## Box 1. Investment Component of the Recovery, Transformation and Resilience Plan

## Source details

**Canonical URL:** [Box 1. Investment Component of the Recovery, Transformation and Resilience Plan](https://www.imf.org/-/media/files/publications/selected-issues-papers/2025/english/sipea2025076.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/selected-issues-papers/2025/english/sipea2025076.pdf.md)
- [Structured JSON version](/-/media/files/publications/selected-issues-papers/2025/english/sipea2025076.pdf.json)

---

### Overview of the RRF and Spain’s RTRP
- Envelope of €163 billion—representing about 10 percent of GDP or over half of Spain’s yearly aggregate investment—split roughly evenly between the grant and loan components.
- Since 2021, about 60 percent of the allocated RRF funds have been disbursed.
- A fifth payment request—comprising grants and, for the first time, loans—is currently under evaluation by the European Commission.
- All milestones and targets of the plan associated with the grant component have to be completed by August 2026.
- Prepared by Ana Lariau (EUR).

### RRF execution and investment disbursement (key operational figures)
- By end-March 2025:
  - Amount called reached almost €77 billion (96 percent of total grants available).
  - About 65 percent of this amount (almost €50 billion) having already been awarded to final private sector beneficiaries.
- Heterogeneity across government levels:
  - Calls award rates for regional and local governments range between 40 and 75 percent.
- Call award rates and assignment:
  - Similar remaining execution gaps between central government and regional/local governments, but significant variation across regions.

### Recent dynamics: key facts on investment performance
- Plan implementation balance:
  - Almost 70 percent of the planned reforms have been completed.
  - Only 12 percent of the investment milestones and targets have been fulfilled.
- Aggregate investment:
  - Despite RRF execution progress, aggregate investment has remained broadly stable as a percentage of GDP.
- Sectoral performance (by end-2024 / recent periods):
  - Investment in transport equipment: nearly 20 percent below its pre-pandemic level by end-2024.
  - Other construction (non-residential construction): remains below its 2019 level.
  - Housing investment and investment in other machinery and equipment: returned to around their pre-pandemic levels but with a modest rebound.
- Possible reasons for muted aggregate investment response to RRF awards:
  - Backloading of investment milestones and targets implying backloading of actual investment effects.
  - Some RRF-funded investments may be classified as intermediate consumption rather than gross fixed capital formation (GFCF) in the National Accounts.
  - Multi-year project implementation smoothing effects (e.g., transportation projects).
  - Delays between approval of calls, distribution of funds to firms, and actual investment execution.
  - Offsetting declines in other investment components in the National Accounts.
- Illustration (Awarded Calls vs. GFCF-to-GDP for 2021–2025):
  - Awarded Calls values: 0.4, 1.2, 2.2, 3.0, 3.0 for 2021–2025 respectively.
  - GFCF-to-GDP values: 20.2, 20.4, 19.7, 19.5, 20.0 for 2021–2025 respectively.

### Comparative performance with peers
- Spain’s GFCF-to-GDP ratio has steadily declined since the pandemic, while:
  - The euro area ratio has remained broadly stable.
  - The 4-country peer group (Germany, France, Italy, and Portugal) has seen a slight rise.
- Weakness is broad-based but most pronounced in:
  - Transport equipment.
  - Other construction.
- Sector-specific challenges for transport investment may include:
  - High costs associated with fleet renewal in the green transition.
  - Limited access to charging or refueling infrastructure.
  - Potential regulatory or licensing hurdles increasing uncertainty or compliance costs.

### Data used for analysis
- Macroeconomic dataset:
  - Spain, 1995 to 2024, quarterly frequency.
  - Includes GDP, GFCF and subcomponents (machinery and equipment: transport and other equipment; construction: housing and other construction), GFCF deflators, 10-year government bond yield.
  - All variables in constant 2015 prices, sourced from INE.
- Firm-level dataset:
  - Covers 5.1 million firms in Spain, Germany, France, and Italy for 2003 to 2022.
  - Sourced from Orbis Bureau Van Dijk (BvD), compiled by IMF’s Research Department.
  - Analysis focuses on ten economic sectors covering approximately 80 percent of total value added and 70 percent of employment in Spain.

