It was not too long ago that I found myself sitting in a law lecture when the professor walked in and proceeded to address us in a very somber manner. When the noise in the lecture theatre subsided, he addressed several complaints students gave concerning the way we were covering lecture material and how the course was being taught. Students were critical of the fact that the course was very theoretical and scientific in explaining what the law is and how to apply it.To address these concerns, he scrolled through a research paper that centered on a recent study concerning the effects of automation on jobs of the future. After reading the report, he explained to us that if he taught the course any other way, he would essentially be giving us skills that are soon to be automated by the time we graduated. And therefore render the whole course redundant.

A recent study done by the Organisation for Economic Cooperation and Development (OECD) shows that 35% of jobs in South Africa are currently at risk of automation. Included in this list are the jobs of clerks, cashiers, tellers, construction workers, mine workers and maintenance workers. In a study done by Stellenbosch University lecturer Danielle Le Roux, he estimated that that 35% added up to about 4.5 million South Africans. As AI becomes more intelligent and machine learning develops at a rapid rate, it is only logical to conclude that this number of automatable jobs will only increase exponentially over time. A key question thus becomes whether South Africa is ready for the effects that automation will have on its country. Before answering that question, a preliminary one would be: what kind of effects will automation have on South Africa? The answer to this question requires a deep dive into the political, economic and social issues that are unique to South Africa contextually, and that ought to change the way we think about the effects of automation and how to respond.

One such contextual issue in South Africa is its historical legacy of racial injustice. It is no secret that pre-1994, South Africa issued a policy of apartheid that created mass racism, injustice and racial segregation. Though post 1994 apartheid was abolished, the effects of that historical legacy still reverberate across the South African society. I personally was shocked to see the spatial differences when I arrived in Cape Town for the first time. I could not help but notice these stark differences on the road between the airport and the university. There is deep racial segregation between suburbs and townships. With the poor townships being predominantly black or mixed race, and the rich and peaceful suburbs being predominantly white.

Aerial view over Hout Bay and Imizamo Yethu in Cape Town

Part of this racial legacy is that many black and mixed race families were starved of the opportunity to obtain high paying jobs that require high expertise and a high level of education and training. This creates generational problems because families cannot afford to pay the tuition fees required for their children to obtain the education necessary to obtain such jobs. And even if they do, the ghost of student loans haunts them from the second they graduate. The reality for many black South Africans, in terms of employment, is only obtaining jobs that do not require specific skills, training or higher education in order to make enough money to survive and feed their families. As a result, this racial legacy has aided in producing an imbalance in the employment of white and black South Africans. Though controversial, this is an undeniable reality irrespective of where one stands ideologically.

How does this legacy affect our understanding of the effects of automation? As mentioned beforehand, 35% of jobs currently are at risk of automation. Business Tech explains that the occupational group that has the highest probability of becoming automated typically do not require specific skills or training. These include food preparation assistants, assemblers, labourers, refuse workers, cleaners and helpers. In addition to this, the jobs in which machines can perform 75% of the activities include clerks, cashiers, tellers, construction work and mining. In a report published by the World Economic Forum, it explained that the use of robots in production affects lower-skilled workers, and leaves professional workers better off. The unintended consequences of this is that the jobs at most risk are predominantly held by black South Africans. Half of all black South Africa workers are in occupations with an 80% or greater probability of automation; so are 47% of coloured workers. For white employees, however, the proportion is only around 30%.  This will therefore exacerbate the current inequality that exists in South Africa due to the effects of automation on South Africa’s demographics.

South Africa currently is rated one of the most unequal countries in the world. If South Africa fails to prepare for the effects of automation, that inequality gap will arguably only increase. This could potentially exacerbate political and racial tensions in the country. In the future, if unaddressed, we could see massive protests from trade unions against the effects of automation on employment and wages. We are already witnessing this issue become a reality as a protest against banking sector retrenchments has been launched by the trade union Sasbo. The reasons given by the banking sector is that the retrenchments come as a consequence of technological developments that flow from the fourth industrial revolution. The union has demanded that the banks consider other options than retrenchments and begin a program that re-skills employees whose positions are at risk. These retrenchments have been the subject of conversation of massive banks such as ABSA, Standard Bank and Nedbank.

