## wpiea2020055-print-pdf

## Source details

**Canonical URL:** [wpiea2020055-print-pdf](https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020055-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2020/english/wpiea2020055-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2020/english/wpiea2020055-print-pdf.pdf.json)

---

### I. INTRODUCTION
- Regional focus: MENAP (Middle East, North Africa, Afghanistan and Pakistan) and CCA (Caucasus and Central Asia) need higher and more inclusive growth to boost incomes and job creation.
- Key constraints for private sector and SME development:
  - Access to finance is a major constraint (Purfield and al, 2018; IMF, 2018).
  - SMEs in MENAP and CCA contribute relatively little in output and employment compared to other regions.
- Comparative indicators:
  - Average share of loans to SMEs in total bank lending in MENAP and CCA: about 7 percent.
  - World Bank Enterprise Survey:
    - About 32 percent of firms in MENAP report access to credit as a major constraint (world average 26 percent).
    - 18 percent in CCA report access to credit as a major constraint.
  - Registered firms are 54 percent more likely to have a bank account and 32 percent more likely to have loans (Farazi (2014)).
  - Introduction of collateral registries for movable assets can increase the likelihood of firms having access to bank financing by 10 percentage points (Love et al. (2016)).
- Paper objective:
  - Identify main constraints to SME financial inclusion across legal, institutional, macroeconomic, financial sector, and business environment dimensions.
  - Provide priorities and policies most likely to influence SME access to finance in MENAP and CCA.

### II. STYLIZED FACTS
- Data sources:
  - World Bank Enterprise Survey (WBES) — firm-level, covers SMEs and large firms, survey roughly every 4 years.
  - IMF Financial Access Survey (FAS) — annual, limited SME lending data and covers only 7 of the MENAP and CCA countries.
- Empirical patterns (WBES, 2018 or latest):
  - Firms using banks to finance investments: MENAP and CCA score 16 percent (world average 30 percent).
  - MENAP has the lowest share of firms with a bank loan and a checking or savings account; CCA slightly better but lags other regions.
  - Both CCA and MENAP have the lowest percentage of firms financing investment and working capital using banks.
  - Almost 80 percent of firms in MENAP and CCA use internal funds instead of banks.
- SME Financial Inclusion Index (data period: 2006-2017):
  - Indicators (six) divided into Access and Usage:
    - Access: Percent of firms with a bank loan/line of credit; Percent of firms with a checking or savings account.
    - Usage: Percent of firms using banks to finance investments; Percent of firms using banks to finance working capital; Proportion of investments financed by banks; Proportion of working capital financed by banks.
  - Construction steps:
    - Normalization: winsorize series, then min-max normalize to [0, 1].
    - Principal Component Analysis (PCA): first principal component used; first component accounts for around 70 percent of variance.
    - Aggregation: sub-indices aggregated using PCA-derived weights; final index re-normalized to [0, 1]; linear aggregation assumes full compensability.
- Coverage and regional results:
  - Index available for 119 countries; 20 are in MENAP and CCA.
  - Highest SME financial inclusion in regions: Lebanon (LBN), Morocco (MAR), Tunisia (TUN).
  - Lowest: Afghanistan (AFG), Iraq (IRQ), Yemen (YEM).
  - Regional result: MENAP has the lowest level of SME financial inclusion; CCA has the third lowest.

### III. EMPIRICAL FRAMEWORK
- Sample and approach:
  - Sample: around 123 countries (Appendix table 9).
  - Objective: identify correlations (not causality) between fundamentals and SME financial inclusion.
  - Baseline specification uses FI = SME financial inclusion index and controls vector X.
- Baseline controls (expected signs):
  - Total investment (percent of GDP) — expected positive.
  - Inflation rate — expected negative.
  - SME share of employment — ambiguous.
  - Income level dummies — FI expected to improve with development.
  - Region_dummy for MENAP or CCA — expected negative.
- Expanded variable groups and expectations:
  - Macroeconomic environment:
    - Diversification (OECD complexity index) — expected positive.
    - Competition (WEF domestic competition index) — expected positive.
    - Informality (shadow economy percent of GDP) — expected negative.
    - Infrastructure (fixed telephone lines per 100 inhabitants proxy) — expected positive.
    - Public investment (percent of total investment) — crowding-out hypothesis implies negative; could be positive if supportive infrastructure.
    - Oil exporter dummy — expected negative.
  - Institutional quality:
    - Voice and accountability; Political stability; Government effectiveness; Control of corruption — expected positive.
  - Financial sector characteristics and regulation:
    - Bank profitability (return on equity) — higher profitability may reduce incentive to lend to SMEs (negative).
    - Asset quality (NPLs to total loans) — higher NPLs expected negative.
    - Bank deposits (%GDP) — expected positive.
    - Bank Z-score — expected positive.
    - Bank concentration (3-bank asset concentration percent) — expected negative.
    - Financial regulation indicators: regulatory agency capacity; regulatory and supervisory capacity of deposit-taking activities; regulatory and supervisory capacity for financial inclusion.
  - Business environment:
    - Tax burden (tax payments % of firm profit) and business start-up cost (% of GNI per capita) — expected negative.
    - Contract enforcement (days) and property rights index — expected positive.
    - Property registration cost (percent of property value) — expected negative.
    - Credit information (credit bureau coverage) — expected positive.
- Estimation strategy and limitations:
  - Equations estimated using OLS panel fixed effects.
  - Variables z and interaction terms added one at a time to limit multicollinearity.
  - Data limitations and endogeneity acknowledged; results interpreted as indicative of direction and strength, not definitive causal estimates.
  - WBES time-series limitations: few non-consecutive years per country limit dynamic methods like GMM.

