## _wp09182

## Source details

**Canonical URL:** [_wp09182](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09182.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09182.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09182.pdf.json)

---

### III. Introduction — scope and approach
- Study objective:
  - Determinants of financial market development in Africa, focusing on banking systems and stock markets using a panel of 53 countries for the period 1990 to 2006.
  - Banking sector modeled on a 53-country panel; stock market modeled on an 18-country panel.
- Theoretical baselines:
  - Banking sector regressions: Mckinnon–Shaw hypothesis (financial repression hinders financial development).
  - Stock market regressions: Calderon–Rossell model (stock market development is a function of income level and stock market liquidity).
- Methodology notes:
  - Banking sector: includes a lagged dependent variable and uses the system generalized method of moments (GMM).
  - Stock market: uses fixed effects and instrumental variable techniques; system GMM used cautiously because of small cross-section (maximum 18).
- Key empirical variables:
  - Banking: credit to the private sector relative to GDP; commercial bank assets relative to total financial assets of the banking sector.
  - Stock market: stock market capitalization relative to GDP; stock market liquidity (value traded/GDP); income level; capital account liberalization; macroeconomic and institutional variables.

### IV. Main empirical findings (summary)
- Banking sector development determinants:
  - Income level is an important determinant.
  - Creditor rights protection is important (creditor rights index ranges from 0 to 4).
  - Financial repression (reserve requirements proxy) matters negatively.
  - Political risk is an important negative determinant.
- Stock market development determinants:
  - Stock market liquidity (value traded as a percentage of GDP) is a main determinant.
  - Gross domestic savings matter.
  - Banking sector development contributes positively to stock market development (banks and markets are complements).
  - Political risk is an important negative determinant.
- Interaction effects:
  - Capital account liberalization promotes financial market development only in countries with high-income levels or well-developed institutions (threshold effects).
- Policy-relevant interpretation:
  - Political risk reduction and institutional improvement are central for both banking and stock market development.
  - Macroeconomic stability and creditor rights protection significantly affect banking sector development.

### V. Stylized facts and selected quantitative points
- Stock exchanges and listings:
  - Number of stock markets in Africa: rose from 5 in 1990 to 18 currently (starting from about 5 in 1989).
- Panels and sample sizes:
  - Banking sector regression: panel of 53 countries (1990 to 2006).
  - Stock market regression: panel of 18 countries (1990 to 2006).
- Banking sector depth and reach:
  - On average bank credit to the private sector represents no more than 15 percent of GDP in Africa.
  - In developed economies, bank credit to the private sector is more than 100 percent of GDP.
  - Banking services penetration is as low as 5 percent in some countries.
  - Capital adequacy ratio averages 16 percent of risk-weighted assets.
- Stock market growth and concentration:
  - Total market capitalization increased by 113 percent between 1995 and 2005.
  - South Africa and Egypt together account for more than 50 percent of all listed companies in the continent.
  - Headline liquidity comparison: "0.04 percent in Swaziland compared with about 31 percent in Mexico."
- Market structure and institutions:
  - Banking systems are generally very concentrated and often dominated by foreign banks.
  - The ratio of M1 to M2 is the highest in the world for Africa (cash remains dominant).
  - Legal and regulatory constraints include slow court proceedings, absence of credit assessment information, weak property-rights protection, and under-resourced regulators.

### VI. Stock market size, liquidity, and country indicators (selected 2007 figures)
- Key observation: "Low liquidity implies more difficulty in supporting a local market with its own trading systems, market analysis, and brokers because business volume is too low."
- Selected table entries (Number of Listed Companies | Stock Market Capitalization/GDP | Value Traded/GDP | Turnover):
  - Botswana: 18 | 41.90 | 0.92 | 2.2
  - Cote d' Ivoire: 40 | 320.82 | 2.5
  - Egypt: 60 | 391.241.4 | 45.4
  - Ghana: 32 | 18.60 | 0.73 | 3.8
  - Kenya: 51 | 42.24 | 4.5 | 10.6
  - Morroco: 65 | 85.535.9 | 41.9
  - Mauritius: 41 | 73.15 | 5.8 | 7.9
  - Namibia: 9 | 9.30 | 0.33 | 3.7
  - Nigeria: 202 | 35.910.1 | 28.2
  - South Africa: 401 | 280.8153.4 | 54.6
  - Tanzania: n.a | 40.12 | 2.1
  - Tunisia: 48 | 14.11 | 9.13 | 13.2
  - Uganda: n.a | 1.20 | 0.15 | 5.2
  - Zambia: n.a | 15.60 | 0.64 | 4.1
  - Argentina: 103 | 31.93 | 3.19 | 9.8
  - Brazil: 392 | 79.344.5 | 56.1
  - Chile: 244 | 118.927.1 | 22.8
  - Mexico: 1314 | 212.930.8
  - Malaysia: 1027 | 1568353.2
  - Thailand: 476 | 64.345.1 | 70.2
- Source for table: Beck et al. (2008).

### VII. Infrastructure, private equity, and bond markets
- Infrastructure and market microstructure constraints:
  - Trading, clearing, and settlement systems are often slow; some transactions can take months (Senbet, 2008); many exchanges still operate manual systems.
  - Most markets lack central depository systems; some restrict foreign participation; these bottlenecks slow information production and turnover.
- Returns and international investor opportunities:
  - Despite small size and low liquidity, returns on African markets have generally been high; Sharpe-adjusted returns similar to Latin America and Asia (Senbet, 2008).
  - African markets offer diversification opportunities with low correlation to the global system.
- Private equity market:
  - OECD: "investments in private equity reached US$2.3 billion in 2006."
  - Notable transactions and funds: buyout of Celtel for "US$3.4 billion"; Pamodzi announced "US$1.3 billion pan-African fund"; Rennaissance Capital launched "US$1 billion pan-African investment fund."
  - About 31 fund managers active in Africa; about seven private equity funds dedicated to infrastructure (Ndiaye, 2008).
  - South Africa: share of 80 percent of sub-Saharan African private capital; Nigeria: 10 percent (Santiso, 2007).
  - South Africa: total funds under management roughly 2.8 percent of GDP in 2007 versus global average 2.1 percent and European average 1.9 percent (KPMG and SAVCA, 2008).
  - Sources of funds: United States provides 50 percent; South Africa provides 25 percent (one-third of that from pension and endowment funds); Europe about 9 percent; China bought a 20 percent stake in Standard Bank for US$56.5 billion (figure as reported).
  - Risk premiums in Africa decreased from 8.9 percent in 2006 to 6.7 percent in 2008.
  - Drivers: better macro environment (high growth, low inflation), high returns driven by cheap labor and low competition. Constraints: weak deal-flow systems, poor governance, political instability.
- Bond market in Africa:
  - Long-term debt market is the least developed segment; growth mainly from government sector.
  - Nongovernment domestic debt increased from US$91 million in 2001 to US$801 million in 2006 (seven sub-Saharan countries referenced).
  - Foreign currency debts dominate; local currency debt mainly short-term (maturities usually less than a year; 3 to 6 months popular).
  - Local commercial banks and institutional investors account for about 70 percent of outstanding debt.
  - Foreign participation example: end-June 2007 foreigners held about 11 percent of Ghana’s domestic currency government debt (more than US$400 million) and 14 percent of Zambia’s.
  - 2007 sovereign issuance: Ghana issued US$750million; Gabon issued US$1 billion (Linn and Nagy, 2008).
  - Problems: small size, low liquidity, lack of long-term maturities, limited investor base; recommended mitigants include sound macro policy, market infrastructure, incentives for participation, and yield curve development.

