Electoral Participation in Iran’s Parliamentary Politics: Between Two Competing Explanations

Abstract

The paper uncovers determinants of turnout in Iran by studying the role of institutional and socio-economic variables in parliamentary politics since the 2000s. The paper argues that Iran’s electoral system has dichotomized the pattern of participation between center and periphery. The dynamic of participation in the center stems primarily from national shifts in the factional rule. However, in provincial peripheries, Iran’s electoral system promotes the personal particularistic demands of voters in the MP-citizen linkage. In this environment, the discretionary power of local state machinery over the daily lives of provincial citizens lays the ground for the role of local bureaus to influence participation. This argument draws on statistical analysis of parliamentary turnout and the study of several Iranian newspapers and official reports. The findings of the paper suggest a new mechanism by which institutional settings may shape the pattern of participation more generally.

Following the rise of competitive authoritarianism, scholars and pundits have actively debated the features of these regimes and the prospect of authoritarian change or durability. Considerable scholarly attention has been paid to the role of elections as a catalyst for change or as an agent of durability in these systems. (Gandhi and Lust-Okar 2009; Lindberg 2009) However, few have addressed the question of political participation in electoral authoritarianism. In other words, why do citizens bother to vote in electoral authoritarianism when their votes do not change the main body of the ruling elite? The conventional wisdom of scholars commonly stresses the role of clientelistic exchanges in mobilizing citizens in electoral authoritarianism. Existing studies generally highlight competitive clientelism (Lust 2009), illiteracy and patronage (Blaydes 2011)), the monopoly over patronage resources and constraining citizens’ choice (Magaloni 2006), elite cooptation (Gandhi and Przeworski 2006) and natural resource spending (Mahdavi, 2015) as the main determinants of electoral mobilization. In the broader context of transitional societies, scholars of comparative politics identify two general approaches to explain the determinants of voter mobilization. (Hagopian 2009) On one hand, some argue that the institutional setting, notably the electoral rule, determines the ways that politicians reach voters. On the other hand, the second approach highlights structural conditions such as the modernization stage as an underlying variable shaping the way that politicians mobilize voters. For example, indigent voters in the early stage of modernization respond favorably to instant material goods instead of the long-term issue-oriented promises of the politicians.

Despite a body of empirical evidence from transitional democracies and authoritarian regimes, these studies highlight one determinant of voter mobilization at the expense of dismissing other elements and consequently overlook the complexity of electoral participation in authoritarian regimes. In particular, they downgrade the significant impact of micro-level power hierarchy and subnational state-building in shaping the pattern of mobilization. The paper addresses this shortcoming by studying the role of institutional and socioeconomic variables in shaping the pattern of participation in Iran’s parliamentary politics since the 2000s. Drawing from a statistical analysis of the parliamentary turnout, the study of several official reports and a historical examination of the pattern of mobilization in Iran, the paper argues that in provincial areas, subnational state machinery is the main driving force in Iran’s parliamentary politics. In contrast, parliamentary politics in urban areas has been mainly affected by national shifts in factional politics which ensue from controlled politics and programmatic policymaking. In local districts, Iran’s institutional setting, notably the electoral rule, accommodates the personal particularistic demands of local voters in MP-citizen linkage. In exercising the enormous discretionary power over the daily lives of provincial citizens, local state machinery impacts Iran’s parliamentary politics. This setting, which has been developed over an extended period of time, mediates the impact of structural variables such as socioeconomic indicators in Iran’s parliamentary turnout and creates institutional channels broader than the electoral system to determine mobilization in electoral authoritarianism. These findings of the paper align with theories that highlight the role of state-building in shaping the pattern of mobilization (Shefter 1994).

The contributions of this paper to the study of electoral authoritarianism are threefold. First, the paper adds a new dimension to the role of institutional settings in shaping the pattern of participation and mobilization in electoral authoritarianism. Although several studies examined the process of institutional development in transitional societies, existing studies generally stress the impact of the electoral rule on legislators and voter incentives. This paper uncovers the significant role of local power hierarchy in shaping the preferences of Iranian citizens and political elites in parliamentary politics. Second, this study contributes to the field by bridging the literature on electoral institutions and patronage politics to address the origin of turnout in electoral authoritarianism; i.e., this study examines competing approaches in the study of participation in the particular context of Iran. The study concludes that the impacts of institutional and socioeconomic variables on the citizen-politician linkage and the pattern of participation are not mutually exclusive. Institutional arrangements may widen or narrow the scope of immaterial (such as personal) connections between citizens and politicians to accommodate particularistic socioeconomic concerns or ethnic and kinship ties. Finally, despite scholarly efforts to study citizen-politician linkage in advanced industrialized nations and transitional democracies (Kitschelt 2000), this linkage in electoral authoritarianism has remained relatively unexamined. The findings of this paper shed light on the relationship between citizen and politician in Iran and can be regarded as a stepping stone for the study of citizen-politician linkage in electoral authoritarianism. Instead of clientelistic exchanges, the paper highlights the broader role of personal particularistic demands in shaping citizen-politician linkage in provincial districts. This finding also partially challenges some recent studies (i.e., (Mahdavi 2015)) on the impact of public spending on the MPs’ re-election by showing that their suggested citizen-MP mechanism is not grounded on a strong empirical basis.

The paper proceeds as follows. First, the paper provides a brief overview of the determinants of Iran’s parliamentary politics. Next, it reviews the existing approaches to the study of turnout in developing countries and explains the important role that the electoral system may play in shaping the pattern of participation in the authoritarian setting of Iran. Lastly, the paper discusses the results of the statistical analysis of Iran’s parliamentary turnout and briefly accounts for the establishment of MP-citizen linkage by studying the impact of subnational state-building on the pattern of mobilization in Iran’s parliamentary politics.

