INTRODUCTION
Administrative lawyers in Kenya are currently showing considerable interest in section 33 of the Constitution of the Republic of South Africa, 1996 and the South African experience of administrative justice under that provision.Footnote 1 The reason is that, as with many other provisions of Kenya's transformative Constitution of 2010 (2010 Constitution), the article 47 rights to fair administrative action were modelled along South African lines. The wording of the rights in article 47 is almost identical to that of the rights to just administrative action in section 33 of South Africa's Constitution, and article 47(3) also mandated the enactment of legislation to give effect to most of those constitutional rights. In view of these similarities is it is not surprising that the drafting of the Kenyan Fair Administrative Action Act 4 of 2015 (FAAA) was strongly influenced by the equivalent South African legislation, the Promotion of Administrative Justice Act 3 of 2000 (PAJA).
More than two decades have passed since South Africans first acquired rights to administrative justice, and the PAJA has been in operation for well over 15 years.Footnote 2 South Africa can thus be regarded as a sort of laboratory for Kenyan administrative justice, and Kenya is in a position to learn from South Africa's experience and mistakes. Indeed, such benefits are already evident. In the drafting of its own legislation, Kenya has been able to avoid many of the problems South Africa's courts encountered in relation to the PAJA; duly forewarned, the Kenyan courts will surely be able to steer clear of others.
The aim of this article is to highlight some of the South African experience in relation to section 33 and the PAJA, particularly those aspects likely to be of interest to a Kenyan legal audience that is starting to grapple with the FAAA. Perhaps the most important lesson South Africa can offer Kenya at this point is to take its constitutionally mandated legislation seriously, and to avoid the example set by the South African courts in making their own administrative justice statute almost redundant.
THE PROVENANCE AND INFLUENCE OF SECTION 33
As is well known, in 1994 the Republic of South Africa successfully made the transition between an undemocratic system characterized by white minority rule and institutionalized racial segregation to a system of constitutional democracy. At the same time, the republic exchanged parliamentary sovereignty for constitutional supremacy. This dramatic change was initially achieved by way of an interim or transitional constitution (Interim Constitution)Footnote 3 with a justiciable Bill of Rights, including (in section 24) rights to administrative justice.
Administrative law had played a dual and paradoxical role in the legal system of pre-democratic South Africa. Administrative law review was the closest thing South African law had to a bill of rights,Footnote 4 but this branch of the law was simultaneously a tool for implementing and enforcing the infamous system of apartheid.Footnote 5 The section 24 rights thus represented an emphatic rejection of South Africa's painful history of administrative injustice and an attempt to “put South Africa at the frontiers of the search for a culture of justification”.Footnote 6
The most immediate inspiration for section 24 was article 18 of the Constitution of the Republic of Namibia, which among other things required administrative officials to act “fairly and reasonably”. As South-West Africa, the territory had been a German protectorate from 1884 but was effectively possessed by South Africa from 1915 to 1990, the year in which Namibia achieved independence.Footnote 7 The two countries had similar legacies of administrative injustice to address, particularly since South Africa had notoriously introduced the apartheid system into South-West Africa.
In turn, the South African provision influenced the inclusion of similar rights in the Constitutions of Malawi and Uganda in 1994 and 1995 respectively,Footnote 8 and seemed to inspire a worldwide trend in this regard.Footnote 9 Recent examples of the trend are the inclusion of rights to administrative justice in article 16 of the Constitution of the Republic of Fiji and section 68 of the Constitution of Zimbabwe, both of which date from 2013.
South Africa's Interim Constitution, a deliberately transitional document, was ultimately replaced by the “final” or 1996 Constitution (1996 Constitution).Footnote 10 It, too, included rights to “just administrative action”, set out in article 33 in remarkably generous terms:
“(1) Everyone has a right to administrative action that is lawful, reasonable and procedurally fair.
(2) Everyone whose rights have been adversely affected by administrative action has the right to be given written reasons.
(3) National legislation must be enacted to give effect to these rights, and must -
(a) provide for the review of administrative action by a court or, where appropriate, an independent and impartial tribunal;
(b) impose a duty on the state to give effect to the rights in subsections (1) and (2); and
(c) promote an efficient administration.”
This provision served as the chief model for Kenya's constitutional rights to “fair administrative action” (as it also did for Zimbabwe's section 68 rights). This is not surprising, for it is well known that Kenya's transformativeFootnote 11 2010 Constitution also borrowed heavily from South Africa's 1996 Constitution more generally, and Kenya certainly needed to address a long history of administrative injustice, suffered not only during the period of British colonial rule but also after independence in 1963.Footnote 12 Article 47 of the Kenyan Constitution on “fair administrative action” reads:
“(1) Every person has a right to administrative action that is expeditious, efficient, lawful, reasonable and procedurally fair.
(2) If a right or fundamental freedom of a person has been or is likely to be adversely affected by administrative action, the person has the right to be given written reasons for the action.
(3) Parliament shall enact legislation to give effect to the rights in clause (1) and that legislation shall
(a) provide for the review of administrative action by a court or, if appropriate, an independent and impartial tribunal; and
(b) promote efficient administration.”
