Abstract This article reviews the trajectory of employment in South Af rica before and during the Covid-19 pandemic and lockdowns. Thereafter, we propose measures for the alleviation of unemployment through the development of a comparative advantage in services as a response to the environment created by existing policies and multiple crises. The policy is the best response to the consequences of existing measures in place. Threats – in the form of the pandemic, advances in Four th Industrial Revolution (4IR) technologies , online retail, and the July 2021 lootings – present the need for strategic investment and fur ther protections for this sector. By Anelisa Mzungulu and Bhaso Ndzendze | Peer Review 18 T H E T H I N K E R Alleviating Unemployment in South Africa: Harnessing Comparative Advantage in the Ser vices Sector PEER REVIEW 19T H E T H I N K E R Introduction In 1994, the newly democratically elected government of South Africa received an economy that had been systematically geared towards the marginalisation of the majority of the population. That is to say that for nearly half a century, the African population had been deliberately subjected to second-class education, labour laws that stifled their economic and political advancement, a business regulatory framework that criminalised forms of firm ownership, and zoning restrictions that kept the black majority outside the metropolitan that served as the focal point of commerce (Levinsohn, 2007). However, the supply of labour grew virtually exponentially in the aftermath of apartheid. In particular, the post-apartheid economy of South Africa witnessed an unprecedented influx of African women into its labour market. On the whole, due to previous marginalisation, the new entrants were relatively unskilled. Simultaneously, while witnessing new entrants, the South African labour market experienced a negative shock in overall labour demand. In particular, the demand for unskilled labour in the mining and agricultural sectors declined. Additionally, the inclination of the economy towards skilled labour contributed as a technical change towards the decline in demand for unskilled labour (Banerjee et al., 2008). Years later in post-apartheid South Africa, with virtually all restrictions of the apartheid system lifted and erased from the law, unemployment over the same period has about doubled, and the same group that bore the brunt of apartheid now disproportionately bears the brunt of unemployment. There are various reasons for the increase in unemployment in post-apartheid South Africa. The growth of the labour force is highly concentrated in the African population. In an eight-year period, after the fall of apartheid, the African women labour force participation rate rose by 15% and the African men labour force participation rose by 5%. The increases were partly due to the removal of apartheid restrictions that prevented movements into urban areas, signalling new possibilities of employment within these areas for the African population. These perceived employment opportunities and expected returns to employment were held superior in contrast to what was previously opened to non-white groups. Educational levels are relatively higher and participation rates traditionally increase with the improvement in educational levels, particularly for women. The much larger increase in the African women labour force participation rate, relative to the African men labour force participation rate, is as a result of the decline in the access of African women to African men’s income, who had traditionally served as breadwinners. HIV/AIDS has had enormous impact in creating single parent households and increasing incidences of female headship. Additionally, the South African labour market has suffered as a result of not being spared, like other countries, of skill-biased technological change. The consequence of all these factors is increased labour supply and declining labour demand that has led to a decline in wage rates. These wage rates have been found unacceptable by society and, as a result, there has been substantial unemployment. The socio-economic costs of high unemployment in South Africa present a three-pronged threat. Firstly, unemployed South Africans of working age represent potential output that could be earning, if the majority of the working age population was put to use. Secondly, South Africa’s high unemployment captures the dynamism of unemployment. Those who are not acquiring the experience and skills are necessarily unemployed and will be incapable of contributing to their own productivity in the future. Thus, the costs of unemployment are amplified over time. A chief consequence of this is the foregone future growth. Furthermore, the dynamics involved, as will be shown by the data available for unemployment, is such that unemployment is not self-correcting as a problem. Lastly, and hard to measure but very important, is that unemployment accelerates the social ills that engender a sense of hopelessness. On top of the list is