THE WORLD BANK GROUP AFRICA REGION POVERTY REDUCTION & ECONOMIC MANAGEMENT J U Ly 2 0 1 1 IssUE 1 South Africa 63539 Focus on savings, Investment, and Inclusive Growth Economic Update South Africa Economic Update Focus on Savings, Investment, and Inclusive Growth © 2011 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street NW Washington, DC 20433 USA All rights reserved This report was prepared by the staff of the Africa Region Poverty Reduction and Economic Man‑ agement. The findings, interpretations, and conclusions expressed herein are those of the authors and do not necessarily reflect the views of the World Bank’s Board of Executive Directors or the countries they represent. Photo credits: John Hogg and Trevor Samson, World Bank. The report was designed, edited, and typeset by Communications Development Incorporated, Washington, DC. Contents Acknowledgments vi Foreword vii Executive summary ix Section 1 Recent economic developments and prospects 1 Global trends: Developing countries lead global recovery 1 Recent trends in South Africa: Broad‑based economic recovery 1 Economic outlook for South Africa 10 Risks to the outlook 11 Section 2 Toward more inclusive growth: Focus on savings and investment 15 Savings, investment, and long‑term growth 15 Investment rates in South Africa: Role of the real returns to capital 20 South Africa’s savings: Required levels and determinants 24 The way forward 29 Annex 1 Computing the real return to capital 33 Annex 2 Empirical relationship between real returns to capital and fixed investment 35 Annex 3 Data and variable construction—Actual vs. predicted values 37 Annex 4 Methodology at a glance 39 References 41 Notes 45 Boxes 1.1 Commodity market developments 3 1.2 South Africa joins the BRICS (Brazil, Russian Federation, India, China, and South Africa) 6 1.3 Low foreign direct investment inflows to South Africa 12 2.1 South Africa’s nongovernment savings: What explains the declining trends? 19 2.2 Main determinants of savings 27 iii SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth 2.3 Saving‑growth for four successful emerging market economies 28 2.4 Stokvels in South Africa 30 2.5 Developing Factory Southern Africa—Lessons from Factory Asia 31 Figures 1 Toward a virtuous cycle for inclusive growth viii 1.1 Output in the secondary sector is below precrisis levels, while output in the primary and tertiary sectors are 5 percent above precrisis levels 4 iv 1.2 Unemployment has been worryingly unresponsive to the economic recovery 5 1.3 Young people and new entrants are the most vulnerable in the job market 7 1.4 In manufacturing, growth in real wages far outstripped growth in total factor productivity between 2005 and 2010 7 1.5 Despite increased public sector borrowing, fiscal sustainability has not yet emerged as a concern 8 1.6 Having benefitted from improved global import demand, South Africa’s exports have picked up briskly 9 1.7 The current account deficit rebounded from an exceptionally low 1 percent of GDP in 2010 Q4 to 3.1 percent in 2011 Q1 10 1.8 Net portfolio flows turned sharply negative in the last quarter of 2010 as foreigners reduced their positions in South Africa while South Africans increased their portfolio positions abroad 10 1.9 Foreigners reduced their net positions in domestic stock and bond markets in the first quarter of 2011 11 2.1 Toward a virtuous cycle for inclusive growth 16 2.2 Real GDP per capita in South Africa has grown at an unremarkable pace over the last three decades 17 2.3 Over 1980–2008, the average South African’s real income increased less than 10 percent, while the average Chinese became 11 times richer 17 2.4 The relationship between investment and growth is positive . . . 18 2.5 . . . as is the relationship between saving and investment . . . 18 2.6 . . . and the relationship between saving and growth 18 2.7 After peaking at 35 percent in 1980, South Africa’s national savings rate has been declining 20 2.8 Since the 1990s, South Africa’s savings rate has been low compared with other emerging economies 20 2.9 South Africa’s investment rate shows three trends since 1960 21 2.10 South Africa’s domestic fixed investment rate has slipped compared with other emerging economies 21 2.11 Real returns to capital in South Africa have risen sharply since the early 1990s . . . 22 2.12 . . . benefiting from the rising real profit rate and moderating price increases for capital since 1993 22 2.13 Real returns to capital in South Africa have been highest in construction and trade 23 2.14 Projected savings rate for South Africa under pessimistic scenario 25 2.15 Projected savings rate for South Africa under business as usual scenario 25 2.16 Projected savings rate for South Africa under high employment growth path scenario 25 2.17 South Africa’s demographic profile (proportion of total population) 27 Tables 1.1 The global outlook, 2009–13 2 1.2 GDP growth by main sectors (value added), 2007–11 Q1 2 1.3 Gross domestic expenditure growth by component, 2007–11 Q1 4 1.4 Growth and unemployment in the BRICS (Brazil, Russian Federation, India, China, and South Africa) and selected OECD economies, 2008–10 (percent) 8 1.5 Consolidated government fiscal framework, 2007–13 (percent of GDP unless otherwise indicated) 8 1.6 Macroeconomic outlook, 2007–13 (percent change unless otherwise indicated) 13 2.1 Statistics for countries with average annual GDP growth of at least 6 percent a year, 1980–2008 17 2.2 Predicted changes in saving rates in South Africa (percentage points) 26 A2.1 Regression results for the relationship between private fixed capital accumulation and real v returns, 1955–2010 36 Acknowledgments This report was prepared by a team com‑ The team is thankful to a large number of prising Fernando Im and Sandeep Mahajan individuals who helped shape the report and (Co‑Task Team Leaders), Allen Dennis, and provided invaluable insights. These include Phindile Ngwenya. Brian Pinto (Senior Advi‑ Noro Andriamihaja, Irina Astrakhan, Claus sor, PRMVP) was a special guest coauthor for Astrup, Thomas Buckley, Alfredo Cuevas this issue. Sarwat Hussain provided communi‑ (IMF), Constantino Hevia, Sarwat Hussain, cations support. Simi Siwisa and Ian Gillson Patrick Kabuya, Nir Klein (IMF), Guo Li, Nor‑ provided useful contributions. David Rosen‑ man Loayza, Catherine Macleod (National blatt, Wolfgang Fengler, and Jane Kiringai were Treasury), Elvis Mtonga, Ashish Narain, Zeinab the peer reviewers for the report. The report Partow, Marco Scuriatti, Renaud Seligmann, was prepared under the guidance of Ruth and Chunlin Zhang. Kagia (Country Director for South Africa) and John Panzer (Sector Manager). vi Foreword The World Bank Office in South Africa is African is gathering strength, with economic pleased to launch a new, biannual series of eco‑ growth expected to pick up to 3.5 percent in nomic reports—the South Africa Economic 2011. The policy emphasis now shifts back to Update. This first issue is anchored in the the longer term structural issues. For South national aspirations of faster and more inclu‑ Africa, this entails raising the level of growth sive growth, with special emphasis on the issues and making it more inclusive. The modest per‑ of savings and investment. formance on savings and investment certainly is By way of background, this genre of eco‑ not consistent with the government’s objective nomic reports constitutes an important aspect of attaining 6–7 percent GDP growth. This is of the World Bank’s analytical program in a clearly brought out by the experience of success‑ number of middle income client countries, ful countries that have maintained high rates of including Brazil, Russia, India, and China— growth over sustained periods and backed by the alliance that South Africa has recently influential studies in the economics literature. joined, collectively now referred to as BRICS. The crucial challenges associated with lifting In keeping with the established tradition, the the savings and investment rates are covered South Africa Economic Update comprises with analytical depth in this inaugural issue. two sections. The first section summarizes the We offer this report for critical review with recent economic developments, juxtaposing the sincere hope that it will contribute to the both global and country‑level trends. The sec‑ national debate on an important topic and ond section takes a deeper analytical look at a help shape informed policy decisions for sus‑ topic of special interest and high relevance to tainable economic recovery in South Africa. the country. In the aftermath of the worst financial crisis Ruth Kagia since the Great Depression that left no coun‑ Country Director for South Africa try untouched, economic recovery in South The World Bank vii Executive summary The firming of the economic recovery is put‑ strengthen consumer spending. Strong gov‑ ting the policy spotlight back on the longer ernment spending, as part of the countercycli‑ With considerable term challenge of faster, more inclusive GDP cal fiscal policy, will boost growth in 2011 and growth. Modest investment rates despite attrac‑ 2012, but is likely to wane thereafter. strengthening tive returns and low savings rates despite In light of South Africa’s low national sav‑ of the economic favorable demographics are important impedi‑ ings, the reemergence of high current account ments. A virtuous cycle of faster capital accu‑ deficits, financed mostly through volatile port‑ recovery, the focus mulation, job creation (especially for the folio flows, will reemerge as the biggest cause shifts back to the youth), and technological advancement needs for macroeconomic concern over the medium to be stimulated. There are no quick fixes that term. challenge of raising can produce the desired stimulus. The quest With considerable strengthening of the eco‑ GDP growth and for inclusive growth calls for a different, bolder nomic recovery and GDP projected to reach approach. Integration of the advanced and its potential by 2014, the focus shifts back to making it much less‑developed economies and more effective the longer term challenge of raising GDP more inclusive integration with the global economy, using Fac‑ growth to 6–7 percent—and making it much tory Southern Africa as a platform, hold con‑ more inclusive to tackle the extremely high siderable potential. unemployment. Broad-based economic Faster inclusive growth: Focus recovery in South Africa on savings and investment South Africa’s medium‑term growth prospects Confronted with widespread and persistent point to a strengthening recovery. GDP growth exclusion and unemployment, policy makers is projected to be 3.5 percent in 2011, 4.1 per‑ in South Africa have rightly set their eyes on a cent in 2012 and 4.4 percent in 2013. The long‑ trajectory of faster, more inclusive GDP growth. term potential growth rate under the current This will require lifting the current low rates policy environment is estimated at 3.5 percent. of investment and savings, more intensive The ongoing global recovery should con‑ use of labor, and heavy doses of productivity tinue to support exports, but strengthening enhancements. domestic demand will increase imports and A suboptimal equilibrium of low rates of moderate net contribution of trade to growth. savings and investment—low employment Consumer spending is expected to remain intensity of production—slow productivity strong, and gross fixed capital formation is growth has emerged in South Africa (figure 1), forecast to grow after declining for two con‑ under mining the quest for inclusive growth. secutive years. As businesses demand more A stimulus to any one of the three elements labor, employment should rise and further can generate a virtuous cycle of faster capital ix SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth Toward a virtuous cycle starts with basic education, where access has Figure for inclusive growth improved, but quality has not. Test scores 1 show South Africa faring miserably relative Higher savings and to global comparators. With intakes largely investment ill‑equipped with cognitive skills, the prob‑ lem only gets compounded at the higher Inclusive and technical education levels. growth Higher • Labor relations are much more contentious in Higher employment South Africa than other emerging market x productivity intensity of production economies. This is an implicit tax on invest‑ ment, partly explaining why global inves‑ tors, armed with options, have eschewed long‑horizon opportunities in South Africa. accumulation–job creation–technological In addition, wage levels relative to worker advancement. productivity are significantly higher than Employ ment‑intensive growth would among South Africa’s peers, also discourag‑ enhance productivity by skilling and tooling ing new investment. the unemployed labor force. Tackling youth • Savings rates are low, as discussed below. unemployment will enhance the benefit from the favorable demographics and raise savings. What explains the low savings rates? Productivity enhancements, in turn, will raise South Africa’s savings rates have been sig‑ GDP growth and attract private investment, nificantly below the potential suggested by its while lessening the burden on investment and economic and structural characteristics. Our saving to support higher growth. Just as higher analysis suggests that visible improvements in savings rates will need to underpin higher the national savings rate will depend most of growth over the long run, higher incomes, all on: especially among the lower end of the income • Resolving the high youth unemployment. The spectrum, will raise the level of savings. young‑age population ratio in South Africa fell from 58 percent in 1996 to 47 percent in What explains the low investment rates? 2009. This demographic group is still under The low private investment rates could mean 30 years of age and, according to employ‑ one of two things: either the real returns ment statistics, facing acute unemployment. to capital are low, or private investors are Resolving the unemployment problem for not responding to changes in real returns this group holds tremendous potential for because of risk perceptions or structural increasing the savings rate. barriers to investment. Our research shows • Ensuring a productivity-led spurt in GDP growth. that returns are high across most major sec‑ Substantial increases in economic growth tors and have been increasing since the mid‑ are typically accompanied by increased 1990s: South Africa ought to be an attractive savings, as consumption patterns tend to place for investment. This suggests that the change more slowly, in line with the habit real problem might be insufficient savings, formation hypothesis. The increase in sav‑ rising risk perceptions or deeper structural ings, in turn, lays the foundation for sus‑ impediments. tained high growth over the long run. Four issues stand out in this regard: • Fiscal consolidation as envisaged by National • Industrial competition is much weaker in Treasury should have a desirable effect on South Africa than its international peers. national savings. Global experience shows This points to entry barriers that discourage that permanent increases in public savings new investment despite high returns. tend to increase national savings. Further‑ • Skills development remains an important more, reductions in recurrent public spend‑ deterrent for firms, new and old, looking ing tend to be more effective than increases to expand their business. The problem in tax collection. The way forward into, not out of, the less‑developed economy, How to jumpstart the virtuous cycle of faster and that labor is more mobile toward the capital accumulation, job creation, and techno‑ advanced economy while entrepreneurs have logical advancement? No quick fixes or small greater access to its markets. Integrating the perturbations will produce dramatic results. two economies will require a big push on pub‑ The quest for inclusive growth will require, lic transport infrastructure while implement‑ above all, a significantly different mindset. ing programs to enhance financial inclusion Two major pushes, both in the spirit of cre‑ and improving the cognitive and technical ative and bold thinking, seem to hold the most skills of youth. xi potential —one requiring a more intensive inward look, and the other an outward look. Smarter regional integration More effective integration with the global More effective internal integration economy is needed based on South Africa’s A big push is needed to better integrate the latent comparative advantages, particularly its advanced economy and the less‑developed two surplus endowments in natural resources economy, marked by the spatially separated and unemployed labor. South Africa should be A big push is needed townships and informal settlements where the an attractive destination for long‑term global to better integrate the bulk of the unemployed live. Faster growth investment but it is not. A clear, consistent, will have to come from the less‑developed and predictable strategy for attracting foreign advanced economy and economy, which has the potential to take off direct investment will be important in this in the same way that other successful emerg‑ regard. Factory Southern Africa can underpin the less-developed ing market economies have. Ways will have to South Africa’s competitiveness in global mar‑ economy be found to exploit the “arbitrage” between kets, based on nimble “win‑win” regional pro‑ the two economies to ensure that capital flows duction supply chains. SECTIoN 1 Recent economic