C7 v Secrtary of State for the Home Department

Section 51 of the Senior Courts Act 1981 (‘the 1981 Act’) is headed ‘Costs in civil division of Court of Appeal, High Court and county courts’. Section 51(1) provides ‘Subject to the provisions of this or any other enactment and to rules of court, the costs of and incidental to all proceedings’ in the courts mentioned in the heading, and in the family court, ‘shall be in the discretion of the court’. Section 51(2) makes further provision about rules of court. Section 51(3) provides that ‘The court shall have full power to determine by whom and to what extent the costs are to be paid’. Nothing in section 51(1) is to ‘alter the practice in any criminal cause or in bankruptcy’ (section 51(5)).

Both parties referred to Aiden Shipping Co Limited v Interbulk Limited [1986] 1 AC 965 in which the House of Lords held that the language of section 51 was wide enough to permit the court to order that costs be paid by a person who was not a party to the litigation, and that that wide power was not limited by any rules of court. As Ms Giovannetti explained, Lord Goff’s speech shows that section 51 has had several predecessors. He said, at p 975 A-C, that section 51(1) is

Lord Goff continued, at 975 E-H, that section 51(1) is concerned with the jurisdiction to make orders about costs. That jurisdiction, it should not be forgotten, is ‘expressed to be subject to rules of court, as was the power conferred by section 5 of the Act of 1890. It is therefore open to the rule-making authority…to make rules which control the exercise of the court’s jurisdiction under section 52(1)’. In that context, it was no surprise that the power was conferred in wide terms. That legislative policy was ‘entirely sensible’. He was surprised that it had been suggested that any limitation should be said to be implied in the provision which itself conferred that jurisdiction.

It is convenient here briefly to mention an article about costs in England and Wales by AL Goodhart (as he then was) in Volume 38 of the Yale Law Journal (May 1929, p 849). I thank Mr Deakin for finding it. The article has a useful summary of the history of costs (pages 851-854). There was a distinction between ‘costs awarded by common-law courts and those given by equity’.

The common law rules were based ‘entirely on statute’. The first such statute, which gave the plaintiff his costs, and which was in force until it was replaced in 1875 (presumably by the Supreme Court of Judicature Act (1873) Amendment Act 1875), was the Statute of Gloucester (1275). The law giving costs to the successful defendant developed more slowly; see for example, 23 HEN. VIII, c 15 (1531), 8 ELIZ C 2 (1566) and 4 JAC 1. c 3 (1607). The Supreme Court of Judicature Acts 1873 and 1875 made an important change to the principle on which costs were awarded. In previous statutes, costs had followed the event, but Order 55 of the Rules of the Supreme Court (Schedule 1 to the Act of 1875) provided that costs should be in the discretion of the court.

The Lord Chancellor’s jurisdiction in costs was ‘essentially different’. According to Professor Goodhart (as he later became), there has been a long dispute whether the Lord Chancellor’s ‘power to award costs was an inherent one or was based on statute, 17 RICH III c 6 (1394). The better view seems to be that the power was inherent, and it is clear that the courts have acted on this view’. The great difference between equity and common law costs ‘lay in the fact that in equity costs were in the discretion of the court while at common law they followed the event’. The 1875 Rules of Court therefore put equity costs on a statutory basis, but did not change the principle on which they were awarded. They did, however, change the principle on which common law costs were awarded.

Section 40(1) of the British Nationality Act 1981 (‘the BNA’) defines ‘citizenship status’. Section 40(2) gives the Secretary of State power by order to deprive a person of his citizenship status if satisfied that deprivation is conducive to the public good. The Secretary of State may not make an order under section 40(2) unless satisfied that the order would make a person stateless. Section 40(5) requires the Secretary of State, before making an order under section 40(2), to serve a notice specifying three things, which include his right of appeal, either under section 40A(1), or under section 2B of the Special Immigration Appeals Commission Act 1997 (‘the 1997 Act’) (see paragraph 31, below).

Section 40A(1) of the BNA provides that a person who is given notice under section 40(5) of a decision to make an order in respect of him under section 40 may appeal to the First-tier Tribunal (‘the F-tT’). Section 40A(1) does not apply to a decision if the Secretary of State certifies, in short, that it was taken wholly or partly in reliance on material which, in his opinion, should not be made public. Section 40A(3) applies to an appeal under section 40A(1) various provisions of the 2002 Act, some of which (such as section 106) have since been repealed,.

