Procedural Rationality Review after Animal Defenders International: A Constructively Critical Approach

Abstract

Procedural rationality review – Quality of law-making – Fundamental rights – Relation to margin of appreciation – Guidelines to avoid the use of double standards

Throughout the years, the European Court of Human Rights has developed a noticeable, albeit not quite coherent practice of procedural rationality review. Procedural rationality review entails the Strasbourg Court’s consideration of the quality of the decision-making process at the legislative, administrative, and judicial stages to assess whether government interference in human rights was proportional.¹ In view of the idea of parliamentary autonomy, especially in jurisdictions organised on the principle of political constitutionalism,² review of the legislative procedure is the most contentious aspect. A milestone case in this respect was Hatton v the UK. The Court acknowledged that is was up to the national authorities to balance environmental and economic interests in the complex topic of aircraft noise pollution. It considered that it could nevertheless investigate the decision-making procedure to make sure that the national authorities had carried out a careful balancing exercise. The first chamber considered that the UK government had violated Article 8 of the Convention for not having made ‘specific and complete studies’ that allowed for the ‘right’ balance to be achieved. The Grand Chamber, however, upheld the legislation after relaxing the standard to be applied to the requirement that the law be based on ‘appropriate’ investigations and studies.³

Earlier research has highlighted the development of the procedural rationality review in the case law of the Strasbourg Court.⁴ The Grand Chamber added to it in Animal Defenders International v the United Kingdom (henceforth ADI),⁵ with the introduction of the general-measure doctrine, the principles of which were repeated in subsequent judgments.⁶ Since then, the fear has been voiced that procedural rationality, rather than enabling supranational oversight where wide discretion has been granted to national authorities, is used to avoid fundamental rights protection even where the margin of appreciation for the national authorities is narrow, and that inconsistency in the use of this type of review leads to double standards. This raises urgent questions as to the method and consequences of procedural rationality review when parliamentary acts are at issue. This article will, therefore, discuss the method of procedural rationality review of parliamentary acts and propose four rules of thumb.

The paper is structured as follows. The first section explains the general measure doctrine as developed in the Strasbourg case law since ADI. Next, a general framework for procedural rationality review of parliamentary acts is proposed, starting with a typology based on the purposes of such review, and then proposing the rules of thumb. These rules of thumb are related to the questions that have arisen from ADI and subsequent case law. One question is whether the Court should first establish a wide margin of appreciation for the legislator or whether, conversely, the quality of the lawmaking process should impact on the scope of such a margin. A related question is whether the quality of the lawmaking procedure justifies an outcome that is highly dubious from a fundamental rights point of view, especially in the light of precedent, or whether it should only be invoked if it is difficult to make such an assessment at first view. The last question elaborates on what procedural rationality review involves: how deeply may the Court interfere with the quality of the lawmaking process in general, and the parliamentary debate in particular, given the principle of parliamentary sovereignty? Throughout the article, the Strasbourg case law will be tested against this framework. The paper winds up with a conclusion.

The general measure doctrine

ADI v the UK revolved around a non-governmental organisation that campaigns against all forms of cruelty to animals. For one of its campaigns, in which it criticised the keeping and exhibition of primates and their use in television advertising, ADI had developed a television advertisement that compared a caged primate with a four-year-old girl in chains. The Broadcast Advertising Clearance Centre declined to allow the advertisement to be broadcast, due to a breach of the Communications Act 2003. As a result, the advertisement was not broadcast on television, although it could be viewed on the internet. ADI initiated proceedings requesting a declaration of incompatibility, but its claim was dismissed successively by the High Court and the House of Lords.

When ADI thereupon turned to the European Court of Human Rights, it was believed that the prohibition would not withstand the test at the Strasbourg Court, considering the importance attached to political speech and in the light of precedents such as Verein gegen Tierfabrike (VgT) and TV Vest.⁷ The applicant in VgT, like ADI, was a non-governmental organisation that campaigned for the protection of animals. It had been prohibited from broadcasting a television advertisement aimed at industrial meat production for the same reasons as those at play in ADI. The European Court of Human Rights acknowledged that the ban on political advertisement had a legitimate aim as it served to prevent powerful groups from obtaining a competitive political profit, thus ensuring the independence of broadcasters and sparing the political process from undergoing undue commercial influence.⁸ It reproached the Swiss authorities, however, for not having demonstrated why the prohibition also applied to VgT, as the latter was by no means a financially powerful group.⁹

In ADI, the UK Courts replied to this reproach by agreeing with the UK Parliament’s assumption that it was difficult to make such distinctions in law; this would create uncertainty, invite litigation, and could be circumvented through the formation of smaller, splintered groups.¹⁰ The UK government played out this argument as an intervening party in the case of TV Vest v Norway. In that case, the broadcasting company TV Vest had been fined for having broadcast a political advertisement for the Pensioners Party. As an intervening party, the UK government argued that ‘The Court appeared to have misunderstood the justification for a ban on political advertising, namely the fact that such a ban could not distinguish between different groups by reference to the power, funds or influence which they happened to have at a particular time’ and submitted a copy of the House of Lord’s ADI judgment.¹¹ This did not convince the Strasbourg Court, which simply stated that the prohibition and the fine were not proportional, because the rationale of the law addressed the major political parties, whereas the Pensioners Party was a small party for whose protection the ban was actually intended, and paid advertising was its only means of reaching a broader public.¹²

