In “Setting the Rules of the Game: The Choice of Electoral Systems in Advanced Democracies” Carles Boix argues that electoral laws can originate endogenously from the strategic decisions of dominant parties.
I wish to make a simple methodological point. Boix's makes a basic mistake in his regression analysis that undermines the robustness of his conclusions. In Table 1 (p. 619), Boix runs a multivariate regression with two interaction terms. "Threat" is an interaction term of the strength of socialism and the number of old parties variables. "Fragmentation X Area Dummy" is an interactive term of the combined index of ethnic and religious fragmentation and geographical area.
When using interactive terms in a multivariate regression, it is important to include both the interactive and the constitutive terms (i.e. the variables that make up the interaction term) in the regression. Not including the constitutive terms in the analysis can create omitted variable bias in the coefficients.
Boix does not include the constitutive terms for "Threat" in Model 1 and 3. Moreover, he fails to include the constitutive terms for "Fragmentation X Area Dummy" in Model 3. This failure to include the constitutive terms has real consequences. Brambor, Clark, and Golder (2006) replicate Boix's analysis, including the proper constitutive terms of the interactions. When the constitutive terms are included in the regression, there is no longer any evidence that the "Fragmentation X Area Dummy" interaction term has a statistically significant effect on the adoption of proportional representation. Conversely, by including the constitutive terms, the statistical significance of the "Threat" interaction term decreases (while remaining significant to the 90% level), while it's coefficient increases by 340%. Therefore, Boix strongly underestimates the interactive relationship between "OldParties" and "Socialism" on the adoption of proportional representation.
Reference:
Thomas Branbor, William Roberts Clark, and Matt Golder, "Understanding Interaction Models: Improving Empirical Analysis," Political Analysis, 14:1 (Winter 2006), 63-82.
Of the readings this week, the one that caught my attention the most was the Berkowitz, Pistor and Richard article on the transplant effect of legal institutions. The idea that legality—or the effectiveness of the legal order—depends more on the country’s receptiveness, focusing on the transplant process, and not on what was transplanted, is refreshing. While they focus on the legal family, their discussion brings forth the question of whether or not institutions can be discussed in a vacuum. The emphasis on the process of institutionalization, and not on individual institutional characteristics, forces us to consider the role of institutions in given contexts, particularly when addressing what Berkowitz, Pistor and Richard call “cognitive institution.” Society must have a demand for the cognitive institution, tweaking it and giving it a bit of local personality, so as to make it their own. It is this idea put forth by Berkowitz, Pistor and Richard that caught my eye. The transplant of institutions only has a chance of being durable and maintaining its integrity if the local culture is willing (and ready) to embrace it, otherwise, it is likely to suffer resistance and an inherent backlash that may even corrupt it from within. This has significant policy implications, as multilaterals and foreign governments tend to push a laundry list of institutions à la Washington Consensus as a panacea for growth, while not considering local needs and culture. In this setting, it seems more reasonable to frame institutions locally, allowing for indigenous alterations, and perhaps even some suboptimal modifications, in exchange for higher likelihood of institutional efficiency. (Of course, the extent of adaptations needs to be carefully considered, so as not to entirely transform the institution.) Furthermore, the importance of the transplant process puts forth on the table whether or not we can openly talk about institutions and their performance, without considering the varieties of socio-economic and cultural contexts they are embedded in. Institutions do not occur in a vacuum. They may be sticky, but at the end of the day, they are still subject to local needs and demands. So when focusing on institutional performance, should we be discussing the set of characteristics individual institutions have (what incentives they provide, how they mitigate collective action problems, what mechanisms they use) or should we focus on the underlying conditions that support their performance?
As much as I enjoyed reading Berkowitz, Pistor and Richard’s argument, I had some difficulty with their methodology. In particular, in their main regression (p.183), they use one dummy variable for receptive and another dummy for unreceptive. From what I had understood, either the country was receptive (1) or not (0), so there’s no need to include both receptive and unreceptive in the regression as dummy variables (especially because that would make the X matrix non-invertible, and the regression wouldn’t work). Since they are able to run the regression, it must mean that receptive and unreceptive are not mutually exclusive, which I don’t yet quite understand.
The Boix reading this week brought to mind the Pamela Paxton article from Week 4 ("Women's Suffrage in the Measurement of Democracy: Problems of Operationalization"). Boix claims that the government has no incentive to change electoral rules until the electoral arena changes dramatically. At this time, the ruling parties consider altering the electoral system to maintain their advantage. He identifies universal suffrage as a key "big change", because it greatly increased the size of the electorate. He also states that countries which already had universal suffrage went through a political realignment at the same time, due to urbanization and industrialization. This realignment (from an urban/rural cleavage to a capital/labor cleavage) transformed the preferences of already enfranchised citizens. This causal story does not sit well with me, as it seems to be a catch-all explanation that defines a "threat" without explaining exactly how these two factors compounded to threaten the old parties. It also seems a bit convenient to say that universal suffrage was the big shock to the system, except where it wasn't, but there were preference changes that happened at the same time with the same overall effect! It is unclear whether increasing the size of the electorate, by itself, can convince ruling parties to consider electoral changes. Incorporating Paxton's argument about the inclusion of female suffrage at this point leads to some interesting questions. Female suffrage certainly increased the electorate size---at these times in history, did the ruling parties consider changing the electoral system? If they didn't, is it because the preference transformation did not occur at the same time, or did the ruling parties not feel threatened for other reasons?
Glaeser and Shliefer’s model of the adoption of civil codes versus common law frameworks is quite striking in its counterintuitive findings on the relationship between power and the imposition of judicial systems. I take issue with their framework because it briefly and uncritically glosses over the deeper commitment problem and necessary transfers that such a system would require to be effective. The framework ignores Moe’s injunction to pay attention to winners and losers and why they accept institutions in a particularly egregious way by noting that the most powerful actors in the system lose some of their powers and are compensated through unspecified transfers. Although they appeal to the Coase Theorem to motivate this argument, it is rather unpersuasive. Indeed, the Coasian framework that they adopt relies on the fact that power asymmetries do not impose an overlarge transaction cost on decisions over judicial institutions.
This manifests itself in the tension between an extremely powerful king, one who likely possesses sufficient resources to engage in large scale coercion of notables, and the choice of granting decentralized justice. Glaeser and Shliefer argue that in such a setting local notables were able to convince the king to set up socially optimal jury systems that were harder for him to manipulate, thus curtailing his own power. It is unclear why a powerful sovereign would ever consider divesting itself of power in exchange for transferred resources that it should be able to expropriate anyway given the implied power asymmetry. In so far as this creates a credibility problem, then, another mechanism is needed to persuasively argue that the monarch is really allowing for local juries because they diminish his power and improve social welfare. A similar problem exists when examining the decision of local notables to call for a system with professional judges. Fear of other local notables may explain a desire for such a system on the part of weaker notables with strong neighbors, but we are still left with the problem of what the stronger neighbor gets out of surrendering power over local outcomes. More damagingly, it is not clear that the king’s justice would be any more safe from strongly coercive local notables than a jury, since Glaeser and Shliefer note that the monarch’s power in such a system is usually circumscribed. It is also even more unclear than in the case of the strong sovereign what local notables are getting out of this arrangement in exchange for allowing others power. The inability of Glaeser and Shliefer’s theory to address why key actors might support a socially efficient outcome in the face of a diminishment of their utility, or even to specify how their utility loss is made up in exchange for supporting the socially efficient outcome, is a good example of the neglect that Moe’s article highlights. By removing the politics of winners and losers we are left with an interesting but ultimately inprobable theory. Moreover, by leaving these elements out of their formal analysis, it is unclear whether Glaeser and Shliefer’s comparative statics hold and much more difficult to test their theory.
