Revista de Derecho Público: Teoría y Método
Marcial Pons Ediciones Jurídicas y Sociales
Vol. 7 | 2023 pp. 83-109
Madrid, 2023
DOI:10.37417/RPD/vol_7_2023_1259
© Jaime Phillips Letelier
Recibido: 09/03/2023 | Aceptado: 11/04/2023
Editado bajo licencia Creative Commons Attribution 4.0 International License.

THE NORMATIVE FOUNDATIONS OF RETROACTIVITY AND LEGITIMATE EXPECTATION IN THE UK ADMINISTRATIVE LAW

Jaime Phillips Letelier*
Assistant Professor, Administrative Law
Universidad Finis Terrae (Chile)

“The real tie lies in the feelings and expectations we have raised in other minds. Else all pledges might be broken when there was no outward penalty. There would be no such thing as faithfulness”

George Eliot
The Mill on the Floss

Abstract: The doctrines of non-retroactivity and legitimate expectation find their normative foundation in the idea that legal norms and administrative decisions should be stable and predictable. Thus both doctrines are intertwined around the idea that the legal system has to allow individual planning and reliance on law. This intertwinement situation is problematic because the protection of individual planning does not fully explain the prohibiting sense often infused in the non-retroactivity principle. Also, extending this principle to genuine legitimate expectation cases may result in unduly and excessive protection to individuals from law or policy changes. Thus, this paper proposes a new normative foundation for the non-retroactivity principle around finality. This proposal will explain the prohibiting sense of the non-retroactivity principle and allow the development of an understanding that retroactivity happens in degrees. Finally, this proposal will enable making sense of the legitimate expectation case law and the justification of compensation for damages caused by detrimental reliance.

Keywords: Retroactivity, Legitimate expectation, Finality, Individual planning

Resumen: Las doctrinas de la no-retroactividad y la confianza legítima encuentran su fundamento normativo en la idea de que las normas legales deben ser estables y predecibles, permitiendo así que los individuos planifiquen su conducta conforme con ellas. Esto es problemático porque la protección de la planificación de conducta no explica en forma completa el sentido prohibitivo con que es formulado el principio de no-retroactividad y los diferentes grados con que la retroactividad tiene lugar. Además, la extensión de este principio a genuinos casos de confianza legítima puede suponer una protección indebida y excesiva de los individuos frente a cambios normativos. Por ello, este trabajo propone una nueva fundamentación normativa del principio de irretroactividad en torno a la idea de finalidad para explicar su sentido prohibitivo y desarrollar la comprensión de que la retroactividad se produce en grados. Por último, esta propuesta permitirá dar un mejor sentido a la jurisprudencia de la confianza legítima y a la justificación de la indemnización de los daños causados por la frustración de la confianza.

Palabras CLAVE: retroactividad, confianza legítima, finalidad, planificación individual

Summary: 1. INTRODUCTION.— 2. STATUS QUO: INTERWOVEN DOCTRINES: 2.1. Reliance in Retroactivity and Legitimate Expectation. 2.2. The overlap.—3. SHORTCOMINGS IN THE STATUS QUO: 3.1. Retroactivity and Individual Planning. 3.2. Apparent Retroactivity is also Retroactivity. 3.3. Unfair Protection of Legitimate Expectations.—4. PROPOSAL: 4.1. Finality and Retroactivity. 4.2. Total and Partial Retroactivity. 4.3. The Temporality Test.—5. CONCLUSIONS.

1. Introduction

Legal certainty embodies a tension between two aspects of the Rule of Law. On the one hand, it requires the legal system to be stable, and that its rules should have only prospective effect, thus it can guide its subject’s behaviour 1. On the other, the Rule of Law demands the legal system to be “pliable and responsive to changing conditions” 2, and thus might require amendments when rules become pointless or when new needs for the public good arise 3. While stability is a critical aspect of the Rule of Law, authorities may have to make decisions which are inconsistent with determinations made in the past or that frustrate plans made by the individuals according to previous norms or decisions, affecting legal positions generated in the past 4.

Circumscribed to this tension between stability and change, we find the doctrines of non-retroactivity and legitimate expectation. Many argue that the fundamental problem of retroactive decision-making is that it doesn’t allow people to rely on current rules, thus undermining a basic tenet of any sound legal system 5 Thus, Raz regards non-retroactivity as a value essential to the Rule of Law 6.

Meanwhile, the doctrine of legitimate expectation states that a “pattern of conduct, or a representation, or a promise” can raise expectations that the authority cannot disappoint without acting unfairly 7, even in the case of decisions not having a retroactive effect, because a citizen may have relied on the expectation induced by the public body 8.

The first section will show that the doctrines mentioned above have a common normative foundation: that the law should allow citizens to rely on rules and decisions made by public bodies. Following Fuller’s approach, decisions having a retroactive effect or frustrating expectations derive their harmfulness from what he calls “legislative inconstancy” 9. In this paper, I will label the problems addressed here as concerned with legislative or administrative temporal inconstancy. The case law on this problem will show that ascertaining differences in applying the non-retroactivity and legitimate expectation doctrines is challenging. As scholarship and courts intertwine them around the same idea, they overlap around the same issues.

In the second section, I will show that this overlapping situation needs revision. First, justifying the non-retroactivity principle on the idea that people should be able to rely on current law can give way to unfair forms of retroactivity because that foundation does not fully explain the prohibiting sense that pervades the principle. The second problem is that decisions having a softer form of retroactivity called “apparent” often generate problems addressed in terms of legitimate expectation. Third, handling legitimate expectation cases in terms of retroactivity may seriously undermine the capability of public bodies to achieve important policy goals.

This analysis will allow two things. First, to make sense of the case law on the problem of temporal inconstancy. Scholars see legitimate expectation as an underdeveloped, disintegrated and confusing doctrine 10 because of its weak normative foundation and its development around cases more concerned with retroactivity, where I think the presence of legitimate expectations is immaterial. Second, to propose some changes in the law regarding how to protect legitimate expectations.

Hence, in the third section, I will propose three things. First, rather than justifying the non-retroactivity principle on the need for reliance and individual planning, I will use the idea of finality. Second, I will propose a distinction between total and partial retroactivity, which will help address non-retroactivity cases and distinguish them from legitimate expectation cases. Third, I will present a method to handle legitimate expectation cases, which I will call the “temporality test”, which will specify the fairness requirements in situations where public bodies are empowered to depart from previous decisions or representation and thus to frustrate.

2. Status quo: interwoven doctrines

2.1. Reliance in Retroactivity and Legitimate Expectation

A “shifting regulatory landscape is a serious problem” 11: people should have fair warning of the legal consequences of their actions and not see their reliance interests disappointed without the authority giving them some consideration 12. Hence, the doctrine of non-retroactivity bans legal norms or decisions from taking effect upon events occurred before they came into force 13.

