equations shown in source), with:
    - coefficients that are common time effects, absorbing common shocks;
    - coefficients indicating the average amount of income shock smoothing contributed by a given channel (measured as the percent of an income shock smoothed), with  (notation preserved as in source).
  - Coefficients can be negative (channels may exacerbate shocks).
  -  indicates the share of income shocks smoothed by international income flows.
  - The last coefficient, , indicates how consumption (private and public) moves with output — larger  implies less income shock smoothing. Full insulation is achieved if  (notation preserved).
- Reference for further details: Furceri and Zdzienicka (2013) (as cited in the source).

### Fiscal Union: Common Practices in Federal States — key elements and main findings (summary of background paper content)
- Motivation and benchmarks:
  - Purpose: Understand how fiscal unions operate in seven federations to distill elements relevant to the euro area.
  - Candidate comparators: federal states (Australia, Belgium, Brazil, Canada, Germany, Switzerland, U.S.) considered more instructive than existing currency unions for the euro area context.
  - Note: The Padoa-Schioppa Group called for a “sui generis fiscal federalism approach for the euro area” (Enderlein et al, 2012) (as cited).
- Table excerpt (as presented in the source):
  - Table 1. Main Features of Federations
    - Area (square km) / Federations / Dispersion of per capita income (coefficient of dispersion) / Number of regions/states (excl. territories and special districts)
    - Australia7,6920.196
    - Belgium310.503
    - Brazil8,5120.6326
    - Canada9,0120.3510
    - Germany3570.2716
    - Switzerland41...26
    - USA9,1620.4050
    - Currency Unions
      - Euro area2,7120.5317
      - WAEMU3,4640.848
    - Source: OECD, Wikipedia, Staff estimates. (table values preserved exactly as in source)
- Main findings (explicit points preserved):
  - Centrally managed budget: A significant amount of risk sharing in federal states is delivered through centralized revenue and expenditures, including through social security systems; expenditure responsibilities are backed by a centralized fiscal authority with power to raise revenues.
    - For the seven federations under study, at least one-third of general government expenditure is assumed by the center, and half or more of general government revenue is collected by the center.
    - Central government involvement in public capital expenditure ranges from about 10 percent in Canada to over 40 percent in Brazil (as described in the source).
  - Social safety net and banking backstop:
    - All federal states in the sample provide insurance against individual income and financial sector shocks via a federation-wide social safety net and an (explicit or implicit) federal backstop for banks.
    - Social protection: Unemployment, pension, and health insurance benefits are harmonized and funded mostly nationally in almost all countries; health services provision is often funded and administered locally. In the U.S., unemployment insurance is mainly state-based within federally-set basic requirements; large parts of health and pension insurance are organized and funded at the federal level; the federal government provides a minimum backstop during exceptionally severe downturns.
    - Banking sector backstops: Responsibility for resolving nationally active banks or providing deposit insurance falls on the federal level in all sample countries. The U.S., Canadian, Australian, and Belgian systems are run by the federal government or have explicit federal fiscal backing when dedicated funds fall short. Swiss and German deposit insurance systems are privately-funded without an explicit ex ante fiscal backstop (though the German resolution system established since 2008 has such a backstop); federal ex post support has occurred when needed.
  - Debt issuance and liability:
    - Public debt is issued separately by each level of government; no federation in the sample allows joint and several liability of central level to borrow backed by sub-national revenues.
    - Example: Federal government borrowing in Belgium is not backed by regional government revenues; sub-national governments finance debt backed by their own revenue and typically without an explicit guarantee from the central level.
  - Fiscal rules and governance:
    - Sub-national fiscal rules exist in most federations to maintain discipline; the most common rule is a balanced-budget requirement, sometimes with constraints on sub-national borrowing. More granular rules (e.g., on spending) are less common.
    - Nexus between solidarity and discipline:
      - Spectrum of approaches: U.S. and Canada rely more on market discipline with regional fiscal responsibility and tax autonomy, accompanied by a history of the center forgoing bailouts in some episodes (no-bailout approach credible because other fiscal risk sharing exists). Brazil and Germany have experienced bailout episodes, leading to a more center-based approach with larger federal tax powers and stricter sub-national constraints. Australia, Belgium, and Switzerland occupy intermediary positions with varying tax autonomy and governance mechanisms (intergovernmental coordination, direct democratic control).
- Policy-relevant implications for the euro area (as articulated in the source):
  - Institutionalized fiscal risk sharing in federations is typically coupled with strong governance frameworks to ensure good fiscal practices; these institutions complement market mechanisms in preventing and buffering idiosyncratic regional shocks and add to union stability.
  - The euro area lacks ex ante fiscal risk sharing mechanisms comparable to federations; lessons from federations suggest reforms that combine centralized fiscal capacity, common social safety nets, banking backstops, and governance frameworks reconciling solidarity with discipline could enhance stability.
  - The paper frames questions for application to the euro area: the role of a central budget; funding of social safety nets and financial backstops; addressing moral hazard in center–subnational financial relations; and modalities for fiscal governance consistent with bailout or no-bailout practices.

