Decoding Neutral Rates: IMF Unpacks Euro Area's Equilibrium Interest Landscape

In a detailed working paper from the International Monetary Fund’s European Department, economists Robert C. M. Beyer and Luis Brandao Marques dive into one of the most pressing questions in monetary economics: what is the real equilibrium interest rate in the euro area? Released in June 2025, the study draws on research tools and findings from the IMF, the Federal Reserve Bank of New York, the Bank for International Settlements, and various academic contributions. The study is particularly timely, as inflation in the eurozone returns toward target and the question of where policy rates should settle, neither stoking inflation nor hampering growth, has become central to European Central Bank (ECB) deliberations. Yet this seemingly straightforward benchmark, the real neutral rate, proves to be a theoretical and empirical minefield. The paper showcases eight major methodologies, both from closed and open economy perspectives, to reveal just how diverse and uncertain estimates of this elusive variable can be.

Eight Models, One Monetary Enigma

The authors present updated euro area estimates from a range of models: from statistical filters such as the univariate stochastic volatility model of Beyer and Milivojevic (2023), to dynamic structural general equilibrium (DSGE) models like those by Neri and Gerali (2019) and Zhang et al. (2021). They also explore the well-known Holston, Laubach, and Williams (HLW) framework, Del Negro’s VAR models, and a forward-looking term-structure model used internally by the IMF. Each offers a different lens on what constitutes a “neutral” rate, some targeting long-term structural drivers like demographics and productivity, others focusing on short-term shocks and global spillovers.

The results are anything but uniform. For the fourth quarter of 2024, estimates range from -1.2 percent to +1.7 percent. Long-term estimates, such as those from univariate filters and VARs, tend to show low and stable neutral rates. By contrast, DSGE models, which incorporate more shocks and frictions, produce high-frequency, volatile rates that reflect shifting risk appetites and economic conditions. Neri and Gerali’s model, for instance, showed a short-run equilibrium rate spiking to over 5 percent during the 2022 inflation surge before falling back to 0.8 percent by end-2024. In open economy models, the influence of global factors is especially pronounced. Ferreira and Shousha (2023), for example, find that safe asset supply and rising convenience yields have pushed equilibrium rates upward since 2017.

The Uncertainty Behind the Numbers

One of the paper’s starkest takeaways is the immense estimation uncertainty. Even in the most sophisticated models, standard errors remain wide, sometimes exceeding five percentage points, as in HLW2023. Estimates are highly sensitive to the dataset used, initial values, and model assumptions. Semi-structural models such as HLW2017 are particularly unstable; simply adding or removing quarters from the sample can lead to sharp shifts in the output. On the other hand, univariate and time-varying VAR models offer greater stability, though they lack deeper economic structure. Term structure models, which reflect market expectations, offer cleaner time-specific rates but are deeply influenced by central bank forward guidance and, as the paper warns, risk becoming circular if used by the central bank to inform its own policy stance.

To navigate these issues, the authors recommend evaluating model estimates using three criteria: conceptual fit for the specific policy question (short-term vs. long-term relevance), robustness to data changes, and consistency with real-world macroeconomic indicators. For example, subdued private investment, weak loan demand, and modest fiscal consolidation in the euro area suggest a lower short-term neutral rate than some models imply. Applying this judgment-based filter, the authors estimate that the short-term real neutral rate at the end of 2024 likely ranged between 0 and 0.25 percent.

Markets Expect Low Rates, But for How Long?

Interestingly, financial markets appear aligned with this muted outlook. Market-implied forward rates, such as the 1y1y and 5y5y real risk-free rates, suggest that investors anticipate a softening of rates in the coming years. While these rates have increased from pre-pandemic lows, they remain negative over longer horizons. That implies skepticism about a sustained rise in equilibrium rates, despite recent inflation shocks. The ECB itself does not publish a formal neutral rate, but its macroeconomic projections hint at a real neutral rate of roughly 0.1 percent by 2026. Still, as the authors caution, relying on market-based signals can be misleading since they are influenced by central bank communication, investor sentiment, and global geopolitical uncertainty.

No Silver Bullet for Central Bankers

For monetary policymakers, the real equilibrium rate remains an essential, but slippery, guidepost. The authors explore three policy approaches to deal with model uncertainty. First, central banks can adopt gap-based rules that respond directly to inflation and output gaps, sidestepping unobservable variables like r*. Second, they can use model averaging, assigning weights to different models based on prior beliefs. Finally, robust control methods favor policies that perform well across worst-case scenarios, even if they are not optimal in any one model. Given the data and policy context of 2024, the paper leans toward viewing the ECB’s stance at that time as broadly neutral.

Looking forward, the direction of the equilibrium rate remains contested. Forces like AI-driven productivity growth, higher defense spending, and reduced precautionary savings could push it upward. Meanwhile, structural trends, aging populations, low potential growth, and financial fragmentation might keep it low. As of now, the markets seem to bet on the latter. In conclusion, the authors call for new unified frameworks that can reconcile these competing models and help guide policy in an age of persistent uncertainty.