crime, disengagement with the political process, and a lack of interest in one’s future wellbeing. Because unemployment is prevalent amongst the youth, their disillusionment with post-apartheid South Africa has implications for the country’s future (Levinsohn, 2007). For all the reasons above and more, unemployment is clamouring for attention as a challenge facing South Africa. This article proposes the alleviation of unemployment through the development of a comparative advantage in the services sector as a response to the environment created by existing policies. The policy is the best response to the consequences of existing policies in place. PEER REVIEW Table 1: Participation, employment, and unemployment rates by year and gender (%) The South African Labour Market Table 1 reflects the labour force participation rate and the unemployment rate over time. Here, and in subsequent tables, the ‘narrow’ definition of unemployment is utilised to define the participation rate and the unemployment rate, but one could have easily used the ‘broad’ definition. The unemployment rate in the South Af rican labour market is high regardless of whether the ‘broad’ or ‘narrow’ definition is used (quotable). The dichotomy between the ‘narrow’ and ‘broad’ definitions of unemployment is that the former does not factor in individuals who have not actively searched for work in the past four weeks, while the latter factors in those who have simply stated that they are available for work, notwithstanding whether or not search activity has been conducted in the past four weeks. In dealing with Table 1, some definitions are useful. Defining those who are employed with N, those who are unemployed with U and those who are economically inactive as NE, the labour force participation rate is given by N + U N + NE + U the employment rate is given by N N + NE + U and the unemployment rate is given by U N + NE + U Table 1 depicts three key points. Firstly, and relative to international standards, the labour force participation rates are low; however, the labour force participation rates have sustained growth over the past six years f rom 2012 to 2017 (before declining for women in 2018 (53.17%) and rebounding to 59.57% in 2019). The labour force participation rate for women has risen f rom 48.3% in 2012 to 53.6% in 2017, a positive change of 5.3% in the six years between 2012 and 2017. The labour force participation rate for men has risen f rom 61.68% in 2012 to 66.13% in 2017, a positive change of 4.45%. This was reversed to 65.62% in 2018 and a slight rebound in 2019 (still below 2017 levels) at 65.8%. Although the increase in the labour force participation rate for women was larger than the labour force participation rate for men over the same period, there are signs that the increases in the labour force participation rate for women are smaller relative to the increases in the labour force participation rate for women experienced after the lifting of apartheid restrictions in the early 90s. The lifting of apartheid restrictions, creating perceived employment opportunities, resulted in an unprecedented influx of relatively unskilled women into the labour market (Banerjee et al., 2008). In the period between 1997 and 2005, the labour force participation rate increased by a larger change of 15% (Levinsohn, 2007). 20 V o l u m e 8 8 / 2 0 2 1 Survey Year Survey Year WOMEN MEN Unemployment Rate Unemployment Rate Participation Rate Participation Rate Employment Rate Employment Rate 2012 2013 2014 2015 2016 2017 2012 2013 2014 2015 2016 2017 48.3 49.58 50.68 52.1 52.25 53.6 61.68 62.18 63.8 65.1 65.35 66.13 34.83 36.13 36.65 37.68 36.98 37.73 47.55 47.85 48.85 49.88 49.2 49.15 27.88 27.2 27.2 27.7 29.15 29.6 22.9 23.08 23.33 23.38 24.73 25.7 (Source: Quarterly Labour Force Survey (2012–2017)) PEER REVIEW Table 2 shows unemployment rates for the youth (ages 15–34 years old) and adults (ages 35–64 years old) according to levels of education. Evidently, adults performed relatively better than the youth at 17.9% unemployment rate compared to the youth’s unemployment rate of 38.8%. Most notably, the unemployment rate for the youth is higher relatively to that of adults, notwithstanding the level of educational attainment. Nevertheless, the unemployment rate is mitigated as the level of education increases. For example, during the Covid-19 pandemic, StatsSA (2020: 10) observes that: [O]f those who continued to receive pay during the lockdown, some had a reduction in their pay/salary during the lockdown. There seems to be some relationship between the level of education and reduction in pay/salary. Those with higher levels of education had higher chances of receiving a full salary than those with lower levels of education in both Q3: 2020 and Q4: 2020. About 9 in every 10 employed graduates (90,2%) continued to receive full salaries, compared to 81,2% of those with less than matric as their highest level of education in Q4: 2020. Table 2 divides the aggregate figures in Table 1 into policy relevant slices, by dividing them into educational