Johnstone v Alpha Insurance A/S (in bankruptcy)
[2024] NZHC 725
•8 April 2024
IN THE HIGH COURT OF NEW ZEALAND AUCKLAND REGISTRY
I TE KŌTI MATUA O AOTEAROA TĀMAKI MAKAURAU ROHE
CIV-2023-404-966
[2024] NZHC 725
UNDER the Companies Act 1993 and subpart 4 of the High Court Rules 2016 IN THE MATTER
of CBL INSURANCE LIMITED (in
liquidation)
BETWEEN
KARE JOHNSTONE and ANDREW JOHN
GRENFELL as liquidators of CBL INSURANCE LIMITED (in liquidation) Applicants
AND
ALPHA INSURANCE A/S (in bankruptcy) Respondent
Hearing: 6 November 2023 Counsel:
D Chisholm KC and A Murray for the Applicants
A Barker KC and J MacGillivray for the Respondent
Judgment:
8 April 2024
Reissued:
11 April 2024
JUDGMENT OF BECROFT J
[Application under s 292 of the Companies Act 1993 to set aside voidable transactions]
This judgment was delivered by me on 8 April 2024 at 9.30am pursuant to r 11.5 of the High Court Rules 2016.
Registrar/Deputy Registrar
……………………………………
Solicitors:
DLA Piper, Auckland Tompkins Wake, Hamilton
JOHNSTONE v ALPHA INSURANCE [2024] NZHC 725 [8 April 2024]
What is this application?
[1] This is an application by the liquidators of CBL Insurance Limited (in liquidation) (CBLI). They apply for orders that certain transactions entered into with Alpha Insurance A/S (in bankruptcy) (Alpha) be set aside as voidable transactions under s 292 of the Companies Act 1993 (the Act) and that Alpha pay the liquidators EUR 25 million and GBP 397,000 plus interest and costs.
[2] As the hearing developed, it became apparent (and was agreed) that there were two issues for determination:
(a)what constitutes “due debts” under s 292(2) of the Act in the context of the business of insurance and reinsurance companies; and
(b)whether in this case CBLI could pay its “due debts” at the time of the relevant transactions.
[3] It was also agreed that there was no New Zealand authority which was directly on point.
CBLI and Alpha
[4] CBLI was one of New Zealand’s largest licensed insurers and it provided reinsurance services mainly overseas. Much of CBLI’s business came from reinsuring European insurance companies. CBLI would cover a proportion of those insurance companies’ claims in exchange for a proportion of the premiums those companies received.
[5] One of the European insurers that CBLI insured was the respondent, Alpha, a Danish Company which provided insurance primarily in the French commercial building construction market including workers’ accident insurance.
[6] As a licensed insurer in New Zealand, CBLI was subject to the scrutiny and oversight of the Reserve Bank of New Zealand (RBNZ) as the insurance industry’s
supervisor. As I will set out in more detail, during the 2010’s, the RBNZ had expressed escalating concern about the management of CBLI’s business and how that would affect its solvency. Some of this concern related to transactions between CBLI and Alpha.
[7] On 23 February 2018, those concerns escalated to the point that CBLI was placed into interim liquidation. This was seven days after the relevant transactions with Alpha had been completed. Final liquidation occurred in November 2018. Meanwhile, Alpha was also declared bankrupt by decree of the Maritime and Commercial Court in Copenhagen on 8 May 2018.
[8] The applicants, Ms Kare Johnstone and Mr Andrew Grenfell, are the appointed liquidators of CBLI (the liquidators).
The transactions
[9] The transactions the liquidators seek to set aside are (together, “the transactions”):
(a)entry into a liability settlement agreement dated 9 February 2018 between CBLI and Alpha (the “LSA”);
(b)a payment under the LSA of EUR 25 million paid by CBLI to Alpha on or about 16 February 2018; and
(c)a “standard” payment under a separate contract—a “Motor Quota Share Reinsurance Treaty” between Alpha and CBLI—of GBP 397,000 paid by CBLI to Alpha on 14 February 2018.
Voidable transactions
[10] Pursuant to s 292(1) of the Act, an insolvent transaction is voidable by the liquidator:
292 Insolvent transaction voidable
(1)A transaction by a company is voidable by the liquidator if it—
(a)is an insolvent transaction; and
(b)is entered into within the restricted period.
[11] A transaction is an insolvent transaction if it is caught by the definition provided in s 292(2):
…
(2)An insolvent transaction is a transaction by a company that—
(a)is entered into at a time when the company is unable to pay its due debts; and
(b)enables another person to receive more towards satisfaction of a debt owed by the company than the person would receive, or would be likely to receive, in the company’s liquidation.
(Emphasis added).
[12] If the transaction was entered into within the “restricted period”, defined below, the presumption is the transaction was “entered into at a time when the company was unable to pay its due debts”, unless the contrary is proved.1
[13]The restricted period is relevantly defined in s 292(4C)(b):
In the case of a company that was put into liquidation by the court, the period of 6 months before the making of the application to the court together with the period commencing on the date of the making of that application and ending on the date on which, and at the time at which, the order of the court was made.
[14] It is not disputed that because RBNZ applied to put CBLI into liquidation on 23 February 2018, the restricted period therefore commenced on 23 August 2017 (that being six months before the RBNZ’s application).
Relevant background information
[15] The following information is required to understand the parties’ respective positions regarding the proper interpretation of s 292(2)—namely, the meaning of “due debts”.
1 Companies Act 1993, s 292(4A).
The nature of insurance companies’ business
[16] Both counsel explained in helpful detail the nature and business of insurance companies and highlighted the particular characteristics of reinsurance in the context of the European building construction market.
[17] For present purposes it is sufficient to point to the affidavit of Mr Geoffrey Atkins, a Sydney-based actuary who explained:2
[11]General insurance is one of few “cash flow positive’ businesses in the economy. Premiums are received in advance at or near the commencement of the insurance policy, and most of the outgo is in the form of claims which are paid out over months or years during and after the period of the insurance policy. In the meantime, the insurance company has a positive cash balance of funds collected from premiums and held to meet claims. These funds are generally invested and the return on investments is additional revenue for the insurer.
[12]This feature – essentially where you get paid for the product before you ‘manufacture’ it – is why insurance accounting and prudential regulation focusses so much on the company holding adequate provisions in respect of liability for insurance policies that have been issued. The liabilities need to be estimated, because the amount and timing of the required payments is not known with certainty … .
[18] The nature of insurance and reinsurance businesses was well described in what counsel regarded as a highly relevant Australian case, New Cap Reinsurance Corp Ltd (in liq) v AE Grant.3 Although the reinsurance in that case was in respect of catastrophe policies, the reasoning applies similarly to building construction reinsurance. The Court explained:
[12] It is a feature of insurance and reinsurance business that there will be a timelag between the occurrence of the event giving rise to a liability under the policy and the insurer’s liability becoming known, and between the liability becoming known and being fully quantifiable. Depending on the type of business, the time required for a full assessment of the extent of liability may be significant. It may be measured in years, not months. By way of example, a significant part of NCRA’s business was providing reinsurance of insurers’ liabilities under catastrophe policies. There were delays between the occurrence of the catastrophe, such as a hurricane, and the making of claims on the original policies of insurance. There would have been further delays
2 Mr Atkins was retained as an expert by the liquidators. He is employed by Finity Consulting Pty Ltd (Finity).
3 New Cap Reinsurance Corp Ltd (in liq) v AE Grant [2008] NSWSC 1015 per White J [New Cap Reinsurance].
before it could be determined that the insurer’s level of claims liabilities was likely to reach the threshold at which the reinsurance policy would be triggered. As the reinsurance policies provided indemnity once the level of paid claims under the original policies reached a threshold, there would be further delays before NCRA became liable to pay. There would be still further delays before the full extent of its liability could become known.