The doctrine of non-retroactivity partly enforces this aspect of the Rule of Law by employing a presumption against giving retroactive effect to statutes 14. However, this is only a rule of statutory interpretation. Primary legislation can have retroactive effects if Parliament provides it 15. Moreover, administrative bodies can apply statutes with retroactive effect as long as legislation authorises them. Hence, when Parliament enacts new legislation, the presumption provides that public bodies have to construe statutes to not impair vested rights or alter accrued obligations under the earlier legislation. The presumption can be drawn from Common Law, and legislation 16.

However, the presumption against retroactivity does not secure the stability of legal norms and individual planning because the regulatory context can constantly change without retroactively applying those changes. This is why some further principle is required to ensure the stability of norms or decisions. In this context, the doctrine of legitimate expectation becomes relevant. Thus, the non-retroactivity principle secures reliance on current law by limiting the possibility of making decisions with effect upon past events. In contrast, legitimate expectation ensures reliance in current law by protecting the expectations of individuals raised by past conduct, decisions or representations. Thus, the presumption against retroactivity and legitimate expectation have a common foundation.

From a doctrinal perspective, we can see this phenomenon in the distinctions made by Craig 17. This author begins the chapter on legitimate expectations of his well-known book by making a triple distinction in which retroactivity can take place, following partly the approach of the EU scholar Schwarze. The latter distinguishes actual from apparent retroactivity. Actual retroactivity is the effect of a norm or decision upon events concluded in the past, while apparent retroactivity is the effect of a norm or decision upon events that originated in the past and are not yet concluded, which are immediately affected by the subsequent norm, decision or policy without being “properly” retroactive (hence, the adjective “apparent”) 18.

Craig embraces this distinction but formulates a slightly different concept of apparent retroactivity. He claims that decisions not having retroactive effect at all can still be harmful to a person “who planned her action on the basis of a policy choice made by the administration”. Thus, Craig labels as “problematic” the effect of a norm or decision upon “plans made in the past”, which he calls “apparent retroactivity”; an approach which leaves the open question of whether he regards the frustration of legitimate expectations as a mitigated form of retroactivity or not.

2.2. The overlap

To understand how the presumption against retroactivity operates, we can analyse Plewa v Chief Adjudication Officer. Mr Plewa started receiving a retirement pension in January 1981. In December of that year, he received a form stating that he had to disclose his wife’s earnings since that affected the amount of his pension. He failed to disclose that information. In April 1987 Parliament enacted new legislation regarding overpayments. It repealed a defence that existed under the Social Security Act 1975 that allowed those who received overpayments to keep the money if they diligently tried to avoid the overpayment. Finally, in October 1987 Mr Plewa’s pension was adjusted.

The Chief Adjudication Officer, the Social Security Appeal Tribunal and the Court of Appeal held that although they were satisfied that Mr Plewa was “quite innocent” 19, the 1987 Act was to be applied retroactively. However, the House of Lord reversed the decision. They held that the removal of the due diligence defence is a situation where the presumption against retroactivity applies 20. Overpayments made before the enactment of the 1987 Act gave the right to the diligence defence. Therefore, in this case, the presumption against retroactivity allowed Mr Plewa to retain the right to the “diligence defence”. Nonetheless, that defence would not be available for overpayments made after that date.

The decision is grounded in Mr Plewa’s reliance in the legislation containing the “diligence defence”. This reliance is essential because people would be incentivised to inquire more about possible overpayments under another statute, such as the one enacted in 1987. Hence, legislation cannot remove it retroactively. In Fuller’s words, judging a person’s conduct by retroactive laws “is to convey to him your indifference to his powers of self-determination” 21.

The presumption against retroactivity will not operate when only apparent retroactivity takes place. For example, in Nigel Rowe, Alec David & Others v The Commissioners for HM Revenue and Customs, the 2014 Finance Act gave the tax authority the power to issue an order to seek an accelerated payment of taxes in dispute 22. The respondents issued payment notices regarding appeals that were not yet decided. The claimants argued, inter alia, that they had an accrued postponement right and that the payment notice was in breach of their legitimate expectations 23. The court decided that a legitimate expectation cannot constrain Parliament’s power to change previous positions and that the 2014 Finance Act “expressly removes rights that previously existed … in respect of all appeals (whenever made)” 24.

Interestingly the 2014 Finance Act did not say expressly that the postponement rights were to be removed in respect of all appeals whenever made. The court construed the statute as meaning that if it did not made distinctions regarding the time the appeal were made, the postponement right would be removed to all pending appeals. Following this criteria, statutes can be generally construed as having an apparent retroactive effect, while the presumption will only operate against giving statutes actual retroactive effect.

The presumption against retroactivity does not disapprove the termination of a right awarded by previous legislation, provided that the decision does not take effect upon events concluded in the past. In this case, the postponement right was only removed for the future 25. Hence, some further principle is required to sustain the decision to maintain for the future legal positions created by repealed legislation. This explains why the claimants argued that the accelerated payment notice was a breach of their legitimate expectations. However, the court considered that a legitimate expectation could not constrain the Parliament for that matter.

In administrative decision-making, the presumption has a stronger effect, amounting to a presumption and a ban on actual and apparent retroactivity. This applies to secondary legislation as well as to individual administrative decisions.

Regarding secondary legislation, an example is Secretary of State for Energy and Climate Change v Friends of the Earth and Others 26. The Secretary of State had a policy of paying a “generation tariff” to electricity supply companies that made payments to small-scale low-carbon electricity producers which had eligible installations for that purpose under the 2008 Energy Act. Once an installation was eligible, the tariff was fixed for 25 years. That policy was changed because the cost of the technology for those installations dropped and there was a risk of paying too much for the generation of low-carbon electricity. The new policy reduced the “generation tariff” from 1 April 2012 on all installations. The proceedings were not concerned with the legality of the policy change nor the legitimate expectations of those who were about to build low-carbon electricity-producing facilities but with the question of whether the Secretary of State had the power to reduce the tariff due to installations becoming eligible before 1 April 2012 27.

Moses LJ said that the presumption against retroactivity is even stronger regarding delegated legislation. The decision to apply the new policy to installations that became eligible before the change was ultra vires. The old scheme created rights that, in the absence of an express provision by the empowering legislation, cannot be removed with (apparent) retroactive effect 28.

The decision of the minister didn’t have an actual retroactive effect. Nonetheless, the court’s decision prevented those installations from being affected by the policy change. It may be true that applying the new scheme to installations becoming eligible would entail frustrating the legitimate expectations of those who relied upon the old policy. However, in this case, the legitimate expectation argument seems pointless because the stronger effect of the presumption against giving retroactive effect to secondary legislation is extensive even to apparent retroactivity.

Regarding individual administrative decisions, a landmark case is Re 56 Denton Road 29. The War Damages Commission sent a letter to the plaintiff’s house to inform her that her house, partly destroyed by enemy action, was previously considered a “total loss”, now was a “non-total loss” (the qualification “non-total loss” entitled the plaintiff to a larger compensation sum). Later on, the Commission sought to change its previous determination and qualify the damages as a total loss because they realised that they had created an “awkward precedent” 30. Vaisey J held that communicated decisions which affect the rights of the subject are final and conclusive and cannot be altered in absence of statutory power 31.

