Information about Sub-Saharan Africa África subsahariana

2. Capital Inflows to Frontier Markets in Sub-Saharan Africa

International Monetary Fund. African Dept.
Published Date:
May 2011
  • ShareShare
Information about Sub-Saharan Africa África subsahariana
Show Summary Details

Introduction and Summary

Before the global financial crisis, there was substantial investor interest in sub-Saharan Africa’s (SSA) frontier markets (FMs),1 reflecting their strong economic performance and a global economy awash with liquidity.

These countries received considerable volumes of capital inflows, mirroring the steep rise in private capital flows to other emerging and developing countries in the middle of the past decade. Although such flows briefly reversed during the apex of the crisis, very low interest rates in advanced countries and an attenuation of global risk aversion have once again prompted investors to scour the globe in search of attractive investment opportunities. Post-crisis, anecdotal evidence—including heightened attention of investment banks—suggests that interest in FMs has also resumed. Against this backdrop, this chapter addresses the following questions:

  • To what extent has the resurgence of global capital flows translated into a resumption of private capital inflows, especially portfolio inflows, to sub-Saharan Africa FMs?
  • Do global push factors or local pull factors dominate in steering investor interest?
  • Why have some sub-Saharan African FM countries garnered investor interest, while others—including some larger countries—have been sidestepped?
  • Which policy options are most suitable for sub-Saharan Africa FM countries to use at this juncture to address any resumption of large capital inflows?

Understanding the determinants of and scope for private capital inflows is important for sub-Saharan African countries for a number of reasons. For one, such flows are increasingly the main source of external financing for many countries in the region. The deterioration in the fiscal accounts of most advanced countries because of the crisis also implies that the prospects for sustaining even current levels of official financing are doubtful. Second, private flows tend to be more volatile. At times flows are large relative to the size of the economy and monetary base, complicating macroeconomic management.

The main findings are the following:

  • The overall trend of capital flows to sub-Saharan Africa’s FMs mirrors trends elsewhere, with strong inflows before the global crisis and a sharp decline during the crisis.
  • Postcrisis, and in 2010 in particular, more differentiation is evident. Private investors, possibly still smarting from the global financial losses of recent years, seem to be distinguishing between markets. Thus, country-specific pull factors govern the pattern of flows across regions and countries. In a few of the region’s FMs (Ghana, Mauritius, and, to a somewhat lesser degree, Zambia) portfolio flows picked up markedly in 2010. But in others, there is little sign of resumption in inflows.
  • For fixed-income investments, market participants identify yields as the key driver of inflows. Thus, the monetary policy easing that was undertaken by many of the FMs may have reduced the incentives for inflows. Exchange rate volatility is another factor that seems to have played a role in some countries. For equity portfolio flows and foreign direct investment (FDI), a range of other factors contribute to the expected return and riskiness of the investment and influence whether inflows have resumed.
  • The region’s FMs outperform other groups—including FMs in other regions, other sub-Saharan African countries, and even select emerging market countries—on a number of indicators of institutional quality, growth prospects, and macroeconomic outcomes. As the contribution of official financing continues to diminish, improvements in many of these areas could help other sub-Saharan African countries to attract private sources of financing for investment and growth.
  • Two of the region’s eleven FMs have opted for capital controls in response to the volatility of portfolio inflows, but most countries have continued to rely on macroeconomic policies and macroprudential measures to respond to pressures from current and prospective inflows.

Box 2.1.What Are Frontier Markets?

The term “frontier markets” (FMs) is commonly used to describe a subset of emerging markets (EMs) that have small financial sectors and/or have low annual turnover and liquidity, but nonetheless demonstrate a relative openness to and accessibility for foreign investors. They are generally in the early stages of financial market development. In most cases the existence of market restrictions make them unsuitable for inclusion in the larger EM indices, such as the Morgan Stanley Capital International (MSCI) Emerging Market Index.

The main attraction of frontier equity markets for investors is that they may offer high, long-term returns and low correlations with other markets. At the same time, frontier market short- and medium-term securities typically have higher yields than in more developed emerging countries. With a few exceptions, because of low liquidity and turnover, the main investors in FMs are typically dedicated funds and hedge (or “boutique”) funds.

There is no generally accepted list of FMs, and different investment banks have their own coverage. For instance, the MSCI Barra has 26 countries classified as FMs, while other indices include 4 to 5 more countries. Among sub-Saharan African countries, Kenya, Mauritius, and Nigeria are typically included in almost all indices.

In this chapter, a looser definition of FMs for sub-Saharan Africa is adopted. Criteria used to select countries include recent growth dynamics and perspectives, financial market development, general institutional conditions and evolution, natural resource richness, and political conditions and perspectives. Although some of the countries are not included in investment bank indices, there has been sufficient outside investor interest over the past five years to warrant their consideration here. The list includes Angola, Ghana, Kenya, Mauritius, Mozambique, Nigeria, Senegal, Tanzania, Uganda, Zambia, and Zimbabwe.

Members of the benchmark groups used as comparators in this chapter have been chosen based on a similar set of criteria, to promote comparability. The group of other FMs consists of Bangladesh, Bulgaria, Jordan, Kazakhstan, Pakistan, Romania, Sri Lanka, Tunisia, and Vietnam. The select group of emerging markets consists of Colombia, Egypt, Malaysia, Peru, South Africa, Thailand, and Uruguay.

This box was prepared by Montfort Mlachila.

The Nature and Volume of Flows to Sub-Saharan African Frontier Markets in a Global Context

Trends in Overall Capital Flows

Flows to sub-Saharan African FMs should be seen against the backdrop of developments in the region more generally. Thus, this section first discusses the trends in overall capital flows to sub-Saharan African countries in recent years. The section then turns to capital flows to the FMs, focusing especially on developments in private portfolio inflows to debt and equity securities.

During the last two decades, external sources of funding for investment and growth in sub-Saharan Africa have undergone a noteworthy transformation (Figure 2.1). First, a sixfold increase has occurred in total flows, especially since 2000. Second, in sharp departure from the previous decade, most of the increase has come from the private sector, even when excluding South Africa and Nigeria.2 Inflows from private capital in the form of both FDI and portfolio flows have increased rapidly, although not all countries in sub-Saharan Africa have participated equally in this transformation, particularly in the ability to attract portfolio inflows. The same trend has occurred in transfers, whereby remittances have overtaken official transfers (grants) that have been declining in importance during the past decade. Total net private inflows amounted to about $41 billion dollars in 2010, with South Africa accounting for more than 40 percent of the total.

Figure 2.1.Private vs. Official Financing to Sub-Saharan African Countries1

Source: IMF, World Economic Outlook database.

1Private versus official refers to destination of flows.

