Trump’s Fed pick Kevin Warsh and AI optimism clash with economists’ doubts

In Washington’s debate over who should replace Fed Chair Jerome Powell, President Donald Trump has pinned his hopes on productivity—arguing that artificial intelligence could help the economy grow faster without stoking inflation. Trump’s Treasury secretary, Scott Bessent, has promoted the idea that a “Greenspan-like” approach could revive an era when businesses benefited from technology and when unemployment fell while inflation stayed contained. The administration has moved toward that vision with Trump’s choice of Kevin Warsh to lead the central bank.

The case Trump and his team have been making relies on how they read history: that then-Fed Chair Alan Greenspan helped fuel the late-1990s boom and kept interest rates from rising despite pressures in wages. In the AP reporting, Warsh is portrayed as aligned with that approach, having argued in speeches and writings that improvements in productivity tied to AI could justify a more dovish stance on rates. The argument is built around the relationship economists often describe between productivity and inflation—where workers can produce more per hour, allowing firms to raise pay without necessarily pushing prices higher.

But the same reporting highlights why skeptics say the analogy may not fit the present. Economist Dario Perkins said the Trump administration was offering what he called “a rather distorted version of what actually happened in the 1990s.” Perkins argued that the 1990s story Trump’s team is telling is incomplete at best, citing economists’ view that the period’s inflation control did not rest on productivity alone. The reporting also places the debate in a different global and policy environment than the one Greenspan faced, with trade and immigration dynamics now described as far less favorable to the kind of disinflation that characterized the 1990s.

Skeptics also question whether today’s productivity gains—at least those visible in 2025—can be attributed primarily to AI. Joe Brusuelas, chief economist at RSM, wrote that the 2025 productivity improvements “are not because of artificial intelligence” and instead reflect investments in automation made when companies could not find enough workers during and after the COVID-19 pandemic, and that “those investments are starting to pay off.” Other economists, the story says, argue that it may take time for AI to affect how companies operate and how much productivity rises.

The Federal Reserve’s own officials have also offered caution against translating a productivity boom directly into rate cuts. Federal Reserve Gov. Michael Barr, in remarks reported by AP earlier this month, said a productivity boom can raise the economy’s “speed limit” for growth without pushing prices higher, but might not justify lowering policy rates. Barr warned that if businesses borrow to invest in AI, that could put upward pressure on interest rates, and that workers and households could save less and borrow more in anticipation of higher wages—further pressuring rates upward. In Barr’s framing, “The AI boom is unlikely to be a reason for lowering policy rates.”

Similarly, Austan Goolsbee, president of the Chicago Fed, suggested that the late-1990s analogy may be harder to apply than Warsh’s supporters believe. Goolsbee said that “the analogy to the late 90s is a little harder for me to understand,” and explained that Greenspan’s insight was about holding off on raising rates when productivity gains meant inflation pressures were manageable—not that the Fed should cut rates because productivity growth was higher. Goolsbee also emphasized the point in terms of what the committee should take from Greenspan’s approach, setting up what the AP story describes as a potential clash if the Senate confirms Warsh as chair.

The dispute is also shaped by how Warsh fits into the Fed’s interest-rate politics. The AP reporting describes Warsh as a former inflation hawk who, when he served as a Fed governor, objected to aspects of the central bank’s efforts to support the economy after the 2007-2009 Great Recession, even while unemployment exceeded 9%. In contrast, the Trump administration’s current pitch ties Warsh to a more permissive rates outlook if productivity improvements from AI can contain inflation.

That contrast extends to the way proponents recall the Greenspan era. Perkins, in the AP story, said that Warsh and Bessent talk only about the “dovish 1995/96 version of Greenspan” while overlooking the “hawkish 1999/2000 variant,” which included the Fed raising its benchmark rate from 4.75% to 6.5% in less than a year. The AP reporting also notes that the rate Trump complains about now is around 3.6%, underscoring how the present debate hinges on whether the productivity-to-inflation link can again justify a softer policy path.

The AP story places the argument against a backdrop of shifting macroeconomic conditions. It says Greenspan avoided rate hikes when government surpluses reduced the pressure to borrow, while today deficits are described as piling up and the Congressional Budget Office expects federal debt to reach 120% of U.S. gross domestic product by 2035. It also points to the 1990s period of trade liberalization and rising immigration as factors that helped keep inflation contained, while describing that trend as now having given way to “de-globalization.” Michael Pearce of Oxford Economics is quoted saying, “That benign era is clearly behind us,” while Perkins wrote that “Trade barriers are going up” and that “Globalization has given way to de-globalization.”