[13] Accounting standards and industry practice require insurers to make provision by way of estimates of liability for claims incurred but not reported (IBNR) or not enough reported (IBNER). Provisions for claims liabilities are not assessed merely by aggregating estimates of liability for individual claims, together with an estimate of IBNR. Actuarial analysis is used to estimate claims liabilities. Actuarial analysis is not based on the aggregation of estimates of individual claims. Rather, typically, claims liabilities are assessed by class of insurance using trends in aggregated data. For example, there may be an historical trend as to how claims liabilities develop over time for a particular class of business. An actuary may use such information in relation to earlier underwriting years to predict the future claims liabilities for later underwriting years. There may be market experience of loss ratios on particular classes of business for certain underwriting years.
[14] The determination of an insurer’s insurance liabilities involves the making of informed estimates of uncertain future outcomes. The determination of an insurer’s insurance liabilities is quite different from, for example, the determination of the liabilities of a retailer or manufacturer, whose debts can usually be precisely ascertained.
[19] For CBLI, it was agreed that a feature of the reinsurance business in the construction market is that there could be a very long tail for claims. At least in the context of the French building construction market, the “tail” could be up to 16 years after construction was complete. This delay was said to be a feature of the reinsurance environment in which CBLI operated. As will be discussed, this long tail will be relevant to the determination of what constitutes a “due debt”.
The regulatory regime under which CBLI operated
[20] Given the large amounts of money obtained “upfront” by insurance companies from premiums paid by policy holders, and the lag between claims being made and then quantified under those policies, it is readily understandable why oversight and supervision of insurance companies should be provided. To state the obvious, otherwise the pool of money paid in advance might dissipate or vanish through criminal or imprudent actions by the insurance company. Put bluntly, without oversight, premium holders could be left high and dry.
[21] In New Zealand, the necessary regulatory and supervisory vehicle is provided by the Insurance (Prudential Supervision) Act 2010 (IPSA). I understand the IPSA to be the first comprehensive regulatory framework for the prudential regulation and supervision of insurers carrying on business in New Zealand.4 Pursuant to the legislation, RBNZ functions as the supervisor and regulator of the industry.
[22] The RBNZ may issue solvency standards,5 and these standards have the status of regulations.6 Under the RBNZ’s Solvency Standard for Non-life Insurance Businesses 2014, insurers are required to hold capital sufficient to enable insurers to absorb potential losses from insurance claims and other risks.7
[23] I was advised by counsel that the general requirement for a non-life insurance company is to maintain a solvency margin, or standard, of 100 per cent. This means that an insurer’s “actual solvency capital” (capital reserves less certain deductions), must be greater than 100 per cent of its minimum solvency capital, (being the complex calculation of the minimum capital required having regard to the insurer’s risks). As will be seen, in July 2017, due to mounting concerns, the RBNZ required CBLI to increase its solvency margin to 170 per cent.
Outstanding Claims Liability (OCL)
[24] Part of calculating the minimum solvency capital (see above) is calculating an insurer’s Outstanding Claims Liability (OCL). The nature of the OCL, and whether in this case it constitutes a “due debt” for the purpose of s 292, is at the absolute centre of this case. Therefore, it is worth explaining the OCL in some detail.
[25] The OCL is an estimate, provided by an insurance company’s approved actuary, of the insurer’s total likely claims exposure for the cover it has written. The
4 A useful overview of the Insurance (Prudential Supervision) Act (IPSA) 2010 and the regulatory obligations of insurance companies, particularly insurance solvency and the operation of the solvency standards, is provided by Robinson J in R v Harris [2023] NZHC 2635. Mr Harris was the CEO and a director of CBLI, and charges laid against him under the Crimes Act 1961 arising out of aspects of his involvement in the affairs of CBLI were addressed (and dismissed) in that judgment. Counsel advised this judgment did not bear on the issues before this Court and it was unnecessary for me to consider it further.
5 IPSA, s 55.
6 See s 233.
7 This capital is also referred to as reserves or sometimes “provisioning”.
calculation of the OCL is governed by the New Zealand Equivalent to International Financial Reporting Standard 4 Insurance Contracts (NZ IFRS 4). This standard is a regulation for the purposes of the now apparently applicable Legislation Act 2019.
[26] The OCL is dealt with by paragraph 5.1 of the NZ IFRS 4 (Appendix D). While the OCL is a reasonably straightforward concept, its definition is a little more complex. It is sufficient to set out the following. At paragraph 5.1, the OCL is referred to as:
An outstanding claims liability shall be recognised in respect of direct business and reinsurance business and shall be measured as the central estimate of the present value of the expected future payments for claims incurred with an additional risk margin to allow for the inherent uncertainty in the central estimate.
(Emphasis in original).
[27] “Claims incurred” is not confined to claims that have been made/lodged prior to the end of a reporting period; they include claims reported and not finalised, and even claims as yet unreported and not known at the end of the reporting period.8
[28] At paragraph 5.2, the term “expected future payments” (as that term used in paragraph 5.1), shall include:9
(a)amounts in relation to unpaid reported claims;
(b)claims incurred but not reported (IBNR);
(c)claims incurred but not enough reported (IBNER); and
(d)costs, including claims handling costs, which the insurer expects to incur in settling these incurred claims.
[29]Counsel drew my attention, in particular, to paragraphs (b) and (c).
[30] Paragraph (b) effectively means prospective or future claims not yet known about, let alone made/reported, at the time of calculating the OCL. In shorthand, this is referred to as IBNR (“incurred but not reported”). This is why the assessment of an
8 See the “Definitions” section at [19.1] in the NZ IFRS 4 (Appendix D).
9 NZ IFRS 4 (Appendix D).
actuary is so important; without actuarial expertise the IBNR would be an insurance company’s educated best guess.
[31] Paragraph (c) refers to claims incurred but insufficiently reported so that exact liability cannot presently be ascertained. In shorthand, this is referred to as IBNER (“incurred by not enough reported”). This also requires actuarial assessment of prospective liability.10
[32] In the context of the central argument in this case as to whether the OCL should be considered a “due debt” payable by CBLI, it is important to note that while paragraph (b) represents an actuarial calculation of claims not even known about— what Mr Chisholm KC, for the liquidators, called “prospective claims”—it is wrong to characterise the whole of the OCL in exactly the same way. The OCL is also made up of other types of claims which are specific and quantifiable, or soon to be quantifiable, so that characterising them as a “debt” is more straightforward. For instance, paragraph (a) is much more concrete as it relates to actual claims; and paragraph (c) relates to claims known about, but not sufficiently known about to make an exact calculation of liability at the time the OCL is finalised. Similarly, (d) refers to costs, which in respect to (a) and (c), will be readily quantifiable.
[33] As counsel explained, the calculation of the OCL is an art not a science. It is not a “nice to have” or simply consistent with good accounting practice—although it is both those things—it is a mandatory requirement. It must be professionally calculated, and it must be included in an insurance company’s annual balance sheets. On the assumption that the OCL has been accurately and reliably calculated, it will assist in determining whether an insurance company is balance sheet solvent. Mr Barker KC, for Alpha, agreed that it might also assist a company to forecast its annual profit. To the extent that the OCL is unreliably calculated, as was exactly the case for CBLI here, it may, when compared with the properly calculated OCL, reveal irrefutable balance sheet insolvency.
[34] To illustrate how inaccurately CBLI’s OCL was recorded on its balance sheet, I refer to Mr Atkin’s affidavit which attached his report (from Finity) that reviewed
10 See also the comments of White J addressing both concepts in New Cap Reinsurance, above n 3.
CBLI’s reserve adequacy as at 31 December 2017. That report, as was explained by Mr Barker, for Alpha, showed the recorded balance sheet figure for the OCL as being NZD 592.1 million. That resulted in net assets for CBLI of $77.7 million. However, on Finity’s calculations the OCL should have been $894.3 million (low) and up to
$1,005.8 million (high). Taking those figures, there were net liabilities of $97.7 million (low) and $213.2 million (high), respectively.
[35] And as further background, the observations of Ms Johnstone, one of the appointed liquidators for CBLI, are revealing. She deposed that based on the actuarial analysis subsequent to CBLI’s liquidation, it was insolvent as at 31 December 2016 and 2017. I assume, but it is not clear, this means balance sheet insolvent. She also deposed that CBLI was largely balance sheet insolvent from not later than 2013.