### IV. RESULTS — Overview
- Main empirical messages (correlational):
  - Macroeconomic stability and an investment-friendly environment are important for SME financial inclusion.
  - Limited public sector size helps avoid crowding out SME access to credit.
  - Financial sector soundness, competitive banking systems, and openness support SME inclusion.
  - Institutional factors (governance, regulatory and supervisory capacity, credit information) matter strongly.
  - Supportive business environment: modern collateral and insolvency frameworks, enforceable property rights and contracts.
- Caution: results are correlational due to methodological constraints.

### V. APPENDIX REGRESSION SUMMARIES — Main empirical findings and selected coefficients
- Baseline model:
  - Investment is positively correlated with SME financial inclusion.
  - Inflation is negatively correlated; negative relationship slightly stronger for CCA countries.
  - SME share of employment is negatively correlated with SME access to finance.
  - MENAP and CCA dummies show negative correlations with SME financial inclusion.
  - SME financial inclusion tends to increase with income level.
- Macrofinancial environment, competition and structure:
  - Strong positive correlations: economic competition, diversification, and SME financial inclusion.
  - Competition relation is significantly stronger for MENAP and CCA compared to sample average.
  - Infrastructure positively correlated; informality and public investment share negatively correlated.
  - Elasticity examples:
    - A 1 percent increase in public investment may lead to a 0.7 percent decline in SME financial inclusion.
    - A 1 percent increase in public investment is associated with a 0.2 percent decline for emerging markets and developing countries, on average.
  - Oil-exporting countries have on average lower SME financial inclusion.
- Institutions and governance:
  - Strong positive correlations: government effectiveness and control of corruption (coefficients almost twice larger for MENAP relative to sample average).
  - Political stability and voice & accountability positively correlated; voice & accountability relationship stronger for CCA.
  - Institutional coefficients generally larger than most other variables, implying substantial potential impact from institutional improvements.
- Financial sector characteristics:
  - Positive correlations: bank Z-score (stability), bank deposits (%GDP), regulatory capacity.
  - Negative correlations: bank return on equity (profitability) and NPLs (% gross loans).
- Business environment and legal/contractual factors:
  - Negative correlations: tax burden (% of profit), time required to enforce a contract, cost to register property, cost of business start-up procedures.
  - Positive correlations: property rights, public credit registry coverage.
- Sample-average coefficient estimates (Figures 8 and 9):
  - Figure 8 (Macrofinancial environment): Infrastructure: 0.279; Competition: 0.141; Diversification: 0.351; Informality: 0.222; Inflation: 0.067; Public investment (%total): 0.093; Government effectiveness: 0.086; Bank deposit (%GDP): 0.086; Bank Z-score: 0.153.
  - Figure 9 (Institutions, business environment): Voice & accountability: 0.07; Political stability: 0.07; Control of corruption: 0.071; Public credit registry coverage: 0.047; Regulatory and supervisory capacity for financial inclusion: 0.239; Regulatory and supervisory capacity of deposit-taking activities: 0.129; Capacity of regulatory agency: 0.119; Cost of business start-up procedures: 0.116; Cost to register a property: 0.102; Time required to enforce a contract: 0.102; Property rights: 0.041; Tax burden (% of profit): 0.122; Additional reported coefficient: 0.064.
  - Note: Coefficients from equations (1) and (2), based on OLS panel fixed effects; statistically significant at a minimum level of 10%, with robust standard errors.

### VI. APPENDIX TABLE HIGHLIGHTS (selected numeric results)
- Appendix table 1 (MENA macroeconomic determinants) — example coefficients:
  - Investment (%GDP): 0.005***, 0.003, 0.003, 0.004*, 0.002, 0.002, 0.005***, 0.002, 0.003, 0.004*, 0.002, 0.004*.
  - Inflation: -0.007***, -0.004*, -0.006***, -0.003, -0.004, -0.004, -0.007***, -0.004*, -0.006***, -0.003, -0.004, -0.007***.
  - MENA dummy examples: -0.167***, -0.115*, -0.184***, -0.106, -0.143**, -0.147**, -0.155***, -0.078, -0.039, -0.270**, -1.592***, -0.127.
  - Eco. Diversification: 0.086*** and 0.084***; Informality: -0.005***; Infrastructure: 0.007***; Public investment (% total): -0.002**; Oil exporters: -0.085**.
  - Interaction: MENA*Competition: 0.351*** (0.133).
  - Observations by column: 190 121 148 124 124 122 190 121 148 124 124 122.
  - R-squared examples: 0.158 0.250 0.210 0.267 0.099 0.273 0.174 0.252 0.211 0.280 0.123 0.275.
- Appendix table 2 (CCA macroeconomic determinants) — example coefficients:
  - Investment (%GDP): 0.005***, 0.003, 0.004*, 0.004**, 0.003, 0.005**, 0.003, 0.004*, 0.004*, 0.003, 0.005**.
  - Inflation: -0.008***, -0.005**, -0.007***, -0.003, -0.005, -0.007***.
  - CCA dummy examples: -0.078**, 0.002, -0.052, -0.128***, -0.071**, -0.014, 0.022, -0.426***, -0.062, -0.734***, -0.014.
  - Eco. Diversification: 0.093***; Informality: -0.004**; Infrastructure: 0.008***; Public investment (% total): -0.002*.
  - Interaction: CCA*Informality: 0.010*** (0.002); CCA*Competition: 0.153** (0.063).
  - Observations by column: 190 121 148 124 124 122 121 148 124 124 122.
  - R-squared examples: 0.118 0.227 0.165 0.264 0.068 0.210 0.227 0.171 0.265 0.070 0.210.
- Appendix tables 3–4 (Financial sector and regulatory characteristics) — selected:
  - Bank return on equity: -0.002* (MENA); -0.001 (CCA).
  - Bank deposit (%GDP): 0.005*** and 0.004*** (MENA); 0.004*** (CCA).
  - NPLs (% gross loans): -0.006* (MENA); -0.008*** (CCA).
  - Bank Z-score: 0.005** (MENA); 0.003 (CCA).
  - MENA interaction significance: MENA*Bank Z-score: 0.009***; MENA*Capacity of regulatory agency: 0.016***; MENA*Reg and Sup capacity for FI: 0.047***; MENA*Reg and Sup capacity of deposit-taking activities: 0.014***.
- Appendix tables 5–6 (Quality of institutions) — selected governance coefficients:
  - Voice & accountability: 0.102*** (MENA); 0.114*** (CCA).
  - Political stability: 0.050** (MENA); 0.064*** (CCA).
  - Government effectiveness: 0.131*** (MENA); 0.152*** (CCA).
  - Control of corruption: 0.123*** (MENA); 0.127*** (CCA).
  - Interaction examples: MENA*Political stability: 0.081**; CCA*Gov effectiveness: 0.138* (one specification).
- Appendix tables 7–8 (Business environment) — selected coefficients:
  - Total tax rate (%profit): -0.000* and -0.001** (MENA).
  - Cost of business start-up procedures: -0.001*** (MENA and CCA).
  - Cost to register property: -0.011*** and -0.013*** (MENA); -0.012*** (CCA).
  - Time required to enforce a contract: -0.000* (MENA); -0.000** (CCA).
  - Public credit registry coverage: 0.003** (MENA and CCA).
  - Property rights: 0.070*** (MENA); 0.069*** (CCA).
  - MENA interaction: MENA*Cost to register property: 0.036*** (0.011).
- Model notes:
  - Robust standard errors in parentheses throughout.
  - ***, **, * indicate statistical significance at 10, 5, and 1 percent levels, respectively.
  - Dependent variable across tables: SME financial inclusion index.

