Steam-powered looms, railroads, the internet: Technological breakthroughs such as these have powered economic growth by increasing efficiency, enabling people to make more with less.
Today, economists don’t know whether artificial intelligence will displace labor or augment it — whether the technology will make workers more efficient in their jobs or replace them. Either way, output per hour — the country’s productivity — is likely to rise. Economic theory and history suggest that one consequence will be slower inflation, even with faster economic growth. Businesses that make more with less face reduced pressure to raise prices to preserve their margins.
If this reasoning sounds familiar, it could be because Kevin Warsh, President Trump’s nominee to be the Federal Reserve chair, uses it to justify the idea of cutting interest rates, even though inflation is still clocking in above the Fed’s 2 percent target. Anticipating productivity gains from A.I. that will lower business costs, Mr. Warsh claims that the Fed can reduce the cost of borrowing without encouraging inflation.
But Mr. Warsh misses an essential dynamic: A.I.’s potential to increase demand, encouraging business investment and consumer spending, both of which tend to boost inflation. This omission makes his argument that A.I. provides a rationale for rate cuts a lot less convincing.
Mr. Warsh would seem to have some recent economic history on his side. A previous Fed chair, Alan Greenspan, argued in the 1990s that tech-driven productivity increases during that era’s internet boom would enable the economy to sustain lower unemployment without faster inflation.
Back then, economists worried that if unemployment fell much below 5 or 6 percent, in-demand workers would force employers to raise their wages, which would, in turn, feed inflation. Mr. Greenspan, however, saw before others that the internet was making work more efficient. Rising productivity meant companies could lift wages without passing on the cost of those wage increases in the prices people paid for their products and services. Mr. Greenspan’s Fed, therefore, didn’t have to raise interest rates when unemployment started to fall to levels that, absent the productivity increase, might have encouraged inflation.
Today’s economic situation is similar, Mr. Warsh argues. “A.I. will be a significant disinflationary force, increasing productivity and bolstering American competitiveness,” he wrote in The Wall Street Journal last fall.
But A.I. will also create demand, which will push the other way on inflation and borrowing costs.
Companies expect A.I. to raise the return on the investments they make in growing their businesses, so they will increase those investments. A recent report from Moody’s Analytics showed that by the end of last year, the United States had more active and planned data centers than the rest of the world combined. Big American tech firms are expected to invest $700 billion this year. That’s over 2 percent of the country’s gross domestic product. These investment dollars will flow to construction companies building data centers, chip firms and other hardware producers, along with the workers they need to sustain the boom.
Americans who own technology stocks have done well recently, as hopes about A.I.’s future returns have buoyed investor optimism. The richest 10 percent of households saw their inflation-adjusted stock wealth appreciate by $7 trillion last year. Those wealth gains are boosting spending. Economists have found that, unsurprisingly, people who feel richer buy more, putting about 3 percent of their gains into new spending. For last year, that 3 percent translates into over $200 billion in extra spending, equivalent to roughly a fifth of the year’s increase in real spending.
This has happened before. After years of productivity expansion during the 1990s, the Fed ultimately concluded in 1999 that it had to raise interest rates, judging that increased spending had, on net, raised the neutral rate of interest — the rate needed to keep inflation from accelerating. In his confirmation hearing last month, Mr. Warsh acknowledged the potential for A.I. to spur demand but downplayed its impact, saying that any such increase would amount to only “a few tenths of 1 percent.”
But A.I. is clearly lifting investment and consumer spending by much more than that. Moreover, inflation remains persistently elevated, and it would be dangerous for the Fed to take seriously only half the story of A.I.’s potential impact. For now, the Fed’s best option is to hold rates where they are and watch how the rise of A.I. affects both productivity and demand.
Jared Bernstein was the chair of President Joe Biden’s Council of Economic Advisers from 2023 to 2025 and is a policy fellow at the Stanford Institute for Economic Policy Research and the Center for American Progress.
Janet L. Yellen is a distinguished fellow at the Brookings Institution. She was the chair of the Federal Reserve from 2014 to 2018, appointed by President Barack Obama, and she was the Treasury secretary during the Biden administration.
Source photographs via Getty Images.
The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: [email protected].
Follow the New York Times Opinion section on Facebook, Instagram, TikTok, Bluesky, WhatsApp and Threads.
The post Kevin Warsh Is Already Getting It Wrong appeared first on New York Times.
