By 2030, there will be 539,200 fewer office and administrative support jobs than there were in 2020, a decline about the size of Atlanta's population. 

There will be 336,000 fewer cashier positions in the United States in 2030 than there were in 2020. Over 250,000 secretary and administrative assistant jobs are expected to become obsolete. About 85,000 fewer fabricators and assemblers in the manufacturing industry will be employed in 2030 compared to 2020. 

That's according to the Bureau of Labor Statistics' most recent employment projections. The government agency reasons that automation or other efficiency gains through technology will contribute to employment declines in 19 out of the 30 occupations it estimates will have the largest decline in employment over the next decade. 

Read on to find out which occupations are most at risk of job loss due to automation and other structural changes in the economy. 

Key findings

  • Automation and other technologies are projected to play a role in reducing the number of jobs in 19 of the 30 occupations BLS estimates will see the largest decline in employment by 2030. 
  • The top three occupations with the largest projected declines in employment by 2030 are as follows:
    • Cashiers, with 336,400 fewer jobs.
    • Secretaries and administrative assistants (except those in legal, medical, and executive fields), with 156,900 fewer jobs.
    • Executive secretaries and executive administrative assistants, with 100,600 fewer jobs. 
  • The occupations that will experience the fastest declines in employment by 2030 are as follows:
    • Word processors and typists, with a 16.3% decline in jobs.
    • Parking enforcement workers, with a 2.8% decline. 
    • Nuclear power reactor operators, with a 1.8% decline. 
  • Three major occupational groups are projected to see declines in employment by 2030: 
    • Office and administrative support occupations, with 539,200 fewer jobs.
    • Sales and related occupations, with 202,900 fewer jobs. 
    • Production occupations, with 39,000 fewer jobs. 
  • Eight of the 30 occupations BLS projects to experience the largest decline in employment by 2030 pay below-average wages. 
  • 26 of the 30 occupations BLS estimated to have the largest decline in employment by 2030 are in office and administrative support, sales, and production.
  • Workers in occupations most at risk of automation are more likely to make less money and have less education than workers in occupations at less risk of automation, according to the Government Accountability Office. 

These are the jobs that will experience the largest decline in employment over the next decade

Three major occupational groups are projected to see declines in employment, per the BLS: sales and related occupations, office and administrative support occupations, and production occupations, which includes manufacturing jobs. 

Office and administrative support occupations are estimated to see the largest decline in employment with over half a million jobs becoming obsolete by 2030. 

Occupational group

Employment change, 2020-30

Percent employment change, 2020-30

Median annual wage, 2019-2020

Sales and related occupations

-202,900

-1.40%

$31,500

Office and administrative support occupations

-539,200

-2.80%

$38,720

Production occupations

-39,000

-0.40%

$37,440

Data source: Bureau of Labor Statistics (2021).

According to the BLS, these are the 10 occupations that will undergo the largest decline in employment -- caused at least in part by automation or technology -- by 2030. Occupations in bold make less than the national median on average.

  • Cashiers (-336,400 positions)
  • Secretaries and administrative assistants, except legal, medical, and executive (-156,900 positions)
  • Executive secretaries and executive administrative assistants (-100,600 positions)
  • Miscellaneous assemblers and fabricators (-84,600 positions)
  • Tellers (-73,100 positions)
  • Inspectors, testers, sorters, samplers, and weighers (-68,100 positions)
  • Office clerks, general (-60,400 positions)
  • Bookkeeping, accounting, and auditing clerks (-48,100 positions)
  • Shipping, receiving, and inventory clerks (-40,600 positions)
  • Data entry keyers (-35,600 positions)

See the tables below for more details, including more occupations, occupational categories, median wages, and other data.

These are the occupations that will see the fastest declines over the next decade -- again, where automation and technology are one of the causes. Occupations in bold make less than the national median wage.

