## 1brnea2023002

## Source details

**Canonical URL:** [1brnea2023002](https://www.imf.org/-/media/files/publications/cr/2023/english/1brnea2023002.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2023/english/1brnea2023002.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2023/english/1brnea2023002.pdf.json)

---

### Behavioral responses promoted by alternative CO2 mitigation policies
- Comprehensive carbon pricing promotes:
  - Power generation: shifting from natural gas and coal to renewables, and potentially to hydrogen and fossil generation with carbon capture and storage.
  - Industry: reducing CO2 and electricity intensity and output levels (examples: mandating carbon reporting, usage-based electricity tariffs, energy-efficient appliances, LED streetlights, electric vehicles).
  - Transportation: shifting to more efficient ICE vehicles, to electric or other zero-emission vehicles, and reducing vehicle miles travelled.
  - Buildings: reducing CO2 intensity, electricity intensity, and energy demand via energy-efficient construction and appliance upgrades.
- Non-pricing instruments:
  - Renewable portfolio standards and feed-in tariffs only promote fossil-to-renewable generation switching.
  - Emission rate regulations/feebates reduce fleet emissions gradually but do not accelerate retirement of old vehicles or reduce vehicle miles travelled.
  - Incentives for net zero new buildings reduce emissions very gradually given <2 percent annual replacement of building stock.
- Practical considerations:
  - Non-pricing instruments are used to complement carbon pricing and may be more politically acceptable by avoiding energy price pass-through.
  - Feebates can kick-start de-carbonization in hard-to-abate sectors (transportation, buildings).
  - Policymakers must balance carbon pricing (most efficient) with reinforcing instruments (less efficient but more acceptable).

*Prepared by: Ian Parry, Simon Black, Karlygash Zhunussova.*

### Momentum, coverage, and price signals for carbon pricing
- Global and regional coverage:
  - Carbon pricing schemes operate in 45 countries (national and sub-national initiatives; EU ETS included for EU countries).
  - Nationally: 21 carbon taxes and 6 ETSs implemented.
  - 28 percent of global GHGs are formally subject to pricing.
  - Countries in Asia with carbon pricing: Indonesia, Japan, Korea, Singapore; under consideration: Thailand, Vietnam.
- Prices and commitments:
  - EU ETS prices are currently around $100 per tonne.
  - Canada has committed to an equivalent US$140 price by 2030.
  - Sweden: $130 per tonne (economywide average price in 2022).
  - Economywide average prices in 2022 varied from below $5 to $130 per tonne.
  - The average price across schemes is $20 per tonne.
- Coverage variation:
  - GHG emissions subject to pricing vary from below 30 percent to over 70 percent (e.g., Canada, Germany, Korea, Sweden).
- Brunei Darussalam context:
  - Under the Income Tax (Petroleum) Act of 1963, oil and gas companies face a fixed 55 percent petroleum income tax rate.
  - No sales or excise tax on fuel.
  - Under its carbon pricing strategy, Brunei Darussalam aims to launch an ETS and from 2025 plans to reduce carbon intensity from all industrial sectors and new power utilities.

### Instrument choice, tradeoffs, and implementation design
- Carbon taxes:
  - Typically under finance ministries; easier to administer and often applied midstream.
  - Provide price certainty; require periodic adjustment to meet emissions goals.
- ETSs:
  - Typically under environment ministries; require MRV systems, trading markets, allowance platforms.
  - Provide emissions certainty if cap set appropriately; can exhibit allowance price volatility.
  - Price stability mechanisms (e.g., price floors) can reduce volatility.
- Price vs emissions certainty:
  - Carbon taxes: price certainty, emissions depend on market response.
  - ETSs: emissions certainty (if cap set), prices determined by markets and may be volatile.
  - ETSs with floors/ceilings can approximate carbon taxes.
- Hybrid and sectoral designs:
  - Example hybrid: ETS for power and industry; carbon tax for transportation and buildings.
  - Cost effectiveness requires aligning prices across instruments; applying both tax and ETS to same base is generally inefficient.
- Revenue use and distribution:
  - Carbon tax revenues more likely to accrue to general budget: fully used for general purposes in 16 schemes; earmarked in only five.
  - ETS auction revenues more likely earmarked: 5 of 7 ETS schemes partially earmarked.
  - Free allowance allocation used early in some ETSs to build support; auctioning used in others.
  - Productive revenue uses (public investment) can yield large efficiency gains; universal lump-sum transfers forgo some efficiency benefits.
  - Distributional burdens are mildly regressive in some cases; free allowance allocation concentrates rents to shareholders.
- Political economy:
  - ETSs can be more politically feasible where free permits build industry support.
  - Carbon taxes can be made more acceptable via revenue recycling, communications, and targeted support.
  - Design levers: phase down free allocation (example: EU ETS from 80 percent in 2013 to 30 percent in 2020); mimic free allocation via targeted revenue recycling.

### Tradeable Performance Standards (TPS) and feebates (design and behavioral effects)
- TPS features:
  - Government sets CO2 emission rates per unit of output by industry; firms meeting standards can sell credits to non-compliant firms.
  - TPSs promote emissions-intensity reductions similar to carbon pricing but do not promote consumer demand responses because there is no price pass-through.
  - Example: Canada’s federal TPS for industry where provincial policies do not apply.
- Feebates:
  - Sliding fees for firms above a pivot emission rate and rebates for firms below it; pivot set to industry average and updated yields approximate revenue neutrality.
  - Fee = CO2 price × (firm CO2/unit output − pivot CO2/unit output) × firm production level.
  - Feebates provide certainty over emissions price and are cost-effective within industries; can be harmonized across industries.
  - Like TPSs, feebates do not charge average firm for remaining emissions and thus do not promote consumer demand responses.
- Comparative points:
  - Feebates provide price certainty and automatic cost effectiveness across firms.
  - TPSs require liquid credit markets to be cost effective; can be made more feebate-like with out-of-compliance fees/subsidies.

### Quantitative assessment methodology (CPAT) and timing assumptions
- CPAT model:
  - Spreadsheet-based projecting fuel use and GHG emissions for major energy sectors in 188 countries.
  - Impacts depend on proportionate impacts on future fuel prices and price responsiveness of fuel use.
  - Fuel and electricity price elasticities over the longer term parameterized generally between -0.5 and -0.8.
  - Carbon emission factors by fuel product from IIASA (Wagner 2020); 2019 emissions calibrated to UNFCCC GHG; 2020-1 calibrated to EC-JRC (Crippa and others 2022), Global Carbon Budget (Friedlingstein and others 2021), and various sources.
  - CPAT linked to input-output tables to infer impacts on production costs, consumer prices, and household burdens.
- Timing with global energy prices:
  - Phasing a $50 carbon price on top of projected prices would imply 2030 O&G prices 16 and 59 percent below mid-2022 levels, while coal prices would be 11 percent higher (note: phrasing indicates alternative projections).

### Brunei-specific emissions baseline, targets, and CPAT scenarios
- Baseline and targets:
  - Authorities submitted an unconditional target: cut GHGs to 20 percent below baseline levels in 2030 in the NDC (reference year 2015, emissions of 11.6 MtCO2e).
  - Energy-related emissions = 97.2 percent of Brunei Darussalam’s 13.6 MtCO2 GHG emissions in 2022.
  - Sector shares of 2022 emissions:
    - Power generation and own-use by the energy industries: 54.6 percent.
    - Power generation – public utilities: 21.5 percent.
    - Fugitive emissions: 8.6 percent.
    - Transport: 8.9 percent.
    - Non-specified industries: 3.0 percent.
    - Industrial processes: 1.0 percent.
    - Waste: 1.5 percent.
    - Agriculture: 0.3 percent.
    - Buildings: 0.6 percent.
  - Authorities announced move towards Net Zero by 2050; linear extrapolation implies economy-wide emissions around 8.8 MtCO2e or 60 percent reduction to BAU scenario.
- IMF staff baseline projection:
  - GHG emissions increasing 73 percent from 12.7 MtCO2e in 2021 to 21.9 MtCO2e in 2030 (i.e., 26 percent lower than authorities’ projection of 29.5 MtCO2e).
  - Drivers: projected GDP increase 21 percent and energy intensity of GDP increase 68 percent.
- Modeled mitigation scenarios (definitions):
  - Carbon tax 50: explicit carbon tax from $20 per tonne in 2024 rising linearly to $50 per tonne in 2030; coverage: power, industry, transport, buildings.
  - ETS: ETS from $20 per tonne in 2024 rising linearly to $50 per tonne in 2030; coverage: power and industry.
  - Feebates: feebates in power and industry from $20 per tonne in 2024 rising linearly to $50 per tonne in 2030.
  - Subsidy reform: phase out consumer-side fossil fuel subsidies starting 2024 linearly over five years.
  - Combination: Carbon tax (as Carbon tax 50) combined with fossil fuel subsidy reform (starting 2024 linearly over five years).

### Emissions impacts by scenario and sector (2030 relative to baseline)
- Aggregate emissions reductions in 2030:
  - Carbon tax 50: 44 percent reduction.
  - ETS: 41 percent reduction.
  - Feebates: 10 percent reduction.
  - Subsidy Reform: 24 percent reduction.
  - Combination: 50 percent reduction.
- Sectoral contributions in 2030:
  - Power sector reductions vary from 24-85 percent across scenarios.
  - Industry reductions vary from 7-76 percent across scenarios.
  - Transport accounts for less than 10 percent of emissions reductions in 2030.
  - Buildings account for 1 percent of reduction in 2030.
- Interpretation:
  - Power and industry account for the most CO2 emissions reductions in all scenarios due to larger emissions shares and higher responsiveness to carbon pricing.

### Electricity generation and renewables (facts and modeled outcomes)
- Historical/current shares:
  - Natural gas share in electricity generation: 78 percent in 2020 (down from nearly 99% in 1990).
  - Coal was 28 percent of power in 2020.
  - Solar generated 2 GWh in 2020, about 0.03 percent of electricity.
  - Authorities target at least 30 percent renewable share of energy power generation mix by 2035 (current less than one percent).
- Model results for 2030 electricity mix:
  - Carbon tax 50 and ETS raise renewables to about 14 percent of electricity generation in 2030.
  - Other scenarios reach about 7-10 percent renewables in 2030.
  - Modeled results fall far short of the government’s 30 percent by 2035 plan.

### Fiscal and macroeconomic implications (revenues, growth, and recycling)
- Potential revenues and budget savings:
  - Carbon pricing could raise up to 7 percent of GDP in 2030 on top of BAU.
  - In 2030:
    - Feebates scenario yields additional 1.6 percent of GDP in budget savings.
    - Combination scenario yields 7.2 percent of additional revenues.
  - Cumulative revenues/budget savings 2024-2030:
    - Feebates: $1.1 billion.
    - Combination: $5.5 billion.
  - Note: Feebates are revenue-neutral by design, but erosion of subsidized base yields budget savings in the model.
- Growth impacts without revenue recycling (percentage point reductions in GDP growth in 2030):
  - Carbon Tax 50: 2.9 percentage points.
  - ETS: 2.5 percentage points.
  - Feebates: 1.0 percentage point.
  - Subsidy Reform: 3.2 percentage points.
  - Combination: 4.3 percentage points.
- Illustrative revenue recycling:
  - 75 percent of carbon tax revenues recycled to productive public investment and 25 percent to targeted cash transfers.
  - Under this recycling, negative impact on GDP growth in 2030 would be almost completely offset, reducing it to just 0.1-0.2 percentage points.
  - Offset depends on Keynesian fiscal multiplier of recycling policies.

