## 1.    Slovenia GDP and  EU Growth Composition, 1997–2022

## Source details

**Canonical URL:** [1.    Slovenia GDP and  EU Growth Composition, 1997–2022](https://www.imf.org/-/media/files/publications/cr/2024/english/1svnea2024002.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2024/english/1svnea2024002.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2024/english/1svnea2024002.pdf.json)

---

### Introduction
- Slovenia converged closer to the EU average ahead of the Global Financial Crisis (GFC) but convergence has slowed since the GFC.
- Between 1995 and 2008, the income gap relative to the EU average narrowed by 14 percentage points to less than 10 percent.
- The GFC and the subsequent banking crisis in Slovenia in 2013 reversed many gains; convergence resumed in 2016 but as of 2022 relative income per capita was still below the earlier peak.
- Given demographic limits on labor contributions, policy focus should be on reinvigorating private investment and pursuing labor and product market reforms to boost total factor productivity (TFP) growth.

### Growth and Productivity Trends
- Composition shift (pre- and post-EU accession vs post-GFC):
  - Pre- and immediately post-EU accession: growth driven by strong capital accumulation and rapid TFP growth; labor contributed very little.
  - After the GFC: capital made negative contributions to growth due to low public and especially private investment; labor played a positive role.
  - TFP remained an important driver of GDP growth apart from 2008–13, but TFP growth has slowed in recent years, contributing to declining labor productivity.
- Decomposition note: for a Cobb-Douglas production function, growth in output per worker = TFP growth + (capital intensity × capital share).
- Staff’s shift-share analysis (Annex I) provides sectoral distribution insights and labour reallocation effects.

### Investment and Capital Stock (Key findings)
- Investment lagged depreciation after the GFC, lowering the capital stock.
- Declines led primarily by private investment.
- Capital stock has not recovered to its earlier peak but is on an upward trend.
- Capital intensity fell in Slovenia and the capital stock per worker is low relative to peers.

### Labor Market Indicators (Key findings)
- Recent labor contributions to growth reflect rising activity and falling unemployment.
- Average hours worked have declined as in other EU countries.
- Overall activity rate in Slovenia is above the EU average, but activity is lower in age groups 20–24 and 60–64.
- An increasing inflow of foreign workers is helping ease demographic constraints; the share of foreign workers in total is increasing.

### Productivity Trends and Determinants
- TFP growth has slowed in recent years, similar to EU peers.
- Labor productivity has also fallen.
- Productivity drivers from literature include:
  - Investment in physical capital and innovation.
  - Education and labor force quality.
  - Supportive environment (institutions, infrastructure, policies).
  - Firm-specific characteristics (technology adoption, R&D, training, management).
- ECB (2021) notes labor regulations, ICT patenting, financial openness and tax structure explain cross-country TFP differences.
- IMAD (2023) provides granular analysis and policy recommendations on Slovenia’s strengths/weaknesses.

### Evidence from Firm-level Data: methodology and caveats
- Data sources: Eurostat structural business statistics, EBRD/World Bank BEEPS, Orbis Bureau Van Dijk (BvD).
- Orbis BvD production function estimated at industry-level across countries following Ackerberg, Caves, and Frazer (2015); measurement caveats noted (e.g., underrepresentation of smaller firms, revenue-dependence).

### Firm Size
- Slovenian firms are smaller than EU peers; prevalence of small and micro firms is high.
- Smaller firms tend to be less productive: firms with more employees are closer to the TFP frontier due to economies of scale, resources, IT and R&D investment.
- World Bank (2021) estimate: productivity of a firm in the highest quartile of the size distribution is about 12 percent and 22 percent closer to the output and value added TFP frontiers, respectively, relative to a firm in the lowest quartile.

### Access to Finance
- Survey evidence does not indicate major issues with access to banks; Slovenian firms are more likely to secure a credit line or loan from financial institutions than EU peers.
- A larger share of financing in Slovenia comes from state-owned banks or government agencies.
- As of end-2022, SID Bank's loans to non-bank customers amounted to €1.4 billion, representing around 13 percent of all bank loans to non-financial corporations (NFCs), mostly to SMEs.
- Slovenian firms rely more on supplier funding compared with peers, particularly small firms; younger companies depend heavily on internal funds.

### Leverage and Productivity (Firm-level econometric evidence)
- Threshold model (Coricelli et al., 2012 methodology) suggests entities with higher leverage levels tend to experience higher productivity growth.
- For Slovenia, it is difficult to distinguish effects across leverage thresholds; point estimates generally point to a positive impact of higher leverage on TFP growth.
- Given low indebtedness of Slovenian firms, there appears to be room to increase leverage without negatively impacting productivity.

Key leverage descriptive statistics (post-truncation sample):
- Truncated sample: about 396 thousand observations.
- Missing long-term debt and loans: 42 percent and 38 percent respectively.
- For leverage defined as total debt to total assets: about 163 thousand observations (45 percent of total).
- Average leverage value: about 26 percent (0.26).
- Percent of total (firms with zero or positive leverage): 44.54 (All), 40.89 (Small <50 empl.), 2.98 (Medium 50-249 empl.), 0.67 (Large 250+ empl.), 7.81 (Young < 5 years).
- Summary percentiles (All): p25 0.08; p50 0.23; p75 0.40; Max 0.76.