### VII. POLICY RECOMMENDATIONS AND IMPLICATIONS
- Key guiding principles for country-customized approaches:
  - (i) a sound macroeconomic environment, in particular economic competition and macroeconomic stability;
  - (ii) better institutional quality, including improved governance;
  - (iii) financial sector soundness, including through strong supervisory and regulatory frameworks and competition;
  - (iv) an enabling business environment, cutting across legal, regulatory and tax issues.
- Additional policy cautions:
  - Higher financial inclusion could be associated with lower safety buffers for banks; additional steps may be needed to guarantee financial stability.
- Suggested future research:
  - Investigate the tradeoff between financial stability and financial inclusion.
  - Examine the role of demand versus supply factors in explaining low levels of SME financial inclusion.

### VIII. SAMPLE (Appendix table 9)
- Countries listed in sample (verbatim as provided): Afghanistan, Albania, Angola, Antigua and Barbuda, Argentina, Armenia, Azerbaijan, Bahamas, The, Bangladesh, Barbados, Belarus, Belize, Benin, Bhutan, Bolivia, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Burkina Faso, Burundi, Cambodia, Cameroon, Central African Republic, Chad, Chile, China, Colombia, Congo, Dem. Rep., Congo, Rep., Costa Rica, Cote d'Ivoire, Croatia, Czech Republic, Djibouti, Dominica, Dominican Republic, Ecuador, Egypt, Arab Rep., El Salvador, Eritrea, Estonia, Ethiopia, Fiji, Gabon, Gambia, The, Georgia, Ghana, Guatemala, Guinea, Guinea-Bissau, Guyana, Honduras, Hungary, India, Indonesia, Israel, Jamaica, Jordan, Kazakhstan, Kenya, Kyrgyz Republic, Latvia, Lebanon, Lesotho, Lithuania, Madagascar, Malawi, Malaysia, Mali, Mauritania, Mauritius, Mexico, Moldova, Mongolia, Montenegro, Morocco, Mozambique, Myanmar, Namibia, Nepal, Nicaragua, Niger, Nigeria, Pakistan, Panama, Paraguay, Peru, Philippines, Poland, Romania, Russian Federation, Rwanda, Samoa (Solomon Islands appears as Solomon Islands), Sierra Leone, Slovak Republic, Slovenia, Solomon Islands, South Africa, Sri Lanka, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, Sudan, Suriname, Swaziland, Tajikistan, Tanzania, Thailand, Timor-Leste, Togo, Tonga (not listed in source), Tunisia, Turkey, Turkmenistan, Uganda, Ukraine, Uruguay, Uzbekistan, Vanuatu, Venezuela, RB, Vietnam, West Bank and Gaza, Yemen, Rep., Zambia, Zimbabwe.

*Source: https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020055-print-pdf.pdf*

### References .............................................................................................................

### wpiea2020055-print-pdf - References

### I. INTRODUCTION
- Regional focus: MENAP (Middle East, North Africa, Afghanistan and Pakistan) and CCA (Caucasus and Central Asia) need higher and more inclusive growth to boost incomes and job creation.
- Key constraints identified for private sector and SME development:
  - Access to finance is a major constraint (Purfield and al, 2018; IMF, 2018).
  - SMEs in MENAP and CCA contribute relatively little in output and employment compared to other regions.
- Comparative indicators and findings:
  - The average share of loans to SMEs in total bank lending in the MENAP and CCA regions is about 7 percent.
  - According to the World Bank Enterprise Survey:
    - About 32 percent of firms in the MENAP region report access to credit as a major constraint (world average 26 percent).
    - 18 percent in the CCA region report access to credit as a major constraint.
  - Registered firms are 54 percent more likely to have a bank account and 32 percent more likely to have loans (Farazi (2014)).
  - Introduction of collateral registries for movable assets can increase the likelihood of firms having access to bank financing by 10 percentage points (Love et al. (2016)).
- Paper objective and contribution:
  - Identify main constraints to SME financial inclusion across legal, institutional, macroeconomic, financial sector, and business environment dimensions.
  - Provide priorities and policies most likely to influence SME access to finance in MENAP and CCA.