### Scenario outcomes and key statistics (Table 4 summary; exact figures)
- Covered Sectors:
  - Carbon Tax 50: All
  - ETS: All
  - Feebates: Power and industry
  - Subsidy Reform: All
  - Combination: All
- Energy-Related CO2 Emissions reduction in 2030, % to a BAU:
  - Carbon Tax 50: 44.2
  - ETS: 40.8
  - Feebates: 10.2
  - Subsidy Reform: 24.3
  - Combination: 50.4
- Cumulative CO2 Emissions Reductions in 2024-2030, MtCO2:
  - Carbon Tax 50: 26.2
  - ETS: 23.8
  - Feebates: 6.4
  - Subsidy Reform: 12.7
  - Combination: 30.2
- Additional Fiscal Revenues Raised in 2030 or Budget Savings, % of GDP:
  - Carbon Tax 50: 4.7
  - ETS: 4.1
  - Feebates: 1.6
  - Subsidy Reform: 5.2
  - Combination: 7.2
- Cumulative Additional Fiscal Revenues Raised in 2024-2030, bn USD:
  - Carbon Tax 50: 3.6
  - ETS: 3.1
  - Feebates: 1.1
  - Subsidy Reform: 3.4
  - Combination: 5.5
- Impact on GDP Growth in 2030, percentage points deviation from the BAU growth:
  - Carbon Tax 50: -0.09
  - ETS: -0.09
  - Feebates: -0.05
  - Subsidy Reform: -0.15
  - Combination: -0.11
- Electricity Price Increase in 2030, percent from the BAU price:
  - Carbon Tax 50: 79
  - ETS: 71
  - Feebates: 3
  - Subsidy Reform: 179
  - Combination: 288
- Electricity Price Increase in 2030, percent from the current price:
  - Carbon Tax 50: -22
  - ETS: -26
  - Feebates: -55
  - Subsidy Reform: 22
  - Combination: 69
- Pure Abatement Costs, % of GDP:
  - Carbon Tax 50: -0.85
  - ETS: -0.57
  - Feebates: -0.09
  - Subsidy Reform: -0.10
  - Combination: -1.24
- Domestic Co-Benefits (transport, air pollution, climate), % of GDP:
  - Carbon Tax 50: 3.4
  - ETS: 3.1
  - Feebates: 0.8
  - Subsidy Reform: 1.9
  - Combination: 4.0
- Note: Table entries include the base erosion effect. Price projections start from 2022 onwards.

### Mitigation costs, co-benefits, and distributional impacts
- Aggregate net economic cost:
  - Mitigation policies impose relatively small economic cost equivalent to about 0.1-1.2 percent of GDP in 2030.
  - Economic costs reflect pure mitigation costs (annualized costs of cleaner but more expensive technologies net of lifetime energy savings).
- Domestic co-benefits:
  - 85-100 percent of domestic environmental co-benefits reflect fewer local air pollution deaths.
  - 9-15 percent reductions in traffic congestion and accident externalities.
- Global climate benefits:
  - Adding global benefits valued at $75 per tonne increases environmental benefits from 0.1-0.5 to 0.8-4.0 percent of GDP.
- Energy price impacts (weighted changes relative to baseline in 2030):
  - Coal prices increase by 41, 37, 41 percent in Carbon tax 50, ETS, Combination scenarios respectively.
  - Coal prices unaffected by Subsidy reform and Feebates.
  - Electricity prices increase about by 2-10 cents per kilowatt hour (or 71-288 percent compared to baseline in 2030) in all scenarios.
  - Gasoline prices grow by 36-94 percent relative to baseline (excluding Feebates).
  - Diesel price increases 54-189 percent relative to baseline.
- Production costs and competitiveness:
  - Production costs increase most for pharmaceuticals, chemicals, and transport in 2030.
  - Competitiveness measures include partial exemptions, output-based rebates, and free allowance allocations (trade-off: reduced government revenue).

### Additional policies and design elements to reach Net Zero by 2050
- Pricing beyond energy sector: extend pricing to non-carbon GHGs (e.g., methane fee for fugitive emissions); agricultural sector most affected.
- Promote low-carbon technologies and deployment:
  - Accelerate capital turnover: targeted subsidies for retrofitting buildings, accelerated depreciation, low-carbon long-life capital goods.
  - Accelerate learning-by-doing: production subsidies (declining over time), Carbon Contracts for Differences (CCfds).
  - Address network externalities: public investment in smart grids, charging stations, public transport.
  - Lower financing costs: feed-in tariffs; power purchase agreements.
  - Boost research: IP protection, R&D subsidies and tax credits, accelerated depreciation for R&D, grants.
- Design elements to enhance acceptability and effectiveness:
  - Balance carbon pricing and other instruments at sectoral level.
  - Recycle revenues to boost the economy (lower taxes or fund productive investments) and ensure equitable household distribution.
  - Recycle revenues to fund clean infrastructure and adaptation investments.
  - Market reforms to enhance competition and investment in energy sectors.
  - Just transition measures: social safety nets, assistance for displaced workers and regions.
  - Measures to limit impacts on industrial competitiveness.
  - Financial sector support for low-carbon transition.
  - Extensive consultations, information campaigns, and phased reforms.

### Use of carbon pricing revenues (Table A1 summary)
- Revenue-use options evaluated on four metrics: Economic Efficiency; Income Distribution; Administrative Burden; Political Feasibility.
- Key assessments (verbatim phrasing preserved where given):
  - Public investment:
    - Economic Efficiency: "Potentially significant (high fiscal multipliers, especially for low-carbon investments)"
    - Income Distribution: "Can disproportionately benefit low-income households (for example, if provides basic education, health, infrastructure), but depends on implementation"
    - Administrative Burden: "Modest; requires strong public investment management"
    - Political Feasibility: "Can be popular, with green investment especially favored in climate-concerned countries"
  - Tax reductions:
    - Economic Efficiency: "Can improve incentives for work effort and investment and reduce incentives for the black economy and tax evasion"
    - Income Distribution: "Can be designed to be progressive (for example, via increases in personal income tax thresholds)"
    - Administrative Burden: "Minimal"
    - Political Feasibility: "Popular with beneficiaries (for example, households for personal cuts, firms for corporate income tax cuts)"
  - Deficit reduction:
    - Economic Efficiency: "Lowers future tax burdens and macro-financial risk"
    - Income Distribution: "Depends on country circumstances"
    - Administrative Burden: "Minimal"
    - Political Feasibility: "Does not garner political support"
  - Universal lump-sum transfers:
    - Economic Efficiency: "Forgoes efficiency benefits (for example, no enhanced incentive for work effort)"
    - Income Distribution: "Progressive (disproportionately benefits the poor)"
    - Administrative Burden: "New capacity may be needed (but should be manageable)"
    - Political Feasibility: "Mixed, with some households/firms favouring or disliking lump-sum transfers"
  - Means-tested cash transfers or social assistance:
    - Economic Efficiency: "Forgoes efficiency benefits, but typically requires only a small share of revenues"
    - Income Distribution: "Effective at helping low-income groups if transfers are well targeted or if social safety nets are comprehensive"
    - Administrative Burden: "Low if builds on existing capacity, otherwise significant"
    - Political Feasibility: "Generally popular"
  - Direct assistance for household energy bills:
    - Economic Efficiency: "Forgoes efficiency benefits; reduction in environmental effectiveness depending on design"
    - Income Distribution: "Provides partial relief for households (but does not help with indirect pricing burden)"
    - Administrative Burden: "Low if builds on existing capacity, otherwise significant"
    - Political Feasibility: "Generally popular"
- Technical point: "Dividends have no efficiency benefits as they do not increase the real return to work effort or investment."

### CPAT overview, data, caveats
- CPAT scope:
  - Provides projections for 200 countries of fuel use and CO2 emissions by major energy sector (power, industrial, transport, residential; excludes international aviation and maritime fuels).
- Baseline inputs:
  - GDP projections from latest IMF forecasts; assumptions on income elasticity and own-price elasticity; technological change rates; future international energy prices; current fuel taxes/subsidies and carbon pricing held constant in real terms.
- Mechanics and data:
  - Fuel demand curves: constant elasticity specification.
  - Parameterization: IEA data, average of IEA and IMF projections for international energy prices, fuel price elasticities typically between -0.5 and -0.8, carbon emissions factors from IEA.
  - Domestic environmental costs based on IMF methodologies.
- Caveats:
  - Abstracts from mitigation actions beyond those implicit in recent data.
  - Fuel price responses may not hold for dramatic price changes leading to technological advances (e.g., CCS, direct air capture).
  - Does not explicitly account for general equilibrium effects or international price changes due to simultaneous reforms in large countries.

### Inflation decomposition (SVAR) for Brunei: findings and methodology
- Recent inflation dynamics:
  - Average headline inflation: -0.4 percent before pandemic, 1.7 percent in 2021, 3.7 percent in 2022.
  - Food price increased 5.1 percent y/y in 2022.
  - Transport price increase peaked in 2021 at 5.9 percent EoY.
  - Foreign workers contributed half of private sector employment pre-pandemic; saw a 20 percent drop after pandemic start.
- SVAR identification:
  - Demand shocks: movements of GDP and CPI in same direction.
  - Supply shocks: movements of GDP and CPI in opposite directions.
  - Sample period: 2011Q1 to 2022Q3; seasonally adjusted CPI and real GDP; two specifications: total real GDP and real non-O&G GDP.
- Key decomposition findings using non-O&G output:
  - Supply factors explained about 38 percent of CPI inflation in 2022.
  - Demand factors explained about 31 percent of CPI inflation in 2022.
  - In 2021: supply factors explained about 45 percent, demand about 7 percent.
- Real output decomposition:
  - Supply disruptions dragged on output growth; demand factors contributed positively after reopening.
  - O&G output held back by infrastructure maintenance despite surge in O&G prices in 2022.
- Caveats:
  - Model assumes domestic factors main drivers; may underweight external import-price effects.
  - Price controls on basic goods may mute headline inflation responses.

