98880 South Africa Economic Update Jobs and South Africa’s Changing Demographics South Africa Economic Update 7 12 aug 15.indd 1 8/12/15 2:07 PM © 2015 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street NW Washington, DC 20433 USA All rights reserved This report was prepared by the staff of the Country Management Unit for Southern Africa; the Trade and Competitiveness, Social Protection and Labor and Development Prospects Group; the Macroeconomics and Fiscal Management Global Practice; and the Poverty Global Practice. The findings, interpretations, and conclusions expressed herein are those of the authors and do not necessarily reflect the views of the World Bank’s Board of Executive Directors or the countries they represent. Cover photos: Kids Left - Håkon Stillingen; Kids Right - Sarah Wiseman; Woman working feath- ers - South Africa Tourism; Man with wheelbarrow - Stephen Wolfe; Other - Flickr @ Creative Commons. The report was designed, edited, and typeset by Communications Development Incorporated, Washington, DC. South Africa Economic Update 7 12 aug 15.indd 2 8/12/15 2:07 PM Contents Foreword   v Acknowledgments   vi Executive Summary   1 Section 1 Recent Economic Developments   9 Global economic developments and prospects    9 Recent trends in South Africa    13 Labor markets   19 Fiscal policy   23 Monetary policy and inflation    24 External sector   25 Economic outlook   26 Notes   29 Section 2  Jobs and South Africa’s Changing Demographics    31 Introduction   31 How far has South Africa progressed in its demographic transition?    35 South Africa’s recent labor market performance amid demographic change    37 Making the most of South Africa’s future demographic changes    42 Conclusions   47 Notes   48 References   51 Boxes 1.1 How have fuel and electricity price changes affected households in South Africa?   16 1.2 What do exports mean for jobs and skills demand in South Africa?    21 1.3 Comparing regulations at the local level: subnational Doing Business in South Africa 2015   28 2.1 Demographic change and the long-term sustainability of social spending    32 2.2 Sector-level trends in employment and productivity    40 Figures 1.1 Global manufacturing purchasing managers’ index, industrial production, and trade all weakened through May 2015    9 iii South Africa Economic Update 7 12 aug 15.indd 3 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S 1.2 Despite a decline in the U.S. rig count, oil markets remain oversupplied    10 1.3 Global commodity prices are expected to remain soft    11 1.4 Gross capital flows to developing countries are robust    12 1.5 Global growth outlook to rise modestly in 2015    12 1.6 Growth outlook in key developing countries    13 1.7 GDP growth is stuttering    13 1.8 Total manufacturing capacity underutilization has deepened since early 2008    18 1.9 The gap between electricity supply and GDP is growing (2006 = 100)    18 1.10 Unemployment trends are set negative    19 iv 1.11 Manufacturing job numbers remain the hardest hit by the crisis    20 1.12 Higher employment growth rates are needed to start bridging the jobs gap    23 1.13 How the debt-to-GDP ratio has evolved with each vintage of the budget    23 1.14 Core inflation is nudging the upper bound of the inflation target    25 1.15 External imbalances haves improved a little recently, but South Africa remains vulnerable   25 2.1 How demographics can affect the economy and spur higher growth and living standards   32 2.2 South Africa’s changing demographic and age profile    34 2.3 South Africa’s demographic window is open    35 2.4 The dependency ratio in South Africa at its trough will be higher than in East Asia   36 2.5 The demographic window is already closed in OECD countries and close to closing in China and East Asia    37 2.6 South Africa is further advanced in its demographic transition than the rest of Sub- Saharan Africa but stands to benefit from a growing region    37 2.7 Skilled vs. unskilled workers    38 2.8 Drivers of 1.7 percent increase in real per capita GDP, 2000–13    39 2.9 Stepping stones in realizing a demographic dividend    43 2.10 Number of people employed in baseline and scenario 1    44 2.11 Rapid improvements in unemployment rates, labor productivity, and educational attainment can accelerate growth in income and consumption    45 2.12 Poverty headcount rates will fall below 5 percent by 2030 and inequality can be reduced by increasing productivity and educational attainment, and generating more jobs   46 Tables 1.1 GDP components   14 1.2 Aggregate demand components   15 1.3 Government and related entities have picked up some of the slack in creating jobs   22 1.4 Economic outlook through 2017    26 2.1 Evolution of the South African labor market between 2000 and 2014    38 2.2 Evolution of ratio of skilled to unskilled workers by sector    39 South Africa Economic Update 7 12 aug 15.indd 4 8/12/15 2:07 PM Foreword Job creation is one of South Africa’s most press- growing working-age population to spur ing challenges. The country is plagued by one growth and improve living standards. It of the highest levels of unemployment and eco- underscores the importance of job-­ intensive nomic inactivity for middle-income countries. growth and the need for the education and More than one-third of its labor force is out of training system to produce a more skilled work or has stopped searching for jobs. The workforce better attuned to evolving labor unemployment rate for young people is close market demands. to 50 percent, and many more of them are nei- This and much more is discussed in ther employed nor in the education system. greater detail in this report, which uses In the coming decades, millions more young ­ evidence-based analysis to draw its conclu- people will join the working age population sions and recommendations. The report also and more jobs will have to be created if South presents recent economic developments in Africa is to harness this demographic oppor- South Africa, where on-going power outages, tunity to accelerate growth and raise living difficult labor relations, and policy uncer- standards. Without this South Africa risks fac- tainty have compounded external headwinds ing a worsening economic situation of rising from lower commodity prices and slowing unemployment and greater dependency where growth in China. those fortunate enough to work and the state I sincerely hope that this report’s find- will have to shoulder a greater burden. Job cre- ings promote informed dialogue and policy ation is at the heart of the Government’s 2012 debate about the country’s development National Development Plan, which aims to cre- priorities pertaining to job creation and ate 11 million jobs by 2030. In an environment economic inclusion in a context of major of slow economic growth, progress towards demographic changes. this goal has been painfully slow. This seventh issue of the World Bank’s Economic Update for South Africa focuses on the challenges posed by changing demo- graphics and its implications for jobs and labor markets in South Africa. Its analyses Guang Zhe Chen center on how job creation, higher savings, Country Director for Botswana, and a sharper focus on improving the human Lesotho, Namibia, South Africa, capital of school leavers and the unemployed Swaziland, Zambia, and Zimbabwe can create the enabling conditions for the World Bank v South Africa Economic Update 7 12 aug 15.indd 5 8/12/15 2:07 PM Acknowledgments This edition was prepared by a core team (DECSN) box 3 in chapter 1. Gerard Kambou comprising Catriona Mary Purfield (AFCS1) (DECPG) and Fernando Im (GMFDR) and Fernando Im (GMFDR). This report prepared chapter 1 and Phindile Ngwenya is the seventh in the series of the South (GMFDR) provided very able research Africa Economic Updates (SAEU) that was assistance. Peer reviewers were Hans Lofgren, launched in July 2011. The SAEUs aim to (DECPG), Maryla Maliszewska (DECPG), position the Bank on critical development Delfin Go (Consultant), Syed Ejaz Ghani issues in South Africa by carrying out timely (GMFDR), Csilla Lakatos (DECPG), and assessments of macroeconomic developments Thomas Moultrie (University of Cape Town). and select structural issues. Each issue The report was prepared under the overall comprises two sections, the first summarizing guidance and supervision of Asad Alam the recent economic developments and the (Country Director, AFCS1) and Mark Roland second analyzing more in depth a topic of Thomas (Practice Manager, GMFDR). special interest. The seventh issue of the We are also grateful for the input received SAEU examines the implications of South from government counterparts. At South Africa’s changing demographics for jobs and Africa’s National Treasury, we would like growth. to thank Ian Stuart, Faaiqa Salie, Theresa Catriona Purfield task lead the focus Anton, and Catherine Macleod for the chapter, leading a team comprising: Lucilla comments received. We would also like Bruni (GSPDR), Sy ud A mer A hmed to thank Chris Loewald, the colleagues (DECPG), Marcio Jose Vargas da Cruz at the South African Reserve Bank and (DECPG), William G. Battaile (GMFDR), John Kruger, Josephilda Nhlapo-Hlope, and Fernando Im. Victor Sulla (GPVDR) and Khulekani Mathe, Percy Moleke of the Mariam Lomaia (GSPDR) prepared box 1, South African Presidency, as well as Kefiloe Thomas Farole (GCJDR) and Claire Honore Masiteng, Statistics South Africa, for their Hollweg (GTCDR) box 2 and Trimor Mici helpful suggestions and inputs. vi South Africa Economic Update 7 12 aug 15.indd 6 8/12/15 2:07 PM Executive Summary Global economic developments first quarter of 2015. But in the second quar- and prospects ter this dampening impact started dissipat- Global growth hit a soft patch at the start ing in a number of oil-importing countries. of 2015. In the first quarter it slowed to 2.0 Long-term inflation expectations have recov- percent, quarter on quarter (q/q) annual- ered in both the Euro Area and the United ized, from 2.3 percent in the fourth quarter States, moving closer to central bank targets. of 2014. The slowdown was reflected in a After a rebound in March, gross capital decline in global manufacturing activity, on flows to developing countries maintained the back of weak industrial production and their robust momentum in the second quar- goods trade data, especially in large emerg- ter of 2015. Capital inflows to developing ing markets. countries averaged $61 billion per month Among high-income countries, economic in the second quarter of 2015. The pick-up activity is strengthening. Euro Area growth stemmed from a surge in international bond picked up to 1.6 percent in the first quarter of issuance, reflecting large corporate issues 2015 from 1.3 percent the previous quarter. from East Asia and Latin America. But equity Growth in Japan accelerated to 2.4 percent in flows have fallen significantly due to a sharp the first quarter from 1.1 percent in the pre- drop in flows to China, mirroring the plunge vious quarter. For the second quarter, despite of the Chinese stock market since mid-June. the uncertainty surrounding Greece, the Looking ahead, global activity should Euro Area Composite PMI in July remained be supported by low commodity prices and close to a four-year high. Confidence indica- generally still-benign financing conditions, tors for the second quarter point to further even as U.S. monetary policy normalizes. improvements, while indicators in Japan also According to the World Bank Global Economic suggest continued, albeit moderate, growth. Prospects of June 2015, global growth is In the United States, growth continues to projected to rise modestly to 2.8 percent in strengthen. The economy contracted by 0.7 2015 from 2.6 percent in 2014 with a further percent in the first quarter, reflecting the moderate pickup in 2016 and 2017 to more impact of temporary factors (bad weather, than 3.2 percent. port disruptions), but recovered in the sec- Growth in high-income countries is ond quarter with real gross domestic product projected to average 2.0 percent in 2015 (GDP) expanding by 2.3 percent on the back and average 2.3 percent in 2016 and 2017, of robust growth in household consumption. up from 1.8 percent in 2014, driven largely Job creation continued at an above-trend by the United States and reflecting gradual pace, and the unemployment rate declined strengthening in the Euro Area and Japan. to 5.3 percent in June, its lowest since 2008. Growth in the United States is expected to The sharp fall in oil prices since June 2014 rise to 2.7 percent this year and 2.8 percent continued to weigh on global inflation in the the next, from 2.4 percent in 2014. In the 1 South Africa Economic Update 7 12 aug 15.indd 1 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Euro Area, growth is expected this year headline growth as maize production fell. at 1.5 percent, up from 0.9 percent in 2014 After the prolonged strike in the platinum with some modest strengthening in 2016. In sector in the first half of 2014, mining pro- Japan, it is projected to reach 1.1 percent in duction gathered momentum over the rest 2015, from zero growth in 2014. of 2014, with growth peaking at 15.2 percent By contrast, growth in developing coun- q/q in the final quarter as producers sought tries is projected to slow this year as they to bring production back to pre-strike levels. adjust to lower commodity prices and tighter Growth eased somewhat in the first quarter financing conditions ahead. Next year should of 2015 to 10.2 percent q/q, reflecting in part 2 see a slight recovery. Their aggregate growth lower mineral prices, but mining still contrib- is expected to be 4.4 percent in 2015, down uted slightly more than half of total headline from 4.6 percent in 2014, reflecting China’s real GDP growth. Prospects for the mining sustained slowdown. Growth in Sub-Saharan sector in the second quarter are dim. In Africa is forecast to slow to 4.2 percent in April and May, mining sector output shrank 2015 and to around 4 percent in 2016 and by 4.7 percent m/m seasonally adjusted in 2017, from 4.6 percent in 2014, as low oil and each month as lower commodity prices and mineral prices reduce growth in commodity- power outages contributed to sharp contrac- exporting African countries. tions in coal, platinum group metals, and In South Africa, domestic factors con- iron-ore production. tinue to impede the recovery in economic Activity in the secondary sector resumed growth. Labor unrest and electricity short- its downward trajectory in the first quarter as ages held headline real GDP growth to just electricity outages took over from strikes as 1.5 percent in 2014, entrenching the modera- the main check on manufacturing recovery. tion in economic activity evident since 2011. The rebound from strike-induced losses in For the first time since 2009, aggregate eco- manufacturing in the last quarter of 2014 nomic growth fell short of population growth proved short-lived. The sector contracted 2.4 in 2014, reducing per capita real GDP by 0.4 percent q/q in the first quarter, subtracting percent to Rand 55,712 ($6,800). Poverty 0.3 percentage points from headline q/q and inequality rates are therefore likely to growth, as it struggled with more frequent have stayed broadly unchanged from those power outages and falling external demand recorded in the 2010/11 household expendi- for steel. After five quarters of robust growth, ture survey where 21.7 percent of the popula- construction activity also decelerated sharply tion lived below the food poverty line (Rand in the first quarter. Manufacturing output 335 per capita per month) and the Gini coef- shrank by 2.0 percent and 0.4 percent ficient of income inequality stood at 0.69. m/m seasonally adjusted in April and May, Growth lost momentum in the first quar- respectively, as severe power outages took ter of 2015. Real GDP expanded by 1.3 per- their toll on production, suggesting that cent q/q seasonally adjusted annual rate (2.0 growth in this sector will remain weak in the percent y/y), significantly below the 4.1 per- second quarter. cent q/q growth rate registered in the final Growth in the tertiary sector also mod- quarter of 2014, which was boosted by the erated, despite a robust performance from recovery in output from the first half of 2014. the financial, real estate, and business ser- Agriculture and manufacturing subtracted vices subsector. Overall sector growth edged 0.7 percent from headline q/q growth, with down to 1.5 percent q/q in the first quarter, mining and financial services the only two its slowest pace since the 2009 global finan- sectors making a significant contribution. cial crisis, reflecting a 0.8 percent q/q con- On a quarterly basis, growth deceler- traction in government services which last ated in the primary sector largely due to the experienced a contraction of this severity in impact of a drought on maize production 2001. The key bright spot in the economy, and moderation in the growth of mining and building on its strong performance in output. After six quarters of steady growth, 2014, was the finance, real estate, and busi- agricultural production contracted sharply ness services subsector: its growth rose to 3.8 (down 16.6 percent q/q) in the first quar- percent q/q in the first quarter and contrib- ter, subtracting 0.4 percentage points from uted 0.7 percentage points to q/q headline South Africa Economic Update 7 12 aug 15.indd 2 8/12/15 2:07 PM growth. Wholesale and retail trade recovered These structural factors are reinforc- somewhat in the first quarter, yet continued ing cyclical weakness in domestic demand. to reflect weakness in private consumption Household consumption is expected to grow demand. only modestly. Lower oil prices provided The current environment of weak growth only temporary relief to household bud- and confidence is making it difficult to gets and headline inflation. The prospect tackle the triple challenges of high unem- of higher electricity tariffs and potentially ployment and fiscal and external imbal- higher pass-through of a more depreciated ances. The current environment has not rand to domestic prices, will limit space for been favorable for any sharp pick up in job households to expand consumption. High 3 creation and the unemployment rate has unemployment and indebtedness, along remained high at 25 percent. On the fiscal with tightening credit standards, continue front, the 2015 budget has targeted a grad- to weigh on consumer sentiment, while gov- ual reduction in the budget deficit from 3.7 ernment consumption is subdued because of percent of GDP in 2014–15 to 2.5 percent of consolidation efforts. GDP by 2017–18 to help stabilize the debt Concerns over electricity supply and rising burden. But achieving these targets in an input and wage costs are being compounded environment of low economic growth and by broader commodity price weakness as well rising financial pressures at state-owned as policy and regulatory uncertainty, and are enterprises will be challenging. Headline likely to dampen the outlook for investment. inflation eased temporarily on the back of Labor relations are expected to remain dif- lower fuel prices. But with core inf lation ficult in an environment of weak growth. close to the upper end of the inflation tar- On the plus side, the recovery in advanced get band and rand weakness increasing the countries and still-robust growth in Sub- risk of pass through to prices, the SARB Saharan Africa should boost demand for has raised policy interest rates. The current South Africa’s non-mineral exports. Still, the account deficit remains wide and is largely anticipated weakness in mineral demand and funded by capital inflows. prices limit the overall improvement in the current account deficit, which is expected Economic outlook to stay elevated at around 5.0–5.2 percent of Real GDP growth is forecast at 2.0 percent in GDP over the medium term. Given the weak 2015, and to slowly strengthen to 2.4 percent recovery, South Africa’s output gap, put at 1.2 in 2017. Agricultural growth is expected to percent of potential growth in 2014, is pro- be pulled down in the near term by the poor jected to narrow only slowly toward the end maize harvest, but finance and business ser- of the forecast horizon. vices are set to grow robustly throughout the Given the weak recovery, little progress forecast period. But overall, growth in South is expected against the triple challenges of Africa will remain largely below the average high unemployment, deep poverty, and wide growth rate of 4.2 percent and 4.0 percent for inequality. Agricultural growth is relatively Sub-Saharan Africa, in 2015 and 2016–2017, weak, while the extractives and metals sectors respectively. continue to struggle against lower minerals Structural impediments will continue prices and rising production costs, causing weighing heavily on growth. Power-supply further job losses in these sectors. With little bottlenecks are one of the main underly- change in growth drivers or their job inten- ing factors in the downgraded revision of sity expected, unemployment is set to remain the growth outlook since the previous eco- sticky and high. Extreme poverty may well nomic update. Relative to our forecast in the remain broadly unchanged at its current November 2014 Economic Update (Volume level, reflecting the low rate of economic 6), real GDP growth has been revised down growth. Our forecasts do not foresee inequal- by 0.5 and 0.8 percentage points for 2015 and ity narrowing either, because a high rate of 2016, respectively. We expect some improve- joblessness means that the gap in incomes ment in the electricity situation only toward between the employed and unemployed per- the end of the forecast period as new power- sists. Growth in the consumption of the bot- generation capacity comes online. tom 40 percent is seen flat at about 1 percent. South Africa Economic Update 7 12 aug 15.indd 3 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Downside risks to this already-weak eco- under one-third