Project: FREE policy brief

Motivated by the unusual patterns and dynamics of the Arab Spring, we construct a model explaining the vulnerability of the newly established incumbent to popular unrest. Using this model for the case of similar protests in Ukraine, we find that the current combination of availability of information, military capacity of the incumbent and his radicalization, together with the opportunity costs of participation in a protest, are likely to result in the formation of new government that is also vulnerable to popular protests. The persistence of the protests after the formation of a temporal government in Ukraine supports this hypothesis. Additionally, as the policy position of Viktor Yanukovych was relatively mild, his potential successor might be more radical. Exponential growth of social media users, reduction of military capacity, relatively high unemployment and the possible radicalization of the Ukrainian President might put the country into an “instability zone” with recurrent protests.

On the night of 21 November 2013 spontaneous protests erupted in Kiev, the capital of Ukraine, after the Ukrainian government suspended preparations for signing an Association Agreement and a Free Trade Agreement with the European Union in favor of agreements with Russia. The movement concentrated on Independence Square (Maidan Naseljenosti) soon took on the name “Euromaidan”. Soon the protest spread to other cities in the country. The initial agenda of closer relations with the EU was soon encompassed in the wider protest against Viktor Yanukovych, elected President in 2010. He fled the country on February 21 under the pressure of popular protests, exacerbating the leadership crisis. Temporary leadership was taken up by the Speaker of the Supreme Rada – Oleksander Turchinov, while new elections were scheduled to take place on May 25.

Despite the successful removal of Victor Yanukovich from power and a promise of new elections in May, the protests on Maidan did not cease. The major factor of uncertainty comes from the very nature of the protests. For many months, it ran without organizers or formal leadership so that the future course of action remains unclear. It is hard to comply with the demands of Maidan, as no clear set of demands are formulated. Though five figures of Maidan: Tymoshenko, Klitschko, Tyagnybok, Yatsenuk and Yarosh remain the most visible, none of them has sufficient support of Maidan. Whichever course prevails – resumed Eurointegration or an alliance with Russia (which became a less likely option)– the number of people who oppose the new course is likely to be enough to fill a new Maidan.

The swift happenings in Maidan are highly reminiscent of the events of the Arab Spring at its crux: it also was a leaderless protest, coordinated mainly with social media, and encompasses people of vastly different socio-economic, political and demographic characteristics.

Using social media technologies, Euromaidan has created an interactive map of logistics (http://maydanneeds.com/) that provides detailed information on and locations of where to eat, makeshift hospitals, information booths, and the barricades. Clicking on the icons of the map, one discovers not only the locations of the facilities but also their needs, which enables coordination of protesters’ efforts to contribute to the common cause. However, just as in case of Tahrir Square or the Tunisian unrest, the common cause is poorly defined: aside from dissatisfaction with Viktor Yanukovych, the protesters exhibited very different preferences for the future course of action, and the three most prominent figures of the protest – Klitschko, Tyagnybok and Yatsenuk – were shunned as they spoke about the common agenda.

The aftermath of the Arab Spring remains unclear for both protesters and the world. The Syrian social unrest has resulted in ongoing violent conflict, while Libyan society still experiences serious problems with the formation of a new government after the murder of Kaddafi and the end of civil war. Tunisia and Egypt were able to choose new Presidents and form new governments. The latter were themselves dismissed soon after they came to power: the first elected post-Mubarak government collapsed in mid-2013 after a year of almost uninterrupted protests. These two cases are especially interesting as constitutional exits of leaders who were in autocratic office for less than one year were generally caused by coups and not protests between 1945 and 2002 (Svolik, 2009).

Nevertheless, we can apply the knowledge acquired there to the new Ukrainian protest we observed on Maidan and try to predict its development by the means of stylized models suggested in Dagaev, Lamberova, Sobolev and Sonin (2013).

Our approach relies on four simple parameters that drive the dynamics of the protests. First, we consider the costs of collective action – the opportunity costs of spending time on Maidan. The second parameter is the military capacity of the incumbent that can be devoted to the suspension of the protest. The higher it is, the more numerous should the protest be to succeed. The third parameter we use is the degree of the radicalization of the incumbent (the difference between his position and the preferred policy of the majority of the population). Finally, we use an information availability parameter (how many people are aware of the place and time of the occurrence of the protest).

With the electric telegraph, a communication tool of the 19th century, information availability was low and many of those who would have been glad to pay the costs of collective actions to replace the incumbent stay at home as they are not aware of the protest taking place. With Facebook and Twitter, the availability of information is much higher. According to our findings, the crucial role in dynamics of contemporary mass actions is played by the ratio of military capacity to the information availability rather than their values per se.

Our framework assumes that each citizen’s decision of whether to participate in the protest or not is based on the difference between her position and the preferred policy of the incumbent. According to this decision, all citizens can be classified into two groups – those who participate in a protest against the incumbent, and those who do not. We define a person, who has the median position among the protesters, as the expected new incumbent. So if the elections were held among the protesters, he would receive the widest support. If the number of citizens participating in the protest is sufficient to overcome the military capacity of the current incumbent, the protest becomes successful, and the expected new incumbent of the protest becomes the new incumbent. The combination of military capacity, opportunity costs and costs of coordination determine the size of the stability zone – a segment of policy space where the incumbent is not vulnerable to mass protest.

