## sdn1803

## Source details

**Canonical URL:** [sdn1803](https://www.imf.org/-/media/files/publications/sdn/2018/sdn1803.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/sdn/2018/sdn1803.pdf.md)
- [Structured JSON version](/-/media/files/publications/sdn/2018/sdn1803.pdf.json)

---

### Executive summary and key proposition
- Proposal: a central fiscal stabilization capacity (CFC) — a macroeconomic stabilization fund financed by regular annual contributions to build assets in good times, make transfers in bad times, and with a borrowing capacity for exceptionally large or persistent common shocks.
- Purpose: smooth the macroeconomic cycle, address country-specific shocks (which monetary policy cannot), and facilitate an appropriate fiscal–monetary policy mix for common shocks, especially when monetary policy is constrained at the effective lower bound and fiscal space is limited in some countries.
- Political context: renewed momentum for euro area reforms provides a window to discuss a CFC alongside Banking Union completion and ESM development; design explicitly aims to address moral hazard and permanent transfer concerns.
- Core design constraints:
  - Link stabilization explicitly to risk reduction.
  - Respect guiding principles of economic effectiveness and political feasibility.
  - Include safeguards to mitigate moral hazard and avoid ex ante permanent transfers.

### Political feasibility constraints and design safeguards
- Design options to limit moral hazard and permanent transfers:
  - Require adherence to strict fiscal rules as precondition for access.
  - Trigger transfers automatically via a transparent cyclical indicator to ensure nondiscretionary support.
  - Require a “usage premium” on cumulative transfers once a recipient has recovered.
  - Cap cumulative net transfers to a country.
- Operational principle: make key operational parameters simple and as automatic as possible to minimize disputes (measurement choices like output gap can undermine support).

### CFC core features (proposal summary)
- Fund type: dedicated macroeconomic stabilization fund (rainy-day fund) financed by regular annual contributions from national budgets; invests in safe, liquid assets.
- Borrowing capacity: available for exceptionally large area-wide shocks that would exhaust fund assets.
- Transfer mechanism: automatic transfers triggered by a cyclical indicator and proportional to cyclical fluctuations in that indicator (deviation of unemployment rate from its moving average proposed).
- Moral hazard mitigation: countries must comply with fiscal rules before receiving transfers in excess of cumulative net contributions.
- Fiscal/statistical treatment: envisaged contributions and transfers are final and treated as expense/revenue in fiscal accounts; exact statistical treatment depends on legal/institutional setup and Eurostat consultation.

### Primary calibration example emphasized throughout
- Annual contribution assumed in baseline: 0.35 percent of GDP per country.
- Transfer calibration example: 0.5 percent of GDP for every 1 percentage point deviation of unemployment above its moving average.
- Trigger moving average recommended: seven-year moving average of unemployment (seven-year balances countercyclicality and risk of structural-change-driven transfers).

### Sizing implication (quantitative illustration)
- With annual contributions of 0.35 percent of GDP, assets in typical good times would have sufficed to finance a large part of automatic stabilizers in past euro area recessions, but accumulated assets likely fall short under tail-risk shocks, indicating need for borrowing capacity.

### Guiding principles
- Primary objective: meaningful macroeconomic stabilization; additional functions risk undermining stabilization effectiveness.
- Size trade-off: balance meaningful stabilization power against affordability so member countries will participate given opportunity costs.

### Limits and complementarities
- Not a substitute for national policy adjustment where imbalances or competitiveness problems exist; structural reforms and, in severe cases, an ESM program with conditionality remain appropriate.
- Complements Banking Union and CMU but does not replace need for deeper private cross-border risk sharing or common deposit insurance.

---

### Trigger design, specification, and distributional implications
- Preferred trigger: deviation of unemployment rate from its moving average (less subject to real-time revisions than output gap or GDP growth).
- Moving average period:
  - Seven-year moving average recommended as balancing responsiveness and limiting structural-change-driven transfers; ECB research finds euro area business cycles range from six to nine years; CEPR dates average cycle about eight years.
  - A 10-year moving average raises correlation with output gap but increases risk of structural transfers; a 5-year average follows unemployment closely and provides insufficient stabilization.
- Alternative triggers and drawbacks:
  - Output gap: theoretically best but unobservable and subject to large revisions (almost 1½ percent of GDP on average for EU countries); poor real-time trigger.
  - Real GDP growth deviations: more volatile and subject to larger revisions; less correlated with output gap.
- Trigger variants:
  - “Double condition” (Carnot and others): unemployment must be both above moving average and increasing to trigger transfers — tightens conditions and limits transfers during improving but still-high unemployment.
  - Two trigger formulations affect distribution across countries:
    - Percentage point deviation (baseline): transfers per percentage point deviation.
    - Percent deviation: favors countries with structurally lower unemployment and reduces transfers to high-structural-unemployment countries.

---

### Transfer rule and calibration examples
- Baseline historical-calibration transfer rule:
  - Transfers equal 0.5 percent of GDP for every 1 percentage point unemployment deviation above its seven-year moving average.
  - Under that calibration, unemployment would need to exceed moving average by 0.75 percentage points to generate net annual transfers (gross transfers net of contributions).
- Suggested parameter ranges:
  - Contribution rates between ¼ and ½ percent of GDP per year.
  - Transfer rate around 0.5 percent of GDP per 1 percentage point unemployment deviation (or slightly higher depending on trigger).
- Earmarking option:
  - Possible to earmark transfers to items with large multipliers (public investment) or cyclical spending (unemployment benefits) but would limit overall stabilization flexibility.

### Mechanisms to avoid permanent transfers (design options)
- Usage premium:
  - Additional contributions based on past use of the fund and cyclical position; charged once recipient has recovered (unemployment falls below moving average).
  - Example: model uses 휆_C = 0.01 (interpretation preserved in technical formulae); careful calibration needed—World Economic Outlook projections show usage premium helps rebuild assets and pay down borrowing.
  - Quantitative illustration: average usage premium payments by countries after crisis about ¼ percent of GDP per year at most; Spain peak example about 1.1 percent of GDP under simple formulation.
- Cap on cumulative net transfers:
  - Limits support to countries with rising structural unemployment; politically may be necessary though economically suboptimal.
- Cap on cumulative net contributions:
  - Limits how large net contributors become; trade-off is reduced asset buildup and greater reliance on borrowing.
- Convex usage premium:
  - Premium that grows faster with cumulative net transfers possible but harder to communicate and calibrate.

### Compliance link to support
- Transfers exceeding a country’s cumulative net contributions could be contingent on compliance with EU fiscal rules.
- Compliance mechanism options:
  - Transfers proportional to country’s compliance rate over past years.
  - Noncompliance reduces transfers progressively; single noncompliance event reduces transfers but not immediate suspension.
- Rule simplification recommended:
  - Shift to a single fiscal anchor (for example, the debt ratio) combined with an expenditure growth rule and a debt-correction mechanism to enable more automatic enforcement and align with CFC functioning.

