3. Private External Financing Flows and the Global Financial Crisis
- International Monetary Fund. African Dept.
- Published Date:
- April 2010
Introduction and Summary
The first decade of the 21st century witnessed a dramatic surge of private financial flows1 into emerging and developing regions, including sub-Saharan Africa. Gross private flows to the region increased fivefold from 2002 to 2007. Although, as in other regions, the global financial crisis caused inflows to plunge, there are tentative signs of renewed interest in the region from foreign investors.
Capital inflows can raise major challenges for policymakers. They can deliver the economic benefits of access to foreign savings and support for financial sector development. However, as their recent volatility has to some extent demonstrated, they must be managed carefully to avoid overheating of the economy, loss of competitiveness, and increased vulnerability to crises. Building on the staff analysis in the April 2008 Regional Economic Outlook: Sub-Saharan Africa, this chapter looks at the effects of the global financial crisis and the policy implications of volatility in flows.2
The chapter first briefly reviews recent research, theoretical and empirical, on the benefits and risks of international financial integration for developing countries. It then addresses the following questions:
What was the scale of private capital inflows to sub-Saharan Africa before, during, and after the global financial crisis? What form did the inflows take, and how did they compare with flows of private capital to other emerging and developing regions? How did the pattern of flows differ by country within sub-Saharan Africa?
How did the size and direction of swings in private capital flows compare with other external shocks that buffeted the region during the global crisis, such as huge swings in commodity prices and reductions in remittances? To what extent have official inflows increased to offset these shocks?
How did macroeconomic policies respond to these developments? Were there policies that helped to mitigate the impact of diminished private capital flows that could offer lessons for how policymakers should manage a resumption in private capital inflows?
Why have some countries been able to attract private capital inflows on a sustained basis and others have not, and what are the implications for policymakers?
The main findings are that
Private capital inflows to sub-Saharan Africa rose sharply during the recent expansion, though they failed to keep pace with the boom experienced in some other emerging and developing regions. However, the reduction caused by the global financial crisis was correspondingly more modest. This partly reflects the composition of these flows and the relatively greater importance in sub-Saharan Africa of foreign direct investment (FDI), which proved more resilient than other forms of private capital.
The boom in private capital flows bypassed over one-third of the countries in sub-Saharan Africa, and much of the region is still not integrated into international capital markets and dependent on official external financing.
For the region as a whole, when measured over a full economic cycle, financial flows have typically been a greater source of volatility than trade flows. However, during the current crisis, for many countries, movements in the terms of trade outweighed the reversal in private capital flows. For oil producers, deterioration in the terms of trade was exacerbated by the reduced availability of private external financing. However, remittance flows to the region have held up surprisingly well.
With multilateral institutions recently scaling up support, an increase in official financing has partially compensated for the reduction in private capital inflows. Bilateral donors also need to increase their support if they are to meet previous aid commitments. While the recent dramatic weakening of public finances and the expectation that economic recoveries will be anemic in donor countries makes this more challenging, the commitments are small relative to total donor budgets.
Among countries that attracted significant capital inflows before the crisis, better macroeconomic management when funds were flowing in was associated with superior performance when the global financial crisis hit and private capital flows diminished. Specifically, countries that had shown more fiscal restraint when inflows were surging experienced less deterioration in economic growth after the crisis. By contrast, resistance to exchange rate appreciation and capital account restrictions do not seem to have made a difference to the slowdown resulting from the crisis.
From a longer-term perspective, especially given the budget woes of traditional donors, official financing is likely to continue declining in importance and competition for external private financing is likely to become more intense. Experience within sub-Saharan Africa suggests that the same sorts of reforms needed to liberate productive potential—promoting trade and financial sector development, encouraging domestic savings and investment, raising standards of governance, and building up institutions—are also likely to help attract sustained private inflows.
International Financial Integration and Developing Countries
What Have We Learned?
Private capital flows to emerging market and developing countries reflect a combination of push and pull factors. Push factors comprise global determinants such as interest rates and market growth. Pull factors are those that affect the relative attractiveness of different destinations for investment opportunities (Figure 3.1). A number of pull factors have proven consistently helpful in attracting capital, among them market size, the quality of institutions, economic stability, and deep and open financial markets (World Bank, 2009a;Levy-Yeyati, Panizza, and Stein, 2007). Fiscal discipline and natural resources have also proven influential in attracting FDI to sub-Saharan Africa (IMF, 2008a). Several of these factors affect not only the size but also the composition of capital inflows. Sound institutions, for instance, may attract more FDI and portfolio flows, which are not only less risky than debt but also more likely to generate technology spillovers (Faria and Mauro, 2004).
Figure 3.1.Determinants of Private Capital Flows
Source: IMF staff.
Current research suggests that countries may need to reach a threshold level of development in order to reap the benefits and avoid the risks of financial integration. In theory, access to foreign capital should help capital-poor and labor-rich developing countries to increase investment and grow faster. Realizing these gains in practice, however, requires more than simply opening up to foreign capital. Inadequate protection of property rights, for example, will deter investors. Capital must also be allocated efficiently once it enters a country. Volatile real exchange rates, weak prudential supervision in the financial sector, output and labor market frictions, and tax policies biased against trade, such as high tariffs, may undermine otherwise viable investments (Kose, Prasad, and Taylor, 2009).3 In the absence of some of the necessary preconditions, opening up to foreign capital may do more harm than good, for example, by causing real exchange rates to appreciate or destabilizing fragile banking sectors (Rodrik and Subramanian, 2008).
