## 1estea2025002-print-pdf

## Source details

**Canonical URL:** [1estea2025002-print-pdf](https://www.imf.org/-/media/files/publications/cr/2025/english/1estea2025002-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2025/english/1estea2025002-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2025/english/1estea2025002-print-pdf.pdf.json)

---

### A. Fiscal Multipliers — Concepts, Methods, and IMF “Bucket Approach”
- Fiscal multipliers: short-term ratio of change in output to an exogenous change in a fiscal instrument; do not imply effects on employment, distribution, or long-term potential GDP.
- Estimation challenges:
  - Two-way causality between fiscal variables and GDP.
  - No consensus on ideal identification; main approaches: SVAR (Blanchard and Perotti 2002) and narrative approaches (Romer and Romer 2010, Devries et al. 2011).
- Aggregate literature evidence (37 papers, 2002–2025):
  - "110 out of 134 estimates of one-year tax or fiscal revenue multipliers ... show negative values, 95 of which are between zero and -1."
  - "Considering 56 estimates of spending multipliers, 46 are positive (33 between zero and 1)."
- IMF “Bucket Approach” (Batini et al, 2014):
  - Six structural characteristics scored 0 or 1 (with occasional 0.5), aggregate score determines multiplier bucket.
  - Bucket ranges (Table 1):
    - Score 0 - 3: Low multiplier 0.1 – 0.3
    - Score 3 - 4: Medium multiplier 0.4 – 0.6
    - Score 4 - 6: High multiplier 0.7 – 1.0
  - Adjustments:
    - If economy near lowest point in cycle (largest negative output gap), increase both bounds by 60 percent.
    - If near peak, reduce both bounds by 40 percent.
    - If monetary policy near zero-lower bound, increase multiplier range by up to 30 percent.
- Application to Estonia (as of 2025Q1):
  - Scores assigned:
    - Low trade openness: 0
    - High Labor market rigidity: 0.5
    - Weak automatic stabilizers: 1
    - Quasi-fixed exchange rate: 1
    - Low/safe public debt level: 1
    - Effective PFM and Revenue Administration: 1
  - Resulting score: 4.5 → high-multiplier range; mid-range estimate 0.85.
  - Output gap adjustment:
    - Staff estimates output gap in 2024 at -2 percent of potential GDP and expects it to remain broadly unchanged in 2025.
    - Applying 60 percent uplift: 0.85 x (1+0.6) = 1.36.
    - Adjusted lower bound: 0.7 x 1.6 = 1.12.

### B. Static Keynesian Multiplier Estimates for Estonia (Average 2024-2030)
- Static multiplier formula preserved:
  - κii = ΔY / ΔAii = (1 − mii) / [1 − c(1 − mcc)] = (1 − mii) / [1 − (1 − t − s)(1 − mcc)]
- Parameter values used:
  - Tax Revenue (ratio to GDP) A: 0.357
  - Transfers and Subsidies (ratio to GDP) B: 0.191
  - A – B = Net Tax Rate (t): 0.166
  - Private Savings Rate (s): 0.218
  - Propensity to consume (c = 1 − t − s): 0.617
  - Import intensity of private consumption (mcc): 0.372
- Static fiscal multipliers (Average 2024-2030, Table 3):
  - Public Consumption (G) static fiscal multiplier: 1.44
    - Import intensity (mii) total: 0.12
      - direct: 0.04
      - indirect: 0.08
  - Public Investment (I) static fiscal multiplier: 0.96
    - Import intensity (mii) total: 0.41
      - direct: 0.25
      - indirect: 0.16
  - Combined G and I static fiscal multiplier: 1.16
    - Import intensity total: 0.29
      - direct: 0.16
      - indirect: 0.13
  - Note: Public investment import intensity uses import intensity of total investment as proxy.

### C. Blanchard-Perotti SVAR Estimates — Impact and Cumulative Multipliers (Estonia, 2001Q1–2024Q2)
- Identification and data:
  - Variables: real GDP, total government spending, total fiscal revenues; normalized by trend in GDP (HP-filtered).
  - Controls: linear/quadratic trends, seasonal dummies, known tax reform indicators, ECB policy rate, terms-of-trade index, trade-weighted foreign partners’ GDP.
  - Impulse responses normalized to euros per one-euro shock.
- Main statistically significant (5 percent) non-cumulative responses to a one-euro shock:
  - Revenues: real GDP falls for 10 quarters; peak immediate non-cumulative response of -0.25 euro after 4 quarters.
  - Spending: real GDP increases by about 0.7 euro immediately; response statistically significant up to 4 quarters.
- Granular cumulative multipliers (selected entries preserved exactly):
  - Total Revenues: 1st quarter -0.07; After 4 quarters -0.19; After 8 quarters -0.52; Peak/Through -0.52 (8th quarter)
  - Direct Taxes: 1st quarter -0.62; After 4 quarters -1.17; After 8 quarters -1.21; Peak/Through -1.32 (7th quarter)
    - PIT: 1st quarter -0.54; After 4 quarters -1.23; After 8 quarters -1.47; Peak/Through -1.59 (7th quarter)
    - CIT: 1st quarter -1.07; After 4 quarters -0.61; After 8 quarters -0.84; Peak/Through -1.07 (1st quarter)
    - VAT: 1st quarter -0.35; After 4 quarters -0.05; After 8 quarters 0.19; Peak/Through -0.35 (2nd quarter)
  - Total Spending: 1st quarter 0.75; After 4 quarters 0.64; After 8 quarters 0.64; Peak/Through 0.75 (1st quarter)
    - Government Consumption: 1st quarter 1.05; After 4 quarters 0.68; After 8 quarters 0.35; Peak/Through 1.05 (1st quarter)
      - Wages and Salaries: 1st quarter 1.89; After 4 quarters 0.91; After 8 quarters 0.38; Peak/Through 1.89 (1st quarter)
      - Intermediate consumption: 1st quarter 0.01; After 4 quarters 0.00; After 8 quarters 0.00; Peak/Through 0.01 (1st quarter)
    - Subsidies: 1st quarter 0.50; After 4 quarters 1.12; After 8 quarters 1.50; Peak/Through 1.50 (8th quarter)
    - Transfers: 1st quarter 0.69; After 4 quarters -0.03; After 8 quarters -0.10; Peak/Through 0.69 (1st quarter)
    - Gross Fixed Capital Formation: 1st quarter -0.02; After 4 quarters 0.25; After 8 quarters 0.17; Peak/Through 0.30 (2nd quarter)
- Dynamics and interpretation:
  - Taxes: negative on impact; direct taxes (especially PIT) become more negative and persistent over two years; VAT and CIT decline in absolute effect after one year.
  - Spending: positive on impact (except public investment initially barely negative); peak effects occur sooner and decay rapidly, except subsidies which build up over two years.
  - Wages and salaries have the largest immediate spending impact (1.89) but fade by 80 percent in two years.

### D. SVECM Estimates — Permanent Shocks and Potential GDP Effects
- Model: SVECM estimated with Bayesian techniques; allows permanent fiscal shocks to affect both output gap and potential GDP growth.
- Calibration: immediate effects of temporary shocks aligned with BP SVAR when converted to euros.
- Quantified effects of 1 percent permanent shocks (preserved exactly):
  - Immediate estimated effect: a little over 0.01 percentage points decline in (quarterly) potential GDP growth for both permanent revenue and spending increases.
  - Over 2 years: about 0.1 percentage point lower GDP.
  - Over 5 years: real GDP is about 0.2 percentage point lower following a permanent tax increase and 0.15 percentage point lower following a permanent spending increase.
  - Long run cumulative effect on GDP: -0.25 percentage point for permanent tax shock and -0.22 percentage point for permanent spending shock.
- Interpretation: permanent revenue and spending increases have small but persistent negative effects on potential GDP growth; negative impact of permanent revenue shocks is larger (in absolute terms) than that of spending increases.

### E. Fiscal Multipliers — Aggregate Conclusions and Short-term Trade-offs
- Aggregate first-year multipliers across three methods (IMF’s BA, static comparative approach, and BP SVAR) summarized:
  - general fiscal variables (1.1)
  - aggregate spending (0.6−1.2)
  - net fiscal revenues (-0.2)
- Trade-offs for consolidation (short-term perspective):
  - Spending cuts (especially wage bill reductions) produce larger immediate negative output effects, but these effects are short-lived.
  - Revenue-based consolidation via direct taxes: slightly lower immediate effects but more persistent output costs; direct tax multipliers show high persistence over two years.
  - Considering two-year multipliers, spending cuts unrelated to subsidies or VAT hikes may be most growth-friendly.
  - Caveats: buoyancy, administrative burden, progressivity, distributional impacts, and tax avoidance risks must be considered.
- Long-term perspective:
  - Permanent increases in revenues and spending reduce level of GDP; permanent revenue shocks have larger negative long-run effects than spending increases, suggesting spending-based consolidation could be less costly in the long run relative to tax-based consolidation.