---

### Model simulations: setup, shocks, and main quantitative findings
- Model: IMF’s EUROMOD (a version of the Flexible System of Global Models, FSGM) with 24 countries/regions (11 largest euro area countries individually plus blocks).
- Baseline pre-shock calibration:
  - Countries pay contributions of 0.35 percent of GDP (paid by raising labor income tax rate in model).
  - Over 5 years before shock, these contributions generate assets of 1.8 percent of GDP (model steady-state aggregate assets target with 0.35 percent contributions is 10 percent of euro area GDP; text-preserved notation preserved in technical appendices).
- Illustrative area-wide shock components:
  1. Private demand shock reducing real private investment by 3 percent over t+1 to t+3 (with associated consumption hit).
  2. Sovereign and corporate risk premium spikes for higher-debt countries (medium group: sovereign +100 bps, corporate +50 bps; high group: sovereign +200 bps, corporate +100 bps; with CFC these shocks are eliminated or greatly reduced).
  3. Procyclical fiscal consolidation similar to 2011–12 tightening (spending cuts larger the higher the public debt ratio).
- Country groupings by 2017 general government debt:
  - Low debt: debt < 85 percent of GDP — Austria, Finland, Germany, Ireland, Netherlands.
  - Medium debt: 85–120 percent — Belgium, France, Spain.
  - High debt: > 120 percent — Greece, Italy, Portugal.
- Monetary policy regimes:
  - Unconstrained: policy rate can respond.
  - Constrained at the effective lower bound (ELB): policy rate fixed at zero over the simulation period.
- Main simulation quantitative results:
  - When monetary policy is unconstrained:
    - The CFC reduces the impact of the shock by nearly one-third.
  - When monetary policy is constrained at the ELB:
    - The CFC reduces the impact of the shock by nearly three-fifths.
  - The CFC improves inflation outcomes by supporting a better macroeconomic policy mix and reducing risk of a debt-deflation spiral when monetary policy is constrained.
  - Increasing transfer rates to smooth more of the shock would cause the CFC to borrow — indicating need for both built-up assets and borrowing capacity.
- Country-specific impacts (monetary policy constrained example):
  - Without CFC, at trough Italy’s output gap is more than 4 percentage points lower than Germany’s.
  - With CFC, difference between Italy and Germany’s output gaps is reduced by half, to just over 2 percentage points.

### Country-specific shock scenario (Spain)
- Modeled Spain-only shock similar to main area-wide shock:
  - The CFC smooths about half of the country-specific shock impact.
  - CFC assets are more than sufficient to cover prescribed transfers for a large shock in a relatively large euro area country.
  - Transfers are used to reduce cuts to public consumption and investment.

---

### Historical simulations and quantitative illustrations (1990–2017 baseline calibration)
- Baseline historical calibration assumptions:
  - CFC starts in 1990.
  - Annual contribution per country: 0.35 percent of GDP.
  - Trigger: deviation of unemployment rate above its seven-year moving average.
  - Transfers: 0.5 percent of GDP per 1 percentage point deviation.
  - Borrowing capacity available.
- Baseline historical outcomes (1990–2017):
  - CFC would have built up assets worth 1.6 percent of euro area GDP before the global financial crisis (alternative statements in text report over 1½ percent or around 1.6 percent).
  - Assets would have financed prescribed transfers until 2012; borrowing capacity used during 2013–17.
  - Total CFC borrowing would have peaked at 1.1 percent of euro area GDP in 2014, declining to 0.6 percent of GDP by 2017.
  - Aggregate annual gross transfers would have peaked at 1.1 percent of euro area GDP in 2013 and averaged almost 0.4 percent of euro area GDP per year.
  - For hardest hit countries, gross annual transfers would have averaged 2½ percent of their respective GDPs over 2009–13.
  - Maximum annual gross transfer to an individual country: 6 percent of that country’s GDP (Greece in 2013).
  - Maximum cumulative net transfers to a single country: around 20 percent of GDP (Greece in 2009–16; Spain in 2008–15).
  - Maximum cumulative net contributions from a single country: 6 percent of GDP (Ireland in 2007).
  - Under baseline, a country receives positive net annual transfers when unemployment exceeds its trend by over 0.7 percentage point.
  - Net asset position of the fund would fluctuate around zero symmetrically between 1990 and 2017; some countries persistent net contributors (Austria, Netherlands), others net beneficiaries (Finland).

### Sensitivity analyses and calibration trade-offs (selected findings)
- Contribution size effects:
  - Contributions of 0.1 percent of GDP per year: CFC would always need to borrow; most countries become net beneficiaries.
  - Contributions over 0.5 percent of GDP per year: excessive asset accumulation, even during GFC.
- Transfer size effects:
  - Transfers of 0.25 percent of GDP per 1 percentage point deviation: fund would always accumulate assets and provide insufficient stabilization.
  - Transfers over 0.75 percent of GDP per 1 percentage point deviation: greater stabilization but may cause perpetual borrowing by CFC.
- Trigger specification effects:
  - Percent-deviation trigger (vs percentage-point): increases transfers to low-structural-unemployment countries (Austria, Netherlands), reduces transfers to high-structural-unemployment countries (Greece, Spain), reduces accumulated assets and raises borrowing likelihood.
- Moving average length:
  - 5-year average: follows unemployment closely but provides insufficient stabilization.
  - 7-year average: judged best balance.
  - 10-year average: raises chance of transfers reflecting structural changes rather than cyclical conditions.
- Start date effect:
  - If CFC began in 1999 (vs 1990), assets before GFC exceed 2 percent of euro area GDP (vs about 1½ percent starting in 1990) and dynamics of net positions across countries shift.
- Usage premia and caps:
  - Usage premia (example defined in technical appendix) can rebuild assets and reduce borrowing post-shock; net beneficiaries would pay additional contributions when recovered, illustrative average about ½ to 1½ percent of GDP per year for a few years after receiving 10–20 percent of GDP in net cumulative transfers.
  - Caps on cumulative contributions (example 2 percent of GDP) limit potential one-way transfers from sustained net contributors but increase reliance on borrowing.

---

### Conclusions and policy recommendations
- A CFC could provide substantial macroeconomic stabilization with modest contributions and risk reduction.
- Key policy recommendations:
  - Design the CFC primarily for macroeconomic stabilization; avoid burdening it with unrelated functions.
  - Start with modest, constant annual contributions (suggested range ¼–½ percent of GDP) and transfer parameters around 0.5 percent of GDP per 1 percentage point unemployment deviation as central calibration.
  - Adopt the unemployment-deviation trigger (seven-year moving average) or tighter “double condition” variant to reduce disputes and improve real-time operability.
  - Include a borrowing capacity to handle tail risks and larger or persistent area-wide shocks.
  - Link transfers exceeding cumulative net contributions to compliance with simplified and more automatic fiscal rules (for example, single fiscal anchor plus expenditure growth rule).
  - Incorporate mechanisms to mitigate permanent transfers (usage premia, caps) and carefully calibrate them to balance insurance features with political acceptability.
  - Introduce the CFC in an expansionary phase of the cycle so assets can be built up before large transfers are needed.
- Implementation caveats:
  - Further technical work required on mechanism details and calibration to address stakeholder concerns.
  - Institutional setup and governance arrangements remain to be settled and present political challenges.

*Source: sdn1803*

### EXECUTIVE SUMMARY ____________________________________________________________________________ 4

### EXECUTIVE SUMMARY

### Key proposition
- A central fiscal stabilization capacity (CFC) is proposed as a macroeconomic stabilization fund financed by regular annual contributions to build assets in good times and make transfers in bad times; it would also have a borrowing capacity for exceptionally large or persistent common shocks.

### Rationale for timing and political context
- Renewed political momentum for euro area reforms creates a window to discuss a CFC alongside completing the Banking Union and further developing the European Stability Mechanism (ESM).
- Concerns remain about moral hazard and permanent transfers associated with fiscal risk sharing; the proposal is intended to address those concerns.

### Core features emphasized
- The CFC’s remit is to smooth the macroeconomic cycle, addressing country-specific shocks (which monetary policy cannot) and facilitating an appropriate fiscal–monetary policy mix for common shocks, especially when monetary policy is constrained at the effective lower bound and fiscal space is limited in some countries.
- To achieve economic effectiveness and political feasibility, the CFC must explicitly link stabilization to risk reduction and respect guiding principles (see Guiding Principles).

