Discretion in property law: a study of judicial correction of registered title

Abstract

This paper examines discretionary decision making in a property law context: the statutory power to correct a mistaken entry on the land register. This discretionary power has importance as an occasion when the allocation of property occurs without fixed rules but is put into the control of the court without guidance on content or goals. The case-law is studied using legal empirical analysis. The judicial approach is described and findings are analysed. The paper relates the conclusions to doctrinal and policy debates in land registration and discusses how the judicial input advances understanding of the registration system as an integrated whole.

Introduction

The English model of land registration was established to bring greater certainty into title by visibly vesting title in a registered proprietor free from unprotected interests and conferring powers of disposition. Yet the defining test of a system's commitment to the register as a source of title – what happens when the register has been changed when it should not have been – is answered by an explicit statutory power to reallocate title as a matter of discretion. That this discretion ever gained traction is surprising since it challenges the central tenet that a register entry is effective to confer good title.¹

The scope of the power to sustain or to correct an entry due to fraud or error, and the place of that power within the registration system, have been the subject of academic scrutiny. Scholarly research has taken a doctrinal perspective which is informative in testing the parameters of the judicial power, promoting internal consistency and addressing the effects of different interpretative options. As a substantial body of case-law has now built up in this area, this paper changes tack by undertaking an analysis of the law in action to supplement the doctrinal literature. The paper reports on a systematic, empirical examination of case-law that was undertaken to investigate the exercise of the discretionary power to correct mistaken entries in the land register. It explores what the judicial component brings to decision making and how that might colour the understanding of land registration.

The following section will give an overview of the legal framework in which the discretion operates before the paper moves on to explain the study undertaken. The findings are then described and analysed with discussion of their implications.

The legal framework for the discretion to correct register entries

In the event of a mistaken entry, the English system has for almost a century avoided a fixed, uniform rule and has instead determined the issue of correction as a matter of statutory discretion. The Land Registration Act 1925 declared: ‘The register may be rectified pursuant to an order of the court or by the registrar.’² The power extended to a collection of specified circumstances, the last of which covered ‘any other case where, by reason of any error or omission in the register, or by reason of any entry made under a mistake, it may be deemed just to rectify the register’.³ The 1925 Act was replaced by the Land Registration Act 2002: ‘The court may make an order for alteration of the register for the purpose of- (a) correcting a mistake …’.⁴ The new correction power sits alongside distinct powers to alter the register for the purposes of bringing it up to date and of giving effect to rights excepted from the effect of registration.⁵

The exercise of the correction power is not left unregulated. It must be exercised in favour of correcting unless there are ‘exceptional circumstances’.⁶ Although this may be described as a discretion,⁷ in comparison to the 1925 Act its scope is rather limited as the starting point is not an even balancing of equities but the imposition of a duty. This leads to a process of adjudication which differs under the 2002 Act:

the court must ask itself two questions: (1) are there exceptional circumstances in this case? and (2) do those exceptional circumstances justify not making the alteration? The first of these questions requires one to know what is meant by ‘exceptional circumstances’ and then to establish whether such circumstances exist as a matter of fact.⁸

Where the correction of a mistake would prejudicially affect the title of a registered proprietor, known as ‘rectification’, two special consequences follow. First, the power is regulated differently again. No order for correction of the entry may be made in relation to land in the proprietor's possession unless he has ‘by fraud or lack of proper care caused or substantially contributed to the mistake’, or it would for any other reason be ‘unjust for the alteration not to be made’.⁹ Those two conditions substantially replicate provisions also found in the 1925 Act.¹⁰ If either is present, then the power is restored and must once more be exercised unless there are exceptional circumstances.¹¹ Secondly, the power is supplemented by state compensation for whichever party suffers from the adverse exercise of discretion, whether the entry is rectified or allowed to stand.¹²

There is scarcely an element of the power to correct that has escaped controversy. This is important to the present exercise because the doctrinal ambiguity might affect the exercise of discretion. In particular, judicial descriptions of the material relevant to the exercise of discretion are likely to depend on the interpretation of the legal parameters within which the discretion operates. The first set of controversies has concerned the scope of the correction power. ‘Mistake’, the criterion which determines whether an entry is liable to correction, has been elusive.¹³ There has been doubt whether a mistake will persist following a disposition or whether the mistake is thereby expunged.¹⁴ There is a related issue over what may be ordered for the purpose of correction: whether an entry could be corrected though it is not itself a mistake,¹⁵ and whether correction operates retrospectively.¹⁶ If there is a power to correct against a transferee, then it raises the issue of the nature of the claim to rectification, whether it might be governed by statutory priority rules and, if so, which of them.¹⁷ These are matters with which case-law has begun to grapple, but which remain far from satisfactorily concluded.

The second subject of debate has been the integration of the correction power into the land registration scheme. This is seen in the enduring discussion of how the correction power interacts with the rule that title vests by virtue of registration. While the rule undoubtedly vests the legal estate,¹⁸ there has been concern over whether it carries the full beneficial interest. That prompts questions over the interface between that provision and residual principles of equity; in particular, whether there is scope left for resulting or constructive trusts¹⁹ and whether the word changes from the 1925 Act to the 2002 Act would affect the availability of trusts.²⁰ If the proprietor has the mere shell of legal title through registration,²¹ then equity's imposition of a trust is taking the place that would otherwise be occupied by the statutory discretion to rectify the register – leaving any alteration proceedings without redistributive effect but only the limited function of bringing the register in line with the equitable entitlement. If encountered in correction proceedings, the lack of any prejudice to title would inevitably channel the decision maker's characterisation of the case and lead to a strong inclination to correct.²²

The specific concerns expressed in the literature reflect two broader issues. First, the degree to which the register may be portrayed as a reflection of existing title as opposed to a free-standing source of title;²³ and, secondly, the degree to which registration is being eroded by reactionary judicial attitudes which resist any passing of title without the owner's consent.²⁴ Those issues are vital in understanding and appraising the registration system. An examination of the decision making which goes beyond studying the evolution of legal principles has the capacity to bring a different perspective to those issues.

