## Appendix 1. Fiscal Sustainability Framework for Resource-Rich Countries

## Source details

**Canonical URL:** [Appendix 1. Fiscal Sustainability Framework for Resource-Rich Countries](https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023230-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2023/english/wpiea2023230-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2023/english/wpiea2023230-print-pdf.pdf.json)

---

### Glossary and context
- Acronyms and terms:
  - CAB: Cyclically-Adjusted Balance
  - CEMAC: Central African Economic and Monetary Community
  - FSF: Fiscal Sustainability Framework
  - FX: Foreign Exchange
  - GDP: Gross Domestic Product
  - MPIH: Modified Personal Income Hypothesis
  - NFA: Net Financial Assets
  - NFW: Net Financial Wealth
  - NFACR: Ratio of Net Financial Assets to Commodity Revenues
  - NRPB: Non-Resource Primary Balance
  - PIH: Permanent Income Hypothesis
  - SWF: Sovereign Wealth Fund
- Empirical observations:
  - Commodity prices rose sharply until 2014 (“2000s commodities boom”) and collapsed by about a third in the second half of 2014.
  - Commodity prices more than doubled between March 2020 and December 2022.
  - About one quarter of resource-rich countries suspended their rules following the commodity price crash of 2014.
  - The Covid-19 crisis caused large fiscal deteriorations and many countries activated escape clauses.

### Objectives of fiscal rules in resource-rich countries
- Core objectives (rules typically aim to achieve one or more):
  - Economic stabilization
    - Delink yearly government expenditure from volatile resource revenues.
    - Save windfalls in high-price periods; allow borrowing or use of assets in low-price periods.
  - Insurance against large and persistent shocks
    - Accumulate precautionary savings when prices are high.
  - Fiscal sustainability and intergenerational equity
    - Ensure spending is affordable and split appropriately between consumption and saving.
  - Mitigation of Dutch disease effects
    - Manage windfalls through the budget to limit real exchange rate appreciation; structural policies remain necessary.
- Trade-offs:
  - No single rule achieves all objectives simultaneously; priority depends on country circumstances and reserve horizons.

### Challenges in natural resource fiscal management
- Volatility and persistence:
  - Prices can move by as much as 40–80 percent over a decade.
  - Commodity revenues can account for up to more than 80 percent of budget revenue in some cases.
  - Average non-oil primary deficits fluctuated around 15 percent of non-oil GDP since mid-1990s, increasing to about 25 percent of non-oil GDP during 2001-08.
- Exhaustibility and uncertainty:
  - Resource horizons vary greatly; estimating future resource revenues is highly uncertain.
- Growth, governance, and Dutch disease:
  - Episodes of long periods of low/negative growth after boom-busts; exceptions include Botswana, Chile, Norway.
  - Resource rents can exacerbate rent-seeking, corruption, and conflict.

### Lessons from international experience with fiscal rules
- Adoption and rule types:
  - Number of resource-rich countries with at least one numerical fiscal rule grew from 5 in 2000 to 23 in 2019, then fell to 17 in 2021 due to COVID-19 suspensions.
  - Four categories: debt, budget balance, expenditure, and revenue rules; revenue rules less frequent.
  - Examples of rule combinations: Peru (budget balance + debt rule), Mongolia (all four types), Liberia (only debt rule).
- Performance and compliance:
  - Empirical evidence mixed: some studies find rules reduce procyclicality; others find limited or no impact.
  - Successful cases: Norway’s non-oil deficit cap linked to a 3 percent expected long-run real return; Chile’s structural balance adjusted for copper supported prudence.
  - Failures/weaknesses: Timor-Leste (overspent during 2003-08), Ecuador (repeated rule changes, procyclical capital spending), Mongolia (persistent non-compliance, debt ~90 percent of GDP after breaches).
- Resource funds:
  - Role: stabilization funds, saving funds, financing funds; many accumulate foreign assets.
  - Evidence mixed: some studies find no significant impact on spending cyclicality; others find SWFs associated with smaller spending volatility, particularly with high institutional quality.
  - Institutional integration matters: success when funds are integrated with fiscal rules (Botswana, Chile, Norway); failures when inconsistent (Ecuador, Venezuela, Chad).

### Fiscal Sustainability Framework (FSF): rationale and mechanics
- Fiscal anchor and operational rule:
  - Fiscal anchor: government “net wealth” = net financial assets + resource wealth in the ground (discounted sum of future resource revenues).
  - Operational rule: non-resource primary balance (NRPB), usually as a share of non-resource GDP.
- NRPB target and preservation of wealth:
  - NRPB target computed to stabilize net wealth; stabilizing can be defined in real terms, as a share of non-resource GDP, or in real per capita terms.
  - Preserving wealth implies government can consume:
    - return on financial wealth already accumulated,
    - implicit return on NPV of future resource revenues,
    - non-resource revenues.
- Cyclical adjustment:
  - FSF can use a cyclically-adjusted NRPB based on structural non-resource revenues (adjust for non-resource output gap and revenue elasticity).
- Appeal:
  - Targeting non-resource balance may avoid forced long-term cuts as resource revenues decline.

### Operational challenges and paradoxes of the FSF
- Calibration difficulties:
  - Financial wealth observable; underground resource wealth depends on long-term commodity price estimates that are hard to predict.
  - Estimated resource wealth can change abruptly when expected prices are revised.
- Volatility of expenditure targets:
  - Medium- to long-term forecasts depend heavily on current prices; re-estimated resource wealth forces revisions to NRPB and expenditure plans.
- Sustainability paradox:
  - Optimistic initial resource wealth estimates can permit high early borrowing and negative financial wealth; a price collapse can leave countries with high debt and vulnerability.
- Primary balance exclusion of interest payments:
  - NRPB excludes interest payments; primary-balance-based rules risk divergence and may need correction mechanisms.
- Public investment omission:
  - FSF excludes physical assets and feedback of public investment on growth; MPIH variant allows front-loaded investment but requires later consolidation if multipliers insufficient.
- Social time preference:
  - Preservation of wealth may clash with high present-time preferences and development needs; negative interest-growth differentials can imply high required NRPB surpluses.

### Variants and alternatives
- Modified PIH (MPIH):
  - Accommodates temporary scaling-up of capital expenditure; assumes scaling-up may not impact growth (simplest case).
  - If higher investment raises potential growth, future NRPBs improve and consolidation needs may be reduced.
  - Trade-off: investing in real assets vs accumulating financial assets; scaling-up often leaves higher real assets but lower net financial assets in long term absent consolidation.
- Approaches that do not stabilize wealth:
  - Finite-horizon depletion models: front-load expenditure with planned gradual adjustment before extraction ends.
  - Bird-in-hand rule: spend only interest income from accumulated financial assets (extreme precautionary saving).

### Medium-term fiscal framework and precautionary buffers
- Horizon and elements:
  - Long-term strategies: horizons of several decades (generally 10 to 30 years); assess resource size, valuation risks, demographics, technology, infrastructure, climate change.
  - Medium-term fiscal frameworks: typically cover 3 to 5 years; should set feasible and stable multiyear projections, provide flexibility, transparent fiscal anchors, risk-based rules, and checks and balances.
- Principle: build fiscal buffers to respond to shocks
  - Buffers should incentivize saving in booms and allow use in busts; often larger than for non-resource-rich economies.
  - Forms of buffer:
    - Asset-side buffer: accumulated liquid financial assets.
    - Liability-side buffer: room to borrow up to a “debt limit” (example: if debt distress threshold = 90 percent of GDP and current debt = 70 percent of GDP, debt buffer = 20 percent of GDP).
  - Overall buffer = financial assets + borrowing capacity.

### Fiscal anchors and NFACR (financial assets floor)
- Anchor types: financial assets-based anchor (floor) or debt-based anchor (ceiling); anchors often formulated as floors/ceilings.
- NFACR definition and function:
  - NFACR = stock of net financial assets (NFA) divided by annual commodity revenues.
  - Example interpretation: NFACR = 3 implies NFA should be at least equivalent to 3 years of commodity revenues.
  - Main function: protect spending plans and stabilize fiscal policy amid commodity revenue shocks.
  - Stabilization mechanism:
    - Price surge → assets-to-revenue ratio declines → rule calls for saving.
    - Price fall → ratio increases → use financial savings to fund budget.
- Calibration (three-step process):
  - Step 1: size of shock and desired degree of protection (e.g., largest fall with 10% probability over given period).
  - Step 2: estimate size of impact on budget = prices * quantities * government take relative to baseline.
  - Step 3: estimate buffer size depending on maximal feasible fiscal adjustment; willingness to adjust reduces required buffer.
- Country-specific calibration factors:
  - Degree of volatility, opportunity cost of self-insurance, desired degree of insurance.

### Illustrative NFACR calibration (Box 2)
- Illustrative assumptions:
  - Country with yearly oil revenues averaging around 30 percent of GDP.
  - Government offsets half of revenue shortfalls via fiscal adjustment; returns of financial assets cover remaining half.
- Key illustrative findings:
  - High protection requires savings equivalent to around 3-4 years of oil revenues.
  - Example statement: “a NFA ratio of around 3 (which represents about 90 percent of a large oil exporter’s GDP) would allow to maintain spending in 90 percent of the most favorable shocks, provided that fiscal adjustment offsets half of the oil revenue losses.”
  - For more extreme shocks, an NFA ratio of 3 would be gradually depleted unless larger fiscal adjustment is accepted.

