Measuring Hybrid Regimes

Measuring Hybrid Regimes: An Alternative Measurement Method and Classification of Post-Soviet Regimes* Farid Guliyev† Abstract The hybrid regime framework provides a useful approach for analysis of (post-)Third wave political regimes, yet there has not been sufficient effort to develop adequate systematic measure of ‘mixed’ regimes. Several measures that exist suffer from various measurement validity problems. This article introduces an alternative method and demonstrates how it can be used to classify post-Soviet regimes in the 1991-2005 period. Future research can employ this new classification to explore the question of change and stability of hybrid regimes in post-communist Eurasia as well as to study causes and consequences of regime change and democratisation in comparative perspective. Keywords: hybrid regimes, electoral authoritarianism, political regimes, measurement, postSoviet regimes * Working paper, last revised Feb 22, 2012. † School of Humanities and Social Sciences, Jacobs University Bremen, Campus Ring 1, 28759, Bremen, Germany [email: [email protected]] 1 Introduction As the advancement of democracy seems to have halted throughout the world (Diamond 2008), it has become clear that the third wave of democratisation (Huntington 1991) has produced a mixed range of political regimes. While some countries succeeded in setting up institutionalised democracy, others failed to democratise and reverted to outright authoritarianism. But a much larger group of polities evolved into intermediary regimes that combine major attributes of both democracy and authoritarianism. These regimes have been dubbed ‘mixed’, ‘gray zone’ or ‘hybrid’ regimes (e.g., Karl 1995; Carothers 2002; Bunce & Wolchik 2008; also see Zinecker 2009), formally defined as those political regimes that are ‘neither clearly democratic nor conventionally authoritarian’ (Diamond 2002, p. 25). Depending on one’s perspective, this category is composed of varieties of electoral authoritarian systems (see, e.g. Diamond 2002; Levitsky & Way 2002; Schedler 2006a; Howard & Roessler 2006) and defective democracies (Croissant & Merkel 2004; also see Bogaards 2009). Most important, at the start of the twenty-first century hybrid regimes represent the most widespread type of political system outside the liberal-democratic world (Howard & Roessler 2006, p. 365; Bogaards 2009, p. 399). In fact, electoral authoritarianism has become ‘the modal type of political regime’ among less developed countries (Schedler 2006b, p. 3). Thus said, how well are we equipped to measure hybrid regimes? Recently, attempts have been made to create, using Dahl’s criteria of Polyarchy (Dahl 1971), a systematic scheme to measure hybrid regimes (Howard & Roessler 2006; Brownlee 2009). However, notwithstanding some of their merits, the existing measurements suffer from some ‘measurement validity’ problems that is the concerns about how adequately the chosen operationalization and the scoring of cases capture the concepts (regime types, in our case) the scholar seeks to measure (Adcock & Collier 2001, pp. 529-530). Consequently, the quality of the available measures precludes their use as definitive guide to hybrid regime measurement. This article advances an alternative measurement framework and illustrates how it can be usefully applied to identify political regime types. This is illustrated with a classification of regimes in the Soviet successor states for the period between 1991-2005. The choice of this particular region is motivated by the consideration that many scholars believe that the region is home to many ‘mixed’ forms of regimes (e.g. Levitsky & Way 2002; Bunce & Wolchik 2008; 2010; Hale 2010). The region’s political regimes, therefore, should serve as a 2 suitable ground for exploring a new measurement method. By introducing it, this article aims to contribute to the ongoing effort to design precise hands-on tools for measuring political regimes in general and hybrid regimes in particular. Furthermore, the typology of post-Soviet regimes hereby advanced can be used to derive hypotheses about change and durability of hybrid regimes that can be tested empirically by future research. It can also benefit scholars interested in understanding causes and consequences of regime change including democratic transitions, autocratic reversals, and the persistence of authoritarian rule. Disaggregating Hybrid Regimes To begin with, hybrid regimes combine considerable features of both authoritarianism and democracy and therefore, typologically, lie in the middle of the continuum between consolidated democracy (e.g. West European countries, USA) and full-scale authoritarianism (e.g. North Korea, Saudi Arabia). They range from authoritarian regimes with democratic aspects to democracies with anti-democratic deficiencies. Authoritarian regimes with democratic aspects have been termed differently but the most prominent concepts include ‘electoral authoritarianism’ (Schedler 2006a), ‘competitive authoritarianism’ (Levitsky & Way 2002), and ‘semi-authoritarianism’ (Ottaway 2003). With regard to democracies with deficiencies, scholars have introduced an entire vocabulary of diminished subtypes of democracy (Collier & Levitsky 1997) including such terms as ‘delegative democracy’ (O’Donnell 1994), ‘illiberal democracy’ (Zakaria 1997) and, more generally, ‘defective democracy’ (Croissant & Merkel 