The study

The special nature of discretionary decision making is such that only limited value can be extracted from applying doctrinal methods. Formal law has little to say here other than that the decision maker may lawfully take account of relevant considerations; analysing indicative leading cases would not convey the breadth of accumulated experience.²⁵ In this study, therefore, judgments were not examined for the purpose of identifying emergent legal rules, but were instead mined for the factors deemed relevant to the exercise of discretion.²⁶ The study was devised to investigate the exercise of discretion in practice: what factors are taken into account in correction proceedings, what weight is accorded to them, what standard of cogency they must attain. To achieve this, the judgments were subjected to ‘content analysis’²⁷ in accordance with the recognised methodological guidelines.²⁸

First, a systematic search was undertaken to unearth judgments on correction by the adjudicators²⁹ and courts.³⁰ The adjudicators’ judgments were taken from the Tribunals Judiciary website³¹ and covered the period from the establishment of the office of Adjudicator until its abolition and the transfer of its jurisdiction to the First-tier Tribunal.³² The final judgment was rendered on 4 November 2014. The court judgments were harvested from a database search from the date of enactment of the Land Registration Act 2002³³ to the same end date.³⁴ Secondly, to bring consistency into the analysis and to allow comparison by standardised units, a scheme of categories of factors was created, and the discretionary factors considered in each judgment to be relevant to the exercise of discretion were systematically recorded and assigned to those categories. The purpose was to record the relevance of factors and consequently if a specific factor was addressed and stated not to be sufficiently potent to reach the relevant threshold then it was nevertheless included in the results since it was inferentially being recognised as potentially relevant. No attempt was made to capture the weight of each factor unless it was specifically reported in the judgment.

Tabulating the factors in that way was intended to facilitate the third step. This involved reviewing the results for the prevalence, scarcity and absence of particular factors, the coalescence of particular factors in judgments, any disproportion between the factual engagement of a factor and its mention in the judgment, the coincidence of factors in particular circumstances, the correlations of factors to judgment outcomes, and the passing of legal thresholds. Using that information as the basis for further inquiry, the fourth step was to return to particular judgments for a more contextualised investigation of features that had prompted interest from the review. Each decision to return for further examination of judgments was taken to seek explanation for the following matters: the profile of the results (prevalence, distribution and variety in factors), the most extreme frequencies in results (highest and lowest tallies), associations amongst factors or between the factors and case properties (correlations or patterns), and results that were noteworthy for being out of line with expectations.³⁵ This stage was intended to impart better contextual understanding of judicial decision making than that revealed from statistics.

The study examined the factors used in three discretionary regimes: the basic discretion under the 1925 Act; the ‘exceptional circumstances’ test of the 2002 Act; and the ‘unjust not to [correct]’ test, which applies where the defendant proprietor was in possession under both Acts and which has the effect of restoring the correction power. All three discretions are apparently open as to the type of factor which the decision maker may accept as relevant and are consequently comparable in regard to the judicial selection and weighting of factors. In contrast, the ‘fraud’ and ‘lack of proper care’ tests, which apply where the defendant proprietor was in possession and which have the effect of restoring the correction power, each appoint one exclusive criterion for adjudication and do not invite the court to supply the relevant factors; unlike the ‘unjust’ test, they do not involve the same judicial role in handling discretionary factors, and this study does not examine their application.

While the three discretions under consideration share the common features noted above, there are differences between them which may influence the results. First, as between the three discretions, the thresholds vary significantly. The 2002 Act prescribes no threshold for the cogency of factors but is merely that deemed sufficient to justify correction; conversely, the ‘just’ test under the 1925 Act suggests an assessment of factors for and against correction on the ‘balance of equities’,³⁶ and it is implicitly less demanding than the ‘unjust not to’ test,³⁷ although in both cases the lack of verbal elaboration leaves much to judicial standard setting. Secondly, under the 1925 Act and the ‘unjust not to’ test, there is no legislative stipulation as to the attributes of the factors, whereas under the 2002 Act the attributes of the factors, singly or collectively, must meet the description of ‘exceptional circumstances’. Finally, the ‘unjust not to’ test serves the purpose of re-opening the door to correction, and so may not be on all fours with an actual determination whether to correct. The study takes the approach that, because of the latitude in determining relevance of factors in all three situations, there is no necessity to differentiate between the discretions for the purpose of examining the relevance of factors; in other matters, however, the three discretions are distinguished where appropriate.

There is also a change of emphasis from the 1925 Act to the 2002 Act which may influence the discretion. The 1925 Act enabled correction in a slightly broader range of enumerated cases,³⁸ and certain provisions suggested that protection of disponees might not be dependent on registration alone,³⁹ whereas the new Act takes a harder line that restricts correctable mistake,⁴⁰ reinforces title by registration⁴¹ and expands the definition of the protected proprietor in possession.⁴² It reflects a conscious policy shift towards enhancing the conclusiveness of title acquired by registration,⁴³ which might permeate the exercise of the discretions. But the movement is not all one way, for the 2002 Act also imposes the hurdle of exceptional circumstances to resist correction.

A number of limitations were inherent in the design of the research. First and foremost, there was selection bias in the data set of adjudicators’ judgments because of their preselection by judicial panel before inclusion in the online database.⁴⁴ Nevertheless, there were numerous judgments by different judicial personnel, ranging over a decade, and across a wide cross-section of disputes, which was felt to be a sufficient spread of judicial reasoning to justify drawing inferences. Secondly, there was a risk of poor processes in cataloguing the factors, which was addressed through a pilot run, cross checking and quality processes. Thirdly, during the qualitative analysis stage there was a risk that counts or patterns might have been overlooked due to the subjective nature of the triggers for further investigation, which rested on the observations and expectations of the lead researcher. The small size of the data collected and the repeated re-presentation of the data in different formats assisted in ensuring that all opportunities to reveal noteworthy aspects were taken.

The findings: results and analysis

Presentation of results

There were 52 judgments dealing with correction. The unit of assessment was the judgment, so that appeal judgments in the same litigation (of which there were six) were coded as distinct units. Each judgment was examined for certain ‘properties’, which comprised various ascertained features of the dispute rather than the judicial expressions of potential influences on discretion. Table 1 shows the case properties, most of which reflect the criteria used in the various statutory terms. The exception is the column dealing with those cases in which the claimant seeking correction had no rival claim of their own. That occurred in cases where the claimant was a concerned member of the public who sought to correct the defendant's register without advancing a claim of their own, or who was a rival claimant putting forward a rival claim that was ultimately found to be misconceived.

Table 1. Case properties

Tables 2–10 record the use of a factor in the exercise of the discretion if it was one to which the court or adjudicator adverted in reaching judgment and did not explicitly reject, even if the factor was ultimately found insufficiently compelling to alter the decision. The factors included are therefore relevant factors rather than persuasive factors.

