In democratic competition, political parties link political elites to citizens. While the nature of the linkages can vary, a prominent argument in the literature is that programmatic linkages, or policy-based relationships between voters and parties, are a function of income (Kitschelt Reference Kitschelt2000; Kitschelt and Wilkinson Reference Kitschelt, Wilkinson, Kitschelt and Wilkinson2007; Kitschelt and Kselman Reference Kitschelt and Kselman2013). This theory is epitomized by Kitschelt and Freeze's claim that “poverty goes with a predominance of clientelistic accountability strategies in competitive politics and high affluence goes with an emphasis on programmatic party competition” (Reference Kitschelt and Freeze2010, 30).
This article addresses three issues. First, despite a growing body of literature dedicated to clientelism and brokers in non-Western democracies (e.g., Stokes et al. Reference Stokes, Dunning, Nazareno and Brusco2013; Wantchekon Reference Wantchekon2003; Szwarcberg Reference Szwarcberg2015), the extent to which their parties are connected to voters through programmatic links or their clientelistic alternatives is still unclear. Furthermore, comparative analyses are lacking in the context of East Asian democracies.Footnote 1 Second, the economic explanation does not offer a firm conclusion as to whether the effect of economic development holds only cross-nationally or within countries as well, and whether individual economic conditions determine these linkage mechanisms. The third concerns the lack of consideration of the motivations of party activists. Those understanding these linkages as parties’ strategic decisions about how to distribute goods tend to implicitly assume that party activists are brokers in non-programmatic distributive politics, or unnecessary players for programmatic distribution.Footnote 2 Accordingly, distributive politics is often treated as a game taking place mostly between party elites and voters.Footnote 3
However, this assumption about parties as unitary actors is hardly warranted when the empirical evidence is considered. From Michels’ (Reference Michels1962 [1911]) pioneering report on the intra-party politics of the Social Democratic Party of Germany and Panebianco's (Reference Panebianco1988) influential volume on party organizations in Europe, to recent survey-based studies on party members and activists (e.g., Whiteley and Seyd Reference Whiteley and Seyd2002; Pederson et al. Reference Pederson, Bille, Buch, Elklit, Hansen and Nielsen2004; Cross and Young Reference Cross and Young2008; Van Haute and Carty Reference Van Haute and Kenneth Carty2012), parties are observed to be organizations of individuals with varying interests and motivations. Most parties in contemporary democracies—whether programmatic or clientelistic—contain a sizable group of activists, especially during elections. As front-line carriers for electoral appeals in local communities, party activists are the political blacksmiths that forge party–voter links. Nevertheless, despite the growing attention to how party activists monitor voters or implement parties’ linkage strategies, the effect of economic well-being on the mechanisms through which voters are attached to parties has not been studied using micro-level data on party activists and their motivations.
This article focuses on the motivations of party activists, investigating whether income shapes these linkage mechanisms. Specifically, this article examines whether party activists in a lower income country (or district) are less policy-motivated compared to those in a higher income country (or district), and whether higher income activists are more policy-motivated than those with lower incomes. To probe this relationship, I first identify and measure the motivations underlying party activism including policy motivations. To do so, I collected and analyzed novel survey data on local party activists from two model third-wave democracies—Mongolia and South Korea. The results lend support for the alternative explanation that linkage mechanisms operationalized as activist motivations are related less to economic well-being and more to the features of parties, specifically their historical origin, ideology, and capacity to deliver clientelistic benefits. I also shed light on the growth of civil society as a possible factor determining party–voter linkage mechanisms by providing citizens with alternative routes to policy influence.
LINKAGE MECHANISM AND ACTIVISTS’ MOTIVATION
The economic explanation claims that, in countries where poverty prevails, it is cheaper for parties to buy votes from the needy than to produce and deliver policy programs. The more developed, the higher the costs parties must pay to monitor citizens who now relocate with increasing frequency or to respond to diverging citizen interests. Consequently, parties in better-off societies will be more likely to cultivate programmatic linkages. This line of thought leads to the conclusion that economic well-being promotes policy-based linkages while weakening non-programmatic ones.
