## 1. GDP per Capita Gaps with CEE Peers

## Source details

**Canonical URL:** [1. GDP per Capita Gaps with CEE Peers](https://www.imf.org/-/media/files/publications/cr/2021/english/1rouea2021002.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2021/english/1rouea2021002.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2021/english/1rouea2021002.pdf.json)

---

### Key findings on gaps and significance
- Romania has made steady progress in income convergence with EU peers, but a persistent income gap remains.
- Romania lags regional EU peers across a broad set of public infrastructure indicators (railtrack density, road density, mobile subscription, internet usage, air passengers, electricity generation capacity).
- Romania exhibits significant gaps in governance rankings versus regional EU peers (WGI): limited progress over the last decade and reversals in regulatory quality, control of corruption and government effectiveness in recent years.
- The governance indicators are perception-based and subject to uncertainty; interpretations should account for opinion-survey and expert-assessment limitations.

### Data and sample scope
- Panel sample for primary regressions: up to 94 EMs during 1996-2015.
- Sample for investment-channel regressions (investment shocks): 78 EMs during 1991-2020.
- Regressions involving public and private investment use 42 EMs over 1997-2015.
- Regressions involving the infrastructure index use 32-42 EMs depending on projection horizons over 1996-2015.

### Methodology
- Framework: Production function Y = F(K_public, K_private, Z(γ)), with governance reforms (γ) affecting output directly via Z(γ) and indirectly via public and private capital accumulation.
- Empirical approach:
  - Local projection methods: regress k-period-ahead percentage changes in dependent variables on changes in governance index (GovReform).
  - Governance index = average of three WGI subindices: government effectiveness, rule of law, regulatory quality.
  - Specification includes time and country fixed effects, two lags of GDP growth, two lags of dependent variable, and two lags of change in governance scores.
  - Identification of exogenous investment shocks: difference between October WEO projections for investment in the current year and actual observations for that year.

### Impact of governance shocks on output (main regression results)
- A two-standard-deviation shock in governance produces an economically and statistically significant and persistent effect on output.
- Impulse response highlights:
  - Maximum impact: 2 percent increase in output level two years after the shock.
  - Six-year horizon: impact stabilizes at 1.5 percent.
- Regressions use up to 1,390 observations and 94 countries for horizons 1–4 years.

### Feasible reform scenario for Romania
- Scenario motivation: pace of governance reforms observed during EU accession period.
- Assumed reform: closing ½ of the governance gap with EU peers by 2026.
- Quantified impacts:
  - Increase in annual medium-term output growth rate by 0.3pp.
  - Closing ¼ of Romania’s income gap with EU peers by 2026, increasing Romania’s GDP level by 2.7 percent.
- Pace of reforms implied: 0.45 standard deviations increase in the governance index per year (consistent with the EU accession period pace of 0.42 standard deviations increase per year).

### Heterogeneity across governance subindices
- Government effectiveness shows the largest potential for increasing output.
- Output responses:
  - Overall constructed governance index: 1.5% increase at a 6-year horizon (cumulated).
  - Government effectiveness shock: output response almost doubles the overall-index response.
  - Regulatory quality: similar pattern to overall index.
  - Rule of law: less significant and diminishing by the 6-year horizon.

### The role of the investment channel
- Governance → Investment:
  - Both public investment and private investment respond positively and significantly to governance shocks.
  - Timing of peaks:
    - Public investment: peaks 3 years after the shock.
    - Private investment: peaks 1 year after the shock and is more persistent in the medium term.
- Investment → Output (investment shocks):
  - A 10 percent increase in public investment increases output by 0.2-0.4 percent depending on horizon.
  - Similar positive effects apply to increases in private investment.
  - Investment-channel regressions use exogenous investment shocks identified from October WEO projection errors.
- Investment channel as transmitter of governance reforms:
  - Investment is the main channel of transmission of improvements in governance to output.
  - Contribution under the reform scenario:
    - Public investment channel accounts for about 1/3 of the total increase in output.
    - Private investment channel accounts for about 1/3 of the total increase in output.
  - Additional channels:
    - Labor channel: Better governance can help retain skilled workers, increasing human capital and output (example: reduced emigration effects on labor productivity).
    - Governance reforms are also positively linked with a reduction in inequality and informality.