developments and prospects Global trends: Developing overhang, and the sovereign debt crisis in countries lead global recovery parts of Europe. Economic recovery The global economy recovered to an impres‑ sive 3.8 percent growth in 2010 after having Recent trends in South Africa: in South Africa contracted 2.2 percent in 2009 (World Bank Broad-based economic recovery has continued to 2011d). Developing countries, with their GDP Economic recovery in South Africa has contin‑ growth accelerating by 5.4 percentage points in ued to gather strength, benefitting from the gather strength 2010, contributed almost half the year’s global positive (albeit risk‑laden) global trends, GDP growth. This was aided by their vibrant domes‑ growth picked up from –1.7 percent in 2009 to tic demand, resurgence in international trade, 2.8 percent in 2010, and the momentum carried increased financial flows, and, for commod‑ into 2011 Q1 with 4.8 percent growth (q/q, sea‑ ity exporters, higher commodity prices. High sonally adjusted and annualized; table 1.2). Of income countries also rebounded strongly the 10 major sectors, 8 posted positive growth in from recession in 2009—to 2.7 percent growth 2011 Q1, with only agriculture and construction in 2010—as countercyclical macroeconomic bucking the trend. We project GDP growth to policies showed effect, business investment and increase to 3.5 percent in 2011 and 4.1 percent manufacturing activity strengthened, and con‑ in 2012, in line with National Treasury projec‑ sumer spending consistently gained. tions. Unlike the majority of other developing Global growth is projected to remain countries, South Africa is still not operating strong until 2013 (table 1.1), shaking off the close to full capacity, so inflation pressures are mild slowdown in trade and industrial pro‑ less severe. duction in the first half of 2011 caused by the The recovery in the real economy has been events in the Middle East and North Africa broad‑based but uneven. Output in the second‑ and the earthquake and tsunami in Japan. ary sector remains below precrisis levels, while Developing countries would see growth eas‑ both the primary and tertiary sectors are 5 ing to a still strong 6.3 percent from 2011 percent above their precrisis levels (figure 1.1). onward, in line with their long‑term potential Of the major sectors, only manufacturing and growth. Policy tightening and events in Japan, agriculture are yet to recover to their precri‑ among other facts, would reduce growth in sis levels, despite manufacturing’s growing by high income countries to 2.2 percent in 2011 a robust 14.5 percent in the 2011 Q1. General before a broader recovery begins in 2012. government services and construction have Downside risks remain in high food prices, been the strongest performers since 2008 Q3, possible oil price spikes, and lingering post‑ and the only sectors not to have experienced crisis difficulties in high income countries, negative growth in any of the interim quarters. including high unemployment rates, fiscal Mining, having benefited from higher metal 1 SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth table The global outlook, 2009–13 1.1 percent change from previous year, unless otherwise indicated indicator 2009 2010a 2011b 2012b 2013b world trade volume (goods and nonfactor services) –11.0 11.5 8.0 7.7 7.7 commodity prices nonoil commodities –24.1 27.6 20.7 –12.0 9.4 oil price (percent change) –36.3 28.0 35.6 –4.8 –3.3 oil price ($ per barrel) 61.8 79.0 107.2 102.1 98.7 2 real gdp growth world –2.2 3.8 3.2 3.6 3.6 high income –3.4 2.7 2.2 2.7 2.6 Euro area –4.1 1.7 1.7 1.8 1.9 Japan –6.3 4.0 0.1 2.6 2.0 United states –2.6 2.8 2.6 2.9 2.7 developing countries 1.9 7.3 6.3 6.2 6.3 East asia and pacific 7.4 9.6 8.5 8.1 8.2 china 9.1 10.3 9.3 8.7 8.8 Europe and central asia –6.4 5.2 4.7 4.4 4.6 russian Federation –7.8 4.0 4.4 4.0 4.1 latin america and the caribbean –2.1 6.0 4.5 4.1 4.0 Brazil –0.7 7.5 4.2 4.1 3.8 middle East and north africa 2.8 3.1 1.9 3.5 4.0 Egypt 4.7 5.2 1.0 3.5 5.0 south asia 6.2 9.3 7.5 7.7 7.9 india 9.1 8.8 8.0 8.4 8.5 sub-saharan africa 2.0 4.8 5.1 5.7 5.7 south africa –1.8 2.8 3.5 4.1 4.4 developing countries excluding china and india –1.8 5.5 4.5 4.5 4.6 a. Estimated. b. Forecast. source: world Bank 2011d. table GDP growth by main sectors (value added), 2007–11 Q1 1.2 sector 2007 2008 2009 2010 2010 Q1 2010 Q2 2010 Q3 2010 Q4 2011 Q1 primary 0.6 –0.1 –3.9 4.3 14.7 –15.0 28.3 15.8 0.5 agriculture 2.7 16.1 –3.0 0.9 4.9 13.6 16.3 12.5 –2.6 mining 0.0 –5.6 –4.2 5.8 18.7 –24.5 33.7 17.1 1.8 secondary 6.2 3.0 –7.1 4.1 6.9 4.3 –3.8 3.6 11.1 manufacturing 5.2 2.6 –10.4 5.0 8.3 5.7 –4.9 4.1 14.5 Electricity 3.4 –3.1 –1.6 2.0 4.9 –1.7 –2.2 5.6 3.3 construction 15.0 9.5 7.4 1.5 1.3 1.0 0.8 0.2 0.0 tertiary 6.1 4.5 0.7 2.2 2.6 4.6 2.0 3.5 3.7 wholesale and retail 5.3 0.8 –2.5 2.2 3.1 6.0 3.3 3.5 4.4 transport 6.6 3.4 0.6 2.9 2.4 4.5 3.0 4.2 3.6 Finance 7.9 7.3 0.9 1.9 3.2 4.0 1.4 1.7 4.8 government services 4.0 4.5 4.1 3.0 1.2 4.6 0.5 5.7 1.8 personal services 5.6 3.9 –0.3 0.6 3.5 3.6 3.1 3.3 2.7 gdp growth 5.6 3.6 –1.7 2.8 4.8 2.8 2.7 4.5 4.8 source: south african reserve Bank. Box Commodity market developments 1.1 commodity prices, up significantly since early 2009, are expected to remain high in 2011 before weakening moderately over the rest of the forecast horizon (2011–13). robust demand from china and its above average metal intensity in production (its gdp is 13 percent of the world’s total and it consumes 40 percent of global metal supplies), recovery in industrial production in high income countries, and some supply constraints have contributed to the strong recovery in metal and mineral prices. most metal prices have at least doubled since their recession lows. in the case of platinum and gold, of which south africa is a major pro- ducer, the price of platinum has doubled and that of gold increased more than 70 percent since early 2009, picking up south africa’s terms of trade. metal prices are expected to rise 17 percent in 2011 before beginning to decline in 2012 as more produc- tion comes on stream. 3 Box figure 1 Commodity price movements, 2000–11 Agriculture Metals and minerals Base metals Energy 400 Commodity prices (index, 2000 = 100) 300 200 100 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 source: world Bank development Economics prospect group and south african reserve Bank. Box figure 2 South Africa’s terms-of-trade, 2000–11 Excluding gold Including gold 130 Terms of trade (index, 2005 = 100) 120 110 100 90 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 source: world Bank development Economics prospect group and south african reserve Bank. despite ample supplies and spare capacity, crude oil prices began rising in 2010 Q4 given strong demand and expectations of future supply tightness. they rose even more toward the end of the year and into 2011 as political turmoil in the middle East and north africa disrupted oil deliveries (notably light sweet crude from libya) and fears of further possible disruptions in the region took hold. Estimates suggest that the turmoil in north africa has added about $15 to the price of oil. assuming no further supply disruptions and a gradual reduction in uncertainty about the political situation in the middle East and north africa, oil prices are expected to ease in the second half of 2011, averaging $107/bbl for the year, before declining further in 2012 and 2013 toward $80/bbl (in 2011 dollars), consistent with long-term demand and supply. By early 2011 the food price index had reached its 2008 peak. maize and wheat prices exceeded their 2008 highs by 2 and 4 percent, respectively, given a fall in wheat production from central Europe due to the heat wave and a disappointing U.s. maize crop. But supplies for rice were abundant, contributing to a subdued price increase. low stocks, delayed plantings, and higher oil prices are expected to sustain wheat and maize prices above their 2010 levels this year, with projected increases of 12–13 per- cent. rice prices are forecast to remain unchanged. SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth output in the secondary sector is below precrisis levels, while output in Figure the primary and tertiary sectors are 5 percent above precrisis levels 1.1 110 Output (index, 2008 Q3 = 100) 105 100 4 95 90 r e ng r er g n r il t ce t s P en or ice to to to rin ta ur tio D at an ini re G ec ec ec sp m ult rv tu uc ,w Fin M rn an ys ys ys d ac se ric tr as an ve Tr uf ns ar ar iar Ag al ,g go an ale im nd Co on ity rt M Te al Pr les co rs ric er Pe The recovery in the Se ho ct en Ele W G real economy has source: south african reserve Bank and world Bank staff calculations. been broad-based prices and improved terms‑of‑trade (box 1.1), an impressive 9.5 percent growth in 2011 Q1, lost momentum in 2011 Q1 on account of wet added in large part by purchases of military but uneven weather and technical problems affecting aircraft (South African Reserve Bank 2011). gold production.1 Crop production was also Gross fixed capital formation gained fur‑ adversely affected by heavy rainfall and spo‑ ther traction in 2011 Q1, with an annualized radic flooding, leading to a small decline after increase of 3.1 percent. Improved business con‑ three quarters of double‑digit growth. fidence lifted private fixed capital formation Gross domestic expenditure continued its a notch, as it, too, recorded positive but mod‑ strong run with 8.3 percent growth in 2011 Q1 est growth for the fourth consecutive quarter. (table 1.3). Looser monetary conditions, robust Recovery in this component is still partial, how‑ wage increases, and rising financial asset and ever, with its level in 2011 Q1 remaining 13 per‑ real estate valuations helped strengthen con‑ cent below the precrisis 2008 Q3 level. Fixed sumer confidence and provide momentum investment by general government, by contrast, to household consumption, which has grown has fallen nine quarters in a row; in 2011 Q1 it seven quarters in a row. Government consump‑ stood 17 percent below precrisis levels. Robust tion has continued to hold firm, recording investment by state‑owned enterprises, with table Gross domestic expenditure growth by component, 2007–11 Q1 1.3 component 2007 2008 2009 2010 2010 Q1 2010 Q2 2010 Q3 2010 Q4 2011 Q1 total final consumption household 5.5 2.2 –2.0 4.4 5.5 4.4 5.7 4.8 5.2 durables 1.9 –9.4 –9.6 24.0 39.2 45.4 13.4 5.0 21.5 semidurables 11.3 4.2 –1.8 6.5 30.2 9.8 –4.8 4.6 8.6 nondurables 5.0 0.8 –2.7 2.1 12.1 –2.3 2.9 0.2 3.4 government 4.1 4.7 4.8 4.6 7.1 7.1 –0.8 3.9 9.5 gross fixed capital formation 14.0 14.1 –2.2 –3.7 –2.8 1.2 1.0 1.5 3.1 private 8.9 9.2 –8.9 –4.4 –2.9 2.2 2.0 1.6 2.7 government 22.2 16.1 –4.0 –10.9 –10.3 –5.3 –3.0 –1.9 –0.5 public corporations 34.8 36.2 26.1 3.5 2.6 2.9 0.7 3.3 6.6 change in inventories (billions of rands) 19.8 –12.4 –34.5 –3.8 –7.9 –7.6 –0.9 1.1 9.3 gross domestic expenditure 6.3 3.4 –1.7 4.2 10.9 3.2 6.6 2.4 8.3 source: south african reserve Bank. emphasis on infrastructure in the utilities and structural issues that keep unemployment high transport sectors, has helped buffer the weaker in South Africa3 as well as the lagged response performances of the private and government of jobs creation to economic recovery in gen‑ sectors, with its 2011 Q1 level 25 percent above eral.4 Young people and new entrants (with a the precrisis level. big overlap between the two groups) are the Despite the improved pace, fixed investment most vulnerable on the job market (figure 1.3). has continued to lag the recovery in total out‑ Although unemployment increased across put. As a result, the ratio of fixed investment to the world after the global crisis struck, the mag‑ GDP slipped to 19.6 percent in 2010 and fur‑ nitude of the increase has been far higher in 5 ther to 19.0 percent in 2011 Q1, down from 23.6 South Africa than most other countries (table percent in 2008. As discussed in section 2, to 1.4). 5 Indeed, the unemployment rate fell in position South Africa for a higher GDP growth each of the other BRICS in 2010, after having trajectory it will be important to ensure signifi‑ risen in 2009, while it continued to increase in cantly faster increases in fixed investment. South Africa. Export volumes have supported the recovery Despite the slack in labor markets, nominal but remain below precrisis levels. A beneficial wages in nonagricultural formal sector have Despite the improved factor has been the growing trade with China, risen well above inflation and productivity in pace, fixed investment which tripled as a share of South Africa’s recent years. Real wage growth far outstripped total trade, from less than 5 percent in 2000 total factor productivity growth between 2005 has continued to lag to roughly 15 percent in 2010. South Africa’s and 2010 in the manufacturing sector, which recent entry in the BRICS (Brazil, Russian would have adversely affected employment in the recovery in output Federation, India, China, and South Africa) the sector (figure 1.4). Between 2004 and 2010 bloc offers an opportunity to further develop productivity rose 18 percent while the real pro‑ its economic relationship with major emerging duction wage rose 55 percent. markets (box 1.2). Fiscal developments: Labor market trends: unemployment an overriding concern unemployment remains stubbornly high The 2011 budget is anchored in a stronger eco‑ Unemployment has been worryingly unre‑ nomic base, with revenue projections much sponsive to the economic recovery (figure 1.2). firmer than a year ago. The nation’s overrid‑ Seven quarters after the end of recession that ing concern with the high unemployment and cost 869,000 jobs, 2 the unemployment rate widespread exclusion understandingly guides stood at 25.0 percent (33.4 percent including expenditure priorities. In addition to the youth discouraged workers) in 2011 Q1, only fraction‑ wage subsidy, a jobs fund and extension of tax ally short of its postcrisis peak of 25.3 percent credits for job‑creating new investments figure in 2010 Q3. This lack of response reflects the prominently among the proposals to stimulate Figure unemployment has been worryingly unresponsive to the economic recovery 1.2 Youth (ages 15–24) unemployment rate Youth (ages 15–24) absoption rate 60 20 55 Percent Percent 50 15 45 40 10 source: statistics south africa and world Bank staff calculations. SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth South Africa joins the BRICS (Brazil, Russian Box Federation, India, China, and South Africa) 1.2 south africa’s entry (formally on april 13, 2011) into the high-profile Brics (with Brazil, russia, india, and china) heralds its eco- nomic prowess and perhaps even more its recognized strategic importance as the gateway to the african continent. the acronym was coined in 2001 by the goldman sachs chairman Jim o’neill to describe the block of non–organisation for Economic co-operation and development emerging countries (thus excluding mexico and the republic of Korea) that would dominate the global economic landscape by 2050. the Brics cover more than 25 percent of the world’s land area and have more than 40 percent of its population. in addition to being large, they are defined by their dynamism and fast-paced growth and transformations. over 2000–09 china, india, and 6 russia had annual gdp growth above 5 percent, while south africa averaged 3.6 percent and Brazil 3.3 percent. while originally clubbed together on the basis of their economic clout and large size, the group is morphing into a geopolitical force. What does South Africa bring to the table? it is the largest economy on the african continent, which has a total population of 1 billion. it is also the continent’s most sophis- ticated economy, with highly advanced financial markets well integrated with the rest of the world. it has a stable political envi- ronment and a proven track record in macroeconomic management. Ernst & Young, the consultancy, showed in its 2011 africa attractiveness survey that investors perceive south africa as the most attractive african country to do business. the 2010 United nations conference on trade and development world investment report ranked south africa in the top 20 priority economies for foreign direct investment inflows in the world. What can South Africa gain? the Brics association should strengthen south africa’s economic and political clout in the global arena. the Brics as a bloc are demanding a more meaningful voice at multilateral institutions such as the United nations, the world Bank, and the international monetary Fund, where they are seeking major reforms; an agenda dear to south africa. there are significant economic ties among the group, with the potential for much more. china is south africa’s largest trading partner country, having overtaken the United states in 2010, and south africa’s trade with asia is higher than with Europe. south africa, like Brazil and russia, has a compara- tive advantage in natural resources, while china and india have huge unmet needs: the two sides are natural trading partners in that respect. the association also has the potential to make south africa more attractive for foreign investors, especially in green- field investments, something lacking to date. on the policy front, the country has the opportunity to tap into Brics’ experience in tackling social challenges, which, in many respects, are broadly similar. Key indicators for the BRICS (Brazil, Russian Federation, India, China, and South Africa), 2009 russian south indicator Brazil Federation india china africa population (million) 194 142 