Berkowitz, Pistor and Richard present an interesting analysis of the so-called “transplant effect” on economic development and “effective” legal institutions. While Berkowitz et al. present a compelling story, their analysis suffers from several methodological flaws that undermine the force of their conclusions. In the first instance, their analysis is deliberately limited to only 49 countries. This list primarily consists of European countries and their colonies in North and South America. Notable absences in the lists include eastern Europe and former Soviet countries, former African colonies, China, Caribbean nations and many Middle Eastern countries. Would the results change if the data included more countries in diverse regions of the world? Further, what is the difference between countries who have transplanted legal systems and ones who do not? They also do not explain anomalies in their data. Countries like Hong Kong, Singapore and Taiwan have comparatively high scores on “legality” despite the fact that they were unreceptive and involuntary transplants according to the data. Berkowitz et al.’s measure of “legality” also raises questions. They are based on survey data dealing with country risk assessments based on five “legality” variables. Is this the best measure? They retest their data using an alternative measure and find the generally the same results, but the issue still remains: Would Berkowitz et al’s results have changed if instead of using the composite index as their dependent variable, they regressed the independent variables on each individual composite score (i.e., efficiency of judicial system, rule of law, etc)? Are the results issue-dependent? Berkowitz et al’s model may also be too parsimonious. They state that they tested their results using other variables, but they fail to disclose exactly what those variables were. Indeed, how would Berkowitz et al’s results have differed if they controlled for whether a country has a common law or civil law system, a point discussed by Glaeser et al.? A host of other variables, including measures of de jure and de facto judicial independence, system of government and measures of bureaucratic efficiency, among others, could have bolstered the robustness of the results. Amy Semet
This week’s readings provide a variety of explanations of institutional origins. As for electoral systems, Boix argues that the choice of electoral systems in advanced democracies is contingent on the power contest between old parties and new parties after the electoral arena changes: if the new parties are weak or if the new parties are strong but the old is still dominant, the plurality/majority electoral rule will persist; otherwise, the old party will shift the plurality/majority rule to proportional representation. In contrast, Cusack et al. propose economic interests as the determinant of present electoral systems. They view that if business and unions had already established cooperative relations marked by co-specific skill investments in the early twentieth century when politics moved from the local to the national level, proportional representation would be adopted; if business and labor failed to cooperate and rather craft unions tried to control labor supply and skills at that time, majority rule would stay. As for legal systems, Glaeser and Shleifer maintain that the power configurations between the local feudal lords and the king in the twelfth and thirteenth centuries accounted for the different legal systems namely the common law system and the civil law system adopted by European countries. Likewise, Berkowitz et al. hold that it is the process of lawmaking and the demand for law in local niches that determined the performance of legal systems.
All of these works have done a good job in exploring the internal origins of institutions in advanced countries. However, as Berkowitz et al. found that, most institutions in many late developed or developing countries were externally transplanted by colonization or imported from early developers by voluntary imitation. For the late developers, the compatibility of externally originated institutions with the preexisting situations profoundly affects the institutions’ performance in the local environment. Many questions thus arise. What factors can systematically explain the variance of institutional compatibility in developing countries? Is the compatibility contingent on culture or history? Did political agents’ strategy or choice affect the result of institution transplant? Can we apply the theories of institutional origins based on the experiences of European countries to other countries? These questions deserve further study if we intend to better understand institutional origins.
While Glaesar et al interpret the adoption of a legal system as a conscious and deliberate decision on the part of the state; Berkowitz et al acknowledge that some legal systems are forcibly imposed. According to Berkowitz, the involuntary enforcement of a legal system could have longstanding effects on a state’s corruption and judicial effectiveness. This is particularly the case when the system imposed is not an efficient “match” with the recipient state’s cognitive understanding of the law. In many ways, Berkowitz establishes the critical link between the legal experience of European countries (which are mainly the focus of Glaeser) and states in the developing world, many of which acquired their legal systems through the process of colonization.
Something that I found particularly troubling about the Berkowitz piece is the possible influence that selection effects may have on many of the study’s findings. For example, the central dependent variable for this analysis was a composite variable measuring levels of corruption, low contract repudiation, low risk of government expropriation, and judicial effectiveness. Countries that were involuntary recipients or had the French civil code were among the worst performers in terms of this measure of ‘legality.’ I would argue that much of the variance in the dependent variable can be explained by the fact that many of these countries are “new states” with weak institutions, regardless of their “legal family or origins.” In addition, it is unclear how much of the results can be attributed to regime type, which interestingly enough is not controlled for in the regression analysis.
Something else to consider is the fact that varying levels of legality in former colonial territories may be attributable to other characteristics of colonial rule, rather than differences across legal systems. For example, recall the approach adopted by Acemoglu et al. that examines the impact of the types of colonial institutions that were established on economic growth (settler vs. extractive).
While the analysis and the empirical results were certainly interesting, it is difficult to disentangle the effects of regime type, colonial institutions/ legacy (aside from legal structures) and recent experiences of statehood on legality.
In contrast to Gleaser and Shleifer, I argue that institutions are not chosen for “efficiency” considerations. The efficiency approach, originating in transaction cost economics, and in its application to politics, holds that institutions and policies are chosen to maximize some measure of “social welfare”, and bargains are struck between players that enable all parties to capture gains from trade. According to this approach, cross country institutional differences are a result of variation in the circumstances/environment that players confront when designing “optimal” institutions (e.g. the presence or absence of powerful local magnates).
The efficiency approach is not a political theory of institutions. Fist, it neglects the commitment problem that is inherent in politics and that is not present in economic life: actors in the political arena cannot reach “Coasian agreements” as in Gleaser and Shleifer because there is no third party to enforce contracts. Second, policies and institutions are the result of distributional and social conflicts. When choosing institutions, there is no such thing as the maximization of social welfare. Instead, actors who hold political power may chose “inefficient” policies/institutions as long as these choices maximize their utility, and this is usually done at the expense of the rest of society. Thus, approaches to institutions based on efficiency considerations not only miss one side of the story (Moe). They miss the most important one to explain institutional origins.
A common theme of this week’s reading is the importance of power dynamics among actors for institutional development. Moe explicitly points out that neoinstitutionalist literature tends to favor analyses of processes within institutions while giving little formal attention to the emergence of the institutions. Thus, institutions are viewed largely as a rational response to collective action problems, rather than a product of the political agency that shapes their structure. In particular, he points out that there exists a fundamental distinction between private and public sector actors that renders the “economics of organization” of limited use for analysis. Clear property rights in private transactions give harmed parties an incentive to exit a potential transaction, while no such exit option exists for losers in the political process. This fits well with the analysis of Cusack, et al., which maintains that PR systems emerged when economic actors were equally inclined to cooperate with each other. Implied in this analysis is that the willingness of the actors to enter into a political contract to establish PR was crucial for the emergence of these systems. Alternatively, majoritarian systems remained in polities where relations between labor and capital were more conflictual, and where non-market coordination was unnecessary. However, note that the inability of actors to simply leave the political process necessarily led to the purposeful “undermining” of policies by the political losers: majoritarian systems were retained by bourgeois parties in order to undermine the goals of the left.