level and age cohort. What we can decipher f rom the two is that unemployment is concentrated in individuals with matric or less. For the youth between ages 15 and 34, the unemployment rate for those with matric is 39.1% and 44.1% for those with less than a matric. For adults between the ages of 35 and 64, the unemployment rate for those with matriculation is 16.1% and even higher for those without a matric at 23.4%. Evidently, the completion of school is not sufficient to guarantee entry into the labour market. However, as stated previously, the rate of unemployment is mitigated by educational attainment: unemployment rates decline with a post-matric qualification to 28.8% for the youth and 8.2% for adults and even further for graduates to 11.9% for the youth and 4.4% for adults – a relatively minimal unemployment rate compared to those with matric or less. Secondly, employment rates are quite low, with only 48.02% of men having been employed in the last quarter of 2020 (StatsSA , 2020: 25). Only close to a half of the male labour force is employed and only 37.73% of women work by 2017. According to the narrow definition, the women labour force is still far out from 50% of its labour force being utilised. Thirdly, before the Covid-19 pandemic, unemployment rates continued to rise for both men and women over time, as the figures in Table 2 stress. Unemployment had stubbornly increased and will likely not retreat on its own. In light of this, the requirement is that it should be addressed by government policy. 21T H E T H I N K E R TotalLess Than MatricMatricGraduate Other Tertiary Youth (15-34) Adult (35-64) 39.1 16.1 11.9 4.4 28.8 8.2 44.1 23.4 38.8 17.9 Table 2: Unemployment Rate (%) by age and educational level in 2018. (Source: Quarterly Labour Force Survey Q2, 2018) Evidently, adults performed relatively better than the youth at 17.9% unemployment rate compared to the youth’s unemployment rate of 38.8%. Most notably, the unemployment rate for the youth is higher relatively to that of adults, notwithstanding the level of educational attainment. Nevertheless, the unemployment rate is mitigated as the level of education increases. PEER REVIEW Total Agriculture Mining Manufacturing Utilities Construction Trade Transport Finance and other business services Community and social services Private households 16 100 835 434 1799 148 1395 3265 954 2395 3560 1311 16 288 843 435 1744 161 1476 3219 1014 2399 3692 1296 16 313 842 381 1789 151 1363 3429 983 2495 3622 1251 14 148 799 373 1456 113 1066 2946 885 2234 3244 1005 22 V o l u m e 8 8 / 2 0 2 1 Industry Q2 2017 Q2 2018 Q2 2019 Q2 2020 Table 3: Participation, employment, and unemployment rates by age group (%). Table 3 captures the trajectory of employment and labour force participation. There is a large jump (48.2%) in the labour force rate f rom ages 15–24 (25.60%) to ages 25–34 (73.80%). This captures the results of the 15–24 age cohort leaving school and entering the labour force market. At a staggering 52.4% rate of unemployment, the age cohort between 15–24 years experiences the highest concentration of unemployment. Relative to this age cohort, the unemployment rate declines to 32% for the age cohort of 25–34 years, declines further to 20.9% for the age cohort of 35–44 years, and ultimately experiences modest figures of 16.1% for the age cohort of 45–54 years and 8.9% for the age cohort of 55–64 years respectively. Age Group Unemployment Rate Participation Rate Employment Rate 15–24 25–34 35–44 45–54 55–64 25.60 73.80 80.90 74.00 45.80 12.20 49.40 64.00 62.10 41.80 52.40 33.00 20.90 16.10 8.90 (Source: Quarterly Labour Force Survey, 2018) Table 4: Employment by industries. (Source: Quarterly Labour Force Survey (figures are in the thousands)) Table 4 shows that between the periods of Q2 2017 and Q2 2019, the South Af rican industries experienced the shedding and creation of jobs. There was a net gain of a total of 213,000 jobs across all industries. This was reversed by the onset of Covid-19. In the first months of the ‘hard lockdown’, 2.2 million people lost their jobs (Smit, 2021). The majority of the job losses are concentrated in the trade sector at 483,000, followed by the community and services industry shedding a total of 378,000 jobs. The manufacturing industry closely trails with 330,000 jobs and manufacturing with 297,000 jobs. PEER REVIEW 23T H E T H I N K E R Tables 5 and 6 depict labour market transitions, a moving picture of the labour market, as opposed to the previous tables which were a view of the economy at a point in time. Tables 5 and 6 isolate the school to work transitions, and are therefore only reflective of youth aged 15–34. Table 5 presents transition rates for cohorts aged 15– 34, and Table 6 presents transition rates for those with at least a matric qualification. The rows represent the state of the individual in September 2017, while the columns represent the state of the individual six months later in March 2018. The rows indicate the labour market state of the candidate in September 2017, while the