of the new entrants to the nomic outlook prevail. On the external working-age cohort since 2000 found a job, side, they include a sharper than expected and by mid-2015 only a little more than 40 slowdown in the Chinese economy, bouts of percent of the working-age population were risk aversion and financial-market volatility employed. Unemployment today is higher due to the normalization of U.S. monetary that it was at the end of apartheid. To the policy and faltering growth in the Euro Area extent that unemployment lowers current amid uncertainty surrounding developments and future earnings potential through the in Greece. China is still a major market for erosion of skills and human capital during 4 South African mineral exports, and Europe periods of prolonged joblessness, lifetime for manufactured exports, so a slowdown earnings of workers are lower and the ability in either would reduce demand for South to generate savings is constrained. Savings as Africa’s exports. The growing size of some a share of GDP have fallen as the working- regional markets may partly counter any hit age population expanded. Real per capita from these markets. But South Africa’s large income in 2014 was only 40 percent higher current account deficit, financed heavily by than in 1994, trailing the increases not only volatile capital flows, makes the economy in East Asia but also in Latin America, two vulnerable to shifts in investor sentiment and regions that experienced similar increases in shifting capital flows that could arise in the share of the working age population as South context of fall-out from the “lift-off” in inter- Africa. est rates in the United States. Our analysis focuses on how job cre- On the domestic front, if labor relations ation, savings, and better human capital can do not improve or if power disruptions improve enabling conditions for the grow- worsen, growth could well disappoint fur- ing working-age population to contribute to ther. And if wage settlements continue to faster economic growth. As the “youth bulge” exceed inflation and productivity gains, com- passes into working age, a country enters its petitiveness will erode, undermining the role “demographic window of opportunity,” but a of net exports in supporting the recovery. dividend does not materialize automatically. While addressing power shortages will To realize the first demographic dividend, be critical to removing a key break on growth must be jobs intensive and the educa- near-term growth, achieving the 5 percent tion system has to prepare graduates with the growth target of the National Development skills demanded by the labor market. Other- Plan requires much more than that. South wise, a country risks a worsening economic Africa needs urgently to boost growth to this situation with rising unemployment and level if it is to provide jobs for young work- increased dependency. The second dividend ers, address its growing social tensions, and accrues when consumption rises more slowly reduce its substantial poverty and inequality. than incomes and requires not only mecha- Improved labor relations, matched by greater nisms to encourage saving but also produc- collaboration between the public and private tive investment in human capital (knowledge sectors and policy certainty to improve the and skills) so labor productivity can improve business environment, are fundamental to and boost workers’ earnings. restoring confidence. South Africa is in its demographic win- dow of opportunity and will remain there Jobs and South Africa’s for around 50 more years. Since 1994, its changing demographics working-age population (15–64 years) has In South Africa, a lack of jobs has stymied increased by 11 million. In the next 50 years, its ability to capitalize on its demographic it will grow by another 9 million. Having opportunity. Since 1994, the working-age such a high share of its population—68.3 population has expanded by more than 11 percent at its peak in 2045—in its working million. Job opportunities created were con- prime presents a tremendous opportunity centrated in services, but not in agriculture, for the country to boost its growth and raise manufacturing, and mining, which shed living standards. But it also presents tremen- workers, with the result that total jobs created dous challenges. In the next 15 years alone, fell far short of the growing labor supply. Just the working-age population will expand by South Africa Economic Update 7 12 aug 15.indd 4 8/12/15 2:07 PM 280,000 a year. These people will have to find people employed grows almost three times productive jobs if South Africa is to harvest faster than in the business-as-usual case, with the potential boost to growth and living stan- jobs growth averaging 2.2 percent a year. By dards before it faces the rising burden associ- 2030, 4.03 million more people are work- ated with a rapidly growing elderly. ing than in the previous case and relative to To illustrate how a growing working 2015 some 5.8 million new jobs are created, age population has the potential to affect which is more than enough to employ the growth, incomes, poverty, and inequality, we new entrants and allow South Africa to make contrast a business as usual scenario where a significant dent in its unemployment rate. current high rates of unemployment and low But even with more jobs, the gains to 5 rates of job creation persist through 2030, growth and income from a growing work- to three other potential futures that evolve ing-age population appear relatively modest around creating a better enabling environ- in South Africa because aging is offsetting ment—one where more jobs are created, a some of the benefits. By 2030 the share of second where greater job creation is accom- the population older than 65 will already panied by training to improve worker skills, have reached 8 percent, 2.25 percentage and a third where job creation and better points more than today. This is limiting the skills are accompanied by better education scope for a potential second demographic attainment by new entrants. dividend. Because the old draw down their If unemployment and employment remain savings faster than the working-age share of constant at current rates, South Africa sees the population can increase national sav- only a very modest boost to growth and ings, the increase in the old-age dependency incomes from its rising working-age popu- ratio, particularly after 2027, is high enough lation. In the baseline scenario, real GDP to limit the improvements in overall savings growth averages 3.7 percent a year and and the secondary boost to economic growth growth in real per capita incomes averages via greater investment. Total savings reach 3.1 percent a year over 2015–30. The number 19.5 percent of GDP by 2030, only about 1 of employed rises modestly from about 14.8 percentage point of GDP higher than in the million workers in 2015 to 16.6 million by business-as-usual scenario. 2030. This implies that only about 40 percent Higher growth in labor productivity will of the 4.3 million increase in the working-age need to accompany job creation if South cohort over the next 15 years will find jobs. Africa is to make greater strides in raising Moreover, the high rates of youth unemploy- growth and incomes as the working-age pop- ment imply that the economy is not only for- ulation moves toward its peak. In the pro- going the benefit of the labor supply growing ductivity-enhancing scenario, we examine faster than the overall population, but also how growth and income evolve if the decline missing the potential benefit that these new in unemployment highlighted above is also entrants have more years of schooling. Nor accompanied by improved labor-productivity do incomes rise enough to create space for a growth that could come about through bet- sharp increase in savings: savings as a share ter training and skills. Here, overall real GDP of GDP rise modestly from 14.5 percent of rises at an average rate of 5.3 percent a year, GDP in 2015 to just under 18.5 percent by above the 5 percent target set in the National 2030. Development Plan (2012). The economy’s The job creation scenario illustrates the size more than doubles between 2015 and necessity for strong job creation to employ 2030 and is one-quarter larger in 2030 than the growing working-age population. Here, in the business-as-usual baseline case. In per South Africa’s unemployment rate converges capita terms, real GDP growth averages 4.7 with the upper middle-income country aver- percent a year and per capita incomes reach age of 5.8 percent by 2030 and the employ- US$12,558 by 2030. The increase in produc- ment rate rises to about 54 percent. Labor tivity helps offset some of the pressures on productivity growth is positive but not as savings and growth from having a growing strong as in other emerging market econo- share of the population older than 65. Sav- mies. In this more optimistic future, growth ings as a share of GDP is 20.8 percent of GDP is far more jobs intensive: the number of in 2030, which allows South Africa to reap South Africa Economic Update 7 12 aug 15.indd 5 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S a somewhat higher dividend from its large growth could accelerate to 5.4 percent a year working-age population. and per capita incomes could double by 2030, The impact of a growing working-age which would virtually eliminate extreme pov- population on growth and incomes would erty and begin to reduce inequality. be further enhanced by efforts to improve Changing the growth and jobs dynamics educational attainment and skills. In the will require urgent action on several mutu- third scenario, job creation and faster labor- ally reinforcing fronts. The government productivity growth is accompanied by has already introduced an employment tax these efforts. In an accelerated educational- incentive to encourage firms to hire young 6 attainment scenario, overall real GDP and workers. Through its Industrial Policy Action real GDP per capita rise at an average of 5.4 Plan, it is also offering incentives to promote percent and 4.8 percent a year, respectively. potentially labor-intense sectors like manu- Per capita incomes reach US$12,766 and sav- facturing and agriculture. Faster and deeper ings rise to 21 percent of GDP by 2030. The global and regional integration in trade in gain in the economy of higher educational goods and services would bolster this effort. attainment is understated by these numbers, Nonetheless, low-cost, labor-intense pro- because 15 years is not long enough for the duction is unlikely to be the main engine full impact of these efforts on growth and for job creation for South Africa, given how incomes to fully materialize. these sectors have shrunk over the past two Extreme poverty—measured by the share decades. Policies also need to focus on devel- of the population living on less than $1.25 oping services, small and medium firms, per day—could be virtually eliminated if and household enterprises, including in the South Africa could combine jobs creation informal economy, as engines for job cre- with improvements in its labor productivity ation. Policies that improve the business envi- and educational attainment. In the business- ronment, especially for small firms, could as-usual baseline the share of the population include reducing the burden of red tape, living on less than $1.25 (PPP adjusted) a day improving access to low-cost finance, and falls from 9.4 percent in 2011 to 2.7 percent securing greater flexibility in labor–market in 2030. However, the extreme poverty rate regulations. falls to about 1 percent of the population The greatest priority on the supply side in the most optimistic scenario. When mea- is to improve levels of educational attain- sured at the $2 a day poverty line, the poverty ment in South Africa. Getting basic school- rate falls by almost three-quarters from 23.7 ing right is the first step to ensuring that percent in 2011 to about 4 percent in 2030 in school leavers and graduates have the foun- the most optimistic scenario. dational skills necessary to function in the Our analysis shows that more job-intensive modern workplace. Educational attainment growth would help tackle unemployment and not only shapes employment opportunities, create jobs for many new labor-force entrants but also provides the foundation for further in the coming 15 years, allowing South Africa on-the-job learning and training. This will to take the first step in harnessing its favor- not be an easy task. South Africa has already able demographics. But the analysis also achieved almost universal school attendance shows that simply increasing the number of and the challenge now is to improve learning jobs will not be enough to allow South Africa outcomes by better training and support of to boost savings and derive the second demo- teachers. graphic dividend. More job-intensive growth Promoting skills, especially of the long- needs to be accompanied as a first priority term unemployed, also calls for complemen- by improving the quality of education so that tary efforts. Steps are being taken to scale better educated youth are entering the work- up technical and vocational education and force. This needs to be complemented by training opportunities. But access remains efforts to improve the productivity of existing a challenge, especially for the young who workers and the unemployed through better lack a high school qualification, while the skills development and training. By creat- system continues to face problems of quality ing a virtuous circle of job-intensive growth, and institutional capacity in supplying the improved productivity, and higher savings, skills demanded by employers. To broaden South Africa Economic Update 7 12 aug 15.indd 6 8/12/15 2:07 PM access, training opportunities should match with the private sector will ensure that they the actual educational levels of young unem- are high quality and better geared to the ployed, while financial support for enrollees needs of the labor market. Employers can could help overcome the hurdles of high also provide internships and other opportu- transport and living costs. Designing training nities for practical training to help overcome programs in consultation and partnership new entrants’ lack of work experience. 7 South Africa Economic Update 7 12 aug 15.indd 7 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S 8 South Africa Economic Update 7 12 aug 15.indd 8 8/12/15 2:07 PM SECTION 1 Recent Economic Developments Global economic developments 2015 from 1.3 percent the previous quarter, and prospects bolstered by a weakening euro, declining oil prices, and more accommodating financing Sources of global growth rotate as activity conditions. Growth in Japan accelerated strengthens in high-income economies to 2.4 percent in the first quarter from 1.1 but slows in developing economies percent in the previous quarter, helped Global growth hit a soft patch at the start by a recovery in private consumption and of 2015. In the first quarter it slowed to investment. For the second quarter, despite 2.0 percent, quarter on quarter (q/q) the uncertainty surrounding Greece, the annualized, from 2.3 percent in the fourth Euro Area Composite PMI in July remained quarter of 2014.1 The slowdown was reflected close to a four-year high. Confidence in a decline in global manufacturing activity, indicators for the second quarter point to on the back of weak industrial production further improvements, while indicators and goods trade data (figure 1.1), especially in Japan also suggest continued, albeit in large emerging markets. moderate, growth. Among high-income countries, economic In the United States, growth continues to activity is strengthening. Euro Area growth strengthen. The economy contracted by 0.7 picked up to 1.6 percent in the first quarter of percent in the first quarter, reflecting the Global manufacturing purchasing managers’ index, industrial production, and Figure trade all weakened through May 2015 1.1 Manufacturing PMI (LHS) Industrial Production Global trade (RHS) 54 6 5 52 Year-over-year, percent 4 Index 50 3 2 48 1 46 0 Jan-12 Jun-12 Nov-12 Apr-13 Sep-13 Feb-14 Jul-14 Dec-14 May-15 Source: World Bank, CPB, Haver Analytics. 9 South Africa Economic Update 7 12 aug 15.indd 9 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Figure Despite a decline in the U.S. rig count, oil markets remain oversupplied 1.2 U.S. oil rig count (LHS) Saudi Arabia oil production (RHS) U.S. oil production (RHS) 1800 11 1600 10 1400 Million barrels per day 9 1200 Count 8 1000 10 7 800 600 6 400 5 Jan-10 Jun-10 Nov-10 Apr-11 Sep-11 Feb-12 Jul-12 Dec-12 May-13 Oct-13 Mar-14 Aug-14 Jan-15 Iron ore prices are Source: World Bank Global Economic Prospects, June 2015, Baker Hughes, International Energy Agency. forecast to decline by impact of temporary factors (bad weather, of future production) of 60 percent since 46 percent in 2015 port disruptions), but recovered in the the peak in October 2014, oil markets have second quarter with real GDP expanding remained oversupplied as U.S. oil production by 2.3 percent on the back of robust growth has adjusted only slowly, oil production by in household consumption. Job creation high-income oil producers rose modestly continued at an above-trend pace, and the with new supply also expected from Iran, unemployment rate declined to 5.3 percent and inventories remained high (figure 1.2). in June, its lowest since 2008. Uncertainty on supply adjustments has In developing countries, economic growth increased price volatility since November continued to weaken, South Africa’s key 2014, but oil prices are expected to remain trading partners included. China’s gross well below their 2013 levels over the next domestic product (GDP) grew by 7.0 percent decade. in 2015Q2, up from 5.3 percent in the first Weak global demand, U.S. dollar quarter as stimulus measures supported appreciation, and low oil prices have put activity. Reflecting the softening of growth pressure on non-oil commodity prices. Metals in China relative to previous years, real (including iron ore) and agricultural raw GDP growth in the first quarter slowed in materials prices—among the commodities East Asia, including Indonesia, Malaysia, most sensitive to the business cycle—were and Thailand. This weakness appears at the end of June about 15 percent lower to have extended to the second quarter, than a year earlier. Metals prices are now 43 where industrial production in several percent below their February 2011 high and emerging-market countries struggled to are projected to decline by 13 percent in gain pace (Indonesia, Malaysia) or turned 2015 relative to 2014, due to excess capacity negative (Brazil, the Russian Federation, the and slowing demand in China. The largest Republic of Korea, and Thailand), reflecting decline is expected for iron ore (down 46 adjustment to lower commodity prices and percent for 2015), reflecting new low-cost weaker exports, as well as domestic factors. mining capacity (mainly in Australia) coming online in 2015 and 2016. Fertilizer prices (a Commodity prices will remain key input to most agricultural commodities) low but volatile and agricultural prices more widely are also Over the second quarter, oil prices recovered expected to fall on weaker demand, ample from their January 2015 lows, but in July they supply, and reduced energy costs (figure 1.3). weakened again and are now projected at an annual average of $57 per barrel in 2015, The global inflation outlook is favorable compared to $96 per barrel in 2014. Despite The sharp fall in oil prices since June 2014 a decline in the U.S. oil-rig count (a measure continued to weigh on global inflation in South Africa Economic Update 7 12 aug 15.indd 10 8/12/15 2:07 PM Figure Global commodity prices are expected to remain soft 1.3 Agriculture Energy Metals and minerals 180 160 Nominal Index, 2010 = 100 140 120 11 100 80 60 40 Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Source: World Bank Global Economic Prospects, June 2015. Global growth is projected to rise the first quarter of 2015, with the proportion quarter of 2015 (figure 1.4). Capital inflows of countries with low inf lation (or even to developing countries averaged $61 billion modestly to 2.8 deflation) unusually high. But in the second per month in the second quarter of 2015, quarter this dampening impact started stronger than the monthly average of $46 percent in 2015 dissipating in a number of oil-importing billion in the first quarter. The pick-up countries. Long-term inflation expectations stemmed from a surge in international have recovered in both the Euro Area and bond issuance, reflecting large corporate the United States, moving closer to central issues from East Asia and Latin America. bank targets. Syndicated bank lending recovered from Divergence in monetary policy persists. weakness in the first four months of the year The European Central Bank launched in but equity flows have fallen significantly due March its quantitative easing program with to a sharp drop in flows to China, mirroring monthly asset purchases of €60 billion until the plunge of the Chinese stock market since at least September 2016, while the Bank of mid-June. Japan maintained its commitment to aggres- sive policy easing. In contrast, the U.S. Fed- High-income countries are set to eral Reserve is expected to start hiking policy drive global growth in 2015 interest rates, albeit gradually, in the second Looking ahead, global activity should be half of 2015. Notwithstanding the recent cor- supported by low commodity prices and rection, expectations of divergent monetary generally still-benign financing conditions, policies have caused the U.S. dollar to appre- even as U.S. monetary policy normalizes. ciate significantly since mid-2014, leading to According to the World Bank Global Economic renewed pressure on emerging and frontier Prospects of June 2015, global growth is market currencies, and increased volatility projected to rise modestly to 2.8 percent in in them. Yet despite these pressures, central 2015 from 2.6 percent in 2014 with a further banks in several large oil-importing develop- moderate pickup in 2016 and 2017 to more ing countries were able to cut interest rates than 3.2 percent (figure 1.5). since the start of the year as inflation moved Growth in high-income countries is closer to policy targets, current account defi- projected to average 2.0 percent in 2015 cits narrowed, and growth remained soft. and average 2.3 percent in 2016 and 2017, up from 1.8 percent in 2014, driven largely Capital inflows to developing by the United States and reflecting gradual countries remain robust strengthening in the Euro Area and Japan. After a rebound in March, gross capital Growth in the United States is expected to flows to developing countries maintained rise to 2.7 percent this year and 2.8 percent their robust momentum in the second the next, from 2.4 percent in 2014. In the South Africa Economic Update 7 12 aug 15.indd 11 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Figure Gross capital flows to developing countries are robust 1.4 Bank Bond Equity 80 70 60 50 US $ billions 40 12 30 20 10 0 Oct-13 Dec-13 Feb-14 Apr-14 Jun-14 Aug-14 Oct-14 Dec-14 Feb-15 Apr-15 Jun-15 Developing country Source: World Bank, Dealogic. growth is expected to slow to 4.4 Figure Global growth outlook to rise modestly in 2015 percent in 2015 1.5 World High-income countries Developing countries 10 8 6 4 Percent 2 0 -2 -4 2007 2009 2011 2013 2015 2017 Source: World Bank Global Economic Prospects, June 2015. Euro Area, growth is expected this year continue accelerating, as lower oil prices at 1.5 percent, up from 0.9 percent in 2014 help boost private consumption. Growth in with some modest strengthening in 2016. In Sub-Saharan Africa is forecast to slow to 4.2 Japan, it is projected to reach 1.1 percent in percent in 2015 and to around 4 percent in 2015, from zero growth in 2014. 