### F. Tax System Overview, Recent Changes (2024–26), and Revenue Challenges
- Key features of Estonia’s tax system:
  - Single-rate corporate income tax (CIT) on distributed profits.
  - Flat personal income tax (PIT) excluding personal dividends.
  - Limited wealth taxation; property tax based on land value only.
  - Territorial tax system exempting foreign profits.
- Main changes introduced in 2024–26 (Box 1, preserved exactly):
  - Income tax:
    - Income tax rate increased from 20% to 22% in January 2025 (payable on distributed profits).
    - Lower 14% rate on regular profit distribution abolished.
    - Quarterly advance CIT rate for credit institutions increased from 14% to 18%.
    - Plan to replace phased-out income tax allowance with a flat €8,400 annual income tax allowance postponed until 2026.
    - Entrepreneur account: Tax rate cut from 40% to 20%.
    - Social contributions: Minimum obligation increased from €239.25 to €270.60.
  - VAT:
    - VAT standard rate increased from 20% to 22% in January 2024.
    - Accommodation services reduced to 13% (from 9%).
    - Lowest 5% rate abolished; VAT system now has three rates: 22%, 13%, and 9%.
    - Special scheme: €40,000/calendar year VAT registration threshold to include supply of real estate transactions and financial/insurance services.
    - Definition of new buildings changed to “less than two-year-old”.
  - Motor vehicle tax:
    - New tax applied to all motor vehicles registered in January 2025; first payment deadline in June.
    - Average tax amount / passenger vehicle: €157.
  - Excise taxes:
    - Alcohol, tobacco: increase by 5% (Jan 2025), 5% (July 2025), and 10% (Jan 2026).
    - Gasoline: increase by 5% (July 2025), and 5%/year (May 2026-28).
    - Diesel: increase up to €493/kl May 2025-27.
  - National security tax:
    - Standard VAT rate further increased from 22% to 24%, starting from July 1, 2025.
    - Additional 2% tax levied on personal incomes (including for the entrepreneur account) from January 1, 2026.
    - Additional 2% tax levied on distributed profits from January 1, 2026.
- Revenue pressures and options:
  - Defense spending: current 3.5 percent of GDP; authorities’ revised target 5.4 percent of GDP.
  - Aging risk scenario (2024 Ageing Report): healthcare and long-term care could be 6 ppts higher in the long run than projected at current policies.
  - Estonia is about 6 ppts of GDP below European AEs and 1.7 percent below CESEE in revenue for the same GDP per capita level.
  - Options to strengthen revenues:
    - Address PIT allowance design; calibrate basic allowance, tax rates, and/or tax brackets.
    - Improve tax administration capacity; explore alternative CIT regimes to broaden base while preserving competitiveness.
    - Streamline remaining VAT exemptions.
    - Limit exemptions on residential land and develop a fiscal cadaster to introduce a modern tax on immovable property.
- Key statistics (2023 and related estimates):
  - Estonia tax revenue: 22 percent of GDP (2023).
  - Indirect taxes (mainly VAT) are 42 percent of tax revenue.
  - Direct taxes (mostly PIT) account for 29 percent of tax revenues.
  - CIT revenue: 1.9 percent of GDP (Estonia) vs. 5.0 percent of GDP (Nordics), 3.5 percent of GDP (European AEs), and 2.4 percent of GDP (European EMEs) (2023).
  - Property tax revenue in Estonia: less than 0.2 percent of GDP.
  - VAT efficiency per unit of final consumption: 74 percent in 2022.
  - Excise revenue: almost 3 percent of GDP.
  - Tax potential estimated at around 35 percent of GDP (Benitez et al., 2023).
  - Tax gap estimated at about 14 percent of GDP (actual revenues could be up to 14 percent of GDP higher at current development).
    - Peers: about 8 percent of GDP in European AEs and Nordic countries; 7.3 percent in European EMEs.

### G. Personal Income Tax (PIT) Design, Distortions, and Reform Scenarios
- Current design and parameters (preserved exactly):
  - Single statutory tax rate of 22 percent on income above the basic allowance.
  - Basic allowance valued annually at €7,848 in 2024.
  - Allowance phased out linearly over annual income range from €14,400 to €25,200.
  - Phase-out rate around 73 percent; implication example preserved: 72%*22%=15.8%.
  - Marginal rate pattern: inverted U-shape with marginal rates reaching about 35 percent at average annual income levels, then declining to 22 percent for higher incomes.
- Status Quo outcomes:
  - PIT revenue at 6.9 percent of GDP.
  - Gini coefficient improves only by 3.2 ppts relative to pre-tax income.
- Announced 2026 reform (as described):
  - Universal basic exemption of €700/month or €8,400/year.
  - Flat rate increases to 24 percent.
  - New system estimated to collect 0.5 percent of GDP less than the status quo.
- Revenue-neutral and alternative reform designs (values preserved exactly):
  - Revenue-neutral options:
    - (i) increase the single statutory rate from 22 to 26 percent, or
    - (ii) lower the universal allowance from €8,400 to €6,750.
    - Redistributive capacity higher in option (i) and lower in option (ii) relative to the status quo.
  - Two-rate scenario:
    - Rates at 22 and 30 percent would yield 0.9 ppts of GDP additional revenue relative to the status quo and 1.4 percent relative to the announced reform.
    - Two-rate design shifts burden slightly to the right of the wage income distribution (from middle 6-8th deciles to top 2 deciles).
    - Average rate increases by 2 ppts relative to the status quo, and by 3.5 ppts relative to the announced reform.
    - If combined with a lower standard CIT rate on overall profits (e.g., 15 percent), a two-rate PIT scenario (15 percent in bracket #2, and 29 percent in bracket #3) should be considered to achieve revenue neutrality while preventing arbitrage.
  - More progressive PIT scenario:
    - Four rates (10-15-25-40) and a lower universal allowance of €6,000.
    - Would bring in 1.4 percent of GDP extra revenue relative to the status quo and almost 2 ppts relative to the announced reform.
    - Would reach 85 percent of wage earners.
    - Average rate of 21.6 percent.
- Policy implications:
  - Higher rates for above-median incomes raise revenues and improve progressivity.
  - Two-rate or multi-rate structures can improve revenue mobilization and redistribution while protecting lower-income groups.
  - Broaden base and strengthen administration to capture untaxed activities and reduce informality.

### H. VAT Performance, Gaps, and Policy Recommendations
- VAT performance:
  - VAT C-efficiency high since 2005; automated digital administration (99 percent of declarations submitted digitally in 2023).
  - Higher-than-EU-average VAT registration threshold: turnover < €40,000 (covers 60 percent of businesses).
  - Exemptions remain widespread and create inefficiencies.
- European Commission 2024 VAT Gap Report (based on 2022):
  - VAT policy gap about one third of theoretically collectable VAT revenue.
  - Closing actionable exemption gap would yield 1.1 percent of GDP.
  - Closing rate gap would yield 0.3 percent of GDP.
  - Strengthening tax compliance could generate up to 0.4 percent of GDP.
- Noncompliance sensitivity:
  - In the EU, noncompliance estimated to worsen by about 0.5 ppts of theoretical revenues for each 1 ppts VAT rate increase (EC, 2024).
- Administrative measures proposed:
  - Require all invoices declared in VAT return annex (currently only transactions above $1,000 included).
  - Expand e-invoicing for all VAT payers (current coverage 7 percent).
- Policy trade-offs:
  - Resist proliferation of reduced rates and exemptions; equity concerns better addressed via income-based targeted transfers.
  - Preferred benchmark: uniform, broad-based consumption tax with few well-specified exemptions.

### I. Property Tax — Current Status, Potential, and Reform Steps
- Current features:
  - Property taxation limited to a land tax; residential land exempt up to generous thresholds.
  - Property taxes collect less than 0.2 percent of GDP.
  - 2022 land revaluation implemented from 2024; property values increased 8 times requiring lower maximum tax rates.
- 2025/2026 specifics:
  - Upper bounds of new tax rate for 2025: 0.1 to 1 percent of market value (residential/agricultural), 0.1 to 0.5 percent (profit-yielding land), 0.1 to 2 percent (other use).
  - 2025 cap on value increase increased to 50 percent/€20 from 10 percent/€5 previously.
  - From 2026 homeowner’s exemption decided by local governments and capped at €1,000.
  - Example: in 2025 City of Tallinn exempted 172,500 homeowners from paying taxes on domestic land, about 80 percent of taxpayers.
- Revenue potential illustrative scenario:
  - Estonia has about 266,475 residential buildings (2021 census).
  - Illustrative property-value tax scenario (primary vs. secondary residences, rates e.g., 0.05 percent and 0.1 percent, exemption threshold 10 percent) could generate as much as 1.7 percent of GDP.
- Administrative constraints:
  - No fully-fledged fiscal cadaster; registries not synchronized; periodic revaluations every four years starting 2022.
- Recommendations:
  - Near term: limit residential primary owner exemption; synchronize population registry with land registry and income statements; design broad base with means-tested compensation.
  - Medium term: prepare for value-based property tax (fiscal cadaster), flat rate, minimum exemptions, uniform treatment of business/residential property, and measures for owners of multiple properties.

### J. Corporate Taxation — DPT Assessment, Options, and Trade-offs
- DPT mechanics and recent changes:
  - DPT levied at 22/78 of net profit distribution (about 28.2% gross-up); from January 2026 nominal rate goes up to 24% (fractional 24/76 = 31.6%).
  - Undistributed corporate profits tax exempt.
  - Abolition of reduced 14% CIT on regular profit distributions from 2025; advance CIT for banks increased from 14% to 18%.
- Advantages of DPT:
  - Less distortive than standard CIT (falls mostly on rents); impulse to corporate investment; neutrality between debt and equity; simplicity in digital context.
- Drawbacks and risks:
  - Potential sizeable revenue loss if statutory rates not increased to offset narrow base.
  - Arbitrage risks (incorporation by high-wealth individuals, shareholder deferral), challenges assessing compliance.
  - Baltic DPTs tax gross distributions and thus capture more than pure rents.
- Policy options presented:
  - Improve DPT design: higher statutory rate; move base toward net distributions (closer to rents) with caution on induced arbitrage.
  - Move toward standard CIT via well-sequenced reform with mitigation (ACE or immediate expensing).
  - Explore cash-flow rent taxes (R-base, R+F-base, S-base) recognizing transition and revenue risks.

