A Learning-from-Prices view of Inflation and Business Cycles

Projektinformationen

LearnInCycle

ID Finanzhilfevereinbarung: 101088411

DOI

10.3030/101088411

EK-Unterschriftsdatum 20 März 2023

Startdatum 1 April 2023

Enddatum 31 März 2028

Finanziert unter

• European Research Council (ERC)

Gesamtkosten

€ 1 147 981,25

EU-Beitrag

€ 1 147 981,00

Koordiniert durch

ETABLISSEMENT D'ENSEIGNEMENT SUPERIEUR CONSULAIRE HAUTES ETUDES COMMERCIALES DE PARIS
France

Periodic Reporting for period 1 - LearnInCycle (A Learning-from-Prices view of Inflation and Business Cycles)

Berichtszeitraum: 2023-04-01 bis 2025-09-30

Zusammenfassung vom Kontext und den Gesamtzielen des Projekts

In recent years, inflation has returned to the forefront of public and policy debate. Following a long period of subdued price dynamics, global disruptions—first from the COVID-19 pandemic, then from geopolitical tensions like the war in Ukraine—have caused a rapid and uneven rise in consumer prices. In the Eurozone, for example, inflation jumped from negative territory in late 2020 to over 5% by early 2022. This abrupt change has raised urgent questions for policymakers: Will inflation persist? How can monetary policy respond without stalling economic growth?

The LearnInCycle project addresses these questions from a novel perspective. Rather than focusing solely on firms and price-setting behavior—as is common in traditional macroeconomic models—this project centers on households. It investigates how everyday consumers form inflation expectations based on the prices they observe when shopping, and how these expectations, in turn, influence broader economic outcomes such as consumption, output, and inflation itself.

This "learning-from-prices" approach posits that market prices are more than just outcomes—they are also information sources. When prices rise, consumers don’t just pay more; they interpret these changes as signals about the economy. This interpretation can lead them to alter their spending habits: some may shop around more, switch stores, or delay purchases. These decisions ripple through the economy, affecting aggregate demand and inflation dynamics—even when firms themselves face no pricing constraints.

LearnInCycle develops this insight through five integrated work packages. It starts with new economic theories where inflation and output co-move because of consumers' responses to price signals. It then tests these theories using large-scale datasets tracking millions of retail transactions and household shopping patterns in the U.S. Finally, the project uses novel economic model to study how people update their expectations and how this feedback into persistent inflation.

The key innovation is to link individual behavior and perception—especially related to inflation—with macroeconomic outcomes. By providing both theoretical models and empirical evidence, LearnInCycle offers a fresh view on how inflation expectations form and spread, and how they influence the effectiveness of monetary policy. The project also explores how central banks might better tailor their actions and communication to address households' information imperfection, thereby improving policy transmission in times of uncertainty.

Arbeit, die ab Beginn des Projekts bis zum Ende des durch den Bericht erfassten Berichtszeitraums geleistet wurde, und die wichtigsten bis dahin erzielten Ergebnisse

In the initial phase of the project, a new theoretical framework was developed to understand how household perceptions of inflation influence macroeconomic dynamics. This work led to a peer-reviewed article, conditionally accepted for publication in the Journal of the European Economic Association. The proposed model demonstrates that households reallocate spending across retailers in response to perceived inflation, generating substantial macroeconomic effects even in the absence of nominal rigidities. This mechanism offers a novel explanation for the observed co-movement between inflation and output following monetary shocks.

Building on this foundation, new empirical measures of household shopping behavior have been constructed using the Nielsen Retail and Panel Scanner Dataset. These measures aim to capture how consumers adjust their spending patterns in response to changes in perceived prices. The resulting empirical evidence represents a key step toward validating the theoretical framework and is being developed into a working paper expected to be released in the near future.

Further extensions of the project are currently being pursued to study the persistence of inflation expectations and their policy implications. Preliminary findings have been presented at leading academic conferences. This research explores how survey-based inflation beliefs can be replicated within a self-referential model incorporating forward-looking information frictions. Ongoing work seeks to embed this framework into a standard New Keynesian environment to examine its relevance for monetary policy design.

A related line of inquiry has expanded the project's core insights to the financial sector, examining how central bank asset purchases shape investor beliefs and market prices. This research shows that such interventions can influence inflation expectations and outcomes even in the absence of price rigidities, thereby extending the learning-from-prices mechanism from consumers to financial markets. The results are publicly available on my web page in the form of a working paper and have been disseminated across multiple academic and policy platforms.

Fortschritte, die über den aktuellen Stand der Technik hinausgehen und voraussichtliche potenzielle Auswirkungen (einschließlich der bis dato erzielten sozioökonomischen Auswirkungen und weiter gefassten gesellschaftlichen Auswirkungen des Projekts)

The project has developed a new theoretical framework, culminating in a peer-reviewed conditionally accepted publication in the Journal of the European Economic Association. This framework introduces new strategies to identify and disentangle the channels through which inflation expectations shape the business cycle. In particular, it reveals that shopping behavior, driven by perceived inflation, can act as an independent channel for the transmission of inflation to real economic activity. Preliminary empirical results, based on the Nielsen Retail and Panel Scanner Dataset, represent a key step toward validating this theory. A novel measure of household shopping activity has been constructed, capturing consumer responses to inflation signals in real-world data. Ongoing work extends the model to examine inflation persistence and policy transmission using a forward-looking representation of information frictions, which significantly improves analytical tractability. A related paper expands the scope of the project by applying the learning-from-prices approach in the context of asset purchasing policies. It shows how central bank asset purchases affect investor beliefs and asset prices, providing new microfoundations for the effectiveness of such policies in managing inflation—particularly under unconventional circumstances. This work highlights the broader applicability of the framework and its relevance for understanding the impact of policy interventions in both household and financial domains.

The results have substantial implications for monetary policy design and communication, emphasizing the importance of accounting for belief-driven behavior in both consumers and investors. To ensure further uptake and impact, future work should focus on deeper empirical validation of the proposed shopping behavior proxy, broader access to high-frequency consumer data, and the integration of these insights into mainstream macroeconomic models. Engagement with policymakers, particularly central banks, will be key to translating these findings into practice. At this stage, the project has produced two major outputs: a peer-reviewed article and a publicly available working paper. Finalizing ongoing empirical work and disseminating results through academic and policy networks will be essential to maximize the project's long-term contribution and support the broader application of its findings in research and policy development.