2004). These regimes are electoral democracies but differ from liberal democracy. While electoral democratic systems meet the procedural criterion of free and fair elections, they do not meet up to the more comprehensive (liberal)democratic standard encompassing the rule of law, an effective separation of powers and the guarantee of basic civil liberties (see, e.g. Zakaria 1997, p. 22). Electoral authoritarian regimes differ from closed authoritarian systems in that they allow for multiparty electoral competition (Diamond 2002, p. 25). Their rulers use such democratic institutions as parties, elections and legislatures. But, typically, these institutions are artfully manipulated (Schedler 2002), functioning as tools to help the ruler maintain his hold on power (Levitsky & Way 2002, p. 54). Manipulation of democratic institutions was not an essential part of the classical concept of authoritarian regime (Linz 1964; 2000). Hence, the introduction of the new term ‘electoral authoritarianism’. 3 Moving down the ‘ladder of abstraction’ (Sartori 1970), following Howard & Roessler (2006), and before them Diamond (2002), electoral authoritarian regimes can be subdivided into hegemonic authoritarian and competitive authoritarian regimes. The crucial distinction between the two subtypes lies in the level of contestation they allow. Hegemonic authoritarian regimes hold elections but their outcome is known in advance whereas competitive authoritarian regimes allow for truly contested elections in which the opposition candidates have real chances to win. But unlike democracy, in which all competing parties have equal chances to win (or lose), competitive authoritarian regimes are ruled by the incumbents who will use their control of state resources and the media to disadvantage opponents and to prevent the opposition’s victory. Existing Measures Among several measures of hybrid regimes, two can be considered relatively more systematic: one developed by Howard & Roessler (2006), the other – by Brownlee (2009). Howard & Roessler’s (2006, p. 368) measurement draws on previous conceptual work on hybrid regimes. Their measurement tries to capture differences between hybrid types by assessing the level of contestation in a polity. This is done by examining the procedures used for selecting the chief executive in a polity. The coding draws information from two sources: the Freedom House (FH) ratings and the Polity dataset. The following coding is employed. Closed authoritarian regimes are those polities that receive a 7 (the worst score) on the FH political rights index or a -8 or worse on the Polity scale that ranges from +10 (full democracy) to -10 (full autocracy). To distinguish competitive authoritarian regimes from hegemonic authoritarian regimes, the authors use the measure of the share of votes for the winning party or candidate, putting the threshold at 70 %. On the other side of the continuum, liberal and electoral democracies are separated from competitive authoritarian regimes by applying the following rule: democracy is any regime-year receiving a FH political rights score of 1 to 2 or a Polity score of 6 to 10. Competitive authoritarian regimes, therefore, are all polities rated 3 to 6 on the FH political rights index or -7 to 6 on the Polity scale and in which the winning party/candidate got no more than 70 % of the vote. Brownlee’s (2009) measurement of hybrid regimes (electoral authoritarianism), in turn, is based on data from the World Bank’s Database of Political Institutions (or DPI, see Keefer 2007). Brownlee uses the DPI indices of Legislative and Executive Electoral 4 Competitiveness (LIEC and EIEC) to classify closed and electoral authoritarian regimes in the following way. Regime-years that received a score between 1-4 are coded as fully closed authoritarian whereas regime-years with a score of 5-7 are coded as electoral authoritarian. The latter are further classified into competitive authoritarian (a score of 7 on one of the indices) and hegemonic authoritarian (with scores no greater than 5 or 6 on the same measures). The principal advantage of Howard & Roessler’s and Brownlee’s approach to regime analysis is that it moves beyond a simplistic dichotomous view of political regime (such as, e.g. Alvarez et al. 1996; Przeworski et al. 2000). In addition, it can be used in combination with well-known typologies of classical authoritarian regimes (Geddes 1999; also Brownlee 2009) and of defective types of democracy (Croissant & Merkel 2004). Most importantly, it recognises the diversity of regimes that reside between liberal consolidated democracy and closed authoritarianism. It also acknowledges that these regimes are neither transitory nor temporary but regime types on their own (see also Carothers 2002; Bogaards 2009). Yet, there are several problems with respect to how the authors use data sources to measure regime types, specifically concerns about measurement validity (Adcock & Collier 2001). First, there seems to be a misfit between the concepts (regime types) the authors try to measure and their choice of measures in the data sets used. Second, the cut-off points are drawn largely arbitrarily and oftentimes without explicit justification. The Discussion section will provide a more elaborate examination of these points. An Alternative Method of Measurement This section introduces an alternative way of measuring hybrid regimes that is