### Empirical strategy (macro and micro)
- Macro: Error Correction Model (ECM)
  - Long-run equation consistent with neoclassical investment theory:
    - ln(i_t) = φ_0 + φ_1 ln(y_t) + φ_2 ucc_t + ε_t
    - ucc_t = (r_t + δ) / P_k with δ set to zero for simplicity.
  - Short-run dynamics include lags of Δln y_t, ucc_t, Δln i_t, and an error correction term ECT_t−1.
  - Model estimated using quarterly data from 1995 to 2020 and simulated out of sample to assess post-pandemic recovery dynamics.
  - Separate estimations for total capital formation and for subcomponents: machinery and equipment, other construction.
  - Model augmented in robustness checks with an Economic Policy Uncertainty (EPU) index for Spain.
- Micro: Firm-level panel regression
  - Baseline specification:
    - NetInv_cji,t = β_0 + β_1 Age_cji,t−1 + β_2 ln(Size_cji,t−1) + β_3 Δln(Sales_cji,t−1) + β_4 Profit_cji,t−1 + β_5 Markup_cji,t−1 + β_6 Leverage_cji,t−1 + firm_i + cty_c × year_t × Nace4_j + ε_cji,t
  - NetInv defined as change in fixed assets relative to previous year’s fixed assets.
  - Controls: firm fixed effects; fully interacted country-year-sector (Nace4) fixed effects.
  - Regression estimated at firm level using annual data, weighted by firm size.
  - Augmented specifications interact age, size, profitability, and leverage with country and post-pandemic dummies; subsample analyses for small firms and middle-aged firms.

### Main results — Macroeconomic analysis
- Aggregate investment vs. fundamentals:
  - As of 2024Q4, the cumulative shortfall in capital formation since the pandemic is estimated at around 15 percent.
  - Investment has gradually moved closer to model-implied fundamentals during 2023–24.
  - Error correction term: if investment is 10 percent below its long-run equilibrium, it would rise by approximately 0.29 percent in the following quarter.
- Component breakdown (as of 2024Q4):
  - Machinery and equipment: cumulative shortfall of approximately 56 percent vis-à-vis model-implied level since the start of the pandemic.
  - Other construction: cumulative shortfall of 16 percent.
- Model fit and interpretation:
  - ECM exhibits strong explanatory power; coefficients statistically significant and consistent with theoretical priors.
  - Model provides a better statistical fit for machinery and equipment than for other construction.
- Role of Economic Policy Uncertainty (EPU):
  - EPU spiked sharply during the pandemic and remains above pre-pandemic levels.
  - Incorporating EPU into the ECM:
    - EPU has a statistically significant long-run impact on total capital formation and on subcategories (other construction and machinery and equipment).
    - In the short run, EPU significantly affects total capital formation and machinery and equipment, with timing of effects varying.
    - Adding EPU reduces the cumulative gap between actual and predicted investment by 2 percentage points overall and by 14 percentage points for machinery and equipment.

### Firm-level drivers of net investment
- Profitability:
  - Lagged profit coefficient is statistically significant across all specifications and robust to changes in fixed effects.
  - L.Profit: 0.0429*** (0.00306); 0.0434*** (0.00298); 0.0474*** (0.00407) in baseline columns.
- Firm size:
  - Larger firms invest less relative to their size.
  - L.lnFixed Assets: -1.024*** (0.0577); -1.022*** (0.0580); -1.020*** (0.0628).
- Leverage and markups:
  - Higher leverage associated with lower investment on average but coefficients small and statistically insignificant.
  - L.Leverage rate: -0.00766 (0.00764); -0.00760 (0.00768); -0.00599 (0.01000).
  - Markups negatively associated with investment; L.Markup: -0.00692*** (0.00201); -0.00709*** (0.00211); -0.00838 (0.00514).
- Other baseline coefficients and fit:
  - L.Age: 0.0113 (0.0136); 0.0106 (0.0135); 0.0101 (0.0179).
  - Constant: 1.740*** (0.0775); 1.738*** (0.0781); 1.735*** (0.0822).
  - Observations: 7,713,968.00; 7,716,057.00; 7,716,330.00.
  - R^2: 0.470; 0.459; 0.442.