The increasing reliance on private external financing poses challenges for macroeconomic management, given the relative size and volatility of these flows. First, although net private capital inflows3 to sub-Saharan Africa constitute only about one tenth of total net private flows to emerging and developing countries, the inflows are large relative to the economic size of the recipient countries (Figure 2.2). Indeed, in recent years, net private capital flows to sub-Saharan Africa have been larger as a percent of GDP than in other developing and emerging countries. Second, the more volatile flows have the potential to cause considerable difficulties in monetary management in sub-Saharan Africa, given limited monetization and shallow financial markets. For instance, in proportion to reserve money, nonresident holdings of government securities have swung from almost nothing to more than 40 percent in some countries. Finally, although FDI flows have been a stable component of private flows to sub-Saharan Africa (Figure 2.3), net portfolio inflows have been volatile, with significant net outflows occurring during 2008 (Figure 2.4). In the last two years, net portfolio inflows to sub-Saharan Africa have been on the rise again, constituting about half of net private inflows in 2010.

Figure 2.2.Net Private Capital Flows to Emerging and Developing Economies

Source: IMF, World Economic Outlook database.

1Private refers to source of flows.

Figure 2.3.Net Private Capital Flows to Sub-Saharan African Countries

Source: IMF, World Economic Outlook database.

1Private refers to source of flows.

Figure 2.4.Net Portfolio Investment in Emerging and Developing Economies

Source: IMF, World Economic Outlook database.

Portfolio Flows to Frontier Markets

Net portfolio investment flows in sub-Saharan Africa FM countries amounted to about ½ billion U.S. dollars in 2010 (Figure 2.5). The main types of portfolio flows are holdings of government securities and private equities, with the latter significantly more important than the former. Portfolio investment has been stable in FM countries relative to emerging markets (Figure 2.6).4 During 2007, all groups received portfolio inflows, with emerging market economies (EMEs) attracting the highest amount as a percent of GDP. During the global crisis, flows to EMEs contracted by 5 percent of GDP, and subsequently rebounded by the same amount through 2010. The modest inflows of portfolio investment to sub-Saharan Africa FMs during 2010 mirrors developments in other FMs, where inflows remain far below their precrisis ratios to GDP.

Figure 2.5.Sub-Saharan Africa Frontier Markets: Portfolio Investments (Net)

Source: IMF, World Economic Outlook database.

Figure 2.6.Portfolio Investment Net

Source: IMF, World Economic Outlook database.

After a sharp decline in the value of portfolio holdings by foreigners in 2008, there has been a recovery in 2010 in most sub-Saharan Africa FMs. Based on information from the IMF Coordinated Portfolio Investment Survey,5 average foreign holdings (excluding Mauritius and Nigeria) were about US$450 million in 2010, already higher than the previous peak in 2007 (Figure 2.7). The value of portfolio holdings in Mauritius was hardly affected by the global crisis, while Nigeria suffered badly during 2008–09 (Figure 2.8).

Figure 2.7.Average Stock of Portfolio Investment Liabilities1

Source: IMF, Coordinated Portfolio Investment Survey (CPIS).

1Excludes Liberia, Mauritius, Nigeria, and South Africa.

Figure 2.8.Stock of Portfolio Investment Liabilities

Source: IMF, Coordinated Portfolio Investment Survey (CPIS).

Holdings of Government Securities

Private capital flows into government coffers through two main channels. Treasury bills are quantitatively the most important source of financing. Most of the inflows have been into short-term (91-day) treasury bills, although increasingly governments are issuing long-dated bonds in domestic currencies—with Kenya and Mauritius taking the lead.6 This is the most traded segment of the market. Because of the limited existence of secondary markets, foreign investors, often wary of rollover and exchange rate risk, generally prefer this segment.

There have been relatively few international sovereign bond issuances by sub-Saharan Africa FMs over the past decade, especially when compared to other FMs (Figures 2.9 and 2.10). Most of the sub-Saharan Africa issuances took place in 2007. Among the FMs, only Ghana, Nigeria, and Senegal have ever issued sovereign bonds.7 In 2007 Ghana issued two bonds for US$950 million and Nigeria issued one for US$525 million. Senegal had its maiden issuance of US$200 million in 2009. More recently, Nigeria issued another bond for US$500 million in the first quarter of 2011. Outside of FMs and South Africa, the only sub-Saharan African countries that have issued sovereign bonds are Gabon and Seychelles.

Figure 2.9.Sub-Saharan Africa Frontier Markets: External Bond Issuance

Source: Dealogic.

Figure 2.10.Other Frontier Markets: External Bond Issuance

Source: Dealogic.


Equity investment has been the largest component of total portfolio investment. Equity prices have recovered more slowly in FMs than in emerging markets, with prices of sub-Saharan Africa FM stocks moving closely in line with those of FMs in other regions (Figure 2.11). Various country-specific factors have held back prices in specific sub-Saharan African FMs (Figures 2.12). In both Kenya and Nigeria, political factors may have been at play. In Kenya, prices were weighed down by uncertainties before the August 2010 constitutional referendum, but since then foreign investors more than doubled their holdings from end-2009. In Nigeria, the combination of the looming election, governance problems at the stock exchange, and a domestic banking crisis seem to have been important factors. On the other hand, after 10 years of economic decline in Zimbabwe, a government of national unity, which was formed in 2009, started to address the economic crisis through adoption of a multicurrency system (for example, full official dollarization). As a result, portfolio flows have resumed, and market capitalization rapidly increased from US$1.5 billion in February 2009 to US$5 billion in October 2010. Indeed, in almost all sub-Saharan Africa FMs (even in Nigeria), estimates for 2009 show a sharp recovery in the value of foreign equity holdings (Figure 2.13).

Figure 2.11.Morgan Stanley Capital International (MSCI) Indices

Source: Bloomberg.

Figure 2.12.Country Stock Market Indices

Source: Bloomberg.

1For Kenya index 2005:Q1 = 100, for Tanzania index 2006:Q4 = 100, for Nigeria index 2007:Q1 = 100.

Figure 2.13.Sub-Saharan Africa Frontier Markets: Foreign-Held Equity Securities

Source: IMF, Coordinated Portfolio Investment Survey (CPIS).

Other Private Flows to Frontier Markets

A steady recovery in external credit lines has occurred for sub-Saharan African FMs. Data available through September 2010 show that after peaking at about US$50 billion in September 2010, foreign bank claims on sub-Saharan Africa FMs (Figure 2.14) have steadily regained vigor. Indeed, the total for all countries excluding Nigeria is now above the 2008 peak. The recovery is particularly remarkable for Angola, Ghana, Mauritius, and Uganda, while Mozambique and Tanzania were hardly scathed. Nigeria and Zimbabwe are the only countries that have yet to register significant recoveries. In Nigeria, for instance, after peaking at about US$12 billion in 2008, credit lines have persistently remained at about US$5–6 billion.

Figure 2.14.Sub-Saharan Africa Frontier Markets: Consolidated Foreign Bank Claims

Source: Bank of International Settlements.