### Responses to deviations and shock responses under NFACR
- Convergence:
  - Deviations from NFACR should be addressed via gradual convergence; floor not expected to be met every year but on average over medium term.
- Positive shocks (Appendix 3 example):
  - Anchor objective: NFACR = 3; initial assets = 30; resource revenues = 10.
  - If revenues rise to 13, illustrative response: spend one fourth of the revenue surprise initially and raise spending gradually if persistent, keeping NFA-to-revenue ratio at 3 as the guide.
- Negative shocks:
  - Small temporary shocks: use buffers to maintain spending for first(s) year(s).
  - Larger or persistent shocks: combine buffer use with fiscal adjustment and plan to rebuild buffers.
  - Very large shocks may require review of framework and calibration.

### Alternative anchors and debt ceilings
- Nominal NFA buffer calibrated to be depleted after a particular shock (value-at-risk, model-based methods) tends to require smaller buffers but risks large adjustment if shocks are more persistent than expected.
- Debt ceiling anchor:
  - Suitable where budget dependence on resource revenues is lower and willingness to self-insure is lower.
  - Design: express as gross debt ratio ceiling with denominator = non-resource GDP; include both domestic and external debt in numerator.
  - Calibration: follow IMF (2018a) principles; tailor to commodity producers by replacing terms-of-trade with commodity prices, differentiating positive/negative gaps, and using commodity-producer sample.

### Dual anchors (debt ceiling + financial asset floor)
- Rationale:
  - Dual anchors may be used temporarily (transition) or permanently; configuration complex and often dominated by one buffer.
  - Benefit: different insurance properties—assets vs debt—allow cost-effective shock response depending on correlation of asset returns and debt costs to shocks.
- Calibration challenges:
  - Risk of double-counting buffers; joint calibration may require portfolio-optimization modeling.

### Selecting and calibrating operational rules
- Options and trade-offs:
  - Overall/primary balance rules: simple, transparent; risk of procyclicality and allowing booms to raise permanent spending.
  - Cyclically-adjusted balance (CAB) rules: conceptually mitigate procyclicality but require capacity to estimate economic and commodity cycles; structural commodity price estimation is difficult and can make CAB resemble nominal balance.
  - Non-resource (primary) balance rules: decouple fiscal policy from commodity volatility; calibration and communication challenges exist.
  - Revenue allocation rules (revenue split, price smoothing) constrain saving vs spending of resource revenues but do not limit borrowing.
  - Expenditure rules:
    - Advantages: simpler to monitor, governments control spending envelope, reduce expenditure volatility, support buffer creation.
    - Disadvantages: do not incentivize revenue mobilization; require capability to tap buffers or borrow when needed.
- Transition and maintenance periods:
  - While building buffers, indicative benchmarks for annual adjustment can be used with priority on realistic political/social feasibility.
  - During maintenance, operational rule should contain and stabilize spending so windfalls are not fully spent in booms.
- Formal conditions for steady-state expenditure calibration (using NFA anchor):
  - Sustainable expenditure ratio to stabilize NFA over the cycle: derived using cyclically-adjusted variables and trend growth parameters.
  - Sustainable expenditure ratio to stabilize gross debt in the longer term: analogous formula using long-term growth parameters.
  - These ratios need not be identical ex ante; periodic updates required.

### Suggested steps to calibrate the sustainable expenditure envelope (Box 3)
- Calibration steps (illustrative example where anchor = gross debt ceiling treated as a target):
  - Step 1: set anchor target (illustration: target = 40 percent of GDP).
  - Step 2: derive anchor-stabilizing overall balance (illustration: nominal GDP growth = 6 percent → fiscal deficit = 2.3 percent of GDP = -40*0.06/1.06).
  - Step 3: derive expenditure ratio using structural variables and prudent revenue assumptions (illustration: projected revenue ratio = 30 percent of GDP → expenditure ratio = 32.3 percent of GDP).
  - Step 4: adjust expenditure ratio per long-term sustainability benchmark if needed (illustration: long-term gross debt = 60 percent of GDP; growth = 6 percent → debt-stabilizing balance ≈ fiscal deficit of about 3.4 percent of GDP; example relation noted: "28.9-32.3=3.4").
- Operational guidance:
  - Recalibrate every time medium-term plans are defined (illustrative cadence: every 4 years).
  - Use long-term price averages for resource revenues; adopt conservative non-resource revenue baselines until reforms bear fruit.

### Choice of operational rules (Appendix 2 highlights)
- Overall/primary balance rule:
  - Advantages: simplicity, transparency, calibration via formulas relating balance to net debt.
  - Risks: procyclicality, temporary windfalls driving permanent spending increases.
- Challenges with commodity-cycle correction (Box A2):
  - Commodity prices display varying frequencies and high persistence; estimating structural commodity prices is complex and may leave CAB similar to nominal balance.
- Non-resource balance rule:
  - Strongly mitigates procyclicality; challenges in calibration, separation of resource-related items, and public communication.
- Revenue allocation rules:
  - Constrain split between saving and spending of resource revenues but do not limit borrowing, potentially insufficient for macro stability.

### Debt-rule calibration for commodity producers (Appendix 4 case study)
- Methodology summary:
  - Assume debt limit = 80 percent of GDP; estimate safety margin via stochastic simulations over 8-year horizon using debt accumulation and fiscal reaction function.
  - Historical shocks distribution estimated using 1990-2020 data on real GDP growth, real effective interest rates, real exchange rates, terms of trade gaps, and external loan disbursements.
- Commodity-producer tailoring:
  - Use commodity-based terms of trade; differentiate positive vs negative gaps; estimate reaction functions on commodity-producer sample.
- Key sample statistics and parameters (Table A2):
  - Last year debt to GDP ratio (%) = 65.7
  - Debt limit (% of GDP) = 80
  - Risk Tolerance (%) = 5
  - Average historical growth rate (% of GDP) = 3.2
  - Average historical primary balance (% of GDP) = 0.38
- Fiscal reaction function estimates (reported coefficients):
  - Lag of primary balance to GDP ratio = 0.359*** (0.0510)
  - Lag of debt to GDP ratio = 0.0160*** (0.00465)
  - Negative output gap = 0.0904** (0.0339)
  - Negative terms of trade gap = 0.119* (0.0621)
  - Positive terms of trade gap = 0.299*** (0.0479)
  - External disbursement = -0.159* (0.0927)
  - Constant = -0.948** (0.459)
  - Observations = 851; R-squared = 0.321; Number of ifscode = 37
- Results and implications:
  - Default 2018 parameterization suggests a safe debt anchor near 70 percent of GDP for Colombia.
  - Incorporating commodity risks and tailoring reduces the prudent debt anchor to around 55 percent of GDP, implying an additional buffer of about 15 percent of GDP relative to default calibration.

*Italic source: Content from IMF Working Papers — A New Fiscal Framework for Resource-Rich Countries (wpiea2023230-print-pdf).*

### Appendix 1. Fiscal Sustainability Framework for Resource-Rich Countries ................................................

### Appendix 1. Fiscal Sustainability Framework for Resource-Rich Countries

### Glossary
- CAB: Cyclically-Adjusted Balance
- CEMAC: Central African Economic and Monetary Community
- FSF: Fiscal Sustainability Framework
- FX: Foreign Exchange
- GDP: Gross Domestic Product
- MPIH: Modified Personal Income Hypothesis
- NFA: Net Financial Assets
- NFW: Net Financial Wealth
- NFACR: Ratio of Net Financial Assets to Commodity Revenues
- NRPB: Non-Resource Primary Balance
- PIH: Permanent Income Hypothesis
- SWF: Sovereign Wealth Fund

### Key context and empirical observations (Introduction)
- Commodity prices exhibited very large fluctuations in the past two decades.
- From the beginning of the new millennium until 2014, prices of most natural resources rose sharply (the “2000s commodities boom”).
- Drivers of the 2000s commodities boom included rising demand from emerging markets, particularly China; concerns over long-term supply availability; and portfolio reallocations towards commodities after the burst of the housing bubble in advanced economies.
- Global commodity prices collapsed by about a third in the second half of 2014 as result of both supply and demand factors.
- Commodity prices more than doubled between March 2020 and December 2022, driven by the Covid-19 crisis and the war in Ukraine.
- The recent upswing in prices has prompted speculation that a new "supercycle" may have started.
- In the longer term, the energy transition in the face of climate change is likely to reduce global demand for fossil fuels, creating significant downside risks to the price of key commodities, especially oil (IEA 2023).
- About one quarter of resource-rich countries suspended their rules following the commodity price crash of 2014 (Mihalyi and Fernández 2018).
- The Covid-19 crisis caused a large deterioration in fiscal deficits and public debt globally, and many countries activated escape clauses to allow deviations from their rules (Gbohoui and Medas 2020, Caselli and others 2022).