*Italic: Content derived from the referenced PDF chapter/section provided in the source content.*

### 9.      Centralized revenue. In all federations in the sample, the central government, consolidated

### 9. Centralized revenue. In all federations in the sample, the central government, consolidated with social security funds, holds the largest share of public finances.

### Centralized revenue and spending
- Central government revenue makes up:
  - "about half of general government revenue in Canada, Switzerland, and the U.S."
  - "two thirds in Germany"
  - "even higher shares in Australia and Belgium."
- Centralization provides insurance: a region’s relative tax contribution may fall during a downturn, but pooling resources across regions to finance continuous central expenditures reduces its impact.
- Expenditure responsibilities align with central revenue authority:
  - Central government own expenditure (net of transfers) is high where central revenue is high (Figure 4 and Table 2).
  - In most cases, revenue levied at the center finances federation-wide health and social protection.
  - Central budgets typically include long-horizon items (support programs and infrastructure) with large economies of scale and positive externalities; these are difficult for sub-national governments to finance alone and are protected by central-level provision.
- Central control of investment spending in the seven federations in the sample:
  - "10 to 45 percent, and 20 percent on average" of total investment spending is controlled by the central level.

### Federal social safety nets — insuring individual risks
- All sample countries have a minimum level of centralization of health care, pension, and unemployment insurance systems to ensure a floor in income and social services regardless of residence.
- The U.S. is the only sample country with a significant degree of decentralization in social protection systems.

Unemployment benefits
- General findings:
  - All countries have at least partial central government funding, a central government backstop, and unemployment benefits that are at least partially harmonized.
  - Some countries (e.g., Belgium) allow regional administration latitude; the U.S. is largely unharmonized across states but receives federally-sponsored support in severe shocks.
- Australia:
  - The Australian unemployment benefit system is fully centralized (Vroman, 2002).
  - Funded through general government revenues; benefit levels uniform across the country.
  - Income- and asset-tested; centrally administered through Centrelink.
  - Eligibility harmonized; payment conditional on job search assisted/assessed by federally-approved and -funded organizations.
- Belgium:
  - Unemployment insurance fund funded through national social contributions (Ohnsorge, 2012).
  - Benefits and eligibility harmonized nationally; compliance assessed by regional employment offices.
  - Sanction rate for non-compliance varies from "1 to 3 per thousand job seekers between regions."
- United States:
  - Decentralized, only partially federally-funded, and not fully harmonized.
  - Each state has its own unemployment insurance scheme funded by state payroll taxes; federal payroll taxes collected into FUTA cover administrative costs and act as a backstop.
  - Eligibility differs by state; qualification may depend on highest quarterly wages in any of the past four—or more, depending on the state—quarters of employment being above a state-specific threshold.
  - Benefits vary: waiting periods in some states; large variability in minimum and maximum benefits; only half of U.S. states allow benefits with partial employment income.
  - Benefit duration ranges from "6 to 30 weeks."
  - Federal role in downturns: ad hoc federally-funded emergency programs supplement state systems (e.g., Emergency Unemployment Compensation, created in 2008, extends benefit payments to "34 weeks and more" in high-unemployment states and was "set to expire in 2014"; earlier program: Temporary Extended Unemployment Compensation in 2002-03).
  - Extended Benefits programs, funded equally by state and federal government, automatically provide an additional "13 weeks" of benefits in states where unemployment exceeds "5 percent" and is one-fifth above the level of the previous two years; in extreme cases emergency legislation can temporarily establish full federal funding.