1,155 1,331 49 gross domestic product (gdp) in purchasing power parity (ppp) terms 2,017 2,690 3,778 9,091 507 gdp per capita, ppp (current international $) 10,412 18,963 3,270 6,828 10,278 land area (million km2) 8.5 16.4 3 9.3 1.2 Urban population (percent of total) 86 73 30 44 61 Under-five mortality rate (per 1,000) 21 12 66 19 62 gross savings rate (percent of gdp) 14.6 22.7 33.6 53.6 15.4 ores and metal ores exports (percent of gdp) 1.7 5.7 6.2 1.2 29.3 ores and metal ores imports (percent of gdp) 2.9 1.6 5.6 13.5 1.3 portfolio inflows, net ($ billions) 37.1 3.4 21.1 28.2 9.4 agriculture (percent of gdp) 6.1 4.7 17.1 10.3 3 manufacturing (percent of gdp) 14.8 15 15.9 33.9 15.1 carbon dioxide emissions (kg per ppp $ of gdp)a 0.2 0.6 0.5 0.9 0.9 Energy use (kg of oil equivalent per capita) 1,239 4,730 529 1,484 2,784 gdp per unit of energy use (ppp $ per kg of oil equivalent)a 7.9 3.6 5.4 3.7 3.6 a. data are for 2007. source: world Bank. demand for labor. There is renewed emphasis Treasury appropriately has kept a watchful eye on further education and skills development to on longer term sustainability even as it carries boost employment generation from the labor out the announced short‑term stimulus. This supply side. Social grants continue to figure is important given the reliance on domestic prominently in the budget, understandably markets for deficit financing, for which retain‑ given the massive exclusion levels in the econ‑ ing the investment‑grade ratings is critical. omy. The public sector (including public enter‑ The sound fiscal position has been enabled by prises) will remain heavily engaged in removing several years of budgetary discipline leading infrastructural bottlenecks to growth. Renew‑ up to the global crisis that yielded low levels 7 able energy, environmental protection, and of public debt. The country’s deep and liquid “green” economy initiatives also find emphasis domestic capital markets and ready access to in the budget, with South Africa hosting the international borrowing have provided added COP17 event toward the end of the year. comfort. With the rising fiscal deficits and Despite increased public sector borrowing, public sector borrowing requirement (table the fiscal framework remains well within the 1.5), projected to moderate over the Medium bounds of sustainability (figure 1.5). National Term Expenditure Framework period, the net Despite increased public sector borrowing, the Figure Young people and new entrants are the most vulnerable in the job market fiscal framework remains 1.3 Unemployed (total) Re-entrants New entrants Job leavers Job losers Other Long-term Short-term well within the bounds 5 of sustainability Unemployed workers (millions) 4 3 2 1 0 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 2008 2009 2010 2011 source: statistics south africa and world Bank staff calculations. In manufacturing, growth in real wages far outstripped growth Figure in total factor productivity between 2005 and 2010 1.4 Real wages Total factor productivity 15 10 Growth (percent) 5 0 –5 –10 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 source: Quantec database and world Bank staff calculations. SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth Growth and unemployment in the BRICS (Brazil, Russian Federation, India, table China, and South Africa) and selected oECD economies, 2008 –10 (percent) 1.4 2008 2009 2010a Economy gdp growth Unemployment gdp growth Unemployment gdp growth Unemployment Brazil 5.2 7.9 –0.7 8.1 7.5 6.7 russian Federation 5.2 6.4 –7.8 8.4 4.0 7.5 india 5.1 10.4 9.1 10.7 9.1 10.8 china 9.6 5.9 9.2 6.3 10.3 6.1 south africa 3.6 22.9 –1.7 24.0 2.8 24.9 8 australia 2.6 4.3 1.3 5.6 2.7 5.2 chile 3.7 7.8 –1.7 9.6 5.2 7.1 germany 0.7 7.3 –4.7 7.5 3.5 6.9 Japan –1.2 4.0 –6.3 5.1 4.0 4.0 United states 0.0 5.8 –2.6 9.3 2.9 9.6 total oEcd 0.2 6.1 –3.6 8.2 2.8 8.4 a. preliminary. source: Economist intelligence Unit. Consolidated government fiscal framework, 2007–13 table (percent of GDP unless otherwise indicated) 1.5 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 revised outcome estimate Forecast revenue 30.1 29.5 27.2 28.3 28.3 28.4 28.8 Expenditure 28.5 30.7 33.8 33.6 33.6 33.2 32.6 Budget balance 1.7 –1.2 –6.6 –5.3 –5.3 –4.8 –3.8 total net government debt 23.2 22.7 27.6 30.8 34.3 37.5 39.3 net domestic government debt 18.6 18.5 24.5 29.4 33.3 35.9 37.1 net foreign government debt 4.6 4.2 3.0 0.14 1.0 1.6 2.1 interest cost 2.5 2.4 2.3 2.5 2.6 2.8 2.9 public sector borrowing requirement 0.2 4.3 8.9 10.5 9.5 8.1 6.3 southern african customs Union transfers (millions of rands) 24,713 28,921 27,915 14,991 21,763 32,432 35,997 source: national treasury of the republic of south africa 2011. Despite increased public sector borrowing, fiscal Figure sustainability has not yet emerged as a concern 1.5 Public sector borrowing requirement Budget de cit Primary de cit Revenue Expenditure 15 40 10 30 Percent of GDP Percent of GDP 5 20 0 10 –5 0 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13a 2013/14a a. Forecast. note: a negative deficit indicates a surplus. source: national treasury of the republic of south africa 2011b and world Bank staff calculations. government debt stock to GDP ratio would External sector: Foreign trade recovers, increase to 39 percent by the end of the period as portfolio inflows slow down and stabilize at roughly 40 percent of GDP in Benefiting from the growing global demands FY2015/16 (50 percent including contingent for imports, South Africa’s exports have picked liabilities), well below the current global norms. up briskly. Merchandise exports in 2011 Q1 Moreover, the debt‑servicing cost to the budget were 20 percent higher (seasonally adjusted) would remain under 3 percent in the medium than in 2010 Q1 and 31 percent higher than term, quite manageable. the low point in 2009 Q2 (figure 1.6). Net exports of gold showed similar increases over 9 Banking sector developments: SME the same periods. Imports have also recovered, financing suffers under the crisis driven by the strong recovery in household The South African banking sector has held up consumption of durables and the pickup in well, having entered from a position of strength fixed investment and, more recently, inventory with little foreign currency exposure and rela‑ accumulation. Dividend and interest payments tively low nonperforming loans. A recent World to nonresident investors have risen, with recov‑ Bank study nevertheless finds that the reces‑ ering profits in domestic firms and increased Benefiting from the sion impaired the access of small and medium holdings of South African debt by foreigners. growing global demands enterprises (SMEs) to finance, of particular The current account deficit rebounded from concern given the many contributions SMEs an exceptionally low 1 percent of GDP in 2010 for imports, South (an estimated 6 million in South Africa) can Q4 to 3.1 percent in 2011 Q1, the average since make to employment and economic growth.6 2009 Q2 (figure 1.7). Africa’s exports have Credit conditions tightened considerably for Capital flows to developing countries slowed picked up briskly SMEs during the economic downturn as the considerably in 2011 Q1, in response to the banks became more risk conscious. From the mounting uncertainty surrounding the global banks’ point of view, SME lending is an attrac‑ economic recovery, sovereign debt crisis in tive proposition, because they see SME bank‑ Europe, high commodity prices, and concerns ing not just as a feeder for future business but about overheating in some large emerging mar‑ also as an attractive and profitable business kets. Equity placements dried up, while bond in its own right. But macroeconomic factors issues by emerging market economies started related to the economic downturn seem to have the year strong thanks to attractive yields but become more constraining. This also implies then slowed considerably as uncertainty pre‑ that as the economy continues to recover and vailed.7 With similar concerns, global portfo‑ the macroeconomic factors improve, lending lio investors also offloaded their positions in to SMEs can be expected to pick up again. emerging market stock and bond markets. having benefitted from improved global import demand, Figure South Africa’s exports have picked up briskly 1.6 Merchandise exports Merchandise imports 800 600 Rand (billions) 400 200 0 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 2007 2008 2009 2010 2011 source: south african reserve Bank and world Bank staff calculations. SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth The current account deficit rebounded from an exceptionally low Figure 1 percent of GDP in 2010 Q4 to 3.1 percent in 2011 Q1 1.7 10 Ratio of current account de cit to GDP 8 6 4 10 2 0 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 2007 2008 2009 2010 2011 source: south african reserve Bank and staff calculations. South Africa’s Foreign portfolio investments in South direct investment remain negligible, a problem medium-term growth Africa followed a similar pattern. Net portfolio that precedes the global crisis (box  1.3) and prospects point to flows turned sharply negative in 2010 Q4 as for‑ persists somewhat paradoxically in light of the eigners reduced their positions in South Africa solid real returns on long‑term investment in a strengthening while South Africans increased their portfolio South Africa (section 2). recovery, with GDP positions abroad (figure 1.8). In 2011 Q1 large issues of foreign currency–denominated bonds Economic outlook for South Africa growth projected at by the government and parastatals offset net South Africa’s medium‑term growth prospects 3.5 percent in 2011 sales of bonds and stocks on the locals markets point to a strengthening recovery. We project by foreigners (figure 1.9). Net portfolio invest‑ GDP growth to be 3.5 percent in 2011, slightly ment was still negative as outward portfolio above the 3.4 percent projected in the 2011/12 investment by South Africans increased signifi‑ Budget released in February. Our baseline pro‑ cantly. At the same time, South African banks jections of 4.1 percent for 2012 and 4.4 percent repatriated significant amounts of foreign cur‑ for 2013 are the same as those of National Trea‑ rency deposits, reflected in the positive “other sury, and consistent with the precrisis average investment” in 2011 Q1 (South African Reserve growth rates (4.2 percent between 2000 and Bank 2011) (figure 1.8). Net inflows of foreign 2008). The long‑run potential growth rate Net portfolio flows turned sharply negative in the last quarter of 2010 as foreigners reduced their positions in South Africa while Figure South Africans increased their portfolio positions abroad 1.8 Portfolio investment Foreign direct investment Other investment Total capital ows 15 10 5 $ billions 0 –5 –10 –15 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 2005 2006 2007 2008 2009 2010 2011 source: south african reserve Bank and staff calculations. Foreigners reduced their net positions in domestic stock Figure and bond markets in the first quarter of 2011 1.9 Bonds Equities Total 3 2 1 $ billions 0 11 –1 –2 –3 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 2005 2006 2007 2008 2009 2010 2011 source: south african reserve Bank and staff calculations. With considerable strengthening of the under the current policy environment is esti‑ high current account deficits, financed mostly mated at 3.5 percent. through volatile portfolio flows, will once again economic recovery, Consumer spending, which has been the become the biggest cause for macroeconomic main driver of growth in the postcrisis period concern over the medium term. the focus shifts back will continue to benefit from wage increases Unlike other middle income countries fac‑ to the challenge of negotiated in 2010 and expansion in domestic ing capacity constraints, South Africa’s output credit. With global recovery taking hold, the remains below potential, taking pressure off raising potential GDP uncertainty that affected private investment inflation. Even so, headline Consumer Price growth and making it spending is expected to abate, allowing busi‑ Index inflation is forecast to accelerate in the ness spending to increase, and resume its posi‑ following years close to the upper bound of the much more inclusive tive contribution to growth. The increase in 3–6 percent inflation targeting band, as the spending on consumer durables should also economy recovers and pressures emerge from support investment demand. As a result, gross the cost‑push factors, including food and oil. fixed capital formation is forecast to grow after With considerable strengthening of the eco‑ declining for two consecutive years. As busi‑ nomic recovery and GDP projected to reach nesses demand more labor, employment should its potential by 2014, 8 the focus shifts back to rise and further strengthen consumer spend‑ the longer term challenge of raising potential ing. Strong government spending, as part of the GDP growth to 6–7 percent—and making it countercyclical fiscal policy, will boost growth much more inclusive to tackle the extremely in 2011 and 2012 but is likely to wane thereafter, high unemployment. The current low sav‑ as projected under the latest Budget document. ings and investment would certainly need to The South African economy is well inte‑ be addressed along with emphasis on faster grated in the global economy, and the ongo‑ productivity growth and more employment‑ ing global recovery should continue to support intensive production. The issues of savings and exports. But strengthening domestic demand investment are the focus of the analysis and dis‑ will increase imports and moderate the trade’s cussion in section 2. net contribution to growth. The current account deficit is expected to Risks to the outlook widen, from 2.8 percent of GDP in 2010 to 3.2 percent in 2011 and further to 4.5 percent in Global risks 2013, as domestic demand picks up and divi‑ First is the possibility of a much more pro‑ dend payments on external borrowing (on nounced global slowdown, problematic for the rise in 2010) grow. In light of South Afri‑ South Africa given its links with the global ca’s low national savings, the reemergence of economy. An escalation of the Euro zone SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth Box Low foreign direct investment inflows to South Africa 1.3 with a shortfall in national savings to cover its domestic investment, south africa has had to rely on fairly large current account deficits. Financing them requires foreign investment, which, by and large, has been adequate though heavily biased toward portfolio flows rather than the more reliable foreign direct investment (Fdi). Between 2002 and 2008 portfolio investment by foreigners aver- aged 2.4 percent of gdp, compared with Fdi of only 1.5 percent (box figures 1 and  2). this contrasts with the other upper middle income countries, where Fdi inflows over 2002–08 were more than 5 percent of gdp, almost four times the portfolio inflows. Box figure 1 Compositions of net capital flows to South Africa, 2002–11 12 Foreign direct investment in ows Portfolio in ows Percent of GDP 10 5 0 –5 2002 2003 2004 2005 2006 2007 2008 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 2009 2010 2011 source: south african reserve Bank, world Bank world development indicators database, and world Bank staff calculations. Box figure 2 Capital flow composition comparisons, 2002–08 (average) Foreign direct investment in ows Portfolio in ows 6 5 Percent of GDP 4 3 2 1 0 South China India Russian Brazil Upper middle income Africa Federation countries source: south african reserve Bank, world Bank world development indicators database, and world Bank staff calculations. the composition of Fdi and portfolio flows matters for the recipient country, especially one as savings-starved as south africa. Fdi is far more reliable as a source of external financing. as seen in the global crisis, in times of financial distress, sharp swings in market sentiment can suddenly reverse portfolio flows, with potentially damaging consequences for income growth (levchenko and mauro 2007; tong and wei 2011). Even more beneficial with Fdi are the accompanying spillovers—such as new technology, advanced managerial skills, and links with global markets and production networks—that, under appropriate conditions, can enhance productivity and lead to higher growth (Borenstein and others 1998). the benefit depends on the recipient country’s absorptive capacities, which can include a minimum threshold stock of human capital (Borenstein and others 1998; Xu 2000), financial market development (alfaro and others 2004, 2006), infrastructure, and market institutions. what explains the composition of capital flows to south africa? the macro/financial risks adversely affect both Fdi and portfo- lio flows, while secure property rights exert a positive effect as well as biasing the composition of capital inflows toward Fdi (gwenhamo and Fedderke 2010). overall, policies to ensure that macro/financial risks remain low and make property rights more secure would aid in south africa’s quest to get higher and more broad-based Fdi inflows, targeting not just its natural-resource and capital-intensive sectors but also greenfield labor-intensive ones. sovereign debt crisis would severely affect the Last, South Africa, like other large emerging ongoing recovery in Europe, South Africa’s market economies, attracted large “hot money” largest trading partner and the main destina‑ capital flows in 2010. The lurking risk is a