I really enjoyed the Magaloni reading and I believe that her model adds a great deal to our understanding of hegemonic parties. I had a bit of trouble, however, with how she treated the role of ethnic identity in her analysis. She applies the same logic to ideology and ethnicity. She states that “opposition coordination requires that voters set aside their ideological (or ethnic) differences in order to dislodge the autocrat” (75). After distinguishing between nonideological (or tactical) opposition voters and ideological opposition voters, she claims that “the same logic can be applied to cases where the opposition is divided along ethnic lines. In these cases, opposition coordination will be harder to achieve when there are serious ethnic rivalries among opponents such that voters would rather waste their votes or support the ruling party than support an opposition party of a different ethnic identity”(76). I understand her general point about coordination, but I think she brushes over the strategic use of ethnicity by both the opposition and the regime. I think we have to explore why ethnic rivalries are significant in some contexts but not others. I think Magaloni is suggesting that the existence and strength of such rivalries are intrinsic characteristics of a particular country or culture (that can be overcome by the opposition or not), but such a view does not have a space for the creation or the modification of rivalries by politicians and parties for strategic reasons.
On a separate note, I would like to discuss the policy implications section of Wintrobe’s article. In discussing the policies of democratic countries toward tin-pot and totalitarian dictatorships, he assumes that “the only aim of Western governments is to promote freedom; that is, that the countries of the West have no strategic or economic interests in these countries other than the promotion of freedom for their people” (868). He notes that such a view is idealistic, but it strikes me as a definition that is so far from reality as to not be useful. I am curious if others had an issue with this section as well.
In last week's readings, Olson suggested that it’s best to be ruled by a stationary bandit, rather than a roving bandit. At least the stationary bandit, while maximizing his rents, provides public goods to improve productivity. The problem is that while providing core public goods to enhance economic productivity and maximize their tax revenue base, dictators threaten their own stability and durability. Finding the balance between providing political institutions that may garner the dictator political cooperation, loyalty and support (be it through policy concessions or rent-sharing) and securing one’s position as dictator in the regime is not simple. The opportunity cost of staying in office (extending policy concessions, increasing repression, etc.) may very well lead to a regime breakdown in the long run (or at least, fewer rents in the future as subjects’ welfare decreases). Przeworski and Gandhi, Bueno de Mesquita and Smith, and Wintrobe all discuss to some extent the mechanisms available to dictators to increase loyalty, cooperation and cooptation and the conditions under which certain mechanisms are preferred to others. But the long-run sustainability of using these mechanisms are not discussed. Dictators don’t have endless resources (although sometimes they may behave as if they did), and thus the options available at different instances to increase cooperation change (particularly when discussing regimes within the context of a fairly globalized world). A lesson to be learned from Kalyvas and Gelhbach and Keefer, is that the best bet for survival of authoritarian regimes (be they led by individuals or a single party) lies heavily on their flexibility and willingness to change, a rare attribute in this regime type.
On an additional note, referring back to Wintrobe’s distinction between tinpot dictatorships and totalitarian dictatorships, I had difficulty in accepting part of his definition of totalitarian regimes. The “motivated by utopian goals” aspect of totalitarian dictatorships is not very convincing. While transitions to authoritarian regimes may have been motivated by utopian goals, individual dictators usually are not. I have to agree with Przeworski and Gandhi, that dictators simply want to maximize rent. Consequently, power is a mechanism through which dictators can maximize. To my knowledge, there are not many poor dictators who pursued their ideology in history books. While the concept of a benevolent, social planner, maximizing aggregate social welfare may be appealing to theorists, in practice, authoritarian leaders are more interested in increasing their wealth, power, and in the Marcos family, Imelda's shoe collection.
Magaloni deemphasizes the argument that hegemonic-party regimes rely mainly on fraud and repression in order to retain political power. She argues instead that electoral autocracies are able to monopolize power by buying support among large parts of the poor population by super-majorities that ensure continued hegemony and repel potential opposition. While the generation of widespread support as the foundation for hegemonic-party regimes is a convincing argument, it is questionable to what extent we can discount the role of fraud and repression. Mexico’s powerful drug cartels are at an important juncture in the political game of vote buying and corruption. In return for turning a blind eye to the drug trade (worth approximately $30 bio per year), regional politicians receive significant financial and community support from the drug cartels based on which elections are won and opponents are repressed. The drug cartels influenced the outcome of elections during the reign of the PRI and their impact has increased steadily since 2000. Utilizing both widespread support as well as fraud and repression, the drug cartels are an important part of the explanation of the outcome of elections and warrants more attention in the context not only of Mexico’s political party landscape but that of other autocracies as well.
Wintrobe’s analysis of tin-pot versus totalitarian dictatorships is an extremely useful thought exercise into the tradeoff between repression and genuine legitimacy, introducing the valuable concept of the tin-pot autocrat. His simplistic use of loyalty and failure to account for changes between totalitarian and other types over time pose problems, however. First, Wintrobe’s concept of loyalty needs to be unpacked into two potentially distinctive elements: loyalty obtained through ideology (legitimacy) and cooption. Although, for tractability Wintrobe assumes that both of these components are unproblematically substitutable for each other, this assumption seriously degrades the results. If one believes, quite plausibly, that loyalty based on ideology (e.g. legitimacy) may result in lower repression costs and that its supply is less elastic to shifts in the “price” of loyalty, this might substantially change behavior of loyalty “supply” in the face of economic changes- one of Wintrobe’s primary findings. Although this is difficult to test, a common feature of the literature on Communist regime collapse (especially in Eastern Germany) focuses on the increases reliance on economic performance and economic cooptation as deligitimation increasingly set in and repression became more difficult. Likewise, willingness of security forces to repress also seems to decline the less ideologically driven they are (within China there is the excellent example of the divide between the less ideological Beijing army units and those called in from the countryside, although regional differences also mattered), thus driving up the cost. Indeed, this idea is central to Linz and Stepan’s valuable concept of Post-totalitarian regimes.
Even if we assume that the loyalty relationship is unproblematic, however, Wintrobe’s totalitarian category, as stated, does not usefully capture empirical reality. For example, Wintrobe is wrong when he claims that economic growth has led to more repression in China, which he labels as totalitarian. Fixing analysis only the pre-1990 era, in order to be perfectly fair to Wintrobe’s argument, the relationship between economic growth and repression is actually the opposite of that proposed. Repression decreases as growth picks up in the post-Mao era, only increasing at levels higher than the pre-Mao period at the time of Tienanmen and the 1987 Democracy Wall movement. Both of these, however, were reactions triggered by increasingly strong social movements pushing for further liberalization rather than deliberate programs launched to take advantage of new resources to repress at a higher level. Rather than suggesting that Wintrobe is completely wrong, however, this may suggest that a useful extension to his theory would allow for movement between the ideal type totalitarian and tin-pot. Incorporating the note made above about the two different components of loyalty, one could argue for a three-fold typological criteria that looks at the trade-offs between repression, co-option, and ideology (legitimacy?) and how each varies with economic changes. I would be especially interested in discussing how to measure the later (ideology/legitimacy) in class, if we have time, as there is an interesting emerging literature on testing legitimacy in autocratic settings.
