SOUTH AFRICA ECONOMIC UPDATE Edition 11 | April 2018 | World Bank JOBS AND INEQUALITY Public Disclosure Authorized © 2018 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street NW Washington, DC 20433 USA All rights reserved Photos: Flickr, Pexels, Shutterstock and World Bank. CONTENTS Contents i Acknowledgments iii Foreword iv Abbreviations v Executive Summary vi CHAPTER 1 CHAPTER 2 Recent Economic Developments 1 Jobs and Inequality 23 Global Economic Developments 2 South Africa Remains Trapped in a Cycle of High Real Sector Developments in South Africa 6 Inequality and Slow Job Creation 24 Labor Market Developments in South Africa 12 Creating More and Better Jobs to Reduce Fiscal Developments in South Africa 14 Income Inequalities 36 Inflation and Monetary Policy in South Africa 16 Conclusion 48 The External Sector in South Africa 17 The Outlook for South Africa 19 References 49 Annex: Modeling Prospective Policy Scenarios 51 FIGURES Figure 1.1: Global activity indicators 2 Figure 1.2: Global financial indicators 3 Figure 1.3: StasSA's GDP revisions - fourth versus third quarter 2017 releases 7 Figure 1.4: Commodity price forecasts 9 Figure 1.5: Historical decomposition of domestic output 10 Figure 1.6: GDP by expenditure 11 Figure 1.7: Gross fixed capital formation and business confidence 12 Figure 1.8: Labor market developments 13 Figure 1.9: Changes in fiscal revenue and expenditure, 2018-2021 15 Figure 1.10: Debt-to-GDP ratio with contingent liability scenarios 16 Figure 1.11: Current account components 17 Figure 1.12: Potential and actual per capita GDP, history and forecasts 21 Contents | South Africa Economic Update 11 | i Figure 1.13: Growth projections for South Africa and other EMDEs 21 Figure 2.1: Inequalities in 101 countries, 2013 25 Figure 2.2: South Africa’s polarization 26 Figure 2.3: South Africa's human opportunities, 2015 28 Figure 2.4: Inequality measurement over time 28 Figure 2.5: Labor market status and skills increasingly contribute to inequality 29 Figure 2.6: Sectors' labor and skills intensity 30 Figure 2.7: Jobs, poverty, and inequality 31 Figure 2.8: Deciles’ labor incomes shares, 2015 and 2030, baseline scenario 40 Figure 2.9: Relative impacts of selected interventions on poverty, inequality, and GDP growth 45 Figure 2.10: Contributions to change in inequality with respect to the baseline scenario 47 TABLES Table 1.1: GDP growth (supply side) 8 Table 1.2: Baseline annual growth forecasts 22 Table 2.1: Intergenerational social mobility 27 Table 2.2: Mean hourly wages in $ by education, purchasing power parity 29 Table 2.3: Progress toward the Vision 2030 in different scenarios 46 Table 2.4: Labor market indicators in 2030 in different scenarios 46 Table 2.5: Selected macroeconomic indicators in 2030 in different scenarios 47 BOXES Box 1.1: The impact of global monetary conditions on South Africa 5 Box 1.2: The exchange rate and South Africa’s integration into the global economy 18 Box 2.1: Building assets for the poor through the pension system 32 Box 2.2: Inequalities and political rights demand larger governments in Sub-Saharan Africa 33 Box 2.3: Education flows and the distribution of skilled labor incomes across household deciles 39 ii | South Africa Economic Update 11 | Contents ACKNOWLEDGEMENTS The 11th edition of the South Africa Economic Update was produced by a World Bank team comprising Marek Hanusch, Precious Zikhali, Victor Sulla, Vincent Dadam, Gerard Kambou, David Stephan, Charl Jooste, Mokgabo Molibeli and Zandile Ratshitanga, led by Sébastien Dessus. It was internally peer reviewed by Dino Merotto (Lead Economist, Jobs Groups) and Maurizio Bussolo (Lead Economist, Eastern and Central Asia Chief Economist Office). It benefited from comments from Konstantin Makrelov (Chief Director, National Treasury), Duncan Pieterse (Chief Director, National Treasury) and overall guidance from Mathew Verghis (Practice Manager, Macroeconomics, Trade and Investment) and Paul Noumba Um (Country Director for South Africa). Chapter 2 draws heavily on two World Bank reports. The first report, the Poverty and Inequality Assessment, was conducted in collaboration with Statistics South Africa and the Department of Planning, Monitoring, and Evaluation. It comprehensively reviews the trends and determinants of poverty and inequality in South Africa for the first time since 1996, with a strong focus on the role of labor markets. The second report, the forthcoming Systematic Country Diagnostic, was developed in deep consultation with several government counterparts, including the National Planning Commission and the National Treasury. It aims to selectively identify the most binding constraints to poverty alleviation and inequality reduction, and how to lift these through public intervention. The report was edited by Clarity Editorial and designed by Cybil Maradza. Acknowledgements | South Africa Economic Update 11 | iii FOREWORD I am pleased to launch this 11th edition of the South Africa Economic Update, which offers a review of the country’s recent economic and social developments and its outlook in the context of global economic prospects. Since the previous Economic Update of September 2017, a number of important events have improved South Africa’s economic outlook. The smooth transition in power, the authorities’ reaffirmed adherence to good governance and fiscal consolidation, and an upward revision in national accounts are all contributing to strengthen citizens and business confidence in South Africa’s future. These recent developments, combined with the strong rebound in the world economy, provide South Africa now with a unique opportunity to progress towards its National Development Plan’s goals of eradicating poverty and reducing inequality by 2030. Most observers, including the World Bank, have been revising their growth projections for 2018 and 2019 upwards. But deep challenges remain. This Update reviews the evolution and nature of South Africa’s inequality – the highest in the world– arguing that it has increasingly been driven by labor market developments that demand skills the country’s poor currently lack. Since democracy, social assistance and fiscal redistribution have more generally played a fundamental role in containing the rise in inequality. But the slow growth that generates a mismatch between labor demand and supply makes fiscal redistribution alone grossly insufficient to address the country’s inequalities. Solutions to break out of the mutually reinforcing cycle of low growth and high inequality lie in taking bold actions to giving poor South Africans better access to good jobs. Simulations done in this Update suggest that increasing the skilled labor supply among poor households (through improved education and spatial integration) and labor demand (mainly through strengthened competition) could bring the number of poor people in South Africa down from more than 10 million today to 4 million by 2030. In doing so, the country would strengthen its social contract, where the political rights gained with democracy are met with people sharing in the nation’s wealth. This Update builds on our solid partnerships with the National Treasury, Statistics South Africa, the National Planning Commission, and the Department of Planning, Monitoring, and Evaluation. As the World Bank, we stand ready to work with all stakeholders and support South Africa to fulfill its development agenda and contribute to ending extreme poverty and promoting shared prosperity. It is our hope that the country will continue to use the World Bank’s knowledge, global experience, and convening power as a platform for peer-to-peer learning to identify evidenced- based, pragmatic solutions that can contribute towards achieving the National Development Plan’s goals. Paul Noumba Um World Bank Country Director for South Africa iv | South Africa Economic Update 11 | Foreword ABBREVIATIONS EMDEs Emerging markets and developing economies GDP Gross domestic product OECD Organisation for Economic Co-operation and Development PMI Purchasing Managers’ Index R South African rand SACU Southern African Customs Union StatsSA Statistics South Africa SME Small and medium-sized enterprise $ United States dollar Abbreviations | South Africa Economic Update 11 | v EXECUTIVE SUMMARY vi | South Africa Economic Update 11 | Executive Summary South Africa’s economic outlook has improved. A rise in inequality in South Africa, on which policy interventions confidence in early 2018 and the recent upward revision of could further build. Previously, inequality was largely national accounts for the period 2015 to 2017 suggest that determined by race and geographical origin (reflecting the country is recovering from a difficult 2015 and 2016, the country’s legacy of exclusion). While race remains which marked the end of the super-commodity cycle and a central determinant of inequality, income inequality severe drought. Gross domestic product (GDP) growth is is now increasingly being determined by jobs status: projected to gather pace, increasing from 1.3 percent in employed versus unemployed, skilled versus unskilled. 2017 to 1.4 percent in 2018, 1.8 percent in 2019, and 1.9 Since 1995, wage inequality has risen sharply, reflecting percent in 2020. This in turn would contribute to a broader a severe mismatch between a labor market that rebound among commodity exporters, emerging markets demands skills and a labor force that is not fully able and developing economies, and overall global growth. to respond to such demand, as mostly unskilled and Although it provides little space for fiscal stimulus, the often located far away from economics centers. This 2018 Budget Review’s reaffirmation of the government’s is concerning as it maintains inequality at such high commitment to debt stabilization objectives is expected levels that fiscal redistribution alone cannot reduce. to generate more private investment. But it is also a trend against which citizens and the government can now act more forcefully through But South Africa remains constrained by its low growth efforts and policy initiatives, as opposed to intangible potential. Slow private investment growth and weak factors like race. As a matter of fact, World Bank integration into global value chains prevent the poverty projections indicate that progress in access country from reaping the new economic opportunities to education since democracy is paying off: by 2030, emerging around the globe, and from catching up inequality should be back down to its 1994 level, and with living standards in peer economies. South Africa South Africa should count 8.3 million poor people (at needs to build on its comparative advantages, that $1.90 a day), down from almost 10.5 million in 2017. of an industrial skilled economy, to develop new domestic and international markets through higher But the number of poor people could be brought productivity and innovation. At this condition will down further, to 4 million by 2030, through selected South Africa reduce its high dependency on commodity policy interventions. They include, in the short term, price movements, which do not look favorable for the continuing to address corruption, getting free higher country in the medium term. education right, restoring policy certainty in mining, improving the competitiveness of strategic state- Building on two World Bank reports – the Poverty owned enterprises, further exposing South Africa’s and Inequality Assessment and the forthcoming large conglomerates to foreign competition, and Systematic Country Diagnostic – this 11th edition facilitating skilled immigration. And, in the longer term, of the South Africa Economic Update argues that improving the quality of basic education delivered to significantly raising South Africa’s economic potential students from poor backgrounds and reinforcing the will require breaking away from the equilibrium of spatial integration between economic hubs, where jobs low growth and high inequality in which the country are located, and underserviced informal settlements. has been trapped for decades. In this equilibrium, The first set of reforms would raise labor demand and slow growth and high inequality reinforce each create the fiscal space needed to eventually build labor other: inequality fuels the contestation of resources supply from the poor population through education (through taxation, expropriation, corruption and and spatial integration. The analysis in this report crime), which discourages the investment needed to suggests that these reforms would reinforce each accelerate job creation and reduce inequality. Fiscal other to generate significant positive effects on growth, redistribution through social assistance, while sizeable inequality, and poverty overall. And as inequalities and effectively targeted, has been unable to redress decline, the social contract would strengthen and likely the rise in inequality since 1994, and is increasingly encourage further private investment – a possibility constrained by narrowing fiscal space. Solutions are not captured in our projections. needed to foster inclusive growth, which in practice means improving the poor’s access to good jobs so Constructing this new South Africa will take time, and they can fully participate in the economy. A credible managing expectations will remain a challenge in a path to sustainably redress inequalities is needed to country where strong political rights combine with reduce policy uncertainty and strengthen the social high inequality to demand rapid transformation. In compact on which authorities plan to build consensus this regard, continued efforts to effectively redistribute with business, labor, and civil society. wealth to the poorest while protecting economic growth will need to complement the reforms discussed A silver lining in this very challenging social, political above to create skilled jobs for the poor. and economic environment is the evolving nature of Executive Summary | South Africa Economic Update 11 | vii CHAPTER 1 Recent Economic Developments Global Economic Developments The global economic recovery continues in early 2018 Global economic activity remains solid. Global PMI to hit a 40-month high at the start of 2018 output expanded by an estimated 3 percent in the (Figure 1.1 A). fourth quarter of 2017.¹ This was substantially above the growth potential of the global economy, Momentum in the global goods trade continues. but weaker than in previous quarters, as growth Despite moderating in the fourth quarter of 2017, moderated in the United States, the Euro area, the global goods trade continues to grow, supported China, and other large emerging markets and by the recovery in manufacturing activity and developing economies (EMDEs). However, global investment. In 2017, growth in the global trade of industrial production accelerated in November goods averaged 4.3 percent, nearly three times the and December, and the global manufacturing pace observed in 2016, and up from an average of 2.6 Purchasing Managers’ Index (PMI) was close to a percent over the last five years. New manufacturing seven-year high in January 2018. A further rise in export orders in January indicate that this services sector activity led the global composite momentum will continue in 2018 (Figure 1.1 B). Figure 1.1: Global activity indicators A. Global GDP growth and global composite PMI B. Global trade in goods growth and manufacturing export orders Percent Index Percent, 3m-3m saar Index 5 56 12 58 10 56 4 8 54 6 54 4 3 52 2 52 0 50 2 -2 -4 48 1 50 -6 46 2012 2013 2014 2015 2016 2017 2018 Jan-15 Apr-15 Jul-15 Oct-15 Jan-16 Apr-16 Jul-16 Oct-16 Jan-17 Apr-17 Jul-17 Oct-17 Jan-18 Global GDP growth Composite PMI (RHS) Global goods trade New export orders (RHS) Source: World Bank staff calculations. A. Global growth in 2017 Q4 (orange bar) is estimated using available country-level data from national sources. Values for PMI above 50 indicate expansion. Last observation is January 2018. B. Values for PMI new export orders above 50 indicate expansion. Last observation is January 2018. The economic recovery in advanced economies remaining steady at a 17-year low of 4.1 percent, strengthened at the start of 2018. U.S. growth and average hourly earnings growth increasing moderated toward the end of 2017, but still maintained to 2.9 percent (year-on-year), demonstrating the a robust quarterly pace of 2.6 percent. Labor market strongest annual gains since 2009. Growth in the conditions were strong in January 2018, with nonfarm Euro area moderated slightly to 2.3 percent in the payrolls rising by 200,000, the unemployment rate last quarter of 2017, following a marked rebound ¹ Unless otherwise indicated, all quarterly growth rates in this report are seasonally adjusted annualized rates. Chapter 1 | South Africa Economic Update 11 | 2 in previous quarters. High-frequency indicators firm up. Recent data point to a continued upturn in suggest a strong start to 2018, with the composite commodity-exporting EMDEs, apart from Russia, PMI nearly reaching a 12-year high in January. Growth where activity decelerated toward the end of the in Japan slowed to 0.5 percent in the fourth quarter year, with quarterly growth in retail sales and of 2017, down from 2.2 percent in the third quarter, industrial production contracting. In Brazil, retail but private consumption and exports strengthened, sales and industrial production are growing. Policy and industrial production remained firm toward the interest rates were cut in February, extending an end of the year. easing cycle that started in October 2016. Nigeria’s recovery continued, with a pickup in consumer In China, economic activity indicators point to confidence and a rise in oil production in December. resilient growth, increasingly led by services. The manufacturing and nonmanufacturing PMIs Growth moderated to 6.3 percent in the last quarter declined slightly in January, but remain elevated, of 2017, bringing overall growth for the year to 6.9 suggesting steady momentum in 2018. Recovery was percent. Recent data suggest a gradual slowdown also observed in Angola as the political transition in 2018, accompanied by a continued shift from boosted consumer and business confidence. manufacturing to service activity, entailing, over Growth also picked up in several large commodity- time, less demand for metals. The nonmanufacturing importing EMDEs as domestic headwinds eased, PMI increased in January to its highest level since apart from Turkey, where growth likely decelerated September 2017, while the manufacturing PMI fell in the fourth quarter of 2017. In India, activity to 51.3 in January – the lowest result since May 2017. continued to recover from the temporary adverse effects of the goods and service tax, which came Economic activity in other major emerging into effect in July 2017. markets and developing economies continues to Figure 1.2: Global financial indicators A. Flows to EMDE equity and bond funds B. EMDE bond spreads and stock market index US$ billions, 4 week sum Basis points Index, 100=Jan. 2, 2017 35 420 150 25 380 140 15 5 340 130 -5 300 120 -15 -25 260 110 Jun-16 Sep-16 Dec-16 Mar-17 Jun-17 Sep-17 Dec-17 Sep-17 Oct-17 Nov-17 Dec-17 Jan-18 Feb-18 Equities Bonds Bond spreads Equity index (RHS) Source: World Bank staff calculations. A. Last observation is February 14, 2018. B. Bond spreads are defined as bond yield spreads over U.S. government securities issued by sovereign and quasi-sovereign entities in emerging market economies (in $). Equity index shows the MSCI emerging market index. The vertical line corresponds to the February 2, 2018, release of U.S. employment and wage growth data. Last observation is February 20, 2018. Global financial market volatility spiked amid rising with a continued rise in U.S. long-term yields, driven U.S. inflation expectations and bond yields. Following by rising inflation expectations and prospects of faster a prolonged period of low and stable long-term yields in normalization of U.S. monetary policy. Following the advanced economies, rallying global equity prices, and country’s stronger-than-expected wage growth, U.S. compressed volatility, financial markets were turbulent and global equity markets tumbled in February, erasing in the first half of February 2018. This was associated year-to-date gains. The effect on U.S. and global equity 3 | South Africa Economic Update 11 | Chapter 1 prices was amplified by ongoing concerns about with portfolio flows to bond and equity mutual funds overstretched valuations. Bond spreads and credit surging in January (Figure 1.2 A) and international bond default swaps increased, but remained close to the sales reaching an all-time high of $71 billion. EMDE low levels seen throughout 2017. A favorable global markets were affected by the global sell-off in early economic backdrop likely helped prevent a broader February, particularly corporate bond funds. Sovereign reassessment of credit risks. bond spreads have risen, although they remain low (Figure 1.2 B). Bond issuance moderated in February, Capital flows to EMDEs remain resilient. EMDE with a few countries including Kenya and Nigeria, financial markets started the year on a strong note, returning to the capital markets. The global outlook remains positive but is not without risks for EMDEs The broad and solid global economic expansion commodity-exporting EMDEs is projected to pick up observed in 2017 is expected to continue in 2018 from 1.8 percent in 2017 to 2.7 percent in 2018, and to and 2019. Global growth is estimated to have reached 3.1 percent in 2019, as headwinds gradually moderate. a stronger-than-expected 3 percent in 2017, a notable Growth in commodity-importing EMDEs is projected recovery from a post-crisis low of 2.4 percent in 2016. to remain robust, at 5.7 percent in 2018 and 2019, In 2018, global growth is projected to edge up to 3.1 underpinned by solid