### Fiscal sustainability, PIH framework, and scenario assumptions
- Fiscal context:
  - O&G revenue accounts for 84 percent of total revenue on average FY2005/06–FY2022/23.
  - Wages/salaries, pensions, royalty payments, recurrent charges = 82 percent of total expenditure average.
  - Brunei mostly ran fiscal deficit since FY2014/15; average deficit -8.3 percent of GDP since FY2014/15.
  - Non-O&G primary deficit averaged -53.9 percent of non-O&G GDP; declined to -39.2 percent by FY2022/23 from -60.0 percent in FY2017/18.
- PIH framework assumptions (Table 1 highlights):
  - Interest rate: 2.6%.
  - Population growth rate: 0.42%.
  - Long-term inflation: 1.0%.
  - Government’s share in O&G revenue: 37.0%.
  - Baseline and downside O&G output and price paths described with exact conditions (see source content).
- Fiscal anchor implications:
  - Most conservative anchor (PIH with deficit as constant share of non-O&G GDP) implies NRFB around 6 percent of non-O&G GDP and nearly 35 percent of non-O&G GDP adjustment in 2023.
  - PIH with real annuity (NRFB constant in real terms) has smallest gap to baseline and is likely most feasible; allows higher deficit in transition but requires later reduction.
  - Current baseline differs significantly from PIH anchors, requiring at least a 15 percent adjustment in 2023 under PIH constant annuity in real terms.

### Gradual consolidation scenario and robustness tests
- Transition scenario (PIH constant annuity in real terms):
  - NRFB at -16 percent of non-O&G GDP in 2028 vs -26 percent in baseline — additional 10 percentage point consolidation in medium-term.
  - Continued gradual adjustment ~0.5 of non-resource GDP annually under gradual transition vs ~0.3 ppt in immediate transition.
- Sensitivities:
  - Framework sensitive to r − g dynamics, public spending efficiency, and data limitations (GLCs, sovereign wealth funds).
- Stress tests (resource revenue shocks):
  - Permanent shock: oil price halved and 10 percent exchange rate reduction.
  - Temporary shock: oil prices halved in 2023 and 2026 combined with 10 percent depreciation in 2023.
  - Adjustment strategies: no adjustment; full adjustment (spending reduced one-to-one); partial adjustment (spending reduced up to 2 percent of total expenditures annually).
  - Assumed financing shortfalls met by drawing liquid assets then borrowing.
  - Results indicate real annuity anchor robust under partial and no adjustment scenarios for tested shocks.
- Net Zero (2050) downside:
  - If world meets Net Zero by 2050, Brunei would need additional 4 percent of GDP adjustment by 2028 under transition scenario.

### Policy recommendations for consolidation and diversification
- Accelerate fiscal consolidation for sustainability and inter-generational equity.
- Adopt broad-based revenue diversification, including carbon pricing and wage/subsidy reforms.
- Prioritize efficient, growth-friendly investments in health, education, and green resilient infrastructure to support diversification.
- Strengthen fiscal frameworks and institutions.
- Consolidation measures: contain public sector wage/employment growth; reduce tax expenditures and untargeted subsidies.
- Sequencing and institutions:
  - Strong MTFF and PFM systems, transparency, realistic forecasts, regular publication of plans, and public debate are success factors.
- Illustrative tax and subsidy reform examples referenced from GCC countries (VAT, excise, corporate tax adoptions) and subsidy reform timelines.

### Financial sector soundness: key indicators and risks
- Financial system size and composition:
  - Total financial system assets ~108 percent of GDP end-2022; banking sector = 82 percent.
  - Seven commercial banks (two domestic, five foreign branches) and one Islamic trust fund.
  - Two largest domestic banks account for 46 percent and 15 percent of total financial assets respectively (end-2022).
- Capital and liquidity (figures):
  - Regulatory capital to risk-weighted assets: 20.2 (2016–2022 series shown).
  - Tier 1 capital to risk-weighted assets: 20.0 (2022).
  - Liquid asset ratio: 43.8 (2022).
- Asset quality and profitability:
  - NPL to total loans: 3.3 (2022).
  - NPL net of provisions to capital increased from 4.9 percent to 5.5 percent (2021 → 2022).
  - Provision coverage (specific provisions to total NPLs): 37.8 (2022).
  - Return on assets (before tax): 1.3 (2022).
  - Return on equity (after tax): 9.5 (2022).
  - Efficiency ratio (non-interest expense to gross income): 56.7 (2022).
- Market, liquidity, and concentration risks:
  - Banks’ investment portfolio ~20 percent of total assets; exposure to market risks.
  - Interest rate stress test could reduce total capital to risk-weighted assets by 3.2 percentage points, to 16.4 under most severe case.
  - Large share of deposits invested offshore; liquidity risk if offshore assets ringfenced.
  - Loan-to-deposit ratio: 36 percent (2022).
  - Domestic household loans represented 50.4 percent of domestic lending in 2022; household debt to banking sector 12.4 percent of GDP.
  - Domestic corporate loans growth driven by downstream O&G manufacturing and construction.
- Macro-linkages:
  - VAR result: a 1 percentage point decline in oil price growth increases NPL growth by 0.455 percentage point after three quarters.
- Macroprudential and supervisory actions:
  - Authorities implementing Basel II pillars; designated two D-SIBs with additional capital buffer from 2023.
  - BDCB strengthening risk-based supervision and moving toward Basel III frameworks.

### Digital, urban, health, and finance initiatives (selected items relevant for modernization)
- Health initiatives: Assistive Technology and Robotics in Healthcare; HealthHub; National Steps Challenge™ & Healthy 365 App; TeleHealth.
- Transport initiatives: Autonomous Vehicles; CETRAN; Contactless Fare Payment; On-Demand Shuttle; Open Data and Analytics for Urban Transportation.
- Urban solutions: Punggol Smart Town; Smart Nation Sensor Platform (SNSP); Smart Water Meter; OneService App/Chatbot; myENV App; Elderly Monitoring System; Dengue Hotspots Survey Drones.
- Finance and digital government:
  - GoBusiness; Corppass; Data Innovation Programme Office; FinTech Sandbox; Networked Trade Platform; SGFinDex; SGTraDex.
  - Digital government services: LifeSG; National Digital Identity; CentEx; CrowdTaskSG; Digital Birth and Death Certificates; HDB Resale Portal.
- Singapore relevance and private-sector focus:
  - Programs: Advanced Digital Solutions; Grow Digital.
  - Workforce training examples: SGUnited Mid-Career Pathways-Company Training (SGUP-CT); SGUnited Skills (SGUS); SkillsFuture Career Transition Program (SCTP); SGUnited Mid-Career Pathways-Company Attachment (SGUP-CA).
  - MAS active areas: digital banks; CBDCs; cross-border payments; crypto assets.
- Robotics adoption statistic:
  - Singapore robot density: increased from "about 1 operating robot per 1,000 employees in 2008 to 45 operating robots per 1,000 employees in 2018."

*Source: IMF staff analysis in 1brnea2023002.*

### 1. Behavioral Responses Promoted by Alternative CO

### Box 1. Behavioral Responses Promoted by Alternative CO2 Mitigation Policies

### Comprehensive carbon pricing: promoted responses
- Power generation: shifting (both in terms of new investment and the daily dispatch mix) from natural gas, from these fuels to renewables, and perhaps to hydrogen and fossil generation with carbon capture and storage (in line with Pillar 4: renewable energy, Pillar 5: power management);
- Industry: reducing CO2 and electricity intensity (e.g., through mandating carbon reporting for industries, implementing a usage-based electricity tariff scheme, energy-efficient appliances, LED streetlights and electric vehicles), and output levels (in line with Pillar 1: industrial emissions);
- Transportation: shifting to more efficient internal combustion engine (ICE) vehicles, from ICE vehicles to electric (or other zero emission) vehicles, and reducing vehicle miles travelled. (however, no policies are in place to support sustainable transport plans, according to the International Renewable Energy Agency (IRENA)) (in line with Pillar 3: electric vehicles); and
- Buildings: reducing CO2 intensity, electricity intensity, and energy demand (e.g., through energy efficient construction, improving the energy efficiency of appliances).

### Non-pricing mitigation instruments: scope and limitations
- Renewable portfolio standards and feed-in tariffs for renewables only promote shifting from fossil to renewable generation;
- Emission rate regulations, or feebates, for new vehicles reduce emissions from the on-road fleet gradually over time as the fleet turns over (e.g., they do not accelerate retirement of old vehicles) and they do not reduce vehicle miles travelled; and
- Incentives for net zero new buildings reduce emissions from the building stock very gradually (given that typically less than 2 percent of the building stock is replaced each year).

### Practical policy considerations
- In practice, non-pricing mitigation instruments will be used to complement and reinforce carbon pricing.
- Although less efficient, non-pricing instruments may have greater acceptability as they avoid significant and politically sensitive increases in energy prices—unlike carbon pricing, they do not involve the pass through of carbon tax revenues or allowance rents into energy prices.
- Non-pricing instruments like feebates may have a key role in kick-starting de-carbonization of hard-to-abate sectors, particularly transportation and buildings.
- Policymakers need to strike a balance between carbon pricing (the most efficient but perhaps most politically challenging instrument) and other (less efficient but frequently more acceptable) reinforcing instruments.

*Prepared by: Ian Parry, Simon Black, Karlygash Zhunussova.*

### 11.      There is increasing momentum for carbon pricing globally and in the Asian region,

### 11.      There is increasing momentum for carbon pricing globally and in the Asian region

### Momentum, coverage, and price signals
- Carbon pricing schemes operate in 45 countries, accounting for national and sub-national initiatives and, for EU countries, the EU ETS.
- At the national level, 21 carbon taxes and 6 ETSs have been implemented.
- Major recent initiatives launched in China and Germany; California’s ETS is the largest sub-national scheme.
- Prices and commitments:
  - EU ETS prices are currently around $100 per tonne.
  - Canada has committed to an equivalent US$140 price by 2030.
  - Sweden: $130 per tonne (economywide average price in 2022).
  - Economywide average prices in 2022 varied from below $5 to $130 per tonne.
  - The average price across schemes is $20 per tonne.
- Coverage:
  - GHG emissions subject to (national and sub-national) carbon pricing vary from below 30 percent in some cases to over 70 percent in others (e.g., Canada, Germany, Korea, Sweden).
  - 28 percent of global GHGs are formally subject to pricing.
- Regional presence in Asia:
  - Countries with carbon pricing include Indonesia, Japan, Korea, and Singapore.
  - Carbon pricing is under consideration in Thailand and Vietnam.
- Brunei Darussalam:
  - Under the Income Tax (Petroleum) Act of 1963, oil and gas companies operating in Brunei Darussalam are subject to a fixed 55 percent petroleum income tax rate.
  - Brunei Darussalam does not have a sales or excise tax on fuel.
  - Under the carbon pricing strategy, Brunei Darussalam aims to launch an ETS.
  - Starting 2025 Brunei Darussalam plans to reduce carbon intensity from all industrial sectors and new power utilities.