further used to create a typology of post-Soviet regimes for the period between 1991-2005. Conceptual Distinctions It has by now been largely accepted that democracies differ from autocracies in that they allow free contestation for the key government offices and respect the outcomes of such contestation. Focusing on political contestation for the key offices to conceptualise and operationalize political regime has a long tradition in comparative research and goes back to Schumpeter (1954) and Dahl (1971). Schumpeter (1954, p. 269) defined democracy as ‘that institutional arrangement for arriving at political decisions in which individuals acquire the 5 power to decide by means of a competitive struggle for the people’s vote’. Dahl thought that in a political democracy, citizens must have channels to formulate their preferences and be able, through their individual and collective action, to influence government decisions. He believed that ‘a key characteristic of a democracy is the continuing responsiveness of the government to the preferences of its citizens, considered as political equals’ (Dahl 1971, p. 1). The most important advantage of the Schumpeterian/Dahlian approach is that it focuses largely on procedural criteria and can therefore be easily operationalized. Most of the current conceptualisations of political regime follow Schumpeter and Dahl to consider contestation as the most salient dimension of democracy/autocracy continuum and concentrate on electoral competition and the protection of basic civil liberties as its two main attributes. Therefore, following this tradition, democracies are defined as those political systems in which there exists an equal level playing field for both opposition and incumbents to compete for central governmental offices. Those political systems that do not provide for ‘free and fair’ competition are treated as autocracies (or authoritarian regimes). Hybrid regimes reside somewhere in the middle between the two. Having said that, it is necessary to first evaluate the extent to which countries conduct open and fair competitive elections for the key government offices or, in other words, to assess the conditions for political contestation (see Dahl 1971). To be precise, ‘to determine whether a country is democratic, we need to know whether it allows for political competition. To assess competition, we need to assess the structure and process of competition’ (Bogaards 2007, p. 1233). Following the conceptual distinctions already outlined, three variants of political competition are possible: - contested elections (competitive system), - restricted elections (authoritarian with restricted but competitive elections), - elections absent or controlled (authoritarian, either ‘closed’ or with uncompetitive elections). In the second step, closed authoritarian and hegemonic authoritarian types, and, separately, electoral and consolidated democratic types, need to be separated one from the other. What distinguished electoral authoritarianism from closed authoritarianism is that electoral authoritarian regimes, both hegemonic and competitive, are open to ‘some form of 6 multiparty or multifactional polling’ (Brownlee 2009, p. 524). What separates closed from hegemonic authoritarian regimes is whether or not contestation (elections and parties) exists. Electoral democracy, as has already been pointed out, is largely a minimalist Schumpeterian democracy whereas liberal democracy is an institutionally consolidated democratic system (Linz & Stepan 1996). Measurement Various data sources are available to the researcher who is interested in measuring competition but some of the most oft-used datasets have some conceptual and measurement problems (for an insightful review, see Munck & Verkuilen 2002). The commonly-used Freedom House index, for example, has been shown to be inconsistent as a source of timeseries data. Hartlyn (2002, p. 128) pointed out that it had been shown that during the 1970s and 1980s, the democracy audit organisation tended to rate favourably countries in some regions, especially Latin America, than in others, and in the 1990s their ratings became comparably more rigid. In general, methodologically the FH index is considered one of the most problematic of all existing indices of democracy (Munck & Verkuilen 2002). Another example is the widely-used combined Polity score (POLITY in the dataset), calculated by subtracting a polity’s autocracy score (AUTOC, up to -10) from its democracy score (DEMOC, up to +10) (Polity IV; Marshall & Jaggers 2009, p. 15). However, the Polity dataset originally was intended to measure authority trends not regime types, and much of its combined score is derived from a single measure of constraints on the executive (Gleditsch & Ward 1997; see also Mainwaring, Brinks & Perez-Linan 2001; Cheibub, Gandhi & Vreeland 2009). As the constraints on the executive is a measure of horizontal accountability rather than of openness and fairness of electoral competition, the use of the unified Polity score for measuring contestation does not always and necessarily fit theoretical foundations of many coding schemes (Mainwaring, Brinks & Perez-Linan 2001). Moreover, the question of establishing unambiguous cut-off points along the single Polity scale remains unresolved alongside the problematic nature of the index’s aggregation rules (Munck & Verkuilen 2002, p. 26). To avoid these pitfalls, it