## 10. Stock Market Turnover and Growth: Heterogeneity Between Oil Exporters and Other Countries

## Source details

**Canonical URL:** [10. Stock Market Turnover and Growth: Heterogeneity Between Oil Exporters and Other Countries](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13130.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13130.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13130.pdf.json)

---

### Introduction — central findings and context
- Financial development improves outcomes through: mobilization of savings, allocation of resources, facilitating transactions and risk management, and corporate control.
- Empirical measures used: liquid liabilities to GDP (M2/GDP), private sector credit to GDP, market capitalization to GDP, and turnover (value of shares traded to GDP or to total capitalization).
- Estimation addresses endogeneity using GMM dynamic panel (Arellano-Bover / Blundell-Bond).
- Documented nonlinearities and heterogeneity:
  - When private sector credit exceeds 110 percent of GDP marginal effect may become negative (Arcand, Berkes, and Panizza (2011)).
  - Resource-exporting economies may exhibit shallower finance and weaker finance-growth transmission.
- Key empirical heterogeneities reported:
  - MENA: banking sector depth yields a lower growth impact than rest of world.
  - Europe & Central Asia: banking depth produces a greater growth impact than average.
  - Oil exporters: banking-depth growth impact is weaker and weakens with higher oil dependence; stock market depth may deliver higher growth impact in oil exporters.
  - LICs: finance-growth nexus weaker; estimated growth impact of credit-GDP about one-half that for other countries at similar depth; appears negative at lowest income levels and significantly positive only at about the 73rd percentile of income per capita for LICs in 2008.

### Data, stylized facts, and representative statistics
- Sample and measures:
  - Annual observations 1975–2005 converted to non-overlapping five-year averages; up to 146 countries in some regressions.
  - Private credit = private credit by deposit money banks to GDP.
  - Turnover = value of total shares traded to average real market capitalization.
  - Additional measures: bank deposits/liquid liabilities to GDP, stock market capitalization to GDP.
  - Oil dependence (Oildep) = ratio of non-oil GDP to total GDP (real terms).
- Representative summary statistics and comparisons:
  - Average private credit to GDP by income group:
    - LICs: just over 24 percent
    - MICs: 47 percent
    - HICs: 110 percent
  - Stock market capitalization to GDP:
    - LICs: 23 percent
    - MICs: 73 percent
    - HICs: 130 percent
  - MENA average real per capita GDP growth (1975–2005): 0.4 percent per year
    - EDCs average: 2.4 percent
    - Developing Asia: 5 percent
    - Latin America and the Caribbean: 1.1 percent
    - Central and Eastern Europe: 2.3 percent
  - MENA in 2008:
    - private credit-GDP ratio: 45 percent (EDC average 38 percent; HIC typical 118 percent)
    - stock market turnover: just under 40 percent (world average 54 percent; EDC average 40 percent)
    - credit represented 69 percent of bank deposits (EDCs on average: 90 percent)
  - Regional bank deepening trends:
    - MEDA stalled after 2005, losing about three percentage points of GDP in bank deepening.
    - Europe & Central Asia gained close to 20 percentage points of GDP in bank deepening.
- Within-region heterogeneity (MENA examples):
  - MEDA turnover ~ half that of rest of MENA.
  - Saudi Arabia turnover >130 percent; Jordan, Egypt, Morocco ~30 percent.
  - Private credit depth varies from >50 percent to <15 percent across MENA members.
- LIC disadvantages: lower financial depth and access but not necessarily lower FDI.

### Empirical methodology
- Core dynamic panel specification:
  - Dependent variable: growth rate of GDP per capita (five-year averages).
  - Key regressors: private credit-GDP (banking depth) and turnover (market depth).
  - Controls X: FDI/GDP and gross secondary school enrollment (logs), lagged initial income for convergence.
  - Interactions: regional dummies, oil-exporter dummy (Oilexp), continuous oil dependence (Oildep), income-group dummies, and triple interactions with openness and bank supervision.
- Estimation:
  - System GMM (Arellano-Bover; Blundell-Bond) on five-year averages to reduce cyclical noise and instrument proliferation.
  - Two-step GMM with robust variance-covariance and clustered standard errors at country level.
  - Two lags of covariates used as internal instruments; option h(2) used to control for heteroskedasticity.

### Main regression results — banking depth (private credit) and heterogeneity
- Baseline associations:
  - Private Credit coefficients frequently positive and significant (examples: 0.013***; 0.011***; 0.017*** in various specifications).
  - Private Credit x Financial Crisis consistently negative and significant (example: -0.006***), implying crises reduce growth impact of private credit by about one-half.
- Regional heterogeneity:
  - MENA: growth effects of private credit are lower; for total GDP growth the same banking depth produces growth effects about one-third smaller; for non-oil growth about one-half that of rest of world.
  - Europe & Central Asia: obtain relatively greater growth benefits from private credit (examples: Private Credit x Europe & Central Asia 0.011**; 0.014**).
  - MEDA: Private Credit x MEDA interaction negative and significant (e.g., -0.007*; -0.008*); GCC interaction not statistically different from zero.
- Oil-exporter heterogeneity:
  - Oil exporters as a group obtain smaller benefit from financial deepening; benefits fall continuously with degree of oil dependence (Private Credit x Oildep examples: -0.030***; -0.044***).
  - Interactions larger in absolute value for non-oil GDP growth, indicating banks ineffective in generating activity outside oil sector.
  - GCC countries tend to fare better; Private Credit x Oildep x GCC examples: 0.031*; 0.025.
- Income-level heterogeneity (LICs):
  - LICs obtain lower growth benefits from same level of private credit; benefits increase with income level.
  - Estimated growth impact of credit-GDP about one-half that for other countries with similar depth.
  - At lowest income levels the banking-depth effect appears negative; becomes significantly positive at per capita income of $810 (roughly the 73rd percentile for LICs in 2008).
  - Within LICs, greater trade openness and higher quality bank supervision enhance growth benefits of private credit (Private Credit x LIC x Openness 0.006***; Private Credit x LIC x Bank Supervision 0.003 to 0.004*).

### Main regression results — stock market turnover and heterogeneity
- Baseline associations:
  - Turnover coefficients positive and in many specs significant (examples: 0.005**; 0.007***; 0.013***).
  - Turnover x Financial Crisis negative and significant (examples: -0.006***; -0.009***), indicating crises reduce turnover’s growth contribution.
- Regional heterogeneity:
  - Cross-region heterogeneity observed for banking depth is largely absent for stock market activity; weak evidence of larger impact in Europe & Central Asia (Turnover x Europe & Central Asia up to 0.012**).
- Oil-exporter heterogeneity:
  - Interactions with Oilexp and Oildep mostly not significant for turnover; weak evidence oil exporters outside GCC might derive greater growth benefits from stock market activity.
- Income-level heterogeneity:
  - LICs obtain less growth benefit from stock market activity; effect mitigated by higher-quality bank supervision (Turnover x LIC x Bank Supervision 0.007* in Table 11).
- Magnitude example (Figure 9):
  - Increasing stock market turnover by 20 percentage points of GDP:
    - Starting at 10 percent turnover gains close to one-half of a percentage point.
    - Starting at 30 percent turnover gains decline to about one-fifth of a percentage point.

### Quantitative magnitudes and illustrative examples
- Examples from Figure 5 (credit-to-GDP increases):
  - For countries with initial private credit-to-GDP below EDC average:
    - Algeria: increasing depth from 10 percent to EDC average of 29 percent raises growth by 112 basis points.
    - A comparable non-MENA country would gain 163 basis points, implying a quality effect of 51 basis points (163 − 112 = 51 basis points).
    - Armenia could obtain 160 basis points if reaching EDC average depth.
- Table-based coefficients (selected):
  - Private Credit baseline examples: 0.013***; 0.016**; 0.011***.
  - Private Credit x Financial Crisis: −0.006*** across multiple columns.
  - Private Credit x Oildep: −0.030*** (Table 7).
  - Turnover baseline examples: 0.005**; 0.009**; 0.007*** (Tables 9–11).
  - Turnover x Financial Crisis: −0.006*** to −0.015*** across tables.
- Summary sample statistics (Table 1a / 1b / 2a / 2b / 3):
  - Full-sample private credit mean (observations 673): 35.951 (Std. Dev. 31.042; Min 0.456; Max 191.697).
  - Turnover full-sample mean (observations 361): 33.487 (Std. Dev. 41.633; Min 0.144; Max 294.096).
  - Growth full-sample mean (observations 696): 1.737 (Std. Dev. 2.852; Min -9.838; Max 9.998).
  - Oil-exporters private credit mean (observations 136): 26.347 (Std. Dev. 21.558; Min 2.004; Max 136.846).
  - Cross-country means private credit (146 countries): 33.753 (Std. Dev. 26.735; Min 2.857; Max 148.269).
  - MENA (sample means, Table 3): Private Credit 31.474; Turnover 21.196; Growth 1.366; Non-Oil Growth 1.974; Oil 0.238; Lerner Index 0.345; H-Stat 0.529.