However, these linkage mechanism can be better understood by incorporating the motivations of party activists. As “ambassadors in the community” (Scarrow Reference Scarrow1996), they mediate party–voter communications. While the policy messages may be general slogans or specific proposals, policy-motivated activists can help parties calibrate their programmatic appeals by transmitting the party's messages to voters and collecting grassroots responses where public opinion surveys and online social networks can't reach.Footnote 4 Several studies point to the impact of these policy-motivated activists on the policy positions of candidates and parties, arguing that these activists cause ideological polarization in the party system (May Reference May1973; Aldrich Reference Aldrich1983; Moon Reference Moon2004). On the other hand, policy-indifferent activists, no matter how clear the policy programs the party presents to the public, are likely to be mute on policy issues when they interact with ordinary voters. Thus, the non-programmatic activists allow the parties and candidates to be relatively autonomous in policy decisions and to pay less attention to producing a well-defined policy program, whereas activists’ programmatic motivations will have the opposite effect. Hence, the extent to which party activists are policy-oriented can be regarded as a key attitudinal dimension of the linkage mechanisms observed at the activist level.
If the association between economic well-being and programmatic linkages exists, what would we expect to observe among party activists? The crux of the economic explanation of linkage mechanisms focuses on national characteristics, such as the industrial structure or a large tax base and administrative apparatus, as important conditions for the build-up of programmatic linkages (Kitschelt and Kselman Reference Kitschelt and Kselman2013). If national economic conditions shape linkage mechanisms, we can expect party activists in a highly industrialized country to be more policy-motivated than those in a less developed country.
However, programmatic linkages may not evenly emerge across less and more developed communities within countries. There can be at least two mechanisms that lead the differential development of programmatic linkages. First, voters in affluent communities tend to have high socioeconomic status and, therefore, are expected to be more policy-attentive. Thus, clientelistic approaches by parties in affluent communities have a high chance of failing. This can lead local activists to acknowledge the policy programs of a party (as a part of knowing and responding to their electorate). Second, poverty allows limited time and financial means for social engagement and civic connectedness. Social ties in the local community can grow from frequent and various meetings among the members, such as parents’ meetings, hobby and sports groups, story time and reading groups for children, organizing charity events, and so on. These interactions potentially encourage reflections on public issues such as social inequalities by fostering a sense of civic connectedness (Watts, Williams, and Jagersal Reference Watts, Williams and Jagers2003). Taken together, if the economic argument holds for party activists, the policy-based motivation is more likely to thrive in richer communities.
Another possible line of inquiry concerns the applicability of the economic explanation to the income and motivations of individual activists. The connection between socioeconomic status and political action has been supported in the voluminous literature (e.g., Almond and Verba Reference Almond and Verba1963; Marsh and Kaase Reference Marsh, Kaase, Barnes and Kaase1979; Kaase Reference Kaase, Kent Jennings and van Deth1990). Moreover, those with high levels of socioeconomic status are more likely to engage in organizational activities (Brady, Verba, and Schlozmanal Reference Brady, Verba and Schlozman1995). Past studies, though limited to the contexts of Western democracies, demonstrated that party members and activists also tend to come disproportionately from those with above average income and education (Seyd and Whiteley Reference Seyd and Whiteley1992; Cross and Young 2004; Scarrow and Gezgor Reference Scarrow and Gezgor2010). If socioeconomic conditions are crucial to the attitudinal and behavioral aspects of party activists, the effect of activists’ income may also hold for the motivation of activism. In other words, as poor voters are often believed to overlook or discount policy programs, being lured by short-term, private incentives, low-income party activists may be less policy-motivated than their more financially fortunate peers.
The hypotheses derived from the theoretical discussion are as follows:
H 1
Party activists in a more socioeconomically modernized country will be more policy-motivated than those in a less modernized country.
H 2
Party activists in socioeconomically better-off districts will be more policy-motivated than those in a less developed districts.
H 3
High-income party activists will be more policy-motivated than low-income party activists.
Nonetheless, if the evidence negates the association between economic well-being and activist motivations, we need to consider other plausible explanations. As an alternative explanation, the characteristics of parties themselves may account for motivation. First, parties rooted in an authoritarian past tend to possess more resources for clientelistic distribution, with a higher level of party institutionalization for the effective distribution of goods (Hicken and Kuhonta Reference Hicken and Kuhonta2011). Furthermore, decades of clientelistic interactions between the authoritarian parties as a state apparatus and ordinary voters are likely to have remained as the organizational custom of these parties, affecting the current motivation of party activists as well.