### Instrument choice and administrative considerations
- Carbon taxes:
  - Generally under the purview of finance ministries.
  - Easier to administer than ETSs; can be integrated midstream into existing fuel tax collection.
  - All but one of the 21 existing national carbon taxes are applied midstream.
  - Carbon taxes provide price certainty and require periodic adjustment to maintain progress on emissions goals.
- Emissions Trading Systems (ETSs):
  - Generally under the purview of environment ministries.
  - Require more sophisticated administration: monitoring downstream emissions, setting up emissions trading markets, MRV systems, and allowance exchange platforms.
  - Often implemented with pilot phases to establish MRV and trading infrastructure.
  - ETSs have often been applied to large power and industrial firms, though ETSs can be applied midstream (examples: German and Korean ETSs; proposed German and Korean expansions to transportation and building fuel suppliers).
  - ETSs provide certainty over emissions (if cap is set appropriately) but can exhibit allowance price volatility (example allowance price volatility observed in California, EU, Korea).
  - Price stability mechanisms (e.g., price floors) can be combined with ETSs to reduce price volatility.

### Price vs emissions certainty, and hybrid designs
- Tradeoffs:
  - Carbon taxes: certainty over price, emissions determined by market response.
  - ETSs: certainty over emissions (if cap set), prices determined by allowance markets and can be volatile.
  - Price volatility in ETSs can deter private innovation and adoption of clean technologies with high upfront costs and long-term payoffs.
- Convergence:
  - ETSs with price floors and/or ceilings can mimic carbon taxes by providing greater price certainty.
  - Carbon taxes and ETSs exist on a continuum and can be designed to approximate each other.
- Hybrid approaches:
  - Example: ETS for power and industry; carbon tax for transportation and buildings.
  - Cost effectiveness requires aligning carbon prices across tax rates and ETSs (e.g., setting ETS price floors equal to carbon tax trajectories).
  - Applying both a carbon tax and an ETS to the same emissions base is generally inefficient and duplicative.

### Revenue use, efficiency, and distributional implications
- Observed practices:
  - Carbon tax revenues are more likely to accrue to the general budget; revenues have been fully used for general purposes in 16 carbon tax schemes and partially or fully earmarked for environmental spending in only five cases.
  - ETS revenues are more likely to be earmarked for environmental purposes; in five of seven ETS schemes, allowances auction revenues are at least partially earmarked.
  - In early phases of some ETSs (e.g., EU, Korea), allowances were freely allocated to affected firms to build support and address competitiveness concerns.
  - In other ETSs (e.g., California, Germany), allowances have been auctioned from the start.
- Efficiency implications:
  - Productive uses of revenues (e.g., public investments for Sustainable Development Goals such as health, education, infrastructure) can produce large gains in economy efficiency and help offset negative effects of higher energy prices.
  - Earmarking for environmental investment can be efficient if integrated into robust public investment management systems.
  - Returning revenues in universal or targeted lump-sum transfers to households or firms forgoes some efficiency benefits.
- Distributional impacts:
  - In principle, a carbon tax and an ETS—if applied to the same sectors, with the same price, and prior to allocation of revenues—would impose the same distributional burdens across household income groups.
  - Distributional burdens measured against households’ annual consumption are mildly regressive in some cases, though the opposite can also apply.
  - Under ETSs with free allowance allocation, policy rents accrue to firms and ultimately to shareholders and workers (shareholders concentrated among higher income households), limiting opportunities to use revenues to achieve progressive outcomes.
  - Germany uses auction revenues for transition assistance to vulnerable households, workers, and regions—improving acceptability but forgoing some efficiency benefits.

### Political economy and implementation strategies
- Political feasibility:
  - ETSs may be more politically feasible than taxes in some contexts, particularly where free permit allocation builds industry support.
  - Carbon taxes can be politically challenging; revenue recycling, communications strategies, and identification of key stakeholders can build support.
  - The anticipation of negative distributional outcomes can generate public opposition, making targeted support measures and careful quantification critical.
- Design levers to enhance acceptability:
  - Free allocation in ETSs can be phased down over time (example: EU ETS free allocation reduced from 80 percent in 2013 to 30 percent in 2020).
  - Carbon taxes can be designed to mimic free allocation effects by using revenues for targeted relief to firms.
  - Effective and inclusive communication alongside pragmatic use of revenues is important for durability.

### Compatibility with other instruments and broader GHG coverage
- Reinforcing instruments:
  - Carbon taxes are compatible with overlapping mitigation instruments (e.g., feebates) that reduce emissions without altering the tax rate.
  - Under a pure ETS, overlapping instruments reduce the emissions price without affecting emissions unless the cap is adjusted.
  - Carbon tax variants can be extended to broader emissions sources like forestry.
- Non-CO2 GHGs:
  - Carbon pricing can be extended to non-carbon GHG emissions (e.g., methane) to equalize abatement costs across GHGs and promote cost-effective mitigation.
  - The agricultural sector would be most affected by extending pricing to non-CO2 GHGs.
  - Proxy taxes could be used in the medium term to address administrative and compliance barriers for non-CO2 GHG pricing.

### International coordination and competitiveness
- International price coordination:
  - Coordination among countries (for example, among Southeast Asian countries) can scale up carbon pricing and address competitiveness and policy uncertainty concerns that deter unilateral action.
  - International coordination would be most naturally met through carbon taxes, but ETSs can be accommodated by underpinning them with floor prices or by setting caps that generate expected domestic emissions prices in line with international requirements (example: prototype federal pricing requirements in Canada).
- Sectoral focus for initial pricing:
  - Establishing pricing initially for power and industry sectors may be sensible given the bulk of emissions reductions come from these sectors.

*Source: IMF staff chapter on carbon pricing (excerpts).

### 27.      A tradeable performance standard (TPS) for the power and industry sectors would not

### 1brnea2023002 - 27.      A tradeable performance standard (TPS) for the power and industry sectors would not

### Tradeable Performance Standards (TPS) and Feebates: design and behavioral effects
- TPS design:
  - Government sets a required CO2 emission rate per unit of output for each major industry and power generation.
  - All firms within the industry are required to meet the industry standard; firms that fall short can buy credits from firms that exceed the standards.
  - Credits could be tradable across firms in different industries, promoting a common credit price and equalization of incremental abatement costs across industries.
  - TPSs promote the same behavioral responses to reduce emissions intensity as carbon pricing but do not promote consumer demand responses because there is no pass through of carbon tax revenue or allowance rents into higher prices for electricity and industrial products.
  - Example: Canada has implemented a federal TPS for the industrial sector where provincial/territorial policies are not applied.
- Feebates as the fiscal analogue of TPS:
  - Feebates apply a sliding scale of fees on firms with emission rates above a pivot point and a sliding scale of rebates for firms below the pivot point.
  - If the pivot point is set equal to the industry average emission rate and updated over time, the feebate will be approximately revenue neutral.
  - Mechanically: firms face a fee equal to a CO2 price times the difference between their CO2 per unit of output and the industry-wide pivot point CO2 per unit of output, times the firm’s production level.
  - Feebates automatically promote cost effectiveness within an industry without trading markets because all firms face the same incremental reward for reducing emissions (the emissions price in the feebate).
  - Emission prices can be harmonized across feebate schemes to promote cross-industry cost-effectiveness.
  - Like TPSs, feebates do not charge the average firm for their remaining emissions and therefore do not promote consumer demand responses.
- Comparison pros/cons (feebates versus TPSs / downstream taxes versus ETSs):
  - Feebates provide certainty over the emissions price, while regulations provide certainty over the industry-wide average emissions rate.
  - Feebates are automatically cost effective across firms within industries and across industries (if feebate prices across schemes are harmonized); tradable emission rate standards require liquid credit trading markets with a significant number of market traders to be cost effective.
  - Feebates are compatible with overlapping policies as they provide ongoing incentives for all firms (regardless of whether they are paying fees or receiving rebates) to cut emissions.
  - TPSs can be made more feebate-like by combining them with out-of-compliance fees, subsidies for going beyond standards, and for firms not participating in credit trading.

### Quantitative assessment methodology (CPAT) and timing considerations
- Model and calibration:
  - The quantitative assessment uses the Climate Policy Assessment Tool (CPAT), a spreadsheet-based model projecting fuel use and GHG emissions for major energy sectors in 188 countries.
  - Impacts depend on (i) proportionate impacts on future fuel prices and (ii) price responsiveness of fuel use in different sectors.
  - Price responsiveness parameterized to the mid-range of existing modelling literature and empirical evidence; fuel and electricity price elasticities over the longer term are generally between -0.5 and -0.8.
  - Carbon emissions factors by fuel product are from IIASA (Wagner 2020); emissions in 2019 are calibrated to UNFCCC GHG and 2020-1 calibrated to EC-JRC (Crippa and others 2022), Global Carbon Budget (Friedlingstein and others 2021), and various sources.
  - CPAT is linked to input-output tables to infer impacts on production costs in different industries, consumer prices, and burdens on household income groups.
- Timing and interaction with global energy prices:
  - Carbon pricing might be phased in as global energy prices recede from peaks experienced mid-2020 to mid-2022.
  - Phasing in a $50 carbon price on top of projected prices would imply 2030 O&G prices would be 16 and 59 percent below mid-2022 levels, while coal prices would be 11 percent higher.

### Emissions baseline, targets, and scenarios
- Baseline and targets:
  - Authorities submitted an unconditional target of cutting GHGs to 20 percent below baseline levels in 2030 in the NDC (reference year 2015, emissions of 11.6 MtCO2e).
  - Energy-related emissions accounted for 97.2 percent of Brunei Darussalam’s 13.6 MtCO2 GHG emissions in 2022.
  - Sector shares of 2022 emissions: power generation and own-use by the energy industries 54.6 percent; power generation – public utilities 21.5 percent; fugitive emissions 8.6 percent; transport 8.9 percent; non-specified industries 3.0 percent; industrial processes 1.0 percent; waste 1.5 percent; agriculture 0.3 percent; buildings 0.6 percent.
  - Brunei Darussalam announced a move towards Net Zero by 2050; linear extrapolation implies economy-wide emissions around 8.8 MtCO2e or 60 percent reduction to BAU scenario.
- Baseline projections and drivers:
  - IMF staff project GHG emissions will increase 73 percent from 12.7 MtCO2e in 2021 to 21.9 MtCO2e in 2030 (i.e., 26 percent lower than the authorities’ projection of 29.5 MtCO2e).
  - Emissions increase driven by projected increase in GDP (21 percent) and increase in energy intensity of GDP (68 percent).
- Modeled mitigation scenarios (definitions):
  - Carbon tax 50: explicit carbon tax from $20 per tonne in 2024 rising linearly to $50 per tonne in 2030, covering power, industry, transport, and buildings.
  - ETS: emissions trading system from $20 per tonne in 2024 rising linearly to $50 per tonne in 2030, covering power and industry.
  - Feebates: feebates in power and industry, from $20 per tonne in 2024 rising linearly to $50 per tonne in 2030.
  - Subsidy reform: phase out consumer-side fossil fuel subsidies starting from 2024 linearly over five years.
  - Combination: carbon tax (same as Carbon tax 50) combined with fossil fuel subsidy reform (starting 2024 linearly over five years).