Table 2. The unwarranted entry

Table 3. Party conduct and knowledge

Table 4. Party conduct in litigation

Table 5. Personal characteristics of a party

Table 6. Emotional or social impact on a party

Table 7. Financial impact on a party

Table 8. Impact on a person who is not a party

Table 9. Compensation

Table 10. Other

The total number of factors considered varied widely from judgment to judgment. The figures are shown in Table 11.

Table 11. Number of factors per judgment

Analysis

The numerical tabulation of the coded factors was designed to bring out features of the exercise of discretion that would prompt further inquiries. They are considered under the following subheadings.

(i) Dispersal of factors

The results show that the number of factors considered per judgment ranges from zero to 11, with the most common number of factors being four, and rapidly tailing off beyond five.⁴⁵ This was not due to the same factors cropping up repeatedly. On the contrary, the identity of the factors varied enormously: a total of 30 different factors were considered across the judgments.⁴⁶ This breadth, coupled with the infrequent usage of each factor, indicates a wide dispersal which suggests that the exercise of discretion is far from a formulaic rehearsal of a group of typical, standard issues.

This finding prompted a return to the judgments to investigate the extent of reliance on precedents, derived from a supposition that such a dispersal of factors pointed away from unimaginative reliance on prior case-law to identify factors. The judgments bore this out, with only nine looking to earlier cases in relation to the assessment of the relevance of factors.⁴⁷ Doubtlessly, this low figure reflects the limited value of precedent here, since the application of discretionary tests is not within stare decisis, a decision maker's failure to adhere to a superior court's guidance on discretion is no error of law.⁴⁸ Advocates have been admonished over citing cases as precedents on discretion,⁴⁹ but such warnings do not appear in any way to have hindered counsel's vision in identifying an impressive range of factors.

The extent of dispersal was affected by the coalescence of factors in certain sequences of interconnected events. In a commonplace tale, the former owner might become aware of the challengeable entry, alerting the new owner to the existence of a potential dispute, generating protracted argument which culminates in the new owner putting up a physical obstruction, which goads the claimant into initiating proceedings.⁵⁰ This might suggest tardiness and inaction (the delay factor), expenditure on fencing (the investment factor) in knowledge of an existing dispute (the carelessness factor). To the extent such multiple factors are causatively connected, this feature must be understood as a minor caveat to the extent of dispersal and the significance of the range in factors.

(ii) Scarcity of personal factors

The courts were willing to look beyond the litigants⁵¹ and examine public impacts,⁵² such as road safety⁵³, car parking⁵⁴ or responsibility for managing land.⁵⁵ But, in the main, the results showed that the courts were primarily concerned with the effect of the decision on the parties. One striking result was the absence of almost half of the anticipated factors. There were two tables where the factors were consistently low – ‘Personal characteristics of a party’ (Table 5) and ‘Emotional or social impact on a party’ (Table 6) – in which the constituent factors had a total of only three encounters.⁵⁶ Of those, only in one case was such a factor actually applied,⁵⁷ and the remaining two counts were in a single judgment⁵⁸ which merely raised them in obiter dicta.

The almost total absence of the personal and psychological factors deserves consideration. It is submitted that the best explanation is not due to judicial reluctance to address such matters, but rather that this is an artefact of the research method. The anticipated factors had been initially compiled with an eye to the factors applied in judicial sales of co-owned land, many of which involved homes and were dominated by the personal impacts of remedial options on family members.⁵⁹ That contextual profile is not, however, seen in the jurisdiction to correct register entries for various reasons. First, around half a dozen judgments were essentially boundary disputes where the issue did not involve rending a proprietor from all connections to the land or displacement from a community. Secondly, cases of intense personal connections are more likely to be settled in anticipation of judgment. Thirdly, the acquirer's personal connections need not be addressed explicitly when in occupation since he will fall within the statutory protection for proprietors in possession.⁶⁰ Fourthly, personal connections would not need to be addressed if the applicant is in occupation and his right is an overriding interest. Fifthly, judges may use all-encompassing labels to describe the circumstances, such as unelaborated references to ‘home’,⁶¹ without attempting to identify the specific personal connections which it connotes. For these reasons, the low rate is unlikely to be representative of either the factual incidence of such factors in disputes or the willingness to recognise these as worthy factors when their factual foundation arises.

(iii) Relative prospective impacts on parties

In contrast to the scarcity of personal factors, there was substantial judicial reliance on factors designed to inflict the least inconvenience or financial loss on the parties. This appeared to carry great importance: ‘the starting point should [be] to consider the consequences of altering or not altering the title’⁶² and ‘the court will wish to consider the effect on the relevant parties of an order for rectification or of a refusal to order rectification.’⁶³

Those statements of principle encapsulated individual factors concerning prospective effects and their relative magnitude for each party. The factor comprising ‘investment on improving land’⁶⁴ embodies the idea of expenditure that would not generate returns if the land were restored to another. Similarly, the factor of ‘costs flowing from judgment compliance’⁶⁵ took account of the remedial expense of restoring the land to its pristine state before returning it. In addition, there were allusions to losses exceeding the loss of land,⁶⁶ and five occasions referring to the factor of disruption to an existing larger estate.⁶⁷

One of the problematic influences on the discretion is the availability of indemnity. The study revealed that the parties’ potential entitlements to state compensation were an explicit factor, although mentioned in only four instances⁶⁸ out of the 34 rectification claims where indemnity could have ensued. On returning to investigate how this factor was employed in decision making, it was observed that its availability to one party would be a factor in exercising the discretion so as to withhold rectification from that party or to order rectification in favour of the other party.⁶⁹ But if the quantification of indemnity would not reflect a party's full loss then that could be a consideration in favour of that party.⁷⁰ Rectification proceedings are not, however, well suited to a forensic examination of indemnity, as the indemnity fund holder is not represented in rectification proceedings and the outcome of any indemnity claim has to be speculated upon, perhaps with the addition of arguments over contribution and recoupment. This speculation is difficult when even the basic criteria of ‘prejudice to title’ (as a precondition to indemnity) and ‘loss’ (prescribing eligibility or quantum) have generated controversy.⁷¹ Heavy reliance on indemnity as a factor in rectification proceedings is unsatisfactory when such fundamental legal questions remain outstanding.