### Heterogeneity: small, young, and middle-aged firms
- Small firms (bottom 25th percentile of size distribution):
  - More profitable small firms generally invest more, but this relationship is significantly weaker in Spain.
  - Leverage is more positively associated with investment in small firms in Spain compared to other countries.
- Young- and middle-aged firms (≤ 10 years):
  - Profitability leads to more investment generally, but this relationship is significantly weaker for Spanish firms in the post-pandemic period.
  - More leveraged young firms invest less; the negative relationship is more pronounced for Spanish firms and even more so after the pandemic.
- Interpretation:
  - Young firms differ from small firms; credit constraints play a stronger negative role for young firms in Spain, especially post-pandemic.

### Robustness checks
- Macro-level:
  - ECM findings remain broadly unchanged when the long-term coefficient on GDP is restricted to unity.
  - Baseline macro results hold when additional controls are included (consumer confidence, term premia, financial constraints, terms of trade).
  - Annex III. Table 1 (first-stage ECM highlights) reports:
    - Ln(Y) coefficients: 0.595*** (0.107); -0.092 (0.183); 0.737*** (0.071) in columns 1–3.
    - EPU (quartiles): -0.086*** (0.011); -0.149*** (0.020); -0.040*** (0.007) in columns 4–6.
    - Observations: 100; 100; 100; 969; 696; 696 (columns 1–6).
    - R-squared: 0.601; 0.034; 0.866; 0.707; 0.388; 0.863 (columns 1–6).
  - Annex III. Table 2 (second-stage ECM highlights) reports:
    - Ln(Y) = D: 1.119*** (0.066); 0.919*** (0.108); 1.825*** (0.115) in columns 1–3.
    - UCC (M&E) = D: -0.013*** (0.004); -0.014*** (0.004).
    - EPU (quartile) = D: -0.005* (0.003); -0.004 (0.005); -0.005 (0.005).
    - EC = L: -0.029* (0.015); -0.012 (0.013); -0.142*** (0.049).
    - Observations: 989; 898; 989; 949; 494; 9494 (as reported across columns).
    - R-squared: 0.834; 0.556; 0.832; 0.845; 0.569; 0.855.
- Firm-level:
  - Baseline findings robust to alternative specifications with different combinations of fixed effects (Annex Table III.3).
  - Results robust across different combinations of explanatory variables.

### Conclusion: interpretation of findings
- Spain’s investment post-pandemic recovery has been weak relative to fundamentals and euro area peers; the shortfall spans all major components of investment, with transport equipment particularly weak.
- Both macro- and firm-level evidence point to elevated uncertainty — including domestic policy uncertainty — as a drag on the recovery.
- Investment among small and young- to middle-aged Spanish firms has become less responsive to profitability, particularly during the post-pandemic period.
- Leverage has been a stronger drag on investment for young- and middle-aged Spanish firms, suggesting greater difficulties accessing external finance during the pandemic.
- The broad improvement in financial conditions since the start of the ECB’s easing cycle may disproportionately benefit Spanish firms’ investment going forward, but persistent domestic and global uncertainty could continue to weigh on the investment outlook.

### Policy considerations and recommendations
- Reduce policy uncertainty through clearer, more predictable tax, spending, and regulatory frameworks to restore firms’ willingness to invest.
- Ease financing constraints for young high-growth firms by strengthening access to credit and equity finance, including promoting the EU’s Capital Markets Union and targeted support as needed.
- Address structural barriers in specific sectors (e.g., construction); in particular:
  - Advance amendments to the Land Law.
  - Streamline construction permitting processes to unlock stalled projects and support investment in infrastructure and housing.
- Ensure full execution of planned NGEU investment projects to continue supporting investment in Spain relative to other euro area peers.

