How Advances in AI and Automation Will Upend Our Traditional Models of Economic Development
The real challenge is not viewing technological progress solely through the lens of advanced economies
Around the world and within every industry, machines are disrupting our occupational structures and bringing into question the nature of human labor at an alarming rate. Previous limitations surrounding robotics and automation have been overcome, with investment into such technologies on an exponential upward trajectory.
Changying Precision Technology Company in China focuses on the production of mobile phones and increasingly uses automated production lines. A few years ago the factory was run by 650 employees, but now just 60 people are required to complete the job, with robot co-workers responsible for the rest.
In 2015 Nike completed pilot testing on automated stitching, reducing production on a typical size run of its iconic Air Force 1 shoe from over 500 components across multiple factories to just a single machine and operator.
Elsewhere, automated harvesting equipment is revolutionizing traditional agricultural practices, reducing the need for labor-intensive tasks normally performed by dozens of farm-workers.
Whether you’re optimistic or alarmed by these emergent trends, one thing is clear — we aren’t getting off the train of technological progress anytime soon. However, within the current debate over AI and automation, both parties repeatedly commit two injustices.
The first is a tendency to dramatically understate the scale of impending skill set disruption on the near horizon.
It is important to note that, while it has done well to serve as the plot-line for countless sci-fi movies and stoke dystopian fears, the complete substitution of human with robot labor is not what we are witnessing. Indeed, a 2017 Mckinsey & Company report found that the proportion of occupations that can be fully automated by adapting currently demonstrated technology is less than five percent. There are still vital cognitive, social, emotional, and critical thinking abilities innate within our human biology that are currently — and perhaps will remain so for a long-time — unattainable to AI.
But these skills compose only a tiny fraction of the work activities that most people are actually engaged in within their chosen occupation. The other components, whether they be physical tasks within a predictable environment, data processing, or administration, can now be done quicker, more efficiently and at a lower-cost by a machine.
“We are not witnessing the utter replacement of humans with robot labor, but rather the increased rate at which technology is assuming a larger share of people’s work activities.”
Therefore, we are not witnessing the utter replacement of humans with robot labor, but rather the increased rate at which technology is assuming a larger share of people’s work activities. The same Mckinsey & Company report estimates that half of all the activities people are paid to do in the world’s workforce — the equivalent of $15 trillion in wages — could potentially be automated by adapting currently demonstrated technologies.
Undoubtedly, if the history of progress has taught us anything, it is that new jobs in industries that do not currently exist will be created off the back of these advances in AI and automation. Up until now, the labor market has managed to adapt to the replacement of jobs with capital, with price effects tending to balance the forces of automation and creating new complex tasks for people to be paid to do.
But two emergent issues threaten to disrupt this trend. Firstly, in a world where machines can teach themselves, there is a distinct chance that new forms of paid work created in the industries that we cannot now envisage will be assumed by machines.
A key difference between current and past episodes of technological advancement is that some forms of automation — for example, those that are based on machine learning techniques — improve performance over time when these AI have access to more data. Given the above, there is no reason to assume that we won’t be competing against more capable and increasingly sophisticated machines for the jobs of the future.
The second issue is that future employment growth is expected to derive disproportionately from smaller, generally high-skilled sectors and industries that will be unable to absorb job losses coming from other parts of the labor market. According to the World Economic Forum’s 2016 ‘The Future of Jobs Report’, by 2020, more than a third of the desired core skill sets of most occupations will be comprised of technical skills that are not yet considered crucial to the job today.
Emerging job categories such as data analysts, engineering specialists and information systems experts all require a high-degree of science, technology, engineering and mathematics (STEM) related skills. There is already a fast-growing need for consummate technicians and specialists to create and manage advanced automated systems. For example, manufacturing and production sectors are rapidly being transformed into highly sophisticated industries where high-skilled engineers are in strong demand and cheap human labor is rendered increasingly obsolete.
“Too often are the opportunities and challenges posed by technological progress viewed solely through the lens of advanced economies.”
This leads to the other injustice that plagues mainstream discourse over technological progress: neither techno-optimists nor alarmists have anything to say about the impact that technological advancement will have on the significant share of the global workforce that remains employed in the agriculture, manufacturing, and production industries of developing countries.