Taken together, the factors collected under the rubric of prospective loss, including its offsetting by indemnity, are ones which point away from any judicial tendency to rely on particular case typologies or party behaviour and the temptation to construct simple, fixed rules. Rather they paint a picture of a genuinely discretionary approach which concentrates on relative hardship to the parties and on securing an equitable result in the light of the parties’ own situations, occupation, expenditure, acts of reliance and so on. This emphasis on the allocation of prospective benefits and burdens not only meets the intention of the reformers who put forward the earliest draft discretionary models,⁷² but also aligns with the doctrine of undue hardship as a bar to equitable remedies⁷³ and, furthermore, attracts policy support insofar as it embodies modern economic theory in minimising the costs flowing from the clash of rival claims.

(iv) Penetration of carelessness factors

Despite the willingness to range widely across factors in the exercise of discretion, one dominating factor was encountered in 19 of the 52 judgments: this was the highest count for any factor and related to ‘carelessness by a party’.⁷⁴ That was the search term used in identifying instances when decisions took account of any behaviour which could be characterised as failing to take the steps a reasonable person would have taken in the circumstances, whether or not the term ‘carelessness’ was used in the judgment. Carelessness was taken in the search as a broad concept and not as a legal term of art: it was not restricted to any particular class of participant, duty relationship, mode of behaviour or sphere of activity. Having collated its appearances and reviewed the judgments for possible common threads in the typology of carelessness, it was discerned that there existed distinct forms that could be used to develop three separate concepts of carelessness for analysis.

The first form was identified as the failure to act prudently in one's own interests. Although the judges recognised a person's investment in developing land as a factor against correction,⁷⁵ that inclination might evaporate if it transpired that he had been on notice of an adverse claim beforehand.⁷⁶ The implication was that the individual had knowingly omitted to safeguard his expenditure against the event of having to restore the land to another. This may be perceived as carelessness in looking after one's own interests. It is in the nature of mitigation of harm to oneself and, unlike the next two types of carelessness, does not proceed from a free-standing legal duty owed to another party.

The second concept of carelessness was failure by a prospective acquirer to act carefully during the course of his application to avoid damaging the proprietor's interests. While the mere lodging of an inaccurate application with the registry was not sufficient to attract any censure for carelessness,⁷⁷ it did where the acquirer had lodged a transfer knowing the vendor's power of attorney had been revoked,⁷⁸ where the plan misrepresented the intended boundaries and the applicant did not enlighten the registry,⁷⁹ where the applicant knew of a dispute over the title and did not warn the rival claimant of the application,⁸⁰ or where an applicant for title by adverse possession overstated his case in the supporting declaration.⁸¹

The third concept of carelessness was failure to safeguard one's own title carefully in order to avoid putative acquirers’ wasted efforts in acquiring or defending title. This idea may appear at first sight to have no place in a registration system designed to give secure title, yet the scope for protective behaviour does exist and was a factor in two judgments. In one, the claimant had not resolved an outstanding dispute and ignored the registry's invitation to object to the rival claim;⁸² in the other, the carelessness comprised a simple failure to respond to a land registry proposal to resolve the dispute.⁸³ It is no surprise that these instances are rare. Preventative action typically costs nothing more than a letter asserting rights, and carelessness by existing owners (unlike some putative acquirers) is not incentivised by any prospect of unbargained gain.

Table 3 included the factor ‘notice or knowledge of the mistake/dispute prior to acquisition’.⁸⁴ This prompted an inquiry into the nature of notice or knowledge and its relationship with carelessness. It has already been seen how several judgments treated as carelessness a party's application for registration when aware that his entry would conflict with a rival ownership and yet did not raise this with the registry or the rival,⁸⁵ but there were further judgments in which the knowledge of rival ownership at the time of acquiring was not expressed in terms that suggested failure to act with appropriate care for the owner. Those cases could nevertheless be explained as carelessness, even though no such classifying label was used in the judgment, since they indicate that the acquirer is (or ought to be) equipped with the information to foresee the risk and take steps to avert the harm to the owner. The significance is that even more cases⁸⁶ swell the statistics which record carelessness.

This particular form of the carelessness factor carries the potential to reintroduce aspects of the doctrine of notice which registration was intended to eliminate. The cases show that actual notice of a defect may be a factor, but this is relatively harmless as it does not necessarily presuppose any obligation to make reasonable inquiries and it need not add to the burdens of land transfer. The cases also show that objective standards may infiltrate carelessness, yet they come nowhere near resembling the taboo doctrine of constructive notice. While they may impose pre-emptive duties, they do not require a purchaser to go behind the veil of the vendor's registration to inquire into possible mistakes in the root of title, but relate only to other matters, such as the effectiveness of the immediate transfer form to the purchaser or the accuracy of statutory declarations. In establishing objective standards for those matters, the case-law is, at most, attributing limited weight to entry in the register as a factor in correction, and readily allowing a factor of morality and social policy to tip the scales.

In addition to the ‘exceptional circumstances’ test, the judgments also reveal carelessness factors operating in the statutory ‘unjust’ test which overcomes the immunity of a proprietor in possession. Where the proprietor ignored a registry letter proposing to register a rival, the inaction overcame his immunity from correction in two cases;⁸⁷ in a third case, it was his inexplicable failure to take elementary steps to block an unauthorised sale.⁸⁸ Carelessness extends to exacerbating effects without causing the error, as where the acquirer ‘compounded the effect’⁸⁹ of another's fraud by persisting in a purchase knowing that the seller had been duped by a prior identity theft. Furthermore, it may be noted that carelessness was also critical in restoring the jurisdiction to correct in several judgments where the protection of a proprietor in possession was defeated by the defendant's ‘lack of proper care’, despite this test not being treated as a discretion within this study. The combined mass of these various circumstances for finding carelessness indicates that it is a powerful and pervasive force in discretionary decision making and may occur without explicit description as such.

The reason for the preoccupation with carelessness in all its forms is not evident from the judgments, but it may be justified as the means to ensure that the party who could most easily have prevented the mistake is the one who suffers the loss of title, thus teaching future generations to modify their behaviour to minimise the incidence of mistakes at the cheapest cost.⁹⁰ This, however, seems to be at odds with the judgments noted earlier which emphasise the inquiry into the prospective impact of the judgment on the parties. There is a recognisable divergence of judicial foci between the ex ante penalising of prior carelessness and ex post minimising overall losses to the parties. The approaches do not have to be perceived as mutually exclusive, however, and the shortest explanation is that the two apparently coexist in a pluralist system, a conclusion that is reinforced by appeal to the implicit statutory mandate to pursue a multifactorial discretionary decision making process.