Pension benefits
- Most non-U.S. sample countries have a single unified, centrally-funded public pension system supplemented by private pensions.
- U.S. differentiation:
  - Federal level provides a means-tested minimum pension which states can supplement (OECD, 2011).
  - Eight states pay only the federal minimum; another 29 states administer their own system; 15 offer supplements operated by the federal Social Security Administration (sometimes jointly with state agencies).
  - These supplements provide an "average supplement of 29 percent of the maximum federal benefit for single pensioners and 50 percent for couples."

Health insurance
- Centralization of health insurance systems in all sample countries, with funding differences and some private/state supplements:
  - Australia and Brazil: funded through general taxes.
  - Belgium and Germany: funded through social contributions.
  - Switzerland: funded through individual specific premiums (Paris, Devaux and Wei, 2010 and OECD 2011).
  - U.S.: large private provision; Medicare (federally-funded) for seniors; Medicaid (state- and federally-funded) for the very poor (recently expanded in the Patient Protection and Affordable Care Act); "35 state-level insurance plans" provide access for people with pre-existing conditions.

### Backstops for the banking sector
- Rationale: banking stress can trigger contagion within federations through integrated financial markets; sub-national governments often cannot rescue banks without endangering creditworthiness.
- Nation-wide schemes: all sample countries have (implicit or explicit) nation-wide bank resolution mechanisms and deposit insurance systems (Appendix 1).
- Fiscal backing:
  - Deposit insurance schemes are designed as privately funded in principle; Belgium channels contributions through the federal budget.
  - All except the Swiss and German systems have an explicit fiscal backstop.
  - Even without explicit backstops, federal levels have stepped in during crises (Belgium, Switzerland, the U.S., and Germany); these programs were temporary.
- Canada:
  - Highly centralized operations with an explicit federal fiscal backstop.
  - Deposit insurance run by Canada Deposit Insurance Corporation (CDIC), funded by the banking sector.
  - CDIC covers most chartered banks but not credit unions (which have provincial insurance plans).
  - At end-2011, available funds stood at "0.4 percent of insured deposits" with an increase to "1 percent" targeted.
  - If insufficient, the fund can borrow up to "C$18 bn (1 percent of GDP)" from the federal government and financial markets and repay via increased member contributions.
  - CDIC is the bank resolution authority, cooperating with OSFI; OFSI would temporarily assume operational control in a resolution while CDIC would initiate winding up. Funding for deposit insurance is the only funding envisaged; creditors not eligible for deposit insurance are satisfied by claim seniority.
- Switzerland:
  - No explicit fiscal backstop for the banking sector; banks are mandated to hold large buffers; federal level has provided ex post support when systemic banks were at risk (e.g., UBS in 2008).
  - Deposit insurance: banks that take insurable deposits must be members of Esisuisse and contribute in proportion to insured deposits.
  - Deposit insurance fund is capped at "CHF 6 bn (just over 1 percent of customer savings and deposits)" and has no explicit fiscal backstop.
  - To ensure sufficiency, banks required to hold:
    - secured domestic assets in the amount of "125 percent of insured deposits", and
    - liquid assets in the amount of "50 percent of maximum potential contributions"—in addition to prudential liquidity ratios.
  - FINMA has single authority over bank supervision, bankruptcy and resolution and can trigger disbursements by the deposit insurance fund.
  - Despite lack of formal fiscal backstop, the federal budget bailed out UBS in 2008 on an ad hoc and temporary basis.