### Fiscal multipliers and output impact
- Estimated fiscal multipliers (from regression equation (2) in Table 4):
  - Short-term fiscal multiplier: 0.45
  - Medium-term fiscal multiplier: 0.85
- A 10 percent increase in public investment:
  - Increases output by 0.23 percent on impact
  - Increases output by 0.42 percent in a 3-year horizon
- Public investment average: Public investment is 5 percent of GDP (average)
- Note: Estimated multipliers determine the size of the investment channel in transmitting the impact of governance reforms on output growth but do not affect the estimated overall impact of governance reforms on output growth.

### Regression estimates for public and private investment effects on output (key coefficients)
- Public investment shocks on output (Table 4, PubInvShock coefficients):
  - Column (1): 0.0225**
  - Column (2): 0.0339***
  - Column (3): 0.0345**
  - Column (4): 0.0423**
  - Column (5): 0.0398*
- Private investment shocks on output (Table 5, PrivInvShock coefficients):
  - Column (1): 0.0156***
  - Column (2): 0.0295***
  - Column (3): 0.0396***
  - Column (4): 0.0455***
  - Column (5): 0.0414***

### Infrastructure gap and catch-up with peers
- Infrastructure index construction:
  - Four indicators: internet usage (percent of population), air passengers carried (passengers carried per population), mobile usage (percent of population), rail track density (kilometer per thousand hectare).
  - Series standardized, then Principal Component Analysis (PCA) used; first principal component forms the infrastructure index.
- Response to governance reforms:
  - Infrastructure indicator responds positively and significantly to governance reforms.
- Catch-up projection:
  - Under the reform scenario, Romania would close 61 percent of the infrastructure gap with CEE peers by 2026.
- Aggregation method:
  - Impacts on infrastructure from each year’s governance catchup shock are aggregated based on regression results to calculate the infrastructure index increase for a given year.

### Comparisons with literature and sensitivity
- Furceri and Li 2017: fiscal multiplier 0.2 in the short-run and 0.6 at a 5-year horizon.
- Ramey 2019: most literature finds fiscal multipliers below 1; larger multipliers possible in downturns (e.g., ZLB) or for investment-specific spending.
- Blanchard and Leigh 2012: short-run fiscal multiplier 0.9 to 1.7 for European countries during recession periods.
- Ramey 2021 (survey): short-run infrastructure investment multipliers can be significantly below 1, long-run can be significantly larger than 1.
- Note: The study’s long-run fiscal multiplier is estimated on the conservative side relative to the literature; a higher multiplier would increase the role of the public investment channel.

### Conclusions and policy recommendations
- Main conclusions:
  - Efforts to improve economic governance need to be re-energized in Romania.
  - Bridging half of the gap in governance rankings with regional peers could add 0.3 percentage points annually to Romania’s medium-term growth.
  - Persistent reform efforts could close 1/4 of Romania’s income gap with its EU peers by 2026.
  - The main channel of impact involves public and private investment, each accounting for 1/3 of the total impact on output.
  - Governance improvements could also help alleviate social challenges such as migration, inequality and informality.
- Policy implications and priorities:
  - Improving government effectiveness should be a key focus of the governance reform agenda.
  - Priority actions for public investment management:
    - Improve all aspects of public investment management from the planning stage to procurement to speed-up project implementation.
    - Strengthen project implementation to accelerate absorption of EU funds.
  - SOE reforms:
    - Strengthen governance and monitoring of SOEs; implement corporate governance reforms and improve monitoring of SOE financial performance to boost public investment and EU funds absorption.
  - Wider governance considerations:
    - Reforms in other governance areas could be pursued if needed for other policy targets.
- Inequality effects:
  - Regression results suggest the Gini index responds negatively and significantly to governance improvements.
  - Under the reform scenario, Romania’s Gini index would be reduced by 3.1 percent by 2026, closing 55 percent of the inequality gap with CEE peers.

*International Monetary Fund — ROMANIA chapter excerpt*

### 1. GDP per Capita Gaps with CEE Peers  _______________________________________________ 3

### 1. GDP per Capita Gaps with CEE Peers

### Key findings on gaps and significance
- Romania has made steady progress in income convergence with EU peers, but a persistent income gap remains.
- Romania lags regional EU peers across a broad set of public infrastructure indicators (railtrack density, road density, mobile subscription, internet usage, air passengers, electricity generation capacity).
- Romania exhibits significant gaps in governance rankings versus regional EU peers (WGI): limited progress over the last decade and reversals in regulatory quality, control of corruption and government effectiveness in recent years.
- The governance indicators are perception-based and subject to uncertainty; interpretations should account for opinion-survey and expert-assessment limitations.

