Summary: Researchers find that people respond more to individual prices than to aggregate statistics, but rational inattention and poor recall lead many to overestimate inflation. Source: MIT.
Consumers tune in to prices selectively, but recall is poor and expectations skew high, study finds.
How well do you know your country’s current inflation rate? What do you expect inflation to be next year? And how do those expectations affect the way you plan your budget and spending?
These questions matter for economists and policymakers because central banks take public expectations about inflation into account when setting interest rates. A new study co-authored by an MIT economist suggests that most people pay attention to inflation only intermittently, rely heavily on memories of the prices they personally see, and—on average—overestimate how high inflation will be.
The research, conducted in the United States and Argentina, shows evidence of “rational inattention”: individuals concentrate on inflation information when it is most relevant to them. But the study also documents widespread memory errors and a tendency for expectations to lean upward.
“We found clear evidence of rational inattention,” says Alberto Cavallo, the Douglas Drane Associate Professor in Information Technology and Management at the MIT Sloan School of Management and a co-author of the paper. “People pay attention when they need to.”
Yet even where inflation is a pressing economic reality, memory errors are common. “People have poor memories of past prices,” Cavallo says. “Even in Argentina, where inflation has been high and price changes are salient, many respondents underestimated past prices and therefore concluded that inflation had been larger than it actually was.”
The paper, titled “Inflation Expectations, Learning, and Supermarket Prices,” appears in the American Economic Journal: Macroeconomics. The study’s other authors are Guillermo Cruces of the National University of La Plata and Ricardo Perez-Truglia of UCLA.
Comparing statistics and store prices
The authors base their conclusions on a set of online and in-person survey experiments conducted in both countries, including interviews taken shortly after supermarket shopping trips. The two countries were selected because they represent contrasting inflation experiences: over the five years preceding the study, inflation was relatively low in the United States and much higher in Argentina. This contrast allows the researchers to test how historical exposure to inflation influences attention and beliefs.
Several patterns emerged. Respondents in Argentina—exposed to consistently higher inflation—tended to hold stronger priors about inflation rates and to give more weight to prior beliefs when updating expectations. In the U.S., where inflation was lower and more stable, people placed less weight on prior beliefs and appeared to have weaker, less entrenched expectations.
Quantitatively, the study estimated that U.S. respondents assigned roughly 15 percent weight to prior beliefs when forming expectations, while Argentinian respondents assigned closer to 50 percent. Cavallo interprets this as consistent with rational inattention: where inflation matters a great deal for wages and savings, people invest more attention and form firmer priors.
Another robust finding is that individuals respond more to the prices of the products they personally buy—supermarket prices—than to abstract, aggregate inflation statistics. Across both countries, people gave relatively greater weight to the price changes of specific goods they encounter in their daily lives than to official inflation measures, which are broader but less tangible.
That reliance on salient, local price information interacts with memory limitations. In the United States, about 29 percent of the variation in inflation expectations could be traced to perceived past inflation; in Argentina that share rose to roughly 60 percent. In other words, differences in people’s recollections of past prices account for a substantial portion of why their expectations diverge.
Cavallo notes that an upward bias in inflation expectations may function as a protective heuristic in high-inflation settings. “In a country like Argentina, assuming inflation will be higher can be adaptive: it encourages people to negotiate wages or adjust savings to avoid the worst effects of rising prices,” he says.
Implications for monetary policy and communication
The paper connects to a larger research agenda by Cavallo and colleagues studying price behavior and expectations. Cavallo is a co-founder of the Billion Prices Project, an initiative that tracks retail prices in real time to evaluate official inflation measures. The current study sheds light on how households form inflation expectations, a key input for understanding real interest rates and consumer behavior.
Real interest rates—the interest consumers effectively face—depend on nominal interest rates and expected inflation. If households expect inflation to be higher than nominal rates, they may prefer to buy goods now rather than hold cash or save at low rates, potentially amplifying demand and influencing broader macroeconomic outcomes.
From a policy perspective, the study suggests two complementary options for improving public understanding of inflation. One is to provide clearer, more accessible aggregate statistics; the other is to offer concrete examples of price changes in goods and services that matter to consumers. “Governments and central banks should communicate clearly and use examples of everyday goods,” Cavallo says. “We need to know how much people learn from the information we give them.”
Source: Peter Dizikes, MIT
Image Source: NeuroscienceNews.com image credited to Christine Daniloff/MIT.
Original Research: “Inflation Expectations, Learning, and Supermarket Prices: Evidence from Survey Experiments” by Alberto Cavallo, Guillermo Cruces, and Ricardo Perez-Truglia. American Economic Journal: Macroeconomics. Published online July 31, 2017. doi:10.1257/mac.20150147
MIT. “Intermittent Attention and Poor Memory Shape Public Perceptions of Inflation.” NeuroscienceNews. July 31, 2017.
Abstract
Inflation Expectations, Learning, and Supermarket Prices: Evidence from Survey Experiments
Information frictions significantly shape how households form inflation expectations, but the origins of those frictions are debated. Using survey experiments, this paper identifies two main sources. First, individuals in low-inflation environments have weaker priors, consistent with rational inattention: when inflation is low and stable, people invest less attention in tracking it. Second, cognitive limitations lead people to overweight inaccurate personal information—such as their memories of supermarket price changes—even when official statistics are available. These findings have implications for macroeconomic modeling and policy design.