Not long ago, I flew into a small Ohio airport. Don’t ask why, sometimes Ohio just happens and you have to deal with it. Arriving late (having flown Delta: “Delayed Every Last Time Always”), I didn’t get in until after 11 pm. And I was worried about picking up my rental car, because all the car company desks were dark.
Then my phone vibrated; I had a notification on the Hertz app: “Slot B17, code 2946,” and some instructions. I went to a box, entered the code, and removed my key from the B17 door. My car was in parking spot B17, and I got in the car and drove off.
In 2011, Marc Andreesen famously said “Software Eats the World.” His claim was that the “problem is even worse than it looks because many workers in existing industries will be stranded on the wrong side of software-based disruption and may never be able to work in their fields again.” Well, my experience in Ohio made me wonder: have the folks who once worked at the Hertz counter found new jobs? Their experience was in public-facing retail, but those jobs are disappearing fast.
You’ve seen it in other industries. Once you walked up to the fast food counter and read some words from the menu board out loud. Giant burger, enormous fries, vat of drink, a combo meal. The person behind the counter then looked for the corresponding printed words on his cash register. Unsurprisingly, software ate this world. All you have to do is turn the cash register around. You can press those buttons yourself on a kiosk touch screen. You can enter the payment yourself, and get your own receipt, because software is recording, charging, and transmitting your order to more software on the back line of the restaurant.
At grocery store, self-checkout is now a replacement for a cashier. When I was growing up, the only place you could “check yourself out” was a mirror, but now some stores don’t have any cashiers at all except for handicapped customers who make a special request.
The coronavirus pandemic also contributed to the decline in service jobs, of course, so we can’t attribute all the employment loss to software. But overall, the number of people working service jobs, and the number of hours worked by those who have jobs, have fallen steadily since 2019.
The overall problem can be put in the form of an analogy, the kind you used to work on when you were studying for the SAT:
[software] is to [_________]
[robotics/automation] is to [manufacturing].
The correct answer to the “fill in the blank” is service jobs. In other words, where in the 1960s and 1970s we were all talking about automation causing “technological unemployment,” the new low-hanging fruit for job replacement is service jobs. Bizarrely, the solution of many current political analysts is that the solution is to raise the minimum wage, which disproportionately affects service work.
Let’s think about that logic for a second.
Service jobs are falling because software is replacing human workers at an extraordinary rate. We all see it in some areas — groceries, rental cars, fast food, movie tickets, and so on — but it is also happening in other areas that we don’t see — accounting, information requests, scheduling, reservations, almost any service that can be routinized. What will happen if we try to fix this problem of disappearing jobs with an increase in the minimum wage?
You don’t have to have a PhD in economics to be able to predict the consequences. But I do have a PhD in economics, so let me give it a shot:Even if it “works,” and does not reduce employment at all, there are still many dimensions on which a minimum wage makes work uncomfortable for the least-well-off, because of competition along other margins. The best interpretation one can attach to the minimum wage’s effect, in the period before software started to eat the world, was that the increase in unemployment was less than you might expect. But it still caused reduced employment or reduced hours, much of the time and for any large increase. The reduction in the amount of work, especially hours, for those who are still employed, has been much more pronounced in recent years. This is particularly true in areas (such as Seattle) where the bump in hourly pay wage was large enough to create a “living wage.” Unsurprisingly, the effects are to 1) raise wages for those who still have jobs (the law requires that), and 2) cut the number of jobs, and the number of hours worked (the laws of economics require that). Even if there were a large positive effect on wages and no effect on unemployment, the increased prices for products and services produced by minimum wage workers would significantly harm the least-well-off, out of proportion to the assumed benefit, which is dubious in the first place.
Recently, I wrote about the conflict between directionalists and destinationists. Some destinationists — such as my good friend Richard Salsman — would argue that the problem with the minimum wage is that it interferes with freedom of contract.
But you don’t have to buy that argument to be persuaded that raising the minimum wage right now is a terrible idea. If you care about the people struggling to keep their jobs in a difficult economy, you should oppose raising the minimum wage: it hurts the very people you want to help. Being able to work, and feel productive, is an important part of the social aspect of our economy. Service jobs are disappearing fast enough, without having the process accelerated by misguided support for increasing the minimum wage.