Besley et al.’s and Gehoback et al.’s works both stem from the observation that economic development could happen in autocracies. They provide different explanations to this “puzzle”. Besley et al. contend that when the selectorate is secure in power, it can institutionalize good policy and remove poorly performing leaders from office. It is such ex post accountability enforced by the selectorate that brings about economic growth in autocracies. Conversely, Gehobach et al. hold that institutionalized ruling parties in some autocracies credibly make ex ante commitment not to expropriate thus promote economic investment. Both studies use the case of Chinese economic growth in the post-Mao era to support their relevant arguments. It seems to me that their arguments regarding the Chinese case would be more convincing if they took China’s broader institutional context into account.
In Besley et al.’s paper, Politburo of Chinese Communist Party, the selectorate in their models, maintained a de facto accountability system in central leaderships and thereby kept good policies and leadership going after 1976 when the charisma dictator, Mao Zedong passed away. This contention suffers from some problems. Most importantly, it is unclear where the power and interest of the selectorate come from. The selectorate in their argument seems to be like a black box and presumed to be cohesive and autonomous once they hold power. This is hardly the case in China: Politburo of Chinese Communist Party is usually comprised of different factions in the central leadership; it is more of a place for interest compromise or power balance than an autonomic force to hold top leaders accountable. A considerable portion of the Politburo members are occasionally recruited from different Chinese provinces which their interests are derived from and shaped by. Considering its fragmented nature, it is problematic to assume that Chinese Politburo has a solid power base and national-oriented interest to impose an accountability system within the central leadership.
Likewise, Gehobach et al.’s argument regarding the institutionalizing ruling party in China is somewhat vague without considering the party’s central-local system. I agree with them that Chinese cadres (insiders, in their models’ language) during the post-Mao era hold dual identities: they are members of the ruling group and shareholders of local enterprises at the same time. Such arrangement induced local government’s incentives to develop local economy and protect property rights in their jurisdictions. However, what made the threat of local cadres’collective action credible to the central government so that the center can restrain itself from confiscating the local’s revenues? Isn’t the insitutionalzed cadre evaluation system claimed by Gehobach et al. to empower local cadres also a weapon of the central state to control the cadres? In addition, it is noted that in the 1990s a much larger portion of capital comes from private entrepreneurs and foreign investors outside the central local government system. For these “outside investors”, it is the local cadres who they directly contact in daily business that are the commitment makers. In this case, whether the de facto property right protection enforced by the institutionalizing ruling party within the central local system, can still effectively make credible commitment not to expropriate investors’ profits is beyond their models’ explanation and prediction.
Kalyvas's criticism of the deterministic quality of some of the literature on the breakdown of single-party states in Eastern Europe seems constructive because it underscores the analytical value of separating persistent equilibria (e.g., "decay") from threshold events (e.g., "break-down"). This conceptual distinction seems obvious, yet as Svolvik points out, scholars of democratic survival make the same mistake by conflating the factors leading to democratic consolidation and those leading to authoritarian reversals.
Magaloni's detailed examination of the demise of the PRI in Mexico is a perfect example of why this distinction matters. She points out that, contrary to the findings of some of the preexisting literature, economic growth both reinforced and undermined the PRI. Crucially, in the face of these economic problems, the PRI's survival hinged on its ability to leverage its control of institutions to reward supporters and punish opponents. Its ultimate demise could not necessarily be seen as the inevitable consequence of a process of decay, but rather the result of some exogenous factor(s) that made the regime's "equilibrium" no longer tenable.
Given that "thresholds" are best understood separately from the processes that describe the functioning and survival of autocratic regimes, it would seem unwise for policymakers to underestimate the powerful equilibria that sustain many autocratic regimes. A perfect example of this is U.S. policy toward Cuba, which is premised on the idea that the Cuban regime will eventually collapse under its own weight. Fifty years later and despite years of economic decay in Cuba, we're still waiting.
Przeworski and Gandhi point out that while monarchies and military dictatorships have their own inherent institutions that "organize their rule and neutralize threats from within", civilian dictatorships do not. Thus, they must 'work harder' to achieve cooperation. Yet, as evidence by the case of Mexico's hegemonic party rule, the lack of such ready-made institutions seems to allow for much greater flexibility of civilian regimes. This flexibility allowed the PRI to create institutional incentives (such as the electoral succession rules) in order to co-opt potential challengers. It also allowed them to create a valuable source of information about the population and about their own strength. What I found most interesting in Magaloni's piece was description of how democratic elements such as elections were used as signals. For the PRI, voter turnout served as a signal indicating which areas should be rewarded for loyalty and which should be punished for disloyalty. For the potential opposition, electoral dominance of the PRI was used to signal the ( usually low) chances of successful opposition. With regards to Wintrobe's article, which uses an explicit economic theory of dictatorship, I was left wondering how the prices/cost of repression versus investments in loyalty are signaled. While he discusses how different dictatorial regime types respond to changes in the price of these, it would be interesting to see if/ how different regimes institutionalize these signals such as the PRI did. Observations of strikes, protests and other kinds of mobilizations obviously are one source of signaling of the weakness of the government, but these seem to be far more susceptible to misinterpretation in terms of their true threat to the regime than the institutional structures devised by the PRI. Would regimes lacking these institutional mechanisms be more likely to misconstrue their strengths and vulnerabilities?
I agree with Gehlback and Keefer’s overall argument that institutionalized ruling parties allow autocrats to make credible commitments to investors. But I am not convinced by the working of this model: solving collective action problems (i.e. making cadres believe the autocratic will not expropriate their investment) because their interpretation of the Chinese case is inappropriate (I think they are wrong but I retain the possibility that I’m the one who’s wrong, since I am not an expert of modeling and might have misunderstood their argument). First, the personnel reform was not designed to limit the size of “designated cadres” so the party leaders can credibly commit to the elites. It was to strengthen the viability of the party, forcing the old to retire and recruiting the young and educated. Second, they neglect the fact that a cadre management system has long existed. What has changed is the criteria of rewards—from ideology (which is subject to the discretion of the paramount leader and unstable) to economic growth. I’m not sure of the role of collective action problem in economic development. Third, TVEs is not a proper example, since they are collectively owned and perfectly legitimate under the communist regime. I don’t see the risk of expropriation (how do we expropriate publicly owned properties?). Fourth, there is a great transformation in the composition of investment, and I am not sure if their model is able to explain this change. For example, in 2007 private investment comprised around 57% of China’s domestic investment, whereas state owned enterprises made up 29.8%, FDI 10%, and TVEs 3.2%. That is a lot of “insiders.”
Whilst all the readings this week examined the mechanisms of authoritarian regime maintenance and decline, I found it difficult to reconcile the respective analyses because alternate explanations seem to be driven largely by choice of regime classification which is difficult to transpose across studies.
Regime typologies aside, I found the Svoliks article on “Authoritarian Reversals and Democratic Consolidation(2008) the most compelling. By distinguishing between ‘consolidated’ and ‘unconsolidated’ (or transitional) democracies he is able to define two discrete causal mechanisms associated with regime survival – (i) the likelihood that a regime consolidates (and thus continues as a democracy) and (ii) the likelihood that a regime reverts to autocracy. Using this simple dichotomy and some innovate statistical approaches he constructs a elegant model that appears to effectively reconcile some of the disciplines most hotly debated controversies – including the role of economic growth on democratization (Boix and Stokes 2003; Przeworksi and Limongi 1997, 2000) and the impact of political institutions (e.g. presidentialism vs parliamentarism, and prior regime type) on the survival of democracies (Boix 2003, Cheibub 2007). Whilst his argument was partially incomplete and unsatisfying for its inability to effectively define ‘consolidation’ and to propose key measures that would distinguish it from a transitional democracy, I could find little fault with the underlying logic and found it intuitively appealing.
