Last week I was speaking at the MD&M West Expo in Anaheim, California on the “future of manufacturing.” With over 2,000 manufacturing exhibitors filling the convention center, there was no small amount of interest in this topic.

With China and the rest of Asia making massive inroads in manufacturing over the past couple decades and automation threatening many of the remaining industries, a huge underlying theme of this event was jobs. Where will our jobs in the future come from?

Job loss is not an idle threat. As everyone attending this conference knows, businesses have an obligation to hire the fewest number of people they can get away with, and when automation eliminates the need for an employee, the employee has to go.

However, while job loss is very real and happening all around us, job creation is also happening, in way that many have not seen coming.

To be sure, the transition period we are in will cause considerable collateral damage, but we will also experience a period of unprecedented opportunity provided we create the right systems for capitalizing on them.

As I mentioned at the conference, the fastest way to create new jobs is to automate them out of existence. Here’s why!

Lifecycle of an Industry

All industries form a bell curve 

As with everything in life, all industries have a starting point and an ending point. Usually the starting point can be traced to an invention or discovery such as Alexander Graham Bell’s invention of the telephone or Henry Bessemer’s discovery of a process for making steel cheaply in large volumes. The end comes when new industry replaces the old, like calculators replacing the slide rule.

In addition, all industries form a bell curve. At some point along the way, every industry will experience a period of peak demand for their goods or service.

Many industry are entering the downside of the curve

Many of our largest industries today are entering the second half of the bell curve.

To understand this, ask the simple question, “What goods and services that we buy today will we be spending less money on in the future?” The list you come up should include energy, transportation, healthcare, publishing, insurance, telecom, education, construction, mining, and many more.

Leading indicators that industries are entering their top-of-the-curve midlife crisis are a growing number of startups attacking key profit centers.

Prior to reaching peak demand for their goods or service, industries experience a period of peak employment.

Peak Steel 

Using “Peak Steel” as an example, the peak employment for the steel industry was reached in the 1980s. The peak demand for steel itself is projected to happen in 2024. This is when composite materials will gain enough of a foothold and the overall demand for steal will begin to decline.

Many industries have already begun reducing headcount, partly because of automation, but primarily because they are about to enter their waning years.

The True Engines of Job Creation

Startup companies are necessarily sloppy in how they grow, requiring additional headcount to manage their still-to-be-defined business operations. Mature industries have well-defined processes and are better able to find efficiencies along the way.

Job Creation and Destruction

A recent Wall Street Journal article tried to correct the perception that this is a small business vs. large business debate. Instead, it’s a young business vs. old business issue.

Economists John Haltiwanger, Ron Jarmin and Javier Miranda showed in a 2010 paper, what really drives job growth is fast-growing start ups.

Their study showed that even before the recession, starting in 2006, government policy shifts caused a key turning point.

Young businesses created an average of 5.5 million jobs per month from 1992 through the end of 2006. Since then, they only created 4.7 million per month. This is appearing to be a long-term trend.

Since stable old businesses tend to be major campaign contributors, policymakers tend to favor them over young startups. While the writers suggest the key to more jobs is a good start-up lobby, the real path to job creation is more automation.

Demographic Shifts

Older Americans are continuing to work, primarily because they can’t afford to retire. The recession has impacted their household budgets, and particularly the value of their investments and retirement funds.

According to new U.S. Census Bureau data, 12.5% of the population is over 65 and that number will grow to 20% by 2040, possibly more if the birthrate continues to decline.

People age 65-69 who are still working grew from 22% in 1990 to 31% in 2010. For men working between age 70-74, the numbers grew from 16.6% in 1990 to 20.9% in 2010. For 70-74 year old women it was a similar pattern, increasing from 8.4% to 13.5%.

Blaming Robots

Even before the real age of robots has begun, they are being blamed for stealing people’s jobs.

That’s the view of economists Henry Siu (University of British Columbia) and Nir Jaimovich (Duke University), who point out in a recent article, the reason for this structural change in labor markets is the rise of automation:

“Automation and the adoption of computing technology is leading to the decline of middle-wage jobs of many stripes, both blue-collar jobs in production and maintenance occupations and white-collar jobs in office and administrative support. It is affecting both male and female dominated professions and it is happening broadly across industries – manufacturing, wholesale and retail trade, financial services, and even public administration.”

