CHAPTER 10
Social media and protest participation: 121
Evidence from Russia
SOCIAL MEDIA AND PROTEST PARTICIPATION: EVIDENCE FROM RUSSIA | ENIKOLOPOV, MAKARIN AND PETROVA
Ruben Enikolopov,ac Alexey Makarinb and Maria Petrovaac
a
Universitat Pompeu Fabra; b MIT Sloan School of Management; cCEPR
Social media is playing an increasingly important role in our lives. The political
consequences of the advent of social media are now a hotly contested topic (Zhuravskaya
et al. 2020). An optimistic view has been that social media would empower ordinary
citizens and serve as a ‘liberation technology’ (Diamond and Plattner 2010) that would lead
to faster democratisation in authoritarian countries (Shirky 2008) and make politicians
more accountable (Besley and Prat 2006). Evidence of the effect of the spread of mobile
phones in Africa seems to support the idea that new communication technologies can
help mass political mobilisation (Manacorda and Tesei 2020). A more pessimistic view
has been that autocratic governments would adjust and learn how to exploit social media
to their own advantage, employing a combination of tools including online censorship,
surveillance, and heavy use of online bots and trolls (Morozov 2011, Roberts 2018). In
democracies, social media is now often blamed for the exacerbated political polarisation,
the spread of xenophobic ideas, the proliferation of fake news and general negative effects
on users’ wellbeing (Tufekci 2018, Allcott et al. 2020, Braghieri et al. 2022).
There is plentiful evidence that traditional media (newspapers, radio, TV) have had an
important impact on political outcomes by providing political news and entertainment
(DellaVigna and La Ferrara 2015, Enikolopov and Petrova 2015). In many respects, online
media resemble traditional media and one should expect their persuasion effects to
mirror those of traditional media. However, certain features of new media – of social
media, in particular – are distinct. The two most important distinguishing features of
new social media are low barriers to entry and reliance on user-generated content. Low
entry barriers make gatekeeping the spread of political information much less effective,
allowing new entrants who were previously sidelined by the political establishment. By
providing an outlet for the opposition and for whistleblowers, social media makes it harder
for political and business actors to hide potentially harmful information, increasing their
accountability (Besley and Prat 2006). Low entry barriers can also have social costs. For
example, social media can be used to spread extremist ideas, increasing their reach and
potentially their influence. Low barriers to entry could also undermine the reputation
mechanisms that serve to guard the quality of information of traditional media outlets
(Gentzkow and Shapiro 2006, Cagé 2020) and lead to the spread of misinformation and
fake news. Low barriers to entry also vastly increase the choice of news sources and,
arguably, allow users to tailor their news sources to their pre-existing preferences more
finely than traditional media allow; this potentially could give rise to ‘echo chambers’
and lead to increased political polarisation.
By allowing horizontal flows of information through two-way communication between
122
users, social media facilitates coordination between people, thus potentially making it
easier to organise collective actions such as street protests. At the same time, online protest
THE POLITICAL ECONOMY OF SOCIAL MEDIA
activity in social media could crowd out offline actions necessary for real political change
in autocracies (Gladwell 2010). User-generated content and two-way communication
in social media could also change the way politicians and citizens interact (Bessone
et al. 2022). Social media allows politicians to receive immediate feedback on policy
actions, discuss policy proposals and measure political discontent. Such feedback could
be used for policy improvements, but it could also be used for oppression and political
surveillance. In addition, the low cost of creating automated accounts and the ability
to post content using anonymous or impersonated accounts enable the manipulation of
online content seen by real users, potentially leading to political persuasion. Also, the
data that online platforms collect about their users could be (and have been) used to
target specific groups of users to make such manipulations more effective.
The combination of low barriers to entry and horizontal flows of information could
make social media especially important in facilitating street protests. Low barriers to
entry in social media make it easier to spread information critical of the government,
which is especially important in autocratic regimes where traditional media is under a
tight control by the government. This increases the number of informed citizens who
are unhappy with their government and thus potentially ready to take part in political
protests. Furthermore, horizontal flows of information between users of social media
allow them to exchange logistical information about the upcoming events and coordinate
their tactics on the spot. This helps solve collective-action problems and increases the
probability that protests actually take place by increasing the probability that people who
are potentially ready to participate in political protests actually do participate.
In our paper (Enikolopov et al. 2020),1 we provide causal estimates of the political effects
of social media in a non-democratic environment, focusing on the effect on participation
in political protests. Testing empirically the effect of social media on political protests
is methodologically challenging because social media usage is endogenous to individual
and community characteristics. In addition, protests are typically concentrated in one or
a few primary locations, as was the case for Tahrir Square in Egypt or Maidan in Ukraine.