The decision made in Denton Road entails that the administrative body cannot reassess its decision once it is final and communicated 32. Seen from the perspective of the different levels in which retroactivity can take place, this prohibition supports the idea that inconsistent administrative decisions cannot alter what was already decided beforehand. Moreover, this decision prompts the idea of a ban extensive to apparent retroactivity in the context of individual administrative decisions because the legal position generated by a favourable adjudication cannot be altered by a subsequent inconsistent decision without express statutory power to do so.

One may object that the decision to change the previous determination was unlawful because it involved terminating a legal right. However, the termination of a legal right is not unlawful in itself. As seen in the previous section, statutes can authorise the revocation of a decision that created legal rights. Nevertheless, Friends of the Earth and Denton Road show that having the power to make a decision does not necessarily entail the power to reassess it every time the authority’s mind changes. Those decisions are not “freely revocable”: a special power is required for that matter 33.

Although these cases are addressed mainly by the courts in terms of retroactivity or legal rights, it is possible to link them to the legitimate expectation doctrine. Similar cases have produced relevant decisions on legitimate expectations. For this matter, I will analyse the leading case on substantive legitimate expectations R v North and East Devon Health Authority, ex p Coughlan 34.

In 1971, Miss Coughlan was left seriously disabled after a car accident and in need of constant nursing care. The NHS provided this care in the Newport Hospital for over 20 years. In 1993, she was encouraged to move to a new nursing house called Mardon House (MH). After she and other patients were promised they could stay there “as long as they chose”, she agreed to move.

Later on, the health authority issued rules regarding the eligibility criteria for long-term NHS care. The authority decided that “specialist” nursing services would be provided by the NHS, while “general” nursing services should be purchased by local authorities. Under the new rules the health authority concluded that none of the patients in MH required “specialist” nursing services. Therefore, in 1998, after a public consultation, the authority decided to close MH and transfer the patients to the local authority’s care.

On one part, the court was satisfied that the applicant had a legitimate expectation grounded on the promise made by the authority and that it was an abuse of power not to honour it 35. This side of the case has a close resemblance to Friends of the Earth and Denton Road. Can the authority change its policy, taking away the right created by the promise that Miss Coughlan could make MH her “home for life”? It is unclear that the authority had the power to make that decision.

Moreover, in R (on the application of BAPIO Action Ltd) v Secretary of State for the Home Department a guidance issued by the Secretary of State for Health was declared unlawful because it introduced “a new term, unwritten and formally unauthorised” into previously granted permissions to remain in the UK; changes that where different from what legislation required 36. However, three of the law lords addressed the case in terms of legitimate expectation 37.

A different situation is the case Rootkin v Kent County Council 38. In 1976 the applicant’s daughter was awarded a season ticket for public transportation. The decision was made because her home was thought to be more than 3 miles away from her school. The authority was under a duty to award the ticket when the student’s house was 3 miles or more away from school and had discretion to award the ticket when the distance was less than 3 miles 39. After a few months, the distance was measured for a second time. It became clear that the distance was less than 3 miles. The authority decided to withdraw the season ticket only prospectively. Mrs Rootkin sought judicial review, but the application was dismissed since that the decision was not irrevocable after the actual distance was ascertained 40.

Lawton LJ said Rootkin was distinguishable from Denton Road and Livingstone because in Rootkin “the citizen has no right to a determination on certain facts being established; but only to the benefit of the exercise of discretion” 41. This quotation deserves two comments. First, Lawton LJ could argue that Mrs Rootkin was entitled only to the benefit of the exercise of discretion just because the authority claimed to have mistaken the facts. Otherwise, the authority would have been bound by its previous determination, as in Denton Road and Livingstone. Second, in Denton Road and Livingstone, the plaintiffs were awarded a benefit after the discretionary assessment of a set of facts. The problem was that the authority could not reassess its decision just because it did not like it.

This case shows that the revocation of favourable decisions requires some further grounding or justification than a change in the authority’s mind 42. I think that in Rootkin the fact that they mistook the distance is essential to the decision reached by the court. A policy change may allow the removal of legal rights created by an older policy. Also, ultra vires acts can be revoked with actual retroactive effect or only prospectively 43. In Rootkin the season ticket was withdrawn only for the future. Hence, the decision only had an apparent retroactive effect. Was that decision made considering the expectations of Mrs Rootkin? Does the apparent retroactive effect of that decision constitute a breach of the applicant’s legitimate expectation? This case also shows the intimacy between retroactivity and legitimate expectations.

3. Shortcomings in the status quo

3.1. Retroactivity and Individual Planning

The first problem I want to address is related to the seemingly crystal-clear connection between retroactivity and individual planning. Retroactive legislation cannot guide human conduct. Yet, the simple idea “that a rule passed today should govern what happens tomorrow, not what happened yesterday” can become a complicated one while deciding particular cases 44.

This complication happens for two reasons. The first is that there are degrees of retroactivity, as shown in the last section. Decision-makers only sometimes take this into account. The second has to do with the nature of retroactive decision-making. Regardless of how retroactive, no authority can reshape the past. A subsequent retroactive decision can only change the legal consequences of facts governed by a previous law or decision to attain some goal in the present.

Fuller says an ex post facto criminal statute is retroactive and unfair. However, he casts some doubt on a tax law enacted, for instance, in 2019, imposing a tax on profits gained in 2015. “Such a statute may be grossly unjust”, but you cannot say that it is, strictly speaking, retroactive”, because, whilst the authority takes into consideration new rules that were not in force to calculate the tax, the obligation to pay operates only prospectively 45.

Under this approach, no law can be retroactive. Any rearrangement of the legal consequences of facts concluded in the past that a retroactive decision makes can only take practical effect from its issuance. Therefore, following Fuller’s example, even if a new rule taxes retroactively past profits not governed by any tax-imposing legislation, he will not consider that a real retroactive effect because public bodies can only enforce that obligation upon the coming into force of the new legislation. Nevertheless, the problem is that the same argument is valid for criminal legislation, which he considers retroactive. A retroactive criminal statute, one can say, does not impose the obligation of “having had to do something in the past”, only the obligation of suffering the penalty for having done something which now the legal system considers an offence.

Thus, it is not true that retroactive laws are unfair because they require someone to do the impossible of doing something in the past. They are unfair because they assign some legal consequence to facts concluded in the past to enforce them at present. As a result, the statute makes an obligation accrue upon past events.

Fuller may be right in the example given if we understand that what sustains the ban on retroactivity were only the calculations and decisions made in the past. However, a sufficiently announced retroactive law or decision satisfies that underlying normative foundation in that case. For instance, a law enacted in 2019 imposing a tax on profits gained in 2015 announcing that the authorities will collect the tax in 2022 provides citizens with a wide range of time to prepare financially, thus not committing any unfairness. Hence, legislation or decisions operating only prospectively or having a vacancy period are not unfair when they allow individual planning, even if they have an ex post facto effect.