Determinants of Iran’s Parliamentary Politics

One of the key central institutions in Iranian’spostrevolutionary politics is the parliament or Majles. Iran’s parliament is among the most effective legislative institutions in the region and Iranian citizens participate widely in parliamentary elections.¹ However, determinants of participation in Iran’s parliamentary politics have remained relatively unknown.² This section reviews the driving forces behind participation and accountability in Iran’s parliamentary politics.

Determinants of parliamentary vote choice in Iran are as diverse as Iranian society. In some districts, tribal affiliation and kinship ties drive parliamentary vote choice.³ Local conflicts, notably tension between neighboring towns or counties of a district, also influence citizens to participate in parliamentary elections.⁴ Cultural and religious elements draw the vote choice in some areas. Heads of local bureaus, intermediaries (such as teachers and influential families), and traditional local institutions (such as Friday prayer), actively influence parliamentary elections in some instances. In contrast to local and traditional determinants, factional politics⁵ and party endorsement are an important determinant of the parliamentary vote choice in urban districts, notably Tehran, Mashhad, Tabriz, Shiraz and Isfahan.⁶

According to MPs’ observations, minor particularistic demands⁷ are the most frequent type of demand by constituents.⁸ Clientelism, which is the exchange of private goods for political support, is a significant variable in Iran’s parliamentary electoral politics and is particularly influential in the re-election of provincial candidates. (Qasemi et al. 2011) MPs distribute cash handouts in deprived areas (Saeid 2010, p174) and vote buying is influential in some cases.⁹ Moreover, clientelism is influential far beyond parliamentary elections and has been pervasive throughout domestic politics in Iran.¹⁰ According to a survey of Iranian values conducted by Iran’s Ministry of Culture and Islamic Guidance, 87% of Iranians believe that illegal connections in state machinery are an “important” problem for the country, and 56.4% think these connections are a “very important” problem.¹¹ In Iran’s parliamentary politics, it is assumed that MPs will deliver private goods in their districts as a routine responsibility. In some provincial districts, club goods such as paving the roads, particularly in rural districts, are among the most frequents demands of the electorate. However, public goods (such as improving macroeconomic indicators and fighting corruption) are among the least frequent favors demanded by MPs¹²; i.e., the policy platform and the allocation of public goods in the districts play a minor role in shaping the parliamentary electoral outcome. Overall, personal particularistic demands has dominated the MP-citizen linkage in the local parliamentary politics of Iran. The prevalence of particularistic demands, however, is not as great across districts, since MPs in urban areas rely less on these demands comparing to those in provincial peripheries.

The disparity between center and periphery has created two fundamentally different patterns of participation in urban and provincial districts. To examine this disparity, this paper considers districts with more than 200,000 voters to be urban and labels the remaining districts as provincial. As Table 1 demonstrates, urban districts are ten times more populous than provincial districts, and the average provincial turnout is approximately 15% higher than that in urban districts. Similarly, Fig. 1 shows that turnout in urban areas is considerably lower than that in provincial districts.¹³ Parliamentary election turnout in Tehran and Ilam can be regarded as an example of this discrepancy. Ilam is one of the smallest provinces of Iran, where most of the population resides in small towns and the countryside.¹⁴ As Fig. 2 illustrates, the percentage of turnout in five parliamentary elections in Ilam is approximately two times higher than that in the Tehran province, which has the highest urban population of Iran.

Table 1 Average turnout and the voter population in 2000, 2004, and 2016 Parliamentary elections

Fig. 1

Turnout in urban and provincial districts; data is obtained from the portal of Iran’s Ministry of Interior

Fig. 2

Parliamentary election turnout in Tehran and Ilam; data is obtained from the portal of Iran’s Ministry of Interior

As explained, factional politics, i.e., party endorsement, is not a significant determinant of parliamentary vote choice in provincial districts such as Ilam. Instead, personal connections between MPs and citizens or intermediaries, endorsement by local magnates, ethnic and regional conflicts and cultural elements shape the vote choice of citizens in provincial districts. In contrast, in urban areas such as Tehran, factional politics primarily shapes the electoral outcome. This discrepancy is one of the key significant indicators of the pattern of participation across the nation. The paper uncovers the origin of the disparity in the pattern of participation between center and periphery by studying the institutional settings that have shaped citizen-politician linkage in Iran’s parliamentary politics.

The Role of Institutional Setting and Socioeconomic Status in Shaping Iran’s Parliamentary Politics

Early studies on voter mobilization highlight the social background of the voters as the underlying causal variable (Scott 1972; Epstein 1967). These studies argue that indigent and lower classes respond favorably to clientelistic exchanges, as opposed to the middle class, who prefer programmatic benefits.¹⁵ Thus, political parties who rely on the support of the middle class are more likely to be programmatically based and to invest in ideological appeals and public goods disbursement.

Drawing from these studies, a rational choice approach suggests that in a deprived province with high unemployment, indigent and uneducated citizens ignore the future and prefer instant clientelistic transaction, whereas in urban areas educated, wealthy and more informed voters consider the effect of clientelism in the long-run; i.e., the scarcity of public goods, and therefore prefer programmatic promises and party politics to particularistic and clientelistic exchanges. Therefore, pocketbook issues outweigh modern party politics and encourage most provincial voters to participate in parliamentary elections. Similarly, some attribute the higher turnout in localities to poverty and a lack of political knowledge and illiteracy. (Blaydes 2006) As Bahar and Cottam argued 60 years ago, the local magnates in provincial Iran are able to herd the illiterate and politically uninformed villagers to vote for their candidates (Abrahamian 1982, p. 121; Cottam 1964). One of the key variables determining the higher turnout in this rational choice-style approach is the level of education and knowledge. The following hypothesis evaluates the impact of literacy and education on the turnout.