(v) Reliance on the register

Reliance on the register barely exists as a factor in correction.⁹¹ Under a regime in which the information communicated by the register itself is guaranteed, and its reversal through correction would undermine a transferee's belief, it is unexpected that it is not a crucial element in decision making. Hayton long ago predicted that the correction discretion would develop a vigorous principle of routine protection for a transferee for value – a ‘registrar's darling’ principle by extension from ‘equity's darling’.⁹² The data from the study give the impression that this has not come to pass.

There is an important doctrinal context which explains in part the dearth of reliance factors. Under the 1925 Act, correction was available against subsequent transferees by virtue of explicit statutory provision.⁹³ No equivalent was included in the 2002 Act, leading to unsettled case-law and the view, prevailing for much of the period covered by the study, that transferees were protected against adverse title clams.⁹⁴ The legislation also shields transferees against unprotected interests affecting title,⁹⁵ and for much of the period it was assumed that this extended to correction claims.⁹⁶ During the period in question, therefore, third parties were widely thought to be outside the correction power. In these circumstances it is not surprising that there are few cases considering a third party's reliance on the register.⁹⁷

Those doctrinal movements account for the limited incidence of the factor of reliance, but they do not tell the full story. Returning to the judgments, there were four occasions when jurisdiction was found to rectify against third parties who would otherwise be protected through the title and priority rules,⁹⁸ and in none of them was reliance on the register expressed to be a factor in decision making. In fact, the sole occasion when reliance was noted occurred in a case where the defendants’ own act of registering title created the mistake and they then ‘arranged their affairs and spent money in reliance on their registered title’.⁹⁹ Such disinclination to consider reliance is difficult to explain. The importance of reliance might have been taken as so obvious that it was not worth mentioning, but that is an unlikely explanation when other factors were so clearly addressed in the judgments. Another explanation would be that the court has little regard for the significance of entry in the register. That would be troubling if it reflected a judicial marginalisation of the fundamental ideal of title by registration; but it does not, as there is full discussion of that policy issue in the parts of the judgments examining the availability of correction.

The better explanation may simply stem from the fact that both parties will have relied on the register at some point. The clash arises because the former proprietor would have relied on the register as a guarantee of the security of his rights, and the latter proprietor would have relied on the register as a guarantee of the new acquisition.¹⁰⁰ Reliance does not differentiate between the parties, and judges may simply regard either party's reliance as sufficiently protected by indemnity. That is certainly a fair inference from Pinto v Lim,¹⁰¹ for example, where the sufficiency of the award of indemnity was a powerful factor in the discretion. Beyond that, reliance on the register of itself may have little role to play as a factor in discretion. While the nature of the subsequent conduct of the mistakenly registered proprietor should be taken into account, that is true whether or not the particular conduct was inspired by knowledge of the register content and a belief in its irreversibility.

(vi) Absence of factors

The results showed that 14 of the 52 judgments involved no consideration at all of any factors.¹⁰² Because the judgments had been chosen for their engagement of the correction power, this result was sufficiently striking to return to the judgments for explanation. In two of the judgments, it was recorded that counsel put forward no factors to justify exceptional circumstances. In the remaining 12 judgments, the court simply determined that, since jurisdiction to correct had been found, it followed that the power should be exercised; it was not explicitly stated whether counsel had conceded that there were no exceptional circumstances, although it can be inferred that counsel had at least not dwelt on them. The adjudicative process requires counsel to raise factors directed towards the relevant threshold, and often they did not.

In these cases, counsel presumably predicted that argument on such factors would be wasted breath.¹⁰³ That advocates make such assessments indicates a degree of predictability. There are three potential aspects for prediction: whether a putative factor will be recognised as relevant, what weight will be attached to it, and how high the court will set the relevant threshold. Returning to the judgments for enlightenment, it was found that guidance was extremely sparse. There were no general dicta to assist in determining whether a hypothetical factor would be relevant. Observations on the relative weight of factors in a given factual context were infrequent, and there were no clear statements about the general attribution of weight to factors.

There was, however, limited comment on the standard of exceptional circumstances from which a hint of judicial attitudes can be drawn. It is ‘not simply a matter of balancing the equities. There have to be exceptional circumstances.’¹⁰⁴ While seemingly trite, this is an important reminder that the discretion is not free or unguided but requires a certain level of intensity to be reached. The implication that it is set high comes from indications that it must be ‘out of the ordinary course, or unusual or special, or uncommon’.¹⁰⁵ Applying the test was said to require two stages: ‘(1) are there exceptional circumstances in this case? and (2) do those exceptional circumstances justify not ordering rectification?’¹⁰⁶ This formulation omits any elaboration of whether the exceptional circumstances refer to the type of circumstances or their severity. This has been a significant issue in other legal contexts,¹⁰⁷ where it appears that factors of a familiar and common type do not pass the test,¹⁰⁸ a formula which has incurred criticism since ‘however disastrous the consequences may be to family life, if they are of the usual kind then they cannot be relied on’.¹⁰⁹ In relation to the context of register correction, there is a slim indication pointing the other way that is found in the formula that circumstances must be ‘more than merely unfortunate or unusual’,¹¹⁰ in which the reference to ‘unfortunate’ suggests that severity might suffice as a criterion for exceptional as much as rarity – an approach that is more readily justified than a test based on frequency alone. Clearly the test should be reformed to eliminate the ambiguity. Even if the exceptional circumstances test does refer to severity, the overall impression is that the exceptional quality is pitched at a high level which is significantly outside the normal range. It is perhaps the height at which this test is set that tends to encourage the capitulation by counsel, rather than a reluctance to entertain particular types of factor.

(vii) Factors, thresholds and outcomes

A typical advantage of systematically surveying judgments is that it enables an examination of relationships between the factors and judgment outcomes, particularly those which might suggest causative effects or predictive power. Here, however, no support for any such correlations can be inferred because the small sample sizes are inadequate for regression analysis and because quantitative analysis would be meaningless where the results record only an expression of a factor's potential relevance and not its actual usage. Even when whittled down to only those judgments in which the factor was actually in play, there were still no perfect one-to-one correlations, and the engaged factors yielded such very low frequencies than it was not worthwhile to carry out tests of statistical significance.