### Emissions impacts by scenario and by sector
- Aggregate emissions reductions in 2030 relative to baseline:
  - Carbon tax 50: 44 percent reduction.
  - ETS: 41 percent reduction.
  - Feebates: 10 percent reduction.
  - Subsidy Reform: 24 percent reduction.
  - Combination: 50 percent reduction.
- Sectoral contributions to emissions reductions in 2030:
  - Power sector reductions vary from 24-85 percent across scenarios.
  - Industry reductions vary from 7-76 percent across scenarios.
  - Transport accounts for less than 10 percent of emissions reductions in 2030.
  - Buildings sector accounts for 1 percent of the reduction in 2030.
- Interpretation:
  - Power and industry account for the most CO2 emissions reductions in all scenarios due to larger shares in emissions and higher responsiveness to carbon pricing.

### Electricity generation and renewables
- Current and historical shares:
  - Natural gas share in electricity generation: 78 percent in 2020 (down from nearly 99% in 1990).
  - Coal was 28 percent of power in 2020.
  - Solar generated 2 GWh in 2020, about 0.03 percent of electricity.
  - Authorities target at least 30 percent renewable share of energy power generation mix by 2035 (current less than one percent).
- Model results for 2030 electricity mix:
  - Carbon tax 50 and ETS scenarios raise the share of renewables to about 14 percent of electricity generation in 2030.
  - Other scenarios reach about 7-10 percent renewables in 2030.
  - These modeled results fall far short of the government’s plan to increase renewables to 30 percent of capacity by 2035.

### Fiscal and macroeconomic implications
- Revenues and budget savings:
  - Carbon pricing could raise up to 7 percent of GDP in 2030 on top of BAU.
  - In 2030:
    - Feebates scenario yields additional 1.6 percent of GDP in savings in the budget.
    - Combination scenario yields 7.2 percent of additional revenues.
  - Cumulative revenues collected/budget savings 2024-2030:
    - Feebates: $1.1 billion.
    - Combination: $5.5 billion.
  - Note: Feebates are revenue-neutral by design, but erosion of subsidized base yields budget savings in the model.
- Growth impacts and revenue recycling:
  - Potential negative impacts on GDP growth in 2030 without revenue recycling (percentage point reductions):
    - Carbon Tax 50: 2.9 percentage points.
    - ETS: 2.5 percentage points.
    - Feebates: 1.0 percentage point.
    - Subsidy Reform: 3.2 percentage points.
    - Combination: 4.3 percentage points.
  - Illustrative revenue recycling scenario: 75 percent of carbon tax revenues recycled to productive public investment and 25 percent to targeted cash transfers.
    - Under this recycling, the negative impact on GDP growth in 2030 would be almost completely offset, reducing it to just 0.1-0.2 percentage points.
  - The extent of offset depends on the value of the Keynesian fiscal multiplier of the chosen recycling policies.

*Source: IMF staff analysis in 1brnea2023002.*

### 43.      Mitigation policies would impose a relatively small economic cost in Brunei

### 43.      Mitigation policies would impose a relatively small economic cost in Brunei Darussalam

### Key findings: aggregate costs and co-benefits
- Mitigation policies would impose a relatively small economic cost equivalent to about 0.1-1.2 percent of GDP in 2030.
- Economic costs reflect pure mitigation costs, primarily the annualized costs of using cleaner but more expensive technologies instead of fossil-based technologies (net of any savings in lifetime energy costs).
- 85-100 percent of the domestic environmental co-benefits reflect fewer local air pollution deaths and 9-15 percent reductions in traffic congestion and accident externalities.
- Adding the global climate benefits valued at $75 per tonne increases environmental benefits from 0.1-0.5 to 0.8-4.0 percent of GDP.

### Impact on energy prices and households (Prices Analysis)
- Direct carbon pricing (carbon taxes and ETS) is likely to put upward pressure on energy prices; Subsidy reform and Feebates have lower impacts.
- Weighted average coal prices would increase by 41, 37, 41 percent compared to baseline in the Carbon tax 50, ETS, and Combination scenarios, respectively.
- Coal prices would not be affected by Subsidy reform and Feebates.
- Electricity prices would increase about by 2-10 cents per kilowatt hour (or 71-288 percent compared to baseline in 2030) in all scenarios.
- Gasoline prices would grow by 36-94 percent relative to baseline (excluding the Feebates scenario).
- Diesel price increases would be 54-189 percent relative to the baseline.

### Impact on production costs and competitiveness
- Mitigation policies would increase the cost of industrial production, with larger increases for energy-intensive, trade-exposed (EITE) industries.
- Production cost increases arise from three components:
  - direct tax payments or allowance purchases for firms’ direct emissions;
  - abatement costs when firms cut emissions (e.g., switching to cleaner but costlier technologies and fuels);
  - indirect payments for carbon charges on emissions embodied in inputs, especially electricity.
- Carbon pricing in 2030 would increase production costs the most for pharmaceuticals, chemicals, and transport.
- Measures to address competitiveness concerns (noting trade-offs) include:
  - imposing carbon charges only on firms’ emissions above a threshold level (partial exemption);
  - returning revenues collected from EITE industries as output-based rebates (operationally similar to TPS or feebates);
  - providing free allowance allocations to EITE industries under an ETS.
- A drawback of these measures is reduced potential government revenue from carbon pricing.

### Scenario outcomes and key statistics (Table 4 summary)
- Covered Sectors:
  - Carbon Tax 50: All
  - ETS: All
  - Feebates: Power and industry
  - Subsidy Reform: All
  - Combination: All
- Energy-Related CO2 Emissions reduction in 2030, % to a BAU:
  - Carbon Tax 50: 44.2
  - ETS: 40.8
  - Feebates: 10.2
  - Subsidy Reform: 24.3
  - Combination: 50.4
- Cumulative CO2 Emissions Reductions in 2024-2030, MtCO2:
  - Carbon Tax 50: 26.2
  - ETS: 23.8
  - Feebates: 6.4
  - Subsidy Reform: 12.7
  - Combination: 30.2
- Additional Fiscal Revenues Raised in 2030 or Budget Savings, % of GDP:
  - Carbon Tax 50: 4.7
  - ETS: 4.1
  - Feebates: 1.6
  - Subsidy Reform: 5.2
  - Combination: 7.2
- Cumulative Additional Fiscal Revenues Raised in 2024-2030, bn USD:
  - Carbon Tax 50: 3.6
  - ETS: 3.1
  - Feebates: 1.1
  - Subsidy Reform: 3.4
  - Combination: 5.5
- Impact on GDP Growth in 2030, percentage points deviation from the BAU growth:
  - Carbon Tax 50: -0.09
  - ETS: -0.09
  - Feebates: -0.05
  - Subsidy Reform: -0.15
  - Combination: -0.11
- Electricity Price Increase in 2030, percent from the BAU price:
  - Carbon Tax 50: 79
  - ETS: 71
  - Feebates: 3
  - Subsidy Reform: 179
  - Combination: 288
- Electricity Price Increase in 2030, percent from the current price:
  - Carbon Tax 50: -22
  - ETS: -26
  - Feebates: -55
  - Subsidy Reform: 22
  - Combination: 69
- Pure Abatement Costs, % of GDP:
  - Carbon Tax 50: -0.85
  - ETS: -0.57
  - Feebates: -0.09
  - Subsidy Reform: -0.10
  - Combination: -1.24
- Domestic Co-Benefits (transport, air pollution, climate), % of GDP:
  - Carbon Tax 50: 3.4
  - ETS: 3.1
  - Feebates: 0.8
  - Subsidy Reform: 1.9
  - Combination: 4.0

Note: Table entries include the base erosion effect. Price projections start from 2022 onwards.

### Additional policies to help reach Net Zero by 2050 (options to close the emissions gap)
- Pricing or similar schemes for GHG emissions beyond the energy sector (for example, methane fee to help reducing fugitive emissions).
- Promoting development, adoption, and production of low-carbon technologies:
  - To accelerate capital turnover: targeted subsidies for retrofitting buildings, accelerated depreciation, and low-carbon, long-life capital goods.
  - To accelerate learning-by-doing externalities: production subsidies (declining over time or with production levels), Carbon Contracts for Differences (CCfds).
  - To address network externalities: public investment in enabling infrastructure (for example, smart grids, charging stations, public transportation).
  - To lower financing costs: feed-in tariffs; power purchase agreements.
  - To boost research: intellectual property protection, R&D subsidies and tax credits, accelerated depreciation for R&D, grants to universities and research labs.

### Design elements to enhance effectiveness and political acceptability
- Balance carbon pricing and other mitigation instruments—especially feebates or TPSs—at the sectoral level.
- Recycle carbon pricing revenues to boost the economy (e.g., lowering taxes on work effort or funding socially productive investments) and ensure equitable household distribution of benefits.
- Recycle revenues to support public investments in clean technology infrastructure networks and/or adaptation investments that would not be provided privately.
- Implement market reforms to enhance competition and investment in the main energy sectors.
- Implement just transition measures to assist vulnerable groups (stronger social safety nets or tax reliefs for low-income households, assistance for displaced workers and at-risk regions).
- Adopt measures to limit impacts of carbon pricing on industrial competitiveness.
- Provide financial sector support for the low-carbon transition.
- Conduct extensive upfront consultations and information campaigns; phase reforms progressively to allow adjustment time for households and firms.

*Source: IMF staff using CPAT; text and figures from the provided IMF chapter.*

### 3.      Table A1 provides more discussion of alternative options for the use of carbon pricing

### 1brnea2023002 - 3.      Table A1 provides more discussion of alternative options for the use of carbon pricing revenues

### Use of carbon pricing revenues (Table A1 summary)
- Table A1 compares revenue uses across four metrics: Impact on Economic Efficiency; Impacts on Income Distribution; Administrative Burden; Political Feasibility.
- Revenue-use options and key assessments:
  - Public investment
    - Impact on Economic Efficiency: "Potentially significant (high fiscal multipliers, especially for low-carbon investments)"
    - Impacts on Income Distribution: "Can disproportionately benefit low-income households (for example, if provides basic education, health, infrastructure), but depends on implementation"
    - Administrative Burden: "Modest; requires strong public investment management"
    - Political Feasibility: "Can be popular, with green investment especially favored in climate-concerned countries"
  - Tax reductions
    - Impact on Economic Efficiency: "Can improve incentives for work effort and investment and reduce incentives for the black economy and tax evasion"
    - Impacts on Income Distribution: "Can be designed to be progressive (for example, via increases in personal income tax thresholds)"
    - Administrative Burden: "Minimal"
    - Political Feasibility: "Popular with beneficiaries (for example, households for personal cuts, firms for corproate income tax cuts)"
  - Deficit reduction
    - Impact on Economic Efficiency: "Lowers future tax burdens and macro-financial risk"
    - Impacts on Income Distribution: "Depends on country circumstances"
    - Administrative Burden: "Minimal"
    - Political Feasibility: "Does not garner politcal support"
  - Universal lump-sum transfers
    - Impact on Economic Efficiency: "Forgoes efficiency benefits (for example, no enhanced incentive for work effort)"
    - Impacts on Income Distribution: "Progressive (disproportionately benefits the poor)"
    - Administrative Burden: "New capacity may be needed (but should be manageable)"
    - Political Feasibility: "Mixed, with some households/firms favouring or disliking lump-sum transfers"
  - Means-tested cash transfers or social assistance
    - Impact on Economic Efficiency: "Forgoes efficiency benefits, but typically requires only a small share of revenues"
    - Impacts on Income Distribution: "Effective at helping low-income groups if transfers are well targeted or if social safety nets are comprehensive"
    - Administrative Burden: "Low if builds on existing capacity, otherwise significant"
    - Political Feasibility: "Generally popular"
  - Direct assistance for household energy bills
    - Impact on Economic Efficiency: "Forgoes efficiency benefits; reduction in environmental effectiveness depending on design"
    - Impacts on Income Distribution: "Provides partial relief for households (but does not help with indirect pricing burden)"
    - Administrative Burden: "Low if builds on existing capacity, otherwise significant"
    - Political Feasibility: "Generally popular"
- Key technical point: "Dividends have no efficiency benefits as they do not increase the real return to work effort or investment."
- Visual/quantitative comparison (Annex II, Figure A1):
  - Title: "Economic Efficiency Costs of Alternative Mitigation Instruments for the United States ($50/Ton Carbon Tax), 2030"
  - Note: "Policies reduce economywide CO2 emissions 22 percent below BAU."
  - Instruments represented include:
    - "Carbon tax: 75% of revenue for cutting labor taxes"
    - "Carbon tax and dividend"
    - "Feebate/regulatory combination"
  - Figure metric: "Average cost, $/ton CO2 reduced" (chart bars shown for alternatives; source: IMF (2019a)).

