Listening to households: expectations, behaviour and monetary policy

Speech by Boris Vujčić, Vice-President of the ECB, at the inauguration of the Alexander von Humboldt Professorship awarded to Professor Michael Weber at the European School of Management and Technology (ESMT) Berlin

Berlin, 1 September 2026

Introduction

For a long time, macroeconomic analysis often relied on strong assumptions that expectations were rational. Over the past two decades, evidence has shown the limits of those assumptions for many policy questions. A foundational contribution was to measure expectations directly rather than infer them from outcomes alone.^[1]

This became much more feasible as the internet and mobile technologies became more widely used. Online survey data collection, better questionnaire design and more effective sampling methods now allow high-quality, population-representative evidence to be collected in a very timely manner. The ECB’s Consumer Expectations Survey is a case in point, as is the recent work by the Bank for International Settlements.^[2] This research has improved our understanding of expectations formation and its role in shaping aggregate economic outcomes.

Alexander von Humboldt would no doubt have approved of this enhanced approach to measuring expectations because it follows his maxim about observing carefully before generalising. Economists have not always followed that sequence as faithfully as they might. By contrast, Michael Weber has been at the forefront of research in this area, particularly in advancing our understanding of household inflation expectations and their implications for monetary policy.

In my remarks today, I will focus on three areas: recent advances in our understanding of household expectations; how these insights have informed policy analysis in a period of inflation shocks and geopolitical stress; and the lessons we can draw for communication, trust and credibility.

Recent advances in our understanding of household expectations

Let me begin with a simple observation. Households do not form expectations in the same way as professional forecasters. Most people do not continuously process the full range of macroeconomic data. Their attention is selective, their information is incomplete and their beliefs are widely dispersed. This dispersion is not simply noise: it reflects differences in income, wealth, housing, financial literacy, lived experience and trust in institutions.^[3]

Research co-authored by Michael has shown how salient prices, especially food and fuel prices, can disproportionately shape inflation perceptions and expectations.^[4] Related evidence helps explain the persistent gender gap in inflation expectations, partly through differences in the information environments and in exposure to shopping-related price signals.^[5]

This is why harmonised, cross-country surveys with a panel component, such as the ECB’s Consumer Expectations Survey, are so important. They enable us to track beliefs over time and across environments, and to connect expectations to behaviour.

And this evidence is valuable from a policymaking perspective because it sheds light on three dimensions of expectations: what households believe, how they update their beliefs and how those beliefs affect their behaviour.

This body of work also has implications for macroeconomic models. If expectations vary and information is incomplete, as the evidence certainly suggests, policy models need to explicitly incorporate these variations and information frictions.^[6]

How research on expectations has informed recent policy analysis

Let me now turn to the usefulness of expectations in policy analysis.

Insights about expectations have recently become directly useful for policy, especially in the current context of elevated uncertainty and geopolitical tension.

First, data about expectations improve the way we interpret economic shocks. Aggregate and market data show what happens to prices and output. Household surveys reveal how people interpret shocks, how persistent they expect the shocks to be, and how those beliefs affect their choices. For example, recent evidence from the Consumer Expectations Survey shows that geopolitical risk can significantly worsen household sentiment and spending intentions in the euro area.^[7]

These data are particularly relevant at the present juncture. For example, the June 2026 Eurosystem staff projections exercise underlined the need to communicate not only a baseline but also the uncertainty surrounding that baseline and the alternative scenarios that may unfold. Survey evidence can provide valuable contributions to the preparation and calibration of such scenarios by helping us to understand how households themselves perceive uncertainty: whether they see a shock as temporary, whether they expect it to persist and whether they may respond by postponing spending and increasing precautionary saving.

Such survey evidence can be particularly important when geopolitical shocks affect energy prices and the broader cost of living. Such shocks can lead to weaker confidence, higher inflation expectations and greater economic uncertainty. Surveys allow us to observe these perceptions directly rather than infer them at a later stage from aggregate outcomes.

Second, uncertainty itself matters for demand. Evidence shows that higher perceived macroeconomic uncertainty can persistently reduce household spending.^[8] This helps policymakers distinguish between temporary caution and broader demand weakness.

Third, transmission through household balance sheets can vary. Monetary tightening affects households differently depending on housing tenure, stock market participation, the type of mortgage contract and the timing of interest rate resets.^[9] Analysis based on the Consumer Expectations Survey shows how mortgage repricing feeds into consumption with long and uneven lags across household groups and countries.^[10] By taking these important sources of heterogeneity into account, it is possible to gain a more accurate understanding of aggregate macroeconomic dynamics.

