How Automation Will Define Workforce Development
By: William Nedds
Analyst, Camoin Associates
Email: william@camoinassociates.com
This article first appeared in Camoin Associates' Economic Development Navigator.
The rise of automation and its potential disruption of the global economy is largely considered to be a given. The growing trend of jobs facing irrelevancy at the hands of hardware integrated with software spans numerous industries, from fast food to finance to manufacturing. In some cases, for example, it is now three times cheaper to buy a robot to perform spot welding than to hire a human. While the human demands $25 an hour plus benefits and breaks, and requires expensive safety equipment, the robot can perform the same task for $64 a day while stopping only for routine maintenance every few weeks.
As such, it is virtually a guarantee that the next few decades will mark a period of deep growing pains as the automation and computerization of increasingly non-routine tasks begins to outpace humanity's natural ability to invent new ways to keep ourselves busy.
For another article in Camoin's latest Navigator newsletter, we looked at what puts a job in jeopardy of automation, and where we will most likely be seeing job losses due to automation in the near future. Based on the data that was studied, we found that the average job has about a 61% chance of being automated at some point in the foreseeable future, and that nearly half of all American jobs have a 75% or higher chance of being automated over that same time horizon. This, of course, represents an unmistakable uncertainty in the ability to ensure job security within the workforce of the future, especially among low-income workers whose occupations tend to be more routine and thus more susceptible to automation and computerization. For more info, on the states where automation may have the heaviest toll, check out our other Navigator article "Featured Indicator: State Occupational Automation Risk."
How to Tell if a Job is Automatable
In 2013 (and again in 2017), Carl Benedikt Frey and Michael A. Osborne attempted to quantify what makes a job automatable, through an Oxford University research paper entitled "The Future of Employment: How Susceptible are Jobs to Computerisation?" For their paper, the authors turned to O*NET for occupational data. O*NET is an online service created by the US Department of Labor which catalogues 5-digit SOC occupations alongside a standardized list of the key skills and abilities needed to perform them. The authors pulled data for over 700 unique occupations, as well as the skills needed for each one.
They then identified certain skills included in the O*NET framework that were bottlenecks to automation. These bottlenecks either require a level of physical or mental flexibility that machines have yet to replicate or they incorporate what could be considered "human elements" like social interaction, emotional support, and creative thought. Of the 67 variables in the O*NET framework that were examined, the three main bottlenecks they identified are:
- Perception and Manipulation: Manual dexterity and gross body coordination. The need for a steady and precise hand in order to operate, or the need to move into awkward or hard-to-reach places. Occupations with high Perception and Manipulation include surgeons, dentists, and archeologists.
- Creative Intelligence: Originality. The ability to think outside the box, understand imaginative works and their composition, and act in a creative capacity. Occupations with high Creative Intelligence include choreographers, fashion designers, and curators.
- Social Intelligence: Service orientation. The ability to act well in a social situation, show empathy or emotional support, and successfully negotiate with or persuade others. Occupations with high Social Intelligence include social workers, psychologists, and athletic trainers.
Based on this, it seems that pretty much any occupation that requires working in unpredictable and non-routine tasks (like painting a fictional landscape using your imagination) or face-to-face contact (like comforting or coaching another person) can generally be considered safe from automation for the time-being, though this may change as programmers become better at designing programs with more nuanced and complex decision-making abilities.
This is where occupations start to break down into two general groups: those that are generally automation-proof given today's tech, and those that have only a matter of time before they succumb to the robot overlords. After analyzing the data, the authors concluded that 47% of all jobs are at a high (70% chance or above) risk of being automated. On the other end of the spectrum, 24% of all jobs have very low (less than 10%) likelihood of being automated. These particularly polarized probabilities form a distinct U-shaped curve, with a stark contrast between the number of jobs that are virtually guaranteed to not be automated, and the rest.
Adapting Workforce Strategies to Automation Trends
Emphasizing the Human Element
Does the looming threat of automation mean that the future of the American workforce is all dentistry, massage therapy, and motivational experts? Not quite, but it means that communities wishing to shore up their workforces against the negative impacts of automation will increasingly need to stay two steps ahead of technological progress. Workforce evolution is only going to pick up in pace. Thanks to automation, strategic plans designed to last 25 years may likely be obsolete within a decade, which will warrant new strategies and considerations. To best prepare for evolving workforce demands, communities and regions need to be both reactive to declines in current trends and proactive to new trends that may be off on the horizon.