Although FDI (Figure 2.15) has been an important and rapidly growing source of finance to sub-Saharan African FMs, it still trails the other benchmark groups over the past five years. Moreover, Angola has experienced unusually large foreign direct investment into its oil extraction industry (for example, FDI peaked at 40 percent of GDP in 2000). Excluding Angola, FDI has averaged only 2.6 percent of GDP in the sub-Saharan African FMs compared with 4 percent of GDP in other FMs during 1991–2009 (Figure 2.16). Given the strong output growth and high returns on FDI (Box 2.2), volumes of inflows are low, perhaps because of other structural factors.8

Figure 2.15.Average FDI as Percent of GDP, 1991–2009

Sources: IMF, World Economic Outlook database; and World Bank, Global Development Finance.

Figure 2.16.FDI as Percent of GDP

Sources: IMF, World Economic Outlook database; and World Bank, Global Development Finance.

How Different Is the Recent Experience of Sub-Saharan Africa’s Frontier Markets?

This section synthesizes information from a survey of country officials, market participants, and desk officers on factors influencing capital flows to the region’s 11 FMs in 2010. It then considers differences between these countries and other comparator groups on the key explanatory factors mentioned in the survey findings.

Recent Capital Flows—Contrasting Narratives

The postcrisis pattern of capital inflows to sub-Saharan African FMs has been diverse. While a few FMs have recaptured investor interest, generating a renewed surge in inflows, others have been bypassed. Country case studies suggest that country-specific factors explain the varied experiences. Table 2.1 summarizes the results, and details are presented in Table 2.2.

Table 2.1.Developments in Private Capital Flows during 20101
Strong RecoveryModerate

Little ChangeDecline
FDIGhana, MauritiusKenya,


PortfolioGhana, MauritiusKenya,


OtherGhana, Mauritius,


Source: IMF African Department staff survey.

There are insufficient quarterly data for Angola, Mozambique, and Senegal.

Source: IMF African Department staff survey.

There are insufficient quarterly data for Angola, Mozambique, and Senegal.

Table 2.2.Factors in Attracting Private Capital Inflows
Commodity developmentsCross-border interestsMacroeconomic factorsPolitical/institutionalBanking soundness/regulations
AngolaHigh oil prices
GhanaInvestments in offshore oil fields and the anticipation of oil production to begin in 2011. High prices for gold and cocoaMultilateral debt relief and approval of an IMF supported programInstitutional reforms, such as those affecting capital and financial markets, have made it easier to investForeign bank inflows in 2009-10 to meet increased minimum statutory capital requirements for domestic subsidiary banks
KenyaSignificant FDI flows to its EAC partner countries, with investments in several sectorsIn the first half of 2010, “wait-and-see” attitude in the run-up to referendum on the new constitution
MauritiusImportant regional financial center (with strong regulatory framework) for many cross-border investments bound for Africa and Asia (especially into India due to a highly advantageous double-taxation agreement)Wide-ranging structural reforms to diversify the economy, and efficient administration and market-friendly regulations support its reputation as a financial safe-haven in the region
MozambiqueFDI to exploit its largely untapped resource baseSustained improvements in macroeconomic fundamentals and policies
NigeriaHigh oil pricesMacroeconomic imbalances, including high inflation, a high fiscal deficit, the depletion of oil savings, and decline in foreign reservesCorruption scandals related to management of the stock exchange. A long delay in finalizing a bill to overhaul oil and gas legislation in several areas relevant to foreign companiesBanking crisis
SenegalWAEMU banks from outside Senegal seeking yield and diversification
UgandaFirst significant oil discoveriesSustained improvements in macroeconomic fundamentals and policies
ZambiaHigh copper prices
ZimbabweHigh gold and platinum prices High gold and platinum pricesRestored political and economic stability from early 2009, reducing investor concerns and reviving the inflow of private capitalUnder full dollarization, bank deposits grew from $314 million to $1,322 million during 2009, of which an estimated $270 million of capital inflows originated from U.S. dollar notes circulating outside the banking system
Table 2.2 (concluded)
Financing need: current account 2010 (Percent of GDP)Infrastructure, megaprojects, privatizationCapital controlsMarket infrastructure
Angola0.6A strict system of capital controls, and all capital account transactions are subject to prior approval by the National Bank of AngolaNo secondary government securities market nor stock exchange
Ghana-11.6Privatization of telecom operator Vodafone
Kenya-7.7Privatization of telecom operators Telkom and Safaricom
Mauritius-9.4Public and private sectors related to the tourism sector, a Chinese integrated industrial project, and various real estate development projects
Mozambique-13.6Plans to finance road project during 2010-13 with financing from a Portuguese public bank. FDI inflows associated with megaprojects, especially in coal mining, natural gas, and metals (titanium)Stock exchange is relatively new, with only two listed equity shares. Government bonds are issued on the stock exchange, but domestic banks accounting for most of the subscriptions.
Senegal-8.2Accessed the international markets in 2009 to finance a road projectNo liquid secondary market and only one Senegalese company is listed on the regional stock market.
Tanzania-8.8Tightened capital controls in an effort to discourage speculative inflowsForeign interest in Tanzania’s stock exchange has been growing, but total turnover on this exchange has been extremely modest
Uganda-6.4Portfolio inflows channeled mostly through commercial banks, which purchased government securities directly in the primary securities market on behalf of their foreign clients.
Zambia-1.6Privatization of telecom operator ZamtelTightened capital controls in an effort to discourage speculative inflowsMost of the equity activity has been in IPOs, the majority of which are sold outside the Lusaka stock exchange (LuSE). The LuSE lists only 20 companies, volumes are small, and the exchange is not well integrated with banking payments systems
Zimbabwe-21.3FDI is mainly driven by one large investment in platinum productionStock exchange was closed under stress from hyperinflation in November 2008, but reopened in February 2009. Market capitalization reached US$4 billion by end-October 2010, and portfolio equity investment generated an estimated net inflow of US$60 million in 2010. The liquidity in the exchange has started to improve since October 2010, with monthly transaction volumes reaching about US$50 million
Attracting inflowsIntermediateDiscouraging inflows
Source: IMF staff estimates.
Source: IMF staff estimates.

Resurgence. Ghana, Mauritius, and Zimbabwe—for quite different reasons—have recorded significant new inflows recently. Ghana recorded a resurgence of capital inflows in 2010, mainly the result of domestic factors (Figure 2.17). Unlike most sub-Saharan countries, Ghana tightened its fiscal and monetary policies in 2009 and 2010, in response to significant macroeconomic imbalances. Robust GDP growth prospects premised on strong performance in the gold and cocoa sectors, and the start of oil production, have increased investor interest. All categories of capital flows have recovered sharply since the third quarter of 2009, especially into government securities, which nearly tripled between 2009 and 2010. The recovery in capital inflows to Mauritius mostly mirrors developments in emerging markets. The country acts as a financial platform for investors bound for Asia and sub-Saharan Africa. With a strong regulatory framework in line with international norms and a resilient economy, Mauritius has benefited from favorable investor sentiment, despite lower domestic interest rates. All categories of capital inflows have surged, and have exceeded precrisis levels. Following the implementation of an economic stabilization program in 2009, Zimbabwe has recorded a sharp increase in portfolio equity investment and foreign currency deposit inflows.