### Political feasibility constraints and design safeguards
- To avoid moral hazard and permanent transfers, the SDN examines options such as:
  - Requiring adherence to strict fiscal rules as a precondition for access.
  - Triggering transfers automatically via a transparent cyclical indicator to ensure nondiscretionary support.
  - Requiring a “usage premium” on cumulative transfers once a recipient has recovered.
  - Capping cumulative net transfers to a country.

### Quantitative illustration of capacity
- With annual contributions of 0.35 percent of GDP, the buildup of assets in typical good times would have sufficed to finance a large part of the operation of automatic stabilizers in past euro area recessions.
- Accumulated assets would likely fall short of transfers required under tail-risk shocks, indicating the need for a borrowing capacity.

*italics: Source: sdn1803 - EXECUTIVE SUMMARY*

### INTRODUCTION

### Window of opportunity and IMF stance
- The euro area is experiencing robust growth and increased public support for the euro; this provides a rare window for advancing euro area architecture reforms.
- The IMF has long supported reforms to strengthen the euro area architecture and argues that further integration would make the euro area more resilient to shocks.

### Remaining gaps in the architecture
- Progress has been made (Banking Union, ESM), but gaps remain: a common deposit insurance scheme and a common fiscal backstop to the Single Resolution Fund are still needed; deeper Capital Markets Union (CMU) development is required to strengthen private cross-border risk sharing.
- Greater fiscal risk sharing—coupled with fiscal rule reform—is still needed to allow fiscal policy to better contain shocks and incentivize national fiscal discipline.

### Political debate and proposals
- Recent policy proposals variously call for a euro zone budget, stabilization instruments, investment protection schemes, and a rainy-day fund or common unemployment insurance; political support varies and is constrained by concerns over risk sharing, moral hazard, and permanent transfers.
- The SDN offers a concrete stabilization-focused CFC proposal designed to balance economic objectives and political considerations; institutional and legal setup issues are intentionally not addressed in this note.

### Box 1 summary (recent proposals)
- Three main setups in recent proposals:
  - Contribution-transfer schemes (individual-level reinsurance or country-level rainy-day funds).
  - Borrowing-lending schemes that borrow from markets and on-lend to public/private sectors.
  - A dedicated euro area budget (managed, for example, by a euro area finance minister) for common public goods and investment.

*italics: Source: sdn1803 - EXECUTIVE SUMMARY*

### ECONOMIC RATIONALE FOR A CENTRAL FISCAL STABILIZATION CAPACITY

### Limited cross-border risk sharing to date
- Only about 20 percent of idiosyncratic shocks to euro area countries’ GDP per capita are smoothed via cross-border risk sharing.
- By contrast, about three-quarters of shocks are smoothed across individual states in the United States, mainly via capital and credit markets and federal transfers.
- Completing the Banking Union and CMU should improve private cross-border risk sharing, but will not eliminate the need for public sector risk sharing.

### Roles a CFC would play
- Smoothing country-specific shocks:
  - A single monetary policy increases reliance on national fiscal policy for country-specific stabilization.
  - Fiscal space is uneven across countries; a CFC would strengthen capacity for countercyclical fiscal responses.
- Cushioning area-wide shocks:
  - A CFC would improve the policy mix between fiscal and monetary policy during area-wide shocks, particularly when monetary policy is constrained (e.g., at the effective lower bound) and fiscal space is limited in some member states.
- Mitigating contagion:
  - By helping countries contain shocks, the CFC would reduce contagion risks and the chance that liquidity events morph into solvency crises.

### Incentives for better fiscal behavior
- A CFC would ensure greater fiscal discipline by requiring countries to build buffers in good times and by conditioning support on past compliance with common fiscal rules.
- The fund would encourage countercyclical policies in good times by requiring net contributions from countries and would facilitate revamping fiscal rules to be simpler and more automatically enforced.

### Limits of a CFC
- A CFC is not a substitute for national policy adjustment where imbalances and competitiveness problems exist; structural reforms and, in severe cases, an ESM program with conditionality remain appropriate.
- The CFC can shrink the “grey zone” between liquidity and solvency crises by redistributing fiscal space intertemporally without permanently enlarging it.

*italics: Source: sdn1803 - EXECUTIVE SUMMARY*

### GUIDING PRINCIPLES

### Overarching requirements
- Economic effectiveness and political feasibility constrain the design and size of the CFC; trade-offs will delimit its feasible features.

### Primary objective and sizing
- The CFC should be designed primarily to provide meaningful macroeconomic stabilization; assigning additional functions risks undermining stabilization effectiveness.
- Size must balance meaningful stabilization power against affordability so that member countries are willing to participate given opportunity costs of contributions.

*italics: Source: sdn1803 - EXECUTIVE SUMMARY*

### 13.      Moral hazard risks must be mitigated. Specifically, the risk is that if countries receive some

### sdn1803 - 13.      Moral hazard risks must be mitigated. Specifically, the risk is that if countries receive some

### Moral hazard and fiscal discipline
- Risk: If countries receive insurance against shocks from the CFC, they may run less sustainable fiscal policies and build up less of their own buffers.
- Mitigation principle: The CFC must provide incentives for countries to maintain fiscal discipline.
- Dependence: Effectiveness depends on common fiscal rules and countries’ medium-term fiscal frameworks (MTFFs).
- Policy implication: Fiscal rules should be reformed in conjunction with developing a CFC to make them less procyclical and to provide better incentives to abide by the rules.
- Operational requirement: Stricter adherence to MTFFs that comply with the fiscal rules, with MTFFs designed so fiscal policy is both sustainable and provides an appropriate degree of stabilization.

### Avoiding permanent transfers and political acceptability
- Concern: Fears that a CFC would become a transfer system in which high-performing countries habitually finance poor performers.
- Political constraint: The euro area is not a political union; any scheme creating ex ante permanent transfers would be a political nonstarter.
- Design goal: The CFC should be designed to avoid permanent transfers ex ante.
- Expected outcome: For any specific period some countries will be net contributors and others net beneficiaries; over a long enough horizon a country’s net position with the CFC should be close to balanced.
- Proposal element: Incorporate design features to help avoid permanent transfers.

### Automatic operation to minimize disputes
- Rationale: Making the CFC as automatic as possible can help prevent disputes over its operation.
- Implementation note: Once political agreement is reached, the CFC’s functioning and key operational parameters should be simple and automatic to minimize scope for disputes and discretion.
- Experience lesson: Significant disagreements can emerge over how key parameters are measured or calculated (for example, output gap calculation), undermining support.

### CFC proposal — main features
- Dedicated macroeconomic stabilization fund (rainy-day fund) financed by regular annual contributions from national budgets to build assets in good times and make transfers in bad times.
- A borrowing capacity available to the CFC for exceptionally large common shocks that would exhaust the fund’s assets.
- Automatic transfers triggered by a cyclical indicator and proportional to cyclical fluctuations in this indicator; the deviation of the unemployment rate from its moving average could be a suitable trigger, with transfers proportional to the unemployment rate deviation.
- Moral hazard mitigation: Countries must comply with fiscal rules before receiving transfers in excess of their cumulative net contributions.
- Political acceptability: Design to avoid permanent transfers; several mechanisms to achieve this are considered.