Extensive empirical research has produced surprisingly little unambiguous evidence that in practice private capital flows lead to higher growth. Aside from the subtle and complex interactions, empirical findings are difficult to interpret because of simultaneity and the fact that other reforms that are likely to accompany financial liberalization may explain both an increase in inflows and the sorts of reforms that will attract inflows. Nevertheless, there is general agreement that the kinds of reform needed to curtail the power of entrenched economic interests and liberate the productive potential of developing economies are also helpful in attracting private capital flows and making these flows more productive (Obstfeld, 2009).
From a practical standpoint, financial sector liberalization and capital market opening need to be carefully managed. A sudden surge of capital inflows can undermine previously protected domestic financial sectors. Thus, to minimize the risk and severity of crises, policymakers must first strengthen prudential regulation and allow vulnerable banking systems time to learn risk management techniques and restructure their balance sheets. There is also an emerging consensus that not only the level but also the composition of financial flows matter for growth, so that sequencing is important. FDI and portfolio equity flows are not only more stable and less prone to reversals but are also more likely to generate technology know-how, managerial spillovers, and productivity growth. Debt flows, especially short-term debt, tend to be more procyclical and volatile and to magnify the negative impact of adverse shocks on economic growth.
The Pattern of Private Financing Flows during the Crisis4
Private financial inflows to developing countries expanded rapidly over the previous decade and sub-Saharan Africa shared in the boom. Globally, total gross private inflows to emerging and developing countries rose from $151 billion in 2002 to a peak of $1.7 trillion in 2007 or from $49 billion to $674 billion in net terms.5 FDI, the mainstay of investment flows to developing countries, nearly tripled. Even more dramatic was the explosive growth of portfolio (both debt and equity) and other flows (mainly bank loans and trade credits), from negligible amounts in 2002 to $1.1 trillion in 2007—65 percent of total capital inflows (Figure 3.2). All developing regions shared in the surge. For sub-Saharan Africa, gross private inflows rose from $10.1 billion to $53.0 billion, though outflows rose from $8.1 billion to $28.0 billion and net inflows rose more than tenfold, from $1.9 billion to $28.2 billion (Figure 3.3).
However, not all regions participated equally. One way to make this clear is to calculate the elasticities of the contributions of various regions to the global expansion—that is, the percentage change in inflows to each region divided by the percentage increase to all developing countries (Figure 3.4). An elasticity of less than one indicates a failure to keep pace with the expansion and a declining share of the global pie.
Figure 3.2.Volume and Composition of Private Financial Flows to Emerging and Developing Countries
Source: IMF, World Economic Outlook.
Figure 3.3.The Private Financing Cycle in Sub-Saharan Africa
Source: IMF, World Economic Outlook.
Figure 3.4.Elasticities of Gross Private Inflows during the Expansion of 2002–07
Source: IMF, World Economic Outlook, based on World Bank regional definitions: East Asia Pacific (EAP); Europe and Central Asia (ECA); Latin America and the Caribbean (LAC); Middle East and North Africa (MNA), South Asia (SAS), and sub-Saharan Africa (SSA).
During the recent boom South Asia, Europe and Central Asia, and Latin America and the Caribbean increased their shares; sub-Saharan Africa, East Asia Pacific, and the Middle East and North Africa lost ground. Indeed the latter two regions were net exporters of private capital. With an elasticity of 0.75, sub-Saharan Africa’s share of global flows declined from 6.0 percent in 2002 to 4.1 percent in 2007 and less than a quarter of countries were able to preserve their share. In terms of composition, inflows into sub-Saharan Africa show a similar pattern to global inflows with proportionately larger gains in portfolio and other inflows, but the increases were less dramatic (Figure 3.5).
Figure 3.5.Composition of Private Financial Flows to Sub-Saharan Africa
Source: IMF, World Economic Outlook.
Although portfolio and other flows had healthy growth, they still comprised only 38 percent of total inflows in 2007, and substantially less if South Africa—by far the region’s largest recipient of portfolio inflows—is excluded.
Within sub-Saharan Africa, access to private external financing is uneven. Two-thirds of total inflows went to the two biggest economies, South Africa and Nigeria, and another one-quarter went to the region’s other oil producers (Box 3.1). At the other end of the spectrum, the bottom eight countries had negative inflows, and the next ten together shared just 1 percent of total inflows to the region.
Similar disparities are apparent in inflows relative to GDP or population, and in the disaggregated components of flows (Table 3.1). Portfolio inflows were particularly concentrated, with South Africa receiving virtually all of it. FDI was somewhat more broadly distributed though still highly concentrated, with the region’s oil producers accounting for two-thirds of total inflows.
Having been highly concentrated on a handful of countries at the start of the decade, there were encouraging signs that access to international capital markets was broadening by the end of the boom. Excluding South Africa, the share of inflows going to the four largest recipients fell from 88 percent in 2002 to 46 percent in 2007, while many of the next 20 largest recipients increased their share (Figure 3.6). Nevertheless, inflows remained concentrated and the increase in flows bypassed nearly a third of the countries in the region. In nine countries, inflows declined during the upswing and in another four the increase was less than one percent of average GDP over the five-year period.
Figure 3.6.Concentration of Gross Private Inflows to Sub-Saharan Africa (Excluding South Africa)1
Source: IMF, World Economic Outlook.
1 Including South Africa, the distribution of flows still became less concentrated; the top four countries received 82 percent of inflows in 2002 and 70 percent in 2007.
Box 3.1.South Africa, Africa’s Largest Capital Market
South Africa is by far the largest and most sophisticated market in the region. By market capitalization, its equity market is among the 20 largest in the world (including advanced economies) and foreign investors trade actively in a large and liquid local debt market. South African companies (both private and public) as well as the government have been able to borrow routinely in international capital markets. Reflecting this, South Africa relies more than its neighbors on portfolio and other more volatile forms of investment and was more exposed to the global financial cycle, accounting for two-thirds of the growth in private capital inflows to the region between 2002 and 2007 and experiencing a larger reversal than the rest of the region during the crisis.