columns indicate the status six months later in March 2018. For example, in Table 5, of those in active search for employment in September 2017, 43.9% were still engaged in search activity six months later, while 15.98% grew discouraged and ceased searching for employment. Only 12.26% landed employment in the formal industry. Together, Table 5 and Table 6 show three key points. Firstly, movement between formal industries and informal industries is quite rare, but being in possession of a matric improves the likelihood. Secondly, once one has landed employment in formal industries, one has the tendency to hold onto work (not the same job, however) in the formal industries. The retention rates are quite high (67.44%) for those aged 18–35, and almost 81.69% for those who have at least a matric. Lastly, there is less likelihood (30.70%) of having a job in the informal industries six months later, which indicates barriers to entry in the informal industries. Table 5: Transition matrix of employment status for all youth. Table 6: Transition matrix of employment status for all youth with at least a matric. Informal Informal Searching Searching Formal Formal Total Total Discouraged Discouraged NEA NEA N N NEA Discouraged Searching Informal Formal Total NEA Discouraged Searching Informal Formal Total 15.30 37.08 19..8 9.68 4.62 15.53 11.10 47.78 15.98 5.85 2.82 11.99 63.87 20.85 19.19 10.71 6.58 27.88 55.21 18.93 19.08 8.56 7.57 19.18 1281796 769545 928749 362967 1361128 4735162 276587 222856 347550 94883 651888 1654954 13.32 28.71 43.97 16.37 8.48 29.29 15.67 35.73 45.82 18.04 6.35 22.38 3.32 5.36 5.63 53.83 4.38 8.88 2.23 4.51 3.84 30.70 5.74 5.97 6.20 10.01 12.26 28.10 67.44 38.22 18.03 8.35 17.28 25.76 81.69 43.46 100 100 100 100 100 100 100 100 100 100 100 100 (Source: September 2017 and March 2018 Quarterly Labour Force Survey panel) (Source: September 2017 and March 2018 Quarterly Labour Force Survey panel) PEER REVIEW 24 T H E T H I N K E R Lessons from India and China Some useful lessons can be learnt f rom the development paths of China and India. In 1978, China and India were two developing countries, experiencing low levels of per capita GDP growth and diametrically opposed political, economic, and social institutions. The real GDP per capita income in 1980 stood at $556 and $917 in China and India respectively (Alm, 2008). The Chinese and Indian economies were characteristic of Gerschenkronian ‘latecomer countries’ experiencing an infinitesimal industrialisation pace, with an inability to extract meaningful value f rom their relative economic backwardness (Saccone, 2009). In the Gerschenkronian sense, the typical situation in a backward country prior to the initiation of considerable industrialisation is the antagonism between the actual state of economic activities and existing obstacles to industrial development, and the great promise inherent in such a development. However, the extent of opportunities intrinsic in industrialisation varies between countries and is predicated on each country’s endowment of natural resources. Furthermore, industrialisation in China and India seemed virtually impossible due to the formidable institutional obstacles that remained (Alm, 2008). China was toiling under a communist yoke and India emphasised labour-intensive agricultural farming (Alm, 2008). Additionally, both countries were lacking in, in the stable sense of the word, a reliable, disciplined industrial labour force that has been detached f rom the land and has become suitable for utilisation in factories or factory-like environments. India, with its legacy of British colonial rule, possessed a large number of English-speaking workers and familiarity with the West. India also offered an ample supply of educated workers, many of them college graduates available at a f raction of what they could earn in the U.S and other advanced economies (Alm, 2008). China was the opposite: in 1978, the Chinese labour force still applied their trade in the agrarian economy, with 70.5% of the labour force and 28.8% of GDP in agriculture, forestry and fishing (Alm, 2008). The primary catalyst for growth in both economies was the retracting of the government in favour of relative f ree-markets, which led to a substantial dependence of export-led growth (Alm, 2008). China and India have espoused different development paths. Development Paths An observation of data reveals the emergence of two different development paths in India and China, as stated above. The Chinese development path, after the loosening of the communist yoke, took on an industrialisation production trajectory f rom 1978 onwards. Achieving notable success in production of input goods and finished goods, a comprehensive list is as reflected in Table 7 below: Table 7: China’s Goods Output Soars. [Note: In 1978, China produced 28,000 ref rigerators, 200 ACs, 400 washing machines and 3,800 colour TVs. Chemical fibre Cloth Paper Plastics Electricity Coal Pig iron Steel Steel products 19901978Types of Goods 2006 Units(Millions) .3 11030.0 4.4 .7 256.6 