### Risk reduction: fiscal governance and bailout practices
- Moral hazard concern: sub-national governments may implement riskier policies knowing federal insurance exists; greater risk sharing typically requires stronger governance.
- Fiscal governance mechanisms:
  - Sub-national fiscal rules commonly take the form of balanced budget rules, often complemented by borrowing controls.
  - Enforcement and coordination can be center-based oversight, market discipline, or cooperative arrangements; some arrangements emerged from crises.
  - Where the center credibly commits to a no-bailout policy, market pressures can enforce discipline; where the center assists, fiscal discipline has been imposed top-down.
- Empirical magnitude of risk sharing:
  - Existing federations exhibit sizeable fiscal risk sharing: "of every one percent of a country-specific income shock, about 0.15 to 0.2 percent is smoothed through federal fiscal transfers (see first background note)."
  - The extent of risk sharing is even higher in unitary countries.
  - Experience shows higher fiscal risk sharing tends to go hand in hand with stronger fiscal governance of subnational policies.
- Budget constraints:
  - Rationale: decentralization improves efficiency by tailoring decisions to local needs, but sub-national governments may not internalize federation-wide negative spillovers; budget constraints help pre-empt fiscal excesses.
  - Types of rules:
    - Balanced budget requirements are present in most federations in the sample.
    - Restrictions on sub-national borrowing are common where market discipline is weak—either via explicit limits (e.g., imposed by the Senate in Brazil; granted constitutional status in Germany) or coordination processes with the federal government (Australia, Belgium).
    - Restraining sub-national spending is less common (Brazil, Belgium) as it directly limits fiscal autonomy.

*Source: _sdn1309tn - 9.      Centralized revenue. In all federations in the sample, the central government, consolidated*

### 24.      Sources of discipline. To be properly implemented, budget constraints are complemented

### 24.      Sources of discipline. To be properly implemented, budget constraints are complemented

### Sources of discipline and enforcement frameworks
- Budget constraints require enforcement frameworks to provide incentives for sub-national governments to comply.
- Modes of enforcement observed in sample federations:
  - Market-sanctioned constraints where states/provinces self-impose fiscal rules in legislation to signal sound policies (examples: U.S., Canada).
  - Federal oversight and stronger involvement leading to subnational budget constraints (examples: Germany, Brazil).
  - Intermediary/cooperative arrangements (Appendix 1).

### Interplay with (no) bailout
- Bailout expectations can undermine incentives for sub-national governments to respect budget constraints.
- Dynamic relationship with the center:
  - If sub-national governments believe the center is committed to a no bailout policy, they are incentivized to run sustainable fiscal policies.
  - If credibility of no-bailout is in doubt, sub-national governments may overspend expecting rescue, increasing likelihood of bailout requests.
  - If bailouts materialize, the center faces a choice: let the sub-national government default (reestablishing credibility) or provide support (switching to a bailout regime). An historical example is provided by the U.S. in the 19th century.
- Credibility of no-bailout commitments is enhanced when risk-sharing mechanisms exist that:
  - Ensure basic public services are still provided in the defaulting entity.
  - Minimize spillovers to other constituencies.

### Box 1 — To Bail Out or Not To Bail Out? (Sequential game)
- Stage 1: Central government announces financing policies and commits not to bail out profligate sub-national governments.
- Stage 2: Sub-national governments observe the center’s commitment and choose either:
  - To spend within means → stable outcome; or
  - To borrow excessively → may lead to bailout request.
- If bailout requested, the center chooses:
  - Provide bailout → switches to a different regime; or
  - Resist and allow default → reinforces credibility of no-bailout.
- Credibility determinants:
  - Scope for spillovers and contagion among sub-national economies increases center's bailout payoffs.
  - Extent of federation-wide risk sharing reduces center's incentives to bail out.
  - Degree of tax authority of sub-national governments: low local tax authority directs market pressure to center and increases probability of bailouts.

### Different arrangements for achieving fiscal discipline
- Two extreme arrangements and intermediary models (Figure 6):
  - Self-imposed budget constraints (U.S., Canada):
    - Market discipline underpinned by credible no-bailout commitment.
    - Sub-national governments granted wide-ranging fiscal autonomy.
  - “Center-based” approach (Brazil, Germany):
    - After past bailout episodes, central government exercises extensive controls; less fiscal autonomy for sub-national governments.
  - Consensual/cooperative approaches (Australia, Belgium, Switzerland):
    - Intergovernmental coordination or direct democracy combined with market discipline to achieve sustainable sub-national fiscal policies.

