Information about Sub-Saharan Africa África subsahariana
Chapter

2. The Impact of Global Financial Stress on Sub-Saharan African Banking Systems

Author(s):
International Monetary Fund. African Dept.
Published Date:
May 2012
Share
  • ShareShare
Information about Sub-Saharan Africa África subsahariana
Show Summary Details

INTRODUCTION AND SUMMARY

In the last few years, the world has experienced three episodes of global financial stress. The most significant took place after the collapse of the U.S. investment bank Lehman Brothers in late 2008. The other episodes, in mid-2010 and late 2011, have involved the deterioration of European public finances and financial conditions. The theme that runs through these three episodes is a generalized increase in global risk aversion (Figure 2.1). This chapter explores the channels through which global financial stress affects sub-Saharan African banking systems,1 reviewing the effects of the 2008–09 financial turmoil and the recent European crisis.

Figure 2.1.Global Risk Aversion, 2005–121

Source: J.P. Morgan (updated on April 2,2012).

1 VIX is the Chicago Board of Options Exchange Volatility Index of the U.S. SP500 stock market index, while the VDAX is the German equivalent.

The main findings are the following:

  • Most sub-Saharan African banking systems have proved resilient to the recent episodes of global financial stress.2 Although there has been pressure on loan quality, profitability, and bank liquidity, most banking systems withstood stress without experiencing a crisis.
  • The channels for transmitting financial stress are primarily indirect.3 Direct linkages are limited by the relatively modest scale of banks’ foreign assets and liabilities. Most banking systems in the region have small cross-border liabilities, rely on stable domestic retail funding, and are typically quite liquid. The key indirect channel is the impact of global financial stress on world growth and trade, with attendant spillovers on commodity prices, external demand, economic activity, and, as a result, on the quality of loan portfolios.
  • The macroeconomic impact of the 2008–09 crisis on sub-Saharan Africa’s economies was marked but short lived. Sharp exchange rate movements posed challenges for banks or firms with net open foreign exchange positions, especially in a few dollarized economies.
  • Recent European financial distress is not expected to significantly disrupt most sub-Saharan African banking systems. First and foremost, baseline projections in the April 2012 World Economic Outlook suggest a much more modest slowing of world growth and trade than was observed in 2008–09 (Chapter 1). Secondly, the low financial integration of sub-Saharan African economies limits the direct exposure of most banking systems to worsening financial conditions abroad. However, intensification of the crisis in the euro area along the lines discussed in Chapter 1 would likely impose further strains on the quality of loan portfolios, albeit without creating severe distress.

Although external developments do not pose an immediate threat to the region’s banking systems, there is no room for complacency in regard to ensuring financial stability.

  • For the 16 sub-Saharan African jurisdictions for which Basel Core Principles have been assessed, observance of these principles compares well to that in other regions, although there are still significant shortcomings in prudential regulation and supervision.4
  • Credit to the private sector in several sub-Saharan African countries has continued to grow at high rates throughout the global financial crisis. Although some of the credit growth may reflect financial deepening from a very low base, it may also reflect a deterioration in the quality of banks’ loan portfolios.
  • During the last few years, Pan-African banking groups have expanded rapidly in the region (see below), and in many host countries they already manage a significant share of domestic deposits. The spread of these banking groups has increased competition in national banking systems, while also contributing to the introduction of new technologies, products, and management techniques (FSB, IMF, and World Bank, 2011). But rapid expansion of these groups may, in some cases, have outpaced supervisory capacity. Under adverse economic conditions across the region, these banking groups could become a channel for cross-border contagion.
  • Banks in sub-Saharan Africa have asset-side exposures to segments of the European financial system that have come under stress, posing a risk to banks’ foreign assets.

These findings have the following policy implications, which should be adapted according to each economy’s level of financial development and implementation capacity:

  • Country authorities need to move ahead with their plans to strengthen supervisory capacity and financial sector resilience. Close bank monitoring in countries experiencing fast credit growth is a priority.
  • The emergence of Pan-African banking groups creates the need for regional supervisory arrangements to ensure that these banks are subject to consolidated supervision. Effective mechanisms for limiting cross-border contagion—such as ring-fencing arrangements aimed at preserving subsidiaries’ resources—could be added to a review of existing banking-resolution frameworks.5
  • Given financial stresses in Europe, supervisory authorities should undertake a review of commercial bank and central bank standards and practices in regard to the placement of foreign investments and deposits.
  • Further actions are warranted to explore the impact of macroeconomic developments on systemic financial risk in Africa. International financial institutions can play a facilitating role in this regard, but enhanced dialogue between the relevant authorities in the region’s largest financial systems would improve the evaluation of systemic risks.

THE REGION’S BANKING SYSTEM EXPERIENCE IN 2008–09

Financial stress in the wake of the Lehman collapse triggered a sizable shift away from risky assets and activities. This, in turn, contributed to a sharp global slowdown.

Direct Financial Impact

Among sub-Saharan African countries, South Africa was most affected by the global financial turmoil because its equity market and deep, liquid bond market attract sizable amounts of foreign portfolio investment and carry trade. After the Lehman collapse, South Africa faced significant declines in stock prices and currency depreciation. It also experienced portfolio investment outflows, which nevertheless look modest compared with the large inflows that it received between mid-2009 and mid-2011.

In addition to South Africa, 10 other financially more advanced sub-Saharan African markets experienced portfolio capital outflows in late 2008 that eventually reverted to inflows in late 2009. Although foreign investors had paid increasing attention to these countries since the mid-2000s, the size and development of their financial markets and the cross-border flows they received remained limited (Figures 2.2 and 2.3). Data from the coordinated portfolio and direct investment surveys confirm that equity investments are a significant portion of GDP only in South Africa (Figure 2.4). In most countries, the European presence is through foreign direct investment, rather than portfolio investment. Although European investors play an important role in many of the region’s countries, South African investors themselves play an important role in some of the smaller sub-Saharan African economies (Figure 2.5).

Figure 2.2.Sub-Saharan Africa: Bond Flows, 2008–121

Source: EPFR Global database.

1 Cummulative since July 2004. Data corresponds to cummulative flows of investment in bonds issued by entities of the corresponding SSA countries by global exchange traded funds and mutual funds, expressed in U.S.dollars, asreportedto the EPFR Global database.

2 Other SSA countries for which data are available include Botswana; Congo, Dem. Rep. of; Côte d’Ivoire; Gabon; Ghana; Nigeria; and Zambia.

Figure 2.3.Sub-Saharan Africa: Equity Flows, 2008–111

Source: EPFR Global database.

1 Cummulative since November 2000. Data corresponds to cummulative daily flows of investment in stocks issued by entities of the corresponding SSA countries by global exchange traded funds and mutual funds, expressed in U.S. dollars, as reported to the EPFR Global database.

2 Other SSA countries for which data are available include Botswana, Ghana, Côte d’Ivoire, Kenya, Malawi, Mauritius, Namibia, Nigeria, Zambia, and Zimbabwe.

Figure 2.4.Sub-Saharan Africa, Select Countries: Portfolio and Foreign Direct Investment, June 20111

Sources: Coordinated Portfolio and Direct Investment Surveys.

1 Each bar represents total portfolio and direct investment liabilities of each SSA country to European, U.S., and South African investors. Figures truncated to 100 percent of GDP and data are only shown for those coutries that breach 10 percent of GDP. Those for Liberia (a shipping registry center) account for about 950 percent of GDP. That for offshore center Mauritius accounts for 317 percent of GDP, while total investment for São Tomé and Príncipe account to over 1,000 percent of GDP.

2 Includes SSA countries whose exchange rate regime falls within a wide dispersion in the degree of flexibility, ranging from de facto crawling pegs to fully floating regimes, according to the IMF’s 2011 Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER).

3 Excludes countries in the rand area, which in the chart are grouped along with South Africa for analytical purposes.

Figure 2.5.Sub-Saharan Africa, Select Countries: Inward Portfolio and Foreign Direct Investment, 20101

Source: Coordinated Portfolio and Direct Investment Surveys.

1 Each bar represents total portfolio and direct investment liabilities of each SSA country to European, U.S., and South African investors. Figures truncated to 100 percent of GDP and data are only shown for those coutries that breach 10% of GDP. Those for Liberia (a shipping registry center) account for about 950 percent of GDP. That for offshore center Mauritius accounts for 317 percent of GDP, while total investment for São Tomé and Príncipe account to over 1,000 percent of GDP.

2 Includes SSA countries whose exchange rate regime is classified as either a conventional peg or a currency board, according to the IMF’s 2011 Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER).

3 Excludes countries in the rand area, which in the chart are grouped along with South Africa for analytical purposes.

The impact of the global shift in risk aversion on asset prices varied significantly between South Africa and frontier sub-Saharan African markets.6 This effect can be measured by estimating the sensitivity of local stock prices and exchange rates to movements in the Chicago Board of Options’ Exchange Volatility Index (VIX), a proxy for global risk aversion (Figure 2.6). While the South African rand reacted more strongly to the VIX relative to most emerging market currencies, its stock prices reacted less. Interestingly, however, stock prices reacted more in South Africa than in frontier sub-Saharan African markets.