618.0 34.8 31.8 22.1 1.7 18880.0 13.7 12.3 621.2 1080.0 62.4 66.4 51.5 20.7 59855.0 68.6 26.0 2865.7 2373.0 412.5 419.1 468.9 tons meters tons tons 1000kwh tons tons tons tons PEER REVIEW Additionally, China’s goods output as a proportion of gross domestic product exceeds the average of nations with similar per capita income by about 12 percentage points, serving as evidence of China’s predilection for goods production (Alm, 2008). Furthermore, this proportion of goods output of domestic production is the equivalence of the country’s lag in the world average. Comparatively, China has outstripped India in industrial building-up: by the same measure of goods output as a percentage of domestic production and relatively to the world, India trails the world average by approximately 8 percentage points (Alm, 2008). This is despite India’s own pursuit of goods manufacturing within a predominately services-based development model. It is not true that while China has focused on goods production and India on services, India is morphing into ‘the office of the world’ and China into ‘the factory of the world’ (Saccone, 2009). The growth of Indian industries accelerated, in a relatively inferior and lagging fashion to growth rates that emerged in China. Table 8 below shows a list of such industries. This adds to any conclusion that development paths are not necessarily unique or that they espouse absolute autonomy in their ideas. Development paths may be a healthy fusion of different developmental emphases or trajectories. Irma Adelman, in a paper titled ‘The Fallacies in Development Theory and Their Implications for Policy’ concluded that: Economic development is a highly non-linear, multifaceted, path dependent, dynamic process involving systematically shifting interaction patterns among different aspects of development and therefore requiring predictably changing policies and institutions over time. The Bank (and Fund) must learn to accept this fact of life. They must start delivering a more state-specific, differentiated message to their clients, difficult as it might be. The cookie-cutter approach to policy is likely to be incorrect or irrelevant at least as often as it is right (Adelman, 1999). 25V o l u m e 8 8 / 2 0 2 1 Cement Plate glass Refrigerators Room ACs Washing machines Colour TVs Motor vehicles Microcomputers Integrated circuits Mobile phones 19901978Types of Goods 2006 Units(Millions) 65.2 17.8 0 0 0 0 .1 0 30.4 0 209.7 80.7 68.5 2 6.6 10.3 .5 .1 108.4 0 1236.8 465.7 35.3 68.5 35.6 83.8 7.3 93.4 33575.0 480.1 tons weight (cases) units units units units units units units units (Source: September 2017 and March 2018 Quarterly Labour Force Survey panel) Some useful lessons can be learnt from the development paths of China and India. In 1978, China and India were two developing countries, experiencing low levels of per capita GDP growth and diametrically opposed political, economic, and social institutions. The real GDP per capita income in 1980 stood at $556 and $917 in China and India respectively (Alm, 2008). PEER REVIEW 26 T H E T H I N K E R Table 8: Output index for the manufacturing industry in India (base 1993–4 = 100; weights* industry = 100) Beverages, tobacco and related 12. 38 products 2 Transport equipment and parts 3.98Machinery and equipment 39.57 other than transport equipment 4 Non-metallic mineral products 4.4 5 Other manufacturing industries 2.56 Textile products (including 62.54 wearing apparel) Basic chemicals and chemical 7 products (except products of 14.0 petroleum & coal) 8 Basic metal and alloy industries 7.45 Wool, silk and man-made fibre 92.26 textiles Paper and paper products and 115.73 coal products 10 printing, publishing, and allied 2.65 industries Rubber, plastic petroleum and 115.73 coal products 12 Food products 9.08 Metal products and parts 13 (except machinery and 2.81 equipment) 14 Cotton textiles 5.52 Leather and leather and fur 2006–072004–05Industry and Weight 192.1 194.1 182.5 220.8 142.5 156.1 164.6 1469 146.9 197.8 180.5 137.2 444.5 367.7 357.1 305.8 298.4 285.0 283.4 278.9 268.4 248.6 226.3 185.2 PEER REVIEW 27V o l u m e 8 8 / 2 0 2 1 India, repelled by the strength of China in the production of goods, wisely adopted a services strategy (Alm, 2008). This development path was cushioned by India’s advantage in communications technology, language, and cultural compatibility (Alm, 2008). Despite China’s impressive dominance in the manufacturing of goods, it was particularly overwhelmed by and has lagged behind India’s ability to deliver services. Thus, it was relegated to an inferior position in services, just as India was relegated to an inferior position in the manufacturing of goods (Alm, 2008). India has taken advantage of cheap transaction costs (specifically transport costs) in the advent of the internet and as such is able to deliver cheap services around the world, but predominately to the U.S and Britain (Alm, 2008). In the past decade, India’s services have grown aggressively, shooting up 18 percentage points of all exports to 38 percentage