The lack of observable correlations between particular factors and the ultimate judgment outcome is ultimately reassuring. Although advisers may find it a hindrance when predicting the outcome of litigation, it reinforces the inference that there is no habitual reliance on oversimplified rosters of preconceived key factors,¹¹¹ but instead a meaningful search for the telling factors in the instant dispute. In the light of this inference, the judgments were revisited to extract comments on the paramountcy, weight or compellingness of any particular factor. Apart from an isolated observation that possession attracts special weight,¹¹² it transpired that there were no broad assertions intended to function as guidance about the abstract strength of factors.¹¹³ Once again this supported the conclusion that no one class of factor has a predetermined weighting but its force must depend on its intensity in the given dispute, fortifying the impression that the jurisprudence is dedicated to factual responsiveness.

A more promising line of inquiry relates to the legal thresholds in discretionary decision making and how they can be tallied with the judgment outcomes as a basis for understanding how readily the judiciary might find that the tipping point has passed.¹¹⁴ The thresholds are as follows: (i) under the 1925 Act, where the defendant proprietor is not protected as a proprietor in possession, then correction follows only if it is determined that it would be ‘just’ to correct; five reached the threshold and one did not;¹¹⁵ (ii) under the 2002 Act, where the defendant proprietor is protected as a proprietor in possession, then correction may occur only if the claimant makes out the criterion of fraud or lack of proper care by the defendant or the threshold of ‘unjust not to’ correct the entry; ten reached the criterion or threshold necessary for correction¹¹⁶ while five did not. In all of the judgments where that threshold was reached, correction of the register was ordered, signifying that the exceptional circumstances test was also fulfilled; (iii) under the 2002 Act, correction must follow unless the defendant makes out the threshold of exceptional circumstances. In circumstances where the defendant proprietor was not in possession, but correction would have prejudicially affected his title, only one reached the threshold of exceptional circumstances so as to block rectification of the register,¹¹⁷ 15 failed, and one was remitted. Where correction would not have prejudicially affected the defendant proprietor's title, one reached the threshold of exceptional circumstances so as to block correction,¹¹⁸ and 13 failed to do so.

Using that conceptual division, the various thresholds were attained in 33% of all judgments.¹¹⁹ From those figures, it is necessary to reject any suggestion that the judges set them at an unattainably high standard. There is no undue judicial entrenchment of the statutory default position, but an evident willingness to accept that the identified factors may reach the threshold, once again reinforcing the picture described above of strongly fact-based justice.

Thresholds aside, one final observation stands out from the inquiry into how frequently judges ordered correction. It was ordered in 43 judgments, remitted in one, and declined in eight. Further investigation revealed the eight refusals to have aligned consistently with certain prominent features. In six of them, the claimants seeking correction had no rival claim of their own, and so the perpetuation of the mistaken entry would not have deprived the claimant of any property right.¹²⁰ In the seventh, the dispute concerned an approximate boundary line which, if corrected, would have shown the boundary with greater accuracy but again without depriving the claimant of any property entitlement.¹²¹ The eighth case was Pinto v Lim,¹²² which involved the most compelling of all scenarios, involving as it did a registered forgery followed by a transfer for value to a bona fide purchaser who had relied on the mistaken entry in the register at the time of buying, who moved in and occupied the premises as a home for himself and his wife for over 4 years, while the claimant was primarily interested in the financial value of the land and would receive state compensation.¹²³

If Pinto v Lim is removed from consideration, either as the utmost extremity or as a relic of a repealed legislative test,¹²⁴ then the remaining judgments in which correction was refused under the 2002 Act (whether the defendant proprietor was or was not in possession) all comprise scenarios in which the mistaken registration of the defendant did not cause the claimant to suffer the deprivation of a property entitlement. The two judgments in which the mistake was allowed to stand by reason of exceptional circumstances indicate the judicial approach.¹²⁵ In neither did the court roam far in identifying the relevant factors. In the first, the claimant had held no prior interest that had been prejudiced by the mistaken change to the register, and the factors relied on all sprang from the claimant's lack of title.¹²⁶ In the second, the claimant had an interest which was safeguarded by the general boundary principle,¹²⁷ and the factors relied on all sprang from the fact that correction would have no impact on the claimant's title or resolve the issue of possession.¹²⁸ The exceptional factors lie within that narrow compass and do not imply any broad judicial willingness to uphold a title acquired by mistaken registration against a former owner.

There is no other judgment where the court or adjudicator preserved a mistaken entry that had deprived a claimant of a property right. That is a remarkable datum which demands consideration. It lends itself to the inference that there is an inclination to correct mistaken entries, and that holds sway regardless of whether the particular threshold must be attained in order to enable correction or to block correction.

In itself, however, it is an insufficient basis for criticism, since the sample of judgments may be unrepresentative and it has already been consistently established that the courts pay close attention to the full range of factors so that the finding should not fuel any allegation of judicial unwillingness to take the exercise of discretion seriously. Nevertheless, the datum is a sufficient basis to raise the concern of a possible judicial tendency to correct which might be giving inadequate weight to vital issues: the status of registered entries, the statutory thresholds and the parliamentary mandate to examine the relevant factors. That concern is only reinforced by cases in which the mistaken status of the entry was expressed to be the very reason for exercising the discretion to correct.¹²⁹

The correction tendency might be explained as a natural persistence of traditional methods of adjudicating property entitlements based not on ad hoc redistribution but upon abstract predetermined events that are used as the criteria for determining entitlements. This paradigm is deeply ingrained by the concepts of title and ejectment which comprise the foundation blocks in unregistered property law.¹³⁰ For legal actors immersed in common law, it should not be surprising to see an inclination towards rule-based adjudication by reference to predictable events understood to affect title, and a disinclination to engage in overtly redistributive adjudication which looks to contexts and consequences.