Hence, geographic variation in protests is often very limited. Temporal variation in
protest intensity can provide evidence on the association between the activity and the
content on social media and subsequent protests (Acemoglu et al. 2017), but not on the
causal impact of social media availability.
1 The results of which hold after correcting for the mistakes in two control variables (Enikolopov et al. 2023).
To understand whether social media can indeed promote protest participation, we
study an unexpected wave of political protests in Russia in December 2011 triggered by
electoral fraud in parliamentary elections, coupled with an analysis of the effect of social
media on support for the government. Our empirical setting allows us to overcome the
limitations of previous studies for two reasons. First, there was substantial geographic 123
and temporal variation in both protest activities and the penetration of the major online
social networks across Russian cities. For example, among the 625 cities in our sample, 133
SOCIAL MEDIA AND PROTEST PARTICIPATION: EVIDENCE FROM RUSSIA | ENIKOLOPOV, MAKARIN AND PETROVA
witnessed at least one protest demonstration on 10–11 December 2011, the first weekend
after the elections. Second, particularities of the development of VKontakte (VK), the
most popular social network in Russia, allow us to exploit quasi-random variation in the
penetration of this platform across cities and ultimately identify the causal effect of social
media penetration on political protests.
Our identification is based on the information about the early stages of VK’s development.
VK was created in 2006 by Pavel Durov, a student at Saint Petersburg State University
(SPbSU). This online social network, analogous to Facebook in functionality and design,
was the first mover in the Russian market and secured its dominant position with a
user share of over 90% by 2011. Initially, users could only join the platform by invitation
through a student forum of the university, which was also created by Durov. As a result,
the vast majority of early users of VK were Durov’s fellow students at SPbSU. This, in
turn, made friends and relatives of these students more likely to open an account early
on. Since SPbSU attracted students from across the country, this sped up the propagation
of VK in the cities these students had come from. Network externalities magnified these
effects and, as a result, the distribution of the home cities of Durov’s classmates had a
long-lasting effect on VK penetration.
We exploit this feature of VK development in our empirical analysis by using the origin
of students who studied at SPbSU in the same five-year cohort as the VK founder as an
instrument for VK penetration in the summer of 2011, controlling for the origin of the
students who studied at SPbSU several years earlier and later. Thus, our identification is
based on the assumption that temporal fluctuations in the number of students coming to
SPbSU from different Russian cities were not related to unobserved city characteristics
correlated with political outcomes. In the first-stage regression, we find that the
distribution of the home cities of the students who studied at SPbSU at the same time as
Durov predicts the penetration of VK across cities in 2011, whereas the distribution of the
home cities of the students who studied at SPbSU several years earlier or later does not
(see Figure 1).
Using this instrument, we estimate the causal impact of VK penetration on the
incidence of protests and protest participation. In the reduced-form analysis, we show
that fluctuations in the student flow from Russian cities to SPbSU over time predict the
incidence of protests in December 2011 (Figure 2) and the number of protest participants.
The corresponding IV estimates indicate that the magnitude of the effect is sizable: a 10%
increase in the number of VK users in a city led to both a 4.6 percentage point increase
in the probability of there being a protest and a 19% increase in the number of protest
participants the first weekend after the elections. Non-parametrically, we document that
there exists a threshold of VK penetration below which there is no relation between VK
penetration and protests (see Figure 3).
124
FIGURE 1 VK PENETRATION IN 2011 AND SPBSU STUDENT COHORTS
THE POLITICAL ECONOMY OF SOCIAL MEDIA
Log (SPbSU students), one
cohort older than VK founder
Log (SPbSU students), same
5-year cohort as VK founder
Log (SPbSU students), one
cohort younger than VK founder
-.2 -.1 0 .1 .2
FIGURE 2 INCIDENCE OF PROTESTS IN 2011 AND SPBSU STUDENT COHORTS
Log (SPbSU students), one
cohort older than VK founder
Log (SPbSU students), same
5-year cohort as VK founder
Log (SPbSU students), one
cohort younger than VK founder
-.1 -.05 0 .05 .1
FIGURE 3 NONPARAMETRIC RELATIONSHIP BETWEEN VK PENETRATION AND NUMBER
OF PROTESTERS
10
Log(Number of Protesters), 2011 125
8
SOCIAL MEDIA AND PROTEST PARTICIPATION: EVIDENCE FROM RUSSIA | ENIKOLOPOV, MAKARIN AND PETROVA
2 4 0 6
6 8 10 12 14
Log(VK users), 2011
As a falsification test, we show that VK penetration in 2011 does not predict protest
participation in the same cities before the creation of VK using three different protest
instances: anti-government protests at the end of the Soviet Union (1987–1992), labour
protests in 1997–2002, and social protests in 2005. We also show that VK penetration in
2011 was not related to voting outcomes before the creation of VK. We also replicate our
first-stage regressions using the information on the cities of origin of the students who
studied in more than 60 other major Russian universities. We find that the coefficient for
our instrument – VK founder’s cohort at SPbSU – lies at the top end of the distribution of
the corresponding coefficients in other universities, while the coefficients for younger and
older cohorts lie close to the medians of the corresponding distributions, consistent with
our identifying assumptions.