sud‑ tion for its manufacturing exports. Even a lim‑ den stop in the inherently volatile portfolio ited default by some European countries could flows. Since much of these flows went to South increase risk aversion on the global markets, African bonds, that would raise the borrowing raising the cost of capital and possibly revers‑ cost for both the corporate and the public sec‑ ing capital flows to developing countries. The tors. And the rand, which has held strong and impact of the earthquake and tsunami in Japan contained the cost of imported inflation (par‑ 13 adds an additional short‑term risk. ticularly internationally priced food and fuel), Binding capacity constraints and growing could weaken and add to inflationary pressure, concerns about inflation‑wage spirals because triggering higher interest rates and dampening of rising fuel and food prices could lead to fur‑ domestic consumption. ther monetary tightening and a growth slow‑ down in some emerging markets. The effects of Domestic risks this playing out in Asia, particularly in China, Business confidence remains low, so invest‑ The possibility of a much would pose downside risks for South Africa as ment decisions by the private sector have been more pronounced global its trade continues to shift toward that conti‑ muted despite low interest rates and a strong nent. With China’s strong demand for com‑ rand favoring the purchase of imported capi‑ slowdown is problematic modities, a cooling of the Chinese economy tal goods. The labor market situation is dis‑ beyond baseline projections, would dampen couraging, with hiring decisions still lagging for South Africa given commodity prices and reduce exports from the economic recovery. A fresh round of wage its links with the South Africa and investments in its mineral demands by the trade unions, thus far ranging sector. from 9 percent for public sector employees to global economy Another global risk for South Africa is higher up to 20  percent for other sectors, add com‑ oil prices. The dramatic political changes that plexity to the situation. have swept parts of the Middle East and North High household indebtedness also pres‑ Africa will likely have limited direct effects on ents a downside risk to the recovery, particu‑ global growth, given the small size of the econo‑ larly by subduing household consumption, a mies. But any contagion to a major oil producer major current driver of domestic absorption. could further escalate the fuel price hikes— Although household debt as a share of dispos‑ with damaging effect. Such a scenario could able income fell in 2011 Q1 to 76.8 percent shave up to half a percentage point from global from an average of 80.9 percent in 2009, it growth (assuming prices increase by an addi‑ remains high relative to the average indebted‑ tional $50/bbl from their March 2011 levels) in ness levels for 2000–04 (54.7 percent of dispos‑ 2011 and 1 point in 2012.9 able income). table Macroeconomic outlook, 2007–13 (percent change unless otherwise indicated) 1.6 actual Estimate Forecast indicator 2007 2008 2009 2010 2011 2012 2013 Final household consumption 5.5 2.2 –2.0 4.6 4.2 4.5 4.5 Final government consumption 4.1 4.7 4.8 4.6 4.3 3.7 3.7 gross fixed capital formation 14.0 14.1 –2.2 –3.6 4.8 4.0 5.2 gross domestic expenditure 6.3 3.4 –1.7 4.1 4.2 4.4 4.6 Exports 6.6 1.8 –19.5 5.3 5..0 6.0 6.0 imports 9.0 1.5 –17.4 10.4 9.0 7.0 7.0 real gdp 5.6 3.6 –1.7 2.8 3.5 4.1 4.4 headline consumer price index 6.1 9.9 7.1 4.3 4.9 5.2 5.5 current account balance (% of gdp) –7.0 –7.1 –4.1 –2.8 –3.2 –3.8 –4.5 source: national treasury of the republic of south africa 2011 and world Bank staff calculations. SECTIoN 2 Toward more inclusive growth: Focus on savings and investment The South African economy has done well The exact transition of countries to faster since turning the corner after the fall of apart‑ growth paths involves an element of mystery, The South African heid in 1994. Macroeconomic management has but one empirical regularity is clear: fast‑grow‑ been exemplary, with inflation nestled in the ing emerging market economies by and large economy has target 3–6 percent range and world‑class bud‑ exhibit high investment and savings rates.12 done well since getary and debt outcomes. GDP growth—albeit At around 16 percent, South Africa’s gross modest by global comparisons —has averaged a national savings as a share of gross national turning the corner credible 3.3 percent a year since 1994. And the income is relatively low, no doubt restraining after the fall of external trade and capital accounts have moved the investment rate, at a modest 19 percent. far deeper into the folds of global integration. How much higher the saving and investment apartheid in 1994, But this record has not been enough to rates will have to be to underpin 6–7 percent but this record has absorb the massive wave of new entrants into growth will be estimated below. the labor market since the mid‑1990s, yield‑ Higher savings and investment alone will not been enough ing unemployment rates persistently above 20 not suffice, however. Success in achieving the to absorb the percent and widespread exclusion. The Gini desired growth will depend on two additional coefficient, the most commonly used measure factors: the extent to which firms use more massive wave of of inequality, stood at 0.67 in 2008, the high‑ employment‑intensive (rather than capital‑ est in the world. Weak integration between the intensive) production methods, and the pace new entrants into formal and, in many parts, ultramodern econ‑ of technological innovation, or total factor the labor market omy and the “second” economy in the peri‑ productivity in technical jargon. Without con‑ urban and rural areas undermines economic comitant improvements in these two factors, efficiency and job creation. It also makes it the required rates of saving and investment will more difficult for the rising tide of economic be impossibly high. growth to lift the unemployed and marginal‑ There is a virtuous reinforcing circular‑ ized masses. ity among employment‑intensive production, The government has appropriately set its higher savings/investment, and higher produc‑ sights on much faster and much more inclusive tivity, with inclusive growth at the center of this economic growth—in the range of 7 percent to interaction. But what emerges today is a subop‑ match the employment challenge.10 The New timal equilibrium that can be broken only by a Growth Path envisions 5 million new jobs over massive productivity push (figure 2.1). the next decade to lower the unemployment rate to 15 percent.11 This part of the report, Savings, investment, and anchored in these vital national objectives, long-term growth focuses on savings and investment and their Real GDP per capita has grown at an unre‑ special role. markable pace over the last three decades, 15 SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth Toward a virtuous cycle and quality of the workforce) and technologi‑ Figure for inclusive growth cal improvements, investment inevitably faces 2.1 diminishing returns, and the payoff in faster Higher savings and growth tapers off.16 investment High rates of investment, in turn, require commensurate domestic savings. Although it Inclusive is not necessary that investment be financed growth Higher through domestic savings if the country has 16 Higher employment ready access to external financing, running productivity intensity of production large external imbalances has proven unsus‑ tainable for countries over long periods. The positive long‑run savings– investment and savings– growth relationships across countries the better performance in the 2000s notwith‑ have been established by a number of studies standing. GDP declined 15 percent between (figures 2.5 and 2.6).17 Consistent with this, South Africa’s modest 1980 and 1994 and then increased 30 percent South Africa’s own savings and investment through 2008. Per capita growth picked up rates have been highly correlated: the coeffi‑ record on savings from –1 percent during 1980–94 to a modest 1 cient of correlation between the two series over and investment needs percent in 1995–2003, and then rose to a more 1960–2010 is 0.72. impactful 3.7 percent during 2004–08 (figure South Africa’s modest record on savings and to improve by a big 2.2).13 Given the sharp decline in per capita investment no doubt needs to improve by a big margin to support the GDP between 1980 and 1994, it is striking that margin to support the desired GDP growth. South Africa did not grow faster after 1994 if The national savings rate has remained caught desired GDP growth only to make up lost ground. Overall, the aver‑ in the 15–17 percent range since the early age South African saw her income rise by less 1990s, averaging just under 16 percent over than 10 percent in real terms between 1980 2006–10. Gross investment, by contrast, has and 2008, when the average Chinese became seen a marked increase since the early 2000s, 11 times richer, the average Indian 3 times, and reaching an average of 20.5 percent over the average Chilean 2.5 times (figure 2.3). 2006–10. The divergence is reflected in a large Set against these trends, the growth tar‑ current account deficit, covered largely by get of 6–7 percent (5–6 percent per capita) international portfolio flows, which cannot be appears ambitious but by no means impossible. relied on over an extended period. It is also ambitious when viewed against other countries. Over 1980–2008, only three econo‑ Long-term savings trends mies achieved average GDP growth of 7 percent The national savings rate has been on the or more; China, Singapore, and neighboring decline since peaking at 35 percent in 1980 on Botswana, which relied heavily on its diamond the back of high gold prices that boosted cor‑ industry.14 Even dropping to 6 percent growth porate profits (figure 2.7). The savings rate fell nets only nine economies. to a trough of about 15 percent during 2005– By and large the nine high‑growth coun‑ 07, before recovering a bit as corporates began tries (with 6 percent or higher growth) enjoyed saving due to the uncertainty in the global cri‑ savings and investment rates above 25 percent sis and as banks became more conservative in (table 2.1).15 The Commission on Growth and lending. Underlying this is a steady decline in Development (2008) concluded that an invest‑ nongovernment (households plus public and ment rate of 25 percent is the minimum to private corporations) savings,18 which fell from ensure sustained long‑term growth. Influential an all‑time high of 29 percent in 1980 to just studies have confirmed that investment deter‑ 11 percent in 2007 (box 2.1). General govern‑ mines how fast countries grow (figure 2.4). ment savings declined sharply in the period They also caution that investment alone is not leading up to 1994, stayed negative until 2000, enough over long periods. Without concomi‑ and then recovered on the back of fiscal con‑ tant increases in human capital (the quantity solidation until 2008. They then fell back into Real GDP per capita in South Africa has grown at an Figure unremarkable pace over the last three decades 2.2 GDP growth GDP per capita growth GDP per capita 7 120 5 100 Index (1980 = 100) 3 80 Percent 1 60 17 –1 40 –3 20 –5 0 1980 1985 1990 1995 2000 2005 2010 source: statistics south africa, world Bank world development indicators data, and world Bank staff calculations. over 1980 –2008, the average South African’s real income increased less Figure than 10 percent, while the average Chinese became 11 times richer 2.3 1,100 Peak value of 1,090 GDP per capita (purchasing power parity, 300 2005 international US$) 200 100 0 China India Malaysia Chile Turkey OECD UMIC Argentina Mexico Brazil South Africa source: statistics south africa, world Bank world development indicators data, and world Bank staff calculations. Statistics for countries with average annual GDP growth table of at least 6 percent a year, 1980 –2008 2.1 taiwan, Botswana china india Korea, rep. lao pdr malaysia singapore china vietnam median gdp growth (percent) 7.4 10.1 6.0 6.6 6.0 6.2 7.1 6.2 6.8 6.6 gdp per capita growth (percent) 4.9 9.1 4.2 5.8 3.5 3.6 4.4 5.1 5.0 4.9 gross domestic savings (percent of gdp) 37.7 40.4 23.4 33.0 12.9 37.7 45.2 28.7 21.0 33.0 gross domestic investment (percent of gdp) 29.1 38.3 25.2 32.0 23.3 29.1 34.0 22.9 26.8 29.1 note: countries that attained average gdp growth of at least 6 percent over 1980–2008. Excludes oil-rich oman and countries with population below 1 million. source: statistics south africa and world Bank world development indicators data. negative territory in 2009 and 2010 as the gov‑ 1980s South Africa’s average national savings ernment embarked on countercyclical policies. rate was third after China’s and Malaysia’s (fig‑ South Africa’s savings rates are lower than ure 2.8), but it fell sharply in the 1990s when those in other emerging economies. In the many comparators saw theirs increase, dipping SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth Figure The relationship between investment and growth is positive . . . 2.4 10 China 8 GDP per capita growth (percent) 6 Korea, Rep. y = 0.1942x – 0.0255 Indonesia Singapore Chile India Malaysia R2 = 0.3568 4 Turkey Argentina 2 18 0 South Brazil Mexico Africa –2 –4 0 10 20 30 40 50 Investment rate (percent of GDP) source: hevia, ikeda, and loayza (2010), based on data from statistics south africa and world Bank world development indicators. Figure . . . as is the relationship between saving and investment . . . 2.5 10 China 8 GDP per capita growth (percent) y = 0.131x – 0.008 6 Korea, Rep. R2 = 0.3492 Chile India Indonesia 4 Turkey Malaysia Singapore 2 Mexico 0 South Argentina Brazil Africa –2 –4 0 10 20 30 40 50 National saving rate (percent of GDP) source: hevia, ikeda, and loayza (2010), based on data from statistics south africa and world Bank world development indicators. Figure . . . and the relationship between saving and growth 2.6 50 Investment rate (percent of GDP) 40 China Malaysia Korea, Rep. Singapore 30 Indonesia Chile India y = 0.5623x + 0.1123 Turkey R2 = 0.68 20 Brazil Mexico South Argentina Africa 10 0 0 10 20 30 40 50 National saving rate (percent of GDP) source: hevia, ikeda, and loayza (2010), based on data from statistics south africa and world Bank world development indicators. Box South Africa’s nongovernment savings: What explains the declining trends? 2.1 the weak link on the savings side appears to be households. household consumption as a share of disposable income has been high and household indebtedness has risen sharply in the 2000s, which contributes to the low savings levels. But is a rising pro- pensity to consume enabled by easier access to consumer and mortgage credit to blame for the decline in household savings, as is widely believed? the evidence suggests otherwise. it appears to be adverse declining trend in the household disposable income to gdp ratio that was the driving force behind the household saving patterns; household consumption, by contrast, stayed within a narrow range between 1990 and 2008, before falling sharply from 2008 onward (box figure 1). at the heart of it is the issue of the country’s exceptionally high unemployment rates that prevent more robust increases in household incomes, thus keeping house- hold savings low in general, while also preventing almost any saving among the multitudes without a job. macroeconomic stability, 19 growing urbanization, and rising government savings have also likely played a role, as discussed later under empirical results. Box figure 1 household incomes, consumption, and savings, 1990–2010 Ratio of nal household consumption to GDP Ratio of household gross disposable income to GDP Ratio of household saving to GDP 64 5 63 4 62 3 61 2 60 1 59 0 58 –1 1990 1995 2000 2005 2010 source: south african reserve Bank data and world Bank staff calculations. Box figure 2 Corporate saving statistics in South Africa, 1960–2010 Gross corporate saving Ratio of corporate retained earning to gross Ratio of dividend payments to national disposable income Depreciation gross national disposable income 25 40 20 35 15 30 Percent 10 25 5 20 0 15 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 note: dividend payments are proxied as gross operating surplus minus depreciation minus net saving of the corporate sector. source: south african reserve Bank data and world Bank staff calculations. gross corporate saving has also been on a downward trend after peaking at 20 percent of gdp in the 1980s (box figure 2). it has two components: retained earnings and depreciation costs. the decline in retained earnings between 1995 and 2007 can be explained by the commensurate increase in dividend payouts. it is possible that improving macroeconomic stability made the corpo- rates less risk-averse, lowering their precautionary saving motive. Easier access to global credit after the end of the apartheid sanc- tions and the liberalization of exchange controls would also have made it cheaper for firms to borrow relative to self-financing. despite the decline, corporate savings remain by far the single largest component of the aggregate gross saving, accounting for more than 90 percent in the 2000s. SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth After peaking at 35 percent in 1980, South Africa’s Figure national savings rate has been declining 2.7 Total savings Corporate savings Household savings Government savings 40 30 Percent of GDP 20 20 10 0 –10 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 source: south african reserve