Figure 2.6.Select Countries: Exchange Rate Flexibility and Stock Market Sensitivity to Changes in Global Risk Aversion1,2

Source: Bloomberg and IMF staff estimates.

1 A positive exchange rate sensitivity to the VIX index indicates the domestic currency depreciaties when the VIX index increases.

2 Bubble size represents R squared in linear regressions on VIX (logs, daily).

Although the effect on stock prices and exchange rates was contained, the flight from risky assets did indirectly affect some of the sub-Saharan African frontier markets. For example, in Zambia the sharp reduction in foreign investors’ demand for Zambian bonds led to a credit crunch when the banking system took over as the primary source of government financing (Portillo, 2011).

Outside of the frontier markets, the global shift away from risky assets did not have a major direct impact on most sub-Saharan African countries, as they are not significantly integrated into global financial markets.7 Most sub-Saharan African countries are still financially underdeveloped; many retain significant capital controls, some of which were tightened at the margin during the crisis (Figure 2.7). The median sub-Saharan African country experienced a decline in nonresident capital inflows; with no systematic resident capital flow activity, foreign direct investment also declined, but remained positive, and foreign exchange reserves (net of the 2009 IMF SDR allocation) fell more markedly in countries with fixed exchange rate regimes.

Figure 2.7.Sub-Saharan Africa: Capital Flows and External Debt, 2005–10

Sources: IMF, World Economic Outlook database; and IMF, Statistics Department, International Financial Statistics database.

1 Excludes 2009 SDR allocation.

Resilient Banking Systems in the Region

Nigeria was the only sub-Saharan African country to experience a banking crisis in 2009 (Figure 2.8). Its origins lay primarily on deficiencies in corporate governance (masked during the boom) rather than on external financial shocks (see Box 2.1). All other sub-Saharan African banking systems generally proved resilient in the face of the global economic crisis. Specifically:

  • Banks’ foreign assets and liabilities were not materially affected over the course of the global financial crisis in most sub-Saharan African countries (Figure 2.9), although some countries (notably South Africa) experienced a drop-off in foreign funding (Appendix Tables 2.12.3).
  • In most banking systems, domestic bank deposits continued to increase in percent of GDP throughout the period.8 That said, banks in some countries needed central bank liquidity support during the crisis period, with deposits declining in relation to GDP in a few countries through June 2009.
  • On average, credit to the private sector in relation to GDP increased throughout the crisis period, albeit by less than deposits. Real credit stocks continued to increase in frontier markets and financially developing economies, but came to a halt in oil-exporting economies by early 2010, later rebounding as oil prices recovered (Figures 2.10 and 2.11). By contrast, real credit has been declining in South Africa since early 2007 by over 10 percent of GDP.9 This decline reflects significant deleveraging in the banking system, which relied heavily on domestic wholesale funding.

Figure 2.8.Sub-Saharan Africa: Systemic Banking Crises, 1980–2010

Sources: Laeven and Valencia (2008); and IMF staff estimates.

Figure 2.9.Sub-Saharan Africa: Banks’ Foreign Assets and Liabilities, 2005–10

Sources: IMF, Statistics Department, International Financial Statistics database.

Figure 2.10.Sub-Saharan Africa: Real Domestic Bank Credit, 2004–11:Q21,2

Source: IMF, Statistics Department, International Financial Statistics database.

1 Index includes subsidiaries of foreign banks.

2 Data availability for certain countries is through 2011:Q1 (Equatorial Guinea and Guinea-Bissau), 2010:Q4 (Central African Republic), or 2010:Q3 (Chad).

3 The sharp increase in frontier markets partly reflects currency depreciation in partially dollarized economies.

Figure 2.11.Sub-Saharan Africa, Select Countries: Real Credit Developments, 2005–111,2

Source: IMF, Statistics Department, International FinancialStatistics database.

1 Index at constant real exchange rates.

2 The index indicates the evolution of real credit to the private sector at constant exchange rates. The credit denominated in domestic currency has been deflated by the CPI and the credit denominated in foreign currency has been converted into domestic currency at the exchange rate of January 2005.

Box 2.1.Nigeria’s Banking Crisis

The 2009 banking crisis in Nigeria followed a period of high credit growth under weak regulation and supervision. Following a consolidation of the sector in 2005–06, and amidst large oil-related inflows and a loose monetary policy stance, loans to the private sector climbed from 18 percent of non-oil GDP to 40 percent in about two years (Figure 1). A large share of this expanding credit was used to purchase equities, in many cases with commercial banks extending the credit. Another significant share financed unhedged oil imports. When the stress generated by the global financial crisis burst the equity bubble and oil prices collapsed (Figure 2), many stock-backed and oil-related loans became nonperforming. Although official statistics showed a relatively resilient banking outlook by March 2009 (see Table 1), special audits undertaken by the Central Bank of Nigeria (CBN) revealed a different picture. Ten banks (managing about 40 percent of banking system assets) were either insolvent or substantially undercapitalized. About one-third of bank loans were nonperforming.

Figure 1.Nigeria: Loans from Commercial Banks to the Private Sector, 2006–111

Source: Country authorities.

1 October 2011.

Figure 2.Nigeria: Oil Price, Stock Exchange Index (All Share), and Nonperforming Loans, 2006–11

Sources: Country authorities; Datastream; and IMF, World Economic Outlook database.

Table 1Nigeria: Financial Soundness Indicators, 2007–11
200720082009201020111
Capital Adequacy
Regulatory capital to risk-weighted assets23.422.620.97.09.9
Regulatory Tier I capital to risk-weighted assets22.022.318.84.16.3
Capital (net worth) to assets217.018.515.23.23.9
Asset quality and composition
Nonperforming loans to total gross loans9.57.236.120.111.6
Nonperforming loans net of loan-loss provision to capital9.98.841.564.234.7
Earnings and profitability
Return on assets0.50.4-1.52.10.2
Return on equity3.01.9-9.665.44.5
Liquidity
Liquid asset to total assets24.218.817.318.023.3
Liquid assets to short-term liabilities30.224.222.119.825.7
Source: Country authorities.

June 2011.

The average ratio for the 14 non-intervened banks is considerably above the minimum regulatory threshold of 10 percent.

Source: Country authorities.

June 2011.

The average ratio for the 14 non-intervened banks is considerably above the minimum regulatory threshold of 10 percent.

The CBN took firm actions to contain the damage and considerably strengthen the system. The CBN promptly: (i) injected the equivalent of U.S.$4.2 billion (2.5 percent of 2009 GDP) into the troubled banks; (ii) guaranteed all interbank transactions, foreign credit lines, and pension deposits; (iii) replaced management in eight of the intervened banks; and (iv) committed to protect all depositors and creditors against losses. The CBN then set up the Asset Management Company of Nigeria (AMCON), which since late 2010 has replaced nonperforming loans (NPLs) with tradable zero coupon bonds, bringing five of the eight insolvent banks to zero equity. These five banks have entered merger/acquisition agreements to meet prudential requirements, while the other three (smaller) banks have been temporarily nationalized and fully recapitalized by AMCON. The equivalent of U.S. $23 billion (face value) in bonds (16 percent of 2011 non-oil GDP) has been issued to fund these operations. All interbank past-due liabilities were removed at the end of 2011.

The Nigerian banking crisis is a reminder of the importance of a strong regulatory and supervisory framework, especially in the context of rapid credit growth affected by external developments. The CBN has implemented several reforms, including stricter regulations on corporate governance and risk management; frequent on-site supervision; programs to improve the CBN’s ability to assess systemic risks; and initiatives to boost cross-agency and cross-border cooperation among regulators.

This box was prepared by Gonzalo Salinas.

Financial soundness indicators worsened moderately with the marked slowdown in global economic activity. Nonperforming loans increased and profitability declined. Most sub-Saharan African financial firms were able to withstand the associated stress, partly because they had built capital and liquidity buffers (Figure 2.12). As often happens during recessions, capital-adequacy ratios tended to improve as the slowdown in credit reduced growth in risk-weighted assets.

Figure 2.12.Sub-Saharan Africa: Financial Soundness Indicators, 2006–101

Sources: Country authorities; and IMF staff estimates.

1 The official definition of soundness indicators varies by country. Includes data for up to 33 out of the 45 countries in SSA, but each chart considers only countries for which data are available for all the years.

FINANCIAL SECTOR VULNERABILITIES TO ONGOING EUROPEAN FINANCIAL STRESS

Notwithstanding significant easing of pressures since December 2011, financial conditions in Europe remain under stress.10 Here we consider the exposure of sub-Saharan African banks to adverse developments in Europe, and examine the potential direct and indirect channels of transmission.

A first observation is that European banks have a sizable presence in sub-Saharan Africa, largely reflecting historical linkages (Figures 2.13 and 2.14). On average, more than 90 percent of all sub-Saharan African liabilities to Bank for International Settlements (BIS)-reporting banks are to European banks.

Figure 2.13.Sub-Saharan Africa: Presence of European Banks, 2011

Source: IMF, African Department database.

Figure 2.14.Sub-Saharan Africa: Liabilities to BIS-Reporting Banks, June 20111

Sources: Bank of International Settlements; and IMF, Statistics Depatment, International Financial Statistics database.