COSATU demonstrators protesting against corruption

Another potential consequence is the increase in xenophobic violence that is already a reality in South Africa and has recently spiked again and has caused hundreds to leave their homes. Xenophobic violence in South Africa largely centres around job insecurity that breeds hatred. If the lack of policy measures in relation to automation continues unaddressed, it is more likely than not that many of those who are privy to the effects of unabated automation will look for someone else to pin the blame on.

A second contextual issue in South Africa is its education system. One of the most contentious issues in South Africa at the moment is access to higher education. This problem has generated mass protests across the country over the last few years. It certainly was the case when I was in my first year at the University of Cape Town. Protests continued for weeks and caused  the university to shut down. This was generally the case across many South African universities at the time. The cause of concern was firstly how expensive tuition fees had become at South African universities. The second cause of concern was student loans that graduates were forced to pay when they finished. Some were forced to find a way to clear their term balance if they wanted to graduate. If that was the case then, it would become even more difficult now to justify the hike in tuition fees if the degree lost its value overtime.

If what students learn during their degree becomes redundant soon after they graduate, then the amount of money they would have paid  would be all for nothing. An obvious way in which the degree might lose its value is if what students were being taught were skills that were easily transferable. It is no longer enough for degree programs to cram information into the heads of students and test them on it. For we live in an age where information is easily accessible, and access will only increase exponentially over time. Degrees may need to be restructured around skills that are not yet easily automatable such as critical generative thinking, programming and problem solving in order to produce the kind of graduates that will be prepared for the future of work. Otherwise, failure to address this issue might lead to future protests decrying the value of the education students are paying for.

A final contextual issue to consider that is unique to South Africa is its political setting. Currently, the ANC is the majority party in South Africa and has been in power since 1994. They have democratically dominated the electoral polls ever since. The two major opposition parties to the ANC are the Democratic Alliance (DA) and the Economic Freedom Fighters (EFF). Arguably, the opposition party that has done most of the heavy-lifting in terms of holding the majority party accountable is the EFF. There was however one particular highlight from the DA which came during the State of the Nation Debate in Parliament when its leader, MP Mmusi Maimane, addressed Parliament with several concerns relating to South Africa’s preparedness for the fourth industrial revolution and automation. Unfortunately, these concerns were quickly dismissed by the ANC. Besides that encounter, the future can only tell of what effects automation will have on South Africa’s political scene. The EFF arguably are far left in their politics.

History shows that they are quick to call policies racist when the occasion arises. If businesses and regulators move towards allowing automation to replace humans, how will the EFF condition the responses of its supporters? Will they dismiss these changes as racist and cause more political divide? Will this offset automation to become an even more contentious issue? If the ANC fails to respond adequately to the demands of the fourth industrial revolution, will the people lose confidence in them even further? The land issue sparked controversial debate among South Africans and forced the ANC to act quickly. Will that be the same case if the rate of automation accelerates? Only time will tell.

What I have argued thus far is that there are major contextual issues that will likely exacerbate the effects of automation in South Africa. However, one crucial point must be made clear in this discussion: the rate at which automation will disrupt South African society as it stands cannot be accurately predicted. Simply because 35% of jobs are automatable does not necessarily mean that they will be automated. Further, factors such as South Africa’s political scene and economic health can equally accelerate or delay the arrival of automation in South Africa. In addition to this, it only takes one Bill for Parliament to decide that employers may only augment and not replace the jobs that employees perform, unless circumstances desperately call for replacement. Currently as it stands, there is a commission that President Ramaphosa set up to tackle the issue of the 4th Industrial Revolution. The potential success of this Commission will be something to watch out for.

What is certain is that going forward, all of these factors must be highlighted and taken into account during discussions and debates around automation. There is a lot of uncertainty surrounding this topic, what is important is that policy formulation recognises that there are issues unique to South Africa that must frame the discussion going forward. Only then, can this issue be effectively dealt with.