### K. Firm Dynamics, Misallocation, and Productivity in the Baltics (Estonia focus)
- Macro message:
  - Russia’s war in Ukraine increased input costs; slow productivity growth has eroded competitiveness.
  - Labor productivity growth decelerated over past two decades in the Baltic economies.
- Decomposition framework (Decker et al. 2017 preserved notation):
  - Labor productivity growth = within-firm productivity growth + allocative efficiency change + contribution of entrants + contribution of exits.
  - Notation preserved: ∆퐺퐺푖푖 = ∆푝푝̅푖푖,푐푐 + ∆푐푐푐푐푐푐푐푐(휃휃표표,푝푝표표) + 휃휃퐸퐸2(퐺퐺퐸퐸2 − 퐺퐺퐶퐶2) + 휃휃푋푋1(퐺퐺퐶퐶1 − 퐺퐺푋𝟏).
- Estonia results (statistical register and Orbis robustness):
  - Aggregate allocative efficiency contribution declined during 2006-2022 and turned negative after 2016.
  - Firms with higher productivity were growing employment during 2001-2015; allocative efficiency worsened after 2006.
  - Contribution of firm entry is negative (entrants have lower labor productivity on average than incumbents).
  - Contribution of firm exit to labor productivity growth is positive throughout and increasingly so after 2015.
  - Net entry contribution to labor productivity growth very small.
- Cross-country observations:
  - Latvia: allocative efficiency turned negative during 2016-20; firm entry contribution negative but narrows; exit contribution positive and outweighs entry for 2016-19.
  - Lithuania: allocative efficiency declined and turned negative during 2011-15; exit failed to fully compensate negative entry contribution during 2001-15.
- Entrant dynamics:
  - Entrants are smaller and less resourced; distribution of entrant labor productivity skewed low at entry but improves over ten years.
  - Fast-growing young firms' employment share in Estonia is "2-3 percent"; corresponding U.S. share "about 6 percent".
- Policy implications and recommendations:
  - Improve allocative efficiency to restore productivity and competitiveness.
  - Targeted R&D support and financing for startups in high-tech sectors.
  - During crises, avoid prolonged subsidies for unviable firms; target support for displaced workers and high-growth sectors.
  - Streamline insolvency to allow efficient exit; limit bailouts for nonviable firms while distinguishing startups with legitimate equity gaps.
  - Reduce regulatory burdens in sector-specific areas (road/air transport, mobile e-communications) and align higher education with market demand (STEM).
  - Deepen domestic and EU capital markets; expand venture capital and equity financing; consider pension fund participation.
  - Further investigation recommended: case studies of successful high-quality entrants and policy effectiveness.

*Source: IMF staff estimations and analysis as presented in the IMF chapter on Estonia (1estea2025002-print-pdf).*

### 1. Advanced Economies: Estimates of Fiscal Multipliers __________________________________ 5

### 1. Advanced Economies: Estimates of Fiscal Multipliers __________________________________ 5

### A. Introduction: Fiscal Multipliers
- Fiscal multipliers measure the short-term impact of discretionary fiscal policy on output; defined as the ratio of a change in output to an exogenous change in a fiscal instrument.
- Fiscal multipliers only refer to short-term effects of changes in fiscal variables on GDP, offering little guidance on the effect on employment, social outcomes, income distribution, and long-term (potential) GDP.
- Estimation challenges:
  - Two-way causality between fiscal variables and GDP due to automatic stabilizers and discretionary policy responses.
  - No consensus on the ideal methodology to isolate exogenous fiscal shocks; literature lacks consensus on multiplier sizes.
- Literature overview:
  - Uses two main identification approaches: Structural Vector Autoregression (SVAR, Blanchard and Perotti 2002) and narrative approaches (Romer and Romer 2010, Devries et al. 2011).
  - Aggregate evidence (37 papers, 2002–2025): tax/revenue multipliers mostly negative; spending multipliers mostly positive.
    - "110 out of 134 estimates of one-year tax or fiscal revenue multipliers ... show negative values, 95 of which are between zero and -1."
    - "Considering 56 estimates of spending multipliers, 46 are positive (33 between zero and 1)."
- Paper focus:
  - Applies the SVAR approach to Estonia and examines other IMF staff methods.

### B. The IMF’s “Bucket Approach”
- Method:
  - Back-of-the-envelope classification of countries into low, medium, or high multiplier "buckets" based on six structural characteristics (Batini et al, 2014).
  - Scores: 1 assigned to low trade openness, low labor market flexibility, weak automatic stabilizers, low exchange rate flexibility, “safe” levels of government debt, and highly effective public financial management; otherwise 0.
  - Aggregate score = sum of individual scores; bucket assigned via Table 1 ranges.
- Table 1 (Bucket ranges):
  - Score 0 - 3: Low multiplier 0.1 – 0.3
  - Score 3 - 4: Medium multiplier 0.4 – 0.6
  - Score 4 - 6: High multiplier 0.7 – 1.0
- Adjustments:
  - If economy is close to lowest point in cycle (largest negative output gap), increase both bounds by 60 percent.
  - If close to peak, reduce both bounds by 40 percent.
  - If monetary policy is close to the effective zero-lower bound, increase multiplier range by up to 30 percent.
- Application to Estonia (as of 2025Q1):
  - Scores assigned (Table 2):
    - Low trade openness: 0
    - High Labor market rigidity: 0.5
    - Weak automatic stabilizers: 1
    - Quasi-fixed exchange rate: 1
    - Low/safe public debt level: 1
    - Effective PFM and Revenue Administration: 1
  - Resulting score: 4.5 → implies value in high-multiplier range; mid-range estimate is 0.85.
  - Adjustment for output gap:
    - Staff estimates output gap in 2024 at -2 percent of potential GDP and expects it to remain broadly unchanged in 2025.
    - Applying the 60 percent uplift: 0.85 x (1+0.6) = 1.36.
    - Adjusted lower bound: 0.7 x 1.6 = 1.12.

### C. Static Keynesian Fiscal Multipliers
- Formula (textbook static multiplier for exogenous expenditure Aii):
  - κii = ΔY / ΔAii = (1 − mii) / [1 − c(1 − mcc)] = (1 − mii) / [1 − (1 − t − s)(1 − mcc)]
  - Definitions:
    - κii: static fiscal multiplier for public consumption or public investment
    - mii: import intensity of Aii
    - mcc: import intensity of private consumption
    - c, t, s: propensity to consume, average tax rate, savings rate
- Static estimates for Estonia (Average 2024-2030, Table 3):
  - Public Consumption (G) static fiscal multiplier: 1.44
    - Import intensity (mii) total: 0.12
      - direct: 0.04
      - indirect: 0.08
  - Public Investment (I) static fiscal multiplier: 0.96
    - Import intensity (mii) total: 0.41
      - direct: 0.25
      - indirect: 0.16
  - Combined G and I static fiscal multiplier: 1.16
    - Import intensity total: 0.29
      - direct: 0.16
      - indirect: 0.13
  - Additional parameters used:
    - A. Tax Revenue (ratio to GDP): 0.357
    - B. Transfers and Subsidies (ratio to GDP): 0.191
    - A – B = Net Tax Rate (t): 0.166
    - Private Savings Rate (s): 0.218
    - Propensity to consume (c = 1 − t − s): 0.617
    - Import intensity of private consumption (mcc): 0.372
  - Note: Public investment import intensity uses import intensity of total investment as proxy.

### D. Blanchard-Perotti SVAR Approach
- Identification logic:
  - SVAR isolates exogenous fiscal changes orthogonal to economic developments by assuming discretionary spending changes are not implemented within short intervals due to implementation lags.
  - Within a quarter, changes in fiscal variables arise from automatic stabilizers or truly exogenous fiscal shocks.
  - Independently estimated/calibrated elasticities of revenue and expenditure w.r.t. output are imposed as non-zero restrictions to identify automatic stabilizers; residual is fiscal shock.
  - For spending shocks: government spending is pre-determined within the quarter; Cholesky decomposition ordering places government spending first.
  - For tax shocks: taxes ordered first (tax decisions assumed to be taken first, spending responds).
- Estimation details for Estonia:
  - Sample: Estonian quarterly data on real GDP, total government spending, and total fiscal revenues from 2001Q1 to 2024Q2.
  - Variable normalization: all variables normalized by the trend in GDP (HP-filtered).
  - Exogenous regressors: linear and quadratic deterministic trends (when significant), seasonal dummies, and indicators of known tax reforms in Estonia.
    - Seasonal dummies are interacted with GDP, taxes, and spending to capture quarter-specific patterns.
    - Tax reform dummies constructed using European Commission’s 2021 Taxation Trends Report and institutional knowledge for 2022–2024.
  - Controls included as exogenous variables: ECB policy rate, a terms-of-trade index, and trade-weighted foreign partners’ GDP.
  - Impulse responses: computed without quarterly dependence dummies and normalized to represent the response, in euros, to a one-euro shock in the selected fiscal instrument.

*1. Advanced Economies: Estimates of Fiscal Multipliers __________________________________ 5*

### 15. Multiplier estimates from Blanchard-Perotti’s SVAR approach suggest negative and

### 15. Multiplier estimates from Blanchard-Perotti’s SVAR approach suggest negative and positive responses of real GDP to exogenous increases in total revenues and spending net of interest income and payments, respectively

### Blanchard-Perotti (BP) SVAR main findings
- Considering only statistically significant responses (at 5 percent) to a one-euro shock:
  - Real GDP falls for 10 quarters following exogenous increases in total net revenues, with a (non-cumulative) peak response of -0.25 euro after 4 quarters.
  - Real GDP increases by about 0.7 euro immediately after a spending shock; the non-cumulative response gradually decreases but remains statistically significant up to 4 quarters.
- The original impulse responses (IRF) are elasticities of GDP to autonomous changes in fiscal instrument i. Normalized impulse responses were calculated by dividing IRF by the average share of the fiscal instrument to GDP such that IIRF* = ∆GDP / ∆A_i.