### Objectives and approach of the paper
- Main objective: to assess alternative systems of fiscal rules for resource-rich countries against their policy objectives, taking into account country characteristics.
- Motivation: recent sophistication of analytical tools for commodity producers has improved analysis but reduced simplicity, transparency, and usability.
- Contribution: proposes a more holistic approach to reframe the discussion and offer practical guidance.
- Key methodological shift: move focus away from reliance on precise estimation of resource wealth toward building fiscal buffers as insurance against a variety of risks.
- The paper comprises eight sections (listings of sections not reproduced here).

*IMF Working Papers — A New Fiscal Framework for Resource-Rich Countries (Appendix 1).*

### section II, the paper reviews and draws lessons from international experience with rules’ design and

### II. What Are the Objectives of Fiscal Rules in Resource-Rich Countries?

### A. Challenges Associated with Natural Resource Management
- Revenues from non-renewable natural resources can finance infrastructure and scale up health and education, but volatility and exhaustibility pose fiscal policy design and implementation challenges.
- Commodity price volatility and persistence:
  - Prices can move by as much as 40–80 percent over a decade (IMF 2015).
  - Example: after 2001-08 increase, oil prices fell by around 40 percent to near $60 per barrel in 2009, then rose above $100 in 2010.
- Fiscal impacts of volatility:
  - Commodity revenues can account for up to more than 80 percent of budget revenue in some cases (Figure 1, Panel C).
  - Government revenues co-move with commodity prices and can vary dramatically in percent of non-resource GDP (Figure 2, Panel A).
  - Policy response tends to be highly procyclical:
    - Average non-oil primary deficits fluctuated around 15 percent of non-oil GDP since the mid-1990s, increased to about 25 percent of non-oil GDP during 2001-08, driven by primary spending that more than doubled (Figure 2, Panel B).
    - During downswings, borrowing costs rise when countries need markets, limiting smoothing and causing disruptive spending cuts (Figure 2, Panel C).
  - Overall, fiscal policy has, on average, not been stabilizing in resource-rich countries.
- Exhaustibility and intergenerational equity:
  - Resource horizons vary greatly across countries and commodity types (Figure 3, Panel A).
  - Many countries accumulate financial assets to offset depletion (Figure 3, Panel B).
  - Estimating future resource revenues is highly uncertain (future prices, production costs, reserve sizes, fiscal regimes, future interest rates); climate strategies affect fossil fuel extraction.
- Growth and Dutch disease risks:
  - Historical episodes: long periods of low/negative growth after boom-busts; some countries (Botswana, Chile, Norway) navigated cycles successfully (Figure 3, Panel C).
  - Dutch disease: higher resource prices raise wages and profits in resource sector, appreciate real exchange rate, hurt non-commodity competitiveness (Corden and Neary 1984).
  - Resource rents can exacerbate governance problems: rent-seeking, voracity, corruption, civil conflict (Alesina and others 2008; April 2019 Fiscal Monitor).

### B. Main Objectives of Fiscal Rules in Resource-Rich Countries
- Fiscal rules typically aim to achieve one or more of four objectives:
  - Economic stabilization:
    - Delink yearly government expenditure from volatile resource revenues.
    - When prices are high: save windfalls (repay past debt or accumulate assets).
    - When prices are low: allow borrowing or tapping accumulated assets.
    - Good stabilization rules should: (1) incentivize saving during high prices; (2) allow borrowing or use of assets during low prices.
  - Insurance against large and persistent shocks:
    - Accumulate precautionary savings when prices are high to insure against regime shifts between high and low price levels.
    - Resource-rich countries need larger buffers given shock size and duration.
  - Fiscal sustainability and intergenerational equity:
    - Ensure spending is affordable, prescribe split between consumption and saving to transfer wealth to future generations, and smooth spending profiles to avoid abrupt cuts when resources are exhausted.
  - Mitigation of Dutch disease effects:
    - Manage use of windfalls through the budget to limit real exchange rate appreciation and competitiveness loss.
    - Fiscal rules are not a complete solution; structural diversification and productivity improvements in nontraded sectors are needed.
- Trade-offs:
  - No fiscal rule can achieve all four objectives simultaneously.
  - Objective prioritization depends on country circumstances and reserve horizons (e.g., stabilization more relevant for long reserve horizons; intergenerational equity more important for short reserve horizons).

### III. Lessons From Resource-Rich Countries’ Experiences with Fiscal Rules
- Multiple complicating factors: revenue sharing in federations, revenue earmarking, and alternative frameworks (medium-term frameworks, resource funds).
- Growth in numerical fiscal rule adoption:
  - In 2000, only Canada, Colombia, Indonesia, Niger, and Peru had rule-based fiscal frameworks.
  - Number of resource-rich countries with at least one numerical fiscal rule grew to 23 in 2019, then fell to 17 in 2021 due to COVID-19 suspensions (Azerbaijan, Colombia, Indonesia, Niger, Peru, Russia among suspensions).
  - Rapid increase from 5 in 2000 to 15 in 2004.
- Rule types and combinations (2019 snapshot):
  - Four categories: debt, budget balance, expenditure, and revenue rules.
  - Most common combination: debt rule plus operational expenditure and/or budget balance limits.
  - Balance budget rules implemented by more than three fourths of resource-rich countries; debt and expenditure rules are also common; revenue rules less frequent.
  - Examples:
    - Revenue rules: Iran, Timor-Leste.
    - Only debt rule: Liberia.
    - Only budget balance rule: Chile, Nigeria, Norway.
    - Peru: budget balance + debt rule.
    - Mongolia: all four types.
  - Some countries rely on resource funds or medium-term frameworks instead of rules (e.g., Kuwait has no fiscal rule but a sovereign wealth fund > five times annual GDP in 2021).
- Design variations:
  - Debt rules usually ceilings on gross debt as percent of GDP; Kazakhstan’s 2016 rule links debt to oil fund foreign exchange assets (net debt concept).
  - Budget balance rules vary: overall balance (Indonesia) to non-resource balance, non-resource current balance, structural balance adjusted for nonrenewable resource prices (Chile, Colombia, Mongolia), and reference balance rules (Republic of Congo).
  - Expenditure rules: ceilings on annual nominal growth (Mongolia), primary or current expenditure growth (Mexico, Colombia), caps on real expenditure growth (Peru), or expenditure-to-GDP ratio (Namibia).

### III.B Performance of Fiscal Rules in Resource-Rich Countries
- Overall assessment:
  - Empirical evidence suggests fiscal rules can help but have had limited success insulating fiscal policy from resource revenue fluctuations and ensuring sustainability.
  - Reasons: fundamental uncertainty/volatility of resource revenues, weak compliance due to political economy, and lack of institutional arrangements (public financial management and transparency).
- Procyclicality:
  - Mixed empirical literature:
    - Cespedes and Velasco (2014): rules linked to larger shift toward fiscal counter-cyclicality across two boom episodes.
    - Pieschacon (2012): rules reduce procyclicality for Mexico and Norway (1986-2006).
    - Other studies (Ossowski and others 2008; Villafuerte and López-Murphy 2010; Arezki and Ismail 2013; IMF 2015; Bova and others 2016; Coutinho and others 2022) find limited or no significant impact of rules on procyclicality.
  - Case studies:
    - Successes: Norway’s non-oil deficit cap linked to expected long-run real return (3 percent of assets of Government Pension Fund-Global) protected budget; Chile’s structural balance rule adjusted for copper supported prudent policy.
    - Failures/weaknesses: Timor-Leste overspent during 2003-08 boom; Ecuador’s repeated rule changes and procyclical capital spending (2003-14); Azerbaijan’s non-resource balance guideline did not prevent procyclical spending; Equatorial Guinea’s non-resource current balance rule accompanied by procyclical capital spending.
  - Expenditure rules:
    - Helpful in containing spending pressures and providing room in downturns, especially when combined with overall balance rules (Peru, Mexico, Colombia).
    - Example: Peru’s expenditure rule mitigated procyclicality; Colombia used expenditure rule to accommodate revenue shortfalls in 2015-16.
- Compliance and resilience:
  - Nominal overall balance rules are easy to comply with in booms but fragile in busts; booms can mask unsustainable spending increases.
  - Examples:
    - Nigeria met 3 percent nominal deficit ceiling during boom but failed after 2015 collapse due to lack of accumulated savings.
    - Peru’s 2000 deficit rule failed to adapt as resource dependency rose, leading to repeated amendments.
    - Mexico invoked escape clauses multiple years under 2006 Fiscal Responsibility Law imposing zero federal balance.
    - CEMAC regional zero-domestic balance floor was not observed after 2014 crash; rules revised in 2017.
  - Non-resource and structural balance rules:
    - Provide more flexibility and better compliance during busts; many countries that complied in 2015-16 used structural or non-resource balance rules.
    - Mongolia: persistent non-compliance despite 2006 Fiscal Stability Law due to spending boom; repeated breaches led to amended targets and higher debt (~90 percent of GDP).
- Sustainability correlations:
  - Post-adoption improvements observed:
    - Average debt-to-GDP fell from 47 percent before rule adoption to 33 percent after (1990-2021 sample of adopters).
    - Overall deficit declined from ~0.9 percent of GDP to 0.2 percent of GDP after adoption (1990-2021).
    - Rule adopters recorded better average deficits and lower debt-to-GDP than non-adopters over 2010-19 (overall deficit/debt ~1.9 percent/38 percent for adopters vs 2.5 percent/34 percent for never-adopters; see Figure 6 Panels C and D for exact sample composition).
  - Causality concerns:
    - Correlations do not establish causality; adoption and outcomes may be driven by underlying preferences for prudence.
    - Literature combining resource and non-resource countries indicates positive correlation likely biased by endogeneity; stringency and supporting institutions (independent monitoring, good PFM) matter for effectiveness.
    - Constitutional budget balance rules associated with substantial debt reduction (Asatryan and others 2018); non-constitutional rules less effective.
- Simplicity and measurement challenges:
  - Resource-specific rules are typically more complex to compute and monitor than expenditure or nominal balance rules.
  - Major problems:
    - Reliance on long-term or projected commodity prices that are hard to estimate and often depend on current prices (reforecasting leads to large revisions).
    - Potential output estimation complicated by commodity cycles in addition to business cycles; cyclical adjustment can be asymmetric and biased (Chile and Colombia examples).
    - Overestimation of structural revenues leads to overspending; technical committees’ price projections underestimated booms and overestimated bursts, leading to larger deficits when prices collapsed.
- Competitiveness and Dutch disease:
  - Limited empirical assessments.
  - Caputo and Valdes (2016): negative/declining correlation between real exchange rate and terms of trade in Chile and Norway until late 1990s; correlation declined after fiscal rules introduction, and real exchange rates remained stable during large terms-of-trade shocks (Chile 2009-10; Norway 2003-2010), suggesting rules helped contain Dutch disease in those cases.