The title paradigm does not, however, explain another aspect of the correction tendency, which is that it consistently prefers former owners as first in time over later acquirers who get onto the register under a mistake. To make sense of the observed tendency, some basis is required that differentiates the earlier proprietor from the later. The first in time preference cannot lie in the impropriety of depriving without consent,¹³¹ as taking property away from the later proprietor by correction would be, in legal terms, just as much a deprivation as the taking from the earlier proprietor by mistaken registration. One possible explanation is that the preference implements an effective preventative policy: that by imposing on acquirers the burden of suffering correction, they will in future exercise greater diligence in the process of acquisition. The weakness of this potential explanation is that it assumes the acquirer to be in the better position to ensure a mistaken entry is averted, but that is far from self-evident, and it may well be that owners themselves could often more easily take the necessary steps to forestall the problem, as indicated by the case-law on contributory negligence.¹³² To avoid that weakness, it is possible to suggest a broader explanation, that there may exist a judicial ethic according to which adjudication ought to focus on the choices and actions of the parties, rather than on some particular status they may enjoy which they might have had no part in procuring. Such an approach is likely to prioritise the legitimacy of the means of acquisition over the mere fact of a register entry. Explicit judicial comments occasionally confirm the marginalisation of register entries in this manner;¹³³ but inevitably this approach would eventually impede the economic policy underlying the principle of title by registration.¹³⁴

There remain important contextual observations which might colour the understanding of the correction tendency. First, it reinforces the institutional independence of the decision maker from the registry, since judgments fully review the registry action without suggesting any deference to the registry's decisions to change entries.¹³⁵ Secondly, the jurisdiction to order a change in the register hinges on the vituperative term ‘mistake’ and the affirming language of ‘correction’. Being redolent of rights, wrongs and redress, this terminology may exert a latent influence on how the desirability of correcting is perceived. Had the statute portrayed register entries as sacrosanct and their alteration in distinctly negative terms, then it is easy to imagine that the discretion would be approached very differently. If a genuinely discretionary element is to be preserved in correction applications, it is submitted that precautionary measures are required to ensure that the correction tendency does not become an entrenched reactionary preference for the first in time. This could be done by dealing with the contributing influences: by replacing the suggestive statutory language with neutral terminology; by creating a presumption against correcting; by rewording the exceptional circumstances test to depend upon factor intensity not factor frequency; by specifying more precisely the statutory thresholds; and by designating the defendant's actual reliance on a register entry as a non-exhaustive statutory factor to which a judge must advert in decision making.

Discussion

When the findings are viewed together it is possible to relate them to several broader themes which concern heuristics, syllogisms, guidance and transparency.

The decisions in this study gave clear and precise statement of the factors relied upon as influencing the exercise of discretion which were substantiated by the fact, and almost entirely devoid of rough proxies or any other imperfect but convenient methods of decision making.¹³⁶ There was no tendency to take short cuts in decision making, such as overgeneralising a category of factors without adequate attention to its constituent elements, or the setting of categories of factors based on unelaborated assumptions.¹³⁷ A solitary example of an influential factor based on unspoken assumptions occurred in the reference to ‘home’,¹³⁸ which should have been substantiated by reference to the specific inhabitant's attachments that would have been lost upon correction of the register and eviction from the home, but they are likely to have comprised a rather obvious set of relations with the property and its vicinity. In general, judges were patently willing to re-imagine afresh in each case what specific factors should be influential. This approach of avoiding simplified heuristic processes preserves rational strength in the decision making, and its costs may be deemed acceptable where it exists to determine conflicting private rights rather than merely to pursue a cost-effective, policy-led redistribution of resources.

The findings are broadly against decision making that is ‘shallow’ or ‘canalised’.¹³⁹ This is another possible manifestation of heuristic reasoning in which decision makers routinely search for the presence or absence of a narrow range of leading factors and then use them to dictate the outcome, leading to oversimplification of a party's circumstances. The correction decisions avoid that characterisation, as the study showed no predetermined weighting for particular factors and no habitual repetition of leading factors. One potential caveat, however, is the pervasive influence of carelessness. It is assumed in the judgments to play a significant role in differentiating one scenario from another, but carelessness possesses certain characteristics which should invite suspicion when applied to discretionary decision making. Carelessness is highly fact dependent, it represents an evaluative standard of great vagueness, it affords much leeway in classifying facts, and it cannot be determined afresh on appeal.¹⁴⁰ Once discovered, it has powerful force in characterising the instant case that creates a logically defensible anchor for decision. There is accordingly a temptation for decision makers to grasp too readily at carelessness as a convenient determinant. Its frequent usage raises the concern that carelessness might take a leading status which subordinates other factors and is liable to rob the decision of its fully discretionary quality.

The study found a willingness to explore the parties’ individual circumstances without adopting preconceived legal formulae. The findings revealed receptiveness to an unlimited array of factors and no appetite to fix the weight of any particular factor.¹⁴¹ This is positive in as much as it avoids the erosion of discretion by the mechanical application of abstract, syllogistic reasoning. The cases show decision makers paying close attention to the facts, identifying those that call for response, and avoiding abstraction in setting advance rules. A possible instance of backsliding occurred in cases where the correction claimant lacked a rival interest. There existed a striking trend that correction would not be ordered, lending itself to the inference that an abstract syllogism might be operating: if the claimant has no rival interest, then correction will be declined on the ground of exceptional circumstances.

One might also consider the correction tendency in this regard. In no case under the 2002 Act did the court or adjudicator preserve a mistake whose entry had deprived a claimant of a property right. At a stretch this could suggest the foundation for a property heuristic that wherever a claimant's prior rights are prejudiced by a mistake, then correction follows. Certainly, the consistent upholding of property rights stands in contrast with the rhetoric of discretion. But the mooted heuristic is not substantiated. First, the study shows that the fact of a claimant suffering no deprivation of title by the mistake is not a sufficient criterion for withholding correction.¹⁴² Secondly, the study does not provide the evidential foundation to prove the negative – that such disputes would never result in withholding correction, however compelling the factual basis for protecting the person taking under the mistake. On the current state of knowledge, the correction tendency cannot be regarded as a tacit rule.