The wording of the two constitutional provisions is strikingly similar and the differences between them are relatively slight.Footnote 13 The heading of article 47 opts for “fair” rather than “just” administrative action. To “lawful, reasonable and procedurally fair”, Kenya has added the adjectives “expeditious and efficient”. Kenya's right to reasons is slightly more generous than South Africa's, as it hinges on the existence of a right that has been or is likely to be adversely affected by administrative action. Furthermore, article 47(3) follows section 33(3) in mandating legislation to give effect to the rights (albeit only those listed in article 47(1)) and in requiring the legislation to provide for review by a court or tribunal, where appropriate, and promote efficient administration. Article 47(3) merely omits to mandate an express duty on the state to give effect to the rights, which arguably goes without saying in any event.
THE ENACTMENT AND SCOPE OF THE MANDATED LEGISLATION
In South Africa's case, an important reason for the mandate in section 33(3) was the broad sweep of the wording in section 33(1). The Constitutional Assembly worried about the burden on the public administration of having to comply with such broad and general principles of good governance, for South African administrators were not accustomed to acting reasonably or fairly in all cases. At common law, duties such as this had only applied to a small percentage of decisions, and there had never been a general duty to give reasons. So there was a definite interest not merely in fleshing out the section 33 rights but also in narrowing the scope of their application.
In terms of item 23 of schedule 6 to the 1996 Constitution, the envisaged law had to be enacted within three years of the 1996 Constitution's coming into force, failing which section 33(3) would lapse and the sweeping rights would become directly applicable.Footnote 14 Despite this incentive, the Department of Justice left the drafting of the PAJA very late and it was only at the very end of 1998 that it gave the job to the South African Law Reform Commission. That body then appointed a project committee to perform the investigative work and prepare a report. It was a tall order, but with the generous assistance it received from Professor (now Sir) Jeffrey Jowell QC and a team of foreign experts, by August 1999 the committee managed to produce a report with draft legislation appended to it.Footnote 15
That legislation was subsequently enacted as the PAJA, only just before the deadline and not without some significant last-minute alterations to the draft bill. Unfortunately, the main motive behind the changes seemed to be to save the government trouble by reducing the scope of the act or its application. Thus the definition of “administrative action”, the gateway to the act, became more elaborate than it had originally been, ensuring that only a small proportion of official action would qualify. A provision was added (in section 2) to allow for ministerial exemptions from the act as well as “permissions to vary” requirements relating to procedural fairness and reason-giving. Also, many reforms envisaged in the draft bill were abandoned, evidently on the basis that they were too burdensome or expensive.Footnote 16
Like the PAJA, Kenya's FAAA also had to be finalized in something of a hurry. This was at least partly the result of a complication that did not arise in South Africa's case: a lack of consensus as to the need for the legislation. The deadline set by article 261 of the 2010 Constitution for the enactment of the mandated legislation (and numerous other statutes) was 27 August 2014, four years from the date of the constitution's promulgation. However, in August 2014 Parliament was not ready to proceed with the draft legislation and the National Assembly had to extend the deadline to 27 May 2015. An important factor was the opinion of the chairperson of the Constitutional Implementation Oversight Committee that the article 47 mandate had already been fulfilled by existing legislation, particularly the Commission on Administrative Justice Act 23 of 2011.Footnote 17 A similar view had been expressed by the Commission on Administrative Justice, which strongly doubted the need for “stand-alone” legislation to flesh out article 47 and believed that existing legislation, suitably amended, would meet the case.Footnote 18 The debate was ultimately resolved in favour of the new legislation, however, and the FAAA was assented to just before the expiry of the extended deadline. In the absence of a more specific commencement date, it entered into force on 17 June 2015, 14 days after its publication in the Gazette.Footnote 19
Currie has accurately described the PAJA as an instance of “codification-reform”: a legislative restatement of the common law principles of administrative law with a view to their reform.Footnote 20 As he also points out, the act did not aspire to be an exhaustive statement of those principles and tended towards legislative minimalism.Footnote 21 The FAAA seems very similar in its scope and ambitions, and covers much the same ground as the PAJA. Both statutes start with the meaning of “administrative action” and other definitions; both address procedural fairness in respect of actions affecting individuals, groups or the general public; and both flesh out the right to reasons (even though that particular right did not form part of the article 47(3) mandate). Both statutes feature a detailed list of grounds of review, insist that review applications be brought promptly but only after internal remedies have been exhausted, and confer similar remedial powers on the courts in question. Both allow for regulations to be made.Footnote 22 The two statutes are also of similar length: the PAJA as amended (chiefly by the addition of sections 9A and 10A) consists of 13 sections and the FAAA has 14.
One apparently significant point of difference is that section 12 of the FAAA expressly preserves the common law,Footnote 23 while the PAJA does not. In practice, however, this difference may not be as great as it may seem: the South African courts accept the continuing relevance of the common law subject to the constitution and the PAJA, while the Kenyan courts seem to accept that the constitution and the FAAA take precedence over the common law. To be more specific, the Constitutional Court of South Africa has held that the common law (to the extent that it is not unconstitutional) is still capable of informing the provisions of both the PAJA and section 33;Footnote 24 meanwhile, in a recent judgment, the Kenyan Court of Appeal echoed the words of the South African Constitutional Court when it observed that, “the extent to which the common law principles remain relevant to administrative review will have to be developed on a case-by-case basis as the courts interpret and apply the provisions of the Fair Administrative Action Act and the Constitution”.Footnote 25 The important relationship between the mandated legislation and the common law is discussed below.