### Key statistics and exact numerical findings (selected)
- Small firms defined as bottom 25th percentile of size distribution.
- Annex II. Table 2, Spain: Summary Statistics
  - Net investment: Mean 0.15; Standard deviation 0.57; 1st Percentile -0.47; 99th Percentile 2.89
  - Age: Mean 3.89; Standard deviation 1.02; 1st Percentile 2.00; 99th Percentile 6.00
  - Ln(Fixed Assets): Mean 11.81; Standard deviation 1.72; 1st Percentile 7.69; 99th Percentile 15.39
  - Sales growth: Mean 1.82; Standard deviation 24.09; 1st Percentile -68.84; 99th Percentile 72.15
  - Profit: Mean 0.11; Standard deviation 0.12; 1st Percentile -0.15; 99th Percentile 0.52
  - Markup: Mean 1.27; Standard deviation 0.46; 1st Percentile 0.86; 99th Percentile 3.27
- Annex III. Table 3 (baseline firm-level coefficients, standardized variables)
  - L.Age: 0.0113 (0.0136); 0.0106 (0.0135); 0.0101 (0.0179)
  - L.lnFixed Assets: -1.024*** (0.0577); -1.022*** (0.0580); -1.020*** (0.0628)
  - L.dlnSales: 0.00215* (0.00122); 0.00181 (0.00124); 0.000976 (0.00165)
  - L.Profit: 0.0429*** (0.00306); 0.0434*** (0.00298); 0.0474*** (0.00407)
  - L.Markup: -0.00692*** (0.00201); -0.00709*** (0.00211); -0.00838 (0.00514)
  - L.Leverage rate: -0.00766 (0.00764); -0.00760 (0.00768); -0.00599 (0.01000)
  - Constant: 1.740*** (0.0775); 1.738*** (0.0781); 1.735*** (0.0822)
  - Observations: 7,713,968.00; 7,716,057.00; 7,716,330.00
  - R^2: 0.470; 0.459; 0.442

*Source: IMF staff Box 1. Investment Component of the Recovery, Transformation and Resilience Plan (sipea2025076).*

### Box 1. Investment Component of the Recovery, Transformation and Resilience Plan

### Box 1. Investment Component of the Recovery, Transformation and Resilience Plan

### Overview of the RRF and Spain’s RTRP
- Envelope of €163 billion—representing about 10 percent of GDP or over half of Spain’s yearly aggregate investment—split roughly evenly between the grant and loan components.
- Since 2021, about 60 percent of the allocated RRF funds have been disbursed.
- A fifth payment request—comprising grants and, for the first time, loans—is currently under evaluation by the European Commission.
- All milestones and targets of the plan associated with the grant component have to be completed by August 2026.
- Prepared by Ana Lariau (EUR).

### RRF execution and investment disbursement (key operational figures)
- By end-March 2025:
  - Amount called reached almost €77 billion (96 percent of total grants available).
  - About 65 percent of this amount (almost €50 billion) having already been awarded to final private sector beneficiaries.
- Heterogeneity across government levels:
  - Calls award rates for regional and local governments range between 40 and 75 percent.
- Call award rates and assignment:
  - Similar remaining execution gaps between central government and regional/local governments, but significant variation across regions.

### Recent Dynamics: Key Facts on investment performance
- Plan implementation balance:
  - Almost 70 percent of the planned reforms have been completed.
  - Only 12 percent of the investment milestones and targets have been fulfilled.
- Aggregate investment:
  - Despite RRF execution progress, aggregate investment has remained broadly stable as a percentage of GDP.
- Sectoral performance (by end-2024 / recent periods):
  - Investment in transport equipment: nearly 20 percent below its pre-pandemic level by end-2024.
  - Other construction (non-residential construction): remains below its 2019 level.
  - Housing investment and investment in other machinery and equipment: returned to around their pre-pandemic levels but with a modest rebound.
- Possible reasons for muted aggregate investment response to RRF awards:
  - Backloading of investment milestones and targets implying backloading of actual investment effects.
  - Some RRF-funded investments may be classified as intermediate consumption rather than gross fixed capital formation (GFCF) in the National Accounts.
  - Multi-year project implementation smoothing effects (e.g., transportation projects).
  - Delays between approval of calls, distribution of funds to firms, and actual investment execution.
  - Offsetting declines in other investment components in the National Accounts.
- Illustration: Awarded calls (ELISA, RHS) vs. Gross Fixed Capital Formation (National Accounts) shown for 2021–2025 with Awarded Calls values (0.4, 1.2, 2.2, 3.0, 3.0) and GFCF-to-GDP values (20.2, 20.4, 19.7, 19.5, 20.0) for 2021–2025 respectively.