1 Amount outstanding of consolidated foreign claims of banks from corresponding BIS-reporting countries on each SSA African country on an ultimate risk basis. Includes credit provided by subsidiaries.

2 Includes SSA countries whose exchange rate regime is classified as either a conventional peg or a currency board, according to the IMF’s 2011 Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER).

3 “Other European countries” often includes French, Portuguese, and British banks where reporting countries do not disaggregate data enough. It usually refers to French banks for which disclosure is limited. Scale truncated to 40 percent. Observation for Cape Verde is 75 percent, mostly to Portuguese banks.

4 Excludes countries in the rand area, which in the chart are grouped along with South Africa for analytical purposes.

European banks conduct cross-border transactions and manage subsidiaries, branches, and representative offices in many sub-Saharan African countries. The importance of these subsidiaries is particularly large in some countries. For instance, subsidiaries of British financial institutions manage more than 30 percent of deposits in Botswana, Mauritius, South Africa, and Zambia, while affiliates of Portuguese banks manage about two-thirds of deposits in Angola. Other European banks have a much smaller presence (Appendix Table 2.4).

A second observation is that the European banks active in sub-Saharan Africa have been affected to different degrees by financial stress in Europe (Figure 2.15). Rating agencies have downgraded several banks in France, Portugal, and the U.K. since September 2011.

Figure 2.15.Credit Default Swap Spreads, 2009–12

Source: Bloomberg (updated on April 2, 2012).

Direct Linkages

Despite the strong presence of European banks, most sub-Saharan African bank credit and funding availability is unlikely to be directly affected for several reasons:

  • Residents’ cross-border liabilities to European banks are a relatively small share of GDP for most of the region’s countries, with a few exceptions (Figure 2.16).11 On an ultimate risk basis, these liabilities amounted to, on average, 5 percent of GDP in June 2011, with a median of 3 percent of GDP. Notable exceptions include Cape Verde, The Gambia, São Tomé and Príncipe, Sierra Leone, and Togo. For these countries, reduced access to cross-border credit lines could cut short funding for ongoing investment projects and other economic activity. This partly reflects prevalent capital controls (Figure 2.17), but it also underscores how these small economies with underdeveloped domestic financial markets often fund their large investment needs in specific sectors (tourism, shipping, mining) with cross-border lending from European banks.
  • Sub-Saharan African banks rely on relatively stable domestic retail funding. On average, about 95 percent of liabilities are to domestic residents, mostly in the form of deposits (Figure 2.18). Financial deepening implies that borrowers are increasingly able to borrow from local banks and fund themselves on local markets. Only in a few countries do banking groups rely on short-term wholesale funding—most notably in South Africa, where wholesale funds are from domestic contractual savings institutions with a strong home bias. But there is room for contagion even with stable domestic funding; subsidiaries of European banks may follow tighter lending behavior in the event of financial difficulties in home markets as part of risk-management guidelines for the group.12 More generally, foreign banks play a significant role in the intermediation of domestic savings, including through their subsidiaries (Figures 2.19 and 2.20).
  • Most of the region’s local banking systems are liquid. Local deposits often exceed domestic credit (i.e., loan-to-deposit ratios are well below one). Reserve requirements are high, and some countries have structural liquidity positions that result in persistent excess reserves. Sub-Saharan African banking systems are thus robust to liquidity shocks, at this stage, although standard liquidity indicators have declined slightly in the last few years.

Figure 2.16.Sub-Saharan Africa: Bank Credit to SSA Residents1

Sources: Bank of International Settlements; and IMF, African Department database.

1 Includes bank credit to central bank and other banks, nonbank private sector, and nonbank public sector. Cross border credit and credit by subsidiaries of BIS-reporting banks to SSA residents add to the amounts outstanding of consolidated foreign claims and other exposures of BIS-reporting banks on SSA African countries on an ultimate risk basis. Credit by subsidiaries and local banks add to the corresponding credit granted by the SSA-country banking system, as reported to the International Monetary Fund for the production of International Financial Statistics. Scale truncated to 100 percent of GDP. Observation for Cape Verde deposits is 115 percent of GDP. Banking system data not available for Ethiopia, Guinea, Rwanda, and Zimbabwe.

2 Includes all SSA countries whose exchange rate regime is classified as either a conventional peg or a currency board, according to the IMF’s 2011 Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER).

3 Excludes countries in the rand area, which in the chart are grouped along with South Africa for analytical purposes.

Figure 2.17.Sub-Saharan Africa: Prevalence of Capital Inflow and Outflow Controls, June 20111

Sources: IMF, Annual Report on Exchange Arrangement and Exchange Restrictions; and IMF staff estimates.

1 The IMF de jure capital control indices average binary indicators of restrictiveness in 62 categories of capital transactions. The categories include foreign exchange and domestic currency accounts of residents and nonresidents, restrictions related to the financial sector, and repatriation and surrender requirements. The index distinguishes between inflows (nonresidents’ investments in the country) and outflows (residents’ investments abroad). This broad restrictiveness index can have a value between zero and 1, and higher values represent more restricted cross-border capital flows. The indices measure the prevalence of controls, not the intensity, severity, or degree of enforcement of these controls. Countries displayed in order of higher to lower indices of outflow controls within each group.

2 Includes all SSA countries whose exchange rate regime is classified as either a conventional peg or a currency board, according to the IMF’s 2011 Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER).

3 Excludes countries in the rand area, which in the chart are grouped along with South Africa for analytical purposes.

Figure 2.18.Sub-Saharan Africa: Banking System Funding, June 20111

Source: IMF, Statistics Depatment, International Financial Statistics database.

1 Includes banking system data from disaggregated information submitted to the IMF for the preparation of monetary statistics and the compilation of summary data presented at the IFS. About half the sample is reporting under the new standardized format.

2 Includes all SSA countries whose exchange rate regime is classified as either a conventional peg or a currency board, according to the IMF’s 2011 Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER).

3 Excludes countries in the rand area, which in the chart are grouped along with South Africa for analytical purposes.

Figure 2.19.Consolidated Foreign Claims of BIS-reporting Banks on African Countries, 2011

Sources: Bank of International Settlements; and IMF staff estimates.

Figure 2.20.Sub-Saharan Africa: Broad Financial Intermediation in the Average SSA Country, June 2011

Sources: Bank of International Settlements; and IMF staff estimates.

Although direct risks to funding may be modest, there is potential vulnerability via the significant assets that sub-Saharan African residents hold in European banks (including central and commercial banks). By June 2011, BIS-reporting banks held the equivalent of 14 percent of sub-Saharan African countries’ GDP in deposits and other financial instruments, while external loans to sub-Saharan African residents averaged only about 6 percent of GDP across the countries considered (Figure 2.21). Sub-Saharan African countries are typically net creditors of BIS reporting banks. By contrast, the typical emerging market net foreign asset position with BIS reporting banks is negative.

Figure 2.21.Sub-Saharan Africa: External Loans from and Deposits at BIS-reporting Banks, June 20111

Sources: Bank of International Settlements; and IMF, African Department database.

1 Total refers to the amount outstanding of external loans and deposits of BIS-reporting banks to all sectors from the BIS locational banking statistics. Nonbanks refers to corporates and nonbank public sector. Scale truncated to 40 and 20 percent of GDP, respectively. Observation for Cape Verde deposits is 88 percent in top panel. Countries displayed in order of higher to lower percentages of deposits at BIS banks within each group, except for South Africa which is shown first.

2 Includes all SSA countries whose exchange rate regime is classified as either a conventional peg or a currency board, according to the IMF’s 2011 Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER).

3 Excludes countries in the rand area, which in the chart are grouped along with South Africa for analytical purposes.

Indirect Linkages

As was the case in 2008–09, European financial stress poses an indirect risk to sub-Saharan African banks, mostly via knock-on effects to the region’s exports and output (Chapter 1).

The econometric exercise summarized in Box 2.2 confirms that increases in global risk aversion and deteriorating credit conditions in Europe hurt African economies via their impact on global economic conditions. The impulse-responses suggest that most of the adverse effect would fall on South Africa. Frontier markets would also be affected, but to a lesser degree.

Impact on the Region’s Banking Soundness: Bank Evidence

Bank soundness indicators are also adversely affected by deterioration in the external environment (Box 2.3).

  • Global risk aversion affects sub-Saharan African banks’ liquidity and profitability.
  • As could be expected, higher export prices increase bank profitability and reduce nonperforming loans, consistent with previous research findings (Flamini, McDonald, and Schumacher, 2009; Hartelius, 2010).
  • Surprisingly, faster credit growth is associated with declining nonperforming loans (NPLs) in sub-Saharan Africa. However, this likely reflects time lags because NPLs normally deteriorate later in the credit cycle. Indeed, although not all credit booms lead to a crisis, most banking crises have been preceded by credit booms (Mendoza and Terrones, 2008). International experience suggests that careful oversight is needed even when credit is growing from a very low base and when further financial deepening is warranted.

The Region’s Capital Buffer Strength

The ability of sub-Saharan African banking systems to withstand global financial stress depends on the strength of their capital buffers. As in many emerging countries, reported capital buffers in sub-Saharan Africa tend to be larger than those prevailing in advanced economies. As shown earlier, this partly reflects the fact that bank profitability largely depends on volatile macrofinancial conditions. Volatility of bank profits may not be a concern if capital buffers are large enough to allow banks to absorb periodic losses.