### Banking system characteristics, access, competition, and ownership
- MENA vs EDCs (performance despite deeper credit):
  - MENA mobilizes about 30 percent greater private sector credit than average EDC but underperforms on access and competition.
  - Outreach to population 20–30 percent lower; proportion of firms citing credit as constraint 10 percent higher; percentage of firms receiving bank financing only four fifths of average EDC; estimated competition 20 percent lower.
- MENA vs Sub-Saharan Africa:
  - MENA depth over 2½ times SSA average; outreach to borrowers only twice as large; share of firms citing credit as constraint only 20 percent lower; percentage receiving credit only 20 percent greater; estimated competition virtually identical.
- Bank ownership:
  - Some MENA countries have near-100 percent state bank asset shares (e.g., Algeria and Libya approaching 100 percent in 2008); Syria about 70 percent; Lebanon and Jordan had zero state bank participation and substantial foreign bank penetration.
  - Evidence on state-bank effects mixed; state ownership combined with low depth and low institutional quality can have negative growth effects.
  - Foreign bank presence often linked to improvements in performance and competition.

### Demand- vs supply-side interpretations
- Demand-side (lack of profitable investment opportunities) may explain weaker bank-financed investment returns in oil exporters (Dutch Disease); regressions with non-oil growth consistent with this channel.
- Demand-side less convincing for non-oil-exporting MENA and LICs; stock markets do not show the same weaker nexus.
- Authors’ reading: supply-side conditions—functioning of banks and regulatory environment—primarily drive weaker growth outcomes in MENA, oil exporters, and LICs.

### Policy implications and recommendations
- Prioritize “high-quality deepening” where finance-growth transmission is weak:
  - Facilitate greater access to finance and competition.
  - Strengthen supervision and regulatory frameworks.
  - Improve entry conditions and efficiency in financial and real sectors to convert existing depth into stronger growth outcomes.
- For oil-dependent economies:
  - Address weaker transmission of banking depth to growth; examine stock market development as complementary channel.
  - GCC experience suggests policy design can preserve benefits despite oil dependence.
- For MENA and MEDA:
  - Investigate causes of stalled bank deepening post-2005 and low conversion of deposits to private credit; target reforms to mobilize untapped deposits toward private-sector lending.
- Institutional and regulatory improvements suggested:
  - Establish credit bureaus and improve borrower information.
  - Enhance legal protection of creditor rights.
  - Improve secured transactions framework.
  - For LICs, pursue improvements in bank supervision.
- Expected outcome: these actions should result in higher and more sustainable long-run growth.

### Figures, tables, and diagnostic highlights
- Key figures referenced:
  - Figure 5: Estimated impact of increases in credit-GDP on real per capita growth (percentage points) illustrating “quality effect” and “pure depth effect”.
  - Figure 6–8: Marginal impacts by income, bank supervision quality, and trade openness; Figure 6 notes growth impact becomes positive at per capita income of $810 for LICs (73rd percentile for LICs in 2008).
  - Figure 9: Estimated increase in long-run growth from increasing stock market turnover by 20 percentage points at different initial turnover levels (e.g., gains ~0.5 percentage point at 10 percent turnover; ~0.2 at 30 percent).
- Estimation diagnostics:
  - AR2 p-values generally indicate absence of second-order autocorrelation; Hansen tests indicate instruments are generally valid (examples: Hansen p-values 0.300; 0.419; 0.273).
  - Number of instruments reported (examples: 76; 100).
- Robustness:
  - Results robust across alternative measures (liquid liabilities, market capitalization) and to inclusion of crisis interactions and non-oil GDP growth specifications.

*Source: _wp13130 - 10. Stock Market Turnover and Growth: Heterogeneity Between Oil Exporters and Other Countries (PDF).*

### 10. Stock Market Turnover and Growth: Heterogeneity Between Oil Exporters and

### 10. Stock Market Turnover and Growth: Heterogeneity Between Oil Exporters and

### Section headings
- 10. Stock Market Turnover and Growth: Heterogeneity Between Oil Exporters and Other Countries ...................................................................................................................36
- 11. Stock Market Turnover and Growth: Heterogeneity Across Income Levels ....................37

### Figures (captions and page numbers)
- 1: Average Real Per Capita GDP Growth Rates Across Regions, 1975-2005 ........................38
- 2. Financial Depth Across Regions and Countries ..................................................................38
- 3. Deepening in the Banking Sector, Across Regions, 1975-2008 ..........................................39
- 4. The Ratio of Private Credit to Deposits, 1975-2008 ............................................................40
- 5. Estimated Impact of Increases in Credit-GDP on Real Per Capita Growth (Percentage Points) ..............................................................................................................40
- 6. Estimated Marginal Impact of Increases in Private Credit-GDP on Growth at Different Income Levels (Percentage Points) .....................................................................41
- 7. Estimated Differences Between LICs and non-LICs in the Growth Impact of Private Credit at Different Levels of Bank Supervision Quality (Percentage Points) .....................41
- 8. Estimated Differences Between LICs and non-LICs in the Growth Impact of Private Credit at Different Levels of Trade Openness (Percentage Points) ....................................42
- 9. Estimated Increase in Long-Run Growth from an Increase in Stock Market Turnover by 20 Percentage Points of GDP, at Different Initial Levels of Turnover ..........................42
- 10. Banking Sector Performance in MENA Countries Relative to Emerging and Developing Country Average and to Sub-Saharan Africa, 2008 ........................................43
- 11. Financial Access, Use of Banking Services, and Depth Across Income Groups, 2008 ....43
- 12. Financial Access and Banking Depth (Privy) Across Countries .......................................44

*Source: _wp13130 - 10. Stock Market Turnover and Growth: Heterogeneity Between Oil Exporters and (PDF).*

### 13. Share of Public and Foreign Banks throughout the World, 2002 ......................................44

### 13. Share of Public and Foreign Banks throughout the World, 2002

### I. Introduction — central findings and context
- Financial development is linked to improved economic outcomes through four functions: mobilization of savings, allocation of resources to productive uses, facilitating transactions and risk management, and exerting corporate control.
- Empirical measures of financial depth used: liquid liabilities to GDP (M2/GDP), private sector credit to GDP, market capitalization to GDP, and turnover measures (value of shares traded to GDP or to total capitalization).
- Cross-country and panel methodologies have been used extensively; GMM dynamic panel estimators (Arellano-Bover / Blundell-Bond) are employed here to address endogeneity and omitted-variable bias.
- Nonlinear and heterogeneous effects are documented in the literature:
  - Arcand, Berkes, and Panizza (2011): when private sector credit exceeds 110 percent of GDP, the marginal effect of additional deepening on economic activity becomes negative.
  - Resource-exporting economies may exhibit shallower finance and weaker finance-growth transmission (Nili and Rastad (2007); Beck (2011)).

- Key empirical heterogeneities found in this paper:
  - Middle East and North Africa (MENA) countries: banking sector depth yields a lower growth impact than in the rest of the world.
  - Europe and Central Asia: banking depth produces a greater growth impact than average.
  - Oil exporters: the growth impact of banking depth is weaker in general and weakens progressively with higher oil dependence; stock market depth may deliver higher growth impact in oil exporters.
  - Low Income Countries (LICs): finance-growth nexus is weaker; estimated growth impact of the credit-GDP ratio is about half as large for LICs relative to other countries with similar depth and appears negative at the lowest income levels, becoming significantly positive at about the 73rd percentile of income per capita for LICs in 2008.

### II. Data and stylized facts (sample, measures, and summary statistics)
- Sample and period:
  - Annual country-specific observations from 1975 to 2005, converted to non-overlapping five-year averages.
  - Up to 146 countries included in some regressions.
- Financial development measures from the World Bank Financial Structure Database:
  - Private credit = ratio of private credit by deposit money banks to GDP.
  - Turnover = ratio of the value of total shares traded to average real market capitalization.
  - Robustness checks use bank deposits/liquid liabilities to GDP and stock market capitalization to GDP.
- Additional data:
  - Non-oil GDP, total GDP, population from WEO April 2010; some real non-oil GDP series supplemented from IMF desk economists.
  - Total real per capita GDP, secondary school enrollment, and FDI/GDP from World Bank open source data.
- Oil dependence (Oildep) defined as ratio of non-oil GDP to total GDP (both in real terms).
- Representative summary statistics and comparisons (2008 or study-period averages where noted):
  - Average private credit to GDP:
    - LICs: just over 24 percent
    - Middle Income Countries (MICs): 47 percent
    - High Income Countries (HICs): 110 percent
  - Stock market capitalization to GDP:
    - LICs: 23 percent
    - MICs: 73 percent
    - HICs: 130 percent
  - MENA average real per capita GDP growth (1975–2005): 0.4 percent per year
    - Emerging and Developing Countries (EDCs) average: 2.4 percent
    - Developing Asia: 5 percent
    - Latin America and the Caribbean: 1.1 percent
    - Central and Eastern Europe: 2.3 percent
  - MENA private credit-GDP ratio in 2008: 45 percent (higher than EDC average of 38 percent but below HIC typical 118 percent).
  - MENA stock market turnover in 2008: just under 40 percent (world average 54 percent; EDC average 40 percent).
  - Credit-to-deposits conversion (2008 averages):
    - MENA banking systems: credit represented 69 percent of bank deposits
    - EDCs on average: 90 percent
  - Regional trends:
    - MEDA subregion stalled after 2005, losing about three percentage points of GDP in bank deepening.
    - Europe and Central Asia gained close to 20 percentage points of GDP in bank deepening over the long run.
- Within-region heterogeneity (MENA):
  - MEDA (Algeria, Egypt, Jordan, Lebanon, Libya, Morocco, Syria, Tunisia, West Bank & Gaza) has lower stock market depth (turnover about half that of rest of MENA).
  - GCC and other MENA members show wide dispersion: e.g., Saudi Arabia turnover >130 percent; Jordan, Egypt, Morocco ~30 percent; some countries have private credit depth >50 percent of GDP while others <15 percent.
- LICs are disadvantaged across most dimensions except FDI: lower financial depth, lower secondary enrollment, and lower growth.