On enactment, section 1(1) of the 1997 Act provided that there should be a commission, known as SIAC, ‘for the purpose of exercising the jurisdiction conferred by this Act’. Section 1(2) enacted Schedule 1, which was to ‘have effect in relation to the Commission’. With effect from 14 December 2001, section 1 was amended (by the Anti-terrorism, Crime and Security Act 2001) by the insertion of subsections (3) and (4). Section 1(3) provided that SIAC should be ‘a superior court of record’. Section 1(4) provided that its decisions could be ‘questioned in legal proceedings only in accordance with’ the two provisions of the 1997 Act which were listed in section 1(4). The first, section 7, provides for an appeal to this Court on a point of law. With effect from 14 March, section 1(4) was amended, by the deletion of the second of those listed methods of challenge. With effect from 28 June 2018, section 1(4) was again amended, to reflect further legislative changes to the methods for challenging a decision of SIAC. Section 1(1) has been amended more recently, but that amendment is not relevant to the issues on this appeal.

Section 2(1), as originally enacted, gave a person a right of appeal against various matters in relation to which he would have had a right of appeal but for the fact that the decisions in question had been made on the grounds that they were conducive to the public good. Section 2(3) enacted Schedule 2, which made further provision about such appeals. Schedule 2 has since been repealed. Section 2 was amended with effect from 2 October 2000 to reflect changes in the legislation governing immigration appeals which were made by Part IV of the Immigration and Asylum Act 1999, and, with effect from 1 April 2003 to reflect similar changes made by the Nationality, Immigration and Asylum Act 2002 (‘the 2002 Act’). Section 2(2) of the 2002 Act provided that various provisions of the 2002 Act should apply to appeals to SIAC. Section 2 was amended with effect from 31 August 2006 to reflect further changes made by the Immigration and Asylum and Nationality Act 2006. Section 2 was further amended with effect from 28 July 2014 to reflect the re-structure of rights of appeal in immigration cases which was brought about by the Immigration Act 2014.

The version of section 2(1) which was in force at the time of decision 2 gave a right of appeal to a person against a decision if he had a right of appeal against that decision under various provisions of the 2002 Act. Section 2(2) provided that several provisions of the 2002 Act, listed in section 2(2), were to apply to such an appeal, ‘with any necessary modifications’.

Section 2B was inserted in the 1997 Act by the 2002 Act with effect from 1 April 2003. It has no heading. It was amended, with effect from 4 April 2005, by the Asylum and Immigration (Treatment of Claimants Etc) Act 2004. It provides that a person may appeal to SIAC against a decision to make an order under section 40 of the BNA if he is not entitled to appeal under section 40A(1) of the BNA because of a certificate under section 40A(2). Section 40A(3) is to have effect in relation to such appeals. The nature of such an appeal was one of the issues in R (Begum) v Special Immigration Appeals Commission [2021] UKSC 7; [2022] 1 WLR 3784. In short, and contrary to the impression which might be conveyed by the terse language of section 2B, it is not straightforwardly an appeal on the merits.

Sections 2C, 2D, 2E and 2F of the 1997 Act confer jurisdiction on SIAC to review certain exclusion directions, naturalisation and citizenship decisions and deportation decisions. In each case, the person who is the subject of the decision may apply to SIAC to set aside the decision. ‘In determining whether the [decision] should be set aside, [SIAC] must apply the principles which would be applied in judicial review proceedings’ (section 2C(3)), 2D(3), 2E(3) and 2F(3)). In each case, if SIAC decides that the decision should be set aside, it ‘may make any such order or give any such relief as may be made or given in judicial review proceedings’ (section 2C(4), 2D(4), 2E(4) and 2F(4)). The heading of each of these sections starts with the words ‘Jurisdiction: review of…’.

Section 3 is headed ‘Jurisdiction: bail’. Section 3(1) now applies some provisions of Schedule 10 to the Immigration Act 2016 to SIAC.

Until it was repealed by the 2002 Act, section 4 made provision about when SIAC was required to allow, or, as the case might be, to dismiss, an appeal. Section 4(2) gave SIAC power, when it allowed an appeal, to give directions for giving effect to its decision on the appeal, and to make recommendations. It was the duty of the Secretary of State to comply with any such directions.

Section 5 is headed ‘Procedure in relation to jurisdiction under sections 2 and 3’. Section 5(1) gives the Lord Chancellor power to make rules for various procedural purposes. From its commencement, section 5 gave the Lord Chancellor wide powers in connection with appeals to SIAC. Section 5(1) is in substantially the same form now. It gives the Lord Chancellor power to make rules ‘(a) for regulating the exercise of the rights of appeal conferred by section 2 or 2B’, ‘(b) for prescribing the practice and procedure to be followed on or in connection with appeals…including the mode and burden of proof and admissibility of evidence on such appeals’ and ‘(c) for other matters preliminary or incidental to or arising out of such appeals, including proof of the decisions of [SIAC]’.

Section 5(2) requires rules made under section 5 to confer on an appellant the right to be legally represented, ‘subject to any power conferred on [SIAC] by such rules’.