Second, minor parties often lack the capacity to deliver material or other selective benefits to voters and party activists. This is especially the case where the constitutional design is close to the majoritarian model of democracy, in which power is concentrated in the hands of the majority party (Lijphart Reference Lijphart1991). In the majoritarian model, while major parties alternate in power, minor parties have little access to state resources except for state subsidies distributed to parties according to legislative seats. Hence, the electoral campaigning of minor parties under such institutional constraints will relatively understate non-programmatic appeals, and those working for minor parties are expected to be more policy-oriented than the activists of major parties.
Lastly, in ideologically leftist parties, policy motivation is likely to be more pronounced than other motivations. Their activists, as well as voters, usually come from both underprivileged classes and those who have leftist identities. For instance, Whiteley and Seyd's (Reference Whiteley and Seyd2002) study on the activism of British parties demonstrated the absent or even negative effect of income on party activism in the Labour Party, while they found the opposite for the Conservative Party. Leftist parties aiming to tackle structural issues such as inequality with state resources are expected to make relatively more investment in policy programs, and their activists will be more policy-motivated than those of rightist or moderate parties.
The discussion about potential party factors leads to three expectations within a party system in the contexts of new democracies:
H 4
The activists of authoritarian successor parties will be less policy-motivated than other parties.
H 5
The activists of minor parties will be more policy-motivated than those of major parties.
H 6
The activists of leftist parties will be more policy-motivated than those of centrist and rightist parties.
METHODOLOGY
To evaluate the economic hypotheses on the effects of national-, local-, and individual-level incomes, I carefully selected the countries and districts where I would collect information on party activists. First, the survey had to be carried out in at least two countries that are similar in many aspects but markedly different at the level of economic development. And second, I needed to find at least two districts in each country that vary in economic development but resembling each other in many other respects. Furthermore, to evaluate whether individual-level income shapes activists’ motivations, it is essential to control for other individual characteristics that may affect this relationship. Lastly, evaluating the hypotheses about party factors requires at least two parties that are different in party ideology, size, and whether the party has deep roots in dictatorship.
Accordingly, I conducted surveys on party activists from multiple districts in Mongolia and South Korea during the official period of campaigning for the 2012 legislative election in each country (June 5–27 in Mongolia, the election held on June 28; March 24–April 10 in South Korea, the election held on April 11). As the subjects of interest are party activists (neither formal party members nor party supporters), the survey was performed when party activists were “visibly in action”; I identified them by their roles in the election campaign of a candidate running with a party label.Footnote 5
In small-N comparisons, cases must be carefully selected to avoid selection bias, since the comparative-case strategy tries to “maximize the variance of key variables and to minimize the variance of the control variables” (Lijphart Reference Lijphart1975, 164). South Korea and Mongolia qualify as “the most similar systems” (Przeworski and Teune Reference Przeworski and Teune1970) in several aspects. Both underwent a regime transition in the “third wave” of democratization. Specifically, for both, a combination of a series of mass demonstrations and elite negotiations paved the way for the introduction of a democratic constitution. Under the new constitution, democratic elections have been held in Korea since 1987 and in Mongolia since 1990. Due to their relatively peaceful processes of transition and uninterrupted democracy for the last quarter-century, the two countries are often referred to as successful models of third-wave democracy.
