The Luddite Fallacy is the belief that automation and technological advancement causes the loss of jobs without seeing beyond the short term truth and acknowledging the wealth and new roles created in the longer term. The premise has always been that automation replaces the jobs of repetitive and onerous nature, freeing up humans to pursue more satisfying vocations and more net jobs are created.
The Luddite movement is named after the allegorical apprentice, Ned Ludd who smashed looms in protest against the loss of jobs in 18th century England. Similar acts were committed during the Swing riots some 20 years later when agricultural workers destroyed threshing machines. Both protest movements were crushed by the Government with executions and penal transportation.
The protesters’ only defender in Parliament was Lord Byron who understood the actions were part of a workers’ movement for employee rights. In a parliamentary speech he decried;
“I have been in some of the most oppressed provinces of Turkey; but never, under the most despotic of infidel governments, did I behold such squalid wretchedness as I have seen since my return, in the very heart of a Christian country.”
In a twist of irony; Lord Byron’s daughter; Ada Lovelace would become the first computer programmer to transcend calculation into computation with her work in combining the Charles Babbage’s Analytical Engine and the Jacquard Loom. Ada’s work was critical to where we are today in the fields of computing, machine learning and AI.
With many protesters sentenced to being transported to Australia; it was amusing to discover I’m living in the spiritual home of the Luddites.
What if it’s different this time?
Entrepreneurs and economists are fond of claiming that since the Industrial Revolution, technological advances have led to an overall increase in the number of jobs and this time will be the same. We have some signs this may not be happening. The rise of part time jobs, freelancing contracts and the gig economy may not be due to a desire for work flexibility but a consequence of full time job loss. Many governments’ claims of low unemployment ignore the underlying challenges of underemployment or involuntary part time employment.
On the surface, the data shows jobs are growing. A 2019 Manpower Group report says that 87% of organisations plan to increase or maintain their headcount as a result of automation. Dun & Bradstreet reported 40% of organisations adding more jobs as a result of deploying AI within their business in a similar study. The consistent caveat to these industry reports is whilst jobs are created overall, AI adoption does cause immediate impact on livelihoods of those replaced and there are limitations in the labour market such as;
- Geographical inflexibility – new jobs created are difficult for displaced workers to reach.
- Skills gap – where displaced workers don’t have the necessary expertise to transition to the newly created roles.
- Demographic inequality – MIT’s Work of the Future report summarises the case for concern with less educated segments and regional workers not receiving the benefits that technology is bringing to the economy. Referring to David Autor’s paper;
- “Even in the wealthiest U.S. cities, the workforce is increasingly bifurcated. On one hand, high-wage professionals enjoy the amenities that thriving urban areas can offer. On the other hand, an underclass of less educated service workers gets by with diminishing purchasing power while attending to the care, comfort, and convenience of the more affluent.”
MIT’s report puts forward a balanced set of cases;
- Optimistic that post war 1940 – 1980 technological advancements led to widespread productivity gains and evenly shared wages growth. The hope is history repeats itself.
- Concerned that post 1980 productivity gains have led to increasingly skewed wages growth towards higher education workers whilst the average wage stagnates.
- Fatalistic that distributed wages growth is not purely dependent on rising productivity as how a country shapes its regulations, market policies and education impacts income distribution. Referring to research from Facundo Alvaredo etal and Stijn Broecke etal;
- “Almost all developed countries have experienced job polarization, widening income distributions, and contraction of traditional manufacturing. But most have done more than the United States to counter these undercurrents by investing in worker skills, strengthening social safety nets where needed, and incentivizing private-sector firms to augment labor rather than simply displace it.”
The World Economic Forum’s (WEF) views in their annual Future of Jobs report is also positive but makes some sobering comments;
- It predicts by 2022, technological advancement driving automation will create 133 million jobs and eliminate 75 million, for a net gain of 58 million.
- The advancements driving change to 2022 are AI adoption, high speed connectivity, cloud infrastructure and big data analytics.
- Availability and cost of skilled talent will drive changes to the distribution of the value chain and decisions on locations.
- Reduction in full-time workforce is expected with more adoption of flexible work practices and remote teams.
- AI’s share of work will expand in the areas of reasoning and decision-making, administration, and managing job-related information tasks.
- At least 54% of all employees will need re-skilling as an imperative. The widening skills gap and lack of coordinated approach to solutions is concerning.
- The report concludes with the reassuring statement that whilst automation is stripping away the repetitive and onerous jobs, new opportunities are opening up that focuses on our uniquely human talents of emotional intelligence, creativity, critical thinking and negotiation.
What’s not addressed is where the speed of technological advancement outpaces the rate of adoption and the ability for workers to retrain to adapt to the shifting needs. Interest and investment in AI has accelerated in recent years as detailed in the AI Index reports. We humans are ill equipped to understand and adapt to the nature of exponential growth. Digitally enabled and created roles trumpeted by proponents of the Luddite Fallacy are themselves vulnerable to being automated. Workers reskilling to take on these roles may find themselves doing so only to train up the next generation of AI or being unable to retrain fast enough to take advantage of the transient opportunities.
Consider the example of autonomous vehicles where drivers find themselves inadvertently building datasets for their AI replacements and testing the business model for ridesharing technology companies planning to replace them. How realistic is it to rely on the average Uber driver being displaced being able to replace their lost livelihood by retraining into jobs created by the autonomous driving sector?
The autonomous driving example is pertinent to the many truck drivers globally too. Whilst they undoubtedly face displacement; employers and employees alike need to attribute the value to the industry beyond the act of driving that humans provide. In the case of transport drivers, loading/unloading, managing inventory, interacting and negotiating with the counterparties and dealing with on road issues in a timely manner all contribute to smooth operations and tasks not likely to be automated in the near future when easier options are present.
The continuing validity of the Luddite fallacy remains to be seen with the data indeed demonstrating jobs growth stemming from AI adoption and automation. Industry reports counsel organizations and authorities to prepare for the transition through investment and policy so as to see rising productivity translate to improvements in worker security, living standards and augment their capabilities rather than perpetuate the trend in rising inequalities. As employers we should be conscious of the ramifications and adopting technologies in a manner aligned with our values.