Bloom & Ors v The Pensions Regulator (Nortel, Re)

At the heart of the insolvency legislation lies the implementation of the public policy objective that the assets of an insolvent person (whether individual or corporate) should be realised and distributed pari passu among that person’s creditors, in proportion to the amount of their claims. The pari passu principle is a fundamental principle of justice, equity and fairness, with application in a wide variety of circumstances. Nowhere is it more fundamental than in the insolvency code.

Any workable process of pari passu distribution requires the identification, quantification and, if necessary, valuation of the creditors’ claims, by reference to a particular date (“the cut-off date”). For the purposes of distribution in insolvency the cut-off date is the date upon which the person is first subjected to the relevant insolvency process. Subject to a puzzling sub-issue in cases where, for example, an administration is immediately followed by a liquidation, the cut-off date in relation to a company is the date when it goes into administration or into liquidation.

Since the mid-19^th century a succession of Bankruptcy and Insolvency Acts have sought to establish a wide and inclusive definition of claims qualifying for pari passu treatment, by providing that provable debts are to include both debts and liabilities, and to extend to debts and liabilities which are, at the cut-off date, both present, future and contingent. In relation to personal bankruptcy, the underlying policy is not merely that creditors should be fairly treated inter se, but that the bankrupt should receive as full as possible a discharge from his debts. In relation to corporate insolvency, the requirement for the fair treatment of the company’s creditors is sharpened by the fact that, (save in exceptional cases) the company will be dissolved at the end of the insolvency process so that, if the claim is not subjected to pari passu treatment by that process, it will not be met at all. It will fall down what was described in argument as a black hole. For a claim to qualify for pari passu treatment it must be a provable debt. Generally speaking, although the precise nature of this requirement has been the subject of intense debate at the hearing, provable debts arise only out of matters which have occurred, or have begun to occur, prior to the cut-off date.

A company does not cease to exist merely because it becomes subject to an insolvency process. Financial liabilities have to be incurred for the benefit of the process. Furthermore, Parliament has in specific areas (such as corporation tax) expressly provided for non-provable liabilities to be payable by companies in liquidation regardless whether they arise from anything done for the benefit of the process. In other circumstances, such as rates, the courts have concluded, albeit in the absence of an express statutory provision, that Parliament intended that certain types of non-provable debt, usually arising from matters occurring after the cut-off date, should nonetheless be paid by the company in an insolvency process. This is usually because the statute creating the liability has used criteria for liability which do not distinguish between persons which are, or are not, in an insolvency process. The criteria are insolvency neutral.

In Re Toshoku Finance plc (In liquidation) [2002] 1 WLR 671 the House of Lords concluded (in relation to a company in liquidation) that such statutory liabilities constituted liquidation expenses because they were “necessary disbursements” of the liquidator, for the simple reason that, as Lord Hoffmann put it at paragraph 30:

“There would be little point in a statute which specifically imposed liabilities upon a company in liquidation if they were payable only in the rare case in which it emerged with all other creditors having been paid.”

The Toshoku analysis has been applied to companies in administration, mainly because of the creation after that decision of a similar expenses regime for such companies: see Exeter City Council v. Bairstow [2007] BCC 236.

The primary case of the Regulator and its supporters (namely that liabilities arising from the FSD regime were payable as a liquidation or administration expense), was therefore as follows:

The Administrators challenge this simple analysis in the following way:

It is no small irony that the primary cases of both camps depended upon an assertion that financial liabilities arising from post cut-off date FSDs and CNs are not provable debts, and that the alternative case of the Regulator and its supporters depended upon asserting the exact opposite. On any view, the question whether such liabilities are provable debts lies at the heart of all the competing alternatives, and the vigour with which Mr Sheldon and Mr Moss in particular pursued the submission that they were provable debts lost none of its force or enthusiasm from the fact that success in that submission would be fatal to their clients’ primary case. In fact, as I have noted, the Lehman Trustees were content to be neutral as between expense and provable debt, even though the Regulator and its other supporters were not.

The above is only a summary of the issues. In particular, it is not a necessary conclusion that the priority afforded to a CN must be the same as that to be afforded to the financial consequences of an FSD. Furthermore, the facts of the cases before the court require the determination of the puzzling issues arising, in relation to insolvency processes which (like these) began before 5^th April 2010, where an administration is, or may be, immediately followed by a liquidation.

Nor is there any single logical order in which to address the various overlapping questions. Ultimately, the court’s task is to weigh up the pros and cons of each of the rival interpretations and to reach a conclusion as to which should be preferred. In that process, although policy plays an important part in a purposive construction of statute, responsibility for balancing the policy objectives of the insolvency and pension regimes is, to the extent that they appear to conflict, a matter for Parliament rather than the court.

I have nonetheless found that the difficult process of balancing the rival interpretations is considerably simplified by addressing first the detailed rival arguments about the effect of the provisions in the Insolvency Act and Rules about provable debts, and then by identifying the precise meaning and effect of what I have called the Toshoku principle, in relation to expenses.

The primary statutory provision defining what are provable debts is Rule 13.12 of the Insolvency Rules 1986. It has been the subject of a number of amendments, and the version of the Rule in force for the purposes of the insolvency processes of all the relevant Nortel and Lehman applicant companies is as follows:

“13.12.— “Debt”, “liability” (winding up)

(1)

“Debt” in relation to the winding up of a company, means (subject to the next paragraph) any of the following—

(a)

any debt or liability to which the company is subject at the date on which it goes into liquidation;

(b)

any debt or liability to which the company may become subject after that date by reason of any obligation incurred before that date; and

(c)

any interest provable as mentioned in Rule 4.93(1).

(2)

For the purposes of any provision of the Act or the Rules about winding up, any liability in tort is a debt provable in the winding up, if either—

(a)

the cause of action has accrued at the date on which the company goes into liquidation; or

(b)

all the elements necessary to establish the cause of action exist at that date except for actionable damage.

(3)

For the purposes of reference in any provision of the Act or the Rules about winding up to a debt or liability, it is immaterial whether the debt or liability is present or future, whether it is certain or contingent, or whether its amount is fixed or liquidated, or is capable of being ascertained by fixed rules or as a matter of opinion; and references in any such provision to owing a debt are to be read accordingly.

(4)

In any provision of the Act or the Rules about winding up, except in so far as the context otherwise requires, “liability” means (subject to paragraph (3) above) a liability to pay money or money’s worth, including any liability under an enactment, any liability for breach of trust, any liability in contract, tort or bailment, and any liability arising out of an obligation to make restitution.

(5)

This Rule shall apply where a company is in administration and shall be read as if references to winding-up were a reference to administration.”

The following points need to be made about the amendments. First, sub-rule (2) about tort liabilities was introduced in 2006 specifically to reverse, on policy grounds, the outcome of Re T & N Limited (No 2) [2005] EWHC 2870 (Ch). Prior to that amendment sub-rule (2) provided as follows:

“In determining for the purposes of any provision of the Act or the Rules about winding up, whether any liability in tort is a debt provable in the winding up, the company is deemed to become subject to that liability by reason of an obligation incurred at the time when the cause of action accrued.”

The result in T & N was that the claims of those who had suffered actionable damage from asbestosis only after the cut-off date could not prove in T & N’s liquidation, even though the alleged negligence which gave rise to their exposure to that condition occurred many years previously. Their common law claims fell, however tragically, down a black hole.

Secondly, sub-rule (5) was introduced in 2003 specifically for the purpose of extending the definition of provable debts to administration, following the introduction by the Enterprise Act 2002 of the new style administration which, for the first time, permitted administrators to call for proof and to make distributions to those with provable debts. Thirdly, sub-rules (1), (2) and (5) have all been substantially amended in 2010, but without retrospective effect upon insolvency processes first instituted before 5^th April 2010, so as to provide in clear terms that, where a liquidation is immediately preceded by an administration, the cut-off date for provable debts in the liquidation is the date of the commencement of the administration, rather than of the liquidation (and vice versa).

Rule 13.12(1), (3) and (4) are substantially derived from the similar provisions for personal bankruptcy in section 382(1), (3) and (4) of the Insolvency Act, being re-enactments of a formula dating back originally to the Bankruptcy Act 1869.

Two issues as to the interpretation and application of Rule 13.12 to the FSD regime fall for determination. The first issue is, in the context of an FSD or CN which (as in the present cases) is issued after the onset of administration of the target company, whether any financial obligation arising from either of them is capable of being a provable debt in that administration. The second issue, which arises if the first is answered in the negative, is if after the issue of an FSD while the target is in administration it then goes into liquidation before being issued with a CN, the financial obligation arising from the CN is a provable debt within the meaning of Rule 13.12 in that subsequent liquidation. It is common ground that, at least at first instance and probably in the Court of Appeal, the outcome of the first issue depends upon whether, within the meaning of sub-rule (1)(b), the target incurred a relevant obligation before the onset of the administration.

Put shortly, the alternative case of the Regulator and its supporters is that each target company incurred a relevant obligation before the cut-off date because the FSD regime is designed to require financial assistance where group structures have historically been such as to leave the employer with insufficient resources to meet its pension obligations, by comparison with better resourced companies within the same group, and in particular where the target company has received benefits from the services of the employees of the employer greater in value than the amounts paid to the employer for the use of those services: see in particular section 43(7)(b) of the 2004 Act. The real objective, so it was submitted, was to require groups to provide that support in the first place, so that the FSD regime could not be applied to them, and then to apply the FSD regime only to groups which fell short of that obligation in the first place.

For the Administrators Mr Dicker submitted that this analysis failed on two grounds. Firstly, nothing in the historic circumstances which might trigger the application of the FSD regime constituted a legal obligation, or a breach of a legal obligation, of the type required by Rule 13.12(1)(b). Secondly, he submitted that since, in any event, the imposition of a financial obligation under the FSD regime always depended upon the exercise by the Regulator of one or more broad discretions, this was on settled authority sufficient on its own to prevent any consequential financial obligation from being a provable debt.

I therefore turn to the authorities, which begin with Re Smith; ex parte Edwards (1886) 3 Morrell 179. That was a case on the substantially identical language of section 37 of the Bankruptcy Act 1883. The debtor became bankrupt after entering into an arbitration agreement with the creditor which provided that the costs of the arbitration were to be in the discretion of the arbitrator. After the debtor became bankrupt, the arbitrator made an award, including a costs award, in favour of the creditor. The Divisional Court held, reversing the County Court judge, that the costs award was provable in the bankruptcy, identifying the contractual submission of the debtor to the arbitrator’s costs discretion as a sufficient pre-cut-off date legal obligation.

In Glenister v.Rowe [2000] Ch 76 the debtor obtained an order striking out a claim by the creditor for breach of trust. The creditor filed a notice of appeal, following which the debtor was made bankrupt. Shortly after the debtor’s discharge, the creditor’s appeal was allowed, with costs. The creditor then sought to serve a statutory demand on the debtor for the taxed costs, which the debtor resisted on the grounds that it was a debt provable in his earlier bankruptcy, from which he had therefore been discharged. The Court of Appeal ruled in favour of the creditor. The Deputy Judge had ruled that the creditor’s claim had been a contingent liability at the commencement of the debtor’s bankruptcy within section 382(1)(a) of the Insolvency Act, but rejected a claim in the alternative based on prior obligation under section 382(1)(b). The case was therefore argued in the Court of Appeal under section 382(1)(a), and a last ditch attempt by the debtor to rely upon section 382(1)(b) was refused, as being too late: see page 85 C to D per Mummery LJ. Nonetheless he observed, but without hearing argument on its merits, that the point would in any event have failed.

The main bone of contention in the Court of Appeal was therefore whether, regardless of the presence or absence of a pre-cut-off date obligation, there existed a liability within sub-subsection (a) of a future or contingent nature (within the meaning of subsection (3)) as to which the debtor unsuccessfully relied upon the majority opinion of the House of Lords in Re Sutherland Dec’d [1963] AC 235, to the effect that contingent liability did not necessarily depend upon an antecedent obligation.