The two countries are similar in key institutional aspects as well. Both have a unicameral legislature, a directly elected president, and a moderate multiparty system with an effective number of parties (Laakso and Taagepera Reference Laakso and Taagepera1979) usually ranging between two and three. Both countries have four-year electoral cycles, and in the 2012 legislative election, a mixed electoral system—a combination of single-member district plurality (SMDP) and nation-wide closed-list proportional representation (PR) systems. Plurality seats greatly outnumber PR seats in the legislature in both countries.Footnote 6
Nonetheless, the two countries differ significantly in their level of development. South Korea is a high-income country and a member of the Organization for Economic Cooperation and Development (OECD), while Mongolia is a low-middle income country. Coupled with the substantial similarity in non-economic dimensions, the marked difference in national economic well-being makes South Korea and Mongolia an effective combination of cases to examine whether policy motivations distinctly emerge across countries at different levels of national economic development.Footnote 7
POLITICAL PARTIES IN SOUTH KOREA AND MONGOLIA
I focus on activists from “relevant” parties that gained at least one seat in the legislature in the previous election in 2008 and maintained some organizational persistence until the 2012 election. Accordingly, I carried out a unique survey of nearly 300 local activists from two Mongolian and five Korean parties: the Mongolian People's Party (MPP), the Democratic Party (DP), the Saenuri Party (SP), the Democratic United Party (DUP), the Liberty Forward Party (LFP), the United Progressive Party (UPP), and the New Progressive Party (NPP).Footnote 8 The MPP and the DP have been the two major parties in the Mongolian party system since the early 1990s, dividing 95 percent of the legislative seats in the 2008 election. The MPP was the only legal party that ruled communist Mongolia from 1921 to 1990. The DP, on the other hand, originated from the democratic coalition during the regime transition. In most legislative elections until 2012, except for the one held in 1996, the MPP maintained its governing party status in the State Great Khural (SGK, the legislature in Mongolia), while the DP remained the main opposition party. As observed in many post-dictatorships, the authoritarian successor party has overwhelmed its opposition in terms of its institutionalization and organizational capacity to mobilize its members and activists.
In South Korea, the party with authoritarian roots is the SP. The merger between a democratic faction and parties of authoritarian elites established the party in 1990, though its party name has changed several times since then. The DUP (whose party name has changed more frequently than the SP), on the other hand, has its roots in pro-democracy movements. The SP and the DUP have alternated control over the presidency in democratized Korea. The SP had been the largest party in the National Assembly (NA) until 2016, except for the period of 2004–2008.
The LFP was the third party in the NA from the 2008 NA election. A faction led by a former authoritarian elite founded the party in 1995. Over time, the LFP distinguished itself from other parties by adding overtly rightist ideology to its regional appeals. Lastly, the UPP and the NPP were the two leftist parties with a shared origin, the Democratic Labor Party (DLP). In 2004, the DLP became the first leftist party entering the NA in democratized Korea. However, it had suffered from factional tensions since its foundation. The intra-party conflicts culminated in early 2008, when those opposing the pro-North Korea sentiment of the majority faction finally formed a splinter party, the NPP. The pro-North DLP, some prominent politicians from the NPP, and some former DUP members merged to create the UPP ahead of the 2012 NA election. Table 1 shows the profiles of Korean and Mongolian parties concerning their links to the authoritarian regime, party ideology, and size and their expected level of policy motivations according to the party hypotheses.
Profiles of Korean and Mongolian Parties and their Expected Level of Policy Motivations
Note: The profiles of the parties are based on their features at the time of the 2012 Legislative Election; PM denotes the expected level of policy motivations according to the party hypotheses. The list of full party names are as follows: the Mongolian People’ Party (MPP), the Democratic Party of Mongolia, the Saenuri Party (SP), the Democratic United Party (DUP), the Liberty Forward Party (LFP), the United Progressive Party (UPP), and the New Progressive Party (NPP).
Sample Districts
One of the hypotheses compares activist motivations across communities with varying degrees of income and economic development. Therefore, I conducted the survey in at least one markedly developed district and one indigent district from each country. There were some additional considerations in sample district selection. First, capital districts were excluded. The capitals of the two countries are megacities, with the national population concentrated in these urban centers (approximately 50 percent of the national population residing in Seoul and its vicinity and 45 percent in Ulaanbaatar).Footnote 9 Along with the population concentration, socioeconomic diversity has sharply increased among capital residents even within a small, single-member electoral district. Obviously, districts in which extreme wealth and poverty coexist are not ideal for comparing activists across districts. Moreover, activists in the capital may engage less in a “local” camp and local activities than in central party organizations or municipal-wide activities. I also excluded the regions whose ethnic and linguistic composition is markedly heterogeneous from the rest of the country, such as the western provinces (aimags) of Mongolia. Lastly, I excluded traditional strongholds of any party to ensure fair within-district comparisons between parties.
Table 2 displays the sample districts selected by the criteria. They are one underdeveloped, agricultural district (Arkhangai) and one developed, mining-boom district (Orkhon) from Mongolia, and two underdeveloped, agricultural districts (Icheon and Boeun/Okcheon/Youngdong), two relatively blue-collar districts (Jungwon and Pyeongtaek), and three developed, affluent districts (Yongin, Boondang, and Daejeon-Seogu) from South Korea. The fourth column of Table 2 lists parties of which activists participated in the survey in each sample constituency.
