Class 14 - 1.11.22.pptx from the year 2023

econ

1. Econ 444 – Institutional economics Politics part 4

2. Overview Why do people decide to protest? What motivates? What does empirical evidence say? Do more people protest if they see others protesting or is it the other way round? We will cover 2 papers for this today (present different evidence) Taking to the Streets: Theory and Evidence on Protests under Authoritarianism by Kricheli Ruth, Livne Yair, Magaloni Beatriz Protests as Strategic Games: Experimental Evidence from Hong Kong's Antiauthoritarian Movement by Davide Cantoni , David Y Yang, Noam Yuchtman , Y Jane Zhang.

3. How do people decide to protest? I should protest because a lot of other people are protesting There is no need for me to protest as there are already a lot of people in protest

4. Taking to the Streets: Theory and Evidence on Protests under Authoritarianism by Kricheli Ruth, Livne Yair, Magaloni Beatriz

5. Taking to the streets Protests are a popular way to bring down authoritative regimes When democratic political institutions like voting are missing people express through protests Paper looks at the conditions that facilitate civil uprisings against autocratic regimes Protests involve a cost – fear of regimes reaction What are the conditions under which protests are successful in bringing down autocratic regimes

6. Theories of mass uprising Timur Kuran (1991a) political actions of one person encourage others to protest Participation of one person in protests reduces the participation cost for others Everyone will have their own threshold in terms of what is the minimum number of people they would like to see on streets to join them Kuran (1991a) shows that if a certain threshold in terms of protest size is crossed regime change will occur Kurans (1991a) model does not consider incomplete information Assumes that everyone knows how everyone else feels about the regime – not true

7. How the model works

8. Model Key mechanism is information based In authoritarian regimes people don’t know about distribution of preferences in society The protest in T1, helps to resolve this uncertainty of preferences I am not the only one who dislikes the regime. There are a lot of other people with similar preferences.

9. Model process

10. Hypothesis Model solved through game theoretic framework results in 2 testable hypothesis More repressive regimes are better able to deter citizens from taking part in protests In repressive regimes cost of protest (risks of being jailed, tortured, or killed) is high >> no protests Protests that take place in a more repressive regime are more likely to lead to regime change If people are willing to protest in a riskier environment, the information signal (in terms of anti govt sentiment) is much stronger >> more likelihood of a successful protest.

11. Implications testable through data More repressive regimes are more resilient against survival risks emanating from civil protest; and Second, that regimes which are more politically repressive are less stable, conditional on an episode of citizen protest actually occurring.

12. Testing empirically More repressive regimes are less likely to collapse Civil protests are more detrimental to more repressive regimes compared to less repressive regimes. Heterogenous effects of protests.  

13. Data Paper uses data on 5082 country-year observations from 183 countries from 1950 to 2000 to test out hypotheses. Civil Protest Protest is coded as 1 in every country-year observation wherein either a riot (a violent demonstration or clash of more than 100 citizens) or an anti-government demonstration (an opposition peaceful public gathering of at least 100 people) took place and as 0 otherwise. Political Repression Coded based on the Polity IV score. This variable equals 1 when the country's Polity score is between -10 and -3 >> AJR Paper Repressive FH, is coded as 1 when the country's Freedom House's political rights score is between 5 and 7.6 Regime Breakdown Based on regime change from one type to another other than a few exceptions

14. Results Results support hypothesis 1 This implies that, as suggested by the theory, more repressive regimes are in fact more stable than less repressive regimes. Results also support hypothesis 2 Civil protests are more dangerous for closed and repressive autocracies, as predicted by the authors theory.

15. Protests as Strategic Games: Experimental Evidence from Hong Kong's Antiauthoritarian Movement by Davide Cantoni , David Y Yang, Noam Yuchtman , Y Jane Zhang.

16. Protests as strategic games What drives individual decision to participate in a protest? Strategic substitutes : People are less willing to show up if they know that others are more likely to show up. Strategic Complementarity : Cost of participation is lower when people know that others are participating. People want to be part of a large successful protest (previous paper) Paper looks at the causal impact of giving information about participation behavior of others on individuals decision to participate

17. How do people decide to protest? I should protest because a lot of other people are protesting There is no need for me to protest as there are already a lot of people in protest Strategic Complements Strategic Substitutes

18. Context Hong Kong was a British colony until 1997 In 1997 was returned to China to be ruled as a special administrative unit until 2047 – left ambiguities Based on political issues annual July 1 marches were started Played an important role in Hong Kong citizens’ political engagement with the Chinese government Marches achieved goals particularly when no of people was large Each march focuses on a different issue/policy e.g. 2012 moral national curriculum

19. Context July 1, 2016, march studied in this paper was organized around important political aims: Denounce the perceived corruption of Beijing-backed Chief Executive C. Y. Leung; To mobilize support for democratic political parties in the run-up to the 2016 Legislative Council Elections Protest in authoritarian regime >> was tolerated within certain limits Protest had a turnout of 100,000 Authors show that based on data collected from subjects People thought that if protest was larger it would be more successful

20. Context Cost of protest: Time, tolerating hot weather etc Possibility of crackdown and being arrested Possibility of crackdown also higher with larger protest Clear mechanisms for substitutes and complements Large protest more likely to achieve goals but also more likely to face a crack down

21. What did the paper do Simple information treatment – tell people about percentage of people from survey who are likely to participate in protest See how: Treatment impacts beliefs Treatment impacts protest turnout (self-reported) Impact of beliefs on protest turnout (using treatment as IV for beliefs)

22. Experiment Part 1: Measure Report on own plan Beliefs on others plan (in survey) Beliefs on others actual Part 2: Treatment Remind treatment about their beliefs plus actual avg from survey Give everyone opportunity to report updated beliefs Part 3: Self Reported Participation Ask people to report on participation in protest

23. Part I

24. Part 2 Everyone Additional for Treatment Only

25. Opportunity to update beliefs - Everyone

26. Part 3 Measuring self report actual protest participation It is a bit problematic as people are unlikely to accurately self report participation

27. Impact of treatment on beliefs If prior was more than truth Updated belief goes down If prior was less than truth Updated belief goes up

28. Regression

29. Impact of treatment on protest turnout Prior below truth Treatment decreases participation Prior above truth Treatment increases participation >> Strategic Substitutes

30. Regression

31. How do people decide to protest? I should protest because a lot of other people are protesting There is no need for me to protest as there are already a lot of people in protest Strategic Complements Strategic Substitutes

32. Impact of beliefs on protest participation Want to look at impact of beliefs on participation Beliefs can be endogenous, so they instrument for beliefs using treatment We have seen that treatment has impact on beliefs Treatment is randomly assigned so this can serve as a valid instrument Using IV they can now look at impact of beliefs on participation

33. results As belief about other participation increases >> own participation decreases Consistent with story of strategic substitutes