export growth. percent as the cyclical upturn in advanced economies continues and EMDE growth strengthens. The risks to the global outlook are becoming more balanced, mainly due to the possibility of stronger- Global financial conditions and commodity prices than-expected growth in the largest advanced are expected to stabilize in 2018. Global financing economies and EMDEs. However, downside risks conditions are likely to tighten in 2018 as monetary remain. A sudden increase in borrowing costs, triggered policy normalizes in major advanced economies. Both by a reassessment of the pace of advanced-economy energy and metal prices are expected to level off in monetary policy normalization or concerns about 2018 (Figure 1.4) after posting significant gains in 2017, asset valuations, could lead to severe financial stress while agricultural prices remain stable. and disrupt capital flows to EMDEs. Escalating trade restrictions could derail the recovery in trade. Over the Growth in EMDEs is projected to rise to 4.5 percent longer term, a more pronounced slowdown in potential in 2018 and 4.7 percent in 2019 as activity recovers growth in both advanced economies and EMDEs would further in commodity-exporting countries and make the global economy more vulnerable to shocks remains robust in commodity importers. Growth in and worsen prospects for improved living standards. Chapter 1 | South Africa Economic Update 11 | 4 Box 1.1: The impact of global monetary conditions on South Africa The world’s natural real rate of interest (the rate in the propensity to invest in developed countries needed to equalize the global supply of savings with the pushed the real rate even lower, to negative rates in global demand for investment) has been declining for some Organisation for Economic Co-operation and the past few decades, led by two connected events. At Development (OECD) countries. the beginning of the millennium, preferences (including relatively accommodative U.S. monetary policy) and In comparison with global trends, the decline in South explicit policies in Asian emerging countries increased Africa’s natural real rate of interest was delayed, as the the supply of global savings, leading to a reduction savings investment gap widened from the early 2000s in the natural rate across developed countries. This (See Box 1.1 Figure 1). During this period, foreign direct was one of the causes behind the financial boom investments surged with the commodity boom, while and the subsequent global financial crisis. After the domestic savings stayed low. financial crisis, an increase in savings and a reduction Box 1.1 Figure 1: Savings – investment gap in South Africa 40 35 30 25 20 15 Global financial crisis 10 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Investment % of GDP Savings % GDP Various estimates of South Africa’s natural real interest same time, the delayed adjustment in the savings- rate suggest that it only started to decelerate after the investment gap led to a significant accumulation of financial crisis (See Box 1.1 Table 1). This adjustment foreign liabilities: South Africa’s external debt to gross was reflected in lower net foreign direct investments, national income rose from 19 percent in 2000 to 51 as domestic savings remained depressed. At the percent in 2016. 5 | South Africa Economic Update 11 | Chapter 1 Box 1.1 Table 1: Estimates of South Africa’s natural rate of interest Hodrick- Christiano- Hamilton Leubach and Structural Prescott Fitzgerald Forecasting based Williams (2003) Estimation - Filter Filter Methods (2017) Semi-Structural Bjørnland, Estimation et al 2011 1990-1999 5.3 5.3 3.1 2.2 4.2 2000-2007 5.4 5.4 5.3 3.1 5.1 2008-2017 0.7 0.6 1.2 1.8 0.8 Source: World Bank (2018b). As an increasing number of advanced economies save more, it may also limit the expected rebound normalize monetary policy, global interest rates are in private investment, particularly given that higher expected to rise. While this, along with improved interest rates will make it more expensive for South policy certainty, may encourage South Africans to Africa to service its external debt. Real Sector Developments in South Africa Despite a modest rebound, South African growth continues to lag behind the rest of the world While global growth accelerates, the South African forecasts have been revised upward throughout the economy has been gathering steam slowly. In 2017, year, a growth of 1.3 percent for the year beat the primary sectors were the main drivers of growth, most recent consensus of about 1 percent. This was particularly in the agricultural and mining sectors. largely due to significant methodological revisions Momentum in other sectors has been weak. This means by Statistics South Africa (StatsSA), which date back that South Africa is diverging from global growth. This to at least 2015. The strong upward revision for the is largely because the country’s main exports continue first quarter of 2017 erased a technical recession (two to be commodities – but they are raw materials that consecutive quarterly contractions in GDP) that had are not highly sought-after internationally. Except previously been recorded, although quarterly GDP still for parts of the services sector, South Africa is only contracted in the fourth quarter of 2016. weakly integrated into global and regional value chains, meaning that it has limited opportunities to On the supply side, growth in agriculture was benefit from global growth. The business cycle has revised down, especially for 2016 and 2017, making been gaining momentum since late 2017 and business the downturn from the drought more pronounced and consumer sentiment improved in early 2018. This in 2016 but keeping the agricultural recovery in 2017 may herald the return of investment that the country strong. Upward revisions mainly focused on the needs to make its firms more competitive, transfer services sector and, to a lesser extent, manufacturing. technology, join global supply networks, and continue The main revisions came from much higher growth in overcoming its historical isolation from the world the finance, real estate, and business services sector, economy (World Bank 2018b). This issue is further accounting for 43 percent of the revision in 2017 discussed in Chapter 2. (Figure 1.3 A). Trade, catering, and accommodation saw the second-largest upward revision in 2017, by Growth in 2017 exceeded expectations. Although 17 percent. This is reflected in demand (Figure 1.3 B), Chapter 1 | South Africa Economic Update 11 | 6 with private consumption being significantly revised consumers appear to have spent more on domestic upward in 2017, suggesting increased momentum in goods. Exports, on the other hand, were weaker in 2017 household spending. In addition, imports were revised than previously thought. downward significantly, also helping headline GDP, as Figure 1.3: StatsSA’s GDP revisions – fourth versus third quarter 2017 releases (difference in seasonally adjusted annualized GDP, R billion, constant 2010 prices) A. Supply B. Demand 30 50 30 10 10 -10 -10 -30 2015 Q1 2015 Q2 2015 Q3 2015 Q4 2016 Q1 2016 Q2 2016 Q3 2016 Q4 2017 Q1 2017 Q2 2017 Q3 2015 Q1 2015 Q2 2015 Q3 2015 Q4 2016 Q1 2016 Q2 2016 Q3 2016 Q4 2017 Q1 2017 Q2 2017 Q3 General government services Discrepancy Finance, real estate and business services Imports of goods and services Transport, storage and communication Exports of goods and services Trade, catering and accommodation Change in inventories Construction Manufacturing Gross fixed capital formation Mining Agriculture, forestry and fishing General government Households Total Total Source: StatsSA and World Bank staff calculations. Despite the revisions, the impact of recent average of 27.8 percent in the previous two quarters), developments has not fundamentally altered. suggesting that the sector has largely recovered and will Although improving somewhat faster than anticipated, not add a similar boost to growth going forward (other growth was still low in 2017. In per capita terms, the than parts of the country still affected by drought). economy stagnated, providing no additional income Policy uncertainty was heightened in the sector when that could help reduce poverty. Moreover, growth was the Parliament voted to review the Constitution to still principally led by agriculture. Rebounding from possibly make it easier to expropriate land without the drought, the sector grew by 17.7 percent in 2017, compensation – although with a disclaimer that this following a contraction of 6.4 percent in 2015 and 10.2 should not undermine food security. Under such percent in 2016 due to drought. This recovery contributed circumstances, additional investment in agriculture may 0.4 percentage points to headline GDP. Without this have become less likely (even though the proposal to rebound, the economy would have grown by 0.9 percent. amend the Constitution is not new). This would further Although agriculture still grew by 37.5 percent in the amplify the diverging trend observed since 2010, fourth quarter of 2017 (Table 1.1), it only expanded by whereas investment growth in agriculture was lower 1.1 percent in year-on-year terms (compared with an than in other sectors (1.7% vs. 2.2% annually). 7 | South Africa Economic Update 11 | Chapter 1 Table 1.1: GDP growth (supply side) (quarter-on-quarter percentage change, seasonally adjusted) General government forestry, and fishing Finance, real estate, and communication Trade, catering, and Transport, storage, Personal services accommodation Electricity, gas, Manufacturing and business Construction Agriculture, Mining and and water quarrying services services 2013 Q1 -0.1 14.1 -7.3 -4.4 0.9 1 3.8 4 1.8 2.4 2013 Q2 2.3 -4.6 12.4 3.2 8.9 2.3 3.1 4.4 4.2 3.7 2013 Q3 9.8 12.1 -6.7 2.2 3.2 0.3 4.6 2.1 4.1 2.2 2013 Q4 11 17 13 -5.6 6.3 1.9 3.1 1.8 5.9 2.1 2014 Q1 3.8 -22.9 -5.3 -2.1 3.1 2.2 2.3 2.6 0.9 1.5 2014 Q2 4.5 -3.1 -4.1 1.5 1.4 -0.6 5 2.5 3.8 1.5 2014 Q3 9.2 3.7 -1.2 -1.7 1.5 3.8 3.3 3.5 2.4 1.4 2014 Q4 7.6 14.3 8.2 1.8 2.6 -0.2 3.7 4.5 1.5 0.8 2015 Q1 -11.5 11.9 -2.3 2.6 2.4 4 0.2 2.1 -0.6 0.9 2015 Q2 -20.4 -7.2 -6.4 -7.9 1.4 0.1 0.2 1.6 0.7 1 2015 Q3 -11.9 -9.9 4.6 -8 1.1 1.9 0 1.8 1 1.3 2015 Q4 -6.9 2 -2.6 0.1 1.3 3.4 -1.6 1 1 0.7 2016 Q1 -12.3 -20.3 1.2 -3.4 1.2 2.2 0.4 2.7 1.6 1.7 2016 Q2 -11.4 16.1 8.2 -1.1 0.5 1.9 2.8 3.5 1.5 2.2 2016 Q3 -4.3 5.6 -2.8 -0.4 1.2 -1.8 2.3 2.4 2.2 1.8 2016 Q4 -4.1 -9.9 -2.5 4.6 0.9 2.5 3.1 2.6 1.2 1.3 2017 Q1 26.2 12.6 -4.1 -5.6 -1.2 -5.6 -1.3 -0.5 -0.5 0.3 2017 Q2 36.8 7.8 2.9 8.1 -0.7 0.9 2.6 3.1 -1.9 1.7 2017 Q3 41.1 6.2 3.7 -6.1 -1.2 -0.1 0.8 1.9 1.1 1.2 2017 Q4 37.5 -4.4 4.3 3.3 -1.4 4.8 2.8 2.5 1.4 1 Source: StatsSA Although there was heightened policy uncertainty conduct a similar analysis for the third Mining Charter, around the third Mining Charter released in early but anecdotal evidence suggests that major mining 2017, mining was a major contributor to growth. investments remain on hold and mining houses are not Short-term fluctuations in mining output can be linked investing as much in South Africa as they are elsewhere. to commodity prices. But, in the longer term, higher The charter was taken to court by the Chamber of Mines, output requires additional investment, which has but President Ramaphosa and his new Cabinet have been hampered by policy uncertainty. The World Bank since improved relations with the chamber, which may (2018b) suggests that the second Mining Charter (2010) improve sentiment for future investment and increase strongly muted the investment response to higher output on a more sustainable basis. This issue is further global demand and prices. There is insufficient data to discussed in Chapter 2. Chapter 1 | South Africa Economic Update 11 | 8 Favorable commodity prices supported mining are projected to weaken over time, but platinum prices output in 2017, but this recovery may be temporary. are on the rise as global demand strengthens. The Coal prices rebounded because China cut its production impact of recent diesel scandals and decisions taken in in 2017. But demand for coal is expected to drop as many European cities to ban diesel vehicles (platinum the world moves toward greener technologies for is used as a catalytic converter in diesel engines) environmental reasons (see Figure 1.4). Similarly, iron could nonetheless weaken such positive prospects. ore prices were buoyant in 2017, given supply shortfalls Overall, mining grew by 4.6 percent in 2017, the second- in Australia and Brazil, but are projected to decline strongest performer after agriculture, adding 0.3 from 2018 (World Bank 2017d). They may, however, percentage points to GDP. Mining production continued not decline as much for South Africa because South to grow by 2.4 percent in January 2018, driven by iron African iron ore is relatively high quality and has been ore and other metallic minerals, and with a marked trading at a higher margin since late 2016. Gold prices contraction in platinum group metals. Figure 1.4: Commodity price forecasts ($ constant, index 2015=100) 170.0 150.0 130.0 110.0 90.0 70.0 50.0 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Coal Iron ore Gold Platinum Crude oil Source: World Bank (2017d). Manufacturing contracted by 0.2 percent in 2017. sector’s prospects are still relatively positive due to The sector gained momentum toward the end of the a favorable global environment and the change in year – accelerating its performance from a decline of political leadership. The manufacturing PMI rose above 4.1 percent in the first quarter of 2017 to 4.3 percent 50 (signaling expansion) in February, with a marked growth in the fourth quarter. According to StatsSA, this acceleration in the expected business conditions strong performance in the last quarter was driven by subcategory, where the index improved from 50 in food and beverages, petroleum, chemical products, November 2017 to 79.1 in February 2018. Readings on rubber and plastics, as well as various metals products purchasing commitments and new sales orders have and machinery. Performance was sustained in January also had a marked improvement since November. If 2018, with manufacturing production growing by 2.5 this improvement is sustained, it could mark the end percent year-on-year, again driven largely by food and of years of stagnation in manufacturing. Relaxing the beverages (especially sugar and processed goods). skills constraint, investing in technological upgrades To an extent, this growth can still be considered part and integrating South Africa into regional and global of the rebound from the historical drought, which is value chains will play a critical role to sustain this expected to wear off in 2018. But the manufacturing momentum, as further discussed in Chapter 2. 9 | South Africa Economic Update 11 | Chapter 1 Finance, real estate, and business services, South Transport, storage, and communication and personal Africa’s strongest growth sectors in the past, services were the only other sectors growing faster expanded by only 1.9 percent in 2017 – the lowest than 1 percent in 2017. Electricity, gas, and water grew rate since 2014. Yet they continued to remain the by 0.2 percent; construction contracted by 0.3 percent; strongest non-primary sectors in the economy. and trade, catering, and accommodation contracted In fact, finance and related sectors had the same by 0.6 percent. General government services only contribution to headline growth as agriculture, expanded by 0.3 percent as the public sector contained although agriculture, forestry, and fishing accounts expenditure growth in times of weak revenue collection for about 2.5 percent of GDP compared to about 20 and rising debt. percent for finance, real estate, and business services. The sector was unusually weak in the first quarter of On the demand side, growth has been held back 2017, contracting by 0.5 percent. It showed stronger by domestic factors since at least 2015 (Figure performance in the remaining quarters, but there was 1.5), including policy uncertainty, low business and no noticeable increase in momentum. A loosening consumer confidence, and supply constraints. The in monetary conditions in South Africa may further end of the commodity super-cycle in 2015 resulted support credit growth and help sustain the increase in falling prices for South Africa’s commodity exports in financial intermediation activity seen toward the that have undermined South African purchasing power, end of the year. On the other hand, greater political which has weakened growth. The global economy has certainty could reduce the volatility of the rand and contributed to South African growth since early 2017, affect income from hedging services – an important but at relatively modest levels. business line for South African banks (see Box 1.2). Figure 1.5: Historical decomposition of domestic output (percentage change) 2% 1% 0% -1% -2% -3% Jun-05 Jun-07 Jun-09 Jun-11 Jun-13 Jun-15 Jun-17 World demand shock Commodity specific shock Globalisation shock Domestic shocks Deterministic Domestic output Source: World Bank staff calculations. Note: World demand shocks are proxied by the export deflator, the import deflator, and trading partner output; domestic shocks are proxied by inflation, domestic output, and the exchange rate. Strengthening household consumption was the main in inflation, due to dissipating drought effects and a driver of growth in 2017. Private consumption rose by stronger exchange rate, put less pressure on household 2.2 percent in 2017, the highest recorded rate since budgets, allowing them to spend more in real terms. 2012, contributing 1.4 percentage points to growth Household credit growth remained relatively low in (while an accompanying increase in imports subtracted 2017, expanding below the rate of inflation. In January 0.6 percentage points to GDP growth). The moderation 2018, nominal consumer credit growth slowed to 3.7 Chapter 1 | South Africa Economic Update 11 | 10 percent, but picked up modestly in February, driven by confidence has improved markedly. Credit growth unsecured credit. Households remain indebted and, is more likely to support business investment when although debt-to-disposable-income ratios have been growth has been more buoyant (7.1 percent in January dropping, there is only so much room for taking on 2018). Government consumption grew modestly by 0.6 additional debt for consumption. However, consumer percent in 2017. Figure 1.6: GDP by expenditure (percentage change and contributions to growth, quarter-on-quarter seasonally adjusted annualized rate) 15 10 5 0 -5 -10 -15 2014 2014 2014 2014 2015 2015 2015 2015 2016 2016 2016 2016 2017 2017 2017 2017 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Private consumption expenditure Government consumption expenditure Investments Exports of goods and services Imports of goods and services GDP by expenditure approach Source: StatsSA. Investment recovered from its 4.1 percent expropriate land without compensation). South African contraction in 2016. Gross fixed capital formation 10-year bond yields have strengthened to 2015 levels. expanded modestly by 0.4 percent in 2017. Investment Stronger confidence is also reflected in various indices. returned to levels observed in 2014 but remains below The South African Chamber of Commerce and Industry’s the highs of 2015. It was sluggish in 2017 apart from business confidence index recovered in January and a relatively strong performance in the last quarter February to levels last seen in 2015 (Figure 1.7 B). The (Figure 1.7 A). According to StatsSA, the increase in this quarterly confidence index published by Rand Merchant quarter was largely the result of growth in acquisition Bank and the Bureau for Economic Research rose from of machinery and other equipment (up by 9.2 percent) 34/100 in the third quarter of 2017 to 45/100 in the first and transport equipment (up by 21.7 percent). Fixed quarter of 2018 – an unusually high increase. As with residential and nonresidential investment fell in the the PMI, this was again largely driven by optimistic fourth quarter, in line with weak performance in the expectations rather than actual conditions. Improved construction sector and a soft housing market, with investor sentiment may translate into higher investment more potential sellers than buyers. in 2018. However, whether higher portfolio flows are mirrored in higher fixed investment (in production Improved business confidence was sustained in capacity, for example) remains to be seen. According South Africa throughout the first quarter of 2018. to the Investment Tracker of the Manufacturing Circle, Between mid-December 2017 and mid-March 2018, the manufacturing firms are mainly looking to invest rand strengthened by 12.5 percent, reflecting improved in replacing or maintaining land and buildings and investor appetite for South African assets (although the expanding plant and equipment. Significantly higher rand lost some of this ground after Parliament voted spending is also expected in research and development to review the Constitution to possibly make it easier to activities in the sector. 