Figure 2.17.Foreign Ownership of Government Securities

Source: Ghanaian and Zambian authorities.

Modest recovery. Zambia has recorded a modest pickup of net capital inflows, mainly in the form of FDI. This reflects increased confidence in the Zambian economy, and privatization-related inflows. Growth has been strong and copper prices have recovered sharply. That said, net portfolio flows have been negative in most of 2009 and 2010, reflecting disinvestment in government securities by foreigners who decreased their share from 11 percent (US$250 million in early 2009 to 1½ percent (US$30 million) at end-September 2010 (Figure 2.18). Private capital inflows recovered strongly in Kenya in 2009, but have subsided in 2010 while official flows picked up.

Figure 2.18.Global Indicators

Source: Bloomberg.

No recovery. Several countries have recorded no recovery in net capital flows. Capital inflows peaked in 2007–08, and the trend sharply reversed during late 2009. In countries such as Tanzania and Uganda, despite improving macroeconomic fundamentals, currencies have come under pressure in part as a result of net portfolio outflows.

Box 2.2.Foreign Direct Investment to Sub-Saharan Frontier Markets

Output growth, which supports returns on equity portfolio and foreign direct investment (FDI), has been higher in sub-Saharan Africa frontier markets (FMs) than in other groups (Figure 1, includes projections). Since the mid-1990s, output growth in sub-Saharan Africa FMs has consistently averaged above 4 percent, whereas growth in the other country groups has been more volatile (Figure 2). Growth was resilient in sub-Saharan African FMs even during 2008–09, while the other groups suffered sharp slowdowns during the global financial crisis. According to WEO forecasts, growth will be higher on average during 2011–15 in sub-Saharan African FMs than in any of the other groups. During the full time span from 1992 to 2015, average output growth, at 5.2 percent, is nearly a full percentage point higher in the sub-Saharan Africa FMs than in emerging market economies (EME) and sub-Saharan African nonfrontier groups, and about a ½ percentage point higher than in other FM countries.

Figure 1.Average Output Growth, 1992–2015

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

Figure 2.Output Growth, 1992–2015

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

The rate of return on FDI was substantially higher for the sub-Saharan African FMs than for other regions during the last few years. Figure 3 shows an estimate of the average rate of return on FDI from 1997 to 2008 for each Country group. Returns on FDI are estimated by dividing the current-year FDI profit repatriation by the sum of FDI during the previous 10 years, using data from the World Bank. This calculation assumes that 10 percent of the original FDI volume depreciates every year.1 Based on these estimates, sub-Saharan African FMs achieved the highest return on FDI across the benchmark groups, with an average rate of 17.8 percent over the period 1997–2008, compared with 16.3 percent for EMEs and 15 percent for other FMs. Return on FDI reached above 40 percent in sub-Saharan African FMs in 2008, well above the returns from the other regions (Figure 4). Similarly high rates of return on FDI flowing into sub-Saharan Africa in the 1990s were calculated by Asiedu (2002)2 using data from the United Nations Conference on Trade and Development (1999).3

Figure 3.Average Foreign Direct Investment Return, 1997–2008

Sources: World Bank, Global Development Finance; and IMF staff calculations.

Figure 4.Estimates of Rates of Return on Foreign Direct Investment

Sources: World Bank, Global Development Finance; and IMF staff calculations.

FDI inflows to sub-Saharan Africa partially reflect interest in resource extraction activities, but a diversification of the export base is also taking place. In sub-Saharan Africa, resource extraction industries generally constitute a large share of economic activity. Moreover, capital is often attracted by tax concessions and favorable contract arrangements. It is difficult to get a, comprehensive picture of the final investment destination of FDI given that a sectoral decomposition of FDI flows is not provided for the country groups analyzed here.

Nevertheless, existing data allow a check on the diversification dynamics of the export industry. The Herfindahl-Hirschmann index, for example, looks at how exports and imports of individual countries or groups of countries are concentrated on several products or otherwise distributed in a more homogeneous manner among a series of products.

Some important lessons can be drawn from the figures on concentration dynamics: On average, sub-Saharan African FMs are undiversified compared to other FMs and EMEs (Figure 5). Particularly the oil-exporting sub-Saharan African FMs (Angola and Nigeria) lack diversity in their export product pallet. From 2000 onward some improvement is noticeable for Nigeria while the concentration increased for Angola. High export concentration volatility is noticeable in the industrial development of Ghana, Mozambique, and Zambia. And for about half the sub-Saharan African FMs—Kenya, Mauritius, Senegal, Tanzania, Uganda, and Zimbabwe—the diversification of exports is comparable to other FMs and EMEs (Figure 5, dotted line). Sub-Saharan African FMs show the fastest average diversification improvements between 1996 and 2009, with a drop of the Herfindahl-Hirschmann concentration index of about 11 percent, followed by the EME group with a change of about -5 percent, while the other FMs had an increase in their concentration index of about 15 percent (Figure 6).

Figure 5.Average Concentration of Exports Index

Source: United Nations Conference on Trade and Development statistics.

Figure 6.Change in Concentration of Exports, 1996–2009

Source: United Nations Conference on Trade and Development statistics.

This box was prepared by Alexis Meyer-Cirkel.1 The calculation for the proxy FDI return is done as follows:100 * FDI Remittancet/(0.1 *FDIt9 + 0.2 * FDIt8 + 0.3 *FDIt7 + 0.4 *FDIt6+ 0.5 *FDIt5 + 0.6 *FDIt4+ 0.7 *FDIt3+ 0.8 *FDIt2+ 0.9*FDIt1+FDIt)2 Asiedu, Elizabeth, 2002, “On the Determinants of Foreign Direct Investment to Developing Countries: Is Africa Different?’ World Development, Vol. 30, No. 1, 107–119.3 United Nations Conference on Trade and Development, 1999, “Foreign Direct Investment in Africa: Performance and Potential” (New York: United Nations).

Sharp decline. For various country-specific reasons, there has been a sharp decline in portfolio flows in Nigeria, although net capital flows recovered in 2009 and have picked up in 2010. About US$3.6 billion of portfolio investment left Nigeria during 2008–09, against a net inflow of US$2 billion during the previous two years. This largely reflected lack of investor confidence in the economy as a result of the banking crisis, corruption scandals related to the stock exchange, and political uncertainty related to forthcoming national elections.

What Explains This Varied Experience?

This section explores the factors that have a bearing on the recent experiences of different forms of capital flows. These can be grouped into global and local factors that influence the supply of funds to FMs (for example, including the relative return between the source country and the recipient) and factors that determine the demand for funds.

Global Factors Determining the Supply of Funds

During the past several years, global factors have largely set the aggregate trends of net capital flows to sub-Saharan Africa. On balance, sub-Saharan African countries received net inflows in the years preceding the global financial crisis, followed by sharp declines in flows during the peak of the crisis. Excess liquidity and low yields in advanced countries provided the impetus for these overall trends, as in other regions (Figure 2.18). Risk appetite has been the other key global factor, as spreads and other measures of risk (such as the Chicago Board Options Exchange Market Volatility Index, VIX) spiked after the Lehman collapse. Risk has since been gradually declining except for periodic reversals that reflect sovereign debt and financial sector concerns in Europe. Global factors also include continuing interest in diversification into new markets, particularly those with high growth rates relative to Organisation for Economic Co-operation and Development (OECD) countries (Figure 2.19).