Selected regression findings (Table A II.2 highlights):
- Initial TFP: -0.00279*** (Model 1) ... -0.00269*** (Model 5).
- Small (dummy): -0.00227** (1) ... -0.00219** (5).
- Leverage (overall) (Model 1): 0.0164*** (t statistic 11.29).
- Leverage (low/medium/high) examples:
  - Leverage (low): 0.0358* (2); 0.0272** (3); 0.00521** (4); -0.00593 (5).
  - Leverage (medium): 0.0158*** (2); 0.0197*** (3); 0.00698*** (4); 0.00303* (5).
  - Leverage (high): 0.0192*** (2); 0.0192*** (3); 0.0124*** (4); 0.00301*** (5).
- Threshold estimates (selected):
  - Model (2) threshold: 0.19127; Confidence interval: [0.09040, 0.60726].
  - Model (3) threshold: 0.16047; Confidence interval: [0.13190, 0.18509].
  - Model (4) threshold: 0.74934; Confidence interval: [0.65288, 0.81394].
  - Model (5) threshold: 0.30789; Confidence interval: [0.26572, 0.86484].
- Sample sizes and fit:
  - N: 107718 (Models 1 and 2), 147474 (3), 206537 (4), 100014 (5).
  - adj. R-sq: 0.002 (1), 0.002 (2), 0.003 (3), 0.003 (4), 0.002 (5).
- Decision tree (descriptive) implied TFP growth by leverage bin:
  - Leverage < 0.168: -0.6 percent
  - Leverage in [0.168, 0.38]: -0.2 percent
  - Leverage in [0.38, 0.604]: 0.1 percent
  - Leverage > 0.604: 0.6 percent

### Entry, Exit and Firms’ Survival Rates
- Productivity dispersion is large: top 10 percent most productive manufacturing firms are six times more productive than the bottom 10 percent after controlling for input size.
- The firm-level TFP distribution is highly skewed with many firms below the average.
- Declining firm entry and exit rates likely hamper aggregate TFP growth:
  - Slovenia previously exhibited higher firm entry rates vs EU peers, but this has reversed recently.
  - Post-entry survival rates are decreasing; significant challenges concentrated in the first two years—conditional on surviving past 2 years, likelihood of surviving between years 3 to 5 is higher in Slovenia than in a median EU country.
  - Exit rates, previously aligned with EU countries, have been declining.

### Innovation and Physical Capital: Constraints and Areas for Improvement
- Low NFC investment has hampered productivity and growth since the GFC; financial sector stress and NFC deleveraging caused a sharp decline in private investment.
- Business investment is still below the EU average despite being well above pre-GFC; fixed assets of NFCs are relatively low in per capita terms compared to more advanced EU members.
- Composition: Slovenia is close to average in share of machinery in total fixed assets but has a lower share of intellectual property products, reflecting a low share of computer software and databases.
- Survey-identified obstacles to investment:
  - Lack of suitably qualified staff, uncertainty, regulations, and high energy costs.
  - Over a fifth of respondents to the European Investment Bank (EIB) investment survey reported they invested too little in the period 2016–22.
  - Managers cite lack of staff with the right skills, uncertainty about the future, labor and business regulations, and high energy costs as key obstacles.

### Public Investment and Infrastructure
- Public investment in Slovenia has been relatively high but volatile; it declined sharply after the banking crisis due to fiscal consolidation and picked up markedly in recent years, including major infrastructure projects (e.g., Divača–Koper railway and Karavanke tunnel).
- Public capital stock per capita is below the EU average.
- Quantitative infrastructure indicators (motorway and railway density, high-speed internet coverage) do not reveal significant gaps; high-speed internet coverage is above the EU average.
- Less than 10 percent of firms in the World Bank enterprise survey identified transportation infrastructure as a major constraint.
- Logistics Performance Index (2023) score has improved.

### Innovation
- Slovenia has a higher share of innovative firms than the OECD average, with over 70 percent of workers employed by such firms.
- Innovators are concentrated in large companies but SMEs also undertake a significant share of innovation activities.
- More than a third of firms report development of new products; more than a fifth developed products new to the market.
- Income from innovative products is below sample average; innovation activity needs to better translate into higher productivity.
- Cooperation with other private companies, universities, public research institutes and international collaboration is generally above average.

### R&D and Innovation (additional details)
- R&D is closely linked to innovation, but the relationship is complex and non-linear.
- Slovenia’s R&D capital stock, while close to the average as percent of total, is well below the technological leaders.
- The shortfall largely reflects relatively low business expenditures on R&D overall, with wide variation across sectors:
  - Pharmaceuticals and electrical equipment overperform peers.
  - Computer programming and ICT services lag behind.
- Public R&D spending matters for the capital stock and for complementarities/catalyzing effects.
- Slovenia’s budget allocates at the EU average for R&D and legislation provides for a 100 percent tax allowance for investments in R&D.

### Export Complexity and Trade
- Slovenia ranks 12th out of 131 countries on the economic complexity index, indicating strong integration of complex and knowledge-intensive products into its export portfolio.
- Findings mainly pertain to firms producing goods, where labor productivity tends to be higher than in non-tradable services.