### VIII. Literature on determinants of financial market development (key themes)
- Financial development promotes economic growth (Levine and Zervos, 1998).
- Determinants identified in literature:
  - Time-invariant fixed factors: colonization, geography (latitude, landlocked status), natural resource endowments.
  - Country characteristics: ethnic fractionalization, cultural differences.
  - Legal origin: common law more conducive to capital market development than civil law (La Porta et al., 1997).
  - Creditor rights and credit-information-sharing affect private credit (Djankov, McLiesh and Schleifer, 2007).
  - Openness: mixed literature; evidence that both trade openness and financial openness are important (Baltagi, Demetriades, and Law, 2008); Chinn and Ito (2002, 2006) show financial openness helps only when legal/institutional quality has reached a certain level.
  - Macroeconomic conditions: inflation above a threshold is harmful; income and savings are positively related to financial development.
- Stock market–specific determinants:
  - Financial liberalization promotes transparency and reduces adverse selection (Mishkin, 2001).
  - GDP growth, domestic investment, and financial intermediary development are important (Garcia and Liu, 1999).
  - Nonlinear relationship with banking development: initial complementarity then competition as markets mature (Yartey, 2008b).
- Research gap: limited theoretical and empirical work on determinants of financial market development in developing economies.

### IX. Methodology for banking sector analysis (details)
- Sample: 53 African countries, 1990–2006.
- Baseline dynamic panel specification (schematic): yi,t = α yi,t-1 + β1 M + β2 L + β3 I + ηi + εi,t (lags included).
- Dependent variables:
  - Credit to the private sector relative to GDP.
  - Bank assets relative to total banking system assets.
- Explanatory variables:
  - Macro: GDP per capita (lagged), inflation volatility, domestic savings and investment.
  - Liberalization: reserve requirements (ratio of bank liquid reserves to bank assets), trade openness, financial openness (Chinn–Ito index).
  - Institutional: creditor rights index (0 to 4), political risk (ICRG 100 point scale), law and order, bureaucratic quality, democratic accountability, corruption.
  - Capital flows tested: FDI, remittances.
- Identification and instruments:
  - Two lags of endogenous variables used as instruments.
  - System GMM used to address endogeneity, predetermined variables, and country fixed effects.

### X. Empirical results — Banking sector (highlights from Tables 2 and 3)
- Dependent = Credit to the private sector relative to GDP (Table 2, System GMM):
  - Lagged dependent (models 1–7): 0.999; 0.998; 0.926; 0.857; 0.843; 0.824; 0.868 (std. errors (0.037)***, (0.026)***, (0.022)***, (0.055)***, (0.082)***, (0.075)***, (0.027)***).
  - GDP per capita (models 1–7): 0.054; 0.056; 0.157; 0.175; 0.192; 0.169; 0.127 (std. errors (0.046), (0.042), (0.056)***, (0.063)***, (0.080)**, (0.077)*(0.080)).
  - Reserve requirements (models 1–7): -0.037; -0.047; -0.130; -0.014; -0.003; -0.041; -0.009 (std. errors (0.021)*(0.020)**(0.035)***(0.023)(0.020)(0.021)*(0.017)).
  - Creditor rights (models 1–7): 0.145; 0.163; 0.137; 0.140; 0.037; 0.140; 0.223 (std. errors (0.023)***(0.035)***(0.040)***(0.035)***(0.048)(0.037)***(0.063)***).
  - Trade openness (models 1–7): 0.088; 0.077; 0.468; 0.074; 0.117; 0.001; -0.082 (std. errors (0.051)*(0.023)***(0.132)***(0.051)(0.059)**(0.052)(0.062)).
  - SD annual inflation (selected): -0.857 (0.389)**; other specifications report varied estimates including -2.993 (1.799)* and -1.040 (0.367)***.
  - Financial Openness coefficients include -0.031; -0.039; 0.036; -0.207 (std. errors (0.011)***(0.014)***(0.028)(0.090)**).
  - Remittances (model 7): 0.058 (0.012)***.
  - Observations by model: 350; 352; 322; 322; 322; 308; 296. Number of groups: 28; 28; 27; 27; 27; 26; 28.
  - AR(2) (p-value) by model: 0.2348; 0.1669; 0.1416; 0.1741; 0.1334; 0.3579; 0.3811.
  - Sargan (p-value) by model: 0.5985; 0.7057; 0.9605; 0.9716; 0.9976; 0.9986; 0.849.
- Dependent = Bank assets relative to total assets (Table 3, System GMM):
  - Lagged dependent (models 1–7): 0.540; 0.689; 0.507; 0.477; 0.485; 0.454; 0.431 (std. errors (0.020)***,(0.009)***,(0.042)***,(0.028)***,(0.038)***,(0.027)***,(0.027)***).
  - GDP per capita (models 1–7): 0.647; 0.112; 0.676; 0.711; 0.714; 0.564; 0.262 (std. errors (0.047)***(0.042)***(0.062)***(0.028)***(0.036)***(0.076)***(0.035)***).
  - Reserve requirements (models 1–7): -0.061; -0.063; 0.017; -0.080; -0.078; -0.041; -0.029 (std. errors (0.007)***(0.006)***(0.018)(0.008)***(0.008)***(0.008)***(0.006)***).
  - SD annual inflation (selected): -1.165 (0.370)***; -2.617 (0.726)***; -2.398 (0.575)***; -2.763 (0.701)***; -3.420 (0.416)*** in various models.
  - Trade Openness * Financial Repression (model 3): 0.151 (0.026)***.
  - Financial openness * Law and Order (model 6): 0.090 (0.036)**.
  - Remittances: 0.036 (0.004)***.
  - Observations by model: 346; 346; 316; 316; 316; 302; 290. Number of groups: 29; 29; 29; 29; 29; 28; 29.
  - AR(2) (p-value): 0.5860; 0.3591; 0.4971; 0.5234; 0.5157; 0.4146; 0.4271. Sargan (p-value): 0.6748; 0.5369; 0.4183; 0.3981; 0.4262; 0.9691; 0.8275.