However, this line of thought makes the non-retroactivity principle rely on individual planning to such an extent that it falls into the logic of the legitimate expectation doctrine. Legitimate expectation casts upon the authority a duty to act fairly when having raised expectations by previous decisions or representations, seeks to frustrate them by a subsequent inconsistent decision. Meanwhile, the doctrine of non-retroactivity has a prohibiting sense (there is a presumption against retroactivity, thus not being favoured by the law), something which is inconsistent with the common grounding of the principle on individual planning: decision-makers should not regard as unfair a retroactive decision provided that it allows individual planning. Moreover, decisions with a retroactive effect that alter some form of unfair planning or reliance on the part of the citizen should not have any problem.

An excellent example of how the consideration of individual planning is only sometimes in line with non-retroactivity is McTier v Secretary of State for Education. The appellant, previously convicted for sex offences in 1985, 1988 and 1995, was affected by an order prohibiting him from teaching at certain schools under legislation coming into force after the last of the offences applying to any person “who is employed o engaged to carry out teaching work”. The court said this is not a case of strong retroactivity where vested rights are taken away, but “only the application of an adjusted sanctions regime”. Therefore, the State Secretary had the power to impose the order 46.

I think “an adjusted sanctions regime” is ex post facto legislation because it entails attaching new legal consequences to past events, thus imposing a new obligation in the present. The appellant could defend himself by arguing that the decision is retroactive because he committed the offence relying upon the old legislation that did not prevent him from teaching. However, that position is counter-intuitive because he was not supposed to commit the offence in the first place. Should the law protect a form of planning that is illegitimate? I think not. Does that mean legislation can always retroactively increase or adjust a sanctions regime? Also, no. Nevertheless, under the paradigm that non-retroactivity protects reliance and individual planning, we could conclude that the planning made in McTier was legitimate or that, being illegitimate, the legal consequences of his crime can always vary.

Considering the path travelled so far, we have two options. First, to abandon the prohibiting sense that pervades the ban on retroactivity and reformulate the principle more coherently with its grounding on the protection of individual planning. Second, we reformulate the foundations of the non-retroactivity principle to give a better explanation for its prohibiting force. There are good reasons for taking the second option and formulating a different normative foundation for the ban on retroactivity. Fuller has an inkling of this when he says that legislation imposing taxes upon past profits may not be retroactive, yet “grossly unfair”. Also, the ruling of the High Court points in the same direction because they considered the decision disproportionately harsh, inter alia, because the minister did not give any weight in the decision to the fact that many years had passed since the last offence 47.

3.1. Apparent Retroactivity is also Retroactivity

The second problem of the status quo is how misleading the idea of “apparent” retroactivity is. What is monstrous about retroactive decision-making is said of ex post facto legislation. As Fuller stated, “[t]o speak of governing or directing conduct today by rules that will be enacted tomorrow is to talk in blank prose. To ask how we should appraise an imaginary legal system consisting exclusively of laws that are retroactive, and retroactive only, is like asking how much air pressure there is in a perfect vacuum” 48. This criticism makes sense only for the operation of regulating human affairs by enacting a rule after the facts upon which that rule takes effect.

A clear example of this form of retroactivity takes place in Marks & Spencer v Customs and Excise Commissioners. Parliament enacted legislation to curtail the limitation period to recover unduly paid taxes. In this case, the company was deprived of the chance to recover unduly paid VAT because the new legislation reduced the 6-year limit to 3. The new rules gave a different legal effect to the time elapsed before its enactment to provide the authority with a reason to disregard, at present, the refund claim. Thus, the new legislation favours an adjudication different from what the tax-payer expected when the old legislation was in force. Here the ECJ held that retrospective legislation that made it impossible for a company entitled under previous legislation to collect a refund for unduly paid taxes is contrary to the principles of effectiveness and legitimate expectation 49.

The new legislation modified the legal effect of facts that had taken place in the past, amounting to actual retroactivity. Under the old legislation, a 3-year lapse did not preclude the company’s right to recover unduly paid taxes since a 6-year period was required. Nevertheless, the new legislation established with a retrospective effect that the 3 years passed under the old legislation were enough to preclude the company’s right.

In this case, the new rules had an ex post facto effect. However, retrospective legislation often presents the problem of taking away rights acquired in the past 50 without effecting concluded events by terminating legal rights with an immediate effect rather than plain retroactivity. I see this situation in the cases analysed in the previous section Nigel Rowe, Alec David & Others v The Commissioners for HM Revenue and Customs and Rootkin v Kent County Council.

Removing a postponement right by new legislation or a season ticket by a new assessment of the facts was no decision having an actual retroactive effect as if by some new criteria, the postponement right or the season ticket should be considered as unlawfully granted in the past. Instead, those benefits were terminated with immediate effect, something that Craig believes to be only “apparent” retroactivity, meaning that it is no “real” retroactivity.

However, this argument is debatable. Of course, the termination of a legal right with immediate effect is not a decision that has an actual retroactive effect because it does not regulate human conduct using an ex post facto rule. Nevertheless, that does not mean that no retroactivity occurs in that decision. Terminating a legal right may not be retroactive because the decision does not take effect upon events concluded in the past. Yet, it is retroactive in that, by some new criteria, it terminates a legal right or legal position originated under legislation that provided for longer over-time endurance.

Decisions terminating legal rights or impairing obligations accrued in the past are retroactive, although to a lesser degree than decisions having ex post facto effect. This is why Waldron makes a distinction between “retroactivity” (meaning ex post facto rules) and “retrospectivity” (meaning decisions having an immediate effect over legal positions developing in the past) 51. And this is also why in Spanish Law scholarship distinguishes three degrees of retroactivity: maximum, medium and minimum 52. The distinction can help in making a difference between cases concerned with retroactivity from those concerned with legitimate expectation.

I believe that decisions having this lesser kind of retroactive effect entail a more significant threat to the Rule of Law than decisions that only frustrate legitimate expectations. The alteration of legal positions originating in the past involves a change, challenge or departure from something “already decided” in the past. This principle sustains the rulings in Denton Road, Friends of the Earth and BAPIO, where the authority sought to change (for the future) a decision made in the past which created a legal right. This conduct is different and more problematic than just frustrating an expectation, however “legitimate” it was.

The distinction between actual and apparent retroactivity is helpful because it explains why it is unlawful for some decisions to take immediate effect upon legal positions generated in the past. Decision-makers should not construe legislation as having an actual retroactive effect. However, it usually can have an apparent retroactive effect. Administrative decisions cannot even have an apparent retroactive effect unless express statutory power exists.

However, the idea of “apparent” retroactivity is misleading because a decision having an immediate effect upon legal positions originating in the past is retroactive, although to a lesser degree. Thus, it is confusing to understand decisions having an apparent retroactive effect as unlawful because they frustrate legitimate expectations. In this sense, in the third section, I will argue that the law should award more protection against decisions having the so-called “apparent” retroactive effect from those that only frustrate legitimate expectations.