Hypothesis 1

The less educated people are more likely to turnout in Iran’s parliamentary elections.

Another way of framing the role of structural variables is by highlighting the impact of poverty; i.e., the indigent and underclass tend to ignore the future and rely on instant clientelistic exchanges in elections. Considering the prevalence of clientelistic exchanges in Iran’s parliamentary politics, we expect that the underclass participates at a higher rate to benefit from clientelistic exchanges in the parliamentary elections. The following hypothesis evaluates this explanation.

Hypothesis 2

Indigent voters are more likely to turnout in Iran’s parliamentary elections.

Similarly, the modernization framework may argue that in less developed and less educated areas, premodern elements such as tribal affiliation, ethnic cleavages and kinship ties motivate voters to participate in electoral contests and lead to higher participation. To examine the impact of development on participation, the paper proposes the following hypothesis.

Hypothesis 3

The participation rate in underdeveloped districts is higher than that in developed areas.

The other approach explains the evolution of patronage politics by examining institutional configurations, notably party politics, the electoral system, bureaucratic organization and legislative institutions in developing countries. A review of the institutionalist view regarding the impact of electoral systems on patron-client linkages shows that the more clientelistic the system is, the more candidate-centered are the electoral systems. The causal link for this phenomenon has been discussed by several scholars. For one thing, in personal vote strategy, MPs cannot claim credit for national policy outcomes (Mayhew 2004; Arnold 1990). Nevertheless, they can earn credit for particularistic goods and for supplying patronage, pork and ombudsman-like services in their districts (Ferejohn 1974; Fiorina 1977). Additionally, personalized political competitions let the politicians bargain with small target groups of voters, facilitating clientelistic linkages (Katz 1980; Ames 2001). Similarly, everything being equal, majoritarian systems experience clientelism more than proportional representation (Buchanan and Tullock 1962, p. 263), and multimember majoritarian systems are even more prone to clientelism (Kitschelt 2000).¹⁶ In contrast, some studies associated clientelism with highly centralized party systems in the proportional representation structure, as these systems block the accountability of MPs to identifiable local constituencies (Coppedge 1994). In response to these institutionalist analyses of patronage in the citizen-politician linkage, some restore the earlier approach by developing a more advanced model of accountability, drawing clientelism from modernization elements and socioeconomic variables. They specifically assert that “overall - formal institutions are not particularly useful in accounting for the strategic dynamic of democratic accountability and responsiveness” (Kitschelt and Wilkinson 2007, p. 44). Their theory incorporates an analysis of the level of development and party competition along with patterns of ethnic heterogeneity to explain patron and client linkage.

Electoral Institutions and Parliamentary Turnout in Iran

The impact of the electoral system on patronage politics in Iran has been twofold. On one hand, Iran’s candidate-centered electoral system has led politicians to rely on personal ties with voters and highlight particularistic demands. As scholarly literature shows, personal demands are correlated with clientelism (Piattoni 2001, p. 17). Consequently, clientelistic exchanges in citizen-politician linkage have been routinized in Iran’s local parliamentary politics. Additionally, the low population an MP represents in a provincial district (compared to urban areas) furthers the personal connection between MPs and citizens in localities. On the other hand, the electoral system has attenuated the influence of political parties in Iran’s polity and therefore weakened programmatic policymaking, which is more party-induced and promoted clientelism and particularistic policymaking. As a former MP argues, “we have a parliamentary system, but we do not have parties. Consequently, the candidates must personally convince people to vote. When they want to convince individuals, they offer tangible promises, and these promises deal with the daily lives of people.”¹⁷ This setting intensifies clientelistic accountability as opposed to democratic accountability and programmatic policymaking in Iran’s local parliamentary politics. To unpack this twofold impact, the paper examines the development of Iran’s electoral law to illuminate how the electoral system created an obstinate pattern of electoral behavior in Iran’s parliamentary politics, underpinning clientelistic exchanges in provincial districts.

In the early constitutional era, Iran’s electoral law had specified the property and educational requirements for the electorate. However, these requirements were abolished in 1911 when suffrage was extended to all male Iranians. Paradoxically, the enfranchisement of the Iranian citizens has harmed the transition to democratic rule in Iran. That is, extending the right to vote to all adult males in the chaotic days of the early twentieth century implanted Iran’s patrimonial structure of power into the parliament. In the mostly rural Iranian society in the early twentieth century, the local magnates were able to mobilize the villagers to vote for their candidates. These conservative candidates commonly favored the status quo, thus weakening democratic reform in Iran (Abrahamian 1982, p. 121). Given the power of landowners in Iran’s provincial politics, they regularly gained the majority in Iran’s parliament.¹⁸ Although the 1979 revolution radically transformed Iran’s political hierarchy, the electoral system remained relatively unaffected. Iran’s current electoral law is based on a multimember majoritarian system in large districts and single-member in small areas in two rounds of elections.¹⁹ This electoral system has regenerated a similar pattern of voting behavior in Iran’s postrevolutionary parliamentary politics. In provincial districts, electoral rule accommodates the center of local power in parliamentary politics. Thus, local bureaucracies with substantial resources to influence the daily life of provincial citizens affect parliamentary elections.