### XI. Empirical results — Stock market development (highlights from Tables 4 and 5)
- Panel data (Table 4; dependent = Stock market capitalization relative to GDP):
  - L. GDP per capita (models 1–7): 0.218; 0.575; 0.116; 1.228; 0.027; 0.189; 0.075 (std. errors (0.114)*(0.331)*(0.135)(0.469)**(0.168)(0.151)(0.147)).
  - Stock Market liquidity (models 1–7): 0.214; 0.255; 0.243; 0.212; 0.202; 0.239; 0.220 (std. errors (0.028)***,(0.029)***,(0.028)***,(0.036)***,(0.030)***,(0.035)***,(0.036)***).
  - Private Credit (models 1–7): 0.521; 0.267; 0.464; 0.567; 0.514; 0.402; 0.523 (std. errors (0.134)***(0.115)**(0.110)***(0.178)***(0.146)***(0.131)***(0.124)***).
  - L. Gross domestic savings: 0.155; 0.287; 0.216; 0.177; 0.145; 0.222; 0.095 (std. errors (0.060)**(0.055)***(0.058)***(0.066)***(0.061)**(0.091)**(0.097)).
  - Observations by model: 134; 184; 146; 122; 114; 108; 99. Number of groups: 14; 16; 15; 14; 13; 12; 12.
  - R-squared by model: 0.69; 0.67; 0.75; 0.65; 0.62; 0.79; 0.78.
  - Income elasticity reported: a 1 percentage point increase in GDP per capita increases stock market development by 0.2 percentage points (panel estimate).
  - Bank–market relationship: a 1 percentage point increase in bank credit/GDP increases stock market development by 0.5 percentage points (panel estimate).
  - Stock market liquidity impact: a 1 percentage point increase in value traded/GDP increases market capitalization by 0.2 percentage points.
  - Domestic savings impact: a 1 percentage point increase in gross domestic savings/GDP increases market capitalization by 0.2 percentage points.
- GMM estimates (Table 5; dependent = Stock market capitalization relative to GDP):
  - GDP per capita (models 1–7): 0.082; 0.208; 0.437; 0.650; -0.006; 0.209; 0.335 (std. errors (0.213),(0.647),(0.500),(0.956),(0.245),(0.283),(0.496)).
  - Stock Market liquidity (models 1–7): 0.216; 0.303; 0.293; 0.233; 0.216; 0.238; 0.190 (std. errors (0.032)***(0.067)***(0.049)***(0.078)***(0.035)***(0.039)***(0.066)***).
  - Private Credit (models 1–7): 0.647; 0.555; 0.466; 0.815; 0.562; 0.406; 0.416 (std. errors (0.202)***(0.237)**(0.207)**(0.346)**(0.230)**(0.212)*(0.250)*).
  - Gross domestic savings (models 1–7): 0.319; 0.720; 0.579; 0.456; 0.337; 0.358; 0.006 (std. errors (0.109)***(0.265)***(0.176)***(0.201)**(0.131)**(0.178)**(0.377)).
  - Inflation (%) (selected): 1.831 (0.811)** in one model.
  - Financial Openness coefficients include -0.047; -0.675; -5.317 (std. errors (0.091),(0.255)***(1.987)***).
  - GDP * Financial Openness (interaction): 0.100 (0.035)***.
  - Political Risk: 0.985; 0.956 (std. errors (0.355)***(0.419)**).
  - Political Risk * Financial Openness: 1.258 (0.460)***.
  - Observations by model: 127; 180; 137; 114; 116; 102; 94. Number of groups: 14; 14; 14; 12; 14; 12; 12.
  - R squared reported: 0.7798; 0.4489; 0.61; 0.5206; 0.7679; 0.7986; 0.8275.
- Threshold and simulation findings:
  - Marginal effect of capital account liberalization on stock market capitalization is positive if real GDP per capita is above 7.2 in logarithm (roughly US$1,300).
  - Effect of financial openness on market capitalization is positive for values of political risk greater than 4.27 in logarithm (corresponding to 71.5).
  - Simulations (window variations and iterations noted) show financial openness has a positive effect on stock market development at higher GDP per capita and lower political risk.

### XII. Summary, conclusions, and policy implications
- Main determinants of banking sector development:
  - Creditor rights protection, income level, trade openness, financial repression (negative), and political risk.
  - Macroeconomic mismanagement and financial repression discourage demand for external finance.
- Main determinants of stock market development:
  - Domestic savings, stock market liquidity, bank credit, and institutional quality (law and order, government stability).
  - Institutional quality reduces political risk and supports external finance mobilization.
- Capital account liberalization:
  - No robust positive relationship overall; positive impacts only above threshold levels of GDP per capita and below thresholds of political risk.
- Bank–market relationship:
  - Evidence that banking sector and stock market development are complements in Africa.
- Policy recommendations (derived from empirical results):
  - Reduce political risk and improve institutions (law and order, government stability).
  - Implement more open trade and sound macroeconomic policies.
  - Protect creditor rights and alleviate financial repression (reduce reserve requirements where appropriate).
  - Increase stock market liquidity and promote domestic savings.
  - Build financial intermediary capacity to deepen banking–market linkages.
- Data limitations noted:
  - Privatization and pension reforms/pension funds likely important but not analyzed due to data limitations.

*Source: _wp09182 — excerpts and figures from the IMF working paper content provided in the supplied PDF.*

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

### _wp09182 - References..............................................................................................................

### III. Introduction — scope and approach
- Study objective:
  - Determinants of financial market development in Africa, focusing on banking systems and stock markets.
  - Uses a panel of 53 countries for the period 1990 to 2006.
  - Separately models banking sector development (53-country panel) and stock market development (18-country panel).
- Theoretical baselines:
  - Banking sector regressions: Mckinnon–Shaw hypothesis (financial repression hinders financial development).
  - Stock market regressions: Calderon Rossell model (stock market development is a function of income level and stock market liquidity).
- Methodology notes:
  - Banking sector: includes a lagged dependent variable to capture dynamics and uses the system generalized method of moments (GMM) as preferred estimator.
  - Stock market: uses fixed effects and instrumental variable techniques because system GMM is not designed for a small number of cross-section units (maximum 18).
- Key empirical variables analyzed:
  - Banking sector indicators: credit to the private sector relative to GDP; commercial bank assets relative to total financial assets of the banking sector.
  - Stock market indicator: stock market capitalization relative to GDP; stock market liquidity; income level; capital account liberalization; macroeconomic and institutional environment.

### IV. Summary of main empirical findings
- Determinants of banking sector development:
  - Income level is an important determinant.
  - Creditor rights protection is important.
  - Financial repression (lack of liberalization) matters.
  - Political risk is an important negative determinant.
- Determinants of stock market development:
  - Stock market liquidity is a main determinant.
  - Domestic savings matter for stock market development.
  - Banking sector development contributes to stock market development.
  - Political risk is an important negative determinant.
- Interaction effects:
  - Capital account liberalization promotes financial market development only in countries with high-income levels, well-developed institutions, or both.
- Policy-relevant interpretation:
  - Political risk strongly affects both banking sector and stock market development, suggesting political risk resolution is important for financial market development in Africa.
  - Macroeconomic stability and creditor rights protection are significant for banking sector development.

### V. Stylized facts and documented trends (selected quantitative points)
- Stock exchanges and listings:
  - Number of stock markets in Africa: rose from 5 in 1990 to 18 currently.
  - Starting from about 5 in 1989, there are now 18 stock exchanges in Africa.
- Panels and sample sizes:
  - Banking sector regression: panel of 53 countries (1990 to 2006).
  - Stock market regression: panel of 18 countries (1990 to 2006).
- Banking sector depth and reach:
  - On average bank credit to the private sector represents no more than 15 percent of GDP in Africa.
  - In developed economies, bank credit to the private sector is more than 100 percent of GDP.
  - Banking services penetration is as low as 5 percent in some countries.
  - Capital adequacy ratio averages 16 percent of risk-weighted assets.
- Stock market growth and concentration:
  - Total market capitalization increased by 113 percent between 1995 and 2005.
  - South Africa and Egypt together account for more than 50 percent of all listed companies in the continent.
  - Turnover ratios are reported as “as little as” (statement truncated in source).
- Market structure and institutions:
  - Banking systems are generally very concentrated and often dominated by foreign banks.
  - The ratio of M1 to M2 is the highest in the world for Africa (cash remains dominant).
  - Legal and regulatory environment constraints: slow court proceedings, absence of credit assessment information, weak property-rights protection, regulators lacking independence and resources.