In contrast, I found it more difficult to compare the respective analysis of Wintrobe (1990), Gandhi and Przeworksi (2006) and de Mesquia and Smith (2009). As mentioned above, this is largely because of their different approaches to regime categorization that seems to also lead to an emphasis on different independent variables. Wintrobe models the behavior of dictatorships along a continuum from ‘tinpot’ (i.e. minimizing cost/maximizing rents) to totalitarianism (maximizing power) showing how they respond differently to changes in economic performance through complementary strategies of loyalty (rentsharing) and repression. His conclusion is that when economic performance declines tinpot regimes increase repression (decrease loyalty/rent sharing) whilst totalitarian regimes decrease repression (increase loyalty/rent sharing). In contrast, Gandhi and Przeworksi (2006) assume an alternate, but similar continuum of autocracy – from purely autocratic, to those that encompass democratic institutions. In this instance however they look at tools employed by leaders to deal with revolution suggesting that they involve a combination of policy concessions and rent sharing (notably there is no explicit treatment of repression in the model). This analysis focuses on how institutions (parties, legislatures, elections) emerge to facilitate cooption and cooperation even under otherwise autocratic regimes. Finally, de Mesquita’s analysis which does not distinguish between different types of autocracy, appears to suggest that the relative balance between rent-sharing and repression in the face of revolutionary pressures is dictated by the regimes source of revenues. ‘Free sources ‘ (eg natural resources or international aid) tend to reduce public good expenditures and increase repression in the face of revolution. In their absence, however regimes respond to revolutionary pressures by increasing provision of pubic goods and democratizing. These three similar but clearly different analyses are illustrative of how highly sensitive empirical assumptions are to typologies (in this case regime classification) suggesting the need for a consensus on regime taxonomies for consistency and enhanced cross study comparison.
In “Cooperation, Cooptation and Rebellion under Dictatorships,” Jennifer Gandhi and Adam Przeworski argues that when dictators need more cooperation, they make more policy concessions and distribute fewer rents; when the threat of rebellion is higher, they make larger concessions and distribute more spoil(1). While they present a thought-provoking analysis, questions remain. First, is it right for them to assume that the mere presence of parties in the legislature is an indicator of policy concessions and that more parties should increase “in the need for cooperation and strength of the opposition” (2)? The opposition, however, may not necessarily come through an institutional structure, like a party. Even if it comes from institutional structures, what make political parties the most likely vehicle? Other organized groups, other governmental institutions or the military could also be institutional structures that could affect regime opposition and the need for the dictator to offer compromise. Further, Gandhi et al. state that they see “no reason to believe that dictators have a lexicographic preference for policy over rents” (13). What are the factors that cause a dictator to choose sharing rents over policy compromise or vice versa? Their analysis on that point is unclear to me. Moreover, how does Gandhi et al’s work influence our reading of Magaloni’s “Voting for Autocracy?” In “Voting,” Magoloni presents a thought-provoking account of why hegemonic parties, particularly “electoral autocracies” like PRI in Mexico, survive. In Chapters 3 and 4, Magaloni explains how the PRI used budget cycles and patronage through the mechanism of poverty programs to enhance support. Can we apply Gandhi et al’s teachings to Magaloni’s work? Would Magaloni’s work be more complete if she had focused more on explaining how policy concessions have affected PRI’s ability to stay in power? Magaloni emphasizes the importance of supermajorities in solidifying the PRI’s power. How should Gandhi et al. have incorporated coalition size in their argument? For instance, Bruce Bueno de Mesquita argues that foreign aid enhances democratization when leaders face a revolutionary threat and when coalition size is substantial; if coalition size is small, aid inhibits democratization. How would the ability of a dictator to get a lot of foreign aid affect the calculus in Gandhi et al’s story?
Milan Svolik uses survival analysis to study what he describes as two separate mechanisms by which democracies resist reversal to authoritarianism: "democratic consolidation, which practically eliminates the risk of an authoritarian reversal," and "a separate mechanism that prevents authoritarian reversals in those democracies that are not consolidated."
Svolik's analysis seems quite helpful in that this technique seems to fit the data (counts of democracy age and timing of reversals) more closely than other analyses have done. He also makes an important point-- analyses that do not distinguish consolidated democracies from transitional ones overestimate the predicted age of the average democratic regime from time of democratization, with the overall estimates skewed by including old and transitional democracies in one analysis.
While I think his purpose is a good one--to use the consolidation factor to separate analyses of old and new democracies, two aspects of the study seem quite problematic to me.
First, it is not really clear that the mechanism(s) helping transitional democracies resist breakdown are different from those making democracies consolidate, and thus (in Svolik's terms) fully resistant to breakdown. In fact, it seems that when a transitional democracy resists reversal in circumstances that cause other such regimes to break down, it is actively consolidating itself--not necessarily advancing it into the surefire 'consolidated' category Svolik advances, but is taking yet another step towards that status.
Second, I am concerned about what consolidation is, and how it is measured here. While Svolik says he shows that consolidation (defined by being fully resistant to reversal) is associated with a democracy's age, the whole concept seems to be quickly defined by age: after estimating the likelihood of being resistant at different regime ages, he seems to use the curve from that aggregate analysis to determine whether a regime can be thought of as consolidated, by combining age and the estimated rate of growth in resistance over time (it increases most in the first 20 years after democratization) to estimate the likelihood that each regime is consolidated. Yet we assume that the mechanism(s) within consolidated regimes that are making them resistant manifest themselves with varying timing in different regimes in different time periods, regions, economic conditions, etc; thus, using a model to analyze what affects regime consolidation seems not very helpful when the coding of consolidated regimes is dependent on aggregates of non-reversal in the first place. (I welcome any defense of the logic of his model any of you have, as I like many things about this approach and would like to understand it better.)
I really like Magaloni’s piece –especially her use of examples in the text and one big case study-. Looking at the hegemonic party survival mechanism, especially the role of elections raised two main questions for me.
First, her discussion of the definition of “hegemonic party autocracy” highlights how the debate of regimes’ classification is still unclear and open. It is not clear to me that she gives a conclusive definition of what she studies, the hegemonic party autocracy. The difference between an hegemonic party autocracy and an “uncommon democracy” is not clear to me, because as she points out, it relies on making the assumption that alternation would happen in one case and not in the other case, which can’t be empirically observed. Thus the transition to democracy from a hegemonic party autocracy can be viewed as an alternation to another hegemonic party autocracy as the example of Kenya (transitional election of 2002 and reelection of Kibaki in 2007) seems to show.
This brings me to my second point. I feel she dismisses quite quickly the Huntington argument of hegemonic party survival: the historic origins. Political identities are constructed through history and I believe that a society whose identities are constructed in social classes is different from whose construction is based on ethnic differences. As she mentioned, the elite is the key element to carry opposition from the large coalition of the hegemonic party. But the elite can be characterized as the “rich” against the “poor” that would be the mass, or as a certain ethnicity that would then only represent a percentage of the mass. It seems to me then, that, given the institutional manipulation to maintain power, hegemonic party autocracies based on ethnic differences would not be able to transition to democracy and would always have to transition to another hegemonic party autocracy based on another ethnic coalition. The African examples (Kenya, Ivory Coast) are quite illustrative of this.