### III. Empirical methodology (specification and estimation)
- Core regression model (dynamic panel) focuses on estimating β, the effect of financial development on growth, in a Solow-convergence augmented framework:
  - Dependent variable: growth rate of GDP per capita (five-year averages).
  - Key regressor: financial depth (private credit-GDP ratio for banking depth; turnover for market depth).
  - Controls X include FDI/GDP and gross secondary school enrollment (logarithms of mean values over five-year periods).
  - Inclusion of lagged initial income to capture convergence effects.
  - Interactions: financial depth interacted with (i) regional dummies (Europe & Central Asia, MENA, South Asia, East Asia & Pacific, Sub-Saharan Africa, Latin America & the Caribbean, rest of world), (ii) oil exporter dummy (Oilexp), and (iii) continuous oil dependence (Oildep).
- Estimation approach:
  - System GMM (Arellano and Bover (1995); Blundell and Bond (1998)) applied to non-overlapping five-year averages to reduce cyclical noise and avoid instrument proliferation.
  - Two-step GMM used with robust variance-covariance and clustered standard errors at country level.
  - Two lags of covariates used to construct internal instruments; option h(2) used to control for heteroskedasticity in STATA’s xtabond2.
- Innovations relative to prior studies:
  - Uses dynamic panel rather than cross-section to examine resource-rich differences.
  - Longer sample: 1975–2005 (versus 1992–2001 in Nili and Rastad).
  - Expanded country coverage: up to 146 countries; oil-exporter sample expanded to 30 countries.
  - Includes regressions for non-oil GDP in addition to total GDP growth.
  - Examines both banking sector and stock market activity.

### IV. Main regression results and interpretation (summarized empirical outcomes)
- Banking depth (private credit-GDP ratio):
  - Banking depth delivers a weaker growth impact in MENA relative to other regions; Europe and Central Asia show a stronger positive impact.
  - For oil exporters:
    - The growth impact of banking depth is weaker in general.
    - The banking-growth effect weakens progressively with higher degrees of oil dependence (Oildep).
  - For LICs:
    - The estimated growth impact of credit-GDP is about one-half that for other countries with similar depth.
    - At the lowest income levels the banking-depth effect appears negative, becoming significantly positive only at about the 73rd percentile of income per capita for LICs in 2008.
    - LICs with higher-quality supervision or greater trade openness derive relatively better growth from financial deepening.
- Stock market depth (turnover):
  - Evidence suggests that stock market depth may have a higher growth impact in oil-exporting countries, in contrast to banking depth results.
- Quality gap interpretation:
  - The weaker finance-growth transmission in MENA and LICs is described as a “quality gap” in financial intermediation, potentially driven by:
    - Strong state ownership of banks
    - Lack of competition
    - Insufficient progress in financial reform
    - Limited access to credit for businesses despite deposit accumulation (untapped deposit potential)
    - Weak supervisory and regulatory frameworks
- Policy implication distilled from results:
  - For LICs and regions with a quality gap, policy should prioritize “high-quality deepening”:
    - Facilitate greater access to finance and competition
    - Strengthen supervision and regulatory frameworks
    - Improve entry conditions and efficiency in financial and real sectors to convert existing depth into stronger growth outcomes
  - For oil-dependent economies:
    - Financial sector reforms should address the weaker transmission of banking depth to growth and examine stock market development as a complementary channel.
  - For MENA and MEDA:
    - Investigate causes of stalled bank deepening post-2005 and the low conversion of deposits to private credit; target reforms to mobilize untapped deposits toward private-sector lending.

### V. Organization and scope
- Paper structure:
  - Section II: data description and stylized facts (summary statistics, mean-difference tests, correlations).
  - Section III: empirical methodology (GMM dynamic panel, interaction terms, controls).
  - Section IV: regression results (banking and market depth interactions by region, oil dependence, and income group).
  - Section V: conclusions and plausible drivers of observed heterogeneity (state ownership, competition, financial reform progress, supervision, access).

*Source: _wp13130 - 13. Share of Public and Foreign Banks throughout the World, 2002 (PDF chapter/section).*

### 8. Specifically, we examine heterogeneity in this relationship across regions (Table 6),

### _wp13130 - 8. Specifically, we examine heterogeneity in this relationship across regions (Table 6)

### Regional heterogeneity in the finance-growth relationship
- Baseline regressions follow the common specification used in the literature (e.g., Beck and Levine (2004); Beck (2008)) with a key modification: the regressions account for the possible effect of financial crises on the finance-growth relationship.
- Using the Laeven and Valencia (2012) definition of systemic banking crises:
  - about 60 percent of all such episodes experienced during the 1970–2007 period occurred in the 1990s.
  - Over the 1970–2010 period, systemic banking crises arose about 13 percent of the time in the Middle East and North African (MENA) countries, compared to 23 percent on average for emerging and developing economies.
  - During 2000–10, this frequency spiked at 60 percent for OECD countries, while the MENA region managed to avoid these episodes altogether.
- Across all specifications, financial crises reduce the growth impact of private credit by about one-half.
- Interacting private credit with regional dummies (Table 6):
  - Growth effects of private credit are lower for the MENA region, and for Latin America and the Caribbean.
  - For total GDP growth, the same level of banking depth in the MENA region produces growth effects that are about one-third smaller than in other regions.
  - For non-oil growth, the MENA region shows a growth impact about one-half that of the rest of the world.
  - Europe and Central Asia obtain relatively greater growth benefits from private credit.
- Subgrouping within MENA (Mediterranean-Associated countries vs. the rest; GCC vs. the rest):
  - The GCC countries behave similarly to high-income countries; the coefficient on the interaction term between private credit and the GCC dummy is not statistically significantly different from zero.
  - The MEDA interaction coefficient with private credit is negative and significant when GCC countries are combined with a set of non-Mediterranean countries; the corresponding coefficient for other MENA countries is not statistically significant.
  - Once GCC countries are accounted for separately, the interaction term for the Latin America and Caribbean region no longer becomes significant (i.e., this region behaves relatively similarly to the full set of high-income countries).

### Tests of instrument validity and autocorrelation
- Arellano-Bond test for autocorrelation:
  - The test is applied to the differenced residuals.
  - First-degree correlation in differences is observed for all regressions, as expected.
  - The second-degree correlation test yields no evidence of significant autocorrelation among the set of instruments.
- Hansen test for over-identifying restrictions:
  - The test checks the correlation between the residuals and exogenous variables to assess the validity of instruments.
  - The null hypothesis that the instruments are exogenous cannot be rejected for the reported regressions.

### Quantitative magnitudes and examples of regional “quality effects”
- Estimated impact on long-term total GDP growth from increasing banking sector depth (Figure 5):
  - Because of the log specification, greater growth benefits accrue to countries that begin deepening from a lower initial level.
  - Figure 5a (countries with current private credit-to-GDP below the EDC average):
    - Example: Algeria increasing depth from an initial level of 10 percent to the EDC average of 29 percent would increase its growth rate by 112 basis points.
    - A non-MENA country starting from the same initial depth could increase its growth rate by 163 basis points, implying a quality effect of 51 basis points (163 − 112 = 51 basis points).
    - Armenia would obtain a full benefit of 160 basis points if it were to reach the EDC average depth.
  - Figure 5a also shows MENA countries with initial depth above the EDC average; for these the figure displays gains from increasing depth by 20 percentage points of GDP (roughly the increase observed in high-income countries from 1995 to 2005), with predicted effects and corresponding non-MENA comparisons to quantify quality effects.

