Judicial Institutionalization and Judicial Activism of the Post-Communist Constitutional Courts

In applying constitutional review, post-communist constitutional courts are affected by the existing political and institutional environments, as well as by their own institutional capabilities. However, our understanding of the activity of the post-communist constitutional courts remains incomplete because the existing research fails to consider how the institutional changes on these courts affect their decision-making behavior. In this study, I examine the activity of nineteen post-communist constitutional courts during the 1992-2006 period. I use an aggregate, time-series measure of judicial institutionalization to show that higher levels of institutionalization enhance these constitutional courts’ ability to pursue their policy goals and influence the degree to which they invalidate policy choices of other major political actors, while lower levels of institutionalization limit the courts’ impact on legal and political issues. The findings of this analysis thus provide the first empirical confirmation of the importance of judicial institutionalization to the policy outputs of the post-communist constitutional courts. I also illustrate how various institutional and contextual influences, such as executive power, legislative fragmentation, economic conditions, EU accession process, the identity of the litigants, and the nature of the litigated issues, influence the activity of post-communist constitutional courts.


Introduction
Constitutional courts (hereafter CCs) can perform important functions in the consolidation and maintenance of democratic regimes.They provide a site for the enforcement of constitutional rights and for the delineation of the powers of government bodies.By adjudicating constitutional questions and enforcing constitutional provisions, CCs can make the constitution a living document that shapes and directs the exercise of political power rather than merely a collection of fine phrases that symbolize internationally-recognized norms.CCs can contribute, in other words, to a vigorous protection of individual rights and to the consolidation of democratic regimes.Yet, not all CCs manage to attain this valuable goal.Some become powerless structures, unable to engender public respect and confidence, compel compliance with their decisions, or restrain the appetites of elected politicians.Much depends on the ability of constitutional designers and those leaders who execute CC design to establish a court that has adequate powers and sufficient autonomy to carry out its duties.
When political actors create a CC, their intent regarding the scope of the court's authority, resources, and accountability to its creators is revealed in the institutional design that is adopted in the founding constitution.Institutional design, in other words, is a formal blueprint that gives a CC a responsibility to adjudicate constitutional disputes and outlines the basic tools to fulfill this responsibility (jurisdiction over cases and litigants, financial resources, and so on).However, simply having a formal responsibility is not enough.In order for a CC to fulfill its role as prescribed by its basic design, and to survive and prosper over time, it must move from the vision of its original founders to a well-defined set of organizational structures, goals, and functions (Bumin, Randazzo, and Walker 2009;McGuire 2004;Thorson 2004;Keohane 1969).Thus, if we want to make sense of the CC's actual stature and influence in the national policy making arena, we must look beyond its basic design in the founding constitution to the dates of its implementation, analyze the precise nature of changes in the court's institutional structure after its creation, and then link these institutional features to the court's policy making activity.
This study builds on the Bumin et al. (2009) by exploring the causal connection between the constitutional courts' institutional arrangements and their policy making activity in 19 post-communist states during the 1992-2006 period.I use a measure of CC institutional development developed by these authors -called the judicial viability score -to show that higher levels of institutionalization enhance CCs' ability to pursue their policy goals and their willingness to invalidate policy choices of other major political actors, while lower levels of institutionalization limit the courts' impact on legal and political issues.I also illustrate how various structural and contextual influences, such as executive power, legislative fragmentation, economic conditions, EU accession process, the identity of the litigants, and the nature of the litigated issues, influence the activity of these courts.

The Impact of Institutional Development on the Policy Making of Constitutional Courts
In general, scholars suggest that as organizations institutionalize, they seek legitimation of their activities through active control or shaping of their political environment (Zucker 1987, 451) and that as a rule institutionalization translates into political power (Stinchcombe 1968;Huntington 1968;Keohane 1969;Ragsdale and Theis 1997).The level of institutional viability acquired by an organization thus represents an underlying set of incentives and resources available to its members to shape the scope of their political influence (McGuire 2004, 135).Accordingly, low levels of institutionalization should limit the scope of the CC's activity and policy influence, while greater levels of institutionalization should enhance the court's ability to pursue its policy goals relatively uninhibited.In politically-sensitive cases, where judges may be fearful of potential retaliation for unfavorable decisions, substantial levels of institutional development mitigate these constraints and allow judges to issue rulings which are closer to their collective policy preferences and their institutional or political objectives.The courts will have such flexibility because the specific factors that comprise judicial autonomy, durability, and differentiation collectively protect both the individual judges and the court as a whole from reprisals for unfavorable decisions.(Note 1) It is important to add that the tools that political actors can use against the courts differ in their severity.All impose some costs on courts, but some impose greater costs than others.Being removed from the court or having the court's jurisdiction reduced, for example, are more costly than being overturned or ignored.Courts will then weigh the costs they might face against the potential benefits of reaching policy outcomes that they prefer but other influential actors might oppose (see e.g., Epstein and Knight 1998;Epstein, Knight, and Shvetsova 2001;Iaryczower, Spiller, and Tommasi 2002;Ginsburg 2003;Ferejohn, Rosenbluth, and Shipan 2004).I argue that the ratio of these costs to these benefits is larger in political systems where the courts are weakly institutionalized and smaller in countries where the courts have attained a relatively high level of viability.(Note 2) Significant levels of CCs' institutionalization therefore minimize the severity and the likelihood of reprisals against the court by powerful political actors, allowing the courts to satisfy their policy objectives and behave relatively independently.
However, as Keohane (1969, 860) notes, the actual impact of an organization on its environment also depends on the behavior of other actors who must respect and, when necessary, enforce its decisions.Judicial institutionalization by itself cannot guarantee faithful compliance by the losing party because even the most powerful courts cannot physically force others to accept their decisions.While overruling the court or attacking its jurisdiction, finances, or personnel is a costly and time-consuming process which often requires cooperation between several political actors, non-compliance (i.e., simply ignoring the court's ruling) can be executed quickly and unilaterally (Carrubba, Gabel, and Hankla 2008).Thus, if CCs care about making efficacious policy -one that is complied with and faithfully enforced by other institutions -they will still remain attentive to other strategic constraints on their own behavior as well as on the behavior of the litigants to the case.This means that a host of other institutional and contextual factors -such as the degree of legislative fragmentation, the relative power of the executive, the degree of public support for the court and its decisions, presence and severity of international pressure, and the degree of political transparency -become important to the analyses of judicial decision making.These factors may affect both the capacity and the incentives of political branches to retaliate against the court for unfavorable decisions and/or to engage in overt non-compliance.In sum, by arguing that institutional development affects CC behavior and policy outputs, I do not deny the importance of other considerations that must be taken into account when modeling judicial behavior.