2016 and 2017, from 4.6 percent in 2014, as low oil and mineral prices reduce growth in Developing countries will slow in 2015 commodity-exporting African countries. before recovering slightly in 2016 By contrast, growth in developing countries Global risks have diminished but is projected to slow this year as they adjust remain tilted to the downside to lower commodity prices and tighter Some preexisting risks, especially of deflation financing conditions ahead. Next year in the Euro Area, have receded somewhat should see a slight recovery. Their aggregate but new financial-stability and growth risks growth is expected to be 4.4 percent in 2015, have emerged. Deteriorating prospects down from 4.6 percent in 2014, reflecting in some developing economies, especially China’s sustained slowdown (figure 1.6). commodity-exporting ones, are eroding their Growth in India, however, is expected to macroeconomic resilience. This, with the South Africa Economic Update 7 12 aug 15.indd 12 8/12/15 2:07 PM Figure Growth outlook in key developing countries 1.6 2014 2015 2016 2017 10 8 6 Percent 4 13 2 0 -2 -4 Developing countries China India Nigeria Mexico Brazil Russian Federation South Africa Real GDP per capita Source: World Bank Global Economic Prospects, June 2015. in South Africa fell in 2014 possibility of volatility once U.S. monetary in 2014, entrenching the moderation in eco- policy starts normalizing, is increasing nomic activity evident since 2011 (figure 1.7, the risk of financial stress. In addition, a left panel). For the first time since 2009, continued broad-based dollar appreciation aggregate economic growth fell short of popu- could slow the U.S. economy more than lation growth in 2014, reducing per capita real currently expected. If these downside risks GDP by 0.4 percent (figure 1.7, right panel) to materialize at the same time, they could Rand 55,712 ($6,800). Since the global finan- severely disrupt developing-country financial cial crisis, South Africa’s real GDP growth has markets and real economies. fallen well short of the 5 percent target set in the 2012 National Development Plan. Poverty Recent trends in South Africa and inequality rates are therefore likely to have stayed broadly unchanged from those Growth is stuck in low gear recorded in the 2010/11 household expendi- Domestic factors continue to impede the ture survey where 21.7 percent of the popula- recovery in economic growth in South Africa. tion lived below the food poverty line (Rand Labor unrest and electricity shortages held 335 per capita per month)2 and the Gini coef- headline real GDP growth to just 1.5 percent ficient of income inequality stood at 0.69. Figure GDP growth is stuttering 1.7 Real GDP growth Real GDP per capita growth 6.0 5.0 Average 2000-08: Average 2000-08: 4.8% per year 5.0 4.0 3.5 % a year Average 2009-14: 4.0 2.4% per year 3.0 Average 2009-14: 0.8 % a year 3.0 2.0 Percent % 2.0 1.0 1.0 0.0 0.0 -1.0 -1.0 Average 1994-99: -2.0 2.1% per year Average 1994-99: 0.4 % a year -2.0 -3.0 1994 1999 2004 2009 2014 1994 1999 2004 2009 2014 Source: Statistics South Africa. South Africa Economic Update 7 12 aug 15.indd 13 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Growth lost momentum in the first quar- slightly more than half of total headline real ter of 2015. Real GDP expanded by 1.3 per- GDP growth. Prospects for the mining sec- cent q/q seasonally adjusted annual rates3 tor in the second quarter are dim. In April (2.0 percent y/y), significantly below the 4.1 and May, mining sector output shrank by percent q/q growth rate registered in the 4.7 percent m/m seasonally adjusted in each final quarter of 2014, which was boosted by month as lower commodity prices and power the recovery in output from the first half outages contributed to sharp contractions in of 2014. Agriculture and manufacturing coal, platinum group metals, and iron-ore subtracted 0.7 percent from headline q/q production. 14 growth, with mining and financial services Activity in the secondary sector resumed the only two sectors making a significant its downward trajectory in the first quarter contribution. as electricity outages took over from strikes On a quarterly basis, growth deceler- as the main check on manufacturing recov- ated in the primary sector largely due to the ery. The rebound from strike-induced losses impact of a drought on maize production in manufacturing in the last quarter of 2014 and moderation in the growth of mining proved short-lived. The sector contracted 2.4 Electricity outages output. Growth in the primary sector peaked percent q/q in the first quarter, subtracting at a four-quarter high of 13.3 percent in the 0.3 percentage points from headline q/q replaced strikes as fourth quarter of 2014 before sliding to 3.3 growth, as it struggled with more frequent manufacturing’s percent the following quarter (table 1.1). power outages and falling external demand After six quarters of steady growth, agricul- for steel. After five quarters of robust growth, key constraint tural production contracted sharply (down construction activity also decelerated sharply 16.6 percent q/q) in the first quarter, sub- in the first quarter. Manufacturing output tracting 0.4 percentage points from head- shrank by 2.0 percent and 0.4 percent m/m line growth as maize production fell. After seasonally adjusted in April and May, respec- the prolonged strike in the platinum sector tively, as severe power outages took their toll in the first half of 2014, mining production on production, suggesting that growth in gathered momentum over the rest of 2014, this sector will remain weak in the second with growth peaking at 15.2 percent q/q in quarter. the final quarter as producers sought to bring Growth in the tertiar y sector also production back to pre-strike levels. Growth moderated, despite a robust performance eased somewhat in the first quarter of 2015 from the financial, real estate, and business to 10.2 percent q/q, reflecting in part lower services subsector. Overall sector growth mineral prices, but mining still contributed edged down to 1.5 percent q/q in the first Table GDP components 1.1 Percent, seaonally adjusted annualized rate Sector 2011 2012 2013q1 2013q2 2013q3 2013q4 2013 2014q1 2014q2 2014q3 2014q4 2014 2015q1 GDP at market prices 3.2 2.2 1.4 3.7 1.2 5.1 2.2 -1.6 0.5 2.1 4.1 1.5 1.3 Primary sector -0.3 -2.1 10.1 -3.7 10.2 14.8 3.4 -17.2 -1.0 5.2 13.3 0.0 3.3 Agriculture, forestry, and fishing 1.3 0.6 -2.9 -1.1 3.6 6.8 1.5 4.8 5.6 9.5 7.5 5.6 -16.6 Mining and quarrying -0.7 -2.9 14.3 -4.5 12.3 17.2 4.0 -22.8 -3.0 3.9 15.2 -1.6 10.2 Secondary sector 2.3 1.6 -6.1 9.3 -4.3 8.2 0.9 -3.8 -2.5 -0.4 7.2 0.4 -1.4 Manufacturing 2.9 1.9 -7.8 11.7 -6.6 12.3 0.7 -6.4 -4.0 -1.0 9.5 0.0 -2.4 Electricity, gas, and water 1.4 -0.1 -4.8 3.0 1.9 -6.0 -0.6 0.2 -0.5 -1.1 0.3 -0.9 0.7 Construction 0.4 2.1 -0.8 5.1 0.6 3.6 2.7 3.7 2.1 2.2 3.5 2.9 0.8 Tertiary sector 3.8 3.1 2.4 3.2 2.1 2.7 2.5 1.7 1.9 2.4 1.8 2.1 1.5 Wholesale and retail trade, catering, and accommodations 3.8 3.6 0.9 2.2 0.3 1.8 1.9 1.5 -0.2 3.4 -0.3 1.3 1.2 Transport, storage, and communication 3.0 2.5 2.0 1.6 2.6 1.6 2.0 1.4 3.9 2.2 2.9 2.3 1.2 Finance, real estate, and business services 4.1 3.0 4.8 5.0 2.8 2.6 3.0 1.4 1.2 2.4 3.5 2.2 3.8 General government services 4.5 3.6 1.3 2.8 2.8 4.6 3.1 2.3 3.9 2.2 1.2 3.0 -0.8 Personal services 2.4 2.1 1.2 2.7 1.2 1.2 1.8 1.5 1.5 1.3 0.8 1.4 0.9 Source: Statistics South Africa. South Africa Economic Update 7 12 aug 15.indd 14 8/12/15 2:07 PM quarter, its slowest pace since the 2009 global and consumption from lower fuel prices financial crisis, reflecting a 0.8 percent q/q (with spending in non-durable goods pick- contraction in government services which ing up to 4.9 percent q/q in 2015Q1) is likely last experienced a contraction of this severity to have been short-lived, given the reversal in 2001. The key bright spot in the economy, in fuel prices and weaker rand in Q2 as well and building on its strong performance as the increases in the fuel levy and elec- in 2014, was the finance, real estate, and tricity tariffs since February 2015 (box 1.1). business services subsector: its growth rose Tight fiscal space and continuing efforts by to 3.8 percent q/q in the first quarter and the government to contain costs also led to contributed 0.7 percentage points to q/q a contraction in general government con- 15 headline growth. Wholesale and retail trade sumption spending in the first quarter. recovered somewhat in the first quarter, yet Growth in final consumption will likely continued to ref lect weakness in private remain modest in the second quarter. Retail consumption demand. sales were flat in April and expanded only From the demand side, final consump- modestly in May by 0.1 percent m/m. Con- tion continued to drive real GDP growth sumer confidence deteriorated further in in 2014, contributing some 1.3 percent- the second quarter of 2015. The Bureau of Growth in the finance, age points to headline growth. Consump- Economic Research (BER) consumer con- real estate, and tion growth continued to strengthen in the fidence index plunged from –4 in Q1 to first quarter of 2015 as household spend- –15 in Q2, even lower than the lowest point business services ing accelerated to 2.8 percent q/q on the recorded in the 2009 global financial crisis. back of lower fuel and food prices (table High household indebtedness (78.4 percent sector continued to 1.2), adding 1.7 percentage points to head- of disposable income in the first quarter of gain momentum in line growth. Yet high levels of household 2015) elevated joblessness (25 percent in indebtedness, tightened credit, and high the second quarter of 2015), and modest the first quarter unemployment continued to sap consumer credit extension to households all contrib- confidence, with household expenditures uted to deteriorating consumer sentiment on durable goods moderating to just 1.1 per- and households’ diminished willingness to cent q/q. Moreover, the boost to households spend. Table Aggregate demand components 1.2 Percent, seaonally adjusted annualized rate Component 2011 2012 2013q1 2013q2 2013q3 2013q4 2013 2014q1 2014q2 2014q3 2014q4 2014 2015q1 Total final consumption 4.1 3.4 2.2 1.7 1.5 2.2 3.0 1.2 1.4 1.2 1.4 1.5 1.6 Final consumption expenditure by household (PCE) 4.9 3.4 1.9 1.9 1.8 1.7 2.9 1.0 1.0 1.1 1.6 1.4 2.8 Durable goods 15.9 11.2 7.2 7.6 7.4 6.6 9.0 4.8 3.4 4.0 5.3 5.3 1.1 Semidurable goods 5.9 5.9 4.3 4.1 4.6 5.2 5.3 1.9 1.8 2.8 3.3 3.2 3.3 Nondurable goods 4.4 3.0 2.1 1.9 0.0 1.4 1.5 0.8 0.9 -0.1 1.5 0.8 4.9 Services 3.0 1.6 -0.1 0.1 1.4 0.1 2.2 0.1 0.4 1.2 0.4 0.5 1.4 Final consumption expenditure by general government 1.7 3.4 3.1 1.0 0.7 3.5 3.3 1.7 2.4 1.4 1.0 1.9 -1.9 Gross fixed capital formation (investment) 5.7 3.6 9.2 9.8 9.7 4.8 7.6 -9.2 -5.4 2.4 2.6 -0.4 1.8 General government 12.7 -0.4 32.4 -10.1 4.6 29.6 11.6 8.7 9.8 7.4 5.9 10.3 5.1 Public corporations 0.8 2.9 -11.4 18.2 3.5 2.4 3.1 -0.6 -3.4 2.3 2.5 1.6 -0.6 Private business enterprises 5.8 4.8 11.1 12.7 12.7 0.4 8.1 -15.4 -9.6 1.2 1.7 -3.4 1.6 Change in inventories (R millions) 26,345 36,623 1,259 35,683 5,939 -35,321 1,890 -4,094 -950 2,858 3,770 396 8,812 Residual item (R millions) -5,287 -279 -24,934 -28,120 -36,307 -60,291 -37,413 -55,859 -52,186 -43,417 -55,106 -51,642 -47,278 Gross domestic expenditure 4.9 3.9 -2.5 7.7 -1.9 -5.8 1.4 3.8 0.9 3.2 0.3 0.6 3.4 Exports of goods and services 4.3 0.1 24.0 -5.0 12.2 10.8 4.6 1.6 -15.7 8.6 22.7 2.6 8.3 Imports of goods and services 10.5 6.0 7.2 8.3 0.5 -23.5 1.8 21.7 -14.2 12.5 7.8 -0.5 15.8 Net exports (R millions) -10,968 -60,346 -49,326 -78,059 -54,705 27,182 -38,727 -12,342 -15,431 -23,667 5,056 -11,596 -10,410 Gross domestic product 3.2 2.2 1.4 3.7 1.2 5.1 2.2 -1.6 0.5 2.1 4.1 1.5 1.3 Source: South African Reserve Bank. South Africa Economic Update 7 12 aug 15.indd 15 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Box How have fuel and electricity price changes affected households in South Africa?4 1.1 Lower oil prices and higher electricity tariffs are the two major price changes hitting South African consumers in 2015. In June 2014, the global oil price exceeded $100/barrel but the slowdown in China and increased supply contributed to a 40 percent price drop, which was felt in South Africa from September 2014, although the rand’s depreciation muted the price decline’s impact. Domestic gasoline (petrol) prices fell 28 percent from peak (June 2014) to trough (February 2015)—but in Q2 almost fully reversed. Electricity tariffs rose 12.6 percent in April 2015 (box figure 1). Household consumption patterns play a major role in how these price changes affect the poor.5 The oil-price shock cut prices for gasoline and related products, directly, but also fed through into prices of other goods and services that use fuel or 16 electricity, indirectly. South African headline inflation fell sharply on lower fuel prices, but the prices of food and transportation hardly moved. Fuel outlays account for almost 3.5 percent of average household consumption, electricity around 4 percent (box figure 2). But for extremely poor households (with individuals living on less than $1.25 a day), fuel occupies less of the consumption basket (1.1 percent) and electricity more (5.7 percent). For the non-poor, fuel and electricity each comprise more than 4 percent of the basket. Thus the relatively better-off benefit more from the fall in gasoline prices. But the poor are left compara- tively worse off by rising electivity tariffs.6 The indirect effect of the price changes depends on how households allocate their budget. Poorer households spend a larger share of their incomes on food and beverages (38.3 percent) and housing utilities (26 percent). Richer households spend more on health, transportation, and communication. This makes poorer households more vulnerable to price rises via the indirect channel, too. Box figure 1. Key price changes Box figure 2. Fuel and electricity out- lays by household Electricity price shock 12.6% 7% 140 5.7% 6% 4.8% 120 4.1% 3.8% Index December 2012=100 5% Share in total expenditures 4.2% 4.2% 4% 100 3.0% 3% 80 2% 1.1% 60 1% World oil price shock 0% 40 Rural Urban Poor Non-poor Rural Urban Poor Non-poor 20 Dec-12 Feb-13 Apr-13 Jun-13 Aug-13 Oct-13 Dec-13 Feb-14 Apr-14 Jun-14 Aug-14 Oct-14 Dec-14 Feb-15 Apr-15 Fuel Electricity SA petrol prices World oil prices SA exchange rate SA electricity prices South Africa average Source: Statistics South Africa, World Bank. Source: IES, 2010/11. World Bank calculations. Fuel component includes Motor car fuel, oil and grease, paraffin, petrol for household use, and diesel for households. Investment contracted on an annual basis in subdued in the first quarter of 2015, growing 2014 in the wake of labor unrest, power short- by just 1.8 percent q/q. On the external side, ages, and policy uncertainty. The rebound net exports made a small positive contribution in gross fixed capital formation growth seen to growth in 2014 as imports contracted. But in in the second half of 2014 was insufficient to the first quarter of 2015, import growth again offset the first-half weakness stemming from began to outpace export growth, net exports strikes in mining and manufacturing. Private- subtracted some 2.0 percentage points from sector investment, which accounts for about headline growth, and the outlook for exports two-thirds of all investment, subtracted about will continue to be challenging in Q2 given the half a percentage point from headline real continued decline in global minerals and met- GDP growth in 2014. Investment remained als prices. South Africa Economic Update 7 12 aug 15.indd 16 8/12/15 2:07 PM How have fuel and electricity price changes affected households in South Africa?4 Box (continued) 1.1 Using the changes in fuel and electricity prices through end-April 2015, we find that the net impact was that budgets of poor households were largely unchanged by the two price changes (box figure 3)—in that period lower fuel prices partly offset higher electricity prices. Across all households, the gains from lower fuel prices (1.8 percent) through the end of April were partially offset by the losses tied to higher electricity prices (–0.8 percent). The poor are disproportionally affected by higher electricity prices (absent changes in the free basic amount of electricity provided to poor households), which cost them 1 percent of their consumption and largely offset the temporary gain of 1.1 percent from lower fuel prices.7 Non-poor households gain by 1.1 percent when the net impact of both price changes is considered. Box figure 4 illustrates how the net effects of the two price changes through the end of April impacted different types of 17 households. It shows that the upper deciles (labeled non-poor in box figure 4) of the income distribution gained the most, along with small and medium households, workers in industry and services, the better educated, and those living in urban areas. But if the higher fuel levy is factored into the calculations, poor households are worse off. Box figure 3. Impact of fuel and elec- Box figure 4. Net impact on household tricity price changes on household consumption consumption Electricity supply 2.5% 1.6% had risen by just 1.9% 1.8% 1.4% 1.5% 0.7% 1.2% 4.8 percent from its 1.1% 0.7% Consumption gain / loss 1.0% 0.8% trough in 2009Q1 while 0.5% 1.1% 1.1% 0.8% 0.3% 0.6% real GDP expanded by -0.5% -0.7% -0.6% 0.4% -0.5% -0.3% -0.3% -0.3% twice that amount 0.2% -0.8% -0.8% -1.0% 0.0% -1.5% Small 1-2 Medium 3-4 Large +5 High Medium Low Rural Urban Non-poor Poor Agril Industry Not working Services Non-poor Non-poor -2.5% Poor Total Poor Total Fuel price Electricity price Household Education Location Poverty Sector shock (gain) shock (loss) size Direct Indirect South Africa average Source: World Bank calculations using IES, 2010/11. Supply constraints are looming larger other factors like fractious labor relations, in capacity underutilization maintenance downtime, and unreliability of The importance of growing supply impedi- power supply—have played a growing role. ments as a brake to growth can be gauged Figure 1.9 shows the widening gap between by the drivers of capacity underutilization electricity supply and GDP. Electricity avail- and the relationship between electricity sup- able for distribution has risen by just 4.8 ply and real GDP. Manufacturing capacity percent from its trough in the first quarter underutilization worsened between the first of 2009 (when South Africa last experienced quarters of 2008 and 2015 as total capac- a severe power crisis) and the fourth quarter ity utilization fell to 80.2 percent from 84.4 of 2014, while real GDP expanded by 14 per- percent (figure 1.8). According to Statistics cent, seasonally adjusted. South Africa’s survey of capacity utilization Power constraints, labor unrest, and policy by large manufacturing firms, enterprises uncertainty conspire to undermine business cite “insufficient demand” as the main driver confidence, constraining investment and job of capacity underutilization, reflecting weak creation. The Rand Merchant Bank/Bureau domestic and external demand. of Economic Research (RMB/BER) Busi- Supply shackles, however—capturing ness Confidence Index (BCI) has fluctuated shortages of skilled labor, raw materials, and around the neutral mark of 50, while the South Africa Economic Update 7 12 aug 15.indd 17 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Figure Total manufacturing capacity underutilization has deepened since early 2008 1.8 Underutilisation Underutilisation due to insuf cient demand 0 -5 -10 18 Percent -15 -20 -25 q1 q3 q1 q3 q1 q3 q1 q3 q1 q3 q1 q3 q1 q3 q1 q3 q1 q3 q1 06 06 07 07 08 08 09 09 10 10 11 11 12 12 13 13 14 14 15 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Power constraints, Source: Statistics South Africa. labor unrest, and policy uncertainty Figure The gap between electricity supply and GDP is growing (2006 = 100) undermine business 1.9 130 Electricity available for distribution GDP confidence 125 120 115 110 105 100 95 90 2006m1 2006m5 2006m9 2007m1 2007m5 2007m9 2008m1 2008m5 2008m9 2009m1 2009m5 2009m9 2010m1 2010m5 2010m9 2011m1 2011m5 2011m9 2012m1 2012m5 2012m9 2013m1 2013m5 2013m9 2014m1 2014m5 2014m9 2015m1 2015m5 Source: Statistics South Africa. South African Chamber of Commerce and growth prospects in the Euro Area (a major Industry (SACCI) BCI points to a deteriorat- market for South African manufactures) and ing business environment since 2011. The rand weakness may have improved manufac- SACCI index average for the first 6 months of turing’s prospects. Yet load shedding, the 2015 has been lower than the average for the prospect of future electricity tariff hikes, preceding year. and volatility of the oil price and the rand In recent months, however, the manu- still have a disproportionally larger impact facturing Purchasing Managers’ Index on the energy-intense manufacturing sector. (PMI) has strengthened. In June and July Moreover, the price sub-index, having risen the Barclays manufacturing PMI unexpect- for four consecutive months, suggests rising edly improved to 51.4 seasonally adjusted, input costs in manufacturing. In addition, building on the gains seen in May. The sub- the fact that the overall economy-wide PMI indexes for business activity and employ- remains below the 50 mark suggests that ment continued to strengthen. Improving business conditions overall remain weak. South Africa Economic Update 7 12 aug 15.indd 18 8/12/15 2:07 PM Labor markets quarter was almost fully matched by a corresponding increase in the number of Unemployment remains the key economically inactive and a fall in labor challenge facing South Africa force participation.9 Employment increased South Africa struggles to generate sufficient by some 198,000 relative to Q1 and half jobs to substantially lower its high levels of of that increase reflecting the expansion unemployment. Labor market indicators that of employment in community and social had deteriorated substantially in the first services. quarter of 2015 showed some improvement Nevertheless, the number of jobs gener- in the second quarter. Unemployment ated continues to fall short of what is needed 19 had reached its highest level since 2008 in to employ South Africa’s growing workforce Q1, standing at 26.4 percent by the narrow (see section 2). Compared to a year ago, the measure and 33.9 percent by the broad number of jobs generated of more than half measure, i.e., including discouraged workers a million still falls short of what is needed to (figure 1.10), but improved in Q2 to 25 employ South Africa’s large existing number percent and 32.9 percent, respectively. 