### Human Capital and Skills
- Slovenia has a well-educated labor force and strong education quality:
  - Above-average share of the population with at least upper secondary education.
  - Above-average share of young people (25–34) with tertiary education.
  - Graduates in STEM are on par with EU peers.
  - PISA scores: Slovenia performs very well and above the OECD average in math and science.
  - A new human capital measure places Slovenia in the upper third of the distribution.
- Skills shortages and mismatches are growing concerns:
  - Over- and underqualification rates in Slovenia are about 12 percent.
  - Horizontal skills mismatches are among the highest in the EU in the age group 25–34.
  - About half of Slovenian companies provide training programs for full-time employees (vs. 36 percent on average in the other EU countries).
  - Programs in Slovenia tend to focus more on language, communication and managerial skills compared to EU peers.
- Authorities have initiated a labor market platform to assess competence gaps and predict labor market needs.

### Financial Markets, Access to Finance, and Firm Financing
- Strong deleveraging over the last decade increased corporate reliance on internal financing and trade credits:
  - Loans reached nearly 50 percent of NFC liabilities in the run-up to the GFC but have declined since then to less than 30 percent.
  - The share of trade credits in liabilities in Slovenia is more than twice as high as the average for the euro area.
- Slovenia’s overall financial development, measured by the IMF’s Financial Development Index (FDI), lags behind the average for advanced economies and the gap has widened since the GFC.
  - FDI values between 0.4 and 0.7 have the greatest positive impact on growth.
  - Slovenia’s latest FDI value of 0.3 places it below the optimal range.
- Equity and bond markets are at an early stage:
  - Market capitalization of the Ljubljana Stock Exchange is low and continues to decline.
  - The AFME indicator places Slovenia second to last in Europe for capital market competitiveness.

### Production Frontier Analysis and Stochastic Frontier Estimation (selected results)
- Stochastic frontier model (Battese and Coelli (1995) approach) estimated for 1995–2019.
- Frontier coefficients (selected across columns (1)–(5)):
  - Log of Capital Stock: 0.580; 0.581; 0.458; 0.471; 0.446.
  - Log of Number of Employed: 0.436; 0.420; 0.539; 0.528; 0.552.
  - Trend: 0.012; 0.01; 0.009; 0.008; 0.007.
  - Constant: 3.616; 3.799; 5.662; 5.265; 5.614.
- Mean inefficiency equation (selected coefficients):
  - Output gap: -0.008; -0.008; -0.009.
  - Regulatory quality: -0.158; -0.12; -0.099.
  - Employee protection index: 0.081; 0.108; 0.081.
  - Financial markets access: -0.175; -0.2; -0.191.
  - Informal sector share: 0.005; 0.01; 0.011.
  - Income tax share: 0.002; 0.003; 0.003.
  - Human capital index: -0.042 (statistically significant).
  - Tertiary education: -0.004 (statistically significant).
- Model statistics:
  - Number of observations: 950; 950; 646; 646; 581.
  - Number of countries: 38; 38; 32; 32; 32.
  - Period: 1995–2019 for all models.
  - R-squared: 0.980 (reported).
  - Log likelihood: 99.63; 134.34; 98.63; 98.73; 82.4.
- Interpretation:
  - Structural variables have significant impacts on technical inefficiency with expected signs in most specifications.
  - Better financial markets access and higher percentage of labor force with tertiary education reduce inefficiency.
  - Simulation using point estimates indicates that closing 5 percent of the gap in indicator levels between Slovenia and the OECD top performers could be associated with substantial productivity gains, particularly from improving regulatory quality and financial market access.
  - Estimates should be interpreted with caution due to model specification and estimation uncertainties.

### Conclusions and Policy Recommendations
- Achievements:
  - Slovenia has high per capita income, strong human development indicators, effective institutions, and one of the lowest inequality rates in the world.
  - Strong private investment pre-GFC spurred productivity growth, increased economic complexity, diversified exports, and integration into European value chains.
- Main areas for improvement and policy directions:
  - Innovation:
    - Bolster innovation and technological development, promote investment in ICT and automation, encourage patents and trademarks, and support innovative startups to bridge the gap with EU leaders.
  - Regulatory quality:
    - Ease administrative burden, streamline procedures such as building permits, and further digitalize public services to help private investment and productivity.
    - Reduce market distortions and barriers to entry/exit to increase business dynamism and reallocation.
  - Labor regulations:
    - Further progress on labor market flexibility (e.g., severance costs and broader employer burdens) to support growth and competitiveness.
  - Taxation:
    - Address the high labor tax wedge; lowering the tax wedge on labor could increase labor supply, help growth, and reduce informality.
    - Shift revenue composition away from income toward indirect and property taxes, and reduce tax expenditures to ensure at least revenue neutrality.
    - IMF (2015) estimate: a revenue-neutral reform involving reduction of the tax wedge by 5 percent leads to high long-run growth by 0.2–0.3 percent.
  - Education and skills mismatch:
    - Increase the flexibility of the education system to respond to evolving market needs, in consultation with employers.
    - Use the labor market platform to feedback results to education policies.
  - Capital market development:
    - Implement the government’s capital market development strategy through 2030 to expand access to debt and equity financing, support innovative SMEs, and broaden investment opportunities.
    - Deepening capital markets over time would support private investment and growth.