### Data and sample scope
- Panel sample for primary regressions: up to 94 EMs during 1996-2015.
- Sample for investment-channel regressions (investment shocks): 78 EMs during 1991-2020.
- Due to data limitations, regressions involving public and private investment use 42 EMs over 1997-2015.
- Regressions involving the infrastructure index use 32-42 EMs depending on projection horizons over 1996-2015.

---

### Methodology

### Framework
- Production function framework: Y = F(K_public, K_private, Z(γ)), where γ represents governance reforms.
- Public capital accumulation: δK_public = G(I_public(γ), X(γ)); private capital accumulation: δK_private = G(I_private(γ), M(γ)).
- Governance reforms (γ) affect output directly via Z(γ) and indirectly via capital accumulation and capital complementarities.

### Empirical approach
- Local projection methods: regress k-period-ahead percentage changes in dependent variables (output, public investment, private investment, infrastructure index) on changes in governance index (GovReform).
- Governance index = average of three WGI subindices: government effectiveness, rule of law, regulatory quality.
- Regression specification includes time and country fixed effects, two lags of GDP growth, two lags of dependent variable, and two lags of change in governance scores.
- Identification of exogenous investment shocks: difference between October WEO projections for investment in the current year and actual observations for that year (following Furceri and Li 2017).

---

### Main regression results and a reform scenario

### Impact of governance shocks on output
- A two-standard-deviation shock in governance produces an economically and statistically significant and persistent effect on output.
- Impulse response highlights:
  - Maximum impact: 2 percent increase in output level two years after the shock.
  - Six-year horizon: impact stabilizes at 1.5 percent.
- Table estimation context: regressions use up to 1,390 observations and 94 countries for horizons 1–4 years (see table notes for standard errors and significance levels).

### Feasible reform scenario for Romania
- Scenario motivation: pace of governance reforms observed during EU accession period.
- Assumed reform: closing ½ of the governance gap with EU peers by 2026.
- Quantified impacts:
  - Increase in annual medium-term output growth rate by 0.3pp.
  - Closing ¼ of Romania’s income gap with EU peers by 2026, increasing Romania’s GDP level by 2.7 percent.
- Pace of reforms implied: 0.45 standard deviations increase in the governance index per year (consistent with the EU accession period pace of 0.42 standard deviations increase per year).

### Heterogeneity across governance subindices
- Government effectiveness shows the largest potential for increasing output.
- Output responses:
  - Overall constructed governance index: 1.5% increase at a 6-year horizon (cumulated).
  - Government effectiveness shock: output response almost doubles the overall-index response.
  - Regulatory quality: similar pattern to overall index.
  - Rule of law: less significant and diminishing by the 6-year horizon.

---

### The role of the investment channel

### Governance → Investment
- Both public investment and private investment respond positively and significantly to governance shocks.
- Timing of peaks:
  - Public investment: peaks 3 years after the shock.
  - Private investment: peaks 1 year after the shock and is more persistent in the medium term.

### Investment → Output (investment shocks)
- A 10 percent increase in public investment increases output by 0.2-0.4 percent depending on horizon.
- Similar positive effects apply to increases in private investment.
- The investment-channel regressions use exogenous investment shocks identified from October WEO projection errors.

---

*Prepared by Rudolfs Bems and Mengxue Wang; IMF staff calculations and sources: WDI, WGI, WEO, WEO October projections, and IMF staff regression analysis.*

### 14.      Based on the results from regression equation (2) in Table 4, we infer that the short-

### 14.      Based on the results from regression equation (2) in Table 4, we infer that the short-

### Fiscal multipliers and output impact
- Estimated fiscal multipliers (from regression equation (2) in Table 4):
  - Short-term fiscal multiplier: 0.45
  - Medium-term fiscal multiplier: 0.85
- A 10 percent increase in public investment:
  - Increases output by 0.23 percent on impact
  - Increases output by 0.42 percent in a 3-year horizon
- Public investment average:
  - Public investment is 5 percent of GDP (average)
- Note on interpretation:
  - The estimated multipliers determine the size of the investment channel in transmitting the impact of governance reforms on output growth but do not affect the estimated overall impact of governance reforms on output growth.