8 of unemployed and the growing working-age Quarter on quarter, the decline in the population, which expanded by more than Two-thirds of those number of unemployed in the second 600,000 in the past year. The rate of youth unemployed have been Figure Unemployment trends are set negative unable to secure a job 1.10 for more than a year Changes in narrow unemployment rate Drivers of the change in unemployment rate 10% 26% 25% 5% Unemployment rate (narrow) 24% 0% 23% 22% -5% 21% -10% 20% 2011q2 2012q2 2013q2 2014q2 2015q2 ▲ in unemployment ▲ in employment ▲ in absorption rate ▲ in labor force participation rate Narrow unemployment rate Changes in broad unemployment rate 15% 35% Drivers of the change in unemployment rate Unemployment rate (broad) 10% 33% 5% 31% 0% 29% -5% -10% 27% 2011q2 2012q2 2013q2 2014q2 2015q2 ▲ in unemployment ▲ in employment ▲ in absorption rate ▲ in broad labor force participation rate Broad unemployment rate Source: Statistics South Africa, Quarterly Labour Force Survey (QLFS) 2011–2015 and World Bank calculations. South Africa Economic Update 7 12 aug 15.indd 19 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S unemployment (for ages 15–24) stood at close The current environment of weak growth to 50 percent at the end of Q2, and those and confidence is not favorable for any sharp without jobs for more than a year account for pickup in private-sector job creation. As table almost two-thirds of the unemployed. 1.3 suggests, job creation since the crisis has The rise in the number of employed been driven by government or government- (198,000 q/q) was largely driven by gains related entities, even though the private sector in employment in services and construc- remains the largest employer overall. In the tion. Community and social services (up fourth quarter of 2008, national, provincial, 98,000), construction (up 79,000) and and local government, as well as state-owned 20 trade (up 73,000) posted the largest jobs enterprises (SOEs), accounted for about 14.6 gains. Finance (down 31,000), manufactur- percent of all jobs. By the last quarter of 2014, ing (down 23,000), and agriculture (down this figure had risen to 17.5 percent. The pri- 22,000) shed jobs. While total employment vate sector, in contrast, saw its share decline has risen above the pre-crisis level (by some from roughly 73.6 percent to 72.1 percent 7.4 percent from the second quarter of 2008), over the period. By the fourth quarter of 2014, manufacturing employment remained sub- only 31 percent of the working-age population dued in the second quarter of 2015 and is were employed in a private enterprise. now some 16.3 percent below seven years South Africa’s jobs gap—defined as earlier (figure 1.11). By the end of June 2015, the difference between the current level some 343,000 jobs had been lost in manufac- of employment and the level of employ- turing since 2008Q2. ment required to return to the pre-crisis Box 1.2 suggests that a greater focus on ratio of employed persons to working-age manufacturing exports can generate indi- population and to absorb new entrants— rect employment (i.e., jobs providing inputs is estimated at 944,000 in 2015Q2. Average to these exports) opportunities. But these employment growth slowed to 2.3 percent for jobs are likely to be more capital and skills 2014Q1–2015Q2 from 2.5 percent y/y over intensive, further obscuring the prospects 2011Q1–2013Q4. Given the somewhat slower of securing a job by the least educated or rates of employment creation, we expect to unskilled/semi-skilled segments of the popu- take longer to close this gap (figure 1.12). At lation who, at the end of 2015 Q2, constituted the lower rate of job creation of 2.3 percent, almost 60 percent and 40 percent, respec- we estimate that the jobs deficit would be tively, of the unemployed. roughly 593,000 jobs by 2020. If employment Figure Manufacturing job numbers remain the hardest hit by the crisis 1.11 Employment levels, by industry (% above 2008q2 employment levels) (% above trough levels) 60 50 40 30 Percent 20 10 0 -10 -20 tal ure g ing ies n e rt ce s ce nin ad tio spo an To ilit tur rvi ult Tr ruc Mi Fin Ut an l se fca ric nst Tr Ag nu cia Co Ma so d an ity un mm Co Source: Statistics South Africa, QLFS 2008–2015. South Africa Economic Update 7 12 aug 15.indd 20 8/12/15 2:07 PM Box What do exports mean for jobs and skills demand in South Africa?10 1.2 Exports are a critical source of growth for South Africa; they are also an important channel for productivity enhancement. But what do they mean for jobs? Drawing on a newly developed World Bank dataset11 on the labor content of exports, we explore the possible implications of exports for employment and skills demand in South Africa, paying particular attention to South Africa’s integration into manufacturing global value chains (GVCs) over the past decade. A critical finding is that manufacturing matters for jobs, but its importance comes mainly through its backward links to the domestic economy (box figure 1). The indirect employment impact of manufacturing exports (jobs in industries providing inputs into manufacturing exports) is nearly 4.5 times greater than the direct employment impact. Indeed, while the minerals sector generates the most direct jobs due to its dominance in the export basket and in relative labor intensity, its weak 21 backward links to the domestic economy stifle its overall employment impact. Box figure 1. Number of jobs in exports in 2011 across sectors South Africa, 2011 Machinery, metal, and ferrous metal Minerals nec Energy, chemicals/rubber/plastic Trade and transport services Transport equipment Agriculture, forestry, and sheries Processed food, beverages, and tobacco Other private services Paper/pub, wood products Manufactures nec Apparel, textiles, leather Public administration/defence/health/education Mineral products Electricity, gas, and water nec Construction Dwellings 0 200 400 600 800 1000 Thousands Direct jobs in export Indirect jobs in export Source: World Bank calculations. How have manufacturing exports and GVC integration affected the level and nature of jobs in South Africa? To understand this, we focus on the automotive sector, where South Africa has grown its exports and has become deeply integrated into GVCs over the past decade (box figures 2 and 3). The first point is that while the rapid growth of exports meant that nominal jobs in the sector increased substantially over the decade (box figure 2), the relative labor intensity of South Africa’s exports declined sharply: where automotive exports contributed around $37 of labor per $100 of exports in 2001, this declined to below $30 in 2011 (box figure 3). This level of labor in exports is well below that of Brazil where the labor content has increased; and India, where it has remained broadly stable. South Africa’s decline in labor intensity of its auto exports is explained almost fully by declining labor content in direct manufacturing—indeed, the labor content of direct manufacturing jobs fell by almost half over the decade. Still, significant job growth resulted from the automotive sector’s extensive backward links to the domestic economy. While each direct job in the sector was linked to one indirect job in 2001, by 2013 it was linked to three. Most important, this came through services sector links, which is unsurprising. In fact, it is exactly what we might expect in an environment of GVC integration: exports make increasing use of foreign inputs (reducing direct labor demand) but coordinating these inputs requires more intensive use of (mainly domestic) services, including financial services, transport and logistics, and other business services GVC integration also appears to have substantial implications for skills demand. Box figure 3 shows that for direct jobs in the automotive sector, the contribution of unskilled labor per $100 of exports has declined sharply over the decade, while that of skilled labor has remained steady. This is, again, perhaps to be expected in the context of GVC integration, where the require- ments for productivity, quality, and consistency from global lead firms likely contribute to increased capital intensity and demand for higher skills. Coupling this with the strong shift toward indirect-services employment, which also tends to have a higher skills bias, implications for demand for skilled versus unskilled labor may be important, perhaps accelerating the underly- ing skills bias in South African production.. South Africa Economic Update 7 12 aug 15.indd 21 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Box What do exports mean for jobs and skills demand in South Africa?10 (continued) 1.2 Box figure 2. Automotive jobs in Box figure 3. Automotive skills compo- exports, 2001 and 2011 sition of direct jobs as a share of gross exports, 2001 and 2011 Motor vehicles Motor vehicles 2001 2001 SA SA 2011 2011 2001 2001 BRA CHL CHN COL IND PER RUS THA TUR BRA CHL COL IND PER RUS THA TUR 2011 2011 22 2001 2001 2011 2011 2001 2001 2011 2011 2001 2001 2011 2011 2001 2001 2011 2011 2001 2001 2011 2001 2011 2011 2001 2001 2011 2011 2001 2001 2011 2011 0 1000 2000 3000 4000 5000 6000 7000 8000 0 10 20 30 40 50 60 70 US$ millions Labor’s US$ value added per $100 of gross exports Direct VA Indirect VA Total VA unskilled Total VA skilled Source: World Bank calculations. Government and related entities have picked up some of the slack in creating Table jobs 1.3 Percent, unless otherwise noted 2008q4 2009q4 2010q4 2011q4 2012q4 2013q4 2014q4 (as a percentage of total number of employed persons) National/provincial/local government 12.9 13.8 14.1 14.6 15.1 15.3 15.4 Government-controlled business (such as Eskom) 1.7 1.9 2.1 1.9 2.2 2.1 2.1 A private enterprise 73.6 73.9 73.4 72.9 72.6 71.6 72.1 Non-profit organization 0.9 0.9 0.9 1.0 1.1 1.3 1.3 A private household 10.8 9.4 9.4 9.4 8.9 9.2 8.9 Do not know 0.1 0.1 0.1 0.2 0.1 0.4 0.2 2008q4 2009q4 2010q4 2011q4 2012q4 2013q4 2014q4 (as a percentage of total working-age population) National/provincial/local government 5.8 5.7 5.7 6.0 6.2 6.7 6.6 Government-controlled business (such as Eskom) 0.8 0.8 0.9 0.8 0.9 0.9 0.9 A private enterprise 33.0 30.7 30.0 30.1 29.7 31.0 31.0 Non-profit organization 0.4 0.4 0.4 0.4 0.4 0.6 0.6 A private household 4.8 3.9 3.8 3.9 3.7 4.0 3.8 Do not know 0.0 0.0 0.0 0.1 0.0 0.2 0.1 Note: 2008–12 QLFS survey weights correspond to the 2001 Census. 2013 and 2014 weights correspond to the 2011 Census. growth could be boosted to the 2011–13 rates working-age population were employed in of 2.5 percent, the jobs deficit would fall to 2008—stood well below the average of mid- 419,000 but would still remain high. dle-income countries. Labor-force participa- South Africa, therefore, needs higher, tion is also low compared to that of peers. more inclusive jobs intensive growth to Unemployment is largely structural, and return to pre-crisis employment, yet even at about two-thirds of the unemployed have not that time the initial absorption or employ- found a job in more than a year and are con- ment rate—only 4.6 persons out of 10 in the centrated among the less educated.13 South Africa Economic Update 7 12 aug 15.indd 22 8/12/15 2:07 PM Figure Higher employment growth rates are needed to start bridging the jobs gap 1.12 1,500 1,000 Number in thousands 500 0 23 2008q4 2009q1 2009q2 2009q3 2009q4 2010q1 2010q2 2010q3 2010q4 2011q1 2011q2 2011q3 2011q4 2012q1 2012q2 2012q3 2012q4 2013q1 2013q2 2013q3 2013q4 2014q1 2014q2 2014q3 2014q4 2015q1 2015q2 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 -500 Jobs gap Estimated jobs gap: employment growth of 2.5% -1,000 Estimated jobs gap: employment growth of 2.3% Source: Statistics South Africa, QLFS 2008–15 and World Bank calculations. For the first time in more than a decade, Fiscal policy Achieving this fiscal consolidation in an environment of low economic growth will personal income-tax Despite weaker growth, the budget keeps be challenging. That the debt burden has to its consolidation commitment risen far higher than forecast five years ago rates were raised for The 2015 budget targets a gradual reduc- (figure 1.13) only underscores the lack of fis- all but the bottom tion in the budget deficit, despite a more cal space and role of low growth in forcing downbeat growth outlook. The general it up. The growth outlook underpinning the tax bracket government budget deficit is projected to 2015/16 budget was revised downward to just gradually narrow from 3.7 percent14 of GDP 2 percent in 2015 and 2.4 percent in 2016, in FY2014/15 to 2.5 percent in FY2017/18, about half of a percentage point lower in as expenditures are contained and revenue each year than at the time of the Medium- collection improved over the course of the Term Budget Policy Statement in October three-year medium-term expenditure frame- 2014—and there is still a risk that growth work (MTEF). This would (if achieved) gen- will come in below even these downturned erate a small primary surplus in FY2016/17 projections. and FY2017/18, stabilizing the gross debt-to- In an attempt to rein in public debt, the GDP ratio at 47.6 percent of GDP. 2015/16 budget introduced a mix of revenue Figure How the debt-to-GDP ratio has evolved with each vintage of the budget 1.13 50 45 Gross debt to GDP (%) 2015 Budget 40 2014 Budget 2013 Budget 2012 Budget 35 2011 Budget 2010 Budget 30 2009 Budget 25 9 0 1 2 3 4 5 6 7 8 /0 /1 /1 /1 /1 /1 /1 /1 /1 /1 08 09 10 11 12 13 14 15 16 17 FY FY FY FY FY FY FY FY FY FY Source: Republic of South Africa National Treasury, Budget Reviews 2009–15. South Africa Economic Update 7 12 aug 15.indd 23 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S and expenditure measures. Revenue mea- Monetary policy and inflation sures included an increase of 1 percentage point in the marginal personal income tax Inflation pressures are poised to rate for all brackets except the lowest—the rise, triggering the resumption first increase in those rates in more than of interest rate increases a decade—and increases in the fuel levy, Weak underlying demand pressures helped excise duties on alcohol and tobacco, and ease inflationary pressures and brought head- transfer duties on houses valued at more line inflation close to the mid-point of the than R2.2 million. They were accompanied inflation-targeting band (figure 1.14). Head- 24 by measures to stimulate economic activity, line CPI inflation bottomed at 3.9 percent such as tax relief to small businesses, while y/y in February 2015—its lowest rate in four social payments were raised broadly in line years—before picking up gradually to reach with inflation. 4.7 percent y/y in June, on the back of higher On the expenditure side, the latest budget petrol prices (which includes the impact the cut the expenditure ceiling for non-essential higher fuel levy implemented in April). Core recurrent outlays over the next two years and inflation, at 5.5 percent in June, remained The pause in the scaled back public-capital investment. It cut close to the upper bound of the inflation tar- spending allocations by 1 percent of pro- get. In percentage points, the key drivers of interest rate jected outlays for the current and next fiscal monthly inflation were housing and utilities tightening cycle year relative to original plans. The budget (1.3), miscellaneous goods and services (1.1), assumed a 7.7 percent increase in the nomi- food and non-alcoholic beverages (0.7), and has ended nal wage bill in 2015/16, or 6.6 percent over alcoholic beverages and tobacco (0.5), which the three years of the framework, to help together contributed almost four-fifths of the bring the wage bill down from 11.5 percent overall increase in headline prices in June. of GDP in FY2014/15 to 11.0 percent in the Risks to the inflation outlook are to the outer year. upside. CPI inflation will likely rise further In contrast, the new three-year pay deal over the second half of 2015, reflecting the subsequently agreed upon provides for a depreciation of the rand and pressures on wage increase of 7 percent in 2015/16, and food prices. Oil prices remain volatile. Crude for wage increases of the consumer price oil prices averaged $60.5/bbl in the second index (CPI) plus 1 percent for each of the quarter, up 17 percent from the first quar- next two years. The pay deal also lifted the ter, but have fallen sharply again in July. The housing allowance from R900 to R1,200 potential of a weaker rand in the context of (subject to conditions) and the medical-aid the U.S. Federal Reserve’s expected path of subsidy from 17.8 percent to 28.5 percent. monetary-policy normalization, wage settle- All these increases will likely test the upper ments above inflation and labor-productivity bounds of the budgeted wage envelope and, gains, plus further electricity-tariff increases if growth also falls short of rates assumed and higher food prices due to the domestic in the MTEF, further adjustment measures drought, will all likely add to upward momen- will likely be required to secure the targeted tum of headline CPI over the remainder of reductions in the fiscal deficit and debt the year and into the first half of 2016, when burden. headline inflation is expected to breach the Financial stress in key SOEs also poses upper band. upside risk to the government’s deficit and Against this backdrop, the Reserve Bank debt targets. Including sovereign guarantees resumed on its path of gradual monetary- to SOEs, public-sector debt stood at 58 per- policy normalization, increasing interest cent of GDP at the end of FY2014/15. The rates in July for the first time in a year. The latest budget targets this broader measure to last hikes in the policy rate were in January fall marginally to 57.3 percent by the end of 2014 (50 basis points) and July 2014 (25 basis FY2017/18, if these SOEs fund their opera- points). The July Monetary Policy Committee tions and investment needs within existing increased its reference-policy rate by 25 bps to guarantee limits. 