In this paper, we study how individuals learn from potentially biased statistics using data from both a natural experiment and a survey experiment during a period (2007-15) when the government of Argentina was manipulating official inflation statistics.
To address these limitations in the observational data, we provide a simple model of Bayesian learners with potentially biased statistics and design a survey experiment to test its predictions.
The experimental data also allow us to directly test the hypothesis that there may be sophisticated learning from potentially biased statistics.
More generally, the study of biased statistics goes back to the seminal contribution by Oskar Morgenstem (1963) on measurement, accuracy, and uncertainty in economics.
The most important prediction of this model is that a Bayesian learner is not expected to ignore biased statistics, but instead rationally adjust to the perceived bias.
Rather than simply ignoring biased statistics or accepting them as unbiased, individuals can effectively adjust for the perceived bias using other available information.