is advised to use component scores from the Polity dataset not the combined autocracy-democracy scale (Gleditsch & Ward 1997). A more direct measure of competition in the Polity dataset is the conceptual variable called ‘executive recruitment’ (EXREC in the data set). It can be used as a baseline criterion in the first step of 7 regime coding. The EXREC variable is a composite scale measuring the extent of contestation for the highest executive office in a country. It is derived from three other variables of the Polity dataset: the extent of institutionalisation of executive transfers, the competitiveness of executive selection and the openness of executive recruitment (Marshall & Jaggers 2009, pp. 49-62). According to Marshall & Jaggers (2009), the concept of executive recruitment is derived from its original conceptualisation by Eckstein and Gurr where executive recruitment was meant to involve ‘…the ways in which social superordinates come to occupy their positions of political authority – that is, how institutionalised, competitive and open are the mechanisms for selecting a political leader. In terms of modern democratic theory, democratic systems are defined as those polities that afford their citizens the opportunity to replace their political representatives through regularly scheduled, competitive and open elections’ (Marshall & Jaggers 2009, p. 49). Table 1. Polity EXREC Variable Description Polity IV: executive Description Score in Polity recruitment dataset Executive selection Recruitment of the chief executive is determined through ascription by hereditary succession. Dual executive: The position of chief executive is shared. ascription + designation Executive recruitment is determined both by 1 2 hereditary succession and designation. Executive recruitment Chief executives are chosen by designation by through designation the ruling political elite, without formal 3 competition by parties or individuals. Executive recruitment ‘Unregulated’ changes in executive power through self-selection through forceful seizures of power by rival political elites. In addition to the use, or threat of use, of force, there are no formal mechanisms by which the transfer of executive power is regulated. 8 4 Gradual transition from Gradual transition (2 or more years) from self-selection ‘unregulated’ executive recruitment to some 5 other form of ‘regulated’ recruitment process. Dual executive: The position of chief executive is shared. ascription + election Executive recruitment is determined both by 6 hereditary succession and election. Transitional or The chief executive is chosen through elections restricted elections which are ‘free’ but not necessarily ‘fair’ in both 7 design and practice. While elections are competitive, nevertheless, the electoral process is stacked – although the results are not predetermined – in favor of one candidate over another. Competitive elections Chief executive is chosen through competitive 8 multi-candidate elections. The electoral process is transparent and its outcomes are institutionally uncertain. Note: Descriptions are adapted from Marshall & Jaggers 2009, pp. 49-62. The EXREC variable is a 7-point scale that ranges from 1 (executive recruitment determined by hereditary succession) to 8 (executive recruitment by competitive elections) (see Table 1). This measure provides a good source to draw upon when assessing the extent of competition and distinguishing between systems in which elections are absent or controlled from those systems in which elections are held but with some restrictions, and also from those in which elections are competitive. These correspond to the following three categories respectively: - authoritarian (‘closed’ or with uncompetitive elections), - authoritarian with restricted but competitive elections (‘free but not fair’), - competitive. Competitive systems are those systems in which the electoral process is ‘free and fair’, and the outcomes of elections are ex ante uncertain. The score matching such description is a score of 8 on the Polity’s EXREC variable and assignment of this score in a particular year means a regime can be considered competitive for that year. The authoritarian 9 type with restricted but competitive elections corresponds to what the literature has described as ‘competitive authoritarianism’ and does not need further classification. This type is easy to capture with the EXREC variable. Its score of 7 stands for multiparty elections that are ‘free’ but not necessarily ‘fair’. Putting the threshold at 7 makes it possible to distinguish competitive authoritarian regimes (country-years coded 7) from authoritarian regimes with either closed or uncompetitive elections (country-years falling below 7, i.e. 1-6). Since neither the EXREC nor any other Polity variables are sufficiently discriminating for the task of distinguishing between closed and hegemonic authoritarianism, a more suitable discriminating variable is the index of electoral competitiveness, more precisely Legislative-Executive Indices of Electoral Competitiveness (LIEC and EIEC in the dataset). The source of data for these indices is the World Bank’s DPI dataset which covers the years 1975-2006 (Keefer 2007). The original scoring was done with one-year lag, so this has to be taken into account when coding. It is suggested to use a combined average score. The preference for the average score of the two indices