What Explains Respect for the Rule of Law

What Explains Respect for the Rule of Law?

Evidence from a Cross-national Analysis of Structural Conditions

Svend-Erik Skaaning, Assistant Professor, PhD

Department of Political Science,

University of Aarhus, Denmark

With Niels Bossen & Jeppe Kehlet Sørensen

Paper prepared for the 66th MPSA Annual National Conference, Chicago, April 3-6, 2008

Abstract

This paper seeks explain the current global variation in respect for the rule of law. The empirical examination makes use of OLS-regression to test different structural explanations expected to facilitate or hinder the development of rule of law. Four different measures of rule of law are used interchangeably in the analysis, namely the rule of law sub-index from the Bertelsmann Transformation Index, Freedom House’s rule of law scores from the Freedom of the World Survey, the law and order index linked to the ICRG dataset provided by Political Risk Services, and the aggregate rule of law governance indicator constructed by the World Bank. Seven factors are included in the analysis to account for the outcome in question: oil production, modernization, country size, ethno-religious diversity, origin of legal system, dominant religion, and inequality. Both noteworthy differences and similarities in the results emerge from the analysis. Oil production, modernization, and inequality tend to be consistently associated with the degree of rule of law in the expected way, while the picture is more blurred with regard to the other factors; it basically depends on the measure used as dependent variable. The findings show that structural background conditions tend account rather well for the country-level variations in rule of law but they also call for more evaluation of the definition and measurement of this key concept and further inquiries into its causes.

Where the law is subject to some other authority and has none of its own, the collapse of the state is not far off; but if the law is the master of the government and the government is its slave, then the situation is full of promise and men enjoy all the blessings that the gods shower on a state

Plato, in Laws

Introduction

Throughout the history of political thought, the troublesome issue recurs, whether the better form of government is one based on the rule of men or on the rule of law (Bobbio, 1987: 138). Thus, the search for the best form of political rule not only concerns the contested question of who governs but also the equal question of how a country is governed. Whereas one side of this distinction has been at the center of attention in a large number of empirical studies, the other has hardly been subjected to general, systematic scrutiny – despite it obvious relevance from the perspective of ordinary citizens as well as the research discipline of political science in general and comparative politics in particular.

Guillermo O’Donnell has argued that, “The most critical aspect of a genuinely liberal state [is] the existence of a universalistic legal order” (1993: 10). He came up with this proposition after making the observation that many of the new regimes emerging in the aftermath of the third wave of democratization meet the criteria of polyarchy, while the rule of law is more frequently violated. In specific, often citizens are not subject to direct coercion when voting, votes are counted fairly, and their freedom of expression, organization, and assembly are observed, yet they cannot expect proper treatment from the police, the state agencies, or the courts (O’Donnell, 1993: 11-12). Such violations are frequent and serious in many countries, especially within the developing world. However, also here strong differences exists in the extent to which rule of law is violated.

This kind of variation naturally stimulates our curiosity to search for explanations. Also the intrinsic (freedom, equality) and functional (stability, efficiency) values of rule of law also underlines the importance of digging deeper into its causes. The rule of law is one of the essential pillars upon which any liberal democracy rests, as it has been emphasized by in more recent literature on the quality of democracy (Diamond & Morlino, 2004; O’Donnell, 2004; O’Donnell et al., 2004). Consequently, the present study has clear connections to the research field engaged in democratization and regime studies.

Furthermore, theoretical propositions and empirical investigations exclusively dedicated to account for the degree of rule of law are fairly underdeveloped. Scholars within the tradition of legal studies – producing the great majority of rule of law studies – have typically engaged in discussions regarding the nature of law, legal reasoning, legal systems, and legal institutions rather than general, empirical tests of theoretical explanations. The number of published studies representing general, statistical assessments of theoretical explanations is thus very limited, namely the works of Andrews and Montinola (2004), Barro (2000), and Joireman (2001, 2004). Moreover, out of these accounts for the rule of law, only two of them (Barro, 2000; Joireman, 2004) do not constrain the cross-sectional scope (i.e., the inclusion of countries) more than limits on data availability forced upon their studies.[1]

Instead of focusing on individual particularities, the attention is directed toward characteristics that vary systematically across a broad range of countries. Concerning the theoretical perspective of this paper, the main focus is thus placed on deep (structural, cultural) explanations, which are largely stable and outside the influence of current actors (Kitschelt, 2003: 63-64). The minimal attention previously directed towards empirical accounts for differences in the degree of rule of law means that no well-established explanatory factors exist within the rule of law field. Hence, it seems reasonable examine a broad range of theoretical factors and to set out from one end of the causal chain and then move forward in future studies addressing proximate conditions in the form of institutional and actor-centered explanations.

