Technology and Unemployment

By Jordan Borchowsky

Economists, politicians and leaders in technology are now, more than ever, of the view that the era of technological unemployment is near. The opinion shift is primarily due to automation that is able to remove the human factor from the supply chain. This move dampens the compensation effects that allowed previous types of technology to be net job creators. As such, society needs to consider ways to either avoid or manage the possibility of technologically caused mass unemployment. 

Classical Theories of Unemployment Caused by Technology

Classical schools of thought on technologically produced unemployment suggest technology will not cause long term unemployment. Instead, any unemployment is short term and in the long term, job losses are more than offset by five mechanisms known as compensation factors. The first factor operates by way of the production, operation and maintenance of the new technology. If an organisation decides to install a new piece of technology, extra labour needs to be employed to build, operate and maintain this technology¹. 

The second compensation factor is an increase in investment. If new technology increases the profitability of firms, these firms will have incentive to expand operations, creating jobs in the process². Third, if the technology increases the productivity of an organisation, thereby reducing its costs (as would typically be the case), the firm increases its supply of the goods and services it offers. This increases aggregate supply across the economy, reducing the overall price level and improving the purchasing power of households³. The fourth compensation effect is the creation of entire new industries. Technological innovations can, and have seen the birth of new fields of work, creating a plethora of fresh employment opportunities⁴. Fifth and finally, technology can increase the productivity of individual workers, making them more valuable to their employers. As a result, their wages may rise, meaning their spending increases, raising aggregate demand and therefore output in the economy. Because output increases, employment will also have to increase to keep up⁵. All of these compensation factors are said to work together throughout the macroeconomy, preventing a net loss of jobs in the long term. Indeed, this mechanism occurred during the 19^thand 20^thcentury as highlighted by economist Wassily Leontief ⁶. 

Changing Views on Lump of Labour Fallacy

There are now challenges to the classical belief that technology is incapable of causing any long term unemployment. The Lump of Labour Fallacy (an economic term used to refute the notion that there is a fixed number of jobs in an economy) meant that technology could not result in long term unemployment⁷.  However, recent observations of increased output coupled with decreasing employment in certain industries challenge this theory⁸. Stanford University professor of economics, Nick Bloom, cites these kinds of observations as the reason for major opinion changes on technology and unemployment among himself and other economists⁹.  Harvard economics professor and former United States Treasury Secretary Lawrence Summers notes there is “prevailing opinion” among economists that the era of technological unemployment is on the horizon¹⁰. Even former US President Barak Obama has weighed in, presenting his belief that technology is the primary threat to employment saying “The next wave of economic dislocations won’t come from overseas. It will come from the relentless pace of automation that makes a lot of good middle-class jobs obsolete.”¹¹

There are other possible reasons for this recent pessimistic shift regarding the future of work. Even if the Lump of Labour fallacy does hold, advancements in automation allow for the direct replacement of the human worker. During the Industrial Revolution the implementation of technology resulted in a rise in labour demand, due to the large number of people required to operate many of these machines⁵. In contrast, modern day technologies (such as automation) look to directly replace the human factor. Consequently, should such technology proliferate in the future, compensation effects erode, as the human link is removed from the chain¹².Essentially, it is possible to fill jobs created by the compensation effects with technology rather than human labour.Therefore, persistent technological unemployment can become a reality. 

Technology has usually replaced work regarded as ‘easy’¹³. However, as technology advances, the definition of ‘easy work’ broadens. The type of work reserved for humans (anything beyond ‘easy work’) requires more brainpower, training and specialisation than before. This acts as a barrier to those displaced by technology from finding safe, long term employment because of the entry requirements of the ‘harder’ jobs (such as higher levels of retraining to update their skills).  These barriers will also make it more difficult for new entrants to the labour market. Some people will simply lack the natural abilities and training required to meet the higher requirements of the new workforce. 

