Q Routledge Taylor &. Francis Groi Eurasian Geography and Economics ISSN: 1538-7216 (Print) 1938-2863 (Online) Journal homepage: https://www.tandfonline.com/loi/rege20 The rise of new populist political parties in Czech parliamentary elections between 2010 and 2017: the geography of party replacement Pavel Maskarinec To cite this article: Pavel Maskarinec (2019) The rise of new populist political parties in Czech parliamentary elections between 2010 and 2017: the geography of party replacement, Eurasian Geography and Economics, 60:5, 511-547, DOI: 10.1080/15387216.2019.1691928 To link to this article: https://doi.Org/10.1080/15387216.2019.1691928 Published online: 25 Nov 2019. or Submit your article to this journal D? IjJiL Article views: 398 5>1 View related articles ß1 View Crossmark data G? Citing articles: 11 View citing articles B" Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journal Info rmation?journalCode=rege20 EURASIAN GEOGRAPHY AND ECONOMICS 2019, VOL 60, NO. 5, 511-547 https://doi.Org/10.1080/15387216.2019.1691928 Check for updates The rise of new populist political parties in Czech parliamentary elections between 2010 and 2017: the geography of party replacement Pavel Maškarinec© Faculty of Arts, Department of Political Science and Philosophy, Jan Evangelista Purkyně University, Ústí nad Labem, Czech Republic ABSTRACT This article examines the territorialization of party replacement in the Czech Republic between 2010 and 2017. Using spatial analytical techniques, we found that despite the significant success of new populist parties, which considerably transformed the parliamentary strength of individual parties, the "new" geographical patterning of party support was far from real transformation. Historically, most of the rightist and center-right parties relied on higher support in areas with a high development potential and the developmental axes related to them, whereas established left-wing parties found support in areas with a low development potential. The elections of 2010 and 2013 were exceptions to this pattern, but the elections of 2017 interrupted this deviation when geographical support for some of the new populist parties (those which had succeed in 2013 already) moved mostly to underdeveloped areas. However, as the programmatics of most of the new populist parties are relatively unclear (especially on the left-right ideological division), one can hardly speak of class realignment, but rather of spatial re-stratification of Czech politics: traditionally leftist and rightist voters largely abandoned "their" (established) parties, but on the other hand, their voting mostly did not contravene the Czech spatial division between leftist and rightist regions. Introduction This paper provides an analysis of the territorialization of new populist political parties and the geographical dynamics of electoral change in recent Czech elections. The Czech Republic is an excellent case for testing the territorial patterns of support for fading and rising parties and the spatial patterns of party replacement. This is because the Czech party system had been (in contrast to many other Central and Eastern European countries) relatively stable during the first two post-communist decades, with low levels of volatility, the failure of CONTACT Pavel Maškarinec @ maskarinec@centrum.cz (^Faculty of Arts, Department of Political Science and Philosophy, Jan Evangelista Purkyně University, Pasteurova 3571/13, Ústí nad Labem 400 96, Czech Republic © 2019 Informa UK Limited, trading as Taylor & Francis Group Routledge Taylor & Francis Group ARTICLE HISTORY Received 26 August 2018 Accepted 8 November 2019 KEYWORDS Czech Republic; parliamentary elections; spatial analysis; party replacement; electoral change; populist parties 512 @ P. MAŠKARINEC most new political parties (Deegan-Krause and Haughton 2010; Hanley 2012), and similar levels of support for political parties across the country (Tiemann 2012; Stauber 2017). However, in the October 2017 election to the lower house of the Czech parliament (the Chamber of Deputies) the number of parliamentary parties rose again to nine (compared to five in the elections of 2006 and 2010), the same number as after the first election to the Chamber of Deputies in 1992, before the consolidation of the Czech party system began. Even more importantly, patterns of political support shifted away from previously fairly stable allegiances as the former main parties of the left and the right together achieved less than one-fifth of votes, compared to over two-thirds in 2006. An important aspect of assessing the geography of party system change is to correctly identify new parties as research has identified considerable diversity in what constitutes "new" in this context (Barnea and Rahat 2011; Litton 2015; Sikk and Koker 2017). Sikk (2012) categorizes new parties into four types: (1) prophets, which introduce genuinely new ideological themes into party competition; (2) prolocutors, which are not linked to specific ideologies but aim to represent neglected or unrepresented issues or interests in a society; (3) purifiers, which seek to articulate existing party traditions in more principled and authentic forms, thus targeting an electorate mobilized by established parties; and (4) ones based on a project of newness, which lack any clear conventional ideology but focus on an already-occupied ideological constituency. This article focuses primarily on two genuinely new populist parties which entered parliament in the 2013 elections and were able to defend their parliamentary representation in the 2017 elections: the ANO2011 ("ano" means "yes" in Czech) of Andrej Babiš, a successful tycoon and the second-richest man in the Czech Republic (see Kopeček 2016; Hloušek and Kopeček 2019), which exemplifies a project of newness (i.e. a party without a stronger ideology); and the Dawn of Direct Democracy of Tomio Okamura or Freedom and Direct Democracy - Tomio Okamura (referred to as "the Dawn" and the SPD respectively), which are examples of prolocutor parties due to their strong emphasis on direct democracy, a theme which is relatively ignored in analyses of Czech political parties (see Maškarinec and Bláha 2014; Havlík 2015). Public Affairs (W) was the only other new Czech party which could be classified as a project of newness. However, following success in 2010 they ceased to exist during the parliamentary term, and therefore our analysis only deals with W's electoral support in the context of the rise of the ANO2011, the Dawn, and the SPD, due to strong links between support for all these political parties (Gregor 2014). The main reason for focusing on these parties is that most of the other new parties that successfully entered the Czech party system after 2010 can be classified as purifiers (Hanley 2012). This is also supported by existing analyses of electoral support: for example, in the case of the Tradition, Responsibility, Prosperity 09 party (TOP09), which succeeded in the 2010 parliamentary elections, a considerable overlap was identified between EURASIAN GEOGRAPHY AND ECONOMICS (Q) 513 TOP09's electorate and the spatial support for traditional Czech rightist and center-right parties (Havlik and Voda 2016; Maskarinec 2017). The results of the 2010, 2013 and 2017 Czech parliamentary elections allow us to address two key questions: did voting patterns change substantially in these elections? If so, how (or to what extent) did new political parties replace old ones and what was the pattern (or type) of party replacement? This paper therefore contributes to the international research literature on spatial analyses of electoral change (Shin and Agnew 2007; Tapiador and Mezo 2009; Morrill, Knopp, and Brown 2011; Johnston, Manley, and Jones 2016), in particular by examining how old parties are replaced by new ones and the underlying geographical patterns of support for the fading and rising parties. The significance of our findings lies in revealing the persistence of the underlying spatial patterns of the electoral process. Although Czech electoral competition has undergone profound changes, with a considerable rise of new parties and decline of established ones, the spatial analyses of voting behavior presented here shows that the new geographical patterns of party support do not represent a real transformation of the patterns which arose after the introduction of Czech democracy in 1990. We show that spatial analysis - especially an analysis of the spatial patterns of party replacement - can serve as a useful diagnostic tool and contribute to identifying underlying socio-spatial processes which influence the territorial processes of party replacement. Electoral geography and voting Geography is an important element of voting. Political behavior and electoral change are traditionally explained by two theoretical approaches that Thrift (1983) defines as compositional and contextual approaches. While the compositional approach focuses on the electoral behavior of objectively identifiable social groups, assuming that members of the same social group behave in similar ways and that socioeconomic structure sufficiently explains electoral behavior (Evans 2004), the contextual approach strives to explain electoral support with reference to the characteristics of the environment in which one lives, drawing links between voting and spatial context (Taylor and Johnston 1979), and draws analytical attention to local milieux as contexts in which political opinions are formed (Johnston and Pattie 2008). In fact, neither of these polar positions is empirically defensible. Every individual is a part of a specific social structure - a social location (Giddens 1984) -and must occupy a physical (spatiotemporal) context and, in turn, every context brings together different individuals. This claim, however, does not imply that social structure, as a sum of individuals, is not relevant to voters' choices -rather, much of one's understanding of the social structure, ergo of individuals as well, is learnt in a local context (Johnston 1987). As Harbers (2017) demonstrates, electoral behavior continues to differ between differing electoral 514 @ P. MASKARINEC districts, whereas electoral districts with similar socio-demographic characteristics are likely to be located near one another. In this context, the Lipset and Rokkan (1967) cleavage model has been one of the main methods used to explain variation in geographical patterns of voting behavior, with Taylor and Johnston (1979) and others emphasizing that its advantage (from the viewpoint of electoral geography) is that it brings a spatial dimension to the analysis (Flint and Taylor 2007). The cleavage model is based on the assumption that European party systems emerge as alignments formed between different groups (segments) of voters and political parties representing their particular interests. Specifically, Lipset and Rokkan identified four societal cleavages at the time of the formation of the European party system in the second half of the 19th and in the early 20th century: the center-periphery, state-church, owner-worker, and urban-rural divisions (cf. Lipset and Rokkan 1967; Dogan 2001). Therefore, where voting behavior aligns with these societal cleavages the geography of support for a particular party should correspond with the geographical distribution of the members of the relevant sectors of society (Johnston and Pattie 2008). Since the 1960s, however, there has been a decline of voter identification with established parties and a growth of volatility in advanced industrial democracies, i.e. a process of dealignment due to the social and political modernization of society (Dalton and Wattenberg 2000). Despite this, social structure and cleavages continue to play an important role in voters' choices. In this regard, one must emphasize social context, including the place in which individual voters grow up and from which they derive their fundamental attitudes and values (Evans 2004). With regard to political change, Nardulli's (1995) analysis of US presidential elections between 1928-84 confirmed that critical realignment (i.e. changing party allegiance) was not a national phenomenon; on the contrary, it was a subnational phenomenon localized in spaces that varied considerably across geographic locations. The importance of space in explaining electoral behavior was also confirmed by Darmofal (2008), who showed that one of the principal American political realignments (the 1928-36 Democratic realignment) was highly localized and subnational. Similarly, Kim, Elliot, and Wang (2003) found that the increasing geographical concentration of partisan voting in US presidential elections and continuing regional alignments went along with an increasingly polarized