How different is the policymaking process of an autocracy with respect to a democratic one? From a veto player perspective, one could argue that autocracies are usually characterized as regimes with only one effective political actor (e.g. the dictator), in contrast to democracies, where the number of veto players is at least two (e.g. the government, an opposition party). From this comparison, all that matters in the policymaking process of an autocracy are the preferences of the single ruler, and thus, there is no need to study institutions within the regime. This simplistic treatment leads one to expect in autocracies outcomes such as high discounts rates, policy instability, uncertainty and poor economic performance relative to democracies (Olson 1993). However, this week’s readings show striking similarities between the policymaking processes of autocracies and democracies. First, autocratic leaders are ambitious: they care about office and survival just as much as democratically elected leaders (Wintrope, Bueno de Mesquita and Smith). Second, they seem to face the same credibility and commitment problems that presidents do: mainly, how to appear credible to important economic actors such as investors and guarantee that their assets will not be expropriated (Gehlbach and Keefer). In spite of the absence of accountability mechanisms, autocratic leaders can overcome these problems through institutional innovations (e.g. parties) that substitute for the presence of formal checks and balances like legislatures. In line with this last point, autocrats don’t usually govern alone, and many do so through political parties where internal power competition is just as fierce as the inter party competion we observe in democracies. Fourth, to maintain power, autocrats need to deliver goods in the form of growth and patronage (Magaloni), policy concessions and rents (Ghandi and Przeworski) that mimic the electoral connection between legislators and voters in the most advanced democratic settings. Finally, as in democracies, policy outcomes depend in part on the details of the institutional scenario: it matters for example whether there is a legislature or not, and the characteristics of parties to explain economic performance in autocracies. Thus, when it comes to the logic of policymaking, autocracies seems to function in a manner that resembles many aspects of democratic regimes.
(I have been experiencing patchy internet at home, so apologies for the late posting).
When reading the literature on regime change and economic development, and related bodies of work, I have often have thought about the ways in which the heterogeneity of democratic regime types affect the general patterns we see empirically or the implications of the models we examine, or both.
I think Svolik makes a valuable contribution with his article in this respect. First and foremost, he further breaks down and elaborates upon the conceptual distinction that Przeworski et al. introduced in their work, namely between what makes democracy emerge and what sustains it. By establishing a dichotomy between consolidated democracies, on the one hand, and transitional democracies on the other, Svolik is able to capture, in my opinion, the heterogeneity between democratic regime types and the important implications this has. Crucially, whether we agree with the finer points of his analysis or not, he shows how encompassing all extant democracies under a single bracket or categorisation in our analyses can be problematic and miss some of what we are trying to explain. I thought it was interesting and useful to see how he could thereby break down the major implication that emerged from the recent literature on democratization and development – that there is a positive association between economic development and democratic survival – and ask whether this is primarily because “(1) wealth improves the odds that a democracy consolidates, (2) a high level of economic development actually lowers the hazard of reversals in transitional democracies, or (3) both effects hold?” (p. 155). Przeworski et al, in their work, argued strongly in favour of (1) and Svolik finds the first alternative to be the correct answer as well. I just like the fact that the conceptual distinction he introduces allows him to posit different alternatives and distinguish between two separate, though related, causal mechanisms for why this might be so, in a manner that I think expands upon Przeworski et al.
Like Emily, however, I also have some concerns with the way consolidation is specified (or not). I think she articulated it well, so I will not say much more about it here. On a separate note, I was also wondering whether it would be constructive to think of the distinction that Svolik models in terms of an continuum between democracies in danger of reversal and those who are highly unlikely to do so. I mention this because I am somewhat sceptical of the way in which he divides democracies quite bluntly into two separate groups (If I have missed some subtle point here, please let me know!). In theoretical terms, at least – and this would have to be ascertained with an examination of the empirical evidence – I think there is a gradation between democracies in terms of both how consolidated they are (depending upon how we would specify that) and their susceptibility to a reversal. Though an improvement upon the analyses which group all democracies together as one, regardless of the differences between them, Svolik’s account may not go far enough in capturing the heterogeneity and the implications of that as we might desire. I think it is a useful simplification for the purpose of his analysis, but it appears to me that there are also some mixed cases, which would lie in the middle of such a spectrum, or at least somewhere along a continuum between the two poles.
I thought that the readings this week were examples of a very dynamic and exciting literature. To me it seems that the study of decision-making and policy outcomes in autocracies will soon be just as rich as that with respect to democracies. I was particularly impressed by the combination of formal theory and empirics in much of the literature. It also seems to me that the absence of electoral institutions (which are a necessary part of models of democracy) has allowed theorists of autocracies to develop a much wider set of models and theories of political incentives than is available or proper for theorists of democracies.
While I think that the renewed focus on autocracies is timely and very important, I do think however that the benefits of the separation of the literature into autocratic and democratic segments will eventually be outweighed by the costs. Specifically, it seems to me that other political institutions, such as the quality of bureaucracy, state capacity, political fractionalisation, and international alliances and commitments can and do have an effect on policy in addition to the extent of political accountability to the electorate. We must be careful that we are not captured by the autocratic/democratic distinction, that this scale is not the only scale of institutional character that we regularly consider with respect to policy outcomes.
The Svolik piece on “Authoritarian Reversals and Democratic Consolidation” was particularly interesting. I agree with the author’s argument regarding the importance of taking into account the diverse reasons that account for the survival of democratic institutions. The author correctly notes in some instances that survival may be attributed to the fact that the democracy has truly consolidated. On the other hand, survival can also be attributed to the fact that the regime is still in transition, but continues to exist because of other self-enforcing conditions. While the causal mechanisms accounting for the survival of each type of democracy may be the same, they will operate differently for each regime type. The author does an excellent job in showing the divergence of results between the ‘comprehensive’ and ‘split population’ ‘democratic’ categories.
Something that I found problematic about this piece was the author’s hesitance to provide a conceptual explanation of what exactly accounts for ‘democratic consolidation.’ I understand the author’s claim of an ‘identification problem’ in this respect. However, I was left wondering what the policy implications of the results would be when the ‘end goal’ of consolidation was never clearly defined. For example, the author finds that states that were military dictatorships have a higher likelihood of an authoritarian reversal and a lower likelihood of democratic consolidation. In another statistical test the author shows that economic growth and development enhance democratic transitions and consolidation. Unfortunately, the significance of the results stops here. From a policy perspective, what steps should be taken to turn economic growth into democratic consolidation? Why are military regimes problematic for the democratic consolidation? What are the implications of a military legacy? In all fairness, this does not appear to have been research agenda of the author. However, it would have made the results more interesting if these causal mechanisms were connected to a clearly defined concept of ‘consolidation.’ Perhaps the inclusion of case studies would have been helpful in answering some of these questions.