### Oil dependence and the finance-growth link (Table 7)
- Distinguishing oil exporters from the rest confirms that oil dependency weakens the finance-growth link, providing evidence of a finance channel for the resource-curse.
- Oil exporters as a group obtain a smaller benefit from financial deepening; the benefits fall continuously with the degree of oil dependence.
- Interaction terms are larger in absolute values in regressions for non-oil GDP growth, indicating banks in these countries have been particularly ineffective in generating productive activity outside the oil sector.
- Further interactions of private credit and oil dependence with the GCC dummy:
  - GCC countries tend to fare better compared to similarly oil-dependent countries outside the region.
  - Example: Saudi Arabia—with an oil dependence of about 33 percent in 2005—would obtain a greater growth benefit from private credit than a similarly oil-dependent country such as Trinidad and Tobago.
  - This result aligns with earlier findings that growth benefits from banking depth in GCC countries are similar to those in high-income countries.

### Income-level heterogeneity and role of openness and bank supervision (Table 8; Figures 6–8)
- Low-income countries (LICs):
  - LICs as a group obtain lower growth benefits from the same level of private credit.
  - These benefits increase continuously with income level.
- Within LICs, two factors enhance the growth benefits of private credit:
  - Greater openness to trade, as measured by the ratio of exports and imports to GDP.
  - Higher quality of bank supervision.
- These two characteristics only appear to affect the growth benefits of private credit in LICs; interaction terms for non-LICs are not statistically significant.
- Figures 6–8 illustrate magnitudes:
  - Figure 6: At very low income levels the growth impact is not statistically significant, and only becomes positive (at a 95 percent confidence level) at a per capita income of $810, or roughly the 73rd percentile for LICs in 2008.
  - Figure 7: The mitigating effect of the quality of bank supervision is evident; at low levels LICs are at a clear disadvantage, but as bank supervision quality improves the growth impact for LICs begins to approximate that of middle and high-income countries.
  - As of 2005, the average value of the bank supervision indicator for a sample of 18 LICs was 1.4.

*Italic: Source: _wp13130 - 8. Specifically, we examine heterogeneity in this relationship across regions (Table 6), PDF chapter/section*

### 1.8 for middle-income countries and over 2.5 for high-income countries. Finally, in

### _wp13130 - 1.8 for middle-income countries and over 2.5 for high-income countries. Finally, in

### Stock market activity: growth effects and heterogeneity
- Regressions using a stock market-based Turnover measure find the coefficient on stock market turnover is positive and significant in normal times; crises have a significant negative impact on the coefficient.
- Cross-region heterogeneity observed for banking depth is largely absent for stock market activity, aside from weak evidence of a slightly larger growth impact in Europe and Central Asia (Table 9).
- Oil exporters: neither the interaction with the oil exporter dummy nor with the degree of oil dependence yield significant coefficients, though there is weak evidence that oil exporters outside of the GCC might derive greater growth benefits from stock market activity (Table 10, fourth column).
- Income levels: evidence that LICs obtain less growth benefits from stock market activity, an effect mitigated by higher quality bank supervision (Table 11, fifth column).
- Figure 9: potential gains from increasing stock market turnover by 20 percentage points (approximately equivalent to the deepening experienced by EDCs on average from 1995 to 2008):
  - Starting at 10 percent turnover, gains are close to one-half of a percentage point.
  - For countries starting at a turnover ratio of 30 percent, gains decline to about one-fifth of a percentage point.

### Heterogeneity in the finance-growth nexus: banking depth versus stock markets
- The finance-growth nexus is heterogeneous across regions, income levels and between oil and non-oil exporters; heterogeneity arises primarily for banking depth rather than for stock market activity.
- Possible drivers of heterogeneity include differences in access to financial services and banking competition, which are not perfectly correlated with banking depth.

### Banking system characteristics, access, and competition (MENA and LIC comparisons)
- MENA vs. EDCs: despite MENA countries mobilizing about 30 percent greater private sector credit than the average EDC, they underperform on access and competition:
  - Outreach of banking services to the population is about 20–30 percent lower.
  - Proportion of firms citing credit as a constraint is 10 percent higher.
  - Percentage of firms receiving bank financing is only four fifths of that in the average EDC.
  - Estimated competition in the banking system is 20 percent lower.
- MENA vs. sub-Saharan Africa:
  - MENA depth is over 2½ times the average in sub-Saharan Africa.
  - Outreach to borrowers is only twice as large.
  - Share of firms indicating credit as a major constraint only 20 percent lower.
  - Percentage of surveyed firms receiving bank credit only 20 percent greater.
  - Average estimated competition in the banking system is virtually identical.
- Result: MENA countries exhibit a "quality gap" in banking intermediation; for the same level of depth, growth benefits are at most two-thirds of those obtained in other regions. Gap appears more pronounced for non-GCC countries.

### Financial access differences: LICs and high-income comparisons
- In 2008, banking depth in the average high-income country was 4½ times the level of the average LIC.
- Access contrasts (2008):
  - Access to bank branches and ATMs was over 50 times as great in high-income countries versus LICs.
  - Coverage of banking services (deposits and loans) among the population was about 7 times as great.
  - Coverage of non-bank institutions was 6–9 times as great.
- Summary: LICs suffer from both shallow financial systems and pronounced access deficits, contributing to weaker finance-growth links.

### Bank ownership, state banks, and foreign bank participation
- Many MENA countries have a relatively high share of state banks and/or a relatively small share of foreign-owned banks; heterogeneity within MENA is large.
  - Examples cited: Algeria and Libya had state bank asset shares approaching 100 percent in 2008; Syria about 70 percent; Lebanon and Jordan had zero state bank participation and substantial foreign bank penetration.
  - Other MENA countries had state bank participation between 37 and 57 percent market share in 2008 and modest foreign bank participation below international averages.
- State-bank effects:
  - Evidence is mixed; Korner and Schnabel (2010) identify significant negative growth effects from state ownership of banks when combined with low levels of financial depth and low institutional quality.
  - Country-level evidence of inefficient or corrupt lending by state-owned banks: Khwaja and Mian (2005) document preferential treatment by state-owned banks in Pakistan that cost up to an estimated 1.9 percent of GDP per year.
- Foreign bank presence is often linked to improvements in banking sector performance and competition, suggesting potential benefits from greater openness.

### Demand- and supply-side interpretations
- Demand-side weakness (lack of profitable investment opportunities) could reduce returns to bank-financed investment; for oil exporters Dutch Disease–type effects are a plausible explanation (regressions with non-oil growth as dependent variable are consistent with this).
- However, demand-side explanations are less convincing for non-oil exporting MENA countries or LICs, and do not explain why stock markets do not show the same weaker finance-growth nexus.
- Authors' reading: primarily supply-side conditions—the functioning of banks and their regulatory environment—drive weaker growth outcomes in MENA, oil exporters, and LICs.

### Policy implications and recommendations
- Beyond macroeconomic stability and continued financial reform, two additional fronts are recommended:
  - Reduce impediments to credit expansion, especially in MENA, to increase credit per unit of deposits. Likely suspects: fiscal dominance or overly restrictive monetary policy diverting bank funds away from the private sector.
  - Enhance the quality of bank intermediation, possibly including reassessment of the role of state banks, to improve access and competition.
- Specific institutional and regulatory improvements suggested:
  - Improvements in information on prospective borrowers, including establishment of credit bureaus.
  - Enhancements to legal protection of creditor rights.
  - Improvements to the framework surrounding secured transactions.
  - For LICs, pursue improvements in bank supervision.
- Expected outcome: these actions should result in higher and more sustainable long-run growth.

*Source: _wp13130 - 1.8 for middle-income countries and over 2.5 for high-income countries. Finally, in*