is due to the consideration that the presence of at least some competition involving several parties for either the legislature or the chief executive office is indicative of hegemonic electoral authoritarianism. Since competitive authoritarianism had already been sorted out by the Polity EXREC measure, the DPI score helps to distinguish closed from hegemonic types. The threshold separating the two is established at the average score of 4. A score below or equal to 4 indicates closed authoritarianism; higher than 4 - hegemonic authoritarianism. The DPI scores between 1 and 4 describe the following conditions: - 1 = without legislature or executive; - 2 = unelected legislature or executive; - 3 = legislature or executive where only one candidate is elected, and - 4 = only one party running (see Table 2). 10 Table 2. Legislative and Executive Indices of Electoral Competitiveness (LIEC and EIEC) no legislature/executive: 1 unelected legislature/executive: 2 elected, one candidate: 3 one party, multiple candidates: 4 multiple parties are legal but only one party won seats: 5 multiple parties did win seats but the largest party received more than 6 75% of the seats: largest party got less than 75%: 7 Note: Source: Keefer 2007, pp. 15-16. The average score of 4 is only possible to obtain if one of the values of the LIEC or EIEC indices is at least 5 indicating that multiple parties are legal but only one party won seats (which implies that there were elections where multiple parties run). Since competitive authoritarian regimes had already been sorted out in the first step, there is no need to worry: all country-years that do not fall into the closed authoritarian category are automatically entered as hegemonic even if a country received the highest score of 7, signifying that the largest party won less than 75 %. For the reasons explained further, a scoring of hegemonic and competitive authoritarian regimes based on the DPI’s LIEC and EIEC indices can lead to a misclassification (e.g. Brownlee 2009). Finally, to distinguish electoral democracies from consolidated ones, the most obvious measure is the Polity’s institutionalised Democracy score (DEMOC in the dataset) which is an eleven-point scale (0-10). It is a good and unequivocal measure of the quality and functionality of democratic institutions. A competitive regime that receives a score of 10 (the highest score) on this scale is a consolidated democracy. Otherwise, it is an electoral democracy. Table 3 shows how, following the procedures explained above, the researcher can create a typology consisting of the following regime types: closed authoritarianism, hegemonic authoritarianism, competitive authoritarianism, electoral democracy and consolidated democracy. 11 Table 3. Coding Rules and Measurement Regime Authoritarian (‘closed’ or with Authoritarian Categories uncompetitive elections) with restricted Fully Competitive Regime but competitive elections (‘free but not fair’) Indicator and its Elections absent or controlled 1 (Polity IV EXREC score = 1-6) Restricted Contested elections elections (Polity IV EXREC score = 8) (Polity IV Measure EXREC score = 7) Further if DPI2 comb. if DPI comb. Specificatio average Leg.- average Leg.- - if Polity if Polity DEMOC DEMOC 3 n Rules and Exec. IEC Exec. IEC score = Measureme score ≤ 4 score = otherwise nt (one-year otherwise lagged) (one-year score = 10 lagged) Regime Closed Hegemonic Competitive Electoral Consolidated Types Authoritarian Authoritarian Authoritarian Democracy Democracy Note: Author’s coding rules (see discussion in the text); Sources: 1Polity IV (Marshall & Jaggers 2009, pp. 49-62); 2 DPI Leg.-Exec. IEC = Legislative and Executive Indices of Electoral Competitiveness (Keefer 2007, pp. 15-16); 3Polity IV (Marshall & Jaggers 2009, pp. 13-14). Application: A Classification of Post-Soviet Regimes, 1991-2005 To illustrate heuristic possibilities of the new method, I use the new measurement strategy to score regimes in the twelve post-Soviet states of Eurasia. It has long been held that this region boasts one of the world’s largest pools of hybrid regimes (e.g. Carothers 2002; Schedler 2006b, p. 3; Bunce & Wolchik 2008; Hale 2010), and it therefore is a suitable ground for testing the new measure. Table 4 shows a classification of post-Soviet regimes obtained. 12 Table 4. Classification of Post-Soviet Regimes, 1991-2005 Country Start End (at 2005) Type Azerbaijan 1993 Armenia 1991 Armenia 1998 Belarus 1991 Belarus 1997 Georgia 1991 Georgia 2004 Kazakhstan 1991 Kazakhstan 1995 Kyrgyzstan 1991 Kyrgyzstan 2005 Moldova 1991 Moldova 2001 ElecDem Russia 1992 CompetA Tajikistan 1994 HegemA Turkmenistan 1991 ClosedA Ukraine 1991 Ukraine 1994 Uzbekistan 1991 1994 HegemA Uzbekistan 1995 1998 ClosedA Uzbekistan 1999 HegemA 1995 ElecDem CompetA 1993 CompetA ClosedA 2003 CompetA ElecDem 1994 ClosedA HegemA 2004 HegemA CompetA 2000 1993 CompetA CompetA ElecDem HegemA Note: ClosedA = Closed Authoritarian; HegemA = Hegemonic Authoritarian; CompetA = Competitive Authoritarian; ElecDem = Electoral Democracy. The coding follows the procedures discussed in the text (also see Table 3). The three-year threshold applies to distinguish regimes, understood as the rules of the political game, from periods of struggle over the rules, turmoil and temporary interventions which are better seen as interregna (Geddes 1999, p. 116, fn. 1). The DPI data on Tajikistan (1991-1993) is not available. Of particular interest is an observation that there is no linear evolution towards democracy and away from closed authoritarianism. This observation is particularly relevant for theory of regime change and democratisation which often inaccurately assumes such 13 universal evolutionary linearity. It also re-confirms Carothers’ justified criticism of the oncedominant transition paradigm and democratic teleology, in general (Carothers 2002). Furthermore, the following three tendencies are noteworthy. First is the trend towards less competitiveness. Belarus, Armenia and, to an extent, Azerbaijan (if we take the year 1992 as the ‘democratizing moment’, followed up by stabilization of a hegemonic authoritarian regime), are in this group. Second is the trend towards greater competitiveness. This group includes Georgia, Ukraine, and Moldova. All three are electoral democracies, albeit of a highly deficient kind (for a somewhat similar view, see Beichelt 2004). Finally, there are three cases in which political regimes have been remarkably stable. The stable regimes are the competitive authoritarian regime in Russia, the closed authoritarian regime in Turkmenistan, and the hegemonic authoritarian regime in Tajikistan. Among post-Soviet competitive authoritarian regimes, Russia’s is of particular interest. While many post-Soviet competitive authoritarian regimes moved towards greater or lesser competitiveness (e.g. Belarus, Moldova, Georgia), Russia’s competitive authoritarianism has remained surprisingly resilient. This seems to defy general belief that competitive authoritarian regimes are inherently unstable (e.g. Bunce & Wolchik 2008, p. 7). Further research may be needed to inquire into the causes and implications of the unusually durable competitive authoritarian system that has formed in post-Soviet Russia. Discussion From the conceptual point of view, the proposed measurement strategy comes close to the classification scheme used by Howard & Roessler (2006) but has several advantages. First and foremost, it is a straightforward, easy-to-replicate and theoretically-grounded measure of political regimes. Second, unlike the oft-used Polity scale (which Howard & Roessler combine with Freedom House scores), in which the constraints on the executive is the major contributor (Gleditsch & Ward 1997; Mainwaring, Brinks & Perez-Linan 2001, pp. 56-57), the EXREC variable, that is proposed as an alternative, provides a direct measure of political contestation. Following Alvarez et al. (1996), it is perhaps more useful to consider the difference in the extent to which the power of the executive is constrained an empirical rather than definitional question, as there is variation on this even among established democracies. 14 Third, it is less arbitrary: it relieves the researcher from the need to make difficult choice about establishing proper thresholds along the 21-point Polity scale. The other commonly-used practice is to use different cut-off points on the 7-point Freedom House indices of political rights and civil liberties. In fact, several authors employed the FH index to identify hybrid regimes (e.g. Diamond 2002; Schedler 2006b; Morlino 2009). Just like the cutting off of the combined Polity score, the use of the FH score is arbitrary unless theoretical justification as to why particular cut-off points have been chosen was provided (also see Bogaards 2009, p. 407). Fourth, it provides a direct measure for competitive authoritarianism (an EXREC score of 7) which makes it unnecessary to identify how much of the popular vote went for the winning party or candidate (the indicator used for this by Howard & Roessler) or the share of seats won by the largest party (Brownlee 2009). The DPI indices of legislative and executive electoral competitiveness, which are sometimes used to identify competitive authoritarian regimes, may not be an accurate measure to gauge the extent of contestation. Like with similar measures, the risk is to confuse competition with competitiveness which may lead to misclassifications of competitive and noncompetitive elections, the point first raised by Giovanni Sartori, and later tested by Bogaards (2007) on African data (also, see Munck & Snyder 2004). Brownlee (2009), for instance, uses the DPI indices of electoral competitiveness to classify closed and hybrid authoritarian regimes in the following way. He codes regime-years that received a score between 1-4 as fully closed authoritarian whereas those that received a score of 5-7 code as electoral authoritarian. The latter are further disaggregated into competitive authoritarian (a score of 7 on one of the indices) and hegemonic authoritarian (with scores no greater than 5 or 6 on the same measures). The problem with such approach is that it does not fully appreciate the skilfulness of the incumbents in hegemonic authoritarian regimes with respect to manipulation of the electoral process. I illustrate this with an example of Azerbaijan. In the 2005 parliamentary elections, which according to international observers were largely rigged, the governing Yeni (New) Azerbaijan Party (YAP) won only 61 seats (48.8 %) in the 125-member Milli Mejlis (the Azerbaijani legislature) while the so called independents won 43 seats (34.4 %). Of the two running opposition blocs, i.e. the Freedom Bloc (Azadliq) and the YeS-Yeni Siyaset (New Policy) Alliance, only the Freedom Bloc