Box 2.2.Global Financial Spillovers into sub-Saharan Africa—A GVAR Approach

Global vector autoregression (GVAR) modeling suggests that shocks to global risk aversion and credit conditions in Europe are systematically transmitted to sub-Saharan Africa. GVAR modeling is well suited for studying the transmission of financial spillovers (Dees and others, 2007; Pesaran and others, 2009a and 2009b; Smith and Galesi, 2011), and it provides insights into the dynamic relationships among real GDP, real private credit, exports, real short-term interest rates, and government spending within a country and across borders. The GVAR for sub-Saharan Africa was estimated using quarterly data from the first quarter of 1991 through the first quarter of 2011 (Meyer-Cirkel and Canales-Kriljenko, forthcoming). Given differences in the level of financial development across the region, sub-Saharan Africa was divided into four groups: South Africa, frontier markets, financially developing economies, and sub-Saharan African commodity exporters, which often have very particular growth dynamics (see the April 2010 Regional Economic Outlook).1

How would a sudden increase in the VIX (as a proxy for global risk aversion) similar to the one observed during the Lehman banking crisis affect sub-Saharan Africa? According to the GVAR, it will affect the region by lowering external demand, reducing commodity prices (Figure 1), and tightening worldwide financial conditions. The net result of these channels would be to contract economic activity in sub-Saharan Africa, but the effect will vary within the region. The contraction in growth will be pronounced in South Africa, mild in developing and frontier sub-Saharan African markets (Figure 2), and wide ranging in other regions of the world (Figure 3).

How would a credit crunch in Europe affect sub-Saharan Africa? Impulse responses suggest that independent shocks to credit conditions in Europe, such as those that would take place under episodes of European banking stress, would tend to contract economic activity in the main sub-Saharan African trading partners. The contraction of credit in Europe would adversely affect credit conditions in South Africa, commodity-exporting, and developing sub-Saharan African economies. Although small, the effect is non-negligible. On the other hand, credit dynamics in sub-Saharan African frontier markets appear to be unrelated to those in Europe. Growth would slow down in all sub-Saharan African countries, but especially in South Africa and emerging sub-Saharan African countries as a result of European credit retrenchment (Figure 4).2

The GVAR exercise provides some empirical evidence that global financial conditions affect economic and financial conditions in sub-Saharan Africa. It shows that the effect is stronger if financial stress in one region leads to an increase in global risk aversion. On the other hand, the exercise suggests that a shock to bank lending, with no further spillover effects onto risk aversion, would in general only have a relatively small impact on sub-Saharan Africa’s bank lending and GDP growth.

Figure 1.World: Increased Risk and Oil Price1

Source: IMF staff estimates.

1 Generalized impulse response function of oil price index to a one standard deviation positive shock to VIX.

Figure 2.Sub-Saharan Africa: Risk Aversion and Output Effects1

Source: IMF staff estimates.

1 Effects of Lehman-default size jump in volatility index in regional output over a 30-quarter horizon.

Figure 3.World: Shock in Risk Perception Reduces Global Output1

Source: IMF staff estimates.

1 Effects of Lehman-default size jump in volatility index in regional output over a 30-quarter horizon.

Figure 4.Sub-Saharan Africa: Spillovers from Credit Shocks in Europe1

Source: IMF staff estimates.

1 Generalized impulse response function of output to a negative shock to European private credit.

This box was prepared by Alexis Meyer-Cirkel.1 Data availability reduces the number of countries used to compute regional aggregates. In the sample, frontier markets include Kenya, Mauritius, and Uganda. Financially developing economies include Benin, Burkina Faso, The Gambia, Mali, Niger, Senegal, and Malawi. Commodity exporters include Cameroon, Chad, Republic of Congo, Equatorial Guinea, Gabon, and Nigeria.2 Comparisons across regions need to be interpreted with caution given differences in data quality and availability (including on banking sector development) and significant structural change in many countries in the sample.

Figure 2.22.Sub-Saharan Africa: Ease of Getting Credit (Rank) and Stregth of Legal Rights

Source: World Bank, Doing Business Indicators.

1 Countries displayed in order of higher to lower rankings of strenght of legal rights within each group.

2 Includes all SSA countries whose exchange rate regime is classified as either a conventional peg or a currency board, according to the IMF’s 2011 Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER).

3 Excludes countries in the rand area, which in the chart are grouped along with South Africa for analytical purposes.

For the time being, capital buffers are reported to be strong for the sample of 33 sub-Saharan African countries for which data are readily available. With a median value of about 19 percent (albeit with wide dispersion), in mid-2011 the capital adequacy ratio was more than twice the minimum 8 percent buffer set by the Basel Committee (Basel I). Although global financial distress very slightly increased median nonperforming loans and lowered median profitability, reported capital buffers remained high.

In reality, however, these capital buffers may not be as strong as they appear for at least three reasons:

  • As a result of supervisory capacity constraints, and shortcomings in accounting and auditing practices, nonperforming loans may not be recognized, leading to underprovisioning.
  • Credit has expanded rapidly in many countries, from both domestic and (in most recent years) foreign banks, making it difficult to detect cases of “evergreening” of bad loans. The ability of some sub-Saharan African banking systems to enforce financial contracts may be limited (Figure 2.22).
  • The aggressive expansion of Pan-African banking groups creates supervisory challenges, especially on those banks incorporated in countries with weak consolidated supervision capacity.

Channels of Pan-African Contagion

Although most sub-Saharan African banking systems are relatively insulated from external financial volatility, it is possible that difficulties in some Pan-African banks could spread throughout the region.

Over the last few years, Pan-African banking groups have become systemically important for sub-Saharan Africa. At least nine sub-Saharan African financial groups operate banks in seven or more other African countries (Figure 2.23 and Appendix Table 2.5).13 For instance, Ecobank (a Togo-based financial group) has the most widespread presence in sub-Saharan Africa (in more than 30 of the 45 countries), and manages one of the five largest banks in at least 18 countries. South African and Nigerian banking groups (most notably Standard Bank/Stanbic of South Africa and United Bank for Africa of Nigeria) have also aggressively expanded. The Bank of Africa, originally from Mali, operates in 11 sub-Saharan African countries. The financial exposures involved are not trivial. Even with partial bank-level information, these groups manage more than 30 percent of deposits in 13 of the 45 sub-Saharan African countries considered.

Figure 2.23.Sub-Saharan Africa: Selected Pan-African Banking Groups, June 2011

Source: IMF, African Department database.

The expansion of these banks has improved competition and given rise to economies of scale, because local markets are very small in some jurisdictions. However, although sub-Saharan Africa fares well compared to other regions in its observance of the Basel I Core Principles (Figure 2.24), it faces two major supervisory weaknesses (Beck and others, 2011; Lukonga, 2010):

  • Consolidated banking supervision needs significant further development. In most countries, financial soundness indicators may say little about the true financial condition of a given entity or subsidiary because transfers of risky positions across balance sheets of conglomerate entities cannot be ruled out or effectively monitored.
  • There are problems of effective coordination between home and host supervisors, although in some cases efforts are underway to strengthen such coordination via the negotiation of memoranda of understanding (MOUs) between countries.

Figure 2.24.World: Observance of Basel Core Principles, 2011

Source: Standards and Codes database.

Given these weaknesses, Pan-African banking groups may be inadequately supervised, potentially giving rise to a possible channel of cross-country contagion.

POLICIES TO ENHANCE FINANCIAL SECTOR RESILIENCY

The global economic outlook, though showing some signs of improvement, remains subject to significant downside risks; should these materialize, there will be adverse spillover effects on sub-Saharan African economies and banking systems. Policy measures are warranted to strengthen oversight of banking systems and ensure that the resiliency observed to date is maintained.

External Liquidity Management and Cross-Border Supervisory Cooperation

Given external financial stress, country authorities may need to monitor the placement of external assets to ensure that sub-Saharan African commercial and central banks effectively implement risk management guidelines aimed at safeguarding continued liquidity and availability of these assets. In addition, regulators should consider the adequacy of ring-fencing arrangements for locally operating subsidiaries of international banks.

Strengthening Effective Supervisory Capacity

Over the past decade many African countries have implemented legislative reforms aimed at strengthening banking regulation. Significant scope exists, however, for further strengthening of supervisory capacity and the actual enforcement of prudential regulations. In a number of countries, bank supervisors lack adequate tools to implement timely corrective measures, and they do not have independent authority to take remedial actions. Because of shortages of qualified resources and expertise in supervisory agencies, bank supervisors often have limited ability to assess the effectiveness of banks’ risk management processes and practices.