A Brief Note About Judicial Activism, Independence, and Power
In the preceding section I argued that institutionalization is an important determinant of the political impact of the constitutional courts.Testing this argument requires a measure of the CC's policy making activity.One observable indicator of such activity is the CC's willingness to overrule the actions of other policy makers.It is quite common in the judicial politics literature to examine the rates of judicial activism -the frequency with which courts rule that policies passed by government institutions are unconstitutional -as a manifestation of independent judicial behavior and political impact.It is useful to consider judicial activism rates precisely because the ability of courts to exercise political power is at the heart of the concept (Galligan 1991, 70; see also Holland 1991;McGuire 2004).Furthermore, as Smithey and Ishiyama (2002, 721) point out, "nullification [of policy choices made by other government institutions] is considered to be the highest form of activism by most commentators."And, as Tate (1995, 32-33) notes, "although independent judges will not always choose to substitute their own policy judgment for that of others, independent judges are in good position to assert themselves in policy-making against or in competition with the legislative and executive branches."Some scholars have noted potential downsides to such a measure and it is important to acknowledge the validity of these criticisms.(Note 3) Nevertheless, the rates of judicial activism are commonly used to assess judicial independence and judicial decision-making behavior under political constraints around the world in general, and in post-communist countries in particular (e.g., Schwartz 2000;Smithey and Ishiayma 2002;Herron and Randazzo 2003;Anderson and Gray 2007), and no alternative cross-national measures of behavioral independence or judicial power have been proposed to date.Still, it is important to treat rates of invalidation reported for CCs with great care.In order to assess whether activism rates by the post-communist CCs can be treated as an indicator of their independence and/or power, we need to systematically consider the factors important to CC institutionalization and examine these characteristics alongside a variety of other contextual and institutional influences on the exercise of judicial review.

Measuring Policy Making of Post-Communist Constitutional Courts: Judicial Activism
I assess the degree of judicial activism of the post-communist CCs using an original dataset of CC decisions.I examine case descriptions of post-communist CCs in nineteen countries: Armenia, Azerbaijan, Belarus, Bosnia, Bulgaria, Croatia, Czech Republic, Estonia, Georgia, Hungary, Latvia, Lithuania, Moldova, Poland, Romania, Russia, Slovakia, Slovenia, and Ukraine.These courts made case descriptions available online, permitting me to code a large number of cases.This sample includes published decisions for all of the post-communist CCs that provide online case descriptions in English or one of the languages familiar to this author.(Note 4) Each caseload is coded from the first year for which published decisions were available through 2006.I follow two specific selection criteria for inclusion of cases into the dataset.First, only the instances of posteriori review are included.(Note 5) As Herron andRandazzo (2003, 429) andStone Sweet (2000, 51) argue, once a policy is promulgated, it becomes more costly and potentially dangerous for courts to nullify it.Second, only decisions on the merits of the case are included.If the case was denied standing, dismissed on procedural grounds, or only concerned a request for clarification of a constitutional provision, it was excluded.(Note 6) The dependent variable for this study is the probability that the CC engaged in constitutional review of a legislative statute, executive decree or order, lower court ruling, administrative regulation, or the decision of a central electoral commission (CEC).If the court upheld the policy and affirmed its constitutionality, then the case is coded zero.Cases are coded one if the court nullified/invalidated, overruled, or declared unconstitutional one of the above mentioned policy types.In practice, this measure also includes invalidations of parts of statutes, regulations, or other policy declarations.In many cases, appellants did not challenge a policy in its entirety, but argued that only a part of the policy was unconstitutional.The descriptive statistics about each court's caseload, including the number of published decisions, proportion of 'activist' rulings, and the time period covered, are found in Table 1.The data show that the mean rate of judicial activism across the sample and over the observed period (1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006) is relatively high -54.4% of the published cases decided by the post-communist CCs resulted in invalidation of a challenged policy.As Figure 1 illustrates, there are substantial differences in the frequency of invalidation across countries; some post-communist CCs have been very 'activist' in their rulings while others have been relatively restrained.Belarusian and Azeri CCs exhibit the highest rates of judicial activism in the sample, overturning 77% and 70% of the challenged policies respectively, while Croatian (38%) and Ukrainian (34%) CCs exhibit the lowest rates.principal factor analysis, the authors then convert raw annual data on these indicators to derive a single, underlying indicator of institutional viability for each court.I hypothesize that courts possessing higher levels of institutional viability will more frequently nullify legislative statutes, presidential decrees, lower court decisions, bureaucratic regulations, and central electoral commission (CEC) rulings.I therefore expect to find a positive relationship.(Note 7) Over the observed period (1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006), the CC viability scores range from -2.01 to 1.31 (centered at zero), with higher values representing greater degree of institutional development (see descriptive statistics reported in Table 2).

