1996 or 1999? Wash's first test is "How to view AI"

Null

Written by: Dong Jing

The primary challenge Waller faces as Federal Reserve Chair is not whether to raise or lower interest rates, but a more fundamental judgment: what kind of prosperity is the current AI boom? This decision will determine the Fed’s policy direction and also define Waller’s place in history.

On June 19th, Nick Timiraos, a reporter known as the “New Federal Reserve News Agency,” stated that there are two starkly opposing interpretations within the economics community regarding the AI construction boom:

First, productivity dividends are about to materialize, supply will catch up with demand, and the Fed can hold steady, waiting for inflation to naturally recede; second, the benefits of productivity improvements are still distant, while demand shocks have already arrived. If the Fed waits for data confirmation, it risks missing the optimal intervention window and may be forced to raise rates more aggressively.

The Fed kept rates unchanged this week, but in the latest dot plot, nearly half of officials still expect rate hikes within the year, while others hold the opposite view—this deep internal division reflects the high uncertainty surrounding this core issue.

Waller’s own inclinations are subtly visible during the press conference. He repeatedly emphasized that “strong productivity-driven growth is not something we fear, but something we embrace,” echoing Greenspan’s thinking from 1996.

However, the macro environment he faces—tariff pressures, expanding fiscal deficits, waning globalization benefits—is far from Greenspan’s smooth sailing back then. Making the right judgment between these two historical scenarios will be Waller’s first real test as Fed Chair.

Two 1990s: The Dual Legacy Left by Greenspan

Timiraos notes that Waller has repeatedly cited the 1990s as a historical reference over the past year, but that decade itself contains two entirely different stories.

In 1996, Greenspan faced rapid economic expansion and chose to hold steady. He judged that fast growth would not ignite inflation, and history proved him right. The economy continued to expand for years, earning him the reputation of a “master.”

By 1999, Greenspan changed his stance. With the stock market soaring and the labor market tightening, he began raising rates consecutively, culminating in the bursting of the dot-com bubble. It was also in that year that the Fed established the forward guidance mechanism—signaling rate hikes in advance—which continues today and is something Waller has explicitly expressed a desire to abolish.

The Trump administration publicly admired the 1996 version of the Fed. Waller, before taking office, also publicly stated his hope to build a “confident, cautious” central bank. But the current economic environment may be handing him a different script.

Waller’s Logic: Believing Narratives, Not Waiting for Data

Before taking office, Waller publicly expressed on Fox Business that he fears the Fed might make its “sixth or seventh major mistake”—prematurely tightening monetary policy during a productivity boom that should be allowed to run its course.

Timiraos states that his core argument is: the productivity gains from AI will not immediately show up in official statistics and may take years to materialize. If the Fed insists on waiting for data confirmation, it risks mistaking a healthy boom for an overheating economy and raising rates—precisely the move that could stifle the growth momentum that could otherwise help contain inflation.

The essence of this logic is advocating for forward-looking narratives to replace lagging data as the basis for decision-making. Waller also continued this line during the press conference: when asked whether AI is currently boosting demand or expanding supply, he only said “demand is easier to measure than supply,” deliberately avoiding a clear stance, while adhering to the principle of “not revealing the next move prematurely.”

Timiraos believes that even if Waller’s judgment proves correct, the analogy to the 1990s is incomplete.

When Greenspan made that famous bet in 1996, there were multiple tailwinds: cheap imports and labor from abroad kept inflation low, and federal deficits were narrowing. These structural factors provided additional safety margins for the Fed’s “wait-and-see” approach.

Waller faces a very different environment: tariffs pushing up import costs, expanding rather than shrinking fiscal deficits, and the retreat of globalization benefits. This means that even if AI productivity dividends eventually materialize as expected, the inflationary pressures Waller endures during the waiting process will be much greater than Greenspan’s experience back then.

Counterargument: Chicago Fed’s “Expectations Overdraft” Model

Timiraos points out that the most systematic challenge to Waller’s logic comes from Chicago Fed President Austan Goolsbee.

According to the Wall Street Journal, Goolsbee, at a conference at Stanford last month, made a key distinction: whether productivity booms can keep the central bank on hold depends on whether the boom exceeds expectations. A boom that everyone anticipates can have the opposite effect—people will pre-spend future wealth, significantly increasing spending before the productivity dividends are realized, leading to overheating.

“Eventually, you’ll have to raise rates sharply, much more than you would have if you acted earlier,” Goolsbee said.

He believes that the current AI boom falls precisely into this “everyone can see it” category. Surveys of economists, tech workers, and the general public all show that markets generally expect AI to bring about about a 1% productivity increase annually, with most benefits still in the future. According to his model, this expectation itself justifies rate hikes rather than cuts.

Goolsbee also pointed to real-world “overheating signals”: AI data center construction is driving up land, electricity, and chip prices, while increasing costs for electricians and equipment, squeezing resources in other industries. Apple’s announcement this week of price hikes due to rising costs is evidence of this mechanism at work.

It’s worth noting that Goolsbee’s framework is not without challengers. Fed Governor Christopher Waller, at the same Stanford conference, argued that the “expectation overdraft” mechanism can work only if people can borrow to pre-spend. But in reality, many households are tightly constrained by current income and cannot easily convert future wealth into present spending.

“If they can’t overdraft that part of their spending, the whole mechanism is cut off,” Waller said.

This rebuttal provides theoretical support for Waller’s “hold steady” stance: if borrowing constraints are widespread, demand pre-spending effects will be limited, and productivity booms are more likely to gently expand supply without triggering inflation.

The Ultimate Paradox: Abolish Forward Guidance or Be Forced to Use It

Additionally, Timiraos suggests that Waller faces a deeper paradox—one that stems precisely from what he most wants to change.

He explicitly states he wants to create a Fed that “does not reveal its hand in advance,” reducing forward guidance to keep markets guessing. Yet, the current forward guidance mechanism was established in 1999—when Greenspan began signaling rate hikes in advance to avoid market surprises.

If the economy unfolds as the Trump administration describes optimistically, Waller might never need to give advance signals. But if the economy takes another turn, he will face a dilemma:

Either continue the practice he hopes to abolish—giving early signals of rate hikes; or remain silent, letting markets guess the size and timing of hikes, risking sharp market volatility.

The resolution of this paradox ultimately depends on the same question: is it 1996 or 1999?