Electoral Districts and Political Parties in the Sample
Note: The list of full party names are as follows: the Mongolian People’ Party (MPP), the Democratic Party of Mongolia, the Saenuri Party (SP), the Democratic United Party (DUP), the Liberty Forward Party (LFP), the United Progressive Party (UPP), and the New Progressive Party (NPP).
Identifying and Measuring Activist Motivations
Before examining the relationship between economic well-being and activist motivations, we need to know first whether policy motivations are empirically distinguishable from other types of motivations and to measure them. A number of studies have asserted the multidimensionality of motivations underlying party activism (e.g., Panebianco Reference Panebianco1988; Costantini and King Reference Costantini and King1984; Whiteley et al. Reference Whiteley, Seyd, Richardson and Bissell1994; Cross and Young Reference Cross and Young2008). First, some activists are driven by selective and tangible rewards, such as incentives that may help to heighten their social, political, or economic status. I regard this category of incentives as including clientelistic goods (access to power, jobs, service accessibility, or money) that would meet most of one's non-programmatic desires.Footnote 10 Second, some may find political activities meaningful due to the fresh and unusual experiences they can provide. Through partisan political activities, these activists could gain the feeling of being connected to a larger group of their fellow citizens. In particular, engaging in a rare, national event such as legislative elections can be an exciting opportunity to escape from the daily routine. Third, some may participate with a policy purpose. These activists are interested in public issues and have a desire to influence public policies, even if they lack a clear picture of how they may want to change society. This type of incentive has been called “moral” (Bruter and Harrison Reference Bruter and Harrison2009), “collective” (Panebianco Reference Panebianco1988), or “ideological” (Whiteley et al. Reference Whiteley, Seyd, Richardson and Bissell1994). Fourth, some may decide to engage in political activities because of their sense of obligation. The belief that participation is their duty as a party member, a citizen in a democratic polity, or a member of a local community may lead them to party activism. Lastly, for some others, these activities are no more than a favor to their friends or family. They work for a party or a candidate because of personal relationships, regardless of their agreement with the policy direction of the party or interest in its clientelistic rewards.
To identify and measure the motivations, I employed a questionnaire devised by Costantini and King (Reference Costantini and King1984). It consists of 19 items that attempt to capture the five conceptual types of motivations just outlined. Respondents were asked to indicate the importance of each item on a four-point scale.Footnote 11 And to determine the dimensionality and estimate the extent to which each dimension motivates an activist, I used an item bifactor model. As a multidimensional item response theory (IRT) model, the bifactor model summarizes a multitude of measurements with a small number of factors, identifying the critical dimensions of a theoretical concept.Footnote 12
Table 3 presents the results of this bifactor analysis. The results identify four distinct dimensions of motivations in the two sample countries: self-enhancement, sociality, purposive, and personal dimensions. Interestingly, the allegiance dimension is indiscernible in the structure of motivations, although previous research on the motivations of US party delegates empirically corroborated all five dimensions (Costantini and King Reference Costantini and King1984). While explaining this finding is beyond the scope of this article, a possible explanation may be found in the relatively low level of trust in parties in the two countries.Footnote 13 The low factor loadings and the reliability score on the allegiance dimension may suggest that political parties are not considered to be the backbone of democracy even by their foot soldiers, but rather thought of at best as an instrument for some other purpose. Another possible explanation is that the items used to capture the allegiance dimension (“to demonstrate my loyalty to the state and party leadership,” “strong party loyalty,” and a “sense of community obligation”) in fact concern motivations too different to be measured as a single construct. New democracies lack the experience of gradual and consistent democratic development in which the state, political parties, and civic communities mature together. Accordingly, these three items may be less compatible with one another in new democracies.Footnote 14
Structure of Activist Motivations in Korean and Mongolian Parties: Full-information Bifactor Analysis
Note: Cell entries are factor loadings. The factor loadings and empirical reliability scores from the unidimensional IRT model for each of the Self-enhancement, Sociality, Purposive, and Personal dimensions are in parentheses.