Against this background, the Grand Chamber’s judgment in ADI, deciding in favour of the prohibition by nine votes to eight, came as a surprise. One might have seen a harbinger in Murphy v Ireland, in which a total ban on religious advertising on radio and television had been deemed proportional, but in that case, religious, not political speech was at stake, and the margin of appreciation had been wide. The Court explicitly mentioned that this distinguished the case from VgT.¹³

In ADI, the Court repeated that for restriction of debate on matters of public interest the margin of appreciation for the national authorities is narrow.¹⁴ However, it then introduced a general-measure doctrine, holding that when a general measure is at stake, ‘the more convincing the general justifications […] are, the less importance the Court will attach to its impact in the particular case’.¹⁵ The central question was, therefore, not whether less restrictive rules should have been adopted, but whether the legislature had acted within its margin of appreciation when striking the balance it did.¹⁶ To that end, the Court held that it needed to assess the underlying legislative choices, taking into account the quality of the parliamentary and judicial review of the necessity of the measure as well as the risk of abuse if a general measure were to be relaxed.¹⁷ The Court emphasised that it considered the quality of the national parliamentary and judicial reviews as of ‘central importance’.¹⁸ It was more than satisfied in this case, praising the examination in Parliament of the cultural, political and legal aspects of the prohibition as ‘exceptional’.¹⁹ It referred to this ‘particular competence of Parliament and the extensive pre-legislative consultation on the Convention compatibility’ to explain the degree of deference shown by the UK courts, and was charmed by the level of detail by which the proportionality of the prohibition was nonetheless debated before the High Court and the House of Lords.²⁰ The Court again underlined the ‘considerable weight’ it attached to ‘exacting and pertinent reviews’ by national parliamentary and judicial bodies, before advancing some more substantial arguments.²¹ It noted that the prohibition only applied to paid political advertising and was confined to the most influential media, radio and television, whereas alternative media remained available.²² Addressing the applicant’s argument that a distinction should be made between political groups and social advocacy groups, the Court agreed with the UK Government that such differentiation was not feasible and would lead to uncertainty and litigation.²³ Finally, the Court did consider the concrete impact in the present case but held that it did not outweigh the justifications for the general measure since the applicant could still make use of alternative media.²⁴

Given this unexpected turn, it was to be expected that there would be sharp criticism of the Grand Chamber’s decision. Points of concern were raised by the dissenting judges and underlined in the scholarly literature. For example, questions were raised as to the origin of the ‘general measure’ doctrine,²⁵ the unclear distinction between a ‘general measure’ on the one hand and a ‘blanket ban’ on the other,²⁶ whereas the designation of a prohibition as one or the other would likely determine the outcome of the case, and the residual authority of precedent such as VgT.²⁷ Most important for this paper, however, is the criticism of the central importance given to the quality of parliamentary debate. The Court was reproached for dramatically reducing the intensity of its examination, whereas ‘it is difficult to see, from the rights-holder’s perspective, why the quality and quantity of debate should have a determinative impact on whether there has been a violation of his or her rights’.²⁸ Or, in other words, ‘the fact that a particular topic is debated (possibly repeatedly) by the legislature [does not] necessarily mean that the conclusion reached by that legislature is Convention compliant’.²⁹ Also, this could lead to a degree of incoherence in fundamental rights protection, since identical measures could be assessed differently, depending on the intensity of the parliamentary debate,³⁰ or depending on the origin of the interference.³¹ This would lead to a ‘double standard’.³²

The general reproach seems to be that the Court turned to procedural rationality review in order to justify what would otherwise have been considered a violation of a Convention right. The suspicion was voiced that the Strasbourg Court, faced with growing hostility in the UK and confronted with calls for a stronger concept of subsidiarity after the Brighton Declaration, was mainly trying to appease the United Kingdom.³³ In this respect, ADI deviates from the rationale for procedural rationality review as explained in the milestone judgments handed down in Hatton v the United Kingdom. In that case, the Court accorded a wide margin of appreciation to the national authorities, although emphasising that the Court nevertheless remains empowered to scrutinise procedural aspects.³⁴ Procedural rationality review was thus invoked as a tool to secure fundamental rights protection even when wide discretion has been granted to the national authorities.³⁵ If the Court is unable to assess the substantive merits of a law, it assesses whether the legislator has, at least, substantively considered the competing interests at stake and whether sufficient safeguards to that end have been built into the decision-making process. By contrast, after ADI, the Court has been reproached for using procedural rationality review to avoid fundamental rights protection even when the margin of appreciation for the national authorities is narrow.

