The Islamist voter base during the Arab Spring: More ideology than protest?

Eva Wegner, University College Dublin

*This memo was originally drafted as a part of the Adaptation Strategies of Islamist Parties and Movements  workshop held at George Washington University January 27, 2017. POMEPS Studies 26 is a collection of the memos from this workshop, available as an open-access PDF here.

From the perspective of an Islamist party supporter in the Arab world, much has changed over the past ten years. In the mid-2000s there was barely any Islamist party with a formalized influence on policy in the region, the only exceptions being Hamas in the Gaza Strip, short-lived stints of Jordanian Muslim Brothers as government ministers and a handful of PJD-led local governments in Morocco since 2003. Voting for an Islamist party was essentially casting a ballot for an opposition movement, untainted by government involvement, without much influence on policies but significant­­­ credentials in grassroots activities and charity work.

Islamists have led governments in several countries, such as Egypt, Tunisia, and Morocco but the fate of these governments has been vastly different. In Egypt, the Freedom and Justice party was ousted by a military coup in 2013 after just a year in power and has been violently suppressed since; in Tunisia, the Ennahda party led the first post-transition government but had to make substantial compromises in terms of power and agenda to make the transition work; in Morocco, the Justice and Development Party (PJD) has been “leading” the government since 2011 in what is still an authoritarian monarchy and was able to increase its vote share in both parliamentary and local elections held since; in Algeria and Jordan, it has been business as usual for Islamists, receiving small vote shares in inconsequential elections. In addition – and strongly related to post-Arab Spring developments – many Islamist movements and organizations have adapted their discourses and practices, while new, more radical groups inspired by some version of political Islam have sprung up to challenge mainstream organizations from the fringes.

In the context of such manifold changes, this short memo discusses voter profiles for Islamist parties in the Arab Spring environment. The main questions I seek to explore are whether Islamist parties represent distinctive sets of voters in the Arab World that display ideological attachment with party values or whether the success of Islamist parties in Arab Spring elections was mainly the outcome of clientelism or protest voting.

What do Islamist voters care about? What do we know?

Understanding drivers of support for Islamist parties has motivated several studies over the past decade. At least implicitly, much work is framed in terms of whether support for Islamist parties follows a clientelistic logic or whether these parties represent certain beliefs and values. This is not simply an academic question. If Islamist voters are motivated by clientelistic inducements, Islamist parties could pursue whatever policies they like with a certain level of detachment from and be unaccountable to their electorate. If voter values mirror Islamist ideological positions, these parties may represent relevant socio-political preferences, implying Islamist ideology is appealing beyond the elite/ activist level and ought to be taken seriously.

The clientelistic narrative originates in observations of the general importance of clientelistic inducements in Arab elections, i.e. the promise and handing out of particularistic benefits to voters such as gifts, jobs, money, or other favors (Blaydes 2010; Corstange 2016; Lust 2009). Islamists, with their myriad welfare/ charity organizations that often deliver jobs, emergency relief, medical care, help with education, and so on (see Clark 2003) should be particularly well positioned to deliver such goods and capitalize on this in elections.[1] As a high level of social desirability bias makes clientelistic voting behavior notoriously difficult to assess through surveys, clientelistic voting for Islamist parties has only been assessed indirectly, that is, via studies of voter demographics in actual elections (e.g Elsayyad and Hanafy, 2014; Gana, Hamme, and Rebah 2012; Pellicer and Wegner 2014). The assumption underlying these studies is that a voter base of mostly poor voters would be an indication of clientelistic party-voter linkages.[2] These studies have not found evidence of a clear-cut clientelistic profile of the Islamist electorate.

In the representation model, hypotheses are being developed about what Islamists are standing for, e.g. anti-democratic values, anti-globalization/ authenticity, piety, regime opposition, and so forth and assessed via public opinion surveys such as the World Value Survey or the Arab Barometer (e.g. Garcia-Rivero and Kotze 2007; Gidengil and Karakoç 2016; Robbins 2010). Most of these studies use different methods, data and analytical frameworks and focus on different sets of countries at different moments in time. Findings on the features of the Islamist electorate have thus not been robust and remained inconclusive.

In addition, many scholars argue that a protest voter rationale drives Islamist electoral support, at least in “founding elections” (e.g. Mecham and Chernov Hwang 2014). This argument posits that, after decades of repression and limited political inclusion by Arab regimes, Islamist parties were well positioned to attract voters dissatisfied with the status quo and looking for credible alternatives. While intuitively compelling, this argument has not, to best of my knowledge, been investigated thoroughly. An important implication of that argument would certainly be that whatever support Islamists might have attracted during elections in the Arab Spring context is shaky and not likely to stay unless Islamists were either able to address some of the citizens’ core grievances or somehow manage to maintain a credible anti-elite discourse.

Islamist ideology resonates with Islamist voters

My current research on voting in the MENA (joint with Francesco Cavatorta) combines three surveys (the Word Values Survey, the Arab Barometer and the Afrobarometer) all implemented between 2011-2013 to examine values of Islamist and secular/left supporters in seven Arab countries (Morocco, Algeria, Tunisia, Egypt, Libya, Yemen, and Palestine).[3] The results can speak to a rather simple question – that interestingly has not been addressed in studies of Islamist voter motives – namely, whether Islamist voters share some of the core values propagated by Islamist parties.

