Building Integrated Economies in West Africa
Chapter

Chapter 20. Financial Development: Level, Depth, and Access

Author(s):
Alexei Kireyev
Published Date:
April 2016
Share
  • ShareShare
Show Summary Details
Author(s)
Calixte Ahokpossi, Patrick Imam, Kareem Ismail, Sudipto Karmakar, Christina Kolerus and Mesmin Koulet-Vickot 

The level of the financial sector development in the West African Economic and Monetary Union (WAEMU) can be assessed with respect to its depth, breadth, and access to financial services by comparing the level within WAEMU countries to sub-Saharan Africa averages and to individual comparator countries. For each country and each key financial sector indicator, we have estimated a structural benchmark based on the country’s economic and structural characteristics. Comparisons were also made with selected countries outside the WAEMU, namely Ghana and Kenya, as well as with the average and median for sub-Saharan Africa. Ghana is a natural comparator for many WAEMU countries given its characteristics and geographic proximity. Kenya is an example of a sub-Saharan African economy with a rapidly developing financial sector. The mean and median for sub-Saharan Africa (including South Africa) reflect the development of the rest of the continent. To get better sense of the progress needed to achieve a higher level of financial development, the WAEMU’s financial depth can be evaluated relative to a group of high-growth, non-oil-exporter countries. Policy and institutional asymmetries between two groups of countries usually explain the gap in performance.

Structural Characteristics

The WAEMU’s banking systems have significantly deepened in all WAEMU countries in recent years and most of them meet or exceed the main statistical benchmarks for depth. Togo, Senegal, and Benin have the deepest banking systems in the region (in absolute terms, not relative to benchmarks), while Guinea-Bissau and Niger have the shallowest ones. Depth of nonbank financial sectors is very heterogeneous across WAEMU countries. The equity market is clearly underdeveloped relative to structural characteristics. Figure 20.1 provides more detail by country and over time.

Figure 20.1.WAEMU: Selected Indicators on Financial Sector Depth

Sources: FinStats database; and IMF staff calculations.

Note: Three-letter International Organization for Standardization abbreviations used for country names.

1Latest available year was 2010 for Benin; 2009 for Burkina Faso, Côte d’Ivoire, Mali, Niger, and Togo; 2008 for Senegal. Guinea-Bissau was omitted due to missing data.

Structural benchmarks were calculated for WAEMU countries using a large dataset of countries. Each financial indicator was regressed on a set of structural characteristics, such as GDP per capita and its square, population size and density, the age-dependency ratio, country-specific dummies, and year fixed effects. The structural benchmarks were calculated based on Beck and others (2008), Feyen and Kibuuka (2012), and FinStats from the World Bank. A negative difference between the observed value and the benchmark suggests scope for policy action, while a positive difference could reflect successful reforms. A positive difference, however, should not be construed as absence of scope for further development. The benchmarks are not optimal levels, but rather an indication of where countries with similar characteristics stand with regard to financial development. The analysis was carried out using data from 1995 onward, where available, and the tool was developed by the World Bank for this purpose.

Breadth of the banking sector—assessed through the range of products, markets, and providers—is generally limited in the WAEMU. Competition in the banking system, proxied by the asset concentration of the three largest banks, appears relatively low in all countries and seems to have decreased in recent years. Credit to the public sector relative to GDP has increased substantially, a trend that is usually not interpreted positively in terms of breadth, but which reflects in the WAEMU the end of central bank advances to governments and the development of the regional market. Life insurance is more developed than is indicated by the benchmarks, suggesting diversification of the range of financial products. The number of companies whose stock is listed on the regional stock exchange, called the Bourse Régionale des Valeurs Mobiliéres (BRVM), is very low compared with the benchmark. In addition, active trading is limited to a few of these countries. Figure 20.2 provides more detail by country and over time.

Figure 20.2.WAEMU: Comparing the WAEMU with Selected Benchmark Countries

Sources: Central Bank of West African States; FinStats database; and IMF staff calculations.

Note: Simple average for all WAEMU countries. Three-letter International Organization for Standardization abbreviations used for country names. SSA = Sub-Saharan Africa; WAEMU = West African Economic and Monetary Union.