6.0 percent, citing the persistence of forecast inflation at elevated levels and upside risks to it from second-round effects from a more South Africa Economic Update 7 12 aug 15.indd 24 8/12/15 2:07 PM Figure Core inflation is nudging the upper bound of the inflation target 1.14 Headline in ation and core in ation 7 6 5 4 Percent 25 3 2 1 0 m1 m3 m5 m7 m9 1 m1 m3 m5 m7 m9 1 m1 m3 m5 m7 m9 1 m1 m3 m5 m1 m1 m1 12 12 12 12 12 13 13 13 13 13 14 14 14 14 14 15 15 15 12 13 14 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 CPI headline Core in ation Trimmed mean index Lower-bound in ation target Upper-bound in ation target Source: Statistics South Africa. depreciated rand. It did, however, signal that of falling oil prices and weak domestic any future moves will be highly data depen- demand (figure 1.15). The deficit had dent given the fragile state of the economy. reached a post-crisis peak in 2013 of 6.0 percent of GDP but narrowed to 5.4 percent External sector in 2014. This trend continued in 2015Q1, as the current account deficit improved Adjustment in the current account to 4.8 percent despite the widening trade deficit advances slowly deficit, which doubled to 1.8 percent in The current account deficit narrowed a the first quarter of 2015 as gross domestic little in 2014, as the trade deficit shrank expenditures picked up and export receipts in the last quarter of the year on the back contracted somewhat. The deficit on the External imbalances haves improved a little recently, but South Africa remains Figure vulnerable 1.15 Current account balance 2 110 Current account balance, trade balance (%) 1 100 Terms of trade (Index, 2010 =100) 90 0 80 -2 70 -4 60 -6 50 -8 40 2008/01 2008/02 2008/03 2008/04 2009/01 2009/02 2009/03 2009/04 2010/01 2010/02 2010/03 2010/04 2011/01 2011/02 2011/03 2011/04 2012/01 2012/02 2012/03 2012/04 2013/01 2013/02 2013/03 2013/04 2014/01 2014/02 2014/03 2014/04 2015/01 Current account balance Trade balance Terms of trade (excluding gold) Source: South African Reserve Bank. South Africa Economic Update 7 12 aug 15.indd 25 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S services, income, and current transfers Economic outlook account improved over 2015Q1, more than offsetting the deterioration in the trade Growth will stay hamstrung balance. by domestic constraints South Africa’s terms of trade have wors- Real GDP growth is forecast at 2.0 percent ened since 2011, declining by about 7.9 per- in 2015, and to slowly strengthen to 2.4 per- cent (excluding gold; including gold, 8.4 cent in 2017 (table 1.4). Agricultural growth percent). Falling international oil prices is expected to be pulled down in the near toward the end of 2014 helped with the term by the poor maize harvest, but finance 26 import bill, but lower mineral-export prices and business services are set to grow robustly cut into export receipts. But in the second throughout the forecast period. But overall, half of 2014 import prices fell faster than growth in South Africa will remain largely export prices. This helped bring the trade below the average growth rate of 4.2 percent deficit to a low of 0.9 percent of GDP in and 4.0 percent for Sub-Saharan Africa, in the final quarter of 2014. The first quarter 2015 and 2016–2017, respectively. of 2015 saw a fall in merchandise-exports Structural impediments will continue Real GDP growth is receipts (despite increases in the volume weighing heavily on growth. Power-supply of merchandise exports). The value of bottlenecks are one of the main underlying forecast at 2.0 percent merchandise imports, on the other hand, factors in the downgraded revision of the in 2015, and to slowly increased as higher import volumes coun- growth outlook since the previous economic tered the subdued international oil prices. update. And shortages are muting the strengthen to 2.4 South Africa’s terms-of-trade (excluding rebound in gross fixed-capital formation percent in 2017 gold) deteriorated by 2.4 percent between that might have been expected with the end the final quarter of 2014 and the first quar- of the platinum- and manufacturing-sector ter of 2015. strikes in 2014. Relative to our forecast in the Despite the recent improvement, the November 2014 Economic Update (Volume current account deficit is still wide, leaving 6), real GDP growth has been revised down South Africa exposed to risk aversion and by 0.5 and 0.8 percentage points for 2015 global financial market volatility. During and 2016, respectively. We expect some 2014, the current account deficit was largely improvement in the electricity situation only funded by volatile capital flows as net foreign toward the end of the forecast period as new direct investment inflows remain very low. power-generation capacity comes online. Unidentified capital inflows (i.e., errors and These structural factors are reinforcing omissions) continue to finance almost 80 cyclical weakness in domestic demand. percent of the overall current deficit, with Household consumption is expected to grow identified portfolio and other capital inflows only modestly. Lower oil prices provided funding the rest. only temporary relief to household budgets Table Economic outlook through 2017 1.4 Percent, unless otherwise noted 2012 2013 2014 2015 2016 2017 Real GDP 2.2 2.2 1.5 2.0 2.0 2.4 Final consumption by households 3.4 2.9 1.4 1.8 1.9 2.5 Government consumption 3.4 3.3 1.9 0.8 0.7 0.7 Gross fixed capital formation 3.6 7.6 -0.4 0.4 1.5 2.6 Gross domestic expenditure 3.9 1.4 0.6 1.3 1.6 2.2 Exports of goods and non-factor services 0.1 4.6 2.6 3.2 4.3 5.1 Imports of goods and non-factor services 6.0 1.8 -0.5 1.0 3.0 4.5 Current account balance (% of GDP) -5.0 -5.8 -5.4 -4.9 -5.2 -5.2 Headline CPI inflation 5.7 5.8 6.1 5.1 5.9 5.5 Poverty rate ($1.25 a day, PPP terms)a 11.4 11.3 11.5 11.3 11.3 .. a Projection using sectoral GDP prediction, micro-simulation method used with pass-through = 1.00 based on GDP constant. Source: World Bank calculations, National Treasury, and South African Reserve Bank. South Africa Economic Update 7 12 aug 15.indd 26 8/12/15 2:07 PM and headline inf lation. The prospect of Risks for South Africa are higher electricity tariffs and potentially not easily discounted higher pass-through of a more depreciated Downside risks to this already-weak rand to domestic prices, will limit space for economic outlook prevail. On the external households to expand consumption. High side, they include a sharper than expected unemployment and indebtedness, along slowdown in the Chinese economy, bouts of with tightening credit standards, continue risk aversion and financial-market volatility to weigh on consumer sentiment, while due to the normalization of U.S. monetary government consumption is subdued because policy and faltering growth in the Euro Area of consolidation efforts. amid uncertainty surrounding developments 27 Concerns over electricity supply and rising in Greece. China is still a major market for input and wage costs are being compounded South African mineral exports, and Europe by broader commodity price weakness as well for manufactured exports, so a slowdown as policy and regulatory uncertainty, and are in either would reduce demand for South likely to dampen the outlook for investment. Africa’s exports. The growing size of some Labor relations are expected to remain regional markets may partly counter any hit difficult in an environment of weak growth. from these markets. But South Africa’s large With growth slow and Moreover, incidents of rising social tension current account deficit, financed heavily by unemployment high, that have received widespread global media volatile capital flows, makes the economy coverage may also keep prospective investors vulnerable to shifts in investor sentiment measures of extreme hesitant. and shifting capital flows that could arise On the plus side, the recovery in advanced in the context of fall-out from the “lift-off” poverty and inequality countries and still-robust growth in Sub- in interest rates in the United States. Major are expected to remain Saharan Africa should boost demand for rating agencies have downgraded South South Africa’s non-mineral exports. Still, the African sovereign ratings over the past few stuck at current levels anticipated weakness in mineral demand and years, and while they have all maintained prices limit the overall improvement in the their ratings in recent reviews, they continue current account deficit, which is expected to cite weak economic performance, to stay elevated at around 5.0–5.2 percent of deteriorating fiscal and external imbalances, GDP over the medium term. Given the weak and unaddressed structural vulnerabilities as recovery, South Africa’s output gap, put at 1.2 key risk factors. percent of potential growth in 2014, is pro- On the domestic front, if labor relations jected to narrow only slowly toward the end do not improve or if power disruptions of the forecast horizon. worsen, growth could well disappoint Given the weak recovery, little progress further. And if wage settlements continue is expected against the triple challenges of to exceed inflation and productivity gains, high unemployment, deep poverty, and wide competitiveness will erode, undermining inequality. Agricultural growth is relatively the role of net exports in supporting the weak, while the extractives and metals recovery. sectors continue to struggle against lower While addressing power shortages will minerals prices and rising production costs, be critical to removing a key break on causing further job losses in these sectors. near-term growth, achieving the 5 percent With little change in growth drivers or their growth target of the National Development job intensity expected, unemployment is set Plan requires much more than that. South to remain sticky and high. Extreme poverty Africa needs urgently to boost growth to may well remain broadly unchanged at this level if it is to provide jobs for young its current level, reflecting the low rate of workers, address its growing social tensions, economic growth. Our forecasts do not and reduce its substantial poverty and foresee inequality narrowing either, because inequality. Improved labor relations for a high rate of joblessness means that the a start, matched by greater collaboration gap in incomes between the employed between the public and private sectors and and unemployed persists. Growth in the policy certainty to improve the business consumption of the bottom 40 percent is environment, are fundamental to restoring seen flat at about 1 percent. investor confidence. South Africa Economic Update 7 12 aug 15.indd 27 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S This faster, more inclusive growth also quality of public services, and remove red calls for initiatives to improve education out- tape for small and medium-sized enterprises comes and the skills of the poor, improve the to help foster job creation (box 1.3). Comparing regulations at the local level: subnational Doing Business in South Box Africa 201515 1.3 The World Bank’s Doing Business project studies the role of government and government policies in the day-to-day life of small and medium-sized enterprises. Its fundamental premise is that economic activity benefits from good institutions and transparent regulation designed to be efficient, accessible to all, and simple. 28 Doing Business reports capture key dimensions of the regulatory environment as it applies to these enterprises and are tools that governments can use in designing sound policies for creating firms and jobs. The global annual Doing Business series benchmarks 189 economies as represented by their most populous city—Johannesburg is the marker for South Africa. Econo- mies with more than 100 million inhabitants are represented by their two largest cities. Subnational Doing Business studies expand the Doing Business analysis further. Doing Business in South Africa 2015 benchmarks eight metropolitan areas, the city of Msunduzi, and four maritime con- tainer ports. Metrics are starting a business, dealing with construction permits, getting electricity, registering property, enforcing contracts, and trading across borders. The report finds that entrepreneurs face different regulatory burdens depending on loca- Real GDP growth is tion. No city performs equally well on all ranked indicators—Ekurhuleni, Johannesburg, and Tshwane lead on starting a busi- ness, Cape Town on dealing with construction permits, Mangaung on getting electricity and enforcing contracts, and Johannesburg forecast at 2.0 percent on registering property (box table 1). in 2015, and to slowly The uneven performance across the indicators shows that there is room for cities to learn from each other and replicate good practices without major legislative changes. Local reforms will not only improve the ranking of one location as compared strengthen to 2.4 to another; they can make a significant difference on the global scale as illustrated by the distance to frontier (DTF) score. For example, if a South African city adopted the good practices found across the nine cities in dealing with construction percent in 2017 permits, getting electricity, and enforcing contracts, it would surpass the average performance of OECD high-income economies in all three areas (box figure 1). To give some examples, by reducing the number of requirements for an electricity connection to four (as in Cape Town and Mangaung) and providing a connection within 66 days at a cost of 257.2 percent of income per capita (as in Buffalo City), South Africa’s global DTF score on getting electricity would climb 31 points to be close to Japan and Malaysia, and ahead of Slovenia or Finland. Box table 1. Doing Business in South Africa 2015: Where is it easier? Municipality Starting a business* Dealing with Getting electricity Registering property Enforcing contracts Municipal seat construction permits Ranking DTF score Ranking DTF score Ranking DTF score Ranking DTF score Ranking DTF score (1-9) (100 (1-9) (100 (1-9) (100 (1-9) (100 (1-9) (100 = best = best = best = best = best result) result) result) result) result) Buffalo City 4 78.67 3 77.50 4 75.32 4 62.84 9 62.54 East London Cape Town 4 78.67 1 78.08 2 81.81 8 59.23 6 67.53 Cape Town Ekurhuleni 1 81.18 4 76.84 5 71.83 3 64.23 4 68.26 Germiston eThekwini 4 78.67 5 76.15 3 75.73 6 62.05 3 69.27 Durban Johannesburg 1 81.18 8 68.52 8 55.74 1 65.82 8 66.14 Johannesburg Mangaung 4 78.67 9 68.22 1 83.88 9 58.41 1 71.04 Bloemfontein Msunduzi 4 78.67 6 74.07 7 63.00 7 59.49 2 70.81 Pietermaritzburg Nelson Mandela Bay 4 78.67 2 78.05 9 53.14 5 62.69 7 66.89 Port Elizabeth Tshwane 1 81.18 7 69.88 6 68.51 2 64.71 5 68.17 Pretoria *On starting a business, each city is ranked either 1 or 4 because the only difference between cities is their proximity to the office of the Compensation Fund (Department of Labor) in Pretoria. Entrepreneurs from Ekurhuleni, Johannesburg and Tshwane conduct the registration of their employees at the Pretoria office and these cities rank equally at the top. In the other cities the process takes longer as the applications are first submitted to the local labor centers, then forwarded to the provincial offices and then to Pretoria. These cities are equally ranked at number 4, just behind the 3 cities ranked at the top. Note: The distance to frontier (DTF) score shows how far on average an economy is at any given point in time from the best performance achieved by any economy on each Doing Business indicator since 2005. The measure is normalized to range between 0 and 100, with 100 representing the frontier of best practices (the higher the score, the better). For details, see the About Doing Business and Doing Business in South Africa 2015 section. Source: Doing Business database. South Africa Economic Update 7 12 aug 15.indd 28 8/12/15 2:07 PM Comparing regulations at the local level: subnational Doing Business in South Box Africa 201515 (continued) 1.3 Reducing the requirements and time to deal with construction permits to 15 procedures (as in Nelson Mandela Bay) and 83 days (as in Cape Town) and lowering the cost to 0.68 percent of the warehouse value (as in Mangaung) would move South Africa 11 points higher. Enforcing contracts—an indicator where most South African cities fare quite well—and handling court cases as efficiently and as inexpensively as in Msunduzi and Mangaung, respectively, would improve the country’s DTF score by 5 points, placing it among the best 25 economies globally on this indicator. Box figure 1. Doing Business in South Africa 2015: Where is it easier? 29 Distance to the frontier score (DTF) South African best practices combined (potential DTF score) 80.05 76.03 86.43 81.83 South Africa (Johannesburg) 68.52 62.00 Doing Business 2015 DTF score or below 60.91 or below 71.15 69.82 OECD high income 55.74 economies average 66.14 Bottom 25% of 189 economies 46.35 measured by or below Doing Business Dealing with construction permits Getting electricity Enforcing contracts Notes 7. The estimates exclude changes to the 1. All data are seasonally adjusted annual fuel levy, which rose by 80.5 cents/liter rates unless otherwise noted. on April 1, 2015. If this higher rate is 2. In February 2011–March 2011 prices. included, the average household loses an 3. Seasonally adjusted annualized rate. All additional 0.8 percent of consumption. data are seasonally adjusted annual rates 8. Estimates for the first and second quar- unless otherwise noted. ter of 2015 correspond to the 2013 Mas- 4. Prepared by Victor Sulla (GPVDR). ter Sample using the newly collected 5. The World Bank’s SUBSIM model at 2011 Census data. Prior estimates for http://www.subsim.org/ is a tool to 2008 Q1–2014 Q4 correspond to the assess the impact of price shocks on 2007 Master Sample using the 2011 Cen- household welfare, poverty, and inequal- sus data. Thus quarterly and annual ity, and the government budget, using comparisons across estimates using dif- household budget survey data and input- ferent master samples should be inter- output matrixes. preted with care. 6. In reality, the impact of rising electric- 9. T he cha nge in na r row unem - ity tariffs on the poor is moderated ploy ment can be expressed as: to the extent that many metros apply ∆Unemployment Rate=(∆Unemployed- cross-subsidization of average rates and ∆Employed)+(∆Absorption Rate-∆Labor amend the calculation of the free basic Force Participation Rate). In the case of electricity allowance. Information was broad unemployment, both the unem- not available on these adjustments, and ployed and the labor-force participation so the calculations should be interpreted rate include discouraged workers. as a maximum impact on the poor in the 10. Prepared by Claire Honore Hollweg absence of these mitigating policies. (GTCDR) and Thomas Farole (GCJDR), South Africa Economic Update 7 12 aug 15.indd 29 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S drawing from a forthcoming work that country’s exports embodied in foreign examines the potential to plug South exports, more than tripled. Africa and its neighbors into Regional 13. In the second quarter of 2015, those with and Global Value Chains. education attainment less than matricu- 11. Calì et al. (2015) use input-output data lation from secondary-level school rep- from GTAP and UNIDO to measure resented about 59.6 percent of those the contribution of labor to the value- unemployed, those with matric repre- added contained in a given country’s sented 32.1 percent, and those with ter- domestic production and exports, as tiary level education represented about 30 well as the number of jobs generated 7.7 percent. by exports, by sector, country, and year; 14. The total outturn for national govern- and by skilled and unskilled workers. ment revenue was R8.7 billion higher The data cover 24 sectors (6 services, in F Y2014/15, helped by better than 3 primar y, and 15 manufacturing) expected collections under taxes on per- in about 100 countries intermittently sonal income (R3.0 billion), corporate between 1995 and 2011. income (R1.8 billion), payroll (R832 mil- 12. Between 2001 and 2011, South Africa’s lion), goods and services (R798 million), participation in the automotive GVC, and international trade and transactions as measured by foreign content embod- (R712 million). ied in the country’s exports and the 15. Prepared by Trimor Mici (DECSN). South Africa Economic Update 7 12 aug 15.indd 30 8/12/15 2:07 PM SECTION 2 Jobs and South Africa’s Changing Demographics Introduction Latin America saw its share of the working- South Africa is undergoing a profound age population increase by more than 20 demographic shift in which the share of percent, the same as in East Asia. However, its working-age population has expanded East Asia experienced a sevenfold increase substantially and will continue to grow for in per capita income, while Latin America another five decades. Shifts of this kind had only a twofold gain, in part reflecting can present a tremendous opportunity to different enabling policies. 5 Indeed, “the accelerate economic growth, raise living dividend period is a window of opportunity standards, and reduce poverty. As the rather than a guarantee of improved growth in the working-age population standards of living.”6 exceeds the growth rate of the overall In South Africa, a lack of jobs has stymied population, the supply of labor rises; and if its ability to capitalize on its demographic the higher number of working-age people opportunity. Since 1994, the working- can be employed productively, higher levels age population has expanded by more of income and faster economic growth can than 11 million. Job opportunities created result—an effect called the first demographic were concentrated in services, but not in dividend.1 However, the combination of agriculture, manufacturing, and mining, higher incomes and fewer non-working which shed workers, with the result that total dependents in the overall population can also jobs created fell far short of the growing create space for higher savings, which in turn labor supply. Just under one-third of the can fund investment in infrastructure and new entrants to the working-age cohort human capital, and give a further and more since 2000 found a job, and by mid-2015 sustained boost to per capita income and only a little more than 40 percent of the economic growth—an effect known as the working-age population were employed. second demographic dividend.2 Unemployment today is higher that it was A growing working-age population has at the end of apartheid.7 To the extent that helped spur higher growth and higher unemployment lowers current and future living standards in many countries, but not earnings potential through the erosion of all. China’s demographic opportunity is skills and human capital during periods of estimated to have contributed one-quarter prolonged joblessness, lifetime earnings of the country’s cumulative real per capita of workers are lower and the ability to gross domestic product (GDP) growth generate savings is constrained.8 Savings as between 1982 and 2000, 3 and two-fifths of a share of GDP have fallen as the working- the increase in East Asia’s income between age population expanded. Real per capita 1982 and 1990.4 Yet a growing working-age income in 2014 was only 40 percent higher population did not have as positive an impact than in 1994, trailing the increases not only in Latin America. Between 1965 and 2010, in East Asia but also in Latin America. 