*Source: Selected Issues Paper — BOOSTING PRODUCTIVITY IN SLOVENIA (Republic of Slovenia / International Monetary Fund), April 10, 2024.*

### 1.    Slovenia GDP and  EU Growth Composition, 1997–2022 _________________________ 4

### 1.    Slovenia GDP and  EU Growth Composition, 1997–2022 _________________________ 4

### Introduction
- Slovenia converged closer to the EU average ahead of the Global Financial Crisis (GFC) but convergence has slowed since the GFC.
- Between 1995 and 2008, the income gap relative to the EU average narrowed by 14 percentage points to less than 10 percent.
- The GFC and the subsequent banking crisis in Slovenia in 2013 reversed many gains from the previous decade; convergence resumed in 2016 but as of 2022 relative income per capita was still below the earlier peak.
- Given demographic limits on labor contributions, policy focus should be on reinvigorating private investment and pursuing labor and product market reforms to boost total factor productivity (TFP) growth.

### Growth and Productivity Trends
- Composition shift (pre- and post-EU accession vs post-GFC):
  - Pre- and immediately post-EU accession: growth driven by strong capital accumulation and rapid TFP growth; labor contributed very little.
  - After the GFC: capital made negative contributions to growth due to low public and especially private investment; labor played a positive role in this period.
  - TFP remained an important driver of GDP growth apart from 2008–13, but TFP growth has slowed in recent years, contributing to declining labor productivity.
- Staff’s shift-share analysis (Annex I) provides sectoral distribution insights and labour reallocation effects.
- Note on decomposition: for a Cobb-Douglas production function, growth in output per worker = TFP growth + (capital intensity × capital share).

### Investment and Capital Stock (Key findings)
- Investment lagged depreciation after the GFC, lowering the capital stock.
- Declines led primarily by private investment.
- Capital stock has not recovered to its earlier peak but is on an upward trend.
- Capital intensity fell in Slovenia and the capital stock per worker is low relative to peers.

### Labor Market Indicators (Key findings)
- Recent labor contributions to growth reflect rising activity and falling unemployment.
- Average hours worked have declined as in other EU countries.
- Overall activity rate in Slovenia is above the EU average, but activity is lower in age groups 20–24 and 60–64.
- An increasing inflow of foreign workers is helping ease demographic constraints; the share of foreign workers in total is increasing.

### Productivity Trends
- TFP growth has slowed in recent years, similar to EU peers.
- Labor productivity has also fallen.

### Looking Ahead: Challenges and Opportunities (Determinants of productivity)
- Productivity drivers from literature include:
  - Investment in physical capital and innovation.
  - Education and labor force quality.
  - Supportive environment (institutions, infrastructure, policies).
  - Firm-specific characteristics (technology adoption, R&D, training, management).
- ECB (2021) notes labor regulations, ICT patenting, financial openness and tax structure explain cross-country TFP differences.
- IMAD (2023) provides granular analysis and policy recommendations on Slovenia’s strengths/weaknesses.

### Evidence from Firm-level Data: What Features Matter for Productivity?
- Data sources: Eurostat structural business statistics, EBRD/World Bank BEEPS, Orbis Bureau Van Dijk (BvD).
- Orbis BvD production function estimated at industry-level across countries following Ackerberg, Caves, and Frazer (2015); measurement caveats noted (e.g., underrepresentation of smaller firms, revenue-dependence).

### Firm Size
- Slovenian firms are smaller than EU peers; prevalence of small and micro firms is high.
- Smaller firms tend to be less productive: firms with more employees are closer to the TFP frontier due to economies of scale, resources, IT and R&D investment.
- World Bank (2021) estimate: productivity of a firm in the highest quartile of the size distribution is about 12 percent and 22 percent closer to the output and value added TFP frontiers, respectively, relative to a firm in the lowest quartile.

### Access to Finance
- Survey evidence does not indicate major issues with access to banks; Slovenian firms are more likely to secure a credit line or loan from financial institutions than EU peers.
- A larger share of financing in Slovenia comes from state-owned banks or government agencies.
- As of end-2022, SID Bank's loans to non-bank customers amounted to €1.4 billion, representing around 13 percent of all bank loans to non-financial corporations (NFCs), mostly to SMEs.
- Slovenian firms rely more on supplier funding compared with peers, particularly small firms; younger companies depend heavily on internal funds.

### Leverage and Productivity (Firm-level econometric evidence)
- A threshold model (Coricelli et al., 2012 methodology; Appendix II) suggests entities with higher leverage levels tend to experience higher productivity growth.
- For Slovenia, it is difficult to distinguish effects across leverage thresholds; point estimates generally point to a positive impact of higher leverage on TFP growth.
- Given low indebtedness of Slovenian firms, there appears to be room to increase leverage without negatively impacting productivity.

### Entry, Exit and Firms’ Survival Rates
- Productivity dispersion is large: top 10 percent most productive manufacturing firms are six times more productive than the bottom 10 percent after controlling for input size—a differential higher than the US (2:1) and higher than China and India (5:1).
- The firm-level TFP distribution is highly skewed with many firms below the average.
- Declining firm entry and exit rates likely hamper aggregate TFP growth:
  - Slovenia previously exhibited higher firm entry rates vs EU peers, but this has reversed recently, possibly indicating rising entry barriers.
  - Post-entry survival rates are decreasing; significant challenges are concentrated in the first two years—conditional on surviving past 2 years, likelihood of surviving between years 3 to 5 is higher in Slovenia than in a median EU country.
  - Exit rates, previously aligned with EU countries, have been declining.