Access to finance has increased substantially in all countries and is in line with or exceeds structural benchmarks in most of them. Both banks and microfinance institutions contributed to this development, with microfinance institutions playing an important role in Benin, Togo, and Senegal. Figure 20.3 provides more detail by country and over time.

Figure 20.3.Selected Indicators on Breadth and Access in the WAEMU

Sources: Central Bank of West African States; FinStats database; and IMF staff estimates.

Note: Three-letter International Organization for Standardization abbreviations used for country names.

While statistical benchmarking shows that the WAEMU is not lagging in terms of financial sector depth and access, direct comparison with selected peer countries suggests substantial scope for further development. An agenda for further research would be to understand what drives differences with comparator countries, and in particular whether certain reforms could be replicated in the WAEMU to further develop the financial system.

  • Depth—Private credit to GDP in the WAEMU is on average comparable to the sub-Saharan African average. It is larger than in Ghana, but significantly lower than in Kenya; in the latter, the increase has also been faster than in the WAEMU.
  • Breadth—While Ghana has managed to reduce significantly asset concentration in the banking system and has reached a level comparable to Kenya’s, concentration remains high in the WAEMU. The stock market is also broader in Kenya.
  • Access—Access is an area where the WAEMU is lagging behind all comparator countries. Kenya and, to a lesser extent, Ghana have made impressive progress in this area in recent years.

Stylized Facts

In the literature, financial depth is generally measured by either the ratio of liquid liabilities to GDP or the ratio of private credit to GDP. Although these two measures are correlated, they have different focuses. The liabilities side measure captures the degree of mobilization of monetary resources as a share of GDP, while the credit measure focuses on the extent to which banks finance economic activity. Figures 20.4, 20.5, and 20.6 compare the WAEMU with the control group along these two basic indicators of financial depth, as well as two measures of efficiency. It appears that the WAEMU is lagging relative to the control group. In the WAEMU, the ratio of private sector credit to GDP increased marginally from 12 percent to 17 percent between 1997–2009, while it accelerated from 17 percent to 37 percent of GDP in the control group (Figure 20.4). The ratio of broad money to GDP, a measure of the degree of monetization, followed a similar profile. Broad money relative to GDP grew from 31 percent to 52 percent in the benchmark countries, while rising from 20 percent to 29 percent in the WAEMU over the period (Figure 20.5).

Figure 20.4.Credit to the Private Sector

(Percent of GDP)

Source: Bankscope database.

Note: HGNOE = high-growth, non-oil-exporter countries; and WAEMU = West African Economic and Monetary Union.

Figure 20.5.Broad Money

(Percent of GDP)

Source: Bankscope database.

Note: HGNOE = high-growth, non-oil-exporter countries; and WAEMU = West African Economic and Military Union.

Figure 20.6.Return on Assets and Return on Equity

Source: Bankscope database.

Note: HGNOE = high-growth, non-oil-exporter countries; and WAEMU = West African Economic and Military Union.

The banking system in the WAEMU is not only shallower compared with the benchmark countries, but it is also less profitable. Return on assets, weighted by bank assets, remained constant at around 1.3 percent in the WAEMU, half of the level found in HGNOEs (Figure 20.6). The contrast is also striking for return on equity (Figure 20.6).

In explaining the differences in financial depth across countries, the empirical literature (Demirgüç-Kunt 2006) distinguishes between structural factors and policy factors. Structural factors are country-specific characteristics that cannot be altered by policies in the short term. These include the overall level of economic development and other characteristics such as population size and density and age dependency. The overall level of development, measured by per-capita income, can affect financial depth by elevating demand for financial services and higher supply of savings. Countries with larger populations and higher population density can have deeper financial penetration and lower cost of financial intermediation from economies of scale. The share of nonworking young and old populations (age dependency) affects savings and lending patterns. Policy factors are those that may impact the banking environment. These include macroeconomic policies (such as inflation, fiscal balance, and debt), institutional policies (regulatory and supervisory frameworks, accounting and disclosures practices, credit information, and contract enforcement), and other financial sector reforms that may liberalize credit markets or enhance market compeititon.