### Forestry, agriculture, and waste pricing schemes (Annex III)
- Forestry
  - Policy objective: promote reducing deforestation, afforestation, and enhanced forest management (e.g., planting larger trees, increasing rotation lengths) to increase carbon storage and co-benefits (reduced water loss, floods, soil erosion, river siltation).
  - Feebate program design:
    - Applies to landowners (especially at agricultural/forestry boundary).
    - Fee formula concept: [CO2 rental price] × [carbon storage on their land in a baseline period ─ stored carbon in the current period]
    - Rewards increased carbon storage via reduced fees or increased subsidies.
    - Periods may be averages over multiple years to account for lumpy carbon storage during harvest years.
    - Feebates can be designed to be revenue-neutral in expected terms through scaling of the baseline over time.
    - Administration could use the registry of landowners used for business tax collection.
  - Payment design: feebates could involve rental payments (avoids one-off upfront payments that risk reversal and complex ex-post re-payment procedures).
  - Monitoring: "Forest carbon inventories are estimated through a combination of satellite monitoring, aerial photography, and on-the-ground tree sampling."
- Agriculture
  - Emissions context: "Around fourth fifths of methane emissions in the Philippines are from the agricultural sector and these emissions account for about 70 percent of total GHGs from agriculture."
  - Source breakdown: "Two thirds of agricultural methane emissions are from rice cultivation and one-third from livestock operations."
  - Other GHGs: nitrous oxide primarily from soils.
  - Mitigation measures:
    - Reduce water intensity in rice paddies (e.g., periodic draining) to cut methane from flooded fields.
    - Increase livestock productivity (e.g., breed switching) and shift to alternative feed (e.g., seaweed additive) to reduce enteric fermentation and manure emissions.
  - Pricing approach: could be based on farm-level output or input data, default emissions factors, and rebates for demonstrated mitigation actions (e.g., drainage of rice paddies). Revenues might be recycled to the sector to address competitiveness concerns.
- Waste
  - Limited behavioral responses: landfill methane collection and flaring; reduction in packaging and food waste; enhanced recycling and composting.
  - Policy view: "The case for pricing methane from waste is less compelling than for pricing GHGs from other sectors."
    - Regulation can mimic tax effects given limited observable mitigation responses.
    - Downstream methane taxes do not reduce waste supply; fiscal or regulatory incentives at household/industrial level are needed.
    - "The 389 waste sites in the Philippines are publicly managed and it is more natural to set standards, rather than apply taxes, to public enterprises."

### Climate Policy Assessment Tool (CPAT) overview (Annex IV)
- Scope and outputs:
  - "CPAT provides, on a country-by-country basis for 200 countries, projections of fuel use and CO2 emissions by major energy sector."
  - Sectors: "power, industrial, transport, and residential sectors" (international aviation and maritime fuels excluded).
- Baseline projection inputs:
  - GDP projections (from latest IMF forecasts).
  - Assumptions about income elasticity of demand and own-price elasticity of demand for electricity and other fuel products.
  - Assumptions about the rate of technological change affecting energy efficiency and productivity of energy sources.
  - Future international energy prices.
  - Current fuel taxes/subsidies and carbon pricing held constant in real terms.
- Model mechanics for carbon pricing impacts:
  - Depend on: (i) proportionate impact on future fuel prices by sector; (ii) simplified model of fuel switching within power generation; (iii) various own-price elasticities for electricity and fuel use.
  - Fuel demand curves: "based on a constant elasticity specification."
- Data and parameterization:
  - Basic model parameterized using IEA data on recent fuel use by country and sector.
  - International energy prices projected using an average of IEA and IMF projections for coal, oil, and natural gas.
  - Fuel price elasticities chosen "broadly consistent with empirical evidence and results from energy models (fuel price elasticities are typically between -0.5 and -0.8)."
  - Carbon emissions factors by fuel product from IEA.
  - Domestic environmental costs of fuel use based on IMF methodologies.
- Caveats and limitations:
  - Model abstracts from mitigation actions beyond those implicit in recent fuel use/price data in the baseline.
  - Fuel price responses may not hold for dramatic price changes that drive major technological advances or rapid adoption of technologies such as carbon capture and storage or direct air capture.
  - Does not explicitly account for general equilibrium effects or changes in international fuel prices due to simultaneous reforms in large countries.
  - Parameter values chosen so results are "broadly consistent with those from far more detailed energy models."

### Decomposing inflation drivers in Brunei (SVAR analysis)
- Context and recent inflation dynamics:
  - "Average headline inflation increased from -0.4 percent before the pandemic, to 1.7 percent in 2021 and 3.7 percent in 2022."
  - Food price: "increased 5.1 percent y/y in 2022."
  - Transport price: "the increase in transport price peaked in 2021 at 5.9 percent EoY."
  - Foreign labor in private sector: "foreign workers contributed to half of the private sector employment before the pandemic" and "saw a 20 percent drop in foreign labor employment after the pandemic started."
  - Inflation uptick attributed to combination of demand and supply factors (global supply-chain disruptions, commodity price shocks from COVID-19 and the Ukraine war, labor market disruptions, fiscal and monetary support, release of pent-up demand after late 2021 reopening).
  - Exchange rate pass-through not analyzed in depth; "depreciation of Brunei dollar against the USD after the pandemic has been mild."
- SVAR methodology:
  - Purpose: separate inflation and GDP growth drivers into supply and demand shocks using quarterly GDP and CPI variation.
  - Identification:
    - Demand shocks: "movements of both GDP and CPI in the same direction" (raise output and raise inflation).
    - Supply shocks: "movements of GDP and CPI in opposite directions" (raise output but reduce inflation, or more precisely: supply shocks increase output but reduce inflation, while demand shocks increase both).
  - Model specification:
    - Yt = c + B1 Yt−1 + B2 Yt−2 + Et
    - Rewritten Yt = c + B1 Yt−1 + B2 Yt−2 + A0 Ut where Ut are orthogonal shocks, VAR(Ut)=I.
    - Sign restrictions imposed so demand shock raises GDP and CPI, supply shock raises GDP and lowers CPI (after controlling for constants and lags).
    - Numerical approach: Cholesky decomposition to generate candidate Ã0, jittered by random orthogonal matrices Q to create Â0 = Ã0 Q, repeated until sign restrictions satisfied.
  - Estimation:
    - Bayesian approach using Gibbs sampling.
    - Procedure: set Σ0 = identity; draw b1 from multivariate normal f(b|Σ) with mean b̂ = vec((X′X)−1 (X′Y)) and variance v̂ = Σ0 ⊗ (X′X)−1; draw Σ1 from inverse Wishart f(Σ|b) with scale (Y − X B1)′(Y − X B1) and T degrees of freedom.
    - Iterations: repeat 1000 times, discard first 100 iterations, take mean of coefficients across remaining iterations for estimates.
- Caveats specific to SVAR for small open economies:
  - Model assumes domestic supply and demand conditions are main drivers, which may overlook external supply/demand factors, especially import-price effects.
  - Extensive price control measures on basic goods (food, fuel) in Brunei could make headline inflation less sensitive to underlying shocks.

*Source: IMF staff (content from 1brnea2023002 - 3.      Table A1 provides more discussion of alternative options for the use of carbon pricing revenues).*

### 6.      We used Brunei’s quarterly data on CPI and real GDP in the estimation. The sample

### 6.      We used Brunei’s quarterly data on CPI and real GDP in the estimation.

### Data and methodology
- Sample period: 2011Q1 to 2022Q3.
- Series: seasonally adjusted CPI and real GDP series; year-on-year change applied to generate inflation and real GDP growth.
- Two model specifications estimated due to a large oil and gas (O&G) sector:
  - Using total real GDP.
  - Using real non-O&G GDP (derived by subtracting real outputs of oil and gas mining and manufacture of liquified gas and methanol industries from the real GDP series, then applying seasonal adjustment).
- Motivation: O&G sector output is mostly exported and does not necessarily respond to Brunei’s domestic supply and demand conditions.

### Inflation decomposition: key findings
- The inflation uptick in recent quarters was driven by both demand and supply factors.
- Using non-O&G real output (arguably more responsive to domestic demand):
  - Supply factors explained about 38 percent of CPI inflation in 2022.
  - Demand factors explained about 31 percent of CPI inflation in 2022.
  - In 2021 (economy mostly in lockdown): supply factors explained about 45 percent of inflation, while demand factors explained about 7 percent.
- The reopening in 2022 increased demand pressure relative to the prior year; post-pandemic reopening tends to foster a temporary surge in domestic demand.
- Supply disruptions continued to contribute significantly to inflation in Q2 and Q3 of 2022.
- Compared with Singapore, supply bottlenecks had a more pronounced impact on inflation in Brunei for 2022—potentially because domestic production capacity (including return of foreign workers powering the domestic service sector) took time to recover.

### Real output decomposition: key findings
- Supply disruptions have been a drag on output growth in recent quarters, while demand factors contributed positively to growth.
- O&G sector output growth was held back by infrastructure maintenance despite the surge in O&G prices during 2022.
- Non-O&G output received a boost after reopening but growth in the non-tradable sector (e.g., construction and services) has been modest so far.