Source: IMF, World Economic Outlook.
|Gross Financial Inflows||Difference|
|(Billions of U.S. dollars)||(Percent of GDP)||(Per capita U.S. dollars)||(Percent of SSA Total)||(Billions of U.S. dollars)|
|Private gross inflows||240.9||4.6||42.6||100.0||42.9||-24.9||-5.3|
|Private gross inflows||113.7||6.1||300.6||47.2||25.9||-19.2||4.6|
|Private gross inflows||75.2||4.3||50.2||31.2||4.9||-6.0||-2.7|
|Private gross inflows||52.0||3.2||13.8||21.6||12.1||0.3||-7.2|
The nature of the region’s access to international capital markets became progressively more diversified as it broadened out beyond FDI and traditional bank lending. Sovereign borrowers in Gabon, Ghana, and Seychelles tapped international bond markets during 2006–07, and a growing number of countries secured sovereign credit ratings in anticipation of eventual issuance (Figures 3.7 and 3.8).6
Figure 3.7.International Sovereign Bond Issuance by Sub-Saharan Africa, 2000–09
Figure 3.8.Sub-Saharan African Countries with Sovereign Ratings1
1 Lists number of countries with a foreign sovereign currency rating from Standard and Poor’s, Moodys, or Fitch.
Foreign participation in local currency debt markets is difficult to track but is thought to have become significant in a number of countries, particularly Ghana, Nigeria, Uganda, and Zambia (IMF, 2008a). A growing number of countries also established equity markets (Box 3.2). Beginning in 2006–07 foreign participation in these markets, which had typically been limited to South Africa (by far the region’s largest market), began to widen to other countries. This was led initially by Nigeria but foreign investors have also ventured into a number of other new markets, including Botswana, Ghana, Kenya, Mauritius, and Zambia.
Private Capital Flows during the Crisis
The global financial crisis triggered a fall in private capital flows to sub-Saharan Africa (Figure 3.9):
Access to international bond markets was closed off during 2008. Spreads on sovereign debt widened dramatically after the crisis hit, and planned bond issues totaling about $3.3 billion were shelved.
Reduced participation by foreign investors immediately after the crisis helped push up yields on local government paper in some markets. Detailed data on Zambian debt show that foreign investors reduced their exposure (mainly by not rolling over their holdings of short-term government paper), though this was more than offset by increased purchases by domestic residents.
Net selling by foreign investors fueled declines in equity prices that generally tracked the price patterns in other developing and advanced markets.
Foreign banks reduced their total loan exposure to the region by about 15 percent ($14.4 billion) between September 2008 and June 2009 (Figure 3.9). Almost half the withdrawal came from a sharp cut in exposure to Nigeria, concentrated on its ailing banking sector, but there were also significant reductions in Ghana, Kenya, Tanzania, and Uganda. Syndicated bank lending commitments declined in South Africa but held relatively steady elsewhere, although this partly reflected the rollover of short-term financing rather than new commitments.
Figure 3.9.Selected Indicators of Access to International Capital Markets
1 Average for Botswana, Ghana, Kenya, Mauritius, Namibia, Nigeria, and WAEMU (weighted by market capitalization).
1 Excludes figures for Liberia and Mauritius, which are distorted by large flows associated with international shipping (Liberia) and offshore financing (Mauritius).
Source: Datastream; Bloomberg; Bank of Zambia; and Bank of International Settlements.
Total inflows to sub-Saharan Africa fell from $53.0 billion in 2007 to $22.8 billion in 2009, a decline of 57 percent, or 3.7 percent of GDP. Though large, the reversal was more modest than elsewhere. Globally, gross inflows to emerging and developing economies plummeted by 72 percent over the same period, with the biggest reversals in Central and Eastern Europe, followed by South Asia and Latin America—the same regions that had experienced the largest inflows before the crisis.
Box 3.2.The Emergence of Sub-Saharan African Stock Markets
The number of stock markets in sub-Saharan African countries has risen from 5 in 1989 to 16 today. Between 2002 and 2007 their value (market capitalization) nearly doubled to 153 percent of GDP before dropping to 83 percent of GDP in 2008 as the global financial crisis took hold (Figure).
While foreign capital flows have helped stimulate this growth (Andrianaivo and Yartey, 2009), in most cases markets remain too small and illiquid to attract more significant foreign investment. Except for South Africa and Nigeria, the markets are small (Table). Most have few listed companies, and at about 20 percent of GDP in 2008 (excluding South Africa) average market capitalization is lower than in most emerging markets. Market liquidity is less than 10 percent of the value of shares actually traded each year. Such low business volumes make it difficult to support a local market with its own trading system, market analysis, and brokers. According to Moss, Ramachandran, and Standley (2007), small size and lack of liquidity also deter foreign investors: the exposure of foreign institutional investors is typically negligible until a market reaches about $50 billion in size or $10 billion in shares traded annually. Reforms in a range of areas could support development of the region’s stock markets and in turn contribute to economic growth. Steps to improve the legal and accounting framework, private sector evaluation capabilities, and public sector regulatory oversight would also be beneficial. Appropriate sequencing of reforms is important: stock markets tend to develop only after financial sectors have reached a certain depth (Yartey, 2008). Opening up to foreign investors tends to be helpful only in countries that have little political risk and sufficiently high income (Andrianaivo and Yartey, 2009). Good-quality institutions, such as rule of law, democratic accountability, and limited corruption, are also important in reducing risk and enhancing the viability of external finance. The development of regional markets may also be a way to promote cost efficiency and overcome small market size.