I really enjoyed the Magaloni reading and I believe that her model adds a great deal to our understanding of hegemonic parties. I had a bit of trouble, however, with how she treated the role of ethnic identity in her analysis. She applies the same logic to ideology and ethnicity. She states that “opposition coordination requires that voters set aside their ideological (or ethnic) differences in order to dislodge the autocrat” (75). After distinguishing between nonideological (or tactical) opposition voters and ideological opposition voters, she claims that “the same logic can be applied to cases where the opposition is divided along ethnic lines. In these cases, opposition coordination will be harder to achieve when there are serious ethnic rivalries among opponents such that voters would rather waste their votes or support the ruling party than support an opposition party of a different ethnic identity”(76). I understand her general point about coordination, but I think she brushes over the strategic use of ethnicity by both the opposition and the regime. I think we have to explore why ethnic rivalries are significant in some contexts but not others. I think Magaloni is suggesting that the existence and strength of such rivalries are intrinsic characteristics of a particular country or culture (that can be overcome by the opposition or not), but such a view does not have a space for the creation or the modification of rivalries by politicians and parties for strategic reasons.
ReplyDeleteOn a separate note, I would like to discuss the policy implications section of Wintrobe’s article. In discussing the policies of democratic countries toward tin-pot and totalitarian dictatorships, he assumes that “the only aim of Western governments is to promote freedom; that is, that the countries of the West have no strategic or economic interests in these countries other than the promotion of freedom for their people” (868). He notes that such a view is idealistic, but it strikes me as a definition that is so far from reality as to not be useful. I am curious if others had an issue with this section as well.
In last week's readings, Olson suggested that it’s best to be ruled by a stationary bandit, rather than a roving bandit. At least the stationary bandit, while maximizing his rents, provides public goods to improve productivity. The problem is that while providing core public goods to enhance economic productivity and maximize their tax revenue base, dictators threaten their own stability and durability. Finding the balance between providing political institutions that may garner the dictator political cooperation, loyalty and support (be it through policy concessions or rent-sharing) and securing one’s position as dictator in the regime is not simple. The opportunity cost of staying in office (extending policy concessions, increasing repression, etc.) may very well lead to a regime breakdown in the long run (or at least, fewer rents in the future as subjects’ welfare decreases). Przeworski and Gandhi, Bueno de Mesquita and Smith, and Wintrobe all discuss to some extent the mechanisms available to dictators to increase loyalty, cooperation and cooptation and the conditions under which certain mechanisms are preferred to others. But the long-run sustainability of using these mechanisms are not discussed. Dictators don’t have endless resources (although sometimes they may behave as if they did), and thus the options available at different instances to increase cooperation change (particularly when discussing regimes within the context of a fairly globalized world). A lesson to be learned from Kalyvas and Gelhbach and Keefer, is that the best bet for survival of authoritarian regimes (be they led by individuals or a single party) lies heavily on their flexibility and willingness to change, a rare attribute in this regime type.
ReplyDeleteOn an additional note, referring back to Wintrobe’s distinction between tinpot dictatorships and totalitarian dictatorships, I had difficulty in accepting part of his definition of totalitarian regimes. The “motivated by utopian goals” aspect of totalitarian dictatorships is not very convincing. While transitions to authoritarian regimes may have been motivated by utopian goals, individual dictators usually are not. I have to agree with Przeworski and Gandhi, that dictators simply want to maximize rent. Consequently, power is a mechanism through which dictators can maximize. To my knowledge, there are not many poor dictators who pursued their ideology in history books. While the concept of a benevolent, social planner, maximizing aggregate social welfare may be appealing to theorists, in practice, authoritarian leaders are more interested in increasing their wealth, power, and in the Marcos family, Imelda's shoe collection.
Magaloni deemphasizes the argument that hegemonic-party regimes rely mainly on fraud and repression in order to retain political power. She argues instead that electoral autocracies are able to monopolize power by buying support among large parts of the poor population by super-majorities that ensure continued hegemony and repel potential opposition. While the generation of widespread support as the foundation for hegemonic-party regimes is a convincing argument, it is questionable to what extent we can discount the role of fraud and repression. Mexico’s powerful drug cartels are at an important juncture in the political game of vote buying and corruption. In return for turning a blind eye to the drug trade (worth approximately $30 bio per year), regional politicians receive significant financial and community support from the drug cartels based on which elections are won and opponents are repressed. The drug cartels influenced the outcome of elections during the reign of the PRI and their impact has increased steadily since 2000. Utilizing both widespread support as well as fraud and repression, the drug cartels are an important part of the explanation of the outcome of elections and warrants more attention in the context not only of Mexico’s political party landscape but that of other autocracies as well.
ReplyDeleteWintrobe’s analysis of tin-pot versus totalitarian dictatorships is an extremely useful thought exercise into the tradeoff between repression and genuine legitimacy, introducing the valuable concept of the tin-pot autocrat. His simplistic use of loyalty and failure to account for changes between totalitarian and other types over time pose problems, however. First, Wintrobe’s concept of loyalty needs to be unpacked into two potentially distinctive elements: loyalty obtained through ideology (legitimacy) and cooption. Although, for tractability Wintrobe assumes that both of these components are unproblematically substitutable for each other, this assumption seriously degrades the results. If one believes, quite plausibly, that loyalty based on ideology (e.g. legitimacy) may result in lower repression costs and that its supply is less elastic to shifts in the “price” of loyalty, this might substantially change behavior of loyalty “supply” in the face of economic changes- one of Wintrobe’s primary findings. Although this is difficult to test, a common feature of the literature on Communist regime collapse (especially in Eastern Germany) focuses on the increases reliance on economic performance and economic cooptation as deligitimation increasingly set in and repression became more difficult. Likewise, willingness of security forces to repress also seems to decline the less ideologically driven they are (within China there is the excellent example of the divide between the less ideological Beijing army units and those called in from the countryside, although regional differences also mattered), thus driving up the cost. Indeed, this idea is central to Linz and Stepan’s valuable concept of Post-totalitarian regimes.
ReplyDeleteEven if we assume that the loyalty relationship is unproblematic, however, Wintrobe’s totalitarian category, as stated, does not usefully capture empirical reality. For example, Wintrobe is wrong when he claims that economic growth has led to more repression in China, which he labels as totalitarian. Fixing analysis only the pre-1990 era, in order to be perfectly fair to Wintrobe’s argument, the relationship between economic growth and repression is actually the opposite of that proposed. Repression decreases as growth picks up in the post-Mao era, only increasing at levels higher than the pre-Mao period at the time of Tienanmen and the 1987 Democracy Wall movement. Both of these, however, were reactions triggered by increasingly strong social movements pushing for further liberalization rather than deliberate programs launched to take advantage of new resources to repress at a higher level. Rather than suggesting that Wintrobe is completely wrong, however, this may suggest that a useful extension to his theory would allow for movement between the ideal type totalitarian and tin-pot. Incorporating the note made above about the two different components of loyalty, one could argue for a three-fold typological criteria that looks at the trade-offs between repression, co-option, and ideology (legitimacy?) and how each varies with economic changes. I would be especially interested in discussing how to measure the later (ideology/legitimacy) in class, if we have time, as there is an interesting emerging literature on testing legitimacy in autocratic settings.
Besley et al.’s and Gehoback et al.’s works both stem from the observation that economic development could happen in autocracies. They provide different explanations to this “puzzle”. Besley et al. contend that when the selectorate is secure in power, it can institutionalize good policy and remove poorly performing leaders from office. It is such ex post accountability enforced by the selectorate that brings about economic growth in autocracies. Conversely, Gehobach et al. hold that institutionalized ruling parties in some autocracies credibly make ex ante commitment not to expropriate thus promote economic investment. Both studies use the case of Chinese economic growth in the post-Mao era to support their relevant arguments. It seems to me that their arguments regarding the Chinese case would be more convincing if they took China’s broader institutional context into account.