### II. STYLIZED FACTS
- Data sources:
  - World Bank Enterprise Survey (WBES) — firm-level, covers SMEs and large firms, survey roughly every 4 years.
  - IMF Financial Access Survey (FAS) — annual, limited SME lending data and covers only 7 of the MENAP and CCA countries.
- Key empirical patterns from WBES (2018 or latest):
  - MENAP and CCA score lowest in number of firms using banks to finance investments: 16 percent (world average 30 percent).
  - MENAP has the lowest share of firms with a bank loan and a checking or savings account; CCA slightly better but still lags Asia, Europe, and Latin America.
  - Both CCA and MENAP have the lowest percentage of firms financing investment and working capital using banks.
  - Almost 80 percent of firms in MENAP and CCA use internal funds instead of banks.
- SME Financial Inclusion Index — construction overview:
  - Data period: 2006-2017.
  - Indicators (six) divided into Access and Usage:
    - Access: Percent of firms with a bank loan/line of credit; Percent of firms with a checking or savings account.
    - Usage: Percent of firms using banks to finance investments; Percent of firms using banks to finance working capital; Proportion of investments financed by banks; Proportion of working capital financed by banks.
  - Steps:
    - Normalization: winsorize series, then min-max normalize to [0, 1].
    - Principal Component Analysis (PCA): first principal component used; first component accounts for around 70 percent of variance.
    - Aggregation: sub-indices aggregated using PCA-derived weights; final index re-normalized to [0, 1]; linear aggregation assumes full compensability.
- Coverage and regional results:
  - Index available for 119 countries worldwide; 20 are in MENAP and CCA.
  - Country-level outcomes in the regions: highest SME financial inclusion — Lebanon (LBN), Morocco (MAR), Tunisia (TUN); lowest — Afghanistan (AFG), Iraq (IRQ), Yemen (YEM).
  - Regional result: MENAP has the lowest level of SME financial inclusion; CCA has the third lowest.

### III. DRIVERS OF SME FINANCIAL INCLUSION — EMPIRICAL FRAMEWORK
- Sample and approach:
  - Sample: around 123 countries (Appendix table 9).
  - Purpose: identify correlations (not necessarily causality) between fundamentals and SME financial inclusion.
  - Baseline empirical specification (equation form presented in source) with FI = SME financial inclusion index.
- Baseline control variables (vector X) and expected signs:
  - Total investment (percent of GDP) — expected positive correlation with FI.
  - Inflation rate — expected negative correlation with FI.
  - SME share of employment (in total employment) — ambiguous a priori.
  - Income level dummies — FI expected to improve with development.
  - Region_dummy for MENAP or CCA — expected negative correlation with FI.
- Expanded variables (z) grouped into four categories with expected relationships:

  - Macroeconomic environment:
    - Diversification (OECD complexity index) — expected positive.
    - Competition (World Economic Forum domestic competition index) — expected positive.
    - Informality (size of the ‘shadow economy’ in percent of GDP) — expected negative.
    - Infrastructure (fixed telephone lines per 100 inhabitants proxy) — expected positive.
    - Public investment (percent of total investment) — crowding-out hypothesis suggests negative correlation; could be positive if public investment supports infrastructure.
    - Oil exporter dummy — expected negative for SME inclusion where oil dominates.
    - Empirical regional patterns (Figure 4):
      - CCA: larger informal sector on average.
      - MENAP and CCA: lower diversification and infrastructure compared to most regions.
      - MENAP: larger public sector on average versus the rest of the sample.

  - Institutional quality:
    - Voice and accountability; Political stability; Government effectiveness; Control of corruption — all expected positive for FI.
    - Expectation: stronger positive correlation for MENAP and CCA given lower institutional scores (Figure 5).

  - Financial sector characteristics and regulation:
    - Bank profitability (return on equity) — higher profitability may reduce incentives to lend to SMEs (negative).
    - Asset quality (NPLs to total loans) — higher NPLs expected negative.
    - Bank deposits (deposit to GDP ratio) — higher deposits expected positive.
    - Banking sector stability (bank Z-score) — higher stability expected positive.
    - Bank concentration (3-bank asset concentration percent) — higher concentration expected negative.
    - Financial regulation indicators controlled: regulatory agency capacity; regulatory and supervisory capacity of deposit-taking activities; regulatory and supervisory capacity for financial inclusion.
    - Regional patterns (Figure 6):
      - MENAP and CCA perform better than APD and WHD on asset quality but lag AFR and EUR.
      - CCA lags in banking stability and deposit ratios.
      - MENAP has the most concentrated banking sectors on average.

  - Business environment:
    - Tax burden (tax payments percent of firm profit) and business start-up cost (percent of GNI per capita) — higher values expected negative.
    - Contract enforcement (days to enforce a contract) and property rights index — better outcomes expected positive.
    - Property registration cost (percent of property value) — higher cost expected negative.
    - Credit information (credit bureau coverage) — broader coverage expected positive.
    - Regional patterns (Figure 7):
      - MENAP and CCA lag in credit information availability and strength of legal rights.
      - MENAP and CCA perform better than peers on business taxation on average.
      - MENAP has higher business start-up costs on average compared to EUR and APD.

- Estimation strategy and limitations:
  - Equations 5 and 6 estimated using OLS fixed effects.
  - Variables z and interaction terms added one at a time to limit multicollinearity.
  - Acknowledged data limitations and endogeneity concerns; results interpreted as indicative of direction and strength of relationships, not definitive causal estimates.
  - Time-series limitations: enterprise survey covers few non-consecutive years per country, limiting dynamic methods like GMM.