11 | South Africa Economic Update 11 | Chapter 1 Figure 1.7: Gross fixed capital formation and business confidence A. Gross fixed capital formation (R billion, 2010) B. Business confidence (Index 100=2015) 660 102.0 100.0 650 98.0 640 96.0 630 94.0 620 92.0 90.0 600 88.0 590 86.0 580 84.0 2014 Q2 2014 Q3 2014 Q4 2015 Q1 2015 Q2 2015 Q3 2015 Q4 2016 Q1 2016 Q2 2016 Q3 2016 Q4 2017 Q1 2017 Q2 2017 Q3 2017 Q4 2016M01 2016M06 2016M11 2017M04 2017M09 2018M02 Source: StatsSA and South African Chamber of Commerce and Industry. Exports contracted by 0.1 percent in 2017, the toward the end of the year reflected relatively strong weakest performance since 2012. Even though performance in mining and manufacturing. Coupled exports had a strong rebound of 12.3 percent in the with the 1.9 percent increase in imports in 2017, net fourth quarter, over the year little momentum was exports reduced headline GDP growth for the year. notable. The strong quarter-on-quarter performance Labor Market Developments in South Africa High unemployment, among unskilled and young people in particular, remains an immense challenge Getting a job is the most promising pathway out of to 58.8 percent between the third and fourth quarters. poverty in South Africa. The official unemployment In addition, the number of employed people declined rate fell from 27.7 percent in the third quarter of 2017 to over that period, with a staggering 21,000 people losing 26.7 percent in the fourth quarter (Figure 1.8). However, their jobs. Consequently, an additional 503,000 people this decrease conceals a large exit in the labor force, were deemed economically inactive, with 102,000 with the participation rate dropping from 59.9 percent categorized as discouraged job seekers. Chapter 1 | South Africa Economic Update 11 | 12 Figure 1.8: Labor market developments (millions [LHS] and percent of labor force [RHS]) 25.0 40.0 35.0 20.0 30.0 15.0 25.0 20.0 12.9 13.8 14.6 14.9 15.0 15.0 15.2 15.0 15.1 14.6 14.9 15.0 15.5 10.0 15.0 10.0 5.0 14.6 4.2 14.2 4.4 13.8 4.6 14.1 4.6 14.4 4.8 14.9 4.9 15.1 5.1 15.7 5.3 15.8 5.8 16.2 6.2 16.1 6.2 16.2 6.2 16.2 5.9 5.0 0.0 0.0 Employed Unemployed Employed Unemployed Employed Unemployed Employed Unemployed Employed Unemployed Employed Unemployed Employed Unemployed Employed Unemployed Employed Unemployed Employed Unemployed Employed Unemployed Employed Unemployed Employed Unemployed 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2017 2017 2017 Q1 Q2 Q3 Q4 Economically inactive Official unemployment rate Broad unemployment rate Source: StatsSA and World Bank staff calculations. Young people are particularly affected by 2017 was 30,000 – at least 140,000 jobs were lost in unemployment. Although the unemployment rate the province in the second quarter alone. among the age group 15 to 24 years has declined since the second quarter of 2017, more than half of South Formal employment performance in 2017 improved Africans in this age category participating in the labor significantly following a disappointing 2016 – market are without jobs. The youth unemployment rate creating 22,000 jobs in 2017 on average, after shedding in the last quarter of 2017 was 51.1 percent – down 6,000 in 2016. However, after adding 186,000 jobs in the from 52.2 percent in the previous quarter. Similar third quarter of 2017, the formal sector shed another to the headline unemployment rate, this decrease 135,000 in the last quarter. Informal employment does not necessarily translate into gains in terms of showed strong performance. Overall, 28,000 jobs were employment. In fact, the labor participation rate for created in 2017, against an average of 2,000 in 2016. people between the ages of 15 and 24 years declined About 118,000 jobs were added in the informal sector from 26.8 percent to 25.9 percent between the third in the fourth quarter, after 71,000 were lost in the third. and fourth quarters. An additional 94,000 young people are now recognized as economically inactive. The mining and construction sectors were the worst At the same time, 17,000 jobs occupied by young people performers in terms of employment by sector, with were lost during the last quarter of 2017. an average contraction in employment in 2017 of 2.7 percent and 2.5 percent respectively. The construction The highest average growth rate in employment sector shows signs of a modest improvement, with in 2017 was recorded in the Northern Cape, with 25,000 jobs created in the last quarter of the year, employment in the province growing at 2.5 percent while the mining sector shed 35,000 jobs. Notably, on average after a contraction of 1.6 percent in 2016. higher mining output did not translate into jobs, as the The Western Cape created 26,000 additional jobs sector becomes increasingly capital intensive (World (net), followed by Mpumalanga with 21,000 jobs. Bank 2017a). The highest growth rates in 2017 were in Gauteng lags behind with an average contraction in the transport (1.3 percent) and community and social employment of 1.3 percent, while the net job loss in services (1.1 percent) sectors. Although employment 13 | South Africa Economic Update 11 | Chapter 1 in the agricultural sector contracted overall in 2017, an As in 2016, skilled employment grew the fastest, additional 39,000 jobs were added in the fourth quarter. at 2.8 percent in 2017, followed by semi-skilled jobs (2.7 percent). Unskilled jobs had the slowest growth With the effects of the drought dissipating, rate (1.8 percent). As further discussed in Chapter skilled agriculture displayed the highest recorded 2, this trend reflects a deep mismatch between employment growth in 2017 at 7.9 percent – adding an economy that demands skills and a labor force 14,000 jobs between the third and fourth quarters that is not fully able to respond to such demand, as alone. Overall agricultural employment remained mostly unskilled and often located far away from below 2015 and even 2016 levels. jobs centers. Fiscal Developments in South Africa The government renewed its commitment to stabilize debt, although at higher levels, and averted further rating downgrades The 2018 Budget put South Africa on a stronger • An increase in ad valorem excise duties for fiscal footing, following the October 2017 Medium luxury goods. Term Budget Policy Statement, which had announced a much wider budget deficit than foreseen. The • An increase in estate duty, targeting luxury estates higher deficit was largely the result of a sharp drop above R30 million with a 25 percent levy. in revenue collection and a weak economy. The public debt trajectory rose significantly, to 60 percent of GDP • Increases in the plastic bag levy, the motor vehicle by 2021/22, suggesting that the long-standing debt emissions tax, and the levy on incandescent light stabilization target had been abandoned. Standard and bulbs to encourage environmentally friendly choices. Poor’s downgraded South Africa’s creditworthiness further, following downgrades to sub-investment Other revenue measures include reviewing the (“junk”) status in April. A return to the debt stabilization VAT zero-rating of spending categories that are target was critical to restore market confidence and not consumed by the poor (such as rye bread or avert a downgrade by Moody’s to sub-investment nontraditional beer). A sugar tax (called the health grade, which was expected to result in significant promotion levy) will also be implemented. Health tax capital outflows from index-tracking funds. The 2018 credits (mostly used by the rich) will be curbed. Budget achieved this. Between 2018 and 2021, most of additional revenue New revenue measures were introduced in the 2018 is expected to come from VAT, followed by higher Budget, and expenditure has been reprioritized, to corporate tax revenue due to stronger economic accommodate new spending priorities, notably higher growth (Figure 1.9 A). Top income tax rates have not education (an additional R57 billion over three years) been increased since being adjusted in the 2017 Budget. and drought relief (R6 billion) for areas of the country Projected higher customs tax collection also means still affected (such as the Western Cape). To stem the higher transfers to the other members of the Southern deterioration in revenue, the 2018 Budget proposes six African Customs Union (SACU). main tax measures: Reprioritizing expenditure means significantly lower • An increase of 1 percentage point in VAT, increasing public investment spending, both at the national level the tax to 15 percent. and through provincial and municipal infrastructure grants (Figure 1.9 B). State-owned enterprises will also • No adjustments for inflation for the top four receive less support from government, including the income brackets, combined with below-inflation infrastructure for which these entities are responsible adjustments for the bottom three brackets. (like the South African National Roads Agency Limited, responsible for roads, and the Passenger Rail Agency • An increase in both general fuel (22c/liter) and of South Africa, responsible for rail). Tertiary education Road Accident Fund (30c/liter) levies. received a higher spending allocation to phase in fee- Chapter 1 | South Africa Economic Update 11 | 14 free education for poor and working-class students. university (see Box 2.3), the large reallocation to higher Interest rate payments are expected to be lower due to education will not benefit many of South Africa’s poor. a reduction in the risk premium, consistent with falling borrowing costs and renewed investor confidence Although main budget expenditure was reduced, following changes in the South African government. overall spending is still likely to be higher than under the Medium Term Budget Policy Statement Social grants were increased by R3 billion, above the because a higher contingency reserve was included, rate of inflation. In addition, many of the items poor which past experience suggests it will be spent. The households consume are zero-rated. In this sense, this contingency reserve has increased from R5 billion is a progressive budget, mostly financed by the rich. But, in previous budgets to R8 billion (and R10 billion in given that only 12 percent of poor students qualify for 2020/21). Overall, the 2018 Budget is expansionary. Figure 1.9: Changes in fiscal revenue and expenditure, 2018 – 2021 A. Sources of additional tax revenue, B. Reprioritization of expenditure, 2018 – 2021, compared to 2017 MTBPS (%) 2018 – 2021, compared to 2017 MTBPS (%) 3.0% 2.0% 2.5% VAT 1.5% Other Higher 2.0% 1.0% Households education 1.5% 0.5% 1.0% 0.0% Corporate tax Goods and Sub-national services 0.5% -0.5% Personal income tax 0.0% -1.0% Capital assets SACU -0.5% -1.5% Source: National Treasury and World Bank staff calculations. Notes: (a) Expenditure excludes additional contingency reserve; (b) MTBPS = Medium Term Budget Policy Statement; (c) Expenditure on households refers to transfers under the National Student Financial Aid Scheme. The 2018 Budget estimates a budget deficit of 4.3 Figure 1.10 shows, contingent liabilities that materialize percent of GDP in 2017/18. Considering new revenue would affect the debt trajectory. State-owned entities measures, the deficit is expected to decrease to 3.6 would default on debt that matures. Remaining liabilities percent in 2018/19 and in 2019/20, and to 3.5 percent mature, but the total amount of sovereign guarantees in 2020/21. Higher revenue and assumed lower interest would not be triggered immediately. In the case of service costs result in a primary deficit of 0.2 percent default, about 0.6 percent of GDP would be added to in 2018/19, 0.1 percent in 2019/20, and 0 percent by national government financing requirements annually 2020/21. This will help decelerate debt dynamics, in the medium term. If additional defaults occur, called expected to stabilize at 56.2 percent in 2021/22. “unexpected losses”² in Figure 1.10, annual additional financing requirements could reach 1.2 percent of GDP. Contingent liabilities in state-owned entities remain In addition, the estimated costs of contingent liabilities a fiscal risk. The debt-to-GDP ratio would increase that materialize would be even higher if the overall between 1 percent and 2 percent of GDP in the medium macro and financial situation further deteriorates term if contingent liabilities materialize, and will become from projections in the 2018 Budget’s medium-term one of the major driving forces of debt accumulation. As expenditure framework. ² Calculated as one standard deviation of losses of a binomial distribution. 15 | South Africa Economic Update 11 | Chapter 1 Figure 1.10: Debt-to-GDP ratio with contingent liability scenarios 59 58 57 56 55 54 53 52 2017 2018 2019 2020 Debt to GDP ratio - baseline projection Debt to GDP ratio - expected losses from contingent liabilities Debt to GDP ratio - expected and unexpected losses from contingent liabilities Source: Bachmair and Bogoev (2018). The StatsSA revisions to GDP, including nominal GDP, Other possible reasons for lower tax buoyancy include have several implications. The drop in tax buoyancy increased tax avoidance³ or the emigration of wealthy raised as a concern in the 2017 Medium Term Budget individuals. But with higher GDP the fiscal deficit and Policy Statement (and countered with additional public debt as a percentage of GDP have both been revenue measures in the 2018 Budget) is likely to reduced by about 0.1 percentage points. Although a have been worse than expected – this means that an minor difference, when coupled with stronger 2016 and increase in nominal GDP is associated with even lower 2017 growth this does suggest that South Africa’s fiscal revenue collections than believed. If the weak buoyancy position is stronger than previously anticipated. All in is due to an administrative deterioration in the South all, the Budget was strong enough to convince Moody’s African Revenue Service (currently under review), the not to downgrade South Africa to sub-investment institutional damage could be worse than expected. grade in March, and move the rating outlook to stable. Inflation and Monetary Policy in South Africa Inflation has eased and provided room for looser monetary policy Reduced inflation pressures have helped reduce important contributing factor to the improved inflation policy rates, as the inflation rate is now well within outlook is the appreciation of the rand because of strong the South African Reserve Bank’s target band. commodity prices, and rebounding investor confidence Consumer price inflation dropped from 6.6 percent in and capital inflows since the ANC elective conference in January 2017 to 4.4 percent in January 2018, and fell December 2017. Core inflation also moderated. further to 4 percent in February. There are several factors that account for this decline. Lower food and These factors influenced the South African fuel price inflation played a major role, particularly as Reserve Bank’s inflation forecast, which was the country started to recover from the drought in 2015, revised down for 2018 and 2019. However, further the effects of which continued to be felt in 2016. Another increases in the international oil price and the ³ Tax avoidance is legal and can be aided by tax advisors to minimize fiscal payments as allowable by law. Tax evasion is illegal. Chapter 1 | South Africa Economic Update 11 | 16 implementation of a considerable number of revenue The South African Reserve Bank cut the policy rate measures announced in the 2018 Budget will put by another 25 basis points in March 2018, bringing pressure on inflation. The increase in VAT is expected the policy rate to 6.5 percent. There may be room for to add 0.6 percentage points to headline inflation additional cuts considering benign inflation and lower within a year, somewhat offset by a stronger rand.4 risks. In March 2018, the Monetary Policy Committee The South African Reserve Bank expects consumer also noted that risks had dissipated, notably with price inflation to average 4.9 percent in 2018 and 5.2 respect to both food price inflation and potential percent in 2019. capital outflows from rating downgrades, following the Moody’s decision in March not to downgrade Inflation expectations are high but remain within South Africa to sub-investment grade. However, the the target range, following a marginal decrease Monetary Policy Committee did identify a faster-than- from 5.8 percent to 5.7 percent, according to a survey expected tightening in US monetary policy as a risk, conducted by the Bureau for Economic Research in the and expressed concern about a potential trade war fourth quarter of 2017. However, average expectations involving the US. While there may be room for further for 2019 and five-year inflation expectations hold at cuts, relatively high inflation expectations would keep 5.9 percent and 5.6 percent respectively. the South African Reserve Bank more cautious. The External Sector in South Africa South Africa’s weak linkages with global value chains reduce the scope for a boost in exports The rand strengthened significantly following the factors also contributed to the recent appreciation: a election of Cyril Ramaphosa as the president of the weak dollar against major currencies and the positive ANC in December 2017, and subsequently as the 2017 rebound in the global price of the main mineral president of South Africa in February 2018. Two other commodities exported by South Africa. Figure 1.11: Current account components (R billion) 50 0 -50 -100 -150 -200 -250 -300 2013 Q1 2013 Q2 2013 Q3 2013 Q4 2014 Q1 2014 Q2 2014 Q3 2014 Q4 2015 Q1 2015 Q2 2015 Q3 2015 Q4 2016 Q1 2016 Q2 2016 Q3 2016 Q4 2017 Q1 2017 Q2 2017 Q3 2017 Q4 Merchandise trade balance Service balance Income balance Source: Haver Analytics and World Bank staff calculations 4 Statement of the Monetary Policy Committee – South African Reserve Bank, March 2018. 17 | South Africa Economic Update 11 | Chapter 1 South Africa’s current account deficit narrowed to (especially non-commodity products) struggle to keep 2.5 percent of GDP in 2017, from 3.1 percent in 2016, pace with global exports. in line with World Bank expectations in the previous Update. The trade balance remained in surplus Portfolio investment continues to dominate the throughout 2017 for the first time in many years. Both financial account. In rand terms, foreign direct investment imports and exports picked up in the fourth quarter, into South Africa declined by 46.3 percent between 2016 resulting in a lower trade surplus overall, while and 2017, with a marked drop in the fourth quarter of 2017. dividend receipts declined and payments increased. Portfolio inflows doubled, remaining the largest financing Weak performance in exports is a structural risk to item in the financial account. Reserve assets strengthened the current account balance, as South African exports from mid-2017, reaching $50 billion in February 2018. Box 1.2: The exchange rate and South Africa’s integration into the global economy A competitive exchange rate allows South Africa to Rand volatility. South Africa has one of the most seize opportunities in the global economy. But the volatile currencies in emerging markets. This is partly rand is chronically undervalued and highly volatile. due to the country’s structural external vulnerability This poses problems for the country: and rising external liabilities, as described in Box 1.1. According to the International Monetary Fund, Rand undervaluation. South African productivity is not the rand-dollar exchange rate is largely driven by catching up with the rest of the world.5 This results commodity prices, and domestic and foreign shocks.7 in a depreciating real exchange rate, particularly in Policy uncertainty in recent years has been a major manufacturing and mining. The undervalued currency driver of rand volatility. This makes the returns from reflects the country’s poor global integration, the level trading and cross-border investment expensive, of protection of the economy, and policy uncertainty requiring firms to hedge, which in turn poses that deters investment. It entails low innovation from potential constraints to investment and trade that technology transfer through global value chains and could help South African firms diversify into other foreign direct investment, and relatively low imports globally traded goods. It may also deter investment of technology-intensive capital goods. A divergence in that could destabilize cartels or support technology productivity between South Africa and the world puts transfer. When interest rates in advanced economies the rand on a depreciating trajectory in real terms, are very low, the costs of hedging are raised, making imports more expensive for both consumers especially over the long term (and in particular for and firms. In addition, World Bank research suggests smaller firms). that the real exchange rate is persistently undervalued6 – an observation that is mirrored in purchasing power Regarding foreign investment, World Bank research parity conversion rates, since, according to the OECD, on a sample of 80 developing countries between the exchange rate that would equalize price levels 1990 and 2015 suggests that reducing exchange rate across South Africa and the United States was R5.90 to volatility by 10 percent over one year could boost the dollar in 2016. The spot exchange rate was R14.70, foreign direct investment inflows by an estimated however. Terms-of-trade effects play a role in this, 0.48 percentage points of GDP.8 The same reduction especially commodity prices, but a more structural over the past five years could boost inflows by 0.27 underlying undervaluation can be linked to the percentage points over the long term. Reducing the differences in productivity in the tradable manufacturing rand’s volatility to that of developing country peers and services sectors. A lack of competitiveness of South could potentially boost South Africa’s foreign direct African manufacturing also explains the weak export investment inflows by about 0.25 percentage points response to real depreciations, further supporting of GDP (see Box 1.2 Figure 1). persistent current account deficits. 5 World Bank (2017c). 6 Nguyen (2018). 7 Maveé et al. (2016). 