While that may be true for what’s happening to existing traditional jobs, they fail to account for the wide range of entrepreneurial and new work opportunities that the same technology is creating.

Job losses are easy to count. Startup businesses, however, are far more difficult to monitor because most tend to fly under the radar until they enter a serious growth phase.

Finding the Seeds of Creation in Automation 

According to a May 2011 study by the McKinsey Global Institute titled “Internet matters: The Net’s sweeping impact on growth, jobs, and prosperity,” the Internet has accounted for 21% of GDP growth over the last five years.

They also concluded the Internet is a key catalyst for job creation. Among 4,800 small and medium-size enterprises surveyed, the Internet created 2.6 jobs for each one lost to technology-related efficiencies.

We are now transitioning from room-size automation, which only large companies could afford, to desktop automation that allows small and even one-person businesses to be part of.

In much the same way that the 1985 Apple LaserWriter gave birth to desktop publishing, the 2010 MakerBot’s Thing-O-Matic 3D printer gave birth to desktop manufacturing.

Automation is no longer to domain of the elite few, and the quicker we can make the transition to all industries, the quicker everyone can participate.

The Economics of Automation

Our economy is based on people. Humans are the buying entities, the connectors, the decision-makers, and the trade partners that make our economy work.

Without humans there can be no economy. So when it comes to automation, consider this:

  • A person with a toolbox is more valuable than a person without one.
  • A person with a computer is more valuable than a person without one.
  • A person with a robot is more valuable than a person without one.

Automation does not happen simply for the sake of automation. It’s intended to benefit people.

If we only look at what automation will eliminate, we’ll be viewing the world through a glass-half-empty lens.

Final Thoughts 

In Feb 2012, I made the prediction that over 2 billion jobs would disappear by 2030. We seem to be on track for that to happen.

Driverless cars will eliminate millions of driving positions. 3D printers will eliminate millions of manufacturing jobs. If we continue down the list, teacherless schools will eliminate teachers, pilotless planes will eliminate pilots, checker-less retail will eliminate checkout clerks, and so on.

Whenever jobs go away, politicians tend to have a kneejerk reaction trying to implement legislation that enables us to hang on to the past for a while longer.

But job losses will happen regardless of whatever overt attempts are made to stop the hands of time.

Contrary to popular opinion, automation creates jobs. As the McKinsey study has shown, we get a 2.6 to 1 benefit from jobs lost through digital automation on the Internet. Since the physical world is 5 times the size of the online world, we may create even more jobs with physical automation.

By Futurist Thomas Frey

Author of “Communicating with the Future” – the book that changes everything


15 Responses to “Fastest Way to Create New Jobs? Automate Them Out of Existence!”

Comments List

  1. <a href='http://www.Market-Engineering.com' rel='external nofollow' class='url'>Gary Lundquist</a>

    Tom You and I both understand this well and have spoken to audiences about the rate of change. Back in 1990, I wrote the Way of Change. My bookmark: Change will never, ever again be as SLOW as it is today. The solution is innovation. We have put on Innovation Summits together to educate others about change. ... Innovation is the driver of change. Innovations improve products, sometimes to the point of excluding people. ... The solution to that is to "innovate people". I have launched a revision of the Colorado Innovation Newsletter blog... About 40 essays on innovation. 2 per month. The basics of innovating a person. ... Innovation is a process of renewal. The same basics of that change applies to people as well as technologies. Don't ever stop innovating yourself. Keep learning, adapting, growing, and keeping up with change. ... You, Tom, are a great example. You were always looking ahead. Now you have remade yourself into the guru of the future. Congrats. I stand in awe. ... Best, Gary
    • admin