CONSTITUTIONALIZATION OF ADMINISTRATIVE JUSTICE: THE GOOD AND THE BAD
In South Africa the entrenchment of constitutional rights to administrative justice brought two great advantagesFootnote 26 that Kenya should also enjoy. First, it provided a constitutional justification and explanation for the practice of judicial review, which until then had relied largely on the ultra vires doctrine: at best an incomplete explanation of the power of review.Footnote 27 South Africans found conceptual relief in section 33 and its place within a justiciable bill of rights and a supreme constitution;Footnote 28 Kenyans will no doubt find the same relief in the firm ground of article 47 of the 2010 Constitution. As acknowledged in its article 2, the 2010 Constitution is “the supreme law of the Republic and binds all persons and State organs at both levels of government”.
Secondly, the rights in section 33 of the 1996 Constitution guarantee the practice of judicial review. Under section 36, limitations or infringements of constitutional rights can be tolerated only if they are “reasonable and justifiable in an open and democratic society based on human dignity, equality and freedom” and if they are effected by means of law of general application. Thus administrative conduct can never justify itself: it has to be authorized by law. Furthermore, ouster clauses and other attempts to interfere with the courts’ review powers infringe not only section 33 but also section 34 of the 1996 Constitution (the right of access to court) and would need to be supported by “very powerful considerations” in order to pass the section 36 test.Footnote 29
The Kenyan limitation provision (article 24) is almost identical to South Africa's section 36, and should bring the same advantages in the context of article 47 (and article 48, the Kenyan right of access to justice). As Gichuhi has observed, judicial review is now “embedded in the Constitution”, and the article 47 rights cannot be taken away by legislation, common law or in any other manner,Footnote 30 at least not without proper justification.
These are considerable advantages. In South Africa's case, however, constitutionalizing administrative law also helped create two problems that are described below. Luckily Kenya has managed to avoid the second one already and it is to be hoped that the Kenyan courts will steer clear of the first as well.
The 1996 Constitution and the common law
The first difficulty emerged during the period before the PAJA came into existence, indeed before its enactment was even mandated. It was the problem of working out the relationship between what were then the two most obvious pathways to judicial review: the constitutional rights to administrative justice and the existing, and abundant, common law principles of administrative law. At the time it was widely thought that there were two separate realms of administrative law, one constitutional and the other governed by the common law, and that a litigant had a free choice to pursue judicial review of administrative action via either of these distinct routes.
This scheme was all the more plausible because it accounted nicely for the existence of what were then South Africa's twin highest courts: the new Constitutional Court, which had been created as the highest court in constitutional matters,Footnote 31 and the existing Appellate Division (renamed the Supreme Court of Appeal by the 1996 Constitution), the highest court in other matters.Footnote 32 The division of labour between the two courts tended to confirm the possibility of distinguishing rigidly between constitutional and other, presumably non-constitutional, matters. Indeed, under the Interim Constitution the Appellate Division had had no jurisdiction whatsoever over constitutional matters,Footnote 33 so the theory of two completely separate realms was essential to that court's raison d'ȇtre.
This “silo” theory of administrative law held sway for a number of years. It received explicit confirmation in a 1999 judgment in which the Supreme Court of Appeal distinguished between “constitutional review”, governed by the 1996 Constitution, and “judicial review under the common law”, which answered merely to the empowering statute and the behests of natural justice.Footnote 34 However, while this conveniently preserved a sphere of sovereignty for the court in question, it failed properly to acknowledge the supremacy of the constitution. A few months later, in Pharmaceutical Manufacturers Association, the Constitutional Court pointed out the implications for administrative law of that crucial supremacy principle. Chaskalson P held for a unanimous court:Footnote 35
“I cannot accept this contention, which treats the common law as a body of law separate and distinct from the Constitution. There are not two systems of law, each dealing with the same subject-matter, each having similar requirements, each operating in its own field with its own highest Court. There is only one system of law. It is shaped by the Constitution which is the supreme law, and all law, including the common law, derives its force from the Constitution and is subject to constitutional control.”