Figure 2.19.Real GDP Growth

Source: IMF, World Economic Outlook.

Local Factors Determining the Supply of Funds

Beyond these global trends, the relative size and composition of inflows among sub-Saharan African FMs and relative to other groups of countries depend on a large number of country-specific pull factors (Table 2.2). For all investment categories, the expected rate of return and risk determine the attractiveness of the investment. For debt securities and loans, changes in the interest rate and exchange rate provide the rate of return, and volatility is a key risk factor.

Policy rates. According to survey information, sharply lower domestic interest rates relative to the period before the crisis stand out as an important factor in the relatively slower recovery of debt portfolio inflows to sub-Saharan Africa.

In order to counter the effects of the global recession, most sub-Saharan countries adopted expansionary fiscal and monetary policies. In particular, policy rates were reduced by nearly 4 percentage points on average between end-2008 and September 2010 (Figure 2.20). Indeed, in sub-Saharan Africa FMs, the average policy rates declined much more than in other frontier and emerging markets (Figure 2.21).9

Figure 2.20.Policy Rates Across Regions

Sources: IMF, Information Notice System; and countries’ central banks.

Figure 2.21.Change in Policy Rate Spreads, end-2008 to end-September 2010

Sources: IMF, Information Notice System; countries’ central banks; and IMF staff calculations.

  • Yields on government securities have largely mirrored developments in policy rates. Although 91-day yields rose by 2½ percentage points on average at the onset of the crisis in 2008, yields subsequently declined 4¼ percentage points—to 7¾ percent on average in December 2010—partly reflecting the easing of monetary policy (Figure 2.22). This implied a decline in spreads vis-à-vis U.S. government securities. Yields on government securities have been a key factor in the distribution of debt portfolio inflows. Before the crisis, Ghana, Tanzania, Uganda, and Zambia received significant carry trade operations owing to high yields. Such carry trade has subsequently subsided in line with monetary easing.
  • Exchange rate depreciation has been more acute in the sub-Saharan African FMs since the global crisis. Indeed, whereas exchange rates have rebounded to precrisis levels in the EMEs, they have remained on a downward trend in sub-Saharan African FMs, which may have affected investor perceptions of the risk of continued depreciation (Figure 2.23).
  • Exchange rate volatility has increased since 2008, raising perceived exchange rate risk and discouraging inflows in many countries. In some countries, the central bank intervenes to smooth exchange rate fluctuations. For example, such intervention encouraged carry trade in Uganda before the crisis. Although hedging markets are underdeveloped or nonexistent, the recent volatility may encourage fledgling development of these markets. For instance, nascent signs of renewed foreign interest in Zambian government securities have been accompanied by short-term currency swaps of kwacha for dollars between nonresidents and commercial banks.

Figure 2.22.Average 91-Day Yield

Source: IMF, International Financial Statistics.

Figure 2.23.Evolution of Average Exchange Rates Against the U.S. Dollar

Sources: IMF, Information Notice System; and IMF, International Financial Statistics.

In addition to these factors with a near-term bearing on returns to debt portfolio investment, survey respondents also noted several other background factors that have influenced the recent volume of equity flows to FMs.

  • Commodity developments. Higher commodity prices—copper in Zambia and gold and platinum in Zimbabwe—were expected to boost economic conditions and may have contributed to higher inflows to these countries. Similarly, the coming onstream of oil production likely has enhanced investor interest in Ghana.
  • Macroeconomic policies have improved in a number of countries, most notably Zimbabwe. In some countries facing the challenge of reducing public debt, such as Ghana, the monetary-fiscal mix generates high yields, while credit worthiness is underpinned by other favorable conditions, such as in the oil sector.
  • Political uncertainty. Prospective elections in Uganda (February 2011) and Nigeria (April 2011) may have contributed to the sluggishness of inflows in these countries.
  • Cross-border interests have been an important reason for capital flows associated with Mauritius. Financial flows, mainly from advanced countries, are intermediated by banks in Mauritius, destined for other African countries or for Asia, especially India (Figure 2.24Figure 2.26).

Figure 2.24.Mauritius: Portfolio Investment by Destination, 2009

(Billions of U.S. dollars)

Source: IMF, Coordinated Portfolio Investment Survey (CPIS).

Figure 2.25.Mauritius: Portfolio Liabilities by Creditor, 2009

(Billions of U.S. dollars)

Source: IMF, Coordinated Portfolio Investment Survey (CPIS).

Figure 2.26.Mauritius: Portfolio Liabilities, 2009

Source: IMF, Coordinated Portfolio Investment Survey (CPIS).

1Excludes Angola and Senegal due to data availability.

Factors Determining Demand for Funds

While risk and return drive the supply of foreign capital, capital inflows also depend on official and private financing needs at the country level and the supply of investable securities and instruments.

  • External financing needs. External borrowing partly reflects the need to finance current account deficits. For instance, Zimbabwe relies on capital flows to finance current account deficits of about 20 percent of GDP and support liquidity in the fully dollarized banking system. The increase in external financing in Ghana from 2006 through 2008 also corresponded to the widening of current account deficits. In addition to financing current account deficits, many countries in sub-Saharan Africa are looking to scale up public investment for large infrastructure projects or other mega projects.
  • Foreign financing is often needed, because local financial markets and concessional resources are typically insufficient. Some FM countries, including Ghana in 2007 and Senegal in 2009, have already accessed international markets through sovereign bond issuance, and many other countries are contemplating issuing a sovereign bond. Some countries have also tapped foreign funds for domestic government debt.
  • Privatization of telecommunication operators recently has generated large capital inflows in a number of countries.

Long-Term Determinants of Private External Financing in Sub-Saharan Africa

The previous section examined a number of key factors underpinning the recent experience of capital inflows to sub-Saharan African FMs. This section considers the formal evidence over a longer period, using data from all of sub-Saharan Africa (excluding Nigeria and South Africa). The regressors include variables to capture “push factors” such as global interest rates and risk; “pull factors” such as local interest rates, growth rates, and exchange rate volatility; and constraints to financial transactions, such as capital account openness and financial development. The specification is broadly similar to Milesi-Ferretti and Tille (2011) and IMF (2011a, 2011b).

Overall, the determinants to capital flows into sub-Saharan Africa are broadly in line with those for flows into other regions. In particular, formal empirical analysis confirms that push and pull factors are important determinants of capital inflows to sub-Saharan African countries (Table 2.3).10 Among push factors, U.S. 10-year Treasury bond yields and the CBOE volatility index have negative relationships with inflows across the great majority of country groups. Thus, high yields and global risk aversion in advanced countries significantly deter global private portfolio flows, including to sub-Saharan African FM countries.