### Comparative performance with peers
- Spain’s GFCF-to-GDP ratio has steadily declined since the pandemic, while:
  - The euro area ratio has remained broadly stable.
  - The 4-country peer group (Germany, France, Italy, and Portugal) has seen a slight rise.
- Weakness is broad-based but most pronounced in:
  - Transport equipment.
  - Other construction.
- Sector-specific challenges for transport investment may include:
  - High costs associated with fleet renewal in the green transition.
  - Limited access to charging or refueling infrastructure.
  - Potential regulatory or licensing hurdles increasing uncertainty or compliance costs.

### Data used for analysis
- Macroeconomic dataset:
  - Spain, 1995 to 2024, quarterly frequency.
  - Includes GDP, GFCF and subcomponents (machinery and equipment: transport and other equipment; construction: housing and other construction), GFCF deflators, 10-year government bond yield.
  - All variables in constant 2015 prices, sourced from INE.
- Firm-level dataset:
  - Covers 5.1 million firms in Spain, Germany, France, and Italy for 2003 to 2022.
  - Sourced from Orbis Bureau Van Dijk (BvD), compiled by IMF’s Research Department.
  - Analysis focuses on ten economic sectors covering approximately 80 percent of total value added and 70 percent of employment in Spain.

### Empirical strategy (macro and micro)
- Macro: Error Correction Model (ECM)
  - Long-run equation consistent with neoclassical investment theory:
    - ln(i_t) = φ_0 + φ_1 ln(y_t) + φ_2 ucc_t + ε_t
    - ucc_t = (r_t + δ) / P_k with δ set to zero for simplicity.
  - Short-run dynamics include lags of Δln y_t, ucc_t, Δln i_t, and an error correction term ECT_t−1.
  - Model estimated using quarterly data from 1995 to 2020 and simulated out of sample to assess post-pandemic recovery dynamics.
  - Separate estimations for total capital formation and for subcomponents: machinery and equipment, other construction.
  - Model augmented in robustness checks with an Economic Policy Uncertainty (EPU) index for Spain.
- Micro: Firm-level panel regression
  - Baseline specification:
    - NetInv_cji,t = β_0 + β_1 Age_cji,t−1 + β_2 ln(Size_cji,t−1) + β_3 Δln(Sales_cji,t−1) + β_4 Profit_cji,t−1 + β_5 Markup_cji,t−1 + β_6 Leverage_cji,t−1 + firm_i + cty_c × year_t × Nace4_j + ε_cji,t
  - NetInv defined as change in fixed assets relative to previous year’s fixed assets.
  - Controls: firm fixed effects; fully interacted country-year-sector (Nace4) fixed effects.
  - Regression estimated at firm level using annual data, weighted by firm size.
  - Augmented specifications interact age, size, profitability, and leverage with country and post-pandemic dummies; subsample analyses for small firms and middle-aged firms.

### Main results — Macroeconomic analysis
- Aggregate investment vs. fundamentals:
  - As of 2024Q4, the cumulative shortfall in capital formation since the pandemic is estimated at around 15 percent.
  - Investment has gradually moved closer to model-implied fundamentals during 2023–24.
  - Error correction term: if investment is 10 percent below its long-run equilibrium, it would rise by approximately 0.29 percent in the following quarter.
- Component breakdown (as of 2024Q4):
  - Machinery and equipment: cumulative shortfall of approximately 56 percent vis-à-vis model-implied level since the start of the pandemic.
  - Other construction: cumulative shortfall of 16 percent.
- Model fit and interpretation:
  - ECM exhibits strong explanatory power; coefficients statistically significant and consistent with theoretical priors.
  - Model provides a better statistical fit for machinery and equipment than for other construction.
- Role of Economic Policy Uncertainty (EPU):
  - EPU spiked sharply during the pandemic and remains above pre-pandemic levels.
  - Incorporating EPU into the ECM:
    - EPU has a statistically significant long-run impact on total capital formation and on subcategories (other construction and machinery and equipment).
    - In the short run, EPU significantly affects total capital formation and machinery and equipment, with timing of effects varying.
    - Adding EPU reduces the cumulative gap between actual and predicted investment by 2 percentage points overall and by 14 percentage points for machinery and equipment.

