Does the Media Help the General Public in Understanding Inflation?
| Date | 01 December 2018 |
| DOI | http://doi.org/10.1111/obes.12252 |
| Published date | 01 December 2018 |
| Author | David‐Jan Jansen,Matthias Neuenkirch |
1185
©2018 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd.
OXFORD BULLETIN OF ECONOMICSAND STATISTICS, 80, 6 (2018) 0305–9049
doi: 10.1111/obes.12252
Does the Media Help the General Public in
Understanding Inflation?*
David-Jan Jansen† and Matthias Neuenkirch‡,§
†De Nederlandsche Bank, Financial Stability Division, 1000 AB Amsterdam, The
Netherlands (e-mail: djansenresearch@gmail.com)
‡Faculty IV – Economics, University of Trier, 54296 Trier, Germany
§CESifo, Munich, Germany
Abstract
This paper studies whether media information helps the general public in understanding
inflation. We combine detailed Dutch household survey data on media usage, inflation
perceptions, and inflation expectations. We find no evidence that more-often informed
members of the general public do better in understanding inflation. In fact, more frequent
readership of popular newspapers is associated with slightly less accurate inflation per-
ceptions. There is also no evidence that usage of non-print media leads to more accurate
of views on inflation. One implication of these results is that central banks might need to
consider more direct ways of engaging with the general public.
I. Introduction
This paper asks whether media usage helps members of the general public in understanding
inflation. Since the seminal work by Jonung (1981), it is well known that inflation percep-
tions and expectations vary across sociodemographic clusters. One of his findings was that
expected rates declined with age, which could be explained by different inflation experi-
ences of cohorts over their lifetime. Recently, similar evidence is provided by Malmendier
and Nagel (2016), who analyse over50 years of data from the Reuters/Michigan Survey of
Consumers. The list of relevantbackground f actors is extensive and includes, in addition to
age, variables such as gender, income, and education.The relevance of sociodemographic
backgrounds for views on inflation has been affirmed in many studies, including Bryan
and Venkatu (2001), Souleles (2004), Bruine de Bruin et al. (2010), and Ehrmann, Pfajfar
and Santoro (2017).
JEL Classification numbers: D1, D8, E3, E5.
*Wethank Mark Boukes, W¨andi Bruine de Bruin, Carin van der Cruijsen, Lena Dr¨ager,Michael Ehrmann, Jakob
de Haan, Bernd Hayo, Richhild Moessner, Edith Neuenkirch, Florian Neumeier, Sanne Peeters, Ricardo Reis, Kalle
Rinne, Maarten van Rooij, Stefania Rossi, Francesco Zanetti (the Editor), three anonymous referees and participants
of the 2016 European Public Choice Conference, the 2016 World Finance Conference, the Workshop Household
Surveys in Macroeconomics in Hamburg, the 2017 ESCB Research Conference at the Banco de Espa˜na, and the
2018 Luxembourg Workshop on Household Finance and Consumption for useful comments and suggestions.
1186 Bulletin
The manner in which the media reports on economic developments could be one factor
that explains heterogeneity in views on inflation, as media consumption also differs across
sociodemographic backgrounds. Papers that study the potential role of economic reporting
often take a perspective that differs from a traditional rational-expectations approach. In
particular, these papers incorporate the idea that economic agents are not necessarily always
fully attuned to the relevant news on the economy. In a general setting, (2003) builds on
insight from information theory. He uses the idea of rational inattention, which follows
from the notion that agents have only a finite capacity to process information. One of the
implications is that the way in which people react to news depends on how the media
presents it. Carroll (2003) builds a theoretical model, using insights from epidemiology.
He assumes that individuals obtain information on the macroeconomy from the media, but
only probabilistically. As a consequence, there is a delay before relevant macroeconomic
news reaches every member of the public. Given that his model generates slow-moving
changes in aggregate expectations, it provides a microfoundation for sticky-information
type models (Mankiw and Reis, 2003). In line with the predictions of his model, Carroll
finds evidence that people are better informed when there is more news coverage.
Although the idea that media coverage affects people’s views on inflation is appealing,
the empirical evidence remains somewhat mixed. Some papers detect evidence supportive
of news effects. Lamla and Lein (2015) find that media reporting has had meaningful
impact on inflation perceptions and contributed to their sharp rise in the aftermath of
the euro cash changeover. Dr¨ager and Lamla (2017) reveal that when people have heard
news on inflation, they are more likely to adjust their forecasts. Lamla and Maag (2012)
document that media coverage affects disagreement of consumers. In particular, the level
of disagreement increases with heterogeneity of media coverageand declines in the amount
of reports. On the other side of the spectrum, Pfajfar and Santoro (2013) find only weak
evidence in support of media effects.They test several predictions of Carroll’s theory using
microdata from the Michigan Survey. Their conclusion is that hearing news on prices
does not necessarily produce better forecasts, although it does increase the likelihood that
people revise their expectations. Using survey data for Dutch households,Van der Cruijsen,
Jansen and De Haan (2015) show that more intensive use of media information improves
respondents’ understanding of the ECB. However, they find no empirical evidence that
media consumption has a direct effect on the likelihood of formulating realistic inflation
expectations. There are also papers wherenews effects only occur under certain conditions.
For instance, Lamla and Lein (2015) do find that intensive news reporting improves the
accuracy of inflation expectations. However, this effect only occurs if the newsis not framed
negatively. Dr¨ager (2015) finds evidence that in Sweden, media effects have occurred, but
then mostly during times of increasing inflation.1
Given the mixed evidenceon news effects, this paper wants to look further at if and how
media consumption matters for views on inflation. In line with this aim, a first distinctive
aspect of this paper is the use of granular data on media usage. Related papers on news
effects either rely on a national measure of media reporting (Carroll, 2003; Lamla and Lein,
2015), or use a general question from surveys on whether people have heard news about
1There is also a related literature that studies knowledge of monetary policy and the role of central bank commu-
nication. See, e.g. Carvalho and Nechio (2014), Dr¨ager, Lamla and Pfajfar (2016), Hayo and Neuenkirch (2018).
©2018 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd
To continue reading
Request your trial