### Regression estimates for public and private investment effects on output (key coefficients)
- Public investment shocks on output (Table 4, PubInvShock coefficients):
  - Column (1): 0.0225**
  - Column (2): 0.0339***
  - Column (3): 0.0345**
  - Column (4): 0.0423**
  - Column (5): 0.0398*
- Private investment shocks on output (Table 5, PrivInvShock coefficients):
  - Column (1): 0.0156***
  - Column (2): 0.0295***
  - Column (3): 0.0396***
  - Column (4): 0.0455***
  - Column (5): 0.0414***

### Investment channel as transmitter of governance reforms
- Main finding:
  - Investment is the main channel of transmission of improvements in governance to output.
- Contribution under the reform scenario:
  - Public investment channel accounts for about 1/3 of the total increase in output.
  - Private investment channel accounts for about 1/3 of the total increase in output.
- Additional channels:
  - Labor channel: Better governance can help retain skilled workers, increasing human capital and output (example: reduced emigration effects on labor productivity).
  - Governance reforms are also positively linked with a reduction in inequality and informality (references to annex III and 2019 October WEO Chapter 3).

### Infrastructure gap and catch-up with peers
- Infrastructure index construction:
  - Four indicators: internet usage (percent of population), air passengers carried (passengers carried per population), mobile usage (percent of population), rail track density (kilometer per thousand hectare).
  - Series standardized, then Principal Component Analysis (PCA) used; first principal component forms the infrastructure index.
- Response to governance reforms (Table 6 and Figure 10):
  - Infrastructure indicator responds positively and significantly to governance reforms.
- Catch-up projection:
  - Under the reform scenario, Romania would close 61 percent of the infrastructure gap with CEE peers by 2026.
- Aggregation method:
  - Impacts on infrastructure from each year’s governance catchup shock are aggregated based on regression results to calculate the infrastructure index increase for a given year.

### Comparisons with literature and sensitivity
- Literature comparisons:
  - Furceri and Li 2017: fiscal multiplier 0.2 in the short-run and 0.6 at a 5-year horizon.
  - Ramey 2019: most literature finds fiscal multipliers below 1; larger multipliers possible in downturns (e.g., ZLB) or for investment-specific spending.
  - Blanchard and Leigh 2012: short-run fiscal multiplier 0.9 to 1.7 for European countries during recession periods.
  - Ramey 2021 (survey): short-run infrastructure investment multipliers can be significantly below 1, long-run can be significantly larger than 1.
- Note:
  - The study’s long-run fiscal multiplier is estimated on the conservative side relative to the literature; a higher multiplier would increase the role of the public investment channel.

### Conclusions and policy recommendations
- Main conclusions:
  - Efforts to improve economic governance need to be re-energized in Romania.
  - Bridging half of the gap in governance rankings with regional peers could add 0.3 percentage points annually to Romania’s medium-term growth.
  - Persistent reform efforts could close 1/4 of Romania’s income gap with its EU peers by 2026.
  - The main channel of impact involves public and private investment, each accounting for 1/3 of the total impact on output.
  - Governance improvements could also help alleviate social challenges such as migration, inequality and informality.
- Policy implication:
  - Re-energized and persistent governance reforms that improve public and private investment conditions are central to raising medium-term growth, closing infrastructure and income gaps, and addressing social challenges.

*Sources: WDI, WGI, WEO and IMF staff calculations*

### 18.      To maximize medium-term growth prospects, improving government effectiveness

### To maximize medium-term growth prospects, improving government effectiveness

### Key findings on governance and growth
- Improving government effectiveness should be a key focus of the governance reform agenda; this is the area of governance where gaps with regional peers are the widest.
- Empirical estimates show that gains in government effectiveness have the largest output impact among governance sub-components investigated.
- The centrality of the investment channel in generating output gains implies that improvements in public investment management should be a priority.

### Public investment management: priority actions
- Improve all aspects of public investment management from the planning stage to procurement to speed-up project implementation.
- Strengthen project implementation to accelerate absorption of EU funds.

