- Shopping Bag ( 0 items )
This book examines the effect of economic conditions on election results in five post-communist countries—Russia, Poland, Hungary, Slovakia, and the Czech Republic—in the first decade of post-communist elections. It is the first book length study of economic voting outside of established democracies, as well as one of the few comparative studies of voting in post-communist countries generally. The study relies on an original database composed of regional level economic, demographic, and electoral data, and the analysis features a broadly based comparative assessment of the findings across all twenty elections as well as more focused case study analysis.
The story of postcommunism is, to a large extent, the story of simultaneous transitions. The speed at which so many countries fundamentally and simultaneously reorganized both their political and economic systems may be unprecedented in human history. To understand the postcommunist experience, therefore, one must understand the interaction between these two different and yet interconnected processes. A great deal of academic energy has been expended in an effort to understand how political factors have affected economic developments in the postcommunist world.1 Less attention, however, has been paid to systematic studies of the manner in which economic conditions have reverberated back into the political process.
One fundamental way in which economic developments affect political developments is through elections and voting. Political science has a rich literature on the topic of economic voting in established democracies but has only begun to scratch the surface of the topic in newer democracies.2 The politics of economic reform literature in particular often seems to sidestep the topic entirely, relying on conclusions generated from the economic voting literature in established democracies to assume that voters will turn against incumbent parties when faced with the economic pain that is an inevitable by-product of a transition to a more market oriented economy.3 The uncertainty paradigm, also applied in discussions of postcommunist transitions, takes this one step further by highlighting the many factors working against the likelihood of identifying any systematic effects on political behavior in the wake of the transition's complete and utter upheaval of citizens' political and economic worlds.4
The study that follows demonstrates that in a time of massive change characterized by the emergence of entirely new political systems and a fundamental reorganization of economic life, systematic patterns of cross-regional variation in economic conditions affecting cross-regional variation in election results at the aggregate level can in fact be identified during the first decade (1990-99) of postcommunist elections in five postcommunist countries: Russia, Poland, Hungary, Slovakia, and the Czech Republic. And although these patterns strengthened as time passed, they were present even in the earliest part of the decade. Even more surprisingly, these patterns do not merely take the simplest form envisioned by the politics of economic reform literature with areas more heavily populated by economic losers attempting to "throw the bums out" by turning against incumbent parties in greater numbers than those living in areas of the country with more economic winners. Instead, I find that the presence of economic losers in a region leads more consistently to more votes for particular types of political parties, regardless of who is currently in power. The same is true of economic winners, although the presence of more economic winners of course helps different types of political parties. I identify these types of parties by taking the transition seriously, although not as an event that renders rational political behavior impossible, but, rather, as a focal point that allows voters to make sense of the often chaotic political environment emerging around them by focusing on the "Transitional Identity" of political parties. Thus, it is parties identified primarily with the "Old Regime" that predated the transition that enjoy more electoral success in regions with more economic losers, whereas the "New Regime" parties that are most closely identified with the movement away from communism consistently enjoy more electoral support in regions with more economic winners. So although postcommunist voters may in part have been motivated by economic conditions to throw the bums out, ultimately economic winners and losers acted in the aggregate as if they were more interested in the type of party they were trying to throw in.
In an effort to provide a systematic test of these claims, I employ a broadly comparative framework in this study that takes two forms. First, the study examines elections results from twenty separate presidential and parliamentary elections that took place over a ten-year period in five different countries. Thus, none of the patterns identified can be said to be a product of a particular set of circumstances peculiar to a given country at a given time, or even in a given country over time. Second, the study looks for concentrations of economic winners and losers and their effects on election results not at the national level - which would have yielded but one observation of economic conditions and election results per election - but, rather, at the regional level. In this manner, I am able to examine on a party-by-party basis whether or not individual parties in each of the twenty elections performed better in areas of the country where economic conditions were better or where economic condition were worse. At the same time, I employ a method of analysis that allows for comparison of these party by party results with one another, allowing me to search for patterns of support for hypotheses across the entire set of cases contained in the study. Thus, the use of regional level data facilitates both election-by-election case study analysis and broadly based comparative analysis across all twenty elections. These two types of analyses complement one another: for the case study analyses, the broader analysis presents an opportunity to see how the results fit in a larger comparative format; for the overall comparative analysis, the case studies provide an opportunity to explore the context in which larger patterns do or do not hold. Taken together, the two forms of analyses offer a window into the question of how regional patterns of political support for postcommunist parties and party systems have developed and the role that economic conditions have played in that process.