American electorate (and see Johnston, Manley, and Jones (2016)). In Europe, Shin and Agnew (2007) showed that political change in Italy between 1987-2001 was geographically dependent and, notwithstanding some overlap of geographical support between old and new parties, there were also significant regional differences and the process of party replacement was not a stationary one. Concerning new democracies, Tapiador and Mezo (2009) showed that, despite evidence for the nationalization thesis, i.e. increasing uniformity of electoral support for individual parties in different regions (Caramani EURASIAN GEOGRAPHY AND ECONOMICS (Q) 515 2004), the nationalization process in Spain was compatible with very strong and differentiated regional and local dynamics when some regions became much more internally homogenous than others. Spatial dependence and importance of the territorialization of political parties for electoral competition was also confirmed by Ozen and Kalkan (2017) in Turkey and Sichinava (2017) in Georgia. The Lipset-Rokkan model has also been successfully adapted to explain party formation or electoral geography in the new democracies of Central and Eastern Europe (Zarycki 2000; Whitefield 2002; Evans 2006; Hlousek and Kopecek 2008; Voda and Pink 2015). For instance, Meszaros, Solymosi, and Speiser (2007) demonstrated that Hungarian politics was highly territorially structured: although the right-wing political bloc was characterized by party replacement, its electoral base remained territorially stable. The left-wing political bloc also remained territorially stable, but was represented by the same parties after democratization. Zarycki (2015) emphasized the domination of the liberal-conservative cleavage in a recent transformation of the Polish political scene that affected the structures of the country's electoral geography, resulting in more visibility of the East-West division of Poland's political space. This division was explained by Marcinkiewicz (2018) primarily as the effect of the urban-rural divide, while the economic cleavage (traditionally dominant in Western democracies) mattered only in certain socio-cultural contexsts. Grabowski (2019) also mentioned historical and cultural factors as the most important determinants of Polish electoral behavior. Similarly, the importance of the urban-rural dimension for political competition was also confirmed by Plesivcak (2013) in Slovakia. Most recently, spatial context had a strong influence on voting in the UK Brexit referendum. Here, Becker, Fetzer, and Novy (2017) clearly show that areas with deprivation in terms of education, income and employment were more likely to vote "leave", and these spatial disparities increased the polarization of England's politics between cosmopolitan areas of growth and "provincial backwaters", i.e. areas experiencing decline which can be characterized as "inward-looking, relatively illiberal, negative about the EU and immigration, nostalgic and more English" in their identity (Jennings and Stoker 2016). More importantly, the influence of the same factors as in the Brexit vote was also found by Becker, Fetzer, and Novy (2017) in French presidential elections, which again demonstrated the necessity to focus on place and space when explaining political change in the context of continuing dealignment and realignment and increasing polarization in many formerly highly nationalized party systems across the globe. Given these findings, it is important to determine, in our case, whether the rise of new parties (i.e. how old parties are being replaced by new ones) is based on the transformation of the underlying geographical patterns of party support. These patterns were, in the Czech case, mostly based on the socioeconomic characteristics of regional populations (see below) and a spatial division between leftist and rightist regions. The question is whether the success of 516 @ P. MAŠKARINEC new populist parties and increasing polarization (in terms of growing number of parliamentary parties) lead to the spatial re-stratification of electoral behavior, and whether the spatial patterns of voting for individual parties alters the former division of Czech party politics. Electoral geography and party voting in the Czech Republic However, before we delve into the spatial analysis of change in the Czech Republic's electoral map brought about by the three parliamentary elections after 2010, we need to outline, at least briefly, the previous nature of Czech electoral geography. This was, in contrast to many Central and Eastern European countries, atypically dominated by a socio-economic cleavage (traditionally dominant in Western democracies), and this was the main long-term determinant of Czech party politics at least until the parliamentary election of 2010. After the fall of the communist regime and reinstatement of free political competition in 1990, Czech politics exhibited a relatively smooth emergence of the left-right axis, which gradually took on a traditional socioeconomic form (a cleavage between owners and the working class, or between the winners and the losers of economic and social transformation, respectively (Hloušek and Kopeček 2008)) and became the main structural cleavage of the Czech political spectrum (Matějů and Řeháková 1997; Hloušek and Kopeček 2008; Chytilekand Eibl 2011). Furthermore, individual-level analyses of the links between different social groups and Czech political parties indicated considerable stability of those links throughout the 1990-2010 time period. At the same time, the effects of social class on voting in the Czech Republic were steadily strengthening (Smith and Matějů 2011), an important finding given the precisely opposite trend observed at the international level. Linek and Lyons's (2013) analysis, covering two decades of Czech party politics (from 1990 to 2010) showed that party choice in the Czech Republic was largely based on three cleavages (social class, religion and generational membership) and on the left-right ideological orientation. This cleavage structure contributed to the stabilization of the classic left-right model of political competition, with the right dominated by the liberal-conservative Civic Democratic Party (ODS) and the left by the Czech Social Democratic Party (ČSSD). Since the elections of 1996, the ODS and the ČSSD combined were regularly receiving more than half of the votes, with a maximum of 67.07% in 2006. Alongside these two large parties two medium-sized ones have also had parliamentary representation since the first democratic election in 1990: the Communist Party of Bohemia and Moravia (KSČM) and the Christian Democratic Union - Czechoslovak People's Party (KDU-ČSL). Voters' support for these four ("traditional") parties exhibited steady growth from 74.47% in 1996 to 87.73% in 2006. In each election, this foursome was supplemented by some smaller parties, none of which, however, was able to maintain parliamentary representation for more than two terms (cf. Hloušek and Kopeček 2008, 526-533).1 EURASIAN GEOGRAPHY AND ECONOMICS (Q) 517 However, the 2010 parliamentary elections interrupted the previous trends of bipolarity and the high degree of stability in the Czech political system. Support for the ODS fell to an historical low (20.22%) and the ČSSD achieved its worst result since 1992 (22.08%). The KDU-ČSL and the Green Party (SZ) lost parliamentary representation, while two new parties - the TOP09 and the VV -reached the 5% electoral threshold, thus obtaining seats in the parliament. Even greater changes were brought about by the 2013 early election. Of the two largest parties, the ČSSD won the elections with the lowest gain of a winning party since 1992 (20.45%) and the ODS, as the former main party of the right, only gained 7.72% of the vote. After three years, the KDU-ČSL regained parliamentary representation, while the W left the Chamber of Deputies. The KSČM and the TOP09 maintained their representation, but two other parties entered the lower house as well - ANO2011 and the Dawn. The 2017 election brought about the final breakdown of the Czech party system in its traditional form (see Balík and Hloušek 2016). The ODS, as the former main party on the center-right, slightly improved its gain to 11.32%, but the ČSSD, as the main leftist party, suffered a dramatic decline to 7.27% (see Appendix A). The combined support for the formerly strongest parties of the left and of the right reached 18.59%, i.e. an almost 50-point drop compared to the elections of 2006. Similarly, the combined support for four parties with long-term parliamentary representation experienced a 55-point drop (from 87.73% to 32.15%) between the elections of 2006 and 2017. In contrast, three new parties entered the parliament. Specifically, the Czech Pirate Party (referred to as "the Pirates") obtained 10.79%, the SPD 10.64%, and the movement of Mayors and Independents (STAN) 5.18% of the vote. To return to determinants of electoral behavior, an analysis of individual data showed that while the ODS was predominantly supported by higher-class right-wing voters, the ČSSD was more successful among left-wing voters of lower social classes; the KSČM mobilized voters especially among blue-collar workers, pensioners and the generation born before 1954, who found themselves to the left of ČSSD voters; and finally, in the case of the KDU-ČSL, religious identity was the primary factor of party support, with all the other factors far behind (Linek and Lyons 2013). Similarly, research conducted at the aggregate level found that socioeconomic factors explained the largest part of inter-regional differences in electoral support, and resulted in the stabilization of the geographical distribution of individual parties' constituencies over time (Kostelecký and Bernard 2014). In short, right-wing parties were more successful in regions that were better off economically, while left-wing parties won more support in regions with structural problems in the economy. In other words, while rightist parties (the ODS, the ODA, the Union of Liberty [US]) were more successful in areas of high development potential (especially the axis connecting Prague with the regional capitals of western Bohemia, Pilsen, and north-eastern Bohemia, Liberec), left- 518 @ P. MAŠKARINEC wing parties (especially the KSČM) were preferred in regions with low development potential. The industrial agglomeration of Ostrava/Karviná/Frýdek-Místek was an exception with much more support for leftist parties (Pink and Voda 2012b; Bernard and Šimon 2014; Bernard, Kostelecký, and Šimon 2014; Maškarinec 2017). More specifically, the territory of highest support for the right-wing leader, ODS, was centered in areas with more self-employed people, college graduates, and white-collar workers with higher salaries, while negative influences on its success included high unemployment or widespread Catholicism (for more on the territorialization of established parties in specific regions, see below). In contrast, the ČSSD and the KSČM mobilized voters primarily in areas with high unemployment or with low populations of the self-employed, college graduates or white-collar workers. The KDU-CSL's patterns of support are relatively easy to explain with regard to their anchoring in the Catholic milieu that was historically spatially bounded, mainly in the area of southern Moravia. The geography of support for the ODA and the US (the parties of the so-called "liberal centre") then largely copied that of the ODS, in terms of better results in units with more college graduates, more self-employed persons, higher purchasing power and lower unemployment, while the determinants of the Greens' support were slightly different. The Green Party won more votes in areas with more college-educated and self-employed residents, but on the other hand, the party had considerably worse electoral gains in areas with stronger Catholic populations, which is possibly associated with the more liberal views of the Greens on a number of traditional issues (Kostelecký 2001; Pink and Voda 2012a; Bernard and Šimon 2014; Kostelelecký and Šimon 2015; Maškarinec 2017). Furthermore, Kostelecký and Šimon, who constructed a typology of Czech municipalities according to political preferences in parliamentary elections between 1992 and 2013, identified four types of municipalities according to prevailing party-ideological orientation (fiscal-conservative, Christian-democratic, social-democratic, communist), which, with a minor exception of the Christian-democratic orientation, all transcend the boundaries between the traditional Czech lands, Bohemia and Moravia (cf. Kostelecký and Šimon 2015). More importantly, this division prevails even when analyzing different types of municipalities separately: once again, the left enjoys higher support in structurally disadvantaged regions, both in small municipalities and in major cities, and support for the left is similarly typical of rural municipalities located in peripheral borderland areas where access is difficult (the Frýdlant Hook in North Bohemia, north-western Moravia, borderland areas between the regions of Vysočina and South Moravia) or in contiguous areas of structurally disadvantaged regions (the North Bohemian Brown Coal Basin and the southern part of the Ústí nad Labem Region in general, other borderland areas between regions without major urban centers and with difficult commuting options - those EURASIAN GEOGRAPHY AND ECONOMICS (Q) 519 between Eastern Bohemia and Central Moravia or between the regions of South Moravia and Zlín) (cf. Bernard 2015). With regard to the socio-historical context, Bernard (2014) presents some hypotheses connecting higher support for the ODS (as well as other smaller right-wing and center-right parties) in some Czech regions (especially in Prague, north-eastern and eastern Bohemia) to the specific distribution of some social groups (employers, self-employed, urban intelligentsia) or regional culture which historically emphasized entrepreneurship and independence. Furthermore, analyses using spatial econometric methods found certain independent effects of contextual variables (spatial regime) on voting decisions, namely the formerly German-inhabited Sudetenland macroregion, on electoral support for some parties in parliamentary elections (Kouba 2007). This spatial regime resulted in considerably higher support for the ODS in central Czech Republic relative to the Czech-German borderland (Sudetenland). That the exact opposite finding applies to the KSČM with its traditional stronghold in Sudentenland (cf. Bernard 2014; Maskarinec 2015) is explained by some authors as a result of the German population's expulsion after WWII and subsequent resettling of this region by new inhabitants (Kouba 2007; Charvát 2014). On the other hand, Bernard (2014) assumes that the influence of Sudetenland is currently due to several other structural factors active in this area, rather than only a separate contextual influence of the Sudetenland variable. This is demonstrated by the findings of Kouba (2008) who showed the importance of contextual explanations of electoral support for the KSČM, when comparing Czech central regions to the Sudetenland. Overall, the weak effect of contextual variables in the Czech party system was explained with reference to the high degree of uniformity of electoral support (nationalization) of Czech parties in different regions (cf. Kouba 2007; Lyons and Linek 2010). This high level of spatiotemporal stability of the territorial patterns of support for established Czech parties raises the interesting question of where new parties won their voters and, therefore, whether the Czech case exhibits, similarly to Western Europe, the weakening effects of the socioeconomic and socio-demographic characteristics of regional populations on their voting behavior. Data and methods We have used several spatial techniques to control for the influence of spatial effects and analyze the dynamics of Czech party competition. First, our exploration of the spatial structure of electoral support begins with the formal detection of spatial autocorrelation using Moran's / statistic (Cliff and Ord 1981). However, Moran's / is an overall measure of linear association, whose single value is valid for the entire study area. Since the aim of this study is to identify potentially different patterns of voting behavior within larger units and their transformation between elections, a local indicator of spatial association (LISA) 520 @ P. MASKARINEC was used to obtain a more detailed insight into the ways electoral support was clustered throughout the Czech Republic's territory. Bearing in mind that our goal is to compare differences in the geographical clustering of party replacement patterns for new and established parties, we will use both univariate LISA indicators, which can show the clustering of support for a party in one election, and bivariate LISA indicators, which allow us to compare the types of party replacement between pairs of elections (Anselin 1995). In the case of bivariate correlation, we present maps of so-called outward diffusion, measuring the ways support for an old party in a given unit in elections t shifts toward a new party in surrounding units in elections t + 1. Thus, we assess the extent to which the value at a location for the x-variable (vote for the old party) is correlated with the weighted average of another variable (y, vote for the new party), namely the mean value of the neighboring locations (Anselin 2005). We operationalize local clustering in accordance with Shin and Agnew (2007), who differentiate between two degrees of spatiotemporal stability and two degrees of spatiotemporal non-stability. Thus, positive spatial autocorrelation (a high level of support for a new party in one unit which is accompanied by a high level of support for an old party in same unit, or a low level of support for a new party in one unit which is accompanied by a low level of support for an old party in same unit) is a sign of spatiotemporal stability, specifically of (1) substitution/switching, as most of the territorial strongholds of the old party have been taken over by the new party, or (2) splitting, as support for the old party has split between several parties. In contrast, spatiotemporal non-stability can be indicated either by (1) colonization, when the new party colonizes areas where the old party was successful some time ago (a low level of support for the new party in one unit which is accompanied by a high level of support for the old party in same unit), or by (2) mobilization, when the new party mobilizes a new segment of voters in areas where the old party was weak (a high level of support for the new party in one unit which is accompanied by a low level of support for the old party in the same unit) (Shin and Agnew 2007). In this paper, we analyze data on Chamber of Deputies elections from 2006 to 2017. The choice of 2006 as the starting point of analysis is based on the fact that it was in that year that the electoral competition between the ODS and the CSSD exhibited the strongest tendencies to bipolarity, and the concentration of voters with these two parties can serve as a good dividing line between the spatial distribution of leftist and rightist voters. Our primary indicator of political preferences is the percentage of votes achieved by the parties at the level of the Czech Republic's 205 microregions (administrative districts of municipalities with extended powers and the capital city of Prague).2 Since these microregions can be considered to be relatively natural regional units, they enable one to appropriately explore detailed geographic patterns of voting behavior in the Czech Republic. EURASIAN GEOGRAPHY AND ECONOMICS (Q) 521 Spatial analysis of party replacement in the Czech Republic, 2010-2017 With regard to the content and structure of this section, we will begin with a formal description of the global (i.e. national) spatial clustering of party support. Second, we will introduce patterns of local spatial clustering of support for established parties and their possible transformation. The third and fourth parts of the section will focus on local spatial clustering of support for new Czech populist parties and patterns of party replacement, and the concluding part will discuss whether the success of new parties resulted in a considerable transformation of voting patterns in the Czech Republic. Global spatial clustering of electoral support Although the most recent elections in 2017 strongly destabilized Czech party politics at the systemic level by changing the relative strength of the different parties, there was, at the same time, no strong transformation of existing patterns of global spatial clustering of electoral support (Appendix B). The nationalization of (both new and established) parties' spatial patterns of electoral support remained relatively high (Appendix C), and more importantly, the global spatial clustering of electoral support did not differ considerably between new and established parties (Appendix D). As we stated in the introduction, there is a strong similarity between the geographical support for the ODS, as the former strongest right-wing party, and other center-right or right-wing parties, i.e. parties of the so-called liberal center or the fiscal-conservative party family. However, there was no similar pattern for other new populist parties (Table 1). More importantly, there was much transformation of inter-party spatial relations between elections, which can be best demonstrated by the case of the ANO2011 movement and both parties of Mr. Okamura (the Dawn and the SPD). In the elections of 2013, regional support for the Dawn was only loosely autocorrelated with blocs of leftist or center-right parties, whereas the opposite was true for the following elections of 2017, when the spatial clustering of SPD support was positively autocorrelated with the patterns of voting for leftist parties and, vice versa, negatively with the right-wing ODS. Finally, the spatial distribution of support for the ANO2011 indicated some form of spatial dependence in 2013, but was very different in the 2017 elections. Table 1. Bivariate Moran's / scores for new and established parties, 2006-2017. CSSD (2006) ODS (2006) KSCM (2006) KDU-CSL (2006) TOP09 (2010) VV (2010) -0.079 0.206 -0.136 -0.261 0.078 ANO2011 (2017) 0.311 -0.089 0.230 -0.348 -0.294 ANO2011 (2013) -0.213 0.371 -0.046 -0.435 0.182 Dawn (2013) 0.065 -0.120 -0.064 0.120 -0.130 SPD (2017) 0.393 -0.353 0.215 0.018 -0.449 522 @ P. MAŠKARINEC In 2013, regional support for the ANO2011 was positively associated with previous strongholds of the ODS and negatively with the ČSSD, whereas this pattern completely changed in 2017. The former positive association between the ANO2011 and the ODS disappeared, and was replaced by positive autocorrelation of clustering with both leftist parties, showing a very similar pattern of clustering as in the case of the SPD, and suggesting that the rise of the ANO2011 and the SPD in 2017 was based on the mobilization of voters in regions formerly dominated by leftist parties. Transformation of local spatial clustering of electoral support for established parties The loss of almost 12 percentage points between the elections of 2006 and 2013 did not result in any major transformation in the spatial patterns of ČSSD electoral support (Figure 1), as clusters of high support for the party remained in central and northern Moravia (especially in the industrial agglomeration of Ostrava/Karviná/Frýdek-Místek) and clusters of low support remained concentrated in the Prague area and extended through central Bohemia to northeastern and eastern Bohemia. However, another 13-point drop between the elections of 2013 and 2017 brought about change. The trend of declining support for the ČSSD was accompanied by a gradual growth of homogeneity in the entire area of low support. More importantly, the party suffered a significant drop in a large part of one of its traditional core constituencies in northern Moravia, as well as in contiguous areas of north-western Bohemia and along the boundary between central and western Bohemia. The remaining area of highest support lies predominantly in Southern Moravia (historically the second of the strongholds of ČSSD support): a contiguous strip of high-support units starts in the northern part of the South Moravian Region, continues to the west to cover a large part of the Vysočina Region and ends on the boundary of South Bohemia. Figure 1. Moran's / cluster maps of the distribution of support for the ČSSD, 2006/2017. EURASIAN GEOGRAPHY AND ECONOMICS (Q) 523 Figure 2. Moran's / cluster maps of the distribution of support for the ODS, 2006/2017. The main party of the right wing, the ODS, was also affected by changes in territorial support (Figure 2), although in 2013 its vote share had already dropped to a level threatening its parliamentary representation. Until the elections of 2006, the main stronghold of right-wing voters was the Prague metropolitan area and contiguous parts of central Bohemia, and from there it stretched to a number of contiguous territorial units in north-eastern and eastern Bohemia. However, by 2010, large parts of north-eastern and eastern Bohemia were lost, and this trend continued in the following elections, too. By 2013, the Prague-centered clusters of units of high support for the ODS no longer continued to the north-east and east, but mostly shifted to western Bohemia, with