Too often are the opportunities and challenges posed by technological progress viewed solely through the lens of advanced economies. It must be raised that the vast majority of people in the developing world do not have the skills — or access to the resources to develop those skills — necessary to prosper in the digital era. For example, the training required to become a data scientist entails an advanced quantitative degree in mathematics and statistics, computer science, or engineering. Acquiring such qualifications is no simple task and requires an investment in education and training for a number of years as well as access to adequate higher learning institutes.
Evidently, advanced economies with the capabilities to develop these skills already have a distinct advantage over their developing country counterparts. Ultimately, these trends will upend our traditional models of economic development and jeopardize the future of those people unable to gain the skills necessary to prosper in the digital age.
For the past hundred and fifty years countries have pursued a very particular model of economic development.
As far back as Japan in the late 1800s, a combination of low-wage agriculture and manufacturing — backed by protectionist policies to encourage import substitution and boost exports — was proven to create jobs and build household income. As workers and systems become more productive and households more prosperous, manufacturing moves up the value chain, producing higher quality products. With the construction of transport infrastructure, including roads and railways, rural populations move to cities to join this industrialization wave, creating urban concentrations of consumers with disposable income that helps generate greater prosperity.
It has been this developmental model that has driven a huge influx of 1.2 billion people joining the global labor market between 1980 and 2010, and that has brought millions out of poverty. It is hard to imagine how some of today’s most influential countries — Japan, Germany, China — would have achieved the level of their success if not for some application of this strategy.
In 2018, many developing countries aspire to advance their economies by employing a similar developmental model. Where poor countries with young populations and high birth-rates tend to suffer from skills shortages due to their lack of educated citizens, their advantage lies in access to cheap abundant labor. Traditionally, both agriculture and manufacturing — the two industries needed to get the ball rolling towards industrialization — have accommodated this advantage. The majority of work activities in these roles require limited training and professional oversight, making it possible for developing countries to leverage their labor advantage and grow these industries.
Globalization added a new dimension to this model by making it easier for foreign companies to tap into cheaper labor markets and establish business operations far from home. This in turn brought more opportunities for wealth to be generated within developing economies through the construction of vital infrastructure and employment of significant numbers of the population by foreign companies.
“If a machine can operate as cheaply in Denver USA as in Chennai India, why pay to ship materials and finished goods around the world?”
The rapid rate at which new technologies are being integrated into contemporary work environments is rapidly tearing down this bridge to prosperity. Uptake in the use of robotics has meant that industries are becoming less labor intensive, while the need for highly skilled workers who are closer to where products and services are used is growing.
A 2017 Mckinsey & Company study looked at the automation potential for specific types of activities and jobs within the manufacturing sector. The report estimates that 68 percent of total working hours spent on manufacturing-related activities in the developing world could potentially be automated using currently available technologies.
There is little incentive for companies from advanced economies to continue business operations abroad if they have the ability to fund the capital expenditure that is needed to build highly automated manufacturing plants. If a machine can operate as cheaply in Denver USA as in Chennai India, why pay to ship materials and finished goods around the world?
The drive to automate may ultimately reverse the outflow of manufacturing jobs from rich to poor countries. Such a hypothesis is supported by a survey of business leaders conducted by the Oxford Martin School with 70 percent of respondents believing that developments in automation and 3D printing will encourage companies to move their manufacturing process closer to home. This new dynamic means that low-cost labor can be expected to lose its edge as an essential developmental tool for emerging economies, as automation drives down the cost of manufacturing globally.
Evidently the opportunities generated by the Fourth Industrial Revolution are matched, if not surpassed by, its challenges. On the one hand, the wide-spread implementation of AI and automated systems might mean that advanced economies experiencing population declines or stagnation will be able to maintain living standards even as the labor force wanes. On the other, this leaves low-income countries with high birth-rates stranded at the developmental midway point — yet unable to foster the skills necessary to prosper in the digital era. In the context of global development, the bridge to prosperity is being inadvertently burned down by those who have already made the crossing.
Alex Trauth-Goik is a PhD candidate at the University of Wollongong, Australia, whose research focuses on the development of surveillance systems in China. He strives to offer fresh perspectives on foreign affairs, tech and China (coupled with the odd analysis of human nature).