Country authorities need to further advance their regulatory and supervisory action plans. This requires:

  • Strengthening supervisory institutions by allocating sufficient funding and personnel, and granting supervisory authorities with adequate legal protection and corrective power in the conduct of their functions.
  • Strengthening loan classification criteria, provisioning requirements, and auditing standards to improve asset valuations and capitalization assessments. In particular, countries experiencing fast credit growth should reinforce supervision.
  • Enhancing their ability to conduct stress tests to determine the adequacy of capital and liquidity buffers.
  • Improving consolidated supervision capacity for Pan-African banking and financial conglomerates operating in several sub-Saharan African countries, including by enhancing the ability of home-country supervisors to obtain information about intra-group transactions and to monitor operations of all entities in the conglomerate group. Colleges of supervisors with an interest in banking groups should be formed from different districts. Such colleges may also require that existing restrictions on sharing banking information be addressed to enable proper communication among supervisors.
  • Promoting better coordination and informtion-sharing between home supervisors of international banks operating in sub-Saharan Africa and host country supervisors of subsidiaries. As recommended in the recent joint report to the Group of 20 (FSB, IMF, WB, 2011), supervisory colleges for large international banks should include host authorities if those banks’ local affiliates are systemically important to the host country’s financial system, even in cases where host country operations make up only a small part of the international group. In this context, it is important to ensure that timely and comprehensive information sharing with host supervisors is underpinned by appropriate legal agreements and confidentiality safeguards.
  • Promoting cross-border coordination of regulatory reforms to reduce the likelihood of arbitrage.

Institutional Frameworks for Macroprudential Policies

To internalize the effects of macroeconomic policies, account for systemic risks on banking system soundness, and better coordinate policy reactions to the changing external environment, sub-Saharan African countries would benefit from adopting institutional arrangements that allow for high-level interaction of monetary, fiscal, and financial sector authorities. Macroprudential oversight aims to proactively mitigate systemic risks, and is therefore particularly important in countries such as oil producers that can be heavily affected by external price or demand fluctuations, or in countries where the presence of systemically important Pan-African banking groups requires factoring in the risk of regional contagion. Specific institutional arrangements depend on country circumstances (Nier and others, 2011). Establishing financial stability units at central banks can be a helpful step.

Countercyclical Prudential Regulation

The cyclical behavior of financial soundness indicators in the region suggests the importance of forward-looking countercyclical prudential regulations. The experience of the region suggests, however, that these may usefully be tied to the commodity price and credit cycles rather than solely GDP growth, as is typically the case in dynamic provisioning (Saurina, 2009). For policy options to deal with credit booms, see also Dell’Ariccia and others, forthcoming).

Crisis Management and Resolution

Sub-Saharan African authorities would benefit from running crisis simulation exercises that may highlight deficiencies in country frameworks for managing banking crises and bank resolution. The legal and operational aspects of resolving banking crises—especially systemically important institutions operating across different sub-Saharan African jurisdictions—require special attention. Given the absence of explicit deposit insurance in most sub-Saharan African countries, central banks may need to take stock of their toolkits for dealing with systemic liquidity crises. Institutional frameworks for taking politically difficult actions with respect to the banking system need to be implemented and tested periodically.

Box 2.3.Determinants of Financial Soundness Indicators—Bank-level Evidence

Bank-level data from 71 sub-Saharan African commercial banks suggest that external environment and real and financial domestic activity affect the financial soundness of the region’s banks (Figure 1). The relationships are strong enough to be significant in panel regressions (Canales-Kriljenko, forthcoming). In particular:

  • An increase in the VIX (a measure of U.S. stock market volatility that proxies for global risk aversion and has been closely related to financial stress), leads after a one-year lag to higher nonperforming and impaired loan ratios, lower returns on assets and capital adequacy ratios, and tighter liquidity indicators.
  • The impact is mainly (but not exclusively) indirect, reflecting the fact that a rising VIX is associated with contracting world economic activity and trade as well as lower commodity prices. The panel regressions show that lower world growth and export prices are associated with higher nonperforming and impaired loan ratios as well as stronger liquidity positions. Lower export prices are also associated with weaker risk-weighted capital adequacy ratios.

Figure 1.Sub-Saharan Africa: Bank-level Panel of Financial Soundness Indicators, 2006–10

Source: BankScope.

Among the domestic variables, higher growth can explain higher returns on bank assets, but it does not seem to explain nonperforming loans, liquidity, or capital adequacy. The higher returns partly reflect banks’ ability to apply higher interest rate margins during a boom, which would suggest market power. Higher real credit is strongly associated with lower impaired and nonperforming loan ratios and higher liquidity indicators (Table 1). This suggests that credit conditions play a strong role in measured bank soundness indicators. It also suggests that rather than adjusting for the GDP cycle, prudential regulations should adjust for the credit cycle in assessing the underlying quality of the loan portfolio.

Table 1.Multivariate Pool Regressions on Bank-level Soundness Indicators1(Beta coefficients, unless otherwise indicated)
UnitsROAROENPLLSTLCTINTMARINTREV
VIX(t-1)Index number-0.050 ***-0.1940.027 ***-0.191 ***-0.191 ***-0.018 ***-0.056 ***
Export price growthPercent a year0.051 **-0.354 **-0.0740.115 **0.115 **-0.0090.005
Real credit growthPercent a year-0.0850.828-0.446 ***1.018 ***1.018 ***-0.032-0.029
Adjusted R-squared0.170.020.600.650.650.770.70
Number of cross sections71717171717171
Pool (balanced) observations284284284284284284284
Prob (F statistic)0.000.310.000.000.000.000.00
Durbin Watson3.333.551.942.022.021.851.62
Source: Canales-Kriljenko, Jorge Iván, forthcoming, (Washington: International Monetary Fund).Note: The acronyms used in the table are the following: ROA (return on assets), ROE (return on equity), NPL (nonperforming loans), LST (liquid assets to deposit and short-term funding), LCT (liquid assets to customer and short-term funding), INTMAR (interest rate margin), and INTREV (interest rate revenue).

The estimation method is pool least squares, using white cross-section on standard errors and covariance.

Source: Canales-Kriljenko, Jorge Iván, forthcoming, (Washington: International Monetary Fund).Note: The acronyms used in the table are the following: ROA (return on assets), ROE (return on equity), NPL (nonperforming loans), LST (liquid assets to deposit and short-term funding), LCT (liquid assets to customer and short-term funding), INTMAR (interest rate margin), and INTREV (interest rate revenue).

The estimation method is pool least squares, using white cross-section on standard errors and covariance.

This box was prepared by Jorge Iván Canales-Kriljenko.
Appendix Table 2.1.Sub-Saharan Africa: Selected Aggregate Bank Balance Sheet Accounts (June 2011)(Percent of GDP)
Foreign

Assets
Foreign

Liabilities
Net

Foreign Assets
CreditDeposits and Bonds
Private

sector
Central

Government
Public

Enterprises
Central

Bank
TotalPrivate

sector
Central

government
CBDTotal
Rand area
South Africa11.48.43.066.58.91.22.178.768.93.40.673.0
Lesotho17.90.517.312.23.30.00.716.328.90.00.028.9
Namibia10.80.99.945.55.20.60.952.355.51.50.157.1
Swaziland8.33.94.324.63.40.52.430.928.41.30.029.7
Large oil exporters
Angola5.15.20.018.59.21.18.036.947.11.50.949.5
Nigeria4.00.73.421.910.30.00.632.827.51.21.129.8
Other countries without
conventional exchange rate pegs
Botswana4.41.13.322.31.10.42.526.436.10.20.036.4
Burundi6.11.24.923.25.60.21.630.726.81.51.029.3
Congo, Dem. Rep. of7.01.15.95.40.00.31.06.722.10.70.122.8
Gambia, The5.11.43.714.015.70.04.934.660.00.00.060.0
Ghana3.01.91.110.98.22.02.924.019.51.70.221.4
Kenya3.53.50.034.014.30.72.751.742.23.00.645.8
Madagascar3.70.82.910.32.50.03.816.623.42.60.026.0
Malawi1.80.81.014.62.81.32.421.134.60.60.035.1
Mauritius225.7128.497.384.016.32.56.5109.291.60.30.192.0
Mozambique7.22.44.823.16.81.93.135.030.13.50.534.1
Seychelles42.030.611.524.314.42.46.447.657.51.30.058.8
Sierra Leone9.50.29.39.46.50.51.517.936.62.00.238.9
Tanzania4.81.43.417.37.11.44.029.828.51.90.030.4
Uganda3.11.71.316.76.40.12.025.221.80.70.422.8
Zambia5.82.63.212.17.50.13.523.219.82.30.022.1
CEMAC
Cameroon4.01.62.412.11.31.06.120.526.72.70.029.4
Central African Republic3.21.51.79.51.50.21.712.913.41.30.014.6
Chad1.60.90.75.01.30.32.18.76.82.20.19.0
Congo, Republic of4.80.54.35.40.10.26.512.218.70.40.019.1
Equatorial Guinea1.40.90.56.90.00.34.812.19.52.20.011.7
Gabon3.51.32.28.03.30.17.418.921.62.60.024.2
WAEMU
Benin12.24.37.921.76.30.04.132.242.06.93.852.6
Burkina Faso9.93.16.817.92.30.03.523.637.82.32.342.4
Côte d’Ivoire3.81.72.217.55.30.03.826.637.32.60.240.1
Guinea-Bissau7.12.74.415.30.00.03.218.522.41.10.223.6
Mali8.64.83.817.32.20.03.823.326.85.50.332.7
Niger4.42.42.012.51.20.02.516.216.12.20.618.9
Senegal5.73.81.926.33.70.05.936.043.03.31.247.5
Togo11.96.75.225.27.20.06.639.052.54.42.058.8
Other countries with conventional exchange rate pegs
Cape Verde3.75.1-1.458.214.41.09.683.168.42.61.272.2
Comoros3.61.71.817.70.30.69.428.024.80.50.025.4
Eritrea11.50.111.414.645.11.612.373.583.33.20.386.8
São Tomé and Príncipe14.25.19.134.50.21.77.744.130.41.10.432.0
Memo:1
Mean13.76.77.022.26.70.74.433.937.01.90.539.5
Median5.41.73.417.65.30.43.628.930.31.80.233.0
Min1.40.1-1.45.40.00.00.66.79.50.00.011.7
Max225.7128.497.384.045.12.512.3109.291.66.93.892.0
Sources: International Financial Statistics; and World Economic Outlook databases.