There is evidence in the literature that both structural factors and macroeconomic policies can have an impact on financial deepening. Levine (2003) and Claessens and Feijen (2006) show the importance of overall economic development, measured by per capita income. On macroeconomic policies, Detragiache, Gupta, and Tressel (2005) find a negative impact of inflation on financial depth, while Boyd, Levine, and Smith (2001) highlight the nonlinear relationship between inflation and financial development.

Looking beyond macroeconomic performance, there is also evidence that contract enforcement, credit infrastructure, and market liberalization play an important role. Using bank-level cross-sectional data, Demetriades and Fielding (2012) investigated the determinants of individual banks’ loans in the WAEMU, and found that banks are reluctant to lend because the infrastructure to screen and monitor borrowers is not developed. Governance in all its aspects (government effectiveness, control of corruption, and rule of law) also plays a role. Detragiache, Gupta, and Tressel (2005) found that contract enforcement and property rights matter in financial development. Sacerdoti (2005) explains the low ratio of credit to private sector to GDP by a deficiency in the supporting institutional framework. Ghura, Kpodar, and Singh (2009) explain low financial depth in the CFA franc zone countries through the weaker legal, contractual, and institutional environment in the region compared with sub-Saharan Africa. Using the financial liberalization index constructed by McDonald and Schumacher (2007) that captures some aspects of financial reforms (credit controls, interest rate controls, informal financial sector), Ghura, Kpodar, and Singh (2009) found that this aggregate index is related to greater depth in financial development.

Two Approaches to Assessing

Where does the WAEMU stand relative to some selected economies in terms of its financial depth, and which factors may help explain this performance? The benchmark group of this case study is comprised of some frontier sub-Sarahan African countries—the high-growth, non-oil-exporter countries.1 Our analysis focuses on a group of African countries with shared monetary and financial policies, and compares it with a group of high-performing countries in sub-Saharan Africa. We use two complementary empirical approaches to compare the two groups of countries: first, a regression analysis to identify the factors that explain the difference in financial depth between the WAEMU and the control group; and second, a case study based on the financial benchmark methodology developed by the World Bank (Beck and others 2008; Feyen and Kibuuka 2012). The results are consistent with the literature: stronger rule of law, infrastructure, and credit information collection and dissemination have a strong impact on financial depth.

Financial Depth Determinants

The difference between financial depth in the WAEMU and in the comparator group can be assessed empirically. The measure of financial sector depth is credit to the private sector as a share of GDP, which reflects the extent of financial intermediation in sub-Saharan Africa and the inter-linkage between economic activity and the financial system better than do other liabilities-based measures. The period is sufficiently long to capture the lagged impact of policy reforms on improving the financial environment. The analysis is based on a panel of 16 countries (eight countries from the WAEMU, and eight high-growth, non-oil-exporter countries) over the period 1997–2009 at annual frequency. The panel analysis allows us to track the WAEMU countries over a relatively long time horizon and compare them with other peer economies.

Credit to the private sector as a share of GDP is estimated using the following model:

where the vector χc,t contains macroeconomic variables specific to country c at time t. Yc,t contains country-specific institutional and policy variables. dt is a vector of dummy variables that takes a value 1 if the country belongs to the WAEMU region and zero otherwise.

The macroeconomic indicators include the log of GDP per capita and the log of inflation. Per capita GDP measures the overall level of economic development and is expected to positively affect credit to the private sector. As income rises, demand for financial services increases and that might lead to better penetration. Also with higher incomes, there might be greater savings, which means the banks will have more resources to lend from. Low inflation is considered a sign of macroeconomic stability, which promotes financial intermediation. Thus, the expected coefficient sign for inflation is negative.