31 South Africa Economic Update 7 12 aug 15.indd 31 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Under what conditions can South Africa population are to secure jobs in a world capitalize on its demographic transition? where skills are in high demand. But The answer depends on a wide range of these are areas where South Africa faces economic variables and how demographic challenges. changes affect these, and vice versa. Job Our analysis focuses on how job cre- creation is just the first stepping stone. ation, savings, and better human capital Better labor productivity, educational can improve enabling conditions for the attainment, and skills are also critical growing working-age population to con- enablers if new entrants to the working-age tribute to faster economic growth (figure 32 How demographics can affect the economy and spur higher growth and living Figure standards 2.1 Evolving age structure Channels of impact Intermediate outcomes Goals Human capital % of South Africa's population 100 65+ Demand for health services and productivity Eradicate 90 Pensions (public and private) With 80 enabling extreme Savings and investment poverty Only a little more 70 60 15-64 years conditions Labor market and jobs and 50 policies Reduce than 40 percent 40 30 inequality 20 Demand for education, <15 years Economic growth of the working- 10 health, and social assistance 0 t=1950 t=2050 age population in Growth, health, and education affect demographic trends over time South Africa work Source: Authors’ own depiction for South Africa based on the analytical framework of the forthcoming World Bank and IMF Global Monitoring Report 2015/16. Box Demographic change and the long-term sustainability of social spending 2.1 Using United Nations Population Projections, the National Treasury of South Africa has prepared long-term estimates of social spending that take account of trends in future demographics and utilization of public services.11 Its analysis assesses whether current social policies are consistent with a stable or falling debt burden. Where growth remains around 3 percent a year, government spending on social services is sustainable because demographic change results in a declining school-age population. If policies remain unchanged and real GDP growth exceeds 3 percent a year, the decline in the pupil–teacher ratio results (box figure 1) in a reduction in basic education spending from about 5 percent of GDP in 2015 to just under 4.5 percent by 2040 (box figure 2). Spending on social grants also falls, because the rapidly declining share of young dependents in the population reduces expenditures on the child-support grant by more than the increase in expenditures on the old age grant linked due to the growing number of old people in the population. But the projections reveal pressures on health outlays due to a high growth in the use of health care services. Box figure 1. Trends in pupil–teacher Box figure 2. Long-term trends in ratio social spending 30 Basic education Post-school education Social protection Health 28 6 Learners per educator 26 5 Percentage of GDP 24 4 3 22 2 20 1 18 0 2010 2015 2020 2025 2030 2035 2040 2010 2015 2020 2025 2030 2035 2040 Source: National Treasury (2014). South Africa Economic Update 7 12 aug 15.indd 32 8/12/15 2:07 PM Box Demographic change and the long-term sustainability of social spending (continued) 2.1 Demographic trends point to favorable fiscal dynamics over the long run, but only if economic growth improves and there are no new spending pressures or initiatives. If growth remains around 2 percent a year, the analysis found that social spending would increase as a share of GDP, causing the debt burden to rise, if there are no changes in benefits or taxation. Pressure to address youth unemployment would require expanded vocational training and public works programs, demanding greater fiscal outlays. New social policies, such as National Health Insurance, would add further pressure (box figure 3). Accommodating these initiatives while safeguarding sustainability would require some combination of higher growth, taxes, and reallocation of resources from other programs. 33 Box figure 3. Projected spending with new policies Basic education Post-school education Social protection Health 8 7 6 Percentage of GDP 5 4 3 2 1 0 2010 2015 2020 2025 2030 2035 2040 Source: National Treasury (2014). 2.1). As the “youth bulge” passes into work- and training, so that inexperienced new ing age, a country enters its “demographic entrants and the unemployed can become window of opportunity,” 9 but as seen a divi- more mobile across occupations and attrac- dend does not materialize automatically. tive to hire. And as the working population To realize the first demographic dividend, ages, public demand for health and pension growth must be jobs intensive and the edu- systems rise. The implications of changing cation system has to prepare graduates with demographics for public finances in South the skills demanded by the labor market. Africa may be demanding (box 2.1). While Otherwise, a country risks a worsening eco- it is necessary to put in place pension and nomic situation with rising unemployment health policies to prepare for a rising elderly and increased dependency.10 The second population, a first priority is to ensure that dividend accrues when consumption rises workers are gainfully employed during their more slowly than incomes and requires working lives. This will help boost incomes, not only mechanisms to encourage saving savings, and growth, better preparing South but also productive investment in human Africa for when the old will be a far larger capital (knowledge and skills) so labor pro- share of the population. ductivity can improve and boost workers’ What can South Africa do to put more of earnings. its working-age population to work and reap High-quality public services are also greater benefits from its historically high enablers. As the structure of the popula- and growing working-age population? We tion changes, demand rises for high-quality tackle this question by focusing on the job- schooling to ensure that young entrants related issues. Section 1 below reviews how are equipped for the workplace. However, far South Africa has progressed in its demo- it may also be necessary to provide ser- graphic transition. Section 2 presents how vices that help support both job searching the country’s labor market has absorbed South Africa Economic Update 7 12 aug 15.indd 33 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S the expanding working-age population environment that generates employment, and how the pattern of economic growth higher savings, and improved productivity inf luenced this performance. Section 3 and educational attainment. The conclu- sheds light on how South Africa’s future sions explore what policies could help the growth, income, poverty, and inequality country to better capitalize on its demo- could evolve under a different enabling graphic opportunity. Figure South Africa’s changing demographic and age profile 34 2.2 South Africa 1950 100+ Elderly DR: 6.2% Prop: 3.6% 90-94 80-84 70-74 60-64 50-54 40-44 Prop 15-64: 57.8% As the youth bulge 30-34 20-24 passes into working 10-14 0-4 Youth DR: 66.7% Prop: 38.6% age, a country enters 2.5 2.0 1.5 1.0 0.5 0.0 0.5 1.0 1.5 2.0 2.5 its demographic Millions Females Males window of opportunity South Africa 2000 100+ Elderly DR: 5.4% Prop: 3.4% 90-94 80-84 70-74 60-64 50-54 40-44 Prop 15-64: 63.6% 30-34 20-24 10-14 0-4 Youth DR: 51.9% Prop: 33% 2.5 2.0 1.5 1.0 0.5 0.0 0.5 1.0 1.5 2.0 2.5 Millions Females Males South Africa 2015 100+ Elderly DR: 8.8% Prop: 5.7% 90-94 80-84 70-74 60-64 50-54 40-44 Prop 15-64: 65% 30-34 20-24 10-14 0-4 Youth DR: 45.1% Prop: 29.3% 2.5 2.0 1.5 1.0 0.5 0.0 0.5 1.0 1.5 2.0 2.5 Millions Females Males DR = dependency ratio, Prop = proportion Source: United Nations (2013). South Africa Economic Update 7 12 aug 15.indd 34 8/12/15 2:07 PM Figure South Africa’s demographic window is open 2.3 0.5 0.45 0.4 0.35 0.3 Percent 0.25 0.2 0.15 35 0.1 0.05 0 1950 1970 1990 2010 2030 2050 2069 Demographic window (2009-2069) Population under age 15 Population over age 65 Source: United Nations (2013). In 2015, the working- age share in the How far has South Africa progressed to Statistics South Africa’s most recent in its demographic transition? population estimate, the total dependency population is about South Africa’s population has increased four- ratio was 53.9 percent in 2015, a decline of fold in the past 65 years to 54.9 million as of more than one-third from its peak. There 65 percent mid-2015.12 Until the end of the 1980s, mor- are now about 1.8 working-age persons per tality and fertility rates declined at roughly young and old dependent. However, because equal rates and the population expanded employment is relatively low, there are only by about 2.4 percent a year. Thereafter, an 0.8 workers on this measure, or 0.4 workers if expansion in the use of contraception and all young, old, and non-working dependents family planning contributed to the gradual are considered.16 decline in fertility rates from an average of The working-age population, 15–64 years 6 children per woman in the 1960s to 4.3 by old, expanded 4.5 times or by more than the late 1980s. Population growth slowed to 27 million between 1950 and 2015. The about 2.2 percent a year between 1985 and working-age share in the population began 1990. to rise, slowly, in 1968 with the expansion Population growth slowed more sharply in taking firm hold after 1972 on the back of the 1990s as the HIV/AIDS epidemic shaved sustained declines in the share of the popula- some eight years from male, and 11 years tion under 15. In 2015, the working-age share from female, life expectancy.13 After 2005, in the population is about 65 percent. United mortality rates resumed their decline as new Nations, 2013 put the working-age popula- HIV/AIDS treatments were disseminated and tion at 34.5 million, or more than 10 million the fertility rate fell toward 2.55 children per higher than in 1994 when apartheid ended. woman by 2015.14 Population growth aver- National data from Statistics South Africa aged 1.4 percent between 2005 and 2015, but show a slightly higher increase since 1994 of with the number of births in 2015 at an all- about 11 million, to 35.5 million in 2015. time high of 1.25 million. Either way, the window of demographic Today almost half of South Africa’s opportunity is open for South Africa (fig- population is under 25; 30 percent are under ure 2.3). The youth-dependency ratio has 15. This “young bulge” has begun to move up declined steadily since the late 1960s, creat- and change the population pyramid (figure ing space for the working-age population to 2.2), lowering the dependency ratio.15 The expand and potentially freeing resources for UN Department of Economic and Social economic growth.17 Affairs, Population Division, which publishes The working-age population, 15–64 years the longest population series for South old, will grow by at least another 9 million Africa, shows that the dependency ratio over the next 50 years, peaking in 2065 at peaked at 83.5 percent in 1970. According about 43.8 million.18 The next 20 years alone South Africa Economic Update 7 12 aug 15.indd 35 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S The dependency ratio in South Africa at its trough will be higher than in East Figure Asia 2.4 100 90 80 East Asia 70 Sub-Saharan Africa Dependency ratio 60 South East Asia 50 36 South Africa 40 30 20 10 0 The working-age 55 5 60 0 65 5 70 0 75 5 80 0 85 5 90 0 95 5 00 0 05 5 10 0 15 5 20 0 25 5 30 0 35 5 40 0 45 5 50 0 55 5 0 19 195 19 196 19 196 19 197 19 197 19 198 19 198 19 199 19 199 20 200 20 200 20 201 20 201 20 202 20 202 20 203 20 203 20 204 20 204 20 205 20 205 06 -2 - - - - - - - - - - - - - - - - - - - - - 50 19 population will " grow by 9 million Source: United Nations (2013). people by 2069 will see an average net increase in the work- countries in Eastern Europe and East Asia. ing-age population of about 280,000 people However, these data show that South Africa’s per year, although this does not necessarily total dependency ratio, even at its projected translate one-for-one into an increase in the trough of 46 percent in 2045, will be higher labor force, given that many young people than East Asia’s equivalent trough of 38 remain in education beyond their 15th birth- percent in 2010, suggesting a more shallow day. As a share of the working-age popula- transition for South Africa (figure 2.4). The tion, United Nations (2013) projects that the reason is that the proportion of the elderly peak will occur in 2045, when it will reach population will rise, offsetting some of the 68.3 percent of the population and there will reduction in the population of young people. be 2.1 working adults to support each young Nonetheless, at a time when most other and old dependent. While South Africa has major regions of the world are already facing completed almost 80 percent of the transi- the headwinds of rapidly aging populations, tion to its peak working-age share, it still has South Africa and Sub-Saharan Africa have a window of some 50 years where the work- relatively young populations (figure 2.5) ing-age share will remain high and before and, as discussed above, should continue to aging becomes a bigger concern: the share have them for around another half century. of the population aged 65 and older is pro- In Sub-Saharan Africa, fertility rates have jected to rise from just under 6 percent of the been much slower to decline and, according total population in 2015 to 9 percent in 2045, to United Nations (2013), its demographic before reaching 14 percent in 2065. window will open only in the 2060s (figure South Africa’s demographic transition to 2.5).20 This means that as South Africa begins a peak working-age population is, however, to face negative growth in its working force shallower and longer than that in countries population (blue bars in figure 2.6 from that have harvested the demographic 2045 onward) implying a rising dependency dividend. Some argue that the faster the ratio because of aging, one of its key export transition, the greater its potential impact markets, Sub-Saharan Africa, could grow on on growth,19 although the flip side is that the back of its larger working-age population the potential benefits are shorter-lived. (pink bars in figure 2.6 from 2045 onward), According to United Nations, 2013, South increasing the demand for South African Africa started its demographic transition exports and potentially cushioning the in the late 1960s at a time similar to that in impact of aging on South Africa’s growth. South Africa Economic Update 7 12 aug 15.indd 36 8/12/15 2:07 PM The demographic window is already closed in OECD countries and close to Figure closing in China and East Asia 2.5 Botswana (2027-2069) Lesotho (2036-2082) Namibia (2030-2070) South Africa (2009-2069) Swaziland (2040-2086) Sub-Saharan Africa (2064-2100) OECD (1950-2003) UMI (1996-2031) 37 EAsia (1996-2031) BRICS (1999-2036) LMI (2018-2069) China (1988-2028) UMI - China (2002-2039) EAsia - China (2004-2044) BRICS - China (2005-2052) 1950 1975 2000 2025 2050 2075 2100 Since 2000 the Source: United Nations (2013). working-age population expanded South Africa is further advanced in its demographic transition than the rest of Figure Sub-Saharan Africa but stands to benefit from a growing region by 8.5 million people, 2.6 1.5 Working-age population share growth - South Africa vs Sub-Saharan Africa but jobs grew by only 2.8 million 1.0 0.5 Percent 0 -0.5 -1.0 1950 1956 1962 1968 1974 1980 1986 1992 1998 2004 2010 2016 2022 2028 2034 2040 2046 2052 2058 2064 2070 2076 2082 2088 2094 2100 South Africa Sub-Saharan Africa Source: United Nations (2013). South Africa’s recent labor percent in 1995 to almost 50 percent by market performance amid 2005.21 demographic change But the labor market did not create enough South Africa’s supply of labor increased jobs to match this unprecedented rise in the sharply from 1995. Between 1995 and 2015, supply of labor. Between 2000 and 2014, the working-age population expanded by for example, the working-age population more than 11 million. However, the early expanded by 8.5 million people but the years after the end of apartheid also saw the number of jobs grew by only 2.8 million mass entry into the labor force of African (table 2.1).22 Some 7.7 million people—about women, who were largely unskilled and 22 percent of the working-age population— had previously been excluded from the were unemployed or had stopped looking labor force under apartheid. According for work by 2014. The unemployment rate to some estimates, the female labor force (on the narrow measure) was 25.4 percent participation rates jumped from about 41 or, if discouraged workers are included, the South Africa Economic Update 7 12 aug 15.indd 37 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Table Evolution of the South African labor market between 2000 and 2014 2.1 Working age Labor force participation Employment Unemployment, narrow Unemployment, broad (thousands) (thousands) % (thousands) % (thousands) % (thousands) % 2000 27,008 16,078 59.5 12,336 45.7 3,742 23.3 5,278 30.0 2001 27,528 15,789 57.4 11,660 42.4 4,130 26.2 6,151 34.5 2002 28,068 16,264 57.9 11,935 42.5 4,329 26.6 6,335 34.7 2003 28,585 15,906 55.6 11,959 41.8 3,947 24.8 6,349 34.7 2004 29,080 15,924 54.8 12,265 42.2 3,659 23 6,231 33.7 38 2005 29,556 17,035 57.6 13,034 44.1 4,001 23.5 6,351 32.8 2006 29,999 17,460 58.2 13,601 45.3 3,859 22.1 6,076 30.9 2007 30,414 17,232 56.7 13,609 44.7 3,623 21 6,226 31.4 2008 31,839 18,848 59.2 14,549 45.7 4,299 22.8 5,392 27.0 2009 32,435 18,306 56.4 13,830 42.6 4,476 24.5 6,122 30.7 2010 33,033 18,303 55.4 13,648 41.3 4,655 25.4 6,734 33.0 2011 33,640 18,818 55.9 14,118 42 4,699 25 6,912 32.9 2012 34,253 19,463 56.8 14,562 42.5 4,901 25.2 7,115 32.8 South Africa now 2013 34,868 19,916 57.1 15,036 43.1 4,880 24.5 7,177 32.3 2014 35,489 20,268 57.1 15,117 42.6 5,151 25.4 7,665 33.6 has one of the Source: Labour Force Surveys 2000–07, revised September series, and Quarterly Labour Force Surveys (revised with 2011 Census weights). lowest employment and labor-force unemployment rate was closer to 34 percent. rates and highest unemployment rates by participation rates and Labor-force participation rates have also upper middle-income country standards.23 remained low. After initially rising post-1994 The supply of unskilled workers in South highest unemployment and peaking at close to 60 percent in 2000, Africa grew when the demand for unskilled rates by upper labor-force participation rates fell to an workers fell and employment became more average of 57.4 percent in the pre-crisis years skills intensive. Agriculture, mining, and middle-income between 2000 and 2008. The global financial manufacturing have shed more than half country standards crisis resulted in the labor-force participation a million mainly unskilled, jobs since 2000 rate falling to 55 percent between 2009 and and now account for only 19 percent of 2011, but it subsequently recovered to the total employment, down from just under 57 percent mark in recent years. Even so, 30 percent in 2000. Total employment also by 2014 South Africa had one of the lowest became more skills intensive. In 2000, across employment and labor-force participation all sectors, there were about four unskilled/ Figure Skilled vs. unskilled workers 2.7 Employed: Education and Skills 0.35 0.30 0.25 0.20 0.15 01 03 05 06 07 08 09 11 13 00 02 04 10 12 14 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Skilled/(semi-skilled and unskilled) Tertiary/(matric or less) Source: Labor Force Surveys 2000–07, September Series, and Quarterly Labor Force Surveys 2008 Q3–2014 Q3, Statistics South Africa. Note: The matriculation exam, or matric, is the South African high school leaving exam. Tertiary is third-level education. South Africa Economic Update 7 12 aug 15.indd 38 8/12/15 2:07 PM Table Evolution of ratio of skilled to unskilled workers by sector 2.2 Average 2000-2007 Average 2008-2014 Skilled / (Unskilled Skilled / (unskilled and semi-skilled) Tertiary /matric or less and semi-skilled) Tertiary /matric or less Agriculture 0.036 0.037 0.057 0.044 Mining 0.077 0.080 0.130 0.172 Manufacturing 0.204 0.123 0.246 0.166 Utilities 0.379 0.357 0.452 0.584 Construction 0.089 0.057 0.162 0.091 39 Wholesale and retail 0.161 0.084 0.189 0.106 Transport/comm 0.330 0.142 0.335 0.199 Financial/real estate/business services 0.731 0.437 0.694 0.447 Community and social services 1.069 0.709 0.922 0.730 Total 0.277 0.186 0.333 0.243 Source: Labour Force Surveys (September series), Quarterly Labour Force Surveys (third quarter), Statistics South Africa, and staff calculations. In 2000 there were semi-skilled workers to every skilled worker; youth facing the prospect of long-term job- around four unskilled by 2014 there were just three. Figure 2.7 shows lessness, and they appear ill equipped for that the ratio of skilled to unskilled and semi- a labor market that demands higher skills. workers for each skilled workers rose by roughly 10 percentage According to the Quarterly Labor Force points from 2000 to 2014.24 Similar trends are Survey for the second quarter of 2015, about skilled worker, but evident using levels of educational attainment two-thirds of the unemployed have been by 2014 this ratio (table 2.2). Other institutional constraints, searching for jobs for a year or more; 40 per- including union activity, spatial disparities cent are new entrants to the labor force and had fallen to three that see many of South Africa’s unemployed thus being just out of school they lack work as the demand for living in townships and informal settlements experience. Of 10.2 million 15–24 year olds, far removed from urban centers, and limited almost a third do not work and are not in unskilled workers fell. opportunities in the informal economy also school. The unemployment rate for 15–24 contributed to the