a small overlap to southern Bohemia. Finally, the outcomes of the 2017 election meant another transformation of party support. The larger Prague metropolitan area and related areas of central Bohemia remained the only major compact territory where the ODS retained its former high level of support, while the second largest territory where the party enjoyed major support shifted from western to southern Bohemia. In contrast, most regional clusters of low support were concentrated in Moravia throughout this period (aside from a few units in north-western Bohemia). Subject to some simplification, the distribution of high electoral support for the ODS indicates the continuity of its primarily "Bohemian" identity; as an exception, the second largest city of the country, Brno, and its immediate surroundings were the only areas in Moravia where the ODS was successful in the long term. Overall, despite these changes, the ODS retained its urban voter appeal, as evidenced by its high vote share in cities such as Prague, Brno, Pilsen of western Bohemia, or České Budějovice of southern Bohemia. In contrast to the formerly strongest Czech parties, the strongholds of communist support changed only in a limited way (Figure 3). Although the KSČM lost more than 5 percentage points in the elections of 2017 against the mean of three previous elections, the clusters of high and low communist support exhibited high spatial stability. The first cluster of high KSČM support was 524 @ P. MAŠKARINEC Figure 3. Moran's / cluster maps of the distribution of support for the KSČM, 2006/2017. centered in north-western Bohemia and extended into parts of central Bohemia, the second cluster was in southern Moravia and the third one in parts of northern Moravia. Generally, the strongholds of KSČM support were concentrated in impoverished peripheries, and were mostly found in (1) the country's borderland that had been re-populated by new settlers after WWII, the formerly German-inhabited Sudetenland (Kouba 2007; Bernard and Šimon 2014; Maškarinec 2017; Novotný 2017), or (2) inner peripheries, areas on the edge of the metropolitan areas and regional centers with relatively low levels of educational attainment, high numbers of commuters, and low proportions of tertiary sector employment (Bernard 2015; Bernard and Šimon 2017). However, for the first time since democratization, there was a significant decrease in the spatial clustering of units with the highest communist support in north-western Bohemia and, at the same time, an increase in the compactness of the contiguous area of low support in central Bohemia between Prague and north-eastern Bohemia. Finally, despite the profound transformation of Czech party politics after 2010, the regionalization of electoral support for the KDU-ČSL remained almost intact (Figure 4). The core areas of KDU-ČSL support were concentrated almost exclusively in southern and south-eastern Moravia (plus small areas in northern Moravia), while the party's electoral success in Bohemia was limited to a minor part of eastern Bohemia. Voting for KDU-ČSL was strongly correlated with the distribution of the Catholic population throughout the country and it was especially weak in all of north-western Bohemia and in parts of north-eastern Bohemia (i.e. Sudetenland). Local spatial clustering of electoral support for the ANO2011 The success of new parties, which can be denoted as right-wing or center-right parties (the TOP09, Pirates, the STAN), did not really transform the geographical patterns of voters' behavior in the bloc of non-left-wing parties, but rather EURASIAN GEOGRAPHY AND ECONOMICS (Q) 525 Figure 4. Moran's / cluster maps of the distribution of support for the KDU-ČSL, 2006/2017. divided these voters into several blocs looking for a new alternative to the ODS as the former main party of the right wing. However, this conclusion is far from reality in the case of the ANO2011 and both of Okamura's parties. Let us demonstrate this first by analyzing the electoral base of the currently strongest Czech party, the ANO2011 of Andrej Babiš, one of the country's wealthiest entrepreneurs. In 2013, when the ANO2011 first entered the electoral race, its voter support was concentrated almost exclusively in Bohemia (there were only a few units with high party support in central Moravia). However, in contrast to established parties, the territory of high ANO2011 support comprised a mixture of areas inclined to the right (central, north-eastern and eastern Bohemia) and to the left (north-western Bohemia), although to a lesser extent in the latter case (Figure 5). At the same time, the amount of spatial clustering suggested predominant vote switching from the ODS and partly from the TOP09 as the prevailing type of party replacement.3 Here, again, we can see relatively large clusters of high ANO2011 support in regions with high support for the ODS in 2006, namely areas where fiscal-conservative parties have historically been successful (Figure 6). Furthermore, the relatively low positive values of global Moran's / statistics between the Figure 5. Inter-regional support for the ANO2011, the years 2013/2017 - quintiles (%). 526 @ P. MAŠKARINEC Figure 6. Bivariate Moran's / cluster maps of the distribution of support for the ODS (2006) with the ANO2011 (2013/2017). ANO2011 and the ODS (as well as the TOP09 as another member of this party family) can be attributed to two factors: (1) the lower support for the ANO2011 in some of the largest urban areas where the Czech right has traditionally enjoyed strong support (Prague, Brno) and (2) ANO2011's high vote share in north-western Bohemia, with its traditionally left-wing voters. The latter case is also supported by bivariate LISA maps of voter support for the ANO2011 and leftist parties. We see some overlaps of high support between the ANO2011 and both the ČSSD and the KSČM in north-western Bohemia (Figures 7 and 8), although they do not fully cover areas with traditionally strong leftist electorates (e.g. the western part of the Central Bohemian region or the Frýdlant Hook in north-eastern Bohemia with high support for the Communists). Similarly, the negative autocorrelation between the ANO2011 and leftist parties (especially the KSČM) was also due to weak ANO2011 support in northern Moravia, more importantly in the Moravian-Silesian region (the traditional ČSSD stronghold), but also in some peripheral (borderland) areas of south Moravia with their traditional high concentrations of KSČM voters. Figure 7. Bivariate Moran's / cluster maps of the distribution of support for the ČSSD (2006) with the ANO2011 (2013/2017). EURASIAN GEOGRAPHY AND ECONOMICS (Q) 527 Figure 8. Bivariate Moran's / cluster maps of the distribution of support for the KSČM (2006) with theANO2011 (2013/2017). In contrast to the new right-wing parties, the ANO2011 was one of two parties (together with the Dawn) for which examination of the bivariate LISA maps revealed some clusters of spatiotemporal stability with the VV, which enjoyed an unprecedented (albeit only temporary) electoral success in 2010. However, in the case of the VV we cannot speak about a switching of its electorate to new parties (the ANO2011 and the Dawn) because the dispersion of spatial clustering of territorial support for each of these three parties was quite different, and overlaps only occurred in some parts of the Czech Republic. However, the 2017 electoral contest saw not only the rise of the ANO2011 as the strongest Czech party, but also a profound transformation of the Czech party system's format. The ANO2011 emerged from the elections as the only large party, and party competition shifted far away from the bipolarity which was most present in the elections of 2006, when the strongest parties, the left-wing ČSSD and the right-wing ODS, received more than 30% of the vote each (cf. Balík and Hloušek 2016). But while the ANO2011 appealed especially to voters in mostly rightist regions in the elections of 2013 (some areas in north-western Bohemia and small areas in Central Moravia were the only ones where the party mobilized some segments of traditionally non-rightist voters), the opposite was true in 2017. The party's rise continued predominantly at the expense of decreasing support for (i.e. splitting from) both leftist parties, especially the ČSSD as the ANO2011's main coalition partner in government. In contrast to 2013, the core areas of ANO2011 support in 2017 concentrated especially in the borderland units of an almost contiguous area that included north-western Bohemia and northern and north-eastern Moravia (interrupted only in north-eastern Bohemia), with a considerable drop of voter support in eastern and central Bohemia. The transformation of the ANO 2011 electorate was reflected in a shift to positive autocorrelation between support for the ANO2011 and the ČSSD (high support for both parties in northern and north-eastern Moravia and low support in central and north-eastern Bohemia), as well as in spatial clustering between 528 @ P. MAŠKARINEC the ANO2011 and the KSČM (much more homogeneous areas of high support for both parties in north-western Bohemia and parts of northern and southern Moravia, and low support in central and north-eastern Bohemia). The ANO2011 movement experienced a significant decrease of electoral gains in all of the largest cities except Ústí and Labem (the regional capital in north-western Bohemia) and the industrial agglomeration of Ostrava/Karviná/Frýdek-Místek. The above-mentioned transformation of geographical patterns of ANO2011 support between the elections of 2013 and 2017 was also reflected in the relation between its electorate and that of the W in 2010. In 2013, the ANO2011 made inroads into former W strongholds, especially those in northeastern and eastern Bohemia (only much less so in central or north-western Bohemia). In 2017, though, this compact area almost completely disappeared and ANO2011 support moved to north-western Bohemia and the Moravian-Silesian region (i.e. leftist regions), with much more continuity in areas of the lowest support, especially in South Moravia. Spatial clustering of electoral support for the VV, the Dawn and the SPD Geographical support for the ANO2011 was, in a sense, analogous to that of traditional parties (especially in the elections of 2017), although without spatio-temporal stability between elections. However, this is not valid for the VV and the Dawn, or the SPD, respectively. The clusters of high or low support for the W were not largely limited to either Bohemia or Moravia (Figure 9). Like the communists, the W, too, obtained many votes from Sudetenland, but unlike them, it was more successful in the traditionally rather right-wing regions of north-eastern and eastern Bohemia, as well as, to some extent, in the rather left-wing areas of northern Moravia. In contrast, the KSČM won more support in north-western Bohemia, where the VV was considerably weaker. Furthermore, low W support was distributed between several (largely southern) regions both in Bohemia and in Moravia. We thus can conclude that Figure 9. Inter-regional support for the W, 2010 - quintiles (%). EURASIAN GEOGRAPHY AND ECONOMICS (Q) 529 Figure 10. Bivariate Moran's / cluster maps of the distribution of support for the ČSSD and ODS (2006) with the VV (2010). support for the VV was relatively weakly rooted in geography as its constituency was relatively indistinct. This conclusion is also confirmed in spatial clustering between the VV and established parties, as indicated by the values of bivariate Moran's / statistic. We found only very limited congruence between votes for the W and leftist parties, but the same was true for the KDU-ČSL (Figure 10).4 There were only a limited number of clusters of low support for the W with the ČSSD, the KSČM and the KDU-ČSL. Similarly, clusters of high support between the W and each of these parties were limited to parts of northern Moravia, together with some (not clustered) units in north-western Bohemia. A much stronger spatial clustering existed between the former strongholds of the rightist ODS, and also the Greens (almost completely mimicing the pattern found for the ODS) in 2006 and for the W in 2010. Here, the VV made some significant inroads into the area of north-eastern and eastern Bohemia which was traditionally the strong base of center-right voters, i.e. areas with a historically specific regional culture emphasizing entrepreneurship and independence.5 With regard to party replacement, the W was another party whose support can be associated with vote splitting from the ODS, but also from the Greens, who lost parliamentary representation in 2010. However, a small part of the Ws votes were received from leftist parties (especially from northern Moravia). Therefore, the