RESULTS
Effect of National and Local Development
To probe the relationship between national economic conditions and activist motivations (H 1), I compared activists’ average trait values in each motivational dimension obtained from the unidimensional IRT analyses of the two countries. For a fair inter-country comparison, only major parties are considered (the MPP and the DP in Mongolia; the SP and the DUP in South Korea). Table 4 displays the country average values on the four identified dimensions. The most intriguing finding is the stark difference between Mongolian and Korean activists in both the levels of self-enhancement and purposive motivations. At a glance, the finding that activists in less-developed Mongolia seek private rewards more than Korean activists seems to confirm the expectation of the economic explanation. However, Mongolian activists are simultaneously more policy-motivated than those in post-industrial Korea, confounding simple explanation. Furthermore, the country difference appears more remarkable in the purposive dimension than in the self-enhancement dimension.
Comparing Activist Motivations in Korean and Mongolian Parties (Major Parties Only)
Note: Entries are mean values of trait scores for each dimensional variable, measured by separate unidimensional IRT models. Two-way t test.
This finding should be interpreted with caution. The significantly low levels of both types of motivations among the Korean activists may suggest that many do not regard their parties as the most important channel for achieving nontrivial goals for themselves (e.g., a public office/job and business benefits) or for the broader society (e.g., policy influence). Their low policy motivation, in particular, may indicate that the major parties are more electoral-professional parties through which citizens are not especially empowered or brought together for political aims, and in which, rather, the activists perform a limited role as policy messengers. This interpretation is supported by the conventional view that Korean parties are electoral vehicles for a few politicians and the weakest link for further democratic development of the country (e.g., Kim Reference Kim, Diamond and Kim2000; Wong Reference Wong, Hicken and Kuhonta2014).
On the other hand, it may be too hasty to conclude that the result confirms the weakness of party-driven democracy in Korea. The finding may instead reflect the different state of civil society in the two countries. After the democratic transition, South Korea witnessed a significant increase in the number and role of civil society organizations during the 1990s and 2000s, in various fields including human rights, gender equality, labor, consumer rights, chaebol reform, and even right-wing issues. The growth of civil society offers diverse opportunities to participate in politics for policy influence. This change resulted in the parties being viewed as one of many instruments to reach programmatic goals, while activists have only gradually responded to the opportunities to socialize and engaged in political “fan-club” activities offered by them (Koo Reference Koo2018). On the contrary, the growth of Mongolian civil society has been limited, mainly due to the legacy of 70 years of Soviet rule and the delayed growth of the middle class. Consequently, political parties there may be perceived as relatively effective tools and credible providers of both private and public goods, vis-à-vis civil society organizations.Footnote 15 In either interpretation, the evidence from the cross-country comparison did not confirm the hypothesis that policy motivation among party activists in a more industrialized country is higher than in less developed comparisons (H 1).
Next, I examined whether the motivational pattern can be predicted by local-level economic well-being. Table 5 presents the results of two-tailed t-tests for the mean difference between affluent and less-affluent districts in each motivational dimension. This cross-district comparison was permitted for five parties whose activists from both types of districts participated in the survey. The results indicate that local-level well-being is conducive to neither programmatic nor non-programmatic motivations, contrary to the hypothesized relationship (H 2). The only exception is found in the UPP; their activists in a relatively urban district tend to be more policy-motivated than their peers in a rural district. Nevertheless, taken as a whole, there is little evidence supporting that programmatic motivations flourish in economically well-to-do districts.
Local Economic Well-being and Activist Motivations
Note: Entries are t-statistics from two-way t-test for mean difference between poor and rich districts. None of the t statistics are statistically significant at the level of 0.05. aThe survey responses for the DUP and the LFP were collected only from wealthy districts; bThe UPP Party activists in Jungwon-Gu are compared to their fellow activists in Icheon, as the former district is relatively wealthy compared to the latter.
Effects of Income
I first explored the bivariate relationships between activist motivations and several demographic, economic, attitudinal, and behavioral attributes. Specifically, I divided party activists into three groups according to their trait scores on the self-enhancement and purposive dimensions and probed differences between these groups.
Table 6 indicates that policy motivation does not increase as income goes up, no matter whether measured by annual household income or subjective evaluation of one's economic situation. Instead, the least policy-motivated group appears to be wealthier than its comparison groups. Furthermore, there is no compelling difference in income across different levels of self-enhancement. Albeit preliminary, the expectation of the economic hypothesis that activists’ income fosters programmatic attitudes (H 3) is not supported, at least not in these bivariate group comparisons of Korean and Mongolian party activists.