In their analysis of airport slots Riker and Sened offer an interesting case study of this dynamic. They reject the application of alternative explanations of the allocation of property rights, which ignore the role (and often the preferences) of the government. By underscoring the importance of agreement among all actors that the property rights should be granted, they offer an affirmation to the logic that power dynamics matter, and that institutions develop with the cooperation (whatever the intentions) of all affected actors.
In explaining the origins of the electoral rules, Boix constructs his theory around the premise of the conflicting interests between the right and the left (which provides the incentives to mitigate or block the “threat” via lowering or raising the electoral threshold); whereas Cusack et al. focuses on the incentives for class coalition (which leads the incumbent to adopt coordinative institutions). These two papers each tell “one side of the story.” As Moe points out, political institutions serve both the actors’ needs to tackle the collective action problem and to pursue their own interests (at the expense of others’). I think this point is helpful because it reminds us to look at the institutional creators’ both logics of control and cooperation. It also provides a useful analytical tool that transcends the left/right framework in Boix’s and Cusack et al.’s pieces, which does not always apply to other democracies, where the left/right cleavage is not the main political struggle (coordination). Moreover, it is not limited to the study of democracy. The development of electoral rules in “hegemonic party system” is an example.
Comment Week 7 – Julie Touber This week readings looked at institutional origins with two main approaches: one looked at legal aspects, the other at electorate structure. The various approaches make one big assumption that did not seem that obvious to me: the legitimacy of institutions as a set of rules enforced by a designated power comes from elections and/or some type of written law. But, it seems to me that legitimacy of rules –what makes rules acceptable to the society- is not necessary legalized or formal. I think this is the problem that arises from “the Transplant Effect”: the authors emphasize the importance of the process as opposed to the particular choice of a legal system. The Glaeser and Shleifer piece shows very clearly how the process itself has shaped the development of a specific legal set, therefore, some generic aspects may be found from one country to another one, but by definition every legal set should be specific –and they are-. “The Transplant Effect”, a very legitimate question in regard to the colonization history, makes some assumptions that, I believe, create a bias in their analysis.
First, the main assumption is that the origin country is homogenous and static. I believe this would be a major bias when it comes to understand the role of the colonies in the “national project”. The idea that France, for example, was a static territory (borders) with an accepted French identity for everyone, and that the colonies were an added territory under different administration is not right. Colonization did have economic interests but was not completely dissociated from a wider nation-building agenda. One can argue that the “transplant effect” was as traumatic from Paris to Quimper as it was from Paris to Algiers.
Second, the “transplanted legal systems” are by definition foreign to the observed country. The assumption is that there was no coherent legitimate organization on a large scale, existing before colonization. Thus little attention is brought toward the interaction of the transplanted legal system that is defined as formal and “legal” by Western standards, and the “native” order, that has became informal and illegal according the transplanted legal system.
This brings me to my final point on the “transplant Effect” article. It is stated that “Transplant countries suffer from the transplant effect”, but it is difficult to differentiate the nature of the legal system from the history of colonization. In other words, “the Transplant Effect” identifies negative impacts at different level of the “transplanted laws” depending on their origins, but I am not sure that it can be attributed to the “transplanted law” itself.
Berkowitz, Pistor and Richard highlight the importance of a country's receptivity to to institutions that are imported for the successful importation. That local 'demand' for an institution and its resonance with local customs is important for its functioning and relevance seems quite clear. Yet, as the authors note, this sound reasoning has been repeatedly ignored as institutional packages have been pushed on often unreceptive countries. The value of any given institution comes primarily form the outcome it is said to engender i.e. the security of property, the resolution of disputes through electoral channels, etc. and not from the specific rules themselves. This allows the possibility that institutions can be adapted to different contexts without the need for the adoption of a specific set of institutions that would otherwise be seen as 'the only way'. Qian's discussion of institutional innovations in China highlight this point. Without extending private property rights China was able to create the same incentives associated with property rights through anonymous banking, Township-Village Enterprises, and the dual track production system. As Rodrik points out in 'Institutions for High-Quality Growth: What they are and How to Acquire Them', suboptimal institutions that are actually followed can have more enduring effects than optimal ones that never leave the page. Also, transplantation of legal systems and other institutions often occurs in countries with limited state reach and capacity. This is likely to exacerbate the gap between the rules on the books and those in practice. Glaeser and Shleifer's article was very interesting in providing a logic for why royal judges and civil codes were created where the king was weaker vis-a-vis local magnates (France), while local juries existed where the king was stronger (England). Indeed, where the king is weaker he has little incentive to allow any further autonomy and power to nobility, yet the stronger king can afford to do so, particularly if he can collect taxes and ensure peace from such an arrangement. Yet their story fails to explain how it is that the king was able to enforce the rulings of a royal judge in a scenario where the king's power was weak. What were the incentives for the local magnates to accept that legal structure? Also, in cases where statutory law developed, such as France, where laws were imposed from a centralized plan is it possible to observe some kind of local transplant effect? The civil code was certainly foreign to many regions of France and could perhaps parallel the results of the transplant effect for other countries and colonies. Where the law developed as a result of the king's will it would indeed be a lot less sensitive to local needs and demands than common law, for example. Can this account for the differences in legality between the two systems that some have identified?
Glaeser and Shleifer (2002) present a well structured if somewhat stylized explanation for the historically observed differences in social and economic performance (w.r.t level of government regulation, corruption and bureaucracy) between common and civil law countries. They argue that the evolution of common law (i.e. independent juries) vs civil law (state appointed judges) systems can be traced to the extent of ‘local level bullying’. Where local bullying is significant – with strong feudal lords and weak central government (e.g. France) – civil law systems are most ‘efficient’ because local magnates are willing to accept the inherent bias of state-employed judges rather than rely on decisions from vulnerable local juries. In contrast, where local bullying is low – or where there is strong central power and weak local magnates (e.g. England)– common law jury systems are preferable because they can be protected from local cooption and thus can be relied on to adjudicate fairly.
Whilst the authors presented a well reasoned argument, I found their explanation flawed on numerous levels. Firstly, its focus on ‘efficiency’ as the singular driver of legal system preference is simplistic and fails to accurately reflect the multiple priorities that ultimately informed the design and adoption of national legal systems. Secondly, the dichotomous choice between civil and common law systems seems an artificial construct that does not adequately reflect the vast range of legal system permutations that have evolved historically. Finally I believe the argument fails to provide a convincing rationale for strong sovereigns to voluntarily devolve power to local level jurors, and/or powerful local lords to cede power to a relatively weak central power. History shows that indeed there are other options – experience in developing countries demonstrating how, even where formal structures dictate otherwise, powerful local magnates insist on their own locally appointed judges (e.g local power structures in India).