In answering the research question, what accounts for the global variation in present-day variation in the rule of law, the study in divided into four parts. The first step sheds light on the concept of rule of law and identifies four measures to be employed as dependent variables. Second, seven different theoretical explanations for differences in rule of law are presented and operationalized. In the third step, the theoretical accounts are examined by the use of multiple-regression analysis, and the results are compared and discussed. Finally, a conclusion sums up the findings and makes an outline for further investigations into the issue.

Definition and Measurement of the Dependent Variable(S)

A.V. Dicey popularized the concept rule of law (Harvey, 1961: 491) and his thoughts have inspired most legal philosophers and political theorists, who have advocated a formal view of rule of law. Within this tradition, Lon Fuller has suggested that in a country characterized by the rule of law, the law has to fulfill eight criteria: generality, publicity, prospectivity, clarity, non-contradictoriness, capability of compliance, stability, and congruence between declared rules and the acts of administrators (1981: 158). Rule of law is thus limited to the so-called formal characteristics of constitutional government or a Rechtsstaat. Hans-Joachim Lauth (2001: 33) has surveyed the literature on rule of law and, not very different from Fuller’s list, accentuates fourteen formal attributes shown in table 1.

Table 1: Principles of the Rule of Law (formal Rechtsstaat)

1.  General laws and not ad personam
2.  Publicly promulgated laws
3.  Prohibition against retrospective law
4.  Clear and comprehensive law
5.  No contradictory or inconsistent law
6.  No law should be impossible to respect
7.  Relative stability in the lawmaking
8.  Proportionality in the law
9.  Equality before the law
10.  Precedence of law, no one is above the law
11.  Independent and effective juridical control
12.  Due process of law
13.  Liability to pay compensation for any damage caused
14.  Absence of arbitrary state action and promotion of justice

Also normative political theory has addressed the rule of law. In the chief work on political philosophy in the last century, A Theory of Justice, John Rawls argues that rule of law is part of his overall scheme of justice as fairness. The rule of law is formal justice – understood as “the regular and impartial administration of law” – applied to the legal system (Rawls, 1971: 235-43). He advances two key arguments for establishing a linkage between the formal concept of rule of law and the basic value of liberty (cf. Radin, 1989: 788). First, in the absence of the formal characteristics of rule of law, that is, if rules are vague, if like cases are not treated alike, or if the judicial process is irregular, then “the boundary of liberty is uncertain and liberty is restricted by reasonable fear of its exercise” (Rawls, 1971: 235-243 [§ 38]). Thus, the rule of law is the natural choice of a legal structure since it promotes liberty through the provision of predictability, determinateness, and certainty of legal consequences. Second, the installation of a coercive sovereign, which has enforcement of penal sanctions as one of its core functions, restraints on the Leviathan are warranted.

There are good reasons to approve all the formal features presented above in a united package. However, it is analytically fruitful to recognize that the overarching concept covers at least three dimensions, namely the functioning of the legal system, respect for personal integrity, and maintenance of order. The first dimension is linked to the quality and functioning of the legal system. An independent and impartial judicial system, respect for the decisions of the courts, due process, and equality before the law constitute core attributes of this dimension. The second dimension covers the extent to which the government respects the personal integrity rights of its citizens, such as arbitrary arrests, police violence, and inhuman or degrading treatment (including torture). Finally, the third dimension concerns the ability of the state to uphold order, that is, reduce violence (including armed conflicts and riots) and crime throughout its territory.