Overall, this book takes advantage of the fact that enough time has now passed since the onset of the postcommunist transitional experience to begin to explore in a detailed, comparative, and systematic manner this important aspect of the relationship between the two distinguishing facets of that transition: the inception of massive economic change combined with the adaptation of competitive multiparty elections (at least for the time period and countries included in this particular study). Unlike previous work that has considered the effect of economic conditions on election results as one of a handful of factors to explain the results of a particular election or series of elections, the study places this crucial intersection between the postcommunist world's simultaneous political and economic transitions at the center of its analysis. In doing so, findings are generated that can shed light on multiple facets of both postcommunist politics specifically and mass political behavior more generally.
Beyond the postcommunist context, the book also offers important contributions to the study of economic voting more generally. Political scientists have long been fascinated by the question of how economic conditions affect election results.5 Yet despite the wide range of questions one can ask on this topic, most analyses have focused on one particular type of relationship between economic conditions and election results - the effect of economic conditions on the vote for incumbent parties and candidates - based largely on findings from one country - the United States6 - and attempts to replicate these findings in other established democracies.7 Put another way, the vast majority of the literature has attempted to answer some variation of the question found in the upper left-hand corner of Table 1.1.
|Table 1.1. Questions Asked by Economic Voting Studies|
|Party Affected by State of the Economy|
|Governing Status||Party "Type"|
|Standard Economic Voting Hypotheses||Do incumbent parties perform better if the economy is better?||Do certain "types" of parties perform better if the economy is better/worse?|
|Conditional Economic Voting Hypotheses||Under what conditions do we find more support for the claim that incumbent parties perform better if the economy is better?||Under what conditions do we find more support for the claim that "types" of parties perform better if the economy is better/worse?|
In addition to shifting the location of the study of economic voting to the more turbulent electoral environment of postcommunist Russia, Poland, Hungary, Slovakia, and the Czech Republic in the 1990s, this study also differs from most that have preceded it by addressing questions found in all four quandrants of Table 1.1. At the most basic level, there are two ways to conceive of economic conditions having a systematic impact on election results. The first is for economic conditions to affect the vote for the parties that comprise the government at the time of the election.8 This question is almost always phrased in terms of whether or not incumbent parties or candidates (hereafter I use the term parties to refer to both for simplicity) are helped by better economic conditions. As noted earlier, this question has dominated the study of economic voting in established democracies. But there is a second way to think about the systematic effect of economic conditions on election results, which is by focusing on a party's "type." In practice, this has involved examining whether or not the economy has different effects on the vote for right-wing or left-wing parties, but, as I will note momentarily (and address in much greater detail in the following chapter), there is no reason why partisan orientation along a traditional left-right spectrum is the only "type" that can be considered.9
Moving our analyses beyond the effect of the economy on the vote for just incumbent parties becomes increasingly important as political systems move farther from the ideal type of a two-party system best exemplified by the United States. Within the confines of a stable two-party system, knowing how the economy affects the vote for the incumbent party tells us almost all we need to know about the effect of the economy on election results; if the incumbent party performs better under certain economic conditions, then the opposition party must perform worse almost by definition when facing these same conditions.10 As we move our analyses to more of a multiparty context, however, the simple dichotomy of an incumbent versus nonincumbent approach disappears as multiple parties can be found both in and out of government (with some parties even moving in and out during a single electoral period). Analyses of the effect of the economy on the vote for parties on the basis of something inherent in the party itself - what I am loosely calling here the party's type or identity - have the potential to expand the range of our understanding of how the economy can affect election results, both in terms of the parties most likely to benefit from better economic conditions but also, perhaps even more crucially, in terms of the parties that are more likely to benefit when economic conditions are worse.