A more compelling and logically intuitive argument is presented by Berkowitz et al (2003) who look at the process of transplanting legal systems and the difference in outcomes between “receptive” versus “non-receptive” countries. This article argues that the type of legal system – i.e. civil versus common law systems - is less significant than the process of legal system development in predicting legal efficacy. It would be interesting to see whether this argument holds for other forms of institutions. A potentially important contribution to the development literature would be to use this methodology to empirically investigate the relative importance of ‘form’ over ‘process’ in the establishment of service delivery agencies and other bureaucratic institutions in a developing world context.
Terry Moe's 1990 article argues that the direction of positive theorists of institutions within political science have neglected perhaps the more crucial of the two purposes of institutions--as structural means by which political winners pursue their own interests (through coercion and redistribution)--while focusing on institutions as structures that 'help mitigate collective action problems'. They have done so, he insists, because the methodologies inherited from economics--specifically from study of social choice and new economics of organizations--have made them focus overwhelmingly on political processes including voting and coalition formation. Even the study of public bureaucracies, he points out, is done by focusing on the role of legislatures. Taking public bureaucracies as a focus area to bring out this ‘neglected’ aspect of institutions, he points out how the structural choices of policy makers are influenced by their beliefs about effective policy implementation, hedging against political uncertainty, compromise through issue trading, ability to get away with shirking the interests of their constituents.
Moe's critique is coherent and useful in understanding the arch of the literature until the time of the article's writing, and in making what seems like a sound argument about the ways political actor interests make ‘winners’ and ‘losers’ through public policy making and institutional structuring. One thing that seems suspect to me, though, is the assertion that—repeated several times in Moe’s piece with various wording—it is problem-free to assume that economic institutions of voluntary transactions leave all participants better off, the analytical methods used for them don’t fully work in politics because of the distinctive nature of public authority… Is it really problem free to assume this in economics? Of course, it is recognized that imperfect information—among other things—hinders market efficiency and increases inequality in many economic relationships, but isn’t it true some of the decision-making and power dynamics that come into Moe’s account of politics also apply to economics? The current economic crisis was caused by gambles by many actors (some knowledgeable and some not) to increase their wealth (and power) by taking on more debt than would ever be justifiable, causing outcomes that threw off the market equilibria. Similarly, inequality can reach extents where pressure that amounts to coercion is exercised over ‘voluntary’ participants in the system. Even while political scientists sometimes adeptly find the flaws or caveats in applying economic models and methods to politics, shouldn’t we cast a more critical eye on the original and overall suitability of those models and methods?
Berkowitz et al. (2003) contribute to the debate on the origins of legal institutions by critically reexamining the (accepted) effect of legal family on institutional effectiveness and drawing attention to the role of “receptiveness” and the “transplant effect” with respect to the way in which a country that is not an origin of formal law received its formal law. Specifically, they argue that adaptation of transplanted law to local conditions and familiarity of the receiving population with the transplanted law renders it more effective, whereas countries on which formal law is imposed without adaptation or absent prior familiarity will suffer from the “transplant effect”.
Berkowitz et al. take an important step by including in the analysis considerations beyond the type of law that is transplanted, specifically how “receptive” countries are for transplanted law. However, several questions arise with respect to their analysis. It seems that in addition to exogenous conditions “on the ground,” institutional effectiveness would also depend on how colonizers, for example, went about introducing formal law to their colonies. The extent to which transplanted law was applicable in the receiving country - initially and as time progressed - should have an effect on its effectiveness in the long run. Where local law prevailed for inheritance, family, and commercial matters while transplanted formal law was applied locally in more universal domains such as administrative and criminal law and gradually extended over time, as in many English colonies, we might expect the phasing in of formal law to affect the role of legal intermediaries and ultimate receptiveness, even though no prior familiarity existed and no explicit adaptation occurred.
Furthermore, looking to Shleifer and Glaeser (2002), the extent to which a country’s approach to transplanting its formal law to its colonies might be a function of the basic principles and values that its legal family reflects might be an important determinant in the analysis of institutional effectiveness. An emphasis on temporary persistence of local law for local issues – as in many English colonies – resembles the decentralized character of common law, while full imposition of all domains of transplanted law reflects the core of civil law. In short, Berkowitz et al.’s criteria for categorizing receptiveness might not only be too restrictive, but focusing on more endogenous features of how formal law was introduced in transplant countries might provide for a richer analysis of how the transplanting process of specific legal families might impact effectiveness.
In “Setting the Rules of the Game”, Boix develops a model of the origins of distinctive electoral systems. These electoral institutions, he argues, were themselves the product of strategic action of parties (primarily the ruling parties when a change in the electoral arena takes place). He models the relationship between mass enfranchisement, political cleavage structures, and electoral law choice, with incumbents’ policy preferences and expected strength in a post-reform political environment playing key roles in determining the degree of electoral law reform.
His approach differs from institutionalist models that describe political outcomes as equilibria and that, somehow trapped in static applications of game theory, hardly reflect on the origins of the institutions they claim constrain political actors. A strength of Boix’ approach is the attention he devotes to the idea of institutional change as the outcome of a political struggle between power-maximizing actors.
Boix' important contribution to studies of constitutional design is this: Rather than arguing about which constitutional design is "best" (as some comparativists do), he simply asks how our view of constitutional engineering changes when we acknowledge that the engineers (i.e. politicians/political parties) are not benevolent planners. In essence, he assumes that elites know and accept Cox's (1997) logic, and then asks what elites would do with this knowledge.
I think there are several problems with Boix’ piece – such as omitted variable bias – but I do find that in drawing our attention to the strategic deliberations of parties when it comes to institutional (electoral) change he forces us to consider how this affects our view of the development of distinctive electoral systems. In my view, it is an important dimension that must be addressed.
Riker and Sened posit that most descriptions of legal origins are not historically or scientifically convincing. They point towards the lack of empirical proofs for and the philosophical character of the natural rights and social contract theories. Unfortunately, Riker and Sened's "positive theory of the origin of property rights" falls into the same trap of philosophizing and abstract theory construction. In particular, it is not clear what Riker and Sened mean when they state that for property rights to emerge right-grantors need to "perceive an advantage from enforcing respect for the right" (p. 951). The vague formulation of this condition allows us to read a wide pool of benefits into right-grantor's utility equation. At the same time, Riker and Sened cite property rights as an example for benefits that right-grantors accord in exchange for popular support. Yet, popular support is not the only means of raising the right-grantor's utility. Property can be taxed. It creates opportunities for rent extraction. Efficient management through property rights can be also a source of pride and satisfaction for government officials – the authors claim, for example, that the Civil Aeronautics Board and the Federal Aviation Authority "clearly desired to improve their stewardship" (p. 957).Perhaps more interesting than the study of the emergence of property rights is the analysis of the determinants of the different government measures that bring about property rights. Riker and Sened provide an interesting discussion of the government's options for allocating airport slots. Researchers could examine in more depth why the CAB decided to ignore more efficient solutions to the scarcity of slots (such as the building of new airports or peak load pricing). Glaeser and Schneider, for instance, provide a good example of such process-oriented research by linking the emergence of different mechanisms for the protection of property rights to the level of conflict and the enforcement challenges prevalent in France and England during the middle ages.
In “Setting the Rules of the Game: The Choice of Electoral Systems in Advanced Democracies” Carles Boix argues that electoral laws can originate endogenously from the strategic decisions of dominant parties.