success of the W arose primarily from splitting, with a large proportion of votes received from center-right voters together with the mobilization (to a much lesser extent) of leftist voters, with a significant inability to colonize rightist voters in the urban areas of two biggest Czech cities. However, as the VV did not stand in the elections of 2013, we can also analyze whether there was any party which received the lion's share of votes from the Ws former electorate.6 Here, the scores of bivariate Moran's / (Table 2) suggest a high amount of spatiotemporal instability between support for the VV and other parties in 2013. In most cases, we found relatively low negative values of bivariate LISA indicators, especially for the ODS and the KSČM, while stronger 530 @ P. MAŠKARINEC Table 2. Bivariate Moran's / scores for the VV and other parties, 2010-2017. ČSSD (2013) ODS (2013) KSČM (2013) KDU-ČSL (2013) VV (2010) -0.143 -0.039 -0.032 -0.228 ČSSD (2017) ODS (2017) KSČM (2017) KDU-ČSL (2017) VV (2010) -0.242 -0.043 -0.135 -0.215 ANO2011(2013) ANO2011(2017) Dawn (2013) SPD (2017) VV (2010) 0.270 0.222 0.190 0.114 TOP09 (2013) TOP09 (2017) STAN (2017) Pirates (2017) VV (2010) 0.097 0.023 0.068 0.070 positive associations existed only between geographical support for the W with the ANO2011 and the Dawn; very similar values were also found in the 2017 elections. This conclusion is also supported by a comparison of geographical clustering. We found only a very limited number of units with positive autocorrelation between voting for the W in 2010 and all four traditional parties in 2013. The rightist ODS was no exception and the low value of bivariate Moran's / (in a negative direction) also had its geographic dimension. Whereas the rise of the VV in 2010 was based on the mobilization of a proportion of ODS voters in north-eastern and eastern Bohemia, the 2013 elections did not bring these voters back to the ODS. The ODS received strong support only in some units in central Bohemia. The party was neither able to appeal to voters in large parts of north-eastern and eastern Bohemia who moved to the W in 2010, nor to mobilize former W voters in northern Moravia. From the perspective of party replacement, we can conclude that the electorate of the W was, to a large extent, divided between the ANO2011 and the Dawn in the elections of 2013 (Figure 11). This is particularly evident in some parts of north-eastern and eastern Bohemia where we find vote splitting from the VV to the ANO2011 and the Dawn. However, the situation was different in other parts of the country: in the borderland between eastern Bohemia and Moravia, the ANO2011 no longer received the highest support but the opposite was true for the Dawn; and electoral support for the VV in 2010 reached high Figure 11. Bivariate Moran's / cluster maps of the distribution of support for the W (2010) with the ANO2011 and the Dawn (2013). EURASIAN GEOGRAPHY AND ECONOMICS (Q) 531 values especially in Sudetenland, starting from the western part of northwestern Bohemia and continuing through north-eastern and eastern Bohemia, northern Moravia, and ending in north-eastern Moravia, on the border with Slovakia. In short, while the ANO2011 took the former VV votes in large areas starting from north-western Bohemia, a major part of north-eastern Bohemia, and small parts of central and eastern Bohemia (inland), the Dawn mobilized parts of the W electorate in eastern Bohemia (borderland) and a large part of Moravia (especially northern and south-eastern). When examining the relationship between support for the W with the ANO2011 and the SPD in 2017 (Figure 12) some very interesting findings emerge. They confirm a significant transformation of the ANO2011's electoral base as clusters between votes for the W and the ANO2011 moved to northwestern Bohemia and northern and north-western Moravia, i.e. almost exclusively leftist regions. One notable exception was the central Bohemian Mladá Boleslav region (an important center of automotive manufacturing and home to Škoda Auto's passenger car production facilities), i.e. one with a high development potential, where support for the ANO2011 grew strongly in 2017. In contrast to this, the spatiotemporal stability between votes for the W in 2010 and the Dawn in 2013, or the SPD in 2017, was much stronger. Moreover, alongside north-western Bohemia, the core strongholds of the SPD support remained, especially in northern Moravia. Furthermore, much more support was received by the SPD in the industrial agglomeration of Ostrava/Karviná/Frýdek-Místek, which further expanded the contiguous area of high electoral support for the party in north-eastern Moravia as well as large parts of central and southern Moravia, while areas of strong party support in north-eastern and especially eastern Bohemia were completely lost. Finally, a closer look at the territorialization of the two parties founded by Tomio Okamura showed that, unlike the ANO2011, support for the Dawn was not as clearly divided between Bohemia and Moravia (Figure 13). The Dawn movement obtained many votes in south-eastern Moravia, the same region Figure 12. Bivariate Moran's / cluster maps of the distribution of support for the W (2010) with the ANO2011 and the SPD (2017). 532 @ P. MAŠKARINEC Figure 13. Inter-regional support for the Dawn (2013) and the SPD (2017) - quintiles (%). where its chairman, Tomio Okamura, had been elected Senator in 2012 (cf. Maškarinec and Bláha 2014). Other clusters of high support for the Dawn were located in peripheral areas of north-western Moravia and, to a lesser extent, eastern Bohemia, or north-western Bohemia, although in the latter area, units with a considerable vote share for the Dawn were not clustered continuously. In contrast, the Dawn obtained very little support in a large cluster of units extending from central Bohemia (including Prague) to the country's western border, as well as in some parts of the Bohemian-Moravian borderland and central Moravia (including the city of Brno). However, with regard to patterns of party replacement, we see only a few clusters with stronger mobilization of voters in regions with predominantly centrist or rightist orientations (Figure 14). Here, comparison of the results for the Dawn in 2013 with votes for the ODS, or other right-wing parties, found an almost identical pattern, with a small contiguous area of high support in the eastern Bohemian borderland and then only in some separate Moravian microregions.7 On the other hand, there was much more spatial clustering between the Dawn and leftist parties as well as with the KDU-ČSL. Areas of high support for Figure 14. Bivariate Moran's / cluster maps of the distribution of support for the ODS (2006) with the Dawn (2013) and the SPD (2017). EURASIAN GEOGRAPHY AND ECONOMICS (Q) 533 Figure 15. Bivariate Moran's / cluster maps of the distribution of support for the ČSSD and the KSČM (2006) with the Dawn (2013). both the KDU-ČSL and the Dawn were restricted to the above-mentioned Moravian areas where party chairman Okamura enjoyed strong support due to his senatorial mandate, while the Dawn's vote share was very weak in other Moravian strongholds of the KDU-ČSL (e.g. central Moravia or the Vysočina Region). However, a different picture is offered by bivariate cluster maps of the vote distribution between the Dawn and leftist parties (Figure 15). Here, the Dawn was an important exception among new Czech parties, as its success was based on the mobilization of voters in the leftist region of northern Moravia, but especially in its north-western part (i.e. the rural peripheral borderland areas of former Sudetenland with their traditionally high support for the KSČM, rather than the ČSSD), compared to a much more moderate success of the Dawn in the north-eastern industrial agglomeration of Ostrava/Karviná/Frýdek-Místek (a traditional stronghold of the ČSSD), as well as parts of central Moravia and southeastern Moravia. Comparing the two new populist parties from the 2013 electoral contest, the success of the ANO2011 was, in some respects, based on the mobilization of rightist voters, while this is not the case with the Dawn and even less so with the SPD in 2017. As we can see in Figure 14, negative spatial autocorrelation was the prevailing form of spatial clustering between the ODS and the SPD (the same was true for clustering between the SPD and the TOP09 or the Greens), with a strong increase of such units in 2017 and only unique occurrences of positive spatial autocorrelation. Generally, territorial support for both the Dawn and, especially, the SPD confirmed its prevailing (increasing) Moravian identity, with limited additional support in some parts of north-western Bohemia. The support which the Dawn enjoyed in territories with a traditionally predominant leftist electorate was even more strengthened in the elections of 2017 (Figure 16). Here, we can see much more homogenization of areas with both strong and weak support for the SPD and leftist parties. In the case of the ČSSD, the SPD mobilized its voters in almost the whole of Moravia. Compared to the previous election, the most important change was the expansion of high SPD 534 @ P. MAŠKARINEC Figure 16. Bivariate Moran's / cluster maps of the distribution of support for the ČSSD and the KSČM (2006) with the SPD (2017). support to areas traditionally dominated by the ČSSD, especially to the northeastern industrial agglomeration of Ostrava/Karviná/Frýdek-Místek but also to parts of central Moravia. This was further supplemented by the strong success of the SPD in southern Moravia and in rural peripheral borderland areas of southeastern Moravia, both traditional strongholds of the KSČM (especially in the case of the latter). Toward a new geography of Czech voting? So, did the voting patterns change substantially in the last three Czech elections? Here, the answer must be divided with regard to two groups of parties. On one hand, there was strong evidence of global spatial clustering of electoral support between some of the new parties (the TOP09, the Pirates, the STAN) and the ODS, as the former main party of the right, which also had its local dimension, with strong evidence of spatiotemporal stability. With regard to party replacement, we can speak about vote switching from the ODS to other center-right parties, as the success of these parties largely stemmed from the prior constituencies of the ODS, but possibly also from the strongholds of minor right-wing parties of the 1990s and the early 2000s. In this context, these new parties brought about only marginal changes to previous voting patterns as they were also more successful in areas of high development potential (especially the axis connecting Prague with the regional capitals of western Bohemia, Pilsen, and north-eastern Bohemia, Liberec), and, except for the STAN, in major urban centers (Prague, Pilsen, České Budějovice, Liberec, Hradec Králové, and Pardubice in Bohemia, or Brno, Olomouc, and Opava in Moravia). For that reason, we focused primarily on two types of genuinely new populist parties, those which exemplify projects of "newness" (the W, the ANO2011) and "prolocutor" parties (the Dawn, the SPD), and found some novel inter-regional patterns of Czech voting behavior. First, the success of the W in 2010 was not EURASIAN GEOGRAPHY AND ECONOMICS (Q) 535 geographically rooted as in the case of established parties or the TOP09, the Pirates, or the STAN. The W mobilized voters in both leftist and rightist regions of northern Moravia and especially north-eastern and eastern Bohemia - in areas with or without development potential. The rise of the W can be described as a mix of vote splitting from center-right parties with mobilization of voters where those parties were traditionally rather weak. Second, the transformation of electoral politics continued in the 2013 elections, when both the ANO2011 and the Dawn mobilized voters in some previously "atypical" areas and, at the same time, became the main recipients of former W votes. The ANO2011 succeeded in large areas starting from northwestern Bohemia and including also a major part of north-eastern Bohemia and small parts of central and (inland) eastern Bohemia (where there was also spatial clustering with the TOP09), while the Dawn mobilized former W voters in (borderland) eastern Bohemia and a large part of Moravia (especially northern and south-eastern). In other words, while the ANO2011 primarily recruited the W's center-right electorate, the Dawn appealed to its former voters in peripheral (often rural and borderland) areas. Third, the geography of support for the Dawn was unique in yet another way, as it was able to obtain votes in the typically left-wing regions of northern, southern and south-eastern