### A Macroeconomic Stabilization Fund — functions and conditions
- Purpose: Improve stabilization by counterbalancing fluctuations in revenue and cyclical spending and helping countries adhere to medium-term spending plans.
- Funding: Contributions come from national budgets; the fund should invest in safe and liquid assets to meet commitments when transfers are triggered.
- Treatment: Envisaged contributions and transfers are final (do not create a right to be fully repaid) and should be treated as expense/revenue in fiscal accounts rather than a below-the-line financing operation; statistical treatment depends on exact legal/institutional setup and consultation with Eurostat.
- Stabilization effects:
  - For countries with limited fiscal space, transfers help avoid procyclical fiscal tightening in downturns.
  - For countries with fiscal space, transfers limit increases in government deficits and debt that would have occurred otherwise.
  - In very large common shocks (for example, those justifying activating the Stability and Growth Pact’s escape clause), coordinated discretionary spending increases (for example, on public investment) could also be financed by transfers from the CFC.
- Precondition: All euro area countries would need full-fledged MTFFs consistent with common fiscal rules; the European Semester can provide external review of plans.

### Alternative CFC types considered
- Common unemployment insurance fund (UIF):
  - Financed by a share of individuals’ social security contributions; makes transfers to unemployed persons.
  - Could provide a basic level of insurance that national schemes could top up.
  - Pros: Serves stabilization and social protection objectives; contributions and benefits move countercyclically.
  - Cons: Less flexibility for countries; stabilization limited to the share of unemployment benefits provided; would require harmonization of labor market regulations/unemployment benefit rules.
- Borrowing-lending scheme (BLS):
  - CFC borrows from markets (backed by capital commitments) and lends to member states; similar to ESM setup.
  - Pros: Less moral hazard risk because countries repay; may be politically easier short term.
  - Cons: Fiscal space benefit equals difference between market borrowing cost and CFC borrowing cost (likely small short run); lending creates debt liability on national balance sheets and may not alleviate debt sustainability concerns; only countries with higher market borrowing costs than the CFC would have incentive to borrow.
- Euro area budget:
  - Stabilization potential depends on delegated revenue and expenditure types.
  - A budget focused on common goods (defense, border protection) would probably not be adequate for stabilization unless revenues paid into it were cyclical and spending large enough to affect the macroeconomy.
  - Embedding a euro area budget line within the EU budget would not provide sufficient resources for stabilization if within the current EU budget envelope (less than 1 percent of GDP).

### CFC size, financing, and contribution design
- Trade-off: Size balances desired stabilization degree against opportunity cost of resources devoted to the CFC.
- Political constraint: Larger annual contributions increase risk that governments cut productive public spending or increase distortionary taxes and make political agreement harder.
- Contribution flexibility:
  - Proposal envisages constant annual CFC contributions.
  - Alternative: Allow contributions to fluctuate over the cycle; cyclically varying contributions can increase countercyclicality in good times and reduce gross transfers in bad times for similar net transfers.
  - Advantages of constant contributions: Transparency, predictability for budgets and MTFFs, assured income flow to service CFC borrowing.
- Financing flexibility: Countries could decide how to fund contributions as long as they respect fiscal rules; assigning direct revenue streams (earmarking taxes) would limit flexibility and is less desirable given country heterogeneity.
- Income allocation: Income earned on CFC assets could be credited against contributions of countries that are net contributors; income likely modest if the fund accumulates at most a couple percent of euro area GDP and invests in safe assets; income would be allocated to net contributor countries based on their share of cumulative net contributions.

### Borrowing capacity
- Recommendation: The CFC should have a borrowing capacity, to be used in exceptionally large or persistent common shocks.
- Activation: Borrowing capacity activated when the fund’s assets are projected to fall below a certain threshold.
- Justification: Ensures prescribed transfers can be provided when assets are exhausted, especially if monetary policy approaches the effective lower bound.

### Trigger design — indicator choice and specification
- Importance: Agreement on an appropriate objective trigger for transfers is essential; trigger specification has been a stumbling block in earlier proposals.
- Output gap: Theoretically best candidate but unobservable and subject to large revisions—by almost 1½ percent of GDP on average for EU countries—making it a poor real-time trigger.
- Unemployment rate preferred:
  - Harmonized across euro area countries and more easily observable than output gap or GDP growth.
  - Ex post revisions to the unemployment rate are relatively smaller than for the output gap or real GDP growth.
- Proxy used: Deviation in the unemployment rate from its long-run moving average is a reasonable proxy for the output gap and is highly correlated with the (ex post) estimated output gap for the euro area overall.
- Benefits of unemployment deviation trigger:
  - Reduces risk that transfers are made because of high structural unemployment (accounts for countries’ labor market policies and differing structural unemployment).
  - Using the deviation from a moving average reduces transfers driven by structural changes rather than cyclical fluctuations.
- Moving average specification:
  - A shorter moving average period reduces countercyclicality; a longer period may generate transfers due to structural changes.
  - A seven-year moving average appears to balance these concerns and aligns with ECB research finding euro area business cycles range from six to nine years.
  - Center estimate: Close to estimates that the average length of the euro area business cycle since 1974 has been about eight years.
- Distributional implications:
  - Definition of the trigger (for example, percentage points deviation versus percent deviation) affects distribution of transfers across countries even if aggregate transfers are similar.
  - Agreement on trigger specification is politically and technically difficult.

*Source: https://www.imf.org/-/media/files/publications/sdn/2018/sdn1803.pdf*

### 32.      A “double condition” on the unemployment rate, as proposed by Carnot and others

### 32.      A “double condition” on the unemployment rate, as proposed by Carnot and others

### Trigger design
- Proposed alternative trigger: a “double condition” on the unemployment rate — the unemployment rate must be both above its moving average and increasing to trigger transfers.
- Rationale:
  - Places more restrictions on the period over which transfers are provided.
  - Balances countercyclicality of CFC support against political concerns that transfers may be perceived as excessive or going to countries that are doing well (for example, unemployment is falling but still above the moving average).
  - The “double condition” places more weight than the report’s primary proposed trigger on the concern about transfers going to countries with improving unemployment.

- Real GDP growth deviations as an alternative:
  - Trigger would be the difference between real GDP growth and its moving average or trend.
  - Downsides: GDP growth is subject to larger revisions and is more volatile than the unemployment rate.
  - If measured as a deviation from its moving average, GDP growth is less countercyclical than the unemployment gap because growth rates start picking up after the trough of a cycle, well before the unemployment gap is closed.
  - Euro area GDP growth is not very well correlated with the output gap and is therefore a poor proxy.

### Transfers and calibration
- Transfer rule example:
  - Transfers proportional to the deviation of the unemployment rate from its moving average.
  - Example calibration: transfers worth ½ percent of the recipient country’s GDP for every 1 percentage point deviation of the unemployment rate above its moving average.
  - Under that calibration, unemployment would need to exceed the moving average by ¾ percentage points to generate net annual transfers (that is, gross transfers net of contributions).
- Implications:
  - Historical data suggest transfers of this size would provide meaningful support.
  - Countries still need their own buffers; at the peak of the cycle unemployment will be below the moving average, so it takes time for unemployment to exceed the average and for transfers to kick in.
  - The CFC is not calibrated to completely compensate countries for the extent of deficit fluctuations due to automatic stabilizers.
- Earmarking:
  - Possible but would limit stabilization extent.
  - Earmarking could prioritize spending on large-multiplier items (for example, maintaining public investment during downturns) or cyclical spending (for example, unemployment benefits).
  - Earmarked spending would need to be large enough to plausibly deliver the targeted stabilization; an earmarked CFC would likely be insufficient for large area-wide shocks.

### Avoiding permanent transfers
- Political feasibility depends on building confidence that the CFC will not become a permanent transfer mechanism.
- Determinants: CFC trigger, size of transfers relative to contributions, income earned on assets credited to net contributor countries.
- Additional features required beyond baseline mechanism to mitigate permanent transfers risk.