### State-owned enterprises (SOEs): governance and monitoring
- SOE performance requires improvements; governance of SOEs needs to be strengthened.
- Consistent implementation of corporate governance reforms and improved monitoring of SOE financial performance are important reform components.
- Better SOE governance will help boost public investment and accelerate absorption of EU funds.

### Wider governance reform considerations
- While the main discussion focuses on the impact on growth, reforms in other governance areas could be favored if needed for other policy targets.

### Definition: Worldwide Governance Index components (as used in the analysis)
- Government Effectiveness: Reflects perceptions of the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government's commitment to such policies.
- Regulatory Quality: Reflects perceptions of the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development.
- Rule of Law: Reflects perceptions of the extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence.
- Voice and Accountability: Reflects perceptions of the extent to which a country's citizens are able to participate in selecting their government, as well as freedom of expression, freedom of association, and a free media.
- Political Stability and Absence of Violence/Terrorism: Reflects perceptions of the likelihood that the government will be destabilized or overthrown by unconstitutional or violent means, including politically-motivated violence and terrorism.
- Control of Corruption: Reflects perceptions of the extent to which public power is exercised for private gain, including both petty and grand forms of corruption, as well as "capture" of the state by elites and private interests.

### Governance reforms and inequality: empirical results and projections
- Regression results suggest the Gini index responds negatively and significantly to governance improvements based on the same sample of countries as in the main text.
- Under the reform scenario defined in section C, Romania’s Gini index would be reduced by 3.1 percent by 2026, closing 55 percent of the inequality gap with CEE peers. This is calculated by measuring the impact of governance reforms on the level of Gini index using the regression equation (1).

Key regression table excerpts (Romania: Regression Results for the Impact of Governance Reforms on Gini Index)
- Dependent variable: Gini index (various dynamic specifications across columns (1)–(5))
- GovShock coefficients:
  - Column (1): GovShock = -0.0446** (standard error (0.0149))
  - Column (2): GovShock = -0.0811** (standard error (0.0256))
  - Column (3): GovShock = -0.141*** (standard error (0.0277))
  - Column (4): GovShock = -0.0971** (standard error (0.0345))
  - Column (5): GovShock = -0.0842 (standard error (0.0635))
- L1.GovShock coefficients:
  - Column (1): L1.GovShock = -0.0112 (standard error (0.0178))
  - Column (2): L1.GovShock = -0.0300 (standard error (0.0237))
  - Column (3): L1.GovShock = -0.0408 (standard error (0.0478))
  - Column (4): L1.GovShock = 0.00528 (standard error (0.0399))
  - Column (5): L1.GovShock = 0.0882*** (standard error (0.0249))
- L2.GovShock coefficients:
  - Column (1): L2.GovShock = -0.0261** (standard error (0.0117))
  - Column (2): L2.GovShock = -0.0383 (standard error (0.0235))
  - Column (3): L2.GovShock = 0.00682 (standard error (0.0375))
  - Column (4): L2.GovShock = 0.102** (standard error (0.0312))
  - Column (5): L2.GovShock = 0.0942* (standard error (0.0439))
- L1.GDP_growth coefficients (examples):
  - Column (1): L1.GDP_growth = 0.00157* (standard error (0.000716))
  - Column (2): L1.GDP_growth = 0.00243* (standard error (0.00114))
- L2.GDP_growth coefficients (examples):
  - Column (1): L2.GDP_growth = 0.000914** (standard error (0.000392))
  - Column (2): L2.GDP_growth = 0.00122** (standard error (0.000514))
- L1.Gini and L2.Gini coefficients vary across specifications; examples:
  - Column (1): L1.Gini = 0.200*** (standard error (0.0556)); L2.Gini = 0.316*** (standard error (0.0395))
  - Column (2): L1.Gini = 0.586*** (standard error (0.164)); L2.Gini = -0.0402 (standard error (0.0997))
- Sample sizes:
  - Observations: Column (1) = 537; Column (2) = 475; Column (3) = 415; Column (4) = 355; Column (5) = 295
  - Number of countries: Column (1) = 62; Column (2) = 60; Column (3) = 60; Column (4) = 60; Column (5) = 58
- Note: Standard errors (asymptotic Driscoll-Kraay SEs are used) are in parentheses. *** p<0.01, ** p<0.05, * p<0.1. Fixed effects are included but not shown.

*International Monetary Fund — ROMANIA chapter excerpt*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2021/english/1rouea2021002.pdf_