Citrine fiasco exposes a market that’s looking for an excuse to fall

Unlock Editor’s Digest for free

FT editor Roula Khalaf selects her favorite stories in this weekly newspaper.

AI-Doomer Fan Fiction FTW

Yesterday, following recent patterns, shares of various companies fell that could be sensitive to AI disruption – software, asset management and, joining the party for the first time, banks. The most popular explanation for the renewed panic was blog post From Citini Research on how AI could put high earners out of work and hurt the economy. From FT:

Investors have recently seized on social media rumors and incremental growth by smaller AI companies to justify further selling, with a widely circulated blog post by Citrini Research over the weekend detailing how AI could hypothetically push the US unemployment rate above 10 percent by 2028 proving to be the latest catalyst.

The most important thing about the post is not what it says. That’s because the stock market has reached the point where blog posts matter stock movesOr at least where people think they do. It is often true that markets sell off for reasons that are unclear or unknown, and people like me have to hunt around for an explanation. This could be one of those cases. Either way, the Citrine fiasco is evidence that we are in an expensive market that is looking for an excuse to fall, for reasons that are probably much broader than AI.

But what attracted the market’s attention to that post? It takes a perspective several years into the future, allowing it to reflect on recessions and financial crises that have already occurred. “What follows is a scenario, not a prediction… The sole purpose of this piece is to model a scenario that has been relatively little explored,” it begins. But the opposite is true. The post became popular because it is an extreme version of one of the most explored scenarios of all time: robots turning on their creators. It has been explored at least since Karel Capek introduced the term robot and predicted at that very moment that robots could cause a lot of trouble. This happened more than a hundred years ago. More than a century before that, some weavers in Nottingham, under the banner of the fictional General Ned Ludd, had explored the subject by breaking into some textile mills.

On the logic of the post. The central scenario is that AI destroys jobs and incomes, but does not replace them, sending economies into recession and markets into crisis:

Computer owners saw their wealth explode as labor costs disappeared. Meanwhile, real wage growth collapsed. . . White-collar workers lost jobs to machines and were forced into lower-paid roles. . . When cracks began to appear in the consumer economy, economic pundits popularized the phrase “ghost GDP”: output that appears in the national accounts but does not circulate through the real economy. . . The velocity of money became stable. The human-centered consumer economy, which accounted for 70 percent of GDP at the time, ended. . . AI capabilities improved, companies needed fewer employees, white-collar layoffs increased, displaced workers spent less, margin pressure drove companies to invest more in AI, AI capabilities improved. . . It was a negative feedback loop with no natural break. . .

The irony of this was that the AI ​​infrastructure complex continued to perform even as the economy it was disrupting began to deteriorate. NVDA was still posting record revenues. TSM was still running at 95 percent+ utilization. Hyperscalers were still spending $150-200 billion per quarter on data center capex. Economies like Taiwan and Korea that were completely convex to this trend largely outperformed.

Something struck me as strange about this picture, although (due to my non-existent formal education in economics) I struggled to articulate what it was. This sounds like financial commentary that considers only one side of the trade (“Stocks went down because sellers outnumbered buyers”, etc.). Citrini depicts a world of massively rising productivity coupled with declining consumption. Did you understand?

Joseph Steinberg, an economist at the University of Toronto, helped me understand that something was wrong. “I expect my economics students to make the first part of the argument about ‘ghost GDP,'” he told me. What does it mean for output to “appear in the national accounts but not circulate in the real economy”? If GDP is rising – all those robots are making stuff faster and faster – then there’s something on the other side of the national account. Identification Will also have to increase. The possibilities are consumption, investment, government spending or net exports. In the Citrine scenario, consumption is falling rapidly. So is government spending increasing (depending on what is actually taxed or borrowed from)? Or export – to other countries going through the same crisis?

None of this makes much sense, so the investment is abandoned. But Steinberg points out that investment is only sustainable on the basis of future consumption; “Investing in AI only makes sense if there is income to buy these things generated by AI”.

It seems to me that there must be a distributive element to this argument that Citrini does not fully bring out. Income and consumption increases, but somewhere along the way there is no benefit to your average American white collar worker. Perhaps this is all “in economies that were purely convex to this trend, such as Taiwan and Korea”. Or maybe it all goes down to some very annoying Silicon Valley giants. In this scenario, we all get very, very expensive lattes for Peter Thiel. Or perhaps Thiel’s assets have been confiscated by the government and redistributed. But it’s hard to see how the Thiels of the world Citrini envisions can remain rich over the long term, as their AI investments stop paying off as consumption declines.

None of this is to say that I believe the AI ​​revolution, like previous tech governance changes, will ultimately create more jobs and make us all richer. Not me. What I am claiming is that Citrini’s description of the risks seems inconsistent (perhaps readers with a better understanding of macroeconomics can understand this better?). This post went viral not because it explains the current situation, but because it speaks to ancient fears.

a good read

“A fire chief may have rights “Closing roads during an emergency, but that doesn’t mean the fire chief has the authority to impose road tolls.”