Public Law for Everyone

The judgment of Chamberlain J in KP v Foreign Secretary revisits and clarifies some fundamentals about the nature of substantive judicial review. In doing so, it reminds us that the ‘proportionality versus rationality’ debate is ultimately a distraction from the real business of substantive review: namely, setting the justificatory burden the decision-maker should be placed under and determining the appropriate role of the reviewing court in deciding whether that burden has been discharged.

Students of administrative law who are grappling with rationality, or Wednesbury, review would be well-advised to pay close attention to Chamberlain J’s recent decision in KP v Secretary of State for Foreign, Commonwealth and Development Affairs [2025] EWHC 370 (Admin). In the course of a relatively concise judgment, the structure of rationality review is explicated with admirable clarity, and its nuances — along with those of its close doctrinal cousin, deference — charted. In this post, I set out Chamberlain J’s approach and argue that it is a welcome statement, or restatement, of the law in this area — one that aligns closely with an approach that I advocated some years ago in ‘From Bifurcation to Calibration: Twin-Track Deference and the Culture of Justification’, my contribution to a collection of essays on substantive judicial review published in honour of the late Professor Michael Taggart.

Background

The claimant was a Sri Lankan national of Tamil ethnicity who was imprisoned for assault, and who had previously been convicted of arson and sexual assault, on the island of Diego Garcia in the British Indian Ocean Territory (‘BIOT’). The claimant had what Chamberlain J described as ‘complex mental health problems’ with a history of self-harm and attempted suicide. The authorities accepted that he could not lawfully be returned to Sri Lanka or permanently remain on Diego Garcia; however, the Home Secretary refused to grant him leave to enter the UK, preferring instead to attempt to find a third country that would take him. Against that background, Chamberlain J explained that:

The sole ground of challenge is that the decisions [of the Home Secretary] are irrational because there is a serious risk to the claimant’s health and life while he remains on Diego Garcia and no realistic prospect of any other country accepting him in the foreseeable future.

Detailed reasons were offered by the Government in support of its decision to refuse the claimant permission to enter the UK; they centred on the risk that it was considered the claimant, given his history of offending, would pose to those in the UK and the importance of maintaining public confidence in the immigration and asylum system. However, Chamberlain J observed that the claimant was in a position that distinguished him from the ‘countless individuals’ who would no doubt benefit from being allowed to enter the UK, including to receive treatment for mental or physical health problems. Not only, noted the judge, was the claimant in ‘an overseas territory for whose external relations the UK Government is responsible as the paramount government’; in addition, BIOT is ‘a territory with no permanent population and the defendants accept it is ‘impracticable for any migrant (let alone one with the claimant’s mental health difficulties) to remain there in the long term’, while ‘the BIOT authorities have decided that it would violate [customary international law] to return him to the country of which he is a national’. It followed, said Chamberlain J, that:

the defendants accept that there are only two options open to them: find a third country willing to take the claimant, or admit him to the UK. Both involve positive action by the UK Government.

The question was whether the Home Secretary’s decision to prefer the first over the second option was irrational (or Wednesbury unreasonable). In the course of addressing that question, Chamberlain J advanced three key propositions relevant to assessing rationality. Each of those propositions is examined in turn in the following three sections of this post.

Normative importance of interests threatened by decision

It has long been recognised that the Wednesbury doctrine is not monolithic, in that the courts’ approach — and, with it, the intensity of review — may vary depending on the circumstances of the case. By the 1990s, the language of ‘sub-Wednesbury‘ review (implying a lower threshold for intervention, and thus more demanding judicial scrutiny) and ‘super-Wednesbury‘ review (connoting less demanding scrutiny) had entered the administrative law lexicon. However, such terminology was often unaccompanied by clarity about exactly how the courts’ approach varied according to the circumstances of the case. In my essay ‘From Bifurcation to Calibration’, I argued that if doctrinal and conceptual clarity were to be achieved in this area, it was essential to distinguish between the burden of justification to which the defendant is put and judicial determination as to whether that burden has been discharged. I also argued that the nature, or weight, of the justificatory burden placed on the defendant was one — but only one — of the factors in play when ‘variable’ standards of judicial review were being debated:

…when we say that an administrative measure must be justified by reference to a given benchmark, this will normally import a requirement other (and lesser) than that the reviewing court agrees with the measure, in the sense of being satisfied that it would have done the same thing had it been standing in the administrator’s shoes. But ruling out this approach—which, if adopted, would cause the distinction between appeal and review to collapse—leaves open a range of possibilities. The court might, for instance, insist that the measure be shown to be strictly necessary and proportionate, or reasonable, or not flagrantly unreasonable, or that it satisfies some other different or interstitial standard. Viewed through the prism of justification, choosing between competing potential standards of review amounts to determining the extent of the burden of justification under which the defendant is to be placed, and by reference to which a decision that is ‘suspect’ falls to be evaluated. This, in turn, raises questions about how courts should go about determining whether—and, if so, to what extent—a decision is ‘suspect’ and how, in the light of that, the extent of the justificatory burden should be determined.

I went on to argue against a blunt approach premised on a binary distinction between requiring the defendant to show either that the decision is not manifestly unreasonable or that it is proportionate, and that setting the justificatory burden must be a more subtle enterprise that accords with the normative importance of whatever it is that is placed at risk by the decision. In this way, I contended that it was inappropriate (for example) to treat human rights cases as the preserve of proportionality while condemning all non-rights cases to the weaker protected afforded by the traditional conception of Wednesbury. The now-orthodox nature of that approach, whereby a binary distinction between rights and non-rights cases is rejected in favour of more subtle calibration of the importance of what is placed at stake by the impugned decision, was acknowledged by Chamberlain J in KP:

the court’s approach to assessing the rationality of a decision varies depending on the importance of the interests affected by it or, to put the point another way, the gravity of its potential consequences. In this connection, it is not necessary to identify a “right” impacted by the challenged decision. It is true that, in [R v Secretary of State for the Home Department, ex parte Bugdaycay [1987] AC 514] … Lord Bridge made reference … to the “right to life”, but what made a “more rigorous examination” appropriate was not the existence of such a right but rather “the gravity of the consequences” flowing from the challenged decision – i.e. the fact that the decision was said to have put the claimant’s life at risk. … The interests in question may be such as to ground a right properly so-called … but not necessarily. In many of the situations in which the heightened standard of review applies, the claimant will have no prior right, whether under statute or at common law, to the benefit which the decision denies him.

Of course, once this point is recognised, other important doctrinal questions arise about the relationship between rationality and proportionality review. However, Chamberlain J, perhaps understandably, did not attempt to grasp that particular nettle, confining himself to the observation that in Kennedy v Charity Commission [2014] UKSC 20, [2015] AC 455, Lord Mance had suggested that (as Chamberlain J put it) ‘in some cases there may be no difference between rationality review and “European” (i.e. proportionality-based) review’, and to quoting from Lord Sumption’s judgment in Pham v Secretary of State for the Home Department [2015] UKSC 19, [2015] 1 WLR 1591, in which Lord Sumption doubted the extent, in certain circumstances, to which rationality and proportionality review differ. In this way, Chamberlain J, without seeking to resolve the debate about the relationship between rationality and proportionality review, recognised that the nature of that relationship is a more subtle one than the debate in this area has sometimes acknowledged.

The burden of justification and intrinsic deference

Chamberlain J’s first proposition, then, is that the variable intensity of substantive judicial review falls to be calibrated by reference to the normative importance of the interests threatened by the decision in question, an approach that eschews the sort of taxonomy that would draw bald distinctions between (for instance) cases in which fundamental rights are and are not implicated, and in which radically different approaches to review are available. His second proposition follows directly from the first:

where it applies, the heightened standard of review has implications for the way the court evaluates complaints of both process and outcome irrationality. In the former case, the court will subject the decision to “more rigorous examination, to ensure that it is in no way flawed” (Bugdaycay, 531) … Where the complaint is of outcome irrationality, more will be required by way of justification ([R v Ministry of Defence, ex parte Smith], 554); and the importance of the interests affected may, in principle, narrow the range of decisions open to the decision-maker, potentially to just one (Pham, [107]).

