Pretty much everyone in tech agrees that generative AI, Large Language Models and ChatGPT are a generational change in what we can do with software, and in what we can automate with software. There isn’t much agreement on anything else about LLMs – indeed, we’re still working out what the arguments are – but everyone agrees there’s a lot more automation coming, and entirely new kinds of automation. Automation means jobs, and people.
This is also happening very fast: ChatGPT has (apparently) over 100m users after just six months, and this data from Productiv suggests it’s already a top-dozen ‘shadow IT’ app. So, how many jobs is this going to take, how fast, and can there be new jobs to replace them?
We should start by remembering that we’ve been automating work for 200 years. Every time we go through a wave of automation, whole classes of jobs go away, but new classes of jobs get created. There is frictional pain and dislocation in that process, and sometimes the new jobs go to different people in different places, but over time the total number of jobs doesn’t go down, and we have all become more prosperous.
When this is happening to your own generation, it seems natural and intuitive to worry that this time, there aren’t going to be those new jobs. We can see the jobs that are going away, but we can’t predict what the new jobs will be, and often they don’t exist yet. We know (or should know), empirically, that there always have been those new jobs in the past, and that they weren’t predictable either: no-one in 1800 would have predicted that in 1900 a million Americans would work on ‘railways’ and no-one in 1900 would have predicted ‘video post-production’ or ‘software engineer’ as employment categories. But it seems insufficient to take it on faith that this will happen now just because it always has in the past. How do you know it will happen this time? Is this different?
At this point, any first-year economics student will tell us that this is answered by, amongst other things, the ‘Lump of Labour’ fallacy.
The Lump of Labour fallacy is the misconception that there is a fixed amount of work to be done, and that if some work is taken by a machine then there will be less work for people. But if it becomes cheaper to use a machine to make, say, a pair of shoes, then the shoes are cheaper, more people can buy shoes and they have more money to spend on other things besides, and we discover new things we need or want, and new jobs. The efficient gain isn’t confined to the shoe: generally, it ripples outward through the economy and creates new prosperity and new jobs. So, we don’t know what the new jobs will be, but we have a model that says, not just that there always have been new jobs, but why that is inherent in the process. Don’t worry about AI!
The most fundamental challenge to this model today, I think, is to say that no, what’s really been happening for the last 200 years of automation is that we’ve been moving up the scale of human capability.
‘Barge haulers on the Volga’, Ilya Repin, 1870-73. (Note the steam boat on the horizon.)
We began with humans as beasts of burden, pulling barges up the river, and moved up: we automated legs, then arms, then fingers, and now brains. We went from farm work to blue collar work to white collar work, and now we’ll automate the white-collar work as well and there’ll be nothing left. Factories were replaced by call centres, but if we automate the call centres, what else is there?
Here, I think it’s useful to look at another piece of economic and tech history: the Jevons Paradox.
In the 19th century the British navy ran on coal. Britain had a lot of coal (it was the Saudi Arabia of the steam age) but people worried what would happen when the coal ran out. Ah, said the engineers: don’t worry, because the steam engines keep getting more efficient, so we’ll use less coal. No, said Jevons: if we make steam engines more efficient, then they will be cheaper to run, and we will use more of them and use them for new and different things, and so we will use more coal.
We’ve been applying the Jevons Paradox to white collar work for 150 years.
It’s hard to imagine jobs of the future that don’t exist yet, but it’s also hard to imagine some of the jobs of the past that have already been automated away. Gogol’s downtrodden clerks in 1830s St Petersburg spent their entire adult lives copying out documents, one at a time, by hand. They were human Xeroxes. By the 1880s, typewriters produced perfectly legible text at twice the words-per-minute, and carbons gave half a dozen free copies as well. Typewriters meant a clerk could produce more than 10 times the output. A few decades later, adding machines from companies like Burroughs did the same for book-keeping and accounting: instead of adding up columns with a pen, the machine does it for you, in 20% of the time, with no mistakes.
What did that do to clerical employment? People hired far more clerks. Automation plus the Jevons Paradox meant more jobs.
If one clerk with a machine can do the work of 10, then you might have fewer clerks, but you might also you might do far more with them. If, Jevons tells us, it becomes much cheaper and more efficient to do something, you might do more of it – you might do more analysis or manage more inventory. You might build a different and more efficient business that is only possible because you can automate their administration with typewriters and adding machines.
This process keeps repeating. This is Jack Lemon as CC Baxter in The Apartment in 1960, using an electro-mechanical adding machine from Friden fifty years after adding machines were new and exciting.