Policy and institutional variables include indices on rule of law, political stability, credit coverage, Internet coverage among adults, and the quality of contract enforcement. The rule of law represents a measure of the extent to which banks have faith in contract enforcement, police, and courts, and the likelihood of crime and violence. A strong rule of law is expected to create an environment conducive to bank lending. Property rights capture the dimension of rule of law related to the strength of collateral entitlements and enforcement, which helps banks extend collateralized credit. Political stability, measured here by the perceptions of the likelihood that the government will be destabilized or overthrown by unconstitutional or violent means (Worldwide Governance Indicators), is another factor that we consider in explaining financial penetration. When the political environment is stable, there is less uncertainty, and banks are more willing to lend. The quality of contract enforcement, measured by the number of days required to enforce a contract, is also an important determinant of bank lending. The longer it takes to enforce a contract, the costlier the borrowers’ default for banks, the smaller the amount of credit disbursed by banks. Credit coverage captures the quality of credit information. Credible credit bureaus encourage the expansion of credit as they enable, on the one hand, lenders to better screen borrowers, assess and manage risks, and on the other hand, borrowers to gain access to finance. Internet coverage is used as a proxy for infrastructure development, which reduces the cost of bank penetration and helps improve banks’ geographical coverage. Good Internet coverage indicates a solid telecommunication infrastructure, which is critical to and for bank transactions and transfers.

Tables 20.1 and 20.2, respectively, show some descriptive statistics and correlations. Table 20.3 shows the definition and sources of the variables used in the analysis. Credit to the private sector relative to GDP is most correlated with infrastructure, rule of law, and GDP per capita. Various measures of the quality of legal environment (rule of law, property rights, and political stability) are highly correlated with each other, suggesting that they may be measuring similar attributes of the credit environment.

Table 20.1Descriptive Statistics, 1997–2009
VariableObservationsMeanStd. Dev.MinMax
Credit to private/GDP1680.160.150.010.85
WAEMU_dummy2080.500.500.001.00
Inflation2060.050.04−0.090.16
GDP per capita2086.281.034.738.94
Rule of law160−0.440.66−1.901.05
Political stability160−0.260.88−2.281.13
Internet per hundred2072.554.360.0025.00
Credit coverage672.764.840.0021.80
Property16038.4116.5110.0075.00
Sources: World Development Indicators (World Bank); World Governance Indicators (World Bank), Heritage Foundation Database.Note: Max = maximum; Min = minimum; Std. Dev. = standard deviation; WAEMU = West African Economic and Monetary Union.
Sources: World Development Indicators (World Bank); World Governance Indicators (World Bank), Heritage Foundation Database.Note: Max = maximum; Min = minimum; Std. Dev. = standard deviation; WAEMU = West African Economic and Monetary Union.
Table 20.2Correlation Matrix
123456789
Credit to private/GDP11.00
WAEMU_dummy2−0.261.00
Inflation30.08−0.471.00
GDP per capita40.76−0.260.121.00
Rule of law50.72−0.450.090.741.00
Political stability60.52−0.10−0.050.580.761.00
Internet per hundred70.86−0.260.110.640.560.391.00
Credit coverage80.610.03−0.340.440.330.370.651.00
Property90.65−0.380.030.770.860.580.530.381.00
Source: Authors’ calculations.
Source: Authors’ calculations.
Table 20.3Definition of Variables
VariableDefinitionSource
Credit to private/GDPCredit to the private sector as percentage of GDPInternational Financial Statistics
WAEMU_dummy InflationTake value 1 if the country belongs to the WAEMU Inflation rateInternational Financial Statistics
GDP per capitaGross domestic product per capitaInternational Financial Statistics
Rule of lawRule of law captures perceptions of the extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence. Estimate gives the country’s score on the aggregate indicator, in units of a standard normal. Distribution, that is, ranging from approximately -2.5 to 2.5Worldwide Governance Indicators

(World Bank)
Political stabilityPolitical stability and absence of violence/terrorism captures perceptions of the likelihood that the government will be destabilized or overthrown by unconstitutional or violent means, including politically motivated violence and terrorism. Estimate gives the country’s score on the aggregate indicator, in units of a standard normal distribution, that is, ranging from approximately -2.5 to 2.5.Worldwide Governance Indicators

(World Bank)
Internet per hundredInternet users are people with access to the worldwide network (number per hundred people).Worldwide Governance Indicators

(World Bank)
Credit coveragePublic credit registry coverage reports the number of individuals and firms listed in a public credit registry with current information on repayment history, unpaid debts, or credit outstanding. The number is expressed as a percentage of adult population.Worldwide Governance Indicators

(World Bank)
PropertyAn index measuring the ability of individuals to accumulate private property, secured by clear laws that are fully enforced by the state.Heritage Foundation Database
Source: Authors’ presentation.
Source: Authors’ presentation.