8 Note, however, the difficulty in clearly delineating cause and effect. Chapter 1 | South Africa Economic Update 11 | 18 Box 1.2 Figure 1: Potential impact of lower rand volatility on foreign direct investment to South Africa 0.08 1-year exchange rate volatility 0.07 0.13% of GDP 0.06 increase in FDI 0.21% 0.05 of GDP 0.23% 0.24% of GDP (2016) increase in FDI of GDP 0.04 increase in FDI increase in FDI 0.03 0.02 0.01 0.00 South Africa Mexico Turkey Korea, Rep. Indonesia Source: Algu et al. (2018). Note: Bars depict currency volatility; bubbles depict potential impact on foreign direct investment to South Africa by lowering rand volatility to volatility levels of peers. The Outlook for South Africa A cyclical rebound is expected, but higher potential growth will require ambitious structural policies The South African economy’s performance is As discussed previously, coal prices are expected to improving. The year 2016 was a trough in the business decline as countries, especially economic giants like cycle and, although growth in 2017 was largely driven China, switch to greener sources of energy. China’s by agriculture’s drought recovery, momentum is being continued transition from an investment- to a gained. The South African Reserve Bank’s business consumption-driven economy also results in stagnant cycle indicators have been improving, lending further iron ore prices. Platinum demand is not expected to support to the economy’s recovery. Business and be buoyant, although it is still projected to increase consumer sentiment has improved. The 2018 Budget from $950 in 2017 to $1,056 by 2020. Gold prices are has signaled to the markets a return to the debt likely to decline as interest rates go up throughout the stabilization target. The economy is growing – the developed world, although global uncertainty shocks question is, how fast? may result in temporary price increases. Rising oil prices will put pressure on firms and consumers – the Tailwinds from the agricultural recovery are World Bank (2017d) projects oil prices to increase from expected to taper off in 2018: by the end of 2017, $53 a barrel in 2017 to $60 by 2020. agricultural production already exceeded the pre- drought production peak in the fourth quarter of 2014. If there is an amicable resolution to the third Drought continues to linger in parts of the country, Mining Charter, the World Bank estimates that including in the Western Cape. Mining may provide investment in the sector may increase by 25 further support to the economy, but this will depend percent (see Chapter 2). Beyond this, much of the on favorable commodity prices. Even though the recovery will depend on manufacturing and services. commodity outlook has improved in general, South South African manufacturing is closely linked to Africa’s raw materials are not highly sought-after. domestic consumption, which will provide some 19 | South Africa Economic Update 11 | Chapter 1 support to the sector, and to global demand through The extent to which investment will improve in the mining value chain. The World Bank (2018b) 2018 and beyond is difficult to predict. In mining, argues that South African manufacturing remains there are several projects (especially in coal) that relatively uncompetitive globally. Whether South have been shelved due to legislation related to African manufacturing can recover and improve the the third Mining Charter. These projects could be country’s growth performance will depend on policies implemented quickly and translate into higher that support competition and competitiveness, growth, as discussed in Chapter 2. Beyond this, investment in research and development, and the effect of the boost to investor certainty on links between South Africa and global and regional actual investment remains to be seen. The World value chains, including the associated foreign direct Bank (2018b) suggests that South Africa’s drop on investment and knowledge transfer. An improved Transparency International’s corruption-control climate for investors increases the likelihood of these indicator between 2001 and 2016 reduced investment improvements taking place. by JSE-listed firms by 10.5 percent over that period. Beyond strengthening institutions, investment may The World Bank (2018b) argues that finance is South also return if the President succeeds in strengthening Africa’s strongest sector. It has a long history of the social compact while accelerating structural intermediating capital in global markets, has benefited reforms. Yet, an element of investor prudence will from reforms in the 1990s, and has adopted global remain ahead of the 2019 elections, given policy financial standards. The sector is expected to continue uncertainty on land reform, the Mining Charter, driving the country’s growth performance: lower policy intellectual property rights, and the information and rates may support an increase in the provision of credit communications technologies regulatory framework domestically, South African companies will continue (it is good news that outstanding purchasing power to seek domestic finance to expand abroad, and a agreements for independent producers were signed growing world economy offers plenty of opportunities in April). In addition, government is cutting back for profitable deals. The second services sector likely to on public investment. Overall, modest investment contribute to growth is retail and wholesale, supported growth is projected between 2018 and 2019, although by growth in household consumption. it could quickly gather steam if strong political will translates into reforms. As consumption, exports, and Looking at demand, household consumption is investment strengthen overall, imports also improve. likely to continue to be the main driver of growth. Inflation is expected to be benign, barring further Overall, GDP is expected to grow at 1.4 percent in shocks from drought or domestic or global politics. As 2018, 1.8 percent in 2019, and 1.9 percent in 2020, the economy is still catching up, supply-side pressures reaching GDP potential in 2020 (Figure 1.12). This are expected to remain moderate. Low inflation will timing is in line with South Africa’s business cycle, support household budgets and provide room for the which tends to stretch across seven years. It is a South African Reserve Bank to loosen monetary policy, conservative estimate, assuming projected potential which will strengthen the credit cycle – especially growth of 1.4 percent, in line with past performance. as household indebtedness has reduced. A stronger The South African Reserve Bank’s leading indicator economy is also likely to contribute to wage growth, rose from 105.8 in December 2017 to 106.1 in January supporting consumption. However, consumers will 2018, supporting the case for further improvement in be affected by the revenue measures set out in the the business cycle in the first half of the year, albeit 2018 Budget. The rich will be hit particularly hard, a modest one. Critically, this optimism will need to while higher transfers provide some relief to the poor. be translated into better business conditions and Overall, fiscal consolidation will dampen consumption investment. Decisive structural reforms, discussed in growth. Government consumption is projected to Chapter 2, could shift this trajectory, bringing potential stagnate in real terms, in line with the 2018 Budget. growth above 2 percent over the medium term. Such Exports are expected to pick up, but only modestly. reforms could ensure that South Africa does not further This is due to relatively soft commodity prices and the fall behind the growth performance of its peers (Figure time it takes to improve the competitiveness of South 1.13) and seizes opportunities to grow the economy for Africa’s export sectors. jobs and higher incomes for its citizens. Chapter 1 | South Africa Economic Update 11 | 20 Figure 1.12: Potential and actual per capita GDP, history and forecasts 59,000 57,000 55,000 53,000 51,000 49,000 47,000 45,000 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Potential GDP GDP Potential GDP (reform scenario) Source: World Bank staff calculations. Note: The increase of potential GDP in the reform scenario is based on simulations described in Chapter 2. If implemented, higher potential growth would also mean higher actual growth. Figure 1.13: Growth projections for South Africa and other EMDEs 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 2017 2018 2019 2020 Emerging markets and developing economies South Africa Source: World Bank (2018c). South Africa’s public debt is expected to stabilize South Africa’s current account deficit is expected over the medium term. A strengthening economy to remain at just over 2 percent of GDP – lower will support government’s debt stabilization target, than in the past but still requiring foreign financing although it is important to acknowledge the significant that adds to South Africa’s external liabilities. The pressures that redistribution places on public spending underlying trade balance is expected to deteriorate in a highly unequal economy like South Africa’s, somewhat, with stronger import pressures and as Chapter 2 will show. Strong political resolve to sluggish exports. The income balance, the main drag sustainably accommodate these pressures will be on the current account balance, is expected to remain needed to stick to the target. negative. But an improved political climate may result 21 | South Africa Economic Update 11 | Chapter 1 in the increased repatriation of profits, especially as South put pressure on the skilled labor force, increasing African operations abroad (including in Africa) mature. inequality, providing incentives to firms to further substitute labor for capital, and further undermining Unemployment is expected to reduce modestly. Job the competitiveness of South African firms. Progress seekers leaving the labor force in a weak economy may in higher education and skilled migration (especially be drawn back into it when the economy improves, in the short to medium term) plays a critical role in while others are unable to participate, as further addressing the skills constraint. Poverty is projected discussed in Chapter 2. Overall, this results in a relatively to decline, although these results will remain modest stable unemployment rate, even if employment until the economy gathers steam beyond the rate of growth accelerates. Given South Africa’s chronic population growth. Structural policies will be central skills constraint, a recovering economy may quickly to achieving this. Table 1.2: Baseline annual growth forecasts 2015 2016 2017 2018e 2019f 2020f Real GDP growth, at constant market prices 1.3 0.6 1.3 1.4 1.8 1.9 Private consumption 1.8 0.7 2.2 2.1 2.0 2.1 Government consumption -0.3 1.9 0.6 -0.3 1.2 1.4 Gross fixed capital investment 3.4 -4.1 0.4 3.0 3.4 2.8 Exports, goods, and services 2.8 1.0 -0.1 1.5 2.2 2.1 Imports, goods, and services 5.4 -3.8 1.9 3.3 3.1 2.8 Real GDP growth, at constant factor prices 1.3 0.8 1.1 1.4 1.8 1.9 Agriculture -6.4 -10.2 17.7 4.0 1.7 1.8 Industry 1.1 -1.3 1.5 1.6 1.9 2.1 Services 1.7 2.0 0.4 1.2 1.8 1.7 Inflation (consumer price index) 4.6 6.3 5.3 4.9 5.1 5.4 Current account balance (% of GDP) -4.4 -3.1 -2.5 -2.1 -2.2 -2.1 Financial and capital account (% of GDP) 4.4 3.1 2.5 2.1 2.2 2.1 Fiscal balance (% of GDP) -3.7 -3.5 -4.3 -3.6 -3.6 -3.5 Debt (% of GDP) 49.0 50.7 53.3 55.1 55.3 56.0 Primary balance (% of GDP) -1.0 -0.5 -1.2 -0.2 -0.1 0.0 Source: World Bank staff calculations. Note: e = estimate, f = forecast. These growth projections are conservative. The to redress inequalities, and by the large contingent confidence boost, if sustained, could result in higher liabilities in state-owned enterprises. President investment and consumption than projected, especially Ramaphosa’s summits on investment, jobs, and social if this investment translates into research, development, issues hold significant potential to forge consensus and value chain integration. However, overconfidence around implementable policy solutions that can bolster in the country’s ability to deliver difficult, long-term growth and South Africa’s resilience to shocks, be they reforms in a relatively short period of time is a risk. And domestic or foreign. Chapter 2 discusses some of these the authorities’ ability to redress fiscal accounts will potential policy solutions, aimed at encouraging higher be challenged by continuous demand for redistribution inclusive growth. Chapter 1 | South Africa Economic Update 11 | 22 CHAPTER 2 Jobs and Inequality This chapter builds on two World Bank official It aims to identify the most binding constraints to reports. The first one, the Poverty and Inequality alleviating poverty and reducing inequality, and Assessment (World Bank 2018a), was conducted in how government could address these constraints. collaboration with StatsSA and the Department of Building on the findings of these two reports, this Planning, Monitoring, and Evaluation. It provides a chapter discusses policy options to reduce inequality comprehensive review of the trends and determinants by creating more and better jobs for the poorest 40 of poverty and inequality in South Africa, with a strong percent of South Africans. As beyond the scope of focus on the role of labor markets. The second one, the this Update, it does not, on the other hand, discusses Systematic Country Diagnostic (World Bank 2018b), comprehensively options to reduce wealth inequality, entailed a deep consultation process with government except in Box 2.1. entities, including the National Planning Commission. South Africa Remains Trapped in a Cycle of High Inequality and Slow Job Creation Inequalities remain extremely high despite effective redistributive policies South Africa is one of the most unequal countries in the same level of resource) to 100 (all resources held the world. Measured by income or consumption, the by one individual) – is higher in South Africa than in all Gini coefficient – measuring the distance to perfect other countries for which comparable data exists, and equality, ranging from 0 (all citizens enjoying exactly by a significant margin (Figure 2.1). Chapter 2 | South Africa Economic Update 11 | 24 Figure 2.1: Inequalities in 101 countries, 2013 Ukraine — Slovenia + Income survey Norway — Slovak Republic + Czech Republic - Consumption survey Kazakhstan — Belarus - Average Gini for all countries Kosovo Iceland — Finland — Sweden - Belgium — Netherlands — Moldova — Kyrgyrz Republic — Albania — Serbia - Denmark + Iraq - Germany — Tajikistan Austria - Hungary + Pakistan — Cambodia — Armenia + Switzerland — Mongolia — Montenegro - Mauritius — Croatia - Ireland - United Kingdom — Poland — France - Estonia + Ethiopia Guinea Bosnia and Herzegovina Niger Sierra Leone Cyprus + Luxembourg + Romania — Lithuania - Italy + India Burkina Faso Latvia — Mauritius - Spain + Bulgaria - Portugal - Greece + Iran. Islamic Rep. — Vietnam — Tanzania — Thailand — Lao PDR + Bhutan - Sri Lana — Indonesia Georgia — Turkey + Senegal + Uganda — United States - Russian Federation - Uruguay — Congo. Dem. Rep China - Argentina — Micronesia. Fed. Sts. Madagascar Philippines — Zimbabwe Chad Benin El Salvador — Djibouti Peru — Togo + Cameroon + Seychelles Nicaragua + Dominican Republic — Ecuador — Bolivia — Paraguay + Guatemala — Congo. Rep. Mexico + Costa Rica - Rwanda - Chile — Panama — Brazil — Columbia — Honduras - Haiti South Africa — 20 30 40 50 60 Gini index Source: World Bank (2016a). 25 | South Africa Economic Update 11 | Chapter 2 South Africa’s levels of inequality reflect its also illustrated by the fact that less than a fourth of the polarized society, with a small elite, a large class South African population did not experience any spell of poor people, and a relatively small middle class. of poverty between 2008 and 2015 (World Bank 2017a), It has the highest polarization index in the world, which and could be considered middle class or elite over the measures the economic difference between “poles” same period. In comparison, close to 80 percent of – concentrations of population according to their the Mauritian population is classified as middle class economic welfare (income or consumption). This is (Figure 2.2). Figure 2.2: South Africa’s polarization A. Polarization indices across countries B. Class sizes, South Africa (2008–2015) (2006–2015) South Africa 0.37 Mexico 0.28 Elite Russia 0.26 US 0.22 Middle 4% UK 0.22 Italy 0.21 class Israel 0.22 20% Hungary 0.21 Poland 0.20 Australia 0.20 France 0.20 Vulnerable Chronic poor Canada 0.19 China 0.19 14% 49% Germany 0.18 Czech Republic 0.19 Netherlands 0.18 Transient Denmark 0.18 Belgium 0.18 poor Norway 0.18 13% Luxembourg 0.18 Sweden 0.17 Finland 0.17 Source: World Bank (2018a). This divide reflects the country’s legacy of racial As a result, intergenerational social mobility exclusion. As detailed in the forthcoming Systematic continues to be low in South Africa. Two-fifths of Country Diagnostic (World Bank 2018b), exclusion all sons born to very poor fathers – those in the first under regimes of segregation and apartheid quintile – will occupy the bottom 40 percent of their manifested itself in labor markets, suppressing generation’s income distribution. Sons of rich fathers – black (meaning here black African, Indian, and those in the fifth quintile – have a 43 percent chance of colored) South Africans’ access to work in many also being in the top quintile of their income distribution sectors; land and freedom of movement; social (Table 2.1). Racial differences may still be a major factor protection, education, health, and infrastructure; in low intergenerational mobility, but they are not the and finance and the ability to build wealth. To this only reason: education, jobs, and internal migration day, historically disadvantaged South Africans strongly affect chances of upward social mobility (World hold fewer assets, have fewer skills and poorer Bank 2018a). This is consistent with the observation health, are still more likely to be unemployed, and, that, since democracy, education levels and labor status if employed, earn lower wages. (participation, occupation, and so on) have become the main determinants of inequality, as discussed below. Chapter 2 | South Africa Economic Update 11 | 26 Table 2.1: Intergenerational social mobility Son Quintile Father Quintile 1 2 3 4 5 1 22.77 19.43 21.22 20.38 16.19 2 15.98 16.61 21.21 22.52 23.67 3 13.19 15.44 20.35 22.25 28.77 4 10.86 14.67 16.79 22.70 34.97 5 9.88 12.42 14.12 20.88 42.71 Source: World Bank (2018a). Public Interventions since democracy have helped disparities persist. The Human Opportunity Index redistribute public resources to the poor. The measures the extent to which the provision of a social wage – government’s investment in education, given public service can redress lack of access to health services, social assistance, public transport, such services due to race, gender, family background, housing, and local amenities – accounts for close or any other personal circumstances beyond a to 60 percent of government expenditure and has child’s control and considered by society to be an played a notable role to reduce poverty and inequality. unjust source of exclusion. In the last decade, most Measuring the redistributive nature of South Africa’s indices in the Human Opportunity Index improved in budget, the World Bank (2014) concluded that fiscal South Africa, reflecting a decrease in inequality of policy (a progressive tax system combined with a opportunity, although the country remains among well-targeted social protection system) in South the most unequal in the world. The country achieved Africa reduces the Gini coefficient by 18 points. More near-universal access to primary education, a recently, the Poverty and Inequality Assessment necessary first step for equalizing opportunities (World Bank 2018a) confirmed these findings, among children. Analysis of matric performance suggesting that social assistance alone (mainly for 2002, 2009, and 2016 reveals that the number child support and old age grants) contributes to of black African learners performing at a level in reducing inequality by 10.5 points (compared with mathematics that would allow them to study, for a situation where households would not receive example, engineering at university increased by 65 such assistance).9 Financial inclusion also strongly percent over the whole period.10 In 2002, more than contributed to reducing inequality in consumption half of high-level mathematics performers in the since 1994. But this conceals a persistently high public examination system were white. By 2016, level of wealth inequality in the country, which is over two-thirds were black. The fast rise in access aggravated when considering the financial liabilities to telecommunications, electricity, sanitation, and of the poorest households. The World Bank (2018a) school infrastructure also improved opportunities estimates that 71 percent of national net wealth for children in South Africa. However, despite these belonged to 10 percent of the population in 2015. improvements, persistent disparities in access to quality basic and higher education, and the ability to Public interventions have also contributed to finish primary school on time, continue to constrain reducing unequal access to opportunities, but progress in this area. (Figure 2.3). 9 The World Bank (2018a) suggests that receiving social assistance has a marginal negative effect on labor force participation. 10 Van der Berg and Gustafsson (2017). 