### Self-imposed budget constraints: U.S. and Canada
- Setup:
  - No bailout rules: federal governments consistently resisted bailouts; U.S. constitutional mechanisms prevent local sovereign obligations from being assumed by other jurisdictions and from extending federal protection for states’ creditors (11th amendment of the Constitution).
  - Fiscal autonomy: sub-national governments can determine bases and rates of broad-base taxes; central governments place no constraints on spending and borrowing decisions.
- Model emerging after crisis:
  - 19th century crises: Defaults in the 1830s and 1840s after investment booms; by 1857 most states had constitutional balanced budget amendments; 49 of 50 states have some version of this restriction today.
  - Recessions in the 1980s and 1990s: Canadian provinces saw rising public debt, higher risk premia and downgrades; federal resistance to bailouts led provinces to adjust tax rates and enact balanced-budget rules (by 1993 a majority of provinces had such rules). U.S. states faced regional downturns and cuts in federal grants and adjusted independently.
  - Rainy-day funds: number of state rainy-day funds rose from 12 in 1982 to 45 in 1995 as a means to avoid pro-cyclical policies under balanced-budget rules.
- Footnotes/notes:
  - (11) Limited depth of sub-national debt markets can constrain scope for purely market-based systems. Mobility of workers/firms across states can increase competitive pressures to maintain discipline while ensuring low taxation and high quality public services.
  - (12) Even though U.S. government is committed to no-bailout for states, it has at times provided financial support to local entities (examples: New York City in 1975; District of Columbia after 1994-95).

### “Center-based” approach: Brazil and Germany
- Brazil:
  - Following a series of sub-national bailouts in the late 1990s, the federal government assumed states’ liabilities in return for bilateral contracts strictly restricting fiscal policies for the following 30 years.
  - States required to maintain expenditures and debt levels below ceilings set by the (federal) Senate.
  - Federal government may deduct any over-spending from tax-sharing transfers.
  - Key elements later legislated in a Fiscal Responsibility Law in 2000.
- Germany:
  - Increases in Länder debts over the 1970s and 1980s led to “supplementary transfers” to Bremen and Saarland in 1987 (augmented in 1992), eroding fiscal conservatism and contributing to noncompliance with the SGP in 2002.
  - To restore no-bailout commitment: constitutional debt-brake rule adopted in 2009 binding all government levels to a structurally balanced budget.
  - A Stability Council created in 2010 to enforce discipline and ensure sub-national fiscal sustainability.
  - Länder with distressed finances receive small transfers (financed by federal and Länder governments in equal parts) and consent to a macroeconomic adjustment program; five Länder are currently under such programs.

### Cooperative process: Australia, Belgium, Switzerland
- Consensus building and intergovernmental coordination; not tested by sub-national fiscal crises.
- Australia:
  - Fiscal transparency: States negotiate borrowing allocations within the Loan Council, which discloses fiscal information and requires public explanations for overruns.
- Belgium:
  - Similar setup within the High Finance Council; borrowing allocations negotiated alongside intergovernmental transfers; debt issuance by subnational entities approved by the Council of Ministers.
  - Federal government can cap regional borrowing for two years in event of misalignments.
- Switzerland:
  - High sub-national autonomy supplemented by direct democracy.
  - Direct democracy secures effective constraints on taxation and loan finance; in some cantons borrowing is restricted to investment projects (golden rule) and subject to popular referenda.

### Lessons for the Euro Area
- Risk sharing combined with strong governance:
  - Federations in the sample have pooling of fiscal resources, centralization of expenditure functions (especially social protection), and an implicit/explicit backstop of the financial sector from the center.
  - Insurance mechanisms are paired with strong governance and enforcement mechanisms.
  - Where bailouts occurred, they were implemented with central coordination or oversight over sub-national fiscal policy.
  - The euro area has few risk sharing mechanisms and weaker governance compared with these federations.
- Improving institutional setting:
  - Market discipline is ineffective when bailout expectations are high and central enforcement is weak.
  - Cooperative approaches have limits in EMU’s first decade; intergovernmental coordination is harder with many jurisdictions.
  - Transitioning to a credible no-bailout regime must be gradual and coordinated; interim enforcement may require more direct central imposition.
  - The Fiscal Compact and the “two-pack” are first steps: Fiscal Compact calls for effective implementation of automatic corrective mechanisms at the national level; the “two-pack” enhances budgetary transparency and monitoring.
- Transition as result of crisis:
  - Historical episodes show fiscal stress and bailouts can trigger rebalancing of power toward the center and momentum for institutional changes restricting sub-national borrowing and enhancing central fiscal monitoring.
  - The current crisis presents the euro area with a similar opportunity.