lack of job opportunities year olds is almost 50 percent (measured as and social networks for unskilled workers and a percent of 2.7 million 15–24 year olds in new labor-market entrants since 2000. the labor force). If we broaden the young to Insufficient job creation has left South also include those up to the age of 34, almost Africa’s unskilled or semi-skilled and its two-thirds of the 5.2 million unemployed are Figure Drivers of 1.7 percent increase in real per capita GDP, 2000 –13 2.8 2.19 Productivity Employment -0.15 Participation rate -0.52 Demographic change 0.14 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 % Source: World Bank calculations based on real GDP per capita and employment from the revised 2000–07 Labor Force Survey (September series) and the Quarterly Labor Force Survey data for 2008 Q3 to 2013 Q3 using consistent census weights. South Africa Economic Update 7 12 aug 15.indd 39 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S in this 15–34 age bracket. While more than from annual growth in per capita incomes, half of those employed have the South Afri- essentially canceling out the positive impact can matric (high school)-level exam qualifi- (0.14 percentage points per year) of the cation or a higher level of education, about increasing working-age population. Were 60 percent of the unemployed have not even it not for the drag from lower-labor partici- achieved the matric qualification. Well-docu- pation and employment, annual growth in mented concerns about the quality of educa- real per capita GDP could have been almost tion in South Africa imply that the increase 30 percent higher than it has been. Box 2.2 in years of schooling that has occurred might examines in more detail the role that the 40 not be translating into the skills needed in its service sector played in driving changes in labor market.25 labor productivity and employment before Moreover, since 2000 economic growth in and after the crisis and highlights the worri- South Africa has largely been driven by gains some trend that, since the end of the global in labor productivity and has not been jobs financial crisis, growth in labor productivity intensive.26 In real terms, per capita GDP rose appears to have petered out. by 1.7 percent a year between 2000 and 2013 The inability to create sufficient job Increases in years of (figure 2.8). Labor productivity, measured by opportunities, especially for its young and output per worker, rose by 2.2 percent a year. unskilled workers, has stymied South Africa’s schooling might not However, the decreased labor-force participa- ability to capitalize on its demographic be translating into tion rate since 2000 subtracted 0.5 percent- opportunity so far. The size of the economy age points a year from headline per capita as measured by real GDP was 80 percent the skills needed in GDP growth. The decline in employment 27 larger and incomes in real per capita terms the labor market. shaved off a further 0.15 percentage points of R55,712 in 2014 were just 40 percent Box Sector-level trends in employment and productivity 2.2 Growth in per capita income has been driven by gains in labor productivity and not job creation. Total labor productivity rose by 2.2 percent a year between 2000 and 2013, driving overall growth in per capita incomes. Agriculture, manufacturing, and mining combined shed more than a half a million jobs and saw their combined labor productivity grow at an average rate of 0.7 percentage points a year. Services added more than 2.2 million jobs over the period, but this was not enough to absorb the growing working-age population plus those who lost their jobs in other sectors.29 But services combined job creation with improved labor productivity, accounting for 60 percent of the annual increase in total labor productivity. Within services, employment rose the most in finance, real estate, and business services (up by almost 890,000) and output per worker remained high at around R300,000 per worker. Box figure 1. Changes in employment and labor productivity 200 Change in labor productivity, 2000-13 (%) Agriculture 150 100 Trade Personal services Construction Transport 50 Financial services Manufacturing Mining 0 -80 -60 -40 -20 0 20 40 60 80 100 120 Community services Utilities -50 -100 Change in employment, 2000-13 (%) Source: World Bank calculations based on real GDP per capita and employment from the revised 2000–07 Labor Force Survey (September series) and the Quarterly Labor Force Survey data for 2008 Q3 to 2013 Q3 using consistent census weights. The size of the bubble corresponds to the rand value of gross value-added of the sector in 2013 reported by Statistics South Africa. South Africa Economic Update 7 12 aug 15.indd 40 8/12/15 2:07 PM Box Sector-level trends in employment and productivity (continued) 2.2 Box figure 2. Annual change in employment (as a share of labor force) by sector, 2000–13 0.50 Change in employment as a share of labor force, 2000-13 (%) 0.40 0.30 41 0.20 0.10 Wholesale and Agriculture Manufacturing Mining retail 0.00 Construction Other Finance and and utilities activities business -0.10 -0.20 The new jobs -0.30 Contribution of employment to overall per capita -0.40 GDP growth = – 0.15% a year needed to reduce -0.50 unemployment and -0.60 absorb the many new entrants to the labor Job creation, especially in services, expanded after the global financial crisis but productivity growth petered out. Prior to the crisis (2000–08), services employment grew by 1.4 million and labor productivity in services rose by 1.7 percent per year. force are unlikely Since 2010, total employment has grown by about 1 million, recouping the jobs lost in the crisis, driven largely by new jobs to materialize in in services. However, total labor productivity contracted by an average of 0.23 percent a year between 2010 and 2013, reflect- ing declines in output per worker in finance, real estate and business services, community services, and transport as employment the coming decades in these sub-sectors expanded. Growth in real per capita GDP slowed sharply to an average rate of just 0.86 a year between 2010 and 2013. without a better enabling environment. Box figure 3. Productivity growth, 2000–08 3.00 Intersectoral reallocation 2.50 Other Services Annual change in labor productivity, 2000–08 (%) 2.00 Services Finance and business 1.50 Wholesale and retail 1.00 Construction and utilities Secondary Manufacturing 0.50 Mining Primary Agriculture 0.00 South Africa Economic Update 7 12 aug 15.indd 41 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Box Sector-level trends in employment and productivity (continued) 2.2 Box figure 4. Productivity growth, 2010–13 0.30 0.20 Intersectoral reallocation Annual change in labor productivty (%) effect 0.10 42 Secondary 0.00 Primary Annual change in -0.10 labor productity = -0.23% 2010-13 -0.20 Services -0.30 -0.40 -0.50 Source: World Bank calculations based on real GDP per capita and employment from the revised 2000–07 Labor Force Survey (September series) and the Quarterly Labor Force Survey data for 2008 Q3 to 2013 Q3 using consistent census weights. higher than in 1994. Savings as a share We use a World Bank model, LINK AGE of GDP had fallen to 14.9 percent of GDP (van der Mensbrugghe, 2011), 30 to examine by end-2014, from 17.7 percent in 1994, how improvements in the enabling environ- reflecting declining household disposable ment could help Africa better seize its demo- income amid South Africa’s chronically high graphic opportunity. The model ties growth unemployment.28 to changes in the labor force, investment in It is hard to envisage a drastic fall in the capital, and total productivity. The impact of number of jobless or enough new jobs to a changing age structure occurs as a rising absorb the many new entrants forecast to join working-age population increases the sup- the labor force in the coming decades with- ply of labor and drives higher per capita out a better enabling environment. income (real GDP). As the labor supply rises, the associated reduction in the number of Making the most of South Africa’s young or old dependents can create space future demographic changes for households to save more. Higher savings Under what conditions can South Africa cap- in turn boosts investment and drives higher italize on its demographic transition? The growth in the model. answer depends on a wide range of economic The model also considers the structure variables and how demographic changes of the labor market, differentiating between affect these, and vice versa. Job creation is skilled and unskilled labor, which in turn just the first stepping stone (figure 2.9). Bet- affects levels of productivity and overall ter labor productivity, educational attain- growth. As countries grow, they see changes ment, and skills are also critical enablers if in their share of skilled and unskilled workers, new entrants to the working-age population and LINKAGE keeps track of the young, work- are to secure jobs in a world where skills are ing-age, and old-age shares of the population in high demand. But, as the previous section each year to calculate dependency ratios. showed, these are areas where South Africa To illustrate how a growing working- faces challenges. age population has the potential to affect South Africa Economic Update 7 12 aug 15.indd 42 8/12/15 2:07 PM Figure Stepping stones in realizing a demographic dividend 2.9 43 Second demographic dividend First demographic dividend Source: Authors’ own depiction. growth, incomes, poverty, and inequality in 2030.32 The employment rate rises to 54 South Africa, we contrast a business-as-usual percent of the working-age population. baseline scenario with three other potential • In scenario 2—“the productivity-enhanc- futures that evolve around the steps involved ing scenario”—the fall in unemployment in creating a better enabling environment from scenario 1 is accompanied by more to realize the first and second demographic rapid gains in labor productivity, whose dividends: growth rates converge with the BRIC • In the business-as-usual baseline group average by 2030, implying a 30 scenario, the labor market’s ability percent higher average growth rate in to absorb new entrants remains labor productivity than in scenario 1. constrained. Labor-force participation • In scenario 3—“the accelerated educa- remains low at its 2014 level of 57 tional attainment scenario”—the lower percent and employment rate stagnates unemployment rate and faster labor- at about 43 percent of the working-age productivity growth of scenario 2—is population.31 Unemployment persists at complemented by faster skills attainment 25.1 percent. The number of employed by school graduates entering the work- increases at the same rate as the working- ing-age population. In scenarios 1 and age population, 0.76 percent a year 2, educational attainment rates stay at (drawn from the UN World Population current levels, but even then the share Prospects, 2013), still faster than the of workers with more than nine years of growth rate of the total population (0.55 schooling rises as younger-age cohorts percent a year). with more years of schooling than older • In scenario 1—“the job - creation cohorts cause the share of employed scenario”—the labor market improves with more than nine years of schooling considerably. Enough new jobs are to rise from 61 percent in 2014 to 68 per- generated not only to absorb all cent by 2030.33 In scenario 3 we assume new entrants but also to reduce the share of the working-age population unemployment to the upper-middle- that has at least nine years of schooling income country average of 5.8 percent by rises to 72 percent—a level on par with South Africa Economic Update 7 12 aug 15.indd 43 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Figure Number of people employed in baseline and scenario 1 2.10 Unemployment rate constant at 25.1% (baseline) Unemployment rate converges with UMC average of 5.8% (Scenario 1) 21 20 19 Millions of employed 18 44 17 16 15 14 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Source: World Bank staff estimates. Without an improved Note: The employment figures in the baseline and scenario 1 are determined by calculating the employment ratio based on the unemployment and labor-force participation rates, and the working-age population projections. The unemployment rate in the Quarterly Labor Force Survey for 2014 was 25.1 percent. The upper middle-income country average unemployment rate of 5.8 percent is from WDI 2014; in scenario 1, it is assumed that the rate converges from 25.1 labor market, percent to 5.8 percent at a uniform pace over 15 years. Labor-force participation is held constant in both scenarios at 57.1 percent. The working-age population share projections are taken from the United Nations World Population Prospects (2013) medium fertility scenario. only 40 percent of the projected skill share of 18 non-Mid- Job-creation scenario those comprising dle Eastern high-income countries—to This scenario illustrates the necessity for the increase in the proxy for improvements in educational strong job creation to employ the growing attainment.34 working-age population. Here, South working-age cohort We now look at each scenario in more detail. A frica’s unemployment rate converges will find jobs with the upper middle-income country Business-as-usual baseline scenario average of 5.8 percent by 2030 and the If unemployment and employment remain employment rate rises to about 54 percent.36 constant at current rates, South Africa sees Labor productivity growth is positive but only a very modest boost to growth and not as strong as in other emerging market incomes from its rising working-age popula- economies. In this more optimistic future, tion. In the baseline, real GDP growth aver- growth is far more jobs intensive: the number ages 3.7 percent a year and growth in real per of people employed grows almost three times capita incomes averages 3.1 percent a year faster than in the business-as-usual case, with over 2015–30. The number of employed rises jobs growth averaging 2.2 percent a year. By modestly from about 14.8 million workers in 2030, 4.03 million more people are working 2015 to 16.6 million by 2030. This implies than in the previous case (figure 2.10) and that only about 40 percent of the 4.3 million relative to 2015 some 5.8 million new jobs increase in the working-age cohort over the are created, which is more than enough to next 15 years will find jobs. Moreover, the employ the new entrants and allow South high rates of youth unemployment 35 imply Africa to make a significant dent in its that the economy is not only forgoing the unemployment rate. benefit of the labor supply growing faster Relative to the baseline, the larger than the overall population, but also is miss- number of workers employed has the effect of ing the potential benefit that these new increasing total economic output. Similarly, entrants have more years of schooling. Nor real GDP is about 13 percent bigger than in do incomes rise enough to create space for a the baseline in 2030 and 95 percent bigger sharp increase in savings: savings as a share than in 2015, where the unemployment rate of GDP rise modestly from 14.5 percent of converges with the upper middle-income GDP in 2015 to just under 18.5 percent by country average. In this scenario, overall GDP 2030. and real GDP per capita rise by an average Ultimately, this scenario falls far short of 4.6 and 4.0 percent a year, respectively, or of what South Africa could achieve from its roughly 1.3 times faster than in the business- demographic tailwinds. as-usual scenario (figure 2.11). By 2030, real South Africa Economic Update 7 12 aug 15.indd 44 8/12/15 2:07 PM Rapid improvements in unemployment rates, labor productivity, and educational Figure attainment can accelerate growth in income and consumption 2.11 Average annual growth rates, 2015–30 (%) Growth, 2015–30 (%) 120 6.0 110 5.5 100 5.0 90 80 4.5 45 70 4.0 60 50 3.5 40 3.0 30 Real GDP pc Real GDP Consumption Real GDP pc Real GDP Consumption If jobs can be created Business as usual baseline Scenario 1: Unemployment converges to 5.8% to absorb new Scenario 2: Unemployment and productivity convergence Scenario 3: Unemployment and productivity convergence plus accelerated educational attainment entrants and lower Source: World Bank staff estimates. the unemployment Note: LINKAGE simulation results for the business-as-usual baseline and scenarios 1, 2, and 3. In scenario 1 the unemployment rate converges from 25.1 percent in 2014 to 5.8 percent in 2030. Scenario 2 is identical to scenario 1, except that the average annual labor productivity growth rate over 2015–30 is 30 percent higher and converges with the BRIC group average productivity growth rate by 2030. Scenario 3 is identical to scenario 2, but with the share of skilled workers rate, GDP growth converging with the average of 18 non-Middle Eastern high-income countries. might average 4.6 GDP per capita would be US$11,356 per population moves toward its peak. In this sec- percent a year. person, about 13 percent more than the ond scenario, we examine how growth and US$10,056 in the business-as-usual case. income evolve if the decline in unemploy- But even with more jobs, the gains to ment highlighted above is also accompanied growth and income from a growing work- by improved labor-productivity growth that ing-age population appear relatively modest could come about through better training in South Africa because aging is offsetting and skills. Here, overall real GDP rises at an some of the benefits. By 2030 the share of average rate of 5.3 percent a year, above the the population older than 65 will already 5 percent target set in the National Devel- have reached 8 percent, 2.25 percentage opment Plan (2012). The economy’s size points more than today. This is limiting the more than doubles between 2015 and 2030 scope for a potential second demographic and is one-quarter larger in 2030 than in dividend. Because the old draw down their the business-as-usual baseline case. In per savings faster than the working-age share of capita terms, real GDP growth averages 4.7 the population can increase national sav- percent a year and per capita incomes reach ings, the increase in the old-age dependency US$12,558 by 2030. The increase in produc- ratio, particularly after 2027, is high enough tivity helps offset some of the pressures on to limit the improvements in overall savings savings and growth from having a growing and the secondary boost to economic growth share of the population older than 65. Sav- via greater investment. Total savings reach ings as a share of GDP is 20.8 percent of GDP 19.5 percent of GDP by 2030, only about 1 in 2030, 37 which allows South Africa to reap percentage point of GDP higher than in the a somewhat higher dividend from its large business-as-usual scenario. working-age population. Productivity-enhancing scenario Accelerated educational-attainment scenario Higher growth in labor productivity will The impact of a growing working-age need to accompany job creation if South population on growth and incomes would Africa is to make greater strides in rais- be further enhanced by efforts to improve ing growth and incomes as the working-age educational attainment and skills. In the South Africa Economic Update 7 12 aug 15.indd 45 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S third scenario, job creation and faster labor- impact on household characteristics, sec- productivity growth is accompanied by these tor of employment, skills, and income; and efforts. In this scenario, overall real GDP and allows the income distribution to evolve over real GDP per capita rise at an average of 5.4 time. This calculation process in turn is used percent and 4.8 percent a year, respectively. to estimate the impact of changing growth Per capita incomes reach US$12,766 and and income trajectories on poverty and savings rise to 21 percent of GDP by 2030. The inequality. gain in the economy of higher educational Extreme poverty—measured by the share attainment is understated by these numbers, of the population living on less than $1.25 46 because 15 years is not long enough for the per day—could be virtually eliminated if full impact of these efforts on growth and South Africa could combine jobs creation incomes to fully materialize. with improvements in its labor productivity and educational attainment (figure 2.12). Implications for poverty and inequality In the business-as-usual baseline the share What are the implications, for poverty reduc- of the population living on less than $1.25 tion and inequality, of capitalizing on South (PPP adjusted) a day falls from 9.4 percent in Extreme poverty could Africa’s demographic trends by creating 2011 (Povcalnet, 2015) to 2.7 percent in 2030. a better enabling environment relative to However, the extreme poverty rate falls to be almost eliminated the business-as-usual baseline? We map the about 1 percent of the population in scenario by 2030 if South Africa different trajectories of per capita income 3. When measured at the $2 a day poverty growth from each simulation to households line, the poverty rate falls by almost three- could generate jobs using the 2010/11 Expenditure Household quarters from 23.7 percent in 2011 to about 4 and improve labor- Survey for South Africa. 