### REFERENCES

### _wp13130 - REFERENCES

### References
- Abdih, Yasser and Anjali Garg (2011) ―Unemployment in the Oil-Importing Countries of the Middle East and North Africa,‖ Unpublished Surveillance Meeting Note (Washington: International Monetary Fund).
- Abiad, Abdul, Enrica Detragiache, and Thierry Tressel (2010) ―A New Database of Financial Reforms,‖ IMF Staff Papers, Vol. 57(2), pp. 281–302.
- Al Hussainy, Ed; Andrea Coppola; Erik Feyen; Alain Ize; Katie Kibbuka; Haocong Ren (2011) ―FinStats 2011: A Ready-to-Use Tool to Benchmark Financial Sectors Across Countries and Over Time,‖ (Washington: World Bank).
- Anzoategui, Diego, María Soledad Martínez Pería, Roberto Rocha (2010) ―Bank Competition in the Middle East and Northern Africa Region,‖ World Bank Policy Research Paper 5363 (Washington: World Bank).
- Arcand, J.L., E. Berkes, and U. Panizza, 2011, ―Too Much Finance?‖ Unpublished (Washington: International Monetary Fund).
- Arellano, M. and O. Bover, 1995, ―Another Look at the Instrumental-Variable Estimation of Error-Components Models‖, Journal of Econometrics, Vol. 68, pp. 29-52.
- Arellano, M. and S. Bond, 1991, ―Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations‖, Review of Economic Studies, Vol. 58, pp. 277-97.
- Beck, Thorsten, 2008, ―The Econometrics of Finance and Growth,‖ World Bank Policy Research Paper 4608 (Washington: World Bank).
- Beck, Thorsten, 2011, ―Finance and Oil: Is There a Resource Curse in Financial Development?‖, unpublished (Tilburg, The Netherlands: Tilburg University).
- Beck, Thorsten. and Asli Demirgüç-Kunt, 2009, ―Financial Institutions and Markets Across Countries and over Time: Data and Analysis‖, World Bank Policy Research Working Paper No. 4943 (Washington: World Bank).
- Beck, Thorsten and Ross Levine (2004), ―Stock Markets, Banks, and Growth: Panel Evidence,‖ Journal of Banking and Finance, 28, pp. 423-442.
- Beck, Thorsten, Ross Levine and Norman Loayza, 2000, ―Finance and the Sources of Growth‖, Journal of Financial Economics, Vol. 58, pp. 261-300.
- Blundell, R. and S. Bond, 1998, ―Initial Conditions and Moment Restrictions in Dynamic Panel Data Models‖, Journal of Econometrics, Vol. 87, pp. 115-43.
- Bhattacharya, Rina and Hirut Wolde (2010), ―Constraints on Growth in the MENA Region,‖ IMF Working Paper 10/30 (Washington: International Monetary Fund).
- Demirguc-Kunt, Asli and Ross Levine (2008), ―Finance, Financial Sector Policies, and Long-Run Growth,‖ Unpublished (Washington: World Bank).
- Demirguc-Kunt, Asli and Ross Levine (2009), ―Finance and Inequality: Theory and Evidence,‖ NBER Working Paper 15275 (Cambridge, Massachusetts: NBER).
- Farazi, Subika, Erik Feyen, and Roberto Rocha (2011), ―Bank Ownership and Performance in the Middle East and North Africa Region,‖ (Washington: World Bank).
- Guillaume, Dominique and Tobias Rasmussen (2011), “Growth, Trade, and Competitiveness of Oil Importing Countries in the Middle East and North Africa (MENAP), Unpublished Surveillance Meeting Note (Washington: International Monetary Fund).
- Hirschberg, J. and J. Lye, 2010, ―A reinterpretation of interactions in regressions,‖ Applied Economics Letters, 17, pp. 427-430.
- Hakura, Dalia (2004), ―Growth in the Middle East and North Africa,‖ IMF Working Paper 04/56 (Washington: International Monetary Fund).
- Khwaja, A.I. and A. Mian, 2005, ―Do Lenders Favor Politically Connected Firms? Rent provision in an Emerging Financial Market,‖ Quarterly Journal of Economics, Vol. 120, Issue 4, pp. 1371-1411.
- Khan, M., A. Senhadji, and B. Smith, 2001, ―Inflation and Financial Depth,‖ IMF Working Paper No. 00/110 (Washington: International Monetary Fund).
- Khan, M., and A. Senhadji, 2000, ―Threshold Effects in the Relationship between Inflation and Growth‖, IMF Working Paper No. 01/44 (Washington: International Monetary Fund).
- King, R. G. and R. Levine, 1993, ―Finance and Growth: Schumpeter Might Be Right‖, Quarterly Journal of Economics, Vol. 108, pp. 717–38.
- Klein, N., 2010, ―The Linkage between the Oil and the Non-oil Sectors – A Panel VAR Approach‖, IMF Working Paper No. 10/118 (Washington: International Monetary Fund).
- Korner, T. and I. Schnabel, 2010, ―Public Ownership of Banks and Economic Growth: The Role of Heterogeneity,‖ CEPR Discussion Paper 8138.
- La Porta, R., F. Lopez-de-Silanes, A. Shleifer and R. W. Vishny, 1998, ―Law and Finance‖, Journal of Political Economy, Vol. 106, pp. 1113-1155.
- Laeven, Luc and Fabian Valencia, 2012, ―Systemic Banking Crises: An Update,‖ IMF Working Paper 12/163 (Washington: International Monetary Fund).
- Levine, R. and S. Zervos, 1998, ―Stock Markets, Banks, and Economic Growth‖, American Economic Review, Vol. 88, pp. 537-58.
- Levine, R., 1997a, ―Financial Development and Economic Growth: Views and Agenda,‖ Journal of Economic Literature, Vol. 35(2), pp. 688-726.
- Levine, R., 1997b, ―Law, Finance, and Economic Growth‖, Journal of Financial Intermediation, Vol. 8, pp. 36–67.
- Levine, R., 1998, ―The Legal Environment, Banks, and Long-Run Economic Growth‖, Journal of Money, Credit, and Banking, Vol. 30, pp. 596-613.
- Levine, R., 2004, ―Finance and Growth: Theory and Evidence‖, NBER Working Paper No. 10766 (Cambridge, Massachusetts: NBER).
- Levine, R., N. Loayza and T. Beck, 2000, ―Financial Intermediation and Growth: Causality and Causes‖, Journal of Monetary Economics, Vol. 46, pp. 31-77.
- Micco, A., U. Panizza, and M. Yáñez, 2007, ―Bank Ownership and Performance: Does Politics Matter?‖ Journal of Banking and Finance, Vo. 31 (1), pp. 219-241.
- Nili, M. and M. Rastad, 2007, ―Addressing the Growth Failure of the Oil Economies: The Role of Financial Development‖, The Quarterly Journal of Economics and Finance, Vol. 46, pp. 726–40.
- Roodman, D. M., 2009a, ―A Note on the Theme of Too Many Instruments‖, Oxford Bulletin of Economics and Statistics, Vol. 71(1), pp. 135–158.
- Roodman, D. M., 2009b, ―How to Do xtabond2: An introduction to difference and system GMM in Stata,‖ The Stata Journal (2009), Vol. 9, no.1, pp. 86-136.
- Rousseau, Peter L. and Paul Wachtel (2011), ―What is Happening to the Impact of Financial Deepening on Economic Growth?‖ Economic Inquiry, Vol. 49, No.1, pp. 276-288.
- Rousseau, P. L. and P. Wachtel, 2000, ―Equity Markets and Growth: Cross-Country Evidence on Timing and Outcomes, 1980-1995‖, Journal of Business and Finance, Vol. 24, pp. 726–40.
- World Bank, 2011, Financial Access and Stability: MENA Financial Flagship Report (Washington: World Bank).