      Vinod Khosla, a well-known venture capitalist and co-founder of Sun Microsystems, recently said that Big Data would eliminate the need for 80% of all doctors. These remarks have been very troubling for the medical community, I tend to agree with what he's saying but it’s helpful to put it into perspective. Sixty years ago it took 5 people in the cockpit to fly a large commercial aircraft - pilot, co-pilot, flight engineer, radio operator, and navigator. Over time, we have found ways to automate (eliminate) 3 of those positions and the FAA is looking closely now at eliminating the co-pilot, making this an 80% reduction in the amount of people needed to fly a plane. As we improve our ability to diagnose the human condition, instead of running 4-5 tests on a person, we will be monitoring 4-5 million data points, something impossible for a human to do. Today, we are on the verge of eliminating 80% of the people in an airplane cockpit, but since far more people are flying, we still need a lot of pilots. In a similar fashion, humans have far more physical problems than we are able to detect today. If we improve our ability to detect problems by 500% (5X) and eliminate 80% of the doctors needed to do 5 times the work, we would end up with the same number of doctors we have today. If I were a doctor, I wouldn't be worried about losing my job any time soon. In fact, it would be in their best interest to welcome these changes. Tom
    • admin

      Grant, In a word, yes, automation creates jobs. But the simplicity of the statement belies the shifting tectonic plates of complexity needed to leverage it. Tom
  2. <a href='http://www.agresourcestrategies.com' rel='external nofollow' class='url'>Tim Gieseke</a>

    Thanks for the positive insights. As a one-person business involved in developing ecocommerce models at the local, state and national levels, I certainly agree with your observations. This led me toward understanding how our networks are changing the governance of policy and where the next wave of jobs lie - here is an illustration: https://prezi.com/lp3tff-pwtw0/shared-governance-gateway-to-entrepreneurism/
    • admin

      Tim, I like the way you think. It looks like you've launched a fascinating movement in the space of shared governance. Implementing it will undoubtedly require considerable time and energy, and considerable human labor along the way. :-) Tom
  3. <a href='http://www.moorinsightsstrategy.com' rel='external nofollow' class='url'>Paul Teich</a>

    Great observations. But there is a lag time for people and culture to adjust to new technology, and new occupations and methods of work for previous occupations. The last time we saw this scale of technology related job displacement, Marx & Engels wrote an interesting set of socioeconomic observations and prescriptions. Perhaps 100s of millions of people died in the revolutions that were spurred by their ideas (right, wrong, or indifferent...it's simply a cause-and-effect observation). A century afterwards the world has settled down a bit, but their redefinition of the contracts between governments, companies, and employees/citizens still reverberates through worldwide geopolitics. What's happening to secretaries, middle management, data entry clerks, etc.? If you are in marketing and you don't understand big data, that's a problem, it's going to change your career permanently...and perhaps for a short few decades the world will need fewer marketing employees. If existing doctors don't understand and course correct for their new career model, they won't have their jobs in another decade or two (self-selection or forced retirement). If you look at doctors as knowledge workers, then I believe the number needed will decline. If you look at doctors as service workers who act as the front-end "human interface" (an interesting repurposing of that definition) for intelligent systems, then the world might need more, but they will have a much different skill set and our assumptions and expectations for them will change. Using airline pilots as another example, the rate of increase for pilot demand coupled with the rate of increase in the quality of cockpit automation means that pilots in general are less qualified for emergency situations than they were a couple of decades ago (when virtually all were former military pilots). Eventually auto-pilots will fill the gap with better AI and big data based knowledge of infrequent events, but for the time being we should all be concerned as an older, more experienced set of pilots retires. We will be experiencing turbulence for the next few decades. Please ensure that your seat belts are fastened and your tray tables are in an upright and locked position... Paul
    • admin

      Paul, Thanks for your observations. There are many fundamental system changes that will need to take place before we can truly leverage the job-creation potential of automation. As an example, our current education system is woefully out of touch with the needs of the future. Education today is far too labor intensive, takes too long, costs too much, and focuses on the wrong skill sets. Our systems for home ownership make every transaction very slow and expensive, forcing people to commute to their job for years rather than relocate. At the same time, we are living in an increasingly mobile society. Our tax code will be mocked by comedians for several decades into the future because it is so brutally incomprehensible. If forces people to dedicate far too much intellectual bandwidth to filling out forms, while distracting them from causes and business dealings that are far more important. As we work our way down the list from our immigrations systems, to our banking and finance systems, to our healthcare systems, and our justice systems, we find ourselves lacking on virtually every front. Problems ahead are enormous, and solving them will require extensive human effort, with many people working long and hard to make it happen. And truth be told, no problem ever gets totally solved. Every solution unleashes more problems along the way. Automation and robots will never be the problem. Rather, what I've described are all system problems that will demand constant human effort to fashion them into something our future selves can work with. Hope this helps, Tom PS - Ironically, it's a lot of work explaining why we will never run out of work.
    • admin