Fourth, there are differences in the anchoring of inflation expectations, which is key to understanding the likely persistence of shocks. Expectations of households are generally less well anchored than professional forecasts, but their longer-term expectations tend to be more centred around our 2% inflation target and less sensitive to surprises than the shorter-term expectations.^[11] This distinction helps us assess whether movements reflect temporary inflation episodes or deeper credibility risks.

The broader lesson is that surveys do not replace aggregate, administrative or market data. They complement them by measuring perceptions, uncertainty and intended behaviour that other sources cannot observe directly. In this respect, the data in consumer surveys complement other evidence from the survey on the access to finance of enterprises and the bank lending survey, allowing policymakers to compare household beliefs, firm conditions and credit supply channels in near real time.

Communication with the public: building trust and credibility

This brings me to communication. Research shows that central bank communication can influence household inflation expectations.^[12] Since expectations affect transmission, communication is not an afterthought. It is part of policy transmission itself.

One important finding in Michael’s research is that, for broad audiences, communication about the target and objective can be more effective than communication about technical instruments.^[13]

In the euro area, evidence indicates that clear communication about the inflation target, paired with simple policy explanations, can improve perceived central bank credibility, including among groups with lower financial literacy.^[14]

These effects matter because better-anchored expectations can reduce the output costs of inflationary or cost-push shocks.

But communication faces substantial constraints. Getting through to households is difficult. Technical language, though perfectly accurate, may not be accessible to all audiences. Monetary policy also competes for attention with work, bills and family life. And official communications by central banks may not always seem consistent with what people see in shops and in their energy bills.

As Michael’s research shows, when inflation is low, reaching the public is particularly hard due to low consumer attention. When inflation is high, people pay more attention and are better informed.^[15] His research therefore paints a picture of two rather different communication environments. When inflation is low, the main challenge is to reach an inattentive public. But when we do get the public’s attention, direct communication can be powerful. Simple messages about our objective and the role of monetary policy can improve public understanding and the central bank’s credibility, particularly among groups with lower financial literacy.^[16]

When inflation is high, by contrast, the public is much easier to reach because they are already paying more attention to prices. Yet it is also harder to convey genuinely new information, since households are closely following developments in shops, at petrol stations and in their energy bills. This can nevertheless create a window of opportunity to explain how the central bank is responding, what it can achieve, and why it must remain focused on price stability. In such circumstances, communication should be clear, accessible and candid about uncertainty.

Another important finding from this area of research is that the messenger – and not just the message – can matter in central bank communication.^[17] Michael’s research has in fact shown that more diverse policy committees and messengers can help build trust among otherwise under-represented and harder-to-reach groups.^[18] In the euro area’s multi-country and multi-cultural setting, this can also strengthen the effectiveness of communication.

As ECB President Christine Lagarde has argued, communication must adapt to changing media environments.^[19] The objective is to remain accessible and rigorous in our messaging, rather than using novel ways of communicating for the sake of it.

Let me give you a concrete example: the ECB’s Valentine’s Day poem about returning inflation to 2% posted on one of our social media channels.^[20] Poetry is not a policy instrument, but the point this illustrated is important: communication sometimes needs to use alternative formats that people can better relate to.

Still, communication cannot replace a central bank’s performance. As the saying goes, you need to walk the walk, not just talk the talk. Trust is gained through competence, consistency, transparency and accountability. When people trust the central bank, their inflation expectations tend to be closer to the inflation target.^[21] In the end, credibility depends on what we do, and whether it matches what we say.

Surveys help in yet another way. They allow central banks to listen to the public. They show which messages reach their intended audience, where there are misunderstandings and how different groups experience economic shocks. Credible communication is therefore a “two-way street”.

Conclusion

Let me close with a few remarks on the importance of data quality.

High-quality surveys need to remain a priority. Representative sampling, harmonisation across countries, repeated interviewing and timely delivery of survey data are essential for tracking the interaction between beliefs and behaviour.

In today’s fast-paced world, the risks to survey data quality must be actively managed. Survey fatigue and the potential rise of inauthentic responses, including AI-related risks, mean that strict quality assurance is needed and the burden on respondents has to be managed carefully. At the ECB these are among the top priorities for the teams running such surveys.

Michael Weber’s work has helped shape this field, and I am confident he and his co-authors will continue delivering valuable insights in the years ahead.

Thank you very much for your attention.