Other Influences on the Policy Making of Post-Communist Constitutional Courts
Although my primary interest is to explore the impact of the institutional viability on the independence and power of the CC (proxied by the rate of judicial activism), it is necessary to include other influences in this analysis to derive robust statistical inferences.Furthermore, inclusion of these additional factors will allow me to present a more comprehensive picture of the determinants of judicial activism, power, and independence in the post-communist countries.I add controls for institutional influences that may undermine or bolster the policy making authority of the courts.I also control for potential effects caused by litigants and legal issues adjudicated by the courts.These factors are important to consider because judicial behavior may be affected when a certain party appears as the litigant before the court, when specific issues are litigated, or when political and economic environment is not conducive to judicial intervention.

Institutional and Economic Environment
First, this study considers whether legislative fragmentation influences the probability of judicial activism by the constitutional courts.Stone Sweet (2000, 54) contends that when parties form broad coalitions to pass legislation, the promulgated statutes are generally less contentious than those produced by a single dominant party.Carrubba, Gabel, and Hankla (2008) similarly argue that legislative statutes that have been crafted through compromise and coalition-building are less likely to face court challenges because the policies satisfy a broad range of political actors.Since a broad coalition supports the promulgated law, the CC's scope of policy influence is circumscribed.
If the court invalidates the law, the legislative coalition could potentially overrule the court.Additionally, the cohesiveness of legislative coalition also increases the legislature's ability to retaliate against the court in case of invalidation of the challenged law by revising CC jurisdiction or limiting the court's budget (Ferejohn et al. 2004).Thus, I expect that the likelihood of activist rulings against legislative statutes will be lower in countries where the legislature is more fragmented.
To assess the degree of legislative fragmentation for the post-communist countries in the sample, this study uses Nordsieck's (2015) elections data.The effective number of legislative parties (as a proxy measure for legislative fragmentation) was calculated using the formula from Laakso and Taagepera (1979).The scores for legislative fragmentation for the post-communist sample range from 1.14 in Azerbaijan from 1998 through 1999 (least fragmented legislature) to 8.79 in Armenia from 2003 through 2004 (most fragmented).
Second, this study considers whether the structure of the executive system may impact the CC's ability to exercise constitutional review.Ackerman (1997) argues that presidentialism is good for courts by providing them with a role as an arbitrator among law-making powers.The reason the presidential and semi-presidential systems support judicial activism is because of the potential for institutional divergences between the executive and the legislative branches.These divergent policy views can be ameliorated by the presence of a capable and active constitutional court (Ginsburg 2003, 82-86).In addition, division of power between branches can allow the courts to exercise greater independence and discretion in their rulings because certeris paribus attacking the court through restrictions on jurisdiction or budget may be more difficult in systems where passage of legislation requires the cooperation between two separate political bodies.Because of the combination of a difficult environment for passing new legislation to retaliate against the CC for unfavorable decisions or for overruling the court, along with high demand from the executive and the legislature for a neutral third party to resolve jurisdictional boundary disputes, semi-presidential and presidential systems may therefore favor the development of an activist and independent CC more so than pure parliamentary systems.
To assess the relative power of the executive, this study relies on the presidential powers index (PPI) derived from the Comparative Political Dataset II.The dataset provides an index of executive-legislative power distribution that is specifically tailored to the capacity of presidents or legislatures to maintain or alter the status quo of policy-making.A significant advantage of this measure is that it incorporates temporal changes in executive power into coding.This provides for a continuous measure of executive power from 1989 through 2006, with pure parliamentarism and strong presidentialism (with a weak, reactive legislature) as endpoints.
Theoretically, the full index runs from zero (pure parliamentarism) to twenty-nine (superpresidentialism).However, the range of actual scores for executive-legislative powers for the post-communist sample is from one (very weak presidency in Bosnia and Herzegovina) to 22.5 (very strong presidency) in Belarus.The mean PPI score in the sample is 9.77, which, in substance, represents semi-presidential regimes with moderately-balanced distribution of power between the president and the legislature.I expect that the likelihood of CC activism will increase in states with semi-presidential and presidential systems.
The third characteristic potentially influencing judicial activism in the post-communist countries is economic performance.The literature posits that economic performance matters for judicial behavior although there is a disagreement on the direction of its influence (for a useful discussion, see Herron and Randazzo 2003, 426).Some suggest that in countries where economic conditions worsen over time, probability of judicial intervention decreases; others argue that the courts will actually exert more influence on national policy during periods of poor economic performance.Yet another group of scholars speculate that in many post-communist countries, privatization and economic growth thrust public officials into uncharted waters and, in response, administrative agencies and legislatures often enacted contradicting legislation.This facilitates judicial activism, as economic actors interested in avoiding technical violations increasingly rely on the courts to clarify and standardize the law.
Since at this point we can only speculate about the relationship between economic growth and judicial activism, I include a measure of GDP growth, using first year of a country's transition as the base year, and rely on a two-tailed test of statistical significance to assess the direction of its influence.Economic data is extracted from the European Bank for Reconstruction and Development.
The fourth environmental characteristic potentially influencing activism rates of the post-communist CCs is the state's respect for political and civil rights.Epp (1998) and Tarrow (1998) argue that regime's recognition of extensive political and civil rights may provide greater opportunities for citizens and politicians to bring cases to the CC, enhancing and solidifying its role in the country's political system.To control for this possibility, this study uses Freedom House's cross-national time-series data, which measure political rights and civil liberties around the world.Political rights and civil liberties indices contain ratings between one and seven, with one for the 'most free' and seven for the 'least free'.This study combines the two indices to construct a single variable.The country scores can thus range from two (most democratic) to fourteen (least democratic).I anticipate a negative relationship -the rates of judicial activism will be higher in countries with more vigorous support for political rights and civil liberties.
Finally, some scholars note that the European Union accession process has generated an unprecedented momentum for reforms in the candidate/applicant countries (Schimmelfennig and Sedelmeier 2005: 226).As part of the accession process, the European Commission regularly evaluates candidate countries in a wide range of areas in the framework of its reports on the progress of each country towards the fulfillment of the 'Copenhagen criteria.' Maveety and Grosskopff (2004, 468) argue that the EU-imposed constraints associated with the accession process reinforce the power of national CCs and promote their distinctive contribution to the policy-making process by allowing the courts to reference or invoke EU legal principles in their rulings and, by doing so, enhance the legitimacy of their rulings (see also Boulanger 2001Boulanger , 2002;;Nyikos 2001).Moreover, by monitoring the countries' commitment to the Copenhagen criteria, the EU accession program may lower the probability of non-compliance with the CC rulings.
This study relies on a dichotomous variable where one represents EU applicant status and zero otherwise, taking into the consideration the official dates when each applicant country obtained her status.By the end of 2006, Bulgaria, Croatia, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovenia, and Slovakia participated in the EU accession program.This study expects a positive relationship between the participation in the EU accession process and CC activism in a candidate country.