### Innovation and Physical Capital: Constraints and Areas for Improvement
- Low NFC investment has hampered productivity and growth since the GFC; financial sector stress and NFC deleveraging caused a sharp decline in private investment.
- Business investment is still below the EU average despite being well above pre-GFC; fixed assets of NFCs are relatively low in per capita terms compared to more advanced EU members.
- Composition matters: Slovenia is close to average in share of machinery in total fixed assets but has a lower share of intellectual property products, reflecting low share of computer software and databases.
- Survey-identified obstacles to investment: lack of suitably qualified staff, uncertainty, regulations, and high energy costs.
  - Over a fifth of respondents to the European Investment Bank (EIB) investment survey reported they invested too little in the period 2016–22.
  - Managers cite lack of staff with the right skills, uncertainty about the future, labor and business regulations, and high energy costs as key obstacles.

### Public Investment and Infrastructure
- Public investment in Slovenia has been relatively high but volatile; it declined sharply after the banking crisis due to fiscal consolidation and picked up markedly in recent years, including major infrastructure projects (e.g., Divača–Koper railway and Karavanke tunnel).
- Public capital stock per capita is below the EU average.
- Quantitative infrastructure indicators (motorway and railway density, high-speed internet coverage) do not reveal significant gaps; high-speed internet coverage is above the EU average.
- Less than 10 percent of firms in the World Bank enterprise survey identified transportation infrastructure as a major constraint.
- Logistics Performance Index (2023) score has improved.

### Innovation
- Slovenia has a higher share of innovative firms than the OECD average, with over 70 percent of workers employed by such firms.
- Innovators are concentrated in large companies but SMEs also undertake a significant share of innovation activities.
- More than a third of firms report development of new products; more than a fifth developed products new to the market.
- Income from innovative products is below sample average; innovation activity needs to better translate into higher productivity.
- Cooperation with other private companies, universities, public research institutes and international collaboration is generally above average.

*Source: Selected Issues Paper — BOOSTING PRODUCTIVITY IN SLOVENIA (Republic of Slovenia / International Monetary Fund), April 10, 2024.*

### 18. The gap in R&D relative to economies at the frontier is, however, substantial. R&D is

### 18. The gap in R&D relative to economies at the frontier is, however, substantial.

### R&D and Innovation
- R&D is closely linked to innovation, but the relationship is complex and non-linear.
- Studies generally find that the R&D capital stock is positively associated with productivity growth.
- Slovenia’s R&D capital stock, while close to the average as percent of total, is well below the technological leaders.
- The shortfall largely reflects relatively low business expenditures on R&D overall, with wide variation across sectors:
  - Pharmaceuticals and electrical equipment significantly overperform peers.
  - Computer programming and ICT services lag behind.
- Public R&D spending matters for the capital stock and for complementarities/catalyzing effects.
- Slovenia’s budget allocates at the EU average for R&D and legislation provides for a 100 percent tax allowance for investments in R&D.

### Export Complexity and Trade
- Firms engaged in international markets face heightened competition that fosters innovation, efficiency, and product quality.
- The economic complexity index (Hidalgo and Hausmann, 2009) captures export diversity and ubiquity and is positively correlated with productivity and growth.
- Slovenia ranks 12th out of 131 countries on the economic complexity index, indicating strong integration of complex and knowledge-intensive products into its export portfolio.
- The findings mainly pertain to firms producing goods, where labor productivity tends to be higher than in non-tradable services.

### Human Capital and Skills
- Slovenia has a well-educated labor force and strong education quality:
  - Above-average share of the population with at least upper secondary education.
  - Above-average share of young people (25–34) with tertiary education.
  - Graduates in STEM are on par with EU peers.
  - PISA scores: Slovenia performs very well and above the OECD average in math and science.
  - A new human capital measure places Slovenia in the upper third of the distribution.
- Skills shortages and mismatches are growing concerns:
  - Over- and underqualification rates in Slovenia are about 12 percent.
  - Horizontal skills mismatches (discrepancy between field of education and occupation) are among the highest in the EU in the age group 25–34.
  - About half of Slovenian companies provide training programs for full-time employees (vs. 36 percent on average in the other EU countries).
  - Programs in Slovenia tend to focus more on language, communication and managerial skills compared to EU peers.
- Authorities have initiated a labor market platform to assess competence gaps and predict labor market needs.

### Financial Markets, Access to Finance, and Firm Financing
- Strong deleveraging over the last decade increased corporate reliance on internal financing and trade credits:
  - Loans reached nearly 50 percent of NFC liabilities in the run-up to the GFC but have declined since then to less than 30 percent.
  - The share of trade credits in liabilities in Slovenia is more than twice as high as the average for the euro area.
- Slovenia’s overall financial development, measured by the IMF’s Financial Development Index (FDI), lags behind the average for advanced economies and the gap has widened since the GFC.
  - Financial institutions access and efficiency broadly in line with peers.
  - Financial institution depth and financial market indicators are significantly weaker (low private credit to GDP ratio, small number of private debt issuers, low stock market capitalization and turnover).
- The relationship between financial development and growth is bell-shaped:
  - FDI values between 0.4 and 0.7 have the greatest positive impact on growth.
  - Slovenia’s latest FDI value of 0.3 places it below the optimal range, implying potential gains from moving up the curve.
- Equity and bond markets are at an early stage:
  - Market capitalization of the Ljubljana Stock Exchange is low and continues to decline.
  - The AFME indicator places Slovenia second to last in Europe for capital market competitiveness, with only marginal improvement since 2018.

