It is easy to forget that inflation was both low and stable across most advanced economies in the decade leading up to the Covid-19 pandemic. Since the crisis, several factors have combined to drive up prices, and inflation risk has changed dramatically. These risks are rarely symmetric and can vary substantially over time. This column takes this idea of variable balance of macroeconomic risks and studies the appropriate response of monetary policy from both a theoretical and quantitative perspective. Central banks should be alert to the ‘skew’ of inflation risk and re-think their strategies to respond dynamically to it.
Inflation risks are rarely symmetric and tend to vary dramatically over time. Following a decade of low inflation across most advanced economies, recent global events – including the Covid-19 pandemic and Russia’s full-scale invasion of Ukraine – have reignited concerns about the prospect of sustained price increases. While central banks have occasionally referenced changes in the balance of risks to explain their monetary policy decisions, there has been little research on what is the optimal strategic response when inflation risks are distributed unevenly (or ‘skewed’).
In our recent paper, we present a new approach for studying the implications of unbalanced inflation risks for monetary policy within state-of-the-art macroeconomic models. We show that optimal monetary policy requires the central bank to actively counteract the direction of inflation risks, especially when such risks are elevated. When upside risk outweighs downside risk, as seen during the inflation surge in 2021 in many countries, the central bank must counteract this by communicating monetary tightening. Conversely, when risk is skewed to the downside, as often happens after major contractions (e.g., 2008–09), additional monetary policy accommodation – often called ‘loosening’ or ‘easing’ – should be communicated.
In this blog, we focus on the part of our paper where we introduce a novel monetary policy strategy – the Risk-Adjusted Inflation Targeting (RAIT) – into the model economy and explore its implications for post-pandemic macroeconomic dynamics. RAIT involves adjusting the monetary policy stance and communications strategy in response to changes in the balance of inflation risks. Our macroeconomic model shows that by incorporating these adjustments into the central bank’s strategy, RAIT effectively shields inflation expectations from the distortions caused by imbalances in inflation risks.
During the post-pandemic inflation surge in 2021–2022, inflation risks were heavily skewed to the upside due to supply chain disruptions and energy price shocks. Under the RAIT framework, the central bank could have explicitly communicated the heightened probability of inflation overshooting its target and signalled an earlier tightening of monetary policy. In theory, this proactive approach could have stabilised inflation expectations and reduced the need for more aggressive rate hikes later.
Asymmetric inflation risks and expectations anchoring
Inflation risks shift in response to economic conditions, including supply chain disruptions, geopolitical shocks and fiscal policies. Importantly, these risks are often uneven (or ‘asymmetric’), meaning the likelihood of inflation accelerating can differ significantly from the likelihood of it falling.
The balance of these risks has critical implications for the key objective of monetary policy – keeping inflation expectations stable (or ‘anchored’). When risks are ‘tilted towards the upside’, inflation expectations may rise above the central bank’s target. On the other hand, a negative balance of risks may lead to deflationary pressures, threatening the de-anchoring of expectations in the opposite direction.
This isn’t just economic theory: it is a statistical property. Think of inflation like the weather forecast. If there’s a higher chance of extreme heat, you’d probably prepare for a hotter day than usual, even if the average forecast hasn’t changed much. The same logic applies to inflation expectations. Consider the post-pandemic period in 2021–2022. As supply chain disruptions and energy price spikes made high inflation outcomes more likely, people (including businesses and households) started adjusting their expectations upward – even before inflation fully materialised.
If people perceive inflation risks as tilted to the upside, they naturally expect higher inflation, influencing their wage demands, pricing decisions and investment choices. This can lead to a self-fulfilling prophecy, as such behaviour drives up prices.
Inflation risks in real time
In our paper, we use an econometric model to estimate the balance of risks to core ‘Personal Consumption Expenditure’ (PCE) inflation in the United States in real-time over the postwar period. Our findings reveal that asymmetry in inflation risks is persistent, with prolonged periods of unbalanced risks that occasionally shift direction abruptly. After a decade of negative balance of risks (i.e., more deflationary pressure), the pandemic brought with it a sharp rise in upside risks (i.e., more inflationary pressure). This has since reached levels comparable to those observed during the 1970s. Allowing for the balance of inflation risks to vary over time within our econometric model achieves superior predictive accuracy compared with standard models used widely within policy institutions (Stock and Watson, 2007). Keeping tabs on risk is crucial to forecast inflation.
So, shifts in the balance of risks are both prevalent and quantitatively significant. But how should a central bank respond? We recommend that the monetary authority adopt a policy strategy focused on consistently centring communications on the path of future interest rates around the evolving balance of risks – particularly during periods of substantial changes in inflation risks accompanied by heightened volatility. Such situations are typical of major inflationary episodes, including the recent inflation surge and after the oil shocks seen during the 1970s. Keeping people’s expectations stable is the most effective way to keep prices stable, particularly in the face of changing risks.