Table 2.3.Formal Evidence on Determinants of Capital Inflows
Sub-Saharan Africa1Sub-Saharan Africa

Frontier Markets
Non-High-Income, non-

Sub-Saharan Africa
All Countries
Net total

Net portfolio

Net total

Net portfolio

Net total

Net portfolio

Net total

Net portfolio

U.S. 10-yr Treasury bond yield-0.25**-0.26**-0.23**-0.01-0.11**-0.28**-0.20**-0.28**
Log (VIX index)0.18-0.390.60**-1.05**0.08-0.78**-0.06-0.74**
Output growth-0.010.07****0.06**0.01**0.05**
Private credit/GDP†4.31**2.164.10**2.902.08**1.181.77**1.00**
Capital account openness-0.06-0.12-**-0.030.10**-0.07
Average size GDP0.060.05-**0.000.03**
(in U.S. dollars, all periods)(1.09)(0.31)(-0.69)(1.09)(0.75)(3.63)(0.72)(3.87)
Time span1991–20101991–20101991–20101991–2010
Sources: Chinn-Ito (2008); IMF, International Financial Statistics, World Economic Outlook database, and IMF staff calculations; and World Bank, World Development Indicators (2010).Note: Dependent variable is the log level of the respective capital inflow concept.** indicate significance at the 5 percent level. † lagged by one year. T-statistics in parenthesis. Estimates obtained by fixed-effects regression.

Excludes South Africa and Nigeria.

Sources: Chinn-Ito (2008); IMF, International Financial Statistics, World Economic Outlook database, and IMF staff calculations; and World Bank, World Development Indicators (2010).Note: Dependent variable is the log level of the respective capital inflow concept.** indicate significance at the 5 percent level. † lagged by one year. T-statistics in parenthesis. Estimates obtained by fixed-effects regression.

Excludes South Africa and Nigeria.

Output growth and financial sector development (using private credit to GDP as a proxy) have a positive and significant association with inflows across the majority of country subsets.11,12 Capital account openness13 is less robust, influencing only total net inflows to non-sub-Saharan African countries. In contrast, these regulatory obstacles appear to have had limited power in steering capital flows to sub-Saharan African countries. The size of an economy, measured as the average GDP in U.S. dollars over the entire panel time span, plays a significant role in attracting capital in some samples.

A few factors expected to affect capital inflows do not seem to have much bearing. Exchange rate volatility14 was only important in constraining net total inflows to the full sample of countries. Although market participants indicate that the spread of the local policy rate vis-à-vis the U.S. rate is an important pull factor for bonds, formal econometric analysis is unable to find a systematic role for short-term interest rates in attracting foreign capital. This result may reflect a combination of poor data quality, simultaneity in the relationship between short-term interest rates and capital flows, and time-varying risk premiums in individual countries that are not adequately proxied by global risk.

Comparison with Other Groups

Sub-Saharan Africa FMs perform well on many indicators of macroeconomic and financial conditions and public sector management (Table 2.4). They outperform their regional peers in almost all indicators and compare favorably to the groups of other FMs and select EMEs in a number of indicators. In the category of macroeconomic conditions, the SSA FM group has the highest projected growth rates and the lowest external debt and related interest rate burden. In public sector management, SSA FMs lead other groups in low levels of informal payments to public officials, strength of legal rights, the least time required for enforcing contracts, and the group has the lowest total tax rate (as a percent of profit). On indicators of financial sector conditions, sub-Saharan Africa FMs’ banking sector performs well and ranks second lowest in cost-income ratio. Sub-Saharan Africa FMs outperform their regional peers in the credit depth information index and capital account openness, but trail the other benchmark groups.

Table 2.4.Comparison of Key Indicators, 2009
SSA Non-frontier MarketsSSA Frontier MarketsOther Frontier MarketsEmerging Markets
Capital Flows
Total net private capital inflows in percent of GDP (2010)
Total net private capital inflows in percent of reserves (2009)18.357.028.71.5
Net portfolio inflows in percent of GDP (2010)
Net portfolio investment in percent of reserve money (2009)-0.1-2.75.3-9.5
Macroeconomic conditions
GDP in billions of U.S. dollars (average)74093226
GDP per capita in U.S. dollars2,0261,7583,9426,542
Current account balance in percent of GDP (2010)-10.4-6.8-3.21.1
GDP growth projection in percent (average 2011–15)
Interest payments on external debt in percent of GNI (2008)3.270.391.721.15
Reserves in months of imports5.
External debt stocks in percent of GNI (2008)70.621.749.225.9
Inflation, consumer prices (annual percentage change)12.812.714.78.5
Public sector management
Informal payments to public officials in percent of firms27.31.613.898.3
Strength of legal rights index (0 = weak to 10 = strong)
Government stability index (higher number = greater stability)
Time required to enforce a contract in days664.7591.2750.1738.3
Time required to register property in days89.270.566.448.9
Total tax rate in percent of profits79.537.041.844.3
Primary schooling completion rate, total (percent of relevant age group)59.373.482.3110.5
Financial sector conditions
Domestic credit to private sector in percent of GDP (2008)
Credit depth of information index (0 = low to 6 = high)
Banking sector cost-income ratio0.
Capital account openness (Chinn-Ito), (-2.5 = closed to 2.5 = open)-
Sources: IMF, World Economic Outlook database, and International Financial Statistics; World Bank, Doing Business Indicators; and World Development Indicators; and PRS Group, International Country Risk Guide.Note: Group ranking in individual series, by color:
Sources: IMF, World Economic Outlook database, and International Financial Statistics; World Bank, Doing Business Indicators; and World Development Indicators; and PRS Group, International Country Risk Guide.Note: Group ranking in individual series, by color:

How does performance in these indicators relate to the size of capital inflows during the last two years? Portfolio flows relative to GDP seem to be the most sensitive to macroeconomic, financial, and public sector performance. The select group of EMEs has the best overall performance and also attracts the most portfolio inflows in percent of GDP. Likewise, SSA non-FM countries have the weakest performance and the least portfolio inflows to GDP. In contrast, portfolio inflows as ratios to reserve money and other types of capital inflows in 2009–10 have shown less of a systematic relationship to macroeconomic, financial sector, and public sector performance.

Exchange rate volatility in the sub-Saharan Africa FM group has roughly moved in lockstep with the other regions (Figure 2.27), although it has been slightly higher during 2009–10. Market participants often argue that increases in volatility may encourage dollarization, including of deposits in the banking sector. However, in spite of higher volatility since 2008, foreign currency deposits have remained a stable share of total banking sector deposits in the sub-Saharan Africa FMs (Figure 2.28). Indeed, no clear relationship is noticeable between changes in exchange rate volatility and the share of foreign currency deposits during 2009–10 (Figure 2.29).

Figure 2.27.Coefficient of Variation for Nominal Effective Exchange Rate, 12-Month Period

Sources: IMF, Information Notice System; and IMF staff calculations.