From its commencement, section 5(3) made, and still makes, provision enabling rules to be made in connection with SIAC’s unusual role, permitting, for example, rules enabling procedures in which the appellant is not given full particulars of the reasons of the decision which is the subject of the appeal, enabling SIAC to hold proceedings in the absence of the appellant and of his legal representative and making provision about the functions of special advocates.

By section 5(4), rules may enable some functions of SIAC to be performed by a single member of SIAC, and may include provision ‘conferring on [SIAC] such ancillary powers as the Lord Chancellor considers necessary for the purposes of the exercise of its functions’. The Lord Chancellor must, when making rules, ‘have regard, in particular’, to the two needs listed in section 5(6).

With effect from 1 April 2003, section 5 was amended (by the 2002 Act) by the addition of subsection (2A). Section 5(2A) provided ‘Rules under this section may, in particular, do anything which may be done by rules under section 106 of [the 2002 Act] (appeals: rules)’. Section 106(1) of the 2002 Act gave the Lord Chancellor power to make rules ‘(a) regulating the exercise of the right of appeal’ under various provisions of the 2002 Act, and (b) ‘prescribing the procedure in connection with proceedings under’ the same provisions, and under section 103. Section 106(2) provided that such rules must ‘in particular’, entitle an appellant to be legally represented (section 106(2)(a)), and included a list of 18 other things which such rules might enable or require. Section 106(3)(a) added that rules made under section 106(1) might also ‘enable an adjudicator or the Tribunal to make an award of costs or expenses’, and conferred five other express enabling powers about costs and expenses, such as to make rules about the taxation or assessment of costs, and for the award of interest on costs or expenses (section 106(3)(b) and (c)) (and see also section 106(3)(d) and (e)).

Section 5 was amended again with effect from 15 February 2010, to reflect the transfer of functions to the Upper Tribunal. From that date, section 5(2A) provided, and still provides, that rules under section 5 ‘may, in particular, do anything which may be done by Tribunal Procedure Rules’. The phrase ‘Tribunal Procedure Rules’ is not defined in the 1997 Act. I consider that it is likely to be a reference to the ‘Tribunal Procedure Rules’ referred to in section 22(1) of the Tribunals Courts and Enforcement Act 2007 (‘the 2007 Act’) (see paragraph 42, below).

Section 3(1) and (2) of the 2007 Act establish the F-tT and the Upper Tribunal (‘the UT’) respectively. They are established ‘for the purpose of exercising the functions conferred’ on them ‘under or by virtue of’ the 2007 Act ‘or any other Act’. By section 3(5), the UT ‘is to be a superior court of record’. Section 7 gives the Lord Chancellor power, with the concurrence of the Senior President of Tribunals, to organise each of the F-tT and of the UT into ‘a number of chambers’. Sections 9 and 10 give the F-tT and the UT power to review some of their own decisions. Section 9(3) and 10(3) provide that ‘Tribunal Procedure Rules’ may make various provisions about the exercise of those powers.

Section 22 is headed ‘Tribunal Procedure Rules’. Section 22(1) provides that there are to be rules, to be called ‘Tribunal Procedure Rules’ governing ‘the practice and procedure to be followed’ in the F-tT and in the UT. They are to be made by the Tribunal Procedure Committee (section 22(2)). Section 22(3) enacts Schedule 5. Schedule 5 has four Parts. Part 1 makes further provision about the content of Tribunal Procedure Rules, and Part 2 about the membership of the Tribunal Procedure Committee. Paragraph 12(1) of Schedule 5 provides that rules may make provision ‘for regulating matters relating to costs…of proceedings before [the F-tT] or [the UT]’. Paragraph 12(2) lists five types of provision which the provision made by paragraph 12(1) ‘includes (in particular)’.

Section 25 is headed ‘Supplementary powers of Upper Tribunal’. Section 25(1) gives the UT ‘the same powers rights privileges and authority as the High Court’ in relation to the three matters listed in section 25(2), which include, at (c), ‘all matters incidental to the [UT’s] functions’. But section 25(2) is not to be taken to limit any power to make Tribunal Procedure Rules’ or to limited by anything other than an express limitation in Tribunal Procedure Rules (section 25(3)(b)).

Section 29 is headed ‘Costs and expenses’. Section 29(1) provides that the costs ‘of and incidental to’ all proceedings in the F-tT and in the UT ‘shall be in the discretion of the Tribunal in which the proceedings take place’. By section 29(2), the relevant tribunal has ‘full power to determine by whom and to what extent the costs are to be paid’. But (by section 29(3)), those two provisions ‘have effect subject to Tribunal Procedure Rules’.