### VI. Policy implications and recommendations (derived from empirical results)
- To stimulate banking sector development:
  - Reduce political risk.
  - Improve institutions and creditor rights protection.
  - Implement more open trade and sound macroeconomic policies.
  - Promote financial liberalization where appropriate.
- To stimulate stock market development:
  - Increase stock market liquidity.
  - Promote domestic savings.
  - Promote financial intermediaries (to deepen banking sector linkages).
  - Reduce political risk.

*Source: _wp09182 - References (extracts and figures) from the IMF chapter content provided.*

### 0.04 percent in Swaziland compared with about 31 percent in Mexico. Low liquidity implies

### _wp09182 - 0.04 percent in Swaziland compared with about 31 percent in Mexico. Low liquidity implies

### Stock market size, liquidity, and country indicators (Africa and selected emerging markets, 2007)
- Key observation: "Low liquidity implies more difficulty in supporting a local market with its own trading systems, market analysis, and brokers because business volume is too low."
- Table of indicators (Number of Listed Companies | Stock Market Capitalization/GDP | Value Traded/GDP Turnover) — Africa and selected countries:
  - Botswana: 18 | 41.90 | 0.92 | 2.2
  - Cote d' Ivoire: 40 | 320.82 | 2.5
  - Egypt: 60 | 391.241.4 | 45.4
  - Ghana: 32 | 18.60 | 0.73 | 3.8
  - Kenya: 51 | 42.24 | 4.5 | 10.6
  - Morroco: 65 | 85.535.9 | 41.9
  - Mauritius: 41 | 73.15 | 5.8 | 7.9
  - Namibia: 9 | 9.30 | 0.33 | 3.7
  - Nigeria: 202 | 35.910.1 | 28.2
  - South Africa: 401 | 280.8153.4 | 54.6
  - Tanzania: n.a | 40.12 | 2.1
  - Tunisia: 48 | 14.11 | 9.13 | 13.2
  - Uganda: n.a | 1.20 | 0.15 | 5.2
  - Zambia: n.a | 15.60 | 0.64 | 4.1
  - Argentina: 103 | 31.93 | 3.19 | 9.8
  - Brazil: 392 | 79.344.5 | 56.1
  - Chile: 244 | 118.927.1 | 22.8
  - Mexico: 1314 | 212.930.8
  - Malaysia: 1027 | 1568353.2
  - Thailand: 476 | 64.345.1 | 70.2
- Source for table: Beck et al. (2008)
- Headline comparison cited: "0.04 percent in Swaziland compared with about 31 percent in Mexico."

### Infrastructure and market microstructure constraints
- Findings:
  - "African stock markets suffer from infrastructural bottlenecks."
  - "Trading, clearing, and settlement systems are so slow it can take months to execute a single transaction (Senbet, 2008), and most of the exchanges still operate manual systems."
  - "Most markets do not have central depository systems, and some restrict foreign participation."
  - These bottlenecks "slow information production, hamper activity and turnover, and render financial integration difficult" and "induce inactivity (Yartey and Adjasi, 2007)."

### Returns and international investor opportunities
- Findings:
  - "Despite the problems of small size and low liquidity, returns on African markets have generally been high."
  - Senbet (2008): after controlling for risk (Sharpe ratio) returns are similar to those realized in Latin America and Asia even when converted into dollars.
  - Implication: African markets "represent unexploited opportunities for international investors" and provide "diversification opportunities that are minimally lowly correlated with the global system and its risk."

### Private equity market in Africa
- Definition: "Private equity refers to any type of investment in an asset in which the equity is not freely tradable on a public stock market" and "refers to how the funds have been raised—on private rather than public markets."
- Recent trends and amounts:
  - OECD: "investments in private equity reached US$2.3 billion in 2006."
  - Buyout of Celtel for "US$3.4 billion."
  - 2007: Pamodzi Investments Holdings announced a "US$1.3 billion pan-African fund"; Rennaissance Capital launched a "US$1 billion pan-African investment fund."
- Market structure and concentration:
  - "About 31 fund managers active in Africa" and "about seven private equity funds are dedicated to infrastructure (Ndiaye, 2008)."
  - South Africa: "share of 80 percent of sub-Saharan African private capital"; Nigeria: "another 10 percent (Santiso, 2007)."
  - South Africa: "Total funds under management were roughly 2.8 percent of GDP in 2007—higher than the global average of 2.1 percent and the European average of 1.9 percent (KPMG and SAVCA, 2008)."
- Sources of funds:
  - "United States provides 50 percent of the capital of private equity funds in Africa."
  - "South Africa is second with 25 percent, of which a third is raised from pension and endowment funds (Santiso, 2007)."
  - "Europe contributes about 9 percent of the capital, primarily from public funds of European development finance institutions such as Proparco (France) and CDC (UK)."
  - "China also recently bought a 20 percent stake in Standard Bank, the largest bank in South Africa, for US$56.5 billion."
- Risk premia:
  - "Risk premiums in Africa decreased from 8.9 percent in 2006 to 6.7 percent in 2008."
- Drivers and constraints:
  - Drivers: "better macroeconomic environment characterized by high growth and low inflation" and "high returns ... due to cheap labor, little competition, low rents, and therefore higher margins."
  - Constraints: "absence of systems and institutions to ease deal flows and the exit of private equity funds, poor governance, and political instability."
  - Institutional and regulatory quality "important to the proper functioning of the private equity market in Africa (Yartey, 2007b)."

### Bond market in Africa
- Overall state:
  - "The market for long-term debt is the least developed segment of Africa’s capital market."
  - Growth momentum mainly from the "government sector."
  - Corporate debt markets have "generally lagged the government bond market."
- Evidence of nongovernment domestic debt expansion:
  - "Increase in the volume of nongovernment domestic debt, from US$91 million in 2001 to US$801 million in 2006" (seven sub-Saharan countries referenced).
- Currency and maturity profile:
  - "Foreign currency debts dominate the debt market in Africa."
  - "Local currency debt is mainly short-term with maturities usually less than a year; maturities of 3 to 6 months are the most popular."
- Issuance and investor base:
  - Issuers include "governments, regional development banks, and corporations."
  - "Issuance of local currency debt is erratic and small in volume, leading to problems in developing fungible and liquid instruments and benchmarks."
  - "Local commercial banks and institutional investors account for about 70 percent of outstanding debt."
- Foreign participation examples:
  - "At the end of June 2007 foreigners held about 11 percent of Ghana’s domestic currency government debt (more than US$400 million) and 14 percent of Zambia’s (Linn and Nagy, 2008)."
  - 2007 sovereign issuance: "Ghana issued US$750million in bonds and Gabon issued US$1 billion (Linn and Nagy, 2008)."
- Market microstructure problems and implications:
  - Problems: "small size, low liquidity, lack of long-term maturities, and limited investor base."
  - Consequences: lack of "reliable yield curves, pricing benchmarks, and financial products to hedge risk."
  - Recommended mitigants: "sound macroeconomic policy and a stable political environment"; "building up market infrastructure, including trading, information dissemination, clearing and settlement systems"; "incentives that reinforce good market participation; and developing a yield curve would also be beneficial."
- Comparative figures (bond market capitalization):
  - International comparison charts referenced showing "Private Bond Market Capitalization" and "Public Bond Market Capitalization" by region, and time series "Africa: Bond Market Capitalization (Percent of GDP)" for 1994–2006 (source: Financial structure database- Beck et al. 2008).