### Granular fiscal multipliers (BP approach, cumulative horizons)
- Method: included fiscal instrument of interest as endogenous and its aggregate counterpart as exogenous control; cumulative multiplier k_i(h) computed following Ramey and Zubairy (2018). Only statistically significant responses at the 5 percent level are considered in multiplier computation.
- Key cumulative and impact multipliers (Table 4 — values preserved as reported):
  - Total Revenues: 1st quarter -0.07; After 4 quarters -0.19; After 8 quarters -0.52; Peak/Through -0.52 (8th quarter)
  - Direct Taxes: 1st quarter -0.62; After 4 quarters -1.17; After 8 quarters -1.21; Peak/Through -1.32 (7th quarter)
    - PIT: 1st quarter -0.54; After 4 quarters -1.23; After 8 quarters -1.47; Peak/Through -1.59 (7th quarter)
    - CIT: 1st quarter -1.07; After 4 quarters -0.61; After 8 quarters -0.84; Peak/Through -1.07 (1st quarter)
    - VAT: 1st quarter -0.35; After 4 quarters -0.05; After 8 quarters 0.19; Peak/Through -0.35 (2nd quarter)
  - Total Spending: 1st quarter 0.75; After 4 quarters 0.64; After 8 quarters 0.64; Peak/Through 0.75 (1st quarter)
    - Government Consumption: 1st quarter 1.05; After 4 quarters 0.68; After 8 quarters 0.35; Peak/Through 1.05 (1st quarter)
      - Wages and Salaries (/1 includes employers’ social contributions): 1st quarter 1.89; After 4 quarters 0.91; After 8 quarters 0.38; Peak/Through 1.89 (1st quarter)
      - Intermediate consumption: 1st quarter 0.01; After 4 quarters 0.00; After 8 quarters 0.00; Peak/Through 0.01 (1st quarter)
    - Subsidies: 1st quarter 0.50; After 4 quarters 1.12; After 8 quarters 1.50; Peak/Through 1.50 (8th quarter)
    - Transfers: 1st quarter 0.69; After 4 quarters -0.03; After 8 quarters -0.10; Peak/Through 0.69 (1st quarter)
    - Gross Fixed Capital Formation: 1st quarter -0.02; After 4 quarters 0.25; After 8 quarters 0.17; Peak/Through 0.30 (2nd quarter)

### Dynamics and interpretation across instruments
- Taxes:
  - Tax shocks are consistently negative on impact (within a quarter) and tend to be more persistent than spending multipliers.
  - Except for VAT and CIT, tax multipliers generally remain negative and become stronger over a two-year horizon.
  - Total net revenues, direct taxes, and PIT multipliers build up over time; within two years the negative multiplier effect of direct taxes doubles, driven by PIT which becomes three times as strong.
  - VAT and CIT decline in absolute terms after one year: VAT to almost zero and CIT by 40 percent.
- Spending:
  - All spending multipliers are positive on impact except public investment (barely negative on impact, becomes positive and peaks after two quarters).
  - Spending multipliers tend to peak sooner (first or second quarter) and rapidly decay, consistent with possible crowding-out and Ricardian-equivalence channels.
  - Subsidies are an exception: multiplier becomes three times stronger after two years, indicating buildup as subsidized programs mature.
  - Wages and salaries have the largest immediate impact but fade by 80 percent in two years; combined with near-zero intermediate consumption multiplier, government consumption multiplier remains above one on impact.
  - Public investment multipliers are smaller than government consumption multipliers but generally have a larger positive and persistent effect than intermediate consumption.

### SVECM estimates: effects on potential GDP
- Model: SVECM estimated with Bayesian techniques using quarterly Estonian data on real GDP, total spending, and total fiscal revenues from 2001Q1 to 2024Q2; allows permanent spending and tax shocks to affect both the output gap and potential GDP growth; temporary fiscal shocks affect only the cycle.
- Calibration: immediate effects of temporary fiscal shocks match BP SVAR results (when converted to euros); priors allowed negative effect of permanent tax shock on GDP and were agnostic on sign for permanent spending shock.
- Quantified effects of 1 percent permanent shocks:
  - Immediate estimated effect: a little over 0.01 percentage points decline in (quarterly) potential GDP growth for both permanent revenue and spending increases.
  - Compounded effects:
    - Over 2 years: about 0.1 percentage point lower GDP.
    - Over 5 years: real GDP is about 0.2 percentage point lower following a permanent tax increase and 0.15 percentage point lower following a permanent spending increase.
    - Long run cumulative effect on GDP: -0.25 percentage point for permanent tax shock and -0.22 percentage point for permanent spending shock.
- Interpretation: permanent increases in both taxes and spending have small but persistent negative effects on potential GDP growth, permanently reducing the level of GDP; negative impact of permanent revenue shocks is larger (in absolute terms) than that of spending increases.

### Conclusions and policy implications
- Aggregate summary:
  - Estimated first-year multipliers across three methods (IMF’s BA, static comparative approach, and BP SVAR) indicate important short-term costs of fiscal consolidation: general fiscal variables (1.1), aggregate spending (0.6−1.2), and net fiscal revenues (-0.2).
- Trade-offs for fiscal consolidation strategies (short-term multiplier perspective):
  - Spending cuts (especially reductions in the wage bill) produce larger immediate negative output effects, but these effects are short-lived.
  - Revenue-based consolidations via direct taxes have slightly lower immediate effects but output costs build over time and are more persistent.
  - Considering two-year multipliers and the high persistence of direct tax multipliers, spending cuts unrelated to subsidies or VAT tax hikes may be the most growth-friendly options.
  - Caveats: policy choices must abstract from other important considerations—buoyancy and administrative burden, progressivity, income distribution, risk of tax avoidance—that may influence policy design.
- Long-term perspective:
  - Permanent shocks to both revenues and spending reduce the level of GDP; negative output effect of permanent revenue shocks is larger than for spending increases, suggesting fiscal consolidations based on spending cuts could be less costly (or even not costly) in the long run relative to tax-based consolidations.

*Source: IMF staff estimations as presented in the IMF chapter on Estonia (Blanchard-Perotti SVAR and SVECM analyses).*

### 2. The growth rates of the trends in GDP, revenues, and spending are modelled to

### 2. The growth rates of the trends in GDP, revenues, and spending are modelled to

### Model Structure and Identification
- Trend growth of GDP, revenues, and spending are modelled to converge to the historical average GDP growth, ∆y.
- Assumption ensures that ratios of both revenues and spending to GDP converge to a constant.
- Cyclical components of revenues and spending embed automatic stabilizers (i.e., elasticities to the business cycle; see equations 5 and 8).
- Trend growth of revenues and spending are related to potential GDP growth (via equations 6 and 9, respectively).
- Identification strategy:
  - Temporary shocks do not affect trends (i.e., no temporary fiscal shocks affecting equations 2, 6, and 9).
  - Permanent shocks affect both cycles and trends (see equations 1-2, 5-6, and 8-9).
- To better identify the business cycle, a composite variable Ωay (훺훺
푎푎
푦푦) is included in the output gap equation; it includes detrended series of:
  - capacity utilization,
  - unemployment rate,
  - goods imports,
  - industrial production,
  - retail sales,
  - an economic confidence indicator.

### Estimation Results
- Estimation results are displayed in Table 1 (Annex II. Table 1: Estonia: SVECM Estimation Results).
- Priors and posterior moments are reported for parameters including AR coefficients, macro HFIs, permanent tax and spending shocks, capacity utilization, unemployment rate, imports of goods, industrial production, income elasticities, LT tax distortions, and LT spending distortions/externalities.
- Selected prior/posterior parameter entries (as shown in the table):
  - Beta prior: 0.8 mean; 0.2 std; mode 0.155 — AR coeff.
  - Gamma prior: 10.5 mean; 0.1 std; mode 0.114 — Macro HFIs.
  - Normal prior: 0.05 mean; 10 std; mode 0.045 — Perm tax shock.
  - Normal prior: -0.051 mean; -0.933 mode — Perm spending shock.
  - Gamma prior: 10.5 mean; 0.450 mode — Capacity Utilization.
  - Gamma prior: 1 mean; 0.5 std; mode 1.214 — Unemployment rate, (-).
  - Gamma prior: 10.5 mean; 0.454 mode — Imports of Goods.
  - Gamma prior: 10.5 mean; 1.182 mode — Industrial Production.
  - Other entries include AR coeff. priors with Beta 0.8, Normal priors for LT tax distortions and LT spending distortions/externalities, Beta and Gamma priors for income elasticities and capacity utilization, and Normal priors for income elasticity terms with values such as 0.7, -0.386 in the table.