If the correction tendency were in future to emerge as a guiding principle, invariably leading to correction wherever the claimant was deprived of rights by the mistake, its effect would be largely to withdraw the discretionary component in decision making on correction. That would leave no judicial room for discretionary manoeuvring and would force attention onto the concept of mistake as the determinant of correction. It would be apt to inspire arguments over the capacity of mistake to act as a safety valve for hard cases, as opposed to following some strict algorithm to determine its application, and would risk judicial interpretations which oscillate between predictable formalism and counteracting perceived unfairness in outcomes.¹⁴³

Despite the judicial inclination against syllogistic reasoning in this context, factual responsiveness makes for an uneasy relationship with other values. In particular, it restricts the guidance on factors available to current and future disputants. While it is clear from the judgments studied that all relevant material will be admitted and its weight will not be prejudged, there is virtually no further elaboration. An absence of guidance suppresses predictability, creates a risk of not treating like cases alike, and smacks of retrospective law making.¹⁴⁴ In other discretionary contexts, there is an expectation,¹⁴⁵ perhaps even a constitutional responsibility,¹⁴⁶ on decision makers to formulate guidance. The lack of guidance in correction cases might be conditioned by concerns about subverting the legislative intent, inflicting on the immediate litigants the costs of setting guidance for posterity, and fettering the court's discretion.¹⁴⁷ Those concerns, however, are also characteristic of other contexts where guidance is forthcoming. It is submitted that litigants in correction proceedings already have the basic guidance they need from the statutory regulation of the discretion, such as the ‘exceptional circumstances’ test, the ‘proprietor in possession’ rule and its provisos, and the known traditions of discretionary adjudication. Further regulation may be undesirable as the specific details concerning the relevance and weight of factors are inherently inapt for prescribing in advance, being infinitely variable in their intensity and incommensurable in quality. Despite the lack of specific judicial guidance, the evidence of litigants not contesting the exercise of discretion suggests that the decision making already supports at least a modicum of predictability and therefore achieves a measure of success in balancing generic guidance against the flexibility to respond to unique situations.

Finally, making explicit the principled basis for decision making is not only an issue of predictability, but also a rule of law issue concerning transparency in adjudication. The judgments show that the range of factors is fully addressed in the chain of reasoning, but the lack of detail in assessing the relative weight of factors is a particularly marked and common deficit in the transparency necessary for accountability. It is not suggested that this is complacency born of the knowledge that the discretion is effectively unappealable.¹⁴⁸ Nevertheless, the consequence is that cases are disposed on the basis of unspoken value judgments about the relative weighting. While there may be a sense that forensic argument over value hierarchies is unnecessary in a collegial discipline of shared values and would be alien to the common law process, this represents an unsolved challenge for correction proceedings and adjudicative discretions in general.

Conclusion

The foremost feature of register correction, when seen as a composite of the legislative framework and the judicial handling of discretion, is its highly discretionary quality. It has not been cut down by the emergence of any rigid, restrictive rules. There is no empty repetition of some set formula; no single factor is universally accepted as decisive; the weight of individual factors is not predetermined but responsive to the context. Shallowness and short cuts have been avoided. The eligible factors and their respective weightings have not been reduced to an abstract syllogism, save for a potential reservation when the claimant lacks a rival title, and little attention is paid to earlier precedents. The judicial component has enriched this part of the registration system by infusing it with consideration of all manner of real-world events. At the same time, there remains sufficient predictability on some occasions to enable counsel to settle in anticipation of the outcome. In addition, the statutory thresholds are seemingly applied at an appropriate level, being neither irrebuttable on the one hand, nor inconsequential on the other. These features of the jurisprudence are collectively indicative of a serious judicial commitment to the importance of responding to individual circumstances in cases of mistake.

The impression of meticulous judicial investigation across a wide range of relevant factors is, however, overshadowed by the lack of substantive effect on outcomes. At first glance the findings seem paradoxical: great pains are taken to ensure a highly fact-specific discretion, yet the observed propensity to correct implies that discretion has little significance in decision making. The latter point must not, however, be overstated as the correction tendency has little claim to represent a judicial principle. Were it to fossilise as a rule, it would demand urgent attention as a judicial negation of the statutory discretion and as a direct challenge to registration of title, downgrading its role from a means to reallocate property rights to a mere financial guarantee. But the observed tendency is currently insufficient to conclude that the rehearsal of discretionary factors is simply window dressing, and it does not yet justify more than a degree of scepticism over the responsiveness claimed for the discretion.

The study provides a view on judicial attitudes towards the significance of register entries. The case outcomes exhibit a preponderance of correction orders, transcending the various statutory thresholds, which suggests that the fact of registration alone is a weak reason for confirming title and that it is overtaken by all manner of circumstances external to the register. This view is supported by other aspects of the judgments; for example, reliance on the register is almost entirely absent from the factors considered, and non-consensual vesting by registration is occasionally given, without more, as a reason to correct. The register is not altogether disregarded, however, as entry on the register was shown to have significance when one party had relied on his own entry and the other party had never had a corresponding opportunity for reliance. In correction proceedings, the influence of a register entry may be slight, yet that does not in any sense signal defiance of the registration system. The correction clause has delegated to judges the power (at least where title would be prejudiced) to select between the property right and the financial guarantee of indemnity. Although correction decisions show a judicial inclination to take advantage of that liberty by tilting decision making heavily in favour of human factors in preference to the register, that choice of emphasis does not stray beyond the delegated power and cannot be stigmatised as a frustration of the legislative scheme.

Judicial decision making has shown extensive reliance on carelessness factors and this, too, may prompt concerns over the registration-mindedness of the judicial input. The risk here is not merely paying too slight attention to the register, but rather that the judicial approach may be directly undermining the register. This could occur because carelessness factors extend to penalising a party who presses ahead despite notice of an off-register issue casting doubt on his acquisition. It raises questions over the doctrine of notice, which registration was designed to eliminate. The study explained how judges have stopped short of inquiring into constructive notice of matters behind a transferor's register entry, and so it is decidedly premature to announce the resurrection of the equitable doctrine of notice and the judicial undermining of the register. This is, however, an important and controversial domain for the carelessness factor which, if left to grow unchecked, could conflict with a primary goal of the registration system.

The study also provides a perspective on the values which are brought into registration when judges resolve title conflicts unconstrained by statutory rules. One cluster of factors points towards the outcome which creates the least prospective suffering by the parties. It corresponds to an objective of limiting the future hardship or cost to the litigants which maps onto a norm of efficient redistribution after a title conflict has arisen. Judicial references to public interest indicate that this value extends beyond the litigants’ positions to the wider costs falling on society. Overtaking those influences, however, is an even more powerful current of factors which dwell on past carelessness. They encapsulate two judicial values: punishing those who squander their opportunities to avoid title conflict, and establishing guidance on standards to instil future precautionary practices. The combination of factors, implying ex post efficiency, moral desert and ex ante care, indicates a panoply of values just as broad as that found in common law property, equally lacking in hierarchy,¹⁴⁹ and curtailed (as illustrated by the approach to constructive notice) only as needed to avert direct conflict with the rules of registration.