### Key Summary Statistics (Tables 1a–3)
- Table 1a (Full-sample observations, 5-year averages unless specified):
  - Private Credit: Number of Observations 673; Mean 35.951; Std. Dev. 31.042; Min 0.456; Max 191.697
  - Bank Deposits: Number of Observations 668; Mean 38.352; Std. Dev. 29.249; Min 1.828; Max 216.983
  - Liquid Liabilities: Number of Observations 655; Mean 44.220; Std. Dev. 28.497; Min 5.212; Max 227.672
  - Market Cap: Number of Observations 357; Mean 32.217; Std. Dev. 38.473; Min 0.038; Max 232.213
  - Turnover: Number of Observations 361; Mean 33.487; Std. Dev. 41.633; Min 0.144; Max 294.096
  - Growth: Number of Observations 696; Mean 1.737; Std. Dev. 2.852; Min -9.838; Max 9.998
  - Non-Oil Growth: Number of Observations 645; Mean 1.749; Std. Dev. 2.923; Min -10.929; Max 9.860
  - Education: Number of Observations 671; Mean 61.825; Std. Dev. 32.998; Min 2.499; Max 158.453
  - FDI: Number of Observations 696; Mean 2.480; Std. Dev. 3.460; Min -3.623; Max 33.540
  - Oil: Number of Observations 652; Mean 0.040; Std. Dev. 0.121; Min 0.000; Max 0.780
  - Lerner Index: Number of Observations 315; Mean 0.242; Std. Dev. 0.096; Min -0.034; Max 0.501
  - H-Stat: Number of Observations 309; Mean 0.653; Std. Dev. 0.185; Min 0.174; Max 1.035
- Table 1b (Cross-country means):
  - Private Credit: Number of Countries 146; Mean 33.753; Std. Dev. 26.735; Min 2.857; Max 148.269
  - Bank Deposits: Number of Countries 144; Mean 36.713; Std. Dev. 26.439; Min 4.595; Max 173.864
  - Liquid Liabilities: Number of Countries 142; Mean 42.783; Std. Dev. 26.072; Min 9.591; Max 182.613
  - Market Cap: Number of Countries 105; Mean 29.916; Std. Dev. 33.343; Min 0.547; Max 156.721
  - Turnover: Number of Countries 104; Mean 29.761; Std. Dev. 29.526; Min 0.742; Max 139.587
  - Growth: Number of Countries 150; Mean 1.894; Std. Dev. 1.673; Min -1.769; Max 7.997
  - Non-Oil Growth: Number of Countries 147; Mean 1.886; Std. Dev. 1.842; Min -3.747; Max 7.997
  - Education: Number of Countries 150; Mean 62.544; Std. Dev. 31.308; Min 5.638; Max 115.638
  - FDI: Number of Countries 150; Mean 2.835; Std. Dev. 2.817; Min 0.060; Max 16.406
  - Oil: Number of Countries 147; Mean 0.056; Std. Dev. 0.144; Min 0.000; Max 0.757
  - Lerner Index: Number of Countries 70; Mean 0.249; Std. Dev. 0.096; Min -0.034; Max 0.501
  - H-Stat: Number of Countries 69; Mean 0.635; Std. Dev. 0.185; Min 0.174; Max 1.035
- Table 2a (Oil exporters, observations):
  - Private Credit: Number of Observations 136; Mean 26.347; Std. Dev. 21.558; Min 2.004; Max 136.846
  - Bank Deposits: Number of Observations 131; Mean 30.526; Std. Dev. 22.080; Min 2.080; Max 115.104
  - Liquid Liabilities: Number of Observations 132; Mean 39.024; Std. Dev. 23.869; Min 5.212; Max 123.680
  - Market Cap: Number of Observations 70; Mean 31.157; Std. Dev. 41.278; Min 0.038; Max 198.713
  - Turnover: Number of Observations 70; Mean 21.639; Std. Dev. 23.446; Min 0.144; Max 100.875
  - Growth: Number of Observations 137; Mean 1.280; Std. Dev. 3.144; Min -9.838; Max 9.998
  - Non-Oil Growth: Number of Observations 97; Mean 1.153; Std. Dev. 3.735; Min -10.929; Max 9.847
  - Education: Number of Observations 131; Mean 55.128; Std. Dev. 26.991; Min 6.043; Max 117.992
  - FDI: Number of Observations 137; Mean 2.496; Std. Dev. 3.537; Min -3.073; Max 28.225
  - Oil: Number of Observations 104; Mean 0.250; Std. Dev. 0.197; Min 0.000; Max 0.780
  - Lerner Index: Number of Observations 88; Mean 0.301; Std. Dev. 0.113; Min 0.063; Max 0.501
  - H-Stat: Number of Observations 88; Mean 0.643; Std. Dev. 0.161; Min 0.299; Max 0.991
- Table 2b (Oil exporters, cross-country means):
  - Private Credit: Number of Countries 31; Mean 24.896; Std. Dev. 17.916; Min 2.857; Max 88.680
  - Bank Deposits: Number of Countries 30; Mean 29.241; Std. Dev. 20.258; Min 4.764; Max 92.135
  - Liquid Liabilities: Number of Countries 30; Mean 37.533; Std. Dev. 21.493; Min 12.796; Max 101.873
  - Market Cap: Number of Countries 19; Mean 37.261; Std. Dev. 40.727; Min 6.892; Max 146.005
  - Turnover: Number of Countries 19; Mean 21.705; Std. Dev. 20.291; Min 0.839; Max 67.584
  - Growth: Number of Countries 31; Mean 1.432; Std. Dev. 1.536; Min -1.278; Max 5.473
  - Non-Oil Growth: Number of Countries 31; Mean 1.370; Std. Dev. 2.259; Min -3.747; Max 6.212
  - Education: Number of Countries 31; Mean 56.913; Std. Dev. 27.417; Min 8.862; Max 106.619
  - FDI: Number of Countries 31; Mean 3.235; Std. Dev. 3.720; Min 0.115; Max 16.406
  - Oil: Number of Countries 31; Mean 0.265; Std. Dev. 0.210; Min 0.031; Max 0.757
  - Lerner Index: Number of Countries 19; Mean 0.320; Std. Dev. 0.116; Min 0.063; Max 0.501
  - H-Stat: Number of Countries 19; Mean 0.620; Std. Dev. 0.168; Min 0.299; Max 0.991
- Table 3 (Sample Means by Region — selected entries):
  - Middle East and North Africa: Private Credit 31.474; Bank Deposits 39.186; Liquid Liabilities 51.399; Market Cap 46.140; Turnover 21.196; Growth 1.366; Non-Oil Growth 1.974; Education 66.539; FDI 2.128; Oil 0.238; Lerner Index 0.345; H-Stat 0.529
  - East Asia and Pacific: Private Credit 31.151; Bank Deposits 36.874; Liquid Liabilities 43.005; Market Cap 27.804; Turnover 26.181; Growth 2.552; Non-Oil Growth 2.272; Education 49.848; FDI 3.044; Oil 0.023; Lerner Index 0.255; H-Stat 0.743
  - Rest of the World, Low-Income, Middle-Income groupings: Low-Income Countries Private Credit 17.516; Middle-Income Countries Private Credit 29.783

### Tests for Differences in Means (Table 4)
- P-values for differences in means:
  - Private Credit: VI. No-oil Exporters vs. Oil Exporters 0.0195; VII. All Other Regions vs. Middle East and North Africa 0.3290; VIII. All Other vs. Low Income Countries 0.0000; IX. High Income vs. Low Income Countries 0.0000
  - Bank Deposits: 0.0426; 0.4485; 0.0000; 0.0000
  - Liquid Liabilities: 0.1100; 0.1464; 0.0000; 0.0000
  - Market Cap: 0.1444; 0.0366; 0.0003; 0.0000
  - Turnover: 0.1170; 0.1438; 0.0012; 0.0000
  - Growth: 0.0406; 0.0577; 0.0902; 0.1627
  - Non-Oil Growth: 0.0395; 0.4994; 0.0403; 0.1162
  - Education: 0.1590; 0.4280; 0.0000; 0.0000
  - FDI: 0.2099; 0.2075; 0.3365; 0.4154

### Pairwise Correlations (Table 5)
- Full-sample pairwise correlations (one observation per country) — selected coefficients (asterisks indicate significance at the 5 percent level or better as in source):
  - Private Credit with Bank Deposits: 0.8909*
  - Private Credit with Liquid Liabilities: 0.8567*
  - Private Credit with Market Cap: 0.6135*
  - Turnover with Market Cap: 0.3484*
  - Growth with Non-Oil Growth: 0.8996*
- Pairwise correlations using 5-year averages (selected):
  - Private Credit with Bank Deposits: 0.8697*
  - Private Credit with Liquid Liabilities: 0.8343*
  - Private Credit with Market Cap: 0.5899*
  - Growth with Non-Oil Growth: 0.9480*

### Regression Findings — Private Credit and Growth (Table 6)
- Dependent variable: Real per capita GDP growth and Real per capita non-oil GDP growth (dynamic panel GMM; Arellano and Bover (1995); data averaged over non-overlapping five year periods beginning in 1980).
- Baseline coefficients and significance (selected columns):
  - Private Credit: 0.013*** (column 1); 0.016** (column 2); 0.012* (column 3); 0.015** (column 4); 0.012*** (column 5); 0.018** (column 6); 0.014** (column 7); 0.012 (column 8)
  - Private Credit x Financial Crisis: -0.006*** (columns shown); -0.005***; -0.006***; -0.006***; -0.007***; -0.005***; -0.006***; -0.006**
  - Private Credit x Middle East and North Africa: -0.005* (column shown); -0.009*** (another specification)
  - Private Credit x MEDA: -0.007*; -0.008*
  - Private Credit x non-MEDA: -0.001; 0.000
  - Private Credit x GCC: 0.002; 0.004
  - Private Credit x non-GCC: -0.012**; -0.009*
  - Private Credit x Europe & Central Asia: 0.011**; 0.014**; 0.011*; 0.009; 0.014**; 0.010
- Controls (selected):
  - Education: 0.021**; 0.022**; 0.017**; 0.018*; 0.018*; 0.026**; 0.018*; 0.021** (t-statistics reported)
  - Initial GDP per capita: -0.015***; -0.021***; -0.016**; -0.020***; -0.013***; -0.023***; -0.016**; -0.018**
  - FDI: 0.348***; 0.234*; 0.238*; 0.223*; 0.261***; 0.138; 0.156; 0.205
- Estimation diagnostics (selected):
  - Observations: 678 (varies by specification)
  - Number of countries: 146 (varies)
  - AR2 (p-value): e.g., 0.927; 0.991; 0.966; 0.968; 0.866; 0.984; 0.965
  - Hansen (p-value): e.g., 0.300; 0.419; 0.273; 0.140; 0.480; 0.340; 0.479
  - Number of instruments: e.g., 76; 100; 100; 76; 92; 100; 100
- Wald test statistics for regional coefficient sums (selected): 0.141; 0.433; 0.070; 0.62 (as reported for specified combinations)