As with other great truths, this proposition seemed blindingly obvious once it had been stated. It is worth stressing, however, that the dictum does not imply that the common law has been jettisoned. The common law not only continues to exist but is capable of informing the interpretation of both the PAJA and section 33;Footnote 36 indeed the common law remains a valid pathway to judicial review for exercises of private power that are not reviewable under the PAJA. The point is rather that the common law is not insulated from the constitution, but is ruled by it and must be interpreted in line with it. Furthermore, the common law does not compete with review under the PAJA, the statute that gives effect to the rights in section 33. Accordingly, where the PAJA is applicable, it is not open to a litigant to sidestep the statute by bringing a review application on common law, instead of PAJA, grounds.Footnote 37
In Kenya there seems to be some judicial support for the idea that administrative law is to be found in separate constitutional and non-constitutional silos. At least one High Court judge, while quoting extensively from the judgment in Pharmaceutical Manufacturers Association, has suggested that under the Law Reform Act and order 53 of the Civil Procedure Rules of 2010 there remains a “purely judicial review application” quite separate from the constitution.Footnote 38 More recently the same judge, Odunga J, described the parameters of administrative law review in purely common law terms, as if untouched by the constitution in general or by article 47 in particular.Footnote 39 By contrast, in a 2014 judgment the Supreme Court of Kenya saw judicial review as arising out of articles 23(3) and 165(3) of the 2010 Constitution rather than the common law, and regarded common law principles relating to legitimate expectations as having been subsumed under the constitution, in line with what was said in Pharmaceutical Manufacturers Association. Footnote 40 In the more recent case of Suchan Investment, the Kenyan Court of Appeal again echoed the words of the South African Constitutional Court when it stated:
“The common law principles of administrative review have now been subsumed under Article 47 [sic] Constitution and Section 7 of the Fair Administrative Action Act. In this regard, there are no two systems of law regulating administrative action - the common law and the Constitution - but only one system grounded in the Constitution. The courts’ power to statutorily review administrative action no longer flows directly from the common law, but inter alia from the constitutionally mandated Fair Administrative Action Act and Article 47 of the Constitution.”Footnote 41
This approach is consonant with constitutional supremacy, whereas the notion of a realm of administrative law untouched by the constitution clearly is not. Worryingly, however, it seems that the concept of a single system of law continues to be questioned by some. Some Kenyan judges evidently continue to entertain judicial review applications “simpliciter”,Footnote 42 notwithstanding the new basis for judicial review established by the 2010 Constitution and the FAAA. Kenyan commentators have also recently lamented that the conduct of judicial review is still deeply rooted in the common lawFootnote 43 and have complained of “routine citation and over-reliance on common law precedents”.Footnote 44 As Professor Gathii has warned, the Kenyan judiciary will have to guard against the development of a “two-tracked system of judicial review”, one governed by the constitution and one (purportedly) governed by the common law.Footnote 45
The concept of administrative action
In South Africa's case, a more persistent disadvantage of the constitutional rights was the prominence acquired by the concept of “administrative action”. Although unknown and evidently unnecessary in the pre-democratic era, this concept now serves to demarcate the administrative justice rights. Since neither the interim nor the 1996 Constitution provided any definition of the term, it was up to the courts to give meaning to it. Soon after 1994 the Constitutional Court was presented with a series of opportunities to describe and define the area of administrative action. In effect, it used these to narrow down the concept, mainly by distinguishing it from legislative, executive and judicial action.Footnote 46 As the author has explained elsewhere,Footnote 47 this narrowing down made sense in the light of the wide range of rights protected by the democratic constitution and the consequent shrinking of the workload of administrative law. The problem, however, was that in the process of definition the concept of administrative action became the focus of South African administrative law, its centre of gravity.Footnote 48 As the Constitutional Court admitted, giving meaning to the concept was no simple matter but involved the drawing of “difficult boundaries”.Footnote 49 That difficulty was indeed shared by all who worked with the demarcation, for it was hard for practitioners to predict when action would be regarded as administrative and when it would not, hard for administrators to grasp the concept and hard for academics to make sense of it and explain it to others.
But for the PAJA, the concept of administrative action would in all probability have become less prominent over time as the case law on the topic gradually accumulated and the law became more certain and more predictable. However, that was not to be. The PAJA's parsimonious and elaborate definition of administrative action ensured that the concept remained at the forefront of South African administrative law, and indeed became the single greatest disadvantage of codification in this area.
By contrast, Kenya has opted for a considerably wider and simpler statutory definition whose effects are likely to be concomitantly less deleterious. However the provision will still require judicial interpretation, particularly on account of its breadth, and the courts will still have to guard against the distracting effects of the concept. This theme is revisited in the next section, which considers the PAJA and the FAAA in more detail.
THE GOOD AND THE BAD OF THE PAJA, AND SOME PITFALLS AVOIDED BY THE FAAA
There is much to be said for having a statute like the PAJA or the FAAA, as opposed to relying on the murkier common law, as codifying the procedures for review, the grounds of review and the remedies available on review is very likely to advance the clarity and accessibility of administrative law. These were certainly aims that the South African Law Reform Commission and its project committee had in mind when producing the proposed administrative justice legislation, even if they were not the primary objects behind the constitutional mandate in section 33(3). From the Law Commission's point of view, codification presented an excellent opportunity to engage in some reform of substantive principles of South African administrative law.
An especially significant area of reform concerns the PAJA provisions in sections 3, 4 and 5 setting out the elements of procedural fairness and spelling out the right to reasons. Commentators seem to agree that, supplemented by the Regulations on Fair Administrative Procedures,Footnote 50 these three provisions are gradually helping to clarify the law, educate administrators and spread a culture of administrative justice in South Africa.Footnote 51 Their counterparts in the FAAA are sections 4, 5 and 6.
Section 4 of the FAAA is comparable to section 3 of the PAJA, which relates to procedural fairness in individual cases and sets out the content of fairness in that context. Section 5 of the FAAA, in turn, resonates with section 4 of the PAJA. It concerns procedural fairness when proposed administrative action is likely “materially and adversely” to affect the legal rights or interests of a group or of the general public.
Section 6 of the FAAA would seem to be the counterpart of section 5 of the PAJA, but is more widely conceived: instead of concerning itself merely with reasons for administrative action (which are also referred to in section 4(2) of the FAAA), section 6 entitles affected persons to “such information as may be necessary to facilitate his or her application for an appeal or review”, including but not limited to reasons for the action. It is encouraging to see in section 6(4) of the FAAA a salutary presumption similar to that in section 5(3) of the PAJA, that action unsupported by reasons was taken for no good reason. Also impressive is that section 6(3) of the FAAA insists that reasons be given within 30 days of a request rather than the longer period of 90 days stipulated in section 5 of the PAJA.