- Options to reduce the risk of permanent transfers:
  - A usage premium:
    - Countries pay additional contributions based on past use of the fund and their cyclical position (similar to insurance surcharges after an accident).
    - After using the CFC, countries pay the usage premium only once the economy has recovered (that is, once the unemployment rate falls below its moving average).
    - Needs careful calibration so it does not nullify fiscal risk-sharing objective but is large enough to mitigate permanent transfer likelihood.
    - Idea inspired by Carnot, Kizior, and Mourre (2017).
  - A cap on cumulative net transfers:
    - Cumulative net transfers to a single country could be capped to address concerns about permanent transfers to countries with increasing structural unemployment and to ensure the CFC is not a substitute for necessary policy adjustment.
    - A cap that limits support for macroeconomic stabilization is economically suboptimal but may be politically necessary.
  - A cap on cumulative net contributions:
    - Better-performing countries could become sizable net contributors; a cap on cumulative net contributions may be necessary to make the CFC politically palatable.
    - Trade-off: a contributions cap limits assets built up in good times and makes borrowing by the CFC more likely.
  - Alternative: convex usage premium that grows faster as cumulative net transfers rise (would make transfers too expensive past a point). More complicated to communicate and calibrate than a linear usage premium plus cap.

### Supporting fiscal discipline
- Link transfers to compliance with EU fiscal rules to reduce moral hazard and create incentives for better compliance.
- Possible mechanism:
  - Transfers that exceed a country’s cumulative net contributions (the cross-country risk sharing component) could be contingent on compliance with EU fiscal rules.
  - Option: set transfers proportional to the country’s rate of compliance with the rules over the past several years.
  - Being assessed as noncompliant once would not lead to immediate suspension but would decrease transfers; continued noncompliance would reduce transfers further and could lead to suspension after an agreed number of years.
- Simplification of rules to allow more automatic enforcement:
  - Current complexity makes compliance assessment challenging.
  - Streamlining and simplifying fiscal rules would help monitoring and transparency and support the compliance–transfers link.
  - IMF staff suggested shifting to a single fiscal anchor (for example, the debt ratio) combined with an expenditure growth rule as the operational target, complemented by a debt-correction mechanism; this would focus on debt sustainability and align with the stabilization fund’s functioning.

### How would the proposed CFC perform? — Assessment approaches
- Two approaches used to assess CFC performance:
  - Multi-country general equilibrium simulations using EUROMOD (a version of the Flexible System of Global Models).
  - Historical simulations using data for 1990–2017 to simulate hypothetical financial flows to and from the CFC as if it had existed throughout recent history.

### General equilibrium model simulations — shock design and results
- Main shock scenario calibrated to roughly resemble the recent euro area crisis; shock has three components:
  1. A drop in private investment and consumption.
  2. Sovereign and corporate risk premium spikes in countries with higher public debt levels.
  3. A procyclical fiscal consolidation similar in size to the tightening in the euro area structural balance in 2011–12, implemented through cuts to public consumption and investment.
- Country differentiation:
  - Some countries have no risk premium shock and adjust less than average (for example, Germany).
  - Intermediate-risk countries see a moderate risk premium rise and intermediate fiscal adjustment (for example, France).
  - Highest debt countries face the largest risk premium shock and do the most fiscal consolidation (for example, Italy).
- Monetary policy settings:
  - Scenarios differentiate between monetary policy unconstrained and constrained at the effective lower bound.
- Model mechanics:
  - In comparative simulations, countries make annual contributions and build up CFC assets before the shock; transfers are triggered by unemployment deviation from its moving average and are used to reduce procyclical fiscal consolidation.
  - Presence of the CFC also mitigates the risk premium shock by design (eliminates risk premium shock for intermediate public debt countries and reduces it by five-sixths for high debt level countries).

- Key quantitative findings:
  - When monetary policy is unconstrained:
    - The CFC reduces the impact of the shock by nearly one-third.
  - When monetary policy is constrained at the effective lower bound:
    - The CFC reduces the impact of the shock by nearly three-fifths.
  - The CFC improves inflation outcomes by supporting a better macroeconomic policy mix and reducing the risk of a debt-deflation spiral when monetary policy is constrained.
  - Increasing the transfer rate to smooth more of the shock would cause the CFC to borrow, indicating the need for built-up assets and a borrowing capacity.
  - Country-specific impact (monetary policy constrained):
    - Without a CFC, at the trough the output gap for Italy is more than 4 percentage points lower than Germany.
    - With a CFC, the difference between Italy and Germany’s output gaps is reduced by half, to just over 2 percentage points.
    - The CFC reduces divergences in shock impacts across countries.

### Additional model notes
- EUROMOD includes 24 countries/regions (the 11 largest euro area countries plus a block for the rest of the euro area and 12 other countries/regions).
- Historical simulation period starts in 1990 to include at least two full economic cycles per country.
- A smaller shock scenario was modeled and exhibited similar dynamics to the larger shock scenario; results left to the Technical Appendix.

*Source: IMF staff chapter "32.      A “double condition” on the unemployment rate, as proposed by Carnot and others" from the PDF content unit provided.*

### 47.       A country-specific shock scenario confirms the benefits of the CFC. Euro area monetary

### sdn1803 - 47.       A country-specific shock scenario confirms the benefits of the CFC.

### Country-specific shock scenario (Spain)
- Euro area monetary policy usually does not respond to shocks in most individual countries, implying a role for a CFC to support macroeconomic stabilization when such shocks occur.
- Modeled a shock to Spain alone similar to the main area-wide shock scenario; with area-wide monetary policy unconstrained, the CFC helps smooth the shock.
- Key quantitative outcomes:
  - The CFC smooths about half of the country-specific shock impact (Figure 5, left panel).
  - CFC assets are more than sufficient to cover the prescribed transfers even with a large shock in a relatively large euro area country.
- Model outputs shown:
  - Spain: Output Gap (Percent difference) — comparison of Full Shock (no CFC) and With CFC across t-4 to t+5.
  - Level of CFC Assets (Percent of euro area GDP) — comparison of Full Shock (no CFC) and With CFC across t-4 to t+5.
- Transfers from the CFC are used to reduce the cuts to public consumption and investment.

### Calculations using historical data — baseline calibration and outcomes
- Baseline calibration assumptions:
  - The CFC starts in 1990.
  - Each country pays annual contributions of 0.35 percent of GDP.
  - Transfers to a country are triggered by the deviation in the unemployment rate above its seven-year moving average.
  - Countries receive transfers of 0.5 percent of GDP for every 1 percentage point deviation.
- Under this calibration:
  - The CFC would have built up assets worth 1.6 percent of euro area GDP before the global financial crisis.
  - The assets would have financed the prescribed transfers until 2012 and the borrowing capacity would have been used during 2013–17.
  - Total CFC borrowing would have peaked at 1.1 percent of euro area GDP in 2014, then declined to 0.6 percent of GDP by 2017.
  - Aggregate annual gross transfers provided by the CFC would have peaked at 1.1 percent of euro area GDP in 2013.
  - For the hardest hit countries, gross annual transfers would have averaged 2½ percent of their respective GDPs over 2009–13.

### Calibration trade-offs and recommended parameter ranges
- Calibration must balance desired fund size and stabilization capacity:
  - Setting the contribution rate too high or the transfer rate too low → potentially excessive accumulation of assets.
  - Setting the contribution rate too low or the transfer rate too high → more frequent reliance on borrowing capacity, politically less palatable.
- Suggested parameter ranges from analysis:
  - Contribution rates between ¼ and ½ percent of GDP per year.
  - Transfer rate should be about 0.5 percent of GDP or slightly higher, depending on the trigger.