### IV. RESULTS (summary statements from source)
- The empirical analysis aims to establish correlations between control variables and SME financial inclusion across the global sample and to assess relative importance for MENAP and CCA via interaction terms.
- Findings emphasize the importance of:
  - Macroeconomic stability and investment-friendly environment.
  - Limited public sector size to avoid crowding out SME access to credit.
  - Financial sector soundness, competitive banking systems, and open economies.
  - Institutional factors: governance, regulatory and supervisory capacity, credit information availability.
  - Supportive business environment: modern collateral and insolvency frameworks, enforceable property rights and contracts.
- Caution: empirical results are correlational due to data and methodological constraints described above.

*Source: https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020055-print-pdf.pdf*

### Appendix tables 1 through 8 provide the regression results. Figures 8 and 9 summarize the

### Appendix tables 1–8 and Figures 8–9 (regression summary)

### Main empirical findings
- Coefficients associated to the variables in the baseline model are statistically significant with the expected sign in most specifications.
- Investment is positively correlated with SME financial inclusion.
- Inflation is negatively correlated with SME financial inclusion; this negative relationship appears slightly stronger for CCA countries.
- The share of employment in the SME sector is negatively correlated with SME access to finance—likely reflecting that low-income countries often have a larger, more financially constrained SME sector.
- MENAP and CCA dummies show negative correlations with SME financial inclusion, indicating SMEs in both regions are on average more constrained in access to formal financial services.
- SME financial inclusion tends to increase with income level (income level dummies show positive associations).

### Macrofinancial environment, competition and structure
- Strong positive and statistically significant correlations:
  - Economic competition, diversification, and SME financial inclusion.
  - The relationship with competition is significantly stronger for MENAP and CCA compared to the sample average.
- Infrastructure is positively correlated with SME access to finance.
- Negative correlations:
  - Informality and the share of public investment are negatively associated with SME financial inclusion.
  - A large public sector is associated with lower SME financial inclusion.
- Elasticity estimates:
  - A 1 percent increase in public investment may lead to a 0.7 percent decline in SME financial inclusion.
  - A 1 percent increase in public investment is associated with a 0.2 percent decline for emerging markets and developing countries, on average.
- Oil-exporting countries are found to have on average a lower degree of SME financial inclusion.

### Institutions and governance
- Institutional variables show strong positive correlations with SME financial inclusion:
  - Government effectiveness and control of corruption are positively correlated; the relationship is stronger for MENAP and CCA (estimated coefficients almost twice larger for MENAP relative to the sample average).
  - Political stability is positively correlated with SME access to credit, particularly relevant for MENAP and CCA.
  - Voice and accountability is positively correlated with SME financial inclusion, with a stronger relationship for CCA.
- Overall, estimated coefficients for institutional variables are larger compared to most other variables in the model, suggesting improvements in institutional quality can have relatively large impacts on SME access to finance.

### Financial sector characteristics
- Positive correlations:
  - Banking sector stability (bank z-score) and SME financial inclusion.
  - Higher bank deposits ratio (bank deposit (%GDP)) associated with increased SME access to formal finance.
  - Regulatory capacity is associated with improved SME access to finance (effective regulatory and supervisory framework contributes to SME financial inclusion).
- Negative correlations:
  - Bank profitability (bank return on equity) tends to be associated with lower SME financing.
  - NPLs (nonperforming loans) are negatively correlated with SME financial inclusion, indicating lower asset quality restricts SME access to credit.

### Business environment and legal/contractual factors
- Negative correlations with SME financial inclusion:
  - Restrictive tax system (Tax burden (% of profit)).
  - Costs for contract enforcement (Time required to enforce a contract).
  - Property registry costs (Cost to register a property).
  - Cost of business start-up procedures (Cost of business start-up procedures).
- Positive correlations:
  - Property rights and availability of credit information (Public credit registry coverage) improve SME access to finance by reducing uncertainties and easing availability of collateral and borrower information.

### Selected coefficient estimates reported in Figures 8 and 9
- Figure 8 (Macrofinancial environment and SME financial inclusion) — sample-average coefficients shown:
  - Infrastructure: 0.279
  - Competition: 0.141
  - Diversification: 0.351
  - Informality: 0.222
  - Inflation: 0.067
  - Public investment (%total): 0.093
  - Government effectiveness: 0.086
  - Bank deposit (%GDP): 0.086
  - Bank Z-score: 0.153
- Figure 9 (Institutions, business environment and SME financial inclusion) — sample-average coefficients shown:
  - Voice & accountability: 0.07
  - Political stability: 0.07
  - Control of corruption: 0.071
  - Public credit registry coverage: 0.047
  - Regulatory and supervisory capacity for financial inclusion: 0.239
  - Regulatory and supervisor capacity of deposit-taking activities: 0.129
  - Capacity of regulatory agency: 0.119
  - Cost of business start-up procedures: 0.116
  - Cost to register a property: 0.102
  - Time required to enforce a contract: 0.102
  - Property rights: 0.041
  - Tax burden (% of profit): 0.122
  - (Additional reported coefficient) 0.064
- Note: Coefficient estimates from equations (1) and (2), based on OLS panel fixed effects. The coefficients are statistically significant at a minimum level of 10%, with robust standard errors.

### Policy recommendations and implications (conclusion)
- A comprehensive, country-customized approach can catalyze SME access to finance; key guiding principles include prioritizing:
  - (i) a sound macroeconomic environment, in particular economic competition and macroeconomic stability;
  - (ii) better institutional quality, including improved governance;
  - (iii) financial sector soundness, including through strong supervisory and regulatory frameworks and competition;
  - (iv) an enabling business environment, cutting across legal, regulatory and tax issues.
- Policymakers should be aware that higher financial inclusion could be associated with lower safety buffers for banks; additional steps may be needed to guarantee financial stability.
- Future research directions suggested:
  - Investigate the tradeoff between financial stability and financial inclusion.
  - Examine the role of demand versus supply factors in explaining low levels of SME financial inclusion.