27 | South Africa Economic Update 11 | Chapter 2 Figure 2.3: South Africa’s human opportunities, 2015 100 90 80 70 60 50 40 30 20 10 0 Adequate infrastructure at school Access to electricity Water of good quality Improved water School Attendance (6 to 11 years) School Attendance (12 to 15 years) Started primary on time School Attendance (18 to 25 years) Adequate teachers Finish 7th grade Finish tertiary grade Access to telecommunications Improved sanitation Water on site No severe overcrowding Refuse removal service No environmental problems Has health insurance Education access Education quality Living conditions Source: World Bank (2018a). Consumption inequalities have grown since 1996, gaps widen between the poor and a small emerging at the expense of the bottom 40 percent of the middle class, and narrow between the middle class population in particular. Between 1996 and 2015, the and rich households. Inequality by source of income Gini coefficient of consumption inequality rose from (grants and others vs. labor income) in turn suggests 61 to 63, peaking at 65 in 2006 (World Bank 2018a). that labor incomes for the middle class are growing, In other words, it is very likely that, without public while the bottom 40 percent of the population intervention, inequality in South Africa would have continues to experience poor access to jobs and been even higher. Since 2006, levels of consumption slowly growing grants, leading to slow growth in inequality have slightly declined, but the nature of the poor’s consumption compared with richer groups the inequality has been significantly evolving, as (Figure 2.4). Figure 2.4: Inequality measurement over time A. Consumption growth incidence, 2006-15 B. Income shares by sources, 2006 and 2015 5 60% 64% 34% 22% 13% 6% 5% 3% 4.5 Annual growth rate (%) 94% 95% 97% 4 78% 87% 3.5 66% 3 40% 36% 2.5 2006 2015 2006 2015 2006 2015 2006 2015 2 1.5 1 0.5 Bottom 40th-75th 75th-90th Top 10 0 40 Percentile Percentile 0 10 20 30 40 50 60 70 80 90 100 Expenditure percentiles Work Grants and others Source: World Bank (2018a). Chapter 2 | South Africa Economic Update 11 | 28 The labor market is central to reducing inequality Labor market developments have become the main income countries such as Bangladesh and in high- drivers of income inequalities. A detailed breakdown income countries such as Austria (World Bank 2018a). of the various factors affecting inequality suggests Skilled workers earn nearly five times the average that education and labor market status have become wage offered to unskilled workers (see Table 2.2), primarily responsible for overall inequality (Figure yet constitute less than a fifth of the total working 2.5 A). Unlike race or geographical location, which population. As discussed later in this chapter, the high were the primary determinants of inequality at the skills premium in South Africa puts skills-intensive onset of democracy, labor status and education sectors (such as manufacturing) at a disadvantage in levels can be changed through public interventions. international competition, unless these sectors can Wage inequality remains extremely high in South reach productivity levels commensurate with higher Africa, with wages matching remuneration in low- wage levels.11 Table 2.2: Mean hourly wages in $ by education, purchasing power parity Brazil Chile Columbia Ecuador Indonesia Mexico South Africa No education 2.4 4.4 1.8 2.6 1.3 2.0 2.1 Primary education 3.1 4.7 2.5 3.2 1.9 2.5 3.4 Secondary education 3.8 5.5 3.1 3.9 2.0 3.3 5.7 Tertiary education 7.9 11.5 6.9 .. 3.2 6.3 11.6 Source: World Bank (2018b). Computations conducted using the International Income Distribution Dataset. Strongly differentiated employment rates across for semi-skilled and 75.3 percent for skilled workers. skills groups combine with high wage inequality to The probability of participating in labor markets and explain the large influence of labor status on income finding a job is strongly linked to individuals’ education inequality. In 2012, the employment rate of unskilled levels (World Bank 2018a). workers was 34 percent, compared with 51.5 percent Figure 2.5: Labor market status and skills increasingly contribute to inequality A. Factors of inequality, 2006–2015 B. Wage inequality, 1995-2014 70 8% 10% 68 44% 50% 66 Gini coefficient 64 48% 40% 62 60 2006 2015 58 Race 56 Education and employment status Location and other demographics 1995 2005 2010 2014 Source: World Bank (2018a). 11 See also World Bank (2017c) for a comparison of salaries between South Africa, India and Malaysia for a selected set of skilled jobs, including ICT developers, engineers, etc. This comparison confirms the high skills premium in South Africa. 29 | South Africa Economic Update 11 | Chapter 2 Wage inequality has increased since 1995, reflecting in capital and skills intensity in most South African a growing mismatch between demand for skilled economic sectors since 1994. Employment growth labor and an excess supply of unskilled labor. was lower than GDP growth, reflecting a shift from Between 1995 and 2014, the real wage Gini coefficient labor to capital and a change in the skills composition – measuring dispersion in wages in South Africa – of the labor force, raising labor productivity (Figure increased from 58 to 69 (with rapid growth from 2005, 2.6 A). Between 1994 and 2015, the number of formal when the real wage Gini coefficient was still at 59, see unskilled and semi-skilled jobs in the South African Figure 2.5 B). This trend was reinforced by an increase economy contracted (Figure 2.6 B). Figure 2.6: Sectors’ labor and skills intensity A. Sectors’ GDP and employment growth, 2000–2016 B. Employment growth by skills, 1995–2015 Average annual employment growth (%) 2000-2016 6 500% CONT FIN 400% 5 300% 4 CSP 200% 3 TRS 100% 2 0% UTI WRT 1 PHH -100% 0 -200% MAN Total Electricity CSP Agriculture Mining Manufacturing Construction Trade Transport Financial services Domestic Services -1 MIN AGR -2 -3 -2 -1 0 1 2 3 4 5 6 7 Average annual GDP growth (%) 2000-2016 Skilled Semi-skilled Unskilled Source: World Bank (2018a). Note: Notes: AGR = Agriculture; MAN = Manufacturing; MIN = Mining; WRT = Wholesale and Retail Trade; TRS = Transport; PHH = Private Households; UTI = Utilities; CSP = Community, Social, Personal Services; FIN = Financial Services; CONT = Construction. Unsurprisingly, getting a good job holds the most the probability of escaping poverty by 10 percent. At promise for escaping poverty (at the individual the national level, increasing employment would level) and reducing inequalities (at the collective have a significant, but not necessarily large, effect level). A household is 21 percent more likely to move on inequalities: creating 1 million jobs would reduce out of poverty if its employment income increases the Gini coefficient by 0.8 to 1.5 points, depending on as a share of total income. Finding a job results in a the sector where jobs were created. This is because 19 percent increase in the likelihood of moving out individuals with the highest probability of being hired of poverty. A change in job skill levels also increases (based on their characteristics) for these new jobs the chance of movement by 8 percent. However, an are not necessarily among the poorest job seekers increase in the share of children in a household lowers (World Bank 2017a). Chapter 2 | South Africa Economic Update 11 | 30 Figure 2.7: Jobs, Poverty and Inequality A. Factors to escape poverty B. Change in Gini from job creation 0 19% -0.2 21% % change in gini, by sector -0.4 8% -0.6 -0.8 -11% -1 -1.2 -1.4 Increase Find Change job Change -1.6 work employment skill level share of 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 income children Increase in employment, millions share Source: World Bank (2018a). Note: bullets in Figure B reflect the different economic sectors. Future progress in the reduction of poverty and higher taxation, borrowing, or by reducing the provision inequality through social assistance is limited by of other public services would generate the additional the current low fiscal space. As discussed in Chapter growth needed to protect fiscal sustainability. Besides, 1, fiscal space in the national budget is constrained by given high levels of household indebtedness, little slow growth and high debt levels. Potential gains from reduction in inequality can be expected from higher improving efficiency in the social assistance system private borrowing, particularly when considering that would be limited, because the system is already well the poorest households have been using borrowing targeted with good coverage (more than 17 million to finance consumption rather than to build assets recipients in 2016). This means that providing more (World Bank 2018b). While redistribution continues to support can only be achieved through higher individual be critical to contain inequality, it increasingly needs to grants. Only stronger economic growth could provide be done in a way that stimulates inclusive GDP growth. such space, and it is doubtful that raising grants through Box 2.1 suggests an option in this respect. 31 | South Africa Economic Update 11 | Chapter 2 Box 2.1: Building assets for the poor through the pension system There are about 32 million people of working age in South spatial economy. Higher property taxes would need Africa. Roughly one-third are covered by an occupational to be phased in incrementally to not overwhelm pension scheme at any given time. This leaves the the middle class. An agreement with municipalities unemployed, informal sector workers, and those not in could facilitate the use of a portion of property tax for the labor force uncovered by a contributory pension. The national expenditure. elderly poor draw on old age grants – a social pension system that most countries have moved beyond. Unlike The World Bank modeled the progressive introduction pensions, grants are not assets. Reconstruction and of a 10-percentage-point increase in the taxation of Development Programme houses and, to an extent, real estate consumption by raising property taxes redistributed or restituted land are the only meaningful (representing an additional fiscal revenue of about 0.6 form of asset transfer in South Africa that may reach the percent of GDP annually), saved in a pension fund. The poor. But they reach few beneficiaries, and sometimes revenue collected would fully subsidize the pension lock poor South Africans into unproductive areas. system for the bottom 40 percent of the population. Extending a contributory pension system to all South The poor would be able to draw a dividend from the Africans and subsidizing it for the poor (gradually pension, allowing them to participate in the economy. reducing the subsidy with income) would help the poor The conversion of consumption into forced savings build assets. As pension assets tend to be invested in would raise investment by 5.2 percent by 2030 and the JSE, it would be broad-based empowerment, giving real GDP by 1.2 percent, lifting an additional 2.9 million all citizens a stake in corporate South Africa. It would South Africans out of poverty. The redistributive effects be important to allow these assets to be registered as would be significant: poverty and the Gini coefficient black-owned on the JSE. In addition, delegating voting would drop to 8.2 percent and 58.1 respectively by rights to pension fund managers – perhaps themselves 2030, down from 12.7 percent and 59.5 in the baseline from historically disadvantaged backgrounds – would scenario discussed in the next section. empower poor people and give them a collective voice in shareholder decisions. An accumulation over a 40-year period with reasonable net returns (assumed here to be 3 percent One way of financing a social contributory pension above inflation) would result in a balance that should could be through a property tax (which is immobile and be sufficient to generate a pension of about twice thus not prone to evasion), which would finance assets the current value of the old age grant, adjusted for for the poor with a tax on the wealthy’s assets. Given inflation. Higher pensions could be generated through that property wealth is often inherited, this would additional voluntary savings in the same account. The also help redress historical inequality. It would also pension could also include provisions for survivor’s provide incentives to reduce home sizes, contributing insurance benefits purchased from the market on a to the much-needed densification of South Africa’s competitive basis. Source: World Bank (2018b). More inclusive, labor-intensive growth is needed. violence), and policy uncertainty (World Bank 2018b). The relationship between growth and inequality, Growing investor uncertainty in the face of strong and its effects, remains inconclusive.12 But it is likely political demand for redistribution of wealth has that severe inequalities in South Africa are affecting contributed to a private investment slowdown in economic growth, leading to contested resources, South Africa (World Bank 2017b and Chapter 1 of this fueling fragility (crime, corruption, gender-based document). Although there is no strong evidence 12 Conceptually, the effect of inequality might go either way: if higher inequality leads to the more rapid accumulation of savings (as richer households typically save more), it may spur growth; if it leads to suboptimal investment in education or health care, it may have a negative effect on growth. But empirical studies remain inconclusive. Motivated by such ambiguity, a recent set of papers decomposes overall inequality into components that may be especially harmful to growth. In particular, it may be expected that inequality of opportunity is harmful for growth, while the effect of inequalities that arise from differences in effort may act in the opposite direction. There is some evidence (in United States and Brazil) that inequality of opportunity may be bad for growth, at least subnationally. Across countries, however, there is no robust evidence of a negative effect (World Bank 2016a). Chapter 2 | South Africa Economic Update 11 | 32 of the negative effects of fiscal redistribution on redistribution reduces the Gini coefficient by 13 points growth overall (International Monetary Fund 2014a), or more – a threshold South Africa now exceeds. As there is a threshold above which fiscal redistribution highlighted in Box 2.2, the combination of inequality negatively affects growth, disturbing the balance and political rights since democracy contributes to between positive effects (demand multipliers, high the high demand for redistribution. As such, it is a returns on investments in education, improved deeply rooted structural factor that is difficult to health, and enhanced social cohesion) and negative tackle in isolation. But it points to the need to develop effects (reduced savings, innovation efforts, and job-related policy interventions that simultaneously investment). The International Monetary Fund (2014a) stimulate growth and reduce inequalities to generate suggests that this threshold is reached when fiscal sustainable and substantial results. Box 2.2: Inequalities and political rights demand larger governments in Sub-Saharan Africa A recent World Bank econometric analysis aims economies of scale (thus lower demand), and to understand differences in government sizes large populations exhibit more heterogeneous (measured based on public expenditure as a share of preferences for public goods and therefore GDP) in Sub-Saharan Africa, with a focus on Southern agree on lower demand for public goods. African Customs Union countries (Botswana, Lesotho, Namibia, South Africa, and Swaziland). • Ethnic and other forms of social fractionalization, Lesotho and Namibia have the largest and second- reflecting different preferences for public goods largest governments respectively among the 37 (and thus lower public spending). countries considered, while the other three Southern African Customs Union countries rank among the • High inequality combined with high political seven largest governments in Sub-Saharan Africa. rights leading to high demand for fiscal redistribution. Government size could be influenced by the following structural factors: • Electoral rules and government types, whereby majoritarian and presidential regimes require • Trade openness, whereby greater openness less expenditure to acquire political power than would be associated with greater exposure to proportional and parliamentarian regimes. external shocks and a related demand from citizens for greater social protection. Cross-country regressions conducted on 37 Sub- Saharan countries confirm the influence of most • High per capita GDP increasing demand of these factors. They help identify the respective for complex and luxury public goods such contributions of the different variables to differences as regulatory services (needed in complex observed across countries in government sizes. economies) or cultural enhancement services. Variables fall under three different groups: inequalities, political systems and rights, and • Country size, whereby sharing non-rivalrous economy and population. public goods across large populations generates 33 | South Africa Economic Update 11 | Chapter 2 Box 2.2 Figure 1: Structural determinants of central government expenditures (difference with Sub-Saharan average, % of GDP) 30 25 20 15 10 5 0 -5 -10 -15 -20 Senegal Madagascar Niger Nigeria Gabon Tanzania Uganda Mali Cameroon Ethiopia Kenya Sierra Leone CAR Burundi Chad Malawi The Gambia Côte d'Ivoire Burkina Faso Ghana Zambia Zimbabwe Guinea-Bissau Mauritius Guinea Benin Angola Comoros South Africa Swaziland Botswana Mozambique Liberia Togo Namibia Seychelles Lesotho Inequalities Political system and rights Economy and population Results suggest that Southern African Customs Union (World Bank 2017b). Among SACU members, South countries’ comparatively large governments are Africa stands out as the country where the combination driven by a combination of inequalities and political of political rights and inequality has the largest systems based on fiscal redistribution. Union members’ contribution to central government expenditures. government sizes are on average about 10 percentage points higher than the Sub-Saharan average. Of this While causal relationships cannot be strictly established difference, more than 8 percentage points of GDP through this simple analysis, results nonetheless suggest can be attributed to the combination of inequalities that the political system’s response to inequalities is (3.5 percentage points of GDP) and political systems strongly correlated with public expenditure, reflecting (4.2 percentage points of GDP), confirming the strong the ongoing social contract in SACU countries, and in redistributive role played by the fiscus in South Africa South Africa in particular. and in Namibia, where similar analysis was conducted Source: Dessus et al. (2018). The impacts of public job-creation interventions because they are not systematically evaluated. have been mixed. As discussed in the previous International evidence suggests that the impact of South Africa Economic Update (World Bank 2017c), active labor market programs is generally limited the country has been engaging in several active labor when they are fragmented and taking place in a slow market programs to raise youth employability by growth environment, as is the case in South Africa. The supporting training and skills development, promoting country also has an active industrial policy. The World entrepreneurship, and providing employment services. Bank (2017a) suggests that tax incentives in South But the results of these efforts are largely unknown Africa have positive effects on investment and jobs Chapter 2 | South Africa Economic Update 11 | 34 at a moderate fiscal cost in about half of the sectors Support for small and medium-sized enterprises covered,13 given the potentially large jobs multiplier (SMEs) is another area of intervention, as most effects associated. However, these incentives are not jobs in South Africa are located in firms of less than sufficient to create the large number of jobs needed, 50 workers. However, the share of employment in given the sectors’ structural binding constraints (labor SMEs declined from 72 percent in 2005 to 67 percent actions, policy uncertainty, trade facilitation costs, in 2015, mirroring weak net job creation in this segment electricity shortages, and lack of competition).14 in the last decade. Small businesses are nonetheless critical to reducing inequalities, providing an entry Another important public intervention lies in point for young people to enter the labor market (World the establishment of sectoral minimum wages, Bank 2018a). While direct support to SMEs, including and their progressive alignment with a national access to finance for startups, may help and could be wage (World Bank 2017c). Here again, the effect on improved, the World Bank (2018b) suggests that only inequalities is not clear cut, given poor enforcement a deep-seated improvement in the broader economic (for informal workers, in particular) and uncertain environment in which SMEs evolve could really make effects on job demand (in sectors shielded from a difference, including in particular: international competition, in particular) and on workers performing better for higher wages. With these caveats • Greater foreign competition for large firms in mind, microeconomic analysis conducted by the (that SMEs could cater to through downstream World Bank (2018a) suggests that the introduction of participation in value chains). the national minimum wage would have a positive, but marginal, impact on reducing inequalities, depending • Greater availability of skills (whose shortage and on its negative effect on employment. Consideration cost particularly affect SMEs). of general equilibrium effects (which notably capture the impact of higher demand for goods and services • Improved spatial integration (many SMEs are located from households seeing their income increasing in townships, far from main economic hubs). 