Incredible technological advancements have made jobs previously believed to be immune to technology now replaceable by automation¹⁴. It was once thought that jobs with a repetitive pattern of predictable behaviour were the ones at risk of automation. Now this risk is not exclusive to these types of jobs. For example, a 2017 study by Oxford University found that the fashion modelling industry has a 98% chance of becoming automated by 2030¹⁵. Professional jobs that many young people are earning qualifications for at the current time, also have a high probability of becoming automated. The same Oxford University study found that the occupations of accounting and auditing have a 94% chance of becoming automated by the year 2030¹⁵.Considering that accounting is the seventh most popular major in the USA, many young people may be dedicating their time to learning skills that may be obsolete in the near future¹⁶. This is not to say that the accounting major does not have any useful transferable skills, but many of the abilities these students are developing may not be as applicable upon graduation. 

All of the above increases the risk of structural unemployment¹⁷. This is where individuals find themselves unemployed because their skills are no longer in demand as the economy changes. In this case, the change is driven by technology.

Ethics Analysis of Technology and Unemployment

Technology has the potential to improve living standards. It happened during the Industrial Revolution. However, the potential for job losses does raise the question of the ethics of technological advancement. Utilitarianism is an ethical theory which states that an action is ethical if it produces more positive outcomes than negative outcomes. The gains that technology bring need to be weighed against the losses that it causes (namely unemployment). If these gains exceed the losses, allowing for the continued advancement of technology is ethical when viewed under utilitarianism lens. 

During the Industrial Revolution, some English textile workers formed a group known as the Luddites. These textile workers were concerned about the proliferation of new weaving machines which they believed would put them out of work. Motivated by this concern, the Luddites broke into factories, destroyed machines, and attacked employers and government officials¹⁸. Even though it is easy to sympathise with Luddite concerns, it also is important to understand what would have happened if they got their way and society gave into the fear of technological unemployment. The fact that we are able to enjoy quality, affordable clothing is as a direct result of innovations to weaving equipment during the Industrial Revolution. Extending the Luddite philosophy further means that the Industrial Revolution would have been halted, as any form of mechanisation would have been seen as a threat to jobs. As a result, the world today would be vastly different. Innovation during the Industrial Revolution is responsible for conveniences we take for granted in modern society. It is unreasonable to suggest the losses endured at that time outweigh the gains that we enjoy today. The technological unemployment endured was short lived and the gains that we enjoy today are undoubtedly worth this cost. Therefore, interfering with technology today does run the risk of causing major harm tomorrow, stopping, or at least pausing the development of society. 

On the other hand, much like a financial institution has a fiduciary duty to act in the best interest of a client, the government of a nation has a duty to act in the best interest of its citizens. Therefore, if there is a large societal transformation such as a radical advancement of technology, government has a role to optimize the change, ensuring to the best of its ability that technology benefits all. 

Take autonomous trucks as an example. In 2016, the self-driving technology company Otto, completed a 212 kilometre journey from Colorado Springs to Fort Collins, Colorado, USA¹⁹.  This was the longest continuous trip ever completed by a self-driving truck. Waymo, Volvo and Daimler have been designing and testing self-driving trucks with successful results²⁰. The era of self-driving trucks is near. Consider that truck driving is the most common occupation in 30 US states, employing 3 million people across the country. The effects automation could have on this industry and the economy as a whole is likely to be severe²¹. Self-driving trucks could make millions of people jobless resulting in broken families and more strain on the welfare system. Accounting for declining compensation effects, the economic risks in allowing the continued development of self-driving technology are high. Yet, there is no way to definitively  know what benefits new technologies will bring and no way of forecasting the costs of regulating or outright banning new technology. This means there is no justifying a broad regulation of technology under the utilitarian framework.

Policy Possibilities: UBI

Some economists propose that instead of trying to prevent the possibility of technological unemployment, society should try and find ways to live with it. One proposed way of managing technological unemployment is a Universal Basic Income (UBI). A UBI is “a periodic cash payment unconditionally delivered to all individuals, without means-test or work requirement” (Basic Income Earth Network)²². Former US Secretary of Labour and political commentator Robert Reich believes the need for a UBI is “almost inevitable”²³. Many leaders in the technology sector, including Facebook founder Mark Zuckerberg have also voiced support for a form of UBI to counter the effects of automation²⁴. Supporters of a UBI agree the rise of automation is likely to result in long term unemployment, poverty and income inequality. 