### III.C Resource Funds
- Objectives and models:
  - Funds used to shield budgets from revenue volatility or save for future generations; models include stabilization funds, saving funds, and financing funds.
  - Many funds accumulate foreign currency assets; some are sovereign wealth funds (SWFs) with large foreign assets.
  - Examples:
    - Chile’s copper stabilization fund (until 2006), Azerbaijan’s State Oil Fund (savings), Norway’s Government Pension Fund Global and Timor-Leste’s Petroleum Fund (financing funds).
- Size and diversity:
  - Figure 7 shows sovereign wealth funds in 2021 as percent of GDP (not reproduced here).
- Evidence on effectiveness:
  - Mixed empirical literature:
    - Davis and others (2001): funds do not significantly impact cyclicality of government spending for selected countries (1965-1999).
    - Ossowski and others (2018); Crain and Devlin (2003): funds may not have meaningful impact and can complicate fiscal management; may increase spending volatility.
    - Sugawara (2014) and Koh (2017): SWFs associated with smaller public spending volatility, particularly with high institutional quality.
    - El Badawi and others (2018): SWFs have robust stabilizing role and contribute to fiscal sustainability over 1985-2015.
- Institutional integration matters:
  - Success when funds are well integrated with broader fiscal rules and policy frameworks (Botswana, Chile, Norway).
  - Failures or suspensions when inconsistent with other policy objectives (Ecuador, Venezuela, Chad).
  - Examples of problematic practices:
    - Gabon: deposits into savings fund with low returns while paying higher interest on external debt.
    - Ecuador: complied with deposit rules but accumulated payment arrears due to earmarking and cashflow fragmentation.
    - Ghana: saved ~$500 million in SWFs 2012-2014 while borrowing ~$7 billion on international markets at ~5 percent higher rates than SWF returns.

### IV. Using A Fiscal Sustainability Anchor: An Appealing but Difficult-To-Implement Approach
#### A. Rationale and Main Aspects of the Fiscal Sustainability Framework (FSF)
- Framework summary:
  - Extension of standard fiscal sustainability to incorporate commodity revenues and resource depletion; related to Permanent Income Hypothesis (PIH) applications to oil countries.
  - Fiscal anchor: government “net wealth” = net financial assets (financial assets minus debt) + resource wealth in the ground (discounted sum of future resource revenues).
  - Operational rule: non-resource primary balance (NRPB), usually as a share of non-resource GDP. (Analytics in Appendix 1.)
- NRPB target:
  - Computed to stabilize net wealth at its current level; stabilizing net wealth can be defined:
    - (1) in real terms,
    - (2) as share of non-resource GDP,
    - (3) in real per capita terms.
  - Choice reflects views on intergenerational equity (e.g., keeping wealth constant as share of non-resource GDP may imply tighter nominal NRPB than stabilizing in real per capita terms).
- Central result:
  - Preserving wealth requires government to consume each year:
    - (1) return on financial wealth already accumulated,
    - (2) implicit return on net present value of future resource revenues,
    - (3) non-resource revenues.
  - In ratio formulation (share of non-resource GDP), current expenditure ratio equals growth-adjusted real return on net wealth ratio plus non-resource revenue ratio (details in Appendix 1).
- Cyclical adjustment:
  - FSF can be adapted for stabilization by using a cyclically-adjusted NRPB based on structural non-resource revenues (replace nominal revenues with structural revenues adjusted for non-resource output gap and revenue elasticity).
- Appeal:
  - Targeting non-resource balance is attractive since it can be broadly constant under unchanged policy, avoiding forced long-term cuts required by targeting fixed overall/primary balances as resource revenues decline.

#### B. Operational Challenges of FSF
- Calibration difficulties:
  - Financial wealth observable; underground resource wealth depends on long-term commodity price estimates that are hard to predict and often current-price dependent; nonlinear relationship between prices and economically viable reserves.
  - Estimated resource wealth can change abruptly when expected prices are revised.
- Volatility of expenditure targets:
  - Medium- to long-term price forecasts depend heavily on current prices; large current price moves lead to re-estimated resource wealth and revised NRPB targets, forcing sizeable revisions to expenditure plans.
  - Separating temporary vs permanent shocks is difficult: treating all shocks as permanent causes spending volatility; treating all as temporary causes insufficient saving.
- Sustainability paradox:
  - FSF could endorse unsustainable trajectories if initial resource wealth estimates are optimistic:
    - Early in extraction, high resource wealth estimates may permit low net financial assets or negative financial wealth financed by borrowing.
    - If prices collapse and resource wealth is revised down, country may face high debt when vulnerable and unable to roll over debt.
  - Result: initial optimism may produce unsustainable borrowing and spending paths.
- Primary balance exclusion of interest payments:
  - NRPB excludes interest payments; any primary-balance-based rule risks divergence because primary balance must offset automatic dynamics for net debt stability.
  - If primary balance falls short for one period, net debt can enter a self-reinforcing explosive loop; consistent period-by-period compliance is unrealistic, so a correction mechanism is needed.
- Public investment and physical assets omission:
  - FSF excludes physical assets and feedback of public investment on growth; criticized for potentially preventing warranted capital spending that could raise potential growth.
  - Alternative models (e.g., modified PIH) can incorporate front-loaded investment before stabilizing deficits.
- Social time preference and distributional realism:
  - Preserving resource-generated wealth across generations may clash with high present-time preferences and pressing development needs, especially in low-income countries.
  - When interest-growth differential is persistently negative, FSF (stabilizing wealth as share of non-resource GDP) may call for NRPB surpluses that seem undesirably high because income from net wealth is insufficient to prevent relative decline; additional surpluses required to maintain net wealth share.

#### C. Variants of the Fiscal Sustainability Approach
- Variants rooted in intertemporal budget constraint exist (details beyond excerpt); non-constant but sustainable NRPB paths can be derived within this family of approaches.

*International Monetary Fund, IMF Working Papers — A New Fiscal Framework for Resource-Rich Countries (sections II–IV excerpt).*

### Appendix 1. The computation of these paths is more demanding. And these mores sophisticated frameworks

### Appendix 1. The computation of these paths is more demanding. And these mores sophisticated frameworks

### Approaches with scaling-up of investment
- The “modified PIH” (MPIH) is designed to accommodate a temporary scaling-up of capital expenditure (IMF 2012a).  
- MPIH features:
  - A more front-loaded spending path financed by resource revenue that may be offset by lower spending (and higher NRPB relative to the standard PIH) in the future.
  - The expenditure path would no longer be smoothed, although fiscal policy would remain anchored within a sustainable framework.
- Simplest MPIH assumptions and implications:
  - Assumes the scaling-up of investment does not impact growth.27
  - In the absence of a fiscal multiplier effect, net wealth would initially decline because the stock of financial assets would not increase as fast as under the PIH and therefore would not fully offset the decline in resource wealth over time.
  - To ensure fiscal sustainability, fiscal consolidation would be needed following the period of investment scaling-up: spending would have to be reduced to improve the NRPB up to the point where net wealth is stabilized at the lower level28—or beyond if the government wishes to bring net wealth back to its initial level for equity reasons (IMF 2012b).
- If higher investment raises potential growth (via fiscal multiplier) and indirectly raises non-resource revenues:
  - Future NRPBs will improve, potentially reducing the need for consolidation on the spending side because NRPB improvement would come “automatically” from the revenue side.
  - The “automatic” improvement in the NRPB could allow sustaining the (lower) level of net wealth.
- Trade-offs emphasized by MPIH:
  - Investing resource revenues in real assets (not recorded in net wealth as defined in the FSF) versus financial assets (recorded in net wealth).
  - Although real-asset investment is likely to raise GDP growth and non-resource revenues and have positive indirect effects on financial asset accumulation in the long term, the multiplier effect may not be sufficient to offset forgone financial assets and associated forgone financial returns during the scaling-up period.
  - Absent future fiscal consolidation, scaling-up investment generally results in a higher stock of real assets but a lower stock of net financial assets (and lower net wealth29) in the long term.

