Expectations surveys, central banks and the economy

Welcome address by Luis de Guindos, Vice-President of the ECB, at the 5th joint ECB, Bank of Canada and Federal Reserve Bank of New York Conference on expectations surveys, central banks and the economy

Frankfurt am Main, 1 October 2024

It is my pleasure to welcome you to this fifth joint conference on expectations surveys organised by the European Central Bank, the Bank of Canada and the Federal Reserve Bank of New York.

In my remarks today, I will delve into the fascinating world of expectations surveys and their relevance to central banks. I will review how useful expectations surveys have proven to be for central banks over the period since 2019, the year we held our first conference in this series. In addition, I will touch on the challenges facing central banks in using surveys. The fact that central banks generally operate under great uncertainty has come to the fore over the past five years. Today, too, we are facing huge uncertainty – not least in view of the many prevailing economic, financial and geopolitical risks. Yet, it is precisely in this unpredictable and highly complex landscape that surveys have come into their own.

The return of survey expectations

Over the past decade, central banks and other policymaking institutions have invested significantly in expectations surveys and have drawn increasingly on survey data for their policy analysis and research. These surveys cover consumers, firms, financial market participants and other experts, including professional forecasters. At the ECB, we can fortunately look to a wide array of such surveys covering diverse topics such as consumer expectations, household finance and consumption, access to finance of enterprises, the payment attitudes of consumers and bank lending. Since 2013, the ECB has also conducted a survey of wholesale market participants on credit terms and conditions, and it recently developed a new survey of monetary analysts to collect expert expectations about key monetary policy parameters and concepts. Finally, the ECB’s Survey of Professional Forecasters was launched back in 1999 at the start of Economic and Monetary Union. Its structured collection of data has supported a rich research programme investigating economic forecasts and expert expectations.^[1]

All ECB surveys can provide insights into how different economic agents form and update their expectations. They can reveal the potential biases in these expectations and the extent to which expectations feed into economic decisions. Surveys were indeed quite central to the economic debate in the 1950s and the 1960s but their role became more marginal when rational expectations were incorporated into economic modelling in the 1970s. Over the past ten years, however, economists have seen survey expectations clearly returning to the mainstream.^[2] One could describe the recent growth in survey-based research as a “counter-revolution” following the earlier “revolution” centred on rational expectations. Today, while models based on rational expectations still form a useful reference point in our analysis and research, they are no longer thought to provide a solid basis for understanding business cycles, for gauging the risks of financial crises or for designing effective economic policies. The central insight gained from this new line of survey-based research is that many economic agents may systematically form expectations by using partial sets of information or by following subjective narratives about how the economy functions – for example by applying simple rules of thumb.^[3] It is important to understand such subjective expectations, because these beliefs often underlie the economic choices and financial decisions that drive the economy.^[4]

Surveys have repeatedly proven their usefulness over the past five years. During the COVID-19 pandemic, they were especially useful in helping to track financial conditions for firms and households, as well as in estimating the labour market response to the pandemic shock. Online surveys were of great benefit during the pandemic as in-person survey interviews were hampered by lockdown restrictions. For example, the ECB’s Consumer Expectations Survey – an online survey which was fortuitously launched in early 2020 – helped us understand the severity of the pandemic-induced collapse in consumption and gauge the overall effectiveness of the major policy interventions by governments and other authorities at the time.^[5]

Insights from surveys during the recent period of high inflation

More recently, the data collected in surveys strongly supported the analysis of the recent inflationary episode in the euro area.^[6] During the early phase of the inflation surge in 2022, survey data helped to inform the central discussion on the likely persistence of the shock. For example, the observed increase in consumers’ medium-term expectations may have interacted with an increase in firms’ pricing power to make the original supply shocks more persistent than they would otherwise have been.^[7]

Forces that would gradually help bring inflation back down to our target were also visible in more recent survey data. For example, we could see how the rise in inflation and inflation expectations was acting as a major constraint on demand and consumer spending owing to its impact on real incomes. In August 2023 respondents to the ECB’s Consumer Expectations Survey were asked what actions they were planning to take in light of their expectations about future inflation. The results clearly showed that a much higher share of consumers planned to reduce their spending in response to the expectations of higher prices.^[8] In addition, consumers indicated that they would start to shop around more and buy cheaper varieties of goods and services than they normally would. In a context where the ECB was taking decisive monetary policy action aimed at restoring price stability, these behavioural responses to higher inflation expectations also contributed to the gradual unwinding of the inflationary pressures across the euro area economy.

Insights for financial stability analysis

In addition to monetary policy, expectations surveys are now increasingly being used for other central bank tasks as well. This includes financial stability analysis. Here, surveys can help identify potential sources of financial risk not only in financial markets and the banking system, but also in the household and non-financial corporate sectors.^[9] Even when there is no discernible financial stress at the aggregate level, the disaggregated or individual-level data typically provided by surveys can help us to identify emerging risks across particular sectors or socio-demographic groups.

In financial stability analysis, the topic of financial literacy is receiving increased attention. In the first keynote lecture of the conference, Professor Annamaria Lusardi from Stanford University will talk about why financial literacy is relevant for central banks. One consideration for financial stability analysis is that less financially literate households may be less prepared to cope with adverse economic and financial shocks. Yet, these households tend to be the most exposed to such shocks and more heavily affected when they occur. Policies seeking to boost financial literacy may help borrowers to source loans that are cheaper to service, thus promoting more efficient and more sustainable debt management. These issues may be particularly relevant for real estate markets and housing, which will be the focus of the second keynote lecture of the conference, given by Professor Tarun Ramadorai from Imperial College London. Professor Ramadorai will discuss the importance of non-rational beliefs in the housing market and how household surveys can help inform policies that can address these frictions.

Sustaining the quality and representativeness of surveys

Our experiences with survey data also highlight the challenges that policymakers face when using these data. Survey data can be volatile and there is evidence of overreaction in both household and firm surveys of expectations. For this reason, surveys may provide a noisy signal for policymaking in practice, which complicates how these data should feed into the policy reaction function. In this respect, I hope the research presented at today’s conference can also help policymakers distinguish the signal from the noise that is always embedded in expectations data. These considerations underline the importance of the quality of the survey design, including the sampling and data collection methods. It is crucial that questions are designed to avoid the framing of responses and that the complexity of the questionnaires is managed appropriately to avoid survey fatigue, which may negatively affect data quality. As central banks are making increasing use of survey data, they need to continuously and carefully monitor these data to ensure responses remain representative of the underlying population’s beliefs and behaviour.

Conclusion

Let me conclude. Today, expectations surveys are an important part of the toolkit available to central banks for their policy analysis. These surveys reveal insights about the economy that would otherwise remain hidden from view. As a result, they can contribute to more robust policy decisions and better policy assessments.

I would like to finish by thanking the presenters and participants in advance for their contributions and the conference organisers for putting together such an impressive programme. I wish you all a productive and successful two days of lively debate and discussion. I am confident that the insights that will emerge from sharing our experiences of different surveys across many countries and institutions will ultimately enhance the way in which we use expectations surveys to help guide policy decisions.