ReplyDeleteI wish to make a simple methodological point. Boix's makes a basic mistake in his regression analysis that undermines the robustness of his conclusions. In Table 1 (p. 619), Boix runs a multivariate regression with two interaction terms. "Threat" is an interaction term of the strength of socialism and the number of old parties variables. "Fragmentation X Area Dummy" is an interactive term of the combined index of ethnic and religious fragmentation and geographical area.
When using interactive terms in a multivariate regression, it is important to include both the interactive and the constitutive terms (i.e. the variables that make up the interaction term) in the regression. Not including the constitutive terms in the analysis can create omitted variable bias in the coefficients.
Boix does not include the constitutive terms for "Threat" in Model 1 and 3. Moreover, he fails to include the constitutive terms for "Fragmentation X Area Dummy" in Model 3. This failure to include the constitutive terms has real consequences. Brambor, Clark, and Golder (2006) replicate Boix's analysis, including the proper constitutive terms of the interactions. When the constitutive terms are included in the regression, there is no longer any evidence that the "Fragmentation X Area Dummy" interaction term has a statistically significant effect on the adoption of proportional representation. Conversely, by including the constitutive terms, the statistical significance of the "Threat" interaction term decreases (while remaining significant to the 90% level), while it's coefficient increases by 340%. Therefore, Boix strongly underestimates the interactive relationship between "OldParties" and "Socialism" on the adoption of proportional representation.
Reference:
Thomas Branbor, William Roberts Clark, and Matt Golder, "Understanding Interaction Models: Improving Empirical Analysis," Political Analysis, 14:1 (Winter 2006), 63-82.
Of the readings this week, the one that caught my attention the most was the Berkowitz, Pistor and Richard article on the transplant effect of legal institutions. The idea that legality—or the effectiveness of the legal order—depends more on the country’s receptiveness, focusing on the transplant process, and not on what was transplanted, is refreshing. While they focus on the legal family, their discussion brings forth the question of whether or not institutions can be discussed in a vacuum. The emphasis on the process of institutionalization, and not on individual institutional characteristics, forces us to consider the role of institutions in given contexts, particularly when addressing what Berkowitz, Pistor and Richard call “cognitive institution.” Society must have a demand for the cognitive institution, tweaking it and giving it a bit of local personality, so as to make it their own. It is this idea put forth by Berkowitz, Pistor and Richard that caught my eye. The transplant of institutions only has a chance of being durable and maintaining its integrity if the local culture is willing (and ready) to embrace it, otherwise, it is likely to suffer resistance and an inherent backlash that may even corrupt it from within. This has significant policy implications, as multilaterals and foreign governments tend to push a laundry list of institutions à la Washington Consensus as a panacea for growth, while not considering local needs and culture. In this setting, it seems more reasonable to frame institutions locally, allowing for indigenous alterations, and perhaps even some suboptimal modifications, in exchange for higher likelihood of institutional efficiency. (Of course, the extent of adaptations needs to be carefully considered, so as not to entirely transform the institution.)
ReplyDeleteFurthermore, the importance of the transplant process puts forth on the table whether or not we can openly talk about institutions and their performance, without considering the varieties of socio-economic and cultural contexts they are embedded in. Institutions do not occur in a vacuum. They may be sticky, but at the end of the day, they are still subject to local needs and demands. So when focusing on institutional performance, should we be discussing the set of characteristics individual institutions have (what incentives they provide, how they mitigate collective action problems, what mechanisms they use) or should we focus on the underlying conditions that support their performance?
As much as I enjoyed reading Berkowitz, Pistor and Richard’s argument, I had some difficulty with their methodology. In particular, in their main regression (p.183), they use one dummy variable for receptive and another dummy for unreceptive. From what I had understood, either the country was receptive (1) or not (0), so there’s no need to include both receptive and unreceptive in the regression as dummy variables (especially because that would make the X matrix non-invertible, and the regression wouldn’t work). Since they are able to run the regression, it must mean that receptive and unreceptive are not mutually exclusive, which I don’t yet quite understand.
The Boix reading this week brought to mind the Pamela Paxton article from Week 4 ("Women's Suffrage in the Measurement of Democracy: Problems of Operationalization"). Boix claims that the government has no incentive to change electoral rules until the electoral arena changes dramatically. At this time, the ruling parties consider altering the electoral system to maintain their advantage. He identifies universal suffrage as a key "big change", because it greatly increased the size of the electorate. He also states that countries which already had universal suffrage went through a political realignment at the same time, due to urbanization and industrialization. This realignment (from an urban/rural cleavage to a capital/labor cleavage) transformed the preferences of already enfranchised citizens. This causal story does not sit well with me, as it seems to be a catch-all explanation that defines a "threat" without explaining exactly how these two factors compounded to threaten the old parties. It also seems a bit convenient to say that universal suffrage was the big shock to the system, except where it wasn't, but there were preference changes that happened at the same time with the same overall effect! It is unclear whether increasing the size of the electorate, by itself, can convince ruling parties to consider electoral changes. Incorporating Paxton's argument about the inclusion of female suffrage at this point leads to some interesting questions. Female suffrage certainly increased the electorate size---at these times in history, did the ruling parties consider changing the electoral system? If they didn't, is it because the preference transformation did not occur at the same time, or did the ruling parties not feel threatened for other reasons?
ReplyDeleteGlaeser and Shliefer’s model of the adoption of civil codes versus common law frameworks is quite striking in its counterintuitive findings on the relationship between power and the imposition of judicial systems. I take issue with their framework because it briefly and uncritically glosses over the deeper commitment problem and necessary transfers that such a system would require to be effective. The framework ignores Moe’s injunction to pay attention to winners and losers and why they accept institutions in a particularly egregious way by noting that the most powerful actors in the system lose some of their powers and are compensated through unspecified transfers. Although they appeal to the Coase Theorem to motivate this argument, it is rather unpersuasive. Indeed, the Coasian framework that they adopt relies on the fact that power asymmetries do not impose an overlarge transaction cost on decisions over judicial institutions.
ReplyDeleteThis manifests itself in the tension between an extremely powerful king, one who likely possesses sufficient resources to engage in large scale coercion of notables, and the choice of granting decentralized justice. Glaeser and Shliefer argue that in such a setting local notables were able to convince the king to set up socially optimal jury systems that were harder for him to manipulate, thus curtailing his own power. It is unclear why a powerful sovereign would ever consider divesting itself of power in exchange for transferred resources that it should be able to expropriate anyway given the implied power asymmetry. In so far as this creates a credibility problem, then, another mechanism is needed to persuasively argue that the monarch is really allowing for local juries because they diminish his power and improve social welfare. A similar problem exists when examining the decision of local notables to call for a system with professional judges. Fear of other local notables may explain a desire for such a system on the part of weaker notables with strong neighbors, but we are still left with the problem of what the stronger neighbor gets out of surrendering power over local outcomes. More damagingly, it is not clear that the king’s justice would be any more safe from strongly coercive local notables than a jury, since Glaeser and Shliefer note that the monarch’s power in such a system is usually circumscribed. It is also even more unclear than in the case of the strong sovereign what local notables are getting out of this arrangement in exchange for allowing others power. The inability of Glaeser and Shliefer’s theory to address why key actors might support a socially efficient outcome in the face of a diminishment of their utility, or even to specify how their utility loss is made up in exchange for supporting the socially efficient outcome, is a good example of the neglect that Moe’s article highlights. By removing the politics of winners and losers we are left with an interesting but ultimately inprobable theory. Moreover, by leaving these elements out of their formal analysis, it is unclear whether Glaeser and Shliefer’s comparative statics hold and much more difficult to test their theory.