It is also encouraging that the FAAA eschews the over-lavish provision made in the PAJA for “departures” from procedural fairness, and confines itself to a single instance: under section 6(5), an administrator may depart from the requirement to give adequate reasons “if it is reasonable and justifiable in the circumstances” and must inform the requester accordingly. However, one commentator has suggested that this provision violates articles 24 and 35 of Kenya's 2010 Constitution (the limitation clause and the right of access to information) as well as amounting to an unconstitutional delegation of unguided discretionary power. Unlike the PAJA, the FAAA fails to provide appropriate factors to guide the administrator in exercising the power to depart from the requirement.Footnote 52 As South Africa's highest court has explained,Footnote 53 such a lack of guidance increases the risk that rights will be violated by the administrator and is thus at odds with the state's duty to “respect, protect, promote and fulfil” constitutional rights and fundamental freedoms: a duty placed equally on the Kenyan state by article 21(1) of the 2010 Constitution.
Section 6 of the PAJA undertook considerable reform of the common law. For example, the list of grounds of review in that section includes a rather thorough ground of rationality, which has certainly proved useful in South AfricaFootnote 54 and which Kenya has sensibly borrowed (section 7(2)(i) of the FAAA). Section 6(2)(e)(ii) of the PAJA contains another reform, which makes it clear that ulterior “motive” and not merely “purpose” is a ground of review: a proposition that was by no means clear under common law.Footnote 55 Again, this also features as a ground in the Kenyan legislation (section 7(2)(e) of the FAAA).
Another PAJA innovation is section 6(3) read with section 6(2)(g). These provisions amount to a ground of unreasonable delay in taking a decision, which is linked to special remedies set out in section 8(2) of the PAJA. The ground of review was not altogether new to South African law, but its existence and status at common law were not as certain as they are now. Kenya now has a similar ground in section 7(2)(j) and 7(3) of the FAAA and a comparable remedial mechanism in section 11(2). These should help to discourage the problem of administrative foot-dragging or “bureaucratic stonewalling”.Footnote 56
By no means all provisions of the PAJA were welcomed, however, and particular criticism has been directed at some of the provisions that the legislature introduced at the last minute. Below are a few examples. It is reassuring to see that, in most instances, Kenya has managed to avoid making the same mistakes.
• Section 2 of the PAJA has been criticized for contemplating ministerial exemptions and variations from sections 3, 4 and 5 over and above the individualized “departure” provisions. These loopholes have always seemed entirely unnecessary given the inherently and deliberately flexible nature of the sections in question, and the author is not aware that they have ever been used. Such a provision was wisely omitted from the Kenyan act.
• Section 4 of the PAJA (concerning procedural fairness in public cases) has been criticized for its optional nature, as decisions made under it were deliberately excluded from the definition of administrative action in section 1(ii).Footnote 57 By contrast, the Kenyan provision (section 5 of the FAAA) appears to be compulsory and is therefore more likely to be taken seriously by administrators.
• In the grounds of review, section 6(2)(h) of the PAJA was criticized for making no mention of proportionality and instead embracing Wednesbury unreasonableness,Footnote 58 which is notoriously unpopular for setting too high a threshold as well as for its tautologous and essentially unhelpful nature. These are two further pitfalls nimbly avoided by the Kenyan act. Section 7(2)(k) and (l) of the FAAA respectively provide for review of an action or decision that is simply “unreasonable” or on the ground that “the administrative action or decision is not proportionate to the interests or rights affected”. Thus the Kenyan courts should not need to engage in the sort of rescue mission that the South African Constitutional Court undertook in relation to section 6(2)(h).Footnote 59
• Section 7(1) of the PAJA caused widespread dismay by introducing a six month time limit for bringing applications for judicial review. The provision thus displaced the more flexible common law principle relating to delay, that applications must be brought within a reasonable time. The time limit received endorsement from the Constitutional Court a few years ago,Footnote 60 but remains one of the reasons why litigants prefer to avoid the PAJA if at all possible (a theme taken up again below). The Kenyan act avoided this unpopular move and opted for “without unreasonable delay” instead of a specific period of time (section 9(1) of the FAAA). This seems appropriate in a jurisdiction that is doing its best to increase access to the courts.Footnote 61 However, there is scope for confusion with order 53, which lays down specific time limits for common law remedies, and with the procedural rules for constitutional litigation,Footnote 62 which allow constitutional petitions to be filed at any time. The Kenyan courts will have to clarify the relationship between section 9(1) and these pre-existing provisions, particularly since the FAAA makes no reference to them.
• Section 7(2) of the PAJA is widely thought to be overly demanding in its requirement that internal remedies be exhausted before review is sought, though this provision, too, was warmly supported by the Constitutional Court a few years ago.Footnote 63 Interestingly, Kenya has here followed rather than avoided South Africa's example. Section 9(2) of the FAAA stands in sharp contrast to the general principle that the availability of alternative remedies is not by itself a bar to the grant of a judicial review order.Footnote 64 The FAAA provision may well be thought too strict, especially the sweeping reference to “all remedies available under any other written law”, and it will have to be interpreted restrictively in order to avoid injustice.
These are some of the PAJA provisions that have attracted criticism. However, there is little doubt that the most problematic and least admired part of the PAJA is its definition of “administrative action”.