  • Parking enforcement workers (-35% employment)
  • Nuclear power reactor operators (-32.9% employment)
  • Telephone operators (-25.4% employment)
  • Switchboard operators, including answering service (-22.7% employment)
  • Data entry keyers (-22.5% employment)
  • Legal secretaries and administrative assistants (-21% employment)
  • Executive secretaries and executive administrative assistants (-18.7% employment)
  • Order clerks (-18.2% employment)
  • Timing device assemblers and adjusters (-17.8% employment)
  • Print binding and finishing workers (-17.5% employment)

Automation and technology will contribute to lower employment in 19 of the 30 jobs projected to have the largest decline in employment

Automation and technology that makes workers more productive are the primary reasons cited by the BLS for job losses in 19 of the 30 occupations projected to have the largest decline in employment by 2030 and 16 of the 30 occupations projected to have the fastest decline in employment by 2030. 

Given that occupations that feature repetitive and predictable tasks are most easily automated, it's logical that occupations in office and administrative support, like secretaries and clerks, data entry keyers, and telephone operators are estimated to see the largest net decline in employment and some of the fastest declines in employment over the next decade. 

Cashiers are projected to have the largest decline in employment by 2030 of any category. BLS estimates that some 336,000 cashier jobs will no longer be necessary by 2030 due to a combination of technology and automation in the form of self-checkout and online commerce. 

A number of production occupations are also expected to see large and fast declines due to improvements in robotics, 3D printing, the Internet of Things, and other related technologies. 

Workers in jobs most at risk of automation are more likely to make less money and have less education than workers in jobs not at risk of automation

While automation presents risks to workers across the economy, those that make less money and have less education tend to hold occupations that are more vulnerable to automation, according to the Government Accountability Office. 

About 60% of workers in jobs susceptible to automation hold a high school diploma or less education, according to the GAO. Meanwhile, almost 90% of workers that hold a graduate degree and nearly 75% of workers that have a bachelor's degree are in occupations not exposed to automation.

The median hourly wage for workers in occupations susceptible to automation is $14.26 while those in roles not susceptible to automation make $22.06 per hour, the GAO found. 19 of the 30 occupations that BLS projects to see the largest employment declines by 2030 provided wages below the national average in 2020. 

Characteristics of workers in jobs susceptible to automation and tech jobs, 2016

Workers in jobs susceptible to automation

Workers in jobs not susceptible to automation

Number of workers

58,700,000

77,700,000

Median age

39

41

Mean hourly wage

$17.37

$26.94

Median hourly wage

$14.26

$22.06

Total workers

43.00%

57.00%

Male

44.10%

55.90%

Female

41.90%

58.10%

Asian, non-Hispanic

35.90%

64.10%

Black, non-Hispanic

46.40%

53.60%

Hispanic

54.10%

45.90%

White, non-Hispanic

40.00%

60.00%

Other, non-Hispanic

45.20%

54.80%

Graduate degree

11.30%

88.70%

Bachelor's degree

26.90%

73.10%

Some college, but less than a BA

46.70%

53.30%

High school degree or less

60.70%

39.30%

Data source: Government Accountability Office (2019.)

Unfortunately, low-wage workers with relatively little formal education are the least likely to be able to afford retraining or upskilling they may need to change careers. Employers may be less willing to upskill lower-wage, lower-skill workers because doing so would require more resources than upskilling a higher-wage, higher-skill worker. 

It's not just low-wage workers: 42,000 supervisor roles are expected to be obsolete by 2030

Low-wage jobs are not the only ones at risk of automation. The Brookings Institution examined the overlap between patents for AI technologies and job descriptions and found that certain roles most commonly occupied by men with bachelor's degrees are most exposed to automation, along with production jobs. The latter category includes managers, supervisors, and analysts, according to Brookings. 

The BLS' projections allude to these roles being at risk to automation. BLS projects almost 23,000 fewer office worker supervisors and almost 19,000 fewer supervisors of non-retail sales workers by 2030, for example. As automation and other technology require fewer workers in the roles those supervisors would manage, a decline in supervisors would logically follow. 

The Brookings Institution also found that women are less exposed to automation because they have tended toward occupations that involve more interpersonal skills, such as education and healthcare. 

A separate Brookings report found that women have historically adapted better to automation than men despite previously being more exposed to automation. Women are also trending toward surpassing men in educational attainment and as a result may be better positioned to cope with and take advantage of automation. 