Bank and world Bank staff calculations. Public and private investment have tended Since the 1990s, South Africa’s saving rate has been low Figure compared with other emerging economies to reinforce each other, 2.8 1980 1990s 2000s though the public has 50 been more influential 40 30 Percent 20 10 0 China Malaysia India Russia Chile Mexico UMIC Argentina OECD Brazil Turkey South Africa source: south african reserve Bank, world Bank world development indicators data, and world Bank staff calculations. its ranking to share last with Brazil in the 1990s Public investment (general government plus and be last in the 2000s. public enterprises) and private investment have tended to reinforce each other, though the pub‑ Long-term investment trends lic has been more influential in shaping aggre‑ Influenced by the savings rates, the investment gate investment. The coefficient of correlation rate shows three distinct phases since 1960 (fig‑ between the public investment rate and the total ure 2.9). investment rate was 0.91 over 1960–2010, and 0.74 • The first phase spans 1960–81, with the between the private and total investment rates. investment rate increasing from 22 percent in From being in the middle of the pack in 1960 to an all‑time high of 33 percent in 1981. the 1980s South Africa slipped to being last in • The second phase saw the investment rate the 1990s and second from last in the 2000s, fall by more than half to 14 percent by 1993 slightly ahead of Brazil (figure 2.10). and then stagnate until 2001. • The third phase saw the investment rate Investment rates in South Africa: increase from 15 percent in 2001 to 22.5 Role of the real returns to capital percent in 2008 before being affected by the Is there enough incentive for the private sector global financial crisis. to invest, create jobs, and contribute to faster Figure South Africa’s investment rate shows three trends since 1960 2.9 Public Private Total 40 Investment rate (percent of GDP) 30 20 21 10 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 source: south african reserve Bank and world Bank staff calculations. South Africa is clearly South Africa’s domestic fixed investment rate has slipped an attractive place Figure compared with other emerging economies for investment 2.10 1980 1990s 2000s 40 Fixed investment (% of GDP) 30 20 10 0 China India Malaysia UMIC Chile Mexico OECD Turkey Russia Argentina South Brazil Africa source: south african reserve Bank, world Bank world development indicators data, and world Bank staff calculations. growth? The low private investment rates could compared with the interest cost, proxied by the mean one of two things: either the real returns prime lending rate. If the returns are substan‑ to capital are low, or private investors are not tially in excess of the interest cost, something responding to changes in real returns because other than returns to investment is impeding of risk perceptions or structural barriers to private investment. investment. Returns are high and have been South Africa is clearly an attractive place increasing since the mid‑1990s, suggesting that for investment. The real returns to capital have the real problem might be insufficient savings risen sharply since the early 1990s, to an aver‑ or deeper structural impediments. age rate of 15 percent between 1994 and 2008, By analogy with a bond, the nominal return or 23 percent in nominal terms (figure 2.11). to capital is simply the nominal profit rate They averaged close to 22 percent in 2005–08, (net of depreciation) per unit of physical capi‑ in the same range as that for China, albeit for tal (“coupon”) plus the change in the price a much longer period (Bai and others 2006). of capital (“capital gain/loss”) divided by the Such high returns are especially striking given price of capital. To measure in real terms the South Africa’s modest GDP growth. They have rate of inflation is subtracted from the nomi‑ also risen comfortably in excess of the borrow‑ nal return measure (annex 1). This can be ing cost, measured by the prime rate, which SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth has declined sharply since the late 1990s (fig‑ apartheid isolation from global markets, which ure 2.11). made capital equipment available at more com‑ The real returns to capital increase with the petitive rates. Also important was the global real profit (or rental) rate per unit of physi‑ moderation of the prices of information and cal capital (or nominal profit rate divided by communication technology–related equip‑ the GDP deflator) and fall when the price of ment, which catered such growing sectors as capital rises relative to the GDP deflator. For finance and telecommunications. The rising example, when the price of buying a piece of profit rate likely reflects the shift to more capi‑ 22 machinery falls, the returns from that invest‑ tal‑intensive technologies and across‑the‑board ment automatically increase. The real profit skills upgrading that enhanced the output gen‑ rate has been rising since 1993, while the price erated by each additional unit of capital. of capital has been falling relative to the GDP The real returns to capital are high across deflator (figure 2.12). Thus, both forces have the major sectors, if with variation (figure reinforced each other in increasing the real 2.13).19 Real returns in the construction and rates of returns to capital. domestic retail and wholesale trade sectors The real returns to The slowdown in the price of capital was have been extraordinarily high—85 percent in likely triggered by the end of South Africa’s the 2000s for the former and almost 60 percent capital are high across the major sectors, Figure Real returns to capital in South Africa have risen sharply since the early 1990s . . . if with variation 2.11 Real returns to capital Real prime lending rate 25 20 15 Percent 1950s 1960s 1970s 10 1980s 5 0 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 note: real prime rate is calculated as the prime rate at the end of the year minus inflation of gdp deflator during the year. source: south african reserve Bank data and world Bank staff calculations. . . . benefiting from the rising real profit rate and Figure moderating price increases for capital since 1993 2.12 Ratio of price of capital to GDP de ator Real pro t rate 1.8 0.30 s 50 s 60 19 19 s 70 s 19 80 1.5 0.25 19 Percent Ratio 1.2 0.20 0.9 0.15 0.6 0.10 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 note: real prime rate is calculated as prime rate at the end of the year minus inflation of gdp deflator during the year. source: south african reserve Bank data and world Bank staff calculations. Figure Real returns to capital in South Africa have been highest in construction and trade 2.13 1993–99 2000–10 90 70 50 Percent 30 23 10 –10 Construction Trade Manufacturing Agriculture Finance Mining Transport Community source: south african reserve Bank data and world Bank staff calculations. It is somewhat for the latter. In finance real returns rose from not been higher, particularly in greenfield paradoxical that, an average of 8 percent in the 1990s to almost areas. South Africa also fares well on the major 20 percent in the 2000s, and in manufactur‑ global competitiveness indices. The World despite high and ing from 17 percent to 25 percent. Real returns Bank’s Doing Business Survey ranked South have been highly volatile in mining and agri‑ Africa a creditable 34th of 183 countries in increasing real returns, culture due to the nature of the sectors, but 2011, ahead of each of the other BRICS and private investment has still averaged 25 percent in the 2000s in agri‑ most upper middle income countries. The culture and 16 percent in mining (up from Global Competitiveness Report of the World not responded with just 6 percent in the 1990s). Private interest in Economic Forum placed South Africa 54th more vigor or FDI has these two sectors seems not to have lived up of 139 countries in 2010 (down 9 ranks from to full potential, even after accounting for the 2009) on its Global Competitiveness Index, the not been higher high volatility in returns. Real returns in the highest rank in Sub‑Saharan Africa. South transport sector doubled from 5 percent in the Africa’s strengths lie in its financial and fiscal 1990s to 10 percent in the 2000s. These reveal‑ institutions, investor and property protection, ing results would benefit from further research tax system, legal and judicial system, and audit‑ to learn more about the underlying causes. ing and reporting standards. Private investment has become less respon‑ Clearly, this favorable aggregate picture sive to the returns on investment since 1994. masks aspects of the investment climate that That is, the increase in private fixed investment disproportionately affect private investors. in response to each percentage point increase Four key issues stand out: high industrial con‑ in real returns has become more subdued centration; significant skills gaps; contentious (annex 2). This highlights a missed oppor‑ labor relations and work stoppages; and low tunity, given the steep increase in returns. savings rates. It is also somewhat puzzling. If anything the 1. Industrial competition is much weaker in response by private investors should have South Africa than its international peers become more favorable as the shackles of the (Aghion and others 2008). This points to apartheid system came off and the economy entry barriers that discourage new invest‑ became better integrated with the world. Some‑ ment despite high returns. The severity of how, worsening risk perceptions and structural the problem is recognized at the highest barriers have become an impediment. levels in South Africa, and the New Growth Path framework targets competition policy Policy issues to consider as one of its core reform areas. It is somewhat paradoxical that, despite high 2. Skills development, thwarted for the blacks and increasing real returns, private investment during apartheid, remains an important has not responded with more vigor or FDI has deterrent for firms, new and old, looking to SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth expand their business. The problem starts South Africa’s savings: Required levels with basic education: access has improved, and determinants but quality has not. 20,21 Test scores show What level of savings is needed to support South Africa faring miserably relative to South Africa’s growth objectives? This section global comparators. Accordingly, the Global tackles that question under varying conditions Competitiveness Report ranks the country of the employment intensity of production— 125th on the Quality of Basic Education, the number of additional workers deployed for 130th on the Quality of the Educational each additional unit of output—and produc‑ 24 System, and 137th (third from last) on the tivity growth. The simulations are based on a Quality of Math and Science Education. simple growth accounting model (annex 4). With intakes largely ill‑equipped with cog‑ nitive (literary and quantitative) skills, Savings needed for sustained 6.5 the problem only gets compounded at the percent GDP growth rate higher and technical education levels. Skills The scenarios here target an increase in GDP are among the top reform areas picked up growth from 2.8 percent in 2010 to 3.4, 4.1, and South Africa will find by the New Growth Path, though there are 4.4 percent in 2011, 2012, and 2013 (National no easy near‑term solutions. Treasury projections) and settling at 6.5 per‑ it hard to increase 3. Labor relations are much more contentious cent from 2016 onward. The investment rate savings to the levels than other emerging market economies, closely follows savings rates to ensure external partly because they are seen more through sustainability. Two sets of assumptions are criti‑ needed for target political and equity lenses rather than a cal to the results: growth rates without pure economic lens. Protracted wage dis‑ • Technological improvements, or total fac- putes and work stoppages, on full display in tor productivity (TFP) growth: Two alter‑ making production 2010, contribute to South Africa’s rank of native assumptions are used. Moderate TFP more employment- 132nd of 139 on the labor relations index growth: TFP growth of 1 percent per year, of the Global Competitiveness Report. estimated to be the average for South Africa intensive and enhancing This is an implicit tax on investment, partly over 1996–2006 (Eyraud 2009). Improved productivity growth explaining why global investors, armed with TFP growth: TFP growth of 1.5 percent a options, have eschewed long‑horizon oppor‑ year, ref lecting economywide structural tunities (as opposed to short‑term securi‑ reforms geared for greater efficiency. ties) in South Africa. In addition, wage • Labor intensity of production: Pessimistic levels relative to worker productivity are sig‑ scenario: Growth in the number of employed nificantly higher than among its peers, 22,23 workers increases to 1 percent, the average also discouraging new investment. rate of increase between 2001 and 2010. 4. Low savings rates are discussed in depth in Business-as-usual scenario: Growth in the the next section. number of employed workers picks up from Among other key issues, crime emerges as 0.8 percent in 2010 to 1.3 percent in 2013, the constraint most frequently cited by the and to 1.9 percent from 2016 onward, but firms covered by the 2010 Investment Climate the employment intensity of growth remains Assessment of the World Bank. Reinforcing the same. 24 High employment–intensity sce- this, the latest Global Competitiveness Report nario: Employment growth picks up from ranked South Africa 137th (third from last) on 0.8 percent in 2010 to 3 percent from 2016 Business Costs of Crime and Violence. Access onward. This is predicated on a reversal of to reliable electricity, corruption, access to the rising capital‑labor trends in key sectors finance by small and medium enterprises, and such as manufacturing and mining, more anticompetitive practices rounded off the top flexible labor markets, a more dynamic five constraints faced by South African firms. informal sector, and lower spatial barri‑ Also influencing investment decisions is policy ers that now raise transportation costs and uncertainty, which the government is address‑ labor search costs. ing in mining, and a volatile and overvalued The main results are presented in figures exchange rate. 2.14–2.16. Figure Projected savings rate for South Africa under pessimistic scenario 2.14 Under 1 percent total factor productivity growth Under 1.5 percent total factor productivity growth 50 40 Savings rate (percent) 30 20 25 10 0 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 source: south african reserve Bank data and world Bank staff calculations. Figure Projected savings rate for South Africa under business as usual scenario 2.15 Under 1 percent total factor productivity growth Under 1.5 percent total factor productivity growth 50 40 Savings rate (percent) 30 20 10 0 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 source: south african reserve Bank data and world Bank staff calculations. Figure Projected savings rate for South Africa under high employment growth path scenario 2.16 Under 1 percent total factor productivity growth Under 1.5 percent total factor productivity growth 50 40 Savings rate (percent) 30 20 10 0 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 source: south african reserve Bank data and world Bank staff calculations. SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth It is clear from the results that South South Africa’s savings rates have been signif‑ Africa will find it hard, if not impossible, to icantly below the potential suggested by its eco‑ increase savings to the levels needed for tar‑ nomic and structural characteristics (table 2.2). get growth rates without making production The national savings rate should have declined more employment‑ intensive and enhancing 4.6 percentage points between the two periods, productivity growth. Without commensurate mainly on account of growing urbanization, improvements in these two reinforcing areas, financial liberalization, and enhanced macro‑ savings would need to rise by more than 10 economic stability (box 2.2). The actual perfor‑ 26 percentage points within a short span of time mance was significantly worse. The national (five years under current scenarios), something savings rate fell from an average of 24 percent rarely seen anywhere, to support sustained 6.5 in the first period to 16 percent in the second, percent GDP growth. Or, more realistically, if a decline of 8 percentage points. In other the savings rate peaks at 25 percent, with the words, South Africa’s savings rate fell short by investment rate only slightly above that for sus‑ 3.4 percentage points relative to what it could tainability reasons, GDP growth would have to have been, given its characteristics. Two crucial factors settle down at less than 4.5 percent. A similar exercise for nongovernment sav‑ ings finds an even larger discrepancy. The aver‑ appear to have kept the What is the savings rate consistent with age nongovernment savings rate for 1968–94 savings rate low: youth South Africa’s economic and structural was 26.4 percent, which declined to 19.3 per‑ characteristics? An empirical investigation cent during 1995–2008. The model predicts unemployment and The previous discussion makes clear the need an increase of 3.1 percentage points—for an slow income growth for higher public and especially private sav‑ underperformance of 10 percentage points. ings. The analysis here compares the actual Two crucial factors appear to have kept the change in savings for two nonoverlapping peri‑ savings rate low: ods against the predicted change for the same • Youth unemployment. With better results period (annex 4).25 For South Africa a natural on youth employment, 26 the savings divi‑ break point to examine structural changes is dend from the sharp decline in the young‑ 1994. Accordingly, the analysis here considers age dependency ratio (the ratio of those subperiods: 1968–94 and 1995–2008. under 15 years of age to the working age table Predicted changes in saving rates in South Africa (percentage points) 2.2 contribution to change contribution to change in in national saving rate nongovernment saving rate positive contributors Young-age dependency ratio 5.0 12.4 real gross national disposable income (gross private domestic investment for private saving) per capita growth 1.2 0.9 real gross national disposable income (gross private domestic investment for private saving) per capita (natural log) 0.1 0.0 government saving to gross private domestic investment 1.4 negative contributors Urbanization ratio –6.5 –5.3 domestic private credit flow to gross national disposable income (gross private domestic investment for private saving) –1.4 –1.7 real interest rate –1.3 –2.8 gdp deflator inflation –1.3 –1.5 old-age dependency ratio –0.3 –0.3 m2 to gross national disposable income –0.02 –0.03 terms of trade –0.02 –0.02 predicted change in national saving rate –4.6 3.1 actual change in saving rate –8.1 –7.0 source: world Bank staff calculations, based on hevia, ikeda, and loayza (2010). Figure South Africa’s demographic profile (proportion of total population) 2.17 Under age 16 Ages 16–65 Age 65 and older Young–age dependency ratio 80 60 Percent 40 27 20 0 1980 1990 2000 2010 note: Young-age dependency ratio is the ratio of those under age 16 to the working age population. source: Un-world Bank population database and world Bank staff calculations. Box Main determinants of savings 2.2 the model applied here picks up the following structural elements as being the most influential for saving: level of income. it is well established in the empirical literature that higher per capita incomes are associated with higher savings rates. the size of the impact depends on the development level of the country and whether the changes to income are permanent. For south africa the difference in average per capita gross national disposable income (gndi) between the two subperiods was neg- ligible; it took until 2006 for per capita gndi to catch up with 1981. this means the income channel did not play a major role, as it could have. income growth. a strong positive relationship has been found between income growth and savings by many studies. the causal direction is not yet clear, however. south africa’s real per capita gndi growth was about 1.7 percentage points higher in the sec- ond period, which, according to the results, would have led to an increase in the savings rate of 1.2 percentage points. demographics. in south africa the old-age dependency ratio (the ratio of people over 65 years old to the working age population) increased slightly in the second period causing a small drop (0.3 percentage points) in national savings. the decline in the young- age dependency ratio (ratio of people aged 15 years or less divided by working age population) was substantial, from 72.6 percent to 52.6 percent, which should have led to a sharp increase in the national savings rate of 5 percentage points. Economic uncertainty. risk-averse agents, faced with uncertain events and unable to perfectly insure risks because of incomplete financial markets, accumulate assets to avoid suffering large adjustment in consumption. a common proxy for economic uncertainty is the inflation rate, which fell in south africa in the second period, lowering the need for precautionary savings. the model pre- dicts that this effect would have caused national savings to be 1.3 percentage points lower over 1995–2008. Urbanization. Uncertainty is also proxied in the literature by the degree of urbanization. rural incomes, often linked to agriculture and thus exposed to unpredictable weather conditions, tend to be less secure, especially in developing countries. an increase in urbanization can thus be expected to lead to more predictable income streams and to lower aggregate savings. the model predicts that the increase in the urbanization ratio in south africa would have resulted in a decline in the savings rate of 6.5 percentage points. Financial development. Financial liberalization often involves lifting controls over interest rates, which then rise. the income effect from this tends to increase savings while the substitution effect tends to lower them; the net effect is uncertain. the modeling applied here finds that the negative substitution effect tends to dominate. thus, the increase in real interest rates in south africa would have caused savings to fall by 1.3 percentage points. Financial sector development also expands the credit to previously credit-constrained private agents, reducing their savings, something established with regularity in the empirical literature. in south africa credit to the private sector contracted in the first period but grew in the second. the net effect of the change in private credit would have been a 1.4 percentage point decline in the savings rate. in addition, financial depth, measured by m2/gdp, remained virtually unchanged between the two periods, thus having only a marginal impact on the savings rate. SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth Box Saving-growth for four successful emerging market economies 2.3 china saw its gross domestic savings rate increase from 40 percent to 52 percent between 2002 and 2009; it stood at 30 percent in 1975 (box figure 1). chile’s rate rose stupendously in the 1980s, from 9 percent in 1982 to 30 percent in 1989 (box figure 2). india’s rate was on a gradually increasing trend from the early 1980s until the early 2000s, when it suddenly spurted from 25 percent in 2003 to 34 percent in 2007 (box figure 3). the jump in indonesia’s savings came much earlier—in the late 1960s and early 1970s (box figure 4). in each case, large positive swings in the savings rates were accompanied by equally large increases in gdp growth rates. it is quite likely that savings and growth reinforced each other while third forces—such as favorable demographics and better policy 28 management—simultaneously helped both processes. high savings were not a prerequisite for growth takeoffs in these cases, but without the savings response, high growth rates would have faltered in the longer term, which these countries largely avoided. Box figure 1 China Box figure 3 India Ratio of gross domestic GDP growth, Ratio of gross domestic GDP growth, savings to GDP, 3-year 3-year moving savings to GDP, 3-year 3-year moving moving average average moving average average 60 15 40 10 30 7.5 Improvements in the 40 10 Percent Percent 20 5 national savings rate 20 5 10 2.5 will depend most of all 0 0 0 0 on resolving the high 1975 1980 1985 1990 1995 2000 2005 2009 1981 1985 1990 1995 2000 2005 2009 Box figure 2 Chile Box figure 4 Indonesia unemployment rates Ratio of gross domestic GDP growth, Ratio of gross domestic GDP growth, savings to GDP, 3-year 3-year moving savings to GDP, 3-year 3-year moving and achieving a mutually 45 moving average average 10 45 moving average average 10 reinforcing process 30 5 30 5 of higher savings and Percent Percent 15 0 15 0 faster GDP growth 0 –5 0 –5 1981 1985 1990 1995 2000 2005 2009 1963 1970 1980 1990 2000 2009 source: world Bank world development indicators data and world Bank staff calculations. population) seen in figure 2.17 would have on resolving the high unemployment rates, been much higher. Even as the young‑age especially among the youth, and ensuring a population bulge came of age and entered GDP growth spurt. According to the United the labor market, the effective dependency on Nations–World Bank population estimates, income earners did not decline much, due the young‑age population ratio in South Africa to pervasive youth unemployment, prevent‑ fell from 58 percent in 1996 to 47 percent in ing the significant improvements in savings 2009. This demographic group is still under that should have resulted. 30 years of age and, according to employment • Slow income growth. Gross national disposable statistics, facing acute unemployment. Resolv‑ income barely changed between the two peri‑ ing the unemployment problem for this group ods, preventing an increase in the savings rate. holds tremendous potential for increasing the If it had risen 40 percent rather than staying savings rate. stationary, national savings would have been 5 A mutually reinforcing process of higher percentage points higher in the second period. savings and faster GDP growth will also be important. When a big increase in the savings Policy issues to consider rate is required, it is unlikely to occur on its Looking ahead, visible improvements in the own. A more plausible view is that if growth national savings rate will depend most of all and disposable income pick up substantially, savings will tend to go up because consumption around the pricing, efficiency, and governance patterns will change more slowly, in line with of the contractual savings institutions and low the habit formation hypothesis.27 This would “preservation” of public pension funds is no lay the foundation for sustained high growth doubt healthy. But it is unlikely to significantly over the long run. increase national savings. The importance of having savings and A final matter is the financial exclusion of income growth reinforce each other is borne small businesses and the poor, which has only out by China, Chile, India, and Indonesia, gotten worse since the start of the global finan‑ which had sharp upticks in savings (box 2.3). cial crisis.31 Despite the presence of world‑class 29 This suggests the need for a productivity‑ financial and savings institutions, a large pro‑ led growth shock, which could be externally portion of the population still relies on nontra‑ funded (preferably by FDI) in the short to ditional mechanisms for saving and financial medium term, accompanied by efforts to security. Participation of low‑income groups ensure that the savings response is robust in the formal system has been estimated at enough to make growth sustainable. A concom‑ less than 15 percent, often through compul‑ itant decline in inequality would only multiply sory workplace schemes (Finmark Trust 2009). A trajectory of faster the benefits, since the opportunity to increase Stokvels and burial societies fill this gap for and more inclusive savings is greater when people break out of low‑ many black South Africans not in the urban income traps. Savings follow only after families mainstream (box 2.4). But such small institu‑ GDP growth will cover basic expenditures on food and shelter. tions lack the organizational capacity for scal‑ Global experience further shows that per‑ ability and long‑term viability, and without require higher rates of manent increases in public savings tend to adequate regulatory supervision, they can even investment and savings, increase national savings; more so in the short be susceptible to fraudulent schemes. So, the run than in the long.28 And reductions in recur‑ policy challenge is to find a way of channeling more intensive use rent public spending tend to be more effective the sophistication of the high‑end formal insti‑ of labor, and heavy in this. With the given stream of government tutions with the innovative drive and reach of expenditures, it is often irrelevant whether the the traditional ones to enhance financial inclu‑ doses of productivity government taxes today or later. This is known sion and bring the masses into the savings fold. enhancements as the Ricardian equivalence proposition, A lack of microfinance institutions with econo‑ which has strong empirical support. The evi‑ mies of scale and scope could be the missing dence on the effectiveness of tax incentives to link.32 raise savings (including those targeting retire‑ ment saving instruments) is also not too prom‑ The way forward ising. So, National Treasury’s envisaged fiscal Confronted with widespread and persistent consolidation over the medium run should exclusion and unemployment, policy makers have a desirable effect on national savings. in South Africa have rightly set their eyes on The contractual savings system appears a trajectory of faster and more inclusive GDP to be functioning well overall and does not growth. This will require higher rates of invest‑ show much scope for contributing to higher ment and savings, more intensive use of labor, national savings. South Africa had converted and heavy doses of productivity enhancements. its public pension system from pay‑as‑you‑go Among these essential ingredients for inclusive to fully funded toward the end of the apart‑ growth for South Africa, a virtuous cycle is seen heid government the latter according to inter‑ at play. Employment‑intensive growth would national experience tends to be associated enhance productivity by skilling and tooling with higher saving.29 Moreover, with 9 million the unemployed labor force. Tackling youth members and more than R2 trillion in assets, unemployment will enhance the benefit from South Africa already has among the world’s the favorable demographics and raise savings. largest pension systems relative to the size of Productivity enhancements, in turn, will raise its economy. 30 The coverage rate for formal‑ GDP growth and attract private investment, sector employees is estimated at about 60 per‑ while lessening the burden on investment and cent, relatively high. The ongoing policy debate saving to support higher growth. SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth Box Stokvels in South Africa 2.4 stokvels constitute an industry with which many are familiar but about which little is known. what follows are some of the key facts, figures, and anecdotes that emerge from the rather scattered and scarce information: the word stokvel originates from “stock fairs” from colonial times, when it was used to refer to rotation cattle auctions. it has evolved over time as an overarching term for informal community-based saving clubs, including (wits 2009): • contributions stokvel—members contribute a fixed amount weekly, fortnightly, or monthly, with the pooled amount allo- cated on a rotational basis. • Basic stokvel—expands the contributions stokvel to cover specific events such as funeral assistance and christmas. 30 • purchasing stokvel—members contribute over a fixed term, at the end of which they purchase big-ticket items such a marquee (big tents for functions) or big gas stoves that can be used to generate income. • grocery stokvel—members contribute a fixed monthly amount for a specific period, such as one year. the pooled amount is then used to buy durable grocery items (sugar, soap, oil) in bulk, thus taking advantage of economies of scale otherwise not possible for individuals. • investment group—contributions are used for investment, and dividends are either shared or reinvested. • Family stokvel—family members pool funds in a formal bank account to buy large items such as cars. • Burial societies or makgotlas—used specifically for funerals, though these are not strictly stokvels. some assist in acquiring and preparing food, and some cover the full cost of the funeral. the composition of stokvel memberships has evolved from predominantly rural black poor to urban black executives. the sophistication of a stokvel depends on the composition of its members. the Finmark trust (2010) estimates that 18 percent of south africans used informal mechanisms such as stokvels and burial societies in 2010 for savings. one in every two black adults is reported to be member of a stokvel. the 11,000 stokvels registered with the national stokvel association of south africa had between 12 and 150 members each, for a total of around 150,000. Estimates of nonregistered stokvels run in the thousands, the majority of them women. Efforts to formalize stokvels have met mixed results. the national stokvels association of south africa was established in 1988 as a self-regulating body. the Banks act (1990) was amended in 2006 to make provision for stokvels. By 2009 all major banks— including absa, FnB, standard Bank, nedbank, and postbank ithala—had a savings product customized to stokvels. the “industry” (including “burial societies”) generated an estimated r12 billion in new funds in 2003 (University of witwatersrand 2009), higher than the r10.7 billion in purchases of bonds during the same year. stokvels have been used in the retail sector to negotiate purchases of goods at a discount, from makro, Unilever, or shoprite. in investment forums, such as the sasol inzalo BEE investment initiative, provision was made for stokvels to subscribe together with the national Empowerment Fund. other BEE deals that favored stokvel investment include telkom, real africa investments, stokvel car rental, stokvel times, and techno spaza. Just as higher savings rates will need to to better integrate the advanced economy underpin higher growth over the long run, part of South Africa and the less‑ developed higher incomes, especially among the lower economy part, marked by the spatially sepa‑ end of the income spectrum, will raise the level rated townships and informal settlements