RDP 2021-05: Central Bank Communication: One Size Does Not Fit All Non-technical summary for ‘Central Bank Communication: One Size Does Not Fit All’

High-quality central bank communication can improve the effectiveness of monetary policy and minimise disruptions when monetary policy changes. There has, however, been very little study on how to make central bank communication more effective. We aim to fill this gap by developing a novel approach that uses survey data and machine learning models to assess the communication quality of central bank publications.

We focus on how different audiences, in particular economists and non-economists, perceive the readability of and the degree of reasoning in various economic communications. We argue that reasoning, not just readability, is an important part of the broader concept of central bank transparency.

We have 3 main findings. First, simple readability indices, such as the Flesch–Kincaid grade level, are barely correlated with people's survey ratings of text readability or reasoning. This suggests that a focus on simple readability metrics may fail to increase a broad measure of readability or the transparency of documents overall.

Second, economists and non-economists understand economic communication differently. There is no correlation between the way the economists and non-economists in our survey perceive the readability and reasoning of the same piece of text (see the figure). What is clear to economists can be unclear to non-economists and vice versa. Our machine learning results reinforce this observation. One size does not fit all.

Finally, there seems to be a trade-off between readability and reasoning. The introductions of documents tend to be more readable but contain less reasoning, while the conclusions tend to be less readable but contain more reasoning. Different parts of a document have different objectives and trying to achieve multiple objectives within a single paragraph is difficult. Consequently, the application of any single metric to an entire document will fail to capture the need for different emphases at different points. Thus, text quality metrics – including our own – should be used to inform rather than prescribe and not be applied indiscriminately.

Our results, while preliminary and subject to a number of limitations, all point in a similar direction: communication needs to be adapted for different audiences and no single measure can do justice to the multiplicity of objectives communication has. Thus, central banks aiming to improve their communications and transparency may need to present their core arguments in a range of formats with different expressions of those arguments in each to reach different audiences.

Back to abstract