Experimental Design
We run a survey among German households and examine whether households' demand for expert narratives ("decoding") differs from households' demand for central bank information. We do this via an incentivized Willingness-to-Pay framework.
Within our survey, we first anchor respondents’ belief about the level of current inflation as well as its past development. We give the respondents the information about the level of the inflation rate in Germany in October 2022. We then ask the respondents about their beliefs on the inflation rate in October 2024. Next, we provide them with the true value of the inflation rate in October 2024.
We then randomize respondents into two groups. We allow the first group to acquire an explanation from the ECB and the second group to acquire an explanation from economic experts on why the inflation rate in Germany has declined over this horizon. Respondents are asked whether they want to acquire the explanation or alternatively get an additional payment. We clearly state that one of the scenarios will be randomly implemented by a computer. We vary the amount of the additional payment – allowing us to derive individual demand curves for each individual.
We also provide the groups with the opportunity to sign up for a cost-free ECB newsletter or a cost-free newsletter from the ifo Institute Munich. The first group will receive the question on the ECB newsletter, the second group the question on the ifo newsletter.
After the computer has decided on a scenario, depending on the respondents' decision in that scenario, respondents' either receive the explanation from the ECB (group 1) or the explanation from experts (group 2) or are informed about the amount of their additional payout (and that they receive no additional information).
Finally, we then ask participants about their posterior expectations on the inflation rate.
Our main hypothesis is that households’ demand for expert narratives is higher compared to households’ demand for central bank explanations.