Litigant Characteristics
I also add controls for potential effects caused by particular litigants before the constitutional courts.As Herron and Randazzo (2003) point it out, this is necessary because judicial behavior may be affected when certain actors appear as litigants before the court.All cases included in this analysis where thus coded for litigant characteristics (appellant and respondent).Each of the six litigant variables discussed below is dichotomous and coded one when the litigant of interest appears and zero otherwise.
In the previous section, I posited that from the perspective of the institutional environment in which the courts operate, presidential and semi-presidential systems support judicial activism.Here, I consider another possibility: when a president is a direct litigant to the case adjudicated by the CC, the court will be deferential and more willing to rule in the president's favor.Schwartz (2000) and Herron and Randazzo (2003), for example, conclude that concentrated executive power is a significant factor in explaining judicial subservience in the post-communist countries.To control for this possible effect, I measure when the president (or a high-ranking member of the presidential administration) appears as an appellant before the CC and asks the court to invalidate legislative statutes, policy decisions of regional/local authorities, or rulings of the courts of general jurisdiction.I hypothesize that the CC will be more likely to nullify policy decisions of other actors when asked directly by the president.I expect that CCs will be deferential to the presidential requests because presidents are more easily able to retaliate against the court or to ignore its rulings.As Ginsburg (2003) argues, presidents can act quickly and often unilaterally in response to unfavorable decisions, and this possibility weights heavily on the minds of CC judges when they consider the ratio of costs to the benefits of upholding the policy challenged by the president.
Second, I control for those cases where individual citizens, private businesses, and civic organizations appeal the court (direct constitutional challenges and constitutional appeals).All post-communist countries adopted CCs during their transitions with the express purpose of protecting political, civil, and basic human rights of individual citizens.If CC judges view their role as protectors of individual rights, they are more likely to engage in judicial review when presented a claim on behalf of these appellants.In short, I expect a higher probability of judicial activism if a private citizen appears as an appellant before the court.
Third, I consider whether CCs are more willing to invalidate legislative statutes if a minority party in the legislature appeals the court.Stone Sweet (2000) and Ginsburg (2003) argue that CCs are inherently "minoritarian institutions" because they provide opposition parties a forum to challenge the policies of political majorities.However, Herron and Randazzo (2003) suggest that CCs are less likely to rule in favor of legislative minorities when the legislative majority is a direct litigant to the case.Since it remains unclear why CCs would systematically favor preferences of legislative minorities, I include the measure but rely on a two-tailed test of statistical significance to assess the direction of its influence on judicial activism.
Fourth, I control for those cases where legislature appears as a respondent before the court.I hypothesized earlier that the probability of invalidation of legislative statutes will be lower in systems characterized by high degree of legislative fragmentation.Here, I consider an alternative hypothesis: if and when the legislature appears as a respondent in a case adjudicated by the CC, the court will be less deferential and more willing to invalidate the statute.While legislatures can retaliate against the courts for negative rulings, the parliamentarians are able to do so only to the extent that they are sufficiently coherent as a group to amass the majorities necessary for retaliation against judicial salaries, terms of office, or jurisdiction.Ceteris paribus, it is more difficult for the fragmented legislatures to affect CC behavior as extraordinary levels of parliamentary coherence are required to punish the court for going against the legislative policies (Tate 1995;Chavez, Ferejohn, and Weingast 2004).
Fifth, I hypothesize that CCs will be particularly active in ruling against the ordinary judiciary and administrative agencies when one of the lower courts or an agency appears as a respondent to a case.As Schwartz (2000, 236-237) indicates, general courts in many post-communist states are either unaware of, or deliberately indifferent to, both the constitution and CC rulings (on the Russian case, see also Trochev 2005).Judicial and bureaucratic incompetence is also widespread; ordinary judges and bureaucrats often misinterpret and incorrectly apply statutes and codes, violating citizens' constitutional rights in the process (see Anderson and Gray 2007).Thus, CCs have many opportunities to review and reverse decisions originating in the bureaucracy and the ordinary judiciary.In sum, I expect a greater probability of activism in cases where a lower court or an administrative agency is a respondent.