Powers of the Reserve Bank
[36] Where the RBNZ has concerns about the financial position of a licensed insurer, IPSA gives it broad powers to act including to apply to liquidate an insurer under s 151:
151 Bank may apply for liquidation of insurers
(1)The Bank may, in the case of a licensed insurer that may be put into liquidation under or in accordance with the Companies Act 1993, apply to the High Court to appoint a liquidator for the insurer.
(2)The High Court may, on an application under subsection (1), appoint a liquidator for the licensed insurer if it is satisfied that—
(a)the insurer is unable to pay its debts (and, for that purpose, section 287 of the Companies Act 1993 applies with all necessary modifications whether or not the insurer is a company); or
(b)the insurer is failing to maintain a solvency margin; or
(c)the insurer has persistently or seriously failed to comply with any direction, condition, or other requirement imposed by or under this Act or the regulations; or
(d)it is just and equitable that the insurer be put into liquidation.
(3)The High Court may, on the application of the Bank, appoint a liquidator for a body corporate that may be put into liquidation under or in accordance with the Companies Act 1993 if it is satisfied that the body corporate is carrying on insurance business in New Zealand without holding a licence in breach of section 15.
(4)Subsection (3) does not limit subsections (1) and (2).
(5)Nothing in this section limits or affects any other enactment that provides for the winding up, liquidation, or dissolution of any body corporate or any class of body corporate.
[37] Subsections (1), (2)(a), and (2)(d) mirror the general rights of various parties to apply to put a company into liquidation.11 However, it will be seen that in addition to the usual rules applying to the liquidation of a company, the RBNZ can apply for liquidation of an insurance company if the company has failed to maintain a solvency margin or failed to comply with a direction, condition or requirement imposed by RBNZ.12
[38] A key point to make is that an insurance company faces additional “solvency” issues and obligations to those faced by a non-insurance company.
Chronology of key events
[39] While many expert affidavits were filed, none of the deponents were required for cross-examination. The facts are essentially agreed between Mr Chisholm, for the liquidators, and Mr Barker, for Alpha. I am grateful for their constructive approach.
[40] The following chronology is drawn essentially from Mr Chisholm’s largely uncontested table and is augmented by relevant additional facts:
Date
Event
1971
CBLI was incorporated. 99 per cent of CBLI’s business was overseas in over 20 countries around the world. CBLI was one of the larger insurance companies in New Zealand.
September 2013
CBLI was granted a licence under the IPSA regime.
October 2015
Its holding company, CBLC, was listed on the NZX and ASX.
11 See the Companies Act, s 241(4). Note that ss 240A and 241(2)(vii) confirm that the ordinary rules relating to insolvency apply to the liquidation of an insurance company.
12 See IPSA, s 151(2)(b) and (c).
2010s
CBLI provided reinsurance for three European “cedant”13 insurers, including Alpha, who provided insurance to the French construction industry.
June 2016
The RBNZ began to have concerns regarding the level of CBLI’s reserves including its OCL and whether CBLI was carrying on business in a prudent manner.
July 2017
Alpha’s Danish Regulator (Danish Financial Supervision Authority) had concerns about CBLI’s ability to meet its reinsurance commitments to Alpha. Alpha was required to substantially increase its reserves.
CBLI’s Regulator, the RBNZ, also had concerns about CBLI’s solvency.
July 2017
The RBNZ required CBLI to increase its reserves to maintain a solvency ratio of 170 per cent.
At the same time, it appointed McGrath Nicol (the liquidators’ firm) to investigate CBLI’s financial position. McGrath Nichol’s conclusion, supported by actuarial analysis from Mr Atkins of Finity, was that CBLI’s appointed actuary (PWC) was underestimating CBLI’s Outstanding Claims Liability (OCL).
23 August 2017
“The restricted period” began under s 292(6) of the Act— being six months before the application for liquidation which was filed on 23 February 2018.
15 November 2017
CBLI, and its appointed actuary, advised the RBNZ that CBLI was likely to breach its solvency ratio as at 31 December 2017.
22 November 2017
RBNZ made directions preventing CBLI from entering any transaction involving payment or transfer of assets in excess of NZD 5 million without consulting RBNZ. This was one of the directions referred to in the SFO prosecution.14
31 December 2017
CBLI’s liabilities far exceeded its assets, from at least December 2017, and so it was “balance sheet” insolvent. This is accepted by all parties.
2 February 2018
NZX and ASX put trading halts on CBLI’s listed parent company, CBLC.
13 “Cedant” is defined in the NZ IFRS 4 (Appendix D) as the policy holder under a reinsurance contract.
14 See R v Harris, above n 4.
5 February 2018
CBLC announced that CBLI had under-reserved its liabilities (including OCL) by NZD 100 million and was
writing off NZD 44 million of receivables and required capital raising.
7 February 2018
CBLI’s credit rating was downgraded from bb+ to b++ (I was provided with no information as to the relative significance of these credit ratings).
9 February 2018
Alpha advised CBLI it believed the Danish Regulator FSA, would prevent Alpha from writing further business with CBLI.
9 February 2018
CBLI and Alpha enter into the Liability Settlement Agreement (LSA). This was done to address concerns by the Danish FSA that Alpha strengthen its reserves. Under the LSA, CBLI was to pay to Alpha reinsurance collateral of EUR 25 million to be used to meet claims and future claims. In return, CBLI’s liability to Alpha was reduced by EUR 27 million.
12 February 2018
RBNZ directed CBLI not to make the payment of EUR 25 million to Alpha. This was another direction referred to in the SFO prosecution. It was found not to be a valid direction under s 143 of IPSA in the criminal prosecution of CBLI’s director.15
However, it is said by the liquidators to reinforce RBNZ’s concerns as to illegitimate creditor preferences.
14 February 2018
CBLI makes a “run of the mill” reinsurance payment of GBP 397,000 to Alpha in respect of CBLI’s share of a claim under a Motor Quota Share Reinsurance Treaty between Alpha and CBLI.
16 February 2018
In contravention of the RBNZ direction, and believing it to be invalid, the directors of CBLI pay EUR 25 million to Alpha pursuant to the LSA, believing that to be in the best interests of CBLI.
23 February 2018
The RBNZ applied to put CBLI into liquidation and also applied for interim liquidators to be appointed under s 246 of the Act.
Interim liquidators were appointed.
(The application had been previously drafted in January 2018).
3 March 2018
Alpha ceased issuing new/reinsuring contracts, and the next day entered into voluntary liquidation.
15 See R v Harris, above n 4.
8 May 2018
Alpha placed into bankruptcy by the Danish Courts following determination that the company was insolvent.
12 November 2018
CBLI put into liquidation by order of High Court. By this time, the application was unopposed by CBLI’s directors.16
25 September 2019
Liquidator’s notice served, regarding alleged voidable transaction of EUR 25 million.
22 October 2019
Notice of Objection to Notice served.
4 September 2020
Notice filed regarding alleged voidable transaction of GBP 397,000. Served on 7 September 2020.
29 September 2020
Notice of Objection served.
22 October 2022
Originating application to set aside both transactions filed.
What is agreed between the parties?
[41] Mr Chisholm and Mr Barker presented their cases with clarity. I acknowledge their assistance, and that of their juniors, Ms Murray and Mr MacGillivray, who were clearly very knowledgeable about the detail of their respective cases. At the outset of their arguments, they agreed that the following matters were accepted and/or conceded:
(a)The transactions took place at the time when CBLI was unarguably and fundamentally “balance sheet” insolvent.
(b)This “balance sheet” insolvency was masked by the inclusion of an inadequate amount reserved for the OCL. A more accurate calculation of the OCL would have meant that CBLI was demonstrably “balance sheet insolvent”—see paragraphs [33]–[34] above.
(c)CBLI could not pay in full its OCL—whether assessed by its own actuary, or Finity—as at the date of the transactions.
(d)The transactions were entered into during the “restricted period”.
16 See Reserve Bank of New Zealand v CBL Insurance Ltd [2018] NZHC 2969.
(e)Alpha has the obligation to rebut the presumption that the transactions were entered into a time when CBLI was “unable to pay its due debts”.