This aspect of Chamberlain J’s judgment underlines the importance of, and goes hand in hand with, his first proposition, in that the justificatory burden the defendant decision-maker finds itself under is, properly understood, a highly variable concept, the extent of which falls to be calibrated by reference to the normative importance of the interests that are in play. This, in turn, recognises that the appropriate extent of the court’s role varies according to the circumstances of the case, the degree of appropriate judicial oversight depending, at least in part, on the normative significance of what is at stake. In my chapter ‘From Bifurcation to Calibration’, I argued that such variability in relation to setting the extent of the justificatory burden amounts to, even though it is not generally described as, a form of deference, the appropriate degree of which will vary from case to case. I dubbed it intrinsic deference, arguing that:

This is an established phenomenon, even if the language of deference is less familiar in this context. Indeed, the notion is hard-­‐wired into the traditional, doctrine-­‐led approach to substantive review. Asking whether a decision is Wednesbury reasonable in the normal sense is less demanding—or more deferential— than asking whether there are cogent reasons for it; asking whether a decision is reasonable is less demanding than asking whether it is proportionate; asking whether a decision is reasonably necessary is less demanding than asking whether it is strictly necessary; and asking whether a decision is manifestly lacking a reasonable justification is less demanding than asking whether it is accompanied by very weighty reasons. Some instruments within the substantive-­review toolbox are, in other words, intrinsically deferential because, in the first place, they involve imposing upon the decision­‐maker a slighter burden of justification. Viewed in this way, setting the burden of justification and setting the level of intrinsic deference are merely two sides of one coin.

Thus Wednesbury review, at least in its traditional guise, is inherently deferential in the sense that the lightness of the justificatory burden forecloses any possibility of intense judicial review, given that the decision-maker need do no more than demonstrate that its decision is justified in the sense of not being manifestly unreasonable. In contrast, proportionality review is inherently less deferential because the justificatory burden imposed — which requires, among other things, the decision-maker to demonstrate the necessity of its decision and the fairness of the balance struck between relevant competing interests — is intrinsically greater. Importantly, however, once we recognise that the crucial animating factor is the normative importance of the interest at stake, the binary distinction that has traditionally been drawn between rationality and proportionality review begins to dissolve and the possibility of a relatively heavy burden of justification arises even if the case is not of a type that would traditionally have attracted the application of the proportionality test. Thus Chamberlain J concluded in KP that the burden of justification to be placed on the decision-maker was a heavy one given the nature of what was at stake, notwithstanding that his analysis was framed in terms of rationality rather than proportionality:

The challenged decision affects the claimant’s vital interests (health and life) and has potentially grave consequences for him (serious injury or death). In principle, this claim therefore falls within the category to which the heightened standard of rationality review applies. This is so notwithstanding that the claimant has no prior legal right to enter the UK and the defendants have no legal duty to admit him, unless rationality compels it.

Adjudicative deference

The third proposition advanced by Chamberlain J relates to deference as it is more traditionally understood:

[T]he importance of the claimant’s interests is not the only factor relevant to the court’s approach to a complaint of outcome irrationality. The nature and importance of the public interests on the other side of the balance may also be important. In some fields, institutional considerations may require the court to recognise that it is less well-placed than a democratically accountable decision-maker to evaluate the impact that a particular decision may have on a particular public interest, or the weight to be accorded to that impact. Equally, constitutional considerations may dictate that the court should pay particular respect to the views of a democratically accountable decision-maker about how to balance the public and private interests. In such cases, even where a decision will have grave consequences for the individual, the court may have to afford the decision-maker a wide margin when considering whether the outcome is irrational …

This aspect of Chamberlain J’s analysis is helpful for two reasons. First, it recognises that deference — in the sense of the degree of respect or weight to be accorded to the decision-maker’s own assessment when it comes to determining whether the burden of justification has been discharged — is potentially relevant not only to proportionality review, in which context deference is a common preoccupation, but also to rationality review. Given that rationality review is traditionally considered to be intrinsically deferential, it may not immediately be obvious why deference per se should be necessary or appropriate. However, once we recognise that rationality review can, as in the KP case, involve the imposition of a relatively weighty burden of justification in acknowledgment of the importance of the interests at stake and the gravity of the potential consequences of the decision, the role that deference might appropriately play becomes clearer.