Issue Characteristics
The rate of CC activism may also be influenced by the specific issues litigated.CCs may be more inclined to invalidate policies in certain issue-areas, such as those concerning separation of powers and individual political and economic rights.Each of the four issue variables described below is dichotomous and coded 1 when the issue of interest appears and 0 otherwise.
First, I argue that CCs are more likely to invalidate legislative statutes or presidential decrees pertaining to separation of powers issues or issues of governmental authority.These issues may involve disputes over authority between executive and legislative branches, or disputes concerning the distribution of political power at the national and subnational levels.Although going against the president or the legislature over such sensitive political issues as the allocation of power significantly increase the prospects of a counterattack and/or non-compliance, courts may take a more active role to ensure lawful action of government.For instance, Schwartz (2000, 228) shows that in separation of power disputes, the post-communist CCs "frequently rule against the party in power… and by and large, they have gotten away with it."In such disputes, the CC always has at least one significant national policy-maker on its side; this somewhat lowers the threat of retaliation by the losing party for an unfavorable decision.Additionally, as Schwartz (2000, 229) argues, "there is a good deal of pressure on these branches to obey the court's judgments in allocating power between them."In the vast majority of cases, the alternative -a collapse of the constitutional order -will leave both parties worse off (Stone Sweet 2000; Shapiro and Stone Sweet 2002;Ginsburg 2003).
Second, I control for instances where the courts decide on matters related to elections (e.g., verification of election results, rulings on electoral disputes, cases concerning denial of eligibility for certain candidates, charges of gerrymandering on the grounds that such schemes violate constitutional rights).I argue that it is especially unwise for the CCs to actively intervene in such issues due to the politicized nature of such disputes and a greater threat of retaliation and/or non-compliance.I thus expect that the courts will be less likely to declare elections invalid or to nullify other election-related policy decisions.
Third, I expect that the CCs may be more active when resolving economic issues.These cases often involve private and public disputes over property rights, complaints about unconstitutional/illegal taxes and customs duties imposed on individual citizens and businesses, and requests for financial reparations for economic losses incurred during the communist era or immediately after the collapse of communism.These issues are highly salient and controversial, and in some instances involve claims against the state for financial damages resulting from ethnic conflicts (e.g., Bosnia and Herzegovina, Georgia), civil wars (e.g., Azerbaijan, Moldova), and ethnic/religious/linguistic discrimination (e.g., Estonia, Russia, Ukraine, Hungary, Bulgaria).
Finally, I hypothesize that the CCs will be more inclined to act in order to preserve and defend individual political rights, civil liberties, and fundamental human rights.As Sadurski (2005) and Maveety and Grosskopff (2004) argue, since most post-communist countries have signed the European Convention on Human Rights and are members of the Council of Europe, there is an additional incentive to enforce these rights.These agreements reinforce the power of the CCs by allowing them to reference or invoke European legal principles in their rulings and, in doing so, enhance the legitimacy of their rulings (Boulanger 2001(Boulanger , 2002;;Nyikos 2001).

Results and Discussion
To explore the factors influencing CC activism in the post-communist countries, I analyze pooled cross-sectional data on judicial decisions.Because the dependent variable is dichotomous, I use multivariate probit estimation.
In order to account for the fact that some constitutional courts publish more decisions than others, I adopt a system of proportional weighing suggested by Herron and Randazzo (2003).This weighing technique measures the proportion of cases from each CC and weighs those cases by the inverse.(Note 8) This is necessary because the unweighted sample generates distorted results since certain courts will have a disproportionate influence on the statistical analysis.The third environmental characteristic considered in this analysis pertained to the regime's general respect for political rights and civil liberties.I expected that CCs engage in activism more frequently in countries where the political system recognizes and respects the rights of its citizens (as measured by the Freedom House).The empirical results do no support this claim.Although the coefficient is negative (indicating that CC activism is more prevalent in countries with extensive protection of rights), the variable is not statistically significant.I also expected that economic performance (proxied by GDP growth rate) influences judicial behavior.However, I did not specify direction of its influence.The results demonstrate that economic performance has a statistically significant and positive influence on CC activism.Furthermore, the impact of economic conditions on judicial activism is substantial.As the county's economic conditions improve, the likelihood of CC invalidating statutes, bureaucratic regulations, and other government acts increases by 10.5%.These results contradict Herron and Randazzo's (2003) finding that judicial activism in the post-communist countries is more prevalent during economic downturns.At this point, I can only speculate alongside Anderson, Bernstein, and Gray (2005) that my results indicate that economic growth leads to an increasing demand for more objective dispute resolution mechanisms and better-functioning regulatory and judicial institutions in the post-communist countries, and that CCs respond to this demand by standardizing the law, clarifying public and private property rights, and invalidating contradictory or illegal policies.More research on the impact of economic conditions on judicial activism is clearly necessary to determine the precise nature of this relationship.
The last environmental factor considered in this analysis concerned a country's participation in the EU accession process.This study expected to find a positive relationship between a country's participation in the EU accession process and CC activism in a candidate country.According to Table 3, however, participation in the EU accession process does not exert a statistically significant influence on the likelihood of CC activism.Moreover, the coefficient is negative, indicating that CC is less likely to invalidate government policies in candidate countries.The data seem to indicate that EU influences are not as important for the CC behavior as some scholars have suggested.