      MF, The iPhone 4 that was assembled by Foxconn in China, required 24 people to handle it from start to finish. The new iPhone 5 being assembled in Brazil on their new automated assembly plant only requires 2 people to touch it. Yes, it also will require more machine techs and maintenance people in the background, but when assembling tens of millions of units, overall labor costs per unit drop appreciably. Foxconn is still winning when it comes to human labor costs, but once a large portion of human labor is removed, location for manufacturing becomes far less important and proximity to the engineering team becomes a higher priority. Tom
  4. <a href='http://www.RetirementSingularity.com' rel='external nofollow' class='url'>Michael Nuschke</a>

    Enjoyed your perspective, thanks! In my mind, the number and types of jobs impacted by robotics and AI in relatively short order will be massive. Meanwhile, the number of effected humans (especially those over 50) that will be able to remake/retrain themselves and keep up with accelerating change may be less. It seems an important part in our job is to present this dilemma convincingly enough that it starts more people preparing now for that future! Cheers, Michael Nuschke
  5. Antonio C

    While I agree that automation creates jobs, it also raises the barrier for entry to each new job created. Without a radical redesign of the school system or some incredibly wonderful technology that boosts a person's intelligence, not everyone can become an AI researcher, for example. Current innovation in online education will certainly reach more people who have the aptitude but not the opportunity, but giving the opportunity to people who simply don't have the aptitude (be it in motivation, or actual cognitive capacity) won't do a thing them. And with each subsequent advance, and a rapidly increasing rate of change, you wind up increasingly closer to the inevitability of massive unemployment. Even taking doctors as an example: Not all doctors are created equally. I agree that you will end up with the same number of POSITIONS for doctors, you most likely won't end up with the same number of doctors, however, as the bar is raised as to what kind of education and intellect is required to perform increasingly complex tasks. (Or you wind up with a massive labor pool for relatively simple but related tasks and wind up devaluing the labor) Intelligent people often fail to recognize, as they tend to associate only with other intelligent people, that there are a LOT of Joe Shmoes out there who simply aren't up to the task of being employed in radically technologically advanced positions. And as automation becomes more and more advanced, the base level of education and intelligence required for a job increases. Obviously, it's more complex than just intelligence, as there are plenty of intelligent people working at Starbucks or Walmart, even in permanent positions. But it is an easily understandable and easily applied metric that is often overlooked when we talk about automation creating jobs. It is the reason there is a shortage of coders. Better education will help and even those people who can't handle coding advanced AI will get lesser jobs in the field... Until those are replaced and the bar is raised even higher. This is not to say that we need to avoid automation. Simply that, without planning for it, a rapidly rising tide does NOT raise all boats. Without retrofitting and planning, a whoooooole lot of the dinghies out there are going to sink.
  6. Demitri M

    While this was a good article, I was a bit disappointed in the lack of concrete statements or predictions about exactly what manner of jobs that start-ups based on automation will create, or how to address the challenge of educating tomorrow's workforce as necessary for such jobs. Most if not all high-tech jobs that produce automatons require years of training and/or a degree in computer science/engineering. And, as we all know, higher education puts people in mountains of debt that may for all we know be impossible to repay, due to how even expensive skills/degrees may soon become useless in the job marketplace.
    • FuturistSpeaker

      Demitri, Here are four other columns where I go into more detail about future jobs. http://www.futuristspeaker.com/2013/11/have-we-reached-peak-employment-24-future-industries-that-will-lead-to-an-era-of-super-employment/ http://www.futuristspeaker.com/2014/03/162-future-jobs-preparing-for-jobs-that-dont-yet-exist/ http://www.futuristspeaker.com/2014/07/the-laws-of-exponential-capabilities/ http://www.futuristspeaker.com/2014/08/2050-and-the-future-of-infrastructure/ Hope this helps, Futurist Thomas Frey
  7. <a href='https://www.greetly.com/blog/3-small-office-automation-steps' rel='external nofollow' class='url'>Dana Mueller</a>

    This is a great article. At least today, most of the automation is in fields that no one really wants to do. So while it might impact jobs, it has the possibility of freeing us up for more interesting work on behalf of mankind.

Leave a Reply