Source: IMF, World Economic Outlook.
|Number of Listed Companies||Stock Market Capitalization||Stock Market Capitalization||Value Traded||Turnover|
|(Percent of GDP)||(Billions of US$)||(Percent of GDP)|
Across countries, there was a significant correlation between the strength of upswings and downswings. Countries where inflows had risen more in 2002–07 experienced relatively greater reversals in 2008–09.7 However, there were some differences by subgroup, the most notable being oil producers, which contributed only 11.5 percent of the growth during the upswing but 28.8 percent of the decline in the downswing (Figure 3.10). That largely reflects an earlier spike in investment in 2001–03 to bring onstream new production capacity in Angola, Chad, and Equatorial Guinea (Figure 3.11).
Figure 3.10.Contributions to Changes in Inflows over the Cycle
Source: IMF, World Economic Outlook.
Figure 3.11.Sub-Saharan Africa: Oil Investment and Production
Source: IMF, World Economic Outlook, and The Energy Information Administration (EIA).
By type of flow, FDI fell by just 9.5 percent in 2008–09, portfolio flows by 41.8 percent, and other flows by 490 percent.
There are tentative signs of renewed foreign investor interest in sub-Saharan Africa but inflows have not yet rebounded as much there as in some other regions. Foreign banks seem to have begun rebuilding their exposure starting in mid-2009. Spreads on the region’s external sovereign bonds have also fallen back to pre-crisis levels. South Africa returned to the international bond markets in mid-2009, Senegal issued its first international bond in December 2009, and Seychelles concluded a successful debt exchange operation in February 2010. Angola, Kenya, Mozambique, Nigeria, and Tanzania are among countries that have indicated their intention to borrow in international markets. However, the recovery in equity prices since early 2009 has been less strong in sub-Saharan Africa than in some other regions. Whereas equity prices in the more advanced markets of Botswana, Mauritius, Namibia, and South Africa have rebounded nicely, in other countries they remain subdued.
The Effect of Other External Shocks
Other external shocks hit a number of countries in sub-Saharan Africa because of the global crisis. Commodity producers, especially oil exporters, experienced sharp swings in their terms of trade, and countries that rely heavily on remittances saw these flows reduced as income and employment opportunities fell in advanced countries. Finally, higher official flows partially offset the decline in private flows.
Commodity producers, especially oil exporters, were subject to sharp price swings which were especially challenging when they added to the reversal of capital flows. For oil producers, massive terms-of-trade losses in 2009, averaging 26.8 percent of GDP, coincided with a reversal in financial flows of 3.8 percent of GDP.
However, for the non-oil-exporter group, terms-of-trade gains in 2008–09 largely offset the financing shock (Figures 3.12 and 3.13). Over the cycle, however, even at the region’s relatively low level of integration into global financial markets, for both oil and non-oil exporters financial accounts are considerably larger sources of volatility in the balance of payments than current accounts.8
Figure 3.12.Terms of Trade and Financial Shocks, Sub-Saharan Africa Non-Oil Exporters
Source: IMF, World Economic Outlook.
Figure 3.13.Terms of Trade and Financial Shocks, Sub-Saharan Africa Oil Exporters
Source: IMF, World Economic Outlook.
Fears that remittance flows would be substantially reduced because of the global financial crisis have so far proven unfounded. After peaking at $18 billion in 2008 (2¼ percent of recipient country GDP on average), officially recorded remittance flows fell by only $0.5 billion (3 percent) in 2009, according to preliminary estimates. While countries that rely more heavily on remittances faced somewhat larger reductions in these flows, in no case did the changes exceed 0.5 percent of GDP (Figure 3.14).9 However, the impact of a potentially “jobless” recovery in advanced economies may only feed through fully to remittances with a lag. A deceleration in construction activities in the Gulf Cooperation Council (GCC) states may also act as a drag on remittances, though sub-Saharan African countries are less dependent on remittances from GCC states than some other countries, especially in Asia.
Figure 3.14.Remittance Flows, 2008–09
Source: IMF, World Economic Outlook.
Official flows remain an important source of financing for many countries in the region.10 Excluding South Africa and Nigeria, official financing made up nearly half of inflows to sub-Saharan Africa over the cycle and virtually all of it for many of the poorest countries. At the individual country level, official flows on average dampened the swings in private capital flows (Table 3.2). A countercyclical pattern was also evident at the regional level, as official flows declined from 2002–06 before support was scaled up in response to the food and fuel price shocks and the global financial crisis (Figure 3.15).
|U.S. Dollar Value||-0.11||-0.09||-0.19|
|Percent of GDP||-0.26||-0.18||-0.20|Figure 3.15.Official and Private Financing to Sub-Saharan Africa Excluding South Africa and Nigeria
Source: IMF, World Economic Outlook.
Multilateral lending to the region increased substantially in response to the crisis. Tasked with leading the global crisis response, the IMF increased concessional lending to sub-Saharan Africa nearly fivefold in 2009, with new commitments of US$3.6 billion in concessional lending and US$1.4 billion in stand-by and extended arrangements. The increase in SDR allocations added a further US$12 billion of new reserve assets that governments can access on nonconcessional terms (see further in Chapter 1). World Bank financing to the region began increasing in 2007–08 in response to the food and fuel price shocks and expanded further in 2009, with new commitments of US$8.2 billion representing an increase of 135 percent over the 2006 level.