could secure some 6 seats in the legislature (Kara 2007). The independents were not only all regime loyalists but 15 were in fact party affiliates. The Organization of Security and Cooperation in Europe (OSCE) Election Observation Mission reported that a large number of self-nominated candidates, which constituted more than half of all candidates, had a party affiliation and 332 members of the YAP were registered as self-nominated ‘independents’ in addition to the 101 candidates officially registered by the party (see OSCE 2006, p. 9; Kara 2007, p. 721). Despite similar problems with previous parliamentary elections in Azerbaijan, the DPI, which is compiled from ‘objective facts’, continued to register a score of 7 for legislative competitiveness for the country since 1995 through 2005. Drawing on this information, Brownlee (2009) coded all these years as ‘competitive authoritarian’ in his dataset. The same is true for legislative indices for Uzbekistan (coded 7 from 1999 through 2005), and Kazakhstan (1995-2005) which also ended up in the competitive authoritarian group for these years although their scores for executive competitiveness were 3 for Kazakhstan (1992-2006), and 6 (1992-95) and later 3 (1996-2006) for Uzbekistan. All this, despite Brownlee’s (2009, p. 524) own coding rules: ‘Regime years that measured 1-4 in the DPI index were coded as fully closed authoritarian’ [It is unclear whether he uses an average score or a score on one of the indices] and despite the following statement on the same page: ‘Levitsky and Way explicitly excluded regimes like Egypt and Uzbekistan, where rulers enjoyed hegemonic electoral dominance’. The proper coding for all three regimes, especially after the year 2000, would be not ‘competitive authoritarian’ but ‘hegemonic authoritarian’. The former, as Brownlee (2009, p. 524) rightly pointed out, requires setting the standard higher. Conclusion As more and more scholars find it fruitful to use the hybrid regime framework in their analyses, it is important to match this interest with an adequate measurement of mixed regimes. This article puts forth a measurement method which can be used to identify various forms of hybrid regimes and applied universally. The alternative measurement method aims to overcome some of the flaws found in the existing measures of hybrid regimes in general and electoral authoritarian regimes in particular. The added value of this analysis is a new classification of post-Soviet regimes based on the measurement introduced. Political scientists have long studied political regimes as both the dependent variable (regime outcomes) and an explanatory factor (comparative 16 performance of various regime types) (Gasiorowski 1990, p. 110). Developing a good typology is the crucial step in any research interested in regime characteristics and trajectories of regime change. The classification of post-Soviet regimes presented in this article can be used to study patterns of regime change between the opposite poles of closed authoritarianism and liberal democracy. Our typology of post-Soviet regime types raises a number of questions that are relevant for theory building and further empirical testing. For example, what accounts for the diversity of hybrid regimes in the states of the former Soviet Union? Are particular (sub)types of hybrid regime more likely than others to democratise? Why has Russia’s competitive authoritarian regime, unlike others in the same region, remained so uniquely stable? What are the implications of different types of regime for socio-economic performance of post-Soviet states? Answering these questions is an important step towards developing a macropolitical theory of change and stability of hybrid regimes in the developing and postcommunist worlds. References - Adcock, R. & Collier, D. (2001) ‘Measurement Validity: A Shared Standard for Qualitative and Quantitative Research’, American Political Science Review, 95, 3, pp. 529-46. - Alvarez, M., Cheibub, J., Limongi, F. & Przeworski, A. (1996) ‘Classifying Political Regimes’, Studies in Comparative International Development, 31, 2, pp. 3-36. - Beichelt, T. (2004) ‘Autocracy and Democracy in Belarus, Russia and Ukraine’, Democratization, 11, 5, pp. 113-132. - Bogaards, M. (2007) ‘Measuring Democracy through Election Outcomes. A Critique with African Data’, Comparative Political Studies, 40, 10, pp. 1211-37. - Bogaards, M. (2009) ‘How to Classify Hybrid Regimes? Defective Democracy and Electoral Authoritarianism’, Democratization, 16, 2, pp. 399-423. - Brownlee, J. (2009) ‘Portents of Pluralism: How Hybrid Regimes Affect Democratic Transitions’, American Journal of Political Science, 53, 3, pp. 515-32. - Bunce, V. & Wolchik, S. (2008) ‘Mixed Regimes in Postcommunist Eurasia: Tipping Democratic and Tipping Authoritarian’, Paper delivered at the conference on ‘Democratization in European Former Soviet Republics: Limits, Obstacles and Perspectives’, June 13-15, Florence, Italy. - Bunce, V. & Wolchik, S. (2010) ‘Defeating Dictators: Electoral Change and Stability in Competitive Authoritarian Regimes’, World Politics, 62, 1, pp. 43-86. - Carothers, T. (2002) ‘The End of the Transition Paradigm’, Journal of Democracy, 13, 1, pp. 5-21. 