Excludes Liberia, Mauritius, and Seychelles.

Sources: International Financial Statistics; and World Economic Outlook databases.

Excludes Liberia, Mauritius, and Seychelles.

Appendix Table 2.2.Sub-Saharan Africa: Selected Aggregate Bank Balance Sheet Accounts (June 2008-June 2009)(Change in percent of GDP over the period)
Foreign AssetsForeign LiabilitiesNet Foreign AssetsCreditDeposits and Bonds
Private sectorCentral GovernmentPublic EnterprisesCentral Bankother banskTotalPrivate sectorCentral governmenCentral BankOther banksTotal
Rand area
South Africa-3.2-1.8-1.31.21.20.40.12.90.5-0.20.20.5
Lesotho0.30.00.30.3-0.3-0.40.50.21.9-1.00.00.9
Namibia12.1-0.412.52.0-0.20.40.22.421.40.3-1.220.5
Swaziland3.40.62.8-0.7-0.10.2-0.4-1.03.40.50.03.9
Large oil exporters
Angola2.43.2-0.86.23.80.25.615.818.30.3-0.118.6
Nigeria0.30.20.25.2-1.00.0-2.02.12.10.40.53.0
Other countries without
conventional exchange rate pegs
Botswana-0.10.9-1.06.1-0.60.10.66.27.00.40.17.5
Burundi-0.4-0.1-0.30.31.1-0.30.82.01.60.2-0.11.7
Congo, Dem. Rep. of-0.80.0-0.81.2-0.10.00.01.1-0.70.00.4-0.3
Gambia, The-0.80.2-1.10.5-1.20.0-0.5-1.33.40.00.03.4
Ghana1.10.20.82.30.00.60.23.12.80.10.53.4
Kenya-2.1-0.7-1.31.41.5-0.1-0.62.30.30.70.41.3
Madagascar0.10.00.10.70.80.00.62.12.70.90.03.6
Malawi-0.20.2-0.34.90.10.01.76.78.60.10.69.2
Mauritius0.21.6-1.45.90.31.10.27.58.30.0-0.18.1
Mozambique-1.40.6-2.05.10.60.30.26.33.80.00.03.8
Seychelles11.36.25.1-0.7-6.7-0.2-4.9-12.4-5.8-0.1-0.5-6.4
Sierra Leone-0.2-0.1-0.12.31.50.20.14.23.31.10.04.4
Tanzania-0.2-0.70.51.6-1.40.71.01.90.70.90.01.6
Uganda-0.20.8-1.00.8-0.40.0-0.10.40.60.0-0.50.1
Zambia0.70.30.41.1-0.30.0-0.40.40.60.10.00.6
CEMAC
Cameroon-0.50.4-0.91.70.00.10.72.51.9-0.20.11.8
Central African Republic0.0-0.10.2-1.00.3-0.20.0-0.9-0.5-0.10.0-0.7
Chad-0.8-0.3-0.60.80.10.20.92.01.30.50.01.8
Congo, Republic of0.00.1-0.12.00.0-0.12.74.75.10.10.05.2
Equatorial Guinea1.5-0.92.45.50.00.3-1.74.13.51.40.04.9
Gabon0.20.20.02.50.60.02.55.75.80.20.06.1
WAEMU
Benin-1.9-0.2-1.71.61.70.0-0.23.25.4-0.7-0.93.7
Burkina Faso1.0-0.11.10.80.50.00.01.32.2-0.20.12.1
Côte d’Ivoire0.10.10.00.3-1.10.01.20.41.1-0.30.10.9
Guinea-Bissau1.5-3.14.5-2.5-0.80.0-0.5-3.8-0.40.70.60.9
Mali0.80.60.20.70.50.00.41.71.30.0-0.21.1
Niger0.31.1-0.82.5-0.10.01.63.92.10.00.42.5
Senegal-0.7-0.2-0.42.2-0.10.00.72.84.1-0.40.44.1
Togo-0.40.2-0.6-0.75.10.01.05.45.70.11.16.9
Other countries with conventional exchange rate pegs
Cape Verde-3.00.2-3.15.2-1.4-0.50.33.5-4.70.50.0-4.1
Comoros-2.40.1-2.63.60.2-0.60.43.61.1-0.10.01.0
Eritrea-1.9-0.1-1.8-3.1-8.4-0.7-1.0-13.2-17.9-0.6-0.1-18.7
São Tomé and Príncipe-5.4-1.8-3.52.30.00.11.43.81.1-0.30.00.8
Memo:1
Mean0.30.20.11.9-0.10.00.32.12.60.10.02.8
Median-0.20.1-0.41.60.00.00.22.42.10.10.02.1
Min-5.4-3.1-3.5-3.1-8.4-0.7-4.9-13.2-17.9-1.0-1.2-18.7
Max12.16.212.56.25.11.15.615.821.41.41.120.5
Sources: International Financial Statistics; and World Economic Outlook databases.

Excludes Liberia (large shipping registry) and Mauritius and Seychelles (large offshore centers). Updated banking system information on Ethiopia, Guinea, Rwanda, and Zimbabwe was not readily available.

Sources: International Financial Statistics; and World Economic Outlook databases.

Excludes Liberia (large shipping registry) and Mauritius and Seychelles (large offshore centers). Updated banking system information on Ethiopia, Guinea, Rwanda, and Zimbabwe was not readily available.

Appendix Table 2.3.Sub-Saharan Africa: Selected Aggregate Bank Balance Sheet Accounts (June 2009-June 2011)(Change in percent of GDP over the period)
Foreign

Assets
Foreign

Liabilities
Net Foreign

Assets
CreditDeposits and Bonds
Private

sector
Central

Government
Public

Enterprises
Central

Bank
TotalPrivate

sector
Central

government
Central

Bank
Total
Rand area
South Africa-5.7-6.00.3-10.91.5-0.1-0.1-9.5-9.50.80.1-8.5
Lesotho-3.70.0-3.81.6-0.50.0-0.60.5-4.2-0.20.0-4.4
Namibia-5.3-0.5-4.71.62.0-0.40.13.3-2.50.60.0-1.9
Swaziland-0.23.1-3.42.32.00.21.05.51.6-0.50.01.1
Large oil exporters
Angola-2.00.1-2.12.9-1.70.2-1.7-0.42.5-1.20.82.1
Nigeria-0.7-0.4-0.3-12.52.90.0-1.2-10.8-4.6-0.20.4-4.4
Other countries without
conventional exchange rate pegs
Botswana-1.0-1.50.5-2.00.20.20.5-1.0-8.7-0.4-0.1-9.1
Burundi-1.6-0.4-1.26.11.4-0.4-1.15.91.90.61.03.5
Congo, Dem. Rep. of3.1-0.13.2-0.5-0.30.20.3-0.36.10.1-0.55.7
Gambia, The0.7-0.20.91.83.60.00.65.97.20.00.07.2
Ghana-0.30.0-0.2-4.43.4-0.50.2-1.2-0.60.3-0.3-0.6
Kenya-1.71.2-3.06.33.40.30.610.55.90.9-0.16.7
Madagascar0.00.1-0.1-0.70.00.00.1-0.60.0-0.20.0-0.2
Malawi0.4-0.20.62.9-1.40.8-0.61.76.20.4-0.66.0
Mauritius50.138.212.01.9-5.3-1.82.0-3.2-3.70.0-0.1-3.8
Mozambique0.00.6-0.64.5-1.60.80.23.91.10.40.51.9
Seychelles-0.6-5.14.53.9-5.9-0.11.7-0.312.4-0.7-0.411.4
Sierra Leone3.70.23.51.70.10.10.01.99.7-0.60.29.3
Tanzania0.90.50.42.01.80.0-0.13.74.4-0.20.04.2
Uganda-0.70.0-0.74.80.80.00.35.95.00.20.25.4
Zambia1.50.01.6-0.73.1-0.10.32.52.30.5-0.12.7
CEMAC
Cameroon1.30.31.01.20.20.2-0.90.72.40.3-0.12.6
Central African Republic0.90.80.13.2-0.90.00.72.93.70.30.04.0
Chad-0.1-0.90.80.40.9-0.9-0.40.0-0.10.60.00.4
Congo, Republic of1.20.11.11.0-0.2-0.4-0.40.12.3-0.20.02.0
Equatorial Guinea-1.10.3-1.40.90.00.0-0.40.5-1.4-0.80.0-2.2
Gabon0.30.10.2-3.01.00.10.8-1.2-3.00.50.0-2.5
WAEMU
Benin3.70.63.11.51.80.00.13.53.41.32.16.8
Burkina Faso4.30.53.81.20.90.01.23.210.3-0.81.911.5
Côte d’Ivoire1.9-0.12.02.82.30.00.65.710.31.1-0.311.1
Guinea-Bissau3.21.02.28.6-1.70.01.18.011.2-0.7-0.410.2
Mali3.21.12.10.50.20.00.20.91.90.50.12.5
Niger1.6-0.31.80.40.60.0-1.2-0.20.50.6-0.11.0
Senegal0.60.9-0.31.9-0.10.00.52.32.50.20.12.8
Togo4.13.11.07.7-2.00.03.69.38.41.40.410.2
Other countries with conventional exchange rate pegs
Cape Verde-0.8-1.10.30.7-0.30.5-3.6-2.7-5.7-0.71.2-5.2
Comoros2.81.21.65.0-0.40.13.58.25.60.40.06.0
Eritrea3.30.03.3-2.8-3.9-0.2-1.0-7.9-6.60.2-0.1-6.5
São Tomé and Príncipe1.8-4.56.39.30.21.0-3.86.75.50.50.46.4
Memo:1
Mean1.80.90.91.40.20.00.11.72.20.10.22.5
Median0.40.00.51.70.20.00.21.92.40.20.02.7
Min-5.7-6.0-4.7-12.5-5.9-1.8-3.8-10.8-9.5-1.2-0.6-9.1
Max50.138.212.09.33.61.03.610.512.41.42.111.5
Sources: International Financial Statistics; and World Economic Outlook databases.