### Tax System Overview and Recent Changes (2024-26)
- Estonia’s tax mix historically relies on consumption taxes—especially VAT—while income taxes are a relatively small share of revenue.
- Key features of Estonia’s tax system:
  - Single-rate corporate income tax (CIT) on distributed profits.
  - Flat personal income tax (PIT) which excludes personal dividends.
  - Limited wealth taxation with a property tax based solely on land value.
  - Territorial tax system exempting foreign profits.
- Main changes introduced in 2024-26 (Box 1):
  - Income tax:
    - Income tax rate increased from 20% to 22% in January 2025 (payable on distributed profits).
    - Lower 14% rate on regular profit distribution was abolished.
    - Quarterly advance CIT rate on overall profits payable by credit institutions and Estonian branches of foreign credit institutions increased from 14% to 18%.
    - Plan to replace phased-out income tax allowance with a flat €8,400 annual income tax allowance postponed until 2026.
    - Entrepreneur account: Tax rate cut from 40% to 20%.
    - Social contributions: Minimum obligation increased from €239.25 to €270.60.
  - VAT:
    - VAT standard rate increased from 20% to 22% in January 2024.
    - Accommodation services subject to reduced 13% VAT, up from 9%.
    - The lowest 5% rate was abolished; VAT system now has three rates: 22%, 13%, and 9%.
    - Special scheme for small businesses: €40,000/calendar year VAT registration threshold to include supply of real estate transactions and financial/insurance services.
    - Definition of new buildings in VAT Act changed to “less than two-year-old” for VAT applicability.
  - Motor vehicle tax:
    - New tax applied to all motor vehicles registered in January 2025; first payment deadline in June.
    - Tax consists of a registration fee and an annual motor vehicle tax determined by the base amount (min. €50), a CO2 component, and total weight.
    - Average tax amount / passenger vehicle: €157.
  - Excise taxes:
    - Alcohol, tobacco: increase by 5% (Jan 2025), 5% (July 2025), and 10% (Jan 2026).
    - Gasoline: increase by 5% (July 2025), and 5%/year (May 2026-28).
    - Diesel: increase up to €493/kl May 2025-27.
  - National security tax:
    - Standard VAT rate further increased from 22% to 24%, starting from July 1, 2025.
    - Additional 2% tax levied on personal incomes (including for the entrepreneur account) from January 1, 2026.
    - Additional 2% tax levied on distributed profits from January 1, 2026.

### Revenue Challenges and Options
- Fiscal pressures and spending:
  - Elevated near-term defense spending and long-term pressures from population aging and climate mitigation.
  - Defense spending at 3.5 percent of GDP (current) with authorities announcing a revised 5.4 percent target.
  - In a risk scenario (2024 Ageing Report), healthcare and long-term care could be 6 ppts higher in the long run than projected at current policies.
- Revenue shortfall and options:
  - For the same level of development (measured by GDP per capita), Estonia is about 6 ppts of GDP below European AEs and 1.7 percent below CESEE in terms of revenue.
  - Options to strengthen revenues include:
    - Addressing PIT revenue shortfall from the introduction of a uniform allowance by considering revenue neutral options (calibrate the basic allowance, tax rates, and/or tax brackets subject to intended progressivity).
    - Improving capacity of tax administration to analyze income statements and exploring alternative CIT regimes that preserve competitiveness while broadening the base.
    - Streamlining remaining VAT exemptions to broaden the tax base.
    - Limiting exemptions on residential land and taking steps to introduce a modern tax on immovable property by developing a fiscal cadaster to ensure fair taxation based on value and use of property.
- Trade-offs and distributional considerations:
  - Estonia’s reliance on indirect taxes yields a high indirect tax share and limits redistributive capacity compared with AEs and Nordics.
  - Post-tax and transfers income inequality in Estonia is reduced by almost one third, less than the euro area average of 38 percent, reflecting limited redistributive capacity from the tax side and heavier reliance on transfers.

### Key Statistics and Estimates
- Revenue and tax structure (2023):
  - Estonia tax revenue: 22 percent of GDP as of 2023.
  - Indirect taxes (mainly VAT) are 42 percent of tax revenue.
  - Direct taxes (mostly PIT) account for 29 percent of tax revenues.
  - CIT revenue: 1.9 percent of GDP (Estonia) vs. 5.0 percent of GDP (Nordics), 3.5 percent of GDP (European AEs), and 2.4 percent of GDP (European EMEs), as of 2023.
  - Property tax revenue in Estonia: less than 0.2 percent of GDP.
  - VAT efficiency per unit of final consumption: 74 percent in 2022 (implying 26 percent of theoretical revenue is lost via policy decision and/or noncompliance).
  - Excise revenue: almost 3 percent of GDP in Estonia.
- Tax potential:
  - Estonia’s tax potential estimated at around 35 percent of GDP using stochastic frontier analysis (Benitez et al., 2023).
- Defense spending:
  - Current defense spending: 3.5 percent of GDP.
  - Authorities’ revised defense target: 5.4 percent of GDP.
- Aging risk scenario:
  - Healthcare and long-term care could be 6 ppts higher in the long run than projected at current policies (2024 Ageing Report).

*Source: IMF staff estimates.*

### 9.      Estonia exhibits a higher tax potential than European AEs as well as a wider tax gap,

### 9.      Estonia exhibits a higher tax potential than European AEs as well as a wider tax gap,

### Tax potential and tax gap
- Tax gap estimated at about 14 percent of GDP (i.e., actual revenues could be up to 14 percent of GDP higher at the current level of development and structural characteristics).
- Tax gap in peers:
  - about 8 percent of GDP in European AEs and Nordic countries.
  - 7.3 percent in European EMEs.
- Chart indicators (as reported): Estonia tax revenue and tax gap values shown among peers (values preserved in source charts).

### Structural factors and tax base
- Drivers of the divergence:
  - Lower tax-to-GDP levels and structural factors related to untaxed sectors and activities.
  - Relatively higher share of informality than European AEs, especially affecting income tax collection.
  - Certain activities/sectors untaxed or lightly taxed by design: real estate, construction, corporate retained profits, wealth, social services, gambling.
  - Generous minimum thresholds contribute to exclusion from the tax net.
- Revenue administration challenges:
  - Construction and hospitality sectors, partially unofficial salaries (for PIT).
  - Hidden profit allocations: tax avoidance through structuring transactions, transfer pricing, taxation of profits attributed to permanent establishments (PEs) (for CIT).
  - Domestic Profit Tax (DPT) system complicates treatment of corporate restructuring, mergers and divisions, tax exemption rights and compliance assessments, with potential tax revenue and tax gap difficult to estimate.
- Policy direction: strengthen core of the tax system, enhance revenue administration capacity to handle transaction complexities, and broaden the tax base.

### Benchmarking the Estonian tax system against best practices
- Role of tax system: generate necessary revenues as efficiently, equitably, and manageably as possible while minimizing disincentives to work, invest, and save; keeping tax laws simple and enforceable.
- Estonia has the highest ratios of indirect to direct taxes among AEs; the ratio will increase further as VAT rate hikes are phased in and the PIT uniform allowance is launched.
- Evidence summarized:
  - Income taxes generally more harmful for growth than consumption and property taxes (references cited in source).
  - Broadening the VAT base (fewer reduced rates and exemptions) more conducive to higher long-run growth than higher rates alone for OECD countries.
  - Exemptions of intermediate goods are especially inefficient by disrupting VAT input tax crediting.
- Equity trade-offs:
  - Income taxes tend to reduce inequality more than consumption taxes; progressive income tax systems predominant in advanced economies.

### Consumption, digitalization, and wealth/property taxation
- VAT and administration:
  - Estonia’s consumption tax is efficiently collected and aligned with best practices.
  - As of 2023, 99 percent of declarations were submitted digitally.
  - Companies can register online through the Estonian Tax and Customs Board; electronic filing and payment options and automated analytics support compliance risk management.
  - Estonia has abolished the super-reduced VAT rate of 5 percent and now has only two reduced rates.
  - Higher-than-EU-average VAT registration threshold: turnover in calendar year less than €40,000 (i.e., for 60 percent of total Estonian businesses).
  - Exemptions remain more widespread and raise inefficiencies.
- Capital and wealth taxation:
  - Capital income is relatively less taxed in Estonia.
  - Estonia has a higher Gini coefficient for wealth than income.
  - Capital gains from sale/exchange of assets generally taxed on a net basis as part of ordinary income; capital losses deductible only against capital gains.
  - Certain qualifying capital gains exempt (e.g., gain from sale of a personal residence); capital gains by non-residents on sale of shares in resident companies generally not taxable.
  - DPT system leads to systematically higher capitalization and a high level of non-distributed profits.
- Real property taxes:
  - Estonia lacks a property tax based on capital improved value.
  - Steps taken in 2022: update of land values based on an automated mass appraisal.

### Personal Income Tax (PIT): design, distortions, and reform scenarios
- Current PIT design and parameters:
  - Single statutory tax rate of 22 percent on income above the basic allowance.
  - Basic allowance (basic tax exemption) valued annually at €7,848 in 2024.
  - Allowance phased out linearly over the annual income range from €14,400 to €25,200.
  - Phase-out rate around 73 percent; implication example preserved: 72%*22%=15.8%.
  - Resulting marginal rate pattern: an inverted U-shape (“hump”) with marginal rates reaching relatively high levels at average annual income levels, about 35 percent, before declining to 22 percent for higher income levels.
- Distributional and revenue outcomes under Status Quo:
  - PIT revenue at 6.9 percent of GDP, considerably lower than the European AEs average and less than half of Nordic countries.
  - Gini coefficient improves only by 3.2 ppts relative to pre-tax income.
  - Fulltime workers earning a median wage are more likely to face higher marginal tax rates.
  - Flat and lower marginal rate for top income earners reduces redistributive capacity and overall revenue mobilization.
- Announced 2026 reform (as described):
  - Introduces a universal basic exemption of €700/month or €8,400/year.
  - Flat rate increases to 24 percent.
  - Despite a 2 ppts increase in the statutory rate, the new system is estimated to collect 0.5 percent of GDP less than the status quo.
  - Share of people paying the tax is reduced due to the higher basic allowance.
- Revenue-neutral and alternative reform designs (parameters and impacts preserved exactly as reported):
  - Revenue-neutral options to align new system with status quo revenues:
    - (i) increasing the single statutory rate from 22 to 26 percent, or
    - (ii) lowering the universal allowance from €8,400 to €6,750.
    - Redistributive capacity is higher in option (i) and lower in option (ii) relative to the status quo.
  - Two-rate scenario:
    - Rates set at 22 and 30 percent would yield 0.9 ppts of GDP additional revenue relative to the status quo and 1.4 percent relative to the announced reform.
    - The two-rate design shifts the burden slightly to the right of the wage income distribution (from the middle 6-8th deciles to the top 2 deciles).
    - Redistributive capacity would improve.
    - Average rate increases by 2 ppts relative to the status quo, and by 3.5 ppts relative to the announced reform.
    - If combined with a lower standard CIT rate on overall profits (e.g., 15 percent), a two-rate PIT scenario (15 percent in bracket #2, and 29 percent in bracket #3) should be considered to achieve revenue neutrality while preventing arbitrage.
    - Such changes would imply lower rates for the bottom 90 percent of taxpayers.
  - More progressive PIT scenario:
    - Four rates (10-15-25-40) and a lower universal allowance of €6,000.
    - Would bring in 1.4 percent of GDP in extra revenues relative to the status quo and almost 2 ppts relative to the announced reform.
    - Would reach 85 percent of wage earners.
    - Average rate of 21.6 percent.
- Policy implications highlighted:
  - Higher rates for above-median incomes would not only raise revenues but also improve PIT progressivity and efficiency.
  - Consider two-rate or multi-rate structures to improve revenue mobilization and redistribution while protecting lower-income groups.
  - Broaden tax base and strengthen administration to capture untaxed activities and reduce informality.