Robots and AI aren't all doom and gloom

Job automation is a double-edged sword. On one side, automation of labor can render certain jobs obsolete and threaten the livelihoods of economically vulnerable individuals. Automation is also a path to productivity gains, lower prices, and new industries. 

It's unlikely that automation will erase entire occupations. Workers in any occupation carry out a range of tasks, not all of which can be automated. And there's certainly a debate about which jobs are most threatened by automation and how quickly technology that allows automation will be developed and adopted.

Still, certain tasks can be more easily automated, such as those that are repetitive and don't rely too much on human-to-human interaction. Those tasks make up a larger amount of work in certain occupations. As a result, some occupations are more at risk of job loss to automation than others.

Notably, over half of those 30 occupations provide below-average wages. In other words, hundreds of thousands of economically vulnerable workers are projected to lose their jobs to automation, and at the same time they are less likely to have the financial means to independently retrain or upskill in an effort to switch careers. This is a vexing problem, the solutions to which are hotly debated. 

But it's not all doom and gloom. Despite the rapid pace of technological change and the COVID-19 pandemic, the Bureau of Labor Statistics projects that total U.S. employment will grow from 153.5 million to 165.4 million jobs from 2020 to 2030. 

And if those in lower-wage occupations that are most exposed to automation are given support to learn new skills and retrain, automation -- and a smart response to it -- can be a force that elevates the livelihoods and competitiveness of otherwise at-risk workers.

Full employment decline tables

Largest employment declines predicted by 2030

Occupation

Occupational category

Employment change, 2020–30

Percent employment change, 2020–30

Median annual wage, 2020

Automation/technology a cause for employment decline?

Below median annual wage?