### Mechanisms to avoid permanent transfers and their effects
- Examined mechanisms: usage premium and cap on cumulative net contributions.
- Usage premium:
  - Premium charged based on a country’s past use of the CFC and its cyclical position.
  - World Economic Outlook projections out to 2022 show the usage premium helps rebuild assets and pay down borrowing after a large shock.
  - Quantitative illustrations:
    - Countries that would pay a usage premium after the crisis: average is about ¼ percent of GDP per year at most.
    - Spain (extreme case): would have to pay at the peak about 1.1 percent of its GDP under the simple usage premium formulation.
  - Highlight: Spain’s peak of 1.1 percent is rather high, stressing the importance of careful design and calibration of the usage premium.
- Cap on cumulative net contributions:
  - A cap limits buildup of assets in good times and increases the likelihood of relying on borrowing capacity.
  - Aggregate effect: cap causes cumulative net transfers to peak at a slightly lower level.
  - Example: cap = 2% of GDP illustrated—without the cap the Netherlands’ net cumulative contributions would have grown with each expansionary phase without fully reversing; with the cap the net position would have reverted to nearly balanced during the crisis.

### Conclusions and policy recommendations
- A CFC could provide substantial macroeconomic stabilization with modest contributions and risk reduction.
- Key summary findings:
  - Even with relatively modest annual contributions, a CFC could provide meaningful stabilization in a downturn.
  - Making transfers conditional on strict compliance with the EU fiscal rules would help address moral hazard risks from fiscal risk sharing.
  - Several design options are available to prevent permanent transfers between countries (usage premium, caps on cumulative net contributions, etc.).
- Benefits of a CFC:
  - Strengthens the euro area’s resilience to shocks.
  - Helps achieve a balanced policy mix in response to shocks and provides policy space when monetary policy is constrained.
  - Requires countries to build buffers in good times and provides a common safety net in bad times, reducing likelihood and severity of future crises and spillovers.
- Implementation considerations:
  - Further technical work required to ensure mechanism and calibration address stakeholder concerns.
  - Political challenges to agree on technical features should not be underestimated.
  - Streamlining and simplifying fiscal rules would help ensure a transparent and automatic link between risk sharing and compliance, but would require strong political will.
  - Institutional setup and governance arrangements were not addressed and would need to be settled.
- Timing recommendation:
  - The expansionary phase of the cycle is the best time to introduce a CFC, especially if assets must be built up through contributions before activating transfers.
  - Policymakers should seize the opportunity created by political support to strengthen the euro area’s resilience.

*Source: IMF staff calculations and analysis in "A Central Fiscal Stabilization Capacity for the Euro Area" (SDN/18/03), March 2018.*

### 1. Technical Details of the CFC Mechanism _________________________________________________________ 8

### 1. Technical Details of the CFC Mechanism

### A. Selection of the CFC Trigger
- Rationale for trigger choice:
  - The deviation of the unemployment rate from its 7-year moving average is selected as the CFC transfer trigger because it is subject to smaller ex-post revisions than quarterly real GDP growth.
  - Revisions (average absolute values after 8 quarters) for the four largest euro area countries:
    - RGDP Growth revisions: 0.26, 0.36, 0.27, 0.30 (Germany, France, Italy, Spain) with average 0.30 (percentage points).
    - Unemployment Rate revisions: 0.35, 0.24, 0.11, 0.54 with average 0.31 (percentage points).
    - RGDP growth Revision/Mean: 1.23, 1.09, 1.49, 1.40 with average 1.31 (percent).
    - Unemployment Rate Revision/Mean: 7.43, 9.10, 9.12, 15.97 with average 10.40 (percent).
    - RGDP Growth Mean: 0.21, 0.33, 0.18, 0.21 with average 0.23 (percent).
    - Unemployment Rate Mean: 0.05, 0.03, 0.01, 0.03 with average 0.03 (percent).
- Comparative performance:
  - Transfers calculated using real-time versus ex-post series are much closer for the unemployment rate trigger than for the real GDP growth trigger (Figure 1).
  - Using real GDP growth deviations would cause more transfers in real time that would not have been made with ex post data, due to larger revisions.
- Relationship to output gap:
  - The inverse of the unemployment deviation from its 7-year moving average is highly correlated with the (ex post) estimated output gap for most euro area countries.
  - For the euro area as a whole, the correlation coefficient is 0.7 for the period 1990 to 2017 using the 7-year moving average.
  - Using a 10-year moving average raises the correlation to 0.8 for the euro area but increases the risk of transfers reflecting structural rather than cyclical unemployment.
  - Average business cycle length in the euro area is around 8 years (CEPR Business Cycle Dating Committee). ECB research finds cycles ranging from 6 to 9 years.
- Note on alternative indicators:
  - The deviation of real GDP growth from trend yields a much lower contemporaneous correlation with the estimated output gap for the euro area (coefficient of 0.1–0.2 for 1990–2017 when using 5–10 year moving averages).

### B. General Equilibrium Model Simulations
- Model and scope:
  - Simulations use the IMF’s EUROMOD model (a version of the Flexible System of Global Models, FSGM), a forward-looking, semi-structural global general equilibrium model with 24 countries/regions (for EUROMOD, 11 largest individual euro area countries plus a block for the rest of the euro area and 12 other countries/regions).
  - Simulations illustrate how a CFC could stabilize output when a shock occurs but do not show behavior over a full cycle or multiple cycles.
- CFC calibration and pre-shock buildup:
  - Countries pay contributions of 0.35 percent of GDP by raising the labor income tax rate.
  - Over the 5 years before the shock, these contributions generate assets of 1.8 percent of GDP.
  - Given the aggregate contribution target of 0.35 percent of euro area GDP per year, the steady state CFC assets target (퐴
푡
∗) is 10 percent of euro area GDP.
- Illustrative shock scenario (area-wide) components:
  - (i) Private demand shock affecting real private investment and consumption.
  - (ii) Risk premium shock for countries with higher public debt levels.
  - (iii) Procyclical fiscal consolidation, with spending cuts larger the higher the public debt ratio.
- Country grouping by 2017 general government debt (Box 2):
  - Low debt group: debt below 85 percent of GDP — Austria, Finland, Germany, Ireland, Netherlands.
  - Medium debt group: debt between 85 and 120 percent of GDP — Belgium, France, Spain.
  - High debt group: debt above 120 percent of GDP — Greece, Italy, Portugal.
- Shock magnitudes and profiles (Box 2):
  - Private sector demand shock: reduces real private investment by 3 percent over t+1 to t+3 (with an implicit consumption hit).
  - Risk premium shocks:
    - Medium group: sovereign risk premium +100 basis points (bps) and corporate risk premium +50 bps in t+1, decaying by half after 2 years.
    - High group: shock doubled (sovereign +200 bps; corporate +100 bps).
    - With the CFC: medium group experiences no risk premium shock; high group sovereign shock reduced to around 30 bps and corporate shock to around 15 bps.
  - Procyclical fiscal consolidation shock profile (cuts increase to trough at t+3 then partly unwind by t+5):
    - Low debt group trough cuts: 휖̅_GC = −0.96 and 휖̅_GI = −0.45 (percentage points of baseline GDP) — 60 percent of medium group.
    - Medium debt group trough cuts (euro area aggregate): 휖̅_GC = −1.6 and 휖̅_GI = −0.75 (percentage points of baseline GDP).
    - High debt group trough cuts: 휖̅_GC = −3.2 and 휖̅_GI = −1.5 (percentage points of baseline GDP) — twice medium group.
- Monetary policy regimes analyzed:
  - Monetary policy unconstrained (policy rate can respond).
  - Monetary policy constrained at the effective lower bound (ELB) — policy rate fixed at zero over the simulation period.
- Main simulation findings:
  - Monetary policy has a substantial stabilizing role:
    - Output gap trough around 2½ percentage points of euro area GDP below baseline when monetary policy can respond.
    - Output gap trough around 6 percent when monetary policy is constrained at the ELB (red lines in Figure 3).
  - The CFC provides meaningful smoothing of the shock (blue lines in Figure 3):
    - Reduces the shock’s output impact by about one-third when monetary policy responds.
    - Reduces the shock’s output impact by about three-fifths when monetary policy is constrained.
  - The CFC improves inflation outcomes by supporting a better macroeconomic policy mix and reducing the risk of a “debt-deflation” spiral.