*Source: IMF staff analysis as presented in the cited chapter.*

### 20.      Considering a slightly more progressive tax structure (e.g., two rates) would bring

### 20.      Considering a slightly more progressive tax structure (e.g., two rates) would bring 

### Personal Income Tax (PIT): main findings and scenarios
- Slightly more progressive tax structure (e.g., two rates) would bring significant advantages in terms of work incentives and revenue mobilization (Text Figure 3).
- Adjusting the marginal rate could improve efficiency and equity while also mobilizing additional revenue.
- Efficiency gains could arise from a reduction of marginal tax rates for median wage workers, incentivizing positive labor supply responses.
- Labor supply elasticities vary across workers and households; non-tax factors (for example, availability of childcare) are important drivers of labor supply decisions.
- Decomposing redistributive capacities shows that the two scenarios (‘Two Rates’ and ‘More Progressive’) achieve higher redistributive capacity with higher average tax rates.

- Key illustrative simulation outputs (as presented in source text):
  - "AverageRedistributiveProgressiveShare Paying  
    RateCapacityCapacityTax  
    Before tax0050.80000"
  - "Status Quo6.90.047.617.93.214.879.2"
  - "Reform6.4-0.547.416.63.417.177.2"
  - "Two-Rates7.80.945.420.25.421.479.2"
  - "More Progressive8.31.444.521.66.322.985.0"
  - "Source: IMF staff calculations."
  - Additional numeric snippets in the table and figure captions as presented:  
    "20.979.26.90.046.218.04.6 ∆PIT/GDP Gini coeff. 6.9 Scenario Text Table 2. Estonia: Comparison of Personal Income Tax Simulations, 2025 (Percent) 13.582.8 PIT/GDP 77.26.9 0.047.917.92.9 0.047.118.03.717.1"

- Qualitative illustration from Text Figure 3:
  - Various PIT design scenarios could correct the marginal tax rate spike and low redistribution of the status quo.
  - Slightly higher average tax rates under more progressive systems generate higher revenue mobilization.
  - Several revenue-neutral options would involve either a lower allowance or a higher rate.
  - Slightly more progressive systems would improve work incentives and revenue mobilization even more.

### Policy implications and considerations for PIT reform
- Revenue-neutral reforms can be implemented by either lowering the allowance or raising the headline rate.
- A two-rate or more progressive schedule can raise redistribution and revenue while improving labor incentives for median wage workers.
- Heterogeneity in labor supply elasticities and non-tax determinants of labor supply (e.g., childcare) should inform policy design and expected behavioral responses.

*Source: IMF staff calculations as presented in the source content.*

### E. Business Income Taxation — objectives and trade-offs
- Rationale for taxing corporate income:
  - Corporations are convenient collection agents, reducing tax administration costs.
  - A source-based CIT allows countries to tax foreign capital owners on capital income earned locally.
- Trade-offs:
  - If CIT falls on the normal return to capital (not only rents), it can adversely impact FDI and distort savings and investment patterns.
  - Taxing rents and fully exempting investments is the justification for the simplified CIT system adopted by Estonia in the early 2000s.

### Classification of CIT bases and implications
- Full return to equity:
  - Standard CIT levies tax on corporate profit as defined by accounting rules; capital expenditure is not deductible but depreciation is based on a schedule.
  - Interest is deductible, creating a “debt bias” (different treatment of debt- and equity-financed investment).
  - Standard CIT systems tax the normal return to capital if investment is equity-financed.
- Full return to capital (CBIT):
  - Disallows interest deductibility, bringing the entire return on capital into the corporate tax base.
  - Larger tax base allows for a lower rate without revenue loss; PIT treatment of interest, dividends, and capital gains should be aligned.
  - CBIT was developed by the US Treasury Department in 1992 and proposed in Sweden in 2014 (proposal not implemented).
- Economic rents / cash flow corporate taxes:
  - Define CIT base such that only economic rents (excess profits over the required normal return) are taxed, avoiding efficiency losses.
  - Cash-flow taxes are based on Sources & Uses of Funds rather than Profit & Loss.
  - Evidence (IMF’s multi-region GIMF model analysis by Carton et al., 2019) suggests replacing a standard CIT with cash-flow taxes boosts output in the reforming country and yields positive long-run spillovers.
  - Estonia’s CIT is close to a share- or ‘S-base’ cash flow tax on pure economic rents, with caveats (not based on net profit distributions).
  - Structural reforms to the CIT base carry significant revenue risks during transition, as statutory rates would need to rise to keep revenues constant with a narrower base.

### Designs of cash flow corporate taxes (three approaches)
- R-base cash flow tax:
  - Full expensing of capital expenditures; disallow interest and dividend deductions.
  - Tax is a levy on firm’s net cash flows from real activities; investment expensing implies rent taxation.
- R+F-base cash flow tax:
  - Full expensing of capital expenditures while retaining interest deductibility and adding net debt to the tax base.
- S-base cash flow tax:
  - Tax on net distributions of corporations (dividends plus share buybacks minus capital increases).

### Allowance for Corporate Equity (ACE)
- ACE maintains interest deductibility and adds a notional interest deduction for equity capital to reduce debt bias.
- Elaboration: IFS Capital Taxes Group in the UK (Institute for Fiscal Studies, 1991).
- Belgium and Italy have recently implemented variants of the ACE.

### Estonia and Latvia: Distributed Profit Taxes (DPTs)
- Estonia and Latvia operate DPTs since 2000 and 2018, respectively.
- DPT mechanics:
  - Upon declaration of a dividend, the DPT is levied at a rate of 22/78 of the net amount of the profit distribution, corresponding to 28 percent on the gross amount distributed.
  - Undistributed corporate profits are tax exempt (covers active and passive income and capital gains from sale of shares or securities).
  - In Latvia, immovable property owned by non-residents is excluded from this exemption; in Estonia it is not.
  - Resident companies and PEs of non-resident companies can benefit from the exemption.
- Recent changes and compliance signals:
  - To raise more revenue from CIT, the reduced 14 percent CIT rate on regular profit distributions (also taxed at 7 percent with a withholding tax) has been abolished from 2025.
  - The advance CIT for credit institutions (‘bank levy’) has been increased from 14 percent to 18 percent.
  - Compliance under DPT is hard to assess; observed spikes in revenues before tax hike announcements point to significant untapped revenue.
- Footnote detail on gross-up and rates:
  - "In Estonia, CIT is computed based on the grossed-up amount. Since January 2025, the tax rate on the net profit distribution is 22/78, i.e., about 28.2%. The amount of tax is equal to (Net Dividend x 22)/78. From January 2026 the nominal rate goes up to 24%. The fractional rate will be 24/76, bringing the gross-up rate up to 31.6%."

### Assessment of the DPT and its implications
- DPT advantages:
  - Likely less distortive than a standard CIT because it falls mostly on rents.
  - Has provided impulse to corporate investment and addresses neutrality between debt and equity.
  - Simplicity: no need to track revenues and deductible expenses; works well in a digitalized environment with high compliance and strong institutions.
- DPT drawbacks and risks:
  - Potential for sizeable loss in tax revenues if statutory rates are not increased to offset a narrower base.
  - Arbitrage risks: incorporation by high-wealth individuals or service sector employees to collect dividends instead of PIT-subject remuneration; transactions between corporates; shareholder deferral of profit distribution.
  - Baltic DPTs tax gross distributions (distributed dividends and share buybacks without accounting for capital increases), capturing more than economic rents and thus not fully aligning with canonical S-base cash flow taxes.

### Policy options to strengthen CIT revenue mobilization (as presented)
- Improving current design of the DPT, including a higher statutory rate.
  - Move the tax base closer to economic rents by shifting from gross to net distributions.
  - Caveat: higher rates may increase incentives for tax arbitrage, avoidance, and evasion; limited experience with DPTs at higher statutory rates implies considerable risks.
- Moving towards a standard CIT through well-sequenced incremental reform.
  - Standard CIT could credibly increase the tax-to-GDP ratio (already applies to banks in Estonia).
  - Mitigation measures to limit adverse investment impacts: Allowance for Corporate Equity (ACE) or immediate expensing for specific asset types.
  - ACE could generate the same revenue at a lower rate and be more robust against avoidance, but would increase complexity and require enhanced tax administration capacity.
- Exploring other rent taxes on efficiency grounds (cash flow taxes on R- or R+F-base).
  - Estonia’s current CIT design, low CIT revenue levels, and experience with DPT may make it a more amendable starting point for reform.
  - A switch to other rent taxes carries considerable risks because such designs have rarely been implemented broadly despite theoretical appeal.

