Photo: Masood Aslami / Pexels
The ECB governing council member flagged that whether AI drives demand or productivity first will shape monetary policy for years to come, and cautioned that tech stock valuations look dangerously optimistic.
Fabio Panetta, Governor of the Bank of Italy and member of the ECB Governing Council, laid out a framework for thinking about AI and inflation that boils down to a deceptively simple question: does AI make people richer before it makes companies more efficient, or the other way around?
Speaking at a research conference co-hosted by the National Bank of Ukraine and Narodowy Bank Polski in Kyiv on September 21, Panetta argued that the sequencing of AI’s economic effects will determine whether the euro area faces persistent inflationary pressure or a disinflationary wave driven by automation. The distinction matters enormously for how the ECB sets interest rates in the years ahead.
The demand-versus-productivity dilemma #
Panetta’s core argument splits AI’s macroeconomic impact into two channels. The first is demand-driven: AI creates new jobs, lifts expected labor income, and puts more money in consumers’ pockets. More spending chases the same goods, and prices rise. The second channel is productivity-driven: AI automates tasks, reduces headcount, and lowers production costs. Output expands without a proportional increase in wages, and prices fall.
If the demand effects arrive first, inflation could prove stickier than central bankers would like. If automation leads, the opposite happens. Consumption weakens as displaced workers pull back spending, and disinflation sets in. Panetta acknowledged that the timeline and scale of AI’s productivity benefits remain “highly uncertain.” A Bank of Italy study he referenced estimated that widespread AI adoption could boost Italy’s labor productivity growth by more than 1 percentage point per year.
Macro, rates, and crypto—what moved markets and what matters next.
Daily. Free. Join 34,000+ readers across crypto, finance, and policy.
The uncertainty is compounded by the fact that capital investment in AI itself generates demand. Companies pouring money into computing infrastructure and energy to power data centers create short-term spending boosts in those sectors. That capital expenditure can elevate relative prices and feed into broader inflation until supply catches up.
A warning shot at tech valuations #
Panetta warned that current technology stock valuations reflect “overly optimistic expectations” about AI profitability. He suggested that the gap between what markets expect AI to deliver and what it will actually produce in near-term profits leaves equity prices vulnerable to sharp corrections, noting recent fluctuations in tech stock prices.
What this means for monetary policy #
The practical implications for ECB policy are significant. If Panetta’s demand-first scenario plays out, the central bank may need to maintain higher interest rates for longer than markets currently expect. Persistent inflation driven by AI-fueled income growth would make rate cuts harder to justify. The automation-first scenario presents a different challenge: rapid displacement of workers could weaken consumer spending and push inflation below the ECB’s 2% target, forcing the central bank into an easing cycle.
Panetta’s framework also highlights a tension that policymakers will need to manage carefully. Governments eager to capture AI’s productivity benefits may accelerate adoption through subsidies and deregulation. But if those policies amplify the demand channel before productivity gains materialize, they could inadvertently make the ECB’s inflation-fighting job harder.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our