ReplyDeleteIn Besley et al.’s paper, Politburo of Chinese Communist Party, the selectorate in their models, maintained a de facto accountability system in central leaderships and thereby kept good policies and leadership going after 1976 when the charisma dictator, Mao Zedong passed away. This contention suffers from some problems. Most importantly, it is unclear where the power and interest of the selectorate come from. The selectorate in their argument seems to be like a black box and presumed to be cohesive and autonomous once they hold power. This is hardly the case in China: Politburo of Chinese Communist Party is usually comprised of different factions in the central leadership; it is more of a place for interest compromise or power balance than an autonomic force to hold top leaders accountable. A considerable portion of the Politburo members are occasionally recruited from different Chinese provinces which their interests are derived from and shaped by. Considering its fragmented nature, it is problematic to assume that Chinese Politburo has a solid power base and national-oriented interest to impose an accountability system within the central leadership.
Likewise, Gehobach et al.’s argument regarding the institutionalizing ruling party in China is somewhat vague without considering the party’s central-local system. I agree with them that Chinese cadres (insiders, in their models’ language) during the post-Mao era hold dual identities: they are members of the ruling group and shareholders of local enterprises at the same time. Such arrangement induced local government’s incentives to develop local economy and protect property rights in their jurisdictions. However, what made the threat of local cadres’collective action credible to the central government so that the center can restrain itself from confiscating the local’s revenues? Isn’t the insitutionalzed cadre evaluation system claimed by Gehobach et al. to empower local cadres also a weapon of the central state to control the cadres? In addition, it is noted that in the 1990s a much larger portion of capital comes from private entrepreneurs and foreign investors outside the central local government system. For these “outside investors”, it is the local cadres who they directly contact in daily business that are the commitment makers. In this case, whether the de facto property right protection enforced by the institutionalizing ruling party within the central local system, can still effectively make credible commitment not to expropriate investors’ profits is beyond their models’ explanation and prediction.
Kalyvas's criticism of the deterministic quality of some of the literature on the breakdown of single-party states in Eastern Europe seems constructive because it underscores the analytical value of separating persistent equilibria (e.g., "decay") from threshold events (e.g., "break-down"). This conceptual distinction seems obvious, yet as Svolvik points out, scholars of democratic survival make the same mistake by conflating the factors leading to democratic consolidation and those leading to authoritarian reversals.
ReplyDeleteMagaloni's detailed examination of the demise of the PRI in Mexico is a perfect example of why this distinction matters. She points out that, contrary to the findings of some of the preexisting literature, economic growth both reinforced and undermined the PRI. Crucially, in the face of these economic problems, the PRI's survival hinged on its ability to leverage its control of institutions to reward supporters and punish opponents. Its ultimate demise could not necessarily be seen as the inevitable consequence of a process of decay, but rather the result of some exogenous factor(s) that made the regime's "equilibrium" no longer tenable.
Given that "thresholds" are best understood separately from the processes that describe the functioning and survival of autocratic regimes, it would seem unwise for policymakers to underestimate the powerful equilibria that sustain many autocratic regimes. A perfect example of this is U.S. policy toward Cuba, which is premised on the idea that the Cuban regime will eventually collapse under its own weight. Fifty years later and despite years of economic decay in Cuba, we're still waiting.
This comment has been removed by the author.
ReplyDeletePrzeworski and Gandhi point out that while monarchies and military dictatorships have their own inherent institutions that "organize their rule and neutralize threats from within", civilian dictatorships do not. Thus, they must 'work harder' to achieve cooperation. Yet, as evidence by the case of Mexico's hegemonic party rule, the lack of such ready-made institutions seems to allow for much greater flexibility of civilian regimes. This flexibility allowed the PRI to create institutional incentives (such as the electoral succession rules) in order to co-opt potential challengers. It also allowed them to create a valuable source of information about the population and about their own strength.
ReplyDeleteWhat I found most interesting in Magaloni's piece was description of how democratic elements such as elections were used as signals. For the PRI, voter turnout served as a signal indicating which areas should be rewarded for loyalty and which should be punished for disloyalty. For the potential opposition, electoral dominance of the PRI was used to signal the ( usually low) chances of successful opposition.
With regards to Wintrobe's article, which uses an explicit economic theory of dictatorship, I was left wondering how the prices/cost of repression versus investments in loyalty are signaled. While he discusses how different dictatorial regime types respond to changes in the price of these, it would be interesting to see if/ how different regimes institutionalize these signals such as the PRI did. Observations of strikes, protests and other kinds of mobilizations obviously are one source of signaling of the weakness of the government, but these seem to be far more susceptible to misinterpretation in terms of their true threat to the regime than the institutional structures devised by the PRI. Would regimes lacking these institutional mechanisms be more likely to misconstrue their strengths and vulnerabilities?
I agree with Gehlback and Keefer’s overall argument that institutionalized ruling parties allow autocrats to make credible commitments to investors. But I am not convinced by the working of this model: solving collective action problems (i.e. making cadres believe the autocratic will not expropriate their investment) because their interpretation of the Chinese case is inappropriate (I think they are wrong but I retain the possibility that I’m the one who’s wrong, since I am not an expert of modeling and might have misunderstood their argument).
ReplyDeleteFirst, the personnel reform was not designed to limit the size of “designated cadres” so the party leaders can credibly commit to the elites. It was to strengthen the viability of the party, forcing the old to retire and recruiting the young and educated. Second, they neglect the fact that a cadre management system has long existed. What has changed is the criteria of rewards—from ideology (which is subject to the discretion of the paramount leader and unstable) to economic growth. I’m not sure of the role of collective action problem in economic development. Third, TVEs is not a proper example, since they are collectively owned and perfectly legitimate under the communist regime. I don’t see the risk of expropriation (how do we expropriate publicly owned properties?). Fourth, there is a great transformation in the composition of investment, and I am not sure if their model is able to explain this change. For example, in 2007 private investment comprised around 57% of China’s domestic investment, whereas state owned enterprises made up 29.8%, FDI 10%, and TVEs 3.2%. That is a lot of “insiders.”
Whilst all the readings this week examined the mechanisms of authoritarian regime maintenance and decline, I found it difficult to reconcile the respective analyses because alternate explanations seem to be driven largely by choice of regime classification which is difficult to transpose across studies.
ReplyDeleteRegime typologies aside, I found the Svoliks article on “Authoritarian Reversals and Democratic Consolidation(2008) the most compelling. By distinguishing between ‘consolidated’ and ‘unconsolidated’ (or transitional) democracies he is able to define two discrete causal mechanisms associated with regime survival – (i) the likelihood that a regime consolidates (and thus continues as a democracy) and (ii) the likelihood that a regime reverts to autocracy. Using this simple dichotomy and some innovate statistical approaches he constructs a elegant model that appears to effectively reconcile some of the disciplines most hotly debated controversies – including the role of economic growth on democratization (Boix and Stokes 2003; Przeworksi and Limongi 1997, 2000) and the impact of political institutions (e.g. presidentialism vs parliamentarism, and prior regime type) on the survival of democracies (Boix 2003, Cheibub 2007). Whilst his argument was partially incomplete and unsatisfying for its inability to effectively define ‘consolidation’ and to propose key measures that would distinguish it from a transitional democracy, I could find little fault with the underlying logic and found it intuitively appealing.