Re T&N (No2) (supra) was a case of corporate rather than personal insolvency, in which a conclusion that a common law liability was not a provable debt meant that it fell down what I have referred to as a black hole, rather than (had the debtor been an individual) being available for enforcement after the bankrupt’s discharge. The lengthy and painstakingly careful judgment of David Richards J may be summarised for present purposes as follows:

Two days later, the Court of Appeal gave judgment in R (Steele) v. Birmingham City Council [2005] EWCA Civ 1824, [2006] 1 WLR 2380, in which a discharged bankrupt sought judicial review of a decision by the Secretary of State under section 71(1) of the Social Security Administration Act 1992 to recover an overpayment of jobseeker’s allowance which had occurred before his bankruptcy. His case was that the Secretary of State’s determination was a contingent liability at the date of his bankruptcy, and therefore a provable debt from which he had been discharged. The case appears to have been argued mainly under section 382(1)(a) of the Insolvency Act, but with an alternative case under sub-subsection (b): see per Sir Martin Nourse, referring to counsel’s submissions, at paragraphs 10 and 16 respectively.

The Court of Appeal rejected the case under section 382(1)(a) following Glenister v. Rowe. As to the alternative case, the court held that any original common law liability to repay the overpaid jobseeker’s allowance was not the origin of the liability being pursued after discharge, which was itself created entirely as a result of the discretionary decision of the Secretary of State to pursue recovery under section 71(1) of the 1992 Act: see per Sir Martin Nourse at paragraph 16 and per Arden LJ at paragraph 28.

Pausing there, a common feature of both Glenister and Steele was that the liabilities in question arose as a result of the exercise of a discretion, in the first case by a court, and in the second by the Secretary of State. In neither case was it preceded by, still less incurred by reason of, an obligation incurred by the debtor before his bankruptcy for which the discretionary power was created as a means of enforcement, or as a remedy for breach. Both liabilities arose by reason of the exercise of discretions conferred otherwise than by contract. By contrast, in ex parte Edwards, although the liability also arose from the exercise of a general discretion, that discretionary power had been conferred by a private arbitration contract to which the debtor was a party.

The critical importance of a pre-cut-off date legal obligation, as the precondition for proving a contingent debt, was emphasised by the High Court of Australia in relation to the substantially similar provisions of section 82 of the Bankruptcy Act 1966 in Foots v. Southern Cross Mine Management Pty Ltd [2007] HCA 56, [2007] BPIR 1498, at paragraphs 35 to 36. In that case the debtor had fought and lost an action at trial before becoming bankrupt, but the (probably inevitable) costs order was only made after his bankruptcy. The High Court held that even fighting and losing at trial gave rise to no relevant pre-cut-off date obligation.

A different conclusion was reached by the Court of Appeal in Haine v. Day [2008] EWCA Civ 626, [2008] BCC 845, in which a company had, shortly before going into administration, dismissed forty of its employees without prior consultation, contrary to section 188 of the Trade Union and Labour Relations (Consolidation) Act 1992. By section 189 the employees were entitled to complain of the lack of consultation to an employment tribunal which, if it found the complaint well-founded “shall make a declaration to that effect and may also make a protective award”. Administration had intervened, and been followed by liquidation, before the employees made their complaints and obtained their protective awards. The question was whether those awards gave rise to provable debts. The judge, Sir Donald Rattee, concluded, following Steele, that the employer’s failure to consult gave the employees no legal right to a protective payment. Rather, the protective award depended upon the exercise of a discretion by the Employment Tribunal.

The Court of Appeal allowed the employees’ appeal on two grounds. The first was that the pre-liquidation breach of the obligation to consult was a sufficient legal obligation under Rule 13.12(1)(b) to give rise to a contingent liability as at the cut-off date, the relevant contingency consisting of the Tribunal later making the protective award. The second was that, on the facts, the Tribunal could only exercise its discretion in favour of the making of an award, such that it had no real discretion to refuse to do so: see paragraph 55. The court distinguished both Glenister and Steele on those grounds and, at paragraph 87, specifically approved the decision in ex parte Edwards, citing Baron Pollock’s reference to the arbitration agreement as containing the necessary pre-cut-off date obligation.

I have already noted that Haine v. Day is of particular importance in the present context because of the clear recognition by the Court of Appeal of the need to address questions as to the provability of a debt not merely as a technical problem of insolvency law, but in the context of the whole of the relevant legislation, including that which creates the liability, and any underlying EU Directive. At paragraph 7 the Court said this:

“7 Although the problem arises in the context of insolvency, it is essentially one of employment law, and particularly employment law in the context of an EU Directive. It seems that before Sir Donald the case was argued primarily as a technical problem of insolvency law rather than in the context of legislation intended to implement an EU Directive. Had the case been argued before him in the way in which it was argued before us, we think it at the lowest possible and, in reality probable, that he would have reached a different conclusion.”

In Casson v. Law Society [2009] EWHC 1943 (Admin) the Administrative Court (Richards LJ and Maddison J) sought to reconcile Glenister, Steele and Haine v. Day in relation to a liability of a solicitor to pay compensation to a client for pre-bankruptcy inadequate professional services, made the subject of a compensation award by an adjudicator after his discharge. At paragraph 36 Maddison J (with whom Richards LJ agreed) said this:

“The cases of Glenister and Steele, though inevitably based on their own facts, established a consistent principle of general application that where a court or tribunal has a discretion whether or not to make an award, any sum awarded in the exercise of that discretion does not exist as a debt or liability until the award is made. That principle was not affected by the decision of the Court of Appeal in Haine. The Court of Appeal in that case was concerned that principles of EU law in relation to collective redundancies should be implemented by the law of the UK. It considered it unreal to describe the making of a protective award in the circumstances of that case as depending on the exercise of a judicial discretion. The Court of Appeal carefully distinguished the cases of Glenister and Steele, and left untouched the general principle emerging from those cases to which I have referred.”

For completeness, I should mention that in El Ajou v. Stern [2006] EWHC 3067 (Ch) Kitchin J held, applying Glenister, that a claim based upon a post bankruptcy award of interest under section 35A of the Supreme Court Act 1981 did not create a provable debt in the bankruptcy, while in T&N (No 3) [2006] EWHC 1447 (Ch) David Richards J decided that a discretionary assessment of the amount of contribution pursuant to section 2 of the Civil Liability (Contribution) Act 1978 was discretionary only as to quantum, so that it did not prevent the contribution liability from being a provable debt: see paragraphs 71-2.

The most recent decision (so far as counsel were aware) on the meaning of provable debts is Unite (the Union) v. Nortel Networks UK Limited [2010] EWHC 826 (Ch), [2010] BCC 706, a decision about the administration of NNUK which began on 14^th January 2009. On 30^th March 2009 the Administrators terminated the employment of a number of NNUK’s employees, and 37 of them commenced proceedings in the Industrial Tribunal in Northern Ireland, for protective awards, unfair dismissal, breach of contract and discrimination. On a contested application for permission to pursue the claims during the administration moratorium, it was relevant for the judge (Norris J) to decide whether those claims constituted provable debts. He decided that all of them did. The breach of contract claim arose out of the obligation in each contract of employment, even though the breach occurred after the cut-off date. The claims for unfair dismissal and discrimination were analysed as claims based on statutory obligations imposed on NNUK in favour of each employee from the moment of the commencement of his or her employment, it being, again, irrelevant that the breaches of those statutory obligations all occurred after the cut-off date: see paragraph 25(b) to (d) inclusive. After a concise review of Glenister, Steele, Haine v. Day and Casson, the judge continued, at paragraph 33 as follows:

“I am not in any doubt about the matter: but if I were, I think the court should incline towards restricting the category of claims which are not provable. The consequence of the claims not being provable in bankruptcy in Glenister (above) Steele (above) and Casson (above) was that the claims could still be pursued against the discharged bankrupt. But a company does not survive its liquidation: so if a claim is not provable in the liquidation it is completely irrecoverable. It does not seem to me desirable (especially in relation to employees) to create a category of claim which cannot be dealt with in the insolvency process and is otherwise irrecoverable.”

It is difficult to treat that considerable line of authority as speaking entirely with one voice. In particular, it is difficult to see how there can be a “consistent principle of general application that where a court or tribunal has the discretion whether or not to make an award, any sum awarded in the exercise of that discretion does not exist as a debt or a liability until the award is made” (Casson) in the light of the approval given in Haine v. Day to the decision in ex parte Edwards, in which the arbitral tribunal’s costs discretion was just as general as the court’s discretion as to costs in Glenister. Similarly, it is difficult to see how Norris J’s view that the creation of an employment relationship by contract means that, from then on, all the employer’s statutory obligations are sufficient to render claims arising from a post cut-off date breach of them provable debts, squares with the much more guarded analysis in Haine v. Day, in which it was the pre-cut-off date breach of the employer’s statutory obligation which was treated as the relevant obligation for the purposes of Rule 13.12(1)(b), rather than the mere existence of the obligation.

It is however clear (and not in dispute) that the combined effect of Glenister and Steele is that without a qualifying legal obligation under sub-paragraph (b), there can be no contingent liability under sub-paragraph (a) sufficient to constitute a provable debt, short of an appeal to the Supreme Court based upon the analysis of the majority in Re Sutherland.

Mr Sheldon for the Nortel Trustees sought valiantly to persuade me that both Glenister and Steele were cases only about the bankruptcy equivalent of Rule 13.12(1)(a), so that nothing in either of those cases binds this court in relation to the analysis of a qualifying obligation under sub-rule (b). While I agree with him in relation to Glenister, I consider that Steele clearly is binding authority because it dealt with the alternative case there advanced under section 382(1)(b). Both Sir Martin Nourse (at paragraph 16) and Arden LJ (at paragraph 28) rejected the case advanced under subsection (b). More generally, and in particular when Arden LJ’s observations at paragraphs 22 to 24 are taken into account, I consider it clear that the obligation to which recourse may be had under Rule 13.12(1)(b) must be both a legal obligation (whether contractual or statutory), and one which gives rise to the liability relied upon. It must, on any interpretation of sub-rule (b) be an obligation owed by the insolvent person itself.

Mr Sheldon submitted that the circumstances which would ordinarily justify the imposition of the FSD regime constituted an a fortiori case of pre-cut-off date obligation, by comparison even with Unite (the Union) v. NNUK. The beneficiaries of the pension fund, to the trustees of which a debt arises under the FSD regime, are all pre-cut-off date employees of the group, and it is conduct by the employer and the target companies prior to the onset of insolvency (such as the under-funding of, and receipt of benefits from, the employer) which ordinarily justifies the imposition of an FSD. The FSD regime implements two European Directives.

That may be so, but there are in my judgment at least two fatal obstacles to the recognition of any qualifying obligation under Rule 13.12(1)(b) in such circumstances. The first is that, pending the issue of an FSD, the only obligations, contractual or statutory, owed to the employee beneficiaries of the pension scheme, or to the scheme trustees, are owed by the employer. It is indeed precisely because of the absence of any equivalent contractual or statutory obligations owed by the target companies that the novel FSD regime was created to plug the gap. The first ground upon which the Court of Appeal in Haine v. Day distinguished Glenister and Steele is therefore absent. At the very highest there was, perhaps, a moral obligation on the targets to provide proper support to the employer in connection with the pension scheme, or a commercial incentive to do so by way of minimising the risk of being subjected to the FSD regime in the future, but not a legal obligation.

The second obstacle is that by no stretch of the imagination can the complex and sophisticated discretionary process created by the FSD regime be described as one in which the Regulator has no real discretion not to issue an FSD or a CN against a target. Accordingly the second ground by which the Court of Appeal in Haine v. Day distinguished both Glenister and Steele is equally missing.

