Proportionality vs Rationality Review: A False Dichotomy?

Rachelle Lam[1]

 

Introduction

In English Administrative Law, much ink has been spilt over the relative intensities of review of a proportionality standard, as opposed to a test of unreasonableness, and whether the former should replace the latter as a general head of substantive judicial review. This article argues that such an assumption is premised upon an inaccurate portrayal of the relationship between the two tests, and proposes an alternative theoretical framework where the focus shifts away from formalist doctrinal veneers to the subject matter and context of each case.

The Wednesbury standard of review involves posing the question of whether the administrative act is “so unreasonable that no reasonable authority would ever have come to it”.[2] In contrast, the test of proportionality involves four distinct stages: (i) whether the objective of the measure is sufficiently important to justify the limitation of a fundamental right; (ii) whether it is rationally connected to that objective; (iii) whether a less intrusive measure could have been used; and (iv) whether a fair balance has been struck between the rights of the individual and the interests of the community.[3]

 

A more “stringent” form of review

On its face, proportionality appears to offer more structure than the Wednesbury test. This is because proportionality requires the judge to examine the normative content of the private rights being vindicated, and the justification for the relative weight accorded to competing public interests.[4] This has the potential to significantly attenuate the administrator’s capacity to make policy choices by circumscribing the range of substantive options available to the decision-maker.

In contrast, the Wednesbury test involves posing a relatively undemanding question on the decision-maker. In instances which concern the formulation and implementation of national political and economic policy, the courts will show considerable deference to the judgment of elected ministers and recognise that there is room for more than one view.[5] As a result, Wednesbury review can often be vague and tautologous, as it fails to expose the structure and underlying values of the judicial reasoning process which necessarily precedes a finding of unreasonableness.

The disparity in the proportionality and Wednesbury processes is illustrated by the decision in Regina v. Ministry of Defence, ex parte Smith,[6] and subsequently, Smith and Grady v. United Kingdom.[7] The applicants were discharged from their service in the armed forces on account of their homosexuality. The domestic court applied the Wednesbury test: the existence of an apparent justification for the qualification on the applicants’ right to respect for private life,[8] namely that of ensuring discipline, morale and unit cohesiveness in the army,[9] was within the range of responses open to a reasonable decision-maker, and hence, in the court’s view, established the legality of the government’s policy. [10] In contrast, the European Court of Human Rights, in applying the proportionality test, questioned whether the UK government’s policy was “necessary in a democratic society”, and concluded that although it may answer a pressing social need, the magnitude of the reservation placed upon human rights was disproportionate.[11]

 

The chameleonic nature of the two tests

The Wednesbury and proportionality tests are not monolithic,[12] and both operate as sliding scales of review.[13] Indeed, Wednesbury unreasonableness may even prove to be a more “stringent” standard of review than proportionality. In Keyu v. Secretary of State for Foreign and Commonwealth Affairs,[14] Lady Hale articulated the respective competing interests that should have been taken into account by a reasonable decision-maker, and held that, in the calculus of pros and cons, the decision-maker had failed to ascribe sufficient weight to the individual’s rights.[15]

In contrast, where proportionality is coupled with a significant degree of judicial deference, there is little analytical weighing of interests, such that the test being applied in effect amounts to a test based on arbitrariness of conduct.[16] Furthermore, what proportionality requires can be malleable. For example, the term “fair balance” can sometimes be taken to be a criterion of its own, whilst at other times it is treated as being synonymous to proportionality as a whole. Understood thus, the two tests are not necessarily as distinct as their paradigm characterizations may imply.

 

Proportionality as a “general principle”?


The wisdom in preserving Wednesbury reasonableness in English administrative law has been questioned both judicially and academically. Dyson LJ professed that he had “difficulty in seeing what justification there now is for retaining the Wednesbury test”. [17] It is argued that although the time has not come for Wednesbury to be expunged yet, the courts should throw off the rigid constitutional division between proportionality and Wednesbury, and adopt a third route distinct from both bifurcation and consolidation.