Figure 2.28.Sub-Saharan Africa Frontier Markets: Foreign Currency Deposits in the Banking System1

Source: IMF, International Financial Statistics.

Figure 2.29.Effective Exchange Rate Volatility and Dollarization, 2009 vs 2010

Sources: IMF, Information Notice System, and International Financial Statistics.

Policy Recommendations

Except for a few countries, the volume of post-crisis inflows has not yet returned to a level that is complicating macroeconomic management. So, most countries’ authorities have not yet been tasked to provide a policy response to large inflows. Indeed, some policymakers may be primarily interested in how to further induce stable and beneficial capital inflows that can support investment and growth. The formal results shown in the previous section point to policies that operate over a longer time horizon, such as developing financial markets and integrating them with global markets, and implementing policies that provide a supportive macroeconomic and institutional environment.

Nonetheless, for the countries that have experienced a rebound in inflows, and for the others in which market participants indicate incipient interest, it still is useful to consider a policy strategy to manage the flows. Indeed, inflows posed challenges for policy in several countries before the crisis, at times leading to undesired pressures and volatility of the exchange rate. During the crisis, outflows led to severe pressure for exchange rate depreciation in many cases.

Capital inflows to sub-Saharan African FMs can also be large relative to the size of the monetary base and financial markets, complicating monetary operations. For instance, foreign holdings of government securities have fluctuated by as much as 50 percent of reserve money, according to data from Ghana, Uganda, and Zambia (Figure 2.30). In contrast to many emerging markets that have adopted inflation targeting as their monetary anchor, many of the sub-Saharan African FM countries target a monetary aggregate. Large and volatile capital inflows may require frequent adjustments to monetary targets, particularly if money demand fluctuates significantly with expected exchange rate changes and interest rate differentials. Central banks may also switch their target from reserve money to broad money for a closer link to inflation objectives or implement a more flexible approach to monetary targets.

Figure 2.30.Foreign Holdings of Government Securities, 2005:Q1–2010:Q31

Source: Country authorities; and IMF staff calculations.

1For Ghana, data available for 2008:Q4–2010:Q3, and for Uganda, 2008:Q1–2010:Q2

Foreign investment into government securities may also create challenges for fiscal policy. A prolonged surge of capital inflows may well induce complacency among country authorities regarding the level of public deficits that can be sustainably financed without crowding out the private sector. If global liquidity remains excessive for an extended time, inflows to FMs may resume with vigor, particularly once concerns over elections and other temporary factors are eliminated. Thus, policymakers will need to weigh and prepare for alternative policy responses. Country options for addressing large capital inflows consist of a mix of macroeconomic policies (including monetary, exchange rate, and fiscal policies), macroprudential measures, structural reforms that increase the capacity of domestic capital markets, and capital controls. Similar to other countries receiving capital inflows, the primary response should focus on macroeconomic policy adjustment and prudential measures to ensure resiliency of the financial system.15

Capital inflows that are large relative to the size of the economy or financial markets may inevitably induce exchange rate appreciation. In some cases, where the exchange rate is undervalued and competitiveness is not a concern, exchange rate flexibility may be desirable. When competitiveness is already weak, however, the monetary authorities may opt to intervene in the foreign exchange market. If the level of foreign reserves is below a target level for reserve adequacy, the intervention may have the salutary side effect of building sufficient reserve buffers.

  • According to the latest exchange rate assessments by IMF staff, only Ghana and Mozambique had overvalued real exchanges relative to measures of their equilibrium values (Table 2.5). Other countries had exchange rates in line with fundamentals, although these estimates typically have a certain degree of uncertainty.
  • Competitiveness is a concern for authorities in Mauritius owing to large recent inflows and had been a concern in Uganda before the crisis. These countries and a few others with low reserve ratios, such as Kenya and Ghana, could take advantage of capital inflows to opportunistically build foreign reserve buffers.
  • Many countries have also intervened to smooth exchange rate fluctuations in both directions. They could also deepen financial markets to better absorb inflows while minimizing volatility.
  • Several countries—Angola, Kenya, Nigeria, and Zambia—have had large cumulative exchange rate appreciations since 2004. Although fundamental factors such as positive terms-of-trade shocks may explain these developments, the appreciation of the exchange rate may have reduced competitiveness of the noncommodity-based sectors.
Table 2.5.Current Macroeconomic Indicators
Exchange rate



Reserves in

months of


2011 CPI

2011 Output

2011 Fiscal

2011 Public



External debt

distress risk

Latest bank

2008 Capital


Zimbabwe0.68.36.5-2.753.127.0In distress-1.83
Sources: IMF, World Economic Outlook database, and African Department database.
Sources: IMF, World Economic Outlook database, and African Department database.

When monetary authorities intervene to prevent exchange rate appreciation, they must also decide whether to sterilize the corresponding increase in base money by offsetting sales of domestic securities. By keeping money growth subdued, sterilized intervention may assuage incipient pressures for high credit growth and asset price inflation. However, sustained sterilization may be ineffective in highly integrated capital markets, where the resulting high interest rates induce further capital inflows. Sterilization may also be difficult in shallow and illiquid markets where sales of domestic securities may not easily be absorbed without large changes in market prices. Also, if the central bank’s balance sheet is weak, it may be reluctant to incur high sterilization costs.

  • The Bank of Zambia has been active in sterilization operations, primarily through open market operations.
  • The Bank of Ghana has sterilized part of its exchange rate interventions, using its own paper. Operational costs, estimated at 0.3 percent of GDP in 2010, have already begun affecting its cash flow. Sterilization to address a future surge in capital inflows would likely have limits and be costly.
  • In some cases, the increase in reserve growth associated with intervention may not feed into credit growth and overheating when the banking system is weak. In Ghana, for instance, intervention was only partly sterilized in 2010. But the increased liquidity did not increase domestic credit growth because of low confidence and high nonperforming loans in banks.
  • Banking systems in several countries were strained by the global slowdown or by domestic factors. Efforts are under way to improve banking soundness in Mozambique and Zimbabwe by restructuring problem loans, cleaning up balance sheets, and improving supervision and the regulatory framework. Nigeria is improving banking soundness through recapitalization and a troubled asset purchase facility. For these countries, capital inflows intermediated by banks may not generate higher credit growth. On the other hand, weak banks may be even more vulnerable to volatility in flows.

A tighter fiscal stance can help stem overheating pressures by reducing aggregate demand. In contrast to restrictive monetary policy, fiscal restraint reduces domestic interest rates, thus deterring capital inflows.

  • Many countries in sub-Saharan Africa implemented countercyclical fiscal policies to counter negative effects of the crisis. The combination of these discretionary measures and the impact of the slowdown on revenues led to a rise in fiscal deficits and debt. As the economic recovery accelerates, countercyclical policy would imply that country authorities would need to tighten fiscal stances.
  • Fiscal consolidation would be important to reduce public debt for many countries, including Kenya, Ghana, and Mauritius. However, the space for tightening is limited by the need for public investment in infrastructure.