In contrast, I found it more difficult to compare the respective analysis of Wintrobe (1990), Gandhi and Przeworksi (2006) and de Mesquia and Smith (2009). As mentioned above, this is largely because of their different approaches to regime categorization that seems to also lead to an emphasis on different independent variables. Wintrobe models the behavior of dictatorships along a continuum from ‘tinpot’ (i.e. minimizing cost/maximizing rents) to totalitarianism (maximizing power) showing how they respond differently to changes in economic performance through complementary strategies of loyalty (rentsharing) and repression. His conclusion is that when economic performance declines tinpot regimes increase repression (decrease loyalty/rent sharing) whilst totalitarian regimes decrease repression (increase loyalty/rent sharing). In contrast, Gandhi and Przeworksi (2006) assume an alternate, but similar continuum of autocracy – from purely autocratic, to those that encompass democratic institutions. In this instance however they look at tools employed by leaders to deal with revolution suggesting that they involve a combination of policy concessions and rent sharing (notably there is no explicit treatment of repression in the model). This analysis focuses on how institutions (parties, legislatures, elections) emerge to facilitate cooption and cooperation even under otherwise autocratic regimes. Finally, de Mesquita’s analysis which does not distinguish between different types of autocracy, appears to suggest that the relative balance between rent-sharing and repression in the face of revolutionary pressures is dictated by the regimes source of revenues. ‘Free sources ‘ (eg natural resources or international aid) tend to reduce public good expenditures and increase repression in the face of revolution. In their absence, however regimes respond to revolutionary pressures by increasing provision of pubic goods and democratizing. These three similar but clearly different analyses are illustrative of how highly sensitive empirical assumptions are to typologies (in this case regime classification) suggesting the need for a consensus on regime taxonomies for consistency and enhanced cross study comparison.
In “Cooperation, Cooptation and Rebellion under Dictatorships,” Jennifer Gandhi and Adam Przeworski argues that when dictators need more cooperation, they make more policy concessions and distribute fewer rents; when the threat of rebellion is higher, they make larger concessions and distribute more spoil(1). While they present a thought-provoking analysis, questions remain. First, is it right for them to assume that the mere presence of parties in the legislature is an indicator of policy concessions and that more parties should increase “in the need for cooperation and strength of the opposition” (2)? The opposition, however, may not necessarily come through an institutional structure, like a party. Even if it comes from institutional structures, what make political parties the most likely vehicle? Other organized groups, other governmental institutions or the military could also be institutional structures that could affect regime opposition and the need for the dictator to offer compromise. Further, Gandhi et al. state that they see “no reason to believe that dictators have a lexicographic preference for policy over rents” (13). What are the factors that cause a dictator to choose sharing rents over policy compromise or vice versa? Their analysis on that point is unclear to me. Moreover, how does Gandhi et al’s work influence our reading of Magaloni’s “Voting for Autocracy?” In “Voting,” Magoloni presents a thought-provoking account of why hegemonic parties, particularly “electoral autocracies” like PRI in Mexico, survive. In Chapters 3 and 4, Magaloni explains how the PRI used budget cycles and patronage through the mechanism of poverty programs to enhance support. Can we apply Gandhi et al’s teachings to Magaloni’s work? Would Magaloni’s work be more complete if she had focused more on explaining how policy concessions have affected PRI’s ability to stay in power? Magaloni emphasizes the importance of supermajorities in solidifying the PRI’s power. How should Gandhi et al. have incorporated coalition size in their argument? For instance, Bruce Bueno de Mesquita argues that foreign aid enhances democratization when leaders face a revolutionary threat and when coalition size is substantial; if coalition size is small, aid inhibits democratization. How would the ability of a dictator to get a lot of foreign aid affect the calculus in Gandhi et al’s story?
ReplyDeleteAmy Semet
(First posted at 5:51 p.m., corrected slightly.)
ReplyDeleteMilan Svolik uses survival analysis to study what he describes as two separate mechanisms by which democracies resist reversal to authoritarianism: "democratic consolidation, which practically eliminates the risk of an authoritarian reversal," and "a separate mechanism that prevents authoritarian reversals in those democracies that are not consolidated."
Svolik's analysis seems quite helpful in that this technique seems to fit the data (counts of democracy age and timing of reversals) more closely than other analyses have done. He also makes an important point-- analyses that do not distinguish consolidated democracies from transitional ones overestimate the predicted age of the average democratic regime from time of democratization, with the overall estimates skewed by including old and transitional democracies in one analysis.
While I think his purpose is a good one--to use the consolidation factor to separate analyses of old and new democracies, two aspects of the study seem quite problematic to me.
First, it is not really clear that the mechanism(s) helping transitional democracies resist breakdown are different from those making democracies consolidate, and thus (in Svolik's terms) fully resistant to breakdown. In fact, it seems that when a transitional democracy resists reversal in circumstances that cause other such regimes to break down, it is actively consolidating itself--not necessarily advancing it into the surefire 'consolidated' category Svolik advances, but is taking yet another step towards that status.
Second, I am concerned about what consolidation is, and how it is measured here. While Svolik says he shows that consolidation (defined by being fully resistant to reversal) is associated with a democracy's age, the whole concept seems to be quickly defined by age: after estimating the likelihood of being resistant at different regime ages, he seems to use the curve from that aggregate analysis to determine whether a regime can be thought of as consolidated, by combining age and the estimated rate of growth in resistance over time (it increases most in the first 20 years after democratization) to estimate the likelihood that each regime is consolidated. Yet we assume that the mechanism(s) within consolidated regimes that are making them resistant manifest themselves with varying timing in different regimes in different time periods, regions, economic conditions, etc; thus, using a model to analyze what affects regime consolidation seems not very helpful when the coding of consolidated regimes is dependent on aggregates of non-reversal in the first place. (I welcome any defense of the logic of his model any of you have, as I like many things about this approach and would like to understand it better.)
Julie Touber – Comment Week 6
ReplyDeleteI really like Magaloni’s piece –especially her use of examples in the text and one big case study-. Looking at the hegemonic party survival mechanism, especially the role of elections raised two main questions for me.
First, her discussion of the definition of “hegemonic party autocracy” highlights how the debate of regimes’ classification is still unclear and open. It is not clear to me that she gives a conclusive definition of what she studies, the hegemonic party autocracy. The difference between an hegemonic party autocracy and an “uncommon democracy” is not clear to me, because as she points out, it relies on making the assumption that alternation would happen in one case and not in the other case, which can’t be empirically observed. Thus the transition to democracy from a hegemonic party autocracy can be viewed as an alternation to another hegemonic party autocracy as the example of Kenya (transitional election of 2002 and reelection of Kibaki in 2007) seems to show.