Moravia. However, with the rare exception of eastern Moravia, its results were very weak in the country's largest cities, suggesting that support for the movement was rooted in the context of small towns and rural (often peripheral) areas. Its predominantly Moravian-rooted support was confirmed by the result of the SPD in 2017, combined with the high level of spatiotemporal stability of regional support between the Dawn and the SPD. Moreover, after the disappearance of its support in eastern Bohemia, the movement appealed almost exclusively to leftist voters. Indeed, it connected Moravian areas with high support for the ČSSD (the industrial agglomeration of Ostrava/Karviná/Frýdek-Místek) and the KSČM (rural peripheral borderland areas of northern as well as south-eastern Moravia), with those where the KDU-ČSL was strong, together with the similarly leftist north-western Bohemia. Thus, the rise of the Dawn and especially the SPD was based on vote splitting from leftist parties, but also on the mobilization of new segment of voters because the SPD also succeeded in Moravian areas where each of these parties was rather weak. Finally, of all the new parties, geographical support for the biggest one, the ANO2011, showed a considerable amount of spatiotemporal instability between elections. The predominant form in which the party won its spatial support was splitting from either the center-right and rightist (2013) or the leftist (2017) electorates. Our spatial analysis reveals that the rise of the ANO2011 in 2013 was triggered by its takeover of especially right-wing voters in central, north-eastern and eastern Bohemia (all, except central Bohemia, strongholds of the W in 2010) but also by mobilization of a new segment of voters in the leftist north-western Bohemia. On 536 @ P. MAŠKARINEC the other hand, in 2017, the core areas of party strength moved almost exclusively to the leftist regions of north-western Bohemia and northern and north-eastern Moravia. From the spatial perspective, the ANO2011 occupied most of the territories traditionally dominated by leftist voters. At the same time, the geographical patterns of ANO2011 and SPD support largely overlapped with underdeveloped areas. For instance, Havlík and Voda (2016) classified the ANO2011 as the main center-right actor after the 2013 election, and similarly, Kopeček (2016) stated that by entering the government in 2013, the movement effectively abandoned its anti-political appeal toward center-right voters. Furthermore, Havlík and Voda explained the success of the VV in 2010 and the ANO2011 in 2013 (as examples of the so-called centrist populist parties) as a shift from class voting (i.e. the left-right division) to non-ideological voting and perceptions of politics, bringing forward the competency of centrist populist parties to solve the most important issues alongside general dissatisfaction with established parties (Havlík and Voda 2018). Cabada and Tomšič (2016) speak about the rise of person-based politics. Other authors then found that support for new parties in the elections of 2010 and 2013 (except the TOP09) was relatively weakly associated with traditional structural factors, including the traditional socioeconomic cleavage as the main cleavage in Czech electoral competition (Linek 2012b; Maškarinec and Bláha 2014; Havlík and Voda 2016; Maškarinec 2017). In contrast, our spatial analysis reveals that the geographies of ANO2011 support significantly changed between the elections of 2013 and 2017 and the party's success in 2017 was based almost exclusively on inroads into leftist regions. In this context, spatial clustering of support for most new parties in the elections of 2017 suggests a possible renewal of the left-right division of Czech politics, even if the different parties on both sides of this divide considerably changed. However, as the programmatic of most of the new parties (all except the TOP09) are relatively unclear (especially on the left-right ideological division), one can hardly speak of class realignment, but rather of spatial re-stratification of Czech party politics: traditionally leftist and rightist voters largely abandoned "their" parties but, on the other hand, their voting mostly did not contravene the Czech spatial division between leftist and rightist regions. Here, some authors emphasized both the role of populism in the 2017 electoral discourse (Krčal and Naxera 2018) and anti-corruption rhetoric as an election success strategy against established parties (Naxera 2018) in an environment of chronic inefficiency and instability of Czech cabinets (Brunclik and Kubát 2016). Overall, we confirmed that political change in the Czech Republic continues to be geographically determined. One might be inclined to consider the existence of some socio-geographic processes underlying party replacement. However, from the perspective of electoral geography, the change was not as dramatic as one might infer from national-level results. At the same time, the clustering of party support in the 2017 elections also had its socio-historical EURASIAN GEOGRAPHY AND ECONOMICS (Q) 537 overlap. The first success of ANO2011 in the year 2013 (similar to that of the W in 2010) was largely based on its appeal to center-right voters in regions with the specific distribution of some social groups (employers, self-employed, urban intelligentsia) or regional culture which historically emphasized entrepreneur-ship and independence. In contrast, the strengthened success of the party of one of the wealthiest Czech entrepreneurs was associated with its appeal to voters in almost exclusively leftist areas found in structurally disadvantaged regions, often borderland (or inland) peripheries with difficult access to urban centers and commuting. It was a similar peripheral space in which support was mobilized for the Dawn in 2013, and even more so for the SPD in 2017, whereas the success of other new parties (the TOP09, the Pirates, the STAN) was again associated with the above-mentioned area of historically specific regional culture, with entrepreneurship and independence at the fore. Generally, while new parties which can be classified as projects of "newness" or "prolocutors" received most votes from established leftist parties (especially in 2017), the other new parties - "purifiers" (i.e. smaller center-right parties) succeeded at the expense of the liberal-conservative ODS, as the former main party of the right, or of former strongholds of "liberal centre" parties.8 In this regard, the new parties and their spatial support do not represent a substantial innovation, although their success in specific regions does motivate further research and the need to analyze possible long-term contextual effects on electoral behavior. This, however, is not necessarily linked to specific parties only, but rather to the different party-ideological families, as suggested by Kostelecky and Simon (2015). Conclusion As electoral processes are geographical in nature, we have presented a spatial analysis of recent trends in Czech parliamentary elections between 2006 and 2017, shedding light on the geographical patterns of voting behavior. The Czech Republic was a promising case for testing spatial patterns of party replacement as its party system had been relatively stable and closed during the first two post-communist decades. However, a series of three electoral earthquakes completely transformed the shape of the Czech party system, and the rise of new parties accompanied by the emergence and growing influence of centrist technocratic populism ends, according to Havlik (2019), the era of the Czech Republic's "exceptionalism" in the Central European region with regard to its resistance to populist illiberal challenges (see Bustikova and Guasti 2019). However, despite the profound changes that the Czech party system has undergone in the last three elections, which significantly changed both the number and the power of individual political parties, the patterns of spatial clustering of support for new Czech parties, as well as the patterns of party 538 @ P. MASKARINEC replacement between the fading and rising parties, generally did not confirm such a level of transformation of geographical patterning of voter support. This suggests that vote splitting from established parties of the right (the ODS) and left (the CSSD, the KSCM) was the main type of geography of party replacement. Parties whose electoral performance exhibits some atypical spatial clustering (i.e. mobilization of new segments of voters or colonization of areas where established parties were weak) collapsed soon after their electoral breakthrough (the VV) or changed their electoral appeal (the ANO2011 and, to a much lesser extent, the Okamura movements). The apparent spatiotemporal stability of the rightist and center-right electorate was only partly disturbed by the success of the VV in 2010 and the ANO2011 in 2013. A working hypothesis can be formulated that the Czech right-wing voter has neither disappeared nor changed preferences, but has increasingly struggled to find an alternative to the ODS, once the major rightist Czech party. At the same time, the spatial distribution of support for the rightist and center-right parties (established as well as new ones) overlaps with areas with a high development potential, i.e. core territories and the developmental axes related to them. In contrast, the main strongholds for leftist parties and the remaining new parties (the Dawn, the SPD, the ANO2011 - the latter only in the 2017 elections) are in areas of low development potential or structurally disadvantaged regions, often borderland (or inland) peripheries. We can then conclude that the socio-economic composition of districts in the Czech Republic further plays an important role in whether or not parties are able to attract voters. The empirical results presented in this paper also resonate with previous findings that the process of political change is geographically dependent, and the process of party replacement is not a stationary one (Shin and Agnew 2007). This was best demonstrated by the case of the strongest current Czech party, the ANO2011, which took a different path in the elections of 2013 and 2017, aligning with both the center-right and rightist (2013) and the leftist (2017) electorates. It is precisely this example that illustrates the need to consider the geographical patterns of party replacement. Although most of the new Czech parties enjoyed similar levels of electoral support throughout the country (i.e. a high level of nationalization), political realignment is a highly localized sub-national phenomenon in the Czech case as well as in different contexts (Nardulli 1995; Kim, Elliot, and Wang 2003; Darmofal 2008). Furthermore, the geographical concentration of partisan voting on the Czech regional level confirmed the general trend of increasing polarization of the electorate (Jennings and Stoker 2016; Johnston, Manley, and Jones 2016), which has its own spatial dimension. Also, in the Czech Republic, we can speak about a division between voter preferences in the more developed areas versus structurally disadvantaged regions. More importantly, in the Czech case, this division is not limited to parliamentary elections, as the EURASIAN GEOGRAPHY AND ECONOMICS (Q) 539 spatial polarization of the Czech party system described above was also reflected in the presidential election where the electoral contest was spatially divided between two blocs (subject to some simplification, see Maskarinec 2013; Lysek and Lebeda 2018) that largely corresponded to spatial support for the political parties presented in this paper. Some authors (Hlousek and Kopecek 2008) when considering the emergence of party systems in post-communist countries identify a socioeconomic cleavage between the winners and losers of the economic and social transformation, and our conclusion suggests that this division has a tendency to rather increase and further polarize the country's electorate. To conclude, the findings of this study may also be relevant for scholars of other post-communist countries of Central and Eastern Europe or Asia as well as of Western European countries. The increasing instability of party systems and the rise of populism in the West, which considerably transformed some long-term "consolidated" party systems, suggest that Western democracies are becoming more similar to the systems of consolidating democracies in Central and Eastern Europe. The results from the Czech case illustrate that spatial insights on party replacement and the geographical dynamics of electoral change may reveal that profound changes of the party-political spectrum (aggregated at the national level) may obscure a high level of permanence of socio-historic patterns of voter behavior, especially when looking beyond individual parties as unitary actors and focusing rather on the different party families. Notes 1. These generally smaller center-right political parties of liberal orientation, sometimes referred to as "the liberal centre" (Hanley 2012) - the Civic Democratic Alliance (ODA); the Union of Liberty (US); and the Green Party (SZ) - were supplemented by the radical right-wing populist Association for the Republic - Republican Party of Czechoslovakia (SPR-RSC) in the 1990s. For a detailed discussion of the formation of new parties in the Czech Republic, see Hanley (2012). 