### Fiscal sustainability context and diagnostics
- Revenue dependence and expenditure rigidity:
  - Revenue from O&G sector accounts for 84 percent of total revenue on average from FY2005/06 to FY2022/23.
  - Wages and salaries, pensions, royalty payments, and annually recurrent charges account for 82 percent of total expenditure on average from FY2005/06 to FY2022/23.
- Fiscal outcomes and trends:
  - Brunei has mostly run a fiscal deficit since FY2014/15; the government has run an average deficit of -8.3 percent of GDP since FY2014/15.
  - The non-O&G primary deficit has averaged around -53.9 percent of non-O&G GDP.
  - The non-O&G deficit declined to -39.2 percent of non-O&G GDP by FY2022/23 from -60.0 percent in FY2017/18.
  - Two notable surplus exceptions: FY2022/23 and FY2018/19 (rebound in oil prices).
- Risks: depletion of proven reserves, domestic production disruptions, reductions in global O&G prices, and global transition to net zero emissions by 2050.

### PIH framework, scenarios, and key assumptions
- Framework: Permanent Income Hypothesis (PIH) used to estimate net wealth (net financial wealth plus resource wealth) and the PIH norm (sustainable annual flow of income).
- Alternative long-term fiscal anchors examined:
  - Non-O&G fiscal balance as a constant share of non-O&G GDP (consumption smoothing across generations).
  - Real annuity (NRFB constant over time in real terms).
  - Real annuity per capita (NRFB constant in real per capita terms).
  - Comparison note: Bird-in-hand (BIH) contrasts with PIH by restricting spending to interest on accumulated financial wealth.
- Table 1 key scenario assumptions (as used in the PIH analysis):
  - Baseline: O&G output — IMF baseline projections (up to 2028), constant from 2029 to 2050.
  - Downside: O&G output — IMF downside projections (up to 2028), decrease 70% by 2050 (From 2029 to 2050).
  - Oil price (USD/barrel): IMF baseline projections (up to 2028); Increase 1% every year (From 2029 to 2050) under baseline; Decrease to $24 by 2050 (From 2029 to 2050) under downside.
  - Gas price (USD/MBtu): IMF baseline projections (up to 2028) & historical averages; Increase 1% every year (From 2029 to 2050) under baseline; Reach to $3.95 by 2050 (From 2029 to 2050) under downside.
  - Interest rate: 2.6%.
  - Population growth rate: 0.42%.
  - Long-term inflation rate: 1.0%.
  - Gov’t share in O&G revenue: 37.0%.
- Downside projection deviations from baseline (notes):
  - a) 1 standard deviation fall in O&G prices 2024-2028;
  - b) 10% reduction in O&G output volume 2023-2028;
  - c) Inflation for 2023 kept at 2022 level (higher than baseline);
  - d) reduction in FDI, imports, taxes (import duty, corporate tax), credit growth 2023-2028.

### Fiscal anchor scenario results and implications
- The most conservative fiscal anchor (PIH norm with deficit as constant share of non-O&G GDP) implies:
  - NRFB around 6 percent of non-O&G GDP.
  - Would impose a nearly 35 percent of non-O&G GDP adjustment in 2023 (likely least feasible).
- A PIH norm with a deficit constant in real terms (real annuity):
  - Has the smallest gap to the baseline and is likely the most feasible.
  - Allows for a much higher deficit during the transition period, enabling financing of development and diversification spending.
  - The deficit would need to be reduced below the PIH norm with a deficit as a constant non-O&G of 6 percent of non-resource GDP in the very long-term.
- Overall assessment:
  - The current baseline differs significantly from all long-term PIH fiscal anchors, requiring at least an adjustment of 15 percent in 2023 (under the PIH norm of a constant annuity in real terms).
  - The gap will need to be closed with a fiscal adjustment or “transition” during the medium-term.

### Policy recommendations and strategic priorities
- Accelerate fiscal consolidation efforts to ensure long-term fiscal sustainability and inter-generational equity.
- Adopt a broad-based revenue diversification strategy, including carbon pricing together with wage and subsidy reforms.
- Prioritize efficient and growth-friendly investments in health, education and green and resilient infrastructure to support diversification.
- Strengthen fiscal frameworks and institutions as a critical complement to reforms.
- Note: The 3-year Fiscal Consolidation Program (FCP) launched in FY2018/19 focuses on efficiency (corporatization, PPPs, consolidation) but its scope is narrow and implementation was limited by COVID-19; government remains committed to FCP initiatives.

*Source: IMF staff estimates and analysis in the supplied content.*

### 8. A gradual and sustained consolidation path could help Brunei feasibly transition to the

### 1brnea2023002 - 8. A gradual and sustained consolidation path could help Brunei feasibly transition to the

### Gradual transition scenario and PIH annuity implications
- Under a transition scenario for the PIH norm of a constant annuity in real terms:
  - NRFB would need to be at -16 percent of non-O&G GDP in 2028, compared to -26 percent in the baseline — requiring an additional 10 percentage point (ppt) of non-O&G in consolidation during the medium-term.
  - Beyond the medium-term, the constant real annuity would imply a continued and gradual adjustment of about 0.5 of non-resource GDP annually under the gradual transition scenario compared to around 0.3 ppt in the immediate transition scenario.
  - This projection is based on the conservative assumption of no long-term growth dividends from the additional spending.
- Trade-offs of the gradual transition:
  - Allows additional public investments in the medium-term and time to design supporting fiscal reforms to reduce the non-O&G deficit.
  - Comes at the cost of reduced fiscal space and more fiscal prudence in the long-term.

### Fiscal risk analysis and sensitivity to macro assumptions
- Key sensitivities and required re-assessments:
  - The fiscal framework should be adjusted based on the materialization of the real interest rate (r) - real GDP growth (g) dynamics.
  - The higher the real GDP growth, the lower the deficit that is sustainable.
  - The higher the real interest rate, the higher the permanent deficit that can be financed by returns from accumulated financial wealth.
  - A reversal to low interest rates in the longer-term would imply that the sustainable deficit path might be lower than expected.
  - Lower-than-envisaged public spending efficiency could result in lower-than-expected growth and affect the assessment of sustainable deficit.
- Data and coverage limitations:
  - Lack of data availability, including on the GLCs and the sovereign wealth funds, limits the scope for a more comprehensive fiscal risk analysis.

### Robustness to resource revenue shocks (stress tests)
- Two shock scenarios tested:
  - i) Permanent reduction in oil price equal to half the price of 2023 and a 10 percent reduction in exchange rate.
  - ii) Temporary reduction in oil prices equal to half the projected prices in 2023 and 2026 combined with a 10 percent depreciation in exchange rate in 2023.
- Three adjustment strategies considered:
  - i) No adjustment: pre-shock nominal path for spending unchanged.
  - ii) Full adjustment: spending reduced one-to-one with the shortfall in resource revenue.
  - iii) Partial adjustment: spending reduced to offset the shortfall but up to a maximum annual limit of 2 percent of total expenditures.
- Assumptions on financing shortfalls:
  - Any resource revenue shortfall not offset with spending reductions is assumed to be met first by drawing down liquid assets and second through new borrowing (once liquid assets are exhausted).
- Results:
  - Panel A (permanent shock) and Panel B (temporary shock) both indicate robustness of the real annuity fiscal anchor with the fiscal path not deviating substantially from fiscal sustainability under the partial and no adjustment scenarios.
- Note:
  - Brunei is not as exposed as its neighbors to losses from natural disasters, and hence, robustness to a natural disaster shock is not tested.

### Net Zero (2050) downside scenario
- If the globe meets its Net Zero target by 2050:
  - In the transition scenario, lower demand for O&G and lower global prices would imply that Brunei would need to adjust by an additional 4 percent of GDP by 2028 and further fiscal prudence in the long-term.
  - These results highlight sensitivity of the anchor to O&G sector assumptions and the importance of regular re-calibration of the framework to reflect market trends and fiscal risk analysis incorporating climate risks.

### Policy recommendations for consolidation and diversification
- A well-designed consolidation program can help attain the fiscal anchor in the medium-term while mitigating short-term growth and distributional impacts. Consolidation measures include:
  - Broad-based revenue diversification, including carbon pricing.
  - Containment of public sector wage and employment.
  - Reduction in tax expenditures and untargeted subsidies.
- Specific reform areas and evidence:
  - Tax reforms—GCC experiences provide guidance for non-O&G revenue mobilization (see examples of VAT, excise, and federal corporate tax adoptions in GCC countries).
  - Expenditure rationalization—composition of adjustment matters:
    - Capital spending cuts are less effective than reductions in transfers and wages according to the empirical literature cited.
    - Contractionary effects of fiscal consolidation in the short-term are largest when it involves cuts in productive expenditure on health, education, infrastructure, public order and safety, and public administration.
    - Estimated multipliers associated with spending on renewable energy (1.1-1.5) are larger than fossil fuel energy investments (0.5-0.6) with over 90 percent probability (Batini et al., 2021), which could also address transition risks.
- Illustrative country tax measures (GCC examples):
  - UAE: Excise tax series and VAT (2018, 5 percent), federal corporate tax for local businesses (2023, 9 percent).
  - Bahrain: VAT (2019, 5 percent → 2022, 10 percent), Excise tax (2017).
  - Saudi Arabia: VAT (2018, 5 percent → 2020, 15 percent), Real Estate Transaction Tax (2020, 5 percent), Excise taxes and fees/taxes on services.
- Examples of subsidy reforms:
  - Saudi Arabia aims to eliminate energy subsidies by 2030; UAE and Qatar link fuel prices to global prices since 2015 and 2016 respectively, resulting in energy subsidies around 1 percent of GDP in 2021 for UAE and Qatar.

### Institutional underpinnings and sequencing
- Success factors for growth-friendly consolidation:
  - Strong institutions, including a credible medium-term fiscal framework (MTFF) and sound public financial management (PFM) systems to strengthen fiscal discipline.
  - Transparency: broad fiscal coverage, realistic macro-forecasts, regular publication of fiscal plans and outcomes, and public debate.
  - Building strong budgetary institutions takes time and is effective when prioritized early and well sequenced.

### Financial sector stability and buffers (overview from appendix)
- Structure and size:
  - Total financial system assets represented around 108 percent of GDP as of the end of 2022; the banking sector accounts for 82 percent.
  - There are seven commercial banks (two domestic banks and five foreign bank branches) and one Islamic trust fund.
  - The two largest domestic banks account for 46 percent and 15 percent of total financial assets, respectively, as of the end of 2022.
- Capital and liquidity (as of end–2022 and historical series):
  - Regulatory capital to risk-weighted assets: 20.2 (2016–2022 series shown).
  - Tier 1 capital to risk weighted assets: 20.0 (2022).
  - Liquid asset ratio (liquid assets to total assets) was 43.8 (2022).
- Asset quality and provisioning:
  - NPL to total loans: 3.3 (2022).
  - NPL net of provisions to capital increased from 4.9 percent to 5.5 percent (2021 → 2022).
  - Provision coverage (specific provisions to total NPLs) fell to 37.8 in 2022.
- Profitability and efficiency:
  - Return on assets (before tax): 1.3 (2022).
  - Return on equity (after tax): 9.5 (2022).
  - Non-interest expense to gross income (efficiency ratio): 56.7 (2022).
- International investment and FX considerations:
  - Loans and financing to domestic borrowers represented 27 percent of total banking assets in 2022.
  - 52 percent of the banking sector assets as of the end of 2022 were invested abroad in the form of offshore investments or placements with financial institutions abroad; many held in Singapore dollar or United States dollar with hedging.
  - According to the largest bank’s 2022 financial statement, it held USD-denominated assets amounting to around 5.2 billion BND, equivalent to around 20 percent of total banking sector assets, with net exposure reported to -0.2 billion BND by use of foreign currency hedge.
- Nonbank financial sector:
  - Finance companies accounted for 8 percent of total financial sector assets in 2022.
  - Eleven insurers and Takaful operators accounted for 8 percent of total financial sector assets in 2022.
- Regulatory and macroprudential developments:
  - Brunei Darussalam Central Bank (BDCB) is strengthening surveillance and regulatory frameworks, implementing risk-based supervision and the three pillars of Basel II, and moving towards wider Basel III frameworks as medium-to-long term targets.
  - Macroprudential tools used include a cap on loan-to-value (LTV) ratio in 2012 (revoked in December 2016) and Total Debt Service Ratio (TDSR) introduced in 2015 (later relaxed), and designation of domestic systemically important bank(s) (D-SIBs) in 2020 with additional capital buffer requirements from 2023.