In urban districts, however, Iran’s electoral system has created a different pattern of participation. Given the large number of voters that an MP represents, establishing personal connections in urban areas is exhausting. For one thing, the voters have difficulty in selecting and establishing personal connections with several candidates. For instance, the Tehrani voters should write the name of 30 candidates on the ballot paper, and the Guardian Council has blocked several attempts to computerize the voting system and slates. In addition to the difficulty in counting the votes in urban districts, this electoral system commonly creates confusion for the voters. After all, selecting 30 candidates from a list of several hundred qualified candidates in Tehran is time-consuming and frustrating. Consequently, the voters usually cannot fill out the entire ballot paper and select less than 30 candidates.²⁰

For another thing, a large number of eligible voters in urban districts compared to provincial districts impedes the establishment of a personal connection between MPs and constituents. As explained, Iran’s electoral system defines an urban district as one multimember district. The population of the district determines the number of MPs. For instance, the Tehran district has 30 MPs in the Iranian parliament. Each of these 30 MPs represents the entire district i.e. 6,800,000 voters. Therefore, instead of personal connections, voters rely on heuristics, such as the prominence of a candidate or factional endorsement.²¹

Considering the inefficiencies of candidate-centered electoral law in Iran, reformists in the 6th Majles attempted to reform electoral law to lessen the effect of the personal vote strategy and increase the role of parties in Iran’s politics. In March 2003, the 6th Majles passed legislation entitled, “Provincialization of Election” (Tarhe Ostani Shodan Entekhabat). According to this legislation, instead of selecting candidates for each district, the electorate was to select a list of candidates for their entire province. In fact, the bill sought to increase the size and population of each district an MP would represent to limit the personal connection between citizens and MPs in provincial districts. Thus, candidates would have to rely more on party endorsement and programmatic promises for all voters in the province. The Guardian Council vetoed this legislation. In the next terms of Majles, some MPs tried to pass similar legislation, but the Guardian Council vetoed other attempts to provincialize the electoral law.²² Therefore, Iran’s electoral system remained candidate-centered, promoting particularistic demands in provincial districts to serve the conservative forces in Iranian politics. This conservative resistance to democratic reform has undermined democratic accountability in Iran’s citizen-politician linkage, and despite drastic shifts in the patrimonial settings of Iranian society, clientelistic accountability has remained in place so far.

Data and Method

The study of Iran’s parliamentary politics is relatively rare and limited to a handful of analyses by policy-oriented communities. The conventional wisdom of analysts generally assumes that lopsided elections and controlled politics by the Guardian Council determine the parliamentary electoral outcome in Iran. That is, the disqualification of reformist and pro-democratic candidates by the Guardian Council discourages the Iranian public and paves the way for conservative victory in Iran’s politics. The vetting rate of candidates in parliamentary elections can be regarded as evidence for this argument. For instance, the vetting rate in the 2000 parliamentary election dropped to 11% of total candidates, the lowest level in the post-Khomeini era, and the reformists won the 6th parliamentary election in 2000 by a wide margin. The 6th Majles election in 2000 was one of the most competitive parliamentary elections in the post-Khomeini era. Leading reformist figures were allowed by the Guardian Council to participate and be nominated in the parliamentary election. On the other hand, in the 2004 parliamentary election, leading reformist figures had been barred from participation in the election and consequently, key reformist parties, namely, the IIPF (Islamic Iran Participation Front) and MIRO (Mojahedin of the Islamic Revolution Organization), boycotted the election,²³ paving the way for the conservative victory.

To consider the impact of controlled politics, this paper studies turnout in three different cases. Given the low disqualification rate in the 2000 parliamentary election (the 6th Majles) and the high disqualification rate in the 2004 election (the 7th Majles) and 2016 election (the 10th Majles), the comparative study of these three elections provides insight into the origin of participation in the presence and the absence of competition and controlled politics in parliamentary elections. The analysis of the 2016 parliamentary election provides further evidence to study participation in a relatively new factional setting in the postgreen movement environment and to test the continuity in the pattern of participation after a decade. For the sake of brevity, this paper sets aside the 2008 and 2012 elections and examines the existing explanations through a comparative study of the voter participation across the 2000, 2004 and 2016 parliamentary elections. The analysis of the 2008 and 2012 parliamentary elections (which are similar to 2004 election in terms of factional setting and the level of controlled politics), however, demonstrates a similar pattern of participation to Iran’s parliamentary elections in recent decade.

Iran’s real and competitive elections and effective institutional setting, as well as drastic socioeconomic change, offer an exceptional opportunity to examine the institutional versus structural approaches in the study of voter mobilization. Nevertheless, each authoritarian regime has its own unique features preventing students of comparative politics from developing general theories about electoral participation. Iran’s political system introduces a unique mixture of modern democratic institutions and religious hierarchy. Therefore, instead of developing highly general theories to discover the covering laws, the paper seeks middle-range theory building to “identify recurring conjunctions of mechanism” and pathways that create the causal outcome. These theories “provide more contingent and specific generalizations and allow researchers to contribute to more nuanced theories”. (George and Bennett 2005, p. 8).