The model allows us to predict the dynamics of the protests that is generated by different combinations of the parameters. For illustrative purposes, the availability of information about the protest is proxied by data on Facebook penetration and military capacity is described by the number of military personnel per capita in 2009 collected by the International Institute for Strategic Studies (Hackett, 2010). The incumbent policy position is proxied by the Legitimacy Index from Polity IV, where a higher index corresponds to lower legitimacy of the incumbent and his regime. Finally, the costs of participation in a protest are proxied by the employment rate. A recent study by Campante and Chor stresses unemployment as important determinants of opportunity costs of taking to the streets during Arab Spring (Campante & Chor, 2012). As unemployed individuals have fewer options of how to spend their time, one should expect that a substantial number of unemployed people corresponds to a relative ease of sparking unrest.

Using these parameters, we can explain success or failure of the protest, and predict some proprieties of its aftermath.

For example, high military capacity, opportunity costs and costs of coordination generate a broad stability zone, so that even a radical incumbent would not face a threat of revolution. The decline of any of three parameters can narrow the zone of stability and make the autocrat vulnerable to mass protest. As the incumbent is highly radical, a significant part of the population takes to the streets. As a result, the new incumbent’s position is sufficiently close to the one of the median voter and is, thus, inside the stability zone. An example of such a scenario is the overturn of Slobodan Milosevic after the fall of communism, when there were eight failed and one successful attempts to form a wide coalition of opposition parties (Spoerri, 2008). The process of finding a common ground started in 1990 with the emergence of the coalition of six parties, the Associated Opposition of Serbia, which broke shortly after a series of power struggles, policy disagreements, and personality clashes. It was only ten years later that the protest which facilitated Milosevic’s downfall took place, as the leader of the united opposition, the Democratic Opposition of Serbia, was able to ensure the non-involvement of the crucial military unit on behalf of Milosevic (Bujosevic & Radanovic, 2003).

In contrast, the events of the Arab Spring had different political dynamics. Low military capacity and high unemployment of Egypt and Tunisia determined a narrow stability zone ex ante. The absence of protests of these long-lived regimes can be explained by the relatively moderate position of the incumbent. In 2010, the Egyptian and Tunisian regimes had scores of 5 and 4 (out of 12) of Political Legitimacy from Polity IV, respectively. However the emergence of social media that enabled the users to coordinate their actions easily narrowed the relatively small stability zone. As the incumbent was less radical, fewer citizens benefit from its replacement and take to the streets. Thus, the new incumbent is defined by protesters, her policy position is more radical and now is out of the stability zone. The new incumbent immediately faces the new social unrest.

Table 1 presents the stylized results of our study. Locating the combination of parameters of the country in the table allows us to make the prediction about the dynamics of the protest. There are several possible courses of events: the protest can be weak and die out soon, with the incumbent staying in place; it can be significant, but yet not large enough to overthrow the incumbent; it can lead to the replacement of the incumbent, followed by the period of stability; and, finally, it can result in the replacement of the incumbent, but not cessation of the protest.

Table 1. Protest Outcome as a Function of Parameters

What do our findings tell us about Ukraine? The previously used proxy for the information index there is not a good choice, as the majority of users prefer the Russian version of Facebook – Vkontakte – as the major social network of the country. Thus, we rely on the Vkontakte penetration data (as of 2013, 18.5 million of people in Ukraine were using the network, constituting 40.6% of the population, see report of Ukranian IT-news agency AIN.UA: http://ain.ua/2013/11/28/503853).

The military parameter, that reduces the likelihood of successful protest, is low, compared to the countries of Arab Spring and constitutes 2.8 active military per 1000 people (see the Ukrainian law “Armed Forces of Ukraine for 2013”, http://w1.c1.rada.gov.ua/pls/zweb2/webproc34?id=&pf3511=49126&pf35401=283834), which is half of the one in Egypt at the beginning of the protests. The unemployment parameter fell to 8.6 during the incumbency of Victor Yanukovych, which corresponds to the pre-protest unemployment in Egypt in 2010.

The legitimacy measure presents a difficulty for comparison with the Arab Spring cases, as the Legitimacy Index has not yet been updated. However, the harsh actions of Victor Yanukovych during the 2013-2014 protests (including the suppression of the protest and the passage of laws denying freedom of assembly and freedom of expression, as well as the refusal to repeal his earlier changes of the constitution towards more presidential form of government) suggest that his legitimacy level fell by approximately 50% (and was about 20% right before he was ousted from power), which is corroborated by polls (“Ukraine’s future in peril under President Yanukovych”, The Washington Post, 2 December, 2013). Thus, the conservative estimate is that his legitimacy index shifted from 3 to 4 or 5, as shown on Figure 1.