To further the analysis, we use a two-step Feasible Generalized Least Squares to estimate the model previously explained. The estimation approach allows us not only to address issues of heteroscedasticity, but also to estimate the impact of time-invariant variables such as the WAEMU dummy, while controlling for country-specific effects. Several model specifications have been estimated and the results are presented in Table 20.4.

Table 20.4Financial Development in the WAEMU and HGNOE in SSA, 1997–2009
VARIABLES(1)(2)(3)(4)(5)(6)(7)(8)
WAEMU_dummy−0.028−0.003−0.029−0.0100.013−0.0230.0120.075
0.005***0.0070.007***0.0060.0140.006***0.0120.021***
Inflation−0.171−0.180−0.273−0.126−0.063−0.164−0.156−0.094
0.067***0.075***0.0760.072*0.1320.068***0.0970.106
GDP per capita0.1140.0830.0890.0560.0740.0990.0440.059
0.003***0.003***0.004***0.006***0.011***0.004***0.008***0.012***
Rule of law0.0780.0450.077
0.006***0.020***0.020***
Political stability0.0330.009−0.008
0.003***0.0090.009
Internet per hundred0.0200.0180.015
0.002***0.002***0.003***
Credit coverage0.0100.003
0.0040.002
Property0.0010.001−0.003
0.000***0.0010.001***
Constant−0.519−0.300−0.349−0.217−0.337−0.479−0.151−0.166
0.021***0.024***0.0270.030***0.067***0.0230.054***0.056***
Observations1671321321675213812251
Number of countries1717171715171714
Chi square (Wald Test)1231.911638.05727.28489.96116.771267.99588.84204.56
Source: Authors’ calculations.Note: SSA = Sub-Saharan Africa; HGNOE = high-growth, non-oil-exporter countries; WAEMU = West African Economic and Monetary Union.
Source: Authors’ calculations.Note: SSA = Sub-Saharan Africa; HGNOE = high-growth, non-oil-exporter countries; WAEMU = West African Economic and Monetary Union.

The WAEMU dummy is negative and significant even after controlling for macroeconomic variables (Table 20.4, column 1), thereby providing evidence that financial deepening is indeed weaker in the WAEMU than it is in high-growth, non-oil-exporter countries in sub-Saharan Africa. The sign and significance of the WAEMU variable are, however, not stable when policy and institutional factors are accounted for.

As expected, macroeconomic variables are important determinants of credit to the private sector relative to GDP. The impact of inflation is negative and significant in most specifications of the model. Per-capita GDP is positively associated with the financial depth in all specifications.

Looking at institutional/policy variables, we find that political stability, availability of information on borrowers, strength of legal framework, and quality of infrastructure are associated with deeper financial markets (Table 20.4, columns 2–6). The introduction of some of these variables substantially weakens the control variable for WAEMU countries, suggesting that they may explain away most of the differences in financial depth between the WAEMU and the non-oil-exporter countries.

Both rule of law and property rights capture the quality of the legal environment and are highly correlated (Table 20.2). They both are significantly associated with financial depth (Table 20.4, columns 2 and 6). However, the introduction of rule of law in the regression weakens the WAEMU control variable, which becomes insignificant. In other words, once rule of law is accounted for, the difference between the WAEMU group and the non-oil-exporter countries becomes statistically insignificant. This result is in line with the literature that finds that a stronger legal framework promotes financial development because depositors can provide longer-term savings and banks can extend more credit, as they have a greater chance at recovering nonperforming assets through courts.