### Box 1 — Technical Details of the CFC Mechanism (model implementation notes)
- Units and targets:
  - All variables are in percent of GDP.
  - Aggregate CFC contribution target C_t^* is defined so that with an aggregate contribution target of 0.35 percent of euro area GDP per year, the steady state CFC assets target 퐴
푡
∗ is 10 percent of euro area GDP.
- Country-level actual contributions formula (text-preserved representation):
  - 퐶
푖,푡
= (퐶
푖,푡
∗ + 휆
퐶
min(푢
푖,푡
푔푎푝
,0) max(푁푇
푖,푡
,0)) − 퐼
푖,푡
  - 푢
푖,푡
푔푎푝 is the unemployment rate deviation from its 7-year moving average.
  - NT_i,t is the cumulative net transfers (sum of transfers minus contributions) country i has received from the CFC up to period t.
  - I_i,t is the country’s share of interest earned on CFC assets if it is a net contributor.
  - The second term on the right-hand side is the “usage premium” to help prevent permanent transfers.
  - Currently in the model 휆
퐶
= 0.01. Interpretation: if unemployment falls 1 percentage point below its moving average and the country has received net transfers of 1 percent of GDP, the contribution will rise by 0.01 percent of GDP for that country in period t.
- Interest on assets:
  - Total interest earned on CFC assets (or paid on borrowing) is defined using the German short-term interest rate denoted by 푖
푡−1
퐷퐸푈 and assets level (text equations preserved in source).
- Assets evolution:
  - 퐴
푡
= 퐴
푡−1
/ (휋
푡
훾
푠푠
) + 퐶
푡
− 푇
푡
+ 퐼
푡
  - Where C_t is sum of contributions, T_t is sum of transfers in period t, and I_t is interest.
- Transfers rule:
  - Transfers to country i: 푇
푖,푡
= 휆
푢
max(푢
푖,푡
푔푎푝
,0)
  - Baseline calibration sets 휆
푢
= 1, which approximates a transfer parameter of 0.5 percent of GDP per 1 percentage point deviation of unemployment from the moving average when analyzing historical data.
- Calibration note:
  - The usage premium and interest-sharing features are designed so that countries accessing the CFC in bad times face higher contributions once they recover, limiting permanent transfers.

*Source: A Central Fiscal Stabilization Capacity for the Euro Area: Technical Appendix (SDN technical appendix).*

### 13.      While the CFC does not quite exhaust its assets in our main calibration, varying the

### sdn1803 - 13.      While the CFC does not quite exhaust its assets in our main calibration, varying the

### Simulation results: aggregate shock scenarios
- When monetary policy is unconstrained, transfers from the proposed CFC use only a modest portion of assets in the main calibration (Figure 4, left panel, blue line).
- In the main calibration, when monetary policy is constrained, assets built up before the shock are just enough to cover prescribed transfers (Figure 4, right panel, blue line).
- An additional simulation with a higher transfer parameter (dashed black line in Figure 4) shows the CFC would need a borrowing capacity to provide more smoothing or to deal with larger shocks.
- For a smaller shock (1/3 the size of the main scenario):
  - When monetary policy is unconstrained, the CFC smooths about 30 percent of the shock.
  - When monetary policy is at the effective lower bound (ELB), the CFC smooths about half the shock impact (Figure 6).
- Aggregate and inflation dynamics are illustrated for full shock scenarios with full monetary policy response (left) and monetary policy constrained at the ELB (right) (Figure 3).

### Country-level impacts and heterogeneity
- Without a CFC (monetary policy constrained), at the trough the output gap for Italy (red dashed line) is more than 4 percentage points lower than for Germany (blue dashed line) because of additional risk premia shock and greater fiscal adjustment.
- With the CFC (monetary policy constrained), the shock impact is smaller for each country and the difference between output gaps for Italy and Germany is cut nearly in half—to around 2 percentage points (Figure 5).
- The difference in shock impact between France and Germany is also reduced with the CFC.
- Note on fiscal multipliers: they are endogenous in the model and effectively larger when monetary policy is constrained because interest rate cuts do not mitigate some of the impact of fiscal tightening.

### Country-specific shock: Spain
- Simulating a country-specific shock affecting only Spain (private demand shock, intermediate risk premium, procyclical fiscal adjustment as in Box 2):
  - Even with euro area monetary policy unconstrained, the policy response to a country-specific shock is muted, increasing the need for fiscal support.
  - The CFC smooths a bit less than half of the Spain-specific shock (Figure 7, left panel).
  - Transfers to a relatively large euro area country like Spain do not substantially dent the level of CFC assets (Figure 7, right panel).

### Historical-data calibration: data and approach
- Uses annual "ex-post" data on unemployment rate, nominal and real GDP, and output gap from the IMF’s WEO database from 1980 to 2017, covering the 19 economies currently in the euro area.
- For brevity, calculations presented for the 11 largest economies (accounting for 97 percent of euro area GDP) and the euro area as a whole (GDP weighted sums/averages from the 19 countries).
- Projections for the unemployment rate from the WEO Database for the period 2018–22 are used in some calculations.
- The exercise computes hypothetical contributions to and transfers from the CFC had it been in place since 1990, to quantify accumulation of funds and distribution across countries under different design/calibration assumptions.

### Baseline calibration and key quantitative outcomes
Baseline assumptions:
- CFC starts operating in 1990.
- Each country pays regular annual contributions to the CFC of 0.35 percent of GDP.
- Transfers triggered by deviation of unemployment rate from its 7-year moving average.
- Countries receive transfers of 0.5 percent of GDP for every 1 percentage point deviation.
- A borrowing capacity is available to the CFC.

Under the baseline:
- The CFC would have accumulated around 1 percent of euro area GDP in assets before the early 1990s recession; sufficient to finance required transfers in 1993–94 but exhausted by 1995 (Figure 8.1, blue line). Borrowing capacity would be utilized from 1995–2000.
- Before the global financial crisis (GFC) the CFC would have built up assets of over 1½ percent of GDP, then exhausted those assets in 2012, relying on borrowing capacity from 2013.
- Total CFC borrowing would have peaked at 1.1 percent of euro area GDP in 2014.
- Aggregate annual gross transfers from the CFC would peak at around 1.1 percent of euro area GDP in 2013.
- Aggregate gross transfers would average almost 0.4 percent of euro area GDP per year.
- Maximum annual gross transfer to an individual country: 6 percent of that country’s GDP (Greece in 2013).
- Maximum cumulative net transfers to a single country: around 20 percent of GDP (Greece in 2009–16, Spain in 2008–15).
- Maximum cumulative net contributions from a single country: 6 percent of GDP (Ireland in 2007).
- Under the baseline a country receives positive net annual transfers when the unemployment rate exceeds its trend by over 0.7 percentage point (i.e., 0.5 x (u- ū)>0.35 when (u- ū)>0.7, where u denotes unemployment rate and ū denotes its long-run moving average).
- The net asset position of the fund would fluctuate around zero symmetrically between 1990 to 2017 (Figure 8.1); some countries would be persistent net contributors (Austria, Netherlands) and others net beneficiaries (Finland) over time.