15 as benefiting from the national minimum wage) nonetheless points to a less favorable conclusion, Raising labor demand by creating new markets will as suggesting that the introduction of the national ultimately be the driver of inequality reduction. As minimum wage would: discussed in the next section, South Africa’s long-term economic growth prospects are weak given the current • Affect the price of goods disproportionally consumed policy mix and foreseeable external circumstances. by the poor, particularly agricultural goods. Future labor demand is unlikely to be high enough to create the number and quality of jobs needed to • Moderate wage increases for unskilled labor reduce inequalities, unless structural reforms are whose remuneration is already above the implemented to stimulate growth and prepare the minimum wage. labor force to respond to new needs in the economy. South Africa will need to build on its comparative • Shift labor demand toward skilled labor. advantages to raise labor productivity and develop new markets for its firms, both locally and abroad. The • Deepen capital intensity at the expense of second section of this chapter explores several policy unskilled labor. options in this respect. 13 Agriculture, construction, manufacturing, trade, and other services. In contrast, tax incentives granted to mining, utilities, transport, and finance do not appear to encourage additional investments and jobs. 14 See International Monetary Fund (2014b) and Hlatshwayo and Saxegaard (2016) on the structural reasons for a lack of export supply response to the large depreciation of the real exchange rate between 2011 and 2014. 15 In the event though, SMEs in townships would also be faced with higher competition from larger firms. 35 | South Africa Economic Update 11 | Chapter 2 Creating More and Better Jobs to Reduce Income Inequalities Root causes of inequalities need to be tackled through job creation South Africa needs to focus on lifting binding • Weak integration into regional and global value constraints to the reduction of inequalities. chains. Many economic sectors in South Africa Analyzing the root causes of inequality – which cause have long been protected from foreign competition slow economic growth, unemployment, and fragility by natural trade barriers, such as distance, and – the forthcoming Systematic Country Diagnostic a history of import substitution, sanctions, and (World Bank 2018b) identifies five key areas for policy industrial policy support. State intervention, before interventions: and after 1994, has supported market structures that thwart competition, and monopolistic positions • Insufficient skills. Skills are critical for both labor are pervasive in South Africa. State-owned entities supply and demand. They raise the productivity of remain dominant and their inefficiency is damaging workers and entrepreneurs (and their ability to the competitiveness of the economy. Product access finance), help firms innovate, and expand markets in South Africa have high barriers to production at competitive prices. This in turn entry and are poorly integrated into the global raises demand in the economy for more goods economy, which means the country is missing out and services, which requires hiring more low- on opportunities to tap into global markets and and high-skilled workers. The legacy of “Bantu grow through technology transfers associated with education” continues to deprive South Africa value chain participation. of the skills it urgently needs,16 resulting in low competitiveness, high unemployment, and wage • Limited or expensive connectivity and inequality. Improving access to quality education underserviced historically disadvantaged for all in South Africa is critical. In the interim, settlements. Many South Africans live relatively bringing in skills from other countries can provide far away from job opportunities, in townships, relief to the skills-intensive economy, which will informal settlements, and the former homelands. help it to grow and create jobs. Those who are closer to opportunities, especially near urban areas, still live on the outskirts. This • Weak property rights and the unequal makes commuting expensive, aggravated by distribution of land and assets. Wealth and a functioning but anticompetitive minibus taxi land are still held by very few South Africans. sector. There has been significant migration from Publicly provided housing is an important rural areas, which supports poverty reduction, but asset for an increasing number of poor South can also put pressure on the sustainability of public Africans. Yet weak titling and tenure security of services and raise social tensions with existing property, especially in poorer and more informal residents competing for the same services, jobs, areas, limits the value of property – including as and business. Interventions to densify and diversify collateral to access finance. This inequality fuels (from a land-use perspective) urban spaces will a contestation over resources, especially as this be key to offering more job opportunities for distribution is rooted in the historical injustice workers from historically disadvantaged areas of apartheid. Policy uncertainty, be it from land and reducing persistent geographical segregation. reform or from principles underlying the third Mining Charter, is a symptom of South Africa’s • Climate change: Low-carbon transition and incomplete transition away from apartheid. water scarcity. South Africa’s carbon and water- To reduce such uncertainty – which remains use intensity will need to be addressed to preserve detrimental to private investment and innovation its development prospects and reduce inequalities. – requires consensus on sustainable interventions South Africa is cognizant of its interest to engage to effectively and equitably redistribute assets. in a low-carbon transition, 17 while adapting to the consequences of climate change, such 16 See World Bank (2017c) for a discussion on the skills in high demand on South Africa’s labor market. 17 Alton et al. (2014). Chapter 2 | South Africa Economic Update 11 | 36 as prolonged droughts that debilitate natural The model combines labor demand and supply ecosystems and amplify water insecurities. This effects to determine labor outcomes: jobs and will call for efforts to better manage the use of real wages. Labor demand results chiefly from these natural resources, including the proper product demand, stemming from households’ higher pricing of environmental externalities and incomes, government’s consumption decisions, and providing targeted support to households whose the price competitiveness of South Africa’s exports. livelihoods depend on low-value, energy- and In responding to these demands, enterprises coal-intensive activities (such as coal miners employ production factors (capital, different types and steelworkers) and water-intensive activities of labor, and intermediate goods and services also (such as small landholders that depend on cheap, partly produced with labor) in different proportions unsustainable irrigated water). depending on their costs (financial costs, wages, indirect taxes, and so on), productivity, and their Effective interventions in these areas should sustainably degree of substitutability. The latter is determined lift most binding constraints to the reduction of by the age of technology – existing installed capacity poverty and inequality, rather than just mitigating their (for which substitutability is low) versus new consequences. Limited time and resources call for technologies bought with new investment. Thus, the selective, persistent, and concentrated interventions, faster investment growth is, the faster the economy as opposed to fragmented ones, for maximum impact. can adjust to changes in relative prices from shocks or policies. Labor demand also accounts for the fact that Building on a solid tradition of prospective policy different sectors offer different remunerations for analysis in South Africa,18 the World Bank used a the same skill levels, possibly reflecting variations in computable general equilibrium model to assess the labor productivity levels and arrangements between potential impact of various policy interventions on unions and employers across sectors.21 jobs, poverty, and inequality by 2030.19 Compared with previous attempts, this model looks at new policy Labor supply is mainly determined by long- options (education, spatial integration), includes a term drivers, such as demographics (working-age microsimulation module based on the recent Living population growth and dependency ratios) and Conditions Survey (2014/15) to generate detailed human capital (education and health, to a lesser poverty and inequality numbers, and reflects recent extent). The World Bank (2018a) suggests that if national and international structural trends, such shorter-term drivers of labor supply, such as real as protracted slow total factor productivity growth wages, play a role, their impact is much more muted, since the financial crisis and low mineral prices since reflecting the low labor supply response to emerging the end of the super commodity cycle. More broadly, job opportunities in South Africa. This fundamentally the model aims to provide a consistent framework to mirrors the mismatch between growing demand for explore possible medium-term developments, based skills not met by adequate supply, and low demand on the main structural features of South Africa’s for unskilled labor in the face of sticky reservation economy. However, like any model attempting to wages22 kept above minimum levels given the high simplify complex and fluid realities, it is not exempt opportunity cost of working (transport costs and from methodological criticisms. Traditional caveats time, insecurity, insufficient childcare support for against computable general equilibrium models 20 single-headed households). Collective bargaining in focus on assumptions regarding markets’ clearance monopolistic or oligopolistic sectors also contributes mechanisms and allocative efficiency, and the model’s to wage stickiness in South Africa (World Bank 2018b). inability to capture externalities. The model used in In this case, unions’ ability to negotiate above market this chapter tries to address some of these points, wages extends inequity.23 including specifying imperfect competition in South Africa’s labor and capital markets and the rigidity in Long-term policy impacts are measured by comparing the allocation of production factors across sectors (as a baseline scenario to alternative policy scenarios. past investments are considered largely immobile), Such a baseline scenario is developed to project the and focusing on the effect of public interventions on economy until 2030 in the absence of any major shock or job creation, poverty, and inequality reduction (three radical shift from the current policy stance. This scenario fundamental externalities not internalized by markets). should not be considered as a projection, but rather as 18 Faulkner et al. (2013). 19 An economic model that uses data to simulate how an economy might react over time to changes in policy and other factors. See the Annex for a detailed presentation of the model’s main characteristics. 20 See Isaacs and Storm (2016). 21 See World Bank (2018a) for a detailed discussion of the impact of unionization on wage distribution within and across sectors. 22 Reservation wages refer to the lowest wage a worker will accept to do a job. 23 Bourguignon and Dessus (2009). 37 | South Africa Economic Update 11 | Chapter 2 a possible future, from which the impact of alternative to redress them, as highlighted in Box 2.2. Tables 2.3, policy stances can be evaluated. It does not prejudge 2.4, and 2.5 show the main results of these simulated the political feasibility of such a future, which can be scenarios. Their comparison, from a social, labor, or considered uncertain given persistent inequalities and macroeconomic perspective, demonstrates the effect people’s access to political and judicial instruments of different policies. In the absence of new policy interventions, prospects to reduce inequalities are limited, but would nonetheless benefit from improved access to education The baseline scenario (BAS) includes several of 1.4 percent between 2018 and 2030, generating assumptions. The population is set to grow at an 215,000 new jobs per year, two-thirds of which annual average of 1.1 percent from 2018 to 2030 (from would be semi-skilled and skilled (see table in Box 57 million to 65 million).24 Keeping constant pass rates 2.3). Overall, real wage levels would remain almost (matric and tertiary education) at 2016 levels, the unchanged between 2018 and 2030, but the skills supply of semi-skilled and skilled labor is projected premium would drop, reflecting a faster supply of to grow faster annually (1.6 percent and 2 percent skilled labor than unskilled labor. Slow investment respectively) than that of unskilled labor (0.7 percent) growth (1.4 percent per year, because of low household between 2018 and 2030 (a total labor supply growth of savings, relatively high and costly external debt, and 1.3 percent). Water supply is assumed to stay constant depressed foreign investment) would generate a at its current level until 2030, as all possible South modest increase of 1.1 percent in productive capacity African water reserves are already being exploited. In (physical capital stock, reflecting the accumulation contrast, mineral reserves (coal, gold, other mining) of past investments). The economy’s slow growth are considered infinite, and their depletion rate is being would be driven by water scarcity (forcing businesses driven by world prices (using World Bank projections, to use less water-intensive technologies as the price see Figure 1.4). Technological progress is set to of distributed water would grow faster than general stagnate over the period 2018 to 2030 (optimistically, inflation every year).27 The introduction of a carbon tax given recent negative trends).25 Net foreign financial would create distortion and negatively affect growth flows are expected to grow at 2 percent annually. (but less so than if South African exports were taxed Except for the progressive introduction of a carbon by importers based on their carbon content),28 but tax, all direct and indirect tax rates (including import these effects would be compensated by larger public tariffs) are assumed to stay unchanged from 2017 investments financed by carbon tax collection. Slow in the baseline scenario.26 Public consumption and gains in competitiveness (from increased investments public transfers (social assistance) to households or technological progress) and modest rebounds in are assumed to stay constant in real per capita terms mineral prices would limit annual exports growth (in between 2018 and 2030. volume) to 1.3 percent, further removing South Africa from global developments. Slow growth in public Rebounding from low levels in 2016 and 2017, consumption, combined with the introduction of a real GDP would grow at an annual average rate carbon tax, would nonetheless help stabilize public 24 United Nations population projections. 25 See World Bank (2017c) on recent trends in technological progress and innovation in South Africa. 26 While the introduction of a carbon tax represents a policy shift, it was preferred to consider it in the baseline scenario given authorities’ commitment to the Paris Agreement, and because its non-introduction would induce consequences (such as trade retaliation from signatories of the Paris Agreement) more difficult to evaluate. The recent announcement to raise fuel levy and environmental taxes in the 2018 Budget confirms authorities’ intentions in this respect. The Carbon Tax Bill was approved by Cabinet in 2017 and is expected to be enacted by Parliament in 2018. 27 This assumes perfect allocation of water across industries. Imperfect allocation (and the possible associated shortages) would have a larger negative impact on economic activity. These simulations do not capture the use of water that is not commercially traded. 28 The mitigation of carbon emissions is obtained in the model by imposing a tax on the direct and indirect consumption of carbon embedded in coal and refined petroleum products. Without this tax (representing 1 percent of GDP in 2030), carbon dioxide emissions would be 17 percent higher in 2030, in line with the announced objective to decrease emissions by 13 percent by 2025 and by 26 percent by 2035 compared with business as usual. Induced distortive effects from the introduction of the carbon tax could cost up to 0.4 percent of GDP by 2025 and 0.7 percent by 2030, and be regressive as the poorest deciles would be most affected. This compares with higher estimates of the cost of introducing carbon taxes (Alton et al. 2014), ranging from 1 percent to 1.2 percent of GDP under more optimistic growth scenarios (3.9 percent GDP growth between 2010 and 2025), higher mitigation ambitions, and possibly higher coal output projections under the business-as-usual scenario with higher anticipated global prices. Chapter 2 | South Africa Economic Update 11 | 38 debt and generate modest fiscal space for additional more slowly than services and agriculture. Jobs would public investment. From a sectoral perspective, mining, be shed in the coal and gold sectors. and manufacturing to a lesser extent, would grow Box 2.3: Education flows and the distribution of skilled labor incomes across household deciles According to the National Income Dynamic Study time, in terms of skills and the distribution of incomes 2014/15, semi-skilled and skilled labor incomes are across deciles. Out of 10 children from the first decile, unequally distributed across household groups. Defined less than two currently pass matric (high school by the education levels of households’ working members diploma) and only 0.2 (or 2 out of 100) graduate from (skilled: tertiary graduate; semi-skilled: matric; unskilled: university. This compares to 2.3 children (out of 10) incomplete secondary education or less), skilled labor from the richest decile graduating from university. income accruing to the four poorest household deciles29 While still extremely unequal, such inequalities are less in 2014/15 was only 0.1 percent of the total (against 84.2 pronounced than those recorded in the distribution of percent accruing to the richest decile). Similarly, only 3.2 skilled and semi-skilled labor incomes. Better skilled percent of semi-skilled labor income accrued to the four entrants from poor backgrounds will eventually lead to poorest deciles (against 45.8 percent to the richest decile). a redistribution of skilled labor income to the poorest households. Policy options to accelerate this transition At the same time, new cohorts entering the labor are discussed in this chapter. market will modify the labor force composition over Box 2.3 Table 1: Education levels among South African household deciles Deciles Cohort Get to Pass the Obtain the Enter Graduate grade 12 matric bachelor pass university 1 10 3.8 1.7 0.3 0.3 0.2 2 10 3.8 1.7 0.3 0.3 0.2 3 10 5.5 2.8 0.5 0.6 0.3 4 10 5.5 2.8 0.5 0.6 0.3 5 10 6.9 3.9 0.8 1.0 0.5 6 10 6.9 3.9 0.8 1.0 0.5 7 10 6.9 5.0 1.5 1.6 0.8 8 10 6.9 5.0 1.5 1.6 0.8 9 10 8.7 8.0 4.6 3.9 2.3 10 10 8.7 8.0 4.6 3.9 2.3 Total 100.0 63.4 43.1 15.5 14.9 8.0 Source: World Bank staff calculations based on Van Broekhuizen et al. (2016) and StatsSA. As a result of past education efforts, poverty (at $1.90 to the poorest 40 percent would increase from 8.6 a day) would decline significantly, from 18.6 percent percent to 10.3 percent. Given projected weak economic in 2017 to 12.7 percent of the population in 2030. The growth and the fact that public transfers to the poor are Gini coefficient would drop from 62.8 in 2017 to 59.5 in assumed to remain constant in per capita terms, most of 2030, and the share of real disposable income accruing the projected reduction in inequalities can be attributed 29 A decile is one of 10 equal groups into which a population can be divided. 39 | South Africa Economic Update 11 | Chapter 2 to the redistribution of semi-skilled and skilled labor progress is consistent with a significant reduction in income over time.30 Progress in education among the inequalities of opportunities in the last decade, which poorest deciles will contribute to this redistribution will eventually affect labor markets. A greater supply (Box 2.3). At current pass rates, the proportion of skilled of skilled labor in a depressed economic environment (university degree level) and semi-skilled labor (matric would prolong the decline in the skills premium recorded level) incomes accruing to the bottom 40 percent would between 2006 and 2015 (World Bank 2018a), though respectively rise to 3.6 and 11.2 percent in 2030. Such from a high initial level. Figure 2.8: Deciles’ labor incomes shares, 2015 and 2030, baseline scenario A. Semi-skilled B. Skilled 40% 80% 35% 70% 30% 60% 25% 50% 20% 40% 15% 30% 10% 20% 5% 10% 0% 0% 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Deciles Deciles 2015 2030 2015 2030 Source: World Bank staff calculations. Thus, by 2030, income inequality in South Africa Increased policy certainty in the mining sector would would be lower than in 1996. While significant, this spur growth and labor demand, and reduce poverty. performance would compare modestly with that of There have been several attempts to measure the impact selected peers: in Brazil the Gini coefficient decreased of policy uncertainty on economic activity in recent years. from 59.3 in 2001 to 53.1 in 2011 – more than twice the This includes using algorithms to detect the frequency speed of South Africa. Nonetheless, out of the three of mentions of “uncertainty” in the press relating to percentages points of reduction in inequality, two South Africa’s economic developments.31 The World would originate from a decline in racial inequality. In Bank (2018b) analyzes the impact of the third Mining other words, looking at the different factors explaining Charter’s provisions32 and concludes that its amicable income differences across households, race would lose resolution could increase investments in mining by about some of its explanatory power, prolonging a trend 25 percent between 2018 and 2030, compared with the observed since 2006 (Figure 2.5 A). This is important baseline scenario and given projected commodity prices. as likely to strengthen a social contract which remains Simulating the impact of higher foreign direct investment weakened by the fact that inequalities are still to a in mining in an alternative scenario (MIN) suggests that large degree determined by races. macroeconomic effects would remain modest, because 30 In contrast, inequalities in the distribution of capital income would increase over time given higher savings rates among the richest households. 