2. The data were obtained from the Czech Statistical Office's Election Server, available at: http://www.volby.cz/, accessed 24 November 2017. 3. The bivariate Moran's / cluster map of distribution of support for the TOP09 in 2010 with the ANO2011 in 2013 was almost identical as in the case of the ANO2011 and the ODS, and the same conclusion is also valid for the overlap of voting patterns between the ANO2011 and the Greens in 2006, which once again testifies to the ANO2011's appeal to a considerable number of liberal rightist voters. 4. We present only a bivariate cluster map of the distribution of support for the CSSD and the W, as the KSCM/KDU-CSL map has an almost identical pattern, especially in the former case. 5. The importance of these areas for the fate of new parties is further evidenced by the later success of the TOP09, Pirates and the STAN (see Appendix 5). It was formerly precisely the Greens' successful appeal to voters in this area (alongside the largest 540 @ P. MAŠKARINEC cities - Prague, Brno, Olomouc) that explains their only successful attempt in parliamentary elections in 2006. 6. For more on the emergence, electoral support and the following fall of the W, as an example of a populist or business-firm party, see Hloušek (2012), Havlík and Hloušek (2014), Maškarinec and Bláha (2014), Just and Charvát (2016). 7. The patterns of spatial clustering between support for the TOP09 in 2010 and the Dawn in 2013, as well as the Greens in 2006 and the Dawn in 2013 were almost the same as in the case of the ODS mentioned in Figure 14. 8. Atypically, the STAN found most of its support outside major cities as it comprised primarily of the mayors of smaller towns in Bohemia and Moravia, but other aspects of its spatial support were similar to the other center parties. Disclosure statement No potential conflict of interest was reported by the author. Funding This study was prepared as part of a grant project supported by the Internal Grant Agency of the Jan Evangelista Purkyně University in Ústí nad Labem, Grant No. UJEP-IGA-TC-2019-63-03-2, "Euroscepticism and populism in the 2019 European Parliament elections". ORCID Pavel Maškarinec (g) http://orcid.org/0000-0002-5336-4073 References Anselin, L. 1995. "Local Indicators of Spatial Association-LISA." Geographical Analysis 27: 93-115. doi:10.1111/J.1538-4632.1995.tb00338.x. Anselin, L. 2005. Exploring Spatial Data with GeoDa™: A Workbook. Urbana: University of Illinois. Balík, S., and V. Hloušek. 2016. "The Development and Transformation of the Czech Party System after 1989." Acta Politologica 8: 103-117. Barnea, S., and G. Rahat. 2011. "Out with the Old, in with the 'New': What Constitutes a New Party?" Party Politics 17: 303-320. doi:10.1177/1354068810369148. Becker, S. O., T. Fetzer, and D. Novy. 2017. "Who Voted for Brexit? A Comprehensive District-level Analysis." Economic Policy 32: 601-650. doi:10.1093/epolic/eix012. Bernard, J. 2014. "Historické faktory jako vysvětlení prostorových vzorců volebního chování v postkomunistickém období (Historical Factors as Explanations of Spatial Patterns of Electoral Behaviour in the Post-communist Era)." In Koho volí vaši sousedé? Prostorové vzorce volebního chování na území Česka od roku 1920 do roku 2006, jejich změny a možné příčiny (Who Do Your Neighbours Vote For? Spatial Patterns of Electoral Behaviour in the Czech Territory between 1920 and 2006, Their Changes and Possible causes), edited by T. Kostelecký, R. Mikešová, M. Poláková, D. Čermák, J. Bernard, and M. Šimon, 133-143. Praha: SLON. EURASIAN GEOGRAPHY AND ECONOMICS (Q) 541 Bernard, J. 2015. "Teorie třídně podmíněného hlasování a její souvislosti s prostorovými vzorci volebního chování (Class Voting Theory in the Context of Spatial Patterns of Electoral behaviour)." In Geografie výsledků parlamentních voleb: vzorce volebního chování v Česku 1992-2013 (Geography of Parliamentary Election Results: Patterns of Electoral Behaviour in Czechia, 1992-2013), edited by T. Kostelecký, R. Mikešová, M. Poláková, D. Čermák, J. Bernard, and M. Šimon, 94-116. Praha: Sociologický ústav AV ČR. Bernard, 1, T. Kostelecký, and M. Šimon. 2014. "Existují prostorové kontextové vlivy na volební chování i v relativně nacionalizovaném stranickém systému? Příklad Česka (Are There Spatially Contextual Influences on Voter Behaviour Even in a Relatively Nationalised Party System? the Case of Czechia)." Geografie 119: 240-258. Bernard, 1, and M. Šimon. 2014. "Prostorové vzorce volebního chování v letech 1992-2006: Popis, proměny a možné příčiny (Spatial Patterns of Electoral Behaviour in the Years 1992-2006: Description, Transformations and Possible causes)." In Koho volí vaši sousedé? Prostorové vzorce volebního chování na území Česka od roku 1920 do roku 2006, jejich změny a možné příčiny (Who Do Your Neighbours Vote For? Spatial Patterns of Electoral Behaviour in the Czech Territory between 1920 and 2006, Their Changes and Possible causes), edited by T. Kostelecký, R. Mikešová, M. Poláková, D. Čermák, J. Bernard, and M. Šimon, 109-132. Praha: SLON. Bernard, 1, and M. Šimon. 2017. "Vnitřní periferie v Česku: Multidimenzionalita sociálního vyloučení ve venkovských oblastech (Inner Peripheries in the Czech Republic: The Multidimensional Nature of Social Exclusion in Rural areas)." Sociologický časopis 53: 3-28. Brunclík, M., and M. Kubát. 2016. "The Czech Parliamentary Regime after 1989: Origins, Developments and Challenges." Acta Politologica 8: 5-29. Buštíková, L, and P. Guasti. 2019. "The State as a Firm: Understanding the Autocratic Roots of Technocratic Populism." East European Politics and Societies and Cultures 33: 302-330. doi:10.1177/0888325418791723. Cabada, L, and M. Tomšič. 2016. "The Rise of Person-based Politics in the New Democracies: The Czech Republic and Slovenia." Politics in Central Europe 12: 29-50. doi:10.1515/pce- 2016-0011. Caramani, D. 2004. The Nationalization of Politics: The Formation of National Electorates and Party Systems in Western Europe. Cambridge: Cambridge University Press. Charvát, J. 2014. "Analýza průběhu a výsledků voleb do Národního shromáždění, konaných v roce 1946 (Analysis of the Procedure and Results of the National Assembly Elections of 1946)." In Květnové volby 1946 - volby osudové? Československo před bouří (The May 1946 Elections - Fateful Elections? Czechoslovakia before the storm), edited by J. Kocian and V. Smetana., 23-58. Praha: Euroslavica. Chytilek, R., and O. Eibl. 2011. "České politické strany v policy prostoru (Czech Political Parties in the Policy space)." Sociologický časopis 47: 61-88. Cliff, A. D., and J. K. Ord. 1981. Spatial Processes: Models and Applications. London: Pion. Dalton, R. 1, and M. P. Wattenberg, ed. 2000. Parties without Partisans. Political Change in Advanced Industrial Democracies. Oxford: Oxford University Press. Darmofal, D. 2008. "The Political Geography of the New Deal Realignment." American Politics Research 36: 934-961. doi:10.1177/1532673X08316591. Deegan-Krause, K., and T. Haughton. 2010. "A Fragile Stability. The Institutional Roots of Low Party System Volatility in the Czech Republic, 1990-2009." Politologický časopis 10: 227-241. Dogan, M. 2001. "Class, Religion, Party. Triple Decline of Electoral Cleavages in Western Europe." In Party Systems and Voter Alignments Revisited, edited by L. Karvonen and S. Kuhnle, 93-114. London/New York: Routledge. 542 @ P. MAŠKARINEC Evans, G. 2006. "The Social Bases of Political Divisions in Post-Communist Eastern Europe." Annual Review of Sociology 32: 245-270. doi:10.1146/annurev.soc.32.061604.123144. Evans, J. A. J. 2004. Voters & Voting. An Introduction. London: SAGE Publications. Flint, C, and P. Taylor. 2007. Political Geography: World-Economy, Nation-State, and Locality. Harlow: Pearson/Prentice Hall. Giddens, A. 1984. The Constitution of Society. Berkeley/Los Angeles: University of California Press. Grabowski, W. 2019. "Determinants of Voting Results in Poland in the 2015 Parliamentary Elections. Analysis of Spatial Differences." Communist and Post-communist Studies. doi:10.1016/j.postcomstud.2019.10.006. Gregor, K. 2014. "Přesuny hlasů (Vote shifting)." In Volby do Poslanecké sněmovny 2013 (The 2013 Chamber of Deputies elections), edited by V. Havlík, 189-214. Brno: Munipress. Hanley, S. 2012. "Dynamics of New Party Formation in the Czech Republic 1996-2010: Looking for the Origins of a 'Political Earthquake." East European Politics 28: 119-143. doi:10.1080/21599165.2012.669733. Harbers, I. 2017. "Spatial Effects and Party Nationalization: The Geography of Partisan Support in Mexico." Electoral Studies 47: 55-66. doi:10.1016/j.electstud.2016.11.016. Havlík, V. 2015. "Stable or Not? Patterns of Party System Dynamics and the Rise of the New Political Parties in the Czech Republic." Romanian Journal of Political Science 15: 180-202. Havlík, V. 2019. 'Technocratic Populism and Political llliberalism in Central Europe." Problems of Post-Communism 1-16. doi:10.1080/10758216.2019.1580590. Havlík, V., and V. Hloušek. 2014. "Dr Jekyll and Mr Hyde: The Story of the Populist Public Affairs Party in the Czech Republic." Perspectives on European Politics and Society 15: 552-570. doi:10.1080/15705854.2014.945254. Havlík, V., and P. Voda. 2016. "The Rise of New Political Parties and Re-Alignment of Party Politics in the Czech Republic." Acta Politologica 8: 119-144. Havlík, V., and P. Voda. 2018. "Cleavages, Protest or Voting for Hope? the Rise of Centrist Populist Parties in the Czech Republic." Swiss Political Science Review 24: 161-186. doi:10.1111 /spsr.2018.24.issue-2. Hloušek, V. 2012. "Věci veřejné: Politické podnikání strany typu firmy (The Public Affairs Party: The Political Enterprise of a Business Firm party)." Politologický časopis 19: 322-340. doi:10.581 7/PC2012-4-322. Hloušek, V., and L. Kopeček. 2008. "Cleavages in the Contemporary Czech and Slovak Politics between Persistence and Change." East European Politics and Societies: and Cultures 22: 518-552. doi:10.1177/0888325408315833. Hloušek, V., and L. Kopeček. 2019. "How to Run an Efficient Political Machine: The Billionaire Andrej Babiš and His Political-business Project." Politics in Central Europe 15: 35-54. doi:10.2478/pce-2019-0002. Jennings, W., and G. Stoker. 2016. "The Bifurcation of Politics: Two Englands." The Political Quarterly 87: 372-382. doi:10.1111/poqu.2016.87.issue-3. Johnston, R. J. 1987. "Dealignment, Volatility, and Electoral Geography." Studies in Comparative International Development 22: 3-25. doi:10.1007/BF02687142. Johnston, R., D. Manley, and K. Jones. 2016. "Spatial Polarization of Presidential Voting in the United States, 1992-2012: The 'big Sort' Revisited." Annals of the American Association of Geographers 106: 1047-1062. doi:10.1080/24694452.2016.1191991. Johnston, R., and C. Pattie. 2008. "Place and Vote." In The SAGE Handbook of Political Geography, edited by K. Cox, M. Low, and J. Robinson, 357-374. London: SAGE Publications. Jones, M. P., and S. Mainwaring. 2003. "The Nationalization of Parties and Party Systems. An Empirical Measure and an Application to the Americas." Party Politics 9: 139-166. doi:10.1177/13540688030092002. EURASIAN GEOGRAPHY AND ECONOMICS (Q) 543 Just, P., and J. Charvát. 2016. "Business-Firm Parties and the Czech Party System after 2010." Politics in Central Europe 12: 83-110. doi:10.1515/pce-2016-0018. Kim, 1, E. Elliot, and D.-M. Wang. 2003. "A Spatial Analysis of County-level Outcomes in US Presidential Elections: 1988-2000." Electoral Studies 22: 741-761. doi:10.1016/S0261- 3794(02)00008-2. Kopeček, L. 2016. "Tm Paying, so I Decide': Czech ANO as an Extreme Form of a Business-Firm Party." East European Politics and Societies: and Cultures 30: 725-749. doi:10.1177/ 0888325416650254. Kostelecký, T. 2001. Vzestup nebo pád politického regionalismu? Změny na politické mapě v letech 1992 až 1998 - srovnání České a Slovenské republiky (Rise or Fall of Political Regionalism? Changes in the Electoral Geography, 1992-1998, the Czech Republic and Slovakia compared). Praha: Sociologický ústav AV ČR. Kostelecký, T., and J. Bernard. 