### Approaches that do not stabilize wealth across generations
- Models that depart from wealth-stabilization assumptions include:
  - Finite-horizon depletion models (Baunsgaard and others, 2012):
    - Limit the horizon to a finite number of years; wealth is assumed to be depleted after a certain period rather than maintained for future generations.
    - Public expenditure is more front-loaded.
    - Requires gradual fiscal adjustment at some point to avoid an abrupt decline in the non-resource primary balance at the end of the extraction period.
  - “Bird-in-hand” policy rule:
    - Only the interest income accruing from accumulated financial assets is spent each year; targets the non-resource primary deficit equal to the anticipated return on existing financial assets that year.30
    - Resource revenues are almost fully saved.
    - Compared with the FSF, which consumes each year the return on accumulated financial wealth plus the implicit return on the NPV of future resource revenues, the bird-in-hand rule is an extreme form of precautionary saving (equivalent to assuming no future resource revenue).
    - While fiscally sustainable, it may create social and political tensions because public spending would be very low while resource revenues are accumulated, with high opportunity costs in early years in terms of foregone social and infrastructure spending.

### V. A New Medium-Term Fiscal Framework for Resource-Rich Countries — overview
- The paper proposes enhancing medium-term fiscal frameworks based on principles and rules centered on fiscal buffer accumulation and insurance against shocks.
- Key elements to be discussed: fiscal anchors (section VI) and operational rules (section VII).

### A. Long-term Fiscal Analysis and Strategy
- Long-term fiscal strategies in resource-rich countries should have horizons of several decades (generally 10 to 30 years).
- Long-term analysis elements include:
  - (1) assessing the economic size of the natural resources and risks around their valuation;31
  - (2) assessing effects of longer-term trends with fiscal impact, like demographics, technological changes, infrastructure needs, and climate change.
- Strategies should be consistent with other strategies addressing climate change and long-term factors; considerations differ for oil and gas producers versus metal producers.
- Country examples of long-term analysis initiatives:
  - Norway: 2017 report on “Long-term Perspectives on the Norwegian Economy” analyzing spending pressures from aging and expected gradual decline in returns from oil wealth over the next decades.
  - New Zealand: legislation requires the Treasury to produce a Statement on the Long-term Fiscal Position, an Investment Statement, and a Wellbeing Report at least every four years, and a Long-term Insights Briefing at least every three years; recent reports discuss climate change and aging impacts.

### B. Medium-Term Fiscal Framework
- Medium-term fiscal frameworks typically cover 3 to 5 years and are used to:
  - Set multiyear projections of key fiscal aggregates (expenditures, revenues, budget balances).
  - Provide costing of new measures.
  - Conduct risk assessments.32
- Effective frameworks typically include:
   - (1) fiscal plans that are feasible and stable,
   - (2) flexibility in response to shocks,
   - (3) transparent fiscal anchors,
   - (4) risk-based rules that ensure a path to debt sustainability and buildup of fiscal buffers,
   - (5) checks and balances to promote accountability (Caselli and others, 2022).

### The need to build precautionary buffers against shocks
- Fundamental principle proposed: build fiscal buffers to respond to shocks to promote economic stability and sustainable growth.
- Buffers should incentivize saving in booms and allow use during busts; size often needs to be larger than in non-resource-rich economies.33
- Considerations by reserve horizon:
  - Long reserve horizons (example given: Saudi Arabia has oil reserves equivalent to more than 200 times its annual consumption) imply exposure to long commodity price cycles and likely periods of durably low prices — strong case for building buffers.
  - Short-lived reserves: priority is ensuring any spending increases financed by resource revenues are sustainable once revenues disappear; an ad hoc framework may be more appropriate.
- Forms of fiscal buffers:
  - Asset-side buffer: accumulated financial assets that are liquid and available to finance the budget when needed.
  - Liability-side buffer: room for additional borrowing measured as the distance between current debt level and a “debt limit” above which debt becomes unsustainable with high probability or hurts growth (see IMF 2018a). Example: if debt distress threshold is 90 percent of GDP and current debt is 70 percent of GDP, the debt buffer equals 20 percent of GDP.
  - Overall buffer comprises financial assets plus capacity to borrow without creating unsustainable debt dynamics.
- When shocks occur, countries may smooth spending by depleting financial reserves or borrowing more — both are observed in practice (Figure 8 referenced).

### Key variables of the fiscal framework: fiscal anchors and operational rules
- A well-designed medium-term framework defines an explicit fiscal anchor: a quantified target to calibrate medium-term fiscal plans consistent with goals and fiscal responsibility.34
- The anchor may or may not be a formal fiscal rule; debt rules often guide medium-term policy.
- The fiscal anchor is not expected to bind in every annual budget but should be achieved over the medium term.
- Preferable design features:
  - Separate clearly the functions of fiscal anchors and operational rules; avoid using the anchor as an operational rule.
  - Use the anchor to provide medium-term direction and to calibrate operational rules in a sequential and comprehensive manner (IMF 2018a).

*Source: IMF Working Papers — A New Fiscal Framework for Resource-Rich Countries (Appendix 1 and sections V–VI excerpts).*

### Box 1. Link Between Insurance and Intergenerational Equity/Sustainability

### Box 1. Link Between Insurance and Intergenerational Equity/Sustainability

### Insurance framework and intergenerational equity
- The proposed fiscal framework for resource-rich countries is based on the principle of insurance: building precautionary buffers to hedge against shocks, especially large and infrequent shocks to the terms of trade.
- Intergenerational equity concern: whether the framework unduly favors the current generation.
- FSF framework definition of intergenerational equity: preservation of wealth across generations.
- Limitations noted:
  - FSF “total wealth” does not capture accumulation of physical assets (e.g., hospitals and schools), which many developing countries prefer.
  - Saving in anticipation of future adverse shocks is costly for current generations and nevertheless benefits future generations.
  - Preventing large fluctuations in output and mitigating disruptive fiscal adjustments can be seen as transferring wealth to the future.
- Conclusion: an insurance-based framework "does achieve some degree of intergenerational equity."

### Sustainability and long-run fiscal trajectory
- Concern: the insurance-motivated buffer may smooth expenditure during shocks but expenditure could remain too high once resource revenues are depleted—potentially forcing massive adjustments later.
- Risk described: fiscal deficit after resource exhaustion could be so high that net debt ratio would not stabilize at its steady state value (ratio could become explosive).
- The paper signals further discussion and a proposed solution in Section VII.

### Operational rules, credibility, and flexibility
- Operational rules should constrain the annual budget and apply to flow variables under government control (e.g., fiscal balance, expenditure envelope).
- Operational rules increase credibility; medium-term anchor may not be binding each year but annual plans should be constrained by binding operational rules to ensure consistency with medium- to long-term objectives.
- Choice of operational rule is country-specific; rules promoting stable expenditure growth in booms and busts are preferable to overall nominal balance floors that could increase volatility and procyclicality.
- Ways to balance credibility and flexibility:
  - Anchor based on risk analysis and reviewed regularly (for example every five years or after large shocks) with clear procedures and objectives.
  - Operational rules set in medium-term budgetary plans for a specified period and re-adjusted periodically (say every three to five years) based on a baseline scenario.
  - Use of escape clauses to accommodate exceptional circumstances and large adverse shocks.
- Note: If debt ratio is used as an operational rule and government tries to adjust annual budgets to compensate deviations, fiscal policy may become highly procyclical.

### Financial assets floor (NFACR) as a fiscal anchor
- Two practical anchor types: financial assets-based anchor (floor) or debt-based anchor (ceiling); anchors are often formulated as floors/ceilings rather than targets.
- Definition of NFACR floor:
  - NFACR = ratio of the stock of net financial assets (financial assets minus liabilities, or NFA) to annual commodity revenues.
  - Example: a ratio of 3 means the stock of NFA should be, at least, equivalent to 3 years of commodity revenues.
  - Financial assets considered: liquid assets of the government—financial investments, cash, or other liquid assets.
  - Preferable to calculate in net terms (deduct government debt) so assets are readily available and unencumbered; prevents compliance by simply borrowing.
- Suitability:
  - Particularly useful for large commodity exporters where economic activity and budget are vulnerable to commodity-related revenue volatility.
  - Commodity price shocks can be longer lasting, requiring larger buffers for gradual fiscal adjustment.
  - Cost of borrowing often sensitive to commodity prices; accumulation of a financial asset buffer, generally in foreign currency, helps smooth negative shocks.

### NFACR function and interaction with fiscal adjustment
- Main function: protect spending plans and ensure a more stable role for fiscal policy in the face of commodity revenue shocks.
- Stabilization effects:
  - When commodity prices surge, assets-to-revenue ratio declines → compliance calls for saving a significant share.
  - When commodity prices fall, ratio initially increases → use financial savings to fund budget and avoid large disruptive fiscal adjustments.
- Practical policy: buffers will often be used in conjunction with fiscal adjustment—revenue shortfalls likely accommodated by both using the buffer and carrying out some fiscal consolidation.
- Full protection without fiscal adjustment would require very large buffers; realistic goal is partial protection via assets plus gradual fiscal adjustment.
- For smaller, short-lived shocks the asset buffer could be sufficient without fiscal adjustment.