31 Hlatshwayo and Saxegaard (2016). 32 The third Mining Charter provides more stringent targets that are proposed to be legally enforceable (for the first time), combined with a range of areas that are uncertain in nature, difficult to measure, and potentially of high cost to shareholders. The main cause of concern relates to a higher cost of compliance, which is particularly affected by the following provisions: (i) increased broad-based black economic empowerment shareholding (which may have to be funded by the current owners of capital in the sector), (ii) the rejection of the “once-empowered, always-empowered” principle (where a legislated minimum of broad-based black economic empowerment shareholding needs to be maintained, even after black economic empowerment investors disinvest), (iii) a proposed dividend of 1 percent of turnover to be paid out to broad-based black economic empowerment shareholders annually, and (iv) increased socioeconomic investment requirements (such as the requirement to spend a minimum of 70 percent of total mining procurement value on goods manufactured in South Africa, with at least 21 percent from black-owned companies), as well as various enterprise development requirements. Each of these factors may erode investor returns and is thus likely to undermine mining investment in South Africa. Chapter 2 | South Africa Economic Update 11 | 40 GDP growth would only accelerate by 0.2 percent per Although the magnitude of economic impacts year because of accelerated export and investment to be expected from improved policy certainty growth between 2018 and 2030. By 2030, GDP would be is unknown, this scenario does illustrate the 2.8 percent higher than in the baseline scenario (about persistent tradeoff between growth and equality 2 years of GDP growth), highlighting the importance under current productive patterns. While this point the mining sector continues to play in South Africa. An is not meant to question the critical role of economic additional 60,000 jobs would be created by 2030, lifting growth in improving the living conditions of the poor, 140,000 people out of poverty. Creating jobs in the mining it underlines the need to develop interventions that sector has a relatively significant effect on reducing promote more inclusive growth. The previous South poverty (as discussed in the World Bank’s Poverty and Africa Economic Update (World Bank 2017c) identified Inequality Assessment 2018a), because these new jobs sectors where innovation efforts would have the are mostly unskilled/semi-skilled, but pay above average. largest payoffs for poor households (through jobs and improved consumption). The remainder of this section But inequalities would widen at the same time, explores options to: illustrating the regressive nature of South Africa’s traditional growth patterns. The Gini coefficient would • Raise labor supply among poor households. rise from 59.5 in the baseline scenario to 59.6 in the alternative mining scenario, and the share of incomes • Create new jobs for poor households. accruing to the bottom 40 percent would decrease from 10.3 percent to 10.2 percent. The skewed distribution of • Improve matching between labor supply and mine ownership and the low skills content of new jobs are demand. some of the main factors driving increased inequalities. Accelerated efforts to improve the quality of basic education and access to tertiary education would lead to a significant reduction in inequalities by 2030 In South Africa’s current environment, it takes first policy scenario (EDU1) assumes that increased time to see education progress reflected in labor labor costs for teachers in public schools (increased markets. Raising education levels may take time but remuneration and training/coaching, along with the will be key to reducing inequalities. Between 2018 introduction of performance-based human resources and 2030, young people entering the working-age management systems and accountability frameworks) population (15–64 years) will increase the population would raise the number of children from the poorest by 2.7 percent annually, while elders exiting this age 40 percent passing matric to 30 percent in 2030 (up group will reduce it by 1.5 percent. This means that from 23 percent in 2012), leading to an increase in only 4.3 percent of the working-age population is being the matric national pass rate from 43 percent in 2012 renewed every year, so any progress in the education to 49 percent in 2030. This would also increase the of the youth will influence the composition of the labor number of children from poor households entering force very slowly. In addition, it currently takes about 10 university, given their improved academic eligibility years for young people who will eventually work to be (from 4 percent in the baseline scenario to 6 percent effectively absorbed by labor markets, confirming the in this scenario). In this case, the additional education importance of the first working experience.33 This slow costs would offset the positive effect of higher skilled entry also delays the impact of educational progress labor supply on GDP growth, but would still have a on the skills composition of the labor force and the pronounced impact on reducing inequalities (by half a distribution of education returns across households. percentage point in 2030 compared with the baseline scenario) and thus poverty (by 1 percentage point). Two broad policy interventions can improve the skills of poor young people: improving teachers’ The second education scenario (EDU2) assumes capacity and accountability to raise primary and increased public financial support for poor university secondary school achievements among the poorest students currently eligible for such support (among deciles; and facilitating access to university for poor the six poorest deciles, covering accommodation and eligible students through financial support. The food in addition to tuition), and for students in the same 33 Anand et al. (2016). 41 | South Africa Economic Update 11 | Chapter 2 income brackets currently not attending university with the projected decline in the school-age population, despite being eligible (see Box 2.3). This would entail which would create fiscal space of about the same increasing the number of university students34 by amount (in terms of public education expenditure).35 It about 85,000 in 2018 (up from 985,000 currently), at could also be reduced by private sector participation in an additional cost to the public sector of close to 0.5 providing financial support to poor university students. percent of GDP annually. This amount would finance With reduced inequalities, the number of poor people higher demand for education services and other goods would further drop by almost 3 million in 2030 compared and services through transfers to the poorest students. with the baseline scenario. It would entail a significant reduction in inequalities (of almost 1 percentage point), but at the expense of The effect of higher skills among the poor would GDP growth and public investment (the higher public be much diminished in a slow growth environment. transfers would exceed the more progressive increase Given persistent low labor demand, the skills premium in teachers’ remunerations in the first education for tertiary education graduates (compared with scenario). In turn, the decline in poverty would be slightly workers not having completed secondary education) less pronounced than in the first scenario. Combining in the combined education scenario would decline by the features of the two scenarios would eventually 1.9 percent annually between 2018 and 2030, against lead to an increase in the number of children from the 1.6 percent in the baseline scenario. But if this faster poorest four deciles with a tertiary degree to 4.6 percent decline should actually reduce inequality (as skilled in 2030, against 2.2 percent in the baseline scenario. In labor constitutes a larger share of the incomes of this combined scenario, inequalities (the Gini coefficient) the richest households), it is possible that some of would narrow by 1.5 points in 2030 compared with these poorer skilled workers would join the young the baseline, while the proportion of income accruing unemployed population, as employability, distance to to the poorest four deciles would increase by more jobs, and access to professional networks may differ than 1 percentage point, to 11.3 percent. The impact of across new entrants, at the expense of those from additional public financing on growth would offset that the poorest households.36 This suggests the need to of the higher skilled labor supply. But this cost, of up intervene on both the supply and demand sides of labor, to 1 percent of GDP by 2030, could be partially covered and on the matching between the two. Increased competition in product markets would open massive job opportunities Product market contestability is low in South Africa. costs, inefficiencies in these sectors can act as barriers Product market regulation indicators suggest that South to trade. The previous South Africa Economic Update Africa performs in the middle of its peer countries.37 (World Bank 2017c) indicated that ports tariffs were State control is the area in which it performs most twice as high for manufacturing goods than the global poorly. This includes aspects such as the scope of average. But the large dispersion of rates of return to state-owned entity involvement in the economy, direct capital across sectors suggests that many other sectors government control of enterprises, price controls, and suffer from poor contestability. This may be the result the use of command and control regulations, followed by of historical reasons, and not only product market barriers to entrepreneurship. Product market regulation regulations,38 or policy uncertainty that particularly indicators show that South Africa’s energy, transport, deter foreign investors. For example, key agricultural and communications regulations are relatively and manufacturing markets were historically tightly restrictive of competition. Given their effect on trading regulated and protected oligopolies or monopolies, 34 See also World Bank (2018b) for options to (i) improve technical and vocational training, with a view notably to better respond to labor markets needs and (ii) raise the overall employability of youth, including through improved public-private partnerships. As qualitative in nature and not often evaluated, it is difficult to simulate quantitatively these reforms. They remain nonetheless very important to address inequalities through jobs creation. 35 World Bank (2016b). 36 Burger and Jafta (2006) provide evidence of the persistent disadvantages for the average black worker. The race gap in employment, occupational attainment, and wages remains unexplained after controlling for differences in observable characteristics such as skills. This gap, however, has been decreasing over time for skilled workers. 37 OECD (2017). Product market regulation indicators convert qualitative information concerning laws and regulations that may affect competition into quantitative indicators of the restrictiveness of regulations to competition. They cover both economy-wide regulations and sector-specific regulations for certain network sectors (energy, transport, and communications), professional services, and retail services. One shortcoming is that they measure only the restrictiveness of regulations “on the books” and do not cover implementation. 38 Product market regulations often listed as preventing competition in South Africa include exclusive lease agreements in retail trade and high switching costs in financial and insurance services. Complying with a complex regulatory framework is also listed as particularly detrimental to SMEs. Chapter 2 | South Africa Economic Update 11 | 42 sometimes under state control. After democracy, the Markups above normal profits are estimated to government undertook a range of market reforms to amount to close to 10 percent of GDP in South privatize many of its state-controlled enterprises or Africa. 40 This means that halving them could boards. However, without consistent complementary generate effects of macroeconomic amplitude. measures to open these markets to trade and This is illustrated in an alternative policy scenario competition, their structures remain (World Bank (MKP), where markup rates would be progressively 2018b). Moreover, previous anticompetitive regulations halved by 2030. In this scenario, investments pick up appear to have facilitated several of South Africa’s (by 10 percent in 2030 compared with the baseline cartels (World Bank 2016c). In the case of cement, for scenario) in line with new investment opportunities, example, the industry was a state-sanctioned cartel as do exports (by 9.1 percent) in line with increased with price controls from the 1940s until 1996. After the competitiveness (from improved capital allocation legal cartel was disbanded, firms continued to divide across sectors and productivity gains within sectors). markets along traditional lines. This results in the net creation of an additional 405,000 jobs by 2030 and higher real wages, Greater domestic competition would generate particularly for semi-skilled and skilled workers, as. massive job opportunities. Substantial market power increased competition would open new opportunities in several sectors is likely to result in: in skills-intensive sectors. Inequalities would be reduced (as capital owners, mostly concentrated in • Suppressed demand for capital and labor; the richest decile, would see their real disposable income grow less rapidly than poorer deciles), with a • Low total factor productivity as maintaining in Gini coefficient at 58.8 by 2030 (down from 59.5 in the business poorly competitive firms;39 baseline scenario). With cheaper goods41 and higher labor incomes, the poverty rate would also be reduced • Reduced price competitiveness for sectors to 11 percent by 2030 – lifting an additional 1.1 million using as inputs the (inflated) goods and services people out of poverty. produced in concentrated sectors. Accelerated skills migration could be contemplated to relax skills constraints in the short term The skills constraint in South Africa is a major The impact of accelerated skills migration is deterrent to innovation in general, and to the simulated in an alternative policy scenario (SKM) vitality of the manufacturing sector in particular. where skilled migrants join the labor force between The latter, if revived, has massive job-creation 2018 and 2030 (an additional 150,000 skilled workers potential. The World Bank (2017a) suggests that for by 2030), resulting in an increase of 5.8 percent in every job directly created in manufacturing, another the supply of skilled labor by 2030 compared with 3.8 jobs are indirectly created. But addressing the skills the baseline scenario, and higher remittances out constraint by improving education will take time, which of South Africa equivalent to an additional outflow is why authorities are considering options to expand of 1.2 percent of GDP. Relaxing the skills constraint the supply of high-level skills through immigration would increase GDP by 2 percent in 2030 compared policy reforms. Options include introducing a points- with the baseline scenario, and lead to a strong based system to determine the eligibility criteria supply response in manufacturing – in automotives, for long-term residence visas with a view to attract machinery equipment, and metallic products notably. investors and skilled migrants, granting critical skills Indirect demand for unskilled and semi-skilled labor and business visas that cater for family members, would be strengthened (an additional 50,000 and allowing international graduates with critical skills to 20,000 jobs respectively), as each new skilled migrant apply for a long-term residence visa, and introducing would create 0.5 unskilled or semi-skilled jobs. a differentiated skills-transfer mechanism to cater for However, greater competition between skilled locals different working conditions. and immigrants would exert a downward pressure on 39 Aghion et al. (2008) estimated that a 1-percentage-point increase in sectoral markups has been reducing total factor productivity in the same sector by 0.1 percentage points in South Africa over the period 1970–2004. 40 This computation excludes extractive industries, whose profits originate to some extent from the rarity of minerals (decoupling global prices from cost of extraction), and not necessarily from insufficient competition. 41 See World Bank (2017c) for a discussion of the potential impact of innovation (new goods and services) on the welfare of poor South Africans. 43 | South Africa Economic Update 11 | Chapter 2 real wages and reduce the domestic supply of skilled the size of the upper middle class. As a result, the Gini workers, resulting in a net gain of 135,000 additional coefficient would decline slightly (to 59.3 in 2030, skilled jobs in the economy.42 With higher GDP and the against 59.5 in the baseline scenario), but the share creation of new unskilled jobs, poverty would decline, of income accruing to the poorest 40 percent would but only modestly, as skilled migrants would increase remain unchanged. Spatial integration will be a necessary, but possibly insufficient, driver of job creation The right to decent shelter is enshrined in South or commuting to urban hubs. In addition, the costs of Africa’s Constitution, and much progress has working, transport, and housing need to be reduced. been made in realizing it. Locating the poor closer This is reflected in an alternative policy scenario (SPA) to economic centers and improving mobility is a where 1 percent of GDP would be invested every year necessary condition for accessing jobs. Although many from 2018 into collective transportation systems South Africans, including in the poorest 40 percent, (mostly consumed by the poorest seven deciles) and own houses, there are still backlogs and 5.1 million social housing to reduce their price. This investment people (about 9 percent of the population) still live in would be financed through domestic borrowing (which shacks. Human settlements programs, often focused means less financing for alternative investment on delivering housing units, have unintentionally projects). While macroeconomic effects would remain perpetuated apartheid spatial patterns – leaving relatively modest, they would be positive. GDP would the poor on the periphery of urban areas with fewer be 0.6 percent higher by 2030 than in the baseline economic opportunities. The single-story home scenario, reflecting the positive effect on labor model originally espoused by the Reconstruction and supply stemming from cheaper collective transport Development Programme propels urban sprawl and and social housing prices (reduced by 37 percent does not encourage mixed land use (for job-generating and 48 percent respectively by 2030 compared commercial activity, for example). It also adds to the with the baseline scenario). Efforts to foster spatial cost of transport and commuting time: the average integration would reduce the poverty rate by 2030 commuter distance in South African metros is 22–27 (at 12 percent, against 12.7 percent in the baseline kilometers.43 This means poor South Africans have scenario), largely because of cheaper consumption (to long travel times and high transportation costs. Long a large extent) and more job opportunities (to a lesser and expensive commutes make it more difficult for the extent). Bringing unskilled labor closer to economic poor to access urban jobs and raise their reservation centers would not necessarily create many jobs wage, that is, the wage that makes it worthwhile to (besides during the construction phase of investment work given the associated costs. projects), given the extremely high unemployment rate in this labor category (at 27 percent in 2017), Economic densification would reduce poverty, but reflecting low current demand. Yet, improved spatial not necessarily create many unskilled jobs. Whether integration would also result in a significant decline to bring people to jobs or jobs to people remains an in the Gini coefficient (to 58.6 by 2030). The fourth area of contention in South Africa. Cities tend to be to seventh deciles would be the main beneficiaries, engines of growth, but the country’s spatial history as these households spend a higher share of their has resulted in large informal settlements without income on housing and transport, and would receive much economic activity. There is significant interest a larger share of labor incomes. However, the share in exploring whether economic activity and jobs can be of incomes accruing to the poorest 40 percent of the brought to townships, rather than relying on migration population would rise by 0.3 percentage points. 42 As assuming perfect substitutability between migrants and natives of the same skills, these results may be considered conservative as not capturing the complementarity between the two categories. As filling a gap in labor demand, skilled migration could generate a larger number of jobs than that computed with the general equilibrium model. 