As opposed to most work on the topic that examines only one of these two questions (and, in almost all cases, this is the incumbency question), my study is framed as an explicit comparison between hypotheses drawn from both of these two approaches. Each of the two quadrants in the top row of Table 1.1, therefore, yields a different model. The Referendum Model, based on the upper left-hand quadrant of Table 1.1, predicts simply that incumbent parties ought to perform better in areas of the country where economic conditions are stronger. In contrast, the Transitional Identity Model, based on the upper right-hand quadrant of Table 1.1, categorizes parties in terms of their relationship to the transition itself. On the basis of this classification, one type of party - the aforementioned New Regime parties - is predicted to receive more electoral support in areas of the country where economic conditions are better, whereas another type of party - the Old Regime parties - is predicted to receive more electoral support in areas of the country where economic conditions are worse.11
It is possible to compare the empirical support for each of these models with one another because they make predictions about the effects of economic condition on the electoral fortunes of different parties. Across the twenty elections, there are forty-nine parties coded as incumbent parties, forty parties coded as New Regime parties, and twenty-nine parties coded as Old Regime parties.12 As will be explained in much greater detail later in this book, the method of analysis I employ generates an estimate of our confidence that each particular party provides support for the relevant hypothesis, thus allowing comparisons of the number and proportion of parties that generate support for each of the three primary hypotheses across all twenty elections. The result of this comparative analysis is remarkably clear: across a wide variety of empirical tests, there is consistently stronger empirical support for the predictions of the Transitional Identity Model than for the predictions of the Referendum Model. Furthermore, there are individual elections in which the two models generate competing hypotheses for the same parties. Strikingly, these head-to-head comparisons also yield the same conclusions as the more general analysis: stronger empirical support for the Transitional Identity Model than the Referendum Model.13
As the empirical tests are conducted at a relatively unexplored level of analysis - the regional level - and the models are based on microlevel arguments developed with the postcommunist political context in mind, these results are not intended in any way to serve as a refutation of existing beliefs about the relationship between economic conditions and election results that have been generated from studies in established democracies. Nevertheless, they do demonstrate clear evidence of a relationship between economic conditions and election results that has a more consistent effect on parties based on their "type" (here, the parties' Transitional Identity) than on whether or not the party is currently an incumbent, thus providing support for the claim that there are other illuminating ways to examine the effect of the economy on election results besides focusing solely on the vote for incumbent parties.
The second important contribution of the study to the economic voting literature is to focus attention on the possibilities for analyses that move beyond the first row of Table 1.1 into questions raised in the second row. In doing so, I am building off of a small strand of recent work on the effect of economic conditions on election results in OECD14 countries that focuses on what has come to be called the "clarity of responsibility" argument.15 The basic idea of the argument is that economic conditions are more likely to have the predicted effect on the vote for incumbent parties in countries that feature institutional arrangements that could lead voters to believe that the government has more control over economic policy than in countries with institutional arrangements that could lead voters to believe that the government has less control over these policies.16 In the terms of this book, such a hypothesis fits squarely in the lower left-hand corner of Table 1.1, as it predicts the context (high responsibility institutional arrangements) in which there ought to be more support for the claim that better economic conditions help incumbent parties.
