The Social Value of Information: A Test of a Beauty and Nonbeauty Contest

• Published date: 01 October 2022
• Author: THOMAS LUSTENBERGER,ENZO ROSSI
• Date: 01 October 2022
• DOI: http://doi.org/10.1111/jmcb.12903

DOI: 10.1111/jmcb.12903

THOMAS LUSTENBERGER

ENZO ROSSI

The Social Value of Information: A Test of a Beauty

and Nonbeauty Contest

We demonstrate how the precision of public and private information can

be measured and their welfare implications can be assessed. To this end,

we develop and apply a procedure to test welfare results from a beauty and

nonbeauty contest model based on surveyforecasts of interest rates in a large

country sample over an extended period of time. In most countries, interest

rate forecasts are unbiased. In half of the countries, a higher precision of

public information regarding interest rates increases welfare. This applies in

particular to small open and emerging market economies. During forward

guidance, public information is less precise than private information. On

average, forecasters place greater weight on public than privateinformation

and rapidly correct errors contained in private information, while errors in

public information last over the whole forecast horizon.

JEL codes D8, E43, E52, E58, G1, G29

Keywords: beauty contest, central bank transparency, forward guidance,

interest rate forecasts, strategic forecasts, value of information

Keynes (1936) argued that nancial markets are excessively

volatile because professional investors are more concerned with forecasting the fore-

casts of others than forecasting the fundamental value of the assets they trade. This

The authors thank Romain Baeriswyl, Alex Cukierman, Sylvia Kaufmann, Mico Loretan, Cyril Mon-

net, Marcel Savioz, Florian Schaffner, Martin Summer, Heinz Zimmermann, and two anonymousrefer-

ees for their valuable comments and suggestions. In addition, the authors are grateful for the comments

from participants in the BuBa-OeNB-SNB Workshop 2016, the SNB BrownBag, the Kolloquium of the

University of Basel, the BrownBag of the Study Center Gerzensee, the SFI Research Days 2017, the IV

Gerzensee Alumni Conference 2017, and the Theories and Methods in Macroeconomics Conference 2018

(T2M). A rst version circulated under the title, “Social Valueof Public Information: Testing Morris and

Shin (2002).” An extended version carrying the paper’s nal title was published as SNB Working Paper

2017-17 and as WWZ Working Paper 2018-05. This research project was conducted while the rst author

was afliated with the Swiss National Bank (SNB) and the University of Basel. The views expressed in

this paper are those of the authors and do not necessarily reect those of the SNB or the Swiss Financial

Market Supervisory Authority FINMA.

T L is a Senior Research Associate/Secretary to the Board,Swiss Financial Mar-

ket Supervisory Authority FINMA (E-mail: thomas.lustenberger@me.com). E R is an Economic

Advisor at Swiss National Bank (E-mail: enzo.rossi@snb.ch).

Received November 25, 2019; and accepted in revised form November8, 2021.

Journal of Money, Credit and Banking, Vol. 54, No. 7 (October 2022)

© 2021 The Ohio State University.

2126 :MONEY,CREDIT AND BANKING

has become known as Keynes’ beauty contest metaphor for nancial markets. Morris

and Shin (2002) (MS) developed a theoretical model nested in a beauty-contest game

in which agents observe private and public information to form their actions. Agents’

utility positively depends on correctly forecasting the true (realized) future state of

the economy on the one hand and on forecasting as close to the other agents as pos-

sible (Keynes’s Beauty Contest) on the other hand. The key implication is that when

individuals have access to heterogeneous private information, more precise public

information may be damaging to welfare. Applied to nancial markets, when partic-

ipants overreact to noisy public news because they help forecast each other’s actions,

public information can contribute to more volatility and lower welfare.1This pos-

sibility has widely been interpreted as an objection to more transparency in central

banks.2

Considerable debate has been prompted by the MS model, for which there is no

agreement. To the best of our knowledge, there has been no attempt to test it. How-

ever, an empirical analysis may assist in clarifying views. This is what we do in this

paper. The empirical problem that must be solved is determining the three exogenous

parameters—the precision of the privateinformation, the precision of the public infor-

mation, and the coordination motive—which are difcult to estimate. We circumvent

this problem by testing three hypotheses derived from the MS model and contrast

the results with empirical tests on two hypotheses derived from a nonbeauty contest

model in which agents do not second-guess other agents’ actions (forecasts). From

several potentially testable models with this feature, we selected the model that was

proposed by Veldkamp(2011). In the MS model, agents have an incentive to make as

good a forecast as possible and to be close to the other forecasts. However, welfare

only depends on how close forecasts are to the fundamental values of the economy.

By contrast, in the Veldkamp (2011) model, forecasters only have an incentive to

make an accurate forecast, while welfare depends on both the quality of forecasts

and their alignment with each other. Both models are nested in the theoretical frame-

work proposed by Angeletos and Pavan (2007) (AP).

Our tests are performed on data of professional forecasts of short-term interest rates

provided by Consensus Economics. The data set includes a large set of countries over

an extended period of time, starting at the end of the 1980s and ending in June 2018.

We use survey forecasts for two reasons: (i) forecasts can be interpreted as agents’

actions and (ii) are important for policymakers and the general public as demonstrated

by the active monitoring of forecasters’ accuracy by market participants, the popular

press, and researchers (Marinovic, Ottaviani, and Sørensen 2013).

1. See Angeletos, Iovino, and La’O (2016)

2. Following MS, beauty-contest models have received close attention in other situations, such as

investment games (Angeletos and Pavan 2004), nancial markets (Allen, Morris, and Shin 2006), and

additional economic questions (Angeletos and Pavan 2007). See Myatt and Wallace(2011) for more ref-

erences.