I have not reached that conclusion without an acute sense of discomfort. There is to my mind real force in Norris J’s analysis of the purpose of the provable debt regime in the paragraph from Unite (the Union) which I have quoted, in particular if the inability to recognise, as provable debts, claims arising out of the pre-cut-off date conduct of an insolvent company means that those claims fall into a black hole. Furthermore, Norris J’s conclusion that provable debts should be given as wide an interpretation as possible is nothing new. As long ago as 1871 James LJ described the policy behind the widening of the definition of provable debt in the Bankruptcy Act 1869 as follows, in Re Hyde (1871) LR 7 Ch App 28 at 31-32:

“The broad purview of this Act is, that the bankrupt is to be a freed man – freed not only from debts, but from contracts, liabilities, engagements and contingencies of every kind. On the other hand, all the persons from whose claims, and from liability to whom he is so freed are to come in with the other creditors and share in the distribution of the assets.”

One view of the many cases which have sought to establish an intelligible and clear dividing line between provable and non-provable debts since then is that the desire for certainty has over many years come to obscure that policy objective. It is, for example, incomprehensible on any purposive view why a post cut-off date costs award arising from a trial lost by the debtor before the cut-off date should not be a provable debt, whereas a costs award by an arbitrator should be, even where the arbitration was itself conducted after the onset of the debtor’s bankruptcy. But these are, alas, matters to be addressed either by Parliament, by the Insolvency Service (which initiates changes to the Rules), or by a higher court.

My conclusion that an FSD issued after the commencement of the applicant companies’ administration cannot give rise to a provable debt in that administration pursuant to the insolvency regime makes it necessary to consider the question whether an FSD issued to a target while in administration would give rise to a provable debt in any subsequent liquidation of the target which followed that administration. The case of the Regulator and its supporters that it would do so runs along the following lines:

To that formidable (if technical) argument Mr Dicker for the Administrators advanced two obstacles. The first was that even the issue of a CN depends upon the exercise of a general discretion by the Regulator: see section 47(3) and (4) of the 2004 Act, thereby falling foul of the supposed general principle established in Glenister and Steele, and recognised in Casson. The second obstacle is that the amendments to Rule 13.12 designed to replace the liquidation date with the administration date as the cut-off date for a liquidation immediately preceded by an administration were only omitted from the earlier amendments in 2003 by an obvious mistake, which the court can and should itself deal with by a purposive construction of the pre-2010 version of Rule 13.12, so that, in effect, it should be read and construed as if it had always contained, from 2003, provisions substantially the same as those belatedly introduced in 2010.

In my judgment, if an FSD is issued to a target company while in administration, and a CN is issued to the same target after the administration had been immediately followed by a liquidation, then the CN would create a provable debt in that target’s liquidation. My reasons follow.

First, I do consider that the issue of an FSD imposes a legal obligation, albeit of an unusual kind, upon the target company of which the beneficiaries are the scheme members employed by the associated employer company, represented by the scheme trustees. The legal obligation is to secure the provision of financial support for the scheme by arrangements which are reasonable in the circumstances: see sections 43(3) and 45(3) of the 2004 Act. The FSD is, in short, precisely that which converts what may have been a previous moral obligation or commercial incentive into a legal obligation of the requisite type for the purposes of Rule 13.12(1)(b).

Secondly, I am not persuaded that the undoubted discretion whether or not to issue a CN, conferred on the Regulator by section 47(3) of the 2004 Act is sufficient to deprive what section 49(3) describes as the consequential “debt due from the person to the trustees or managers of the scheme” of the status of a provable debt. In my judgment it comes on the same side of the difficult dividing line as is occupied by the arbitrator’s costs discretion in Re Edwards and the Tribunal’s discretion in Haine v. Day. This is not so much because it is a discretion only capable of being exercised one way, but because the discretion is in terms about how to deal with a previous non-compliance with the pre-existing legal obligation: see section 47(1) of the 2004 Act. While it is true that, read literally, section 47(1) speaks in terms of non-compliance with an FSD generally, rather than necessarily by the intended target of the CN, I consider it clear from a reading of sections 47 to 50 as a whole that the Regulator’s focus in relation to the contemplated target of a CN will inevitably be upon the extent of that target’s non-compliance with the FSD: see in particular section 47(4)(a) and (d), section 49(8) and section 50(9).

Put another way, I consider that the two grounds given in Haine v. Day for distinguishing both Glenister and Steele are each self-sufficient. Where, as in Haine v. Day and the present case, the discretion arises in the context of the enforcement of compliance, or the provision of a remedy for non-compliance, with a pre-existing legal obligation, then Rule 13.12(1)(b) is satisfied. Where by contrast it is the affirmative exercise of the discretion which first creates any legal obligation of the relevant type (as in Glenister, Steele and Foots v. Southern Cross) then its existence necessarily prevents a provable debt arising under Rule 13.12, even if the discretion is exercised by reference to matters which occurred prior to the cut-off date. I also consider that this analysis plainly better serves the underlying rationale for the provable debt regime, as explained by Norris J in Unite (the Union) v. NNUK.

I do not consider that I am bound by Glenister or Steele, still less by Casson, to reach a contrary conclusion. In Glenister there was no pre-cut-off date legal obligation. In Steele the only possible legal obligation was a pure common law duty to make restitution which was not the obligation which gave rise to the liability in question. To the extent that my conclusion is incompatible with the analysis in Casson of the ratio of Glenister and Steele, I am not bound by it and, with great respect, I do not agree with it.

Mr Dicker’s submission that I should construe the pre-2010 version of Rule 13.12 as if it had always provided for an administration cut-off date in the context of proof of debts in an immediately following liquidation calls for further analysis. Proof of debt and distribution to creditors within administration was introduced by the Enterprise Act 2002. This led directly to the amendment of Rule 13.12 by the addition of sub-paragraph (5), to which I have already referred. The Enterprise Act also introduced the facility for a company to move from administration into liquidation, and from liquidation into administration. This led to further amendments to the regime for quantification and proof of debt, in the Insolvency (Amendment) Rules 2005 (SI 2005 No 527).

In an Explanatory Note, described in its heading as not being part of the Rules, the Insolvency Service said this:

“As a result of the changes made to the law on administration by the Enterprise Act 2002 (c.40) a company can move between liquidation and administration or between administration and liquidation. Both of these procedures enable creditors to prove their debts at the date of the administration or liquidation respectively. By way of clarification of the existing rules, the amendments provide that the relevant date is the date of the first insolvency procedure commenced. The Rules affected are:-

• Rules 2.86, 2.87, 2.88, 2.89, 4.91, 4.92, 4.93 and 4.94”

Part 2 of the Rules concerns administration procedure. Part 4 concerns winding up. Rules 2.86 and 4.91 both concern the debts in foreign currency, and provide for conversion into sterling on the relevant cut-off date. Rules 2.87 and 4.92 concern periodic payments, which are provable in relation to amounts due and unpaid up to the relevant cut-off date. Rules 2.88 and 4.93 concern interest, and prohibits proof of interest in respect of any period after the relevant cut-off date. Rules 2.89 and 4.94 concern future debts, and enable a creditor to prove for such a debt even if not due on the relevant cut-off date. In each case, the amendments introduced in 2005 do, as the Explanatory Note states, provide that, in relation to any proof of debt in an insolvency process (be it liquidation or administration) which is immediately preceded by the other type of insolvency process (administration or liquidation) the cut-off date is that applicable to the first of those two consecutive insolvency processes.

Foreign currency, periodical payments, interest and future debts may fairly be described as subordinate parts of the proof of debt regime. Inexplicably, the 2005 Amendments made no equivalent adjustment to the main rule concerning proof of debt, which is of course buried away in the interpretation section, at Rule 13.12. It took five years before an equivalent amendment was made to Rule 13.12, in the form of Schedule 1 to the Insolvency (Amendment) Rules 2010, to which I have already referred. Furthermore, although many of the amendments introduced in 2010 were, by the transitional provisions in paragraph 2 of Schedule 4, made applicable quasi-retrospectively to all insolvencies, even if commenced before April 2010, the amendments to Rule 13.12 were made applicable only where the first of two consecutive insolvency processes commenced after that date.

The failure to amend Rule 13.12 in 2005, coupled with the non-retrospectivity of the 2010 amendment, is at least odd in relation to companies first going into successive insolvency processes between those two dates. Rule 13.12 unmistakably contemplates that, in the subsequent process, the relevant cut-off date is the commencement of that process, whereas the 2005 Amendments contemplate that, for particular aspects of the process of proof and quantification, the relevant cut-off date is that applicable to the earlier process. I can think of no obvious reason for having two different cut-off dates in relation to the same process of proof.

Mr Dicker for the Administrators suggested that I should take the bull by the horns and construe Rule 13.12, as from 2005, as if it had always contained the amendment introduced non-retrospectively in 2010, on the ground that the failure to introduce it earlier was an obvious drafting mistake. He referred me to the following passage in Lord Nicholls’s speech in Inco Europe Limited v. First Choice Distribution [2000] 1 WLR 586, at 592:

“It has long been established that the role of the courts in construing legislation is not confined to resolving ambiguities in statutory language. The court must be able to correct obvious drafting errors. In suitable cases, in discharging its interpretative function the court will add words, or omit words or substitute words. Some notable instances are given in Professor Sir Rupert Cross’s admirable opuscule, Statutory Interpretation, 3^rd ed. (1995), pp. 93-105. He comments, at p. 103:

“In omitting or inserting words the judge is not really engaged in a hypothetical reconstruction of the intentions of the drafter or the legislature, but is simply making as much sense as he can of the text of the statutory provision read in its appropriate context and within the limits of the judicial role.”

This power is confined to plain cases of drafting mistakes. The courts are ever mindful that their constitutional role in this field is interpretative. They must abstain from any course which might have the appearance of judicial legislation. A statute is expressed in language approved and enacted by the legislature. So the courts exercise considerable caution before adding or omitting or substituting words. Before interpreting a statute in this way the court must be abundantly sure of three matters: (1) the intended purpose of the statute or provision in question.; (2) that by inadvertence the draftsman and Parliament failed to give effect to that purpose in the provision in question; and (3) the substance of the provision Parliament would have made, although not necessarily the precise words Parliament would have used, had the error in the Bill been noticed. The third of these conditions is of crucial importance. Otherwise any attempt to determine the meaning of the enactment would cross the boundary between construction and legislation: see per Lord Diplock in Jones v. Wrotham Park Settled Estates [1980] A.C. 74, 105-106. In the present case these three conditions are fulfilled.

Sometimes, even when these conditions are met, the court may find itself inhibited from interpreting the statutory provision in accordance with what it is satisfied was the underlying intention of Parliament. The alteration in language may be too far-reaching. In Western Bank Ltd v. Schindler [1977] Ch. 1, 18, Scarman L.J. observed that the insertion must not be too big, or too much at variance with the language used by the legislature. Or the subject matter may call for a strict interpretation of the statutory language, as in penal legislation.”

Sorely tempted though I am to accede to Mr Dicker’s invitation, I do not consider that it would be a permissible exercise of the court’s interpretative function. My reasons follow.

First, while I think it probable that the failure to amend Rule 13.12 in 2005, in the form subsequently achieved in 2010, was a mistake, the result was that, viewed as a distinct (and very important) provision in the Rules, Rule 13.12 was simply not amended at all in 2005. On its face, it continued to make exactly the same provision as to cut-off date both in administration and liquidation as it had done since 2003, when sub-paragraph (5) was added. The conclusion that the omission of any amendment was a mistake derives only from an appreciation of the purpose behind the amendments to Parts 2 and 4 of the Rules, as explained by the Explanatory Note. This is not therefore a question of correcting a drafting mistake in an amendment actually made to Rule 13.12, but rather the insertion of a whole new and important provision which is, quite simply, not there.