2014. "Závěr (Conclusion)." In Koho volí vaši sousedé? Prostorové vzorce volebního chování na území Česka od roku 1920 do roku 2006, jejich změny a možné příčiny (Who Do Your Neighbours Vote For? Spatial Patterns of Electoral Behaviour in the Czech Territory between 1920 and 2006, Their Changes and Possible causes), edited by T. Kostelecký, R. Mikešová, M. Poláková, D. Čermák, J. Bernard, and M. Šimon, 139-143. Praha: SLON. Kostelecký, T., and M. Šimon. 2015. "Geografie politických orientací voličů v letech 1992-2013 (Geography of Voters' Political Orientations, 1992-2013)." In Geografie výsledků parlamentních voleb: Vzorce volebního chování v Česku 1992-2013 (Geography of Parliamentary Election Results: Patterns of Electoral Behaviour in Czechia, 1992-2013), edited by T. Kostelecký, R. Mikešová, M. Poláková, D. Čermák, J. Bernard, and M. Šimon, 69-93. Praha: Sociologický ústav AV ČR. Kouba, K. 2007. "Prostorová analýza českého stranického systému. Institucionalizace a prostorové režimy (Spatial Analysis of the Czech Party System. Institutionalisation and Spatial regimes)." Sociologický časopis 43: 1017-1037. Kouba, K. 2008. "Konfigurativní analýza jako nový metodologický přístup ve volební geografii? Příklad analýzy volebního zisku KSČM v roce 2006 (Configurative Analysis as a New Methodological Approach in Electoral Geography? Case Study of the KSČM's Vote Share in 2006)." Working Paper No. 08-02. Plzeň: FF ZČU. Krčál, P., and V. Naxera. 2018. "Populistická konstrukce národa ohroženého migrací: CAQDAS volebního diskurzu českých parlamentních voleb v roce 2017 (Populistic Construction of Nation Threatened by Migration: CAQDAS of Election Discourse within 2017 Czech Parliamentary elections)." Sociológia 50: 491-523. Linek, L. 2012b. "Volební podpora nových stran TOP 09 a W (Electoral Support for New Parties: The TOP09 and the W)." In Voliči a volby 2010 (Voters and the 2010 elections), edited by L. Linek, 166-188. Praha: SLON. Linek, L., and P. Lyons. 2013. Dočasná stabilita? Volební podpora politických stran V České republice v letech 1990-2010 (Temporary Stability? Electoral Support for Political Parties in the Czech Republic, 1990-2010). Praha: SLON. Lipset, S. M., and S. Rokkan. 1967. "Party Systems and Voter Alignments: Cross-National Perspectives." In Clevage Structures, Party Systems, and Voter Alignments: An Intoduction, edited by M. L. Seymour and S. Rokkan, 1-64. New York: Free Press. Litton, K. 2015. "Party Novelty: Conceptualization and Measurement of Party Change." Party Politics 21: 712-725. doi:10.1177/1354068813499866. Lyons, P., and L. Linek. 2010. "Party System Nationalisation and Non-uniform Vote Switching. Evidence from the Czech Republic." Sociologický časopis 46: 375-399. 544 @ P. MAŠKARINEC Lysek, 1, and T. Lebeda. 2018. "Kdo rozhodl volby? (Who Decided the elections?)." Accessed 15 November 2018. https://www.arcgis.com/apps/Cascade/index.html?appid = 28414bc133e542db930764cc5c11 ee33 Marcinkiewicz, K. 2018. "The Economy or an Urban-Rural Divide? Explaining Spatial Patterns of Voting Behaviour in Poland." East European Politics and Societies 32: 693-719. doi:10.1177/0888325417739955. Maškarinec, P. 2013. "Prostorová analýza prezidentských voleb v České republice v roce 2013 (Spatial Analysis of the 2013 Czech Presidential elections)." Sociológia 45: 435-469. Maškarinec, P. 2015. "Nekonečný příběh. Pokračující stabilita voličské podpory Komunistické strany Čech a Moravy v parlamentních volbách 2006 - 2013 (The Neverending Story. The Continuing Stability of the Communist Party of Bohemia and Moravia Voter Support in the 2006 - 2013 Parliamentary elections)." Politické vedy 18: 135-160. Maškarinec, P. 2017. "A Spatial Analysis of Czech Parliamentary Elections, 2006-2013." Europe-Asia Studies 69: 426-457. doi:10.1080/09668136.2017.1313962. Maškarinec, P., and P. Bláha. 2014. "For Whom the Bell Tolls: Grievance Theory and the Rise of New Political Parties in the 2010 and 2013 Czech Parliamentary Elections." Sociológia 46: 706-731. Matějů, P., and B. Řeháková. 1997. 'Turning Left or Class Realignment? Analysis of the Changing Relationship between Class and Party in the Czech Republic, 1992-96." East European Politics and Societies: and Cultures 11: 501-542. doi:10.1177/ 0888325497011003004. Mészáros, 1, N. Solymosi, and F. Speiser. 2007. "Spatial Distribution of Political Parties in Hungary 1990-2006." Political Geography 26: 804-823. doi:10.1016/j.polgeo.2007.06.002. Morrill, R., L. Knopp, and M. Brown. 2011. "Anomalies in Red and Blue II: Towards an Understanding of the Roles of Setting, Values, and Demography in the 2004 and 2008 U.S. Presidential Elections." Political Geography 30: 153-168. doi:10.1016/j.polgeo.2011.03.006. Nardulli, P. F. 1995. "The Concept of a Critical Realignment, Electoral Behavior, and Political Change." American Political Science Review 89: 10-22. doi:10.2307/2083071. Naxera, V. 2018. "The Neverending Story: Czech Governments, Corruption and Populist AntiCorruption Rhetoric (2010-2018)." Politics in Central Europe 14: 31-54. doi:10.2478/ pce-2018-0017. Novotný, L. 2017. "Der allgegenwärtige Schatten des Kommunismus im tschechischen Parteiensystem (The omnipresent shadow of communism in the Czech party system)." Austrian Journal of Political Science 46: 17-26. Ozen, I. C, and K. O. Kalkan. 2017. "Spatial Analysis of Contemporary Turkish Elections: A Comprehensive Approach." Turkish Studies 18: 358-377. doi:10.1080/ 14683849.2016.1259576. Pink, M., and P. Voda. 2012a. "Vysvětlení prostorových vzorců volebního chování v parlamentních volbách v České republice a na Slovensku v letech 1996-2010 (Explaining the Spatial Patterns of Electoral Behaviour in Czech and Slovak Parliamentary Elections, 1996-2010)." In Volební mapy České a Slovenské republiky po roce 1993: Vzorce, trendy, proměny (Electoral Geographies of the Czech and Slovak Republics after 1993: Patterns, Trends, changes), edited by M. Pink, 201-242. Brno: CDK. Pink, M., and P. Voda. 2012b. "Základní vymezení voličské základny politických stran v České republice v letech 1996-2010 (Defining the Electorates of Political Parties in the Czech Republic, 1996-2010)." In Volební mapy České a Slovenské republiky po roce 1993: Vzorce, trendy, proměny (Electoral Geographies of the Czech and Slovak Republics after 1993: Patterns, Trends, changes), edited by M. Pink, 17-46. Brno: CDK. Plešivčák, M. 2013. "The Regional Dimension of the Socio-political Urban-rural Conflict in Slovakia." AUC Geographica 48: 47-58. doi:10.14712/23361980.2015.10. EURASIAN GEOGRAPHY AND ECONOMICS (Q) 545 Shin, M. E., and J. Agnew. 2007. "The Geographical Dynamics of Italian Electoral Change, 1987-2001." Electoral Studies 26: 287-302. doi:10.1016/j.electstud.2006.05.002. Sichinava, D. 2017. "Cleavages, Electoral Geography, and the Territorialization of Political Parties in the Republic of Georgia." Eurasian Geography and Economics 58: 670-690. doi:10.1080/15387216.2018.1442734. Sikk, A. 2012. "Newness as a Winning Formula for New Political Parties." Party Politics 18: 465-486. doi:10.1177/1354068810389631. Sikk, A., and P. Kóker. 2017. "Party Novelty and Congruence: A New Approach to Measuring Party Change and Volatility." Party Politics. doi:10.1177/1354068817747512. Šimon, M. 2015. "Measuring Phantom Borders: The Case of Czech/Czechoslovakian Electoral Geography." Erdkunde 69: 139-150. doi:10.3112/erdkunde.2015.02.04. Smith, M. L, and P. Matějů. 2011. "Restratifikace české politiky. Vývoj třídně podmíněného volebního chování v České republice v letech 1992-2010 (The Re-stratification of Czech Politics: Class Voting in the Czech Republic between 1992 and 2010)." Sociologický časopis 47: 33-59. Stauber, J. 2017. "Institutionalization of Nationalized Party System: The Czech Case." Acta Politologica 9: 46-65. Tapiador, F. 1, and J. Mezo. 2009. "Vote Evolution in Spain, 1977-2007: A Spatial Analysis at the Municipal Scale." Political Geography 28: 319-328. doi:10.1016/j.polgeo.2009.07.007. Taylor, P. 1, and R. J. Johnston. 1979. Geography of Elections. Harmondsworth: Penguin Books. Thrift, N. 1983. "On the Determination of Social Action in Space and Time." Environment and Planning D: Society and Space 1: 23-57. doi:10.1068/d010023. Tiemann, G. 2012. "The Nationalization of Political Parties and Party Systems in Postcommunist Eastern Europe." Communist and Post-communist Studies 45: 77-89. doi:10.1016/j.postcomstud.2012.02.009. Voda, P., and M. Pink. 2015. "Explanation of Spatial Differentiation of Electoral Results in the Czech Republic and Slovak Republic." Communist and Post-communist Studies 48: 301-316. doi:10.1016/j.postcomstud.2015.09.002. Whitefield, S. 2002. "Political Cleavages and Post-Communist Politics." Annual Review of Political Science 5: 181-200. doi:10.1146/annurev.polisci.5.112601.144242. Zarycki, T. 2000. "Politics in the Periphery: Political Cleavages in Poland Interpreted in Their Historical and International Context." Europe-Asia Studies 52: 851-873. doi:10.1080/ 713663091. Zarycki, T. 2015. "The Electoral Geography of Poland: Between Stable Spatial Structures and Their Changing Interpretations." Erdkunde 69: 107-124. doi:10.3112/erdkunde.2015.02.02. 546 @ P. MAŠKARINEC Appendix A. Vote shares of Czech political parties, 1990-2017 1990 1992 1996 1998 2002 2006 2010 2013 2017 OF 49.50 ČSSD 4.11 6.53 26.44 32.31 30.20 32.32 22.08 20.45 7.27 ODS 29.73 29.62 27.74 24.47 35.38 20.22 7.72 11.32 KSČM 13.24 14.05 10.33 11.03 18.51 12.81 11.27 14.91 7.76 KDU-ČSL 8.42 6.28 8.08 9.00 14.27 7.22 4.39 6.78 5.80 us 8.60 0.30 sz 4.10 6.52 1.12 2.36 6.29 2.44 3.19 1.46 SPR-RSČ 5.98 8.01 3.90 0.97 0.03 0.19 ODA 5.93 6.36 0.50 0.15 HSD-SMS 10.03 5.87 0.42 TOP09 16.70 11.99 5.31 VV 10.88 Pirates 0.80 2.66 10.79 AN02011 18.65 29.64 Dawn (SPD) 6.88 10.64 STAN 5.18 Others 10.60 19.11 10.74 6.30 8.72 5.68 11.19 6.77 4.49 Legend: ANO2011 (populist); CSSD: Czech Social Democratic Party (social democratic); Dawn: The Dawn of Direct Democracy of Tomio Okamura (radical right-wing populist); HSD-SMS: Movement for Self-Governing Democracy - Association for Moravia and Silesia (regionalist); KDU-ČSL: Christian Democratic Union -Czechoslovak People's Party (Christian democratic); KSČM: Communist Party of Bohemia and Moravia (radical left); ODA: Civic Democratic Alliance (liberal); ODS: Civic Democratic Party (liberal-conservative); OF: Civic Forum (umbrella movement); Pirates: Czech Pirate Party (social-liberal); SPD: Freedom and Direct Democracy - Tomio Okamura (radical right-wing populist); SPR-RSČ: Association for the Republic - Republican Party of Czechoslovakia (radical right-wing populist), in the 2002 election as Republicans of Miroslav Sládek (RMS); STAN: Mayors and Independents (centrist); SZ: Green Party (ecologist), in the 1992 election as a member of the Liberal-Social Union (LSU) coalition; TOP09: Tradition, Responsibility, Prosperity 09 (liberal-conservative); US: Union of Liberty (liberal); VV: Public Affairs (populist). Appendix B. Moran's / scores for parties with parliamentary representation, 2006-2017. 2006 2010 2013 2017 ČSSD 0.632 0.596 0.595 0.503 ODS 0.647 0.557 0.591 0.530 KSČM 0.494 0.438 0.485 0.454 KDU-ČSL 0.677 0.724 0.698 0.701 sz 0.571 0.403 0.426 0.777 TOP09 0.572 0.651 0.530 Pirates 0.278 0.427 0.545 ANO2011 0.456 0.591 Dawn (SPD) 0.562 0.647 VV 0.454 STAN 0.535 Note: Values in elections when the party did not receive parliamentary representation are highlighted in italics. EURASIAN GEOGRAPHY AND ECONOMICS (Q) 547 Appendix C. Party nationalization scores (PNS) for the main political parties, 2006-2017. 2006 2010 2013 2017 ČSSD 0.931 0.909 0.910 0.888 ODS 0.897 0.903 0.853 0.864 KSČM 0.871 0.868 0.877 0.866 KDU-ČSL 0.703 0.616 0.683 0.652 SZ 0.867 0.842 0.843 0.859 TOP09 0.854 0.799 0.785 VV 0.922 Pirates 0.859 0.882 0.887 ANO2011 0.924 0.927 Dawn (SPD) 0.881 0.891 STAN 0.716 Note: PNS is measured by the Gini coefficient (see Jones and Mainwaring 2003). Data are aggregated on the level of 206 microregions - 205 administrative districts of municipalities with extended powers and the capital city of Prague. Appendix D. Bivariate Moran's / scores for new and established parties, 2006-2017. ČSSD (2006) ODS (2006) KSČM (2006) KDU-ČSL (2006) TOP09 (2010) ANO2011 (2013) -0.213 0.371 -0.046 -0.435 0.182 ANO2011 (2017) 0.311 -0.089 0.230 -0.348 -0.294 VV (2010) -0.079 0.206 -0.136 -0.261 0.078 Dawn (2013) 0.065 -0.120 -0.064 0.120 -0.130 SPD (2017) 0.393 -0.353 0.215 0.018 -0.449 Pirates (2010) -0.171 0.180 0.012 -0.196 0.180 Pirates (2013) -0.268 0.339 -0.069 -0.295 0.306 Pirates (2017) -0.436 0.468 -0.258 -0.178 0.469 TOP09 (2010) -0.548 0.516 -0.256 -0.176 1.000 TOP09 (2013) -0.569 0.582 -0.310 -0.207 0.602 TOP09 (2017) -0.446 0.455 -0.179 -0.165 0.463 STAN (2017) -0.367 0.413 -0.232 -0.210 0.424 Appendix E. Bivariate Moran's / cluster maps of the distribution of support for the ODS (2006) with the TOP09 (2010), the Pirates (2017) and the STAN (2017).