(f)The evidential standard required to rebut the presumption is on the balance of probabilities. In this case it is agreed that the presumption can either be rebutted or it cannot. No questions as to degrees of probability arise.
(g)The transactions enabled Alpha “… to receive more towards satisfaction” of debts owed to it by CBLI “than it would receive, or be likely to receive,” as a result of CBLI’s liquidation, thereby satisfying the wording of s 292(2)(b).
(h)Alpha concedes it is not running a good faith defence under s 296(3) of the Act.
(i)If the transactions are declared voidable then the relevant orders can be made in terms of the approach and wording that has been agreed between the parties.
[42] Given these concessions, there are two issues left for determination, to which I now turn. First, it will be helpful to briefly outline the arguments of the two parties.
Summary of the parties’ positions
The liquidators’ position
[43] For Mr Chisholm, the central issue in this case is whether, and to what extent, CBLI’s OCL constituted an actual liability, currently accrued, making it a “due debt” which CBLI could not pay at the time of the transactions.
[44] Mr Chisholm was emphatic that the OCL must be considered a “due debt”. The phrase should be given a wide meaning. In this way, the OCL is an actual accrued claims liability (comprising both current and prospective liabilities). It is not a contingent liability, and it is wrong to characterise it in that way. If the OCL is not characterised as a debt, it would be contrary to the policy of the Act and would often
allow transactions by insurance companies to escape the “net” cast over companies under s 292.
[45] Also in this case, using a perfectly legitimate hindsight analysis, CBLI’s cash flow solvency was doomed. Seven days after the transactions were completed it was placed into interim liquidation, and at the time of final liquidation. all its liabilities effectively fell due, whether actual or prospective. To say that CBLI could have technically continued trading for several years (perhaps up to eight years) thereafter and could have paid its debts as they fell due during that time, was a “hypothetical fantasy.” Cash flow insolvency, as well as the obvious balance sheet insolvency, was clear and inevitable at the time of the transactions, as shortly proved to be the case.
Alpha’s position
[46] Mr Barker accepted that CBLI’s OCL was necessary for compliance with the Solvency Standard, and its RBNZ obligations, and as a transparent accounting practice. Nevertheless, in his view the key issue is whether the OCL is “sufficiently certain to crystalise into a legally due debt sufficiently proximate and temporally connected to the transactions sought to be voided”. In framing the issue this way, he adopts the language used by the Supreme Court in David Browne Contractors v Petterson.17 Put more simply, he submits that the issue is whether in these circumstances, the OCL was a “due debt” for the purpose of s 292(2)(a) at the time of the transactions.
[47] Mr Barker’s equally emphatic view was that the OCL was at best an assessment of forward liabilities, which would not crystalise as “debts”, let alone “due debts”, until 10 to 15 years’ time, or even later. He argued that insurance companies are a species of company where creditor protection is primarily delivered by the RBNZ with its close statutory scrutiny and regulatory regime.
17 David Browne Contractors Ltd v Petterson [2017] NZSC 116, [2018] 1 NZLR 112 [David Browne Contractors].
[48] He accepted that on the logic of his argument, s 292 would seldom apply to insurance companies. He reasoned that this provision does not need to apply in this context because invariably the RBNZ would step in and provide protection before insolvent transactions ever occurred, as indeed it tried to do here by issuing a notice forbidding them. The fact that CBLI subsequently failed and was put into liquidation eight months later is said to be irrelevant in establishing whether the OCL should be treated as a debt due at the time of the transactions.
Both parties
[49] Both counsel accepted that the NZ IRFS 4 Standard applied, and that accountancy standards necessitated the inclusion of the OCL in CBLI’s balance sheet. But their strong point of difference is whether the OCL could be said to constitute part of its “due debts”.
[50] Counsel have also agreed there is no relevant New Zealand authority directly on the point of due debts in the context of insurance companies. It was agreed there were three relevant Australian authorities,18 of which New Cap Reinsurance was the most important. It was also agreed that the starting point was the David Browne New Zealand Supreme Court case. As it happened Mr Chisholm and Mr Barker relied on the same cases; they just reached a different view as to the extent to which the cases supported their arguments.
The issues for determination
[51] Given the way counsel framed their submissions, it became clear that there were two key issues for determination:
(a)should the OCL be properly considered a “due debt” for the purposes of s 292(2)(a); and
(b)was CBLI “unable to pay its due debts”?
18 New Cap Reinsurance, above n 3; Taylor v ANZ Banking Group Ltd (1988) 6 ACLC 808; and Insurance Commissioner v Associated Dominions Assurance Society Proprietary Ltd [1958] 89 CLR 78, [1953] HCA 94.
[52] As this is apparently a novel case, I will set out the competing arguments in respect of each issue in some detail.
Should the OCL be properly considered a “due debt” for the purposes of s 292(2)(a)?
Liquidators’ arguments
[53] For Mr Chisholm, a proper understanding of the OCL is the starting point. In his view the OCL is an estimate of an insurer’s present/accrued claims liability (i.e., “expected future payments required to settle claims that have occurred”—even if not known about). This wording is consistent with the liquidators’ expert accountant, Mr David Pacey. It is also quite consistent with paragraph 5.1 of the NZ IFRS 4 (Appendix D).
[54] Mr Chisholm submitted that the OCL cannot be conflated with contingent liabilities and/or possible future liabilities, which he argues is Alpha’s incorrect description of the OCL and ignores the nature of an insurance business.
[55] Mr Chisholm’s strong submission is that the OCL is not a contingent debt/liability at all. It is an estimate of all CBLI’s unpaid claims liability that has accrued. He relies on the definition of OCL in the NZ IFRS 4 (Appendix D), that OCL “[m]eans all unpaid claims and related handling expenses relating to claims incurred prior to the end of the reporting period” (emphasis added).
[56] The OCL is accordingly a present and accrued liability on the “balance sheet” (incorporating both current and prospective liabilities)—but which may not necessarily have immediate cashflow consequences because, for example, the insurer and the insured are not yet aware of the liability/claim or the claim has not yet been processed or fully processed.
[57] The second step in Mr Chisholm’s argument, as I understand it, is that the need for recognising the OCL as a due debt reflects the reality of an insurance business. He relies on the comments of White J in New Cap Reinsurance in the context of s 95A of the Australian Corporations Act 2001 which provides:
95A Solvency and insolvency
(1)A person is solvent if, and only if, the person is able to pay all the person’s debts, as and when they become due and payable.
(2)A person who is not solvent is insolvent.
(Emphasis added).
[58] In that case, a determination as to whether New Cap Reinsurance was solvent was required. White J emphasised:19
Unlike most businesses, an insurer receives its revenue, in the form of premiums, before incurring its major expenses, in the form of claims under the policies of insurance for which the premiums are received. An insurer who is forced to use premium income from newly written policies to meet its existing liabilities, leaving it without resources to meet future claims on those policies, is insolvent as a matter of ‘commercial common sense’… even if there would be a considerable lapse of time before those claims arise …
[59] In the context of an insurance business, Mr Chisholm says it lacks both practical business sense and commercial reality to ignore accrued claims liabilities (which he describes as existing liabilities) in determining whether an insurer can meet its due debts at a particular point in time. The fact that some of these liabilities may not need to be immediately paid does not change the nature of the estimate of the present claims liability already accrued. They are still due and represent actual debts and actual liabilities. Otherwise, for insurance companies, there is no accurate or reliable way to establish whether due debts can be paid.
[60] Contrary to Alpha’s argument, Mr Chisholm maintains that insurers do not make a distinction between current and non-current liabilities. In this way the OCL includes both current and prospective liabilities, not contingent liabilities. Again, relying on the comments of White J in New Cap Reinsurance that “… a prospective debt is not one immediately payable but which will certainly become due in the future
…”.20
19 New Cap Reinsurance, above n 3, at [45] (citation omitted).
20 At [75].
[61] A third plank of Mr Chisholm’s argument is that characterising the OCL as a due debt is consistent with the New Zealand Supreme Court’s purposive approach in David Browne, which is the leading authority on the meaning of “unable to pay due debts”. The Supreme Court confirmed that a key purpose of the voidable transaction regime is to protect an insolvent company’s creditors (as a whole) against the diminution in assets resulting from a transaction which confers an inappropriate advantage on one or more creditors by allowing that creditor to recover more than it would on liquidation.21
[62] In that light, the Court adopted a wide meaning of due debts in finding that “debts” includes both present and contingent debts relying on authorities under similar Australian legislation, including New Cap Reinsurance.