3.2. Unfair protection of legitimate expectations

The third problem is that the overlapping of retroactivity and legitimate expectation can prompt excessive protection of legitimate expectations, thus undermining the authority’s capability of introducing changes in past decisions for the common good. As I said before, a vital element of the non-retroactivity principle is its “prohibiting sense”. If decision-makers combine this prohibiting sense with the expansive notion of “legitimate expectation”, we can obtain a dangerous blend: a ban on the frustration of legitimate expectations.

To show how this may happen, I will analyse Odelola v Secretary of State for the Home Department 53. The claimant applied for leave to remain in the UK and paid an application fee of £335. The migration rules in force at that time provided that people with “acceptable” training could remain as postgraduate doctors. While the application was pending, the migration rules changed and provided that “acceptable” training was insufficient to obtain leave to remain in the UK because a recognised UK degree was required. Then the application was refused under the new rules since the applicant did not comply with the new regulation.

The decision made by the Home Secretary (upheld by the Asylum and Immigration Tribunal and the Court of Appeal) was challenged for taking away accrued rights since the application was made under the previous policy 54. The House of Lords dismissed the appeal. They considered that no right existed at the time of the application and that the Home Secretary had to decide with whatever rules were in force at the time of the decision 55. He could lawfully change the immigration rules. Also, no legal right was involved, for there was no final decision when the new rules came into force.

If we follow Denton Road, the House of Lords was right in saying no legal right was involved in this case. However, it is remarkable that four of the law lords agreed that it was unfair for the Home Secretary to keep the fee paid while the modification of the rules doomed the application to fail 56. In particular, Lord Scott of Foscote said that the applicant had a “justified expectation that her application would be successful” and that “[s]he paid the money on what turned out to be a false and misleading prospectus” 57.

At most, the applicant had a legitimate expectation created by the old policy. She relied upon that policy by paying the application fee. Nonetheless, courts should not enforce this expectation under the umbrella of legal rights or non-retroactivity hence not allowing their frustration at all, as the claimant pretended. Otherwise, the authority would not be able to apply the new policy to anyone who, in any way, had relied on the old one. Rather, the court can afford some protection to legitimate expectations but only sometimes give them full enforcement.

Another interesting case in this respect is R v Minister of Agriculture Food and Fisheries, ex p. Cox 58. The applicant bought a grazing license and a milk quota. The Milk Marketing Board (MMB) registered the transfer. The applicant never entered into physical occupation of the land because it was thought not necessary for the transfer of the milk quota to take place. Two years later, the seller asked the MMB to declare that the quota transfer was invalid had occurred. Still, the MMB dismissed that petition. Tland occupationUpon this decision, correction of the,

Popplewell J considered the case to fall squarely within the principle laid down in Denton Road. Therefore, the MBB could not take back the decision not to alter the registry 59. Nonetheless, he afterwards said that transferring a milk quota requires physical land occupation. Therefore, despite the infringement of the ECC regulations, the registry was not altered at all 60.

Although Denton Road was applied, it cannot be contended that Mrs Cox had a legal right because there was an ultra vires act. At best Mrs Cox’s legal position is an ultra vires legitimate expectation. However, the court decided to give full protection to that position as if it was a validly acquired legal right.

4. Proposal

4.1. Finality and Retroactivity

Parliament can dismiss the presumption against retroactivity and alter the effect of facts concluded in the past using clear and unambiguous terms. Nevertheless, the so-called actual retroactivity may even be contrary to a formal conception of the Rule of Law which demands “that individuals should be able to plan their lives on the basis of clear, open and general laws” 61. This leaves open the question of how far Parliament can go in enacting legislation with full retroactive effect. In these cases, the HRA 1998 may limit the ex post facto effect of primary legislation 62.

However, the current normative foundation of the non-retroactivity principle on individual planning only partially explains the prohibiting sense of the principle. Hence, a more robust foundation is required to explain the prohibiting sense of the non-retroactivity principle.

For this purpose, I will turn to the notion of finality. The idea of finality entails that lawful decisions made by public authorities should have some conclusiveness; otherwise, the Rule of Law would not receive enforcement since, without conclusiveness, no issue would be indeed decided. This idea had developed mainly for the case of judicial bodies through the doctrine of res judicata 63. The rationale is as follows: once a judge decides a case, he cannot reopen it, thus to secure that, at some point, the procedure comes to an end 64. This finality is regardless of the correctness in law or fact of the decision 65. Thus, judicial rulings are final and conclusive, and the possibilities to reverse them should be restricted to fraud or some exceptional kind of impropriety 66.

This doctrine is consistent with the understanding that the primary function of the Judiciary is deciding isolated disputes 67. Finality in judicial decision-making is necessary to achieve its purpose because conflict is solved only once by a final decision binding the parties. Moreover, there is a public interest in having disputes or issues settled conclusively 68.

Bearing this in mind, I think the idea of finality can also be applied, mutatis mutandi, to legislation and administrative decisions. Finality in public law is a key principle and element of sovereignty 69.

The function of Parliament is to be a representative body that regulates human conduct with prospective general norms. Meanwhile, the role of the Executive is to implement laws regulating human behaviour based on expertise and skill to achieve policy goals 70. Additionally, to effectively regulate human conduct, Parliament and the Executive not only have to enact and apply rules allowing individual planning, but, most essentially, public bodies should decide issues according to the legal instruments in force during their occurrence.

Doing this is critical for the Rule of Law because extensive ex post facto decision-making can turn a legal system based on rules into one where public bodies decide on a case-by-case basis.

Having general ex post facto decision-making powers means that a public body can enact rules purposed to govern facts verified before they came into force. Moreover, such power allow changing what previous rules provided in the past, for facts swallowed by time. Thus, the mere existence of powers to make decisions with actual retroactive effect jeopardizes the fundamental capability of rules to decide issues occurring during their operation, allowing public bodies to decide over past situations on a case-by-case basis.

The same as no legal dispute is concluded until the court reaches a final decision; no issue is finally governed by legislation or any general rule until the legal effect attached to some event becomes immutable to change. Of course, this does not mean that rules can never vary because they never become “final”. It only assures that the norms applicable to rule about specific facts remain unaltered, notwithstanding they can change for ruling about facts occurring after the enactment.

Hence, phrases like that a “system of law composed exclusively of retrospective rules could only exist only as a grotesque conceit worthy of Lewis Carroll or Franz Kafka” 71 (which portray abhorrence for actual retroactivity) cannot be fully explained only under the umbrella of individual predictability of the law. The ban on actual retroactivity is more crucial to the Rule of Law than that: it ensures the efficacy of rules as a regulatory instrument, shielding them from being rendered useless by subsequent ex post facto decisions. For this reason, authorities should handle with the utmost care legislation having actual retroactive effect, even when enacted for “curative” or “welfare” grounds 72.

4.2. Total and Partial Retroactivity

In the previous section, I argued that the distinction between actual and apparent retroactivity helps differentiate between decisions having an ex post facto effect from those which remove or alter decisions or rights acquired in the past. However, the label “apparent” is misleading because it suggests that those decisions are not retroactive while seeming to be. However, I think these decisions have some (less evil) real retroactive effect. When a public body makes decisions with immediate effect over legal rights acquired in the past or issues settled by past administrative decisions, we may also find trouble with the requirement for finality embedded in the Rule of Law.