Berkowitz, Pistor and Richard present an interesting analysis of the so-called “transplant effect” on economic development and “effective” legal institutions. While Berkowitz et al. present a compelling story, their analysis suffers from several methodological flaws that undermine the force of their conclusions. In the first instance, their analysis is deliberately limited to only 49 countries. This list primarily consists of European countries and their colonies in North and South America. Notable absences in the lists include eastern Europe and former Soviet countries, former African colonies, China, Caribbean nations and many Middle Eastern countries. Would the results change if the data included more countries in diverse regions of the world? Further, what is the difference between countries who have transplanted legal systems and ones who do not? They also do not explain anomalies in their data. Countries like Hong Kong, Singapore and Taiwan have comparatively high scores on “legality” despite the fact that they were unreceptive and involuntary transplants according to the data. Berkowitz et al.’s measure of “legality” also raises questions. They are based on survey data dealing with country risk assessments based on five “legality” variables. Is this the best measure? They retest their data using an alternative measure and find the generally the same results, but the issue still remains: Would Berkowitz et al’s results have changed if instead of using the composite index as their dependent variable, they regressed the independent variables on each individual composite score (i.e., efficiency of judicial system, rule of law, etc)? Are the results issue-dependent? Berkowitz et al’s model may also be too parsimonious. They state that they tested their results using other variables, but they fail to disclose exactly what those variables were. Indeed, how would Berkowitz et al’s results have differed if they controlled for whether a country has a common law or civil law system, a point discussed by Glaeser et al.? A host of other variables, including measures of de jure and de facto judicial independence, system of government and measures of bureaucratic efficiency, among others, could have bolstered the robustness of the results.
ReplyDeleteAmy Semet
This week’s readings provide a variety of explanations of institutional origins. As for electoral systems, Boix argues that the choice of electoral systems in advanced democracies is contingent on the power contest between old parties and new parties after the electoral arena changes: if the new parties are weak or if the new parties are strong but the old is still dominant, the plurality/majority electoral rule will persist; otherwise, the old party will shift the plurality/majority rule to proportional representation. In contrast, Cusack et al. propose economic interests as the determinant of present electoral systems. They view that if business and unions had already established cooperative relations marked by co-specific skill investments in the early twentieth century when politics moved from the local to the national level, proportional representation would be adopted; if business and labor failed to cooperate and rather craft unions tried to control labor supply and skills at that time, majority rule would stay. As for legal systems, Glaeser and Shleifer maintain that the power configurations between the local feudal lords and the king in the twelfth and thirteenth centuries accounted for the different legal systems namely the common law system and the civil law system adopted by European countries. Likewise, Berkowitz et al. hold that it is the process of lawmaking and the demand for law in local niches that determined the performance of legal systems.
ReplyDeleteAll of these works have done a good job in exploring the internal origins of institutions in advanced countries. However, as Berkowitz et al. found that, most institutions in many late developed or developing countries were externally transplanted by colonization or imported from early developers by voluntary imitation. For the late developers, the compatibility of externally originated institutions with the preexisting situations profoundly affects the institutions’ performance in the local environment. Many questions thus arise. What factors can systematically explain the variance of institutional compatibility in developing countries? Is the compatibility contingent on culture or history? Did political agents’ strategy or choice affect the result of institution transplant? Can we apply the theories of institutional origins based on the experiences of European countries to other countries? These questions deserve further study if we intend to better understand institutional origins.
While Glaesar et al interpret the adoption of a legal system as a conscious and deliberate decision on the part of the state; Berkowitz et al acknowledge that some legal systems are forcibly imposed. According to Berkowitz, the involuntary enforcement of a legal system could have longstanding effects on a state’s corruption and judicial effectiveness. This is particularly the case when the system imposed is not an efficient “match” with the recipient state’s cognitive understanding of the law. In many ways, Berkowitz establishes the critical link between the legal experience of European countries (which are mainly the focus of Glaeser) and states in the developing world, many of which acquired their legal systems through the process of colonization.
ReplyDeleteSomething that I found particularly troubling about the Berkowitz piece is the possible influence that selection effects may have on many of the study’s findings. For example, the central dependent variable for this analysis was a composite variable measuring levels of corruption, low contract repudiation, low risk of government expropriation, and judicial effectiveness. Countries that were involuntary recipients or had the French civil code were among the worst performers in terms of this measure of ‘legality.’ I would argue that much of the variance in the dependent variable can be explained by the fact that many of these countries are “new states” with weak institutions, regardless of their “legal family or origins.” In addition, it is unclear how much of the results can be attributed to regime type, which interestingly enough is not controlled for in the regression analysis.
Something else to consider is the fact that varying levels of legality in former colonial territories may be attributable to other characteristics of colonial rule, rather than differences across legal systems. For example, recall the approach adopted by Acemoglu et al. that examines the impact of the types of colonial institutions that were established on economic growth (settler vs. extractive).
While the analysis and the empirical results were certainly interesting, it is difficult to disentangle the effects of regime type, colonial institutions/ legacy (aside from legal structures) and recent experiences of statehood on legality.
In contrast to Gleaser and Shleifer, I argue that institutions are not chosen for “efficiency” considerations. The efficiency approach, originating in transaction cost economics, and in its application to politics, holds that institutions and policies are chosen to maximize some measure of “social welfare”, and bargains are struck between players that enable all parties to capture gains from trade. According to this approach, cross country institutional differences are a result of variation in the circumstances/environment that players confront when designing “optimal” institutions (e.g. the presence or absence of powerful local magnates).
ReplyDeleteThe efficiency approach is not a political theory of institutions. Fist, it neglects the commitment problem that is inherent in politics and that is not present in economic life: actors in the political arena cannot reach “Coasian agreements” as in Gleaser and Shleifer because there is no third party to enforce contracts. Second, policies and institutions are the result of distributional and social conflicts. When choosing institutions, there is no such thing as the maximization of social welfare. Instead, actors who hold political power may chose “inefficient” policies/institutions as long as these choices maximize their utility, and this is usually done at the expense of the rest of society. Thus, approaches to institutions based on efficiency considerations not only miss one side of the story (Moe). They miss the most important one to explain institutional origins.
A common theme of this week’s reading is the importance of power dynamics among actors for institutional development. Moe explicitly points out that neoinstitutionalist literature tends to favor analyses of processes within institutions while giving little formal attention to the emergence of the institutions. Thus, institutions are viewed largely as a rational response to collective action problems, rather than a product of the political agency that shapes their structure. In particular, he points out that there exists a fundamental distinction between private and public sector actors that renders the “economics of organization” of limited use for analysis. Clear property rights in private transactions give harmed parties an incentive to exit a potential transaction, while no such exit option exists for losers in the political process. This fits well with the analysis of Cusack, et al., which maintains that PR systems emerged when economic actors were equally inclined to cooperate with each other. Implied in this analysis is that the willingness of the actors to enter into a political contract to establish PR was crucial for the emergence of these systems. Alternatively, majoritarian systems remained in polities where relations between labor and capital were more conflictual, and where non-market coordination was unnecessary. However, note that the inability of actors to simply leave the political process necessarily led to the purposeful “undermining” of policies by the political losers: majoritarian systems were retained by bourgeois parties in order to undermine the goals of the left.