*Source: Excerpts from the IMF staff text provided in the source content.*

### 32.      VAT C-efficiency has been high in Estonia since 2005. This reflects strong revenue

### VAT C-efficiency has been high in Estonia since 2005

### VAT performance and gaps
- VAT C-efficiency has been high in Estonia since 2005, reflecting:
  - broad coverage with fewer reduced rates compared to EU/EA averages (examples listed: pharmaceuticals, hotel accommodation, periodicals, books);
  - high reliance on automated, transparent, and efficient tax administration processes enabling very quick VAT returns;
  - a low compliance gap, despite some variability over time.
- Main tax-administration risks identified:
  - turnover concealment;
  - fraudulent tax returns and input VAT;
  - real estate-related tax frauds.
- European Commission’s 2024 VAT Gap Report estimates for Estonia (based on 2022 data):
  - the VAT policy gap is about one third of the theoretically collectable VAT revenue;
  - closing the actionable exemption gap would yield 1.1 percent of GDP;
  - closing the rate gap would yield 0.3 percent of GDP;
  - strengthening tax compliance could generate up to 0.4 percent of GDP.
- Higher statutory VAT rates are associated with greater incentives for noncompliance:
  - in the EU, on average noncompliance is estimated to worsen by about 0.5 ppts of theoretical revenues for each 1 ppts VAT rate increase (EC, 2024).
- Administrative measures proposed to improve VAT collection:
  - require that all invoices be declared in the VAT return annex (currently only transactions above $1,000 are included);
  - expand e-invoice for all VAT payers (current coverage is 7 percent).

### Policy trade-offs and recommendations on VAT design
- Resist proliferation of reduced rates and exemptions as statutory rates rise.
  - Example proposal and fiscal cost: tax food at a lower 5 percent VAT rate (estimated cost of around €400 million) or introduce exemptions to improve distributional effects — described as a costly approach to pursue equity.
- Equity-focused alternatives recommended:
  - income-based targeted transfers or other forms of cash support to the neediest.
- Preferred benchmark:
  - a uniform, broad-based consumption tax, with very few, well-specified exemptions and limited rate differentiation.
- Rationale:
  - increasing VAT revenue through a broader base and fewer reduced rates and exemptions is presented as more growth friendly relative to standard rate increases;
  - exemptions create cascading effects through denied input tax credits, distortions in production decisions, increased administrative burden, and lowered compliance.
- Fiscal-administration measures reiterated:
  - declare all invoices in VAT return annex;
  - expand e-invoicing beyond current 7 percent coverage.

### Property taxes — current status and potential
- Wealth and distributional context:
  - in Estonia the top 10 percent of households hold about 1/3 and the bottom 40 percent hold less than 5 percent of total wealth (OECD comparison referenced).
- Current Estonian property taxation features:
  - taxes on property in Estonia are limited to a land tax; residential land is exempt up to a generous threshold;
  - taxes on property collect less than 0.2 percent of GDP and share has been on a downward trend in the last decade;
  - no assessment of buildings’ value for taxation purposes and there is no real property transfer tax.
- Land tax specifics and exemptions:
  - the Land Tax Act exempts residential land up to 1,500 m2 in cities and other densely populated areas and up to 2 hectares elsewhere;
  - local authorities may additionally exempt pensioners and other categories;
  - local government discretion includes exemptions up to €1,000 for residential land and reduced rates to attract investment.
- Land revaluation and rates:
  - no property tax revaluation for over 20 years prior to the 2022 comprehensive land revaluation;
  - with the 2022 revaluation implemented from 2024, property values have increased 8 times, requiring a lowering of the maximum tax rate;
  - upper bounds of the new tax rate for 2025 range from 0.1 to 1 percent of market value for residential and agricultural land, from 0.1 to 0.5 percent for profit-yielding land, and from 0.1 to 2 percent for land with other intended use;
  - municipalities may fix rates and exemptions by October 1st annually;
  - the 2022 values are applied progressively; in 2025 the cap on the value increase was increased to 50 percent/€20 from 10 percent/€5 previously;
  - from 2026 the homeowner’s exemption will be decided by local governments and capped at €1,000;
  - homeowners remain exempt from land tax for up to 1,500 m² of domestic land;
  - example: in 2025 the City of Tallinn exempted 172,500 homeowners from paying taxes on domestic land, accounting for about 80 percent of total taxpayers.
- Revenue potential from a new property tax:
  - Estonia has about 266,475 residential buildings (2021 census); about ¾ are one-family dwellings, 18 percent are blocks of flats, the remainder are semi-detached houses and non-residential buildings with at least one living space;
  - about ¾ of dwellings are occupied and ¼ are without permanent residents, and the share of unoccupied dwellings has been rising;
  - nearly 39 million square meters of residential space, an average of 30.1 m2 per person;
  - illustrative scenario: assuming a tax based on property value with differentiation between primary and secondary residence (e.g., 0.05 percent and 0.1 percent, respectively), up to 5 times higher depending on the municipality, and an exemption threshold of 10 percent, a new tax could generate as much as 1.7 percent of GDP.
- Administrative and technical constraints:
  - no fully-fledged fiscal cadaster currently; underlying registries are not synchronized (land registry records legal ownership and areas; building register contains buildings and permits; transaction data forwarded to the Land Board; population registry kept at the Ministry of Interior);
  - no property tax revaluation for over 20 years prior to 2022 revaluation (amendments now provide for periodic revaluations every four years starting in 2022).
- Recommendations for property tax reform:
  - near term: encourage municipalities to limit residential primary owner exemption; tighten mandatory residence registration at local level; synchronize population registry with land registry, real estate transactions, and income statements;
  - design broad tax base with means-tested compensation mechanisms for low-income households;
  - prepare for a property registry of house valuations (‘fiscal cadaster’) using digital mapping technologies (drones, satellite imagery) to create property maps and identify parcels/buildings;
  - prepare measures to shield low-income households (tax deferment schemes for “asset-rich but cash-poor” taxpayers; ensure arrears become due in full when property is alienated);
  - incentivize local governments for maximum property tax effort (e.g., reduce allocations of shared taxes over time or design inter-governmental grants to reward greater municipal tax collection);
  - allow local flexibility to select a property tax rate within a centrally determined narrow range;
  - medium term: consider a full-fledged value-based property tax following technical and legal preparations, applying a flat rate to a base with a minimum of exemptions and uniform treatment of business and residential property; consider provisions for owners of multiple properties.

### Conclusion and consolidated policy options
- Context:
  - Estonia’s tax mix is reliant on consumption taxes—especially VAT;
  - Estonia exhibits low tax effort relative to peers alongside imminent and longer-term spending pressures.
- Staff recommendations (options highlighted in the SIP):
  - address the PIT revenue shortfall by considering revenue neutral options, i.e., calibrate the basic allowance, the tax rates, and/or the tax brackets subject to the intended degree of progressivity;
  - improve the capacity of the tax administration to analyze income statements and explore alternative CIT regimes that would preserve Estonia’s tax competitiveness while reaching a broader tax base;
  - streamline remaining VAT exemptions to broaden the tax base;
  - limit exemptions on residential land and take steps to introduce a modern tax on immovable property by developing a fiscal cadaster.

*Source: Excerpts from the IMF Selected Issues Paper chapter on taxation for the Republic of Estonia (pages 32–44).*

### 1.      The Baltic economies have faced remarkable challenges in recent years. Russia’s war in

### The Baltic economies have faced remarkable challenges in recent years

### Overview and main message
- Russia’s war in Ukraine led to supply disruptions and a sharp increase in input costs for firms; input costs have remained high for the region and, in conjunction with slow productivity growth, have led to erosion of competitiveness (Armendariz and others 2024).
- Improving productivity growth is critical to restore competitiveness.
- Labor productivity growth in the Baltic economies has decelerated over the past two decades.
- Diminishing allocative efficiency contributed to the decline in labor productivity growth over time; the net effect of firm entry and exit has been generally limited.

### Evidence on misallocation and dispersion
- Previous IMF studies and related literature find resource misallocation hindered productivity growth in the last two decades (Figure 1).
- Studies find rising dispersion in the marginal revenue product of capital, especially for Estonia and Lithuania, indicating capital misallocation (Figure 2).

### Literature context and policy-relevant findings
- Olley and Pakes (1996): surviving firms are systematically more productive than exiting firms; productivity growth driven by within-firm improvements and reallocation toward more productive firms.
- Hsieh and Klenow (2009): reducing misallocation and equalizing marginal products of capital and labor across firms could dramatically boost productivity.
- Melitz and Polanec (2015): market share reallocation among surviving firms accounted for about half of productivity growth; firm entry and exit contributed about 30 to 40 percent of productivity growth (Slovenia, 1995-2000).
- Decker and others (2017, 2020): declining business dynamism and weaker responsiveness of employment to productivity have negatively contributed to aggregate productivity growth in the United States; 30 to 50 percent of the post-2000 US productivity slowdown is attributable to declining dynamism.
- Masso and Tiwari (2021) (Estonia, 2000-2017): new firms exhibit higher productivity gains from R&D and innovation compared to incumbents in high-tech sectors; entrants face high exit rates but survivors rapidly close the productivity gap; recommend targeted R&D support and access to financing for scaling up.
- Ascari and others (2023): pandemic crisis had sector-specific “cleansing” effects; supportive fiscal policies such as wage subsidies may delay necessary reallocation and preserve unviable firms; recommend targeted support for displaced workers and incentives for high-growth sectors and avoiding prolonged subsidies to unviable firms.
- Merikull and Paulus (2024) (Estonia, 2004-2020): generous job retention schemes slowed within-industry reallocation toward more productive firms during COVID-19; positive employment effects offset negative productivity impact but net gains from the scheme are limited.