The requirement of finality also explains why the doctrines of res judicata or functus officio may have a role in the administrative context, ideas related to cases connected to the legitimate expectation doctrine 73. Regarding res judicata, Wade & Forsyth state: “[t]he same arguments which require finality for the decisions of courts of law apply to the decisions of statutory tribunals, ministers and other authorities” 74. Functus officio points in the same direction. Meaning “having discharged his duty”, this doctrine expresses the idea that a public authority makes a decision, the power to make that decision ceases to exist 75.

Nevertheless, this immutability of decisions prompted by these doctrines has many exceptions in administrative if compared to judicial decision-making. The case law analysed in Collin’s paper can show that. In R v Dorset Police Authority ex. p. Vaughan the authority awarded the applicant the right to a pension after a medical assessment. The respondent asked for a second assessment, which was adverse for the applicant. In this context, the court decided that in the absence of fraud the first decision the authority had to consider the first decision as final because the regulation did not provide for a second assessment 76. Moreover, two other cases are cited supporting the idea that an unfavourable decision does not extinguish the power to make a new one 77.

Hence, res judicata and functus officio entail a degree of immutability incompatible with the law allowing public authorities to review their own decisions in several situations. Meanwhile, the law, as laid down by Denton Road, shows that it is forbidden for the authority to go back on a decision just because its mind changed about it, whilst it permits public bodies to alter a decision when statute allows it, or there is no final decision 78. This also supports the idea that there is an unquestionable requirement of conclusiveness in administrative decision-making which is however less crucial for the Rule of Law than finality requirements in judicial decision-making. More reasons may justify reopening a decision-making process. Ganz mentions five: ultra vires, error on the facts, change of circumstances, error on law or policy, and change of law or policy 79.

Apparent retroactivity requires special powers from Parliament or good reasons to take place. Without that, a decision having an apparent retroactive effect is unlawful. Because of this, the presence of a legitimate expectation is immaterial in cases concerned with apparent retroactivity. For instance, much of the discussion in Coughlan was about whether the health authority had the power to remove Miss Coughlan from the continuing care awarded since the day of the accident. Although the court decided that they did not have the power to remove her from Mardon House 80, the case (with clear support in the court’s reasoning, yet very much arguably) is largely considered as the leading substantive legitimate expectation case 81.

In other cases, there is a link between legitimate expectation and the duty to act consistently, giving equal treatment to every individual. A court may rule that a decision is unlawful if a public body departs from its general policy in a particular case or acts differently from what it decided for other individuals. In R (Rashid) v Home Secretary, the authority administered his asylum policy inconsistently in the case of one applicant resulting in the refusal of the refugee status. The applicant, however, did not know about the existence of that policy. Nevertheless, Pill LJ said that every individual could expect that the authority will administer its policy, and Dyson LJ noted that the vital issue in a legitimate expectation case is to determine what the authority committed itself to, even if the individual was unaware of that 82. A similar approach is in Mandalia 83.

It is a foremost strange thing to say that an expectation was frustrated when the individual did not expect 84. Hence, like in Coughlan, I think legitimate expectations are immaterial to these cases. The courts are mixing two things in this type of case: the requirement of stability and the idea that the law should be administered consistently in an equal form 85. The main concern of a consistency-equality case should be whether it is lawful for a public body to depart from its general policy or precedent in a particular case. Of course, a public body may “depart” from a policy when the case can be distinguished from others simply because the policy is not applicable.

However, a public body can act inconsistently with the support of good reasons for doing so 86. Absent those reasons, the unequal decision will be unlawful, and the public body will have to apply the general policy in force or act consistently with previous decisions. In such a situation, expectations are irrelevant because equality, as a ground for review, does not require them to work. Consequently, retroactivity (actual or apparent) and the “equality-consistency principle” should be regarded as independents grounds for review 87. An unlawful decision cannot alter a previously lawfully made decision or frustrate a legitimate expectation originating in a public body’s past conduct. This analysis shows that in Coughlan and the equality-consistency cases, the importance of the legitimate expectation doctrine decreases when the policy change or the departure from the general policy is unlawful on other grounds, such as ultra vires or the existence of unequal treatment.

The distinction between actual and apparent retroactivity is helpful. It shows that retroactivity can take place on different levels. While actual retroactivity is a more brutal form of retroactivity, apparent retroactivity consists of attaching new legal effects to legal positions originating in the past. Therefore, it is a “softer” form of retroactivity, although still real retroactivity.

Therefore, I will propose the following terminology. I will call “total or full retroactivity” the effect of a norm or decision upon events concluded in the past because the legal effect of the new norm or decision is attached to a fact prior to their coming into force. Meanwhile, I will call “partial retroactivity” the effect of a norm or decision that attaches a new legal consequence, in the present, to events or transactions originating in the past. I think this wording is better because, although partial retroactivity is concerned with events originating in the past, the norm or decision only takes effect from the day of its enactment and on. Under these terms, norms or decisions that affect the planning undertaken in the past are not retroactive in any sense. Nevertheless, such decisions may attract the protection of the legitimate expectation doctrine.

4.3. The Temporality Test

We can find a better explanation for the prohibiting sense of the non-retroactivity principle under the umbrella of finality. Thus, the legal system can work with rules and not have to decide everything on a case-by-case basis. Meanwhile, legitimate expectation is connected to the requirement of stability, individual planning and transition relief when the authority has legal powers to change its mind on a previous decision or representation. Because of these different foundations, the legitimate expectation doctrine is more likely to allow change whilst protecting individual planning via transitional measures or different forms of protection.

So far, courts have been doing this on a case-by-case basis bearing in mind detrimental reliance, the public interest and the hardship caused to the expectation-bearer, among other considerations. Elliot calls this “palm tree justice” 88. This is very clear in the case of unlawful decisions and unlawful representations 89. I believe that the courts are, implicitly, making a sort of public interest-expectation compatibility analysis, which I will formulate explicitly. Sometimes the fulfilment of an expectation does not seriously undermine the public interest pursued by an inconsistent norm or decision. In that case, courts usually fulfil expectations affected by an inconsistent decision. In other situations, fulfilling an expectation undermines the public interest. In that case, courts give less protection to expectations.To show how fulfilling an expectation may not necessarily undermine the public interest behind a subsequent norm or decision, we can go back to Cox. First, it was contended that the ECC regulations aimed to avoid the creation of a trade by requiring a transfer land use for a milk quota transfer to take place 90. Yet, there was a practice among farmers to use the grazing agreement, without actual occupation taking place, to be able to transfer the quota without refusal by the authority 91. Second, it is clear that no trade was made out of the milk quota since Mrs Cox was producing milk and not trying to resell the quota for a higher price. If the court had declared the transfer ab initio void, that would have entailed the payment of a special levy for producing milk without a quota 92. Therefore, the legitimate expectation created by the unlawful decision can be upheld, bearing in mind that the decision did not harm the public interest.