Secondly, the alteration in language required by Mr Dicker is, as is apparent from the widespread amendments to Rule 13.12 actually introduced in 2010, very far reaching in its effect. Rule 13.12 is the primary rule governing the identification of the cut-off date for the purposes of proof in insolvency processes. To construe the primary rule as having been implicitly amended by reference to amendments made only to subsidiary aspects of the same regime would be, to my mind, to allow the tail to wag the dog.

Thirdly, although I think that the omission was probably a mistake, I am by no means “abundantly sure” of that, which is the requisite standard laid down by Lord Nicholls. It may be that the very experienced and sophisticated draftsman of the 2005 Amendments (within the Insolvency Service) thought it sensible first to prescribe a unified cut-off date in relation to particular aspects of the proof of debt regime, and leave a wholesale amendment of the primary rule to be decided upon after reviewing the feedback about the cautious start, in the light of experience.

Fourthly, there is in my view a subtle dividing line between dealing with drafting mistakes by construction, which is a task for the court, and dealing with them by subsequent amendment, which is a task for the legislature. In my judgment the present task falls clearly on the legislative side of that dividing line. The statutory provisions in question consist of important parts of a highly sophisticated and technical code set out in rules which have to be applied on a daily basis by busy insolvency practitioners who ought, in principle, to be entitled to read the rules and apply them in accordance with their language, at least where it is unambiguous. Rule 13.12, in its pre-April 2010 form, is unambiguous in that respect. It clearly provides that the cut-off date in a liquidation, and in an administration, is in each case the date of the commencement of that insolvency process. No ordinary process of construction would lead to the conclusion that, where for example liquidation immediately follows an administration, proof in that liquidation is governed by the administration cut-off date. On the contrary, a careful comparison between Rule 13.12 and the amended rules in Parts 2 and 4 to which I have referred would suggest that the legislature had quite deliberately made a cautious and limited alteration to the general principle, in relation only to specific aspects of the proof regime.

I have no doubt that there will have been numerous cases between 2005 and 2010 of administrations followed by liquidation, in which the liquidators have dealt with creditors’ claims upon the basis that the relevant cut-off date was the commencement of the liquidation, as unambiguously described by Rule 13.12 in its then form. The decision not to make the 2010 Amendments to Rule 13.12 applicable to existing insolvency processes demonstrates to me that the draftsman (and therefore the legislature) must have thought so as well, and proceeded on the basis that, from 2005 until 2010, Rule 13.12 did indeed mean what it clearly said.

To treat Rule 13.12 as having always contained, since 2005, the 2010 Amendments would, in relation to such intervening insolvency processes, cause an unacceptable degree of potential chaos. In particular, creditors to whom distributions were made on an assumption that their common law claims satisfied Rule 13.12 because they flowed from obligations incurred before the liquidation cut-off date, might find themselves subjected to recovery claims by reference to a conclusion about construction which, by backdating the cut-off date to the commencement of any immediately prior administration, left them out in the cold. Similarly, statutory creditors whose claims were, under the Toshoku principle, converted from provable claims to expense claims (such as local authority rates creditors) upon the cut-off date, but who received a small dividend in respect of liabilities falling due between the onset of administration and the commencement of a subsequent liquidation, would be enabled to claim that the earlier administration cut-off date now declared to be applicable to their claims, as a matter of necessarily retrospective construction of Rule 13.12, meant that they had all been underpaid. The non-retrospectivity of the 2010 Amendments neatly deals with all those difficulties. It would in my judgment be irresponsible for the court, by a process of construction, now to interfere with that sensible result.

Mr Dicker submitted that orthodox principles of construction required the 2005 Amendments, and their effect on the Insolvency Rules as a whole, to be addressed without regard to what the legislature later did in 2010. That may be strictly true, but the legislature’s approach to the 2010 Amendments demonstrates to my mind, even if only by way of example, why there is that dividing line between the court’s and the legislature’s responsibilities in relation to the correction of mistakes.

I have thus far set out in detail what I conceive to be the relevant technical insolvency rules as to provable debt, both now and (since they have not significantly changed since 1865) in 2004. They form the main part of the necessary backdrop to the insolvency priority aspect of the interpretation of the FSD regime. I have also described, as the second part of that necessary backdrop, the consequences, from a technical insolvency perspective, of a debt being neither provable nor an expense. As described by David Richards J in T & N (No 2), it could be pursued only in the very unlikely event that there was a surplus available after payment in full to all unsecured creditors. That is what I have called the black hole, although on analysis it might perhaps better be described as a dark grey hole.

Since no one suggests that financial liabilities arising from the FSD regime could possibly be preferential debts, the remaining part of the necessary backdrop is that part of the technical insolvency regime which deals with liquidation expenses. As to this, there was also a keen and protracted debate between counsel as to the precise meaning and effect of the House of Lords’ decision in Toshoku. That decision was handed down in February 2002, and must be taken to have formed a relevant part of Parliament’s understanding of the technical insolvency legislation when formulating the FSD regime in 2004.

The issue between counsel may be summarised as follows. For the Regulator and its supporters, it was submitted that the House of Lords laid down a clear and simple rule that, wherever statute created a monetary liability which applied either expressly to a company in liquidation, or to a company regardless whether it was or was not in liquidation (and now, after Exeter v. Bairstow, administration), then if that obligation was not a provable debt, it must be a liquidation (or administration) expense.

For the Administrators Mr Trower and Mr Dicker both submitted that the true principle was narrower than that. It was not sufficient that the statutory liability should merely be a liability of a company in an insolvency process, and not provable. It must also be capable of being described as a “necessary disbursement”. It will be an expense of that type if, but only if, Parliament can be shown to have intended that the debt should not merely be a liability of a company in an insolvency process, but also payable by the company while in that insolvency process, and therefore payable by the office-holder (liquidator or administrator) with all the super-priority consequences that would flow from that analysis. The Administrators maintain that this has been identified as the true ratio of Toshoku by two subsequent first instance cases in the Companies Court, namely re Allders Department Stores Limited [2005] EWHC 172 (Ch); [2005] BCC 289 by Lawrence Collins J and Exeter v. Bairstow (supra) by David Richards J. It is submitted that I should therefore follow those two decisions, unless convinced that they are wrong.

The issue in Toshoku was whether the list of liquidation expenses in Rule 4.218(1) was determinative of what was a liquidation expense (as the Crown submitted) or was merely an “outer envelope within which expenses were contained” (as submitted for the liquidators): see and contrast paragraphs 5 and 12. It was common ground in the House of Lords (but not in the courts below) that, as a matter of construction, the relevant liability to corporation tax fell within Rule 4.218(1)(m): see paragraphs 10 and 11. The question was whether this compelled the liquidator to pay that tax as an expense, or merely gave rise to a second question, namely whether it also had to pass a judge-made test which Nicholls LJ in Re Atlantic Computer Systems plc [1992] Ch 505, at 520, called the “liquidation expenses” principle: see again paragraph 12. The House of Lords held that no such second question arose, expressly overruling Re Atlantic Computer Systems plc (on this point) and Re Kentish Homes Limited [1993] BCLC 1375.

The statutory liability under review in Toshoku was one of that type in relation to which, as I have already described, Parliament expressly directed its mind to the question whether the liability should be paid by a company in liquidation. So was the liability to pay corporation tax on chargeable gains arising in the winding up, reviewed in Re Mesco Properties Limited [1979] 1 WLR 558 by Brightman J which, at paragraph 42 in Toshoku, Lord Hoffmann adopted as the basis for his opinion. He continued:

“The statute expressly enacts that a company is chargeable to corporation tax on property of gains arising in the winding up. It follows that the tax is a post-liquidation liability which the liquidator is bound to discharge and it is therefore a “necessary disbursement” within the meaning of the Insolvency Rules.”

I consider that Lord Hoffmann used the expression “post-liquidation liability” as meaning a liability which, because it arose after the cut-off date, was not a provable debt within the meaning of Rule 13.12. Thus far, Lord Hoffmann’s speech is consistent with either of the rival interpretations of it advanced before me.

But Lord Hoffmann’s reasoning deserves closer inspection. The Crown’s submission was that: “it is a sum which by statute is payable by a company in respect of profits or gains arising during a winding up. The liquidator is obliged to pay it. It is therefore a “necessary disbursement” which the liquidator has to make in the course of his administration. That is an end of the matter.” (paragraph 7). At paragraph 8 Lord Hoffmann described that approach as supported by high authority, beginning with Re Mesco Properties Limited. At paragraphs 29 and 30 he identified from Exhall Coal Mining Co Ltd 4 De G J & S 377 and Lundy Granite Co LR6 Ch App 462 a principle which was:

“One which permits, on equitable grounds, the concept of a liability incurred as an expense of the liquidation to be expanded to include liabilities incurred before the liquidation in respect of property afterwards retained by the liquidator for the benefit of the insolvent estate.”

He continued, in paragraph 30:

“It was not, however, a general test for deciding what counted as an expense of the liquidation. Expenses incurred after the liquidation date need no further equitable reason why they should be paid. Of course it will generally be true that such expenses will have been incurred by the liquidator for the purposes of the liquidation. It is not the business of the liquidator to incur expenses for any other purpose. But this is not at all the same thing as saying that the expenses will necessarily be for the benefit of [the] estate. They may simply be liabilities which, as liquidator, he has to pay. For example, there will be the fees payable to fund the Insolvency Service, ranking as paragraph (c) in Rule 4.218(1), where the benefit to the estate may seem somewhat remote. There would be little point in a statute which specifically imposed liabilities upon a company in liquidation if they were payable only in the rare case in which it emerged with all other creditors having been paid.”

At paragraphs 31 to 34 he tested that analysis by reference to established authorities on rates, where liability depends upon rateable occupation, pursuant to statutory criteria which make no express reference to whether a company is, or is not, in liquidation. He noted that the decision of Kay J in Re Watson Kipling & Co (1883) 23 Ch D 500 that liability for post cut-off date rates depended on the Lundy Granite principle had been overruled in Re National Arms and Ammunition Co (1885) 28 Ch D 474, at 480 and 482 which was itself followed in Re Blazer Fire Lighter [1895] 1 Ch 402. At paragraph 34 he concluded, by reference to examples of cases in which a company had remained, after the cut-off date, in rateable occupation otherwise than for the benefit of the insolvent estate that:

“The rates would have been an obligation incurred after the liquidation which (unlike the rent) was not provable and was therefore payable in full.” (my underlining)

He concluded his analysis of the authorities by reference to Re Kentish Homes (supra) in which the liability in question was a post-liquidation liability to community charge on empty flats. Again, the statute made no express reference to whether the liability was payable by a company in liquidation, but it was common ground in Re Kentish Homes that the company was the chargeable person in respect of the relevant flats for the period in question (paragraph 40), and that this was the criterion for liability. Applying the Lundy Granite principle, Sir Donald Nicholls V-C concluded that, because the flats were merely owned by the company, and had not in any sense been retained or used for the benefit of the liquidation, the community charge was not a liquidation expense. In concluding that Re Kentish Homes should be overruled, Lord Hoffmann said this at paragraph 41:

“In the first place, the question of whether the community charge should count as an expense of the liquidation was not a matter for the judge’s discretion. It depended upon whether it came within one of the paragraphs of Rule 4.218. In my opinion if, as was common ground, the company was the chargeable person, it was a necessary expense which came within (m). If, therefore, the liquidator had sufficient assets after satisfying the liabilities coming within paragraphs (a) to (l), he was obliged to pay it. Secondly, the Lundy v. Granite Co principle had no relevance. The liability did not arise out of a pre-liquidation obligation. If it came within the language of paragraph (m), it was a liquidation expense.”