Litigant Characteristics
Earlier, I assumed that the courts care about making efficacious policy and hypothesized that CC behavior in reviewing government policies will be affected by the presence of certain litigants before the court.I expected that when president appears as an appellant before the court, requesting it to invalidate a legislative statute, a lower court ruling, or a policy decision of a local/regional government, the CC will defer to the presidential request.The empirical results support this claim unequivocally.According to the marginal effects reported in Table 3, when a president appears as an appellant, CCs are 13.9% more likely to acquiesce to president's appeal and invalidate the challenged policy.The post-communist CCs clearly favor presidents in their rulings.Given that some post-communist states concentrate enormous power in the presidential office (e.g.Belarus, Russia, Georgia, Azerbaijan, and to a somewhat lesser extent, Armenia, Croatia, and Ukraine), these results are unsurprising.The CCs realize that unlike the legislatures, the presidents are able to act quickly and unilaterally and can sometimes impose substantial costs on courts if they do not get their way.I also proposed that when individual citizens, private businesses, and civic organizations lodge direct constitutional challenges and constitutional appeals, the CC judges may be more likely to engage in judicial review and strike down policies challenged by the ordinary citizens.The results show that CCs are in fact attentive to the requests for invalidation made by private actors.According to the marginal effects for the Individual (Appellant) variable, CCs are 8.6% more likely to invalidate statutes, executive orders, bureaucratic regulations, or lower court rulings when asked to do so by ordinary citizens.This finding thus supports the argument that CC judges view their role as protectors of individual rights and liberties.And, it is reasonable to interpret CCs' activism on the behalf of citizens is an indication of their independence.
Third, I considered whether CCs are more willing to invalidate legislative statutes if a minority party in the legislature appeals the court.However, I did not hypothesize the direction of its influence on the probability of judicial activism.The results show that CC behavior is not influenced by the appeals made by the legislative minorities.Although the variable fails to achieve statistical significance, the negative sign on the coefficient seems to support Herron and Randazzo's (2003) argument that CCs are less likely to rule in favor of legislative minority appellants when the legislative majority or executive is a respondent to the case.Fourth, I controlled for those cases where legislature appeared as a respondent before the court.I hypothesized that when the legislature appears as a respondent in a case adjudicated by the CC, the court may be less deferential and more willing to invalidate the statute.I speculated that the pressures on the CC to rule in favor of legislative majorities (i.e., to uphold the statute) are mitigated by the fact that legislative counterattacks for unfavorable rulings require extraordinary levels parliamentary coherence and significant amount of time to execute.I therefore expected that this may invite more invalidations from the CC.The results support this argument.The CCs are 5.7% more likely to invalidate a legislative statute if and when the legislature appears a respondent to the case.In light of the findings for President (Appellant) and Individual (Appellant) variables reported above, a discovery that CCs are more willing to invalidate legislative statutes when the legislature is a respondent make a good deal of sense.
Finally, I hypothesized that CCs will be particularly active in ruling against the ordinary judiciary and bureaucratic agencies.Since ruling against these actors does not pose an immediate threat to CCs, I expected a greater probability of activism in cases where these actors appear as respondents.The empirical results support my claims in regard to the bureaucratic agencies only.The CCs are 8.1% more likely to invalidate agency decisions if an agency is a respondent to the case.However, as Table 3 shows, post-communist CCs do not systematically rule against the decisions of the lower courts.The Ordinary Judiciary (Respondent) variable is not statistically significant and the sign on the coefficient is in the direction opposite to my hypothesis.

Issue Characteristics
The final set of control variables pertained to the influence of specific issues litigated by the constitutional courts.I hypothesized that judges are more likely to strike down policies that concern allocation of powers (horizontal or vertical), economic issues, and individual rights.Conversely, I expected that CCs will be less likely to intervene in electoral disputes.The empirical data show that the elections-related issues do not exert a significant influence on judicial behavior.On the other hand, my claims regarding the separation of powers, economic issues, and individual rights are confirmed (see Table 3).
Marginal effects show that CCs are 7.7% more likely to invalidate presidential decrees or legislative statutes pertaining to separation of powers issues or to the issues of governmental authority.Moreover, CCs are more active when resolving economic issues involving private and public disputes over property rights, requests for financial reparations for economic losses, and constitutional complaints about taxation and customs duties imposed on individual citizens and businesses.Over the observed period, CCs were 4.7% more likely to strike down policy in these types of cases.Finally, I hypothesized that the CCs will be more inclined to act in order to preserve and defend individual political rights, civil liberties, and fundamental human rights.The results show that courts were 6.3% more likely to exert their influence on rights-related issues and to invalidate challenged policies.Combined with the results for the Individual (Appellant) variable, it is clear that post-communist CCs perceive their role as defenders of rights and seek to actively shape public policy in this area.