points of all exports. In the process, India has surpassed 30 percentage of all exports of the U.S, the largest providers of services in the global market place (Alm, 2008). As evidence of the fact that development does not unfold in a linear fashion and often takes different development paths: China has experienced a 5% decline in services sales. In China, services went f rom 15 percentage points of all exports to 8 percentage points of all exports (Alm, 2008). The development paths of China and India are also evidence of a Gerschenkronian development process that is characterised by initial endowments that determine the trajectory that a country undertakes in pursuit of escaping economic backwardness. The Economic Argument for the Alleviation of Unemployment through Development of a Comparative Advantage in Services In order to understand what this article is proposing it is important to draw a distinction between the informal and formal economies, which can also be referred to as the formal and informal sectors. The formal sector consists of South Af rica’s heavy industries (for example, the mining sector and the metals sector), in addition to commercial agriculture and financial services. The informal sector differs in the sense that it consists of subsistence agriculture, lighter manufacturing, for example food and clothing, and a substantial share of trade and transport services (Davies, 2009). Some are of the view that the informal sector is an ever-evolving sector with the ability to lead to job creation, and lead to economy-wide growth in output. The distinction between formal and informal activities is that informal sectors are largely unregistered firms. Thus, the consequences of policy are largely varying for the informal sector and formal sector. According to the developmental history of China and India, it is crucial for a country to pick a developmental path suitable for the capabilities of its labour force and economic conditions. South Af rica, institutionally and economically, adopted the f ree market principle of trade liberalisation in the early 1990s. Simultaneously, unemployment and labour force participation rate grew rapidly. Trade liberalisation led to a reduction in tariffs charged on imports into South Af rica. Consequently, this lowered the import prices and led to the improvement of the demand for imported products. A look at the initial tariff rates and collections in 2002 reveals that the highest tariffs stemmed f rom clothing and textiles. Therefore, the textiles industry was the most adversely affected and faced unprecedented import competition which led to the production of textiles and clothing declining by 4.7% (Davies, 2009). That is to say neither the informal nor formal sector were spared the assault brought on by cheap imports. However, the challenge of cheaper substitutes was avoided in the formal sector, for example in processed foods and other manufactured goods. Whereas formal sector production increased in some industries due to access to cheaper imports, production declined for informal producers, who face higher import competition without access to foreign export markets. Thus, trade liberalisation caused employment to fall, despite an overall increase in production. There is great reason to believe that jobs have continued to be lost in this fashion in the manufacturing industry. According to the statistical section above, the majority of the job losses, as has been the trend in recent years, are concentrated in the manufacturing industry, which accounted for 108,000 job losses in 2018. In comparison to 2017, manufacturing persisted in job losses with 55,000 jobs lost. However, these national results hide the different outcomes for the formal and informal sectors. The true consequences have been that informal sector production drops significantly, leading to a virtually equal decline in informal employment. However, the formal sector workers and producers make gains, especially in services as a result of improved access to foreign markets. Consequently, there is a shift away PEER REVIEW T H E T H I N K E R f rom the informal sector to the formal sector, which has contributed to the disproportionate smallness informal sector in South Af rica (Davies, 2009). Thus evidently, the informal sector, as well as the formal sector, can gain f rom specialisation in services in the face of cheaper goods f rom Asian Economies because trade liberalisation significantly reduces informal employment by increasing import competition without providing additional opportunities through the access of foreign markets. The government can create access to foreign markets by specialising in services and promoting a services- oriented economy, ancillary to the natural resources competitive and dependent economy. Whereas the government has opted for trade liberalisation, the best complement to that policy would be a focus on services, following a similar growth path as that of India. The assessment of the South Af rican labour market statistical section found that the