Values and policies promoted by Islamists may of course differ across (and within) countries – it is particularly difficult to pinpoint if there are Islamist-specific economic policies (probably not).[4] However, there are two sets of values that are core to any Islamist group. The first, and most important, is a preference for a greater role of religion in political and social life. The second is social conservatism, in particular when it comes to gender roles.[5] Essentially, if voter values and party ideology were aligned on these matters, some form of programmatic party identification is likely even if it might co-exist with other support motives. In turn, if voter values and party ideology do not match, the rationale for casting a ballot for that party is unrelated to core ideology, for example it could follow a clientelistic logic or signal more trust in that party compared to others.

Figure 1 displays the values of respondents indicating that they would vote for an Islamist party “if there were elections tomorrow” compared to any other respondent in the surveys.[6] As can be seen clearly in the figure, Islamist voters do indeed display values in tune with Islamist core messages, most strongly when it comes to the greater role of religion in politics. In contrast, there are no discernible differences between Islamist voters and others regarding economic issues such as preferences for redistribution or economic competition.[7] Figure 2 shows a few demographic features of Islamist voters during the Arab Spring period. To some extent, Islamist parties attracted a slightly better off group compared to the average citizen, with less poverty and more education. At the same time, Islamist voters in the seven countries were, on average, not more likely to have fulltime employment. They were slightly younger but had no gender defined profile. All in all, this does not look like to typical profile of clientelistic voters nor like a pronounced economic grievance profile – at least no more than the average MENA citizen at that time.

In addition, these data contain some items that speak to the protest voter argument, at least partly. It is a sensible assumption that protest voters would be those especially concerned with corruption, given its pervasiveness and the involvement of the old political elites in it. A protest vote can be viewed partly a vote against the corrupted status quo. Figure 3 shows the results of different survey items that address these questions. Contrary to expectations, Islamist voters are not more likely than others to mention corruption as their most important problem. In fact, on all other items relating to corruption, they are less likely than other respondents to believe that corruption is pervasive, that it has increased in recent years or that the government handles it badly. As these items come from three surveys, the general finding that Islamist supporters do not match the protest voter profile when it comes to corruption appears quite robust.


The implication of these findings is relatively straightforward. To some extent an important motive for Islamist support is a very simple one, namely that Islamist ideology resonates with voters. This is particularly interesting if one considers that these surveys were taken during the Arab Spring at a moment when Islamist parties attracted particularly high voter shares. Moreover, the share of respondents indicating a preference for Islamist parties in the surveys exceeds the share these parties won in the Arab Spring (or other nearby elections), showing that these are not simply the values of a number of very committed sympathizers. The finding seems intriguing given that scholars (including myself) often explain Islamist support in terms of things other than Islamist ideology, such as their stance against corruption, the fact that they have not been tested in office or their grassroots activities. These factors might of course matter but should not lead us to overlook the attractiveness of the “Islamist part” of Islamist parties for a substantial group.[8] In this regard, it is worth mentioning that the ideological congruence result is fully robust across survey, country, and also holds for earlier surveys by the Arab Barometer and World Values Survey.

How salient the “Islamist” part of Islamist parties is for voters is another question. Economic problems such as unemployment, poverty, or general management of the economy consistently feature as “the most important problem of the country” for three quarters of respondents in surveys. Islamist voters are no different from the average citizen in this regard and could possibly be shifting if parties with clear economic visions emerged.


To date, it is not clear how Islamist voter profiles will evolve, and one could imagine vastly different paths that mirror national experiences. For example, when Islamist parties participate in government, more spoils can be distributed which could attract new voter groups. Or, more optimistically, voters could be satisfied with the performance of Islamist national and local governments and shift their support to them.[9] The outcome in both cases would be a further broadening of the voter group with the party attracting people that are substantially different from the original group. Alternatively, Islamist parties could have adapted their positions on certain aspects, which could either lead to an adaptation of voter positions (if voters follow the party’s lead) or to a changing voter base. In yet another scenario, being held at least partial responsibility for economic problems could reduce supporters to a core group of staunch ideologues.

The latest electoral results in Morocco show that the latter scenario has not (yet?) materialized there; to the contrary, support has been increasing strongly. [10] In Tunisia, it is too soon to tell but if surveys are any indication, support hoovers around 20 to 25 percent, something similar to the latest election. In Egypt, it is hard to say anything about who would today be supporting a party emerging from the Muslim Brotherhood – an unlikely event for years to come at any rate.

Eva Wegner is an assistant professor at the School of Politics and International Relations (SPIRe) at University College Dublin.

[1] The most well-positioned actors for clientelistic electoral strategies are probably regime parties or well-connected individuals.

[2] That poorer voters are more prone to respond to clientelistic offers is shown theoretically and empirically by a considerable body of literature (see Stokes et al. 2013 among many others).