We highlight two channels through which social media could lead to protest participation
in a non-democracy. On the one hand, low barriers to entry make it much more difficult
for the regime to limit the spread of potentially harmful information that would lead
to more anti-government sentiments in the population. We call this the information
channel. On the other hand, the fact that social media relies on user-generated content
facilitates horizontal information flows, which could lower the costs of coordination and
thus alleviate the collective action problem (Ostrom 1990). We call this the coordination
channel. If this channel is at work, the chances that people take out to the streets could
go up even if the number of individuals in opposition does not increase. In our context,
VK was used heavily for tactical coordination of protests in Russia in 2011-2012. For
instance, for almost every city with a protest, activists created VK protest communities,
where people could exchange the logistical details.
We find no evidence that supports the information channel being at work in our context.
In particular, we study VK’s impact on the pro-government vote and attitudes toward
the regime. We show that, consistently across all elections after the creation of the social
network, VK led, if anything, to a higher, not lower, pro-government vote. We also do
126 not find any evidence for increased political polarisation since there was no jump in
negative attitudes toward the regime or in the opposition vote. Finally, we analyse the
political content on VK and find that, on average, it was neutral or positive towards the
THE POLITICAL ECONOMY OF SOCIAL MEDIA
government.
However, we do find evidence in favour of coordination. We find that the number of VK
users in online protest communities was positively associated with the incidence and
the size of the protests. We also find that the impact of social media on protests was
stronger in larger cities, where logistical coordination tends to be more critical. Finally,
we show that protests tend to be smaller in cities where, conditional on the total number
of social media users, the user base was more fractionalized between Facebook and VK.
A more divided user base matters because it may lead to less horizontal information
flows between users of different social networks and, as a result, more difficult logistical
coordination. These findings may also be consistent with the importance of peer
pressure and social image, and we explore this hypothesis further in a companion project
(Enikolopov et al. 2022).
Overall, our paper provides evidence that social media penetration had a causal effect on
both the incidence and the size of the protest demonstrations in Russia in December 2011.
Additional evidence suggests that social media affects protest activity by reducing the
costs of collective action, rather than by spreading information critical of the government
or by increasing political polarisation. Thus, our results imply that social media induces
coordination and alleviates the collective action problem.
While our results confirm the earlier claims of the digital optimists, we note that these
results may not generalise to other settings. The Russian protests of 2011 were unexpected,
and the government did not have time to prepare for them. If a government is aware of
this effect of social media on political protests, it may counteract it by censoring online
content related to collective action (King et al. 2014, Ananyev et al. 2019) or manipulating
the information in social media (King et al. 2017). Furthermore, the reduction in
coordination costs can also have its dark sides, for example by possibly leading to more
extremism and hate crimes (Bursztyn et al. 2019). More research is needed to understand
whether similar results hold for other outcomes and in different contexts.
REFERENCES
Acemoglu, D, T A Hassan and A Tahoun (2017), “The Power of the Street: Evidence From
Egypt’s Arab Spring”, Review of Financial Studies 31(1): 1–42.
Allcott, H, L Braghieri, S Eichmeyer, and M Gentzkow (2020), “The Welfare Effects of
Social Media”, American Economic Review, 110(3): 629-676.
Ananyev, M, D Xefteris, G Zudenkova, and M Petrova (2019), “Information and
Communication Technologies, Protests, and Censorship”, working paper.
127
Besley, T, and A Prat (2006), “Handcuffs for the Grabbing Hand? Media Capture and
Government Accountability”, American Economic Review 96(3): 720–736.
SOCIAL MEDIA AND PROTEST PARTICIPATION: EVIDENCE FROM RUSSIA | ENIKOLOPOV, MAKARIN AND PETROVA
Bessone, P, F R Campante, C Ferraz and P Souza (2022), “Social Media and the Behavior
of Politicians: Evidence from Facebook in Brazil”, NBER Working Paper No. 30306.
Braghieri, L, R E Levy and A Makarin (2022), “Social media and mental health|,
American Economic Review 112(11): 3660-3693.
Bursztyn, L, G Egorov, R Enikolopov, and M Petrova (2019), “Social Media and
Xenophobia: Evidence from Russia”, Working paper.
