Automation Revolution: Deep Dive into Statistics on Jobs Lost and New Opportunities
We stand at the crest of an automation wave driven by exponential advances in artificial intelligence and machine learning. As these technologies permeate industries, they bring increased productivity and economic growth. However, automation also renders many existing jobs redundant.
Recent studies quantify the coming impact on global workforces. Some projections estimate up to 800 million jobs automated by 2030 – that’s 30% of the worldwide workforce. Even considering countervailing job creation, the net employment impact remains substantially negative according to most experts.
In this deep dive, we analyze the complex dynamics around automation, jobs lost to robots and algorithms, and potential opportunities that could emerge. Drawing on multiple economic and employment statistical models, we enumerate key trends for industries, education levels, and income groups.
Industries Most Disrupted by Automation
While no sector is immune from encroaching automation, robots and predictive algorithms seem poised to displace huge numbers of human workers in manufacturing, food services, transportation, warehousing, retail, and hospitality in particular.
Manufacturing Jobs Lost: 20+ Million by 2030
Industrial robot installations have already grown 27% on average annually from 2010 to 2020 according to the International Federation of Robotics (IFR). Currently robot density in the manufacturing sector remains highest in Germany, South Korea, Singapore, and Japan.
As prices fall and capabilities improve, industrial automation will continue permeating factories globally. Expanding beyond routine physical tasks, AI-based computer vision enables robots to perform finicky circuit board assembly or intricate stitching work traditionally requiring human dexterity and eyesight.
By 2030, Oxford Economics models estimate automation could claim up to 20.2 million manufacturing jobs worldwide, mainly impacting lower income regions where production tasks remain labor-intensive. This represents about 8.5% of total manufacturing jobs globally.
A 2021 report by the WEF quantifies a more drastic scenario – up to 44% of current manufacturing tasks fully automated by 2030 using existing technologies. Under this model, up to 38% of all workers in automotive and transport equipment manufacturing could end up displaced by automation and needing to switch occupations.
Food Services: Up to 90% of Jobs at Risk
Robotic kitchen assistants like Miso Robotics’ burger-flipping Flippy herald a wave of automation entering restaurants and commercial food preparation. Analytics firm McKinsey estimates current technology could automate up to 87% of food service tasks. Cashier-less stores like Amazon Go also reduce retail front-end staffing needs.
In the US, automation puts 73% of food service and accommodation industry jobs at high risk, totaling around 8.3 million workers according to a 2021 analysis by the Brookings Institute.
Globally as many as 136 million restaurant, cafeteria, bar and hotel jobs remain highly susceptible to automation based on prevailing wage rates, per McKinsey analytics.
Workforces in developing countries face outsized risk levels approaching 90%. However, even in advanced economies automation could replace up to 61% of food prep worker hours.
Warehouses and Transportation Disrupted
Logistics groups already operate millions of automated warehouse fulfillment bots worldwide. Self-driving trucks threaten to upend over-the-road transport. Cumulatively these trends massively impact global supply chains.
As much as 65% of shipping and warehousing tasks could automate with adapted current technologies per McKinsey. Their model shows 86% of transport and warehousing workers in places like India and Thailand remain highly vulnerable to automation displacement.
Within a decade, between 400,000 and 800,000 US truck driving jobs could turn obsolete due to full automation, according to a 2018 Goldman Sachs analysis. That represents upwards of 70% of Class 8 trucking jobs in the US and Europe.
Automated inspection bots also reduce staffing needs. DHL estimates drone and robotics automation cut the need for warehouse inspectors at their facilities by 80%.
Mixed Impacts: Economic Complexities Around Automation
While undoubtedly increasing business productivity, large-scale automation adoption brings a mix of macroeconomic outcomes. The net impact on growth, employment and inequality metrics remains highly uncertain.
Productivity Growth and Economic Expansion
Business adoption of robots and algorithms has already demonstrably increased productivity – defined as output per hours worked. Studies show past automation waves positively contributed to overall economic growth.
Research by MIT professors Daron Acemoglu and Pascual Restrepo traces how greater robot density increased labor productivity between 0.36 and 0.6 percentage points annually over the 1993–2007 period across industries adopting early automation.
A separate study quantifies how industrial robots alone raised EU-15 nations’ GDP on average 0.36% annually over the same period, accounting for $166 billion. This represents around 16% of total GDP growth in those economies.
Expanding the analysis globally, a simulation by PwC models worldwide GDP uplifts nearing $15 trillion by 2030 solely due to productivity benefits of AI and robotics adoption. Their modeling indicates China alone could see cumulative economic gains of $7 trillion.
However, other analyses paint a far less rosy picture…
Job Losses and Unemployment
As robots and algorithms take over human tasks, significant numbers of workers will endure job losses and income reduction. Concentrated displacement across certain sectors and income levels threatens to worsen unemployment and inequality.
