- Olivier Basdevant, Andrew Jonelis, Borislava Mircheva, and Slavi Slavov
- Published Date:
- June 2014
© 2014 International Monetary Fund
Joint Bank-Fund Library
The mystery of missing real spillovers in Southern Africa [electronic resource]: some facts and possible explanations / by Oliver Basdevant… [et al.]. – Washington, D.C.: International Monetary Fund, c2014.
1 online resource.
Includes bibliographical references.
Description based on online resource; title from digital title page.
1. Economic development—South Africa—Econometric models. 2. Economic development—Swaziland—Econometric models. I. Basdevant, Olivier. II. International Monetary Fund.
Disclaimer: The views expressed in this book are those of the authors and should not be reported as or attributed to the International Monetary Fund, its Executive Board, or the governments of any of its member countries.
Please send orders to:
International Monetary Fund, Publication Services
P.O. Box 92780, Washington, DC 20090, U.S.A.
Tel.: (202) 623-7430 Fax: (202) 623-7201
South Africa has strong ties with other countries in the region, particularly with its neighbors—Botswana, Lesotho, Mozambique, Namibia, Swaziland, and Zimbabwe (henceforth, BLMNSZ). South Africa’s economy is by far the largest in the region. It accounts for about 85 to 90 percent of these seven countries’ total GDP at market exchange rates. It is often described as an “engine of growth” for the region. Thus, it is reasonable to expect that South Africa’s business cycle and its long-term growth rate will affect those of neighboring countries. However, Figure 1 and the associated correlation coefficients in Table 1 show that there has been little or no correlation among GDP growth rates in Southern Africa in the past three decades. Mozambique is the only country whose GDP growth rate has been positively and significantly correlated with South Africa’s (looking at either contemporaneous growth rates or at five-year moving averages).
Figure 1.Real GDP Growth in Southern Africa, 1986–2012
Source: IMF, International Financial Statistics and African Department databases.
|Correlation with SA’s GDP||0.18||0.16||0.43||0.24||−0.20||−0.24|
|Correlation with SA’s GDP||−0.37||−0.54||0.72||0.20||−0.48||−0.75|
|(5-year averages, 1986–2012)|
|Exports to South Africa||13.2||38.9||21.4||31.4||58.2||55.4|
|(% of total exports, 2010)|
|Exports to South Africa||4.1||15.3||4.7||11.0||27.4||18.8|
|(% of GDP, 2010)|
|Imports from South Africa||86.3||93.7||41.6||71.0||96.5||48.1|
|(% of total imports, 2010)|
|Imports from South Africa||27.7||78.1||12.8||31.4||56.8||29.9|
|(% of GDP, 2010)|
|Exports of commodities||93.11||n.a.||43.32||77.31||90.13||68.61|
|(% of total goods exports)|
|Exports of manufactured goods||6.91||n.a.||56.72||22.71||9.93||31.41|
|(% of total goods exports)|
|Stocks of SA FDI||3.3||2.7||12.3||4.3||11.1||12.4|
|(% of destination country GDP, 2010)|
|Stocks of FDI in SA||1.9||2.8||n.a.||3.8||12.2||n.a.|
|(% of source country GDP, 2011)|
|Deposits in SA||n.a.||20.0||1.3||3.3||9.7||1.1|
|(% of source country GDP, 2010)|
|Market share of SA banks|
|(% of deposits, 2010)|
Based on 2001–12 trade data.
Based on 2001–11 trade data.
Based on 2005–12 trade data.
Recent data suggest a slowdown in South Africa’s economy, which is projected to persist by private-sector analysts as well as by the IMF’s October 2013 World Economic Outlook. Figure 2 compares the latest projections for South Africa’s growth rate to those in the October 2012 World Economic Outlook. It also offers a long-term perspective on South Africa’s growth rate. It is apparent that in addition to a weaker cyclical recovery, the projections have taken a more conservative view of the country’s long-term growth prospects, with the growth rate of potential GDP revised downwards by a percentage point, to around 3 percent. A weaker cyclical recovery is projected because recent data suggest softer consumption and private investment. Regarding South Africa’s long-term growth prospects, the more pessimistic projections are driven by slow progress on the structural reform agenda and by delays in implementing the public infrastructure program (aimed at easing electricity and transportation bottlenecks). The lower potential growth in South Africa appears to be part of a broader trend among emerging markets following the global financial crisis, which has been extensively documented in IMF (2013).
Figure 2.South Africa’s Real GDP Growth, 1980–2018
Source: IMF, International Financial Statistics and World Economic Outlook databases.
Note: WEO = World Economic Outlook.
This paper studies the potential spillovers, both cyclical and long-term, from a slowdown in South Africa’s economy to the other economies in the region and beyond. Multiple linkages exist between South Africa’s economy and those of its neighbors. South Africa forms the Common Monetary Area together with Lesotho, Namibia, and Swaziland, with the rand serving as legal tender throughout and Lesotho, Namibia, and Swaziland currencies pegged to the rand at parity. There are strong fiscal channels at work in the region as well, due to the customs revenue-sharing mechanism under the Southern African Customs Union (SACU). South Africa is a key export destination for most countries in the region, and an even more important source of imports. It is also an important source of foreign direct investment (FDI). The branches and subsidiaries of South Africa–based banks dominate the banking sector in all neighboring countries. South African financial and nonfinancial firms also have a strong presence, while households and firms in the neighboring countries own financial assets in South Africa. South Africa is home to an estimated almost two million workers from elsewhere in the Southern African Development Community (SADC), or just under 1 percent of SADC’s total population (excluding South Africa). The associated flows of remittances are significant. Given all these linkages, it is logical to expect that South Africa’s GDP growth rate should affect those of its neighbors. In other words, there should be real spillovers.
This is an important research question. The IMF has made spillovers a priority in its surveillance of member countries. Existing studies of spillovers in Southern Africa have reached seemingly contradictory conclusions. While Arora and Vamvakidis (2005) found statistical evidence of real spillovers, Kabundi and Loots (2007), IMF (2012), and Canales-Kriljenko, Gwenhamo, and Thomas (2013) found no strong evidence of real spillovers from economic developments in South Africa to the region and the continent.
Chapter 1 reviews the institutional arrangements and stylized facts that would lead us to expect a high degree of real spillovers from South Africa to the region. Chapter 2 illustrates this point by numerically simulating the channels of transmission for Swaziland using a calibrated dynamic stochastic general equilibrium model (the IMF’s Global Integrated Monetary and Fiscal [GIMF] model). Swaziland was chosen for this exercise as it is a good example of the relationship between a small open economy in the region and South Africa. Chapter 3 reviews and updates the available econometric evidence and concludes that after 1994 there is no strong evidence of real spillovers from economic developments in South Africa, either cyclical or structural, to the region and the continent after controlling for global growth.
Chapter 4 investigates a number of plausible explanations for this surprising and counterintuitive finding. First, the extent of trade, financial, and institutional integration in Southern Africa might be more limited than conventional wisdom and anecdotal evidence suggest. Second, as international sanctions ended and South Africa reintegrated with the world economy, the impact of South Africa’s growth on the region and the continent is no longer detectable separately from that of world growth. Finally, the multiple spillover channels might operate in ways that offset each other, so that their net impact is relatively small.