Cagé, J (2020), “Media Competition, Information Provision and Political Participa- tion:
Evidence from French Local Newspapers and Elections, 1944-2014”, Journal of Public
Economics 185: 104077.
DellaVigna, S and E La Ferrara (2015), “Economic and Social Impacts of the Media”, in
Handbook of Media Economics, Vol. 1, Elsevier.
Diamond, L and M Plattner (2010), “Liberation Technology”, Journal of Democracy 21:
69–83.
Gentzkow, M and J M Shapiro (2006), “Media Bias and Reputation”, Journal of Political
Economy 114(2): 280–316.
Enikolopov, R, and M Petrova (2015), “Media Capture”, in Handbook of Media Economics,
Vol. 1, Elsevier.
Enikolopov, R, A Makarin, and M Petrova (2020), “Social Media and Protest Participation:
Evidence from Russia“, Econometrica 88(4): 1479–1514.
Enikolopov, R, A Makarin, M Petrova, and L Polishchuk (2022), “Social Image, Networks,
and Protest Participation”, working Paper.
Enikolopov, R, A Makarin, and M Petrova (2023), “Online Corrigendum to “Social Media
and Protest Participation: Evidence From Russia”, Econometrica 91(3): 1-24.
Gladwell, M (2010), “Small Change”, The New Yorker, 27 September.
King, G, J Pan, and M E Roberts (2014), “Reverse-Engineering Censorship in China:
Randomized Experimentation and Participant Observation”, Science 345(6199): 1–10.
King, G, J Pan, and M E Roberts (2017), “How the Chinese Government Fabricates Social
Media Posts for Strategic Distraction, Not Engaged Argument”, American Political
Science Review 111(3): 484–501.
Manacorda, M and A Tesei (2020), “Liberation Technology: Mobile Phones and Political
128
Mobilization in Africa”, Econometrica 88(2): 533-567.
Morozov, E (2011), The Net Delusion: The Dark Side of Internet Freedom, Perseus Books.
THE POLITICAL ECONOMY OF SOCIAL MEDIA
Roberts, M (2018), Censored: Distraction and Diversion Inside China’s Great Firewall,
Princeton University Press.
Shirky, C (2008), Here Comes Everybody: The Power of Organizing Without Organizations,
AllenLane.
Tufekci, Z (2018), “How Social Media Took Us from Tahrir Square to Donald Trump”,
MIT Technology Review.
Zhuravskaya, E, R Enikolopov, and M Petrova (2020), “Political Effects of the Internet
and Social Media”, Annual Review of Economics 12: 415–438.
ABOUT THE AUTHORS
Maria Petrova is ICREA Research Professor at the Universitat Pompeu Fabra and
Barcelona Institute for Political Economy and Governance (IPEG), and an Affiliate
Professor at Barcelona School of Economics. Maria is a Co-Editor of the Journal of Public
Economics and a member of the Editorial Board of the Review of Economic Studies since
2016. She received PhD from Harvard University in 2008. She spent 2012 - 2013 as a
Visiting Associate Research Scholar at the Center for the Study of Democratic Politics at
Princeton University. In 2012-2013, she was also the Research Director at the Center for
New Media and Society. Her research interests include mass media economics, political
economy, and corporate governance. Her paper have been published in leading journals,
including American Economic Review, Quarterly Journal of Economcs, Econometrica,
American Political Science Review, and Management Science.
Alexey Makarin is an Assistant Professor of Applied Economics and a Mitsubishi Career
Development Assistant Professor in International Management at the MIT Sloan School
of Management. He earned his PhD in Economics from Northwestern University in
June 2019. Alexey’s research centres around political economy and the economics of
digitization, with a special focus on the causal effects of social media platforms.
Maria Petrova is ICREA Research Professor at the Universitat Pompeu Fabra, Researcher
at the Barcelona Institute for Political Economy and Governance (IPEG), Affiliate
Professor at Barcelona Graduate School of Economics, and the Alfa-Bank Associate
Professor of Economics at the New Economic School. Maria is a Co-Editor of the Journal
of Public Economics and a member of the Editorial Board of the Review of Economic
Studies since 2016. She received PhD from Harvard University in 2008. She spent 2012 -
2013 as a Visiting Associate Research Scholar at the Center for the Study of Democratic
Politics at Princeton University. In 2012-2013, she was also the Research Director at the
Center for New Media and Society. Her research interests include mass media economics,
political economy, and corporate governance. 129
SOCIAL MEDIA AND PROTEST PARTICIPATION: EVIDENCE FROM RUSSIA | ENIKOLOPOV, MAKARIN AND PETROVA