[63] Mr Chisholm submitted, on the authority of David Browne, whether a debt is due is a fact sensitive question that depends on the nature of the company’s business and, if known, its future liabilities. He drew my attention to the following passage in that case:22
Concentrating only on debts due at the relevant time could fail to distinguish between those companies suffering a temporary liquidity problem and those that are on any commercial view insolvent even though they may continue to pay their debts for the next few days or weeks or even months before an inevitable failure.
[64] Mr Chisholm noted the Court’s comments that solvency, in the cashflow sense, must be assessed objectively taking a practical business perspective.
[65] Given the particular characteristics and context of this insurance business, Mr Chisholm submitted that in applying the David Browne principles the OCL must be considered a due debt.
[66] Mr Chisholm also strongly emphasised and relied on the conclusion in New Cap Reinsurance that in assessing an insurance company’s ability to pay all its “… debts, as and when they become due and payable”, there is no reason to exclude
21 David Browne Contractors, above n 17, at [94].
22 At [90].
contingent or prospective debts.23 The final conclusion reached by White J in New Cap Reinsurance is said to be compelling:
[112] For these reasons I consider that the insurance liabilities of NCRA to indemnify its reinsureds in respect of paid losses are “debts” within the meaning of s 95A. Once it is found that NCRA’s liability to indemnify its reinsureds in respect of paid losses were “debts” within the meaning of s 95A, it follows that its future liabilities to pay claims once the level of paid losses reach the required level to trigger a layer of insurance were contingent debts, assuming of course that the relevant event which triggered the reinsured’s liability to its insured, and hence its contingent right to claim on its policy of reinsurance, had occurred. For the reasons already given, I consider that such contingent debts can be taken into account in determining solvency.
[113] It follows that NCRA’s insurance liabilities can be taken into account in determining its solvency.
[67] In Mr Chisholm’s view, New Cap Reinsurance is essentially on all fours with this case and there is no good reason not to follow it.
Alpha’s submissions
[68] Mr Barker’s effective starting point is to focus on the wording of s 292(2)(a) and what is meant by the words “the company is unable to pay its due debts” (emphasis added). In his view, this is a strict “cashflow” insolvency test: an ability to pay debts as and when they become due in the normal course of business. It is not a “balance sheet” insolvency test. He submits that the critical difference from solvency more generally within the Act—being unable to pay debts—is the crucial additional requirement of the word “due”.
[69] Mr Barker emphasised that “balance sheet” insolvency does not satisfy the s 292 test. He noted that it is well-established that the test for whether a company is able to pay its due debts is one focussed on the cash position of the company and not with its position on a “balance sheet” basis.24
23 New Cap Reinsurance, above n 3, at [75].
24 Re Gladding King Real Estate Ltd (in liq) (1993) 6 NZCLC 68,261 (HC) at 68,275.
[70] A key point in Alpha’s position is that the liquidators (and Mr Chisholm) have failed to understand the true nature of the OCL. Mr Barker submitted it is a raw assessment of the level of anticipated future claims. It is not determined by reference to any individual claim, or any particular liability. It is not an amount that CBLI could “pay” at that time, or within any reasonably approximate timeframe and, therefore, remove its liability. Part of the OCL is essentially a provisioning exercise in respect of future uncertain liability.
[71] In elaboration of this argument, Mr Barker explained that the Supreme Court in David Browne relaxed the previous strict approach which generally meant that a legally due debt was a debt due for payment (and would generally exclude contingent debts). He accepted that “due debts” are now not so limited and can include contingent debts where a legal obligation to pay is both “sufficiently certain and reasonably temporarily proximate”, which adopts the words used in David Browne.
[72] In these circumstances the OCL did not, and could not, satisfy that test. He noted that the NZ IFRS 4 (Appendix D) is the actuary’s best estimate of an insurer’s total likely claims exposure for the cover it has written. The Appendix allows for an additional risk margin to reflect the inherent uncertainty in the estimate. Moreover, Appendix D does not even refer to debts or due debts.
[73] Mr Barker emphasises that in estimating the OCL, no distinction is drawn between current or non-current liabilities. Tellingly, he says, the OCL will include elements that might need to be paid in two months and others that might not crystalise into an actual claim for 10 years. It might also include estimated claims that never arise at all.
[74] The high degree of uncertainty in the calculation of the OCL is illustrated by the actuarial assessment of CBLI’s OCL completed for the liquidators in the present case by Mr Atkins’ firm, Finity. There was a difference in range of NZD 436 million. The high estimate was 60 per cent higher than the low end of the range. On the liquidators’ case, potentially millions of dollars of projected exposure that may never eventuate as actual insurance claims would have to be classed as debts.
[75] In all these circumstances, Mr Barker submitted it is difficult to see how the OCL could ever be classified as a due debt. At most it is an estimate of the level of debts that may or may not become due in the future.
[76] Mr Barker accepts there are good prudential reasons for requiring insurers to provide an OCL and for it to be recorded in the balance sheet. However, it does not follow that an OCL represents a contingent liability that is sufficiently certain to become a due debt for the purpose of an assessment under s 292. In his view, the approach suggested by the liquidators stretches the words of s 292, and the test set out by the Supreme Court in David Browne, to breaking point.
[77] Even if the OCL could itself be characterised as a debt, it could not be a debt that would become due within a reasonably proximate time. In his submission, there is clearly insufficient certainty that a yet to be filed claim would crystalise into an actual due debt within any reasonable or temporally proximate period. Indeed, given the particular characteristics of claims in respect of French building construction, it might be as long as 16 years. Even the timeframe provided by Mr Atkins (that CBLI would have been able to continue to pay its debts for the next three to five years) simply cannot be reconciled with the David Browne requirement that a liability must become due within a relevant, proximate period.
[78] I have already noted Mr Barker’s policy argument that, properly applied, the IPSA regulatory regime will avoid the need for s 292 arguments (because they will not arise). He also added another policy argument: the simplicity of the cashflow solvency test under s 292 will be defeated if the Court “is drawn into speculative assessments of the future prospects of any company with a deficiency of assets against liability on its balance sheet”.
Discussion: Is the OCL a “due debt?”
[79] As I observed at the conclusion of the oral submissions, both arguments are strong and compelling, which is no more than to explain why this issue is before the Court and why the decision is difficult.
[80] The starting point must be the language of s 292(2)(a) and the words that a “company is unable to pay its due debts”. I accept that this effectively imposes a cashflow solvency test. Clearly Parliament could have additionally imposed, in some circumstances, a balance sheet solvency test. But it chose not to. The words must be interpreted as they appear.
[81] In approaching the interpretation of these words, I remind myself of s 10(1) of the Legislation Act 2019, which provides that the meaning of legislation must be ascertained from its text and in the light of its purpose and its context. The section does not permit the Court to re-write the provisions of an enactment, even if it may be suspected that a particular result is unintended. If the words are plain and unambiguous, they must be applied as they stand. It is a well-established principle that a court of construction is obliged to give effect to the clearly expressed intention of the legislature, even if the result is not altogether satisfying from a lawyer’s point of view.25
[82] The real question here is whether the words due debts are “plain and unambiguous” so that they can only bear one interpretation, as argued for by Mr Barker, or if a wider view can be taken of those words, as argued by Mr Chisholm.