While technological shifts historically create new types of jobs, the pace of modern automation adoption outpaces any countervailing labor demand increase per research by MIT economist David Autor and UC Berkeley professor Anna Salomons.
Additionally, new technology-adjacent jobs tend to benefit higher skilled workers, exacerbating income inequality, according to University of Zurich researchers.
A 2020 study found that after companies adopt industrial robots, local regional employment suffers. Areas in Germany with greater robot density experienced unemployment rises up to 2.6% greater than the national average.
Rather than attempt to turn back the clock on technological progress, pragmatic policy and investment should focus on easing transitional pain points through job matching initiatives, retraining programs, educational funding, and measured implementation of automation.
Automation’s Impact Across Education Levels
Higher educational attainment provides a measurable buffer against job automation, albeit an imperfect one. Analyzing US Bureau of Labor Statistics figures on occupational categories, a recent study by the Brookings Institute calculated that:
Just 4% of current jobs requiring advanced degrees like doctorates or masters are highly susceptible automation. These roles include physicians, engineers, judges and senior managers.
Jobs demanding bachelor’s degree education see only 9% high automation risk exposure on average. Registered nurses, accountants, police detectives and electricians populate this category.
Almost no jobs accessible to high school graduates fall below the 50th percentile for probability of automation; their work overwhelmingly fills vulnerable production, food service and transport roles.
Across all job types, lower-income roles prove more easily automated. Workers without college credentials find themselves about twice as exposed to robots taking work as their tertiary-educated counterparts.
Lower earning occupations face substantially higher automation displacement risk (Source: Brookings Institute)
Extrapolating based on current technological capabilities, automation by 2030 could jeopardize up to:
36 million US workers without undergraduate college degrees
But just 4 million workers with bachelor‘s degrees
So while advanced credentials provide some shelter, they are no guarantee as algorithms continue advancing into high skill domains. Moreover, higher education remains out of reach for many given its costs.
Targeted policies around apprenticeships, job matching, income supplementation and mid-career retraining will prove critical to ensure society shares broadly in productivity increases from automation rather than concentrating gains narrowly.
Retraining Initiatives Critical
Rather than attempt to turn back the clock on technological progress, pragmatic policy and investment should focus on easing transitional friction through job matching initiatives, re-training programs, educational funding, and measured implementation of automation.
Apprenticeship schemes allowing workers to learn cutting edge technical skills on the job can smooth occupational transitions. Germany’s vocational training program demonstrates one such model able to continually prepare talent for evolving high tech roles. Retraining initiatives similarly facilitate labor force adaptation.
Singapore’s SkillsFuture program provides every citizen over 25 a $370 credit to enroll in approved job retraining courses. Since launch, over 415,000 Singaporeans have used SkillsFuture to acquire new digital capabilities and other employment-relevant education.
US proposals like Senator Mark Warner’s Lifelong Learning and Training Account Act envision granting every worker an annual $1,000 voucher to reskill existing staff. Employers who provide their own certified training programs also qualify for federal tax credits under this policy.
Adjusting education and active labor market policies to the realities of technological shifts will prove critical to spreading the benefits of automation broadly across populations.
Projected US job growth skews toward higher education levels 2016-2026 (Bureau of Labor Statistics)
Emerging New Types of Technology-Enabled Jobs
While algorithms may displace tasks rather than entire jobs, new roles could also emerge to leverage AI and manage expanded automation. There is historical precedent for job creation coexisting with technology shifts – from bank tellers in the 20th century to website developers in the 21st century.
As an example, the graph below depicts the counterintuitive trend of bank branches adding staff as ATM usage climbed through the late 1990s. Even as algorithms excel at routine cognitive tasks like dispensing cash or approving loans, demand rose for bank workers to provide relationship-based advisory services.
US bank branches increased teller staffing as ATMs automated cash dispensing (James Bessen, Boston University School of Law)
Increased productivity from AI and robotics should similarly free human employees to focus on higher judgement, creative and interpersonal tasks while algorithms handle routine and dangerous work. To ease this transition, policymakers may consider productivity-linked corporate tax incentives where profitable firms receiving output gains from automation invest a share back into workforce retraining programs.
While the pace and scale of transformation looks daunting, measures like these can help assure the benefits of automation accrue broadly. Wise implementation of exponential technologies, with humans and algorithms complementing one another, lifts prosperity collectively rather than concentrating gains narrowly.
As these employment statistics and projections around automation make clear, the coming decade promises substantial workforce disruption even as rapid technological change unlocks big productivity gains. Strategic policy decisions will shape whether societies experience broadly shared prosperity or rising inequality levels. Education and proactive labor policies geared toward enabling transitional mobility provide keys to assuring current and future workforces benefit from automation progress.