The general measure doctrine was repeatedly invoked in subsequent cases. In several, the Court was satisfied by the quality of the parliamentary review, in isolation or in conjunction with judiciary review.³⁶ It stressed that the central question under the general measure doctrine was not whether less restrictive rules should have been adopted and that, without the contested rule, the legitimate aim could never have been achieved, but whether there had been relevant and sufficient grounds for the legislature’s choice.³⁷

The criticism kept resonating, to the point that the focus on the legitimacy of a measure that follows from the general measure doctrine, made the proportionality test ‘cosmetic’ and the outcome of the assessment a ‘foregone conclusion, if not a fatality’.³⁸ Nonetheless, the general measure doctrine does not by definition give Contracting Parties licence to constrain fundamental rights. In some cases, the general measure doctrine did not prevent the Court from finding a violation. In Bayev v Russia, the Court reiterated that, under the doctrine, the more convincing the general justifications for a general measure are, the less importance the Court will attach to its impact in the particular case. Here, however, the government had provided no sound justification whatsoever for the legislative ban on ‘propaganda of non-traditional sexual relations aimed at minors’.³⁹ In Ognevenko, the Strasbourg Court argued that the dismissal of a locomotive driver, which interfered with the applicant’s right to strike based on the right of association in Article 11 ECHR, did not rise to the level of being necessary in a democratic society. To come to that conclusion, it first repeated the principles laid down in ADI that govern the assessment of legislation.⁴⁰ Next, it expressed dissatisfaction with the quality of the parliamentary procedure: the absence of information explaining the general policy choice of a ban on the railway workers’ right to strike; the absence of a risk assessment to assess potential abuse were the prohibition to be removed; the absence of any consideration of alternatives to imposing a ban, such as minimum services.⁴¹

The general measure doctrine, however, provides grounds to accuse the Court of and criticise if for applying a double standard. In Ognevenko, the dissenting judge accused the Court of sloppiness, of taking on ‘an enlightenment role’ by imposing and even inventing European standards and by finding a violation ‘for pedagogical purposes’, which is ‘quite humiliating for the national authorities’.⁴² Importantly, he reproached the Court for deviating from the principles under the general measure doctrine by which the question should not be whether less restrictive rules should have been adopted, or whether the State could prove that without the measure the legitimate aim would not have been achieved.⁴³

This illustrates the growing concern that the increasingly prominent role of the margin of appreciation and the subsidiarity principle is only being applied to certain Contracting Parties, at the risk of this use of double standards ‘downgrading’ the role of the Strasbourg Court and diminishing its authority to that of a non-judicial advisory committee of experts.⁴⁴ Consistency of purpose and method is thus of utmost importance. The remainder of this article will, therefore, discuss the method of procedural rationality review when parliamentary statutes are at issue.

Models for procedural rationality review

The procedural rationality review of parliamentary procedures is, although it can be observed in the case law of European as well as national courts,⁴⁵ a relatively recent practice. This raises the question of why courts should suddenly feel the urge to assess the quality of parliamentary procedures to adjudicate fundamental rights issues. Courts do not always articulate their reasons for engaging in it, but in fact, procedural rationality review may serve as: (i) a substitute for substantive review; (ii) an escape route; or (iii) a tool to strengthen the proportionality analysis. In what follows, these are labelled, respectively, the substitute model, the escape route model, and the compensatory model.⁴⁶

In its most radical version, the promotion of procedural rationality review as a substitute for substantive review disguises the aversion towards the very notion of judicial review of parliamentary acts – in general, but especially by a supranational court.⁴⁷ Such scepticism is based upon the conviction that Parliament is in all circumstances ‘better placed’ to adjudicate fundamental rights issues. Scrutiny of the decision-making procedure, then, only resolves the tension if it serves to support the parliamentary act, not if it is used to criticise the parliamentary procedure. On the contrary, a judicial critique of parliamentary processes is regarded as even more intrusive than a substantive review.⁴⁸ Strikingly, while the UK Courts remain reluctant to conduct procedural inquiries, even when reviewing Acts of Parliament under the Human Rights Act 1998,⁴⁹ in ADI they did refer to the parliamentary process, but only to praise its quality and uphold the law. More generally, as Kavanagh demonstrates, on the occasions that the UK Courts do take the parliamentary debate into account, this is most often done to support the compatibility of the parliamentary Act with fundamental rights obligations; only rarely is a lack of consideration of parliamentary debate referred to when criticising a parliamentary Act.⁵⁰ Strikingly, proponents of this view tend to distinguish ‘well-functioning democracies’ from underperforming democracies, pleading that procedural rationality review should replace substantive review only in the former – undefined – cases.⁵¹ This not only opens the door to double standards; it flatly advocates it.