For the purpose of comparison, we plot the Arab Spring countries and Ukraine in the space of our variables in Figure 1. The X-axis shows the incumbent’s departure from the median population-preferred policy (proxied by the Legitimacy Index from Polity IV). We use the value of the index in the year prior to the start of unrest, and the index value in 2010 for countries with no protests (Morocco, Oman, Djibouti). The Y-axis corresponds to the employment level. The size of the bubble corresponds to the ratio of military capacity and Facebook (or Vkontakte) penetration (data from the Arab Spring Social Media Report).

The shading of the bubble reflects the type that country belongs to: striped (no significant protest), light gray (continuing protest), and dark grey (multiple protests). Syria is excluded from the classification and is marked white, because of the civil war and international intervention.

Figure 1. Legitimacy index (X-axis), Employment (Y-axis), Military capacity / Social media penetration (size of the bubble) in Arab Spring countries and Ukraine

Figure 1 illustrates that countries appear in tight clusters in line with our theoretical predictions. The countries with continuing protests that did not lead to the downfall of the incumbent are divided into two groups. The first group (Kuwait, Jordan, Lebanon, and Bahrain) has relatively a moderate incumbent policy position and extremely high level of development of new media. The reasons why these countries are not «striped» is high military capacity of the government that could be employed against protesters, combined with high opportunity costs of protesting (low unemployment rates).

The second group of countries (Syria, Algeria, Iraq, Yemen, and Mauritania) has more radical incumbents and higher unemployment rates (so the incentives to protests are higher there), but is poor in terms of IT development. The ratio of military capacity and Facebook (Vkontakte) penetration is high, which is reflected by the size of bubbles that are much larger than in the first groups. That is why in a country with a small military capacity (such as Yemen), the protests did not lead to the incumbent’s replacement.

Two Arab Spring countries belong to the “multiple protests” group. Both Egypt and Tunisia had relatively mild incumbents in the pre-protest era, with Tunisia’s Bashar al-Assad being the milder of the two. Both countries had relatively high unemployment rates and wide Facebook coverage, both factors alleviating the problem of organizing a collective action. Despite the fact that before the start of the protests Facebook coverage in Egypt had close to average values among the countries of Arab Spring, they grew at exponential rates and Egypt attained leading positions in the region in usage of new media several months later. Moreover, low rates of military capacity made protest activity less risky in both Tunisia and Egypt. The remaining differences in Facebook coverage and employment rates in Egypt and Tunisia account for the different structure of recurrent protests, predicted by our model.

Comparison of the Arab Spring countries data with the Ukrainian case shows that high military capacity, new media penetration and unemployment generate even more narrow stability zones than one observes in the cases of Egypt and Tunisia. The reason why the incumbent was not vulnerable to mass protests can be explained by the political legitimacy of Yanukovych as the winner of relatively free elections in 2010.

But if the Arab Spring protests were triggered by rapid growth of new (cheap) communication technologies, the successful protest against Yanukovych can be explained by his radicalization. The radicalization took the form of parliamentary acts that put significant constraints on political rights and civil liberties and violent suppression of dissident actions.

Employing the proposed approach and contrasting the Ukrainian case with the countries of the Arab Spring allows us to draw several conclusions.

Firstly, the current combination of availability of information, military capacity of the incumbent and his radicalization, together with the opportunity costs of staying on Maidan, are likely to result in successful and recurrent protest. The persistence of the protests after the formation of a temporal government supports this hypothesis.
Secondly, it is worthwhile to note that as the policy position of Viktor Yanukovych was relatively mild, his potential successor might be more radical.

Thirdly, the exponential growth of social media, the reduction of military capacity and relatively high unemployment puts Ukraine into an “instability zone”. This implies that the 2004 scenario of the Orange Revolution is unlikely to repeat. The protest of 2004 resulted in a general election, and the elected president Viktor Yushchenko served his term without interruption. The protests of 2013 are more likely to result in a rapid change of incumbents and a period of instability.

One factor can strengthen the possible incumbent’s vulnerability. The external pressure of the Russian government reduces costs of collective resistance to the new Ukrainian authorities among pro-Russian citizens, while the promise of Western countries to support fast EU integration can incentivize politicians to accelerate reforms opposed by significant parts of the population.

References

• Bujosevic, D., & Radanovic, I. (2003). The fall of Milosevic: the October 5th revolution. Palgrave Macmillan.
• Campante, F. R., & Chor, D. (2012). Why was the Arab world poised for revolution? Schooling, economic opportunities, and the Arab Spring. The Journal of Economic Perspectives, 167–187.
• Dagaev, D., Lamberova, N., Sobolev, A., & Sonin, K. (2013). Technological Foundations of Political Instability. Centre for Economic Policy Research Working Paper Series
• Hackett, J. (2010). The Military Balance 2010: The Annual Assessment of Global Military Capabilities and Defense E conomics. London: The International Institute for Strategic Studies.
• Spoerri, M. (2008). Uniting the opposition in the run – up to electoral revolution – Lessons from Serbia 1990 – 2000. Totalitarismus Und Demokratie, 5(2005), 67–85.
• Svolik, M. (2009). Power sharing and leadership dynamics in authoritarian regimes. American Journal of Political Science, 53(2), 477–494