I consider that in overruling Re Kentish Homes, Lord Hoffmann clearly did not regard the question whether a statutory liability was a necessary disbursement, and therefore an expense, as turning on the question whether the statutory language imposing the liability made express reference to it being payable by a company in liquidation. It was sufficient for Lord Hoffmann’s analysis that the statute imposing the community charge, like the statutes imposing a liability for rates, merely defined the person liable, without any reference to liquidation, as the person chargeable by virtue of its ownership of the property (community charge) or the person in rateable occupation (rates). It was enough that the statute imposed a liability on a company, whether or not in liquidation.

Lord Hoffmann concluded with a comment upon the submission of counsel for the liquidators that, both in the instant case and more generally, the imposition of post-liquidation liabilities as liquidation expenses was capable of causing injustice. At paragraph 45 he said:

“The injustice, if any, does not arise because liabilities imposed upon a company in liquidation have priority as expenses of the liquidation, but because it may be unjust to impose certain liabilities upon companies in liquidation.”

After reference to liabilities under environmental legislation, and by reference to Re Mineral Resources Limited [1999] 1 All ER 746, he concluded at paragraph 46:

“In my opinion, the question of whether such liabilities should be imposed upon companies in liquidation is a legislative decision which will depend upon the particular liability in question.”

Pausing there, my reading of Lord Hoffmann’s speech is that, although on the particular facts in Toshoku the statutory liability was imposed expressly on a company in liquidation, he regarded the “necessary disbursements” category of liquidation expense as including all statutory liabilities imposed on a company in liquidation whether expressly, or by a criterion for liability (such as rateable occupation or property ownership) which made no distinction between companies which were, and were not, in liquidation. It remains to consider whether my interpretation is to be displaced by the analysis in Allders, and in Exeter v. Bairstow.

The question in Allders (supra) was whether, if administrators of companies in a group terminated contracts of certain company employees, the consequential redundancy and unfair dismissal liabilities imposed by statute would be expenses of the administration. Mr Trower and Mr Sheldon, both of whom appeared in Allders, told me that the administrators needed, and indeed obtained, the court’s directions as a matter of urgency. The administrators had only been appointed on 26^th January and 4^th February 2005 respectively, and judgment was delivered on 16^th February 2005.

At paragraph 20 Lawrence Collins J noted that, pursuant to paragraphs 9 to 16 of Schedule 6 to the Insolvency Act, certain liabilities to employees are treated as preferential debts, (including protective awards), but only in respect of the period of four months before the cut-off date. At paragraph 21 he noted that liabilities payable by virtue of paragraph 99 of Schedule B1 to the Insolvency Act in priority to the administrators’ expenses are those liabilities adopted by the administrators after fourteen days from appointment and which are “wages or salary”. He concluded that neither the redundancy nor unfair dismissal payments were wages or salary.

He then decided the question before him in paragraph 24, almost every sentence of which is relied upon by the Administrators in their various submissions, and it is therefore worth quoting in full:

“I do not consider that the statutory liabilities for redundancy payments or unfair dismissal claims would be “necessary disbursements” for the purposes of r.2.67(1)(f). First, it would be inconsistent with the scheme of the legislation if the payments referred to in Sch.6 were to be treated as preferential, and yet all other employee-related payments are to be paid as an expense of the administration. That would be to give the Sch.6 payments (which include protective awards) a lesser priority than other types of payments, when the policy appears to have been to give them a greater priority. Secondly, there is nothing in my judgment in Re Toshoku Finance (UK) plc [2002] B.C.C. 110; [2002] 1 WLR 671 which requires a different conclusion. It is not the ratio of that case that any liability imposed on a company which is not provable as a debt is thereby rendered a “necessary disbursement”. The context in which Lord Hoffmann referred to the fact that certain debts could not be proved shows that he was justifying the treatment of certain debts as expenses, and not offering a definition of what liabilities were disbursements. Even if that were the crucial test for winding up, there would be no reason to apply it to administration. Although the current regime envisages a distribution to secured and preferential creditors without a subsequent liquidation, in the normal case of an administration which does not succeed in rescuing the company, the company will go into liquidation and the statutory payment obligation will be a provable debt under r. 13.12. Finally, a construction of r.2.67(1)(f) which applied it to statutory redundancy payment liabilities and other statutory liabilities would have such adverse policy consequences on the administration regime that it is impossible to see that such a result could have been intended.”

A number of points emerge from that paragraph. The first is that Lawrence Collins J was certainly not assuming that the redundancy and unfair dismissal liabilities would, if not administration expenses, fall down the black hole which would follow from them not being provable debts either. On the contrary, he appears to have assumed, (by an interpretation of Rule 13.12 in its then form with which, after lengthy analysis, I concur), that a post administration dismissal immediately followed by a liquidation would mean that the liabilities arising from the dismissal would be provable debts in the subsequent liquidation. He noted that distribution could be made in the administration itself, but regarded distribution in a subsequent liquidation as the “normal case”.

Secondly, his decision preceded by some years Norris J’s much simpler conclusion in Unite (the Union) that redundancy and unfair dismissal payments resulting from a post cut-off date dismissal were always provable debts, even in the insolvency process in which the dismissal occurred, because they arose out of statutory obligations incurred by the company from the moment when it employed the relevant employees. Thirdly, it is literally true that it was not the ratio of Toshoku that “any liability imposed on a company which is not provable as a debt is thereby rendered a necessary disbursement” because, as appears from examples provided by David Richards J in Exeter v. Bairstow, there was, at least then, a major class of common law liability which was neither provable as a debt nor recoverable as a liquidation expense, namely contingent liability in tort where the cause of action accrued after the cut-off date.

Fourthly, it is also true that, if the ratio of a case is identified by reference to that which it specifically decided, Toshoku was in fact a case about a statutory liability where Parliament expressly declared it to be payable by a company in liquidation. Fifthly, Lawrence Collins J’s view that there would be no reason to apply the Toshoku principle to administration has, of course, since been rejected, by David Richards J in Exeter v. Bairstow, at least where the language of the relevant administration expenses rule is substantially the same as that used in the corresponding rule for liquidation. In that context Rule 2.67(1)(f) (for administration) is substantially identical to Rule 4.218(1)(m) (for liquidation), in relation to necessary disbursements. Finally, it is undoubtedly correct that the issue in Toshoku did not require a comprehensive definition of all types of liability which might constitute disbursements.

Nonetheless, I have not been persuaded by anything in Allders that I am wrong in my analysis that the route by which Lord Hoffmann reached his decision in Toshoku stemmed from his conclusion that the “necessary disbursements” category of expense is applicable to all cases in which statute imposes a liability on a company in liquidation, regardless whether the legislation in question did so expressly, or merely by using a criterion for liability that did not distinguish between companies which are, or are not, in an insolvency process.

In Exeter City Council v. Bairstow (supra) the question was whether non-domestic rates in respect of premises occupied by a company in administration were payable as an administration expense. The central issue was whether the Toshoku principle, derived from a case about liquidation, had any application to administration. David Richards J’s conclusion was that it did, at least where the language of both expense regimes was the same. He made two relevant references to Toshoku. First, at paragraph 77, having quoted most of paragraph 24 of Lawrence Collins J’s judgment in Allders, he continued:

“77 It appears to me that the principal basis of the decision was that because of the special treatment of certain categories of employment-related claims under para.99 and under Sch.6, it would be inconsistent to treat further categories as expenses under r.2.67. I concur with that analysis. I also concur with the view that it is not the ratio in Re Toshoku Finance (UK) Plc that any liability imposed on a company which is not provable as a debt is thereby rendered a necessary disbursement. Examples of liabilities which fall into neither category are liabilities in tort where the cause of action arises after the liquidation (see Re T&N Ltd [2005] EWHC 2870; [2006] 1 W.L.R. 1728, subject now to the limited exception in r.13.12(2)(b)) and, it would seem, orders for costs made after the commencement of liquidation in respect of pre-liquidation litigation (Glenister v Rowe (Costs) [2000] Ch. 76, a decision on the rules for proof in bankruptcy).”

Secondly, after concluding in paragraph 83 that, both in liquidation and in administration, rates were payable by the company, he said, at paragraph 84:

“The observations on rates, as being necessary disbursements within r.4.218(1)(m), in Re Toshoku Finance (UK) Plc may, strictly speaking, be obiter but all members of the House of Lords concurred in Lord Hoffmann’s speech and Mr Briggs did not suggest they were wrong. Just as rates are payable in a liquidation as a necessary disbursement, so in my judgment they are payable in an administration.”

In relation to paragraph 77 it is to be noted that neither of the examples there given of non-provable liabilities which are nonetheless not expenses (and which therefore fall into a black hole unless rescued by a subsequent insolvency process) were cases of statutory liability. The first was a common law liability in tort, and the second was a liability to pay a costs order ordered by the Court of Appeal. As to paragraph 84, David Richards J’s analysis broadly accords with my own, in relation to what Lord Hoffmann said about rates. Furthermore, the judge’s decision in Exeter v. Bairstow was that rates, a liability imposed by reference to a criterion that did not distinguish between a company which was or was not in an insolvency process, was nonetheless payable as an administration expense, pursuant to the Toshoku principle. It follows that I read David Richards J’s judgment in Exeter v. Bairstow as confirming, rather than detracting from, my own analysis of Lord Hoffmann’s reasoning in Toshoku, even if that part of the reasoning may, strictly, not have been part of the ratio.

Before leaving Exeter v. Bairstow, I note that David Richards J went out of his way to acknowledge that, when construing legislation for the purpose of understanding whether a particular form of liability is imposed as an expense upon a company in an insolvency process, it will be appropriate to have regard to the underlying objectives of the modern insolvency code, including in particular the rescue culture: see in particular paragraphs 56 to 59. In that case the judge acknowledged that, in the light of substantial evidence deployed by Mr Nicholas Briggs as advocate to the court, his conclusion that rates were indeed an expense would probably have an adverse effect on the rescue culture in at least some cases: see paragraph 68. Nonetheless, this did not dissuade him from that conclusion. There may be circumstances where one public policy (not limited to that which underpins rates) comes into collision with the policy objectives of the insolvency legislation. It is for the legislature, not the courts, to decide how that collision is to be resolved.

I therefore conclude that the Toshoku principle does indeed establish as a general rule that where by statute Parliament imposes a financial liability which is not a provable debt on a company in an insolvency process then, unless it constitutes an expense under any other sub-paragraph in the twin expenses regimes for liquidation and administration, it will constitute a necessary disbursement of the liquidator or administrator. That is the general rule, whether the statute expressly refers to companies in an insolvency process as being subject to the liability, or whether the statute achieves the same result by using a criterion for liability which is insolvency neutral. Any other conclusion would in my judgment attribute an excessive weight to the linguistic method by which different legislation achieved the same result, namely that the statutory obligation in question is a liability of a company in an insolvency process.

Nonetheless, an understanding that this is in general the consequence of the technical provisions of insolvency legislation, and that the imposition of the liability as an expense may be detrimental to the achievement of the rescue objective behind that legislation, will be material to any process of statutory construction designed to ascertain whether the statutory provisions, taken as a whole, do impose liability on a company in an insolvency process and, if they do, the priority thereby afforded to that liability.

It is worth noting in passing that, so far as I have been able to ascertain, in no decided case has it actually been concluded that a statutory liability imposed on a company in an insolvency process nonetheless falls down the black hole constituted by it being neither an expense nor a provable debt. The costs liabilities in Glenister and Foots v. Southern Cross were not, in any real sense, statutory. They arose from court orders, even though the jurisdiction of the relevant courts to make those orders may have been codified in statute. The liabilities in Steele and Casson v. The Law Society were statutory, but they could only be imposed on individuals rather than companies, with no consequential black hole. The liability in Allders was (or at least turned out to be) a provable debt. No case has yet confounded Lord Hoffmann’s commonsense view that there would be little point in statute imposing liabilities on a company in liquidation if they were payable only in the rare case in which it emerged with all other creditors having been paid.