17 - - - - - Cheibub, J.A., Gandhi, J. & Vreeland, J.R. (2009) ‘Democracy and Dictatorship Revisited’, Public Choice, 143, 1-2, pp. 67-101. Collier, D. & Adcock, R. (1999) ‘Democracy and Dichotomies: a Pragmatic Approach to Choices about Concepts’, Annual Review of Political Science, 2, pp. 537-65. Collier, D. & Levitsky, S. (1997) ‘Democracy with Adjectives: Conceptual Innovation in Comparative Research’, World Politics, 49, 3, pp. 430-51. Croissant, A. & Merkel, M. (2004) ‘Consolidated or Defective Democracy? Problems of Regime Change’, Democratization, Special Issue, 11, 5. Dahl, R.A. (1971) Polyarchy: Participation and Opposition (New Haven: Yale University Press). Diamond, L. (2002) ‘Thinking About Hybrid Regimes’, Journal of Democracy, 13, 2, pp. 21-35. Diamond, L. (2008) ‘The Democratic Rollback: The Resurgence of the Predatory State’, Foreign Affairs, March/April. Gasiorowski, M.J. (1990) ‘The Political Regimes Project’, Studies in Comparative International Development, 25, 1, pp. 109-25. Geddes, B. (1999) ‘What Do We Know about Democratization after Twenty Years?’ Annual Review of Political Science, 2, pp. 115-44. Gleditsch, K.S. & Ward, M.D. (1997). ‘Double Take. A Reexamination of Democracy and Autocracy in Modern Polities’, Journal of Conflict Resolution, 41, 3, pp. 361-83. Hartlyn, J. (2002) ‘Democracy and Consolidation in Contemporary Latin America: Current Thinking and Future Challenges’, in J.S. Tulchin and A. Brown (ed.) Democratic Governance and Social Inequality (Boulder, CO: Lynne Rienner Publishers), pp. 103-30. Hale, H. (2010) ‘Eurasian Polities as Hybrid Regimes: The Case of Putin’s Russia’, Journal of Eurasian Studies 1, pp. 33-41. Howard, M.M. & Roessler, P.G. (2006) ‘Liberalizing Electoral Outcomes in Competitive Authoritarian Regimes’, American Journal of Political Science, 50, 2, pp. 365-81. Huntington, S. (1991) The Third Wave: Democratization in the Late Twentieth Century (Norman; London: University of Oklahoma Press). Kara, A. (2007) ‘The Parliamentary Elections in Azerbaijan, November 2005 and May 2006’, Electoral Studies, 26, pp. 720-24. Karl, T.L. (1995) ‘The Hybrid Regimes of Central America’, Journal of Democracy, 6, 3, pp. 72-86. Keefer, P. (2007) DPI2006: Database of Political Institutions: Changes and Variable Definitions (New York: The World Bank). Levitsky, S. & Way, L.A. (2002) ‘The Rise of Competitive Authoritarianism’, Journal of Democracy, 13, 2, pp. 51-65. 18 - - - - - - - Linz, J.J. (1964) ‘An Authoritarian Regime: Spain’, in E. Allardt & Littunen, Y. (eds) Cleavages, Ideologies and Party Systems. Contributions to Comparative Political Sociology (Helsinki: Westermarck Society), pp. 291-341. Linz, J.J. (2000) Totalitarian and Authoritarian Regimes (Boulder, CO: Lynne Rienner Publishers). Linz, J. J. & Stepan, A. (1996) ‘Toward Consolidated Democracies’, Journal of Democracy, 7, 2, pp. 14-33. Linz, J.J. & Stepan, A. (1996) Problems of Democratic Transition and Consolidation: Southern Europe, South America, and Post-Communist Europe (Baltimore, MD: Johns Hopkins University Press). Mainwaring, S, Brinks, D. & Perez-Linan, A. (2001) ‘Classifying Political Regimes in Latin America, 1945-1999’, Studies in Comparative International Development, 36, 1, pp. 37-65. Marshall, M. & Jaggers, K. (2009) Polity IV Project, Political Regime Characteristics and Transitions, 1800-2007. Dataset Users’ Manual. Centre for Systemic Peace, available at www.systemicpeace.org/polity4 Munck, G. L. & Verkuilen, J. (2002) ‘Conceptualizing and Measuring Democracy. Evaluating Alternative Indices’, Comparative Political Studies, 35, 1, pp. 5-34. Munck, G.L. & Snyder, R. (2004) ‘Mapping Political Regimes: How Concepts We Use and the Way We Measure Them Shape the World We See’, Paper presented at the 2004 Annual APSA Meeting, Chicago, Sept.2-5, 2004. Morlino, L. (2009) ‘Are There Hybrid Regimes? Or Are They Just an Optical Illusion?’ European Political Science Review, 1, 2, pp. 273-96. O’Donnell, G. (1994) ‘Delegative Democracy’, Journal of Democracy, 5, 1, pp. 5569. OSCE. (2006) Final Report. Election Observation Mission, Republic of Azerbaijan, Parliamentary Elections 6 November 2005 (OSCE/ODIHR, Baku). Ottaway, M. (2003) Democracy Challenged: The Rise of Semi-Authoritarianism (Washington, DC: Carnegie Endowment for International Peace). Polity IV. Available at: http://www.systemicpeace.org/polity/polity4.htm Przeworski, A., Alvarez, M.E., Cheibub, J.A. & Limongi, F. (2000) Democracy and Development: Political Institutions and Well-Being in the World, 1950-1990 (Cambridge: Cambridge University Press). Sartori, G. (1970) ‘Concept Misformation in Comparative Politics’, American Political Science Review, 64, pp. 1033-53. Schedler, A. (2002) ‘The Menu of Manipulation’, Journal of Democracy, 13, 2, pp. 36-50. Schedler, A. (ed.) (2006a) Electoral Authoritarianism: The Dynamics of Unfree Competition (Boulder: Lynne Rienner Publishers). Schedler, A. (2006b) ‘The Logic of Electoral Authoritarianism’, in A. Schedler (ed.) (2006a), pp. 1-23. Schumpeter, J.A. (1954) Capitalism, Socialism and Democracy (London: Unwin University Books). 19 - Schmitter, P. & Karl, T.L. (1991) ‘What Democracy Is ... and Is Not’, Journal of Democracy, 2, 3, pp. 74-88. Snyder, R. (2006) ‘Beyond Electoral Authoritarianism: The Spectrum of Nondemocratic Regimes’, in A. Schedler (ed.) (2006a), pp. 219-231. Zakaria, F. (1997) ‘The Rise of Illiberal Democracy’, Foreign Affairs, 76, pp. 22-41. Zinecker, H. (2009) ‘Regime-Hybridity in Developing Countries: Achievements and Limitations of New Research on Transitions’, International Studies Review, 11, pp. 302-31. 20