of savings. This interplay between higher sav‑ where the bulk of the unemployed live. ings and investment, employment‑intensive The advanced economy borders on levels growth, and productivity seems to be operating of development seen in industrialized coun‑ today at a suboptimal equilibrium, undermin‑ tries, which also means that the scope for ing the quest for inclusive growth. fast growth is limited. Faster growth will How to move to a more favorable equilib‑ have to come from the less‑developed econ‑ rium? No quick fixes or small perturbations will omy, which has the potential to take off in produce dramatic results. The quest for inclusive the same way that other successful emerging growth may require, above all, a significantly dif‑ market economies have. Ways will have to ferent mindset. Two major pushes, both in the be found to exploit the “arbitrage” between spirit of creative and bold thinking, seem to hold the two economies, which remain largely the most potential—one requiring a more inten‑ unlinked, to ensure that capital flows into, sive inward look, and the other an outward look. not out of, the less‑developed economy— • More effective internal integration holds tre- and that labor is more mobile toward the mendous potential. This involves a big push advanced economy while entrepreneurs Box Developing Factory Southern Africa— Lessons from Factory Asia 2.5 the story of developing East asia’s dramatic rise from an underdeveloped, mainly agrarian region to becoming the world’s factory in a span of a few decades is regarded by many as an economic miracle. Job creation at an unprecedented scale is among the important outcomes of this experience. things looked bleak at the start, with high levels of poverty and a low endowment base of natural resources. But this assessment overlooked the ample supplies of cost-effective labor, much like in southern african countries today, many saddled with high unemployment rates. multinationals set up effective production value chains across East asia, with each link in the value chain located according to the comparative advantages of the host countries. the labor-abundant countries got the labor-intensive parts of the value chain while the more specialized, skill-intensive and capital-intensive parts stayed in Japan, to begin with. But the supply chain system 31 was nimble and adapted to the countries’ emerging strengths in worker skills, wage competitiveness, and capital intensities. this production agility has served the region well. among the nonwestern economic blocs, East asia has the highest intrare- gional trade, comprising largely intermediate goods, underpinning the region’s global trade and competitiveness agenda, and attract- ing ample Fdi. in other words, Factory asia has worked well. Factory southern africa has not. southern africa remains the least integrated region in the world (box figure 1), despite the presence of a customs union (such as the southern african customs Union) and a free trade area (such as the southern african development community), and sufficient variations within the region—whether comparing south africa’s population of 50 million and seychelles’ population of 85,000, or mauritius’s per capita gni of $7,250 in 2009 and democratic republic of congo’s $160— to offer substantial benefits from trade and production specialization. Unlike the nimble production systems in asia, however, the norm in southern africa is for single location production centers with distribution networks in neighboring countries. Box figure 1 Intraindustry trade by region, 2008 (Grubel-Lloyd index) North America Western Europe Southeast Asia and Paci c Northeast Asia Eastern Europe South America Central America and the Caribbean Western Asia Eastern Asia Central Africa Central Asia, Caucasus, and Turkey South Asia Northern Africa Western Africa Southern Africa 0.0 0.1 0.2 0.3 0.4 0.5 0.6 source: world Bank 2011e. Falling regional trade barriers and logistics costs as well as rising labor costs at core production locations spurred the decentral- ization of asian regional production networks to more cost-effective locations in the region. southern africa, by contrast, has seen the persistence of regional trade barriers, including restrictive regulatory policies, that raise trade costs and create uncertainty (box table 1). while tariffs have been lowered within the southern african development community, significant nontariff barriers remain covering one-fifth of regional trade. that undermines regional trade, lowers the region’s attractiveness to foreign investors, and eventually undercuts regional supply chains. the recent world Bank report referenced below catalogues the costs to specific southern african industries arising from nontariff receipts. moving aggressively and strategically on nontariff barriers would open new production and export opportunities for south africa and scale up existing production, with commensurate benefits for job creation. Box table 1 Impacts of nontariff barriers on Southern African trade southern africa regional trade potentially affected Barrier Examples of products affected (% of total) import bans, quotas, and levies wheat, poultry, flour, meat, maize, Uht milk, sugar 6.1 import permits and levies Uht milk, bread, eggs, sugar, cooking oils, maize, oysters 5.4 single marketing channels wheat, meat, dairy, maize, tea, tobacco 5.3 rules of origin textiles and clothing, palm oil, soap, cake decorations, curry powder, wheat flour 3.0 Export taxes dried beans, sheep, wood 4.8 source: world Bank 2011e. SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth have greater access to its markets. Integrat‑ be an attractive destination for long‑term ing the two economies will require a big global investment based on the returns its push on public transport infrastructure offers, but it is not. A clear, consistent, and while implementing programs to enhance predictable strategy for attracting FDI will financial inclusion and improving the cog‑ be important in this regard, which will also nitive and technical skills of youth. The guide attention to skills development, com‑ “2nd Economy” project of the Presidency is petition policy, industrial relations, and an encouraging start in promoting a better wage moderation in line with productivity 32 understanding of the issue.33 growth, in addition to access to finance for • More effective integration with the global econ- SMEs for stronger backward linkages. Fac‑ omy is needed based on South Africa’s latent tory Southern Africa can underpin South comparative advantages, particularly its two Africa’s competitiveness in global markets, surplus endowments in natural resources based on nimble “win‑win” regional produc‑ and unemployed labor. South Africa should tion supply chains (box 2.5). A clear, consistent, and predictable strategy for attracting FDI will be important to increase global investment and guide skills development, competition policy, industrial relations, and wage moderation in line with productivity growth ANNEx 1 Computing the real return to capital The focus section follows the work by Bai and α(t) ˆ r(t) = i(t) – PY(t) = + (ˆ K(t) – PY(t)) – δ(t) P ˆ others (2006). The return to capital is calcu‑ PK(t)K(t) lated using the Hall‑Jorgenson rental price PY(t)Y(t) equation (1967). The firm’s decision between selling the capital at a given point in time or where α(t) is the capital’s share of income in continuing using it is based on three factors: total output. (See Bai and others 2006 for fur‑ the forgone interest it would receive if it sells ther details.) If we assume that the price of out‑ the machinery and invests the proceeds of such put and capital as well as their rates of change transaction, the depreciation rate, and the are the same, the equation translates into the changes in the price of capital. Efficient asset familiar expression equating the real return to markets would equalize the rates of return for the marginal product of physical capital minus each type of capital. the depreciation rate. The return to capital is computed under the Finally, this equation, using the definition assumption that firms take the output price as of α (ratio gross nominal operating surplus to given. The nominal return then is determined nominal output) can be rewritten as follows: by the following formula: Gross nom. oper. surplus PY(t)MPKj(t) i(t) = ˆ – δ(t) + PK(t) PK(t)K(t) i(t) = ˆ – δj + PKj(t) PKj(t) Profit rate per unit of physical capital where i(t) is the nominal return to capital, and = ˆ – δ(t) + PK(t) PK(t) PY and PK are the prices for output and capi‑ tal, MPK is the marginal product of physical Profit rate per unit of physical capital capital, and PY(t)MPK(t) is the marginal revenue r(t) = PK(t) product of capital. The hats over a variable indicate rate of change. Then the real return – δ(t) + (ˆ K(t) – PY(t)). P ˆ to capital is given by: Hence, the above equations indicate that ˆ r(t) = i(t) – PY(t). we can compute the aggregate (sectoral) real One problem with the above expression is return to capital using aggregate (sectoral) that the marginal product of physical capital is output, aggregate (sectoral) depreciation, and not directly observable. But it can be inferred aggregate (sectoral) capital’s share of income using the capital’s share of income in total out‑ in total output. The South African Reserve put. After some algebraic manipulations, the Bank provides series with the aggregate (sec‑ following expression can be used to compute toral) real stock of capital, nominal and real the real return: aggregate (sectoral) output, nominal and real 33 SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth aggregate (sectoral) gross fixed capital forma‑ we assume an annual depreciation rate of 6 tion, and aggregate (sectoral) gross operat‑ percent. Finally, we also calculated the rate of ing surplus and compensation to employees, return using capital stock series computed fol‑ as well as an aggregate depreciation series on lowing the perpetual inventory method, with nominal currency units. Sectoral depreciation similar results. series are not available and, for those instances, 34 ANNEx 2 Empirical relationship between real returns to capital and fixed investment An empirical test of the relationship between investment is also more robust. The coefficient private fixed investment and the real returns to on the standard deviation of inflation is also capital is shown in table A2.1. positive but statistically not significant. Regression 1: Here the growth rate in private Regression 2: In regression 2 we introduce fixed capital formation (PFCF) in period t is a time dummy that takes a value of 1 for the regressed on its lagged value (to capture per‑ post‑apartheid years, 1994–2010, and 0 before sistence), a constant, the average real returns that. The time dummy is interacted with the for periods t‑6 to t‑1, the U.S. federal funds real returns variable and the standard devia‑ rate (FFR) to capture the global business cycle, tion of inflation variable to test whether there and the standard deviation of the GDP defla‑ is a noticeable break in private sector response tor inflation for period t‑6 to capture macro‑ from 1994 onward. The results are striking. Pri‑ economic uncertainty. The coefficient on the vate fixed investment was highly responsive to real returns is positive and statistically signifi‑ real returns in the pre‑1994 period: a 1 percent‑ cant, highlighting that higher real returns lead age point increase in real returns tended to to higher growth in private fixed investment. increase the growth in private fixed investment Essentially, a 1 percentage point increase in rate by 0.31 percentage points. Just as impor‑ real returns is associated with a 0.13 percent‑ tant, this relationship got considerably weaker age point increase in private fixed investment. in the post‑1993 period: that is, during this There is a positive and significant coefficient latter period a 1 percentage point increase in on the lagged value of PFCF growth, showing real returns led to only a 0.15 percentage point substantial persistence. All else being equal, increase in private investment growth, half the a 1 percentage point increase in private fixed magnitude of response in the earlier period investment in period t‑1 is followed by an before. Somewhat surprisingly, a positive and increase of almost half a percentage points significant relationship is seen between private in the same variable in period t. The coeffi‑ fixed investment and the standard deviation of cient on FFR variable also has a positive and inflation in 1994–2010. This may be on account significant coefficient. This likely captures the of an omitted third variable jointly driving fact that FFR is high during the positive phase fixed investment and macro volatility in the of the business cycle when the private fixed same direction. 35 SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth Regression results for the relationship between private fixed table capital accumulation and real returns, 1955–2010 A 2.1 regression 1 regression 2 dependent variable = private standard standard fixed capital formation growth coefficient error t-statistic probability coefficient error t-statistic probability independent variables constant –0.50 0.75 –0.69 0.51 –1.98 0.92 –2.14 0.04 real returns (moving average, t-1 to t-6) 0.13 0.56 2.41 0.02 0.31 0.08 3.70 0.00 Federal funds rate 0.13 0.06 2.1 0.04 0.13 0.06 2.29 0.03 36 standard deviation of gdp deflator inflation 0.04 0.13 0.27 0.79 0.04 0.13 0.28 0.77 lagged value of producer fixed capital formation growth 0.47 0.11 4.12 0.00 0.38 0.12 3.15 0.00 real returns (moving average, t-1 to t-6) dummy variable for post-1993 period –0.16 0.06 –2.72 0.01 standard deviation of gdp deflator inflation dummy variable for post-1993 period 0.71 0.31 2.29 0.03 r-squared 0.46 F-statistic 10.77 r-squared 0.53 F-statistic 9.17 adjusted prob adjusted prob r-squared 0.42 (F-statistic) 0.00 r-squared 0.47 (F-statistic) 0.00 durbin- durbin- watson stat 1.97 watson stat 2.03 source: world Bank staff calculations. ANNEx 3 Data and variable construction—Actual vs. predicted values We follow the work by Hevia, Ikeda, and and the RGNDI in LCU series will grow at the Loayza (2010) and Loayza and others (2000) same rate. to construct the variables used to calculate Finally, we have that: the predicted change in the national and pri‑ RGNDI per capita in USD = RGNDI in USD/ vate savings rate. Data have been obtained Population mainly from the World Development Indi‑ Growth of RGNDI per capita = ln(RGNDI per cators and the South African Reserve Bank. capita in USDt) – ln(RGNDI per capita in All the variables needed for this exercise are USDt–1) available for 1968–2008. We consider two Variables were constructed using the follow‑ subperiods, pre‑ and post‑apartheid (that is, ing series from the World Development Indica‑ 1968–94 and 1995–2008). Below we present tors: NY.GDP.MKTP.KN, NY.GDP.MKTP.KD, a brief description on how each variable has N Y.TRF.NCTR.KN, N Y.GSR.NFCY.KN, and been constructed: SP.POP.TOTL. Gross national disposable income (GNDI): Gross private disposable income (GPDI), This exercise requires a measure of real GNDI Gross national savings (GNS), and gross in constant U.S. dollars. Since the World Devel‑ public and private savings (GPrivSav and opment Indicators does not report all of the GPubSav): series needed to construct this measure, we GNS to GNDI = (GNDI in current LCU – Final first calculate the real GNDI in local currency consumption expenditure, etc in current units (RGNDI in LCU) as: LCU)/GNDI in current LCU RGNDI in LCU = Real GDP in LCU + Real Note: The correlation between GNS to GNDI net income from abroad in LCU + Real net computed using World Development Indicators current transfers from abroad in LCU and Reserve Bank data is 0.99. We then proceed to compute the ratio of GPrivSav in current LCU = Depreciation Private RGNDI in LCU and real GDP in LCU. We use Businesses + Savings Households + Savings this ratio to compute the RGNDI in USD as Corporate + Depreciation Public Corporations follows: GPubSav in current LCU = Depreciation RGNDI in LCU/Real GDP in LCU = RGNDI in Government + Savings Government USD/Real GDP in USD GPrivDI in current LCU = Final consumption RGNDI in USD = Real GDP in USD x RGNDI expenditures (private) + Residual × Final in LCU/Real GDP in LCU consumption expenditures (private)/(Final Since the World Development Indicators series consumption expenditures (private) + on real GDP in LCU and in USD grow at the Final consumption expenditures (public)) + same rate, this implies that the RGNDI in USD GPrivSav 37 SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth GPubDI in current LCU = Final consumption Measures of financial depth and develop- expenditures (public) + Residual × Final ment: The ratio of M2 to GNI and domestic consumption expenditures (public)/(Final and private credit flows to GNDI are used as consumption expenditures (private) + measures of financial development (FM.LBL. Final consumption expenditures (public)) + MQMY.CN and NY.GNP.MKTP.CN from the GPubSav World Development Indicators): We compute the gross public saving to gross M2 to GNI = Money and quasi money (M2) in private disposable income as: current LCU/GNI in current LCU GPubSav to GPrivDI = GPubSav in current To compute the domestic and private credit 38 LCU/GPrivDI in current LCU flow, we take the average of the current and Finally, real gross private disposable income in previous year stock (brought to current prices) USD is constructed as follows: and