The paper utilizes a mixed methodology. After a range of statistical analyses to scrutinize the determinants of parliamentary turnout in three election cases, the paper employs a historical examination of the pattern of participation to explain the underlying causal mechanism. The statistical analysis passes over the contextual and intervening variables. The historical examination identifies these variables and accommodates complex causal mechanisms such as path dependency. The statistical analysis relies on the district level analysis of parliamentary turnout to bypass the shortcomings of existing surveys in the authoritarian context of Iran. To address scholarly concerns (King 1997; Tam Cho and Gaines 2004) on ecological inference, the paper also controls the findings of the linear regression analysis by EI estimation based on King’s solution to the ecological inference problem. Additionally, the paper cross-checks the findings of this study with qualitative evidence drawing from several newspaper archives and government reports, and the strict assumption of EI estimation procedure is also investigated. (Tam Cho and Gaines 2004) The independent variables are obtained primarily from different censuses conducted by the Statistical Center of Iran. The main source for election results is Iran’s Ministry of Interior. The division of cities and counties conforms to the division of data in the Statistical Center of Iran and the corresponding participation rate is extracted from the Ministry of Interior data. The dependent variable in this study is the parliamentary turnout in three elections. There are three sets of independent variables in this study. First, the literacy and the level of education as well as the proxies for poverty test the impact of education and the lower class on participation as suggested by the structuralist view. To examine the impact of modernization, the paper studies the correlation between the level of development of the district and turnout. Additionally, ethnic heterogeneity investigates the impact of some premodern elements, i.e., ethnic tensions, on the turnout. Finally, the logged eligible voters as a proxy for district size examines the role of electoral institutions in participation. In addition, the paper controls the impact of the district configuration by adding some dummy variables such as the variable ‘multicounty’ to measure the regional tension between neighboring counties. The detailed description of the variables is available in the “Appendix”. For the sake of brevity, the impact of some controlling variables investigated in this study is not discussed in the paper.

Result and Discussion

The results of statistical analysis of the parliamentary turnout of the 2000, 2004 and 2016 elections at the district level are mixed across alternative indicators of demographics and socioeconomic status. The population of eligible voters in the district (the logged population in this model), as a proxy for district size, correlates with the percentage of electors who voted and has the highest impact on the turnout; one unit of increase in the log of eligible voters in the district (roughly 900,000 eligible voters in this model) corresponds with a 20% drop, approximately, in the turnout percentage in 2000–2004 and a 14% decline in the 2016 turnout.

As Hypotheses 1 and 2 demonstrate, the rational choice view argues that in a small and underprivileged province such as the Ilam indigent and uneducated citizens prefer instant clientelistic transactions, whereas in urban areas such as Tehran, educated and more informed voters consider the effect of clientelism in the long-run. Therefore, we expect that less educated individuals are more likely to turnout (Tables 2, 3, 4).

Table 2 OLS regression estimates of parliamentary turnout in Iran’s 6th Majles (2000 election)

Table 3 OLS regression estimates of parliamentary turnout in Iran’s 7th Majles (2004 election)

Table 4 OLS regression estimates of parliamentary turnout in Iran’s 10th Majles (2016 election)

The result of the statistical analysis of the three parliamentary elections does not support Hypothesis 1. Illiteracy negatively correlates with participation in the 2000 and 2004 elections and there is no statistically significant correlation between the turnout and illiteracy in the 2016 election. The impact of other proxy variables for knowledge and education also does not support the hypothesis. The percentage of college degree holders positively correlates with the turnout in most of the models. The positive impact of college degree holders is the highest in the 2016 parliamentary election. This finding challenge the study by Crock et al. (2016) on the impact of education on participation in authoritarian regimes. Raisi (2019) argues that this phenomenon results from the development of long-term political past experience in Iran’s electoral politics. Similarly, the percentage of employed individuals in the education sector (a proxy for teachers) with a relatively higher education positively correlates with the turnout and offers further evidence against the hypothesis.

Another way of framing the role of structural variables is to examine the role of poverty; i.e., the underclass tends to rely on instant clientelistic exchanges in elections and consequently is more likely to vote (Hypothesis 2). Similarly, the modernization view argues that less developed districts experience higher turnout (Hypothesis 3).

The result of the statistical analysis of parliamentary elections does not support the role of the lower class in higher participation (Hypothesis 2). Industrial workers, or the percentage of those employed in the industry sector as a proxy for some segments of Iran’s urban underclass, does not correlate with the participation in the 2004 and 2016 elections, and industrial workers even negatively impacted the turnout in the 2000 election. The result of the EI estimate in Table 5 confirms this finding. The likelihood of the participation of the industrial labor in the district is less than 30%, which is less than half of the overall participation rate in elections. The rural population, however, positively affects the turnout in all elections. Hypothesis 2 falls short of explaining this inconsistency in the participation of the lower class in rural and urban areas. One may argue that the level of development and premodern elements such as kinship ties and ethnic tensions explain the higher participation in rural areas. In fact, the impact of tribal affiliation and kinship ties is understandably hard to measure and test at the district level of analysis. Nevertheless, the level of development (Hypothesis 3) in a district may serve as a proxy for the influence of premodern elements. However, the result of the statistical analysis does not support the third hypothesis. The level of development does not correlate with the turnout in districts and the EI estimate in Table 5 does not demonstrate a considerable difference in the turnout of underdeveloped and highly developed areas. The ethnic diversity in the districts also does not impact the turnout since the variable EthnicHetro does not correlate with the level of participation in the district. The impact of denser social networks in rural areas (Lust 2009, p128) is also not a sufficient explanation for higher turnout in the particular case of Iran, as the market reform policies (which have been implemented in Iran to some degree in the post-Khomeini era) atomize citizens in the countryside (Kurtz 2004).