### Production Frontier Analysis and Stochastic Frontier Estimation
- Improving efficiency is key for enhancing productivity growth; TFP growth is the residual not explained by input increases.
- A stochastic frontier model (Battese and Coelli (1995) approach) is used in a panel of advanced countries with PWT data on GDP, labor and capital.
- Mean inefficiency is modeled as a function of: regulatory quality, an employee protection index, financial market access, the share of income taxes in total tax revenue, the informal sector share, and measures of education; the output gap is included to control for the business cycle.
- Key estimation results (dependent variable: log of GDP):
  - Frontier coefficients (models reported as columns (1)–(5)):
    - Log of Capital Stock: 0.580; 0.581; 0.458; 0.471; 0.446 (standard errors reported in source).
    - Log of Number of Employed: 0.436; 0.420; 0.539; 0.528; 0.552.
    - Trend: 0.012; 0.01; 0.009; 0.008; 0.007.
    - Constant: 3.616; 3.799; 5.662; 5.265; 5.614.
  - Mean inefficiency equation (selected coefficients across specifications):
    - Output gap: -0.008; -0.008; -0.009.
    - Regulatory quality: -0.158; -0.12; -0.099.
    - Employee protection index: 0.081; 0.108; 0.081.
    - Financial markets access: -0.175; -0.2; -0.191.
    - Informal sector share: 0.005; 0.01; 0.011.
    - Income tax share: 0.002; 0.003; 0.003.
    - Advanced education: 0.001 (not statistically significant in reported specification).
    - Human capital index: -0.042 (reported as statistically significant).
    - Tertiary education: -0.004 (reported as statistically significant).
  - Model statistics and sample:
    - Number of observations: 950; 950; 646; 646; 581 (by model).
    - Number of countries: 38; 38; 32; 32; 32.
    - Period: 1995–2019 for all models.
    - R-squared: 0.980 (reported).
    - Log likelihood: 99.63; 134.34; 98.63; 98.73; 82.4 (by model).
- Interpretation of estimation:
  - Structural variables have significant impacts on technical inefficiency with expected signs in most specifications.
  - Lower values of the employee protection index, informal sector share and income taxes would contribute to reducing technical inefficiency.
  - Better financial markets access and higher percentage of labor force with tertiary education would reduce inefficiency.
  - A simple simulation using point estimates shows that closing 5 percent of the gap in the indicator levels between Slovenia and the OECD top performers could be associated with substantial productivity gains, particularly from improving regulatory quality and financial market access.
  - The high growth impact from reducing informality could in part reflect the statistical effect of recording previously unrecorded activity.
  - All estimates should be interpreted with caution due to model specification and estimation uncertainties.

### Conclusions and Policy Recommendations
- Achievements:
  - Slovenia has high per capita income, strong human development indicators, effective institutions, and one of the lowest inequality rates in the world.
  - Strong private investment pre-GFC spurred productivity growth, increased economic complexity, diversified exports, and integration into European value chains.
- Main areas for improvement and policy directions:
  - Innovation:
    - Bolster innovation and technological development, promote investment in ICT and automation, encourage patents and trademarks, and support innovative startups to bridge the gap with EU leaders.
  - Regulatory quality:
    - Make the regulatory framework more growth-friendly by easing administrative burden, streamlining procedures such as building permits, and further digitalizing public services to help private investment and productivity.
    - Reduce market distortions and barriers to entry/exit to increase business dynamism and reallocation.
  - Labor regulations:
    - Further progress on labor market flexibility (e.g., severance costs and broader employer burdens) to support growth and competitiveness.
  - Taxation:
    - Address the high labor tax wedge; lowering the tax wedge on labor could increase labor supply, help growth, and reduce informality.
    - Shift revenue composition away from income toward indirect and property taxes, and reduce tax expenditures to ensure at least revenue neutrality.
    - IMF (2015) estimate: a revenue-neutral reform involving reduction of the tax wedge by 5 percent leads to high long-run growth by 0.2–0.3 percent.
  - Education and skills mismatch:
    - Increase the flexibility of the education system to respond to evolving market needs, in consultation with employers.
    - Use the labor market platform to feedback results to education policies.
  - Capital market development:
    - Implement the government’s capital market development strategy through 2030 to expand access to debt and equity financing, support innovative SMEs, and broaden investment opportunities.
    - Deepening capital markets over time would support private investment and growth.

*Source: IMF staff analysis as presented in the Republic of Slovenia country note.*

### References

### 1svnea2024002 - References

### Major thematic areas in the references
- Production function estimation and identification (Ackerberg et al., 2015; De Loecker and Collard-Wexler, 2016).
- Empirical growth and productivity literature, including cross-country growth, convergence, and endogenous technological change (Barro, 1991; DeLong, 1992; Mankiw et al., 1992; Romer, 1990; Jorgenson, 2011; Jorgenson and Stiroh, 2000).
- Firm-level determinants of productivity: misallocation, market structure, finance, and leverage (Hsieh and Klenow, 2009; Syverson, 2004; Syverson, 2011; Beck et al., 2005; Coricelli et al., 2012; Rajan and Zingales, 2001).
- R&D, human capital, and innovation as drivers of productivity (Guellec and Van Pottelsberghe de la Potterie, 2001; Mc Morrow and Röger, 2009; Égert et al., 2022).
- Policy and institutional influences on productivity: governance, fiscal policy, state-owned enterprises, and public investment efficiency (Kaufmann et al., 2010; International Monetary Fund, 2015; World Bank, 2021; Schwartz et al., 2020).
- Measurement and international perspectives on productivity trends (Dieppe, ed., 2021; Díez et al., 2021; European Central Bank, 2021).

