In the ongoing online economics debate over technology, jobs, and automation, the Luddite Fallacy and Lump of Labor fallacies are often invoked.
However, the there is a point of common confusion, which is if the goal of technology is to replace labor, then over the long-term labor cannot compete with technology. I believe technology will not displace all jobs. By lowering costs for one firm (such as a firm installing a robot to replace employees), that automation creates new jobs for another firm. Cars replaced the horse and buggy industry, but look at all the new jobs cars created. What has to be understood is that when a new productivity-enhancing technology is successful, it spawns an industry, creating jobs in the process even if the original intent of the technology is to eliminate jobs. As more companies want this new technology to save money, the company that makes this technology is swamped with demand and needs to expand, and this means jobs are created. And then the new technology may occasionally break, and then repairmen need to be hired, ad infinitum.
Although Amazon’s business model strives to automate as much of the shopping experience as possible, its employee count has surged, owing company’s enormous success:
As another example, despite the decline domestic auto manufacturing, the auto industry employs vastly more people than the horse and buggy industry ever did, including auto dealers, accessories, body shops, car insurance…even accident lawyers.
Although computers (called tabulating machines) were originally intended to help tabulate census results, it eventually gave birth to an industry larger than anyone could have ever imagined, as computers became faster and omnipresent.
In the case of robots, a whole new set of jobs will be created revolving around the robots, such as maintaining the robots. And of course, new jobs are always being created around news technologies. For example, major brands hire people to screen their Facebook pages for spam and to answer inquiries, as well as update social media, which is an example of a job that just a decade ago didn’t exist.
In other instances, automation and employees coexist, meaning that the total employee count does not fall despite automation. Rather than food stores eliminating all cashiers, self-checkouts coexist with them, and also employees are assigned to monitor the self-checkout machines.
And old technologies have a habit of being persistent. Like cashiers coexisting with self-checkouts, horse and buggies still exist, as do typewriters. In many instances, there is a demand for old technology that fills some sort of niche, and this demand, although much smaller than the new technology, is still large enough to keep the old technology going.
Another solution is retraining the workforce to learn better skills, but Gwern is not as optimistic, invoking the IQ argument that many may not be smart enough, in which I agree:
I see. So all we have to do with all the people with <120 IQs, who struggled with algebra and never made it to calculus (when they had the self-discipline to learn it at all), is just to train them into world-class software engineers and managers who can satisfy Silicon Valley standards; and we have to do this for the first time in human history. Gosh, is that all? Why didn’t you say so before - we’ll get on that right away! Or an anonymous “data scientist” recorded in the NYT: “He found my concerns to be amusing. People can get work creating SEO-optimized niche blogs, he said. Or they can learn to code.” Thomas Friedman:
Maybe the issue is the cognitive requirements to perform entry-level labor may rise to a sufficiently high threshold that it excludes too many people from the labor market. This may mean that while the total number of jobs does not fall, the IQ requirement rises, or as I explain in Paul Graham: Economic Inequality and The Refragmentation:
The trend, I predict, is that wealth inequality will become almost the same as IQ inequality. IQ is more much important today than a generation ago in influencing individual economic outcomes. 100 years ago in an economy dominated by manual labor, the difference between a 90 IQ and 120 IQ wasn’t that important, but now it is.
This may be manifested in the ‘hollowing out‘ of the middle, where there too many low-paying jobs, lucrative creative class jobs, but not enough middle-income ones, due to less intelligent people being able to find good-paying jobs. Luddite Fallacy and Lump of Labor may ignore job pay or cognitive demands. However, cars are much more complicated the carriages , yet there are jobs for all intelligence levels, from people who clean the interiors of cars to those who solve differential equations to model airflow.
Ultimately, in our Great Economics Debate, the problem is no one knows of solutions or can predict what will happen, but these issues will impact everyone, hence why there so much discussion about the topic of jobs and automation.