Macroprudential measures complement macroeconomic policies, focusing on regulations to ensure the soundness of the financial system. Measures could include, for example, an increase in capital adequacy requirements and a tightening of lending standards, particularly for loans to the real estate market.

  • In Mauritius, for example, net open positions in foreign exchange are limited to 20 percent of Tier 1 capital. Based on data through mid-2010, banks remained well within these limits because capital inflows resumed after the global crisis.
  • In some countries, such as Zimbabwe, a tightening of banking regulations for prudential reasons correspondingly may constrain net capital inflows.

The authorities may also complement macroeconomic and financial stability measures with temporary controls on capital inflows. The merits of such controls depend on their effectiveness and whether they are consistent with objectives for financial integration. The long-term trend has been toward more open capital accounts: Kenya, Mauritius, Nigeria, Uganda, and Zambia increased their openness (Chinn and Ito, 2008) from 1995 through 2008. However, Tanzania and Zambia tightened capital controls in an effort to discourage speculative inflows following their experience with abrupt reversal of flows during the global crisis.

Box 2.3.Do Sub-Saharan Africa Exchange Rates Respond to Changes in Foreign Interest Rates?

Given the increasing integration of African economies into the world economy, it is worth investigating whether their exchange rates actually respond to key global financial shocks. In particular, it is worth investigating whether exchange rates have become more sensitive to movements in global interest rates and risk patterns in recent years.

To do this, the standard workhorse model of exchange rate determination based on the work of Dornbusch (1980) and Frankel (1979) is used. The Dornbusch model is based on the purchasing power parity principle which assumes that the nominal exchange rate adjusts to equalize prices in a common currency. In the model, the exchange rate adjusts immediately so there is an expectation of an exchange rate adjustment in the opposite direction. The Frankel model supplements the Dornbusch model with the assumption of interest rate parity in which bonds of different countries are perfect substitutes and offer the same return in a common currency in equilibrium.

Theoretically, the exchange rate model can be presented as


s = U.S. dollar exchange rate

i and i* = domestic and U.S. nominal interest rates

R = risk premium

X = control variables

An empirical estimation is done on 12 SSA countries1 with floating exchange rates and sufficient data for 2003–09, distinguishing between countries with open and closed capital accounts (Table 1).1 For countries with open capital accounts, the figures reveal that the bilateral exchange rates moved in tandem with the decline in the Emerging Market Bond Index (EMBI) spread through end-2005 at which point the exchange rates began to depreciate against the U.S. dollar as the U.S. Treasury bill rate reached its peak (Figure 1). The exchange rates began to appreciate as the United States eased interest rates in late 2007 through 2008 but then depreciated sharply when the EMBI spread shot up to 7 percent. The exchange rates have appreciated somewhat since mid-2009 but not to the extent of the decline in the EMBI spread. The exchange rates of countries with less open capital accounts have followed movements in the U.S. Treasury bill rate, depreciating through end-2006 and appreciating subsequently until the relationship was broken by the sharp rise in the EMBI spread in mid-2008. Since then the exchange rates have depreciated at a fairly even rate.

Table 1.Determinants of Bilateral U.S. Dollar Exchange Rate Change
Countries with Open Capital AccountsAll Countries
Interest rate differential-0.030.00
U.S. Treasury bill rate0.34 ***0.18 **
Domestic t-bill-0.02
Policy rate0.01
EMBI interest rate spread0.35 ***0.64 ***0.20 ***0.40 ***
Domestic money growth (t-1)
Domestic inflation
Domestic inflation (t-1)
U.S. inflation-1.14 ***-0.99 **-0.77***-0.59 **
PPP variable (t-1)-0.03 ***-0.04 ***-0.01 ***-0.02***
Exchange rate change (t-1)0.22 ***0.21***0.29***0.29***
DW stat1.921.911.961.94
Source: IMF staff calculations.Note: ** and *** indicate significance at the 5 percent and 1 percent levels, respectively.
Source: IMF staff calculations.Note: ** and *** indicate significance at the 5 percent and 1 percent levels, respectively.

Figure 1.Interest Rates and Dollar Exchange Rates

Sources: IMF, International Financial Statistics; and Bloomberg.

This box was prepared by Alun Thomas.1 Six countries with open capital accounts (Ghana, Kenya, South Africa, Tanzania, Uganda, and Zambia) and six countries with less open capital accounts (Gambia, Madagascar, Malawi, Mozambique, Rwanda, and Sierra Leone).
This chapter was prepared by Valerie Cerra, Montfort Mlachila, and Alexis Meyer Cirkel, with research assistance from Duval Guimarães and Cleary Haines, and administrative assistance from Natasha Minges and Anne O’Donoghue.
1FM generally refers to a subset of emerging markets at an early stage of financial development with potentially high returns and low correlations to other markets (Box 2.1).
2These two large countries typically account for 50 to 60 percent of total flows.
3Private capital inflows exclude nonresident official creditors (source), but include securities and instruments issued by the government and monetary authorities (destination) of the recipient countries.
4Portfolio investment reflects balance of payments information from the World Economic Outlook database, as reported by recipient countries.
5The CPIS measures the global stock and geographical distribution of portfolio investment holdings, as reported by creditor countries. The survey may have gaps in coverage owing to nonparticipation of some important investing countries and international financial centers, as well as difficulties faced by many participating countries in capturing cross-border portfolio investment by households (and in some cases, enterprises) that do not use the services of resident custodians. The stocks are measured at market value; thus, annual changes reflect valuation effects and flows.
6For instance, Kenya has issued infrastructure bonds of up to 25 years.
7Gabon and Seychelles have also issued bonds in 2006 and 2007.
8Indeed, Asiedu (2002) also finds that factors that drive FDI to other developing countries, including returns on investment and infrastructure, are less important in driving FDI to sub-Saharan Africa.
9Countries included in the average discount rate for SSA frontier markets are Ghana, Mauritius, Nigeria, Senegal, Tanzania, Uganda, and Zambia. Other frontier markets are Bangladesh, Bulgaria, Jordan, Kazakhstan, Pakistan, Sri Lanka and Vietnam. Selected EMEs are Colombia, Egypt, Malaysia, Peru, South Africa, Thailand, and Uruguay.
10The results should be interpreted with caution, however, owing to the usual shortcomings associated with endogeneity of regressors and omitted variable bias. The dependent variable is the log level of net total and portfolio flows—thereby the estimations are run using only net inflows and treating net outflows as missing.
11In an alternative specification lagged output growth was used as the regressor, given potential endogeneity issues, and coefficients were very similar.
12This result is robust to the use of the first principal component of three variables (private credit over GDP, stock market capitalization, and cost-to-income ratio of the banking sector) as an alternative proxy for financial development.
13The capital account openness index by Chinn-Ito “is based on the binary dummy variables that codify the tabulation of restrictions on cross-border financial transactions reported in the IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER)” (Chinn and Ito, 2008).
14Measured as the coefficient of variation.
15The policy response hierarchy is outlined in IMF (2011c).

    Other Resources Citing This Publication