*Source: IMF staff Box 1. Investment Component of the Recovery, Transformation and Resilience Plan (sipea2025076).*

### 18.      Net investment is shaped by a range of firm-level characteristics. Estimates from

### 18.      Net investment is shaped by a range of firm-level characteristics. Estimates from

### Firm-level drivers of net investment
- Profitability is a key driver of a firm’s net investment rate: the coefficient on lagged profit is statistically significant across all specifications and robust to changes in the fixed effects structure.
- Firm size (proxied by fixed assets) is negatively associated with the net investment rate with large and significant coefficients, indicating that larger firms invest less relative to their size.
- Higher leverage is associated with lower investment, though estimated coefficients are small and statistically insignificant on average, implying a limited average effect of balance sheet constraints.
- Markups are negatively associated with investment; the negative effect becomes more pronounced in tighter specifications with more granular fixed effects.
- Annex Table III.3 provides coefficient estimates across models with varying combinations of fixed effects.

### Small firms: Spain and post-pandemic interactions
- Sample definition: small firms are those in the bottom 25th percentile of the size distribution within each country.
- Interaction terms included for Spain and the post-pandemic period for profitability and leverage.
- Findings:
  - More profitable small firms generally invest more, but this relationship is significantly weaker in Spain, suggesting profits translate less into investment in Spain (possible drivers: heightened uncertainty or short-termism).
  - Leverage is more positively associated with investment in small firms in Spain compared to other countries, indicating debt may act more as a signal of financial capacity rather than as a constraint for this group.

### Young- and middle-aged firms (≤ 10 years): Spain and post-pandemic interactions
- Sample definition: firms 10 years old or younger.
- Interaction terms for Spain and the post-pandemic period for profitability and leverage.
- Findings:
  - More profitable young- and middle-aged firms generally invest more, but this relationship is significantly weaker for Spanish firms in the post-pandemic period — profits translate less into investment in Spain in this group.
  - More leveraged young firms invest less, consistent with binding credit constraints for this group; the negative relationship is more pronounced for Spanish firms in general and even more so after the pandemic.
  - Implication: young firms differ from small firms; credit constraints play a stronger negative role for young firms in Spain, especially post-pandemic.

### Robustness checks
- Macro-level: Error-correction model findings remain broadly unchanged when the long-term coefficient on GDP is restricted to unity. Baseline macro results hold when additional controls (consumer confidence, term premia, financial constraints, terms of trade) are included.
- Firm-level: Baseline findings for the unconstrained sample are robust to alternative specifications with different combinations of fixed effects (see Annex Table III.3). Results are robust across different combinations of explanatory variables and do not hinge on any single variable’s inclusion.

### Conclusion: interpretation of findings
- Spain’s investment post-pandemic recovery has been weak relative to fundamentals and euro area peers; the shortfall spans all major components of investment, with transport equipment particularly weak.
- Both macro- and firm-level evidence point to elevated uncertainty — including domestic policy uncertainty — as a drag on the recovery.
- Investment among small and young- to middle-aged Spanish firms has become less responsive to profitability, particularly during the post-pandemic period, suggesting uncertainty has curbed willingness to invest even when profitability was strong.
- Leverage has been a stronger drag on investment for young- and middle-aged Spanish firms, hinting at greater difficulties accessing external finance during the pandemic.
- The broad improvement in financial conditions since the start of the ECB’s easing cycle may disproportionately benefit Spanish firms’ investment going forward, but persistent domestic and global uncertainty could continue to weigh on the investment outlook.