*Sources: Appendix tables 1–8 and Figures 8–9, IMF staff estimates.*

### Appendix table 1: SME financial inclusion and the macroeconomic characteristics -

### Appendix table 1: SME financial inclusion and the macroeconomic characteristics - MENA

### Key findings (macroeconomic determinants)
- Investment (%GDP): coefficients reported across columns include 0.005***, 0.003, 0.003, 0.004*, 0.002, 0.002, 0.005***, 0.002, 0.003, 0.004*, 0.002, 0.004* (robust standard errors: ( 0.002)( 0.003)( 0.002)( 0.002)( 0.002)( 0.002)( 0.002)( 0.003)( 0.002)( 0.002)( 0.002)( 0.002))
- Inflation: coefficients include -0.007***, -0.004*, -0.006***, -0.003, -0.004, -0.004, -0.007***, -0.004*, -0.006***, -0.003, -0.004, -0.007*** (robust standard errors: ( 0.002)( 0.002)( 0.002)( 0.002)( 0.003)( 0.003)( 0.002)( 0.002)( 0.002)( 0.002)( 0.003)( 0.002))
- SME share of employment: coefficients include -0.001, -0.002, -0.002**, -0.002**, -0.001, -0.001, -0.001, -0.001, -0.002**, -0.002**, -0.001, -0.003*** (robust standard errors: ( 0.001) repeated)
- MENA dummy: coefficients include -0.167***, -0.115*, -0.184***, -0.106, -0.143**, -0.147**, -0.155***, -0.078, -0.039, -0.270**, -1.592***, -0.127 (robust standard errors: ( 0.044)( 0.061)( 0.062)( 0.065)( 0.066)( 0.063)( 0.044)( 0.091)( 0.231)( 0.114)( 0.566)( 0.093))
- Additional controls reported in later columns: Eco. Diversification 0.086*** and 0.084*** (0.023)(0.024); Informality -0.005***-0.005*** (0.002)(0.002); Infrastructure 0.007***0.007*** (0.001)(0.001); Eco. Competition 0.067*0.053 (0.039)(0.040); Public investment (% total) -0.002**-0.002 (0.001)(0.001); Oil exporters -0.085** (0.037)

### Interaction effects (MENA interactions)
- MENA*Eco. Diversification: 0.043 (0.076)
- MENA*Informality: -0.005 (0.009)
- MENA*Infrastructure: 0.026 (0.018)
- MENA*Competition: 0.351*** (0.133)
- MENA*Public investment: -0.002 (0.002)

### Model fit and sample
- Constant terms across columns: 0.431***, 0.509***, 0.659***, 0.388***, 0.206, 0.572***, 0.439***, 0.505***, 0.657***, 0.385***, 0.257, 0.567*** (robust SEs reported)
- Observations by column: 190 121 148 124 124 122 190 121 148 124 124 122
- R-squared by column: 0.158 0.250 0.210 0.267 0.099 0.273 0.174 0.252 0.211 0.280 0.123 0.275
- Adjusted R-squared by column: 0.140 0.218 0.182 0.236 0.0608 0.242 0.151 0.212 0.178 0.243 0.0784 0.238

### Notes
- Robust standard errors in parentheses. Income levels dummies included but not reported.
- ***, **, * indicate statistical significance at 10, 5, and 1 percent levels, respectively.
- Dependent variable: SME financial inclusion index

*Source: Appendix table 1 of the provided PDF content.*

---

### Appendix table 2: SME financial inclusion and the macroeconomic characteristics - CCA

### Key findings (macroeconomic determinants for CCA)
- Investment (%GDP): coefficients include 0.005***, 0.003, 0.004*, 0.004**, 0.003, 0.005**, 0.003, 0.004*, 0.004*, 0.003, 0.005** (robust SEs reported as ( 0.002)( 0.003) etc.)
- Inflation: coefficients include -0.008***, -0.005**, -0.007***, -0.003, -0.005, -0.007***, -0.005**, -0.007***, -0.003, -0.005, -0.007*** (robust SEs repeated)
- SME share of employment: coefficients include -0.001*, -0.002, -0.002**, -0.002*, -0.001, -0.003***, -0.002, -0.002**, -0.002*, -0.001, -0.003*** (robust SEs repeated)
- CCA dummy: coefficients include -0.078**, 0.002, -0.052, -0.128***, -0.071**, -0.014, 0.022, -0.426***, -0.062, -0.734***, -0.014 (robust SEs: ( 0.031)( 0.043)( 0.046)( 0.038)( 0.034)( 0.031)( 0.065)( 0.092)( 0.066)( 0.266)( 0.031))
- Additional controls: Eco. Diversification 0.093***0.093*** (0.023)(0.023); Informality -0.004**-0.004** (0.002)(0.002); Infrastructure 0.008***0.008*** (0.001)(0.001); Eco. Competition 0.069*0.067* (0.040)(0.040); Public investment (% total) -0.002*-0.002* (0.001)(0.001)

### Interaction effects (CCA interactions)
- CCA*Eco. Diversification: 0.022 (0.062)
- CCA*Informality: 0.010*** (0.002)
- CCA*Infrastructure: -0.003 (0.004)
- CCA*Competition: 0.153** (0.063)
- CCA*Public investment: 0.000 (0.000)

### Model fit and sample
- Constant terms: 0.422***, 0.482***, 0.604***, 0.365***, 0.174, 0.539***, 0.482***, 0.617***, 0.365***, 0.182, 0.539*** (robust SEs reported)
- Observations by column: 190 121 148 124 124 122 121 148 124 124 122
- R-squared by column: 0.118 0.227 0.165 0.264 0.068 0.210 0.227 0.171 0.265 0.070 0.210
- Adjusted R-squared by column: 0.0989 0.194 0.135 0.233 0.0288 0.176 0.187 0.136 0.227 0.0220 0.176