The prospects for continued scaling up of bilateral aid flows are not favorable; indeed reductions are a serious risk. Aid flows are vulnerable to severe recessions in donor countries, especially those where there has been a substantial deterioration in public finances. Preliminary indications are that the aggregate aid flows to Sub-Saharan Africa are likely to fall short of the Gleneagles target in 2010 (Box 3.3).
Policies to Manage Capital Inflows—Avoiding Hard Landings
There is variation in the recent performance of countries in sub-Saharan Africa that had experienced large capital inflows before the crisis. Some countries experienced relatively large postcrisis declines in output growth, but others seem to have escaped relatively unscathed. Some macroeconomic policy responses during the boom period seem to have been helpful in avoiding a hard landing when external financing conditions tightened and offer lessons for policymakers as capital inflows to the region resume.
Box 3.3.Official Aid during the Global Economic Crisis
Though aid flows to sub-Saharan Africa have increased significantly in recent years, they remain short of the commitments made at the G8 Summit in Gleneagles in 2005. According to the Organisation for Economic Co-operation and Development (OECD) Development Assistance Committee (DAC), aid flows (net of debt relief) from traditional donors tripled, from $8 billion to $24 billion, between 2000 and 2008. Aid flows from nontraditional donors reporting to the DAC (including Korea, Mexico, Turkey, and Saudi Arabia) increased from $1 billion in 2003 to $5 billion in 2008. Brazil, India, and China have also emerged as important sources of aid but do not report official numbers. While total financing flows (including FDI and commercial loans) from China to Africa are reportedly several billion dollars, one conservative estimate puts aid flows at about $1.4 billion in 2007 (Brautigam, 2010). China has committed to substantially increasing its aid over the next few years.
Source: OECD DAC.
1 Traditional donors refers to members of OECD DAC. Flows include debt relief, which was exceptionally large in 2005 and 2006.
Source: IMF, World Economic Outlook, and OECD DAC
Deep recessions in most advanced economies, which have severely strained their public finances, will make a further scaling up of aid flows more challenging. The average decline in GDP in DAC countries reached 3.7 percent in 2009, and only modest growth of 2 percent is anticipated in 2010. Even if countries commit to keeping aid programs constant as a share of GDP, this would translate into lower aid flows. The recession has also been accompanied by a precipitous deterioration in fiscal positions: budget deficits in DAC countries widened to an average of 9.2 percent of GDP in 2009– 10. In this environment, aid programs are vulnerable to cutbacks. A DAC survey of spending plans indicates that a majority of countries are on track to meet promises to increase aid made five years ago at the Gleneagles summit. However, the aggregate level of aid flows is likely to fall short of the Gleneagles target because of shortfalls from several large donors (OECD, 2010).
Empirical studies confirm the link between donor economic cycles and aid flows, especially during severe downturns. While in short and mild crises aid does not seem procyclical with respect to real growth or fiscal positions in donor countries, there is evidence that aid flows respond negatively and with a lag to severe downturns in donor countries (see, for example, World Bank, 2009c, and Hallet, 2009). Model-based approaches (for example, Faini, 2006;Bertoli and others, 2008; and Dabla-Norris and others, 2010b) relate aid flows to such economic fundamentals as the fiscal stance, output, and debt in donor countries. They generally find that aid declines with lower growth, a worsening of the fiscal stance, and higher debt in donor countries, although the statistical relationships are not always strong. Simple correlations between GDP growth in donor countries and aid disbursements tend to confirm this. For 1970– 2008, the correlation between real growth and real aid is low and negative (–0.13) but becomes positive and increases (to 0.24) when aid disbursements are lagged by two years (Figure). During severe downturns, when real GDP fell by over 2 percent, Canada, Finland, Source : IMF, World Economic Outlook ; and OECD DAC. Sweden, and New Zealand reacted by cutting aid significantly, whereas in the UK cuts in aid were relatively small and in Ireland, Italy, and Japan were nonexistent (Table).
|t, GDP < -2%||t+1||t+2||t+3||t+4||t+5|
The appropriate response to large capital inflows will depend on country-specific circumstances, including the nature of the inflows, the stage of the business cycle, and the strength of public finances and foreign reserves. However, experience elsewhere11 suggests that (1) maintaining fiscal restraint rather than allowing procyclical increases in public spending during periods of large inflows can help limit currency appreciation and reduce the risk of a hard landing when the flows reverse; (2) resisting nominal exchange rate appreciation tends to be ineffective if there is a persistent surge in capital inflows and can lead to excessive increases in domestic demand if the monetary impact of intervention cannot be neutralized; and (3) tightening capital controls does not seem to deliver better outcomes except perhaps where an economy is operating at near full potential, the level of reserves is adequate, the exchange rate is not undervalued, and flows are likely to be transitory. The rest of this section seeks to assess how much lessons like this applied in sub-Saharan African countries that experienced large capital inflows before the global financial crisis.
Macroeconomic Policy Responses
Quantitative indicators can be used to characterize the response of macroeconomic policies in sub-Saharan Africa to the recent rise and fall of capital inflows:12
Exchange rate policy can be measured by an index of exchange market pressure (EMP) that represents a combination of movements in exchange rates and international reserves.13 Dividing the changes in foreign reserves by EMP yields a ratio measuring the proportion of EMP that is resisted through intervention. This ratio is then standardized to create an index of the degree of resistance to changes in exchange rates—the resistance index (RI)—that has values between 0 and 1, where values closer to 1 imply more resistance to exchange rate fluctuations.14
Sterilization policy is captured by an index that measures the extent to which the monetary authorities are able to insulate domestic liquidity from foreign exchange market intervention. It measures the degree to which monetary authorities contract domestic credit to offset the expansion of the monetary base associated with reserve accumulation. A value of unity or above indicates full sterilization; a value of zero or below indicates no sterilization.
