I was on a panel about automation and the future of work a couple of days ago, giving me a reason to read some of the latest articles on the subject. It’s a bright picture!
Nearly half of Ohio workers hold jobs that a University of Oxford study has identified as likely to be automated in the future, including tens of thousands of cashiers, truck drivers, fast-food workers and warehouse laborers.
In a 2013 study, researchers Carl Benedikt Frey and Michael A. Osborne rated 702 jobs on a scale from zero, meaning they can’t be automated, to one, meaning they certainly will be automated. Of those, 297 jobs rated a 0.7 or higher, meaning there is at least a 70 percent chance the jobs will face disruption from computerization and automation.
The researchers gave a higher likelihood of automation to lower-paid, repetitive jobs that require little or uncomplicated personal interaction and lower levels of education.
To gauge the potential effects of automation on Ohio’s employment, The Dispatch compared the high-risk jobs in the study with U.S. Bureau of Labor statistics surveys that show how many Ohioans hold them. As of March 2016, the last time the bureau released statewide job statistics, Ohio had 163,790 people engaged in food preparation and serving jobs, including fast-food jobs. There were 117,390 cashiers, 111,230 laborers and freight handlers, including warehouse workers, and 70,740 drivers of heavy trucks.
Adding up all of the jobs that are at least 70 percent at risk shows that about 2.5 million jobs in Ohio are at stake. Total employment in the state was about 5.5 million in December.
Taken together, the average pay of all of the Ohio workers in all 297 jobs rated as the most likely to be automated added up to nearly $83.6 billion in 2016, according to the labor bureau’s statistics. That includes 169,300 office clerks, secretaries and administrative assistants, 98,150 waiters and waitresses, and 61,590 bookkeeping, accounting and auditing clerks.
This is a very serious problem. To me, as a nation, we face 3 existential crises. First is climate change. Second is the decline of American democracy. Third is automation. This region is already the center of the opioid crisis because there simply is no hope in a lot of communities in Ohio, Pennsylvania, upstate New York, West Virginia, Michigan, etc., for working-class people where the jobs have disappeared. Moreover, the decline of jobs and hope helps convince white people to blame black and brown people for their problems and make the appeal of someone like Donald Trump that much higher. If you liked the 2016 election, you’ll love a future with half of Ohio workers lacking jobs!
The problem is not that automation is good or bad. Any technology is a tool and its impact depends on how we use it. I am a strong critic of our technofuturism and people who think technology will save us. It won’t. Automation is a great example. There are undoubtedly good things about some parts of it. There are also undoubtedly bad things. Perhaps there are more bad things if it leads to widespread instability. On last night’s panel, there was myself, a politically conservative business professor, and two physicists who work on automation. All of us agreed that a major crisis was in front of us that could have horrible impacts on society. I was in full apocalyptic mode for my talk and the other panelists agreed with all of it, or at least seemed to in our conversation during Q&A.
However, discussions of what to do about this problem I don’t think forward real solutions. If you want to have automation while also having a stable society, you have to provide people with jobs or income that allow them to live in dignity and to more or less live where they want to live. So when I read pieces like this Harvard Business Review article that try to grapple with the problem, I find myself frustrated, not because I disagree with all the points, but because I don’t feel they go nearly far enough.
On the supply side, the key will be to address a range of issues that will help us through the transitions. As noted above, a prerequisite will be to ensure robust GDP growth, since without that there will be no job growth. Three other priorities stand out:
First, a much sharper focus on skills and training. That means reversing the trend of declining government spending on training that is apparent in many OECD countries. It also means a stepped-up role for companies, which will be on the front line of automation adoption and will know better and faster which skills are required.
Second, we should take another look at making the labor market more fluid, including by more active use of digital technologies for job matching and for stimulating the rise of independent work. In fact, the dynamism of labor markets is waning: in the United States, for example, the job reallocation rate dropped by 25% between 1990 and 2013, and the share of workers relocating across state lines annually has fallen by half, to close to 1.5%.
Government, businesses, educational institutions, and labor organizations need to collaborate to ensure that incumbents and new entrants to the labor market have accurate forward-looking knowledge of the evolving mix of skill and experience requirements.
The third priority should be a reevaluation of income and transition support to help displaced workers or those struggling with transitions to new occupations. Germany set an example here by revamping its labor agency and putting an emphasis on acquiring skills. Its labor participation rate has risen by 10 percentage points since reunification, to above the U.S. level.
While I agree that we need more government spending on training, as well as more corporate training (instead of demanding that universities train their entry level employees for jobs that may not even exist in 10 years), with automation moving so quickly, we may very well move beyond the point where this is even a worthy idea. If entire fields become obsolete, how do you even train people for something likely to disappear. The bit on digital technologies to match employer and employee is a complete non-solution. I mean, fine, but this is not an answer. And transition support to new work is a good thing–we need that–but again, what if there just aren’t working-class jobs for people to transition into?
All this discussion on education and retraining shows just how clueless the educated classes are about the working class and working class culture. If you want automation to take place without massive social upheaval, you simply have to provide a decent living and hope to working class people. You have to go way farther than is commonly discussed. A lot of people are simply not capable of going to college. This has to be recognized as something that is OK. We are not going to educate our workforce out of this problem; it’s literally not possible. We all know people in our families who are not college material. They have to have the opportunity for a dignified life.
Of course, this is where Universal Basic Income comes in. And while I’m fine with UBI on principle and would certainly support any serious proposal around it, I maintain that most of its adherents are educated people from non-working class backgrounds who simply don’t understand either the centrality of work to American culture and the connection between work and the welfare state in American politics. Call Horatio Alger a corporate propagandist if you want. He was a corporate propagandist, in addition to being a pedophile. But even if the working class are suckers for believing propaganda about pulling yourself up by your bootstraps and not accepting government handouts, the reality is that tens if not hundreds of millions of Americans believe this very strongly. The idea of the job well done and working to support your family are ideas taught to children, boys especially, from a very young age. We ignore this at our peril. Will working class people even support UBI if the future really is jobless?
Moreover, the limited welfare state that Americans have is largely predicated on work. Social Security is popular because it is widespread, albeit not universal. But the core of that popularity is the idea that we earn our Social Security benefits through a lifetime of work. It’s the same with Medicare. The welfare state programs that are most vulnerable to right-wing attacks–AFDC, Medicaid, Food Stamps–are all disconnected from work. Many Americans feel resentful of these programs. They are wrong to do so, but they also elect politicians who are now creating work requirements for Medicaid recipients and they won’t punish them for that.
This is the root of my skepticism about UBI as a political winner. It’s also why I propose a real full employment program based on the original drafts of the Humphrey-Hawkins Full Employment Act that was eviscerated by the Carter administration before he signed what was left of it in 1978. The federal guarantee of a job is the real answer here. It doesn’t have to be in steel mill. Teaching guitar to kids is a job. Home care work is a job, and a bloody hard one. Carving wood to sell at markets can be a job too. But we have to both have financial stability in the working class and respect the idea of work. I know that for some people, work sucks and they like UBI as an emancipatory measure. That’s fine and all, but for many, and I think a clear majority, of Americans, they want to work. That has to be central to our solutions to the automation crisis.