In Cox there was a ground for change. However, fulfilling that ultra vires expectation did not undermine (or only in a minimal form) the public interest because the applicant did not buy the quota to create trade but to produce milk. This intuition contradicts the assumption that every illegality has a necessary overriding effect 93. The fulfilment of an expectation, although formally incompatible with a regulation, can be compatible with the goals pursued by that regulation given the particular circumstances of the case.

Suppose the public interest is incompatible with the fulfilment of the expectation. That does not mean that the expectation should not be protected at all. “Grandfathering” an expectation is only one possibility. In other cases, legitimate expectations have received less protection. An example of this is R (Bibi) v Newham London Borough Council 94. The applicants to a housing benefit were considered in 1994 to be unintentionally homeless and were provided with temporary accommodation, while the housing authority promised to secure permanent accommodation within 18 months. Later on, the House of Lords ruled in a similar case that the housing authority was not obliged to provide permanent accommodation to homeless people 95. In addition, in 1996 an Act of Parliament came into force which restricted the duty of the authority by creating a new allocation scheme for that purpose with different priorities of who should receive secure tenancy. Finally, the authority failed to fulfil its promise on time.

The court decided that not giving any consideration to the promise was unlawful and then referred to the authority the weight it ought to give to the legitimate expectation 96. In this case, the 1996 Act makes it possible to say that the applicant’s position is a legitimate expectation. However, although the promise is not rendered unlawful, the authority must apply the new allocation scheme. This case is distinguishable from Coughlan because the authority had to construe the 1996 Act as having a partial retroactive effect. Hence, there was a ground for change. Also, the court did not say the applicant’s legal position had to be fully protected. They referred the decisions to the housing authority and suggested financial compensation or amending the allocation scheme in order to give some weight to the expectation 97.

In Bibi it was not reasonable for the authority to fulfil the applicant’s expectation because there were many in the same situation. Thus, it was impossible to comply with everyone’s expectations 98. Nonetheless, as the court suggested, the authority could have given some weight to the expectation in applying the new allocation scheme.

This implicit public interest and expectation approach has three problems:

1) For the sake of transparency, the courts should explicitly address these issues and state in a clear form how compatible the fulfilment of an expectation is with the public interest.

2) The courts should not assess by themselves how compatible the public goal pursued by a norm or decision is with an individual’s legitimate expectations. Making that decision may be an infringement of the duty of comity of the courts towards the administrative authority 99.

3) Suppose the public interest does not tolerate the frustration nor any transitional measure to give some protection to the frustrated expectation. In that case, courts should consider the possibility of giving monetary compensation to those who have detrimentally relied on a previous norm or decision made by the authority.

Considering these problems, I will propose changes to the approach made so far by the courts. Some of these propositions may entail a change in the law or have significant institutional consequences. However, I believe that fairness requires restating problems associated with the tension between stability and reform have been addressed so far. These propositions are what follows.

First, to identify a legitimate expectation case, we must check if the authority can frustrate the expectation. A public body may not have the power to make a decision with a partial retroactive effect, or some other ground for judicial review may prevent that decision from taking effect. If such power exists, legitimate expectations will be material to the case. Next, we must decide whether the public interest is compatible with fulfilling the expectation. The case law shows that there are situations in which the fulfilment of a legitimate expectation brings no danger to the objective of the new policy. Hence, estoppel may seem an appropriate solution in some cases 100. However, courts should avoid giving pre-established solutions in legitimate expectation cases by applying estoppel, res judicata, and functus officio or retroactivity. Instead, they should analyse whether the decision to frustrate an expectation was justified adequately in the public interest, even without detrimental reliance.

Legitimate expectations should be fulfilled as long the public interest can tolerate them in the administrative authority’s criteria. Consequently, the administrative authority has the burden of justifying the decision, which frustrates a legitimate expectation. From a temporal perspective, to frustrate a legitimate expectation, the administrative authority should deal with the following questions: (1) is the fulfilment of the expectation compatible with the public goal pursued by the subsequent norm or decision which seeks to frustrate it? If yes, then the expectation ought to be fulfilled. If no, then the authority should consider (2) whether there is any form of transitional relief compatible with the public interest that the authority could award to the individual. Thus the expectation receives enforcement intending to reduce the hardship caused to the individual.

Answering these questions is, in the first place, a job for the administrative authority 101. Because of this, courts should quash decisions which have not adequately considered the individuals’ legitimate expectations and order the authority to address the questions because administrative bodies are in general better suited to assess whether the public interest can tolerate the fulfilment of an expectation. Also, under the proposed approach, this kind of judicial scrutiny of the Executive can help improve the administrative decision-making process 102. The administrative body has to address these three questions when making its decision. However, it may be more challenging for the public body to identify all the expectation-bearers wronged by the decision and their particular circumstances. In this context, judicial proceedings can help because they put in front of the public body situations that went unnoticed or it did not consider in the decision-making process. Hence, judicial proceedings can assist authorities in seeing the case in a new light to ascertain the compatibility of certain expectations with a public goal or what transitional measures may be adopted.

Once the public body adequately answers the questions, the courts should refrain from further scrutiny to not breach their duty of comity towards the administrative body. Of course, the idea of a “proper” consideration may entail some interference in the exercise of discretion, yet to deny the possibility of review of this ground may purport to the denial of all forms of judicial control of the Executive 103. A good balance is made if the court refers the public body for a new decision to be made instead of assessing what is to be considered compatible with the public interest. The questions only need a reasonable answer, and this may even entail a negative answer for both of them 104.

If no transitional relief is possible, then (3) the authority should consider awarding enough monetary compensation for the expectation bearer to allow adapting to the new policy. This solution entails a change in the law. Still, scholars have suggested it 105. Compensation is a good idea because the policies and representations made by the authority impact economic decisions. Therefore, the frustration of the expectations induced by those representations or policies may entail economic loss, derived from detrimental reliance.

However, courts have to strike a delicate balance in awarding compensation. It will result in encouraging undesirable behaviour if courts overcompensate investors. Hence, but not securing the future possible profits the investor would have gained had the regulation remained unaltered. Compensation entails, indeed, a very complex issue. On the one hand, not awarding compensation is abusive because the authority can change legal rules or policies without considering the effects of previous decisions in the individuals. On the other hand, overcompensating may prevent the state from introducing crucial changes. For instance, in a case like, R v Ministry of Agriculture, Fisheries and Food, ex parte Hamble (Offshore) Fisheries Ltd, the existence of investments rendered useless is real damage caused by a policy shift 106.

The tension described in the introduction brings us to focus on a temporal dimension not explicitly considered by the other approaches. Certain aspects of a reasonability or proportionality approach can be identified in the expectation temporality test 107. However, although not totally different, it is free-standing. The legitimate expectation doctrine, as understood in this paper, is concerned with the over-time compatibility of an expectation with a public goal. This emphasis on the temporal side of things is not essential to reasonability and proportionality and justifies taking a different approach.

5. Conclusions

Legal scholarship has found the normative foundation of non-retroactivity and legitimate expectation in the Rule of Law requirement of securing reliance and individual planning. Courts have used both principles indistinctly while addressing legislative or administrative temporal inconstancy cases. However, I concluded that this overlapping needs to be revised because:

1) The foundation of individual planning does not fully explain the prohibiting sense and the “horror” often attributed to ex post facto decisions or legislation.