43 To compare this with selected other cities: average commuting distances are 7, 11, and 19 kilometers in Lima, Bogotá, and Mexico City, respectively, according to Munoz-Raskin and Scorcia (2017). Chapter 2 | South Africa Economic Update 11 | 44 Combined reforms would help South Africa meet its vision for 2030 The various reforms discussed above would Combining all reforms in one scenario (CMB) suggests reinforce each other to generate a larger impact that poverty could be halved by 2030 compared with on GDP growth, jobs creation and the reduction the baseline scenario (reducing to 6.4 percent of the of poverty. The impact of free access to tertiary population, against 12.7 percent, see Table 2.3). This education for students from low-income households would lift an additional 4 million people out of poverty, would be amplified by efforts to ensure poor students as inequalities reduce (to a Gini coefficient of 56.5) and are academically eligible for university. Greater growth accelerates (to 2.2 percent annually, against competition would increase job opportunities for these 1.4 percent in the baseline scenario). Job creation (an skilled new entrants, and their ability to reap these additional 810,000 jobs), higher wages for workers opportunities would be enhanced by spatial integration from poor households, and cheaper goods and services efforts. Greater competition would also favor spatial would contribute to these outcomes. integration by reducing the price of services (such as public transport and information and communications When combined, individual interventions work technology). Higher growth and foreign investment together and have a much greater impact. Compared through the clarification of the third Mining Charter’s with the sum of individual interventions, combining all the intentions, increased competition, and accelerated reforms would lift 1.2 million more people out of poverty, skills migration would mitigate the skills constraint raise the share of income in the bottom 40 percent by an and generate additional public revenue to finance additional 0.2 percentage points, and further accelerate education and urbanization programs. GDP growth by 0.1 percentage points (Figure 2.9). Figure 2.9: Relative impacts of selected interventions on poverty, inequality, and GDP growth 100% 0.1 0.2 1.2 80% 60% 40% 20% 0% -20% Poverty Bottom 40% share of income GDP Growth MIN EDU1 EDU2 MKP SKM SPA Synergetic Source: World Bank staff calculations. 45 | South Africa Economic Update 11 | Chapter 2 Table 2.3: Progress toward the Vision 2030 in different scenarios Poverty rate Number of poor Gini Bottom 40% share of ($1.90 day) (million) Coefficient total consumption 2017 18.6 10.51 62.8 8.6 BAS 12.7 8.30 59.5 10.3 MIN 12.5 8.16 59.6 10.2 EDU1 11.7 7.66 59.1 10.6 EDU2 12.0 7.80 58.7 10.7 MKP 11.0 7.19 58.8 10.5 SKM 12.7 8.26 59.3 10.3 SPA 12.0 7.80 58.6 10.6 CMB 6.4 4.15 56.5 11.7 Source: World Bank staff calculations. Results from the second row are for 2030. Table 2.4: Labor market indicators in 2030 in different scenarios Jobs (thousands) Unskilled Semi-skilled Skilled Unemployment rate 2017 8,140 5,229 1,981 27.3% BAS 9,158 6,474 2,538 26.7% MIN 9,197 6,490 2,541 26.5% EDU1 9,109 6,557 2,547 26.7% EDU2 9,161 6,461 2,564 26.7% MKP 9,410 6,604 2,561 25.1% SKM 9,207 6,494 2,674 26.3% SPA 9,181 6,475 2,537 26.7% CMB 9,517 6,722 2,741 24.1% Source: World Bank staff calculations. Results from the second row are for 2030. Chapter 2 | South Africa Economic Update 11 | 46 Table 2.5: Selected macroeconomic indicators in 2030 in different scenarios (index 2017: 100; constant prices) Gross domestic Private Gross fixed Exports Imports product consumption investment BAS 119.5 117.4 141.5 116.1 119.8 MIN 122.9 120.0 152.5 119.9 123.6 EDU1 119.5 118.0 138.6 115.8 119.5 EDU2 119.1 117.3 139.1 115.2 119.0 MKP1 125.4 123.6 155.6 126.6 131.1 SKM1 121.9 117.5 147.1 122.3 121.4 SPA1 120.2 118.8 140.4 115.0 118.8 CMB2 133.1 130.4 168.9 136.3 135.6 Source: World Bank staff calculations. Reforms would further reduce racial inequalities. inequality, contributing for about half to the total Figure 2.10 decomposes the changes in inequality reduction in inequality.44 In the combined reforms from reforms with respect to the baseline scenario. scenario (CMB), racial inequalities would explain 38% But reduced uncertainty in the mining sectors, all of total inequalities by 2030. This compares with 48% simulated reforms would entail a decline in racial in 2006, as reflected in Figure 2.5 A. Figure 2.10: Contributions to change in inequality with respect to the baseline scenario 0.5 0.0 -0.5 Charge in Gini index -1.0 -1.5 -2.0 -2.5 -3.0 MIN1 EDU1 EDU2 MKP SKM SPA CMB Race Education Labor status Source: World Bank staff calculations. 44 See Fields (2003) for the method of decomposing inequality across various factors and over time. 47 | South Africa Economic Update 11 | Chapter 2 The overall positive impact of these combined – at about 24 percent in 2030, because only 5.3 million reforms could be underestimated, because they jobs would have been created since 2012, against are likely to improve policy certainty. Reduced the 11 million targeted in the National Development inequality – of racial origin in particular - could lead Plan. This underlines the need to continue to support to less resource contestation and broader support redistribution to the many chronic poor South Africans for investor-friendly reforms. At the same time, the who may not be able to reap the opportunities that unemployment rate would remain extremely elevated such reforms would have created by 2030. Conclusion South Africa’s current economic rebound could be Progress in reducing inequalities since the 2000s, short-lived if the fundamental factors undermining notably in access to education, is slowly leading to its growth potential are not addressed. Stubbornly more skilled workers from the poorest backgrounds. high levels of inequality reflect the weak capacity of This trend needs to be nurtured and amplified, many South Africans to contribute to skills-intensive benefiting from renewed confidence in South Africa economic development. Inequalities fuel contestation as governance issues are addressed and from a and policy uncertainty, deterring the investments and strengthened global outlook. financial resources needed to innovate and expand productive capacities, and to redress historical injustice The analysis presented in this report suggests through targeted public interventions. that selective, persistent efforts to increase domestic competition and improve the poor’s Inequalities remain extremely high, however, their access to university would help South Africa nature is also changing. Inequalities are widening eliminate extreme poverty and significantly reduce between the poor and a small emerging middle class, inequalities by 2030. In turn, this would strengthen and narrowing between the middle class and rich the social contract between the country’s people, households. Today, inequalities are driven by education which would improve investors’ confidence in South and labor status – conditions that policies and public Africa’s future. interventions can improve. Chapter 2 | South Africa Economic Update 11 | 48 REFERENCES Aghion, P., M. Braun, and J. Fedderke. 2008. “Competition and productivity growth in South Africa”, Economics of Transition 16(4): 741–768. Alatas, V. and F. Bourguignon. 2005. “The Evolution of income distribution during Indonesia’s fast growth 1980–96”, in The Microeconomics of Income Distribution Dynamics in East Asia and Latin America, F. Bourguignon, F. Ferreira, and N. Lustig (eds.), Washington, DC: Oxford University Press and World Bank. Algu, Y., M. Hanusch, and H. Nguyen. 2018. “Exchange rate volatility and FDI inflows: Evidence from cross-country panel data”, MTI Working Paper, forthcoming, Washington, DC: World Bank. Alton, T., C. Arndt, R. Davies, F. Hartley, K. Makrelov, J. Thurlow, and D. Ubogu. 2014. “Introducing carbon taxes in South Africa”, Applied Energy 116, 344–54. Anand, R., S. Kothari, and N. Kumar. 2016. “South Africa: Labor market dynamics and inequality”, Working Paper 16/137, Washington, DC: International Monetary Fund. Bachmair, F. and J. Bogoev. 2018. “Assessment of contingent liabilities and their impact on debt dynamics in South Africa”, MTI Working Paper, forthcoming, Washington, DC: World Bank. Bourguignon, F. and S. Dessus. 2009. “Equity and development: Political economy considerations”, in No Growth without Equity? S. Levy and M. Walton (eds.), 45-69, New York and Washington, DC: Palgrave Macmillan and World Bank. Burger, R. and R. Jafta. 2006. “Returns to race: Labour market discrimination in post-apartheid South Africa”, Economic Working Papers 4, no. 06. Stellenbosch: University of Stellenbosch. Bussolo, M. and J. Lay. 2005. “Globalization and poverty changes in Colombia”, in Globalization and Poverty channels and Policy Responses, M. Bussolo and J. Round (eds.), 195–227. London: Routledge. Chitiga-Mabugu, M. 2016. “SAM for South Africa 2012”. Mimeo, Pretoria: University of Pretoria. Dessus, S., J. Herderschee, M. Kwaramba, and N. Munditi. 2018. “Identifying the drivers of demand for government in Southern African Customs Union countries”, MTI practice notes #1, February, Washington, DC: World Bank. Faulkner, D., C. Loewald, and K. Makrelov. 2013. “Achieving higher growth and employment: Policy options for South Africa”, South African Reserve Bank Working Paper, July, Pretoria: South African Reserve Bank. Fields, G. S. 2003. "Accounting for income inequality and its change: a new method, with application to the distribution of earnings in the United States." Research in Labor Economics 22: 1–38. Hlatshwayo, S. and M. Saxegaard (2016). “The consequences of policy uncertainty: Disconnects and dilutions in the South African real effective exchange rate-export relationship”, Working Paper 16/113, Washington, DC: International Monetary Fund. International Monetary Fund. 2014a. “Redistribution, inequality, and growth”, staff discussion note, SDN/14/02, Washington, DC: International Monetary Fund. 49 | South Africa Economic Update 11 | References International Monetary Fund. 2014b. “South Africa’s exports performance: Any role for structural factors”, Selected Issues Paper, Report #14/339, Washington, DC: International Monetary Fund. Isaacs, G., and S. Storm. 2016. “Predicting the impact of national minimum wage: Are the general equilibrium models up to the task?”, Econ3x3, Cape Town: University of Cape Town. Lay, J., R. Thiele, and M. Weibelt. 2008. “Pro-poor growth in Bolivia: Accounting for external shocks and policy reforms”, Review of Development Economics 12(1): 37–56. Lay, J. 2010. “Sequential macro-micro modeling with behavioral microsimulations”. International Journal of Microsimulation 3(1): 24–34. Maveé, N., R. Perrelli, and A. Schimmelpfennig. 2016. “Surprise, surprise: What drives the rand/U.S. dollar exchange rate volatility?” Working Paper 16/205, Washington, DC: International Monetary Fund. Munoz-Raskin, R. and H. Scorcia, 2017. “Time for a tailored approach to South African BRTs”, World Bank Other Operational Studies 28352, Washington, DC: World Bank, and Paris: Organisation for Economic Co-operation and Development. Organisation for Economic Co-operation and Development. 2017. Indicators of product market regulation, oecd. org. Paris: Organisation for Economic Co-operation and Development. Quantec. 2017. SA Standardised industry data. Pretoria. Van Broekhuizen, H., S. van der Berg, and H. Hofmeyr. 2016. “Higher education access and outcomes for the 2008 national matric”, Economic Working Papers 16/16, Stellenbosch: University of Stellenbosch. Van der Berg, S. and M. Gustafsson. 2017. “Quality of basic education. A report to working group 1 of the High Level Panel on the Assessment of Key Legislation”. Unpublished. Stellenbosch: University of Stellenbosch. World Bank. 2016a. “Poverty and shared prosperity 2016: Taking on inequalities”, Washington, DC: World Bank. World Bank. 2016b. “Forever young: Southern Africa demographic opportunity”, Washington, DC: World Bank. World Bank. 2016c. “South Africa Economic Update #8: Promoting faster growth and poverty alleviation through competition”, Washington, DC: World Bank. World Bank. 2017a. “South Africa Economic Update #9: Private investment for jobs”, Washington, DC: World Bank. World Bank. 2017b. “Does fiscal policy benefit the poor and reduce inequality in Namibia?”, Washington, DC: World Bank. World Bank. 2017c. “South Africa Economic Update #10: Innovation for productivity and inclusiveness”, Washington, DC: World Bank. World Bank. 2017d. “Commodities price forecast”, October 26, 2017 http://pubdocs.worldbank.org/ en/678421508960789762/CMO-October-2017-Forecasts.pdf Washington, DC: World Bank. World Bank. 2018a. “South Africa Poverty and Inequality Assessment 2018. Overcoming Poverty and Inequality in South Africa: Drivers, Constraints and Opportunities”, Washington, DC: World Bank. World Bank. 2018b. “Republic of South Africa Systematic Country Diagnostic. An incomplete transition: Overcoming the legacy of exclusion in South Africa”, forthcoming, Washington, DC: World Bank. World Bank. 2018c. “Global economic prospects”, Washington, DC: World Bank. References | South Africa Economic Update 11 | 50 ANNEX Modeling Prospective Policy Scenarios A computable general equilibrium model for South Supply is modeled using nested constant elasticity of Africa is used in this report to develop prospective substitution functions, which describe the substitution policy scenarios. The model aims to provide a and complement relations among the various inputs. consistent framework to explore possible medium-term Producers are profit-maximizers and constant developments, based on the main structural features returns to scale are assumed. Output results from of South Africa’s economy. It is a typical neoclassical two composite goods: intermediate consumption model with endogenous prices, market clearing, and (excluding energy) and value-added plus energy, imperfect substitution between domestic and foreign which are imperfectly substitutable. The intermediate goods, allowing for endogenous factor accumulation. aggregate is obtained by combining all products in fixed As in any computable general equilibrium, prices are proportions (Leontief structure). The value-added is endogenous for each market (goods and factors) and then decomposed in two substitutable parts: labor and equalize supplies (imports; South African production a capital-water-mineral-energy bundle. Energy types for the domestic market; factors supply) and demands (electricity and refined petroleum) are also imperfectly (final demand from households, the government, substitutable. Demand for capital makes a distinction investors, and the rest of the world; intermediate between “old” capital and “new” capital. The model demand from producers; factors demand), to obtain integrates the notion of vintage capital to distinguish the equilibrium. The equilibrium concerns all markets the process of allocating capital already installed from simultaneously. This type of modeling allows detailed that resulting from contemporary investment (putty/ databases to be combined, with a sound micro-based semi-putty production function). “New” capital can be theoretical framework capturing their interdependence allocated more flexibly than “old” capital. Accelerating and linkages. With such characteristics, computable investment therefore strengthens the capacity for general equilibrium models are useful tools to assess adjustment of the productive sectors to changes in the long-term impact of shocks (such as world prices) relative prices.45 Flexible factors prices clear markets and structural reforms. The underlying assumption of through the equalization of demand and supply, but market clearance and monetary neutrality means that factors markets also include sectoral rigidities: some these models are not suited to assessing the short-term sectors pay workers (of the same type) better than effects of macroeconomic policies in economies with others; and in some sectors, capital remuneration well-developed financial markets such as South Africa. exceeds its marginal productivity. Consequently, before-tax prices of goods domestically produced The model is calibrated for the year 2012, based on often exceed their marginal costs.46 a social accounting matrix built by Chitiga-Mabugu (2016). The matrix and the model comprise 55 sectors Income from labor and capital accrue to the different of activity (and corresponding products), 10 household households, in proportion of their initial share in types (corresponding to the 10 income deciles), 12 total incomes. Households pay direct taxes, receive trading partners, and 7 factors of production (informal grants from government, and transfer/receive labor, unskilled, semi-skilled, skilled, capital, mineral, funds from other households47 and trade partners. and water resources). Their net disposable income is allocated to final 45 For all sectors, elasticities of substitution between intermediate consumption and value-added plus energy are set to 0.9 for old capital and 1.8 for new capital; elasticities of substitution between labor and the capital-water-mineral-energy bundle are set at 0.9 for old capital and 1.8 for new capital; elasticities of substitution between labor types are set at 0.1; elasticities of substitution between capital, water, mineral, and energy are set at 0.9 for old capital and 1.8 for new capital. Elasticities of substitution between energy types are set at 0.9 for old capital and 1.8 for new capital. 46 Sectoral wage premia are calibrated using actual labor remuneration per labor types in the different sectors. Super profits are calibrated using actual sectoral gross operating surpluses over capital stocks. Source: Quantec (2017). 47 Net transfers received by the poorest 40 percent of households from richer deciles average 8 percent of their disposable incomes. 51 | South Africa Economic Update 11 | Annex consumption and savings, in fixed proportions of Endogenous savings/investment determines the total their nominal values. Government and investment physical capital stock of the next year.48 Labor supply demands are disaggregated in sectoral demands results from an exogenous trend (reflecting working- once their total value is determined according to age population growth and education projections, see fixed coefficient functions. Box 2.3) and a response (with a supply elasticity ranging from 0.1 for skilled workers to 0.3 for unskilled workers) The model assumes imperfect substitution among to real wages (average wage per labor type divided by goods originating from different geographical areas. the typical consumer price index of the labor type). Import demand results from a Constant Elasticity of Participation rates thus depend on labor remuneration Substitution function of domestic and imported goods and cost of working. Water supply is constant and (with a substitution elasticity of 2.0 between imports mineral supplies (or depletion rates) respond to and domestic products; and 5.0 between imports origin). changes in their remunerations. Population growth is Export supply is symmetrically modeled as a constant set exogenously, as is total factor productivity growth. elasticity of transformation function (with a substitution World prices and net financial transfers from the rest elasticity of 2.0 between domestic products and exports; of the world are also set exogenously. The model is run and 5.0 between export destination). Producers decide annually from 2012 to 2030. to allocate their output to domestic or foreign markets responding to relative prices. A microsimulation module using computable general equilibrium results as inputs projects poverty in Several macroeconomic constraints are introduced in its various dimensions. 49 The module simulates this model. First, the small country assumption holds, trajectories of poverty and distributional change the South African economy being unable to change under several growth and policy scenarios drawing world prices; thus, its imports and exports prices on the macroeconomic from the computable general on world markets are exogenous. Capital transfers equilibrium, including demographic variables are exogenous as well, and therefore the current (composition of the population by age and education); account balance is fixed, so as to achieve the balance labor market variables (employment by sector, wages, of payments equilibrium. Second, the model imposes and farm profits); and exogenous income variables fixed real public expenditures to reflect government’s (public transfers and taxes, private transfers). choice of delivering a given amount and quality of Thus, it simulates the poverty and distributional public services and ability to borrow. Tax rates and implications of demographic change, changes in foreign transfers to government are exogenously occupations and labor incomes, as well as public determined and thus government savings are residually transfers. Demographic changes are simulated determined. Third, investment is determined by the using reweighting techniques, while the core of the availability of savings from government, households, simulation model is an empirical representation of the and the rest of the world. The numeraire of the model income generation process with household income is the exchange rate. Balance of payments equilibrium being composed of individual labor incomes and net is thus obtained through quantities and real exchange public and private (including remittances) transfer rate adjustments. income and other income sources. The analysis is based on the latest South Africa Living Condition The dynamic path of the model depends on several Survey 2014/15. endogenous behaviors and exogenous factors. Annex | South Africa Economic Update 11 | 52