### Regression Findings — Oil Exporters Heterogeneity (Table 7)
- Private Credit baseline coefficients (selected):
  - Private Credit: 0.011***; 0.012***; 0.011***; 0.011***; 0.010*; 0.009**; 0.010*; 0.008*
  - Private Credit x Financial Crisis: -0.006*** (all shown columns)
- Interactions with oil exporter variables:
  - Private Credit x Oilexp: -0.007**; -0.004; -0.010**; -0.010
  - Private Credit x Oildep: -0.030***; -0.030**; -0.044***; -0.044***
  - Private Credit x Oilexp x GCC: 0.001; 0.003
  - Private Credit x Oildep x GCC: 0.031*; 0.025
- Controls (selected):
  - Education: 0.017**; 0.015*; 0.017**; 0.016*; 0.015; 0.011; 0.013; 0.012
  - Initial GDP per capita: -0.012***; -0.013***; -0.012***; -0.012**; -0.011**; -0.009*; -0.010**; -0.008*
  - FDI: 0.357***; 0.276***; 0.341***; 0.288***; 0.284***; 0.186; 0.295***; 0.208*
- Estimation diagnostics (selected):
  - Observations: 678; 637; 678; 637; 630; 630; 630; 630
  - Number of countries: 146; 144; 146; 144; 144; 144; 144; 144
  - AR2: 0.832; 0.928; 0.880; 0.928; 0.969; 0.946; 0.950; 0.929
  - Hansen: 0.278; 0.098; 0.328; 0.299; 0.096; 0.066; 0.255; 0.218

### Regression Findings — Income Heterogeneity (Table 8)
- Private Credit and interactions (selected):
  - Private Credit: 0.017*** (col 1); -0.047** (col 2); 0.017***; 0.011**; 0.013***; 0.013**; 0.019*; 0.027**
  - Private Credit x Financial Crisis: -0.006***; -0.006***; -0.006***; -0.010***; -0.010***; -0.009***; -0.006***; -0.006
  - Private Credit x LIC: -0.006; -0.033***; -0.006; -0.011***; -0.011***; -0.006*; -0.041***
  - Private Credit x Income: 0.009***
  - Private Credit x Openness: -0.001; -0.003
  - Private Credit x LIC x Openness: 0.006***; 0.009**
  - Private Credit x Bank Supervision: 0.001; 0.001
  - Private Credit x LIC x Bank Supervision: 0.003; 0.004*
- Controls (selected):
  - Education: e.g., 0.028***; 0.035***; 0.024**; 0.023**; 0.017***; 0.019*; 0.021**; 0.019**
  - Initial GDP per capita: e.g., -0.024***; -0.054***; -0.023***; -0.020***; -0.019***; -0.019***; -0.020***; -0.020***
  - FDI: 0.298**; 0.275***; 0.362**; 0.225; 0.270; 0.227; 0.389***; 0.373***
- Diagnostics (selected):
  - Observations range: 678; 677; 652; 407; 407; 407; 652; 652
  - Number of countries listed examples: 146; 146; 142; 80; 80; 80; 142; 142
  - AR2 and Hansen p-values provided per column in source

### Regression Findings — Stock Market Turnover and Growth (Tables 9–10)
- Table 9 (Heterogeneity Across Regions; dependent variable: Real per capita GDP growth and Real per capita non-oil GDP growth):
  - Turnover baseline coefficients: 0.005**; 0.009**; 0.008**; 0.009**; 0.005**; 0.007*; 0.008**; 0.008*
  - Turnover x Financial Crisis: -0.006***; -0.010***; -0.010***; -0.009***; -0.009***; -0.012***; -0.013***; -0.012***
  - Regional interactions: Turnover x Middle East and North Africa: -0.001; 0.000 (not significant in shown specs)
  - Turnover x Europe & Central Asia: 0.009; 0.008; 0.006; 0.012**; 0.012*; 0.011**
  - Controls: Education and Initial GDP per capita negative and significant as in tables; FDI positive coefficients in some specs (e.g., 0.266*, 0.405**, 0.353*, 0.333*, 0.247*)
- Table 10 (Heterogeneity Between Oil Exporters and Other Countries):
  - Turnover baseline coefficients: 0.004**; 0.004*; 0.004*; 0.004; 0.005**; 0.005**; 0.005**; 0.005**
  - Turnover x Financial Crisis: -0.007***; -0.007***; -0.007***; -0.008***; -0.010***; -0.009***; -0.010***; -0.010***
  - Interactions with oil exporter variables: Turnover x Oilexp: 0.000; 0.002; -0.001; 0.000; Turnover x Oildep: -0.006; 0.015**; -0.006; 0.018
  - Turnover x Oildep x GCC: -0.028***; -0.032
  - Controls: Education generally positive (e.g., 0.023***; 0.022*; 0.021**); Initial GDP per capita negative and significant (e.g., -0.012***; -0.011***); FDI coefficients mixed
  - Diagnostics: Observations around 363; Number of countries around 104; AR2 and Hansen p-values provided per specification

### Notable patterns and empirical takeaways reported in tables
- Private credit is often positively associated with real per capita GDP growth in baseline specifications (coefficients such as 0.013***, 0.011***, 0.017*** in various tables).
- The interaction Private Credit x Financial Crisis consistently shows a negative and statistically significant coefficient (e.g., -0.006***), indicating adverse conditional effects of financial crises on the growth benefits of private credit.
- Heterogeneity across regions and oil-exporter status: interactions with Middle East and North Africa, MEDA, GCC, Oilexp, and Oildep often reduce or reverse the positive association between private credit and growth in specific subsamples (e.g., Private Credit x Oildep -0.030***).
- Stock market turnover shows positive baseline associations with growth (e.g., 0.005**), but the Turnover x Financial Crisis interaction is negative and significant (e.g., -0.007***), indicating financial turmoil reduces the growth contribution of turnover.
- Education and initial GDP per capita are robust controls: Education typically has a positive and often significant coefficient (e.g., 0.021**, 0.028***), while Initial GDP per capita typically has a negative and significant coefficient (e.g., -0.015***, -0.024***), consistent with conditional convergence patterns.

*Source: _wp13130 - REFERENCES (PDF chapter/section).*

### 1980. Robust t-statistics are shown in parentheses, and significance at the 1 percent (***), 5 percent (**), and 10 perc

### _wp13130 - 1980. Robust t-statistics are shown in parentheses, and significance at the 1 percent (***), 5 percent (**), and 10 perc

### Table 11 — Stock Market Turnover and Growth: Heterogeneity Across Income Levels (dynamic panel GMM results)
- Dependent variable: Real per capita GDP growth
- Estimation method: GMM procedure following Arellano and Bover (1995); data averaged over non-overlapping five year periods beginning in 1980.
- Key regressors and reported coefficients (standard errors shown as robust t-statistics in parentheses; significance at the 1 percent (***), 5 percent (**), and 10 percent (*) levels):
  - Turnover:
    - Column (1): 0.007*** (2.771)
    - Column (2): 0.006 (0.563)
    - Column (3): 0.007*** (2.458)
    - Column (4): 0.013*** (2.799)
    - Column (5): 0.011*** (3.768)
    - Column (6): 0.012*** (3.409)
    - Column (7): 0.002 (0.225)
    - Column (8): 0.004 (0.446)
  - Turnover x Financial Crisis:
    - Column (1): -0.009*** (-3.628)
    - Column (2): -0.008*** (-3.374)
    - Column (3): -0.007*** (-3.207)
    - Column (4): -0.015*** (-4.106)
    - Column (5): -0.011*** (-3.628)
    - Column (6): -0.014 (-4.371)
    - Column (7): -0.008*** (-3.840)
    - Column (8): -0.007*** (-2.916)
  - Interactions with variables related to income:
    - Turnover x LIC:
      - Column (1): -0.003 (-0.884)
      - Column (2): 0.019 (0.668)
      - Column (3): -0.004 (-1.761)
      - Column (4): -0.011** (-2.463)
      - Column (5): -0.010* (-1.904)
      - Column (6): -0.002 (-0.674)
      - Column (7): 0.024 (0.930)
    - Turnover x Income:
      - Column (3): 0.000 (0.066)
    - Turnover x Openness:
      - Column (6): 0.002 (0.848)
      - Column (7): 0.001 (0.432)
    - Turnover x LIC x Openness:
      - Column (6): -0.006 (-0.743)
      - Column (7): -0.007 (-1.002)
    - Turnover x Bank Supervision:
      - Column (6): -0.001 (-0.718)
      - Column (7): -0.001 (-0.910)
    - Turnover x LIC x Bank Supervision:
      - Column (6): 0.007* (1.970)
      - Column (7): 0.007 (1.407)
- Controls:
  - Education:
    - Column (1): 0.009 (0.748)
    - Column (2): 0.012 (0.889)
    - Column (3): 0.012 (1.330)
    - Column (4): 0.022** (2.026)
    - Column (5): 0.021** (2.626)
    - Column (6): 0.019*** (2.653)
    - Column (7): 0.003 (0.220)
    - Column (8): 0.009 (0.754)
  - Initial GDP per capita:
    - Column (1): -0.011** (-2.187)
    - Column (2): -0.010 (-1.597)
    - Column (3): -0.011*** (-3.082)
    - Column (4): -0.017*** (-4.721)
    - Column (5): -0.017*** (-4.889)
    - Column (6): -0.016*** (-4.605)
    - Column (7): -0.010** (-2.128)
    - Column (8): -0.011*** (-2.880)
  - FDI (percentage of GDP):
    - Column (1): 0.312** (2.008)
    - Column (2): 0.299* (1.799)
    - Column (3): 0.612*** (5.396)
    - Column (4): 0.008 (1.165)
    - Column (5): 0.283* (1.727)
    - Column (6): 0.296 (1.381)
    - Column (7): 0.533*** (4.734)
    - Column (8): 0.557*** (5.470)
  - Constant:
    - Column (1): -1.389* (-1.931)
    - Column (2): -1.342* (-1.755)
    - Column (3): -2.787*** (-5.337)
    - Column (4): 0.000 (0.000)
    - Column (5): -1.265* (-1.661)
    - Column (6): -1.327 (-1.341)
    - Column (7): -2.397*** (-4.638)
    - Column (8): -2.523 (-5.449)
- Sample and diagnostics:
  - Observations:
    - Columns (1)-(3): 363, 363, 349
    - Columns (4)-(6): 292, 292, 292
    - Columns (7)-(8): 349, 349
  - Number of countries by column:
    - Columns (1)-(3): 104, 104, 100
    - Columns (4)-(6): 74, 74, 100
    - Columns (7)-(8): 100, 100
  - AR2 statistics:
    - Columns (1)-(8): 0.890, 0.820, 0.930, 0.950, 0.978, 0.943, 0.840, 0.891
  - Hansen statistics:
    - Columns (1)-(8): 0.793, 0.834, 0.868, 0.014, 0.638, 0.653, 0.963, 0.975
  - Number of instruments:
    - Columns (1)-(8): 96, 96, 103, 68, 63, 71, 108, 116
- Notes on variables included in some specifications:
  - Interactions between private credit and a Low-Income Country (LIC) dummy variable and/or either the quality of bank supervision (from Abiad, et al, 2008) and the degree of trade openness (ratio of exports plus imports to GDP).
- Interpretation highlights implied by coefficients:
  - Positive and statistically significant coefficients on Turnover in several specifications (e.g., 0.007***, 0.013***, 0.011***, 0.012***) indicate a positive association between stock market turnover (ratio of stock market value traded to GDP) and real per capita GDP growth in those specifications.
  - The negative and statistically significant coefficients on Turnover x Financial Crisis (e.g., -0.009***, -0.008***, -0.007***, -0.015***) indicate that the growth effect of turnover is reduced during financial crises in those specifications.
  - Interaction terms with LIC, openness, and bank supervision show heterogeneous effects across income levels and institutional contexts, with some coefficients marginally significant (e.g., Turnover x LIC x Bank Supervision 0.007*).