Fiscal policy is represented by the growth of real primary government spending.
Capital controls are measured through an index based on the IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions (Chinn and Ito, 2008).
Movements in the RI suggest that exchange rate policies have tended to be asymmetric and to lean against the wind to prevent the exchange rate from appreciating but not from depreciating. Rising capital inflows and favorable movements in the terms of trade before the global crisis led to upward pressure on foreign exchange markets, which countries tended to resist by accumulating foreign reserves. The crisis, however, brought about downward pressure on foreign exchange markets in late 2008 and early 2009. Most countries with flexible exchange rates allowed substantial downward adjustment and intervened very little to prop up exchange rates. There were, however, variations in policy responses. Kenya, Mauritius, South Africa, Uganda, and Zambia, for instance, allowed relatively significant upward movements in their exchange rates before the crisis, but all of them were simultaneously accumulating foreign reserves. During the crisis Angola, Malawi, and Nigeria, which had nominally flexible exchange rate regimes, intervened quite heavily to limit downward pressure on their foreign exchange markets. Malawi also rationed the provision of foreign exchange when reserves fell to critically low levels. Nigeria tightened exchange restrictions in order to limit pressure on the exchange rate.
Before the crisis the degree of sterilization increased as the region began to attract substantial amounts of foreign inflows. However, a dip in the sterilization index in 2007—the year of peak inflows for most countries—suggests that sterilization may have become more costly over time, or more difficult as increasing financial integration led to more substitutability between domestic and foreign assets.
Real government spending growth accelerated during the boom period though by less than previous economic cycles (IMF, 2009b).
Sub-Saharan Africa’s capital account regimes opened up in the 1990s although they are still more restrictive on average than in other regions. There has been little change in the average degree of restrictiveness in recent years.
Figure 3.16.Sub-Saharan Africa: Exchange Market Pressure Index1
Sources: IMF, International Financial Statistics; and IMF staff calculations.
1 Unweighted averages of country-specific índices (excluding the Democratic Republic of the Congo, Eritrea, Liberia, and Zimbabwe). The index is the weighted average of quarterly changes in foreign reserves and quarterly changes in nomical bilateral exchange rates, using the inverse of their standard deviations as weights. Changes in foreign reserves are normalized on base money.
Avoiding Hard Landings
A central question is whether these policies contributed to better outcomes when the global financial crisis struck and external financing tightened. Here we look at the countries in sub-Saharan Africa that received substantial capital inflows—specifically, those receiving above the median level of inflows for the region before the crisis.15
Figure 3.17.Sub-Saharan Africa: Selected Macroeconomic Policy Indicators1
Sources: IMF, World Economic Outlook; International Financial Statistics; and IMF, staff calculations.
1 Unweighted averages of country-specific indices (excluding Democratic Republic of the Congo, Eritrea, Liberia, and Zimbabwe) except where stated.
2 Calculated as the change in foreign reserves divided by the index of exchange market pressure. The results are then standardized with values between 0 and 1, where values closer to 1 imply a greater degree of resistance to exchange rate fluctuations. Results are shown for countries with floating exchange rate regimes.
3 Median value.
4 Measures the degree to which monetary authorities contract (expand) domestic credit to offset the expansion (contraction) of the monetary base associated with the accumulation (decumulation) of foreign reserves. Coefficient of sterilization estimated by regressing changes in central bank net domestic assets on changes in net foreign assets. A value of unity (or above) indicates full sterilization and a value of zero (or below) indicates no sterilization.
5 Average value of Chinn–Ito de jure index of capital account restrictiveness, normalized to between 0 and 100, with 100 indicating the most open regime.
This sample is then divided according to the degree of fiscal restraint, resistance to exchange market pressure, and the restrictiveness of capital account regimes observed in recipient countries during the precrisis inflow period. The aim is to assess whether policy differences in these areas had a bearing on how individual countries fared in terms of their growth following the crisis.16 The results (Figure 3.18) suggest that:
Figure 3.18.Sub-Saharan Africa: Postcrisis GDP Growth Deceleration and Selected Policy Indicators during the Precrisis Capital Inflow Period1
Sources: IMF, World Economic Outlook, International Financial Statistics, Chinn and Ito (2008), and IMF staff calculations.
1 Values reported are medians for the two groups. Countries with lower real primary spending growth are those with below the median level of real primary spending growth during the precrisis inflows period (2003-07). Similarly, countries with greater resistance to exchange market pressure are those with above median levels of the index of resistance to exchange market pressure during the precrisis inflow period. And countries with less open capital accounts are defined according to their average de jure measure of capital account restrictiveness during 2003-07 using the Chinn–Ito index.
Countries that exhibited greater fiscal restraint during the precrisis inflow period (captured by below median increases in real primary spending) experienced more modest slowdowns in GDP growth following the crisis. This may be because fiscal restraint during the upswing created room for a more robust countercyclical response during the downswing; countries that showed greater spending restraint during the upswing were able to increase real primary spending in 2009 by 7.2 percent compared with an increase of 3.4 percent in countries that had ramped up spending the most during the upswing.
By contrast, intervention in the foreign exchange market to resist upward pressure and the restrictiveness of capital account regimes during the upswing do not seem to have made much of a difference to the scale of the slowdown resulting from the crisis. The postcrisis growth deceleration was broadly similar in countries with above and below median levels of resistance to exchange market pressure, and also in countries with above and below median levels of capital account restrictiveness.
Policies to Attract Private Capital Flows
Given deteriorating public finances and the prospects of an anemic recovery in donor countries, countries in sub-Saharan Africa are likely to have to rely increasingly on private financing. This section explores which structural, institutional, and policy pull factors have been important in attracting private capital inflows in a sub-Saharan African context.