After the conceptualization, the next move is the operationalization of the dependent variable. The rule of law measures have to fulfill a range of criteria to be used. First, the focus should be directed toward the actual level of rule of law, which excludes indicators on formal commitment. Second, the measures should primarily build on expert assessments in the form of standards-based data. Third, the scope of the data should fit the study’s scope, namely the contemporary degree of rule of law in all or, at least, most countries of the world.

A survey of available datasets points out four measures that (more or less) live up to these conditions and two falling short of the scope requisite. They are all presented in table 2, where the datasets and country-years to be used in the subsequent analysis have been emphasized.

Table 2: Rule of Law Measures (Generators and Empirical Scope)

Generator(s) / Dataset / Measure / Countries / Years
Bertelsmann Foundation (BF) / Bertelsmann Transformation Index / Rule of Law / 125 / 2003, 2005, 2007
Freedom House (FH) / Freedom of the World / Rule of Law / 193 / 2005-2006
Freedom House (FH) / Countries at the Crossroads / Rule of Law / 30(60) / 2003-2006
Global Integrity (GI) / Global Integrity Index / Rule of Law and Access to Justice / 55 / 2003, 2006, 2007
Political Risk Services (PRS) / International Country Risk Guide / Law and Order / 139 / 1984-2006
World Bank (WB) / Worldwide Governance Indicators / Rule of Law / 211 / 1996, 1998, 2000, 2002-2006

The first is the rule of law sub-index from the Bertelsmann Transformation Index[2] covering most non-OECD countries with more than three million citizens. The second is Freedom House’s scores for rule of law linked to the Freedom of the World survey.[3] The third is the law and order scores provided by Political Risk Services in relation to its International Country Risk Guide.[4] The fourth is the aggregate rule of law indicator constructed by the World Bank[5] based on many different data sources, including all those listed in the table.

Table 3: Rule of Law Measures (Attributes and Attribute Components)

Measure / Attributes / Attribute Components
Rule of Law
(BF) / Checks and balances / Working separation of powers
Independent judiciary
Legal or political penalties for power abusing officials
Civil rights / Protection of civil liberties and possibility to seek redress for violations
Rule of Law
(FH) / Independent judiciary / Judiciary not subject to interference
Judges appointed and dismissed in a fair and unbiased manner
Judges rule fairly and impartially
Governmental authorities comply with judicial decisions and these are enforced
Private concerns comply with judicial decisions and these are enforced
Rule of law in civil and criminal matters / Protection of defendants’ rights, including presumption of innocence
Access to independent, competent legal counsel
Fair, public, and timely hearing by competent, independent, and impartial tribunal
Prosecutors independent of political control and influence
Prosecutors independent of powerful private legal/illegal interests
Effective and democratic control of law enforcement officials
Law enforcement officials free from the influence of non-state actors
Protection from political terror, unjustified imprisonment, exile, torture, and freedom from war and insurgencies / No arbitrary arrests and detentions and no fabrication or planting of evidence
No beating of detainees or use of excessive force or torture
Conditions in pretrial facilities and prisons humane
Citizens have means of effective petition and redress
Population not subjected to physical harm, forced removal etc. due to conflict/war
Laws, policies, and practices guarantee equal treatment / Various distinct groups have equality before the law
Violence against such groups not widespread and perpetrators brought to justice
No legal/de facto discrimination in employment, housing etc. against such groups
Women enjoy full equality in law and practice
Non-citizens enjoy basic human rights
Law and Order
(PRS) / Strength and impartiality of the legal system
Popular observance of the law
Rule of Law
(WB) / Agents have confidence in the rules of society
Agents abide by the rules of society
High quality contract enforcement
High quality police
High quality courts
Low likelihood of crime
Low likelihood of violence

Although they all intend to measure the rule of law, their conceptualization of the concept is fairly different, as we see from table 3. The noteworthy differences (and similarities make it a reasonable choice to employ more than one, indeed all four measures – interchangeably – as dependent variable in the regressions. This procedure, not used in any of the previous studies, is followed in order to compare, and check the robustness of, the results. Notice, that the scores of the four measures have been standardized to go from 0 to 100, 100 signifying the highest rule of law level.