There are strong grounds for supporting the endorsement of proportionality as a general principle. A key attractiveness of proportionality is in its requirement of a reasoned explanation from the decision maker. Once a claimant has established that there has been an interference in his right, it is for the court to weigh up competing considerations and articulate why it came to its conclusion. For example, Miller (No. 2)[18] appeared to employ proportionality review in scrutinizing the exercise of prerogative powers:[19] although the court was satisfied by the government’s pursuit of the objective of ushering in a new legislative agenda,[20] it held that the Prime Minister failed to discharge the burden[21] of showing that the prorogation was necessary, due to a failure to consider the alternative of a Parliamentary recess,[22] and a fair balance was not struck due to its unjustified length.[23] It is argued (and elaborated below) that an explicit switch to utilising proportionality review across the board should be accompanied  by variable application, which would involve adjusting the intensity of proportionality according to the context and the doctrine of deference.

An objection to the adoption of proportionality as a general principle is that proportionality would have to mean different things in different contexts. If review is variable according to the normative weight to be attached to the relevant interest, proportionality review could simply amount to identifying whether there has been a “manifest error”,[24] and will no longer exhibit the intensity of review nor the structure which defines it. This argument against adopting proportionality as a general head of review suggests that the wholesale elimination of Wednesbury is altogether unsatisfactory.

On the other hand, those who advocate for the retention of separate heads of review argue that the proportionality methodology is only useful where there is a yardstick to which a public body’s actions can be compared with. Without the “anchor” of rights, proportionality will become an indeterminate standard which conceals unconstrained judicial discretion when controlling exercises of administrative discretion.[25] Taggart has proposed a “rainbow of review”: proportionality should replace Wednesbury where human and other “fundamental” rights are directly engaged, and in cases involving “public wrongs” Wednesbury unreasonableness should be applied.[26]

However, this doctrinally-bifurcated vision of substantial judicial review is flawed, because a clean division between “rights” and “public wrongs” is impossible. It is rarely the case that only one of the two is categorically engaged, as the range of circumstances in which the proportionality doctrine applies is broader than has generally been acknowledged. For example, in Secretary of State for the Home Department v. Pham,[27] the court held that a requirement to act proportionately was normatively warranted when a statutory power threatened a sufficiently important interest (British citizenship). While the ground for complaint may appear to be one of maladministration, an applicant is motivated to litigate precisely because the decision impinges upon an underlying right or interest of fundamental importance.[28] Other well-established substantive principles of good administration, such as consistency of treatment, non-retrospectivity and access to court, are just as capable of buttressing the proportionality methodology as are fundamental rights.

 

The Future

Ultimately, to debate whether proportionality should replace Wednesbury is to begin from a false proposition, as this postulates a bright-line distinction between the two principles which does not, in reality, exist. The two principles need not inevitably be conceptualised as competing forms of review which must be chosen between.

A new framework for substantive judicial review is proposed as follows: firstly, the applicant must identify a “badge of unreasonableness” which identifies a flaw in the decision-maker’s decision and is capable of justifying judicial intervention.[29] Then, the court should decide how heavy a burden of justification should be placed on the administrative body. Judicial deference and the rigour and scope of review should be flexibly modulated by reference to normative, institutional and constitutional considerations,[30] such as the inherent significance of the right being vindicated, the character of the policy, and the expertise of the decision-maker. Finally, the court should examine whether that burden has been discharged;[31] in other words, whether the explanation provided by the decision-maker is justified.

This approach has three key advantages. Firstly, it simplifies English law. Rather than leaving the vindication of individual interests to hang upon arbitrary classifications, the framework provides analytical clarity and transparency. Secondly, it acknowledges the fact that courts are not necessarily well placed to assess policy decisions. The flexible and context-sensitive process means that the court will be better able to accord due respect to the institutional competence of decision-makers when need be. Thirdly, the framework promotes more effective judicial supervision of administrative decisions, facilitating legitimate government action whilst upholding the rule of law. It exemplifies constitutional collaboration and counter-balancing at its best.

 

 

[1] Rachelle Lam is an LLM Candidate at University of Cambridge, having recently completed her undergraduate studies in Law with First Class Honours, also at Cambridge. Her areas of interest include Administrative Law, Company Law, and International Commercial Arbitration.