ReplyDeleteIn their analysis of airport slots Riker and Sened offer an interesting case study of this dynamic. They reject the application of alternative explanations of the allocation of property rights, which ignore the role (and often the preferences) of the government. By underscoring the importance of agreement among all actors that the property rights should be granted, they offer an affirmation to the logic that power dynamics matter, and that institutions develop with the cooperation (whatever the intentions) of all affected actors.
In explaining the origins of the electoral rules, Boix constructs his theory around the premise of the conflicting interests between the right and the left (which provides the incentives to mitigate or block the “threat” via lowering or raising the electoral threshold); whereas Cusack et al. focuses on the incentives for class coalition (which leads the incumbent to adopt coordinative institutions). These two papers each tell “one side of the story.” As Moe points out, political institutions serve both the actors’ needs to tackle the collective action problem and to pursue their own interests (at the expense of others’). I think this point is helpful because it reminds us to look at the institutional creators’ both logics of control and cooperation. It also provides a useful analytical tool that transcends the left/right framework in Boix’s and Cusack et al.’s pieces, which does not always apply to other democracies, where the left/right cleavage is not the main political struggle (coordination). Moreover, it is not limited to the study of democracy. The development of electoral rules in “hegemonic party system” is an example.
ReplyDeleteComment Week 7 – Julie Touber
ReplyDeleteThis week readings looked at institutional origins with two main approaches: one looked at legal aspects, the other at electorate structure. The various approaches make one big assumption that did not seem that obvious to me: the legitimacy of institutions as a set of rules enforced by a designated power comes from elections and/or some type of written law. But, it seems to me that legitimacy of rules –what makes rules acceptable to the society- is not necessary legalized or formal. I think this is the problem that arises from “the Transplant Effect”: the authors emphasize the importance of the process as opposed to the particular choice of a legal system. The Glaeser and Shleifer piece shows very clearly how the process itself has shaped the development of a specific legal set, therefore, some generic aspects may be found from one country to another one, but by definition every legal set should be specific –and they are-. “The Transplant Effect”, a very legitimate question in regard to the colonization history, makes some assumptions that, I believe, create a bias in their analysis.
First, the main assumption is that the origin country is homogenous and static. I believe this would be a major bias when it comes to understand the role of the colonies in the “national project”. The idea that France, for example, was a static territory (borders) with an accepted French identity for everyone, and that the colonies were an added territory under different administration is not right. Colonization did have economic interests but was not completely dissociated from a wider nation-building agenda. One can argue that the “transplant effect” was as traumatic from Paris to Quimper as it was from Paris to Algiers.
Second, the “transplanted legal systems” are by definition foreign to the observed country. The assumption is that there was no coherent legitimate organization on a large scale, existing before colonization. Thus little attention is brought toward the interaction of the transplanted legal system that is defined as formal and “legal” by Western standards, and the “native” order, that has became informal and illegal according the transplanted legal system.
This brings me to my final point on the “transplant Effect” article. It is stated that “Transplant countries suffer from the transplant effect”, but it is difficult to differentiate the nature of the legal system from the history of colonization. In other words, “the Transplant Effect” identifies negative impacts at different level of the “transplanted laws” depending on their origins, but I am not sure that it can be attributed to the “transplanted law” itself.
Berkowitz, Pistor and Richard highlight the importance of a country's receptivity to to institutions that are imported for the successful importation. That local 'demand' for an institution and its resonance with local customs is important for its functioning and relevance seems quite clear. Yet, as the authors note, this sound reasoning has been repeatedly ignored as institutional packages have been pushed on often unreceptive countries. The value of any given institution comes primarily form the outcome it is said to engender i.e. the security of property, the resolution of disputes through electoral channels, etc. and not from the specific rules themselves. This allows the possibility that institutions can be adapted to different contexts without the need for the adoption of a specific set of institutions that would otherwise be seen as 'the only way'. Qian's discussion of institutional innovations in China highlight this point. Without extending private property rights China was able to create the same incentives associated with property rights through anonymous banking, Township-Village Enterprises, and the dual track production system.
ReplyDeleteAs Rodrik points out in 'Institutions for High-Quality Growth: What they are and How to Acquire Them', suboptimal institutions that are actually followed can have more enduring effects than optimal ones that never leave the page. Also, transplantation of legal systems and other institutions often occurs in countries with limited state reach and capacity. This is likely to exacerbate the gap between the rules on the books and those in practice.
Glaeser and Shleifer's article was very interesting in providing a logic for why royal judges and civil codes were created where the king was weaker vis-a-vis local magnates (France), while local juries existed where the king was stronger (England). Indeed, where the king is weaker he has little incentive to allow any further autonomy and power to nobility, yet the stronger king can afford to do so, particularly if he can collect taxes and ensure peace from such an arrangement. Yet their story fails to explain how it is that the king was able to enforce the rulings of a royal judge in a scenario where the king's power was weak. What were the incentives for the local magnates to accept that legal structure?
Also, in cases where statutory law developed, such as France, where laws were imposed from a centralized plan is it possible to observe some kind of local transplant effect? The civil code was certainly foreign to many regions of France and could perhaps parallel the results of the transplant effect for other countries and colonies. Where the law developed as a result of the king's will it would indeed be a lot less sensitive to local needs and demands than common law, for example. Can this account for the differences in legality between the two systems that some have identified?
Glaeser and Shleifer (2002) present a well structured if somewhat stylized explanation for the historically observed differences in social and economic performance (w.r.t level of government regulation, corruption and bureaucracy) between common and civil law countries. They argue that the evolution of common law (i.e. independent juries) vs civil law (state appointed judges) systems can be traced to the extent of ‘local level bullying’. Where local bullying is significant – with strong feudal lords and weak central government (e.g. France) – civil law systems are most ‘efficient’ because local magnates are willing to accept the inherent bias of state-employed judges rather than rely on decisions from vulnerable local juries. In contrast, where local bullying is low – or where there is strong central power and weak local magnates (e.g. England)– common law jury systems are preferable because they can be protected from local cooption and thus can be relied on to adjudicate fairly.
ReplyDeleteWhilst the authors presented a well reasoned argument, I found their explanation flawed on numerous levels. Firstly, its focus on ‘efficiency’ as the singular driver of legal system preference is simplistic and fails to accurately reflect the multiple priorities that ultimately informed the design and adoption of national legal systems. Secondly, the dichotomous choice between civil and common law systems seems an artificial construct that does not adequately reflect the vast range of legal system permutations that have evolved historically. Finally I believe the argument fails to provide a convincing rationale for strong sovereigns to voluntarily devolve power to local level jurors, and/or powerful local lords to cede power to a relatively weak central power. History shows that indeed there are other options – experience in developing countries demonstrating how, even where formal structures dictate otherwise, powerful local magnates insist on their own locally appointed judges (e.g local power structures in India).
