At Quartz Cris Tolomia reports on a joint statement released by 200 economists:
More than 200 economists and researchers, including 16 Nobel laureates, released a joint statement on Monday warning that artificial intelligence could reshape the economy at a speed and scale exceeding the Industrial Revolution, and calling on policymakers and technology leaders to begin building policies and institutions to address the disruption.
The statement, titled “We Must Act Now,” warns that AI “could bring risks, including large-scale job displacement, as well as opportunities such as major gains in living standards.” Among its core demands, the statement urges economists, policymakers, and technology leaders to expand their understanding of how AI is reshaping the economy and to develop guardrails ensuring the technology augments rather than displaces human workers.
The statement’s significance lies partly in who signed it. Erik Brynjolfsson, a Stanford economist who helped organize the effort, said there has been “a notable change in the profession,” according to The New York Times. The economics profession has long pushed back on warnings of swift AI-driven displacement, with most researchers arguing that the timeline for technological disruption is routinely overstated. Among those who put their names to the document are Daron Acemoglu and Simon Johnson — both MIT professors and 2024 Nobel economics laureates — whose earlier public skepticism about AI’s disruptive potential made their participation particularly striking, according to the Times.
with this punchline:
The statement arrives as white-collar payrolls have contracted for dozens of consecutive months, a stretch that Aaron Terrazas, a former chief economist at Glassdoor, has called without precedent outside of a recession.
The situation is even more dire in India (although it’s hard to ferret that out of the official statistics).
The timing is noteworthy. Whether or not AI is already replacing large numbers of workers, corporate managers increasingly appear to believe that it soon will, and that expectation is beginning to influence hiring decisions.
OpenAI and Anthropic are not building most of the world’s new computing infrastructure. The principal investors are the hyperscalers—Amazon, Microsoft, Google, Meta, and others. Their cloud businesses supply AI computation to everyone else. Those firms also happen to account for much of the stock market’s recent gains. And those firms are likely to be the greatest beneficiaries of artificial intelligence.
It may be the case that LLM AI will result in creating more jobs than it destroys. As John Maynard Keynes quipped, in the long run we’re all dead. Workers displaced over the next five years receive little comfort from predictions about the labor market twenty years from now. In the shorter term managers of large companies will see trimming their payrolls as a strategy for boosting their stock values. Whether those productivity gains ultimately materialize is almost beside the point. If executives believe AI allows them to operate with smaller staffs, they have every incentive to reduce payrolls now and explain the decision to investors later.
We are in the fifth month of little or no exports of oil, gas, nitrogen fertilizer, sulfur, helium, and aluminum from the Persian Gulf. The Gulf normally supplies 20% of the world’s oil supply and 33% of its nitrogen fertilizer. But the world’s economists think AI is the problem, just when some AI companies are shelving data center projects and local authorities are beginning to ban them. And there is no electricity to run them.
Economic output is more or less proportional to energy consumption, so losing 20% of the oil and gas supply will lead to a significant reduction in economic activity, especially transportation services, which would result in a major recession. The loss of one-third of the fertilizer supply, right at the beginning of the northern hemisphere planting season will lead to major reductions in crop yields this fall, resulting in very much higher food costs for everyone and famine for some.
Yet AI is the problem.
Later this fall, can we please shut down every economics department and every business school and put their faculty, students, and graduates in prison, preferably in one of Russias unused Gulags?
PS. The US military is not only running down its missile inventories (now significantly less than half the starting quantity a year ago), it has also nearly emptied the strategic oil reserve and is running out of diesel oil and jet fuel, which we largely import.
For those of us (“we few, we happy few,…”) who want a demilitarized America, the exhaustion of the US’ military is a good thing, even if not optimal.
“In the shorter term managers of large companies will see trimming their payrolls as a strategy for boosting their stock values.”
I think that’s materially overstated. In any reasonably competitive market productivity gains are fleeting, and competed away fairly rapidly. The Holy Grail of increased equity value is growth.
“…calling on policymakers and technology leaders to begin building policies and institutions to address the disruption.”
Sounds great. But how. By fiat? A crystal ball? The wise men of the Senate? (snicker)
“…corporate managers increasingly appear to believe that it soon will, and that expectation is beginning to influence hiring decisions.”
It seems to be working both ways. Everything we hear or see (the Firm) is that its well underway, often ill thought out, and rehiring is significant when things don’t go as advertised. I’m no AI expert and don’t know how things will ultimately shake out. But institutions will muddle through and mistakes will be made. Same as it ever was. How long ago was it that certain people disdainfully told coal miners to learn to code?
“Later this fall, can we please shut down every economics department and every business school and put their faculty, students, and graduates in prison, preferably in one of Russias unused Gulags?”
If you could kindly indulge me, Bob, I’d rather not spend my later years in prison. And remember, next time a bridge collapses or there is an environmental disaster they might come looking for you.
There are many sectors in which we have never had “reasonably competitive markets”, others in which consolidation has rendered wat were previously competitive markets non-competitive. Behavior in those sectors is more comparable to that of cartels now.
I certainly agree with fighting monopoly or near monopoly, Dave, although in the past 50 years I suspect that the “never” aspect is fairly rare. (Maybe mostly first movers; then people want in on the game.) If we go back to “ancient” times, we had the railroads, steel, oil and newspapers etc as classic examples of monopoly seeking. (and today, Warren Buffet in rail and other businesses) I worry most about lobby/regulatory issues affecting competition. But that’s a separate discussion.
I readily admit that my first-hand experience is in $500MM revenue and below companies, not goliaths, but that’s a lot of GDP, and job creation. Competition is fierce, and consolidation is generally a survival response to that, not monopoly seeking. But a 3-4 player (plus bit players) market is deep enough to keep you awake at night. And beware the 2 player market. It sounds nice, but it is generally a knife fight in a closet, not a license to make margin.
Today? I worry about monopoly in other sectors or venues. Media and academia immediately come to mind. Government services of course. Health care.
A last point: some companies have a “last man standing” strategy. They make a product that has declining unit demand, and they have it. They charge through the nose for it; huge margins. But enterprise valuations? In my world 5-6 x earnings. But a company making half the margin but growing 10%+ a year? 10-15x valuations.
Equity valuation is driven by perceived growth potential. Period, full stop. You don’t cut staff to make stock prices rise, at least not for long.
To be honest, I’m more concerned about the government’s response to the impact of AI than I am about AI itself.
It doesn’t need to be for long. Just up.
Dave
Perhaps that’s our point of departure. In my world, filled with very sophisticated and experienced investors, no one buys into temporary gimmicks to boost equity valuations. If one does, you just made a bad investment.
Maybe in your world, as I understand it consulting with large corporations, the retail or herd investor might buy into such things, at least for a while. I don’t believe it, but it might be your experience.