In a milder version, procedural rationality review substitutes for substantive review in a model characterised by judicial restraint. The thoroughness of parliamentary debate signals the quality of the democratic process, and this reduces the role of the courts. Here as well, a model of political constitutionalism that keeps the organisation of checks and balances within the political sphere⁵² is favoured, although the possibility that Parliament may not have fulfilled its function remains an option, thus leaving room for more intensive judicial scrutiny. It has been noted that the political process can sometimes be lacking, and that a parliamentary majority can sometimes mask regulatory commandeering by special interest groups.⁵³ In this model, the quality of parliamentary debate is used as a lock on the margin of appreciation: if the parliamentary debate was rigorous, the margin is wide, and the Strasbourg Court will remain aloof, and vice versa. The crossing point between the mild and radical versions lies in the criteria used to assess the quality of parliamentary debate.⁵⁴ In the milder version, the criteria are more detailed as to which safeguards should be expected to be found in the process, with more weight given to whether individuals or minority groups have had a voice in the debate. In the more radical version, the criteria are vague, which increases the risk of arbitrary use.

From a more tactical point of view, procedural rationality review can serve as an escape route. Messerschmidt has noted that ‘in the beginning’, even the German Constitutional Court, which has the longest track record when it comes to this type of review, used procedural rationality review ‘as a stopgap’: ‘in some hard cases the Court did not want to sustain a motion nor did he dare to overrule it’.⁵⁵ Tactical considerations fit in with a case-to-case approach, but even then, the Court will, at some point, need to indicate when and how this type of review should be used, if only to bring a bit of coherency to its case law and to avoid being labelled arbitrary.

For a fundamental rights court, the purpose of procedural rationality review as a tool to protect fundamental rights makes the most sense when the Court is unable to substantively assess the merits of a case. In the literature, this is called the ‘compensatory function’.⁵⁶ If priority is given to national authorities in a specific case, this is not based simply on irrefutable grounds of national or parliamentary sovereignty. Instead, the European Court of Human Rights names three reasons for giving priority to national authorities: their ‘direct democratic legitimation’; the fact that they are ‘better placed to evaluate local needs and conditions’; and so that they have special weight in matters ‘of general policy, on which opinions within a democratic society may reasonably differ widely’.⁵⁷ This corresponds to the three main reasons identified in the literature for determining the margin of appreciation: democratic legitimacy; expertise; and the (absence of) common practice of states.⁵⁸ Democratic legitimacy presupposes that a substantive debate has taken place; expertise presupposes that this was an informed debate, based on evidence and knowledge rather than mere assumptions.⁵⁹ Procedural rationality review, then, tests whether the assumption that the national Parliament was better placed to make a legitimate, informed and expert-based decision, was correct. While, in practice, the European Court of Human Rights never relies on procedural assessment alone,⁶⁰ in this model it, in essence, merges it with its substantive procedural review.⁶¹

A finding that the Strasbourg Court shifts back and forth between these rationales is understandable, considering the increasing criticism voiced over the past few years of its interference in domestic affairs. It is important for the Court to be perceived as a legitimate body in the interest of staving off demise or reform.⁶² The recourse to procedural rationality review could, therefore, be explained as the result of the emphasis on the subsidiarity principle and the margin of appreciation, laid down in Protocol No. 15, to ensure more respect for the local dimension of human rights.⁶³ Whether the (milder) substitute model or the compensatory model is to be preferred is, thus, open to discussion. In taking a position in this discussion, legal culture will probably play some role of significance,⁶⁴ although strategic considerations may also carry weight. Abundant empirical research has shown the importance of strategic considerations in explaining the behaviour of courts.⁶⁵ Gerard’s observation that the European Court of Human Rights tends to turn to procedural review especially in cases with ‘a high degree of sensitivity’⁶⁶ would seem to indicate strategic considerations. Yet, if a substitute model is to ease relations between Court and Contracting Parties, it will not be very successful unless it elides into a more radical version. As Angelika Nussberger, a judge at the Strasbourg Court, has noted, if the Court’s review results in criticism of domestic procedures, this may cause even more tension in that relationship.⁶⁷ At the same time, the discretionary use of process review by the European Court of Human Rights has been identified as a point of concern.⁶⁸ Guidelines for more consistent use of procedural rationality review should, therefore, be welcomed.

In this paper, the position is taken that for a fundamental rights court, the compensatory function is the most promising model; it would seem to offer more protection for individuals and under-represented groups against majority groups and entail less risk of arbitrary use, while still respecting the local dimension of human rights adjudication.⁶⁹ By contrast, the substitute model proceeds from the assumption that a diligent parliamentary process leads to a reasonable, Convention-compliant outcome. However, quoting Nussberger once more, ‘while there might be a presumption that an unfair procedure leads to an unfair result, there is no guarantee that sound procedures result in fair outcomes’.⁷⁰ Elsewhere as well, it has been noted that ‘when procedural review is not added to or integrated into substantive scrutiny, but is instead used to replace substantive review, there is a real risk that a tendency towards proceduralisation of review would lead to weakening substantive rights protection’.⁷¹ Therefore, this paper presents four rules of thumb as a general framework for the consistent use of procedural rationality review with the purpose of furthering the protection of fundamental rights. Additionally, the last two rules of thumb can also serve as guidelines for use of the mild substitute model. This links, in this respect, in with the ‘modest role’ for procedural review by courts favoured by Sathanapally.⁷²

The rules of thumb developed in the next sections are:

1. (1) The Court should turn to procedural rationality review when it is unable to substantively assess the merits of a case.