### Sensitivity analyses and design trade-offs
- Trigger specification (percent vs percentage point deviation):
  - Using percent deviation of unemployment from its 7-year moving average would:
    - Increase gross transfers to countries with structurally lower and less volatile unemployment (e.g., Austria, Netherlands).
    - Decrease gross transfers to countries with structurally higher and more volatile unemployment (e.g., Greece, Spain).
    - Cause Austria and the Netherlands to switch from net contributors to net beneficiaries, and reduce net cumulative transfers to Greece and Spain.
    - Reduce accumulated assets in the fund and lead to earlier/higher resort to borrowing capacity (Figure 8.1).
  - Formula under percent-deviation schedule: gross transfers of 0.5 percent of GDP for every 10 percent deviation of unemployment from its trend (i.e., 0.05 x (u-ū)/ū).
- Choice of trend period for unemployment moving average:
  - A longer moving average period leads to higher and longer gross transfers for most countries and increases and prolongs CFC borrowing.
  - A 5-year moving average follows unemployment more closely and would provide insufficient stabilization.
  - A 10-year moving average would yield higher gross transfers that may reflect structural changes rather than cyclical conditions.
  - A 7-year moving average is judged to balance these concerns.
- Sizes of contributions and transfers (calibration trade-offs):
  - Contributions of 0.1 percent of GDP per year: CFC would always need to borrow to finance transfers; most countries become net beneficiaries throughout the simulation (Figure 12.1).
  - Contributions over 0.5 percent of GDP per year: excessive asset accumulation, even during the GFC, producing several permanent net contributors.
  - Transfers of 0.25 percent of GDP per 1 percentage point deviation: the fund would always accumulate assets and provide insufficient stabilization.
  - Transfers over 0.75 percent of GDP per 1 percentage point deviation: greater stabilization but may cause perpetual borrowing by the CFC.
- Start year of the CFC:
  - If the CFC began in 1999 (inception of the euro) instead of 1990, it would have accumulated over 2 percent of euro area GDP before the GFC (vs 1½ percent if starting 1990), exhausted one year later in 2013, and would switch from borrowing to accumulating assets in 2017.
  - Net cumulative transfers to Austria and the Netherlands would be higher, Spain lower, and Finland would switch from net beneficiary to net contributor.
- Mechanisms to avoid permanent transfers (usage premia, bonuses, caps):
  - Require countries to pay additional contributions when unemployment deviation becomes negative, proportional to past cumulative transfers and the strength of cyclical recovery (example: payments equivalent to 1 percent of net cumulative transfers per 1 percentage point decline in unemployment below its moving average) leads to faster asset buildup post-GFC (Figure 16.1).
  - With usage premia as defined, net beneficiaries (Finland, Greece, Portugal, Spain) would contribute in total ½ to 1½ percent of GDP per year to the fund for a few years when recovered, after receiving about 10 to 20 percent of GDP in net cumulative transfers during bust periods (Figures 15 and 16).
  - Caps on cumulative transfers could limit excessive use of funds (and therefore large usage premia).
  - Caps on net cumulative contributions (e.g., 2 percent of GDP) would limit potential one-way transfers from countries with prolonged favorable unemployment dynamics (e.g., Austria, France, Netherlands in baseline) (Figure 17).

*Source: IMF WEO Database and IMF staff calculations (as presented in the technical appendix).*

### 1. Euro area

### 1. Euro area

### Overview of simulation setup and data sources
- Figures use data from the IMF WEO Database and IMF staff calculations.
- Main policy instrument analyzed: a Central Fiscal Stabilization Capacity (CFC) for the euro area.
- Key trigger variables: unemployment rate deviations from moving averages (5-year, 7-year, 10-year).
- Contribution and transfer design parameters explored across figures include:
  - Annual contribution sizes: 0.35 percent of GDP; 0.5 percent of GDP per year; 0.1 percent of GDP per year.
  - Transfer sizes per 1 percentage point unemployment deviation: 0.75 percent of GDP; 0.5 percent of GDP; 0.25 percent of GDP.
  - Trigger definitions: percentage point deviation of unemployment rate from its 7-year moving average; percent deviation of unemployment rate from its 7-year moving average; percentage point deviation from 5-year, 7-year, and 10-year moving averages.
  - Timing scenarios: contributions and transfers start simultaneously in 1990 vs start simultaneously in 1999.
  - Usage premia scenarios: no usage premium; usage premium of 1 percent on cumulative net transfers per 1 ppt unemployment rate deviation.
  - Contribution cap scenarios: no cap on cumulative contributions; 2 percent of GDP cap on cumulative contributions.

### Annual gross transfers from the CFC (figures and triggers)
- Using percentage point deviation of unemployment rate from its 7-year moving average as trigger:
  - Annual contribution size of 0.35 percent of GDP analyzed.
  - Country-level panels include: 1. Euro area; 2. Austria; 3. Belgium; 4. Finland; 5. France; 6. Germany; 7. Greece; 8. Ireland; 9. Italy; 10. Netherlands; 11. Portugal; 12. Spain.
- Using percent deviation of unemployment rate from its 7-year moving average as trigger:
  - Country-level panels shown for same 12 jurisdictions.
- Using percentage point deviation from 5-year, 7-year, and 10-year moving averages as triggers:
  - Comparative annual gross transfers illustrated under each moving-average trigger for the 12 jurisdictions.

### Cumulative net transfers from the CFC (sensitivity to triggers and sizes)
- Cumulative net transfers analyzed under triggers defined as percentage point deviations from 5-year, 7-year, and 10-year moving averages.
- Sensitivity to contribution size (Figure 12):
  - Contributions equal to 0.5 percent of GDP per year.
  - Contributions equal to 0.35 percent of GDP per year.
  - Contributions equal to 0.1 percent of GDP per year.
- Sensitivity to transfer size per 1 percentage point unemployment deviation (Figure 13):
  - Transfers of 0.75 percent of GDP per 1 percentage point unemployment rate deviation.
  - Transfers of 0.5 percent of GDP per 1 percentage point unemployment rate deviation.
  - Transfers of 0.25 percent of GDP per 1 percentage point unemployment rate deviation.
- Sensitivity to start date (Figure 14):
  - Contributions and transfers start simultaneously in 1990.
  - Contributions and transfers start simultaneously in 1999.

### Usage premia and contribution caps
- Usage premia (Figure 15 and 16):
  - Total annual contributions shown under:
    - No usage premium.
    - Usage premium of 1 percent on cumulative net transfers per 1 ppt unemployment rate deviation.
  - Cumulative net transfers compared under the two usage-premia scenarios.
- Contribution cap (Figure 17):
  - Cumulative net transfers presented for:
    - No cap on cumulative contributions to the CFC.
    - 2 percent of GDP cap on cumulative contributions to the CFC.

### Country panels and axis ranges (visual scales preserved from figures)
- Country-specific panels appear for: 1. Euro area; 2. Austria; 3. Belgium; 4. Finland; 5. France; 6. Germany; 7. Greece; 8. Ireland; 9. Italy; 10. Netherlands; 11. Portugal; 12. Spain.
- Axis tick values displayed in the figures include (examples preserved exactly as shown in the source):
  - Negative and positive ranges such as -10.0, -8.0, -6.0, -5.0, -4.0, -3.0, -2.0, -1.0, 0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 1.0, 1.2, 1.4, 1.5, 2.0, 2.5, 3.0, 4.0, 5.0, 6.0, 8.0, 10.0, 14.0, 15.0, 20.0, 22.0, 25.0, 30.0, 40.0.
  - Time axis markers displayed as: 1990 1993 1996 1999 2002 2005 2008 2011 2014 2017 (and in some panels 1990 1995 2000 2005 2010 2015 2020).

*Source: IMF WEO Database and IMF staff calculations.*

---


_Source: https://www.imf.org/-/media/files/publications/sdn/2018/sdn1803.pdf_