### Appendix I — Shift Share Methodology: key definitions and findings
- Methodology: aggregate labor productivity decomposed into three components:
  - (i) Intra-industry productivity growth effect: sum of productivity growth rates of individual sectors in the absence of changes in employment shares.
  - (ii) Structural shift effect (changes in labor shares): contribution from shifts of labor across sectors (positive when labor moves from low- to high-productivity growth sectors).
  - (iii) Interaction effect: residual term capturing impacts such as TFP changes on labor productivity not measured in the first two components.
- Formal expressions (as presented):
  - LPt = (Σ Yi,t) / (Σ Li,t) (aggregate definition)
  - LPt = Σ (LPi,t * Li,t / Lt) (weighted sum representation)
  - First-difference decomposition and growth-rate expression given in the source (full algebraic expressions preserved in the source text).
- Empirical results for Slovenia:
  - Sectoral labor productivity varies widely; dispersion narrowed over two decades but significant differences remain.
  - The intra-industry effect dominates overall labor productivity dynamics.
  - Reallocation of labor between sectors contributed negatively to productivity growth in Slovenia since 2014, though the effect is very small.
  - Declining employment share in manufacturing (generally more productive) had a negative impact on aggregate productivity growth.
  - Major exception: the pharmaceutical sector—high productivity accompanied by increasing employment share.
  - In services, the shift effect turned negative after the GFC, driven by Professional, Scientific & Technical Services and Finance & Insurance, which recorded higher relative productivity growth but decreasing labor shares.