2. (2) Consequently, substantive arguments should prevail if there are serious grounds to argue either conformity or violation of the challenged Act. Procedural rationality review can play a more important role when there are doubts, i.e. so-called hard cases.

3. (3) Evidence used by Parliament should only be questioned if there are serious reasons to doubt its quality.

4. (4) If the Court praises the quality of Parliamentary debate as a means of justifying a questionable measure, it should, in particular, make sure that there was not only extensive debate on the subject in general but that there was also an informed discussion of the relevant legal questions in particular.

When should procedural rationality review be used? the margin of appreciation

If procedural rationality review is to serve as a tool to strengthen fundamental rights protection, then it would make sense to turn to it only in those cases in which the court is unable to substantively assess the merits of a case. From this angle, the margin of appreciation becomes increasingly relevant. Procedural rationality review should play a more important role when the margin of appreciation is wider and become less prominent when the margin of appreciation is narrow. This is also how the current president of the European Court of Justice explains why the Court is following the current trend of engaging in procedural rationality review: it is an ‘interesting way’ of ensuring judicial scrutiny ‘in areas where the […] legislator enjoys broad discretion’ without ‘intruding into the realm of politics’.⁷³ In this respect, it is important to understand the rationale behind the margin of appreciation as a tool for making the principle of subsidiarity operational.

In its case law, the European Court of Human Rights expresses some ambiguity about this relationship. Even though it often leaves a bit of uncertainty as to the scope of the margin of appreciation, it usually turns to procedural rationality review when the margin is relatively wide.⁷⁴ For example, in cases like Hirst,⁷⁵ Hatton,⁷⁶ Lecarpentier,⁷⁷ etc., the Court allowed a broad margin of appreciation. ADI broke with this practice; the Court recognised – in line with its previous case law – that in matters of political speech, ‘the margin of appreciation to be accorded to the State in the present context is, in principle, a narrow one’.⁷⁸ Even though the Court went on to state that the lack of European consensus had broadened the margin of appreciation,⁷⁹ this only served to make it less narrow, not broader. In Ognevenko the Court left the question open, stating merely, and more generally, that states have only a limited margin of appreciation in assessing whether there is a ‘pressing social need’ to adopt certain measures.⁸⁰

This does not mean that procedural rationality review has no role at all when the margin of appreciation is narrow; it can be invoked whenever the court is unable to test the legislative assumption on which the law is based. However, it cannot take a prominent place; it only helps underpin the substantial assessment of the case. For example, in Ognevenko the Court explained, referring to international standards and discussion, why it was not convinced that railway transport should be considered an essential service for which the right to strike could be restricted. The Court, looking for evidence to the contrary, implicitly held that the Russian parliamentary debate had been an uninformed deliberation rather than examine it at any depth; the Government had not provided any evidence to substantiate the assumption that a complete ban on the right to strike for certain railway workers was necessary. Nor had it considered alternatives to the ban or safeguards for those prevented from exercising their right to strike.⁸¹ In ADI, by contrast, the margin of appreciation was relatively broad; the quality of parliamentary debate was nevertheless considered ‘of central importance’ in deciding the case.⁸²

More ambiguity stems from the way the Court describes the relationship between the margin of appreciation and procedural rationality review. In Sukhovetskyy it held that ‘the extent of the State’s margin of appreciation depends on the quality of the decision-making process’.⁸³ This was applied in ADI and other cases.⁸⁴ Phrased this way, the margin of appreciation should be broad if the decision-making process is of high quality and narrow if it is not. The Court would then find itself using the mild version of the substitute model, which, however, may easily flow into its more radical version. By contrast, if procedural rationality review is to serve as a tool to secure fundamental rights review when the Court is unable to assess the merits of a case, it should first determine the scope of the margin of appreciation, and only then resort to a review of the quality of the decision-making process. The latter review should be used as part of the proportionality analysis, rather than to determine the scope of the margin of appreciation.