### Notes
- Robust standard errors in parentheses. Income levels dummies included but not reported.
- ***, **, * indicate statistical significance at 10, 5, and 1 percent levels, respectively.
- Dependent variable: SME financial inclusion index

*Source: Appendix table 2 of the provided PDF content.*

---

### Appendix tables 3–4: Financial sector and regulatory characteristics (MENA and CCA)

### MENA (Appendix table 3) — selected coefficients and interactions
- Investment (%GDP): many columns report 0.005***, 0.005***, 0.003, 0.003, 0.003*, 0.007*, 0.006*, 0.005, 0.005***, 0.003, 0.003, 0.003*, 0.004, 0.003, 0.003 (with reported robust SEs)
- Inflation: examples include -0.007***, -0.006**, -0.001, -0.010**, -0.007***, and other coefficients with various SEs
- Bank return on equity: -0.002* and -0.002 (0.001)(0.001)
- Bank deposit (%GDP): 0.005*** and 0.004*** (0.001)(0.001)
- NPLs (% gross loans): -0.006* and -0.005 (0.003)(0.003)
- Bank Z-score: 0.005** and 0.004* (0.002)(0.002)
- Regulatory capacities: Capacity of regulatory agency -0.001-0.001 (0.001)(0.001); Reg and Sup capacity for FI -0.000-0.000 (0.002)(0.002); Reg and Sup capacity of deposit-taking activities 0.003**0.003** (0.001)(0.001)
- MENA interaction effects notable and significant:
  - MENA*Bank Z-score: 0.009*** (0.003)
  - MENA*Capacity of regulatory agency: 0.016*** (0.003)
  - MENA*Reg and Sup capacity for FI: 0.047*** (0.008)
  - MENA*Reg and Sup capacity of deposit-taking activities: 0.014*** (0.002)
- Observations vary by specification (examples: 190 159 161 112 160 323 232 ...)
- R-squared examples: 0.158 0.165 0.360 0.143 0.192 0.309 0.306 0.420 0.165 0.360 0.145 0.204 0.417 0.409 0.501

### CCA (Appendix table 4) — selected coefficients and interactions
- Investment (%GDP): columns report 0.005***, 0.005***, 0.004**, 0.003, 0.004**, 0.007*, 0.006*, 0.005, 0.005***, 0.004**, 0.003, 0.004**, 0.007**, 0.007*, 0.003 (robust SEs reported)
- Inflation: -0.008***, -0.006**, -0.003, -0.011**, -0.008***, -0.004, and others
- Bank return on equity: -0.001-0.001 (0.001)(0.001)
- Bank deposit (%GDP): 0.004***0.004*** (0.001)(0.001)
- NPLs (% gross loans): -0.008***-0.008*** (0.003)(0.003)
- Bank Z-score: 0.0030.003 (0.002)(0.002)
- Reg and Sup capacity of deposit-taking activities: 0.003**0.004** (0.001)(0.002)
- Interaction coefficients with CCA reported but generally not significant in many specifications (examples: CCA*Bank return on equity 0.002 (0.006); CCA*Bank deposit (%GDP)0.002 (0.004))
- Observations and R-squared vary across specifications (examples: Observations 190 159 161 112 160 323 232 159 161 112 160 323 232; R-squared examples 0.118 0.111 0.285 0.128 0.117 0.319 0.311 0.416 0.112 0.285 0.128 0.118 0.382 0.341 0.457)

### Notes
- Robust standard errors in parentheses. Income levels dummies included but not reported.
- ***, **, * indicate statistical significance at 10, 5, and 1 percent levels, respectively.
- Dependent variable: SME financial inclusion index

*Source: Appendix tables 3–4 of the provided PDF content.*

---

### Appendix tables 5–6: Quality of institutions (MENA and CCA)

### MENA (selected institutional coefficients)
- Investment (%GDP): 0.004** 0.005*** 0.003** 0.003** 0.003** 0.005*** 0.003 0.003** 0.003* (with SEs)
- Inflation: -0.005*** -0.002 -0.003* -0.003 -0.003* -0.002 -0.003* -0.002 -0.003
- SME share of employment: mostly -0.001 to -0.000 range
- MENA dummy: -0.156*** -0.087** -0.107** -0.130*** -0.134*** -0.002 -0.001 -0.031 -0.048 (SEs reported)
- Governance indicators with positive, significant coefficients:
  - Voice & accountability: 0.102*** 0.101*** (0.020)(0.020)
  - Political stability: 0.050** 0.041* (0.021)(0.022)
  - Gov effectiveness: 0.131*** 0.126*** (0.029)(0.029)
  - Control of corruption: 0.123*** 0.119*** (0.024)(0.025)
- MENA interaction terms: MENA*Political stability 0.081** (0.036); MENA*Gov effectiveness 0.133* (0.075); MENA*Control of corruption 0.120* (0.069)

### CCA (selected institutional coefficients)
- Investment (%GDP): 0.005** 0.005*** 0.003* 0.004*** 0.003** 0.005*** 0.003* 0.004** 0.003** (with SEs)
- Inflation: -0.006*** -0.002 -0.003* -0.003 -0.004** -0.002 -0.003* -0.002 -0.004**
- SME share of employment: generally -0.001 to -0.001** (small negative)
- CCA dummy: -0.081* 0.008 -0.068* -0.064*** -0.027 -0.043 -0.077 -0.013 -0.058* (SEs reported)
- Governance indicators with positive, significant coefficients:
  - Voice & accountability: 0.114*** 0.116*** (0.019)(0.020)
  - Political stability: 0.064*** 0.064*** (0.019)(0.019)
  - Gov effectiveness: 0.152*** 0.141*** (0.018)(0.019)
  - Control of corruption: 0.127*** 0.129*** (0.025)(0.026)
- CCA*Gov effectiveness: 0.138* (0.082) reported in one specification