They also argue it is illogical to create jobs at the expense of automation. Brad Stollery of Mediumhighlights the fact that if job creation is the main goal, why not ban the use of shovels and hire twice the number of construction workers to complete the same task with spoons? Doing so makes no sense, and a parallel can be drawn between this and the banning of future technologies. Instead, supporters of a UBI believe that the original workers should be given their shovels back and the rest should, essentially, be paid to stay at home²⁵. 

One way the ethics of a UBI can be analysed is by using a consequentialist framework. Consequentialism determines an action to be ethical should that action produce more positive than negative outcomes. The primary goal of a UBI as it relates to technology is to cope with the effects of unemployment that could result. If UBI achieves this goal more effectively than current welfare systems, it is likely an ethical policy. There has never been a trial of what can truly be defined as a UBI. For example, Finland’s current Basic Income trial only includes 2,000 unemployed individuals, which means it is not an accurate reflection of how a UBI would operate (as it was not provided on an unconditional basis to a cross-section of Finland’s population)²⁶. However, empirical studies of trials that are similar to a UBI do indicate that it may have positive results. For example, the charity GiveDirectly makes Unconditional Cash Transfers to individuals in Kenya. Villages which received these Unconditional Cash Transfers saw significant rises in consumption and investment expenditure when compared to those that did not receive the Unconditional Cash Transfers. In 2011, a similar trial in India called the Madhya Pradesh Unconditional Cash Transfers Project,  involved giving 6,000 people a basic income.  Those areas that received the basic income saw higher levels of enrolment in schools for 14 to18 year olds, and  a 12% greater increase in incomes.

Even though these studies suggest a UBI may help offset some of the negative impacts that technological unemployment may bring, in a utilitarian calculation, the costs of the program must also be taken into account. To provide a $12,000 payment to all US citizens would cost $3.8 trillion²⁷. Co-chairman of Bridgewater Associates Ray Dalio  highlights the fact that “even the most generous welfare states would struggle to cover the cost of a poverty-line basic income.²⁷” Dalio estimates that if the entire social spending budget of the US, apart from infrastructure and education, isreallocated to a Universal Basic Income program, the money  would only cover 92% of the total cost²⁷. 

Therefore, if the costs of running a UBI program are this high, it is difficult for the benefits of this program to outweigh the costs. At the same time, we can not be sure of this result until a comprehensive study of a UBI is completed and the benefits are apparent. Until such time, it remains unclear from a utilitarian view whether UBI is an ethical policy.

Alleviating Technological Unemployment

Due to the views of economists and technology thought leaders, society may have the benefit of advanced warning with regards to the effects of technology on unemployment. Society needs to come together and formulate a plan about how to deal with technological unemployment because as of now, we have no viable policy. Simply allowing for the unfettered advancement and dissemination of technology runs the risk of large scale unemployment. At the same time, stopping technology may halt the progress of humankind. We do not have any sustainable way of managing large scale unemployment. Even though small scale studies of UBI-like programs do show promising results, the cost of a nationwide UBI is immense. The question of how to manage technological innovations ought to be at the forefront of public debate. Generating more ideas presents viable policy options that will ensure technology does not end up pink slipping us all.

Reference List:

¹ Khan, Mynul. “ Robots Won’t Just Take Jobs, They’ll Create Them.” Tech Crunch, 2016.

²Pettinger, Tejvan. “Investment and Economic Growth.” Economics Help, 6 May 2017, www.economicshelp.org/blog/495/economics/investment-and-economic-growth/.

³Kenton, Will. “Aggregate Supply.” Investopedia, 12 Feb. 2018, 

⁴Eswaran, Vijay. “How Emerging Technology Is Driving Job Creation, New Industries.” Chief Executive, 9 Aug. 2018.

⁵Wallis, W. Allen. “Wages, Productivity and Prices.” Foundation For Economic Education, 1 Mar. 1960.

⁶Leontief, Wassily. “Technological Advance, Economic Growth, and the Distribution of Income.” Population and Development Review, vol. 9, no. 3, 1983, pp. 403–410., doi:10.2307/1973315

⁷Economics A-Z.” The Economist, 2018, www.economist.com/economics-a-to-z/l.

