The lady who took my order must have been about 19, as were all the other employees I could see, and pretty clearly new on the job. Getting the order right took some effort. I made the mistake of paying cash. The bill was something like $7.62. I first offered a $10, and she rang it up. Then I found 12 cents in my pocket, and offered it. This was a big mistake, as the cash register had already computed my change, and adjusting to my offer of 12 cents was beyond her abilities.
Most people might have been annoyed, but as an economist and an educator, I’m happy to see human capital building. OK, I was a little annoyed.
Which brings me, of course, to the proposals for a sharply increased minimum wage.
In the end, there really isn’t much argument about what a substantially higher minimum wage will do.
Let us not deny the benefit. For the few people who work at minimum wage, but have worked their way up the ladder enough that they will keep their hours; who are actually trying to support themselves and a few children on these meagre wages, it will mean a modest rise in income. The rise may be more modest once you account for taxes and reductions in transfers. There weren’t any such people at my McDonald’s, but NPR and the New York Times seem able to find them.
That transfer comes from somewhere. Some of it comes from a wealth levy on existing McDonald’s shareholders. If a regulation lowers a company’s profits, the stock price declines. Then the rate of return going forward is the same as always. So it’s a one-time wealth tax on the existing shareholders. Economists are supposed to like wealth taxes, with an asterisk that it makes future investors a bit skittish.
Some of it comes from higher prices. I read estimates that a Big Mac might go up from about $3.00 to about $3.50, and dismissed those price increases as a small burden to bear. Looking around my McDonald’s, I found this argument less persuasive. Because, of course, the kind of people who work at McDonald’s are also the same kind of people who eat at McDonald’s. If you’re working at minimum wage in the middle of Oklahoma, you don’t go out to a nice Greenwich Village restaurant to sample organic free range locally grown non-GMO gluten-free artisanal nuts and berries. McDonald’s is a treat. And a pretty nice one at that. It’s clean, healthy – yes, some offerings are full of sugar and fat, but not of e-coli, and you can get the grilled chicken if you want — and reasonably tasty. Raising prices from $3.00 to $3.50 is not a small matter if you earn under $10 per hour and you’re feeding a few kids too.
Still, that is the benefit.
The cost is just as easy to forecast. McDonald’s cuts hours and uses its most experienced and efficient workers more, and fewer people like my hapless server. And they don’t get the oh-so-needed on-the-job training. The biggest impact of minimum wages is not so much on existing workers, but on new workers entering the labour force. (See a nice new NBER working paper by Jonathan Meer and Jeremy West.)
The effects fall heaviest on low-skilled teenagers, especially minorities. Tom Sowell is eloquent on this point, for example in a recent New York Post op-ed. I was unaware until reading it that minimum wage laws were initially backed in part as conscious efforts to discriminate against minorities and preserve jobs for white people. Sometimes, I guess, policies do have their intended effects.
This much is pretty obvious. Looking around my McDonald’s, though, I could see a deeper possibility — an unexplored avenue for substitution away from low-skilled labour.
Why, I wondered, after 10 minutes in line and the third effort to get my simple order right, did I not simply enter my order on my iPhone, and then it’s ready for me when I step up to the counter? Or why not enter it on a tablet provided right there? Why should ordering at McDonald’s be any different than getting money from a bank, or getting a boarding pass at an airport? High end restaurants answer this question by saying they think their customers value the personal attention of a waiter. Maybe, but certainly not at McDonald’s.
The answer, for now, is certainly that it’s cheaper the way it is. But perhaps not for long.
McDonald’s is reportedly testing an ordering and cellphone payment app. ‘Currently being tested at locations in Salt Lake City and in Austin, Texas, the app lets users order a meal remotely then collect it in person from a store or drive-thru window.’ My server’s job days may already be numbered.
Looking more inquisitively behind the counter, it struck me that the technology overall has changed little since the 1960s when my parents took me there as a child. The fry-o-lator beeps, a teenager picks the basket up and dumps it out, sprinkles salt, and uses a cute little piece of aluminium to neatly line them up in bags, just as they did back then. The main change I could see is that they annoyingly don’t let you put your own sugar in your coffee any more.
It’s clearly only a matter of time before this whole thing is automated. Industrial robots can assemble cars; designing a robot to operate the fry-o-lator, or even to cook and assemble the whole hamburger doesn’t look that hard. Mechanisation usually increases quality: your burger and fries could easily be cooked to order. Swipe your phone or card to pay and off you go. Or, a little drone helicopter delivers it automatically to your table.
(Update: The machine is here already. And planning a new chain to use it, rather than sell to McDonald’s, as predicted. Thanks to Michael Ward for pointing it out.)
Reflecting on it, though, it’s unlikely to be McDonald’s. McDonald’s has an amazing technology when you look hard at it: They have figured out how to run restaurants in a way that dramatically conserves on the world’s scarcest resource, human capital. To run a McDonald’s, you don’t have to know how to cook, how to order food, how to buy kitchen equipment, or all the other hundreds of bits of tough knowledge and skill that it takes to run a restaurant. Hamburger U trains the rest.
The whole operation is about taking low-skill teenagers living typically unstructured lives, and training them to what it takes to work. Peering around the side of the cash register at an earlier trip, I noticed there were pictures on the buttons! You can work at McDonald’s and operate its cash registers even if you’re functionally illiterate! To say nothing of not knowing what to do when offered $10.12 to pay a $7.62 bill. And McDonald’s has a big investment in that technology.
In the face of technical change, it is seldom the successful incumbents who adapt, even when they innovate. Kodak did not bring us digital cameras, trying to protect their film advantage. Print media did not bring us the internet, and are floundering at it. Walmart tries to go online, but Amazon.com is displacing it. The major airlines flop in every attempt to imitate Southwest.
So, as I gaze around the familiar golden arches, it strikes me that the automated fast food restaurant — and the rapid decline in low-skill employment that it implies – will likely not come from McDonald’s itself. Rather, new competitors will arise that perfect the automated, people-less technology. In the same way that McDonald’s displaced the previous era of fast-food restaurants, by perfecting a technology that brilliantly used lots of low-skilled people and conserves on scarce human capital. For McDonald’s to go automatic would be for it to throw away the key innovation that defines it and has made it such a success.
So we may be past the point that McDonald’s sticks with 1950s technology because it’s still cheaper to use people. We may just be waiting for the tipping point.
But robot repair technician is a high-skilled job. McDonald’s provided a positive social externality – it gave young people their first experience of work, of showing up on time, in a uniform, of learning to be pleasant to customers, to work within a hierarchical organisation, and so on. Young people who work at McDonalds don’t get internships at NPR, the New York Times, or Goldman Sachs to develop work experience. As McDonald’s goes, so will that process. All that will be left is cleaning.
A sturdy hike in the minimum wage, in today’s economy, is basically an industrial policy subsidising the transition to low-skill service industry automation.
This article was originally published on John Cochrane’s blog.