AI and why the working week won't reduceThis analysis is provided by Eden Bradfeld at BlackBull Research—sign up for their Substack to receive the latest market insights straight to your inbox.
Tinder, Bumble and so on were once feted as the “new thing”. Here’s how Bumble is doing now.
That’s — not great! That’s pretty bad! The world moved on from dating apps, by and large — dating app consumption is actually down as Gen Z prefers to meet in person. Textile mills of the 21st century. So-long, and thanks for the fish.
There’s two things I’m sure of here:
People will not work less. This has been proven throughout history.
Many currently high-margin, stable businesses will not be are stable or as high-margin.
One of the great economic fallacies is that of optimism — specifically, that the working week will reduce. Here is Keynes, in 1930 —
We may be able to perform all the operations of agriculture, mining, and manufacture with a quarter of the human effort to which we have been accustomed.
Keynes was writing in the wake the Great Depression — it’s fairly remarkable foresight, as the US embarked on several golden decades — $1.00 invested in the S&P in 1929, at the peak of economic gloom, would be worth around $7,622 — you’d have an inflation adjusted return of 41,690.91%. Ne bad, as they say in Scots.
But here’s where he’s wrong — he had hoped for a quarter of human effort — predicting a 15 hour workweek. That hasn’t happened. If anything, the work culture in America and many western countries has become something of a religion — work hard and glorify it. That work has transmuted for many of us from factory jobs and field labour to office jobs and such, but it remains work — we are there to create a surplus of capital, as Marx wrote long ago.
History doesn’t rhyme but it repeats — similar suggestions of the end of work have been made with the advent of AI. Now, it is likely that AI will be able to replace many jobs — especially those that were traditionally protected (you probably don’t need a lawyer to draft up a basic contract, etc…). If we look at the various other revolutions, though, especially the industrial, what we find is that work ends up being something else.
What might it be? Will we have offices filled with people slaving away to Chat GPT, typing in prompts at their terminals? Essentially, will we become part cyborg, delivering commands to our AI counterparts?
It’s interesting to think about what this will do economically. The Industrial Revolution saw vast progress and economies expand rapidly — areas like the North of England, which were traditionally poor, saw riches prosper, while the old class of aristocrats found themselves taxed by both lack of economic progress and real taxes,³ which saw the economic picture turn — at least for a while. And yet — even those economic realities change — the once-rich textile barons of the Industrial Age, with factories in France and England, saw their businesses fall into disrepair as the world moved on. Automated looms, once cutting-edge, found themselves surpassed.
Here’s another example, Chegg Inc, which makes study tools. Of course, Chat GPT has surpassed that and tends to do a better job. Just ask your teens.
That’s also — not very good!
Let me now think about industries that we all think are safe but may be disrupted (don’t you hate that word?) — lawyers, accountants, coders. Uh oh. Whatever happened to “just learn to code, bro”. What happens to the “big four” accounting firms when AI gets good enough to perform most of the functions?
Let’s invert — what are companies and industries that (should) remain impervious:
Luxury — Hermes specialises in the handmade, and that’s part of the brand. The human desire for scarcity and to signal status has not changed in all of history.
Toll-booth businesses — think exchanges (NZX, CBOE, LSE), literal toll booths (Channel Infrastructure), payment operators (Visa, Stripe, etc).
Companies which command mindshare — CostCo, Amazon, etc.
Booze. Duh. AI doesn’t drink booze; humans do.