Second, it follows from this that Chamberlain J’s analysis helpfully recognises the distinction between intrinsic deference and what I have previously called adjudicative deference, describing it in the following way:

In this guise, deference may call for the ascription of respect to the decision­‐maker’s view when the reviewing court assesses whether the burden of justification should be taken to have been discharged. The court might, for instance, be willing to attach greater weight to the view of the decision-‐maker when deciding whether imposing a particular restriction upon a right was necessary, on account of the fact that the decision­‐maker was peculiarly institutionally well-‐placed to form such a view and deployed its expertise in doing so. Or a reviewing court might ascribe a degree of respect to the view of the decision-­maker when deciding whether the restriction placed by the impugned measure upon the relevant right (or other value) is justified by a public policy gain, such that the measure can be considered to have struck a ‘fair balance’, or to be proportionate in the ‘strict’ or ‘narrow’ sense. I have argued elsewhere that adjudicative deference is appropriate in such circumstances, and it is not the purpose of this chapter to rehearse that argument. The thrust of the present argument is not that adjudicative deference is appropriate—which, for the purpose of this chapter, I take to be a given—but that it is distinct from intrinsic deference, and that there is merit in acknowledging and making explicit that point of distinction.

The distinction between intrinsic and adjudicative deference helps to explain an apparent paradox in Chamberlain J’s judgment, in the sense that it combines a heavy justificatory burden with a high level of deference. Once the distinction between the two forms of deference is acknowledged, however, it is possible to understand why a given factual matrix, such as that which arose in KP, might combine a heavy justificatory burden (ie little intrinsic deference) with substantial adjudicative deference. The possibility — and importance — of disaggregating those matters was clearly recognised by Chamberlain J:

[T]he claimant invites the court to conclude, on the basis of a detailed examination of the disclosed materials, that – contrary to the judgement of the defendants – there is no realistic prospect that a third country will take the claimant within any reasonable timescale. The conclusion that the case attracts a heightened standard of review does not displace the well-established institutional reasons for the court to afford the defendant a wide margin when making a judgment of this kind. The judgment involves a prediction about what foreign states might do in the context of a diplomatic negotiation. When evaluating such judgments, the courts will accord great respect to the view of the [Foreign, Commonwealth and Development Office], because of its institutional experience and expertise …

Similarly, Chamberlain J considered, in arriving at his conclusion that the Government’s position was not irrational, that adjudicative deference was warranted on constitutional (or democratic) was well as institutional grounds:

[T]he claimant complains about the balance struck between his interests and those of the general public when excluding him because of his history of criminality and because of the effect admitting him would have on public confidence in the immigration system. Courts have expertise and experience in evaluating evidence relevant to risk and in conducting balancing exercises at the micro-level. But, especially where Parliament has allocated the decision to the Secretary of State, there are sound constitutional reasons why courts should afford respect to Ministers’ views about the weight to be attached to the public safety risks flowing from a positive decision. The same is true, a fortiori, when evaluating the risk that a decision in an individual case would undermine public confidence in the immigration system as a whole.

Conclusion

It is not the purpose of this post to suggest that Chamberlain J’s judgment in KP breaks new ground. However, the judgment is very valuable thanks to the way in which it rigorously analyses and synthesises the case law on rationality review, demonstrating the importance of conceptually distinguishing between different forms of deference, recognising the possibility of combining a heavy burden of justification with substantial adjudicative deference, and acknowledging that the role of the court in substantive review cases must be understood by reference to due analysis of what is at stake in the case rather than by resorting to crude distinctions between (for example) cases that do and do not involve rights.

For longer than was necessary, English administrative law (as well as administrative law elsewhere) was unduly preoccupied by taxonomical debate that focussed, in particular, on the distinction, or relationship, between rationality and proportionality review. KP is a salutary reminder about what constitutes the true business of substantive review: namely, determining the weight of the justificatory burden that is warranted by the case and the extent to which courts, in their adjudicatory role, ought to defer to the decision-maker’s own assessment when deciding whether that burden has been discharged. Once those fundamentals are acknowledged, hoary chestnuts like the ‘rationality versus proportionality’ debate can be seen for what they are: at worst, a distraction from what is really important and, at best, shorthand for the real issues that are in play, and with which KP grapples with admirable clarity.