### Appendix II — Firm Leverage and Productivity Growth: motivation, methods, data, and results
- Motivation:
  - Leverage supplies external resources enabling investment in technologies, expansion, and market entry; trade-off theory predicts a non-monotonic (hump-shaped) relationship between leverage and growth (positive at low leverage, negative beyond a threshold due to debt overhang) (Coricelli et al., 2012).
- Methodology:
  - Threshold regression framework based on Hansen (2000) to identify endogenous threshold γ splitting sample into regimes:
    - yi,t+1 = α1 Li,t + β′Xi,t + εi,t if Li,t ≤ γ
    - yi,t+1 = α2 Li,t + β′Xi,t + εi,t if Li,t > γ
  - Coricelli et al. (2012) approach extended to identify three leverage regions using confidence bounds γ1 and γ2 and indicator functions:
    - yi,t+1 = α1 Li,t I(Li,t ≤ γ1) + α2 Li,t I(γ1 < Li,t ≤ γ2) + α3 Li,t I(Li,t > γ2) + β′Xi,t + εi,t
  - Explanatory variables: intangible fixed assets to total assets, dummy for small firms (employment < 50), dummy for young firms (age ≤ 5).
- Data:
  - Firm-level data from Orbis Bureau van Dijk compiled by IMF Research Department (Díez et al., 2021).
  - Database includes balance sheet, income statement, and employment data for about 100 thousand firms on average since 2010.
  - In 2020, these companies accounted for about 60 percent of registered employment.
  - For leverage measures three definitions used:
    - (i) debt (sum of loans and long-term debt) to total assets;
    - (ii) total liabilities to total assets;
    - (iii) debt to equity.
- Descriptive statistics (post-truncation sample and leverage distribution):
  - Truncated sample contains about 396 thousand observations.
  - Missing long-term debt and loans: 42 percent and 38 percent respectively.
  - For leverage defined as total debt to total assets: about 163 thousand observations (45 percent of total).
  - Average leverage value: about 26 percent (0.26).
  - Summary table excerpts (Firms with zero or positive leverage):
    - Percent of total: 44.54 (All), 40.89 (Small <50 empl.), 2.98 (Medium 50-249 empl.), 0.67 (Large 250+ empl.), 7.81 (Young < 5 years)
    - Mean: 0.26 (All), 0.26 (Small), 0.23 (Medium), 0.24 (Large), 0.27 (Young)
    - St. dev.: 0.20 (All), 0.20 (Small), 0.19 (Medium), 0.18 (Large), 0.21 (Young)
    - p25: 0.08 (All), 0.09 (Small), 0.06 (Medium), 0.09 (Large), 0.08 (Young)
    - p50: 0.23 (All), 0.23 (Small), 0.21 (Medium), 0.22 (Large), 0.24 (Young)
    - p75: 0.40 (All), 0.41 (Small), 0.37 (Medium), 0.37 (Large), 0.44 (Young)
    - Max: 0.76 (All and all size groups)
  - Firms with positive leverage (subset) mean and percent of total:
    - Percent of total: 40.73 (All), 37.4 (Small), 2.7 (Medium), 0.6 (Large), 7.0 (Young)
    - Mean: 0.28 (All), 0.28 (Small), 0.26 (Medium), 0.26 (Large), 0.30 (Young)
    - St. dev.: 0.19 (All), 0.19 (Small), 0.18 (Medium), 0.17 (Large), 0.20 (Young)
    - p25: 0.12 (All), 0.12 (Small), 0.11 (Medium), 0.12 (Large), 0.13 (Young)
    - p50: 0.26 (All), 0.26 (Small), 0.24 (Medium), 0.25 (Large), 0.28 (Young)
    - p75: 0.42 (All), 0.42 (Small), 0.39 (Medium), 0.38 (Large), 0.46 (Young)
    - Max: 0.76 (All and all size groups)
  - Correlations shown in source (excerpt):
    - Leverage — Number of employees: 0.00
    - Leverage — Intangible to total assets: 0.04
    - Leverage — EBIT to total assets: -0.01
- Regression and threshold results (Table A II.2 highlights):
  - OLS and threshold model specifications estimated; main reported coefficients (selected):
    - Initial TFP: -0.00279*** (Model 1), -0.00280*** (2), -0.00380*** (3), -0.00405*** (4), -0.00269*** (5)
    - Small (dummy): -0.00227** (1), -0.00233** (2), -0.00212** (3), -0.00233*** (4), -0.00219** (5)
    - Leverage (overall) (Model 1): 0.0164*** (t statistic 11.29)
    - Leverage (low/medium/high) point estimates (Models 2–5) (selected examples):
      - Leverage (low): 0.0358* (2), 0.0272** (3), 0.00521** (4), -0.00593 (5)
      - Leverage (medium): 0.0158*** (2), 0.0197*** (3), 0.00698*** (4), 0.00303* (5)
      - Leverage (high): 0.0192*** (2), 0.0192*** (3), 0.0124*** (4), 0.00301*** (5)
  - Threshold estimates and confidence intervals (selected):
    - Model (2) threshold estimate: 0.19127; Confidence interval: [0.09040, 0.60726]
    - Model (3) threshold estimate: 0.16047; Confidence interval: [0.13190, 0.18509]
    - Model (4) threshold estimate: 0.74934; Confidence interval: [0.65288, 0.81394]
    - Model (5) threshold estimate: 0.30789; Confidence interval: [0.26572, 0.86484]
  - Sample sizes (N) and adjusted R-squared:
    - N: 107718 (Models 1 and 2), 147474 (3), 206537 (4), 100014 (5)
    - adj. R-sq: 0.002 (1), 0.002 (2), 0.003 (3), 0.003 (4), 0.002 (5)
  - Key empirical interpretations:
    - Simple OLS suggests a positive contribution of leverage to TFP growth in Slovenia and significant convergence and size effects (small firms record lower productivity growth).
    - Threshold models indicate it is difficult to distinguish the differentiated impact of leverage levels robustly; point estimates across leverage bands are often close and equality cannot be rejected at the 5 percent level in many specifications.
    - Exceptions: Model (4) (total liabilities to total assets) shows the coefficient on high leverage significantly larger than low and medium; Model (5) (debt to equity) shows low debt-to-equity ratios associated with weaker productivity growth.
    - Results robust to alternative leverage band splits (e.g., symmetric interval [0.14, 0.24]).
    - Contrasts with Coricelli et al. (2012) who find negative effect of excessive leverage beyond a higher threshold (0.39).
    - Conclusion: given relatively low indebtedness of Slovenian firms, room exists to increase leverage without necessarily harming productivity.
- Decision tree alternative (debt ratio thresholds and implied TFP growth):
  - Decision tree (maximum depth 2) splits at leverage ratios 0.38, then 0.168 and 0.604.
  - Estimated TFP growth by leverage bin:
    - Leverage < 0.168: -0.6 percent
    - Leverage in [0.168, 0.38]: -0.2 percent
    - Leverage in [0.38, 0.604]: 0.1 percent
    - Leverage > 0.604: 0.6 percent
  - Overall finding from the decision tree: productivity growth improves with leverage, though this descriptive result does not condition on other explanatory variables.

### Appendix III — Data and sources for the Stochastic Frontier Analysis (variables and data sources)
- Output: Expenditure-side real GDP at chained PPPs (in million 2017 US$) — Penn World Tables 10.1
- Labor: Number of persons engaged (millions) — Penn World Tables 10.1
- Capital: Capital stock in constant 2017 national prices (in million 2017 US$) — Penn World Tables 10.1
- Output gap: Percent deviation of actual GDP from potential — World Economic Outlook database
- Regulatory quality: Regulatory quality score — Kaufmann, D., Kraay, A. and M. Mastruzzi (2010)
- Employee protection index: Strictness of employment protection – individual and collective dismissals (regular contracts) — OECD
- Financial market access: Financial markets access index — Financial Development Index, IMF
- Income tax share: Share of individual and corporate income taxes in total tax revenue — OECD
- Informal sector share: Share of informal sector as percent of official GDP (MIMIC method) — Elgin et al. (2021)
- Advanced education: Labor force with advanced education (% of total working-age population with advanced education) — World Development Indicators
- Human capital index: Human capital index — Penn World Tables 10.1
- Tertiary education: Percent of adult population with tertiary education — OECD

*Source: 1svnea2024002 - References (PDF chapter/section).*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2024/english/1svnea2024002.pdf_