### Literature on determinants of financial market development
- Themes and findings:
  - Financial development promotes economic growth (e.g., Levine and Zervos, 1998).
  - Determinants of financial market development include:
    - Time-invariant fixed factors: "colonization" and geography (latitude, landlocked status, natural resource endowments).
    - Country characteristics: "ethnic fractionalization" and cultural differences.
    - Legal origin: "common law basis is more conducive to the development of capital markets than a civil law basis" (La Porta et al., 1997).
    - Creditor rights and credit-information-sharing affect private credit (Djankov, McLiesh and Schleifer, 2007).
    - Openness: mixed views—sequencing literature (McKinnon, 1991) vs. simultaneous openness (Rajan and Zingales, 2003a, 2003b); evidence that "both trade openness and financial openness are important" (Baltagi, Demetriades, and Law, 2008); Chinn and Ito (2002, 2006) show financial openness helps only when legal/institutional quality has reached a certain level.
    - Macroeconomic conditions: "inflation above a certain level brings adverse consequences" while "income level and savings are positively related to financial development."
  - Stock market–specific determinants:
    - Financial liberalization promotes transparency and accountability, reducing adverse selection and moral hazard (Mishkin, 2001).
    - GDP growth, domestic investment, and financial intermediary development are important (Garcia and Liu, 1999).
    - "Economic growth, financial liberalization, and foreign portfolio investments" lead emerging stock market growth (El-Wassal, 2005).
    - Nonlinear relationship with banking sector development: stock markets initially supported by banking development, then compete as they mature (Yartey, 2008b).
- Research gap: "little work, theoretical or empirical, on what determines financial market development in developing economies."

### Methodology for banking sector development analysis
- Data and sample:
  - Panel data techniques used for "53 African countries" covering "1990 to 2006."
- Baseline empirical model (structure):
  - Dependent variable modeled as function of lagged dependent variables, macroeconomic variables (M), liberalization variables (L), institutional variables (I), and fixed effects:
    - Equation schematic provided: ,0,11,2,3,, itititititiit yy MLI  
- Dependent variables:
  - Two alternative measures of bank development:
    - "Credit to the private sector relative to GDP"
    - "Bank assets relative to total banking system assets"
  - Rationale:
    - "Credit to the private sector ... isolates credit issued to the private sector."
    - "Bank assets relative to GDP measures the degree to which commercial banks rather than central banks allocate savings."
- Explanatory variables and identification:
  - Macro variables: "GDP per capita, inflation volatility, and domestic savings and investment."
  - Liberalization variables: "reserve requirements, trade openness, and financial openness."
  - Institutional variables: "creditor rights and political risk rating."
  - Two lags of endogenous variables included to address reverse causality and endogeneity.
- Financial repression proxy:
  - "Bank reserve requirements" used as proxy; measured by "the ratio of bank liquid reserves to bank assets."
  - Expected sign: negative (reserve requirements impede financial deepening).
- Creditor rights:
  - "Creditor rights index developed by Djankov, McLiesh, and Schleifer (2005)." Index ranges "from 0 (weak creditor rights) to 4 (strong)."

*Italic: Source: _wp09182 (excerpt) — IMF working paper content provided in the supplied PDF content.*

### 2003. The measure aims at assessing collateral and bankruptcy laws, following La Porta

### _wp09182 - 2003. The measure aims at assessing collateral and bankruptcy laws, following La Porta et al.

### Scope and purpose
- Objective: assess collateral and bankruptcy laws as part of examining the impact of institutional quality on financial development (following La Porta et al., 1997).
- Institutional focus: country performance in providing an environment for secure financial transactions.

### Key explanatory variable groups and expectations
- Income level
  - Use: lagged values of real GDP per capita.
  - Rationale: higher incomes associated with better education, better property rights, improved business environment, enforcement of legal rights, and accounting quality — all supporting financial and stock market development (La Porta et al., 1997).

- Macroeconomic environment
  - Use: lagged values of savings or investments.
  - Expectation: positive determinants of banking sector development because intermediaries and stock markets allocate savings to investment projects.
  - Note: "The correlation of savings and investments with incomes is low in our sample."

- Macroeconomic stability
  - Use: lagged values of inflation, inflation volatility, and real interest rates.
  - Expectation: inflation and inflation volatility negatively impact financial market development; instability reduces profitability and willingness to lend.

- Financial openness
  - Measure: index of capital account liberalization developed by Chin and Ito (2006).
  - Expectation: financial openness reduces cost of capital, increases availability, and can improve financial infrastructure by weeding out inefficient institutions.

- Trade openness
  - Measure: lagged value of (imports + exports) relative to GDP.
  - Expectation: openness affects demand for external finance through specialization, sectoral structure, and technology transfer; can raise risks so well-developed financial markets that diversify risk may evolve with trade activities.

- Capital flows
  - Variables tested: foreign direct investment and remittances.
  - Reference: Aggarwal, Demirguc-Kunt, and Martinez Peria (2006) — worker remittances significant determinant of financial development.

- Political and legal institutions
  - Measure: political risk composite index from the International Country Risk Guide (ICRG).
  - Scale: 100 point scale; higher rating (maximum, theoretically, is 100) indicates lower risk.
  - Expectation: higher institutional quality positively correlated with financial development.
  - Limitation: the composite indicator gives limited guidance on which institutional aspects to target for policy.
  - Additional analysis: four ICRG components studied for stock market development—law and order, bureaucratic quality, democratic accountability, and corruption.

### Estimation strategy and identification
- Panel data techniques used to handle:
  - joint endogeneity of many explanatory variables,
  - unobserved country-specific effects correlated with regressors,
  - dynamics from lagged endogenous variables.
- Problems with standard estimators:
  - OLS, GLS, and fixed effects inconsistent when explanatory variables are endogenous or correlated with country-specific effects.
  - First-differencing creates correlation between the differenced error term and lagged dependent variables.
- Dynamic panel GMM (Arellano and Bond, 1991)
  - Uses lagged values of endogenous regressors and lagged/current values of strictly exogenous regressors as instruments.
  - Variables can be estimated in levels or first differences; difference estimator uses first-differenced variables with lagged instruments.
  - Identifying assumption required: no second-order serial correlation in first differences of the error term.
  - Specification tests: Sargan test of over-identifying restrictions and test of lack of residual serial correlation.
  - Note: first-order serial correlation expected by construction in differenced residuals; only second and higher order serial correlation indicate misspecification.
- System GMM (Blundell and Bond, 1998)
  - Proposed when lagged levels are weak instruments for current differenced variables (series near random walk).
  - Combines difference and levels equations; uses suitably lagged differenced variables as instruments in levels equation.
  - Consistent and more efficient than difference GMM if differenced regressors are not significantly correlated with country fixed effects.
  - Efficiency gains depend on how close series are to a random walk.
- Practical implementation details:
  - Two lags of endogenous variables are used to address reverse causality and endogeneity biases.
  - Usual practice: estimate with two-step estimator but base hypothesis tests on one-step estimator’s statistics due to potential underestimation of standard errors in two-step.