*Source: IMF staff estimates.*

### 5.       The following trends are observed upon analyzing the financial soundness indicators

### 5.       The following trends are observed upon analyzing the financial soundness indicators and balance sheet of banks

### a. Capital adequacy and NPLs
- CAR declined from 21.5 percent in Q4 2021 to 20.2 percent in Q4 2022, mostly due to an increase in risk-weighted assets.
- NPL (net of provisions) to capital ratio increased from 4.9 percent to 5.5 percent, reflecting a decline in provision coverage due mainly to the write-off of NPL accounts.

### b. Sensitivity to oil prices and credit risk dynamics
- Banks’ regulatory and excess reserve balance at the BDCB and deposits offshore increase in line with oil price increases.
- Deposits of banks have a high correlation with oil price movements; credit to the private sector shows a weak correlation, reflecting that banks place or invest increased deposits offshore rather than extend domestic credit.
- VAR results: NPL growth increases 0.455 percentage point with a lag of three quarters after one percentage point decline in oil price growth, indicating credit risk increases when oil price decreases with some lag.

### c. Market and liquidity risks from foreign currency-denominated assets and investment portfolios
- Banks’ investment portfolio amounts to around 20 percent of total assets and they are exposed to potential market risks.
- Interest rate stress test on investment portfolios (approximate asset maturity data from published financial statements) suggests total capital to risk-weighted assets of banks could be reduced by 3.2 percent, resulting in 16.4 percent under the most severe case.
- A large share of assets is held offshore; during a global crisis foreign banks might ringfence these assets and prevent Brunei banks from accessing them, implying possible liquidity risk.

### d. Liquidity structure and tail risks
- Banks are predominantly funded by deposits; loan-to-deposit ratio is 36 percent in 2022.
- A large part of deposits is invested offshore in the forms of deposits to foreign financial institutions and investment in securities.
- A liquidity stress test for seven commercial banks suggests that with an assumed 10 percent run on domestic liabilities (mostly deposits from non-financial domestic customers) and 5 percent run on other liabilities, banks may not be able to maintain an LCR of 100 percent if liquidity placed offshore is not available; while the ratios become over 300 percent if offshore liquidity is available.

### e. Credit growth composition and sectoral concentration
- Credit growth to corporates (non-households) has been in an upward trend, while credit growth to households is moderate.
- Loans to the household sector represented 50.4 percent of domestic lending in 2022.
- Domestic credit growth to households was 1.3 percent in 2022.
- Total household debt to the banking sector is 12.4 percent of GDP (does not include government mortgage lending to employees).
- Domestic corporate loans have shown higher growth recently, mainly driven by the downstream O&G sector in the manufacturing sector and the construction sector.
- Manufacturing and construction sectors represent the largest and the third largest share in banks’ exposures to the domestic corporate sector.
- Given their high NPL share in the past, credit risk may increase, especially if oil prices decline and the economy deteriorates.

### 6. Authorities’ actions and macroprudential context
- Authorities have implemented all three pillars of the Basel II framework.
- Risk-based supervision was strengthened by designating domestically systemic banks (D-SIBs — two so far) and imposing additional capital buffer requirement for D-SIBs starting in 2023.
- BDCB has been using macroprudential tools to address sectoral risks, but instruments currently available and used in Brunei are fewer than those for peer countries.

### Annex I — VAR model applied (key points)
- Data: oil price, NPL, and credit to the private sector by banks from 2011 to 2019 at quarterly frequency.
- VAR specification: 푌푡 = 푐 + ∑퐴푖 푌푡−푖 (i=1..5) + 휀푡 ; 휀푡 ~ W.N.(Σ). Endogenous variables: oil price growth y/y, growth y/y of gross NPL, and growth of bank credit to the private sector y/y.
- Key result restated: As oil price growth increases by one percentage point, it results in a decrease of 0.455 percentage points in NPLs growth after three quarters.
- Limitations: quarterly data between 2011 and 2019 may not be sufficiently long; excluded data after the COVID-19 pandemic; limited stability due to limited data size and possibly omitted explanatory variables.

### Annex II — Interest rate stress test (key points)
- Stress scenarios on two largest banks with interest rate hikes of 50, 200, and 400 basis points.
- Scenario of largest interest rate hike aligned with Singapore’s interbank overnight rate change in 2022 from 0.3 percent in February to 4.4 percent in September.
- Method: investment portfolio broken down by maturities (up to 1 year, 1-3 years, 4-5 years, 6-10 years), assumed final redemption dates at end of 1st, 3rd, 5th, and 10th years; discount and coupon rates set at average of Singapore’s interbank overnight rate during 2022; durations calculated to derive sensitivities to interest rate increases.
- Finding: possible vulnerabilities to investment portfolios in terms of interest rate risks.

### Annex III — Liquidity stress test (key points)
- Stress scenarios on seven commercial banks: scenario X = outflow of 10 percent of domestic liabilities plus 5 percent of foreign liabilities; scenario Y = outflow of 10 percent of domestic liabilities plus 20 percent of foreign liabilities during the next month.
- LCR formula used: LCR = Available Liquid Assets (under a preshock, a postshock1 or a postshock2) / Assumed Outflow (under a scenario X or Y).
- Available liquid assets defined as: 1) domestic and foreign liquid assets with maturity up to one month for baseline (pre-shock); 2) domestic liquid assets with maturity up to one month as a severe case (post-shock 1); 3) 80 percent of domestic liquid assets with maturity up to one month as a more severe case (post-shock 2).
- Domestic liquid assets include cash-in-hands, deposit balance with BDCB, investment in government sukuks. Foreign liquid assets include placements with foreign banks including deposits with parent banks located offshore, and investment offshore.
- Note: For banks that do not disclose detailed maturity information, liquid assets with maturity up to one month are estimated (e.g., placement with banks with maturity up to 3 months are divided by 3 for the amount with maturity up to one month).

*Source: IMF staff analysis in the chapter titled “5.       The following trends are observed upon analyzing the financial soundness indicators” from the provided PDF content.*

### 1. Health initiatives: Assistive Technology and Robotics in Healthcare, HealthHub,

### 1. Health initiatives: Assistive Technology and Robotics in Healthcare, HealthHub,

### Health initiatives
- Named initiatives and tools:
  - Assistive Technology and Robotics in Healthcare
  - HealthHub
  - National Steps Challenge™ & Healthy 365 App
  - TeleHealth
- Implicit focus: digital health platforms, wearable-driven wellness challenges, remote care tools.

### Transport initiatives
- Named initiatives and tools:
  - Autonomous Vehicles
  - CETRAN (test circuit enabling research and testing of self-driving vehicles before these officially hit the roads)
  - Contactless Fare Payment
  - On-Demand Shuttle
  - Open Data and Analytics for Urban Transportation

### Urban solutions initiatives
- Named initiatives and tools:
  - Punggol Smart Town
  - Smart Nation Sensor Platform (SNSP: an integrated, nationwide platform that uses sensors to collect essential data that can be analyzed to create smart solutions)
  - Smart Water Meter
  - OneService App
  - OneService Chatbot
  - Smart Urban Planning
  - Smart Towns
  - myENV App (a convenient tool to update the public on the latest environmental news, including PSI readings and dengue outbreak alerts)
  - Elderly Monitoring System
  - Dengue Hotspots Survey Drones

### Finance initiatives
- Named initiatives and tools:
  - GoBusiness (the go-to platform for businesses in Singapore to access Government e-services and resources)
  - Corppass (a secure log-in method for business to transact with the Government online)
  - Data Innovation Programme Office
  - FinTech Sandbox
  - Networked Trade Platform
  - SGFinDex (the world's first public digital infrastructure to use a national digital identity and centrally managed online consent system to enable individuals to access their financial information held across different government agencies and financial institutions)
  - SGTraDex (a digital infrastructure that facilitates trusted and secure sharing of data between supply chain ecosystem partners)

### Digital Government Services initiatives
- Named initiatives and tools:
  - LifeSG (allows you to easily access Government services, keep up with the latest news and updates, track your applications and more)
  - National Digital Identity
  - CentEx (The Centre of Excellence allows us to innovate quickly and develop citizen-centric services effectively)
  - CrowdTaskSG (a web portal for government agencies in Singapore to engage citizens and gather insights through crowdsourcing tasks)
  - Digital Birth and Death Certificates
  - HDB Resale Portal

### Singapore’s approach and private-sector focus (relevance for Brunei)
- Strategic emphasis: focus more on digital capabilities and business opportunities of the private sector to unleash digital capabilities.
- Program examples supporting private-sector digital adoption and workforce upskilling:
  - Advanced Digital Solutions: supports adoption of advanced technologies (e.g., Artificial intelligence, Robotics, Blockchain and Internet of Things) and integrated digital solutions (e.g., Business-to-business solutions that integrate inventory management, e-invoicing, and digital payments).
  - Grow Digital: offers small and medium-sized enterprises (SMEs) digital solutions to expand their businesses through e-commerce platforms, both locally and overseas, creating economies of scale and scope.
  - Workforce training and reskilling initiatives:
    - SGUnited Mid-Career Pathways-Company Training (SGUP-CT)
    - SGUnited Skills (SGUS)
    - SkillsFuture Career Transition Program (SCTP)
    - SGUnited Mid-Career Pathways-Company Attachment (SGUP-CA)

### Monetary Authority of Singapore (MAS) — digital financial innovation areas
- Active areas:
  - Digital banks
  - Central bank digital currency (CBDCs)
  - Cross-border payment systems
  - Use of crypto assets

### Key statistic on robotics adoption
- Singapore robot density: increased from "about 1 operating robot per 1,000 employees in 2008 to 45 operating robots per 1,000 employees in 2018."

*Source: 1brnea2023002 - 1. Health initiatives: Assistive Technology and Robotics in Healthcare, HealthHub,*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2023/english/1brnea2023002.pdf_