⁸Waters, Richard. “Technology: Rise of the Replicants.” Financial Times, 4 Mar. 2014, www.ft.com/content/dc895d54-a2bf-11e3-9685-00144feab7de.

⁹“A Mighty Contest.” The Economist, 29 Mar. 2014.

¹⁰Summers, Lawrence H. “Lawrence H. Summers on the Economic Challenge of the Future: Jobs.” The Wall Street Journal, 7 July 2014, www.wsj.com/articles/lawrence-h-summers-on-the-economic-challenge-of-the-future-jobs-1404762501.

¹¹Rotman, David. “The Relentless Pace of Automation.” MIT Technology Review, 13 Feb. 2017.

¹²Meyer, David. “Robots May Steal As Many As 800 Million Jobs in the Next 13 Years” Fortune,19 Nov. 2017, http://fortune.com/2017/11/29/robots-automation-replace-jobs-mckinsey-report-800-million/

¹³Chui, Michael, et al. “Where Machines Could Replace Humans—and Where They Can’t (Yet).” McKinsey Quarterly, July 2016, www.mckinsey.com/business-functions/digital-mckinsey/our-insights/where-machines-could-replace-humans-and-where-they-cant-yet.

¹⁴Henry-Nickie, Makada. “AI Should Worry Skilled Workers Too.” Brookings, 8 Nov. 2017, www.brookings.edu/blog/techtank/2017/11/08/ai-should-worry-skilled-knowledge-workers-too/.

¹⁵Scott, Patrick. “These Are the Jobs Most at Risk of Automation According to Oxford University: Is Yours One of Them?” The Telegraph, 27 Sept. 2017, www.telegraph.co.uk/news/2017/09/27/jobs-risk-automation-according-oxford-university-one/.

¹⁶Tetzlaff, Ryan. “Top 10 Most Popular Majors in the U.S.” College Factual,1 Dec. 2016,

¹⁷Kenton, Will. “Structural Unemployment.” Investopedia,22 Mar. 2018,

¹⁸Conniff, Richard. “What the Luddites Really Fought Against.” Smithsonian.com, Mar. 2011, www.smithsonianmag.com/history/what-the-luddites-really-fought-against-264412/.

¹⁹ “Longest Continuous Journey by a Driverless and Autonomous Lorry.” Guinness World Records, 2016, www.guinnessworldrecords.com/world-records/361214-​longest-continuous-journey-by-an-autonomous-vehicle.

²⁰ Freedman, David H. “Self-Driving Trucks Tractor Trailers without a Human at the Wheel Will Soon Barrel onto Highways near You. What Will This Mean for the Nation’s 1.7 Million Truck Drivers?” MIT Technology Review, 2017. http://www.technologyreview.com/s/603493/10-breakthrough-technologies-2017-self-driving-trucks/.

²¹Stewart, Jack. “Robots and Us:Self-Driving Trucks Are Coming to Save Lives and Kill Jobs.” Wired, 5 May 2017, www.wired.com/2017/05/robot-us-self-driving-trucks-coming-save-lives-kill-jobs/.

²²“About Basic Income.” Basic Income Earth Network.

²³Pitts, Harry. “GleninCA, ‘Robert Reich: Universal Basic Income In The US ‘Almost Inevitable‘.’” Basic Income Arts Network, 24 May 2014.

²⁴Lui, Kevin. “Mark Zuckerberg Calls for Universal Basic Income in His Harvard Commencement Speech.” Fortune, 26 May 2017.

²⁵Stollery, Brad. “Universal Basic Income Is an Inevitable Part of Our Automated Future.” Medium, 21 Feb. 2018.

²⁶Henley, Jon. “Finland to End Basic Income Trial after Two Years.” The Guardian , 23 Apr. 2018, www.theguardian.com/world/2018/apr/23/finland-to-end-basic-income-trial-after-two-years.

²⁷ Jaeger Max “Universal Basic Income Would Cost Taxpayers $3.8T per year: Study” New York Post, 12 July 2018.