Rule 9(1) of the Tribunal Procedure (First-tier Tribunal) (Immigration and Asylum Chamber) Rules (2014 SI No 2604) provides, in short, that if the F-tT allows an appeal, it may order the respondent to pay any fee which the appellant has paid or for which the appellant may be liable. Otherwise, the F-tT may only make an order for costs which is a wasted costs order, or if a person has ‘acted unreasonably in bringing, defending or conducting the proceedings’ (rule 9(2)).

Mr Southey relied, principally, on three authorities.

The first authority is The Guardians of West Ham Union v the Churchwardens and Overseers and Guardians of the Poor of the Parish of St Matthew, Bethnal Green [1896] AC 477. The House of Lords had allowed an appeal by the appellants and had ordered the respondents to pay the costs of the appeal, without specifying the amount of those costs. The Clerk of the Parliaments did not certify the amount of those costs until 3 August 1894. The appellants had threatened to levy an execution on the property of the respondents. The Queen's Bench Division granted a stay of execution, on the grounds that the six-month period allowed for payment of the costs under section 1 of the Poor Law (Payment of Debts) Act 1859 (‘the 1859 Act’) had elapsed, and payment of the costs could no longer be enforced. The Court of Appeal affirmed that decision.

One of the issues on the further appeal to the House of Lords was when the payment of costs had been ‘incurred or become due’ under section 1 of the 1859 Act; when the costs order was made, or when the costs were taxed. The respondents also argued that costs at common law were entirely statutory and that the House of Lords had no statutory power to make an order for costs.

Lord Herschell’s and Lord Macnaghten’s are the only reasoned speeches. Lord Herschell said that the power of the House of Lords to award costs had nothing to do with the Lord Chancellor. Costs had been awarded in such appeals for ‘upwards of two centuries’. He added, ‘I can see no other foundation on which the power to order their payment can be rested except the inherent authority of this House as the ultimate Court of Appeal’. If it had such authority, there was no reason to limit the power to order costs to Chancery appeals, or for excluding any appeals from its exercise (p 483). He considered that the Appellate Jurisdiction Act 1876 assumed the existence of such a power (p 484). Lord Macnaghten said that the argument that the House of Lords had no power to award costs was ‘wholly destitute of foundation’. In his view, ‘…the House of Lords, as the highest Court of Appeal, has and necessarily must have an inherent jurisdiction as regards costs. That this inherent jurisdiction is the sole authority for the action of the House of Lords in dealing with the costs of appeals’ was, in his view, shown by a recent change to its practice in relation to costs. That change was ‘of its own motion, without any statutory authority, simply on the principle which then commended itself to this House…’ (p 489).

The second case is the decision of the Divisional Court in R v Chief Metropolitan Magistrate ex p Osman (1990) 90 Cr App R 313. The government of Hong Kong applied for the costs of resisting the applicant’s unsuccessful application for a writ of habeas corpus, after the respondent had committed him to await extradition to Hong Kong. The hearing of the application took four weeks. The applicant argued, among other things, that the Divisional Court had no power to order award costs in favour of a respondent to an application for habeas corpus. Lloyd LJ, with whom French J agreed, referred to section 51(1) of the 1981 Act as the source of the court’s jurisdiction to order costs in that case. The relevant rule of court was RSC Ord 62, rule 2(1). It expressly applied to ‘all criminal proceedings in the High Court’. The applicant relied on section ‘51(2)’ which provided that ‘Nothing in subsection (1) shall alter the practice in any criminal cause or matter…’. I consider that this is probably a misprint for section 51(5).

The respondent conceded that the proceedings were in a ‘criminal cause or matter’. Lloyd LJ then explained why that concession was correct. The ‘crucial question’ was ‘what is the practice in habeas corpus with regard to costs?’. The respondent relied on three reported cases in which costs had been awarded to the applicant (two cases), and to the respondent (one case). He considered that two of the cases were ‘sufficient evidence to establish a practice of awarding costs in the Divisional Court’ (original emphasis).

The third case is R (Cart) v Upper Tribunal. Three claimants applied for judicial review of decisions of, in Cart’s case, the UT, and, in the case of the two other claimants, decisions of SIAC about bail. Neither decision was appealable under the relevant statutory scheme. The Divisional Court held that the UT and SIAC were amenable to judicial review. There was no appeal to this Court in the two SIAC cases, but Cart did appeal to this Court.

Laws LJ, giving a judgment with which Owen J agreed, referred to the statutory provisions which declared that SIAC and the UT were superior courts of record. The defendants argued that those provisions immunised SIAC and the UT from judicial review. In paragraph 6 he said that section 2 of the 1997 Act ‘defines SIAC’s jurisdiction’. The ‘general effect’ of section 2(1), in particular, was ‘of the first importance’. He noted the general similarity between SIAC’s jurisdiction and that of the Asylum and Immigration Tribunal (‘the AIT’), but also that the ‘process of a SIAC appeal’ was ‘very distinct’. He noted the amendment of section 1 by the insertion of section 1(3) and (4).