[83] In this respect, Mr Barker’s argument is very attractive. He takes an elegantly simple approach to analysing the nature of the OCL. Given his submission that the OCL is effectively an ‘estimate’ of the level of debt that may or may not become due in the future, it is simple to conclude that it cannot be an actual debt. This is because it does not describe, on Mr Barker’s argument, any present liability in the sense of a liability that can and must now be paid or quantified. It is a fair question to ask, how could a claim that has not even been advised to the insurer be characterised as a debt that is presently due? It is neither certain nor temporally connected to the transactions sought to be set aside. As he put it more bluntly, Alpha has not even been involved in calculating the OCL. It is solely CBLI’s actuary’s assessment of its liability. How can it be a debt when one party (Alpha) to the debt has not even been involved in its
25 See Williams v Hutt Valley & Bays Fire Board [1967] NZLR 123 (CA) at 129 per North P.
calculation, or ascertained which claim(s) it might relate to, or even more fundamentally, has not even made a claim?
[84] Mr Barker’s argument can be summarised very simply: while the OCL may be a regulatory requirement (and for good reason)—notwithstanding that the estimated liabilities are best understood as being contingent—they are not sufficiently certain liabilities to become a due debt for the purpose of assessment under s 292.
[85] Resolution of this issue is not easy. I was initially attracted to Mr Barker’s reasoning. However, upon consideration and reflection, I have come to the view that I should not accept it. I accept that the argument as detailed by Mr Chisholm is the better view. I conclude that a wider interpretation of due debts is possible and justified. To do so does not involve stretching the language of s 292(2)(a) beyond breaking point. But it does take the section further than a New Zealand court has so far needed to go. There are five principal reasons upon which I rely. I list them and then address each in turn:
(a)Treating the OCL as a due debt is the only way to truly and reliably establish whether a reinsurance company is cash flow solvent, i.e., able to pay its due debts.
(b)Expert accountants/actuaries believe the OCL should be regarded as a due debt.
(c)Classifying the OCL as a due debt is consistent with the Supreme Court’s wide interpretation of “due debt”.
(d)This approach is consistent with the Australian jurisprudence.
(e)The purpose of the Act is best served by treating the OCL as a due debt, and there are sound policy reasons for doing so.
(a) Treating the OCL as a due debt is the only way to truly and reliably establish whether a reinsurance company is cashflow solvent
[86] It is critical to understand the nature of the reinsurance business. Particularly, the fact that insurance companies receive vast amounts of money (premiums) “up front” and that there is a significant time-lag until insurance claims become payable. This is particularly the case for building construction reinsurance policies in France, where the builders’ warranty policies cover builder’s defects for up to ten years. Context is everything.
[87] In that sense, reinsurance companies, and probably insurance companies generally, are a particular species of company to which a conventional assessment of ability to pay due debts is not easily carried out. If Mr Barker is right, then virtually every reinsurance company would be able to claim it was cashflow positive and able to pay its due debts for years to come—even though it is cash flow doomed and must inevitably fail for that reason. I agree that cannot be a justification to mangle the words of s 292(2)(a). However, it does aid a contextual understanding of the nature of the business (here the reinsurance business for European builders’ construction insurance) and how most of its debts will fall due. This approach is consistent with David Browne, which in the general sense holds that it is legitimate, and indeed necessary, in undertaking the inquiry required by s 292, to take into account the nature of the company’s business.26
[88] In my view, treating the OCL as a present, accrued liability (albeit with a significant component of prospective liability included), as argued by Mr Chisholm, is the only effective way of determining a reinsurance company’s ability to pay its due debts. I accept this concept is somewhat of an accounting fiction. But without it, the “due debts” assessment is even more fictitious and would be virtually meaningless.
[89] In short, the due debt assessment should be grounded in the reality of the business involved and should surely reflect the way debts for insurance companies will inevitably fall due. In this context, the OCL should not be regarded as an off- balance sheet contingent liability.
26 See David Browne Contractors, above n 17, at [90].
(b)Expert accountants/actuaries believe the OCL should be regarded as a due debt
[90] Another reason for treating the OCL as a “due debt” is that the expert accountants/actuaries believe this is not only good practice, but is required to understand cashflow liability for reinsurance companies. Mr Atkins comments are illuminating:
[13] A general insurance company is not regarded as solvent merely because it has cash in the bank at any particular point in time. Insolvency for an insurer is always assessed based on the difference between the assets of the company and the liabilities including those for insurance policies previously issued.
[14] I am not aware of any jurisdiction or regulatory system in the world that would regard a general insurance company as solvent just because it has cash in the bank.
[15] For CBLI, the situation is exacerbated because so much of its insurance is ‘long tail’ business, with many policies covering a period of ten years. The insurance liabilities are larger relative to premium income and more difficult to estimate, but the underlying issue is the same.
[16] Another indication that all of the insurance liabilities must be taken into account derives from the observation that most insurers do not distinguish ‘current’ and ‘non-current’ liabilities for outstanding claims on the balance sheet. This is because there is no relevant distinction between whether the claim might be expected to be paid in two months or two years or ten years. The situation is very different for a trading debt, for example, where it is important for users of the accounts to know whether and when such debts needs to be paid or refinanced.
…
[18] However, given the significant outstanding claims liabilities, CBLI was not in a position to pay its due liabilities. The Outstanding Claim Liabilities (OCL) are due liabilities, albeit timing and quantum is uncertain. The OCL have been incurred and CBLI is legally and contractually obliged to pay the OCL as they arise and there is no ability to avoid that liability. It is clear from the balance sheet that CBLI would not be in a position to pay these liabilities.
[91] It is fair for Mr Barker to point out that Mr Atkins, at a previous stage (in his first affidavit), deposed that CBLI was cashflow solvent. However, that was before the s 292(2)(a) issue had arisen. When he was later asked to reflect on that conclusion in the light of the s 292 test and the particular features of a “long tail” reinsurance company, Mr Atkins revised his view. In fairness, even Mr Ross Simmonds (one of the experts retained by Alpha) appropriately acknowledged that was the case. In his second affidavit, Mr Simmonds noted:
[11] As Mr Atkins did not address or attempt to address the issue of cashflow solvency, the analysis and information contained in his July 2018 affidavit is of little assistance to me in addressing the cashflow insolvency issue now.
[92] The expert opinion of Mr David Pacey, an eminently qualified New Zealand accountant, who provided an affidavit as an expert witness, is also persuasive. He deposed:
… I address the general principles of NZ IFRS 4 in relation to the determination of OCL in the paragraphs which follow to support my view that OCL, which includes IBNR, is a liability of CBLI at each reporting date. That is, OCL represents an amount owed and is thus a debt due at the reporting date. In this matter, I concur with paragraph 5 of Ms Johnstone’s affidavit, in responding to the affidavits of Mr Simmonds and Mr Frederiksen, that “OCL is not an off-balance sheet item or a contingent obligation”. In concurring on this matter, I make no comment on the quantum or sufficiency of the OCL disclosed by CBLI in any financial report.
[93]He also specifically concluded:
[22] Based on the preceding summary guidance from NZ IFRS 4 it is my view that the determination of OCL should include an assessment of any expected future payments required to settle claims that have occurred, but which are not yet reported to the insurer as at the balance date or where not enough has been reported to the insurer, i.e., a due debt at the balance date.
[94] To similar effect is the opinion of Ms Johnstone (referred to by Mr Pacey, above). Admittedly, she is one of the two appointed liquidators, and not therefore truly independent, but she is a chartered accountant and licensed insolvency practitioner nonetheless. She deposes:
[5] For the purposes of determining the solvency of a company, you need to include OCL as due obligations. This is important for an insurance company which has existing accrued liabilities which it is liable to pay. OCL is not an off-balance sheet item or a contingent obligation. It is an estimate of a liability which is due but has not been quantified. Under IFRS 4 those existing obligations are required to be quantified and accounted for as they are actual liabilities that arise from a contractual obligation that is a due debt.
[95] The relevant affidavits on behalf of Alpha are not so compelling on this point. The first is from Mr Boris Frederiksen, who is the Danish “insolvency administrator” for Alpha. Quite responsibly, he does not attempt to give an opinion on whether the OCL should be treated as a due debt for s 292 purposes, recognising that is an issue for the New Zealand courts to resolve.