South Af rican labour market is strongly dominated by a relatively unskilled labour force of Af rican women and youth. The majority of the men and women have at least a matric qualification. Therefore, the development of the services comparative advantage should be done through a services provider who will utilise divisions of labour such that those involved in services in South Af rica can upskill themselves and progressively gain the necessary skills. However, it will have to be a move away f rom production. It is worth noting the nature of the South Af rican formal sector. The formal sector in South Af rica is characterised by the same qualities of the formal sector of middle-income countries. However, the informal sector has those of a poor, less developed country. Like middle income countries rich in minerals and competitive in capital and labour intensiveness, it simply cannot compete with East Asian economies in low wage, unskilled labour- intensive manufacturing. Above that, any successful labour-intensive manufacturing will conclude in formalisation, working backwards. As shown in the South Af rican labour market statistical section, South Af rica has experienced, since apartheid, an increase in relatively unskilled female labour force and the hope to get these women into employment is stifled by a virtually disappearing low wage unskilled labour- intensive manufacturing that has relocated to Asia. Thus, it becomes paramount to upskill this labour force through services where we can carve out a market for services produced, including those ancillary services to the manufacturing process. The idea, and indeed the only hope, is to upskill the labour force through a creation of a comparative advantage in services where they participate through the divisions of labour. It was also shown in the statistical section that the youth is largely unemployed and encounter barriers to entry in the informal sector but are likely to hold a job (not necessarily the same job) in the formal sector. However, we saw through the developmental path of India that the advantage of the services sector is that there are minimal transportation costs, if at all. The barriers to entry in the informal sector in the form of transportation costs can be relaxed if the informal activity undertaken is the production of services and the government is serving as patron to such informal firms. This can create entrepreneurial tendencies amongst the youth. Thus, a focus on services can go a long way to reducing the barriers of entry in the informal sector and to creating employment opportunities. To be sure, the services sector faces numerous threats. Two have become pronounced over 2020 and 2021: one structural and another contingent but possibly recurrent in the long term. Firstly, the onset of the Fourth Industrial Revolution (4IR), which threatens to bring automation and online retail to the expense of job-creating brick-and-mortar stores, makes informed investment in this sector that much more imperative. Secondly, the looting which took place in July 2021, in the wake of the arrest of former president Jacob Zuma, evinces the need to provide security to the retail sector. References Abhijit Banerjee, S.G., Galiani, S., Levinsohn, J., McLaren, Z., and Woolard, I. (2008). ‘Why has unemployment risen in the New South Af rica?’ National Bureau of Economic Research, Working Paper No. 13167 [online]. Available at: http://www.nber.org/papers/w13167 Adelman, I. (1999). ‘The fallacies in development theory and their implications for policy,’ California Agricultural Experiment Station Working Paper No. 887 [online]. Accessed at: https://mulyanto. staff.uns.ac.id/wp-content/blogs.dir/58/f iles/2009/12/fallacies-in- development-theory.pdf Alm, W.C. (2008). ‘China and India: Two Paths to Economic Power,’ Economic Letter, 3(8): 1–7. Davies, Rob J. T. (2009). ‘Formal-Informal economy linkages and unemployment in South Af rica,’ Policy Research Institute. 28 PEER REVIEW V o l u m e 8 8 / 2 0 2 1 Gerschenkron, A. (1951). Economic Backwardness in Historical Perspective: A Book of Essays. Cambridge, MA: Belknap Press of Harvard University Press. Jun, F. (2010). ‘Explaining economic growth in India and China: An institutional approach,’ China Papers, 1–15. Levisohn, J. (2007). ‘Two Policies to Alleviate Unemployment in South Af rica,’ Ford School of Public Policy. Saccone, V. V. (2009). ‘Structural Change and Economic Development in China and India,’ The European Journal of Comparative Economics, 6(1): 105–106. Shaikh, A. (1974). ‘Laws of Production and Laws of Algebra: The Humbug Production Function,’ The Review of Economics and Statistics, 56(1): 155–120. Smit, S. (2021) (May 8). ‘Covid kills a decade of employment growth in SA,’ Mail & Guardian [online]. Available at: https://mg.co.za/ business/2021-05-08-covid-kills-a-decade-of-employment-growth- in-sa/ StatsSA, (2020). Quarterly Labour Force Survey Quarter 4: 2020. StatsSA [online]. Available at: http://www.statssa.gov.za/publications/ P0211/P02114thQuarter2020.pdf 29 PEER REVIEW