I therefore approach the construction of the FSD regime on the necessary (albeit hypothetical) assumption that Parliament must have been aware when imposing it that, if it intended to impose financial liabilities on a target company in an insolvency process, but left the question as to the priority of that obligation in the insolvency to be decided by the technical provisions of the Insolvency Act and the Rules, then:

I have described that assumption about Parliament’s knowledge of the law as hypothetical because, in reality, the FSD regime was almost certainly drafted by pensions experts rather than insolvency experts, and the complete failure of the 2004 Act to make any reference to the effect of the regime upon insolvent companies suggests that issues as to priority of consequential financial obligations of a target company in an insolvency process may never have crossed the draftsman’s mind. A number of the relevant cases on the subject had yet to be decided. Even if he had sought and obtained advice as to the effect of leaving the consequences of such issues to the technical provisions of the insolvency legislation, and received advice along the lines that I have just described, he might well be forgiven for having found it difficult to suspend his disbelief since, on any view, the matters giving rise to the imposition of the FSD regime upon a target would almost inevitably have occurred, as I have explained, during a period when the target was a going concern, as part of a group which included the under-funded employer or service company.

My analysis thus far suggests that the outcome of the central questions in this case turns on the answers to the following two questions, asked consecutively:

Does the FSD regime apply so as to impose financial liabilities on target companies in an insolvency process?

In my judgment, this first question is clearly to be answered in the affirmative, and counsel for the Administrators did not, at least with any vigour, argue the contrary. Rather, their submission was that the FSD regime was primarily focused upon solvent targets, as a stepping stone towards their submission that the 2004 Act, by implication, left any consequential financial obligations of an insolvent target to fall down a black hole, being neither expenses nor provable debts.

In my judgment, the FSD regime is a paradigm example of legislation which imposes, as criteria for corporate liability, matters which do not distinguish between companies which are, and which are not, in an insolvency process. As I have described in detail, section 43 of the 2004 Act imposes conditions upon the power to issue an FSD at all, by reference to essentially historic facts about the employer and its relationship with other group companies, as at the look-back date: see section 43(2) and section 44. It imposes a further condition in relation to the issue of an FSD to a particular target which is also historical, by reference to the same look-back date, namely whether it was then either the employer in relation to the scheme, or (in a corporate context) a company which was then connected with, or an associate of, the employer: see section 43(5)(a) and (6)(a) and (c). The critical provision is that this condition needs to be satisfied at “the relevant time” which is the look-back date. Even at the look-back date, it is not necessarily fatal that either the employer or the target might be in an insolvency process, least of all in administration, when its business may still be being carried on under the direction of its administrators. More importantly, since the look-back date may be as much as two years before the determination by the Regulator to issue the FSD, in the context of a group which has attracted the Regulator’s attention because of its financial difficulties, it is eminently possible that it will have moved from being a going concern at the look-back date to a state of multiple insolvency processes by the FSD issue date. That is, in fact, what happened in all three of the only cases in which, thus far, the Regulator has determined to issue an FSD.

It would to my mind be quite bizarre to attribute to Parliament an intention that, for example, an FSD process embarked upon by the Regulator in relation to a group in apparent financial difficulty could be rendered wholly ineffective by the targets’ descent into insolvency processes before the Regulator had conducted the necessary research, issued warning notices, considered representations from directly interested parties, held a determination hearing, and survived a reference to the Tribunal, so as to be in a position for the first time to issue an FSD.

Furthermore, the fact that potential targets might by then be in an insolvency process by no means leads to the conclusion that, in order to achieve the objective of protecting the PPF from excessive exposure, the target’s assets should not be made available to an extent reasonable in all the circumstances for the purposes of securing financial support to the under-funded scheme. A target may, for example, be in administration due to commercial insolvency arising from illiquidity of assets, while nonetheless balance sheet solvent, with a view to the creation of a moratorium during which the liquid assets can be realised at full value for the purpose of paying its creditors in full. Secondly, the target may, even if unable to pay its creditors in full, have a much greater asset base with which to meet creditors’ claims than the employer has, with which to meet its creditors’ claims, including the section 75 debt.

Furthermore, as the Regulator alleges is the case in relation both to Nortel and Lehman, the targets may have received benefits from the employer which are disproportionate to the staff costs recharged to the target by the employer. That sort of situation was described by Ms Agnello for the Regulator as giving rise to a windfall for the unsecured creditors of the target. While I would expect most creditors of an insolvent company to regard themselves as victims rather than the recipients of a windfall, the underlying point that the creditors may receive distributions which are, in the context of the moral hazard at which the FSD regime is aimed, in a sense obtained at the expense of the employer’s pension scheme, is nonetheless a point of real substance.

The only factor sensitive to insolvency employed in the criteria which identify those companies which may be made the targets of an FSD is the requirement, as part of the reasonableness condition in section 43(5)(b) and (7)(d), to consider the financial circumstances of the target, and, perhaps, subject to the analysis which follows, the interests of its creditors, under section 100. But this aspect of the reasonableness condition comes nowhere near imposing a blanket ban on the issue of an FSD to a target in an insolvency process. It is simply one of a potentially limitless number of factors of infinitely variable relevance which the Regulator is directed to consider.

Does the 2004 Act make specific provision as to the priority of the financial obligations imposed by the FSD regime on a target in an insolvency process, or leave that question to be decided by the Insolvency Act and Rules?

This is the question to which a careful weighing up of the pros and cons is both appropriate and necessary. Specifically, the alternatives are:

It is convenient at the outset to deal with what I regard as the least attractive alternative, namely (iv). The argument in its favour was essentially that, since the FSD regime was focused primarily on solvent targets, it would not be surprising to find it implicit in the 2004 Act that Parliament was content for the financial consequences of the imposition of the FSD regime on a target in an insolvency process to be such that it would be ineffective save in most unusual circumstances, where all unsecured creditors were first paid in full. The Administrators’ case was based upon an interpretation of Toshoku which I have rejected, that statutory liabilities of a company in an insolvency process which are not provable debts are nonetheless only expenses if the statute expressly requires the company actually to pay them while in an insolvency process, rather than merely exposing the company to a liability. If the statute in question fell short of that prescription, then, counsel argued, the only remaining issue was whether the liabilities were a provable debt or fell down a black hole, and their submission that they did exactly that was based upon their analysis of the case law on provable debts with which I have in the event largely agreed.

The Administrators supported that submission by pointing, with the benefit of evidence, to the fact that a conclusion that the liabilities were expenses would have a gravely damaging effect, both in the present cases and generally, upon the rescue culture. Their submission was that if an administrator of a potential target was faced with the prospect of an expense liability having priority under Rule 2.67(1)(f) higher even than that of the remuneration of the administrators and their employees (under (g) and (h)), then the administrators would be fatally hampered in making any of the financial judgments and predictions necessary for the achievement either of the rescue of the company itself or, more realistically, of the rescue of all or part of its business, or even its beneficial realisation. The potential FSD and CN expense liability would hang like an enormous sword of Damocles above the administration, paralyzing it in all relevant respects.

The evidence of Mr Bloom, who is a highly experienced insolvency practitioner, as to the general effect upon the rescue culture of a conclusion that the FSD regime imposed financial obligations by way of expense may be summarised as follows:

In relation to the ongoing Nortel administrations Mr Bloom gave a detailed description of the adverse effect of the uncertainty constituted by the impending FSD, if its financial consequences were to rank as administration expenses. He pointed to the maximum estimated exposure of £2.1 billion (the full estimated section 75 debt owed by the Nortel employer to the scheme trustees) and suggested that if the FSD and/or any subsequent CN were to impose financial liabilities as expenses of that magnitude, there might well be an expenses deficiency within the administration, in the context that there is already in place a business sale process which involves the incurring of substantial ongoing expenses in order, in particular, for transitional services to be provided to purchasers. He said that if a potential expense liability of that magnitude cannot be ruled out, then the consequence will be to put the present administrations on hold.

To this undoubted tale of woe counsel for the Administrators added the submission that, in practice, the clearance procedure under section 46 of the 2004 Act offered no real solution, save in cases (unlike the present) where it could easily be shown that the employer was neither a service company nor insufficiently resourced within the meaning of sections 43(2) and 44. In real life, it was submitted, administrators needed to be able to resolve uncertainties of that magnitude much more quickly than the Regulator could be expected to do the necessary research and decision-making sufficient to be able to give clearance under section 46(2)(c), namely that it would not be reasonable to impose the requirements of an FSD on the target seeking clearance. I accept that submission.

The Regulator and its supporters sought to meet the Administrators’ case about the damage which the FSD regime would do to the rescue culture, if its financial consequences constituted administration expenses, by two main submissions. The first was that it was nothing short of alarmist to suggest that financial consequences of anything approaching the full £2.1 billion section 75 debt would be visited upon a target in administration. The target’s liability was no more than to provide reasonable financial support, having regard to the circumstances, including its financial condition. Section 100 required the Regulator to have regard to the interests of the insolvent target’s creditors, so that the Regulator would be unlikely to impose financial obligations at a level, or at a time, which would prevent a beneficial business rescue, and thereby kill the goose that might still lay the golden egg. Any threatened failure on the part of the Regulator to act sensibly could be controlled by a reference to the Tribunal.

Secondly, the Regulator and its supporters submitted that the Companies Court could in any event intervene, on an application by the Administrators, to prevent uncertainties of the type feared by Mr Bloom from disrupting an otherwise beneficial administration, by altering the priority of any expenses created by the FSD regime so as to ensure that the expenses needing to be incurred in promoting and implementing a successful business rescue were not thereby put at risk. Rule 2.67(2) and (3) provide that the court may, in the event of the assets being insufficient to satisfy the expense liabilities, make an order as to the payment out of the assets of the expenses incurred in the administration in such order of priority as the court thinks just. Any such order overrides the prima facie order of priorities laid down in Rule 2.67(1). Similar provision is made in relation to liquidation by Rule 4.220(1) and section 156 of the Insolvency Act.

Mr Trower submitted, but without much enthusiasm, that the language of those provisions did not enable the court to make a prospective order altering expense priorities, but only an order once, at the end of the day, it was shown that there was an expenses shortfall. If the power of the court was thus confined as he submitted, then I agree that it would do little to assist administrators in dealing with the crippling uncertainties identified by Mr Bloom.

But in my judgment the court can make a prospective order under Rule 2.67(3). Altering expense priorities only matters if there is an expenses shortfall. Thus, an order made prospectively altering expense priorities will cause no injustice if it turns out that the assets are sufficient to pay the expenses in full. It requires only a slightly purposive reading of Rule 2.67(3) to conclude that the court may make an order that “in the event of the assets being insufficient to satisfy the liabilities, the expenses incurred by reason of the implementation of the FSD regime against the company are to be paid only out of such surplus as shall remain after all other expenses have been paid”.

I suggested to Mr Trower during argument that a prospective priority order of that kind might largely solve the uncertainties and other serious problems identified by Mr Bloom’s evidence. It would have the effect of placing the Scheme Trustees, as the obligees of any financial obligations imposed by the FSD regime upon the target, at the head of the queue of creditors awaiting distribution of the proceeds of any successful business rescue, without in any way putting at risk the expenditure needing to be incurred to bring that rescue to a successful conclusion. Mr Trower could offer no persuasive argument that this analysis was wrong, and I consider it to be correct.

It will not, of course, be every administration threatened by an FSD in which the court will make such a prospective priority order. The question whether the threat of an FSD is sufficient to undermine the beneficial conduct of an administration is likely to be a fact sensitive question in every case. There will be some cases, and the Lehman administration may well be one of them, where even the maximum quantum of a threatened FSD may pale into relative insignificance when measured against the battalions of other troubles affecting the administrators. By contrast, the very large section 75 debt in the Nortel administrations may, upon examination, lead to the opposite conclusion.