Excludes Liberia (large shipping registry) and Mauritius and Seychelles (large offshore centers). Updated banking system information on Ethiopia, Guinea, Rwanda, and Zimbabwe was not readily available.

Sources: International Financial Statistics; and World Economic Outlook databases.

Excludes Liberia (large shipping registry) and Mauritius and Seychelles (large offshore centers). Updated banking system information on Ethiopia, Guinea, Rwanda, and Zimbabwe was not readily available.

Appendix Table 2.4.Sub-Saharan Africa: Selected Foreign Banking Groups Operating in Africa, 2011(Share of deposits, 2010, where BankScope data available)
Foreign Banks Outside Africa
UKFrenchPortuguese
Barclays GroupABSA (South Africa)1Standard CharteredHSBCSociete GeneraleBPCECredit AgricoleBNP ParibasBanco Espirito SantoCaixa Geral de Deposits (nationalized)Banco Comercial PortuguesBanco BPIOther PortugueseBanco Africano de lnvestimentoBanco Formento de AngolaBanco Privado AtlanticoCaixa Economica de Cabo VerdeBank of Baroda (lndia)Deutsche BankUBSCiti Grouplnternational Commercial BankProcredit Bank (Germany)Orabank
Rand area
South AfricaX18XXXXXXXX
Lesotho
Namibia7
Swaziland
Large oil exporters
Angola1025XX25185
NigeriaX2
Other countries without conventional exchange rate pegs
Botswana3123X
Burundi
Congo, Dem. Rep. of37
Ethiopia
Gambia, The20
Ghana9940
Guinea25236
Kenya10823
Liberia
MadagascarXXX
Malawi
Mauritius12424X34
Mozambique72834X
Rwanda
Seychelles469
Sierra Leone18
Tanzania411815
Uganda81144
Zambia2020XX
CEMAC
Cameroon71620123
Central African Republic
Chad
Congo, Republic of8
Equatorial Guinea24
Gabon8
WAEMU
Benin106
Burkina Faso1210
Côte d’Ivoire21893
Guinea-Bissau
Mali3
Niger
Senegal2211XX
Togo
Other countries with conventional exchange rate pegs
Cape Verde0501030
Comoros
Eritrea
São Tomé and Príncipe
Zimbabwe10
Total number of subsidiaries or branches941321043534213111182111112
Number in percent of 44 SSA countries209305239711795272222185225225
Sources: Annual reports; BankScope; Bankers Almanac; bank websites; country authorities; and IMF staff.Note: An “X” represents bank presence for which specific data is not readily available.

Absa is a subsidiary of Barclays.

Sources: Annual reports; BankScope; Bankers Almanac; bank websites; country authorities; and IMF staff.Note: An “X” represents bank presence for which specific data is not readily available.

Absa is a subsidiary of Barclays.

Appendix Table 2.5.Selected Pan-African Banking Groups, 2011(Share of deposits, 2010, where BankScope data available)
Pan African banks
EcobankOceanic (absorbed by Ecobank)United Bank for AfricaBank of AfricaAccess BankAfrilandBGFI bankBanque AtlantiqueCommercial BankABCSociete Ivoirienne de BanqueBanco EquadorIsland BankCompagnie bancaire de l’Afrique occidentaleBanque Sahélo-Saharienne pour l’Investissement et le CommerceSouth African
Standard Bank Group/StanbicABSA1Nedbank2First Rand
Rand area
South AfricaX27182113
LesothoX30X
NamibiaX7XX
Swaziland432321
Large oil exporters
AngolaX
Nigeria116125X
Other countries without conventional exchange rate pegs
Botswana14X16X
Burundi8X
Congo, Dem. Rep. of3003115
Ethiopia
Gambia, The6X7
Ghana9216
Guinea25X
Kenya10212
Liberia50XX
MadagascarX
MalawiX20X
Mauritius7
Mozambique032073
Rwanda774
Seychelles
Sierra Leone1223
Tanzania00202X5110
Uganda11316
Zambia1XXXX15X
CEMAC
Cameroon94218X25
Central African Republic55227
Chad27X13
Congo, Republic of7X51
Equatorial Guinea4513X
GabonXX46
WAEMU
Benin1830X6
Burkina Faso16X18X
Côte d’Ivoire12151118
Guinea-Bissau10
Mali14125
Niger161815
Senegal12X4XXX
Togo25X
Other countries with conventional exchange rate pegs
Cape Verde0
Comoros
Eritrea
São Tomé and Príncipe325294
ZimbabweX1114
Total number of subsidiaries or branches31417129768761111217458
Sources: Annual Reports, BankScope, Bankers Almanac, and bank websites.Note: An “X” represents bank presence for which specific data is not readily available.

Absa is a subsidiary of Barclays.

Nedbank has a cooperation agreement with Ecobank. It is a subsidiary of a British financial entity, Old Mutual.

Sources: Annual Reports, BankScope, Bankers Almanac, and bank websites.Note: An “X” represents bank presence for which specific data is not readily available.

Absa is a subsidiary of Barclays.

Nedbank has a cooperation agreement with Ecobank. It is a subsidiary of a British financial entity, Old Mutual.