Cashiers

Sales and related occupations

-336,400

-10.00%

$25,020

Yes

Yes

Secretaries and administrative assistants, except legal, medical, and executive

Office and administrative support

-156,900

-7.60%

$38,850

Yes

Yes

Executive secretaries and executive administrative assistants

Office and administrative support

-100,600

-18.70%

$63,110

Yes

No

First-line supervisors of retail sales workers

Sales and related occupations

-90,200

-6.50%

$41,580

N/A

Yes

Miscellaneous assemblers and fabricators

Production

-84,600

-6.70%

$33,550

Yes

Yes

Tellers

Office and administrative support

-73,100

-16.90%

$32,620

Yes

Yes

Inspectors, testers, sorters, samplers, and weighers

Production

-68,100

-12.20%

$40,460

Yes

Yes

Office clerks, general

Office and administrative support

-60,400

-2.10%

$35,330

Yes

Yes

Bookkeeping, accounting, and auditing clerks

Office and administrative support

-48,100

-3.00%

$42,410

Yes

No

Shipping, receiving, and inventory clerks

Office and administrative support

-40,600

-5.50%

$35,260

Yes

Yes

Retail salespersons

Sales and related occupations

-39,300

-1.00%

$27,080

No

Yes

Data entry keyers

Office and administrative support

-35,600

-22.50%

$34,440

Yes

Yes

Customer service representatives

Office and administrative support

-34,500

-1.20%

$35,830

Yes

Yes

Legal secretaries and administrative assistants

Office and administrative support

-33,600

-21.00%

$48,980

Yes

No

Correctional officers and jailers

Protective service

-30,000

-7.20%

$47,410

No

No

Postal service mail carriers

Office and administrative support

-24,700

-7.80%

$51,080

N/A

No

Order clerks

Office and administrative support

-24,400

-18.20%

$35,590

Yes

Yes

Buyers and purchasing agents

Business and financial operations

-23,600

-5.40%

$66,690

Yes

No

First-line supervisors of office and administrative support workers

Office and administrative support

-22,900

-1.50%

$58,450

Yes

No

Telemarketers

Sales and related occupations

-21,900

-18.30%

$27,920

Yes

Yes

Farmworkers, farm, ranch, and aquacultural animals

Farming, fishing, and forestry

-21,300

-8.30%

$29,130

Yes

Yes

Bill and account collectors

Office and administrative support

-20,500

-9.20%

$38,100

Yes

Yes

First-line supervisors of non-retail sales workers

Sales and related occupations

-18,800

-5.00%

$78,560

N/A

No

Sewing machine operators

Production

-18,600

-14.00%

$28,230

N/A

Yes

Computer programmers

Computer and mathematical occupations

-18,300

-9.90%

$89,190

No

No

Payroll and timekeeping clerks

Office and administrative support

-17,700

-12.90%

$47,020

Yes

No

Chief executives

Management occupations

-16,800

-5.70%

$185,950

No

No

Word processors and typists

Office and administrative support

-16,300

-36.00%

$41,050

No

Yes

Printing press operators

Production

-15,900

-9.90%

$37,880

N/A

Yes

Switchboard operators, including answering service

Office and administrative support

-13,600

-22.70%

$31,430

Yes

Yes

Data source: Bureau of Labor Statistics (2021).

Fastest employment declines predicted by 2030

Occupation

Occupational category

Employment change, 2020–30

Percent employment change, 2020–30

Median annual wage, 2020

Automation/technology a cause for employment decline?

Below median annual wage?

Word processors and typists

Office and administrative support

-16,300

-36.00%

$41,050

No

Yes

Parking enforcement workers

Protective service occupations

-2,800

-35.00%

$42,070

Yes

No

Nuclear power reactor operators

Production

-1,800

-32.90%

$104,040

Yes

No

Cutters and trimmers, hand

Production

-2,400

-29.70%

$31,630

No

Yes

Telephone operators

Office and administrative support

-1,200

-25.40%

$37,710

Yes

Yes

Watch and clock repairers

Installation, maintenance, and repair occupations

-700

-24.90%

$45,290

No

No

Door-to-door sales workers, news and street vendors, and related workers

Sales and related occupations

-13,000

-24.10%

$29,730

N/A

Yes

Switchboard operators, including answering service

Office and administrative support

-13,600

-22.70%

$31,430

Yes

Yes

Data entry keyers

Office and administrative support

-35,600

-22.50%

$34,440

Yes

Yes

Shoe machine operators and tenders

Production

-1,100

-21.60%

$30,630

N/A

Yes

Legal secretaries and administrative assistants

Office and administrative support

-33,600

-21.00%

$48,980

Yes

No

Floral designers

Arts, design, entertainment, sports, and media

-8,500

-20.10%

$29,140

No

Yes

Executive secretaries and executive administrative assistants

Office and administrative support

-100,600

-18.70%

$63,110

Yes

No

Manufactured building and mobile home installers

Installation, maintenance, and repair occupations

-600

-18.40%

$35,120

No

Yes

Telemarketers

Sales and related occupations

-21,900

-18.30%

$27,920

No

Yes

Order clerks

Office and administrative support

-24,400

-18.20%

$35,590

Yes

Yes

Timing device assemblers and adjusters

Production

-200

-17.80%

$36,170

Yes

Yes

Print binding and finishing workers

Production

-7,300

-17.50%

$34,260

Yes

Yes

Prepress technicians and workers

Production

-4,800

-17.10%

$41,410

Yes

Yes

Tellers

Office and administrative support

-73,100

-16.90%

$32,620

Yes

Yes

New accounts clerks

Office and administrative support

-7,700

-16.70%

$37,750

Yes

Yes

Electronic equipment installers and repairers, motor vehicles

Installation, maintenance, and repair

-1,600

-16.20%

$39,570

No

Yes

Aircraft structure, surfaces, rigging, and systems assemblers

Production

-6,000

-15.90%

$53,160

Yes

No

Drilling and boring machine tool setters, operators, and tenders, metal and plastic

Production

-1,400

-15.50%

$39,780

Yes

Yes

Milling and planing machine setters, operators, and tenders, metal and plastic

Production

-2,400

-15.10%

$43,150

Yes

No

Textile knitting and weaving machine setters, operators, and tenders

Production

-2,900

-14.80%

$31,090

N/A

Yes

Pharmacy aides

Healthcare support

-5,700

-14.70%

$29,280

No

Yes

Refractory materials repairers, except brickmasons

Installation, maintenance, and repair

-100

-14.50%

$54,610

Yes

No

Shoe and leather workers and repairers

Production

-1,200

-14.00%

$30,550

N/A

Yes

Sewing machine operators

Production

-18,600

-14.00%

$28,230

N/A

Yes

Data source: Bureau of Labor Statistics (2021).