This brings me to my second point. I feel she dismisses quite quickly the Huntington argument of hegemonic party survival: the historic origins. Political identities are constructed through history and I believe that a society whose identities are constructed in social classes is different from whose construction is based on ethnic differences. As she mentioned, the elite is the key element to carry opposition from the large coalition of the hegemonic party. But the elite can be characterized as the “rich” against the “poor” that would be the mass, or as a certain ethnicity that would then only represent a percentage of the mass. It seems to me then, that, given the institutional manipulation to maintain power, hegemonic party autocracies based on ethnic differences would not be able to transition to democracy and would always have to transition to another hegemonic party autocracy based on another ethnic coalition. The African examples (Kenya, Ivory Coast) are quite illustrative of this.
How different is the policymaking process of an autocracy with respect to a democratic one? From a veto player perspective, one could argue that autocracies are usually characterized as regimes with only one effective political actor (e.g. the dictator), in contrast to democracies, where the number of veto players is at least two (e.g. the government, an opposition party). From this comparison, all that matters in the policymaking process of an autocracy are the preferences of the single ruler, and thus, there is no need to study institutions within the regime. This simplistic treatment leads one to expect in autocracies outcomes such as high discounts rates, policy instability, uncertainty and poor economic performance relative to democracies (Olson 1993).
ReplyDeleteHowever, this week’s readings show striking similarities between the policymaking processes of autocracies and democracies. First, autocratic leaders are ambitious: they care about office and survival just as much as democratically elected leaders (Wintrope, Bueno de Mesquita and Smith). Second, they seem to face the same credibility and commitment problems that presidents do: mainly, how to appear credible to important economic actors such as investors and guarantee that their assets will not be expropriated (Gehlbach and Keefer). In spite of the absence of accountability mechanisms, autocratic leaders can overcome these problems through institutional innovations (e.g. parties) that substitute for the presence of formal checks and balances like legislatures. In line with this last point, autocrats don’t usually govern alone, and many do so through political parties where internal power competition is just as fierce as the inter party competion we observe in democracies. Fourth, to maintain power, autocrats need to deliver goods in the form of growth and patronage (Magaloni), policy concessions and rents (Ghandi and Przeworski) that mimic the electoral connection between legislators and voters in the most advanced democratic settings. Finally, as in democracies, policy outcomes depend in part on the details of the institutional scenario: it matters for example whether there is a legislature or not, and the characteristics of parties to explain economic performance in autocracies. Thus, when it comes to the logic of policymaking, autocracies seems to function in a manner that resembles many aspects of democratic regimes.
I posted my paper on Courseworks.
ReplyDelete(I have been experiencing patchy internet at home, so apologies for the late posting).
ReplyDeleteWhen reading the literature on regime change and economic development, and related bodies of work, I have often have thought about the ways in which the heterogeneity of democratic regime types affect the general patterns we see empirically or the implications of the models we examine, or both.
I think Svolik makes a valuable contribution with his article in this respect. First and foremost, he further breaks down and elaborates upon the conceptual distinction that Przeworski et al. introduced in their work, namely between what makes democracy emerge and what sustains it. By establishing a dichotomy between consolidated democracies, on the one hand, and transitional democracies on the other, Svolik is able to capture, in my opinion, the heterogeneity between democratic regime types and the important implications this has. Crucially, whether we agree with the finer points of his analysis or not, he shows how encompassing all extant democracies under a single bracket or categorisation in our analyses can be problematic and miss some of what we are trying to explain. I thought it was interesting and useful to see how he could thereby break down the major implication that emerged from the recent literature on democratization and development – that there is a positive association between economic development and democratic survival – and ask whether this is primarily because “(1) wealth improves the odds that a democracy consolidates, (2) a high level of economic development actually lowers the hazard of reversals in transitional democracies, or (3) both effects hold?” (p. 155). Przeworski et al, in their work, argued strongly in favour of (1) and Svolik finds the first alternative to be the correct answer as well. I just like the fact that the conceptual distinction he introduces allows him to posit different alternatives and distinguish between two separate, though related, causal mechanisms for why this might be so, in a manner that I think expands upon Przeworski et al.
Like Emily, however, I also have some concerns with the way consolidation is specified (or not). I think she articulated it well, so I will not say much more about it here. On a separate note, I was also wondering whether it would be constructive to think of the distinction that Svolik models in terms of an continuum between democracies in danger of reversal and those who are highly unlikely to do so. I mention this because I am somewhat sceptical of the way in which he divides democracies quite bluntly into two separate groups (If I have missed some subtle point here, please let me know!). In theoretical terms, at least – and this would have to be ascertained with an examination of the empirical evidence – I think there is a gradation between democracies in terms of both how consolidated they are (depending upon how we would specify that) and their susceptibility to a reversal. Though an improvement upon the analyses which group all democracies together as one, regardless of the differences between them, Svolik’s account may not go far enough in capturing the heterogeneity and the implications of that as we might desire. I think it is a useful simplification for the purpose of his analysis, but it appears to me that there are also some mixed cases, which would lie in the middle of such a spectrum, or at least somewhere along a continuum between the two poles.
I thought that the readings this week were examples of a very dynamic and exciting literature. To me it seems that the study of decision-making and policy outcomes in autocracies will soon be just as rich as that with respect to democracies. I was particularly impressed by the combination of formal theory and empirics in much of the literature. It also seems to me that the absence of electoral institutions (which are a necessary part of models of democracy) has allowed theorists of autocracies to develop a much wider set of models and theories of political incentives than is available or proper for theorists of democracies.
ReplyDeleteWhile I think that the renewed focus on autocracies is timely and very important, I do think however that the benefits of the separation of the literature into autocratic and democratic segments will eventually be outweighed by the costs. Specifically, it seems to me that other political institutions, such as the quality of bureaucracy, state capacity, political fractionalisation, and international alliances and commitments can and do have an effect on policy in addition to the extent of political accountability to the electorate. We must be careful that we are not captured by the autocratic/democratic distinction, that this scale is not the only scale of institutional character that we regularly consider with respect to policy outcomes.
The Svolik piece on “Authoritarian Reversals and Democratic Consolidation” was particularly interesting. I agree with the author’s argument regarding the importance of taking into account the diverse reasons that account for the survival of democratic institutions. The author correctly notes in some instances that survival may be attributed to the fact that the democracy has truly consolidated. On the other hand, survival can also be attributed to the fact that the regime is still in transition, but continues to exist because of other self-enforcing conditions. While the causal mechanisms accounting for the survival of each type of democracy may be the same, they will operate differently for each regime type. The author does an excellent job in showing the divergence of results between the ‘comprehensive’ and ‘split population’ ‘democratic’ categories.
ReplyDeleteSomething that I found problematic about this piece was the author’s hesitance to provide a conceptual explanation of what exactly accounts for ‘democratic consolidation.’ I understand the author’s claim of an ‘identification problem’ in this respect. However, I was left wondering what the policy implications of the results would be when the ‘end goal’ of consolidation was never clearly defined. For example, the author finds that states that were military dictatorships have a higher likelihood of an authoritarian reversal and a lower likelihood of democratic consolidation. In another statistical test the author shows that economic growth and development enhance democratic transitions and consolidation. Unfortunately, the significance of the results stops here. From a policy perspective, what steps should be taken to turn economic growth into democratic consolidation? Why are military regimes problematic for the democratic consolidation? What are the implications of a military legacy? In all fairness, this does not appear to have been research agenda of the author. However, it would have made the results more interesting if these causal mechanisms were connected to a clearly defined concept of ‘consolidation.’ Perhaps the inclusion of case studies would have been helpful in answering some of these questions.