*Source: IMF — TOWARD A FISCAL UNION FOR THE EURO AREA—BACKGROUND PAPERS (content unit: _sdn1309tn - 24.      Sources of discipline. To be properly implemented, budget constraints are complemented)*

### Appendix 1. Features of Banking Union, Common Borrowing, and Governance in the Sample’s Federations

### Appendix 1. Features of Banking Union, Common Borrowing, and Governance in the Sample’s Federations

### Deposit insurance and bank resolution/supervision — key features by federation
- Australia
  - Deposit insurance: National with government guarantee
  - Bank resolution/supervision: State governments separate
  - Borrowing arrangements: No
  - Joint and several: 64
  - Share of CG debt in GG debt 1/: State level rules
  - Central oversight: None...
  - Bailout Practices 2/: (ellipsis in source)

- Belgium
  - Deposit insurance: Deposit and financial instrument protection fund, funded by bank contributions with fiscal backstop
  - Bank resolution/supervision: Central bank separate
  - Borrowing arrangements: No
  - Joint and several: 92
  - Share of CG debt in GG debt 1/: Ad hoc political commitments
  - Central oversight: None...

- Brazil
  - Deposit insurance: National (FGC)
  - Bank resolution/supervision: Central bank separate
  - Borrowing arrangements: No
  - Joint and several: 70
  - Share of CG debt in GG debt 1/: yes
  - Central oversight: Yes, through contracts
  - Bailout Practices 2/: Bailouts in 1990s

- Canada
  - Deposit insurance: CDIC with fiscal backstop for all but two provinces
  - Bank resolution/supervision: CDIC separate
  - Borrowing arrangements: No
  - Joint and several: 48
  - Share of CG debt in GG debt 1/: Provincial budget rules and unemployment insurance rule, but no federal budget rule
  - Central oversight: None
  - Bailout Practices 2/: Federal governments resisted bailouts

- Germany
  - Deposit insurance: National, but private with ad hoc government guarantee during crisis
  - Bank resolution/supervision: National joint BaFIN (micro) / Buba (macro)
  - Borrowing arrangements: separate
  - Joint and several: No
  - Share of CG debt in GG debt 1/: 63
  - Central oversight: yes (debt brake rule)
  - Bailout Practices 2/: Some, stability council, fiscal rule; Bailout over 1970s and 1980s

- Switzerland
  - Deposit insurance: Privately operated national, without explicit fiscal backstop
  - Bank resolution/supervision: Supervision national, resolution cantonal for cantonal banks
  - Borrowing arrangements: separate
  - Joint and several: No
  - Share of CG debt in GG debt 1/: 57
  - Central oversight: Cantonal rules, usually balanced budget with qualifiers
  - Bailout Practices 2/: None...

- USA
  - Deposit insurance: FDIC, with fiscal backstop
  - Bank resolution/supervision: FDIC separate
  - Borrowing arrangements: No
  - Joint and several: 79
  - Share of CG debt in GG debt 1/: yes, golden rule balanced budgets
  - Central oversight: No
  - Bailout Practices 2/: Federal governments resisted bailouts

### Currency unions — features summarized
- Euro area
  - Banking union: none
  - Deposit insurance/bank resolution/supervision: no
  - Borrowing arrangements: separate
  - Joint and several: No
  - Share of CG debt in GG debt 1/: ...
  - Central oversight: SGP "two" and "six" pack
  - Bailout Practices 2/: Ongoing programs

- WAEMU
  - Deposit insurance: Uno
  - Bank resolution/supervision: Country-level resolution, de jure regional supervision
  - Borrowing arrangements: Separate
  - Joint and several: No
  - Share of CG debt in GG debt 1/: 0
  - Central oversight: Country specific; and WAEMU-wide convergence criteria
  - Bailout Practices 2/: Convergence reports, discussed by Council of Ministers
  - Additional notes: ...