Outside experts weigh in

 

Jason D. Schloetzer

Jason D. Schloetzer, Associate Professor of Business Administration, Georgetown University

Jason D. Schloetzer, Associate Professor of Business Administration, Georgetown University

 

What can Americans do (or what skills can they learn) to have successful careers in an economy that is more and more driven by automation?

The automation of work has been around since before the Industrial Revolution. We no longer think about how many fewer farmers we have today than 100 years ago. We no longer wring our hands when we see robots on factory floors doing jobs that, 50 years ago, would have been done by highly skilled people. And, today, we are no longer surprised when a chatbot answers our questions instead of a customer service agent or when an algorithm reviews our resume instead of a human resources manager. It is simply impossible to fully plan for the changes that this new wave of automation will create.

One approach is to develop skills that require a human in the loop: discernment, judgement, empathy, intuition, creativity, ethics, reasoning, and the ability to understand complex interactions among people, to name a few. These are the skills that I'm helping my students -- and my 11-year-old son -- develop rather than, for instance, rote memorization of facts that Google Home, Siri, or Alexa already know the answer to. 

Are there certain occupations or industries that Americans should try to avoid due to growing automation?

This is difficult to forecast given how quickly technologies evolve and the uncertainties inherent in the public policy arena to intervene with regulating tech or preventing larger-scale job losses. I'd say nearly all occupations and industries will continue to be affected by automation. There is a view that automation only affects blue-collar jobs, but that certainly isn't the case, with front-office, middle-office and back-office white-collar corporate jobs already facing automation pressures. Algorithms are increasingly automating roles in the medical profession, the legal profession, and other non-corporate positions. There is little way to avoid automation; as one tech founder recently said to me, "Very few islands will remain as the AI sea rises."

In your opinion, what are some of the under-discussed challenges that automation will create? On the other hand, what are some of the under-discussed benefits?

In terms of under-discussed challenges, I don't see enough attention given to the public policy issues that automation will create for society. Eventually, a few companies or uniquely positioned individuals will behave badly and society will need to rethink regulation and develop new rules. And eventually, people will become dissatisfied with their working environment or dwindling job prospects and public policy will need to address the human aspect of automation.

In terms of under-discussed benefits, I don't see enough creativity in the discussion around how automation could change the way we live and work. There is so much emphasis on enhancing productivity, reducing costs, reducing prices and the like, but there is less attention given to thinking about the types of new jobs that could be created, the new services that we could experience, or the potential for improvements in healthcare, education, and scientific discovery. We should hold onto hope that the continued rise of automation does not diminish the importance of humanity.

 

Ravin Jesuthasan

Ravin Jesuthasan, Global Transformation Leader, Mercer

Ravin Jesuthasan, Global Transformation Leader, Mercer

 

What can Americans do (or what skills can they learn) to have successful careers in an economy that is more and more driven by automation? 

The COVID crisis has accelerated the adoption of digital technologies like never before and it’s important for US employees to understand that a lot of the work they’re doing now will be substituted, augmented, and transformed by automation. However, that’s not to say everyone will lose their jobs -- the traditional narrative that automation replaces jobs has been proven to be incorrect. The tasks that will be substituted by automation are typically those that are repetitive, rules-based, and increasingly part of white-collar jobs. However, this will create demand for work requiring judgement, creativity, and empathy (i.e., the things that make us truly human).

How can companies better support employees at risk of losing their job to automation?

Companies should first establish a mindset that work and jobs are constantly evolving. Knowing that, they should give workers the tools, resources and incentives to reinvent their own jobs. Finally, to ensure that each of their employees is on a path to grow their career, they will need to evaluate their skills and provide resources to close the identified skill gaps. This could mean upskilling for their reinvented job, reskilling for another job within the company, or outskilling for a job with a different company.