The reproach voiced after ADI, i.e. that the Strasbourg Court focuses on the quality of parliamentary debate in order to avoid substantive scrutiny of a law, which at first sight might seem highly dubious in the light of earlier precedent, should be taken seriously. This highlights the importance of defining the relationship between procedural rationality review and the margin of appreciation; the latter ‘determines the strength with which the state’s activity is scrutinized in a particular instance’.⁸⁵

Hard cases

Animal Defenders International brings into sharp focus the discussion of whether substance or procedure should prevail.⁸⁶ The question is usually put as follows: should a law be set aside as unconstitutional if it suffers from insufficient deliberation and consideration of the facts, yet fully conforms with substantive constitutional law? In the scholarly literature, it has been argued that it should not: courts should only turn to procedural rationality review when there are doubts about the constitutionality of the law⁸⁷ or if the court has established an infringement but still has to determine whether the infringement was justified.⁸⁸ Hence, in Ognevenko, continued criticism by international organisations of the overly restrictive labour regulations in Russia were able to raise doubts about the treaty-conformity of the law. In ADI, however, the question was put in reverse: can a law be deemed consistent with the terms of a treaty if it was the result of an adequate lawmaking process, although its conformity with substantive law can be severely doubted in the light of earlier precedent? The literature has warned of such a weakening of the traditional standards of substantive review.⁸⁹

According to a minimalist view of judicial review, what matters is whether the court can establish that the law infringes upon a fundamental right. The question of whether the infringement was justifiable should be examined from an exclusively procedural point of view. Or, in a more modest proposal, the ‘proportionality stricto sensu test’ should at least be replaced by procedural review.⁹⁰ ADI conforms perfectly to this model. Whether the prohibition amounted to an infringement of Article 10 ECHR was not discussed.⁹¹ In a second step, the quality of parliamentary and judiciary scrutiny was praised, which is why, in a third step, the Court’s own proportionality test was less intensive. In Ognevenko, the Court had not been convinced of the quality of domestic scrutiny and therefore performed a more intensive proportionality test as a third step.

The criticism with which ADI was met, however, shows the risks of taking such an approach; see the problems listed above. In the light of previous case law, the expectations were that the European Court of Human Rights would find the prohibition to violate the Convention and that the prominent place given to the quality of the legislative process suggested a double standard by which the review of identical measures could lead to different outcomes. In this light, it is concerning that the Strasbourg Court chose to implicitly and not explicitly overrule precedents like VgT and TV Vest. By neither giving reasons for overruling those precedents nor explicitly defining their residual authority, the Court fed the suspicion that it had used arbitrary grounds in order to appease the United Kingdom. The dissenting judge’s criticism in Ognevenko adds to this impression. Whereas in ADI UK institutions were praised for their ‘exceptional’ level of scrutiny and in Satakunnan Markkinapörssi Oy and Satamedia Oy the parliamentary review was considered ‘both exacting and pertinent’,⁹² the dissenting (Russian) judge in Ognevenko felt that higher standards were being imposed on the Russian legislature.⁹³

The recommendation that results from this section, and which follows naturally from the first rule of thumb, is that procedural rationality review should only be allowed to play a prominent role in hard cases.⁹⁴ A case is ‘hard’ if settling it demands a balancing of interests the outcome of which cannot easily be predicted on the basis of e.g. legal provisions, precedents or European consensus, or because of the complexity of the case. If there are no serious grounds to doubt a law’s conformity with fundamental rights obligations, there is no need to turn to procedural rationality review. Likewise, if such serious grounds do exist, procedural rationality review may be used to support the proportionality analysis, although substantive arguments should be allowed to prevail. The Ognevenko case lived up to this recommendation, but the ADI case did not.

The intensity of procedural rationality review: assessing evidence

As a dissenting judge in the Finnish case stressed, an appraisal of the quality of parliamentary review can seriously constrict the ability of the court to contradict the balance of competing interests struck by the legislature.⁹⁵ In the end, then, the question is which standards the Court should use to assess the quality of the legislative and judicial decision-making process.

The cheers for domestic scrutiny in certain cases (ADI; Satakunnan Markkinapörssi Oy and Satamedia Oy) as opposed to its neglect in another (Ognevenko) leaves the contracting parties with little guidance and also carries the risk of the Court being perceived as biased with regard to the quality of work performed by domestic institutions.⁹⁶ One way to alleviate that risk would be to adopt standards that can be applied uniformly in all cases. This raises the question of the intensity with which a court should review the quality of the legislative process. Is it up to the court to lay down certain standards or impose a blueprint for rational lawmaking and judicial review? Or should it be satisfied if there is evidence that the issue has been extensively debated and that several considerations have been taken into account? Should the Court simply check whether an extensive debate has been held and that evidence was used, or should it also assess the value of that evidence?

