On Wednesday, the automation-focused U.S. meme candidate Andrew Yang tweeted, “Fast food may be first.” He was commenting on a new CNBC report that reported annual employee turnover rates of 100 per cent at the Panera Bread chain in the U.S. — a figure that is low for the fast food industry, which can see annual turnover of up to 150 per cent.
Those figures may seem ridiculous, but they’re a reality: The fast food restaurants regularly see more than their entire workforce turn over every year. And that is why industry experts — and Andrew Yang — warn that it’s ripe for automation and may be the first field to become entirely automated.
They’re right that fast food is as ripe as any industry is for transformative automation. Typically, one of the major sources of resistance to automating a process, task, or entire job is the impact it will have on a salaried employee.
Layoffs look bad for the company doing the automating, there are myriad social factors in play that create resistance — management will be reluctant to fire longtime employees, for one — and there is risk involved in setting up new machinery, which may take years to get running smoothly.
But in an industry that turns its entire staff over every year — especially one in which the bulk of its jobs are intended to be an agglomeration of repetitive tasks like taking and inputting orders, adding and arranging ingredients to a dish, and cleaning floors and tables — corporations and middle management will spend a lot less time weighing social factors and nursing concerns about optics.
Hell, fast food is already one of the worst-regarded jobs in the U.S. because workers are openly treated with so little dignity, the benefits range from threadbare to nonexistent, and the wages are so low.
All of which is to say: As soon as the fast food companies can automate those jobs, they will. The only things preventing those companies from doing so are the projected costs and the functionality of the automated systems. That’s it.
This, by the way, is why it’s so specious of restaurant industry CEOs to claim that if they’re made to pay restaurant workers a few more bucks an hour it will force them to adopt automation. (The Fight for 15 campaign has been organising workers to raise restaurant wages, and threats of automation have been a persistent bit of propaganda deployed to counter their efforts.)
For one thing, fast food executives are already trying as hard as they can to do exactly that — McDonald’s, for instance, is reportedly spending $US1 billion ($1.5 billion) on automated ordering kiosks this year alone.
As it stands, wages would have to spike exorbitantly high to put execs in a position where they’d be willing to bet on costly, untested back-of-house automation technologies for food prep, even more kiosks — which, let’s be clear, it remains unclear just how much labour savings they offer — and an extensive training program to familiarise employees with all of the above.
(Somewhat hilariously, these sky-high turnover rates that have analysts predicting automation madness to solve the fast food industry’s retention problem — it’s expensive to constantly recruit and train workers — can alternatively be combatted with relatively simple means: offering your employees a modicum of dignity, a little time off, and some benefits. Starbucks, a mega-profitable franchise chain, has a retention rate of 65 per cent precisely because it offers those things, and it is still mega-profitable.)
For all of these reasons, fast food is a powerful bellwether for automation — a canary in the coal mine for the phenomenon more generally. I don’t know if fast food will be “first” — I’d probably still put my money on some subset of the manufacturing industry, somewhere bosses can justify more expensive machinery because the workers they replace are often skilled and, therefore, expensive, too. But the sector provides something close to the ideal conditions for the total automation of corporate service work.
This matters, because there’s still a lot of ambiguity about what automation can reasonably accomplish at reasonably affordable rates right now. There’s so much grey area between what business-to-business enterprise automation solutions companies promise Fortune 500 firms and what automation can actually be achieved that sometimes it’s hard to get a picture of how capable and cost-effective corporate automation really is.
Retail chains have been trying for decades to automate ordering and checkout, for instance, and humans are still doing the vast majority of the labour. Yet there’s no doubt companies are relentlessly developing and pitching new automation systems, some aimed squarely at the fast food industry.
This is why I do think it’s fair to say that the day you see a successful, fully functional and fully automated marquee franchised fast food restaurant — if that day ever comes — that day will indeed be a harbinger for the rest of the economy. (Before then, I’d watch out for more fauxtomation; more jobs outsourced to gig economy workers at places like Deliveroo; delivery is one of the few growth spots in the fast food economy, and I bet we’ll soon see fast food including those gigs in its job creation figures).
That will be the day that we’ll know that automation really is capable of replacing low-paying human labour at scale. The day there are no humans working at McDonald’s — that’s the day the so-called robot jobs apocalypse will have truly begun.