Notes:
- 1/ OECD, national authorities, QPSD database. 2011 or latest available
- 2/ See text.

---

### Appendix 2. Sub-National Fiscal Frameworks in the Sample’s Federations

### Australia — Cooperative approach
- Fiscal Rules: Most states have adopted balanced budget rules
- Borrowing: Commonwealth and state borrowing is coordinated by the Loan Council taking into account each jurisdiction's fiscal position, infrastructure needs and macroeconomic objectives.
- Enforcement:
  - Peer pressure. The states are required to explain publicly overruns in the borrowing allocations set by the Council.
  - Market discipline. States borrow on their own account and the Loan Council provides information to markets on borrowing plans. Enhanced fiscal transparency has sharpened market discipline.

### Belgium — Cooperative approach
- Fiscal Rules: Targets for the overall balance and expenditure growth are set for communities and regions
- Borrowing: Regions and communities can borrow subject to government approval.
- Enforcement:
  - Peer pressure: fiscal targets are negotiated and published in annual reports
  - Administrative sanctions: The federal government is allowed by law to cap regional borrowing for two years.

### Brazil — Hierarchical oversight
- Fiscal Rules and borrowing: After the Fiscal Responsibility Law (FRL), states are required to maintain debt below ceilings set by the Senate. The law requires annual targets for revenues, expenditures, the primary balance and the indebtedness of all government levels.
- Borrowing: Contracts between the federal government and states remain in place from subnational bailout episodes and determine subnational fiscal policies.
- Enforcement:
  - Sanctions: The FRL foresees financial sanctions for sub-national governments breaching debt ceilings, personnel ceilings or transparency requirements; contracts or administrative decisions that violate the FRL are nullified; Governors responsible for violation risk imprisonment.
  - Escape clauses: Debt limits established by the Senate can be revised in event of a severe recession/natural disasters.

### Canada — Market-based approach
- Fiscal Rules: Many provincial governments have introduced some sort of balanced-budget legislation, while municipal governments must post balanced positions.
- Borrowing: No restriction on provincial borrowing but municipal borrowing is limited to investment projects (golden rule).
- Enforcement:
  - Market discipline. The federal government does not guarantee sub-national debts.
  - Escape clauses. In most provinces, surpluses can be carried over to finance a deficit in a subsequent year. Legislation in many provinces builds in escape clauses for special events.

### Germany — Hierarchical oversight
- Fiscal Rules: Lander must gradually converge towards a structural balance position by 2020 and maintain it thereof (Constitutional debt brake).
- Borrowing: Borrowing linked to investment projects until 2020 (golden rule) and cyclical developments thereof. Some states will feature a quicker transition.
- Enforcement:
  - Constitutional debt brake. Enforcement mechanisms are not automatic, but compliance with the structural balanced budget rule can be challenged to Courts and financial assistance to failing states can be cut off if they do not adhere to targets.
  - Escape clauses activated by two thirds majority of Parliament for emergencies outside the control of the government.

### Switzerland — Democracy-based approach
- Fiscal Rules: Informal coordination takes place between the Confederation and the cantons for the preparation of the budget. Most of the cantons have fiscal rules, but these vary with respect to the target.
- Enforcement: Sanctioning mechanisms vary across cantons. Municipal finances are subject to oversight by the respective canton. Close links between spending decisions and the responsibility for their financing.

### United States — Market-based approach
- Fiscal Rules: Most states have constitutional balanced budget rules.
- Borrowing: No restrictions.
- Enforcement:
  - Market discipline. The federal government does not guarantee sub-national debts.

Source: Joumard and Kongsrud (2003); and Fund Staff.

---


_Source: https://www.imf.org/-/media/websites/imf/imported-publications/external/pubs/ft/sdn/2013/_sdn1309tn.pdf_