I expand this insight in three important directions. First, I adapt and augment the clarity of responsibility argument for testing in the postcommunist context. Second, I move the clarity of responsibility argument out of just the lower left-hand corner of Table 1.1 and show how responsibility-based arguments can also be applied in the lower right-hand corner as well. In doing so, I test whether the effect of economic conditions on election results for New Regime and Old Regime parties are also more consistent when the parties are more likely to be able to affect policy outcomes than when they are less likely to be able to do so. Third, and perhaps most importantly, I move beyond the framework of the clarity of responsibility argument and provide a whole set of new theoretical approaches for generating hypotheses that belong in the second row of Table 1.1, or, put another way, new arguments about the conditions under which there should be more empirical support for predictions about the effect of economic conditions on election results. These "conditional economic voting" hypotheses are presented in detail in the second half of Chapter 2, but they include a supply-side approach (does the economy matter more when other factors matter less?), a time-based approach (does the effect of the economy change as time passes?), and a partisan orientation approach (do different types of political orientations lead to more or less consistent effects for the economy within the New Regime and Old Regime categories?). Moreover, the emphasis on exploring questions in the second row of Table 1.1 also allows me to focus on results disaggregated to intuitively interesting levels of analysis, such as whether there are differences in the findings across parliamentary and presidential elections, while at the same time placing such analyses within a coherent theoretical framework.
The findings based on these types of questions are too numerous to address in detail here, but it is worth noting that they offer interesting observations for both those interested in the general topic of economic voting and those concerned more specifically with the development of political behavior in the postcommunist context. In terms of making a contribution to the general economic voting literature, evidence from a variety of tests points toward at least limited support for the "clarity of responsibility" approach, even in the turbulent environment of postcommunist elections. At the same time, analysis of these approaches leads to more focused findings such as the fact that there is little difference between the effect of regional variation in economic conditions on regional variation in the vote for reformed communist successor parties and their unreformed counterparts - and that this relationship changes little over time - a point likely to be of interest to those concerned with postcommunist political behavior.
Examining these conditional economic voting hypotheses also provides an opportunity for retesting the primary conclusion of the study - that there is more empirical support for the Transitional Identity Model than the Referendum Model - across alternative subsets of cases. For example, does this conclusion hold up for elections that took place early in the decade? Or only for political parties that received a particularly large percentage of the overall vote? Each subset of cases therefore provides another opportunity for assessing the comparative empirical support for the Referendum and Transitional Identity Models, but identifying theoretically relevant categories for these subsets of cases requires the types of questions found in the second row of Table 1.1.
Beyond providing additional insight into the effect of economic conditions on the vote in the elections included in this study, the conditional economic voting hypotheses I present here can also hopefully provide a blueprint for a wider incorporation of these types of hypotheses into the general economic voting literature. Most specifically, this book offers a host of new hypotheses - detailed in the following chapter - about the conditions under which there ought to be a more consistent effect for economic conditions on election results. At a more general level, I present a number of new categories of these types of approaches that move beyond the responsibility-based approach. Others who might not want to test the specific hypotheses presented here could still use the basic approaches (e.g., supply-side, time-based) to generate their own (and perhaps more country specific) hypotheses. And at the most general level, the whole idea of generating frameworks for thinking about the context in which economic conditions affect election results hopefully will spur others to identify new theoretical approaches beyond those offered in either this book or in the existing work on the clarity of responsibility argument.
|2||Economic conditions and election results||27|
|3||Comparative cross-regional analysis||78|
|4||Paired case studies||126|
|5||The incumbency hypothesis||152|
|App. 5.1||Coding of primary and other coalition partners||178|
|6||The new regime hypothesis||184|
|App. 6.1||Borderline new regime parties||210|
|7||The old regime hypothesis||215|
|App. 7.1||Supplementary tables||242|
|App. 8.1||Supplementary table||274|
|9||Economic voting and postcommunist politics||276|
|App. I||National election results||309|
|App. II||Regression results and documentation||319|
|App. III||Estimated distributions of first differences||356|
|App. IV||Percentage of positive simulations by party||375|