The results also indicate that political stability matters for financial depth (Table 20.4, column 3). In a politically stable environment, banks have more confidence to lend because there is less uncertainty and a greater chance of recovering their outlays. While political stability matters, it does not explain away the differences between the WAEMU and the non-oil-exporter countries, as the WAEMU control variable remains significant.

Quality of infrastructure, as measured by Internet coverage, positively affects financial depth (Table 20.4, column 4). It also explains away the difference in financial depth between the WAEMU and non-oil-exporter countries. Better infrastructure allows financial institutions to reach borrowers at a lower cost, thereby encouraging more lending.

Information on borrowers as measured by credit coverage is another important factor that is positively associated with financial depth (Table 20.4, column 5). The introduction of credit coverage in the regression also explains away the difference between the WAEMU and non-oil-exporter countries. If banks have more information on their customers, they will screen them better and will engage in greater lending, as they will be less exposed to defaults. Small and medium-sized enterprises and other entrepreneurs with good quality projects and good track records will have access to credit and be able to undertake profitable investment projects.

Overall, the results show that the difference in financial depth between the WAEMU and non-oil-exporter countries is explained by the quality of the legal environment, infrastructure, and information available on borrowers. When rule of law, property rights, and political stability are simultaneously introduced in the model, only rule of law is significant (Table 20.4, column 7), indicating that there is some overlap in the dimension of the institutional environment measured by these variables.

Financial Possibilities Frontier

The concept of financial benchmarking is predicated on the idea that there are structural factors that determine the level of financial development a country can attain. Some factors are conducive to financial sector development (for example, income levels) and others inhibit it (for example, low density of population, which makes infrastructure deployment costly relative to the population served and minimizes the benefit of economies of scale in banking). The benchmarking allows for cross-country comparisons to see how a specific country is doing relative to other countries with similar structural characteristics and at similar stages of development. The benchmarking approach assumes that once appropriate controls are introduced, the process of financial development is broadly comparable across countries and stages of development. Financial sector development is affected by three types of factors: economic development, other structural characteristics, and the policy environment.

where X is an indicator of financial sector development, Y is an indicator of economic development, P represents the policy environment, Z is a vector of structural characteristics, and ε is a residual.

Economic development is captured by income per capita. Demand for financial services increases as income grows. On the supply side, richer countries have better infrastructures and higher competition, which lower the price of financial services. Income per capita is endogenous, but financial sector development affects income per capita with a delay. The reason is that changes in the policy environment affect the financial sector first, and the financial sector in turn then affects economic growth. Therefore, we can write:

The policy environment does not change radically overnight. Good policies today are generally linked to the good policies of yesterday but also to today’s innovations.

The structural variables included in the benchmarking analysis are a set of factors that are considered as external to policy, at least in the short term. These factors include: population, age dependency, a time factor, and special circumstances. Countries with larger populations and higher population density can have deeper and lower cost of providing financial services, thanks to economies of scale. The share of nonworking young and old populations (age dependency) affects saving and lending patterns. Over time, all financial systems tend to improve, albeit at different speed, because of global factors that “lift all boats.” To account for this, a time trend is included in the regression. Many special factors affect financial sector development. In oil exporting countries, for instance, income per capita can be out of proportion with the financial and overall economic development of the country. In contrast, offshore financial centers have a financial sector that is disproportionately larger than the overall economy.

When one runs a regression of financial development on economic development and structural factors only, policy innovations are captured by the residual. To see this, one can substitute equations (2) and (3) into (1), and get the following reduced form expression:

The policy innovation factor is now in the residual. When the benchmark is constructed using the economic and structural variables (Y and Z), the distance between the benchmark and the actual level of financial development is assumed to reflect the country’s policy environment. Countries with better policies (higher u) would tend to have more developed financial sectors compared with countries with worse policies.

FinStats estimates equation (4) via quantile regressions, using data from 177 countries. It then compares a given country to its own potential (benchmark) or to its comparator countries. In the first approach, the country’s benchmark is calculated using its economic and structural variables in equation (4). In the second approach, comparator countries are chosen based on their similarity with the reference country on two dimensions: GDP per capita and populations. The comparator countries are those with the smallest distance to the reference country, where distance is calculated as follows:

where PR is the percentile rank of the country. FinStats uses the expected 25th and 75th percentiles (see Annex 1).