### Modeling determinants of stock market development
- Behavioral model lineage:
  - Calderon-Rossell model: economic growth and stock market liquidity as main determinants of stock market development (annual observations, 1980–87, 42 markets).
  - Yartey (2008) modified model to include additional financial, economic, and institutional variables.
- Baseline empirical specification (following Yartey, 2008):
  - General form presented: ,1,2,3,, ititititiit yM FIβββηε=+ + + + 
    - where Mt is vector of macroeconomic variables (including GDP per capita, real interest rate, domestic savings and investment),
    - Ft is vector of financial variables (including stock market liquidity, private credit, financial openness),
    - It is vector of institutional variables (including political risk rating).
  - Sample: 17 African countries with stock market activity between 1990 and 2006.

- Dependent variable
  - Stock market capitalization: deflated value of listed shares relative to GDP.
  - Interpretation: market size positively correlated with ability to mobilize capital and diversify risk economy-wide.
  - Stock-flow adjustment: use average of two consecutive year-end market capitalizations to estimate mid-year value.
  - Deflation formula provided exactly: {(0.5)*[Ft/P_et + Ft-1/P_et-1]}/[GDPt/P_at] where F is stock market capitalization, P_e is end-of period CPI, and P_a is average annual CPI.

- Explanatory variables for stock market regressions
  - Macroeconomic: lagged gross domestic savings as a percentage of GDP; lagged values of real interest rates; in further estimations real interest rate replaced by lagged inflation and budget deficits as a percentage of GDP.
  - Institutional: political risk.
  - Financial: stock market liquidity, private credit, financial openness.
  - Liquidity measure: lagged total value traded as a percentage of GDP; expectation: positive correlation with stock market development.
  - Banking sector development: deposit money bank credit to the private sector as a percentage of GDP; prior evidence (Yartey, 2008b) indicates a non-linear relationship with stock market development in emerging markets.
  - Identification: two lags of endogenous variables used as instruments in GMM specifications; lagged values of explanatory variables used as instruments.

### Data properties and stationarity notes
- Unit root tests cited: Levin et al. (2002) and Im et al. (2003) show most variables are I(0); for some variables results vary, complicating panel cointegration analysis.
- System GMM requirement note: series must be mean stationary (implying constant mean across time for each country).

*Source: _wp09182 - 2003. The measure aims at assessing collateral and bankruptcy laws, following La Porta et al., IMF working paper PDF content unit provided.*

### appendix for further details).

### V. EMPIRICAL RESULTS

### Banking Sector Development
- Dependent variable: credit to the private sector relative to GDP (results in Table 2).
- Baseline (Model 1) explanatory variables: GDP per capita, reserve requirements, trade openness, creditor rights, standard deviation of inflation, and political risk.
- Key empirical findings:
  - Political risks, creditor rights protection, and trade openness are positively associated with banking sector development.
  - Reserve requirements and inflation volatility are negatively associated with banking sector development.
  - Economic instability (inflation volatility) reduces bank activities and assets.
  - Financial repression (restrictions on interest rates and mandated credit allocation) negatively impacts banking sector development.
- Inflation effects:
  - Current inflation initially found positive and significant; after removing very high inflation outliers (Model 2) current inflation becomes negative and statistically significant.
  - Further analysis recommended (threshold regressions) to clarify the inflation–financial development relationship.
- Capital account liberalization:
  - Using the Chinn-Ito index (Model 4), capital account liberalization negatively affects banking sector development.
  - Interaction results: capital account liberalization positively influences banking sector development only in countries with good legal systems (Model 6 interaction with law and order).
- Trade and openness interactions:
  - Trade openness is positive and statistically significant in Model 1.
  - Simultaneous opening of trade and financial accounts tends to promote banking sector development (support for the simultaneous openness hypothesis).
  - However, trade openness negatively affects bank credit when reserve requirements are high (Models 3 and 5).
- Institutional components:
  - Law and order and government stability are important in explaining banking sector development (components of political risk).
- Other determinants:
  - Private capital flows, such as inward remittances, are positively associated with banking sector development (Model 7).
- Robustness:
  - Alternative dependent variable — commercial bank assets relative to total assets of the banking system (Table 3) — yields similar results to private credit relative to GDP.

### Determinants of Stock Market Development
- Dependent variable: market capitalization as a percentage of GDP (fixed effects estimation; results in Table 4).
- Baseline (Model 1) explanatory variables: GDP per capita, bank credit, stock market liquidity, real interest rate, gross domestic savings.
- Key empirical findings:
  - Bank credit, stock market liquidity (value traded as a percentage of GDP), gross domestic savings, and GDP per capita are significant and have positive effects on stock market development.
  - Real interest rate has a negative effect but is not statistically significant.
  - Income elasticity: a 1 percentage point increase in GDP per capita increases stock market development by 0.2 percentage points.
  - Bank–market relationship: a 1 percentage point increase in bank credit as a percentage of GDP increases stock market development by 0.5 percentage points — evidence that banks and stock markets are complements.
  - Stock market liquidity: a 1 percentage point increase in value traded as a percentage of GDP increases market capitalization by 0.2 percentage points.
  - Domestic savings: a 1 percentage point increase in gross domestic savings as a percentage of GDP increases market capitalization by 0.2 percentage points.
- Inflation effects:
  - Replacing real interest rate with last year’s inflation (Model 2) yields inflation with an unexpected positive and statistically significant sign.
  - After removing inflation swings larger than 20 percent (Model 3), inflation levels do not significantly affect stock market capitalization.
  - Inflation volatility and extreme inflation (>100) found not statistically significant in explaining stock market development.
- Budget deficits:
  - Replacing inflation with budget deficits as a percentage of GDP yields no conclusive evidence of impact on stock market development.
- Capital account liberalization (Chinn-Ito index):
  - Financial openness has a negative effect on stock market development in Model 4.
  - Interaction analyses (Models 5 and 7) show capital account liberalization has a negative impact generally, but in countries with sufficiently high income and low political risk, capital account liberalization has a positive effect.
  - Thresholds and simulations:
    - Marginal effect of capital account liberalization on stock market capitalization is positive if real GDP per capita is above 7.2 in logarithm (corresponding to roughly US$1,300).
    - Effect of financial openness on market capitalization is positive for values of political risk greater than 4.27 in logarithm (corresponding to 71.5).
    - Rolling and recursive simulations (window from median to 95th percentile) indicate interaction effects: financial openness has a positive effect on stock market development for higher GDP per capita; the negative effect of financial openness on stock market development decreases with higher values of risk (i.e., lower political risk).
- Political risk and institutional components:
  - Political risk appears positive and significant: a 1 percentage point improvement in the ICRG political risk rating increases stock market development by 0.9 percentage points (Model 6).
  - Components most important for Africa: law and order and government stability.
- Overall determinants emphasized:
  - Banking sector development, domestic savings, GDP per capita, and stock market liquidity are important determinants.
  - Institutional development (government stability, law and order) reduces political risk, enhances regulatory capacity, and supports external finance.
  - No conclusive evidence that macroeconomic stability more broadly has an impact.
  - Capital account liberalization has a positive impact on stock market development only in higher-income and low political risk countries.