As mentioned earlier, another common explanation for the participation in Iran’s elections is the impact of controlled politics. Conventional wisdom commonly assumes that the Guardian Council determines parliamentary turnout and electoral outcome in Iran. Considering the substantial increase in the number of college degree holders in the electorate and other structural changes, one may argue that Iranians’ attitudes have shifted toward democratic values and the eligible voters normally turnout and vote for prodemocratic candidates in a fair and competitive election. Therefore, the disqualification of the reformists and prodemocratic candidates by the Guardian Council discourages the Iranian public from voting and spawns conservative victories in Iran’s electoral politics. Considering the low disqualification rate in the 2000 parliamentary election and the high disqualification rate in the 2004 and 2016 elections, this comparative study provides insight into the way that controlled politics influences parliamentary turnout. The result of the statistical analysis of the parliamentary turnout in Tables 2, 3 and 4 at the district level does not support the controlled politics explanation. Despite a noticeable decline in the participation rate of the urban areas in the 2004 election,²⁴ the turnout roughly follows the same pattern across the 2000, 2004 and 2016 parliamentary elections, challenging the controlled politics argument on parliamentary turnout. The logged voters as a proxy for district size and the electoral system is the most significant variable in all three models, and educated voters, as the main advocates of free and fair election, generally participated at a higher rate compared to less educated voters. Thus, the Guardian Council and controlled politics is not a significant variable in shaping the pattern of turnout and dissuading the general electorate from participation in parliamentary politics.

Overall, statistical analysis of turnout in Iran’s parliamentary elections does not provide empirical evidence for the role of socioeconomic variables and the controlled politics explanation. In fact, despite decades of development and drastic changes in Iran’s socioeconomic structure, Iranian parliamentary elections face a recurring disparity between electoral outcomes in urban areas and provincial peripheries. It appears that changes in modernization elements and socioeconomic indicators have not transformed the pattern of parliamentary voting behavior in provincial peripheries. Therefore, a significant question remains about the voters’ motivation in smaller districts. In short, what does motivate provincial voters to participate at a higher rate compared to urban citizens?

Table 5 Ecological inference result (dependent variable is parliamentary turnout)

Thus far, the statistical analysis shows a statistically significant correlation between the voter population and the turnout in Iran’s parliamentary elections. One may argue that the impact of the population does not differ between urban and provincial districts and consequently, the lower population in provincial districts does not create a different pattern of participation based on the personal particularistic demands. To investigate the difference in turnout between urban and provincial districts, the paper divides the pooled sample into two sets and runs separate regressions in urban and provincial districts. Afterward, the difference between regression coefficients is examined. According to the institutionalist explanation, Iran’s candidate-centered electoral rule creates two fundamentally different modes of electoral participation. The analysis of the regression coefficients in urban and provincial districts aligns with this explanation. The result of the Chi-squared test in Table 6 indicates a statistically significant difference between the coefficients of the voter population variable in urban (model A of Tables 8, 9 and 10 in “Appendix”) versus provincial (model B of Tables 8, 9, 10 in “Appendix”) districts. These findings imply that the population’s impact on turnout significantly differs between urban and provincial districts. The statistical analysis of the interaction between the dichotomous variable of urban/provincial and the rest of the variables in the pooled sample offers further evidence for these findings. The interaction analysis demonstrates a disparity in the pattern of participation between urban and provincial districts. As Table 7 demonstrates, the interaction between the dummy variable of urban areas and all other variables is statistically significant in the pooled regression model of electoral turnout.

Table 6 The result of coefficient difference test (suest) based on simultaneous regression of urban and provincial samples

Table 7 OLS regression estimates of parliamentary turnout including interaction with the dichotomous variable of urban/provincial

A survey of the MP-citizen linkage in Iran’s parliamentary politics sheds light on this disparity between urban and provincial districts. The analysis of the constituents’ demands by the Research Center of Majles indicates that minor particularistic demands or private goods shape the MP-citizen linkage in small districts. According to this analysis, most of the provincial voters request personal favors. Jobs, loans and cash handouts are the three most frequent demands of voters from MPs. The demand for club goods such as development projects in the district is far below these top three requests.²⁵ The research shows that approximately 60% of the constituents who visit the office of the MPs demand personal financial favors and jobs for themselves or their family members. On the other hand, less than 10% of the clients request development projects in the district. According to this assessment, 40% of MPs did not even work on a development proposal for their districts in the entire incumbency term.

Instead of the district size, one may attribute the MP-citizen linkage to the impact of single-member districts versus multimember districts (Mahdavi 2015) In fact, a significant body of scholarship argues that legislators from single-member districts invest more on the personal vote as opposed to the multimember districts’ MPs who rely on party endorsement and collective reputation. However, the statistical analysis of Iran’s parliamentary turnout does not support this argument. In addition, a multimember district with a low population can experience a considerably higher turnout percentage compared to populous districts. In less populated districts of the periphery, the personal connection between MPs and citizens (resulting from district size) mediates the particularistic economic concerns of the voters (as suggested by the rational choice framework), and traditional elements such as ethnic and kinship ties (as a proxy for variables in the modernization paradigm).

The analysis of the interaction between the dummy variable of urban/provincial and socioeconomic indicators confirms this assertion. The negative interaction between the dichotomous variable of urban/provincial districts and the variable of illiteracy, along with the variable of industry employed (as a proxy for the proportion of the lower class) in Table 7, indicates that the impact of these variables decreases from provincial to urban districts. Thus, economic concerns and modernization indicators have a higher impact in provincial districts. However, in urban districts, the lack of personal connection reduces the impact of socioeconomic variables on turnout.