There is also much force in the first of the Regulator’s submissions, in the sense that the obligation imposed on a target by an FSD is only to provide that level of financial support to the scheme as is reasonable in all the circumstances. Nonetheless the breadth of the regulator’s discretion is such that a mere probability that the financial obligation imposed by the FSD regime on a particular target in administration will be very much less than the theoretical maximum does little of substance to resolve the uncertainties to which Mr Bloom tellingly refers. This submission is, to my mind, of greater relevance when addressing other aspects of the priority question than those thrown up by the effect of the FSD regime upon the rescue culture, and I will therefore return to address it in more detail in due course.

My conclusion in relation to the submission based upon the threat to the rescue culture is that the FSD regime does indeed have a potentially adverse impact but, because it can to a large extent be kept under control by the making of prospective priority orders in appropriate cases, the adverse effect is by no means sufficient to force the court to a conclusion that Parliament must have intended that the financial obligations imposed by the FSD regime should be neither an expense nor a provable debt, so that they fall down a black hole. Even if the choice was between priority as an expense, and subordination to the provable debts of unsecured creditors, I would still not regard the black hole case as made out. In my judgment the prospect that an otherwise powerful case for the implementation of the FSD regime upon a particular target could be rendered a complete brutum fulmen by the onset of that target’s insolvency seems to me to fly in the face of commonsense, and Parliament cannot possibly have intended thus to legislate in vain.

I consider that the real and much more difficult choice therefore lies between priority as an expense, and priority as a provable debt. I have already referred to the irony that, in this context, all the parties’ primary cases involved the assertion that the FSD regime did not create provable debts, an assertion which (save in the pre-April 2010 administration followed by liquidation context) I have found to be correct, if Parliament intended when making the Pensions Act 2004 to leave all issues of priority in insolvency processes to be resolved by reference to the insolvency legislation.

There are nonetheless powerful considerations pointing to a conclusion that Parliament did not intend that the financial consequences of the FSD regime should be to afford the pension trustees super-priority in the administration or liquidation of a target. The starting point is that, as between the pension trustees and the employer, and after consultation and careful thought, it was decided that the ordinary provable debt priority given to the section 75 debt by the 1995 Act (and its predecessor) should not be upgraded in 2004. I was shown ministerial statements made during debate on the 2004 Pensions Bill which demonstrated beyond doubt that the retention of the section 75 debt’s ordinary priority as a provable debt was both deliberate and carefully thought through, despite submissions at the time from the majority of the Government’s consultees that it should be promoted.

There was, in particular, no suggestion in 2004 that the Regulator should be given a discretionary power to promote the priority of that debt by issuing an FSD or a CN for an amount similar to the section 75 debt to an insolvent employer, so as by a side-wind to achieve substantially that result, for example by making the debt payable as the result of the issue of a CN an expense in the employer’s liquidation. Nonetheless section 43(6)(a) of the 2004 Act makes it clear that an FSD and a CN may be issued to an insolvent employer and, if the consequence is that the financial obligation or debt thus created is an expense in the employer’s insolvency process, that is the outcome which would be produced as a result of that interpretation.

The Administrators submitted that it was even more extraordinary that Parliament should have intended to confer a higher priority for FSD obligations imposed on a target, than the priority given to the section 75 debt in the insolvency of the employer. This is, at least at first sight, a formidable reason for searching for any avenue of construction which would avoid that result.

A further peculiarity inherent in a conclusion that the FSD regime, if applied to targets in an insolvency process, will give rise to the super-priority inherent in expense liabilities is that it would produce an entirely different result to that which would flow from the issue of an FSD to the same target at any time (however short) before the commencement of its insolvency process. For reasons which I have set out in full, once an FSD has been issued to a target, it creates a qualifying legal obligation under Rule 13.12(1)(b) such that any subsequent consequences, including the debt which would flow from the issue of a CN, would be provable debts. Why, it may be asked, should the Regulator be able to promote the priority of the financial obligations imposed by the FSD regime above those which would flow from its application before insolvency, by waiting for the onset of insolvency before issuing an FSD? It might, furthermore, be a matter of happenstance which of the two occurred first, in a case where, for example, the Regulator’s preparatory steps were being taken while the group of which the target formed part was veering towards collapse.

The force of this point is, if anything, strengthened by the fact that the general description of the invariable requirements imposed by the issue of an FSD under section 44(3) are more readily applicable to a target which is a going concern, with a business future, than to a target which is in the death throes normally associated with being in an insolvency process. It is in that context no answer for the Regulator to point to the fact that, thus far, all cases of the implementation of the FSD regime have occurred after the collapse into insolvency of the target group. My reading of the FSD regime, taken as a whole, is that Parliament expected it to be applied, if possible, to targets which, because they were still trading, had some prospect of being able to provide ongoing support to the relevant pension scheme during the rest of the scheme’s natural life. The fact that the Regulator’s resources and access to relevant information may make that a difficult task in practice, does not significantly impinge upon the perception to be derived from section 43(3).

It is at this point that Ms Agnello’s submission that, when considering whether to approve proposals by an insolvent target, and whether to impose a CN in the absence of acceptable proposals, the Regulator must have regard to the interests of the target’s creditors, takes on critical significance. Her submission, fully supported by Mr Tennet, was that it by no means follows from a conclusion that the financial obligations flowing from the FSD regime are an expense rather than a provable debt that the creditors of the insolvent target will thereby be unfairly prejudiced, or that the lower priority afforded to the section 75 debt is thereby undermined. The Regulator both can and should, they submitted, identify a level of financial support which properly takes account of the creditors’ claims and which may, in an appropriate case, be set at a level which achieves broad equity between the pension trustees on the one hand, and the other creditors of the target on the other. Parliament had, they said, entrusted that balancing exercise to the Regulator, and conferred an unfettered right on any person whose interests were directly affected to refer the matter for a complete re-evaluation before the Upper Tribunal, which may turn out to be presided over by a Chancery judge. By contrast the section 75 debt was automatically imposed, in full, upon the employer.

In reply, Mr Trower made three points. First, he submitted that since the Regulator’s relevant objectives (under section 5(1)) were to protect pension benefits and, in particular, reduce the risk of situations which may lead to compensation being payable from the PPF, it followed that in any such balancing exercise attempted by the Regulator, the dice would from first to last be heavily loaded against the creditors. An obligation to “have regard” to the interests of persons while pursuing specific objectives which are likely to be adverse to their interests is not the same as an obligation to balance those interests as against the interests of the pension scheme members, in a neutral or even-handed fashion, by reference to the pari passu principle.

Second, he submitted that it was by no means clear whether, and if so how, section 100(2)(b) of the 2004 Act worked in relation to companies in an insolvency process. Third, he submitted that the scope for reference to the Upper Tribunal afforded no real protection, since the Tribunal’s duty to determine under section 103(4) of the 2004 Act what is the appropriate action for the Regulator to take must be equally referable to the Regulator’s objectives.

Mr Trower bolstered his submissions by reference to the fact that, in all the available texts whereby the Regulator had, thus far, given reasons for determinations to issue FSDs (in the Sea Containers, Nortel and Lehman cases) there was not to be found any indication of the carrying out of a balancing exercise between the interests of the pension scheme members and the interests of the insolvent target’s creditors, nor had any such creditor or even class of creditors ever been served with a warning notice, as required in relation to a person whose interests are directly affected by the proposed regulatory action. In fairness, a balancing exercise of the type in question will arise more directly at the stages where the Regulator considers the reasonableness of proposed arrangements, or whether and if so in what amount to issue a CN, than whether to include a particular company as the target of the FSD. No written reasons as to those later processes have yet been published. Mr Trower submitted that the Regulator was, in any event, not qualified to carry out the potentially daunting task of striking such a balance, by comparison either with administrators, or the Companies Court, on an application for directions.

In choosing between these powerful submissions, the starting point is, in my view, a closer examination of the operation of section 100 in relation to a target in an insolvency process. Mr Trower submitted that any analysis of the FSD regime which (whether by way of imposing expense liabilities or provable debts) was capable of defeating or diminishing recoveries by creditors of the target, directly affected each and every creditor’s rights under insolvency law, so that section 100(2)(b) would mean that every creditor was a person whose interests were directly affected by any implementation of the FSD regime against the target. The purpose of that submission was of course to fortify the Administrators’ case that the financial consequences of the FSD regime did not in any way affect creditors of an insolvent target, but I have, after giving due consideration to section 100 as part of that submission, rejected it.

I accept that Mr Trower’s analysis of the creditors’ rights is strictly correct, but I consider it inconceivable that Parliament intended to confer upon every one of the potential army of creditors of an insolvent target a separate and distinct right to be served warning notices, to make representations, and to make references to the Tribunal. It was precisely to exclude armies of that kind from participation in those processes that section 100(2)(b) is limited to the interests of persons “directly” affected by the exercise of the relevant regulatory function.

While it is, from a strict insolvency perspective, probably correct to say that, once a company enters into an insolvency process, its interests are for the most part replaced by direct interests of its creditors, I consider it implicit in section 100 that a target in an insolvency process remains the person “directly affected” by the proposed exercise of the FSD regulatory powers against it, but that precisely because it is an insolvency process, that requires the Regulator to have regard to the interests of its creditors as a body. Generally speaking, those interests will be sufficiently represented by the office-holders, in any process of seeking and obtaining representations, and in any reference to the Tribunal.

There may, as Mr Trower submitted, occasionally be cases where the interests of particular classes of creditors are differentially affected by a proposed application of the FSD regime, so as to place the office-holder in real difficulty. In my judgment, such cases would have to be met by the application of procedural common-sense, so as to permit representatives of differentially affected classes to have their say. I therefore accept the broad thrust of the Regulator’s submission that section 100 does require the Regulator (or the Tribunal on any reference) to have regard to the interests of an insolvent target’s creditors as a body, not because each of them is a person directly affected, but because their interests are, collectively, to be taken as the interests of the insolvent target.

That analysis diminishes, but by no means wholly removes, the force of the Administrators’ submissions that the Regulator’s objectives (transferred to the Tribunal on any reference) afford an unsatisfactory basis for a conclusion that the interests of an insolvent target’s creditors will be fairly or neutrally balanced with those of the members of the relevant pension scheme, merely because of an obligation to have regard to them. The question is not however whether I regard that outcome as an unsatisfactory way of balancing competing policy objectives (i.e. pension scheme protection as against pari passu distribution to creditors), but whether it is so obviously unsatisfactory that Parliament cannot have intended it when creating the FSD regime, bearing in mind that the resolution of such policy clashes is a matter for Parliament, rather than for the court.

The first question is whether any alternative interpretation than an expense priority is obviously more satisfactory for that purpose. Having dispensed with the Administrators’ black hole solution, the only remaining candidate is an interpretation that, for all purposes, and regardless of any insolvency cut-off date, the FSD regime creates provable debts when applied to an insolvent target.

That solution would, to my mind, undoubtedly produce a less unsatisfactory resolution of the policy clash. Once the Regulator had decided, either by approving proposals or by issuing a CN, upon a specified quantum of financial support which the target company ought to provide, if otherwise financially able to do so, the inclusion of that amount as a debt provable by the scheme trustees would automatically qualify on a pari passu basis in the distribution of the assets of the target company, after payment of expenses, preferential debts, and secured creditors. There would, to that extent, be an automatic balancing of the interests of the scheme members and the unsecured creditors, and the Regulator would be, to that extent, relieved of a formidable task for the performance of which its pension objectives make it a less than impartial arbiter. It would also avoid any significant impact on the rescue culture. A solution which required the scheme members to share pari passu with the unsecured creditors would, as Mr Moss memorably described it in argument, be “a gut-feel fair solution”. It is in fact the solution that arises under insolvency law whenever the FSD is issued before the target’s insolvency cut-off date.