[96] The other expert for Alpha is Mr Ross Simmonds, who is a New Zealand expert actuary with extensive experience in the calculation of insurance liabilities. Neither does he appear to express a firm and detailed opinion about whether the OCL constitutes a due debt. The thrust of his actuarial opinion seems to be that it does not, or perhaps, that it is too imprecise a concept to qualify as a due debt.
[97] The difficulty for me is all the affidavit evidence was accepted without cross- examination. On the material before me, the accounting experts in particular, make a strong case as to why in the circumstances of CBLI’s business, the OCL should be classified as a due debt.
[98] Now, I accept that this issue is not to be determined by accountants and actuaries. It is a legal test which requires a court ruling. But that is not to say that the views of duly qualified experts are irrelevant. It would be unsatisfactory if the legal test diverged completely from an expert accounting analysis.
(c) Classifying the OCL as a due debt is consistent with the Supreme Court’s wide interpretation of “due debt”
[99] There are two words to consider: “debt” and “due”. I deal with them separately, as follows, beginning with “debt”.
[100] As was explained in David Browne, “debt” can have a wide meaning and there is no reason to restrict it:
[88] Debt can be a word of wide import and its exact meaning will depend on the context. With regard to similar legislation in Australia, it has been held that the term debt encompasses both present and contingent debts. The same result has been reached in the United Kingdom. We see no reason to take a different approach in New Zealand.
[101] In my view, the “wide import” noted by the Court can stretch sufficiently so that the OCL comes withing its meaning, even though the OCL contains a significant prospective component. As will be clear, I have reached the view that the OCL should probably not be considered a contingent debt in the circumstances of CBLI’s reinsurance business. I accept therefore, that including the OCL within the meaning of “debt”, may be seen as taking the words of the Supreme Court too far given that the
Court referred only to “present and contingent” debts. However, the issue in this case was not before the Court in David Browne and now further refinement is necessary. In any case, as I have already indicated, I accept that the OCL should be seen as a “present” accrued debt even if it contains a significant element of estimated prospective liability. As the Supreme Court emphasised, the “… exact meaning [of the word debt] will depend on the context”.27 I have already sufficiently explained the context here which justifies the conclusion that the OCL is a debt.
[102] I acknowledge that perhaps the biggest problem with my conclusion is that the OCL is by its nature an estimate, encompassing “expected future payments”.28 As Mr Barker emphasised, here Mr Atkins from Finity provided a high and low estimate of the appropriate OCL for CBLI. How is it, he asks, that a figure representing such a range could ever be called a debt? (Those estimates were after the event and clearly without the full business information available). The first answer to Mr Barker’s concerns is that RBNZ approves the appointment of an actuary who makes a prospective analysis—compared to the retrospective analysis provided by Mr Atkins—and with all the details available as to trends for insurance claims, pay- outs, and time frames. But the second and more important answer is that the OCL consists of many component parts, not just prospective claims not yet known. It includes actual claims known about but not yet paid. The OCL should be understood as a whole, as being the amount required to be included in the accounts for that year to ensure the insurance company can continue to operate. In that sense the OCL is a “debt” and it does not stretch the words of s 292, according to the provision’s purpose, to reach that conclusion.
[103] I turn to the word “due”. The short answer is that if, as I have held, the OCL is treated as a present accrued liability (a debt, albeit including some prospective future liabilities) then it is “due” now. At the least it is due in the year for which it is assessed. I consider Mr Barker falls into the trap of breaking down the OCL into its individual parts and focusing just on claims that may, or may not, eventuate. That is an overly refined approach. The OCL also includes claims that are very much known about and are awaiting final quantification. It is the OCL globally, as fixed by the approved
27 At [88].
28 See NZ IFRS 4, at [5.2], as discussed in more detail above from [24].
actuary, which is the “debt”. And it is that debt which a company must be able to pay in that year. It is thus very much “due”. On that interpretation, this is the end of the matter.
[104] If I am wrong on that point, and distant claims liabilities must be considered— so that a much longer time frame for the word “due” is necessary—then I am prepared to bite that interpretive bullet. Again, the comments in David Browne are helpful:
[89] The next issue is what is meant by the term “due”. Contractors and Mechanical accept that, as a matter of commercial practicality when assessing solvency, it is appropriate to take into account both recent past events and those subsequent to the transactions under scrutiny. They argue, however, that the emphasis is on a very short period of time in the future. The liquidator, on the other hand, argues that future debts should be taken into account if they are “reasonably temporally proximate”.
[90] We accept the liquidator’s submission on this point. As was said in Re Cheyne Finance Plc (No 2), the issue of “how far into the future the inquiry as to present solvency is to go … is a fact sensitive question depending upon the nature of the company’s business and if known, of its future liabilities”. Concentrating only on debts due at the relevant time could fail to distinguish between those companies suffering a temporary liquidity problem and those that are, on any commercial view, insolvent even though able to continue to pay their debts “for the next few days, weeks or even months before an inevitable failure”.
[91] Solvency in a cash flow sense must be assessed objectively, taking a practical business prospective. What is reasonably temporally proximate will, as indicated above, fall to be considered in light of the facts of the particular case. If a reasonable and prudent business person would be satisfied that there is sufficient certainty that a contingent debt will, within that relevant period, become legally due then it must be taken into account.
[92] We accept that there is a difference between debts and damages but, once liability and quantum have been established in any claim for damages, the resulting judgment sum will be a debt owing. This means that the question is the same – if there is sufficient certainty that a claim will crystallize in the relevant period, then it must be taken into account. This approach is consistent with the approach taken by the High Court of Australia in Bank of Australasia v Hall.
[105] Mr Barker is right to emphasise that at least part of the OCL includes claims not yet reported which may not crystalise for over a decade. There will be no temporal connection or certainty about such individual claims. But there are two answers to that. One is that there must be a common sense and practical business perspective in light of the way an insurance company operates. More importantly, as I have observed, the OCL is assessed as an overall liability. Even if not payable now, as an overall
amount, actuarily assessed, it will eventually, and in a general sense, become payable. I do not think that involves inappropriate mental gymnastics. The point of the OCL is that it represents a global figure for all claims.
[106] It is also worth emphasising, as with the interpretation of “debt”, that the OCL must, under the NZ IFRS 4 (Appendix D), comprise liabilities that are known, crystallised and which will be soon quantified. It is too sweeping for Mr Barker to categorise the whole OCL as only being a contingent and future, uncrystallised liability. I was not given clear information as to how the OCL for CBLI was made up, in its inadequate or adequate form. But it certainly contains a significant element of specific individual claims that will become finally certain within an appropriately proximate time frame. I cannot accept that just because some of the OCL includes unknown claims years into the future, then the whole of the OCL cannot be considered “due”.
[107] This purposive approach again is consistent with David Browne. As was mentioned in David Browne, the issue of “how far into the future the inquiry as to present solvency is to go … is a fact sensitive question depending upon the nature of the company’s business and if known, of its future liabilities”.29
[108] Given the particular nature of the reinsurance business, it is necessary to go a long way into the future and I see no way around this. Otherwise, a totally artificial picture is painted of a company’s ability to pay its due debts.
(d)The approach is consistent with the Australian jurisprudence
[109] Next, it is necessary to say that treating the OCL as a “due debt” is consistent with the Australian approach as set out in New Cap Reinsurance.30 While Mr Barker is right to say that New Cap Reinsurance was not, on its face, a case about illegal or voidable transactions, it was about the determination of whether New Cap Reinsurance was solvent. Therefore, given the very similar definition of solvency, effectively the same issue arises in this case. In my view, to distinguish New Cap Reinsurance and
29 David Browne Contractors, above n 17, at [90], quoting Re Cheyne Finance Plc (No 2) [2007] EWHC 2402, [2008] 2 All ER 987 (Ch) at [50].
30 New Cap Reinsurance, above n 3.
to say that it is not on all fours with this case, is akin to “dancing on the head of a pin”. I have derived considerable assistance from the reasoning in New Cap Reinsurance, and I note that it was referred to with approval by our Supreme Court in David Browne.