Conclusion
One of the most striking features of the transitions in the post-communist region has been the spectacular growth in the role and prominence of constitutional courts.The situation today is one in which all post-communist countries have CCs, and while the viability and effectiveness of these courts varies greatly, many of them stamped their authority on the process of constitutional transition and on important public policies.Some of their decisions have had enormous financial and budgetary implications, some transgressed clear and strong popular feelings, and others invalidated delicately crafted political compromises.Important aspects of policies on abortion, adoption, the death penalty, lustration and criminal prosecution of former communist officials, conduct of elections, taxation, referenda, citizenship requirements, government control over the ordinary courts, prohibition of political parties, distribution of power, and entrepreneurial activity have all been struck down.On multiple occasions, post-communist CCs affirmed citizens' constitutional rights, such as the right to fair trial, private property, pensions and retirement benefits, housing, and freedom of movement.
My research indicates that these courts' activism is significantly influenced by their level of institutionalization.I find that the post-communist CCs with higher levels of institutional development are, on average, 15.8% more likely to overturn, invalidate, or reverse policy decisions of other government institutions than their less institutionally-developed siblings.My analysis also reveals that the impact of CC viability on judicial activism is larger than the impact of any other contextual or environmental variable included in the statistical model.Although I discover that CCs are sensitive to the appeals by particular litigants, more attentive to certain litigated issues, and more likely to actively engage in policy making in certain political environments, the level of institutional development still has a distinct, positive, and substantial influence on the post-communist CCs' ability to actively shape public policies.Simply put, as post-communist CCs institutionalize, they become a distinctive, respectable, and influential force within the post-communist societies.
Nevertheless, as Keohane (1969) notes, the actual impact of an organization depends upon the interactions between its own organizational characteristics and the surrounding environment.Constitutional courts do not exist in a vacuum and my analysis shows that the institutional environment in which these courts operate is an important determinant of their policy influence.Some institutional environments provide CC judges with more opportunities to shape national policies while other institutional settings limit the scope of judicial impact.I find that presidential and semi-presidential systems are conducive to the activism of CCs.Due to a combination of a difficult environment for passing new legislation to retaliate against the CC for unfavorable decisions, along with high demand from the president and the legislature for a neutral third party to resolve jurisdictional boundary problems, semi-presidential and presidential systems favor the development of activist and powerful CCs in the post-communist region.
Alternatively, I find that pure parliamentary systems limit CC opportunities to actively shape national policies.This discovery seems to support the 'insurance theory' of judicial power proposed by Ginsburg (2003).The author argues that the demand for an active and independent CC is greatest in the political systems where legislative power is evenly split between two dominant parties.In such scenarios, neither party can be confident of winning future elections and would therefore prefer to limit majority rule and value institutional mechanisms that check the power of future electoral winners.My findings similarly indicate that the probability of judicial activism is negatively related to the number of viable parties in the legislature.Although I make no strong claims in support of Ginsburg's theory at this time, the data hint at a possibility that the 'insurance theory' can help explain judicial behavior in the post-communist countries.
Taken together, the results for presidentialism and legislative fragmentation perhaps indicate that the most fortuitous environments for active and independent CCs are those where the presidential branch is independent and relatively influential, and the parliamentary power is evenly split between two or three major parties.This constitutional configuration may be ideal for the development of active judicial power.In future analyses it is necessary to address the interactive effects of these two variables on judicial policy making in a more rigorous manner.
Additionally, I discovered that the impact of economic conditions on judicial activism is positive and substantial.Unfortunately, my data do not provide a satisfactory answer for this phenomenon.I can only speculate that my results indicate that economic growth leads to an increasing demand for more objective dispute resolution mechanisms and better-functioning regulatory and judicial institutions in the post-communist countries, and that CCs respond to this demand by actively intervening in matters related to national economy.This possibility is further supported by the finding that constitutional courts are more active when deciding economic issues involving private and public disputes over property rights, complaints about unfair taxes and customs duties imposed on individual citizens and businesses, and requests for financial reparations for economic losses.A more rigorous and theoretically-grounded investigation of the economy-judicial activism relationship would thus be especially fruitful.
Finally, it is important to emphasize that while structural/institutional environments shape the range of policy options available to the CC judges they do not dictate specific case outcomes.As the empirical results show, in the real world of CC litigation, parties to the specific case can and do influence the extent to which courts take advantage of the existing structural and institutional opportunities.In separation of power cases, the CC is most clearly serving in its role as dispute resolver rather than policy maker.But these kinds of cases are fraught with danger for the court, especially where there is an asymmetrical distribution of power among the litigants to the case.The losing party may respond with non-compliance and/or retaliation against the court.Thus, if the court sides with the more powerful litigant, its decision becomes largely self-enforcing and the risk of reprisal is substantially lowered.This study shows that in the post-communist region CCs clearly favor the presidents in disputes over governmental authority.When a president appears as an appellant before the court and requests it to invalidate policies of another government body (the legislature and local/regional governments are typically the respondents to such cases), the court is likely to rule in the president's favor.
However, where CCs side with private citizens against government institutions in political, civil, or economic rights cases, the reason cannot be either the lower potential for reprisal or the higher likelihood of compliance.The power of government institutions to retaliate against the court or to ignore its decisions is always greater than that of the individuals.Rather, CCs' activism on the behalf of private citizens is a clear indication of their relative independence.The CCs in post-communist countries obviously perceive themselves as protectors of individual rights and liberties and, dependant on their levels of viability, are (more or less) willing to challenge powerful political actors in such cases.It is possible that in deciding in favor of individual rights, the CCs are striving to legitimize their existence in the eyes of the general public and by doing so to raise the costs of future political attacks or non-compliance.
Though I have amassed one of the largest collections of data ever assembled on constitutional court decisions, my findings are nonetheless limited to the post-communist region and only to the published decisions over the course of fifteen years.And my efforts to tease cross-national (and a few longitudinal) inferences out of these data will surely not please everyone.Still, I am convinced of three things: Institutionalization is an important determinant of the constitutional court's ability to contribute to the national policy making and, consequently, to the process of democratic consolidation in transitional societies; it is necessary for scholars to carefully formulate their hypotheses and differentiate between the effects of broad institutional configurations and the effects of peculiar contexts in which actual judicial choices are made; and, the increasing availability of cross-national and longitudinal quantitative data on courts provides an enormous opportunity to test and refine our theories of judicial behavior.