then we divide it by the GNDI, also in cur‑ Real Gross Private Disposable Income in USD = rent prices. The flows are given by the differ‑ RGNDI in USD × GPrivDI in current LCU/ ence between two consecutive years (series GNDI in current LCU K BP1347J, K BP1368J, K BP6018J from the The following series from the South African South African Reserve Bank, and NY.GDP. Reserve Bank have been used here KBP6007J, DEFL.ZS from the World Development KBP6008J, KBP6011J, KBP6018J, KBP6184J, Indicators): KBP6185J, KBP6186J, NRI6200J, NRI6201J, Total domestic (private) credit to GNDI = [(total and NRI6202J. domestic (private) creditt + total domestic (private) creditt–1 × (1 + GDP deflator Urbanization, old dependency ratio, and inflation))/2)]/GNDI young dependency ratios: These series have Total domestic (private) credit flow to GNDIt = been obtained from the World Develop‑ Total domestic (private) credit to GNDIt – ment Indicators (series SP.URB.TOTL.IN.ZS, Total domestic (private) credit to GNDIt–1 SP.POP.DPND.OL, and SP.POP.DPND.YG). Urbanization ratio = Urban population/Total Real interest rate and inflation rate: We com‑ population pute the inflation rate using the GDP deflator Old dependency ratio = Old (above 65)/Working- from the World Development Indicators (NY. age population GDP.DEFL.ZS). Real interest rates are com‑ Young dependency ratio = Young (below 15)/ puted using the bank/discount rate from the Working-age population IFS (60ZF end of period). ln (GDP deflator inflation) = ln [1 + (GDP Terms of trade: This series has been con‑ deflator t – GDP deflator t–1)/GDP deflator t–1] structed using the World Development Indica‑ ln (real interest rate) = ln [(1 + discount rate)/ tors (series NE.TRM.TRAD.XU). (GDP deflator t /GDP deflator t–1)] Terms of trade = ln (terms of trade index/100) ANNEx 4 Methodology at a glance To illustrate the links between savings and of schooling) and the growth rate of the growth, we follow the study done by Hevia workforce. The following equation that and Loayza (2011) for Egypt. The setup they links growth rate of output per worker to use is very simple and consists of a single the national savings ratio, the growth rate sector open‑economy model where output of productivity, the growth rate of the work‑ is produced by combining capital and effec‑ force, the increase in human capital, and the tive units of labor inputs. Effective labor capital‑ output ratio is parameterized using grows by increases in human capital (years South Africa’s data: . where γγt is the growth rate of output, γAt is the in income, which is roughly 0.5. We also let growth rate of productivity, γARt is the growth the depreciation rate be 0.06 (Bernanke rate of the absorption rate, γ15–65 is the growth and Gurkaynak 2001), which is consistent rate of the working age population, Et is years with depreciation series reported by the of schooling, e ΦEt is productivity per worker, y/k South African Reserve Bank. We use the is the output‑to capital ratio, α is the share of perpetual inventory method to calculate payments to capital in total income, δ is the the capital stock series, which gives us a depreciation rate, σ is the national savings capital‑output ratio of 1.85 for 2010, not sig‑ to output ratio, and β is ratio of foreign debt nificantly different from the South African to GDP. This equation indicates that output Reserve Bank’s series. growth is positively correlated with productiv‑ • Education and human capital. We use Barro ity growth, working age population growth, and Lee (2010) to compute the annual human capital growth, and the national sav‑ average increase in education for 1990– ings ratio. 2010 (0.08807). We assume that the returns Below we include a brief description of the to education are approximately 7 percent. sources and the calculations behind the param‑ • Current account sustainability. We require eters required to simulate the model for South the economy to maintain at all times a Africa’s economy: given ratio of foreign debt to GDP constant • Capital accumulation and depreciation rate. to reduce external vulnerability (that is, The capital share in output is computed as safe level of foreign borrowing). The aver‑ one minus the compensation of employees age level of net foreign liabilities to GDP 39 SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth was calculated using data from the South both pieces of information and assume African Reserve Bank (2011) for 2001–09 1 percent growth for our simulations. (0.12). The net income plus transfers from • TFP, real GDP growth and savings rate. We abroad to GDP was obtained using World target a real growth rate of 6.5  percent, Development Indicators data and is the a figure that has been mentioned in bud‑ average for 2001–09 (–0.033). get speeches and in the discussion of the • Labor market. Labor market parameters New Growth Path (6–7  percent growth). are calculated using the March Quarterly We take 16  percent for the savings rate, 40 Labor Force Survey for 2001–11 and popu‑ since South Africa’s gross national savings lation projections for South Africa from to GNDI is roughly 15–16 percent. Finally, the World Bank. According to the sur‑ we report simulation results for a 1 percent vey, working‑age population growth aver‑ annual productivity growth (normal TFP aged about 1.7  percent over this period. growth), which is consistent with Eyraud By contrast, the World Bank projects that (2009). We also report results under a rela‑ working‑age population will grow at about tively optimistic productivity growth sce‑ 0.5 percent over the next 20 years. We use nario (1.5 percent TFP growth). References Aghion, Philippe, Matías Braun, and Johannes Investment Affect Growth?” Journal of Inter- Fedderke. 2008. “Competition and Produc‑ national Economics 45: 115–35. tivity Growth in South Africa.” Economics of Carroll, Christopher, Jody Overland, and David Transition 16 (4): 741–68. Weil. 2000. “Saving and Growth with Habit Alfaro, Laura, Areendam Chanda, Sebnem Formation.” American Economic Review 90 (3): Ozcan‑Kalemli, and Selin Sayek. 2004. “FDI 341–55. and Economic Growth, the Role of Local Caselli, Francesco, and James Feyrer. 2006. “The Financial Markets.” Journal of International Marginal Product of Capital.” Presented Economics 64: 113–34. at the 7th Jacques Polak Annual Research ———. 2006. How Does Foreign Direct Investment Conference, International Monetary Fund, Promote Economic Growth? Exploring the Effects Washington, DC, November 9–10. of Financial Markets on Linkages. NBER Work‑ Centre for Development and Enterprise. ing Paper 12522. Cambridge, MA: National 2011. “A Fresh Look at Unemployment: A Bureau of Economic Research. Conversation among Experts.” Centre for Attanasio, Orazio, Lucio Picci, and Antonello Development and Enterprise Workshop, Scurco. 2000. “Saving, Growth and Invest‑ Johannesburg, South Africa, June. ment: A Macroeconomic Analysis Using a Commission on Growth and Development. 2008. Panel of Countries.” Review of Economics and The Growth Report: Strategies for Sustained Growth Statistics 82 (2): 182–211. and Inclusive Development. Washington, DC: Bai, Chong‑en, Chang‑Tai Hsieh, and Yin‑ Commission on Growth and Development. gyi Qian. 2006. “The Return to Capital in Economic Development Department. 2010. China.” Brookings Papers on Economic Activity “The New Growth Path: The Framework.” 37 (2): 61–102. Economic Development Department, Barro, Robert, and Jong‑Wha Lee. 2010. A New Johannesburg. Data Set of Educational Attainment in the World, Eyraud, Luc. 2009. “Why Isn’t South Africa 1950–2010. NBER Working Papers 15902. Growing Faster? A Comparative Approach.” Cambridge, MA: National Bureau of Eco‑ IMF Working Paper WP/09/25. Interna‑ nomic Research. tional Monetary Fund, Washington, DC. Bernanke, Ben, and Refet Gürkaynak. 2001. Is Feldstein, Martin, and Charles Horioka. 1980. Growth Exogenous? Taking Mankiw, Romer, and “Domestic Saving and International Capital Weil Seriously. NBER Working Paper 8365. Flows.” The Economic Journal 90: 314–29. Cambridge, MA: National Bureau of Eco‑ FinMark Trust. 2009. “Saving against a Storm.” nomic Research. FinMark Trust, Johannesburg. Borenstein, Eduardo, José De Gregorio, and Finmark Trust. 2010. “FinScope South Africa Jong‑wha Lee. 1998. “How Does Foreign 2010.” Finmark Trust, Johannesburg. 41 SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth Gwenhamo, Farayi, and Johannes Fedderke. Serve South Africa Better. Pretoria: National 2010. “The Composition of Foreign Capital Treasury. Stocks in South Africa: The Role of Institu‑ National Treasury of the Republic of South tions, Domestic Risk and Neighbourhood Africa. 2011b. Budget Review 2011. Pretoria: Effects.” Working Papers 163. Economic National Treasury. Research Southern Africa, Cape Town, Schwab, Klaus, ed. 2010. The Global Competitive- South Africa. ness Report 2010–2011. Geneva: World Eco‑ Hall, Robert, and Dale Jorgenson. 1967. “Tax nomic Outlook. Policy and Investment Behavior.” American Schmidt‑Hebberl, Klaus, and Luis Serven. 1999. 42 Economic Review 57 (3): 391–414. The Economies of Saving and Growth. Cam‑ Hevia, Constantino, and Norman Loayza. bridge, UK: Cambridge University Press. 2011. “Saving and Growth in Egypt.” Policy South African Reserve Bank. 2011. Quarterly Research Working Paper 5529. World Bank, Bulletin June 2011 No. 260. Pretoria: South Washington, DC. African Reserve Bank. Hevia, Constantino, Yuki Ikeda, and Norman Statistics South Africa. 2008. Labour Force Survey: Loayza. 2010. “Saving in Egypt: Trends and Historical Revision March Series 2001 to 2007.” Explanations.” World Bank, Washington, Pretoria: Statistics South Africa. DC. Statistics South Africa. 2008. “Quarterly Labour International Labour Office. 2011. Global Force Surveys, several issues.” Pretoria: Statis‑ Employment Trends 2011: The Challenge of a tics South Africa. Jobs Recovery. Geneva: International Labour Statistics South Africa. 2009. “Quarterly Labour Organization. Force Surveys, several issues.” Pretoria: Statis‑ International Monetary Fund. 2010. World Eco- tics South Africa. nomic Outlook: Rebalancing Growth. Washing‑ Statistics South Africa. 2010. “Quarterly Labour ton, DC: International Monetary Fund. Force Surveys, several issues.” Pretoria: Statis‑ ———. 2011. International Financial Statistics tics South Africa. online. International Monetary Fund, Wash‑ Statistics South Africa. 2011. “Quarterly Labour ington, DC. http://www.imfstatistics.org/ Force Surveys, several issues.” Pretoria: Statis‑ imf/. tics South Africa. Levchenko, Andrei, and Paolo Mauro. 2007. Tong, Hui, and Shang‑Jin Wei. 2011. “The “Do Some Forms of Financial Flows Help Composition Matters: Capital Inflows and Protect Against ‘Sudden Stops’?” World Bank Liquidity Crunch During a Global Economic Economic Review 21 (3): 389–411. Crisis.” Review of Financial Studies 24 (6): Levine, Ross, and David Renelt. 1992 “A Sen‑ 2023–52. sitivity Analysis of Cross‑Country Growth Trade and Industrial Policy Strategies. 2009. Regressions.” American Economic Review 82 “Second Economy Strategy: Addressing (4): 942–63. Inequality and Economic Marginalisation.” Loayza, Norman, Klaus Schmidt‐Hebbel, and Trade and Industrial Policy Strategies, Luis Servén. 2000. “What Drives Private Sav‑ Pretoria. ings Across the World.” Review of Economics University of the Witwatersrand. 2009. “The and Statistics 82 (2): 165–81. Stokvel Sector in South Africa: Opportuni‑ Mankiw, Gregory, David Romer, and David Weil. ties and Challenges.” University of the Wit‑ 1992. “A Contribution to the Empirics of watersrand, Graduate School of Business Economic Growth.” Quarterly Journal of Eco- Administration, Johannesburg. nomics 107 (2): 407–37. World Bank. 2010. South Africa: Second Investment National Planning Commission of the Repub‑ Climate Assessment. Business Environment Issues lic of South Africa. 2011. Diagnostic Over- in Shared Growth. Washington, DC: World view. Johannesburg: National Planning Bank. Commission. World Bank. 2011a. Accountability in Public Ser- National Treasury of the Republic of South vice Delivery in South Africa. Washington, DC: Africa. 2011a. A Safer Financial Sector to World Bank World Bank. 2011b. Doing Business 2011: Making World Bank. 2011. World Development Indica‑ a Difference for Entrepreneurs. Washington, DC: tors online. World Bank, Washington, DC. World Bank and the International Finance http://data.worldbank.org/data‑catalog/ Corporation. world‑development‑indicators. World Bank. 2011c. “Financing of Small and Wright, Graham. 2011. “The Power of Suc‑ Medium Enterprises (SMEs) in the Republic cessful, Market‑led Savings Mobilization.” of South Africa.” World Bank, Washington, Consultative Group to Assist the Poor DC. blog, April 19. http://microfinance.cgap. World Bank. 2011d. Global Economic Prospects: org/2011/04/19/the‑power‑of‑successful 43 Maintaining Progress amid Turmoil, vol. 3. ‑market‑led‑savings‑mobilisation/. Washington, DC: World Bank. Xu, Bin. 2000. “Multinational Enterprises, World Bank. 2011e. “Harnessing Regional Inte‑ Technology Diffusion, and Host Country gration for Trade and Growth in Southern Productivity Growth.” Journal of Development Africa.” World Bank, Washington, DC. Economics 62: 477–93. Notes 1. South African Reserve Bank 2011. developing are in order for us to achieve 2. Difference between number of workers a target growth rate of at least 7 percent employed between 2008 Q4 and 2010 Q3 per annum in the near future.” Similarly, (Statistics South Africa 2011). Finance Minister Pravin Gordhan called 3. See, for example, “A fresh look at unem‑ for 7 percent growth over 20 years in a ployment: A conversation among experts” speech at the University of Johannesburg by the Centre for Development and Enter‑ on March 10, 2011. These aspirations prise (2011) for a brief description and are consistent with National Treasury discussion of some of the main structural estimates that GDP growth of 7 percent issues behind the high unemployment in would be needed for 10 years to create 5.5 South Africa. million jobs, consistent with the employ‑ 4. Analysis in Schwab (2010) indicates that it ment target set forth by the New Growth takes three quarters after recession is over Path. for employment to show signs of recovery 11. National Treasury estimates that 9 million (and five quarters for unemployment rate jobs need to be created in the next 10 years to peak). The recovery process in the labor to reach the emerging‑market absorption markets is generally longer if the recession ratio of 56 percent. is associated either with a financial crisis 12. This is one of the core conclusions of the or a housing bust. Commission on Growth and Development 5. According to the International Labor (2008). Organization (2011), global unemploy‑ 13. It then declined by 2.7 percent in 2009 on ment increased from 5.7 percent in 2008 to account of the impact of the global finan‑ 6.3 percent in 2009. Preliminary estimates cial crisis before recovering to 1.8 percent show that it declined marginally to 6.2 per‑ growth in 2010. cent in 2010. 14. The comparator group comprises coun‑ 6. World Bank 2011c. tries with population of at least 1 million 7. World Bank 2011c. for which data were available for at least 20 8. Our estimates show the output gap to be years. around 2.3 percent of potential GDP in 15. Lao PDR was able to get by with rela‑ 2010 and 2011, which the projections show tive low levels of savings because of heavy to close only by 2014. donor financing. 9. World Bank 2011d. 16. See, for example, Levine and Renelt 10. For example, at a lecture at Beijing Uni‑ (1992). versity on August 24, 2010, President 17. Feldstein and Horioka 1980; Mankiw and Jacob Zuma said, “The plans we are now others 1992; Attanasio and others 2000. 45 SouTh AFRICA Economic UpdatE—FocUs on savings, invEstmEnt, and inclUsivE growth 18. The breakdown between savings of public 25. Hevia, Ikeda, and Loayza (2010) recently and private corporations is not available. applied this framework to a country case 19. At the sectoral level the data needed for study on Egypt. the calculations are not available before 26. According to the labor force data, the 1993. unemployment rate is 50 percent among 20. National Planning Commission of the those between the ages of 15 and 24, and Republic of South Africa 2011. 40 percent among those between the ages 21. World Bank 2011a. of 15 and 30. 46 22. World Bank 2010. 27. Carroll and others 2000. 23. The Global Competitiveness Report ranks 28. Schmidt‑Hebberl and Serven 1999. South Africa 112th on its index on Pay and 29. Schmidt‑Hebberl and Serven 1999. Productivity. 30. National Treasury of the Republic of South 24. GDP growth averaged 3.5 percent between Africa 2011. 2001 and 2010 while employment growth 31. World Bank 2011c. averaged 1 percent. The baseline scenario 32. Since 2002, the world‑famous Grameen assumes that the ratio of employment Bank in Bangladesh has successfully issued growth to GDP growth (or the employment market‑based savings instruments– varying intensity of growth) remains the same. from individual passbook savings accounts Using the GDP projections from the 2011 to contractual products such as the hugely Budget for 2010–13 and setting growth of 5 popular Grameen Pension Scheme recur‑ percent and 5.7 percent for 2014 and 2015 ring deposit product to good effect to and 6.5 percent for the rest of the simula‑ millions of poor Bangladeshi’s. A good tion horizon, gives the employment growth account may be found in Wright (2011). rates described in the text. 33. Trade and Industrial Policy Strategies 2009.