### Summary and Conclusion
- Methods: panel data techniques and dynamic panel estimator.
- Main determinants of banking sector development:
  - Creditor rights protection, income level, trade openness, financial repression (negative), and political risk.
  - Good institutions stimulate efficient supply of external finance; poor institutions form structural impediments.
  - Macroeconomic mismanagement and financial repression discourage demand for external finance and derail banking system development.
- Main determinants of stock market development:
  - Domestic savings, stock market liquidity, bank credit, and institutional quality (La Porta et al. framework supported).
  - Institutional quality reduces political risk and supports investment decisions.
- Capital account liberalization:
  - No positive and robust relationship with stock market development overall.
  - Existence of threshold levels of real GDP per capita and political risk below which capital liberalization negatively affects market capitalization.
  - Simulation results: overall marginal effect of capital account liberalization on market capitalization is positive in high-income and low-political risk countries.
- Bank–market relationship:
  - Banking sector and stock market development in Africa are complements, not substitutes.
- Policy implications:
  - Reduce political risk and improve institutions (law and order, government stability).
  - Implement more open trade and sound macroeconomic policies.
  - Better protect creditor rights.
  - Increase stock market liquidity, promote domestic savings, and build the financial intermediary sector to stimulate stock market development.
- Data limitations and additional factors noted:
  - Privatization and pension reforms/pension funds likely important but not analyzed due to data limitations.

*Source: appendix for further details).*

### REFERENCES

### _wp09182 - REFERENCES

### References (bibliography)
- The section compiles cited works including:
  - Acemoglu, D., S. Johnson, and J.A. Robinson (2001), “The Colonial Origins of Comparative Development: An Empirical Investigation,” American Economic Review, Vol. 9, No. 5, pp.1369–1401.
  - Aggarwal, R., A. Demirguc-Kunt, and M.S. Martinez Peria (2006), “Do Workers’ Remittances Promote Financial Development?” Policy Research Working Paper Series 3957 (Washington: World Bank).
  - 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, No. 2, pp. 277–97.
  - Blundell, R., and S. Bond (1998), “Initial Conditions and Moment Restrictions in Dynamic Panel Data Models, Journal of Econometrics, Vol. 87, No. 1, pp. 115–43.
  - Beck, T., A. Demirgüç-Kunt, and R. Levine (2000), "A New Database on Financial Development and Structure," World Bank Economic Review, Vol. 14, pp. 597–605 (Updated 2008).
  - La Porta, R., F.L. de Silanes, A. Shleifer, and R.W. Vishny (1997), “Legal Determinants of External Finance,” Journal of Finance, Vol. 52, No. 3, pp. 1131–50.
  - Levine, R. (1997), “Financial Development and Economic Growth: Views and Agenda,” Journal of Economic Literature, Vol. 35, No. 2, pp. 688–726.
  - Arellano and Bond (1991), Anderson and Hsiao (1981), Blundell and Bond (1998) and related methodological references on dynamic panel estimation.
- (The full References list contains additional entries cited in the paper.)

### Appendix — Variable lists (Tables 1a and 1b)
- Table 1a. Explanatory Variables Used for Banking Sector Regression
  - Income: GDP per capita
  - Macroeconomic environment: Inflation; Inflation volatility; Real interest rate
  - Domestic investment; Domestic savings
  - Financial openness + capital flows: Financial openness; Reserve requirements; Remittances; FDI; Capital flows
  - Institutions / Trade / Openness: Creditor rights; Risk; Trade openness; Law and order; Corruption; Bureaucratic quality; Government stability
- Table 1b. Explanatory Variables Used for Stock Market Regression
  - Income and Macro Environment: GDP per capita; L. Gross domestic savings; Investments
  - Macroeconomic environment: Inflation; Inflation volatility; L. Real interest rate
  - Financial development: Stock market liquidity; Private credit/GDP
  - Financial openness: Financial openness; FDI; Capital flows; Remittances
  - Institutions / Trade / Openness: Trade openness; Risk; Law and order; Corruption; Bureaucratic quality; Government stability

### Key empirical estimates — Banking sector (Table 2: System GMM, dependent = Credit to the private sector relative to GDP)
- Lagged dependant (models 1–7): 0.999; 0.998; 0.926; 0.857; 0.843; 0.824; 0.868 (standard errors in parentheses: (0.037)***, (0.026)***, (0.022)***, (0.055)***, (0.082)***, (0.075)***, (0.027)***).
- GDP per capita (models 1–7): 0.054; 0.056; 0.157; 0.175; 0.192; 0.169; 0.127 (standard errors: (0.046), (0.042), (0.056)***, (0.063)***, (0.080)**, (0.077)*(0.080)).
- Reserve requirements (models 1–7): -0.037; -0.047; -0.130; -0.014; -0.003; -0.041; -0.009 (standard errors: (0.021)*(0.020)**(0.035)***(0.023)(0.020)(0.021)*(0.017)).
- Creditor rights (models 1–7): 0.145; 0.163; 0.137; 0.140; 0.037; 0.140; 0.223 (standard errors: (0.023)***(0.035)***(0.040)***(0.035)***(0.048)(0.037)***(0.063)***).
- Trade openness (models 1–7): 0.088; 0.077; 0.468; 0.074; 0.117; 0.001; -0.082 (standard errors: (0.051)*(0.023)***(0.132)***(0.051)(0.059)**(0.052)(0.062)).
- SD annual inflation (selected models): -0.857 (0.389)**, 0.291 (0.630), -0.174 (0.608), -2.993 (1.799)*(model 4), -2.702 (1.184)**(model 5), -1.040 (0.367)***.
- Financial openness and interaction terms included in later models: Financial Openness coefficients include -0.031; -0.039; 0.036; -0.207 (with standard errors (0.011)***(0.014)***(0.028)(0.090)**). Trade Openness * Financial Openness coefficient 0.126 (0.045)***.
- Remittances coefficient (model 7): 0.058 (0.012)***.
- Observations by model: 350; 352; 322; 322; 322; 308; 296.
- Number of groups by model: 28; 28; 27; 27; 27; 26; 28.
- AR(2) (p-value) by model: 0.2348; 0.1669; 0.1416; 0.1741; 0.1334; 0.3579; 0.3811.
- Sargan (p-value) by model: 0.5985; 0.7057; 0.9605; 0.9716; 0.9976; 0.9986; 0.849.