[2] Associated Provincial Picture Houses Ltd v Wednesbury Corporation [1948] 1 KB 223 (CA), 234 (Lord Greene MR). Lord Diplock’s formulation of the principle was that a decision may be irrational, and hence unlawful, if it ‘is so outrageous in its defiance of logic or of accepted moral standards that no sensible person who had applied his mind to the question could have arrived at it’: Council of Civil Service Unions v Minister for the Civil Service [1985] AC 374 (HL), 410.

[3] Bank Mellat v Her Majesty’s Treasury [2013] UKSC 38 [20] (Lord Sumption).

[4] R (on the application of Daly) v Secretary of State for the Home Department [2001] 2 AC 532 (HL) [27], [30] (Lord Steyn).

[5] R v London Borough of Hammersmith and Fulham and Others, ex p Burkett and Another [1991] 1 AC 521 (HL), 597 (Lord Bridge); see also: Jeffrey Jowell, Anthony Lester QC, ‘Beyond Wednesbury:  Substantive Principles of Administrative Law’ [1987] PL 368.

[6] [1996] QB 51.

[7] (1999) 29 EHRR 493.

[8] Article 8 of the European Convention on Human Rights.

[9] Smith (n 6) 529E-H (Brown LJ).

[10] Smith (n 6) 558A-C (Lord Bingham MR).

[11] For instance, a code of conduct governing relationships between military personnel (regardless of their sexual orientation) would have been sufficient to secure the government’s objective; see also: Mark Elliott, ‘The Human Rights Act 1998 and the Standard of Substantive Review’ [2001] 60 CLJ 301.

[12] Gráinne de Búrca, ‘The principle of proportionality and its application in EC Law’ (1993) 13 Yearbook of European Law 105.

[13] Sir John Laws, ‘The Limitations of Human Rights’ [1998] PL 254, 259-260.

[14] [2015] UKSC 69.

[15] Ibid [309]-[312].

[16] Takis Tridimas, ‘Proportionality in European Community Law: Searching for the Appropriate Standard of Scrutiny’ in Evelyn Ellis (ed.), The Principle of Proportionality in the Laws of Europe (Oxford 1999), 70-72.

[17] R (British Civilian Internees (Far East Region)) v Secretary of State for Defence [2003] EWCA Civ 473 [34].

[18] R (Miller) v The Prime Minister and Cherry v Advocate General for Scotland UKSC 41.

[19] Ibid [56], [59] (Lord Reed).

[20] Ibid [17], [51].

[21] Ibid [51].

[22] Ibid [60].

[23] Ibid [60]-[61].

[24] R v Secretary of State for Health ex parte Eastside Cheese Co [1999] 3 CMLR 123 [47] (Lord Bingham).

[25] Michael Taggart, ‘Proportionality, Deference, Wednesbury’ [2008] NZ L Rev 423.

[26] It is also worth noting that similarly distinct modes of review manage to co-exist in US public law, embraces both “rational basis” and strict scrutiny” review: Ian Loveland, ‘A Fundamental Right to be Gay under the Fourteenth Amendment?’ [1996] PL 601.

[27] [2015] UKSC 19.

[28] Murray Hunt, ‘Against Bifurcation’ in David Dyzenhaus, Murray Hunt and Grant Huscroft (eds), A Simple Common Lawyer: Essays in Honour of Michael Taggart (Oxford 2009).

[29] See: Delios v Canada (Attorney General) 2015 FCA 117 [27]; Paul Daly, ‘Substantive Review in the Common Law World: AAA v Minister for Justice [2017] IESC 80 in Comparative Perspective’ [2019] 1 ISCR 105.

[30] Mark Elliott, ‘Proportionality and Deference: The Importance of a Structured Approach’ in Christopher Forsyth, Mark Elliott, Swati Jhaveri, Michael Ramsden and Anne Scully Hill (eds), Effective Judicial Review: A Cornerstone of Good Governance (Oxford 2010).

[31] Rebecca Williams, ‘Structuring Substantive Review’ [2017] PL 99.