The definition of administrative action
As already noted, shortly before the PAJA was enacted, the South African Parliament did all it could to narrow its scope. The result of its efforts is a “palisade of qualifications”,Footnote 65 all of which have to be satisfied. In order to fall within the PAJA, the power or function in question must be of a public nature and must take the form of a “decision” or “failure to take a decision”. The concept of a “decision” is defined along Australian linesFootnote 66 so as to bring in two further requirements: that the decision be “of an administrative nature” and made “under an empowering provision”, a term that is itself further defined.Footnote 67 The decision must not only affect rights adversely but must also have a “direct, external legal effect”,Footnote 68 a phrase borrowed rather arbitrarily and at the last minute from German federal law.Footnote 69 The purpose of this qualification is seemingly to rule out decisions that are not final, as well as those that are internal to the administration, while the idea of a “legal” effect seems to restate the requirement that rights be affected. Notably, these last two requirements (and not only these) have no resonance with the meaning attributed to “administrative action” in the pre-PAJA jurisprudence dealing with constitutional rights.Footnote 70 There are also nine particular exclusions from the definition,Footnote 71 some of which attempt to take into account that very jurisprudence.
The upshot of all this is a gateway that is unnecessarily complicated and unfriendly to its users, out of step with the “constitutional” meaning of administrative action, and far too narrow.Footnote 72 Very few actions actually qualify as administrative action under the PAJA; rather, very few would have qualified if the courts had not intervened and if the various elements of the definition had instead been taken literally, particularly the notion of “rights” being adversely affected.Footnote 73 If that had happened, the PAJA might have been a complete failure. As things are, the courts have been obliged to waste a great deal of time and energy on the administrative action inquiry. In the early years of the PAJA, what was essentially a threshold issue tended to become the focus of almost every administrative law matter, often at the expense of substance. In effect, cases that ought to have been about reasonableness or procedural fairness became cases about administrative action: cases about the meaning of concepts such as “rights” and “direct, external legal effect”.Footnote 74
Another problem has been the sheer difficulty of applying all these qualifications and making a diagnosis under the PAJA definition. It is telling that, a decade ago, the Constitutional Court was unable to decide whether regulations qualified as administrative action.Footnote 75 A few years ago the same court was notably reluctant to decide whether an exercise of the president's constitutional power to pardon offenders amounted to administrative action.Footnote 76 (In fact, as explained below, the court ended up avoiding the inquiry altogether.)
The irony is that the deliberate narrowness and complexity of the PAJA definition failed to achieve the aim of reducing the legal burden on administrators. One could have predicted this, since it is fairly obvious that constitutional democracy abhors an accountability vacuum. In practice the courts responded to the definition in the PAJA by doing two things. One was to interpret elements of the definition rather liberally. This trend started in earnest in 2005, when the Supreme Court of Appeal speculated that, together with “direct, external legal effect”, the requirement of adversely affected rights probably denoted merely the capacity to affect legal rights.Footnote 77 This dictum, although obiter, had a very powerful effect on subsequent jurisprudence. For example, an applicant for a job or a promotion in the public sector clearly does not have a right to the job or promotion sought, but the courts held that such decisions constituted administrative action because they clearly had the capacity to affect legal rights.Footnote 78 Similarly, the requirement of “direct, external legal effect” has not prevented a court from diagnosing as administrative action a decision merely to recommend the suspension of certain transport licences.Footnote 79
Another response of the courts, and one that revealed itself considerably earlier, was to find a way to compensate for the possible non-applicability of the constitutional rights and, later, of the PAJA, for the latter is supposed to be the main highway to judicial review.Footnote 80 This is described below.
In the light of all these complications, Kenya is to be congratulated on its comparatively simple and inclusive definition of administration.Footnote 81 Section 2 of the FAAA provides:
“[i]n this Act, unless the context otherwise requires –
‘administrative action’ includes –
(i) the powers, functions and duties exercised by authorities or quasi-judicial tribunals; or
(ii) any act, omission or decision of any person, body, or authority that affects the legal rights or interests of any person to whom such action relates.”
Part (ii) of the definition accords with section 3 of the FAAA, which makes the act applicable to “all state and non-state agencies”, including any person exercising administrative authority, performing a judicial or quasi-judicial function or “whose action, omission or decision affects the legal rights or interests of any person to whom such action, omission or decision relates”. In line with this formulation, administrative action extends beyond the acts and omissions of government agencies to those of private administrators. While that is entirely appropriate in the era of outsourcing, the wording of (ii) seems overly broad: it casts the net of judicial review so widely as to include private conduct that may be entirely unconnected with administrative activity. Together, the definition and section 3 would seem to make almost anyone's conduct subject to judicial review under the FAAA for, as Dudley has pointed out, it is difficult to conceive of an act or omission that does not affect a legal right or interest of a person in some manner.Footnote 82
A solution proposed by Dudley is to interpret “legal rights or interests” to mean rights or interests derived from the Bill of Rights.Footnote 83 He reasons that this would resonate with the language of article 47(2) and would effectively reserve judicial review for the vindication of constitutional rights. However, that solution would not necessarily serve to prevent the administrative law review of a range of private disputes having nothing to do with administration. A more effective approach might be to read in a qualification such as “of an administrative nature” so as to limit the range of the acts, omissions and decisions envisaged by both provisions of the FAAA, thus preventing overreach without adding undue complexity or rigidity. A bonus is that South African jurisprudence on the meaning of “of an administrative nature” is conveniently available to be drawn on (or rejected) by the Kenyan courts.Footnote 84
The growth of the principle of legality as a general alternative pathway to review
The story of the growth of the principle of legality should be of interest to a Kenyan audience, for it shows how something resembling Gathii's “two-tracked system of judicial review”Footnote 85 developed in South Africa, and how what was intended to be the secondary track has come almost to dominate the system. The situation in Kenya is not identical, of course, for the main tension there is evidently between constitutional petition and common law review. However, the principle is the same and the South African story should add force to Gathii's warning.