Do you believe there are positive aspects of automation that are often overlooked?

Automation is ideally suited for replacing the dirty, dull, or dangerous work that humans do. It can help eliminate the risks of losing life and limb while also freeing us up to do the things we enjoy most. It can also change the supply side of the talent equation by bringing more talent into the game (e.g., when the work of a driller on an oil rig is moved to a digital control center in a major city; broader, diverse pools of talent can now have access to the work).

What are some of the under-discussed challenges that automation will create?

Automation doesn’t destroy jobs. It differentially affects tasks. It substitutes, augments, and creates new human work at the task level. Our key to achieving the optimal combinations of humans and machines is to ensure we are continuously deconstructing jobs, identifying how automation will affect the component tasks and reconstructing new jobs where humans have a comparative advantage.

 

Mark Muro

Mark Muro, Senior Fellow and Policy Director, Brookings Institution

Mark Muro, Senior Fellow and Policy Director, Brookings Institution

 

What can Americans do (or what skills can they learn) to have successful careers in an economy that is more and more driven by automation?

Automation is most effective at substituting for repetitive, simple physical work that lacks “human” touch. Think taking fast food orders and providing the food. Or think of pushing simple forms in an office. Such things can be done easily by robots and rule-based software.

That said, humans should try to stay away from simple, repetitive, routine work. If a job seems boring and repetitive, workers should worry about its longevity.

Meanwhile, workers should seek work that has a creative, face-to-face, human element. Personalized work, team work, and creative work will last. Humans should get better at being human! High-education work like business management as well as human-service jobs like social services and education may do well.

Just stay away from those rote, menial, jobs.

Are there certain occupations or industries that Americans should try to avoid due to growing automation?

Routine factory work and food service face absolutely the highest threats from automation. Fulfillment -- warehouse work -- could be threatened. And in fact, we know automation has accelerated in the pandemic and is hitting foodservice hard.

In your opinion, what are some of the under-discussed challenges that automation will create? On the other hand, what are some of the under-discussed benefits?

Automation could further undercut the power of human workers, especially in industries like food service. Workers could find themselves competing for jobs with these very cheap, reliable machines.

With that said, the great benefit of automation -- the great hope it raises -- is for it to replace unpleasant, unrewarding work, and to boost productivity to the point that it creates more returns to enterprise, and more prosperity. In that way, automation really can boost firms' output, allowing them to grow, hire more, and pay more workers higher wages. That virtuous outcome -- which complements humans, rather than substitutes for them -- is still not discussed enough.

 

Daniel Zhao, Senior Economist, Glassdoor

 

What can Americans do (or what skills can they learn) to have successful careers in an economy that is more and more driven by automation?

It’s important to remember that automation very rarely completely wipes out careers, but rather, it allows jobs to be done more efficiently. To be successful working in an increasingly automated world, workers need to build skills that are complementary to technology and learn how to run the machine rather than doing the same work the machine automates. It’s also advantageous to sharpen up on soft skills that are irreplaceable by machines. Skills like public speaking, organization and stellar customer service are a few examples of soft skills that can always be improved.

Are there certain occupations that Americans should try to avoid due to automation?

While it may seem like blue-collar workers are the most likely to be replaced by automation, white-collar workers are not immune, either. Just as machines are consistently appearing on assembly lines in factories, new apps and software are transforming routine tasks that were common among traditional office workers. It’s important to continuously adapt to ensure the new technology is helping workers accomplish even more in their day-to-day roles rather than replacing them. And if your job is eliminated due to automation, then it’s never too late to reassess your skills and prepare for a career pivot.

In your opinion, what are some of the under-discussed challenges that automation will create? On the other hand, what are some of the under-discussed benefits?

Automation poses many benefits for a company, from increased productivity to streamlining processes. But it also poses many challenges, from the high cost of implementation to potentially eliminating jobs. Employers should ask themselves if an automated process will benefit their business and help their employees. Or will it keep business the same and replace them?

 

Sources