For instance, in deprived provinces such as Ilam, the daily lives of citizens hinge on state resources more than they do in central urban areas. Therefore, the candidate-centered electoral system provides a channel for provincial citizens to demand their daily needs from MPs who have controlling power over the local bureaucracy. As a result, the disbursement of bureaucratic resources through welfare organizations enables politicians and bureaucrats to build clientelistic networks and later utilize these networks in parliamentary politics, similar to Ilam’s MPs who used to be the heads of the local IKRC²⁶ (a charity-styled revolutionary foundation) bureaus in the province.²⁷

MP-Citizen Linkage in Iran

Thus far, the paper examined the impact of socioeconomic variables and institutional setting on the demands of the constituents. The analysis of the supply side of the equation also explains the institutional setting that shapes the MP-citizen linkage and provides further support for the role of particularistic demands in Iran’s parliamentary politics. Some recent studies on the supply side of the electoral participation (e.g., Mahdavi (2015)) stress the role of natural resources and public spending on maintaining MP-citizen linkage in Iran parliamentary politics. Mahdavi argues that Iranian MPs in single-member oil-rich districts enjoy a clear incumbency advantage over MPs in multimember districts in both types of oil-rich and oil-poor districts. This is because MPs in single-member oil-rich districts have access to higher levels of discretionary spending that translates into higher public good provision for their constituencies. However, as previously explained, Iranian voters in small districts do not truly demand public good provisions from their representatives. Rather, they primarily seek private favors. The statistical analysis of constituent demand by the Research Center of Majles also confirms this explanation. Therefore, one of the key mechanisms identified by Mahdavi as being responsible for the “oil effect” does not have a strong empirical basis. It appears that Mahdavi overlooks the impact of other important determinants of parliamentary politics such as tribal affiliation and kinship ties in some oil-rich provinces such as Khuzestan.²⁸ These determinants are permanent and consequently, could affect the incumbency rate. Moreover, the power of MPs to influence national policies and budget allocation is limited. That is, their power to divert resources and national projects to their district is rare and limited to a handful of influential MPs, as some executive organizations—notably the former Management and Planning Organization²⁹-resisted political pressure for budget allocation (AhmadīAmūyī 2008, p. 77). A report by the Research Center of the Majles indicates that the rate of approval for MPs’ proposals to change the annual budget allocation is limited to the negligible number of 3%.³⁰ Thus, despite higher budget allocation for oil-rich provinces,³¹ MPs do not have control over spending this share of the natural resources in their districts and therefore, cannot claim credit for public spending. Instead, MPs exploit other resources namely local bureaucracy to address the particularistic demands of the constituents. In fact, a historical overview of local power development shows that the local bureaucracy significantly influences the daily lives of local citizens through the disbursement of private particularistic goods. Considering the impact of private particularistic goods in the MP-citizen linkage, one may argue that Iran’s subnational state machinery shapes the local parliamentary turnout.

In the particular case of Iran, local state-building fundamentally transformed the citizen-politician linkage in parliamentary elections. The underlying causal process for this phenomenon has gradually developed from the early constitutional period of Iran to the present. The electoral system and the extension of suffrage to adult males assisted the local magnates, namely the landlords, in controlling Iran’s parliamentary politics in the prerevolutionary era. This setting has been regenerated in the form of bureaucratic clientelism in the postrevolutionary parliamentary politics of Iran.³² The main drivers of this transformation were land reform and postrevolutionary state-building. As a result of these transformations, the daily lives of local citizens hinge upon the acquisition of bureaucratic resources, and consequently, the center of local power shifted from landlords to local bureaucrats. Farazmand’s study of state bureaucracy before the revolution shows by the late 1960s and as a result of land reform, Iran’s rural population was under the dominance of various state bureaucracies (Farazmand 1989). In fact, the people’s expectations of the “landlords were directed toward the local bureaucracy” (Zonis 1971, p. 31), and “villagers had to deal with different bureaucratic entities of the state that had replaced the feudal landlords” (Farazmand 1989, p. 141). The postrevolutionary power transformation led the citizens in provincial peripheries to rely on state resources even more than in the prerevolutionary era (Abrahamian 2008; Ehsani 2009; Lob 2013). Therefore, landowners’ power over parliamentary politics in the prerevolutionary period transformed into the bureaucratic clientelism in the postrevolutionary era.³³ In this setting, disbursement of bureaucratic resources plays a significant role in shaping parliamentary politics in localities.³⁴

Thepostrevolutionary refurbishment of bureaucratic positions and welfare state-building enabled revolutionary factions to fill positions and establish personal networks with intermediaries and citizens through resource allocation and service providing. These networks would later facilitate the mobilization of citizens in parliamentary politics. This setting has created a circular shift in local power hierarchy, since MPs influence the appointments of heads of local bureaus in their districts after winning elections. The lack of bureaucratic autonomy has eroded the boundary between politics and civil service and therefore has allowed state machinery to impact and also be influenced by parliamentary politics. The primary mechanism for exploiting bureaucratic resources is pressuring the ministers by exercising the power of oversight and impeachment over the executive branch.³⁵ The governors of the cities and counties are appointed by the province governor and the Ministry of Interior in Tehran, and the heads of local bureaus are appointed by the corresponding ministry in Tehran. This centralized system of administration helps MPs to influence local bureaucratic appointments by pressuring the ministers in Tehran. Thus, centralized state-building and legal setting guarantee the dominance of the MPs in supervision over local executive agencies, and this institutional setting facilitates MPs’ access to bureaucratic resources to mobilize citizens in local parliamentary elections.