Notes
Note 1. Constitutional provisions and enabling legislation outline the institutional safeguards for CCs and include nomination/appointment procedures, professional requirements to serve on the courts, terms of office, procedures for financing the court, court's jurisdiction over litigants and issues, as well as internal rules of procedure for the consideration of cases and the rendering of decisions (for extensive discussion, see Bumin et al. 2009).
Note 2. Elected branches have a variety of tools they can use to retaliate against the actions of courts, such as appointing sympathetic judges, passing legislation that overrides court rulings, removing cases from the jurisdiction of the court, impeaching judges, refusing to raise salaries, underbudgeting for material factors related to the proper functioning of the CC (such as buildings and staff), or even amending the constitution.But politicians are able to do so only to the extent that they are sufficiently coherent as a group to amass the support necessary to attack the court, and the level of judicial institutionalization intervenes crucially in this regard.
Note 3. In particular, some scholars are concerned that the judicial activism rates do not accurately reflect either the degree of judicial power or the degree to which courts (or individual judges) act 'independently', and disagreements persist about the best way to measure independence empirically.For example, scholars note that it is possible for even the most independent courts to engage in judicial restraint.In addition, infrequent declarations of unconstitutionality may indicate that the governments are actually doing a good job of conforming to the law, and that the courts are finding in their favor because few violations of the constitution really occur (Larkins 1996, 616-617).Other scholars note that in some countries the judiciary may be used by the president or legislature as a tool for legitimizing unlawful actions or for advancing short-term partisan agendas.If this occurs, the court may be extremely effective in reviewing and nullifying certain types of legislation, but it by no means indicates that the court is independent of external influence (see Herron and Randazzo 2003, 423-425).
Note 4. Cases are coded in English and Russian (Armenia, Azerbaijan, Belarus, Latvia, Moldova, Russia, Slovakia, and Ukraine), Belarusian, Ukrainian, and Bulgarian to provide the most comprehensive overview of caseload for each court.I also use English translations provided by the courts to fill any gaps in coverage.For Bosnia, Croatia, Czech Republic, Estonia, Georgia, Hungary, Lithuania, Moldova, Poland, Romania, and Slovenia, I use English translations of cases only (available on the courts' websites).The coverage for these courts is therefore less complete due to my sole reliance on case information translated by the courts.In some cases, this precludes assessment of a vast number of decisions.For example, Hungarian Constitutional Court provided English translations for only 69 decisions over the 1992-2006 period; however, the website provides case descriptions for over 600 cases in Hungarian.
Note 5.An application of posteriori review includes preliminary questions, direct constitutional challenges of legal norms, and constitutional complaints.
Note 6.This choice has significant ramifications for analysis of some CC caseloads.For instance, between 1992 and 2006, the Russian CC issued more than 1,400 decisions, but only 246 of these decisions were on the case merits.Some post-communist CCs therefore make an explicit distinction between judgments and rulings (or decisions and rulings) to indicate to the reader which case descriptions concern decisions on the merits.Hereafter, any reference to decisions or rulings by the CCs is meant to reflect decisions on case merits only.
Note 7. CC viability scores are lagged by one year.Thus, for example, the level of CC viability attained by the end of 2005 is expected to determine the scope of the court's policy influence in 2006.
Note 8. Thus, if one country contains 1/6 of the total number of cases and another country 1/10 of the cases, the former would have a weight of 6 and the latter a weight of 10.This technique ensures comparability across courts in terms of representation within the overall dataset.
Note 9.An examination of the marginal effects allows one to assess how much influence a particular coefficient has on the probability of the dependent variable registering a 1 instead of a 0. This allows one to make judgments of the variable's impact relative to the baseline probability (calculated holding all variables at their minimum levels) for both categories of the dependent variable.The interpretation of marginal effects is similar to OLS regression coefficients.
Note 10.An absolute difference in BIC' is used to evaluate the overall fit of the model (see Long and Freese 2003: 94).An absolute difference in BIC' that is greater than 10 indicates very strong support for the inclusion of the CC viability.
Note 11.It is interesting that the CC are more likely to invalidate legislative statutes and local ordinances on presidential appeal both in countries with relatively weak presidents (e.g., Hungary, Bulgaria, Moldova), as well as in countries with very powerful presidents (e.g., Russia, Azerbaijan, Armenia, Belarus).This observation suggests that presidential and semi-presidential systems provide political environments that are highly conducive to judicial intervention and policy-making.While it may be true that presidentialism is not good for democratic consolidation (see e.g., Linz 1990, Power andGasiorowski 1997), my findings confirm Ackerman's (1997) argument that separation of powers is conducive to judicial activism.

Figure 1 .
Figure 1.Proportion of activist rulings by court (from the 1st year of operation through 2006)

Table 3 .
Table 3 displays the results of the multivariate probit model and marginal effects for the explanatory variables.(Note 9) Institutional and contextual influences on CC activism (probit results) System of proportional weighing is used to ensure comparability across countries in terms of representation within the overall dataset; robust standard errors are reported in parentheses; all significance tests are two-tailed; a difference of BIC' parameters indicates that model with the judicial viability measure is more likely to have Tate, C. N., & Vallinder, T. (Eds.).(1995).The Global Expansion of Judicial Power.New York, NY: New York University Press.Thorson, C. (2004).Why Politicians Want Constitutional Courts: The Russian Case.Communist and Post-Communist Studies, 37, 187-211.http://dx.doi.org/10.1016/j.postcomstud.2004.03.003Trochev, A. (n.d.).Distrusted Courts: The Impact of State (In)capacity on Judicial Power in Post-Communist Countries.paper presented at the annual meeting of the Canadian Political Science Association in Toronto, Ontario, June 1-2, 2005.Zucker, L. G. (1987).Institutional Theories of Organizations.Annual Review of Sociology, 13, 443-464.