As already indicated, generous interpretation of the statutory definition of administrative action was not the only strategy adopted by the South African courts. In fact, the problem of what to do about non-administrative action predated the PAJA by some years, so the courts’ priority was to find another reliable pathway for the judicial review of conduct that might not amount to administrative action. While other pathways to review already existed, they were of a specific nature. Special statutory review operated whenever the legislature made specific provision for judicial review, either as an alternative to ordinary review or to the exclusion of such review; common-law review was confined to exercises of private power such as disciplinary action by private colleges or religious institutions.Footnote 86 The Constitutional Court appreciated early on that it would have to develop a general safety net for exercises of public power that deserved review on at least some administrative law grounds. The problem became more acute with the appearance of the PAJA, whose narrow definition only enlarged the realm of non-administrative action.
In Fedsure in 1998, the Constitutional Court identified the constitutional principle of legality as a sort of backstop for non-administrative action.Footnote 87 It described the principle as an aspect of the rule of law, a concept that was implicit in the Interim Constitution and later became an explicit founding value in section 1(c) of the 1996 Constitution. Crucially, the principle was held to apply to all exercises of public power; it was subsequently developed, mainly by the Constitutional Court, so that it came to replicate much of the content of ordinary administrative law in the form of section 33 and the PAJA.
This development was initially very rapid. By the year 2000, it was clear that any person or body exercising public power would have to act with lawful authorityFootnote 88 and within its jurisdiction.Footnote 89 Even more significantly, the court had also held that the principle of legality requires a minimum level of rationality for all exercises of public power.Footnote 90 So an adequate safety net was in place even before the PAJA came into operation. To that extent it mattered much less whether action qualified as administrative action.
In a remarkable development in 2003, material mistake of (non-jurisdictional) fact was recognized as a ground of review under the principle of legality and possibly also under the PAJA.Footnote 91 The next major development came in Albutt in 2010,Footnote 92 where it was held that non-administrative action in the form of executive power (the president's power to pardon convicted offenders) had to be taken in a procedurally fair manner by virtue of the existing requirement of rationality. In other words, the Constitutional Court did not import procedural fairness as such into the principle of legality, but opened the door by recognizing that it might be irrational in some cases not to hear both sides. More recently the same court has hinted, obiter, that procedural fairness may also operate as a requirement, independently of rationality.Footnote 93
Meanwhile, the Supreme Court of Appeal has established that the principle of legality may also demand the giving of reasons as a matter of rationality.Footnote 94 So, as things currently stand, the only ground of review not potentially encompassed by the principle of legality is proportionality. That situation is starting to change, for there have been cases in which the supposedly basic standard of rationality has resembled something far more rigorous.Footnote 95
The upshot of this rapid and seemingly unstoppable development of the principle of legality is that there is increasingly little reason today to bring one's review application under the PAJA. The principle of legality often offers the same relief without all the disadvantages of the PAJA: the difficult administrative action inquiry, the six-month time limit and the strict duty to exhaust internal remedies. Under the principle of legality, the more generous common law rules apply, so one is required merely to bring one's case within a reasonable time and the court will apply the considerably more relaxed common law duty to exhaust internal remedies. It should come as no surprise, therefore, that practitioners and litigants favour the principle of legality because it is more user-friendly than the PAJA, and that the courts favour it because it is so general and flexible. These advantages, in turn, have led to widespread avoidance of the PAJA on the part of litigants, as they sidestep the statute and bring a challenge under the principle of legality. For some time it has been standard procedure for litigants to argue the PAJA and the principle of legality in the alternative, just in case the action turns out not to be administrative action. However, it seems more and more often that litigants are not bothering even to mention the PAJA in their papers. The courts seem increasingly to be going along with this.