Some countries in sub-Saharan Africa have been consistently more successful than others in attracting capital inflows. To see what could explain this, we identify a sample of 24 countries, 12 of which have consistently been near the bottom of the distribution of gross private inflows to GDP and 12 consistently near the top (Box 3.4).
Both groups are highly diverse along most dimensions. Both contain low- and middle-income countries, small island and large landlocked states, and exporters of oil and other agricultural and mineral commodities.17 Virtually all countries in the sample score well in terms of some performance indicators and poorly in others and it is difficult to identify characteristic typologies. Perhaps the one exception is oil exporters, which combine high trade openness with low governance and human capital development indicators.
Nevertheless, systematic and important differences emerge between the groups. While the situations of individual countries seem to reflect idiosyncratic factors—specific binding constraints or competitive advantages—a comparison of sample means sharply differentiates the two groups in a revealing way (Table 3.3). Better-performing countries:
Were more integrated into the global economy with respect to financial and trade flows. Top performers had more open capital accounts and total gross private inflows were significantly higher, which is hardly surprising as that was the selection criterion. However, top performers also had higher financial outflows and trade shares. There were no significant differences in the composition of inflows (FDI, portfolio, and other) or in trade and current account balances as a percent of GDP.
Had bigger and more developed financial sectors. Broad money was significantly higher relative to GDP in top-performing countries, while private sector credit was higher though not significant.
Had higher measures of institutional quality. Top performers had better institutions as measured by the World Bank’s governance indicators. The rule of law and regulatory quality were both highly significant, and control of corruption was marginally significant.
|Indicator||Bottom 12||Top 12||Significance|
|Integration into global economy|
|Gross private inflows (percent of GDP)||-0.4||10.2||p < .001|
|Gross private outflows (percent of GDP)||1.0||5.3||p = .052|
|Trade (X+M) (percent of GDP)||73.2||114.2||p = .053|
|Capital account openness (de jure)||-0.9||0.1||p = .095|
|Financial sector development|
|Broad money (percent of GDP)||24.6||45.3||p = .043|
|Private sector credit (percent of GDP)||13.3||39.9||p = .337|
|Regulatory quality||-0.9||-0.5||p = .034|
|Rule of law||-1.0||-0.3||p = .004|
|Control of corruption||-0.8||-0.4||p = .056|
|Adult literacy (percent of population)||47.1||78.3||p = .010|
|Internet users (per 100 population)||1.3||7.1||p = .020|
|CPI Inflation||5.9||9.2||p = .304|
|GDP per capita, average 2002–07 (U.S. dollars)||358.1||1478.1||…|
|GDP per capita growth, 2002–07 (percent)||10.2||17.8||p = .036|
|Investment rate||0.2||0.2||p = .025|
|National savings rate (percent of GDP)||14.0||21.9||p = .067|Box 3.4.Consistent Winners and Losers in the Competition for Investment Inflows
Success in attracting inflows can be measured by the ratio of gross private inflows to GDP. Almost all countries move up and down in the distribution, but over the past two decades around half spent a preponderance of their time near the top or bottom. Using this as a selection criterion while weighting recent experience more heavily identifies two reasonably well-defined groups which on average were in the top or bottom quartiles of the distribution and were infrequently near the other extreme. The two groups comprise the countries shown in the table.
While the approach stresses consistency over time, experimenting with different rules such as splitting the sample at the median, or using a shorter time frame, or even using a completely different measure such as GDP growth, gives substantially the same results.
|Bottom 12||Top 12|
|Benin, Burkina Faso, Burundi, Comoros, Côte d'Ivoire, Ethiopia, Gabon, Guinea, Liberia, Rwanda, Sierra Leone, Togo||Angola, Cape Verde, Equatorial Guinea, The Gambia, Lesotho, Mauritius, Namibia, São Tomé and Príncipe, Senegal, Seychelles, Swaziland, Zambia|
Interestingly, dropping the four oil producers from the sample sharply increased the significance levels for all categories. One interpretation might be that investors in enclave sectors are able to find alternative ways of protecting their property rights, but that institutions matter in more general settings.
Had higher levels of human capital. Top performers had higher adult literacy rates and more Internet access, indicating a greater ability to supply human capital complementary to foreign investment flows.
Did not exhibit significantly better macromanagement as illustrated by the level of inflation, which was actually higher in the top 12. A possible explanation is that the top 12 countries were confronted by a more challenging environment, with faster growth and more volatility in capital inflows and the terms of trade.
Had better macroeconomic outcomes. The top 12 had higher investment and savings rates and enjoyed significantly higher real growth.
As elsewhere, the relationship between capital inflows and economic performance in sub-Saharan Africa is difficult to disentangle. While causality is hard to pin down, the stylized facts associate greater openness and higher levels of private financial flows with stronger institutions, higher savings and investment, and faster growth. Subject to the previous caveats in the second section of this chapter, “International Financial Integration and Developing Countries,” about the importance of consistent policies and careful sequencing of reforms, the findings suggest that the same policy frameworks that promote growth and development can also attract private investment flows in a mutually reinforcing way. Specific evidence on the impact of FDI further supports this conclusion (Box 3.5).