What makes technological progress so exciting is also what makes it so potentially devastating to even the best economic development strategies. Shifts in automation hammer home the increasingly important need to boost workforce adaptability to address it. Strong workforce development no longer means planning for current demands and practices; rather, it means planning for the emerging demands and practices that will arise as the American economy adapts and evolves around the jobs that succumb to automation.
Abstract, Adaptive, Agile
The void that will be left by automation is where entrepreneurship, creativity, and critical thinking will thrive. In the Deloitte publication "Agiletown," Osborne and Frey performed a review of 100 London businesses to gauge how these firms have reacted and will continue to adapt to automation trends. The top skills that these businesses saw is increasingly required were digital know-how, management, creativity, entrepreneurship, and problem solving. This is where the foundations for the workforce of the future will rest: not just on the knowledge of how to use the tools at our disposal, but on the ability to step back and look at these tools from new, creative, pragmatic angles.
This represents a strange paradox: the modern workforce must be prepared for two equally divergent and equally imminent realities: that future labor demand will favor increasingly specialized positions, and that accelerating automation warrants training in increasingly high-level, malleable, and abstract skills that can easily be repurposed whenever new technology disrupts the status quo.
And flexibility does not apply only to skills, but to almost every aspect of a worker's career. The authors of "Agiletown" say it themselves: "portfolio careers, whereby workers switch jobs on a regular basis rather than build a career in one area, are likely to become increasingly common. Research already shows an increasing trend in the number of times that Gen Y change jobs in the first five to ten years of their working lives." The flexibility and ease of project startup and completion, brought about by advances in automation and communication and set free by a heavier reliance on more high-level creative work, represents a complete decoupling of the worker from the anchors that typically accompany employment. This ease of movement will make attracting workers even more of a moving target, and further exacerbate the need for flexible and proactive workforce strategies.
Consequently, the most effective tool in a community's workforce development tool belt is improving and maintaining a forward-thinking educational and skills pipeline for workers, and ensuring that the industries it attracts are the ones that can take full advantage of the skills these workers have acquired. Many occupations rely on skills that are both automatable and not automatable, and placing greater emphasis on skills that will survive automation (like the three listed above) as well as skills that are cross-compatible across industries, will smooth any industry transitions if and when they occur.
The finance industry is an excellent example of this. A report by Mckinsey estimates that roughly 50 percent of work in the finance industry, from data entry to verification to processing, can be automated. But finance professionals also rely heavily on customer interaction and socialization to create tailored financial packages for their clients, a skill which is difficult to replace. The push of automation would therefore result in many finance positions relying more on client interaction as a percent of the total work completed. As a result, a worker who is excellent at data entry and poor at customer interaction may be more ill-suited to their job in 2025 than in 2018. An educational program that trains workers for finance jobs, logically, might be poorly preparing its pupils by placing too great a focus on data entry and too little a focus on customer interaction and problem-solving. Optimally, the material provided to students should be either specialized enough to fill a niche unreachable by automation, or abstract enough to allow students to mold their knowledge to new innovations and concepts. As automation becomes more pervasive, both the focus and quality of educational pipelines will become paramount to keeping flesh-and-blood workers competitive within a given community's economy.
A Match Made in Heaven
It is important to mention that, on the whole, automation represents an opportunity for never-before-seen progress in value and wealth creation. This article is by no means an attempt to denounce or detract from what automation has to offer, but aims moreso at ensuring that the human element of the economy maintains a strong position of relevance and usefulness and importance in whatever the economy will become 10, 20, or 50 years from now. Workers are an integral part of ensuring economic health and stability, and ensuring they have a place at the table is one of, if not the most integral part of economic development and workforce strategy.
Sources
Sirkin, Hal (2015). "The Robotic Revolution: The Next Great Leap in Manufacturing" Retrieved from https://www.bcg.com/publications/2015/lean-manufacturing-innovation-robotics-revolution-next-great-leap-manufacturing.aspx
Frey and Osborne (2017). "The Future of Employment: How Susceptible are Jobs to Computerization?" Retrieved from https://www.oxfordmartin.ox.ac.uk/downloads/academic/The_Future_of_Employment.pdf DOI: https://doi.org/10.1016/j.techfore.2016.08.019
Cutler et al. (2014). "Agiletown: the relentless march of technology and London's response" Retrieved from www2.deloitte.com/uk/en/pages/growth/articles/agiletown-the-relentless-march-of-technology-and-londons-response.html
Chui and Manyika (2016). "Where machines could replace humans—and where they can't (yet)" Retrieved from https://www.mckinsey.com/business-functions/digital-mckinsey/our-insights/where-machines-could-replace-humans-and-where-they-cant-yet