2) The so-called “apparent” retroactivity is a form of real retroactivity, and its extensive use can seriously undermine the Rule of Law.

3) The extensive use of the legitimate expectation doctrine can freeze public bodies’ capability to introduce changes to achieve important policy goals.

For these reasons, to make sense of the case law on temporal inconstancy and to avoid unfairness, I propose three things:

1) Finality is a better normative foundation for the non-retroactivity principle than individual planning. Non-retroactivity enforces a more crucial aspect of the Rule of Law, which is having general rules and not deciding cases on a case-by-case basis. The extensive use of ex post facto decision-making powers can make rules or decisions which govern past events irrelevant because their effect upon past events is always subject to change. Furthermore, finality better explains the prohibiting sense that the non-retroactivity principle has: it secures the power of legislation or decisions to take definitive effect upon the events they govern.

2) Since retroactivity can take place in different degrees, it is helpful to distinguish between decisions having total retroactive effects from those just having partial retroactive effects. Partial retroactivity is concerned with decisions having an immediate effect which are temporally inconsistent with legal positions generated in the past by some other decision made by a public body. A public body requires special powers to make decisions having partial retroactive effects.

3) Beyond total and partial retroactivity (or when the law allows retroactivity), the legitimate expectation doctrine is essential to secure stability and the possibility of relying on the legal order. For that purpose, we can apply the expectation temporality test. Even without reliance, the authority had to justify the decision from a temporal perspective to frustrate a legitimate expectation lawfully. This involves the consideration of (1) whether the expectation is compatible with the public interest pursued by his subsequent decision. If not, the authority should consider (2) whether transitional measures can be adopted to protect the individual’s expectations. Finally, suppose the public interest cannot tolerate the fulfilment of the expectation nor any other transitional measure. In that case, (3) monetary compensation should be awarded to those who made investments rendered useless by the inconsistent norm or decision.

6. Bibliography

Jaime ARANCIBIA, Judicial Review of Commercial Regulation, Oxford University Press, Oxford, 2011.

Jonathan AUBURN, Judicial Review: Principles and Procedure, Oxford University Press, Oxford, 2013.

Joanna BELL, “The Doctrine of Legitimate Expectations: Power-Constraining or Right-Conferring Legal Standard?”, Public Law, 2016, 3, pp. 437-455.

Alexander BROWN, A Theory of Legitimate Expectations for Public Administration, Oxford University Press, Oxford, 2017.

Eoin CAROLAN, The New Separation of Powers: A Theory for the Modern State, Oxford University Press, Oxford, 2009.

Ben COLLINS, “When Are Public Bodies Functus Officio Law?”, Judicial Review, 2001, 6, pp. 61-66.

Paul CRAIG, Administrative Law, 8th edition, Sweet & Maxwell, London, 2016.

Timothy ENDICOTT, Vagueness in the Law, Oxford University Press, Oxford, 2000.

Timothy ENDICOTT, Administrative Law, 5th edition, Oxford University Press, Oxford, 2021.

Timothy ENDICOTT, “The Purpose of a State”, American Journal of Jurisprudence, vol. 66(1), 2021, pp. 69-83.

Mark ELLIOTT, “Legitimate Expectation: Reliance, Process, Substance”, Cambridge Law Journal, 260, 2019, pp. 260-264.

Mark ELLIOTT, “British Jobs for British Doctors: Legitimate Expectations and Interdepartamental Decision-making”, Cambridge Law Journal, 2008, 67(3), pp. 453-456.

Mark ELLIOTT, “Legitimate Expectation, Consistency and Abuse of Power”, Judicial Review, 2005, 10(4), pp. 281-288.

Mark ELLIOTT, “Legitimate Expectations and Unlawful Representations”, Cambridge Law Journal, 2004, 63(2), pp. 261-264.

Mark ELLIOTT, “Unlawful Representations, Legitimate Expectations and Estoppel”, Judicial Review, 2003, 8(2), pp. 71-80.

Lon FULLER, The Morality of Law, Yale University Press, New Haven and London, 1969.

Gabriel GANZ, “Estoppel and Res Judicata in Administrative Law”, Public Law, 1965, 1, pp. 237-255.

Daniel GREENBERG, Craies on Legislation: a Practitioners’ Guide to the Nature, Process, Effect and Interpretation of Legislation, 11th edition, Sweet & Maxwell, London, 2017.

KR HANDLEY, Spencer Bower and Handley: Res Judicata, 4th edition, Lexis Nexis, London, 2009.

F. E. LANCELOT and E. STRODE, Everest and Strode’s Law of Estoppel, 3rd edition, Stevens & Sons, London, 1923.

Clive LEWIS, “Retrospective and Prospective Rulings in Administrative Law”, Public Law, 1988, 1, pp. 78-105.

Richard MOULES, Actions Against Public Officials: Legitimate Expectations, Misstatements and Misconduct, Sweet & Maxwell, London, 2009.

Ben JURATOWITCH, Retroactivity and the Common Law, Hart Publishing, Oxford, 2008.

Joseph RAZ, The Authority of Law: Essays on Law and Morality, Oxford University Press, Oxford, 2009.

Paul REYNOLDS, “Legitimate Expectations and the Protection of Trust in Public Officials”, Public Law, 2011, pp. 330-352.

José María RODRÍGUEZ, “Sobre la retroactividad de las normas a los cuarenta años de la Constitución Española”, Revista española de Derecho Administrativo”, 202, 2019, pp. 51-82.

Søren SCHØNBERG, Legitimate Expectations in Administrative Law, Oxford University Press, Oxford, 2000.

Maya SIGRON, Legitimate Expectations Under Article 1 of Protocol No. 1 to the European Convention on Human Rights, Intersentia, Cambridge, 2014.

Jürgen SCHWARZE, European Administrative Law, Rev. edition, Sweet & Maxwell, London, 2006.

Cass SUSTEIN and Adrian VERMEULE, Law & Leviathan: Redeeming the Administrative State, The Belknap Press of Harvard University Press, Cambridge Massachusetts and London, 2020.

Robert THOMAS, “Legitimate Expectations and the Separation of Powers”, in Mathew Groves and Greg Weeks (edits.), Legitimate Expectations in the Common Law World, Hart Publishing, Oxford, 2017, pp. 53-77.

Iain STEELE, “Substantive Legitimate Expectations: Striking the Right Balance?”, Law Quarterly Review, 121(2), 2005, pp. 300-328.

William WADE and Christopher FORSYTH, Administrative Law, 11th edition, Oxford University Press, Oxford, 2014.

Jeremmy WALDRON, “Retroactive Law: How Dodgy Was Duyhoven?”, Otago Law Review, 10, 2004, pp. 631-654.

Rebecca WILLIAMS, “The Multiple Doctrines of Legitimate Expectations”, Law Quarterly Review, 132(4), 2016, pp. 639-663.

---