### Policy considerations and recommendations
- Reduce policy uncertainty through clearer, more predictable tax, spending, and regulatory frameworks to restore firms’ willingness to invest.
- Ease financing constraints for young high-growth firms by strengthening access to credit and equity finance, including promoting the EU’s Capital Markets Union and targeted support as needed.
- Address structural barriers in specific sectors (e.g., construction); in particular:
  - Advance amendments to the Land Law.
  - Streamline construction permitting processes to unlock stalled projects and support investment in infrastructure and housing.
- Ensure full execution of planned NGEU investment projects to continue supporting investment in Spain relative to other euro area peers.

### Key statistics and exact numerical findings (selected)
- Small firms defined as bottom 25th percentile of size distribution.
- Annex II. Table 2, Spain: Summary Statistics
  - Net investment: Mean 0.15; Standard deviation 0.57; 1st Percentile -0.47; 99th Percentile 2.89
  - Age: Mean 3.89; Standard deviation 1.02; 1st Percentile 2.00; 99th Percentile 6.00
  - Ln(Fixed Assets): Mean 11.81; Standard deviation 1.72; 1st Percentile 7.69; 99th Percentile 15.39
  - Sales growth: Mean 1.82; Standard deviation 24.09; 1st Percentile -68.84; 99th Percentile 72.15
  - Profit: Mean 0.11; Standard deviation 0.12; 1st Percentile -0.15; 99th Percentile 0.52
  - Markup: Mean 1.27; Standard deviation 0.46; 1st Percentile 0.86; 99th Percentile 3.27
- Annex III. Table 3 (baseline firm-level coefficients, standardized variables)
  - L.Age: 0.0113 (0.0136); 0.0106 (0.0135); 0.0101 (0.0179)
  - L.lnFixed Assets: -1.024*** (0.0577); -1.022*** (0.0580); -1.020*** (0.0628)
  - L.dlnSales: 0.00215* (0.00122); 0.00181 (0.00124); 0.000976 (0.00165)
  - L.Profit: 0.0429*** (0.00306); 0.0434*** (0.00298); 0.0474*** (0.00407)
  - L.Markup: -0.00692*** (0.00201); -0.00709*** (0.00211); -0.00838 (0.00514)
  - L.Leverage rate: -0.00766 (0.00764); -0.00760 (0.00768); -0.00599 (0.01000)
  - Constant: 1.740*** (0.0775); 1.738*** (0.0781); 1.735*** (0.0822)
  - Observations: 7,713,968.00; 7,716,057.00; 7,716,330.00
  - R^2: 0.470; 0.459; 0.442
- Annex III. Table 1 (first-stage ECM highlights)
  - Ln(Y) coefficients: 0.595*** (0.107); -0.092 (0.183); 0.737*** (0.071) in columns 1–3
  - EPU (quartiles): -0.086*** (0.011); -0.149*** (0.020); -0.040*** (0.007) in columns 4–6
  - Observations: 100; 100; 100; 969; 696; 696 (columns 1–6)
  - R-squared reported: 0.601; 0.034; 0.866; 0.707; 0.388; 0.863 (columns 1–6)
- Annex III. Table 2 (second-stage ECM highlights)
  - Ln(Y) = D: 1.119*** (0.066); 0.919*** (0.108); 1.825*** (0.115) in columns 1–3
  - UCC (M&E) = D: -0.013*** (0.004); -0.014*** (0.004)
  - EPU (quartile) = D: -0.005* (0.003); -0.004 (0.005); -0.005 (0.005)
  - EC = L: -0.029* (0.015); -0.012 (0.013); -0.142*** (0.049)
  - Observations: 989; 898; 989; 949; 494; 9494 (as reported across columns)
  - R-squared: 0.834; 0.556; 0.832; 0.845; 0.569; 0.855

*Source: IMF staff calculations and Orbis (as reported in the chapter).*

---


_Source: https://www.imf.org/-/media/files/publications/selected-issues-papers/2025/english/sipea2025076.pdf_