If deference is given to the legislature on the basis of its assumed expertise then the court may safely assume that reliable evidence was used in the law-making process.⁹⁷ This approach is also given credence by the fact that judges often lack the capacity to assess the quality of scientific evidence.⁹⁸ The presumption may then be reversed only if there are strong indications that the evidence is inadequate, e.g. if it is manifestly flawed or if the court has been given strong evidence to the contrary. An example of the former situation can be found in Smith and Grady v the United Kingdom. Whereas the Court, as a rule, refuses to discuss the quality of evidence used by the national authorities,⁹⁹ there it raised explicit doubts about the quality of the report that underpinned the policy of discharging homosexual members of the armed forces. The Court argued that ‘the independence of the assessment contained in the report is open to question given that it was completed by Minister of Defence civil servants’, that the number of respondents was too low, that the consultations, one-to-one interviews and focus group discussions had not been kept anonymous, and that many of the questions in the survey were suggestive in nature.¹⁰⁰ The latter is illustrated by Kiyutin, in which the Court noted a consensus among expert and international bodies active in the field of public health that travel restrictions on people living with HIV could not be justified by health concerns; it, therefore, required that such travel restrictions were based on ‘expert opinions or scientific analysis […] capable of gainsaying the unanimous view of international experts’.¹⁰¹ From Zelenchuk and Tsytsyura it follows that when the applicant and third-party interveners provide evidence to counter legislative assumptions, this strengthens the government’s duty to prove that the law rests on sound ground.¹⁰²

In ADI, the Courts relied on the Government’s evidence as a matter of principle. The High Court of Justice was prepared to consider the evidence for the applicants given by an expert regarding the pervasiveness of broadcast media, an issue that had not been considered by the Government. However, the report was dismissed for comparing only the impact of messages provided in televisual versus print form, but not analysing the effect of advertising on persons to whom television or radio is the only or main source of information.¹⁰³ The European Court of Human Rights simply stated that there was ‘no evidence of a sufficiently serious shift in the respective influences of the new and of the broadcast media’.¹⁰⁴ By contrast, in Ognevenko, no evidence dealt with in the parliamentary debate was put forward to argue that the ban on the right to strike relied on solid grounds. When Parliament remains silent, the Court is still willing to weigh the evidence that the concrete decision-making process was proportional. Here the requirements can be stricter, as more detailed data is usually available in concrete cases. In this case, the Court was not satisfied with the report introduced by the government because no precise data was given on the damage caused by delayed arrivals and the danger caused by overcrowded platforms.

The intensity of procedural rationality review: assessing parliamentary debate

On the basis of the sovereignty-of-parliament principle, the autonomy of Parliament in organising the law-making process should be acknowledged. This is why judicial interference in political processes is considered a very sensitive issue, to the point that a scholar commenting on Hirst v the United Kingdom explained that judicial evaluation of Parliamentary debate is ‘a constitutional anathema in the domestic context’.¹⁰⁵ Hence, if procedural rationality review prompts the court to assess the quality of the parliamentary law-making process, a minimal approach is required. This should not result in the court prescribing specific procedures or tracing out in great detail which steps to take, but merely in establishing whether the procedure guaranteed an extensive and well-informed examination of facts and interests. The Court, thus, rightfully shows its reluctance to define how the national parliamentary process should unfold.¹⁰⁶ The matter of what should be debated in that process is another question entirely.

In ADI, the European Court was charmed by the fact that parliamentary bodies had examined ‘the cultural, political and legal aspects of the prohibition as part of the broader regulatory system governing broadcast public interest expression in the United Kingdom’.¹⁰⁷ As for the legal aspects in ADI as well as in earlier judgments, the Court’s focus has been in particular on whether a proposed measure had been examined in the light of the European Court of Human Rights’ case law.

Dissenting judges, however, reproached the UK authorities for not having examined in detail the impact on non-wealthy social advocacy groups and the possibility of less restricted options in that regard: ‘the less restrictive options envisaged were dismissed in general terms on the ground that they would be potentially “difficult” to apply without arbitrariness’ and ‘the Government were not able to refer to any expert report which examined whether there existed other practical solutions enabling both the scope of the prohibition to be reduced and its objectives to be conserved’.¹⁰⁸ Both the UK courts and the Strasbourg Court accepted Parliament’s defeat without urging substantiation of the assumption that differentiation in the law was ‘difficult’. Justice Ousley of the High Court acknowledged that ‘the ECtHR decision in VGT could be seen as permitting a restriction based on the size, wealth or responsibility of the group, or its inability otherwise to reach all those whom it wished to reach’, but then went on to say that he did not see how a law could be drafted without being susceptible to abuse nor how ‘rational, practicable distinctions’ could be drawn between various types of advertiser and types of advertisement.¹⁰⁹ Hence, the judge agreed that there was no alternative with an evidentiary basis. The House of Lords referred to the Government’s departmental note, which stated that ‘consideration was given to an alternative regime’,¹¹⁰ although that alternative involved restricting the ban to election (or referendum) time and rules to avoid any particular point of view from dominating and to control the scale of political advertising in terms of broadcasting time and advertising revenue; a distinction based on the type of party or group was never considered. By contrast, ADI’s advocate referred to legislative models in other states and a large body of academic commentary supporting the argument for differentiation.¹¹¹ Here as well, the court replied, on the basis of its own assumptions, that ‘it is difficult to see how any system of rationing or capping could be devised which could not be circumvented’, that deciding on a case-by-case basis would accord ‘excessive discretion to officials, and give rise to many challenges’ – and therefore saw ‘no reason to challenge’ the Government’s assertion that ‘no fair and workable compromise solution could be found’.¹¹² Finally, the Strasbourg Court found that the Government’s ‘fear that the proposed alternative option was not feasible’ was ‘reasonable’.¹¹³ Here too, the Government was not asked to give evidence that a solution had been sought. Nor was any reference made to other countries, although the comparative overview did identify several countries in which a ‘restricted’ ban based on the type of advertiser had indeed been implemented. By contrast, in Ognevenko, the Russian authorities were reproached for not having sought alternatives for a complete ban on the right to strike for certain categories of railway workers.