### Calibration of the NFACR anchor (three-step process)
- Calibration aims to manage larger and more persistent shocks and reflect country-specific assumptions.
- Step 1: Size of shock and desired degree of protection
  - Based on scenarios for future commodity prices, estimate the size of the largest negative shock that a country wishes to protect against (e.g., largest fall with 10% probability) over a given time period (e.g., 5 years).
- Step 2: Size of impact on budget
  - Estimate budgetary impact as: prices * quantities * government take relative to baseline over the period; could include production risk.
- Step 3: Estimate size of buffer (and extent of fiscal adjustment)
  - Asset buffer required depends on the maximum annual feasible fiscal adjustment in response to the shock.
  - If unwilling to adjust, require larger buffers to protect budget for full size of shock in every period; if willing to adjust, shortfall accommodated partly by adjustment and partly by asset buffer.
- Country-specific assumptions affecting calibration:
  - The degree of volatility of commodity revenues (historical volatility of prices and reliance of the budget on commodities).
  - The opportunity cost of building self-insurance (resources diverted from other policy objectives).
  - The desired degree of insurance against shocks (more protection → more upfront saving; less saving → more fiscal adjustment when shocks occur).

### Box 2 — Illustrative NFACR calibration for a large oil exporter
- Assumptions and setup:
  - Country assumed to generate yearly oil revenues averaging around 30 percent of GDP.
  - Government responds to large shortfalls by combining fiscal adjustment (cover half of revenue shortfalls via spending cuts or increases in non-oil revenue) and using returns of financial assets to cover the remaining half.
- Simulation details:
  - Oil price simulations use a random walk distribution around a baseline over the next 5 years.
  - Example year-by-year comparisons: baseline oil price at $86 vs downside $74 in 2023; baseline $80 vs downside $63 in 2024.
  - Initial asset level estimated in nominal terms at end-2022 and then expressed in percent of the 2022 oil revenue or GDP.
- Key illustrative findings:
  - A high degree of protection requires savings equivalent to around 3-4 years of oil revenues.
  - Example: "a NFA ratio of around 3 (which represents about 90 percent of a large oil exporter’s GDP) would allow to maintain spending in 90 percent of the most favorable shocks, provided that fiscal adjustment offsets half of the oil revenue losses."
  - For more extreme shocks (e.g., 5 percent worse shocks), an NFA ratio of 3 would be gradually depleted unless the government accepts larger fiscal adjustment.
  - For smaller shocks, less fiscal adjustment would be needed.
- Figure note: chart shows buffer size to insure against 75%, 90% or 95% of the most favorable shocks for a country with large oil revenues (about 30 percent of GDP), assuming fiscal adjustment equals half the loss in revenue in any given year.

### Responses to deviations from the NFACR anchor
- When the NFA ratio deviates from its anchor value, the country should aim at gradual convergence toward it.
- The NFACR floor is not expected to be met every single year; compliance should be achieved on average over the medium term.
- In large shocks, authorities can adopt an approach ensuring a gradual return to target as persistence or temporariness of the shock becomes clearer.

*Source: wpiea2023230-print-pdf - Box 1. Link Between Insurance and Intergenerational Equity/Sustainability*

### Appendix 3 describes in detail a possible adjustment

### wpiea2023230-print-pdf - Appendix 3 describes in detail a possible adjustment

### Adjustment mechanism for positive shocks (Appendix 3 example)
- Anchor objective: protect spending plans partly funded from annual commodity revenues by anchoring net financial assets to the commodity revenues used to fund the annual budget (NFACR).
- Steady-state assumption vs. outside steady state:
  - In prior sections, all annual commodity revenues were assumed to fund annual spending.
  - Outside the steady state, part of annual commodity revenues may be saved (e.g., transition to rebuild buffers); NFACR can be based on the commodity revenues actually used to fund the annual budget.

- Example calibration and mechanics (as developed in Appendix 3):
  - Anchor: an NFACR ratio of 3.
  - Initial steady state: assets = 30; resource revenues = 10 (initially fully used to fund the annual budget).
  - Positive shock: annual revenues increase to 13.
  - One possible response: spend one fourth of the revenue surprise initially (if anchor = 3, spend 1/4 of surprise), raise spending gradually if the shock proves persistent.
  - Under this path the ratio of NFA to commodity revenues (used to fund spending) is always kept at 3.
  - If the revenue rise proves persistent, eventual outcomes:
    - Annual spending would increase by 3, matching the increase in commodity revenues.
    - Buffers would go to 39.
  - Operational caveats:
    - The “one fourth of the revenue surprise” rule is illustrative; exact path depends on country circumstances and operational rules.
    - Speed of scaling up spending depends on needs and capacity to spend efficiently.

### Response to negative shocks
- NFACR framework flexibility:
  - Allows use of financial savings to fund the budget and avoid large, disruptive fiscal contraction.
  - Choice between using buffers versus tightening the budget (cutting spending or increasing non-commodity revenue) depends on shock type and country circumstances.
- Suggested approaches by shock severity:
  - Small, temporary shocks: use buffers to maintain spending plans during the first(s) year(s).
  - Larger or more persistent shocks: fiscal adjustment generally needed; governments should prepare plans to rebuild buffers over time.
- Post-shock buffer restoration:
  - Key question: pursue active fiscal adjustment to restore the buffer or wait for economic improvement and cyclical revenue increases to generate savings.
  - For moderate and frequent shocks, NFA likely fluctuate cyclically and buffers may replenish automatically.
  - For large negative tail risks, buffers may not replenish automatically; active fiscal adjustment (compress spending for a few years) would be needed.
- Resilience note:
  - Very large shocks may require countries to review framework and calibration of rules to ensure adequacy.

### Other possible financial asset anchors
- Alternative to NFACR: calibrate nominal NFA buffer to be fully depleted after a particular shock.
  - Initial NFA should cover revenue losses over a certain horizon.
  - IMF (2012a, b) approaches: value-at-risk and model-based methods to estimate minimum buffer to absorb tail risks in resource revenue volatility.
  - Target: buffer large enough to ensure, with high probability, resources to respond to shocks in the forecast horizon.
- Trade-offs:
  - This approach tends to require smaller buffer accumulation compared to NFACR.
  - If shock is more persistent than expected, significant fiscal adjustment will be required once buffers are depleted.
  - Countries must assess benefits versus costs of different buffer sizes.

### Anchoring on a debt ceiling
- When suitable:
  - Best for countries with lower budgetary dependence on resource revenues and lower willingness to insure against risks.
  - Practical and politically acceptable where moderate shocks can be accommodated through temporary rise in debt and gradual fiscal adjustment; windfalls used to reduce debt.
- Design principles:
  - Express anchor as gross debt ratio ceiling with denominator = non-resource GDP to reduce volatility.
  - Numerator should cover both domestic and external debt.
  - Governments may still hold small financial assets (government deposits) for liquidity management; such assets would not be primary shock-absorption buffers.
- Conditions for adoption:
  - Appropriate where most negative commodity revenue shocks have moderate budgetary impact and can be accommodated via borrowing and adjustment.
- Debt vs. assets as shock absorbers:
  - Debt could be a “better” shock-absorber if, under negative shock, (1) debt costs do not increase much and/or (2) financial asset prices decline significantly.
  - Example: access to concessional debt whose cost is uncorrelated to shocks can favor debt issuance over selling assets.
  - Typical case: debt costs of resource-rich countries often surge when commodity prices collapse, while financial asset returns may be more resilient (often held abroad and supported by exchange rate depreciation).
- Other rationale for a debt ceiling:
  - Manage debt vulnerabilities (elevated default risk, crowding out, sovereign-debt nexus).
  - If public debt rises significantly, priority may be to lower debt compatible with macro stability and support with a prudently calibrated debt anchor.

### Calibration of the debt anchor
- Follow IMF (2018a) principles:
  - Compute a “safe” debt level by setting a maximum debt limit and deducting a safety margin reflecting volatility of macro shocks (GDP, exchange rate).
  - Tailor methodology for commodity producers (see Appendix 4 suggestions):
    - Sample: re-estimate fiscal reaction function focusing on commodity producers subsample.
    - Specification: replace terms-of-trade variable with a commodity prices variable (e.g., “commodity terms of trade”).
    - Differentiate elasticities between positive versus negative commodity price gaps due to non-linear fiscal balance response.
    - Macroeconomic shocks: generate shocks that capture persistence and skewed distribution with non-zero tail-event probability; multivariate normal draws may be inadequate.