### Model fit and sample
- Observations: 189 184 repeated across specifications
- R-squared examples (MENA): 0.321 0.392 0.335 0.384 0.406 0.394 0.344 0.390 0.411
- R-squared examples (CCA): 0.288 0.381 0.323 0.363 0.379 0.382 0.323 0.363 0.379
- Adjusted R-squared reported for each specification

### Notes
- Robust standard errors in parentheses. Income levels dummies included but not reported.
- ***, **, * indicate statistical significance at 10, 5, and 1 percent levels, respectively.
- Dependent variable: SME financial inclusion index

*Source: Appendix tables 5–6 of the provided PDF content.*

---

### Appendix tables 7–8: Business environment (MENA and CCA)

### MENA (selected business environment coefficients)
- Investment (%GDP): coefficients include 0.005***, 0.006***, 0.005***, 0.005***, 0.006***, 0.006***, 0.002, 0.005***, 0.005***, 0.004**, 0.005***, 0.006***, 0.002 (SEs reported)
- Inflation: consistently negative and often significant (examples: -0.007*** repeated across specs)
- SME share of employment: typically around -0.001 to -0.002**
- MENA dummy: values include -0.167***, -0.163***, -0.176***, -0.190***, -0.148***, -0.152***, -0.147**, -0.259***, -0.177** etc.
- Business environment indicators:
  - Total tax rate (%profit): -0.000* and -0.001** (0.000)(0.000)
  - Cost of business start-up procedures: -0.001*** -0.001*** (0.000)(0.000)
  - Cost to register property: -0.011*** -0.013*** (0.003)(0.003)
  - Time required to enforce a contract: -0.000* -0.000 (0.000)(0.000)
  - Public credit registry coverage: 0.003** 0.003** (0.001)(0.001)
  - Property rights: 0.070*** 0.062*** (0.022)(0.023)
- MENA interaction: MENA*Cost to register property 0.036*** (0.011)

### CCA (selected business environment coefficients)
- Investment (%GDP): similar positive and significant coefficients as MENA (examples: 0.005***, 0.006***, 0.005***)
- Inflation: consistently negative and often significant (examples: -0.008***)
- SME share of employment: small negative coefficients, some significant
- CCA dummy: coefficients include -0.078**, -0.085***, -0.112***, -0.145***, -0.122***, -0.080***, -0.054, -0.084, -0.100***, -0.140***, -0.172*, -0.079** , 0.166 (varies by specification)
- Business environment indicators:
  - Cost of business start-up procedures: -0.001*** -0.001*** (0.000)(0.000)
  - Cost to register property: -0.012*** -0.012*** (0.003)(0.003)
  - Time required to enforce a contract: -0.000** -0.000** (0.000)(0.000)
  - Public credit registry coverage: 0.003** 0.003** (0.001)(0.001)
  - Property rights: 0.069*** 0.071*** (0.022)(0.022)

### Model fit and sample
- Observations: typically 190 182 182 187 182 182 124 182 182 187 182 182 124
- R-squared examples (MENA): 0.158 0.189 0.252 0.220 0.197 0.211 0.149 0.192 0.252 0.238 0.200 0.212 0.158
- R-squared examples (CCA): 0.118 0.153 0.214 0.178 0.174 0.179 0.114 0.153 0.214 0.178 0.174 0.179 0.115

### Notes
- Robust standard errors in parentheses. Income levels dummies included but not reported.
- ***, **, * indicate statistical significance at 10, 5, and 1 percent levels, respectively.
- Dependent variable: SME financial inclusion index

*Source: Appendix tables 7–8 of the provided PDF content.*

---

### Appendix table 9: Sample (countries listed)
- The sample list as provided includes (verbatim, presented here as a contiguous list by source): Afghanistan, Albania, Angola, Antigua and Barbuda, Argentina, Armenia, Azerbaijan, Bahamas, The, Bangladesh, Barbados, Belarus, Belize, Benin, Bhutan, Bolivia, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Burkina Faso, Burundi, Cambodia, Cameroon, Central African Republic, Chad, Chile, China, Colombia, Congo, Dem. Rep., Congo, Rep., Costa Rica, Cote d'Ivoire, Croatia, Czech Republic, Djibouti, Dominica, Dominican Republic, Ecuador, Egypt, Arab Rep., El Salvador, Eritrea, Estonia, Ethiopia, Fiji, Gabon, Gambia, The, Georgia, Ghana, Guatemala, Guinea, Guinea-Bissau, Guyana, Honduras, Hungary, India, Indonesia, Israel, Jamaica, Jordan, Kazakhstan, Kenya, Kyrgyz Republic, Latvia, Lebanon, Lesotho, Lithuania, Madagascar, Malawi, Malaysia, Mali, Mauritania, Mauritius, Mexico, Moldova, Mongolia, Montenegro, Morocco, Mozambique, Myanmar, Namibia, Nepal, Nicaragua, Niger, Nigeria, Pakistan, Panama, Paraguay, Peru, Philippines, Poland, Romania, Russian Federation, Rwanda, Samoa (Solomon Islands appears as Solomon Islands), Sierra Leone, Slovak Republic, Slovenia, Solomon Islands, South Africa, Sri Lanka, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, Sudan, Suriname, Swaziland, Tajikistan, Tanzania, Thailand, Timor-Leste, Togo, Tonga (not listed in source), Tunisia, Turkey, Turkmenistan, Uganda, Ukraine, Uruguay, Uzbekistan, Vanuatu, Venezuela, RB, Vietnam, West Bank and Gaza, Yemen, Rep., Zambia, Zimbabwe.  
- Note: The source lists countries in a continuous block; the above preserves the country names as provided in the source content.

*Source: Appendix table 9 of the provided PDF content.*

*Source: Provided PDF content (Appendix tables 1–9).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020055-print-pdf.pdf_