If, in the Pensions Act 2004, Parliament left the priority of the financial consequences of the FSD regime upon an insolvent target to be resolved by reference to the insolvency legislation, then if the FSD is issued after the cut-off date, for the reasons which I have described in detail, that gut-feel fair solution is not achieved. The question therefore, is has Parliament expressly or by necessary implication nonetheless legislated for that solution in the 2004 Act?

I make it clear at this stage that I would very much have preferred to be able to reach the conclusion that it did just that. Indeed, mindful of the difficulties which I expected to encounter in concluding that the application of Rule 13.12 to the FSD regime could lead to that result, I invited all counsel to consider whether a provable debt outcome could be identified as a matter of construction of the 2004 Act itself. Formally, but without any evident enthusiasm, counsel for the Regulator and its supporters submitted that it could. Nonetheless their detailed submissions in favour of the provable debt solution were entirely focused upon the routes, all of which I have rejected, whereby it might be found by way of analysis of Rule 13.12 and the substantial case law by which it has been interpreted.

Notwithstanding that a provable debt solution is, to my mind, obviously fairer as between scheme members and unsecured creditors and preferable as a means of resolving the underlying policy clash, I find myself driven with reluctance to the conclusion that, when formulating the 2004 Act, Parliament did in fact choose to leave the priority questions which would inevitably flow from the application of the FSD regime to companies in an insolvency process to be resolved purely by the insolvency legislation. My reasons follow.

First, as Mr Dicker submitted, Parliament must be taken to have been cognisant of that insolvency regime when enacting the 2004 Act, and of the ability, should it choose to do so, to define a pension-related obligation in such a way as to ensure that it would be a provable debt in the administration or liquidation of the corporate obligor regardless of the date of issue of the FSD. That is precisely what it did in reformulating by amendment section 75 of the 1995 Act so as to ensure that the section 75 debt was deemed to fall due immediately prior to an insolvency event. Parliament could have chosen to make similar specific provision in relation to financial obligations arising from the brand new FSD regime, either by a similar deeming provision, or by identifying some earlier pre-insolvency obligation for the purposes of Rule 13.12(1)(b). It did not do so.

More generally, there is simply not to be found anywhere in sections 43 to 50 of the 2004 Act, or in the FSD Rules, within which the statutory FSD regime is entirely contained, any reference to insolvency at all. This does not mean that the regime was inapplicable to companies in an insolvency process, for reasons already given, but it does point strongly to a conclusion that Parliament was content that, to the extent that the regime was in fact applied in an insolvency context, the insolvency legislation would sufficiently deal with its consequences, in terms of priority.

Put even more simply, there is nothing in those sections to which a conclusion that, as a matter of construction, the FSD regime creates provable debts when first applied after an insolvency cut-off date can sensibly be attached, or from which such a conclusion can be derived. It is no answer to say that a promotion in priority where insolvency precedes the issue of the FSD is so illogical that it cannot have been intended. It may be illogical, but that is what happens for example in relation to corporation tax and rates, where there exists no discretion to reduce the amount payable so as to reduce or nullify the apparent unfairness of the promotion in priority. At least the discretion of the Regulator under the FSD regime gives scope for the mitigation of that unfairness.

The conclusions which I have reached on the main question thus far make it unnecessary to address Mr Isaacs’ submissions based upon ex parte James (supra). They were designed to identify an escape from the black hole conclusion, in the event that the financial obligations flowing from the FSD process were neither expenses nor provable debts. I should add that, had I been of a different view on construction, namely that Parliament had been content that those obligations should fall down a black hole, I would have found it very difficult to attempt a court fashioned remedy via Mr Isaacs’ escape route, imaginative though it was.

I have therefore been driven to the conclusion, in conformity with the analysis of Rule 13.12 in repeated decisions by which I am bound, and in accordance with what I conceive to be the true principle in Toshoku, that this is a case in which Parliament has legislated to create financial obligations applicable to and payable by a company in an insolvency process which may be triggered (after the cut-off date) in such a way that, rather than creating provable debts, they create administration or liquidation expenses, as the case may be. That conclusion is likely to be to an extent an impediment to the achievement of the objectives of the rescue culture, but the ability of the court to make a prospective priority order will, I think, keep that potential for damage to a minimum, in cases where the uncertainties might otherwise be fatal.

The outcome is, in my view, likely to prove unfair to the creditors of an insolvent target, unless perhaps the Regulator and the Upper Tribunal treat the pari passu principle as a cardinal aspect of the very broad discretions which arise at the three consecutive stages of (1) the determination whether to issue an FSD to an insolvent target, (2) the decision whether particular proposals for financial support by the insolvent target are reasonable in all the circumstances, and (3) the decision whether to issue a CN to an insolvent target, and if so, in what amount. It may be that, by bearing in mind that a pari passu sharing would be the automatic result of any FSD issued before a target went into an insolvency process, this approach to the exercise of the discretion in post cut-off cases will actually occur, but it is no part of my task to bind either the Regulator or the Tribunal.

Obligations arising from the FSD regime are as far removed as it is possible to imagine from the types of liability which can have been in the minds of the draftsmen when, from 1865 onwards, creating the cut-off date based regime for the identification of provable debts now to be found in Rule 13.12. Nonetheless, Parliament has, either through deliberate intent or (I suspect in reality) inadvertence, legislated in such a way as to leave the priority problems associated with the implementation of the FSD regime on insolvent companies to be dealt with by a technical formula which was neither designed nor, in my view, fit for that special purpose. In the sharpest contrast with almost all other forms of administration or liquidation expense (which relate to matters arising after the cut-off date) financial obligations triggered by the FSD regime are justifiable in policy terms almost entirely by reference to matters which will have happened (by commission or omission) before the target’s cut-off date. Furthermore the expense liabilities created by that regime are likely in many cases to exceed by orders of magnitude the expenses of any other type arising in a typical administration or liquidation.

I hope that a higher court may find a way through or around the existing authorities on Rule 13.12 which enables it to identify a sufficient pre-cut-off date obligation in the factors about the target’s relationship with its group employer that qualifies under Rule 13.12(1)(b). That would produce a result consistent with the way in which the FSD regime works if the FSD itself is issued even a day before the cut-off date, and the way in which it will still work (because of the late introduction of the 2010 Amendments to Rule 13.12) if in the present cases an FSD precedes a transformation of the insolvency processes affecting its targets from the present administrations into liquidation. More than that, such a conclusion would produce fairness and justice out of what is, in my view, currently a legislative mess. It may be that the Insolvency Service or Parliament might wish to consider a suitable amendment, either to the Rules or to the 2004 Act, if persuaded as I have been that the conferring of super-priority as expenses upon the financial liabilities arising from the FSD regime is both potentially unfair to the target’s creditors and inconsistent with a decision taken in 2004 not generally to elevate employees’ pension claims above the claims of those creditors. That course would probably be both quicker and cheaper than further hugely expensive litigation on this issue, the cost of which will ultimately have to be borne either by the creditors or by the relevant schemes and their members, or both.

The specific questions framed in these applications, albeit in slightly different language, focus first upon the consequences of the issue of an FSD to the applicant target companies, and secondly upon the consequences of the subsequent issue of a CN. I shall now deal with them separately. It is convenient to deal first with a CN, because at the FSD stage the question whether a subsequent CN would create a provable debt or a liquidation expense may be of real importance in the Administrators’ decision-making about the proposals by way of financial support that may be reasonable in the circumstances. The priority consequences of a CN are, in any event, simpler than those which may flow from the issue of an FSD, because a CN invariably creates a financial obligation in the form of a liquidated debt. By contrast, an FSD may give rise to a number of different types of financial obligation, or none at all, depending upon the Administrators’ response to it.

If a CN comes to be issued to any of the applicant target companies while it is still in its present administration, then the cost of complying with it will be an administration expense. That does not necessarily mean that the Administrators should immediately cause it to be paid, or even that the Scheme Trustees should be allowed immediately to bring proceedings for its enforcement: see Toshoku (supra) at paragraph 39, where Lord Hoffmann said that:

“The fact that a debt counts as an expense of the liquidation does not necessarily mean that the creditor should be allowed immediately to bring proceedings or levy execution. The order of priorities under Rule 4.218(1) may mean that if he is paid at once, the assets to satisfy prior expense claims may be insufficient. So the question of remedy is entirely a matter of discretion.”

I can envisage no reason why that analysis is not fully applicable to a debt created by a CN under section 49(3) of the 2004 Act, although the point has not been argued.

If however the CN is issued to an applicant target company after it has moved from its present administration into liquidation, for non-compliance with an FSD issued while that target was in administration, then the amount specified in the CN will rank as a provable debt. This is because it will be a debt to which the target company has become subject after the cut-off date by reason of a legal obligation previously incurred by reason of the FSD, pursuant to Rule 13.12(1)(b). This is entirely the adventitious result of the lateness of the introduction of the 2010 Amendments to Rule 13.12 which, in relation to any company suffering an insolvency event after 5^th April 2010, will provide for a single cut-off date in relation to all subsequent insolvency processes, namely the date of commencement of the first of them.

It is contemplated by the terms of the Nortel Administrators’ Application that the financial consequences of an FSD might vary, according to whether the FSD is issued before or after notice under Rule 2.95, or before or after a Nortel applicant target moved from administration into liquidation. In my judgment, the consequences are substantially the same, and are as follows. First, the issue of an FSD does not, of itself, create any immediate financial obligation. It imposes a legal obligation on the target company, to be performed at the direction of the Administrators, to secure that reasonable financial support for the pension scheme is put in place within the period specified in the FSD, either by that target acting alone, or by that target acting in cooperation with one or more of the other targets of the FSD. For that purpose the Administrators may need to consult their stakeholders, negotiate with the administrators of other insolvent targets, and the management of any solvent targets (such as LBAM), and no doubt communicate with the Regulator with a view to ascertaining what it might regard as reasonable in the circumstances. The Administrators may well wish to seek further directions from the court, and it would be inappropriate at this stage, and in the absence of any argument about the detail, for me to be any more specific as to the type of proposal which the Administrators ought to make.

Nonetheless, since the FSD regime contemplates a theoretically infinite variety of potentially reasonable proposals, it is appropriate to specify at this stage the priorities which would attach to certain particular kinds. Thus, if the Administrators proposed simply to make a payment of a sum of money, then the cost of making that payment would be an administration expense.

If, alternatively, the Administrators propose to enter into a legally binding contract with the pension scheme trustees for the provision of support, for example, by way of instalments, sums payable in respect of debts or liabilities arising out of that contract would, on the face of it, be chargeable upon the property of which the Administrators had custody immediately before leaving office, pursuant to section 99(4) of the Insolvency Act. If it was so, then those sums or liabilities would be payable in priority to any charge arising under section 99(3) in respect of the Administrators’ remuneration and expenses.

As a further alternative, and as happened in Sea Containers, the office-holders might consider it reasonable to transfer, or promise to transfer in future, certain property to the scheme trustees, such as shares in a special purpose company formed for the purpose of carrying on a business rescued in the course of the administration. That would not of itself give rise to any form of purely financial obligation, although the incidental cost of making that transfer might be.

It would be understandable if the Administrators were to have regard, in addressing the question as to the reasonable proposals to be made, to the question whether any subsequent CN issued because the Regulator refused its approval would itself create a financial liability ranking as an expense, or as a provable debt. That possibility (the no doubt unintended consequence of the late introduction of the 2010 Amendments to Rule 13.12), gives rise to the question whether the Administrators should, for the protection of the target company’s unsecured creditors, seek to arrange matters in a way calculated to maximise the chances that the CN would generate a provable debt in a subsequent liquidation, rather than an administration expense. Since the numerous ramifications of that potentially very complicated question have not been the subject of any argument before me, I propose to say no more than that they may arise for determination in the future, on a further application to the court for directions.

Finally, the Administrators may well wish, if the facts justify it, to seek a prospective priority order, so as to ensure that the uncertainties created by the ranking of any FSD financial obligations as an expense do not undermine the beneficial outcome of the present administrations.