### Response to shocks under the debt anchor
- Debt-to-GDP ratio reacts strongly to GDP shocks (more than fiscal deficit ratio).
- If debt ratio is at anchor, a negative GDP shock can breach the ceiling.
  - Maintaining deficit at or below its ceiling should bring debt back asymptotically, but may take long and be noncredible.
- Policy options:
  - Introduce a “correction mechanism” in operational rule: fiscal deficit or expenditure ceilings include an adjustor that tightens rules proportionally to excess debt ratio (turns operational rule into a fiscal reaction function).
  - Example: Poland (2013) reduced expenditure growth rate gradually if public debt-to-GDP thresholds exceeded.
- Responses are asymmetric:
  - Positive GDP shock lowers debt ratio, but relaxing fiscal policy may be inappropriate if shock is temporary.
  - Large negative shocks may warrant initial debt increases (e.g., Covid-19); a credible transition path is needed to return debt below the ceiling.

### Dual anchor systems (debt ceiling and financial asset floor)
- Rationale and complexity:
  - Dual anchors may co-exist temporarily (transition) or permanently; configuration is more complex and often one buffer will dominate.
  - Transitional examples:
    - While building a financial asset buffer, a debt ceiling can limit excessive debt used to smooth shocks.
    - Developing country with future revenue increases may use concessional borrowing within debt ceiling while promoting asset accumulation as revenues rise.
- Permanent dual anchors:
  - Countries may accumulate both buffers to minimize insurance cost given different correlations of debt costs and asset returns to shocks.
  - Holding both allows selecting cost-effective option when shocks hit (e.g., liquid assets to weather loss of market access).
- Calibration challenges:
  - Existing calibration methods for debt and financial anchors risk double-counting buffers.
  - Joint calibration likely requires a small calibrated portfolio-optimization model to determine optimal mix of debt and FX assets, accounting for shock trade-offs.
  - Countries lacking asset-liability management capacity and with expensive debt should focus on keeping debt low rather than accumulating large financial assets.

### Selecting and calibrating operational rules
- Purpose:
  - Operational rules guide yearly budgets and link them to medium-term fiscal plans.
  - Common options: overall balance rule, non-resource balance rule, expenditure rule.
- Expenditure rule focus (advantages and considerations):
  - Key challenge: prevent excessive expenditure growth during booms to allow buffer accumulation and support stable growth.
  - Advantages:
    - Simpler to operate and monitor; governments control spending envelope.
    - Reduces expenditure volatility and supports buffer creation by saving windfalls.
    - Forces consideration of sustainability and spending prioritization/efficiency.
  - Disadvantages/limitations:
    - Doesn’t incentivize revenue mobilization.
    - Requires good calibration of threshold and ability to tap buffers; infeasible if country cannot offset cyclical revenue shortfalls by borrowing or depleting assets.
- Period of accumulation of the precautionary buffer:
  - During buffer constitution, a formal operational rule is possible but not necessary; priority is an indicative benchmark for annual fiscal adjustment to reach target by a given date.
  - IMF (2018a) provides a formula for transition path; two parameters: date to reach target and duration of adjustment (could be equal for simplicity).
  - Two considerations for transition timing:
    - Annual fiscal adjustment must be realistic politically and socially.
    - Adjustment could be achieved via cyclical upturns in business or commodity cycles rather than structural tightening.
- Period of maintenance and use of the precautionary buffer:
  - Operational rule’s main function: contain and stabilize spending so resource revenues are not fully spent in booms, preserving insurance and sustainability.
  - Expenditure envelope calibration:
    - Short-to-medium term: calibrate to ensure anchor compliance and buffer maintenance.
    - Long term: ensure expenditure envelope does not place gross debt on an explosive path once resources are depleted.
  - Two formal conditions (using NFA anchor):
    - Sustainable expenditure ratio to stabilize NFA over the cycle: formula given using cyclically-adjusted variables, trend GDP growth (푔푔
푇푇푇푇푇푇), and trend GDP (푌푌
푡푡
푇푇푇푇푇푇).
    - Sustainable expenditure ratio to stabilize gross debt in the longer term: analogous formula using long-term growth parameters (푔푔
푇푇푇푇푇푇
퐿퐿푇푇).
    - Note: these two ratios need not be identical ex ante; medium-term expenditure envelopes may need adjustment for long-term sustainability.
  - Calibration caveat:
    - Any estimate of perpetually financeable expenditure is imprecise; periodic updates required as new information (commodity prices, persistence, revenue mobilization prospects) becomes available.
- Design of the expenditure rule in steady state:
  - Maintain expenditure ratio by an expenditure growth rule where real expenditure grows like real potential growth: ∆E/E = ∆Y(PTT)/Y(PTT).
  - Some designs allow expenditure net of new revenue measures, but for commodity producers priority is to keep a firm expenditure path and avoid upward revisions when commodity prices rise that cannot be reversed in downturns.

*Italic source: Content from the specified PDF chapter/section (wpiea2023230-print-pdf).*

### Box 3. Suggested Steps to Calibrate the Sustainable Expenditure Envelope

### Box 3. Suggested Steps to Calibrate the Sustainable Expenditure Envelope

### Calibration steps (anchor = gross debt ceiling treated as a target)
- Step 1: set the anchor’s target.
  - Anchor and operational rule must be calibrated consistently.
  - Example assumption: debt rule expressed as a share of GDP with targeted ratio = 40 percent of GDP.
- Step 2: derive the anchor-stabilizing overall balance.
  - For a given anchor, infer the fiscal balance that would stabilize it.
  - Example assumption: nominal GDP growth = 6 percent.
  - Fiscal balance consistent with debt = 40 percent of GDP = fiscal deficit of 2.3 percent of GDP (= -40*0.06/1.06).
- Step 3: derive the expenditure ratio.
  - Estimate the spending ratio that can be sustained in the steady state, expressed in percent of trend GDP and with all variables in structural terms.
  - Make an assumption regarding the steady-state revenue ratio.
  - For resource revenue ratio, use long-term average of resource prices to avoid dependence on commodity cycles and transitory shocks.
  - For non-resource revenue, prudently assume no revenue mobilization gains going forward (reforms can be factored in when they have borne fruit).
  - Example: projected revenue ratio = 30 percent of GDP (10 percent of GDP resource revenues; 20 percent of GDP non-resource revenues) → expenditure ratio = 32.3 percent of GDP (difference between the debt-stabilizing balance and the revenue ratio).
- Step 4: adjust the expenditure ratio according to the long-term sustainability benchmark (if needed).
  - The expenditure ratio from step 3 may still be high from a long-term perspective.
  - Example long-term assumptions: long-term gross debt anchor = 60 percent of GDP; growth = 6 percent → debt-stabilizing balance ≈ fiscal deficit of about 3.4 percent of GDP.
  - Example long-term non-resource ratio projected to increase from 20 percent of GDP to 28.9 percent of GDP; implication asserted in text: "28.9-32.3=3.4". If this increase is deemed unachievable, the expenditure ratio must be adjusted downward, perhaps during an adjustment period preceding the exhaustion of resources.
- Process frequency and flexibility:
  - Recalibrate every time the country defines medium-term fiscal plans (say, every 4 years) within a well-defined fiscal framework.
  - Countries can adjust expenditure ceilings and anchors depending on prevailing circumstances.

### Illustrative numerical highlights and relationships
- Anchor target (illustration): 40 percent of GDP.
- Nominal GDP growth (illustration): 6 percent.
- Debt-stabilizing fiscal balance for 40 percent of GDP: fiscal deficit of 2.3 percent of GDP (= -40*0.06/1.06).
- Projected revenue ratio (illustration): 30 percent of GDP, comprising:
  - resource revenues = 10 percent of GDP
  - non-resource revenues = 20 percent of GDP
- Resulting steady-state expenditure ratio (illustration): 32.3 percent of GDP.
- Long-term anchor (illustration): gross debt = 60 percent of GDP; growth = 6 percent → debt-stabilizing balance ≈ fiscal deficit of about 3.4 percent of GDP.
- Example long-term non-resource ratio projection: increase from 20 percent of GDP to 28.9 percent of GDP; text notes numerical relation "28.9-32.3=3.4" in the calibration discussion.

### Policy implications and operational guidance
- Treat gross debt ceilings (or financial asset floors) as anchors/targets for calibration.
- Use structural measures and trend GDP for steady-state expenditure ratio estimation to avoid volatility from actual nominal spending.
- For resource revenues, rely on long-term averages of prices to avoid overreacting to commodity cycles and transitory shocks.
- For non-resource revenues, adopt conservative baseline assumptions (no automatic revenue mobilization gains) and incorporate realized reforms when they have effectively borne fruit.
- Where long-term sustainability benchmarks imply tighter outcomes than short-term calibration, plan downward adjustments in the expenditure ratio—potentially phased in before resource exhaustion.
- Embed this calibration exercise within periodic medium-term fiscal planning (illustrative cadence: every 4 years) to allow for anchor and ceiling adjustments in response to evolving circumstances.