Conclusion

The election’s role in authoritarian change or durability at the national level has crowded out the understanding of the election’s impact in the long-term, and the role that microlevel power hierarchy in localities may play. In the particular case of Iran, the pattern of participation has been influenced by the institutional setting, namely, the candidate-centered electoral system, local power expansion, legislative-executive arrangement and politicized bureaucracy. Iran’s electoral system promotes personal connections between MPs and citizens in localities and enables subnational state machinery with controlling power over resources to influence the pattern of mobilization and accountability in local parliamentary elections. Furthermore, subnational state machinery provides necessary resources and communication networks to influence local parliamentary politics and consequently determines the form that citizen-politician linkage takes in provincial peripheries. The underlying causal process for this phenomenon has gradually developed from the early constitutional period to the present. That is, the electoral system and the extension of suffrage to adult males assisted the local magnates, namely, the landlords, in controlling Iran’s parliamentary politics in the prerevolutionary era. This setting has been regenerated in the form of bureaucratic clientelism in the postrevolutionary local parliamentary politics of Iran.

The findings of this paper have important ramifications for the study of electoral authoritarianism. This study stresses the role of the existing institutional setting in the fate of electoral authoritarianism. That is, instead of deliberately designing the rule of the game (as has been highlighted by some studies on electoral authoritarianism), Iran’s conservative ruling elite primarily relies upon the existing institutional setting, which has gradually developed over an extended period of time. This institutional setting, which lays the groundwork for the rise of clientelism in localities, has created a rigid politics of subsystem. The politics of subsystem has been employed by the conservatives to resist attempts for reform, and negative feedback from this subsystem helps them to maintain the stability of the whole system. Thus, the Majles has proved inefficient in performing its democratic function, to the extent that national programmatic accountability has been overshadowed by the local particularistic demands of citizens in Iran’s parliamentary politics. The conservative ruling elites also have a vested interest in preserving this arrangement as it impedes national democratic change in Iran’s parliamentary agenda. Not only did the institutional setting lay the groundwork for the rise of clientelism in Iran’s citizen-politician linkage, but it also shaped the preferences of citizens and political elites in both localities and centers. This institutional setting created a fundamental disparity in the pattern of the voting behavior of center and periphery. In urban areas, the electoral system dissuades citizens from voting since parliamentary elections neither impact the daily lives of citizens in urban areas nor reflect the political and social demands of individuals in these areas. Thus, the parliamentary election turnout in provincial peripheries is considerably higher than that in urban areas in this institutional setting. This institutional setting also has influenced the preferences of political elites in Iran through undermining factional and party politics, as Iran’s electoral system led the candidates in localities to invest in personal votes instead of party endorsement. Consequently, the role of political parties in Iranian politics has been seriously undermined. In this setting, politicians rely on the networks that have been mostly shaped by the disbursement of bureaucratic resources to mobilize voters in provincial districts. Examining the evolution of these networks and their impact on the pattern of participation points to a rich agenda for future research.

Appendix

Data and Codebook

Data Availability

Replication materials can be found at: https://araisi.wordpress.com/2020/02/10/20/

Description of Variables Used at District-Level Analysis of Parliamentary Turnout

Dependent variable:

Turnout: The percentage of voters to total eligible voters in the district in the 6th Majles (2000) and 7th Majles (2004) parliamentary elections, Source: Iran’s Ministry of Interior.

Turnout 2016: The percentage of voters to total eligible voters in the district in the 10th Majles (2016) parliamentary election, Source: The head of election commission in the district or province.

Independent variables:

Development: The percentage of households with access to piped water to the total households in the district, Source: Statistical Center of Iran; 2006 and 2011 census.

Logvoters: Logarithm total eligible voters in the district, Source: Iran’s Ministry of Interior.

Illiteracy: The percentage of people who cannot read and write to total population over 6 years old, Source: Statistical Center of Iran; 2006 and 2011 census.

Totalemployed: The percentage of employed individuals to the total workforce in the district, Source: Statistical Center of Iran; 2006 and 2011 census.

industEmploy: The percentage of employed in the industrial sector to total employed individuals in the district, Source: Statistical Center of Iran; 2006 and 2011 census.

PublicEmploy: The percentage of employed in the public sector to total employed individuals in the district, Source: Statistical Center of Iran; 2006 and 2011 census.

EduEmploy: The percentage of employed in the educational service sector to total employed individuals in the district, Source: Statistical Center of Iran; 2006 and 2011 census.

CollegeDeg: The percentage of individuals with a college degree or above to total population above 21 years old in the district, Source: Statistical Center of Iran; 2006 and 2011 census.

RuralPop: The percentage of the rural population to total population of the district, Source: Statistical Center of Iran; 2006 and 2011 census.

EthnicHetro: Ethnic Heterogeneity (1 = where the ethnic majority is less than 80% of the total population of the district, 0 = other. This study does not include immigrant ethnicities notably in Tehran and Karaj in the coding.), Source: Obtained from University of Texas Libraries, Middle East and Asia Maps, and Vaezi Mansur, 2012, “Survey and Measurement of Public Culture Indexes”: [Tarhe Barressi va sanjeshe shakheshayeh farhange omumi keshvar] in Farsi, Ketab Nashr Press.

MultiCounties: (1 = The districts with more than one county where the eligible voters in the smaller county are at least 10% of the total eligible voters of the district = 1, 0 = other), Source: Iran’s Ministry of Interior.

See Tables 8, 9, and 10.

Table 8 OLS regression estimates of parliamentary turnout in Iran’s 6th Majles (2000 election) for urban and provincial samples

Table 9 OLS regression estimates of parliamentary turnout in Iran’s 7th Majles (2004 election) for urban and provincial samples

Table 10 OLS regression estimates of parliamentary turnout in Iran’s 10th Majles (2016 election) for urban and provincial samples