[3] Not all countries are included in all three surveys but we have at least two surveys – around 3,000 observations – per country. Combining these three surveys enables us to check the robustness of findings across surveys.

[4] Several scholars argue that Islamists have generally liberal economic agendas with a strong private charity component, akin to “compassionate conservatism.” Others highlight rather distributive policies. Above all, this underscores the fact that there is no distinctive “Islamic” economic model (Gerges 2012).

[5] This is perhaps more controversial. In several countries, Islamist parties were found to have more female candidates for office than other parties before such countries introduced gender quotas in parliaments. At the same time, there is a) evidence that these women were put on electoral lists for image reasons, such as in the PJD’s case in the 2000s, (see Wegner 2011) and b) evidence that a lot of Islamist parties/ movements oppose pro gender equality political reforms regarding divorce or inheritance and favor a “special role for women”.

[6] Please see Wegner and Cavatorta 2016 for information on measurement and coding.

[7] Notice that these results are robust to using other empirical strategies. Most importantly, results are robust to using OLS regression of each individual value index on an Islamist voter dummy (plus controls).

[8] How this ideology translates into policies and what level of radicalism of such policies would resonate with Islamist voters is another, mainly empirical, question that cannot be answered at this point given that no Arab Islamist party has remained in office long enough (or with enough power) to implement much of an Islamist agenda (and in a degree of variation that would allow exploring which type of policy voters endorse, on average).

[9] The Afro Barometer survey gives some suggestive evidence that in Morocco (and to a lesser extent Tunisia) stated support for Islamists goes together with positive evaluations of Islamist government participation. In these countries, Islamist supporters were more likely to state that corruption has been decreasing since the Arab Spring. Similarly, Islamist supporters in Morocco and Tunisia are more likely to state that the government does a good job handling “my most important” problem. This was not the case in Algeria, were Islamists had remained in the same – powerless – situation as before the Arab Spring. Relatedly, in research conducted joint with Miquel Pellicer on PJD-led local governments in Morocco we find that support in parliamentary elections increased dramatically in those places where the PJD led the local government and had a small governing coalition (see Pellicer and Wegner 2015).

[10] The PJD doubled the number of votes it received between 2007 and 2011 and again by another 50% in 2016. The PJD is the largest party in parliament but has only about 30% of seats and votes. Since the election last year, the party has been unable to form a government – reminiscent of what happened to the USFP in the 2002 election where, ultimately, the King “stepped in” and appointed Driss Jettou as Prime Minister.


Blaydes, L.., 2010. Elections and distributive politics in Mubarak’s Egypt. Cambridge University Press.

Clark, J. A., 2003. Faith, Networks, and Charity: Islamic Social Welfare Activism and the Middle Class in Egypt, Yemen, and Jordan.

Mecham, Q. and Chernov Hwang, J, 2014. “Conclusion. The New Dynamism of Islamist Parties, in Mecham, Q. and Hwang, J.C. eds., Islamist parties and political normalization in the Muslim world. University of Pennsylvania Press.

Corstange, Daniel. 2016. The Price of a Vote in the Middle East, Cambridge, CUP Press.

Elsayyad, M., and S. Hanafy. 2014. “Voting Islamist or Voting Secular? An Empirical Analysis of Voting Outcomes in Egypt’s ‘Arab Spring.’” Public Choice 160(1–2): 109–30.

Gana, A., G. Van Hamme, and M. Ben Rebah. 2012. “Géographie Électorale et Disparités Socio-Territoriales: Les Enseignements Des Élections Pour L’assemblée Constituante En Tunisie.” L’Espace Politique. Revue en ligne de géographie politique et de géopolitique (18).

Garcia-Rivero, C., and H. Kotze. 2007. “Electoral Support for Islamic Parties in the Middle East and North Africa.” Party Politics 13(5): 611–36.

Gerges, F. A. 2012. “The New Capitalists: Islamists’ Political Economy.” (June 2, 2015).

Gidengil, E., and E. Karakoç. 2016 “Which matters more in the electoral success of Islamist (successor) parties–religion or performance? The Turkish case.” Party Politics 2(3): 325-338.

Lust, E. 2009. “Competitive Clientelism in the Middle East.” Journal of Democracy 20(3): 122–35.

Pellicer, M., and E. Wegner. 2014. “Socio-Economic Voter Profile and Motives for Islamist Support in Morocco.” Party Politics 20(1): 116–33.

Pellicer, M. and E. Wegner. 2015. The justice and development party in Moroccan local politics. The Middle East Journal69(1), pp.32-50.

Robbins, M. D. 2010. “What Accounts for the Success of Islamist Parties in the Arab World?” Dubai School of Government Working Paper (10–01).

Stokes, S., T. Dunning, M. Nazareno, and V. Brusco. 2013. Brokers, Voters, and Clientelism: The Puzzle of Distributive Politics. Cambridge University Press.

Wegner, E. and F. Cavatorta. 2016. Revisiting the Islamist-Secular Divide: Parties and Voters in the Arab World, R&R. Manuscript available from the authors.

Wegner, E., 2011. Islamist opposition in authoritarian regimes: The Party of Justice and Development in Morocco. Syracuse University Press.