*Source: Box 3. Suggested Steps to Calibrate the Sustainable Expenditure Envelope (wpiea2023230-print-pdf).*

### Appendix 2. Choice of Operational Rules

### Appendix 2. Choice of Operational Rules

### Overall or primary balance rule
- Description:
  - The overall balance rule sets a minimum fiscal surplus (if positive) or a deficit ceiling (if negative). Fiscal balance is total government revenue minus total government spending.
  - The primary balance rule differs only by excluding net interest payments from expenditure; arguments are broadly similar.
- Advantages:
  - Simplicity and transparency.
  - Relatively easy to calibrate consistent with the fiscal anchor via formulas relating the overall balance to net debt (or net financial assets). Provided estimates of potential growth are available, these formulas can set the overall balance target that would bring net debt to its anchor level at the end of a predetermined period.
  - Commonly combined with a debt-to-GDP ceiling associated with a deficit-to-GDP ceiling.
- Disadvantages and risks:
  - Can prescribe a very procyclical and unstable fiscal stance: short-term revenue volatility must be offset on the spending side to maintain the nominal balance target.
  - May allow temporary increases in revenue to drive permanent increases in expenditure.
  - Adjusting the rule for business and commodity price cycles can mitigate procyclicality but raises additional problems (see Box A2).

### Box A2: Challenges with Correcting the Fiscal Balance for the Commodity Price Cycle
- Concept:
  - Government revenues decomposed into a “structural” (trend) component, a cyclical component related to the non-resource business cycle, and a cyclical component related to the commodity prices’ short-term cycle.
  - The cyclically-adjusted balance (CAB) excludes the two cyclical components by filtering out high-frequency shocks.
- Implementation and measurement challenges:
  - Implementing a CAB rule requires capacity to estimate both economic and commodity cycles.
  - Commodity prices exhibit cycles of various frequencies, including very long “super-cycles,” making separation of trend, short cycle, and long cycle difficult.
  - Commodity price shocks display high persistence; for oil, prices have been found to behave like a random walk without drift, with most shocks “permanent.”
  - Structural commodity prices are often computed as moving averages of past prices or a combination of past spot and futures prices. If shocks are persistent and smoothing is strong, average forecasting errors and lengths of same-sign forecast errors will be high.
  - If estimates of structural resource prices are highly correlated with actual prices, the CAB will closely resemble the nominal balance and fiscal policy may remain highly procyclical despite the adjustment.
- Example operationalization:
  - The “CAB rule” is a ceiling imposed on the CAB, generally expressed in percent of potential GDP (example: CAB should not exceed 3 percent of potential GDP).

### Non-resource (primary) balance rule
- Description:
  - The non-resource balance rule delinks expenditure from resource revenues by focusing on the fiscal balance excluding resource-related revenue (and possibly excluding net interest if primary).
- Advantages:
  - Strongly mitigates procyclicality by decoupling fiscal policy from volatile commodity prices and volumes in the short run.
  - As financial buffers are built, also insulates policy from less frequent and larger shocks.
  - Residual procyclicality from the non-commodity economy can be addressed by applying a cyclical adjustment.
- Key challenges:
  - Calibration: excluding selected expenditure or revenue items does not guarantee convergence to the fiscal anchor because excluded volatile items still affect financing needs.
  - May be more appropriate when the main objective is macroeconomic stability rather than building buffers or reducing debt vulnerabilities; the non-resource balance could be re-calibrated every 3-5 years (or another frequency) to ensure convergence toward the anchor.
  - Practical difficulty in separating revenue and spending associated with resources versus “other” revenue and spending.
  - Communication challenges: the public may not understand why all resource revenues are removed from the indicator targeted by the government; it can be confusing if the rule is expressed as a deficit ceiling on the non-resource balance while the overall fiscal position (including resource revenues) typically generates a surplus.
- Implication:
  - These communication and technical issues argue in favor of using an expenditure rule instead, but choice depends on country circumstances.

### Revenue allocation rules
- Description:
  - Rules that allocate resource revenues between a portion that is saved and a portion that is used in the budget; excess (deficit) of actual resource revenues relative to budgeted revenues is accumulated in (withdrawn from) a resource fund.
- Types:
  - Revenue split rules:
    - Set aside a certain percentage of revenues using an ad hoc criterion (e.g., save revenues above the amount initially budgeted or the average of past revenues; or save a predetermined percentage of commodity revenues).
  - Price smoothing rules:
    - Allocate revenues based on a reference price. If actual resource revenues exceed reference-revenue-consistent levels (given the reference price), the difference is saved and used in periods of shortfall. Reference prices derive from an automatic formula (e.g., average of past and future prices) or are provided by an independent committee.
- Limitations:
  - These rules constrain only the split between saving and spending of resource revenues and do not limit borrowing.
  - In extreme scenarios, setting aside all commodity revenues could still coincide with fiscal deterioration if high spending is financed through borrowing.
  - Consequently, revenue allocation rules may not be sufficient to achieve macroeconomic stability or guarantee eventual attainment of the fiscal anchor.

*Source: wpiea2023230-print-pdf - Appendix 2. Choice of Operational Rules*

### Appendix 4. Debt Rule Calibration for

### Appendix 4. Debt Rule Calibration for Commodity Producers: A Case Study

### Methodology
- Standard calibration exercise consists of three main steps:
  - Assume a debt limit (threshold beyond which debt distress is likely) of 80 percent of GDP, taking into account Colombia’s debt carrying capacity and the IMF Debt Sustainability Framework for market access countries.
  - Estimate the required safety margin (borrowing space between the debt anchor and the debt limit) using stochastic simulations of macroeconomic and fiscal shocks over an 8-year horizon with a system of simultaneous equations (debt accumulation equation and a fiscal reaction function where primary balance responds to macro variables, including past debt).
  - Summarize simulated debt trajectories in a fan chart to set the debt anchor and calculate the probability that public debt remains below the debt limit in the medium term.
- Historical joint distribution of shocks estimated using data for 1990-2020 on:
  - real GDP growth
  - real effective interest rates
  - real exchange rates
  - terms of trade gaps
  - external loan disbursements

### Tailoring the Calibration for Commodity Producers
- Three main changes to the 2018 framework to reflect commodity-producer specificities:
  - Use a commodity-based terms of trade (Gruss and Kebhaj 2019) instead of overall goods and service terms of trade.
  - Differentiate between negative and positive terms of trade gaps to capture non-linearity in fiscal responses (primary balances often improve during booms and deteriorate during busts; improvements during booms are usually less significant than deteriorations during crashes).
  - Estimate fiscal reaction functions on a sample focused solely on commodity producers (excluding economies where resource revenue > 50 percent of general government revenue in 2020) to better represent fiscal policy response for commodity-dependent countries.

### Parameterization and Key Statistics (from Table A2)
- List of countries used in the sample includes: Algeria, Angola, Bolivia, Botswana, Burkina Faso, Cameroon, Central African Republic, Chad, Colombia, Republic of Congo, Cote d'Ivoire, Democratic Republic of the Congo, Ecuador, Gabon, Ghana, Guinea, Guyana, Indonesia, Iran, Liberia, Mali, Mauritania, Mexico, Mongolia, Mozambique, Niger, Nigeria, Papua New Guinea, Peru, Russian Federation, Sierra Leone, South Africa, Sudan, Yemen, Zambia, Zimbabwe.
- Parameters and values:
  - Last year debt to GDP ratio (%) 65.7
  - Debt limit (% of GDP) 80
  - Risk Tolerance (%) 5
  - Average historical growth rate (% of GDP) 3.2
  - Average historical primary balance (% of GDP) 0.38

### Fiscal Reaction Function Estimates (Table A3)
- Estimated specification (notation as in source):
  - ppbb_it = α_i + ρ ppbb_it−1 + β1 G_it aNaagcp_it + β2 (1−G_it) aNaagcp_it + θ a_t−1 + φ X_it + ε_it
  - where ppbb is primary balance to GDP, aNaagcp is deviation of commodity terms of trade from trend, G is indicator for positive terms of trade gaps, a_t−1 is lagged debt-to-GDP, X includes controls such as public and publicly-guaranteed external disbursements.
- Estimated coefficients (robust standard errors in parentheses; significance stars as reported):
  - Lag of primary balance to GDP ratio 0.359*** (0.0510)
  - Lag of debt to GDP ratio 0.0160*** (0.00465)
  - Negative output gap 0.0904** (0.0339)
  - Negative terms of trade gap 0.119* (0.0621)
  - Postive terms of trade gap 0.299*** (0.0479)
  - External disbursement -0.159* (0.0927)
  - Constant -0.948** (0.459)
  - Observations 851
  - R-squared 0.321
  - Number of ifscode 37
  - Number of groups (not quantified in table)

### Results and Implications
- Baseline with default 2018 parameterization (including a 5 percent risk tolerance) implies:
  - Without commodity-specific adjustments, a debt-to-GDP ratio of around 70 percent would be considered a safe debt anchor for Colombia (Figure A1, Panel A).
- Accounting for commodity revenue risks and tailoring the toolkit:
  - The debt anchor is revised downward to around 55 percent of GDP (Figure A1, Panel B).
  - This implies Colombia would need an additional buffer of around 15 percent of GDP compared to the default calibration to comfortably withstand repeated negative shocks without endangering fiscal sustainability.

### Figure A1 Summary
- Panel A (Default Toolkit): Fan chart based on FAD (2018) default parameterization suggests a safe debt anchor near 70 percent of GDP with 5 percent risk tolerance.
- Panel B (Accounting for Commodity Revenue Risks): Tailored calibration for commodity producers revises the prudent debt anchor to around 55 percent of GDP.

*Source: Appendix 4, wpiea2023230-print-pdf.*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023230-print-pdf.pdf_