### Decomposition framework (following Decker and others 2017)
- Labor productivity growth decomposed into four components:
  1. Sector-level average productivity growth for all continuing firms.
  2. Allocative efficiency: covariance of firm-level labor productivity and the share of industry employment (measure of allocative efficiency).
  3. Contribution of entrants: employment share of entrants × (productivity of entrants − productivity of continuers).
  4. Contribution of exits: employment share of exiters × (productivity of continuers − productivity of exiters).
- Notation preserved from source: ∆퐺퐺푖푖 = ∆푝푝̅푖푖,푐푐 + ∆푐푐푐푐푐푐푐푐(휃휃표표,푝푝표표) + 휃휃퐸퐸2(퐺퐺퐸퐸2 − 퐺퐺퐶퐶2) + 휃휃푋푋1(퐺퐺퐶퐶1 − 퐺퐺푋푋1).
- Calculation: decomposition for each industry-year, aggregate annual components at country level using sector-level employment shares in the initial year.
- Interpretation: first term = within-firm average productivity growth; second term = change in allocative efficiency among continuing firms; remaining terms = aggregate contribution of net entry.

### Results — Estonia (statistical register data and Orbis robustness)
- Aggregate-level allocative efficiency contribution declined during 2006-2022 and turned negative after 2016.
- Firms with higher productivity were growing in employment during 2001-2015; allocative efficiency worsened after 2006.
- Contribution of firm entry is negative (entrants have lower labor productivity on average than incumbents).
- Contribution of firm exit to labor productivity growth is positive throughout the sample period and increasingly so after 2015.
- Net entry contribution to labor productivity growth has been very small.
- Findings are generally consistent across industries (manufacturing, ICT, professional services) and with Orbis data for Estonia.
- Note: value added proxied by firms’ turnover per employee in Estonia’s statistical register data.

Key time bands and observations for Estonia (preserved exactly as in figures):
- 2001-2005; 2006-2010; 2011-2015; 2016-2019; 2020-2022.

### Results — Latvia (administrative data and Orbis robustness)
- Allocative efficiency contribution turned negative during 2016-20.
- Contribution of firm entry to labor productivity growth is negative, though it narrows over time.
- Contribution of firm exit is positive and outweighs that of firm entry for 2016-19.
- Industry-level results consistent with aggregate: allocative efficiency negative for agriculture, manufacturing, construction, wholesale, and retail trade; firm entry negative throughout; firm exit positive and more than compensates entry during 2016-19.

Key time bands and observations for Latvia (preserved exactly as in figures):
- 2011-2015; 2016-2019; 2020-2021.

### Results — Lithuania (administrative data)
- Contribution by allocative efficiency to labor productivity growth declined and turned negative during 2011-15.
- Contribution by firm exit failed to compensate the negative contribution by firm entry during 2001-15.
- Across the three Baltic economies:
  - Industry-level average labor productivity growth plays an important role in aggregate growth.
  - Allocative efficiency and firm dynamics also matter.
  - Overall, allocative efficiency contribution to productivity growth has declined over time.
  - Net contribution by firm dynamics is close to zero.

Key time bands and observations for Lithuania (preserved exactly as in figures):
- 2001-2005; 2006-2010; 2011-2015; 2016-2019; 2020-2022.

### Entrant firm dynamics and productivity evolution
- Entrant firms tend to be smaller, less experienced, and have less access to resources and networks than incumbents.
- Evidence from the 2010 entrant cohort (Estonia, Latvia, Lithuania):
  - The distribution of labor productivity is skewed toward the low end at entry but shifts toward the center over time.
  - Within ten years, average labor productivity increased significantly and became more evenly distributed.
  - Entrant firms’ productivity levels improve over time, but average entrant productivity may remain lower than incumbents for longer due to barriers to growth (e.g., limited access to finance because of lack of collateral; limited access to skilled labor).

### Policy implications and recommendations (as discussed in the source)
- Improve allocative efficiency to restore productivity and competitiveness (implied by findings and literature cited).
- Targeted R&D support for startups in high-tech sectors and ensure new firms can access financing for scaling up (Masso and Tiwari 2021).
- During crises, provide targeted support for displaced workers and incentives for high-growth sectors; avoid prolonged subsidies to unviable firms to prevent productivity stagnation (Ascari and others 2023).
- Design job retention schemes carefully: while preserving employment, generous schemes can slow within-industry reallocation toward more productive firms and limit net productivity gains (Merikull and Paulus 2024).

*Source: https://www.imf.org/-/media/files/publications/cr/2025/english/1estea2025002-print-pdf.pdf*

### 12.      We find that the employment share of micro firms increased over time in Estonia and

### 1estea2025002-print-pdf - 12.      We find that the employment share of micro firms increased over time in Estonia and

### Key empirical findings on firm dynamics and productivity
- The employment share of micro firms increased over time in Estonia and the other two Baltic economies over the past few decades.
- Labor productivity growth slowed down during the same period; with labor trapped in stagnant micro firms, aggregate growth may have also slowed.
- Fast-growing young firms take up a bigger share of employment than slow-growing young firms; in the case of Estonia this share is "2-3 percent".
- The footprint of fast-growing young firms in the aggregate economy remains small compared to more advanced economies; the corresponding employment share in the United States is "about 6 percent".
- Note from data construction: Firms with only one employees are dropped from the sample.

### Policy implications — summary
- Policy makers need to address constraints faced by young firms to promote productivity growth.
- Firm-level data on productivity can help distinguish temporary low productivity of startups from persistently low productivity of nonviable firms.
- Government programs should target innovative young firms which support long-term economic growth, while implementing measures to improve the allocation of capital and labor.

### Supporting high-quality entry (targeted measures)
- Findings:
  - Firm entry rates in Estonia are higher than the EU average, while entry rates for firms with more than 10 employees are lower.
- Recommended policies:
  - Targeted subsidies funding R&D intensive startups with high growth potential to foster productive new firms.
  - Support for learning processes: skilled workforce training and programs facilitating the adoption of new technologies.
  - Make higher-education programs more relevant to market demand (focus on STEM); consider charging a reasonable tuition fee and provide scholarships for students from low-income families and in STEM programs.

### Facilitating efficient exit (insolvency and bailout policy)
- Recommendations:
  - Streamline insolvency procedures to allow unviable firms to exit quickly and free up resources.
  - Limit subsidies or bailouts for firms with no viable path to profitability, while distinguishing unviable firms from startups with equity gaps in early growth due to R&D or other investments.
  - Target support toward innovative firms and focus on adoption of new technologies in traditional sectors.
- Evidence and specific constraints:
  - Asset recovery rates during insolvency processes are relatively low for Estonia and the Baltic region compared to other OECD economies.
  - Lenders often require substantial collateral, limiting access to finance for SMEs and startups that lack sufficient assets.
  - Policy option: introduce more standardized valuation of collateral and allow use of movable properties as collateral.

### Reducing regulatory burdens (product market regulation)
- Estonia features more flexibility than the OECD average in economy-wide product market regulation indicators, reflecting a relatively competition-friendly regulatory framework.
- Areas for improvement:
  - Reduce sector-specific barriers to entry in road and air transport, and mobile e-communications.
  - Simplify licensing processes for new transport operators and reduce regulatory burdens that disproportionately affect small operators.
  - For mobile e-communications: expand spectrum availability and encourage existing operators to share infrastructure to allow new operators to enter the market.
  - Simplify licensing processes and strengthen lobbying transparency.

### Improving allocative efficiency of capital and labor (finance and talent)
- Evidence of constraints:
  - About "70 percent" of firms surveyed by the 2024 European Investment Bank reported dissatisfaction with external finance received in the last financial year.
  - The 2023 survey reports that "30 percent" of firms report such dissatisfaction.
- Policy recommendations:
  - Provide targeted grants or subsidies to innovative firms expected to become more productive than incumbents or for activities that enhance productivity, such as R&D.
  - Endorse EU-wide reforms to deepen and integrate capital markets to give innovative firms access to diversified funding sources.
  - Domestic capital market reforms to ease financing for firms with high intangible-asset shares and limited collateral.
  - Expand availability of venture capital and equity financing, including by facilitating investments by second-pillar pension funds.
  - Consider measures to accelerate integration of high-skill migrants and enhance active labor market policies.
  - Make higher-education more aligned with market demand and STEM needs.

### Summary conclusions and priorities
- Policies should aim to:
  - Lower barriers to scale-up for high-potential startups and allow more competition in product markets.
  - Speed up the learning curve for young firms and facilitate exit of unviable firms under an efficient framework.
  - Reduce distortions in capital and labor markets to allocate resources toward high-productivity firms.
  - Ensure productive new firms can access financing for scaling up, including via venture capital or public-private partnerships.
  - Support deepening of the EU single market to offer more opportunities for small companies to scale up and become more productive, including through leveraging the Savings and Investment Union.
- Further investigation: explore specific case studies of successful high-quality entrants in Estonia and assess the effectiveness of different policies aimed at promoting competition.

*Source: REPUBLIC OF ESTONIA — INTERNATIONAL MONETARY FUND (excerpted content).*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2025/english/1estea2025002-print-pdf.pdf_