Appendix Table 2.6.Sub-Saharan Africa: Capital Adequacy Ratios, 2006–11(Percent of risk-weighted assets)
200620072008200920102011
Rand area
South Africa12.312.813.014.114.914.9
Lesotho19.114.113.713.815.315.3
Namibia14.215.815.515.015.314.5
Swaziland26.323.633.828.419.818.9
Large oil exporters
Angola18.521.919.519.518.614.8
Nigeria40.923.422.620.97.0
Other countries without
conventional exchange rate pegs
Botswana16.818.516.919.620.5
Burundi13.513.515.219.119.7
Congo, Dem. Rep. of10.512.810.914.0
Ethiopia11.320.015.918.317.5
Gambia, The22.519.618.826.025.0
Ghana15.814.813.818.219.117.4
Guinea12.415.915.918.521.417.7
Kenya17.018.018.919.520.819.4
Liberia12.328.422.028.127.422.8
Madagascar
Malawi21.819.621.522.620.320.7
Mauritius15.813.315.315.415.815.8
Mozambique12.514.213.915.114.417.1
Rwanda13.716.615.919.024.427.2
Seychelles19.615.412.021.621.524.2
Sierra Leone33.335.043.534.030.727.0
Tanzania16.316.217.018.318.517.8
Uganda17.919.320.720.920.220.3
Zambia20.418.618.622.322.119.2
CEMAC
Cameroon20.117.718.617.117.915.8
Central African Republic22.725.526.636.530.944.2
Chad20.122.319.323.726.625.8
Congo, Republic of25.429.726.529.721.016.9
Equatorial Guinea19.519.013.731.526.624.2
Gabon17.814.319.424.022.616.4
WAEMU
Benin1.43.010.9
Burkina Faso8.48.37.46.3
Côte d’Ivoire12.49.59.3
Guinea-Bissau18.214.911.717.8
Mali9.210.811.49.916.917.4
Niger10.911.04.99.16.5
Senegal13.113.613.916.518.217.2
Togo2.8-13.713.512.314.4
Other countries with conventional exchange rate pegs
Cape Verde11.011.412.011.415.613.6
Comoros
Eritrea14.016.014.019.0
São Tomé and Príncipe19.816.724.723.227.9
Zimbabwe26.519.243.521.615.3
Source: Country authorities.
Source: Country authorities.
Appendix Table 2.7.Sub-Saharan Africa: Nonperforming Loans, 2006–11(Percent of loans)
200620072008200920102011
Rand area
South Africa1.11.43.95.95.84.7
Lesotho2.11.42.13.23.12.4
Namibia2.62.83.12.72.01.9
Swaziland7.77.57.68.69.37.5
Large oil exporters
Angola4.82.92.42.68.62.4
Nigeria10.09.57.236.120.1
Other countries without
conventional exchange rate pegs
Botswana1.51.31.63.26.1
Burundi1.41.61.45.44.3
Congo, Dem. Rep. of
Ethiopia
Gambia, The10.413.712.314.512.9
Ghana7.96.47.716.217.614.1
Guinea10.510.94.59.55.52.8
Kenya20.210.69.07.96.24.4
Liberia42.416.417.410.925.120.8
Madagascar9.18.911.312.413.9
Malawi5.03.43.23.23.93.9
Mauritius3.02.52.03.32.82.6
Mozambique3.12.61.91.81.92.6
Rwanda25.018.112.613.110.88.0
Seychelles4.42.32.03.85.58.1
Sierra Leone26.925.617.910.615.615.1
Tanzania6.86.36.26.69.86.8
Uganda2.94.12.24.22.12.2
Zambia11.38.87.212.614.810.4
CEMAC
Cameroon12.312.511.512.714.814.8
Central African Republic32.930.421.321.313.914.7
Chad12.411.38.010.412.19.8
Congo, Republic of1.42.71.61.52.11.5
Equatorial Guinea14.311.39.911.87.25.5
Gabon10.77.68.57.29.95.5
WAEMU
Benin19.417.118.716.615.5
Burkina Faso9.4
Côte d’Ivoire
Guinea-Bissau2.38.25.07.23.5
Mali25.025.125.321.921.518.5
Niger21.816.716.014.78.9
Senegal16.818.617.418.720.2
Togo29.022.314.914.6
Other countries with conventional exchange rate pegs
Cape Verde3.95.03.14.14.26.0
Comoros
Eritrea
São Tomé and Príncipe44.326.619.310.29.7
Zimbabwe4.25.36.51.83.1
Source: Country authorities.
Source: Country authorities.
Appendix Table 2.8.Sub-Saharan Africa: Return on Assets, 2006–11(Percent a year)
200620072008200920102011
Rand area
South Africa11.41.41.51.11.21.5
Lesotho0.31.72.72.82.72.8
Namibia1.53.54.23.03.52.9
Swaziland2.91.94.02.52.42.4
Large oil exporters
Angola2.72.72.43.43.02.7
Nigeria0.30.50.4-1.52.1
Other countries without
conventional exchange rate pegs
Botswana6.12.62.92.80.9
Burundi1.72.32.32.62.5
Congo, Dem. Rep. of
Ethiopia2.92.73.23.43.4
Gambia, The4.50.1-1.6-0.50.3
Ghana4.83.73.22.83.83.9
Guinea5.34.13.22.01.92.5
Kenya2.73.02.82.93.73.3
Liberia-0.50.60.70.60.5-0.7
Madagascar3.32.91.65.36.5
Malawi4.24.94.84.74.23.2
Mauritius1.71.91.71.61.41.6
Mozambique4.03.83.53.02.62.5
Rwanda2.41.52.40.72.02.2
Seychelles2.74.24.93.53.73.7
Sierra Leone5.83.12.21.63.43.8
Tanzania3.94.73.83.22.22.7
Uganda3.33.44.02.03.04.0
Zambia5.14.73.62.12.93.7
CEMAC
Cameroon1.31.11.1-0.30.40.2
Central African Republic3.83.92.31.62.92.7
Chad1.81.71.90.11.60.4
Congo, Republic of2.12.92.51.92.30.9
Equatorial Guinea1.21.01.71.31.50.6
Gabon2.52.71.82.80.50.8
WAEMU
Benin0.7
Burkina Faso
Côte d’Ivoire
Guinea-Bissau
Mali13.5-0.80.80.61.41.4
Niger1.7
Senegal1.61.61.41.3
Togo
Other countries with conventional exchange rate pegs
Cape Verde0.81.21.30.70.70.3
Comoros
Eritrea3.03.02.0
São Tomé and Príncipe-2.8-0.9-1.00.10.7
Zimbabwe16.710.335.60.51.9
Source: Country authorities.

As in the IMF’s Global Financial Stability Report,the figures correspond to returns before taxes, as opposed to those published in South African Reserve Bank’s financial stability report.

Source: Country authorities.

As in the IMF’s Global Financial Stability Report,the figures correspond to returns before taxes, as opposed to those published in South African Reserve Bank’s financial stability report.

Appendix Table 2.9.Sub-Saharan Africa: Return on Equity, 2006–11(Percent a year)
200620072008200920102011
Rand area
South Africa118.318.126.918.018.321.0
Lesotho4.521.036.233.330.928.9
Namibia19.944.952.138.441.933.9
Swaziland21.214.822.714.914.413.8
Large oil exporters
Angola28.823.626.536.532.127.2
Nigeria1.93.01.9-9.665.4
Other countries without
conventional exchange rate pegs
Botswana66.943.245.456.59.1
Burundi17.526.429.722.821.8
Congo, Dem. Rep. of
Ethiopia
Gambia, The36.40.8-24.94.81.8
Ghana39.635.830.123.628.627.2
Guinea73.244.839.624.525.432.5
Kenya27.228.325.224.830.732.2
Liberia-4.14.55.43.40.3-4.7
Madagascar0.40.40.20.50.5
Malawi36.039.935.134.229.424.2
Mauritius22.426.424.321.020.022.1
Mozambique60.850.744.736.632.526.5
Rwanda27.015.518.55.011.210.6
Seychelles41.864.268.438.840.141.0
Sierra Leone17.010.37.24.012.115.6
Tanzania26.729.023.218.412.115.1
Uganda27.327.827.712.721.327.4
Zambia30.635.120.89.412.125.5
CEMAC
Cameroon13.812.813.0-2.95.22.2
Central African Republic20.622.814.08.916.014.7
Chad15.515.417.31.012.73.1
Congo, Republic of35.340.832.523.328.014.2
Equatorial Guinea17.713.923.918.514.95.4
Gabon23.532.320.817.25.18.6
WAEMU
Benin-4.81.012.56.08.2
Burkina Faso
Côte d’Ivoire
Guinea-Bissau14.4-12.0-3.6-2.56.2
Mali8.7-14.012.910.015.215.2
Niger3.84.60.223.414.5
Senegal14.514.611.618.6
Togo4.88.512.75.8
Other countries with conventional exchange rate pegs
Cape Verde15.621.721.39.68.23.7
Comoros
Eritrea
São Tomé and Príncipe-2.80.9-3.71.8
Zimbabwe66.341.155.41.911.0
Source: Country authorities.

As in the IMF’s Global Financial Stability Report,the figures correspond to returns before taxes, as opposed to those published in South African Reserve Bank’s financial stability report.

Source: Country authorities.

As in the IMF’s Global Financial Stability Report,the figures correspond to returns before taxes, as opposed to those published in South African Reserve Bank’s financial stability report.

This chapter was prepared by Jorge Iván Canales-Kriljenko, Mauro Mecagni, and Alexis Meyer-Cirkel; with research assistance from Sandra Donnally, Samuel Fahlberg, Emily Forrest, Cleary Haines, Brian Moon, and Luiz Edgard R. Oliveira; editorial assistance from Jenny Kletzin DiBiase; and administrative assistance from Natasha Minges and Anne O’Donoghue.
1Banks and other deposit-taking institutions dominate the financial systems in most of sub-Saharan Africa, with a few exceptions that include countries in southern Africa and offshore financial centers like Mauritius and Seychelles. Data availability prevents examination of the evolution of microfinance institutions, which play a useful and expanding financial role, and of Islamic banks.
2Banking systems in the CFA franc zone have also been resilient to political stress in Côte d’Ivoire during 2010.
3Direct linkages refer to relations between countries’ financial systems, while indirect linkages refer to macroeconomic channels of contagion.
4Financial assessments by the World Bank and IMF under the FSAP provide a rich source of information on banking systems and the quality of bank supervision and oversight in the countries assessed.
5Ring-fencing arrangements could also be of value in preventing the spillover of problems of non-African parent banks into their African subsidiaries.
6Frontier markets include Angola, Ghana, Kenya, Mauritius, Mozambique, Nigeria, Senegal, Tanzania, Uganda, Zambia, and Zimbabwe (see IMF, 2011).
7For an early assessment, see IMF 2009 (Chapter II).
8In particular, sub-Saharan African bank deposits increased, on average, by 3 percentage points of GDP between June 2008 and June 2009, and by another 3 percentage points between June 2009 and June 2011.
9This cannot be entirely attributed to the crisis; the National Credit Act, starting in mid-2007, was also a factor. This legislation imposed stricter criteria for granting bank loans, especially for mortgage financing.
10See the April 2012 Global Financial Statistical Report (IMF 2012).
11Although aggregate banking system data could mask some large exposures at the institutional level, it is unlikely that these are systemically important.
12This risk may be lower in cases where the sub-Saharan African subsidiaries contribute a significant share of the global bank’s profit and growth prospects.
13Some of these regional banks have centralized treasury management, credit decisions, and operational support (Lukonga, 2010).

    Other Resources Citing This Publication