Individual Characteristics of Activists, by Varying Degrees of Seeking Selective Incentives and Policy Incentives (Pooled Sample)
Note: Entries are the mean value of each variable for the activists in the given category; a -1 for “having participated in the campaigning activities for a different party”; 0 for “this is the first participation”; 1 for “having participated in the campaigning activities for the same party”.
The results also yield several other interesting patterns. To briefly summarize, policy-seeking activists read, and have more opportunities to engage in discussions about, the policy documents of their party. Moreover, policy-motivated activists have higher political efficacy and actively engage in both conventional and unconventional forms of political participation. They appear to be desirable citizens in light of the principle of participatory democracy.
Table 7 (self-enhancement as a dependent variable) and Table 8 (purposive as a dependent variable) present the results of multivariate regression analyses. These analyses allow us to determine the factors affecting activist motivations, holding both national- and district-level economic well-being, political party, and other individual factors constant. The regression results trace the patterns outlined in the bivariate analysis, showing that income has little to do with either self-enhancement or policy motivations. Specifically, household income and financial satisfaction have a significant effect on self-enhancement (Table 7) only for two (the SP and the NPP) among seven of the parties in the two countries, although the negative coefficients conform to the expectation of the economic explanation. Furthermore, in the model explaining policy motivation (Table 8), the effects of the objective and subjective income variables are inconsistent across parties. This finding suggests that an activist with clientelistic motivations would not become more policy-oriented as her financial situation improves. Overall, the regression analyses fail to support the argument that poor activists are myopic self-seekers while affluent activists are more programmatic policy-oriented (H 3).Footnote 16
Individual-level Regression Analyses, by Party and by Local Economic Well-Being (Attraction of Selective Incentives)
Note: ***p < .01; **p < .05; *p < .1.
Individual-level Regression Analyses, by Party and by Local Economic Well-Being (Attraction of Policy Incentives)
Note: ***p < .01; **p < .05; *p < .1
PARTY FACTORS AND ACTIVIST MOTIVATION
A closer examination reveals considerable variation across parties in levels of self-enhancement and purposive motivations. Figure 1 compares the average values of the two motivations across the two Mongolian and five Korean parties. Among several interesting findings in this cross-party comparison, it is notable that the activists in a party rooted in dictatorship are the most self-seeking and the least policy-seeking in a given party system. The least programmatic parties are the MPP (the former dominant party in power) in Mongolia and the SP (the party founded by a merger between authoritarian and some democratic elites) in South Korea. This finding supports H 4 that the activists of an authoritarian successor party are less likely to be policy-motivated compared to those of parties with democratic origin.
Comparing Activist Motivations in Mongolian and Korean Parties
Next, in the Korean multiparty system, activists of the minor parties (the LFP, the UPP, and the NPP) appear more policy-motivated and less self-seeking than those of the major parties (the SP and the DUP). This finding lends support to H 5, or the second hypothesis, that the programmatic type of activist characterizes minor rather than mainstream parties. The Korean presidential system and its mixed electoral system, in which PR seats account for only one in six seats in the legislature, inherently favors two major parties, giving small parties little leverage to influence significant national decisions regarding resource allocation.
Lastly, the parties with the most policy-motivated activists in South Korea are the leftist parties, the UPP and the NPP. Proclaiming to represent marginalized sectors of society, the UPP and the NPP have the structure of the mass party whose primary source of financing is membership fees. In this country founded as an anti-communist state and technically still in a war, leftist ideologies have long been taboo. For the politically marginalized leftist parties, collective identities and progressive policy programs have been the primary resource used to motivate their activists and supporters. Hence, the last hypothesis about party factors, H 6, is validated as well.
CONCLUSIONS
The conventional wisdom concerning political parties in new democracies is that they rarely compete over policies, while valence issues prevail in election campaigning in a way that “our party can do anything better than others.” Indeed, East Asian democracies have often been viewed as “unlikely places” from which programmatic parties and party systems would emerge. A range of factors have been advanced as reasons, including the legacy of colonialism and the subsequent rise of nationalism, the limited role of organized labor in nation-building and industrialization, the Confucian emphasis on the “golden mean,” the legacy of the Cold War, and the region's relatively short experience with democracy. These assumptions allowed East Asian democracies to escape close scholarly inquiry.
