Both consumers and business representatives systematically overestimate the level of inflation this phenomenon is typical for economies around the world. Economists offer different explanations for this phenomenon: for example, authors from the IMF, using the example of consumers in the eurozone, showed that people react disproportionately strongly to news about inflation. In principle, the rise in prices causes a stronger reaction among consumers than their decline or stabilization, and prices for frequently purchased goods are more important for the perception of inflation, explained the authors from Statistics Canada. Economists from the National Bank of Denmark, in turn, linked the shifted ideas about inflation, including with the personal characteristics of consumers, such as a tendency to pessimism.

There is another explanation: inflation expectations are influenced by the quality of information, or rather, the source on which people rely when forming their estimates and forecasts, Francesco D'Acunto from Georgetown University and Michael Weber from the University of Chicago concluded in a new study. After conducting surveys in 47 countries covering about 75% of the world's population, they came to the conclusion that, firstly, consumers' biased ideas about inflation are truly a global phenomenon. But, secondly, there are noticeable differences between countries in how strongly these ideas are shifted. These differences, as well as the very reason for the displacements, are explained by the sources of information, the authors concluded.

All sources can be combined into two main groups: local and aggregated. Local data is data that people observe and receive in their immediate environment when they shop, receive utility bills, discuss prices with friends and family, read social networks. Aggregated data comes from sources such as government and central bank reports, official statistics, traditional media. Those who rely mainly on local data form systematically more shifted expectations compared to other respondents.

Consumers in countries with historically higher or volatile inflation follow it more closely, but their receipt of more information paradoxically involves an increase in errors in estimates and forecasts. This is due to the fact that periodic surges in inflation or generally higher inflation background can lead to a decrease in confidence in economic institutions and, as a result, in the aggregated information they provide.

Therefore, in countries with higher inflation, consumers often rely on personal observations when forming estimates and expectations, which leads to an increase in errors.

Global survey

Economists conducted a survey in 2023, it involved almost 47,000 respondents in 47 countries, which together account for about 75% of the world's population and 90% of global GDP. There were four parts in the survey. In the first, people noted their demographic characteristics, such as age, gender or employment. In the second part, the questions were devoted to the financial characteristics of households the availability of savings for a rainy day, real estate, the distribution of financial responsibility within the family. At the third stage, the researchers found out what the current perception and expectations of respondents are regarding macroeconomic variables: inflation, the level of real estate prices, interest rates.

Finally, in the final, the authors asked the participants to assess the importance of various sources of information for them: statements and reports of governments, central banks, statistical services, media (aggregated data), as well as household observations, purchases - their own and those of their acquaintances, utility bills, social media (local data).

Local priorities

The analysis of questionnaires showed that in general, people in the world attach more importance to local sources than aggregated ones. For example, more than 80% of all respondents named prices in stores as important sources of information, also more than 80% - utility bills, 70% - information of their acquaintances. However, the importance of aggregated sources, according to polls, is also quite high: more than 70% named official messages of the government and the central bank as important sources, about 60% - information from traditional media.

However, some aggregated sources may be mentioned by respondents simply because they perceive such an answer as more approved, the authors note. By comparing the estimates of the significance of local and aggregated data for respondents of each country, economists built a "source map" reflecting how much local data is prioritized in each of the 47 countries than aggregated data. And then, based on the respondents' estimates of current inflation (for the last 12 months) and its actual values, they built the same "map of deviations" of inflation expectations by country. By comparing both maps, the researchers found that in those countries where inflation expectations are more overestimated, the dominance of local data sources is usually stronger.

Two data sources

Sources of aggregated data give signals about economic parameters obtained from full distribution, that is, for example, about price changes in general for a wide range of goods and services. This implies that the information from such sources is impartial, not shifted towards any separate market or product. Signals from local sources reflect price changes only a small part of the consumer basket, they are non-representative, representing only a subset (part) of the total consumer set.

Therefore, conclusions about the whole based on information about its part turn out to be biased. Local information is easier to obtain: although aggregated data is publicly available, some search efforts have to be made to obtain it (for example, to find and read an article in a newspaper), while local data can be obtained "automatically" simply in the course of your daily activities.

However, although local information is easier to find, it is more difficult to process: you will have to try to take into account a whole series of prices observed on the shelves, while in the case of aggregated data, everything has already been taken into account and it is enough to open an official report.

In the case of local data, consumers often simplify the analysis, relying, for example, on marker products. When inflation is consistently low, people stop paying attention to it at all and, accordingly, update their inflation expectations but as inflation increases, attention to it also begins to increase.

This is consistent with the concept of rational inattention: people make decisions based on limited information and ignore information, the search or processing of which is associated with higher costs than the benefits derived from it. With consistently low inflation, the probability of errors in economic decisions made on the basis of "old" information is low.

But the probability and price of such errors increase if the inflation rate increases: in such a situation, people begin to make efforts to monitor price dynamics.

Attention and trust

According to the concept of rational inattention, it would be logical if more attention to inflation and information about it led to greater accuracy of its estimates and forecasts made by consumers.

However, it turns out to be the other way around: in countries with relatively high inflation, more information received by consumers does not lead to an increase in the accuracy of their forecasts - on the contrary, their estimates and expectations are more inaccurate than in countries where inflation is relatively lower.

This is due to the fact that where ignorance entails higher costs, consumers prefer to receive information mainly from local sources despite the fact that the processing costs of such signals are higher.

Local sources do not give representative signals about future inflation and, accordingly, lead to an increase in forecast errors. But how to explain the fact that in the presence of publicly available aggregated data, people rely on local observations?

At least partially, this can be explained by a decrease in consumer confidence in sources of aggregated information, the authors found. During the survey, in most countries they could directly assess the respondents' trust in economic institutions, central banks and governments that produce aggregated economic information.

Where trust is higher, respondents are more likely to turn to aggregated sources - both official reports and television and newspapers. In those economies where inflation has historically been at a high level for a longer time, consumers are more likely to turn to their immediate environment for information in order to extract signals from it and form expectations on their basis. Even if the costs associated with obtaining and processing information from both types of sources are the same, consumers prefer the source they trust more. These results mean that building higher trust in institutions including through direct communication with consumers can encourage consumers to use aggregated sources of information more often when shaping their expectations. Since these sources are more accurate and representative, the shift of preferences in their favor can increase the effectiveness of monetary and fiscal policy.