### Key empirical estimates — Banking sector composition (Table 3: System GMM, dependent = Bank assets relative to total assets)
- Lagged dependant (models 1–7): 0.540; 0.689; 0.507; 0.477; 0.485; 0.454; 0.431 (std. errors (0.020)***,(0.009)***,(0.042)***,(0.028)***,(0.038)***,(0.027)***,(0.027)***).
- GDP per capita (models 1–7): 0.647; 0.112; 0.676; 0.711; 0.714; 0.564; 0.262 (std. errors (0.047)***(0.042)***(0.062)***(0.028)***(0.036)***(0.076)***(0.035)***).
- Reserve requirements (models 1–7): -0.061; -0.063; 0.017; -0.080; -0.078; -0.041; -0.029 (std. errors (0.007)***(0.006)***(0.018)(0.008)***(0.008)***(0.008)***(0.006)***).
- SD annual inflation (selected): -1.165 (0.370)***(model 1), -2.617 (0.726)***(model 2), -2.398 (0.575)***(model 3), -2.763 (0.701)***(model 4), -3.420 (0.416)***(model 5), -0.341 (0.238)(model 6).
- Trade Openness * Financial Repression (model 3): 0.151 (0.026)***.
- Financial openness * Law and Order (model 6): 0.090 (0.036)**.
- Remittances (models): 0.036 (0.004)***.
- Observations by model: 346; 346; 316; 316; 316; 302; 290.
- Number of groups: 29; 29; 29; 29; 29; 28; 29.
- AR(2) (p-value) by model: 0.5860; 0.3591; 0.4971; 0.5234; 0.5157; 0.4146; 0.4271.
- Sargan (p-value) by model: 0.6748; 0.5369; 0.4183; 0.3981; 0.4262; 0.9691; 0.8275.

### Key empirical estimates — Stock market development (Panel data, Table 4; dependent = Stock market capitalization relative to GDP)
- L. GDP per capita (models 1–7): 0.218; 0.575; 0.116; 1.228; 0.027; 0.189; 0.075 (std. errors (0.114)*(0.331)*(0.135)(0.469)**(0.168)(0.151)(0.147)).
- Stock Market liquidity (models 1–7): 0.214; 0.255; 0.243; 0.212; 0.202; 0.239; 0.220 (std. errors (0.028)***,(0.029)***,(0.028)***,(0.036)***,(0.030)***,(0.035)***,(0.036)***).
- Private Credit (models 1–7): 0.521; 0.267; 0.464; 0.567; 0.514; 0.402; 0.523 (std. errors (0.134)***(0.115)**(0.110)***(0.178)***(0.146)***(0.131)***(0.124)***).
- L. Gross domestic savings series: 0.155; 0.287; 0.216; 0.177; 0.145; 0.222; 0.095 (std. errors (0.060)**(0.055)***(0.058)***(0.066)***(0.061)**(0.091)**(0.097)).
- Observations by model: 134; 184; 146; 122; 114; 108; 99.
- Number of groups: 14; 16; 15; 14; 13; 12; 12.
- R-squared reported by model: 0.69; 0.67; 0.75; 0.65; 0.62; 0.79; 0.78.
- Hausman test (p-value) across models: 0.59; 0.000; 0.000; 0.03; 0.82; 0.28; 0.31.

### Key empirical estimates — Stock market development (GMM, Table 5; dependent = Stock market capitalization relative to GDP)
- GDP per capita (models 1–7): 0.082; 0.208; 0.437; 0.650; -0.006; 0.209; 0.335 (std. errors (0.213),(0.647),(0.500),(0.956),(0.245),(0.283),(0.496)).
- Stock Market liquidity (models 1–7): 0.216; 0.303; 0.293; 0.233; 0.216; 0.238; 0.190 (std. errors (0.032)***(0.067)***(0.049)***(0.078)***(0.035)***(0.039)***(0.066)***).
- Private Credit (models 1–7): 0.647; 0.555; 0.466; 0.815; 0.562; 0.406; 0.416 (std. errors (0.202)***(0.237)**(0.207)**(0.346)**(0.230)**(0.212)*(0.250)*).
- Gross domestic savings (models 1–7): 0.319; 0.720; 0.579; 0.456; 0.337; 0.358; 0.006 (std. errors (0.109)***(0.265)***(0.176)***(0.201)**(0.131)**(0.178)**(0.377)).
- Inflation (%) (selected): 1.831 (0.811)** (one model).
- Financial Openness coefficients include -0.047; -0.675; -5.317 (std. errors (0.091),(0.255)***(1.987)***).
- GDP * Financial Openness (model): 0.100 (0.035)***.
- Political Risk coefficients: 0.985; 0.956 (std. errors (0.355)***(0.419)**).
- Political Risk * Financial Openness: 1.258 (0.460)***.
- Observations by model: 127; 180; 137; 114; 116; 102; 94.
- Number of groups: 14; 14; 14; 12; 14; 12; 12.
- R squared (reported): 0.7798; 0.4489; 0.61; 0.5206; 0.7679; 0.7986; 0.8275.
- Hansen (equation exactly identified) noted for instrument validity.

### Dynamic System GMM — methodological notes (appendix)
- The dynamic equation estimated: yi,t = α yi,t-1 + β' Xi,t + ηi + εi,t (presented in the source as a dynamic panel specification).
- Estimator comparisons and properties:
  - OLS inconsistent because ηi is correlated with yi,t-1.
  - LSDV (fixed effects) eliminates ηi but is inconsistent for finite T with N → ∞ due to correlation between transformed errors and lagged dependent variables.
  - Anderson and Hsiao (1981) propose differencing and IV using yi,t-2 and yi,t-3 as instruments — consistent but inefficient.
  - Arellano and Bond (1991) difference GMM (AB) uses orthogonality conditions between lagged levels and differenced errors; can treat explanatory variables as predetermined or endogenous; known to generate finite-sample bias and weak instruments in some settings.
  - Kiviet (1995), Bun and Kiviet (2003), Bruno (2005a) derive small-sample bias approximations and propose bias-corrected LSDV (Bruno extends to unbalanced panels).
  - Blundell and Bond (1998) system GMM (BB) augments AB by:
    - using lagged differences as instruments for level equations, and
    - using lagged levels as instruments for differenced equations.
  - Moment conditions for levels and conditions for strictly exogenous, predetermined, and endogenous regressors are specified; combining levels and differences may introduce redundant moment conditions.

### Figures and Simulation results — summary of simulations
- Figure 1a: Political Risk, Income Level and Capital Account Liberalization — shows stock market capitalization (Log) against Capital account liberalization with linear fits and 95% CI for subgroups "when lagged GDP below threshold", "when lagged GDP greater than threshold", "when Political risk rating below threshold", and "when Political risk rating greater than threshold".
- Simulation 1: Financial openness and stock market capitalization with income level as the threshold variable
  - Simulation methodology equation provided in the source (including DUMMYKAOPEN, KAOPEN, Z, X, y, η, ε).
  - Window variation: [5th percentile, 95th percentile]
  - Number of iterations: 20
  - Estimation: robust random effects (including RISK in the control variables)
  - Dotted lines are 90% confidence intervals
- Simulation 2: Financial openness and stock market capitalization with political risk as the threshold variable: Recursive simulation including interaction variable
  - Simulation methodology equation provided (THRESHOLDKAOPEN etc.)
  - Window variation: [50th percentile, 95th percentile]
  - Number of iterations: 50
  - Estimation: robust random effects
  - Dotted lines are 90% confidence intervals
- Simulation 3: Financial openness and stock market capitalization with political risk as the threshold variable: Threshold simulation
  - Simulation methodology equation provided (DUMMYKAOPEN etc.)
  - Window variation: [5th percentile, 95th percentile]
  - Number of iterations: 50
  - Estimation: robust random effects
  - Dotted lines are 90% confidence intervals

*Source: _wp09182 - REFERENCES (PDF chapter/section).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09182.pdf_