The defendants’ primary argument was that a superior court of record was ‘ipso facto’ immune from judicial review. Laws LJ considered that there was a ‘prior, and in some ways greater question’. The mere designation of a court as a superior court of record did not exclude judicial review. The question, rather, was whether the court had ‘characteristics which have been objectively recognised as the marks of a superior court of record’. The defendants’ case, in short, was that judicial review could ‘be ousted by an implication’, and not only that, but ‘one contained in a formula which amounts in effect to a deeming provision’ (paragraph 32). Laws LJ did not accept that, as a matter of statutory construction (paragraph 33).

There must be an independent and authoritative source for interpreting the law. The High Court was such a source. ‘To offer the same guarantee of properly mediated law, any alternative source must amount to the alter ego of the High Court’ (paragraph 39). In paragraphs 54-69, Laws LJ considered three distinctions between inferior courts, which were subject to the supervision of superior courts (in particular, the Court of King’s Bench) and superior courts, which were not. The first was between courts which are presumed to act within their powers until the contrary was shown, and those which were not. The second was between courts which are subject to, or immune from, supervision by judicial review. The third was between courts of limited and courts of unlimited jurisdiction (paragraph 54). An analysis of the cases showed that the name ‘superior court of record’ was not a reliable guide to which courts are immune from judicial review, still less decisive of that issue (paragraph 56).

In paragraph 57, he cited a paragraph from Halsbury’s Laws of England about the contrasting jurisdictions of superior courts and inferior courts. This passage refers to the first distinction described by Laws LJ (see the previous paragraph). In paragraph 65, he said that weight of authority pointed to excess of jurisdiction as the foundation for the issue of the prerogative writs. The Court of King’s Bench, by contrast ‘out of which the writs issue, is a court of unlimited jurisdiction and therefore not itself subject to any such supervision’. Courts ‘whose jurisdiction is limited’ will generally be subject to judicial review (paragraph 68). In paragraph 72 he concluded that the second distinction was the explanation for the first. Even if he had been satisfied that section 1(3) of the 1997 Act could immunise SIAC from judicial review, he would have held that, ‘in the particular case’ they did not (paragraph 73).

In paragraph 75, he described the characteristics which, by using the expression ‘superior court of record’, Parliament might be taken to have attributed to SIAC. SIAC would be presumed to have acted within its powers unless the contrary was shown, its decisions would be precedents for lower courts and tribunals (no doubt because of the records it kept) and such a court has a power to punish for contempt. The conclusion that section 1(3) did not immunise SIAC from judicial review did not ‘deprive it of content’.

SIAC had a limited jurisdiction (narrower than that of the UT). Laws LJ then asked whether SIAC ‘constitute[d] in effect an alter ego of the High Court’ (paragraph 77). That involved considering whether, in effect, the court in question has a final power (subject to an appeal) to interpret for itself the law which it must apply (paragraph 81). SIAC was reviewable for exceeding the boundaries of its permitted subject matter. It was not in dispute that, before the insertion of section 1(3), it was amenable to judicial review. SIAC’s decisions were also reviewable for an error of law within its jurisdiction (paragraph 82). The consequences were limited. If a decision was appealable, judicial review would not be available (as there would be a suitable alternative remedy). Judicial review would not be available in respect of interlocutory decisions on the way to an appealable decision, ‘at least without some gross and florid error’. SIAC’s decisions about bail would not generally be challengeable by judicial review, given the fine judgments involved in such decisions (paragraph 85).

On the Secretary of State’s appeal in Cart, this Court agreed with the reasoning of Laws LJ in paragraphs 28-75 (paragraph 17). It disagreed with Laws LJ’s characterisation of the UT as the alter ego of the High Court (paragraph 19). It agreed with the Divisional Court that the supervisory jurisdiction of the High Court could only be ‘ousted by the plainest possible statutory language’. In the event, it dismissed the appeal.

Mr Southey, who argued the appeal with great tenacity, submitted that designation of SIAC as a superior court of record must have been intended to have some significance. He argued that the designation of the UT and of SIAC as superior courts of record carried with it an inherent jurisdiction to make orders for costs. Section 29 of the 2007 Act was not the source of the UT’s power to make costs orders, but a recognition of the power which flowed independently from that designation. Moreover, it was to be presumed that Parliament had conferred on SIAC all the powers which were necessary to enable it to fulfil its statutory functions, and that must include a power to award costs. That was not contradicted by the provisions of section 5; as rules were necessary, for example, to ensure consistency.

