THE POWER OF WHISPERS: A THEORY OF RUMOR, COMMUNICATION, AND REVOLUTION*

• Author: Heng Chen,Yang K. Lu,Wing Suen
• Date: 01 February 2016
• DOI: http://doi.org/10.1111/iere.12149
• Published date: 01 February 2016

INTERNATIONAL ECONOMIC REVIEW

Vol. 57, No. 1, February 2016

THE POWER OF WHISPERS: A THEORY OF RUMOR, COMMUNICATION,

AND REVOLUTION∗

BYHENG CHEN,YANG K. LU,AND WING SUEN1

University of Hong Kong, Hong Kong; Hong Kong University of Science and Technology,

Hong Kong; University of Hong Kong, Hong Kong

We study how rumors mobilize individuals who take collective action. Rumors may or may not be informative,

but they create public topics on which people can exchange their views. Individuals with diverse private information

rationally evaluate the informativeness of rumors about regime strength. A rumor against the regime can coordinate a

larger mass of attackers if individuals can discuss its veracity than if they cannot. Communication can be so effective

that a rumor can have an even greater impact on mobilization than when the same story is fully believed by everybody.

However, an extreme rumor can backfire and discourage mobilization.

1. INTRODUCTION

Collective actions, such as riots, currency attacks, and bank runs, are often immersed in

rumors. Perhaps the most dramatic place to witness rumors in action is a political revolution.

Amid the recent Tunisian revolution, Ben Ali, the ex-Tunisian leader, was said to have fled his

country. This was confirmed after conflicting rumors about his whereabouts and finally led to

the end of street protests. A while later in Egypt, it was widely reported that Mubarak’s family

had left for London, which was believed by many as a clear sign of fragility of the regime. Similar

rumors about Qaddafi and his family appeared in Libya when the battle between the opposition

and the regime intensified. Rumors are not limited to the series of revolutions in the Arab

Spring. During the 1989 democracy movement in China, rumors repeatedly surfaced about

the death of the leaders, Deng Xiaoping and Li Peng, as well as the divide among communist

leaders.2

Are rumors just rumors? In many cases, yes. Rumors that spread during turmoils often

disappear quickly without a trace. This seems to be natural, as rational individuals may discount

unreliable information they receive in those situations. However, many historical incidents

suggest that rumors often turn out to be particularly effective in mobilization. The Velvet

Revolution in Czechoslovakia was described as a “revolution with roots in a rumor” (Bilefsky,

2009). At the dawn of the revolution, a prominent (false) rumor that a 19-year old student was

brutally killed by the police triggered many otherwise hesitant citizens to take to the streets. The

revolution gained huge momentum right after that and the regime collapsed a few days later.

∗Manuscript received May 2013; revised August 2014.

1We thank Roland B´

enabou, Matthias Doepke, Christian Hellwig, Satoru Takahashi, Jianrong Tian, Yikai Wang,

Fabrizio Zilibotti, and seminar participants at NBER Summer Institute 2012, UBC-HKU Microeconomics Workshop,

Society for Economic Dynamics 2013 Meeting, University of Hong Kong, Toulouse School of Economics, European

University Institute, and National University of Singapore for their suggestions. Guangyu Pei deserves praise for

providing excellent research assistance to this project. Our research is partially funded by the Research Grants Council

of Hong Kong (Project No. HKU 742112B). Please address correspondence to: Heng Chen, School of Economics and

Finance, University of Hong Kong, Room 915, K.K. Leung Building, Pokfulam Road, Hong Kong, Hong Kong. E-mail:

hengchen@hku.hk.

2There were widespread rumors of many variants that Deng died of illness during the protest and that Li was shot to

death. It was also widely rumored in the media that some senior leaders in the Communist Party wrote an open letter

to oppose taking any action against students. See, for example, the news story in the daily newspaper Ming Paoon June

6, 1989.

89

C

(2016) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social

and Economic Research Association

90 CHEN,LU,AND SUEN

In the Arab Spring, the news about Mubarak’s family proved to be false, yet the opposition

credited it for “mark[ing] a new phase” in their campaign.3Chinese history also offers many

anecdotes in which rumors mobilized mass participation, including the Boxer Uprising, the

Republican Revolution, and the May Fourth Movement (Zhang, 2009). Similarly, riots are

often amplified or even sparked by rumors as well: the 1921 Tulsa race riot, the 1967 Newark

riot, and the 2004 Rome riot provide dramatic examples.

A common interpretation of the role of rumors in mass movements is that individuals are just

blindly herded by them. However, we adopt the position that individuals are fully aware that

rumors circulating in times of turmoil may or may not be well founded and that they update

their beliefs in a Bayesian manner. Since rumors are widely circulated and commonly observed,

they may serve as a coordination device just like a public signal in a coordination game. We

explain why some rumors are effective in mobilizing participation in collective actions whereas

others are not.

In this article, we focus on two key aspects of rumors: that they may be true or false and

that people talk about them. Individuals in times of uncertainty and crisis often seek others’

opinions and discuss with peers about their judgment and evaluation of rumors. Information

from fellow citizens can influence their beliefs and even actions. The core of our article is to

show that communication among individuals centering around a public topic can substantially

change outcomes of collective actions.

Specifically, we model political revolution as a global game. Citizens are uncertain about the

regime’s strength and possess dispersed private information about it. A citizen’s incentive to

revolt increases with the aggregate action of all other citizens. If there are sufficient participants,

the regime collapses; otherwise, it survives. Before citizens take actions, they hear a rumor about

the regime. This rumor is a publicly observed message, which could be either an informative

signal about the regime’s strength or an uninformative noise unrelated to fundamentals. Citizens

assess the informativeness of the rumor based on their private information. As a consequence of

diverse private information among citizens, their assessments may also differ. Further, citizens

communicate with one another and tell their peers whether they believe the rumor or not.

In this model, the degree of skepticism is endogenous: Citizens whose private information

differs more from the rumor are more skeptical of it. Due to this skepticism, rumors against the

regime mobilize fewer attackers than when such news is known to be trustworthy. If a rumor

is far different from the fundamental, it will also differ from most citizens’ private information

and therefore be heavily discounted by them. As a result, extreme rumors have little impact on

equilibrium outcomes.

When citizens communicate, those whose private information is close to the rumor will tell

their peers that the rumor is informative. Recipients of confirmatory messages treat what their

peers say as evidence for the truth being close to what the rumor suggests and therefore become

more responsive to the rumor. Consider, for example, a rumor against the regime. A fraction

of the population (those with intermediate private information) will attack the regime if their

peers tell them that the rumor is informative and will not attack otherwise. If the rumor is

indeed near the true strength, more citizens will receive confirmatory messages from their

peers. Therefore, communication helps such a rumor to mobilize more attackers. By the same

mechanism, if the rumor is far from the truth, most citizens will express disbelief to their peers,

which discourages attacking. Therefore, communication overcomes or reinforces skepticism

about a rumor, depending on whether the rumor is close to the fundamental or not.

Interestingly and surprisingly, we find that communication could make rumors even more

powerful than trustworthy news in mobilizing individuals. For the same news against a regime,

it is possible that the regime would survive if all citizens believe that the news is informative and

fully trustworthy, but would collapse if citizens are skeptical about its veracity and talk about

3World Tribunereported on January 28, 2011, that “confirmed by a Western diplomat, ... Mubarak’s wife, Suzanne,

son, Gamal, and granddaughter arrived in London on a private jet as Egypt’s defense minister secretly flew to the

United States.”