Paranoid Androids: Will They Leave Us High and Dry?

It is not only Andrew Yang who expects automation to massively disrupt the labor market, economists are wary, too.

Automation is regarded as a ticking time bomb; robots running AI algorithms taking over most of the jobs. Economists studying automation and the labour market generally agree with this view. According to a 2013 study,[1] almost half of all jobs in the U.S could potentially be automated in the next two decades. But innovation and technological improvement has changed the nature of work continuously for two centuries.

Historically, the loss of employment due to technological innovation has been offset by newer, more productive jobs in a Schumpetarian process of creative destruction. Since the 1950s, technological advances have resulted in the job loss of millions of farmers, factory workers, and railroad workers, as well as the complete elimination of professions like telephone operators, gas pumpers, and elevator attendants. And yet, according to the U.S Bureau of Labour Statistics, outside of recessions, the US employment to population ratio for people ages 25 to 54 years has stabilized at around 80%. This means that workers who lost their jobs were able to find new ones. 

Technological innovation balances the loss of jobs by creating new jobs for the suppliers of the technology or in the new industries spawned from innovation. Often, innovation also results in higher worker productivity saving firms money in labour costs. This enables firms to expand and even add new products resulting in further creation of jobs. Moreover, firms will be able to lower prices to compete, enabling consumers to buy more products, which translates into a further increase in new jobs. This process is the reason for the continuous improvement in living standard seen in our times. According to Heidi Shierholz, a senior economist at the Economic Policy Institute, “Automation does indeed displace workers… but it does not affect the total number of jobs in the economy because of these offsetting effects.”

But history may not be a teacher these days. New jobs being created as a result of technological innovation do not seem to hold in the face of what we now know as the ‘Information Age’ or ‘the Fourth Industrial Revolution’. In 1979, General Motors employed more than 800,000 people and made over $18 billion US dollars. In comparison, Google earned $14 billion US dollars by employing just 58,000 people. Furthermore, as the authors of the book The Second Machine Age point out: “microchip density, processing speed, storage capacity, energy efficiency, download speed and so on—have been improving at exponential rates.” We do not know what innovations will occur in the future, and the disruptive effect it will have on labour markets. [2]

Studies have also suggested the possibility of ‘underemployment’. According to a 2018 working paper on the potential effects of automation on global labor markets, [3] automation may cause a stagnation of wages and polarization of the labor market. While there may still be available employment, they will increasingly be low-paid and volatile without benefits like paid vacation or health insurance. This dynamic has already started to play out in the rise of the Gig Economy, the normalization of freelance work and the swell of the side-hustle culture. 

But does the future of work align with the bleak predictions of automation’s effect on the labour market? We can take some relief in the economic theory called Polanyi’s Paradox that suggests a limit to automation. [4] The paradox refers to the difficulty of automating tasks that cannot be transformed into a set of instructions that are used to tell computers how to perform a given task. This renders creative process out of the sphere of automation. Also jobs with a human element like doctors, teachers and social workers have a very low chance of being automated. 

So the question remains, will automation cause mass unemployment? With the accelerating pace of innovation, one would assume that there would be a corresponding rise in labour productivity. But this is not the case. According to data published by the Conference Board, labour productivity has been declining since 2005 in the U.S, Germany, Japan and UK. Therefore technological innovation does not seem to be making human work irrelevant. But we do not know if future technological innovations will result in massive job loss. 

Our society can take steps to limit the negative impacts of automation while enjoying the benefits of an increase in productivity. Training programs to re-skill workers can be implemented to get them back into the workforce. Educational reform can prepare younger generations for the future of work. An emphasis on education in the burgeoning STEM fields must be encouraged. The acquisition of “soft skills”: creativity, resilience and a commitment to lifelong learning, will equip students with the basic skills necessary to succeed in a volatile job market. Finally, more resources need to be put into research and experiments on bold policies like the Universal Basic Income. Although automation spells dizzying change for the labour market, the future of work doesn’t have to be bleak. 


[1] Carl Benedikt Frey, Michael A. Osborne, “THE FUTURE OF EMPLOYMENT: HOW SUSCEPTIBLE ARE JOBS TO COMPUTERISATION?” (Oxford Martin School, September 17, 2013)

[2] T.L. Andrews, “Robots won’t take your job—they’ll help make room for meaningful work instead”, Available at:

[3] Lukas Schlogl, Andy Sumner, “The Rise of the Robot Reserve Army: Automation and the Future of Economic Development, Work, and Wages in Developing Countries”, 

[4] David H. Autor, “Why Are There Still So Many Jobs?The History and Future of Workplace Automation”, Available at:

Leave a Reply

Your email address will not be published.



Recent Posts

[wpgmza id="1"]
[contact-form-7 id="4" title="Contact form 1"]