The Judge had erred in law in deciding that SIAC had no inherent power to award costs. A superior court of record has inherent powers unless they are excluded expressly or by necessary implication. The West Ham case showed that a superior court of record has an inherent power to award costs, as did Ex p Osman. He referred to other cases about ‘inherent powers’, but to no other decision about an inherent power to award costs. He accepted that this is, fundamentally, a question of statutory interpretation. He also accepted that the inherent jurisdiction to award costs could be constrained by express provisions in the Rules, but that an express statutory provision would be necessary to exclude SIAC’s inherent power to award costs. SIAC’s inherent power to award costs, by analogy with that of the High Court (see Aiden Shipping) existed unless and until it was constrained by any relevant rules.

He submitted that the language of section 29 of the 2007 Act was not consistent with the view that section 29 was the source of the UT’s power to award costs. He was unable to explain how that submission explained the purpose of section 29 as it applies to the F-tT, which is not a superior court of record. The language of section 5(1)(b) and section 5(2A) was not inconsistent with and did not exclude SIAC’s inherent power, which already existed. That power existed before SIAC was designated a superior court of record, but that designation made the argument that SIAC had that power stronger.

Mr Southey also referred to two academic articles about the inherent jurisdiction. Both were expressed in very general terms, and there was only one passing reference to costs. I did not find either article helpful.

His next submission was that if SIAC did have a power to award costs, the Judge’s alternative reasoning in paragraph 16 (see paragraph 16, above) was wrong. His submission appeared to be that if SIAC allowed an appeal, it should award the appellant his costs, but that it did not follow that if the appeal was dismissed, the Secretary of State should be awarded her costs. When pressed he accepted, in effect, that the Secretary of State might be entitled to her costs, but only if all the material was in OPEN and the appellant’s conduct of an appeal was unreasonable. The Secretary of State would not be entitled to her costs, he submitted, if the CLOSED material showed that the appellant’s case was false, but only if all the material which showed that it was false was OPEN.

Finally, he submitted that the Secretary of State’s conduct in changing experts in the course of several appeals about Bangladeshi citizenship, so that her expert in this case was the sixth such expert, was plainly unreasonable, and that SIAC was wrong to hold otherwise.

Ms Giovannetti’s first point was that it was not clear from Mr Southey’s submissions whether his case was that all courts have an inherent power to award costs, or whether it was that all superior courts of record have that power. If all superior courts of record have that power, section 51 of the 1981 Act was unnecessary. She referred to the legislative history as summarised in Aiden Shipping. Section 51 did not define powers which superior courts of record already had. On the contrary, it was the source of those powers. She referred to the article by AL Goodhart. That showed that the source of the powers of common law courts to make orders for costs was statutory. Osman was an unusual case. It decided no more than that, before the enactment of section 51, the Divisional Court had had a practice of making orders for costs in applications for habeas corpus in criminal causes or matters. It is authority for no more than the proposition that the Divisional Court had had such a practice in that narrow class of cases. Moreover, in the cases which illustrated the existence of that practice, there had been no argument, and no reasoning by the Divisional Court, which explained the source of its assumed power to make an order for costs.

She accepted, by reference to the West Ham case, that the House of Lords might be in a different position. All that that case established was that the House of Lords had, in practice, awarded costs for about 200 years, and that the power to do so was expressed to come from its unique position as the ultimate court of appeal. Despite their efforts, counsel’s research had identified no case, apart from cases which dealt with the position in equity, in which it had been held that a superior court of record had, by virtue of that designation alone, a power to award costs. This Court should be very cautious before developing the law in that way. Moreover, it should not do so where, as here, there is a detailed legislative scheme. None of those submissions was inconsistent with the reasoning of the Divisional Court in Cart. The presumption that a superior court of record has acted within its power is not the same as a decision that a superior court of record has power to do everything which is not expressly prohibited. It was always essential to consider the relevant statutory framework.

She submitted that it was necessary to distinguish between a court’s inherent jurisdiction and its implied jurisdiction. A court had such implied powers as were necessary to enable it to do justice, but a power to award costs is not such a power. It was open to the Lord Chancellor, in the exercise of the discretion conferred by section 5, to decide not to confer on SIAC a power to award costs.

If that was wrong, the Judge was entitled to decide that, if SIAC did have a power to award costs, that power should be exercised on the same principles as the power which the F-tT has in appeals which have not been certified. An alternative approach, consistent with Taylor v Lawrence [2002] EWCA Civ 90; [2003] QB 528, would be that a power to award costs is to be exercised in exceptional cases in which the exercise of such a power is necessary to avoid injustice. It was clear from the Judge’s reasoning that this was not such a case. Any challenge to the Judge’s decision that the Secretary of State’s conduct of the appeal had not been unreasonable was ‘obviously hopeless’. The Judge had not focussed exclusively on abuse of process, and had distinctly considered the argument that the Secretary of State’s conduct of the appeal had been unreasonable. Having heard the appeal over several days, he was in a unique position to make that assessment. The appeal had also raised a novel point based on Devaseelan v Secretary of State for the Home Department [2002] UKIAT 702.