In the first few years of the PAJA there was a tendency for litigants to sidestep the statute in favour of constitutional rights or the common law. The Constitutional Court soon put a stop to this practice, for it made no sense to bypass the legislation that was specifically mandated to give effect to those rights and the court was alive to the undesirability of creating parallel systems of law.Footnote 96 Since then, the highest court has continued to enforce the principle of subsidiarity (or one aspect of it)Footnote 97 by not permitting direct reliance on section 33, a higher norm, where the PAJA ought to be used.Footnote 98 It has insisted that, where legislation such as the PAJA gives effect to a constitutional right, a litigant must either use the legislation or challenge it.Footnote 99 However the court has not always extended the same logic to the principle of legality which, as an aspect of the rule of law, is at an even higher level of abstraction than section 33.Footnote 100 It has often allowed litigants to sidestep the PAJA by invoking the principle of legality instead.Footnote 101
Indeed, in its breathtaking 2010 judgment in Albutt,Footnote 102 the Constitutional Court did more than that. In this case, as in many others, the administrative action inquiry was fraught with “difficult questions”;Footnote 103 but now the chief justice held for a unanimous court that it was not necessary or even desirable to answer them. The court reasoned that, because the case could be resolved by the principle of legality on its own unique facts, it was not necessary to “reach” the question of the applicability of the PAJA and that the court below had actually been wrong to engage in the administrative action inquiry, which was merely an ancillary issue.Footnote 104
Albutt dealt the PAJA a near-fatal blow, for logic tells us that, to the extent that all administrative law cases have unique facts, they are all are capable of being resolved by the principle of legality.Footnote 105 A few years later the Constitutional Court conceded that the correct order of inquiry is to consider whether the PAJA applies and, only if it does not, to resort to a more general principle such as the rule of law.Footnote 106 This concession was made quietly, in a footnote, and was surely too unobtrusive as well as rather too late, for bad habits are hard to break, and many judgments since then have effectively demonstrated a continuing tendency to sidestep the PAJA.Footnote 107
More recently a majority of the Supreme Court of Appeal seized an opportunity to express itself emphatically on this topic. In Gijima Footnote 108 it was argued that the PAJA did not apply to an organ of state seeking to have its own decision reviewed and that, if it did, the organ of state could elect to use the principle of legality instead. Cachalia JA acknowledged that litigants sometimes sidestep the PAJA in favour of the principle of legality, and (rightly) blamed this on the difficulty of working out what is and is not administrative action under the PAJA.Footnote 109 If this bypassing were allowed to continue, he said, “PAJA would fall into desuetude”, a result that the framers of the 1996 Constitution and the legislature would not have contemplated.Footnote 110 In his view, therefore, “the proper place of the principle of legality in our law is to act as a safety-net or as a measure of last resort when the law allows no other avenues to challenge the unlawful exercise of public power. It cannot be the first port of call or an alternative path to review, when PAJA applies.”Footnote 111
On appeal, however, the Constitutional Court avoided this larger issue.Footnote 112 It held simply that the PAJA is not applicable when an organ of state seeks review of its own conduct, since such organs are not the primary beneficiaries of the section 33 rights, and that the legality principle must be used instead.Footnote 113 The court thus dealt the PAJA another blow while allowing the more general contest to continue.
CONCLUSION
For South Africa the benefits of having constitutional rights to just administrative action have been considerable. The PAJA, too, has undeniably done some good by making the law clearer and more accessible, and by facilitating some reform. Such benefits will surely also be enjoyed by Kenya. That jurisdiction has the added advantage of being able to learn from South Africa's experience and thus avoid the problems that occurred there. This has already been demonstrated in relation to the FAAA, for Kenya sensibly chose not to replicate the most serious flaws of the PAJA, particularly its overly elaborate definition of administrative action.
As for the story of the PAJA and the constitutional principle of legality, this should serve as a cautionary tale for Kenya as regards its treatment of the FAAA. There appears to have been some early resistance to the Kenyan act for, as noted above, some constituencies doubted that such legislation was desirable or necessary in order to give effect to article 47. Furthermore, the statute does not seem to be proving universally popular with Kenyan lawyers and judges steeped in the common law tradition. In October 2015, almost three months after the FAAA was enacted, one commentator lamented the fact that neither litigants nor the courts seemed to have acknowledged its existence.Footnote 114 Since then, evidence has been mounting of a definite tendency to ignore or bypass the FAAA in favour of the common law, and of reliance on common law doctrines such as Wednesbury unreasonablenessFootnote 115 instead of the more progressive grounds listed in the act. Commentators such as MwangiFootnote 116 and TererFootnote 117 have complained of over-reliance on the common law at the expense of the 2010 Constitution and the FAAA. In short, there are signs that, notwithstanding judicial statements such as those in CCK Footnote 118 and Suchan Investment,Footnote 119 some Kenyan lawyers and judges still regard the FAAA as an unnecessary and unwanted reform of the common law and would prefer to ignore its existence.
This sort of attitude is understandable enough and was certainly part of the South African experience. A decade after the enactment of the PAJA, the author wrote that “[m]any lawyers, especially those educated in the common law era, consider it a spurious refinement and an unnecessary fixing of what was not broken”.Footnote 120 A few years earlier, Currie had described the PAJA as “[u]nloved, disrespected, misunderstood, ignored”.Footnote 121 However, as already observed, ignoring the constitution in favour of the common law is incompatible with the principle of constitutional supremacy. It is also bound to be inimical to the 2010 Constitution's transformative aims.Footnote 122 For example, it is doubtful whether the common law of Kenya contemplates judicial review of the conduct of private bodies, whereas articles 22 and 23 of the 2010 Constitution do invite such scrutiny, and sections 2 and 3 of the FAAA confirm that any entity engaging in quasi-judicial powers or functions will be subject to judicial review.Footnote 123
No doubt the FAAA has some imperfections. Apart from those already identified in this article, a notable flaw is the absence of a catch-all ground of review, for there is nothing in the FAAA to match the “otherwise unconstitutional or unlawful” ground in section 6(2)(f) of the PAJA. Another is the unnecessary repetition of certain grounds of review: procedural fairness, for instance, appears in section 7 of the FAAA at least five times in different guises.Footnote 124 However, flaws such as this should not be blown out of proportion. The absence of a catch-all ground could be cured either by way of a legislative amendment or by means of assiduous judicial interpretation, and the latter tendency to repetition is more of an irritation than a problem. As suggested in this article, there is also much to be grateful for in the statute, including a relatively simple and inclusive definition of administrative action. The author would strongly caution against disrespecting the legislation (not to mention the legislature and the principle of subsidiarity) by allowing the FAAA to be bypassed, whether in favour of the common law, article 47 or some other constitutional principle, as has been the position in South Africa. Kenya should certainly avoid following the South African courts’ example of making its constitutionally mandated legislation almost redundant. In the end, there is no sense in having such legislation without a genuine and concerted commitment to making it work.
