Chicago Federal Reserve President Goolsbee warned at a Stanford Hoover Institution conference that widespread market expectations for AI productivity could itself drive up interest rates and lead to stagflation when the technological boon fails to materialize.

Written by: Bao Yilong

Source: Wall Street Journal

The Chicago Federal Reserve President questioned the narrative surrounding AI productivity, challenging the rate-cut rationale of the Trump administration and the incoming Federal Reserve chairman.

On Friday, Chicago Fed President Austan Goolsbee warned that the broad expectations for artificial intelligence productivity could in itself push up interest rates. If this technological revolution disappoints, the results could be even worse: stagflation.

Goolsbee stated at the annual monetary policy conference of the Hoover Institution at Stanford University, "The greater the hype, the greater the danger."

He cited survey data from the Chicago Fed indicating that economists, technology professionals, and the general public all expect an additional annual productivity growth of about one percentage point over the next decade.

This widespread expectation poses a risk of overheating in the economy. This statement challenges the narrative of "AI-driven rate cuts" promoted by Walsh, who is set to succeed as chair of the Federal Reserve, and the Trump administration.

According to reports, Walsh is expected to be confirmed by the Senate on Monday as the 17th Federal Reserve Chairman. He has previously stated that AI will unleash "the biggest wave of productivity improvement in our lifetime" and characterized it as a "structural disinflation" factor, implying the Federal Reserve has more room for rate cuts because of this.

U.S. Treasury Secretary Yellen shares the same view, likening the current situation to "the budding stage of a productivity boom, reminiscent of the 1990s."

Expectations Themselves Are a Risk

Goolsbee's main argument is that the macroeconomic effects of productivity improvements depend on whether they come as a "surprise" or are "anticipated."

He explained that when productivity improves beyond expectations, inflation decreases, and interest rates can fall accordingly. However, when the market has fully priced in the technological boom, as the current enthusiasm for AI is widely reflected in financial markets and corporate balance sheets, households and businesses may prematurely ramp up spending and investment before productivity genuinely materializes.

This "over-drawing the future" behavior will result in overheating in the economy, thereby pushing up interest rates.

He cited the tech boom of the 1990s as an example, pointing out that the Federal Reserve, led by then-Chairman Greenspan, actually raised interest rates six times between 1999 and 2000 to address the pressures arising from the anticipated productivity-driven demand release.

Goolsbee noted that referencing the 1990s analogy for rate cuts by Walsh and others is "somewhat difficult for me to understand."

If AI Fails, Stagflation Risk Emerges

When pressed by former St. Louis Fed President James Bullard on what would happen if AI productivity expectations fail, Goolsbee provided a more dire assessment.

He stated that if the market continues to expect prosperity and continually overdraws consumption and investment, and if the technological dividends ultimately do not materialize, the economy will fall into a recession against the backdrop of overheated demand and persistently high inflation. He said: You easily get stagflation; it's not a bubble; it's fundamentals.

Goolsbee also listed several leading indicators he is monitoring:

• The wealth effect reflected in housing prices driving consumer spending;
• The surge in data center construction raising land and chip costs, which has already spilled over into industries unrelated to AI;
• And changes in the number of workers exiting the labor force due to expectations of future wealth increases.

Internal Discrepancies: Other Voices Question This Logic

Goolsbee's judgment is not without dissent. At the same forum, Federal Reserve Governor Waller countered his core argument.

Waller stated that the wealth effect channel described by Goolsbee "has existed in many models for a long time," but has "not persisted in real data." He added that if real-world factors such as households' difficulty in easily mortgaging future income or gradual adjustments in spending are considered in the model, that effect would diminish significantly.

Atlanta Fed visiting scholar Steven Davis also raised concerns from another dimension. He cited recent analyses from the Atlanta Fed indicating that the average AI investment by businesses is 14 times the median, suggesting that this investment boom is highly concentrated in a few firms rather than widely spread.

University of Chicago economist Luigi Zingales presented another perspective. A New York Fed survey shows that an increasing number of residents expect to become unemployed due to AI, which could actually raise savings rates rather than boost premature consumer spending.

This runs contrary to Goolsbee's concerns. Goolsbee himself acknowledged that this dynamic could indeed point to the opposite conclusion.