### Figures — captions and featured comparisons
- Figure 1: Average Real Per Capita GDP Growth Rates Across Regions, 1975–2005
- Figure 2: Financial Depth Across Regions and Countries
  - Figure 2a: Financial Depth by Region — Stock Market Turnover Ratio, 2007; Private Credit by Deposit Money Banks/GDP, 2008 (Source: World Bank Database on Financial Structure, 2010, and International Financial Statistics)
  - Figure 2b: Financial Depth in Individual MENA Countries, 2007-08 — Stock Market Turnover Ratio, 2007; Private Credit by Deposit Money Banks/GDP, 2008 (Source: World Bank Database on Financial Structure, 2008, and IFS)
  - Numeric labels shown in Figure 2 include values such as 3.6, 5.0, 0.5, 2.4, 1.1, 0.4, 2.3 (region/country markers in the figure)
- Figure 3: Deepening in the Banking Sector, Across Regions, 1975-2008 (time series markers: 1975, 1985, 1995, 2005, 2008)
- Figure 4: The Ratio of Private Credit to Deposits, 1975–2008
- Figure 5: Estimated Impact of Increases in Credit-to-GDP on Real Per Capita Growth (Percentage Points)
  - Panel comparisons include "Quality Effect" and "Pure Depth Effect"
  - Scenarios shown: "Low Banking Depth Countries" and "Mid-to-High Banking Depth Countries"
- Figure 6: Estimated Marginal Impact of Increases in Private Credit-to-GDP on Growth at Different Income Levels (Percentage Points)
- Figure 7: Estimated Differences between LICs and non-LICs in the Growth Impact of Private Credit at Different Levels of Bank Supervision Quality (Percentage Points)
  - Axis labels and markers include values such as -0.04, -0.02, 0.00, 0.02, 0.04, 0.06, 0.08 and log-income ticks 4 5 6 7 8 9 10 11; a 95% confidence band is shown.
- Figure 8: Estimated Differences between LICs and non-LICs in the Growth Impact of Private Credit at Different Levels of Trade Openness (Percentage Points)
  - Axis and ticks include log trade openness values such as 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5
- Figure 9: Estimated Increase in Long-Run Growth from an Increase in Stock Market Turnover by 20 Percentage Points of GDP, at Different Initial Levels of Turnover
  - Horizontal axis annotated with Initial Levels of Stock Market Turnover values including 0.00, 0.05, 0.10, 0.15, 0.20, 0.25, 0.30, 0.35, 0.40, 0.45; other axis markers include 0.1, 0.2, 0.3, ... up to 1.0
- Figure 10: Banking Sector Performance in MENA Countries Relative to Emerging and Developing Country Average and to Sub-Saharan Africa, 2008
- Figure 11: Financial Access, Use of Banking Services, and Depth across Income Groups, 2008
  - Indicators and numeric scales shown include Private credit/GDP (percent), Borrowers From Commercial Banks Per 1000 Adults, Depositors Of Commercial Banks Per 1000 Adults, Number Of ATMs Per 100,000 Adults, Number Of Commercial Bank Branches Per 1000 Km2, Number Of ODC Branches Per 1000 Km2, and depositors/borrowers per 1000 adults series up to values such as 3,500.
  - Note: The last two indicators in the figure are obtained from the World Bank Enterprise Surveys, most responses between 2006 and 2009; some as early as 2003 for a few countries (China mentioned in source note).
- Figure 12: Financial Access and Banking Depth (Privy) Across Countries
- Figure 13: Share of Public and Foreign Banks throughout the World, 2002
  - Country markers include Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, United Arab Emirates, Algeria, Egypt, Jordan, Lebanon, Libya, Morocco, Sudan, Syria, Tunisia, Yemen and axis ranges 0 to 1 for shares.

### Appendix — Country list by region
- Total listed: 150 countries (oil dependent and low income economies are marked by * and ° respectively in the original list).
- East Asia & Pacific (selected entries shown in source):
  - Cambodia°, Fiji, Indonesia*, Lao PDR°, Malaysia*, Mongolia°, Papua New Guinea*°, Philippines, Samoa°, Solomon Islands°, Thailand, Tonga°, Vanuatu°, Vietnam°
- Europe & Central Asia (selected entries):
  - Albania, Armenia°, Bulgaria, Georgia°, Kazakhstan*, Lithuania, Moldova°, Romania, Russian Federation*, Serbia, Turkey, Ukraine, Uzbekistan*°
- Latin America & Caribbean (selected entries):
  - Argentina, Belize, Bolivia*°, Brazil, Chile, Colombia, Costa Rica, Dominica, Dominican Republic°, Ecuador*, El Salvador, Grenada°, Guatemala, Guyana°, Haiti°, Honduras°, Jamaica, Mexico*, Panama, Paraguay, Peru, St. Kitts and Nevis, St. Lucia°, St. Vincent and the Grenadines°, Uruguay, Venezuela, RB*
- Middle East & North Africa (selected entries):
  - Algeria*, Bahrain*, Egypt, Arab Rep.*, Iran, Islamic Rep.*, Jordan, Kuwait*, Lebanon, Libya*, Morocco, Oman*, Qatar*, Saudi Arabia*, Sudan*°, Syrian Arab Republic*, Tunisia*, United Arab Emirates*, Yemen*°
- South Asia:
  - Bangladesh°, Bhutan°, India, Nepal°, Pakistan, Sri Lanka
- Sub-Saharan Africa (selected entries):
  - Angola*, Benin°, Botswana, Burkina Faso°, Burundi°, Cameroon*°, Cape Verde°, Central African Republic°, Chad*°, Congo, Rep.*°, Cote d'Ivoire°, Ethiopia°, Gabon*, Gambia°, Ghana°, Kenya, Lesotho°, Madagascar°, Malawi°, Mali°, Mauritania°, Mauritius, Mozambique°, Namibia, Niger°, Nigeria*°, Rwanda°, Senegal°, South Africa, Swaziland, Tanzania°, Togo°, Uganda°, Zambia°, Zimbabwe
- High-Income Countries (selected entries):
  - Australia, Austria, Bahamas, The, Barbados, Belgium, Brunei Darussalam, Canada*, Croatia, Cyprus, Czech Republic, Denmark, Equatorial Guinea*, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, Korea, Rep., Latvia, Malta, Netherlands, New Zealand, Norway*, Poland, Portugal, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Trinidad and Tobago*, United Kingdom, United States

*Italic source attribution line.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2013/_wp13130.pdf_