In his reply, Mr Southey said that it was not necessary to decide whether all courts have a power to award costs, or just superior courts of record. The question was whether SIAC had such a power. Section 51 created a ‘playing field’ for the Rules Committee. The key point was that courts of equity had always had a power to award costs. Where legislation was silent, the position adopted by courts with no express powers to award costs is significant. The necessity threshold was too a high a threshold to set for the existence of an inherent power. If the Lord Chancellor had made no rules about costs, it was open to SIAC to decide what its practice about costs should be.

There are four issues.

The historical material suggests that, for many centuries, and subject to the probable exception of the courts of equity, the power to award costs had a statutory origin. Section 51 of the 1981 Act, like its statutory predecessors, did two things. It continued the statutory basis of the common law courts’ power to award costs, but changed it, by substituting, for a rule that costs followed the event, a rule that costs were discretionary. It also put the power of courts of equity to award costs, which had always been discretionary, on a statutory footing.

Chronologically, the first question is whether or not SIAC had any power to award costs on enactment, that is before the 1997 Act was amended by the insertion of sections 1(3) and (4), or 5(2A). I consider that, on the basis of the materials which this Court was shown, it did not. On enactment, SIAC was not a superior court of record. In name, it was not even a court, but a ‘commission’. It had a comparatively limited and wholly statutory jurisdiction. Although this is not decisive, it is also relevant that the draftsman of the 2002 Act (see paragraph 76, below) considered that the formula in section 5(1)(a) and the first part of 5(1)(b) was apt to give the rule-maker a power to enable the relevant body to make an award of costs.

If my understanding of the history is right, a court with a limited statutory jurisdiction would have had no power to award costs unless, like that court’s jurisdiction, that power was conferred by, or under, statute. I also consider that, even without the insertion of section 5(2A), the language of section 5, which conferred on the Lord Chancellor, in very broad terms, powers to regulate SIAC’s procedure, evinced a clear intention that, if SIAC was to have any power to award costs, it could only be conferred by rules made by the Lord Chancellor. An inherent power to award costs (or, indeed, to make any rules about its own procedure) could not co-exist with the powers conferred on the Lord Chancellor by section 5, which occupied the relevant field.

The next question, chronologically, is whether, by amending section 1 of the 1997 Act and making SIAC a superior court of record, Parliament also gave it an inherent power to award costs. I do not consider that it did. The Court was not shown any authority which decides that, in and of itself, the designation of SIAC as a superior court of record gave it an inherent power to award costs. The West Ham Union case is the only case this Court was shown in which a court has decided that it has an inherent power to award costs. On any view, that was a special case. The express reasoning of the House of Lords self-evidently cannot be generalised to superior courts of record. Osman does not concern any inherent jurisdiction to award costs. It was concerned with a different, and narrow, factual question, which was whether or not, before the commencement of the 1981 Act, the Divisional Court had a practice of awarding costs in habeas corpus applications in criminal causes or matters. This conclusion does not, contrary to Mr Southey’s submission, deprive section 1(3) of any meaning. I accept Ms Giovannetti’s submission that it means something, in any event, for the reasons given by Laws LJ in paragraph 75 of Cart (see paragraph 57, above). My reasoning about the effect of section 5, before it was amended, also applies during this period (see paragraph 73, above).

Whatever the position was during these two periods, there was a potentially significant change when section 5 was amended, by the insertion of section 5(2A), with effect from 1 April 2003 (see paragraph 39, above). Section 106(1)(a) and (b) of the 2002 Act used the same formula as section 5(1)(a) and the first part of section 5(1)(b). Section 106(3) provided that rules made under the power conferred by that formula might enable or require various things in connection with an award of costs. I consider that, whatever the position was before this amendment, the words of section 5(2A) show Parliament’s clear intention that, if SIAC was to have a power to award costs, the source of any such power, as in the case of the tribunal and adjudicators, was to be rules made by the Lord Chancellor. It would follow that, if no such rules were made, SIAC would have no such power. I note again that the draftsman of the 2002 Act, whether rightly or wrongly, assumed that that formula, which was in section 5 as enacted, and was repeated in section 106(1) of the 2002 Act, enabled the rule maker to confer a power to award costs. By this stage, SIAC was a superior court of record, but a superior court of record of limited jurisdiction, and it was amenable to judicial review.

Did the further amendment of section 5 with effect from 15 February 2010 (see paragraph 40, above) change that position? The change was a limited one. In effect, the new provisions which govern the Upper and the First-tier Tribunals, as enacted by the 2007 Act, were substituted for the provisions which governed immigration tribunals before their functions were transferred into that new structure. I do not consider that they did change the position.