Annex 1.Benchmarking WAEMU Countries Against Their Potential

Source: Author’s calculations.

The results in Annex 1 show that most WAEMU countries are lagging relative to their potential. A closer analysis of the biggest and most financially advanced economy in the WAEMU (Côte d’Ivoire) against a comparable country in the control group (Mozambique) indicates that strength of contract enforcement and credit information constitute the main difference between Côte d’Ivoire and Mozambique over the later part of the observed period. Mozambique’s government took forceful actions in strengthening contract enforcement, with establishment of a specialized commercial court and the introduction of performance measures. As a result, the time taken to resolve a dispute fell by 72 percent (Doing Business 2008). Furthermore, a new legal framework for credit registries has been enacted in Mozambique, which resulted in expanding the scope and accessibility of credit information. Several other reforms undertaken by Mozambique’s authorities over this period helped increase the flow of credit (FSAP 2009). These include: (1) transitioning from an overall compliance-supervisory regime to risk-based supervision; (2) enhancing the financial infrastructure by significantly improving the national payments system; and (3) new legislation on microfinance.

References

    Beck, T., E.Feyen, AlainIze, and F.Moizeszowicz.2008. “Benchmarking Financial Development.” Policy Research Working Paper, World Bank, Washington.

    • Search Google Scholar
    • Export Citation

    Boyd, J. H., R.Levine, and B. D.Smith.2001. “The Impact of Inflation on Financial Sector Performance.” Journal of Monetary Economics47 (2): 221–48.

    • Search Google Scholar
    • Export Citation

    Claessens, S., and E.Feijen.2006. “Finance and Hunger: Empirical Evidence of the Agricultural Productivity Channel.” Policy Research Working Paper Series 4080, World Bank, Washington.

    • Search Google Scholar
    • Export Citation

    Demetriades, P., and D.Fielding.2012. “Information, Institutions, and Banking Sector Development in West Africa.” Economic Inquiry50: (3).

    • Search Google Scholar
    • Export Citation

    Demirgüç-Kunt, A.2006. “Finance and Economic Development: Policy Choices for Developing Countries.” Policy Research Working Paper Series 3955, World Bank, Washington.

    • Search Google Scholar
    • Export Citation

    Detragiache, E., P.Gupta, and T.Tressel.2005. “Finance in Lower Income Countries: An Empirical Exploration.” Working Paper WP/05/167, International Monetary Fund, Washington.

    • Search Google Scholar
    • Export Citation

    Feyen, E., and K.Kibuuka. (2012). “FinStats 2012: A Ready-to-Use Tool to Benchmark Financial Sectors Across Countries and Time.” Unpublished, World Bank.

    • Search Google Scholar
    • Export Citation

    Ghura, D., K.Kpodar, and R. J.Singh.2009. “Financial Deepening in the CFA Franc Zone: The Role of Institutions.” Working Paper WP/09/113, International Monetary Fund, Washington.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund (IMF). 2010. “Sub-Saharan Africa: Resilience and Risks.” In Regional Economic Outlook, October.

    Levine, R.2003. “More on Finance and Growth: More Finance, More Growth?” Federal Reserve Bank of St. Louis Review85 (4): 31–46.

    McDonald, C., and L.Schumacher.2007. “Financial Deepening in Sub-Saharan Africa: Empirical Evidence on the Role of Creditor Rights Protection and Information Sharing.” Working Paper WP/07/203, International Monetary Fund, Washington.

    • Search Google Scholar
    • Export Citation

    Sacerdoti, E.2005. “Access to Bank Credit in Sub-Saharan Africa.” Working Paper WP/05/166, International Monetary Fund, Washington.

1

High-growth, non-oil-exporter countries are the countries with an average per capita growth rate of at least 3 percent during 1995–2009. Eight frontier sub-Saharan African countries fall into this category: Botswana, Cape Verde, Ethiopia, Mauritius, Mozambique, Rwanda, Tanzania, and Uganda (IMF 2010).

    Other Resources Citing This Publication