A more compelling and logically intuitive argument is presented by Berkowitz et al (2003) who look at the process of transplanting legal systems and the difference in outcomes between “receptive” versus “non-receptive” countries. This article argues that the type of legal system – i.e. civil versus common law systems - is less significant than the process of legal system development in predicting legal efficacy. It would be interesting to see whether this argument holds for other forms of institutions. A potentially important contribution to the development literature would be to use this methodology to empirically investigate the relative importance of ‘form’ over ‘process’ in the establishment of service delivery agencies and other bureaucratic institutions in a developing world context.
Terry Moe's 1990 article argues that the direction of positive theorists of institutions within political science have neglected perhaps the more crucial of the two purposes of institutions--as structural means by which political winners pursue their own interests (through coercion and redistribution)--while focusing on institutions as structures that 'help mitigate collective action problems'. They have done so, he insists, because the methodologies inherited from economics--specifically from study of social choice and new economics of organizations--have made them focus overwhelmingly on political processes including voting and coalition formation. Even the study of public bureaucracies, he points out, is done by focusing on the role of legislatures. Taking public bureaucracies as a focus area to bring out this ‘neglected’ aspect of institutions, he points out how the structural choices of policy makers are influenced by their beliefs about effective policy implementation, hedging against political uncertainty, compromise through issue trading, ability to get away with shirking the interests of their constituents.
ReplyDeleteMoe's critique is coherent and useful in understanding the arch of the literature until the time of the article's writing, and in making what seems like a sound argument about the ways political actor interests make ‘winners’ and ‘losers’ through public policy making and institutional structuring. One thing that seems suspect to me, though, is the assertion that—repeated several times in Moe’s piece with various wording—it is problem-free to assume that economic institutions of voluntary transactions leave all participants better off, the analytical methods used for them don’t fully work in politics because of the distinctive nature of public authority… Is it really problem free to assume this in economics? Of course, it is recognized that imperfect information—among other things—hinders market efficiency and increases inequality in many economic relationships, but isn’t it true some of the decision-making and power dynamics that come into Moe’s account of politics also apply to economics? The current economic crisis was caused by gambles by many actors (some knowledgeable and some not) to increase their wealth (and power) by taking on more debt than would ever be justifiable, causing outcomes that threw off the market equilibria. Similarly, inequality can reach extents where pressure that amounts to coercion is exercised over ‘voluntary’ participants in the system. Even while political scientists sometimes adeptly find the flaws or caveats in applying economic models and methods to politics, shouldn’t we cast a more critical eye on the original and overall suitability of those models and methods?
Berkowitz et al. (2003) contribute to the debate on the origins of legal institutions by critically reexamining the (accepted) effect of legal family on institutional effectiveness and drawing attention to the role of “receptiveness” and the “transplant effect” with respect to the way in which a country that is not an origin of formal law received its formal law. Specifically, they argue that adaptation of transplanted law to local conditions and familiarity of the receiving population with the transplanted law renders it more effective, whereas countries on which formal law is imposed without adaptation or absent prior familiarity will suffer from the “transplant effect”.
ReplyDeleteBerkowitz et al. take an important step by including in the analysis considerations beyond the type of law that is transplanted, specifically how “receptive” countries are for transplanted law. However, several questions arise with respect to their analysis. It seems that in addition to exogenous conditions “on the ground,” institutional effectiveness would also depend on how colonizers, for example, went about introducing formal law to their colonies. The extent to which transplanted law was applicable in the receiving country - initially and as time progressed - should have an effect on its effectiveness in the long run. Where local law prevailed for inheritance, family, and commercial matters while transplanted formal law was applied locally in more universal domains such as administrative and criminal law and gradually extended over time, as in many English colonies, we might expect the phasing in of formal law to affect the role of legal intermediaries and ultimate receptiveness, even though no prior familiarity existed and no explicit adaptation occurred.
Furthermore, looking to Shleifer and Glaeser (2002), the extent to which a country’s approach to transplanting its formal law to its colonies might be a function of the basic principles and values that its legal family reflects might be an important determinant in the analysis of institutional effectiveness. An emphasis on temporary persistence of local law for local issues – as in many English colonies – resembles the decentralized character of common law, while full imposition of all domains of transplanted law reflects the core of civil law. In short, Berkowitz et al.’s criteria for categorizing receptiveness might not only be too restrictive, but focusing on more endogenous features of how formal law was introduced in transplant countries might provide for a richer analysis of how the transplanting process of specific legal families might impact effectiveness.
In “Setting the Rules of the Game”, Boix develops a model of the origins of distinctive electoral systems. These electoral institutions, he argues, were themselves the product of strategic action of parties (primarily the ruling parties when a change in the electoral arena takes place). He models the relationship between mass enfranchisement, political cleavage structures, and electoral law choice, with incumbents’ policy preferences and expected strength in a post-reform political environment playing key roles in determining the degree of electoral law reform.
ReplyDeleteHis approach differs from institutionalist models that describe political outcomes as equilibria and that, somehow trapped in static applications of game theory, hardly reflect on the origins of the institutions they claim constrain political actors. A strength of Boix’ approach is the attention he devotes to the idea of institutional change as the outcome of a political struggle between power-maximizing actors.
Boix' important contribution to studies of constitutional design is this: Rather than arguing about which constitutional design is "best" (as some comparativists do), he simply asks how our view of constitutional engineering changes when we acknowledge that the engineers (i.e. politicians/political parties) are not benevolent planners. In essence, he assumes that elites know and accept Cox's (1997) logic, and then asks what elites would do with this knowledge.
I think there are several problems with Boix’ piece – such as omitted variable bias – but I do find that in drawing our attention to the strategic deliberations of parties when it comes to institutional (electoral) change he forces us to consider how this affects our view of the development of distinctive electoral systems. In my view, it is an important dimension that must be addressed.
Riker and Sened posit that most descriptions of legal origins are not historically or scientifically convincing. They point towards the lack of empirical proofs for and the philosophical character of the natural rights and social contract theories. Unfortunately, Riker and Sened's "positive theory of the origin of property rights" falls into the same trap of philosophizing and abstract theory construction. In particular, it is not clear what Riker and Sened mean when they state that for property rights to emerge right-grantors need to "perceive an advantage from enforcing respect for the right" (p. 951). The vague formulation of this condition allows us to read a wide pool of benefits into right-grantor's utility equation. At the same time, Riker and Sened cite property rights as an example for benefits that right-grantors accord in exchange for popular support. Yet, popular support is not the only means of raising the right-grantor's utility. Property can be taxed. It creates opportunities for rent extraction. Efficient management through property rights can be also a source of pride and satisfaction for government officials – the authors claim, for example, that the Civil Aeronautics Board and the Federal Aviation Authority "clearly desired to improve their stewardship" (p. 957).Perhaps more interesting than the study of the emergence of property rights is the analysis of the determinants of the different government measures that bring about property rights. Riker and Sened provide an interesting discussion of the government's options for allocating airport slots. Researchers could examine in more depth why the CAB decided to ignore more efficient solutions to the scarcity of slots (such as the building of new airports or peak load pricing). Glaeser and Schneider, for instance, provide a good example of such process-oriented research by linking the emergence of different mechanisms for the protection of property rights to the level of conflict and the enforcement challenges prevalent in France and England during the middle ages.
ReplyDelete