Mr Southey suggested that the UT has an inherent power to award costs. That suggestion was based on the words of section 29(1) of the 2007 Act (see paragraph 44, above). He could not explain how that theory explained the purpose of section 29 as respects the F-tT. Moreover, those words are similar to the language of section 51(1) of the 1981 Act (see paragraph 20, above). As I have explained above, that language did not, in the case of common law courts, recognise an inherent jurisdiction to award costs. On the contrary, that power had always been, and continued to be, statutory. But whether or not the UT has such a power is irrelevant. Even if, which I doubt, the UT does have such a power, that assumed fact could cast no light on Parliament’s intentions in amending section 5 of the 1997 Act in 2010. The point is that section 5(2A) does not give SIAC all the powers which the Upper Tribunal has. What it does, instead, is to give the Lord Chancellor, when he makes rules for SIAC, power which enables such rules to ‘do anything which may be done by Tribunal Procedure Rules’. That clearly includes a power to make rules about costs (see paragraphs 42-45, above). For the reasons I have given in the previous paragraph, this amendment, as did the amendment with effect from 1 April 2003, clearly excludes any inherent power to make an award of costs.

One consequence of Mr Southey’s argument, if it is right, is that SIAC, perhaps uniquely among courts and tribunals, would have a completely unfettered discretion to award costs. I describe one difficulty which the logic of this submission caused for Mr Southey’s argument in paragraph 84, below.

Mr Southey also submitted that SIAC has an implied, rather than an inherent, power to award costs. There are two points.

First, I agree with the Judge the test for an implied power in this context is whether a power to award costs is necessary to enable SIAC to do justice. I also agree with him that such a power is not necessary for that purpose. As he said, not all courts or tribunals have a power to award costs. Mr Southey did not show us any material which supported an argument that, in SIAC, a power to award costs is necessary to enable SIAC to do justice. Apart from an isolated and unreasoned statement in Al-Jedda by Irwin J (as he then was) that SIAC has power to award costs, and a consent order which showed that the Secretary of State had agreed to pay costs in one case, there was no material which showed that SIAC, or the parties, had considered that question.

Second, whether such a power is necessary to enable SIAC to do justice is not the only question. There is a more fundamental question. That is whether such a power can be implied in this statutory scheme. Local authorities, the powers of which are wholly statutory, do not have implied powers to do things in a field which is governed by a detailed statutory code, except to the extent that those are authorised by section 111 of the Local Government Act 1972, which codifies the common law about implied statutory powers (see Hazell v Hammersmith and Fulham London Borough Council [1992] 2 AC 1, and R (Kalonga) v Croydon London Borough Council [2022] EWCA Civ 670; [2022] PTSR 1568, paragraphs 30-33 and 73). I consider that the position of SIAC is analogous. There is a detailed statutory code governing SIAC’s procedural powers. The rule-maker under that code is the Lord Chancellor, not SIAC. If the Lord Chancellor has not made a rule authorising SIAC to make an award of costs, SIAC does not have an implied power to do so.

I should make clear that, in reaching this conclusion, I have not been influenced by the amendments to the 1997 Act which gave SIAC power to set aside certain decisions on a statutory review (sections 2C-E). I do not consider that these changes, which arguably gave SIAC a power to award costs in those contexts, can cast light on the meaning of section 5, which, for present purposes, was in its current form before the statutory review amendments were made.

There is a paradox in Mr Southey’s argument on this part of the case. The article 14 argument which we did not permit him to argue was based on a comparison between the F-tT and SIAC, presumably because, absent the certificate, A would have been able to appeal to the F-tT. The F-tT only has power to award costs if the conduct of a party has been unreasonable (see paragraph 45, above). I have recorded Mr Southey’s submission in paragraph 64, above. It was at odds with his case on article 14. If SIAC does have a broad discretion, however, it is difficult to see how, in the absence of any guidance in rules or elsewhere, it could err in law in deciding for itself the principles on which that power should be exercised, or that it could err in law in rejecting Mr Southey’s notably one-sided approach, which is not based on any relevant analogy. If contrary to my view, SIAC does have a power to award costs, I do not consider that the Judge erred in law in deciding that the power should be exercised in accordance with the principle which applies to the exercise of that power by the F-tT and the UT.

The Judge was part of the panel which, over five days, heard the preliminary issues in A’s appeal and in the appeals of C3 and C4. He was steeped in the litigation history, and familiar with the nuances of the legal and factual issues, as his careful judgment on preliminary issues shows. In decision 2, he considered, and succinctly rejected, two arguments: that the Secretary of State’s conduct of the appeal had been an abuse of process and that it had been unreasonable. Each assessment by the Judge is an assessment with which this Court should be reluctant to interfere on an appeal on a point of law. I do not consider that the Judge erred in law in deciding not to order the Secretary of State to pay the costs of A’s appeal.

For those reasons, I would dismiss this appeal.