38 This mapping percent in 2030 in the most optimistic third allows us to generate income distributions for scenario. productivity growth each scenario that account for the changes The Gini coefficient narrows the most in in South Africa’s demographics and their the third scenario from 0.645 in 2011 to 0.630 Poverty headcount rates will fall below 5 percent by 2030 and inequality can be reduced by increasing productivity and educational attainment, and generating Figure more jobs 2.12 Poverty headcounts, 2011 and 2030 (%) Gini coef cient, 2015 and 2030 25.0% 0.650 0.645 20.0% 0.640 15.0% 0.635 10.0% 0.630 5.0% 0.625 0.0% 0.620 Poverty ($1.25) Poverty ($2.00) 2011 2030 2030 2030 2030 Gini Business Scenario 1 Scenario 2 Scenario 3 as usual Unemployment Increase Educational reduction productivity improvement 2011 Poverty rate 2030 Business as usual baseline 2030 Scenario 1 2030 Scenario 2 2030 Scenario 3 Source: World Bank staff estimates. Note: LINKAGE simulation results for the business as usual baseline. In scenario 1, the unemployment rate converges from 25.1 percent in 2014 to 5.8 percent in 2030. Scenario 2 is identical to scenario 1, except that the average annual labor productivity growth rate over 2015–30 is 30 percent higher and converges with the BRIC average productivity growth rate by 2030. Scenario 3 is identical to scenario 2, but with the share of skilled workers converging to the average of 18 non-Middle Eastern high-income countries. The poverty headcount rate for 2011 is taken from Povcalnet. The poverty headcount rates in 2030 under different scenarios are based on GIDD simulation results. The poverty headcount rate is based on the $1.25/day (PPP-adjusted) and $2/day (PPP-adjusted) poverty line. South Africa Economic Update 7 12 aug 15.indd 46 8/12/15 2:07 PM by 2030. This modest reduction is driven by the young unemployed ill equipped for a two effects related to demographic change. labor market that demands more skills. First, an increase in the number of work- More job-intensive growth would help ers in general, especially skilled workers, in tackle unemployment and create jobs for poorer households helps reduce inequality.39 many new labor-force entrants in the com- Second, wages of skilled workers are growing ing 15 years, allowing South Africa to take more slowly than those of unskilled workers the first step in harnessing its favorable due to the relatively faster growth in skilled demographics. But simply increasing the labor supply, which leads to a reduction in number of jobs will not be enough to allow the wage skill premium.40 These changes are South Africa to boost savings and derive the 47 less significant between scenarios 1 and 2, second demographic dividend. More job- where both the number of workers and the intensive growth needs to be accompanied relative growth of skilled workers are similar. as a first priority by improving the quality of education so that better educated youth Conclusions are entering the workforce. This needs to be South Africa is in its demographic window of complemented by efforts to improve the pro- opportunity and will remain there for around ductivity of existing workers and the unem- 50 more years. Since 1994, its working-age ployed through better skills development and population (15–64 years) has increased by training. By creating a virtuous circle of job- 11 million. In the next 50 years, it will grow intensive growth, improved productivity and by another 9 million. Having such a high educational attainment, and higher savings, share of its population—68.3 percent at its growth could accelerate to 5.4 percent a year peak in 2045—in its working prime presents and per capita incomes could double by 2030, a tremendous opportunity for the country to which would virtually eliminate extreme pov- boost its growth and raise living standards. erty and begin to reduce inequality. But it also presents tremendous challenges. Changing the growth and jobs dynamics In the next 15 years alone, the working-age will require urgent action on several mutu- population will expand by 280,000 a year. ally reinforcing fronts. The government These people will have to find productive has already introduced an employment tax jobs if South Africa is to harvest the potential incentive to encourage firms to hire young boost to growth and living standards before workers. Through its Industrial Policy Action it faces the rising burden associated with a Plan, it is also offering incentives to promote rapidly growing elderly population. potentially labor-intense sectors like manu- South Africa has, however, struggled facturing and agriculture. Faster and deeper with a high burden of unemployment as global and regional integration in trade in the number of jobs created has fallen far goods and services would bolster this effort. short of what is needed to absorb into the Nonetheless, low-cost, labor-intense pro- labor force the many new young people of duction is unlikely to be the main engine working age. Since 2000, only one-third of for job creation for South Africa, given how the 8.5 million additional working age found these sectors have shrunk over the past two jobs, mainly in the services sector. Many decades. Policies also need to focus on devel- new entrants were unskilled or lacked basic oping services, small and medium firms, education qualifications, and they joined the and household enterprises, including in the workforce when labor-intense sectors such informal economy, as engines for job cre- as agriculture, mining, and manufacturing ation. Policies that improve the business envi- shed jobs and employment became more ronment, especially for small firms, could skills intensive. Unemployment is now include reducing the burden of red tape (see higher than it was at the end of apartheid, box 1.3 in chapter 1), improving access to with almost one-third of the labor force low-cost finance, and securing greater flex- out of work or discouraged. At a time when ibility in labor–market regulations. almost half the population is under 25, The greatest priority on the supply side is unemployment among the young (15–24) is to improve levels of educational attainment almost 50 percent, double the national rate. in South Africa. Getting basic schooling The education system appears to have left right is the first step to ensuring that school South Africa Economic Update 7 12 aug 15.indd 47 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S leavers and graduates have the foundational 8. Oosthuizen, 2014 finds that South skills necessary to function in the modern Africans enjoy surplus earnings between workplace. Educational attainment not only ages 30 and 59. High rates and the long shapes employment opportunities,41 but also duration of youth unemployment means provides the foundation for further on-the- that the transition to surplus occurs late job learning and training. This will not be an and surplus earnings at one-quarter of easy task. South Africa has already achieved mean labor income for 30–49-year-olds almost universal school attendance and are low. the challenge now is to improve learning 9. The United Nations defines this window 48 outcomes by better training and support of as opening when a country’s youth share teachers. in the population falls below 30 percent Promoting skills, especially of the long-term and its elderly share remains below 15 unemployed, also calls for complementary percent (UN, 2004). efforts. Steps are being taken to scale up 10. Lin, 2012. technical and vocational education and 11. See National Treasury, 2014. training opportunities. But access remains a 12. The population figures for 1965 are challenge, especially for the young who lack derived from United Nations, 2013. a high school qualification, while the system The population for 2015 is the recently continues to face problems of quality and released mid-year estimate from Sta- institutional capacity in supplying the skills tistics South Africa, 2015. After our demanded by employers. To broaden access, analysis was completed, the UN World training opportunities should match the actual Population Prospects (2013) released educational levels of young unemployed, while updated population projections for the financial support for enrollees could help end of July 2015. However, the projec- overcome the hurdles of high transport and tions for South Africa were not funda- living costs. Designing training programs in mentally altered from what is presented consultation and partnership with the private in this paper. sector will ensure that they are high quality and 13. See Moultrie (2015). better geared to the needs of the labor market. 14. According to the UNAIDS Spectrum Employers can also provide internships and model, which incorporates the effects other opportunities for practical training of HIV/AIDS, HIV prevalence among to help overcome new entrants’ lack of work adults of both sexes aged 15–49 in South experience. Africa is close to its peak, and is expected to decline slowly from the current level Notes reported by Statistics South Africa, 2015 1. This effect can be ­reinforced when  the of 16.59 percent in 2015. decline in fertility also stimulates in- 15. The ratio of number of children (under creased female labor-force participation. 15 years) and those older than 64 to the 2. Bloom, Canning, and Sevilla, 2003. total number of people between the ages 3. See Cai and Wang, 2006. of 15 and 64 years. 4. See Bloom et al., 2003. 16. Authors’ calculations based on the Quar- 5. See Bloom et al., 1999, and Drummond, terly Labor Force Survey, Q4 2014. Thakoor, and Yu, 2014. 17. According to the United Nations’s defi- 6. Lee and Mason, 2006. nition in its World Population Prospects 7. See Banerjee, Galian, Levinsohn, (2013) (which defines the window of McLaren, and Woolard, 2008 and Wit- demographic opportunity as opening tenberg, 2014. Before 2000, Statistics when the youth dependency ratio falls South Africa collected labor-force indi- below 30 percent), the window opened in cators using the annual October House- South Africa in 2009 and will not close hold Survey (OHS). In 2000, Statistics until 2069. South Africa switched to a Labor Force 18. For our discussion of population projec- Survey (LFS), causing a break in series, tions, we use United Nations, 2013, the so comparisons with the earlier period median fertility scenario. At the time of should be made with care. publication, Statistics South Africa had South Africa Economic Update 7 12 aug 15.indd 48 8/12/15 2:07 PM not yet published national population on employment (for 2000–07) and the projections that capture the updated Quarterly Labor Force Survey data for 2015 population estimate. Moreover, 2008 Q3 to 2013 Q3, which are reported the national projections do not extend by Statistics South Africa using consis- beyond 2030. tent census weights. From 2014 Q3, the 19. Mason and Lee, 2012. Quarterly Labor Force survey data were 20. See McKinsey Global Institute, 2010, compiled using the new 2011 census Ahmed, Cruz, Go, Maliszewska and weights and so are excluded from the Osorio-Rodarte, 2014, and International Shapley decomposition for consistency Monetary Fund, 2015. purposes. 49 21. See Banerjee et al., 2008. 27. Measured in the Shapley decomposition 22. Before 2000, labor-force data were as a percentage of the labor force. gathered from the annual October 28. See World Bank, 2011. Household Sur vey and are not 29. South Africa is not alone in experienc- comparable to the data collected from ing rapid growth driven by the services 2000 onward. sector. Ghani and O’Connell, 2014 find 23. According to the World Development that other countries in Africa such as Indicators, the average employment Ethiopia, Kenya, and Zambia have expe- ratio was about 54 percent for upper rienced rapid growth in real GDP driven middle-income countries, whereas the by the services sector. unemployment rate and labor-force par- 30. LINK AGE is a dynamic simulation ticipation rate were about 11 percent and model that captures global behaviors in 65 percent, respectively. general equilibrium. It is a global model 24. Statistics South Africa define semi- that also incorporates the changing skilled occupations as jobs such as clerks, nature of trade and investment linkages sales and services, skilled agriculture, between South Africa and its key trading craft and related trades, and plant and partners, including the BRIC group. The machine operators. Low-skilled are ele- model is explained in detail in van der mentary and domestic workers. Manag- Mensbrugghe (2011). Data from Global ers, professionals, and technicians are Economics Prospects 2015, January skilled occupations. (World Bank, 2015) are used to determine 25. See, for example, van der Berg, S., Tay- real GDP per capita growth through 2017. lor, S., Gustafsson, M., Spaull, N., and After 2017 through 2030, these figures Armstrong, P., 2011. Trends in the Inter- come from the OECD. Also after 2017, national Math and Science Study showed such growth is endogenously determined that the average scores on math and sci- as the model solves for different equilibria ence for South Africa’s best-performing over time. Domestic savings as a share of students (those in the 95th percentile) GDP are parameterized following the were below the average scores achieved empirical estimates by Loayza et al. (2000) by students in Singapore; Taiwan, China; for the elasticity of savings to growth in the Republic of Korea; Japan; Finland; GDP per capita, and child dependency Slovenia; and the Russian Federation and old-age dependency ratios. (HSRC, 2012). 31. On the basis of the Quarterly Labor 26. A Shapley decomposition is used to Force Survey at the end of 2014, the decompose per capita real GDP growth unemployment rate is taken to be 25.1 into contributions from labor-produc- percent. tivity growth, employment growth, and 32. The upper middle-income country aver- labor-force growth. The aim is to assess age unemployment rate is the ILO-mod- their roles in driving overall economic eled average from the World Bank’s WDI growth. Our decomposition uses data database. from 2000–13 (revised Labor Force 33. The education levels of the working- Survey) to ensure a consistent employ- age population projected by this sim- ment data series. It uses the revised ple assumption and the demographic Labor Force Survey (September series) effect is consistent with the Constant South Africa Economic Update 7 12 aug 15.indd 49 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Enrolment Ratio scenario of the KC and must equal labor demand, the increase Lutz (2010) education projections data- in the capital-to-labor ratio implies that base, which is the least optimistic of the wage growth is constrained to ensure education projections in that database. that all workers remain employed. 34. The 18 non-Middle Eastern high-income 38. Estimates are derived using the GIDD countries report survey data to I2D2/ micro-simulation framework developed GIDD (Global Income Distribution by Bussolo, Hoyos, and Medvedev (2010). Dynamics) database. The average of The GIDD draws on the 2010/11 House- 72 percent is based on the conservative hold Survey data for South Africa to esti- 50 assumption that educational attain- mate the distribution of incomes across ment rates will remain stable in these households. In addition to incorporat- countries. ing the key changes in our variables that 35. Taken as the average in the 2014 Quar- are derived from our three scenarios terly Labor Force Surveys: about 50 per- conducted in the LINK AGE model, cent for 15–24 year olds and 30 percent the GIDD methodology updates the for 25–29 year olds. household survey data for the end year 36. The 5.8 percent unemployment rate of our simulation, 2030. This is done by assumption is consistent with the National reweighting the population character- Development Plan’s (NDP) employment– ized by the 2010/11 Household Survey unemployment target for the 2010–30 using no-parametric cross-entropy meth- period. However, the NDP also incorpo- ods but keeping this process consistent rated an increase in the labor-force par- with the UN population projections. For ticipation rate to 65 percent on top of the skill–unskilled breakdown, and to the fall in the unemployment rates to ensure comparability across countries, 6 percent by 2030. This would yield an the GIDD defines as skilled anyone with employment ratio of 61 percent. The more than nine years of education. assumptions considered in our scenarios 39. The Gini coefficient declines from 0.645 are more conservative, since labor-force to 0.6393 when the income distribution participation rates are assumed to remain of 2011 is estimated on the basis of the constant at 57.1 percent over the time new weights of 2030, an approach in line horizon. As a result, the employment with the UN population projection for ratio for 2030 is 54 percent in the three 2030. non-baseline scenarios. 40. The wage–skill premium is defined as 37. The increase in savings boosts invest- the ratio between the average wage of ment, which in turn requires more labor. skilled and of unskilled workers. Under the model, since labor supply 41. See World Bank (2012). South Africa Economic Update 7 12 aug 15.indd 50 8/12/15 2:07 PM References Ahmed, S.A., M. Cruz, D.S. Go, M. Content of Export Labor (database). Was- Maliszewska, and I. Osorio-Rodarte. 2014. hington, DC: World Bank. “How Significant is Africa’s Demographic Drummond, P., V. Thakoor, and S. Yu. 2014. Dividend for its Future Growth?” Policy “Africa Rising: Harnessing the Demogra- Research Working Paper 7134, World phic Dividend.” IMF Working Paper, 14 Bank, Washington, DC. (143): 1. Banerjee, A., S. Galiani, J. Levinsohn, Z. Ghani, E., and S. O’Connell. 2014. “Can Ser- McLaren, and I. Woolard. 2008. “Why Has vices be a Growth Escalator in Low Income Unemployment Risen in the New South Countries?” Policy Research Working Africa?” Economics of Transition 16 (4): Paper 6971, World Bank, Washington, DC. 715–40. HSRC (Human Sciences Research Council of Bloom, D., D. Canning, B. Evans, B. Graham, South Africa). 2012. “Towards Equity and P. Lynch, and E. Murphy. 1999. “Popula- Excellence. Highlights from TIMSS 2011: tion Change and Human Development The South African Perspective.” Pretoria, in Latin America.” Background paper for South Africa. IPES 2000. Cambridge, MA: Harvard Insti- International Monetary Fund. 2015. Regional tute for International Development. Economic Outlook: Sub-Saharan Africa. April. Bloom, D., D. Canning, and J. Sevilla. 2003. Washington, DC: International Monetary “The Demographic Dividend—A New Fund. Perspective on the Economic Consequen- KC, S., and W. Lutz. 2014. “The Human Core ces of Population Change.” RAND. of the Shared Socioeconomic Pathways: Bussolo, M., R. de Hoyos, and D. Medvedev. Population Scenarios by Age, Sex and 2010. “Economic Growth and Income Level of Education for all Countries to Distribution: Linking Macro Economic 2100.” Global Environmental Change. Models with Household Survey Data at the Lee, R., and A. Mason. 2006. “What is the Global Level.” International Journal of Micro- Demographic Dividend?” Finance and Deve- simulation 3 (1): 92–102. lopment 43 (3). International Monetary Cai, F., Y. Du, and M. Wang. 2006. “Emplo- Fund. yment in China’s Fast Growing Region: Lin, J.Y. 2012. “Youth Bulge: A Demographic What Can We Learn from it to Make Dividend or a Demographic Bomb in Employment Expansion Keeping Up with Developing Countries?” Let’s Talk Develop- Economic Growth?” Background paper ment (Chief Economist Blog). Washington, submitted to the World Bank AAA Program DC: World Bank. on China’s Labor Market Development. Loayza, N., K. Schmitt-Hebel, L. and Servén. Calì, M., J. François, M. Manchin, D.A. Ober- 2000. “What Drives Private Saving Across dabernig, H. Rojas-Romagosa, S. Rubi- the World? Review of Economics and Statistics nova, and P. Tomberger. 2015. The Labor 82 (2): 165–81. 51 South Africa Economic Update 7 12 aug 15.indd 51 8/12/15 2:07 PM SO U TH A FRICA E C ONOMIC U P D AT E — J OB S A N D S O UT H A FR I C A ’S C H A N G I N G D EM O G R A P H I C S Mason, A., and R.D. Lee. 2012. “Demographic ———. 2004. “World Population to 2300.” Dividends and Aging in Lower-Income Department of Economic and Social Countries.” National Transfer Accounts Affairs, Population Division. New York: Working Paper 12-01. United Nations. McKinsey Global Institute. 2010. “Lions on van der Berg, S., S. Taylor, M. Gustafsson, B, the Move: The Progress and Potential Spaull, and P. Armstrong. 2011. “Impro- of African Economies.” McKinsey and ving Education Quality in South Africa.” Company. Report for the National Planning Com- Moultrie, T.A. 2015. “Demographic Profi- mission. Stellenbosch, South Africa: Uni- 52 les of Five Countries in Southern Africa versity of Stellenbosch, Department of and Implications for the Demographic Economics. Dividend.” Washington, DC: World Bank van der Mensbrugghe, D. 2011. LINKAGE mimeo. Technical Reference Document Version National Treasury. 2014. “South Africa’s Long- 7.1. Washington, DC: World Bank. Term Fiscal Choices.” Report presented at Wittenberg, M. 2014. “Analysis of Employ- Conference on Fiscal Policy and Equality, ment, Real Wage and Productivity Trends Pretoria, November 5. in South Africa since 1999.” Conditions Oosthuizen, M. 2014. “Capturing South Afri- of Work and Employment Series 45. ca’s Demographic Dividend.” In Oxford Geneva, Switzerland: International Labor Companion to Economics in South Africa, Office. edited by Haroon Bhorat, Alan Hirsch, World Bank. 2015a. Global Economic Prospects Ravi Kanbur, and Mthuli Ncube, 231–35. 2015 January: Having Fiscal Space and Using Oxford: Oxford University Press. It. Washington, DC: World Bank. Povcalnet. 2015. World Bank Povcalnet Data- ———. 2015b. Global Monitoring Report bank, 2015 edition. http://iresearch.­ 2015/16: Development in an Era of Demogra- worldbank.org/povcalnet. phic Change. Internal Concept Note. Was- Statistics SA. 2015. Mid-Year Population Estima- hington, DC: World Bank. tes, 2015. Pretoria, South Africa. ———. 2012. South Africa Economic Update: United Nations. 2013. “United Nations World Focus on Inequality of Opportunity. July. Was- Population Prospects.” Rev. DVD ed. 2012. hington, DC: World Bank. Department of Economic and Social ———. 2011. South Africa Economic Update: Affairs, Population Division. New York: Focus on Savings, Investment and Inclusive United Nations. Growth. July. Washington, DC: World Bank. South Africa Economic Update 7 12 aug 15.indd 52 8/12/15 2:07 PM South Africa Economic Update 7 12 aug 15.indd 53 8/12/15 2:07 PM South Africa Economic Update 7 12 aug 15.indd 54 8/12/15 2:07 PM