Limitations of current monetary policy strategies
Our recommendations differ from current policy strategies adopted by central banks all over the world. For example, the ‘Flexible Average Inflation Targeting’ (FAIT) – introduced by the Federal Reserve in August 2020 – represents a significant breakthrough in monetary policy strategy (Clarida, 2022). Under this strategy, the central bank aims to maintain average inflation at its target by permitting temporary deviations to offset past inflation shortfalls or overshoots. This strategy is inherently backward-looking, relying exclusively on past inflation dynamics. As a result, FAIT fails to adapt to sudden reversals in the balance of risks.
In a world where shocks like pandemics or supply chain disruptions (caused by wars and other geopolitical tensions, for example) can rapidly alter the balance of inflation risks, FAIT may prove too rigid. This highlights the need for a more agile framework to address such shifts effectively in real time. Only looking backwards can leave policymakers vulnerable to challenges on the road ahead.
Risk-Adjusted Inflation Targeting
As outlined above, RAIT is a novel strategy requiring the central bank to consistently adjust the monetary policy stance in response to real-time changes in the balance of inflation risks. Specifically, this framework prescribes the central bank to communicate a tighter policy stance in response to upside risks and more accommodative measures to address risks leaning to the downside. In simple terms, this approach would mean that the central bank regularly updates the public on where inflation risks are leaning – whether there’s a bigger chance of inflation running too high or too low – and explains how that affects its interest rate decisions.
How can the central bank determine the appropriate response to unbalanced inflation risks under RAIT and what are the effects of this response? We address these questions using a quantitative Dynamic Stochastic General Equilibrium (DSGE) model – widely used for monetary policy analysis. To do so, we propose a method that is flexible and can be combined with alternative approaches (including judgment) to evaluate the balance of inflation risks.
The 2021-2022 inflation rally
An advantage of the model-based approach proposed in our paper is that it allows for counterfactual analyses. For example, it can be used to address questions like: what if the Federal Reserve had followed the RAIT system during the recent inflation surge? We illustrate the answer to this question in Figure 1, which compares actual data (black lines) for the Federal Funds rate (first panel), core PCE inflation (second panel) and hours worked (third panel) with counterfactual values (grey lines with pink points) predicted under the assumption that the central bank had followed RAIT.
Figure 1: Fed fund rate, core PCE and hours worked (RAIT vs actual data)
Source: Authors’ calculations
According to our analysis, adopting RAIT would have led the Federal Reserve to anticipate the post-pandemic inflation lift-off around three quarters sooner (as shown in the first panel). Notably, the pace of rate increases and the end value of the Federal Funds rate remain broadly aligned with the Federal Reserve’s actual actions. The second panel shows that under RAIT inflation would have declined more quickly, reaching the target in early 2023 and stabilising thereafter. The third panel illustrates that by initiating monetary easing in early 2023, following RAIT would have supported the labour market, partially offsetting earlier tightening effects.
The 2010–2019 deflationary bias
The decade preceding the Covid-19 pandemic was marked by persistently low inflation. This is reflected in our model as a sustained shift toward ‘negative inflation asymmetry’. Despite this, the remarkable stability of the macroeconomic environment resulted in fairly contained inflation volatility.
Against this backdrop, RAIT would have recommended ‘looking through’ (i.e., ignoring) the deflationary bias. This underscores RAIT’s focus on the effects of unbalanced inflation risks on expectations. These effects are muted when inflation volatility is low. In contrast, FAIT would have required the Federal Reserve to promise overshooting the target to compensate for past inflation misses.
Conclusions
Monetary policy is most effective when it is forward-looking and responsive to evolving risks. The RAIT framework offers an adaptive approach by incorporating real-time assessments of inflation risks, allowing central banks to fine-tune their responses to shifting economic conditions.
Our analysis suggests that had RAIT been in place during recent inflationary episodes, central banks could have responded more swiftly to inflationary pressures while containing the effects of interest rate changes on the labour market. Compared to backward-looking strategies like FAIT, RAIT offers a more data-driven and proactive policy framework.
Inflation risks are rarely symmetric, and failing to account for their imbalance can lead to suboptimal decisions. As central banks navigate an increasingly uncertain economic landscape, adopting a framework that actively responds to shifts in inflation risks can strengthen price stability.
Author: Andrea De Polis, Leonardo Melosi, Ivan Petrella
Authors’ note: Any views expressed in this paper are those of the authors and do not necessarily reflect the views of De Nederlandsche Bank or any other person associated with the Eurosystem.