Skip to main content

Demand for Democracy

Periodic Reporting for period 2 - DemandDemoc (Demand for Democracy)

Reporting period: 2018-09-01 to 2020-02-29

"Our research agenda sets out to understand the dynamics of political protests in a ""partly free"" society: how do individuals mobilize to protest for democratization, what are their backgrounds, what is the role of beliefs and social interactions? We study this in the context of Hong Kong, a society with the largest divergence between civil rights (it has high legal standards and protection of fundamental rights) and political rights (citizens of Hong Kong have only very limited say in electing their leaders and determining public affairs).

Most studies of democratic movements occur ex post; such analyses, however, obscure the causes of individuals’ decisions to participate. Coordination has been achieved by the time the study takes place, and both beliefs and behavior will have largely converged on a consensus -- either widespread political participation in the case of a successful democratic movement, or extremely limited participation in the case of a democratic movement that has failed. Democracy movements do not begin with consensus, but ex post studies elicit attitudes, beliefs, and behavior after such consensus has already emerged.

In this project, we propose to overcome these limitations through four key innovations:

First, we study a democratic movement in real time, observing the population of Hong Kong as it expresses its demand for democratic institutions. The “Occupy Central” and “Umbrella Revolution” movements of 2014–15 marked the first moves in a strategic game that will likely last for the coming years and decades, and is reaching its temporary apogee in this year's (2019) continued protests. The constitutional promise to introduce universal suffrage in some near future, and the looming end of the ""one country, two systems"" solution in 2047 are the crucial points of tension. This unique setting allows us to collect panel survey data with a population selected ex ante, before the eventual convergence of beliefs.

Second, through the use of panel surveys, this research will innovate by documenting how beliefs, expectations, and preferences about the democratization process evolve over time, especially in reaction to political events.

Third, this research will use high-quality survey data from representative samples of the general population, as well as studying in more detail a population most involved in the political struggles, university students. We incentivize accurate reporting in our belief elicitation, employ sophisticated survey techniques to estimate the true support for potentially contentious issues, and put special emphasis on the elicitation of second-order beliefs (beliefs about other citizens' beliefs), and how these may drive actual behavior.

Our findings will be the first studies in the literature to document the factors driving a process of democratization in the making, and to provide evidence of causal mechanisms at work. We expect our research to speak to economists, political scientists, and historians; to shed light on both past and current instances of democratization; and, to contribute to the public debate about political change in East Asia."
"The ERC grant, starting in 2017, has allowed us to conduct surveys with a large number of students every year, both before and after the pro-democracy marches that take part regularly every 1st of July. Our research proposal contained two parts that we try to address with these studies.

First, we aimed to obtain a full and comprehensive picture of the social, economic, and ideological background of politically active students. In particular, our focus was on eliciting preferences (also providing cover in case students would not respond truthfully to open questions), beliefs, and second-order beliefs (beliefs about other students' beliefs and preferences). Another important dimension of interest was the role of classic psychological preferences (""big 5 traits"") vs. economic preferences (such as time discounting, risk aversion, reciprocity, inequality aversion) in determining political participation. The panel survey set up with the means of the ERC has allowed us to investigate these traits in every yearly round of interviews on a very broad number of students (~2000). This resulted in a first working paper (""The Fundamental Determinants of Anti-Authoritarianism""), which we are now constantly revising in light of the results that come in with every yearly survey wave. In particular, we are interested in observing how the relationships of interest (in particular, with respect to beliefs) change as the nature and size of the protests changes through time, from the small protests of 2016 and 2017 until now. In the context of these surveys, we have also elicited preferences for redistribution among Hong Kong students, which we plan to use for a comprehensive study of preferences for redistribution in East Asian populations.

Second, we aimed at better understanding the role of information, especially information about peers' beliefs and their potential participation in political actions, on one's own political activities. The ERC grant has allowed us to embed an experimental component in the regularly scheduled yearly surveys. So far, we have implemented two experiments. In the first experiment, we provide (a randomly selected subset of) students information about their peers' plans to attend the upcoming protest march. This allows us to understand whether one's own participation in a movement is a complement or a substitute to other people's participation. Our experimental results consistently point toward a ""game"" of substitutes, in which students reduce their participation when informed that the planned participation is higher than they expected (and vice versa). The results from this experiment has published in the paper ""Protests as Strategic Games"" (Quarterly Journal of Economics, 2019). In the second experiment, we provide students with an indirect monetary incentive (to take pictures of crowds), which indirectly makes them more likely to attend the protest march. We want to investigate whether this (marginal) incentive, by increasing attendance to the march, has the power to change students' beliefs and political attitudes. We do not find an effect on beliefs and attitudes, but, one year later, we notice a higher participation in the political march among the students who received the incentive one year prior. This points to the crucial role of social interactions (making new, politically engaged friends) in determining political activism.

Third, in cooperation with the Hong Kong Panel Study of Social Dynamics (HKPSSD), we are collecting evidence on beliefs and preferences (analogous to the ones collected for our panel sample of university students) for a representative sample of the Hong Kong population."
"Our research aims at advancing our understanding of political dynamics in societies with limited political participation. In our papers, we have tried to bring together economics, political science, and more broadly the psychological and sociological study of social movements. In our research of the role of preferences and beliefs among politically engaged students, we bring, for the first time, both psychological traits and economic preferences to the study of political activism. Our research is also novel in its consistent focus on the role of second order beliefs and how they determine political action, both descriptively and in an experimental setting. Our finding of ""substitutability"" in political activism (people reduce their own activity if told that other people are more politically active than expected) contrasts with a theoretical political economy literature that usually assumes complementarity, and poses new challenges to our understanding of the dynamics of political movements.

We expect our research to produce further insights into these topics, especially on the role of social dynamics and the varying effects of beliefs over the political cycle of protests -- which started of small in 2016 and 2017, and have currently reached a peak."