Using new survey data collected in two third-wave democracies in East Asia, I examined the correlates of activist motivations, focusing on the effect of income. Incorporating party activists into the notion of linkage mechanisms led to the expectation that economic well-being would be associated with activists’ increased role as policy messengers. The evidence from Mongolia and South Korea, however, undermines the premise of the economic account, suggesting that national, local, and individual economic development are not powerful determinants of programmatic motivations, though my results do provide considerable insight into linkage mechanisms. While the voluminous literature highlights income as being crucial in understanding party (and other political) activism, my study casts doubt on the direct relevance of income in explaining activist motivations. Indeed, the evidence tells us that wealthy individuals are not more policy-motivated than their less affluent peers, and that high-income activists seek clientelistic incentives just as much as do their lower-income comparison groups. This finding calls for further research to ask and answer more nuanced questions, such as whether there are different types of clientelistic desires that may in fact be associated with differences in wealth. For example, whereas low-income activists might seek short-term materialistic rewards, the clientelistic rewards that high-income activists pursue may be relatively long-term.
The most surprising finding is the markedly high desire for instrumental utilities, especially policy influence, in less-developed Mongolia, compared to the low level of programmatic motivations in post-industrial Korea. This study preliminarily proposed the state of Mongolian and Korean civil society as a possible explanation for this finding. Economic growth and industrialization may be a “pull factor” inducing policy-motivated activists to attach themselves to parties and, therefore, to contribute to programmatic linkages, as the literature claims. However, development and the resulting entry to post-industrialization are usually accompanied by shifts in the socio-political environment. Among such changes is the rise of civil society organizations, as well as many new forms of political participation enabled by technological advances. These act as rivals to political parties, which were once the primary channel to reach instrumental utilities. Despite the unprecedented growth accompanying the mining boom of the 2000s and early 2010s, the legacy of Communist rule has delayed the emergence of an active civil society in Mongolia, except for a few non-governmental organizations receiving Western aid (Fish and Seeberg Reference Fish and Seeberg2017). In contrast, South Korea witnessed the growth of a middle class prior to its democratic transition, and, after the transition, the burgeoning of civil society organizations led by former activists of the democratic movements. While it is arguable that the major Korean parties are dysfunctional compared to the new growing channels of policy influence, the provision of a broader range of non-party – seemingly cooler – democratic instruments may be a “push factor” distracting the most policy-motivated segment of the citizenry from party-driven activism.
More importantly, this study sheds light on the characteristics of parties to explain within-system variations in activist motivations. While parties in the two countries have been assumed to lack programmatic elements, the empirical evidence reveals remarkable distinctions across parties in the extent to which they are programmatic in the intermediary layers between parties and voters. Authoritarian successor parties tend to attract activists with clientelistic interests more than democratic opposition parties do, echoing the importance of the authoritarian past in analyzing politics in new democracies. Also, small and leftist parties, rather than their larger and conservative rivals, are more favored by those with programmatic interests. This finding suggests that civil society organizations, the possible push factor, and minor/leftist parties may share similarities in their mechanisms of linking citizens to the state, which is an area worthy of further investigation.
As a whole, this study demonstrates that our understanding of linkage mechanisms can be made richer by explicitly incorporating activists into the discussion. Notwithstanding that larger parties will obviously be more credible providers of both programmatic and clientelistic goods, some policy-motivated activists choose parties unable or unlikely to crystallize their policy desires into national policies in the foreseeable future. The inflow of policy-motivated activists to small parties may in turn shape their strategic decisions over whether to invest more on programmatic or clientelistic linkages. These activists may also be able to influence mainstream parties’ policy programs, using these small parties as a means.
Important questions to resolve for future studies are whether different activist motivations lead to different patterns of involvement in intra-party decision-making and, more specifically, whether policy-motivated activists are more engaged and influential in internal party affairs. This study suggests that the correlation between policy motivation and party activism may be a feature of small parties more so than those in the mainstream and of parties with democratic origins more so than those with authoritarian roots. In this regard, linkage mechanisms should be regarded as multi-level phenomena operated and observed not only at the party elite and voter levels, or at the interactions between the two, but also at the activist level, or at the activists’ interactions with another stratum of the party.
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit https://doi.org/10.1017/jea.2018.20.