Box 3.5.Attracting and Reaping the Benefits of FDI in Sub-Saharan Africa
The business cycle in advanced economies tends to have a major impact on the volume of foreign direct investment (FDI) flows to developing countries. Estimates based on a modified gravity model suggest that (1) tighter monetary conditions in advanced economies tend to reduce FDI flows to developing countries, including in sub-Saharan Africa; and (2) the business cycle in advanced economies has a more pronounced negative impact on FDI flows to developing countries, especially to non-fuel exporters, during synchronized slowdowns.
|Countries with better economic|
fundamentals/more economic reforms/more
|Countries with worse economic|
fundamentals/less economic reforms/less stable
|Financial sector depth||Larger||0.36||***||Smaller||0.21|
|Trade openness||More open||0.24||Less open||0.35|
|Infrastructure (phone diffusion)||Higher||0.34||*||Lower||0.3||***|
|Control of corruption||Higher||0.33||**||Lower||-0.02|
|More noncommodity exports||Larger||0.49||***||Smaller||-0.34||*|
|Current account liberalization||Higher||0.16||**||Lower||-0.08|
|Capital account liberalization||Higher||0.14||Lower||0.01|
|Consumer Price Index||Lower||0.31||***||Higher||0.37|
|Real per capita growth||Higher||0.25||*||Lower||0.14|
FDI and GDP Growth by Degree of Financial Sector Depth
Source: IMF, staff estimates.
Standard growth regressions also suggest that FDI has a significantly positive effect on per capita growth in recipient developing countries, though the effect appears to be smaller in sub-Saharan Africa than in other developing regions.
To examine why the relationship between FDI and growth appears to be stronger in some countries than in others, the model is re-estimated using subsamples. Countries are separated into groups according to whether selected indicators are above or below the median value for the sample for (1) economic fundamentals; (2) economic reforms; and (3) macroeconomic stability. It appears that differences in economic fundamentals, the strength of reform, and commitment to macroeconomic discipline are important for explaining cross-country variations in the growth benefits of FDI (Table): FDI has more impact on growth in countries with (1) developed financial sectors (Figure); (2) better institutional quality; (3) more liberal current account regimes; and (4) a more stable macroeconomic environment (stable prices and steady growth).Note: This box was prepared by Jiro Honda, Amina Lahreche, and Genevieve Verdier and is based on Dabla-Norris and others (2010b).
Notes: This chapter was prepared by Robert Burgess, Robert Keyfitz, and Yanliang Miao, with research assistance by Gustavo Ramirez and Duval Guimarães.
Flows can be classified as public or private on the basis of either the source or the recipient. For instance, the purchase of a government bond by a foreign private investor would be identified as an official flow according to the borrower but a private flow according to the lender. For the rest of this chapter, flows will be designated on the basis of the creditor.
The Regional Economic Outlook: Sub-Saharan Africa (IMF, 2008a) assessed the growing importance of private capital flows before the global financial crisis.
For the most part, countries in sub-Saharan Africa typically lie below the thresholds for financial sector depth and institutional development that have been estimated for samples of emerging market countries. However, this does not mean that the region cannot benefit from financial integration. Estimated thresholds are, for example, sensitive to sample, model, and estimation techniques. And, as discussed in Box 3.5, there is some evidence that even within sub-Saharan Africa, the relationship between private capital inflows and growth tends to be stronger in countries with better institutions and deeper financial sectors.
Data on financial flows in this chapter are from the IMF World Economic Outlook database. For the most part they are in line with official series from country authorities. IMF staff estimates are used where official series are unavailable or inadequate. Capacity to monitor private financing flows in many countries remains weak and, as a result, there are significant shortcomings in the quality of some series.
The concept of gross inflows used in this chapter refers to the net acquisition of domestic assets by nonresidents. The sale of a domestic asset by a nonresident is then a negative gross inflow. The concept of net inflows refers to the net acquisition of domestic assets by nonresidents minus the net acquisition of foreign assets by domestic residents.
As of February 2010, 18 countries in sub-Saharan Africa have a sovereign credit rating from one or more of Fitch Ratings, Moody’s Investors Service, and Standard & Poor’s: Benin, Botswana, Burkina Faso, Cameroon, Cape Verde, Gabon, Ghana, Kenya, Lesotho, Mauritius, Mozambique, Namibia, Nigeria, Rwanda, Senegal, Seychelles, South Africa, and Uganda.
The correlation across the region between growth of inflows-to-GDP during 2002–07 and 2007–09 is -0.7.
Over the past decade, the average across the region of standard deviations of year-over-year changes in gross private inflows is higher than that of either the terms of trade or gross exports. The standard deviation of gross private inflows in 2002–09 averaged 8.1 percent, compared with 5.9 percent for the terms of trade and 6.0 percent for gross exports, all expressed as percent of GDP.
Based on World Bank (2009b). Information from other sources suggests a more mixed picture. For example, according to ODI (2010), remittances in Ethiopia fell by 10–20 percent in the first half of 2009.
For the present purpose, official flows are defined as lending by official creditors plus current official transfers.
The approach taken follows that in IMF (2007b), which examines policy responses in emerging markets from 1987–2006.
Changes in nominal interest rates are not considered here. They are unlikely to represent a powerful mechanism for attracting (or deterring) cross-border financial flows in most sub-Saharan African countries given the shallowness of domestic debt markets.
A critical step is the weighting of the two components of the EMP. An obvious option is an unweighted average, but since the volatility of reserve and exchange rate movements is very different, we weight the components to prevent one of them from dominating the index. Another question is whether to use country-specific or region-wide weights. Following IMF (2007b), we use region-wide weights to avoid the risk that countries whose exchange rates barely change would be seen as having a flexible exchange rate policy because of the very small standard deviation of the changes.
Based on this criterion, each of the recipients of large inflows received gross capital inflows of at least 3½ percent of GDP on average during 2003–07.
Postcrisis growth is defined here as the difference between average GDP growth in 2009 and the average during the boom period of 2003–07.
Also notable is the absence of several successful developing countries that did not satisfy the selection criteria set out in Box 3.4.