The ‘exceptional’ examination by parliamentary bodies in ADI, thus, mainly concerned the intent behind the ban, the influence of radio and television broadcasting, and the risk of dominance by wealthy political parties; one crucial aspect, the possibility of less restricted measures, was dealt with rather superficially. Regarding this aspect of the legal problem at hand, the law did not seem to meet the appropriateness standard established by the Grand Chamber in Hatton: the law must be based on ‘appropriate’ investigations and studies.¹¹⁴ Thus, even from the angle of a substitute model, the standards used to assess the quality of parliamentary debate would seem too vague to avoid the more radical version from creeping in.

A similar observation can be made in Parillo v Italy. The prohibition against releasing cryopreserved embryos, at the mother’s request, so that they might be used for stem-cell research, was based on a statute that, according to the Court, had been well-prepared. The parliamentary debate ‘had taken account of the different scientific and ethical opinions and questions on the subject’ and ‘doctors, specialists and associations working in the field of assisted reproduction had contributed to the discussions’, so that ‘the legislature had already taken account of the different interests at stake’.¹¹⁵ Yet, the dissenting judge pointed out that the debate had not touched upon the matter most relevant to the case before the Court, namely the fate of embryos already in cryopreservation at the time the new law came into effect.¹¹⁶

Hence, if procedural rationality review is to have a compensatory function, e.g. by invoking procedural guarantees when the court cannot make a substantial assessment, then the court should not merely look at how detailed the examination of the matter was in globo, but only at those aspects that are relevant in light of the proportionality analysis.

In reality, the Court rarely goes this far. Even in Hatton, the case in which the appropriateness test was established, dissenting judges reproached the majority for stating that the law was based on appropriate research, even though the studies had not covered the relevant issues and were instead ‘limited to sleep disturbances […] not taking into account the problems of those who had been unable to get to sleep in the first place’. No evidence was given that the government had ‘explored all the alternatives, such as using more distant airports’ – with the dissenters noting that the government’s claims had been based on reports prepared by the aviation industry, thus implying that they were biased.¹¹⁷

At first sight, the easier case is the one in which no parliamentary discussion has been held, as in Hirst, or no evidence-based debate, as in Lecarpentier ¹¹⁸ or, seemingly, in Ognevenko. However, the absence of any such debate might be explained by the existence of a widespread, cross-party consensus; as in Hirst, the Strasbourg Court then risks trespassing on sensitive territory – precisely where it puts its legitimacy at risk and should, therefore, be cautious about reprimanding national authorities exclusively on the basis of its own detailed examination. On the other hand, the intensity of parliamentary procedures could potentially be triggered by the fact that a proposed measure is problematic in the light of the ECHR, as was the case in ADI, and should, therefore, make the Court more cautious than restrained with regard to the proportionality of the measure.

Conclusion

Proceeding from the Animal Defenders International case and, more specifically, the critique it encountered, several rules of thumb are formulated in this paper to guide the Court’s use of procedural rationality review. The first rule of thumb is that if procedural rationality review is to serve to protect fundamental rights, the Court should resort to it only when it is unable to substantively assess the merits of a case. If the margin of appreciation is broad, procedural rationality review and the quality of parliamentary debate may be allowed to assume a prominent place in the analysis. If the margin of appreciation is narrow, procedural rationality review can be used to underpin the substantial assessment. The second rule of thumb is that substantive arguments should prevail if there are serious grounds to argue either conformity or violation of the challenged Act. Procedural rationality review can play a more substantial role in case of doubt, i.e. in so-called hard cases. Third, the evidence used by Parliament should only be questioned if there are serious reasons to doubt its quality. And fourth, if the Court praises the quality of Parliamentary debate to justify a questionable measure, it should make particularly sure that there was, in addition to extensive debate on the subject matter in general, informed discussion of the relevant legal questions in particular.

In Animal Defenders International, the Court respected the third rule of thumb but ignored the other three. As a result, it ended up being suspected of applying double standards to favour the United Kingdom. Similar suspicions were echoed in the dissenting judge’s opinion in Ognevenko, in which the Court was accused of careless balancing and a lack of respect for the reasons of general interest underpinning the legislative choices made and the domestic judicial scrutiny applied. To avoid such recriminations, courts, when they resort to procedural rationality review, should develop a more consistent and theoretically underpinned framework of review. The four rules of thumb identified in this paper could serve as a guideline for such a framework.