This paper is part of the Fall 2025 edition of the Brookings Papers on Economic Activity (BPEA), the leading conference series and journal in economics for timely, cutting-edge research about real-world policy issues. The summary below was originally drafted based on the paper’s conference draft presented at the Fall 2025 BPEA Conference on September 25-26, 2025. (Conference drafts, recordings, and presentation slides are available on the conference page.) Find all papers in this edition here.
*Final version posted: June 2026
Investment in artificial intelligence, increased government borrowing, deglobalization, and other factors could in coming years push up the neutral rate of interest that guides both monetary and fiscal policy, suggests a paper in the Fall 2025 edition of the Brookings Papers on Economic Activity (BPEA).
The paper, by Lukasz Rachel of University College London, uses a macroeconomic model of supply and demand for savings and wealth to consider where the natural (or neutral) rate, known as r*, is headed over the longer term in advanced economies such as the United States, Germany, France, the United Kingdom, Canada, and Japan.
The natural rate is the real interest rate (the nominal interest rate minus inflation) that prevails over the medium-to-long term when the supply of savings, or capital, from households is in balance with the demand for funds by businesses and government. It cannot be measured directly, but economists can infer it from models that incorporate key forces affecting the capital market equilibrium.
The estimate serves as a guidepost for monetary policymakers, who must judge whether the interest rates they target are stimulating the economy (thus supporting job growth) or restraining it (thus cooling price pressures). Fiscal policymakers also need to think about the neutral rate in estimating the cost of government debt over time.
Real interest rates in advanced economies trended steadily downward for more than three decades, until central banks sharply raised short-term policy rates to tame the post-COVID inflation surge. But long-term interest rates have risen as well, raising the prospect that r* has increased. The paper estimates the current real natural rate in advanced economies for safe assets (such as government debt) at around 0% to 1.2%. That’s down substantially from more than 7% in the early 1980s, but mostly below what the market expects.
The long descent has been driven by several factors, including aging populations and low productivity growth. Increases in life expectancy motivate people to save more for retirement, thus increasing the supply of available funds in the economy, while low productivity expectations decrease the demand for investment. Under Rachel’s “business-as-usual” scenario, these trends continue and the model predicts a gentle half percentage point decline in the neutral rate between now and 2050.
The paper examines several upside-risk scenarios that might plausibly contribute to the reversal of the long-run trend. Artificial intelligence, if it boosts economic productivity as many experts expect, would increase businesses’ demand for capital and reduce households’ desired saving, raising r* steadily, perhaps by 1 percentage point over the next five years. More immediately, investors might push the neutral rate up by a similar amount if they infer from the large recent losses on their portfolios of nominal government bonds that “safe” assets are not so safe after all.
Increased government borrowing to finance Social Security would increase governments’ demand for funds. Deglobalization, fueled by increased tariffs and rising geopolitical tensions, could cause Asian exporters such as China and oil-producing economies to invest less in Western assets. Plausibly calibrated, these scenarios might boost r* by about half a percentage point each.
All in all, a turnaround and a break from the business-as-usual era of low rates is possible, the paper notes. But for r* to return to levels last seen before the global financial crisis, several of the upside risks would need to materialize all at once, which might not be very likely.
Authors
Author
Discussants
Citations
Rachel, Lukasz. 2025. “What Next for r*? A Capital Market Equilibrium Perspective on the Natural Rate of Interest.” Brookings Papers on Economic Activity, Fall: 131–187.
Auclert, Adrien. 2025. “Comment on ‘What Next for r*? A Capital Market Equilibrium Perspective on the Natural Rate of Interest’.” Brookings Papers on Economic Activity, Fall: 188–201.
Del Negro, Marco. 2025. “Comment on ‘What Next for r*? A Capital Market Equilibrium Perspective on the Natural Rate of Interest’.” Brookings Papers on Economic Activity, Fall: 201–210.
-
Acknowledgements and disclosures
Lukasz Rachel is immensely grateful to Antoine Ding, who provided outstanding research assistance. He thanks the editors, Janice Eberly and Jón Steinsson, and the discussants, Adrien Auclert and Marco Del Negro, for extremely helpful and detailed comments and suggestions. He is also grateful to Ben Moll and Morten Ravn, and for the participants at the 2025 Federal Reserve Bank of New York research symposium on r* where he presented some of the early ideas contained in this paper.
The discussant, Marco Del Negro is very grateful to Elena Elbarmi and Michael Pham for excellent research assistance, and to Maurice Obstfeld for helpful feedback. The views expressed in his comments are his own and not necessarily those of the Federal Reserve Bank of New York or the Federal Reserve System.
David Skidmore authored the summary language for this paper.
The Brookings Institution is committed to quality, independence, and impact.
We are supported by a diverse array of funders. In line with our values and policies, each Brookings publication represents the sole views of its author(s).