[110] Commonality of approach with Australia, where possible, is commercially desirable. It would be unfortunate if on such a significant issue, Australian and New Zealand courts were divided. There is no compelling reason to do so. In this case, without doing violence to the wording of s 292, the same general conclusion as in New Cap Reinsurance can and, in my view should, be reached.
(e)Policy reasons for treating the OCL as a due debt
[111] There are also sound policy reasons supporting my conclusion. To start with, the ramifications of Mr Barker’s argument are not appealing. It would mean that virtually every insurance company would be immune from the provisions of s 292. That is, an insurance company would always be able to say that even though we cannot pay our OCL, we have sufficient cash resources for us to be able to meet our due debts for several years to come—just as Mr Barker asserts here. That would sever any proximate or temporal relationship with the transactions sought to be voided and render s 292 inoperable. Realistically, if the OCL is not to be considered a due debt, the heading to s 292 should be re-written to say, “Insolvent transaction voidable — except in the case of reinsurance companies”.
[112] During the time an insurance company could technically pay its due debts, other than its OCL, it could prefer certain creditors deliberately, imprudently, and even dishonestly. To that extent, s 292 would be rendered nugatory. It would also mean that an insurance company, knowing it was balance sheet insolvent, as here, could continue to operate, while all the while knowing that any preferential transactions it made could not be voided. This is not consistent with the purposes of the Act. Nor does it promote commercial certainty.
[113] Mr Barker’s argument that robust oversight by the RBNZ would prevent this sort of situation happening, is not borne out by these facts. Section 292 protections will still sometimes be necessary, and it is important that, wherever possible, s 292 be interpreted so that this can be provided.
[114] All this is not to say that persuasive policy reasons should mangle the interpretation of an Act or stretch it beyond breaking point. But this is not necessary to reach the interpretation of s 292 in this case.
[115] It is tempting in this discussion to make general comments about the OCL that are beyond what is required for resolution of this case. I conclude that in this case, and in these circumstances, the OCL should be considered and included in the “due debts” analysis for CBLI.
Conclusion as to due debts
[116] In this case the OCL should be considered a due debt because the specific nature of the reinsurance business requires it. The preponderance of expert accounting evidence is in support. To include the OCL is the only way to accurately assess whether a reinsurance company’s debts can be paid. It is consistent with a wide interpretation of due debt employed in David Browne; it is consistent with the conclusions reached in Australia, and there are sound policy reasons to consider the OCL as a due debt.
[117] I accept that the conclusion I have reached significantly widens the s 292 test that has so far been applied. While it is at the outer limits of what the wording of s 292 allows, I do not think it exceeds them. And it does not do violence to the language used in the provision to treat the OCL in this way. I now turn to the second issue.
Was CBLI “unable to pay its due debts”?
The short answer
[118] As I understand, counsel’s agreed position, upon a finding that the OCL was a due debt, is that CBLI could not pay it, and that the transactions must be void. Strictly speaking I need go no further.
The long answer
[119] However, if I am wrong, then in my view Alpha has not discharged its onus to rebut the presumption on the balance of probabilities that CBLI was unable to pay its due debts at the time of the transactions. What follows is why I reach that conclusion. Counsel made detailed submissions on this point, and I first set them out.
Liquidators’ submissions
[120] Mr Chisholm’s argument is that hindsight is a wonderful thing and it can quite properly be relied upon in this case. He noted the comments in New Cap Reinsurance as to the “inestimable advantage of hindsight”,31 where, as here, solvency (ability to pay all debts as and when they become due and payable) falls to be assessed retrospectively. Mr Chisholm is clear that there is no need to speculate as to whether CBLI was able to pay its due debts. In Mr Chisholm’s view, we know that for a stone- cold certainty. That is because, within eight days of the transactions when CBLI was placed into interim liquidation, it was unable to pay its debts. For Mr Chisholm, it is a “hypothetical fantasy” to suggest that CBLI could have continued trading for up to eight years given its reserves and could have paid its debts as and when they fell due during that period.
[121] The commercial reality was all CBLI’s liabilities, whether current or prospective, fell due for payment within a few days after the transactions and at least by the time of the final liquidation, eight months later. And by that time, it was certain that CBLI could not pay them.
[122] Mr Chisholm emphasises that the transactions in issue were not made in the ordinary course of trading, but were made after months of concern by the RBNZ and Alpha’s overseas regulator. These included serious concerns regarding CBLI’s reserves and solvency. Its known financial position had deteriorated markedly. In November 2017, CBLI advised it was likely to breach its solvency ratio as at 31 December 2017. RBNZ had directed CBLI not to make payments under the LSA, which demonstrates its concerns about CBLI’s ongoing “trading”. Thus, the chance
31 New Cap Reinsurance, above n 3, at [51].
of CBLI continuing in business under the control of its directors was virtually non- existent. In Mr Chisholm’s view, it is ludicrous to suggest the company could have continued to pay its due debts. It is clear beyond argument, given its liquidation, that it was unable to pay its due debts during the restricted period.
Alpha’s submissions
[123] Mr Barker argued that the evidence clearly establishes that CBLI was “cash- flow” solvent. He points to the evidence of Mr Simmonds, an expert actuary retained by Alpha. In his affidavit he deposes that “CBLI was able to pay its due debts at the time it was placed into interim liquidation”. He explains how CBLI’s ability to continue to pay its debts would diminish over time. Mr Simmonds deposes that exactly when CBLI reached the stage that it could not meet its due debts would be dependent on how its insurance liabilities in relation to its French builders’ warranty policies developed over time. Mr Simmonds estimated that CBLI would be able to meet its due debts for the next eight years.
[124] Mr Barker also noted Mr Atkins’ (the applicant’s expert) “rough estimate” that CBLI might only have been able to continue to pay claims from between three to five years.
[125] Even more revealingly, Mr Barker emphasised Mr Atkins own evidence— where Mr Atkins confirmed he had previously provided an opinion that CBLI was not cashflow insolvent in 2017 or 2018 “… because the company had cash and investments at the time.” Mr Barker accepted that was a position Mr Atkins has now altered. In his most recent evidence on the issue as to whether the OCL is a due liability, he said that it was, and on whether CBLI would have been in a position to pay those liabilities at the time of the transactions, he said that it could not.
[126] Mr Barker’s strong submission was that CBLI was cashflow solvent. It had cash in hand in excess, and significantly in excess, of insurance payables and other payables—as at 31 December 2017 and on 28 February 2018.
[127] As a further plank to the argument that CBLI was able to pay its due debts, Mr Barker referred in some detail to the decision of Courtney J giving reasons for the order she made placing CBLI into liquidation.32 Mr Barker noted that RBNZ did not argue at the time that CBLI was unable to pay its debts. Indeed, Courtney J observed that:
[30] The company’s financial position would not have brought the case within s 151(2)(a), which requires the company not to be able to pay its debts (cashflow insolvency). But the Bank submitted that, in the context of an insurance company the test of cashflow insolvency is of limited use; an immediate cashflow shortage is rarely the reason for an insurer’s insolvency and an insurer that is able to meet its day-to-day debts immediately may nevertheless be insolvent. The Bank asserted, however, that CBLI’s liabilities substantially exceeded its assets, so that it was balance sheet insolvent, which was significant given that its largest exposure lies in future long-tail claims.
[128] As Mr Barker pointed out, instead, the RBNZ relied on s 151(b)–(d): breaches of the solvency standard; non-compliance with the RBNZ’s directions; and the just and equitable ground.33
Discussion
[129] The first hurdle is the comments by Courtney J at the time CBLI was placed into liquidation. With respect, it is easily jumped. Those comments were made in passing and strictly speaking are obiter. In any case, no specific argument was directed to that point. Indeed, Mr Atkins completely changed his mind on that point when he specifically addressed it.
[130] Secondly, it is obvious that the experts are divided, and in any case, they are all speaking hypothetically. I also get the sense from their affidavits that they are focussing primarily on the nature of the OCL and to what extent that contributes to CBLI’s indebtedness. In the absence of cross-examination, it is very difficult to assess and compare their evidence.
32 See Reserve Bank of New Zealand v CBL Insurance Ltd, above n 16.
33 See [14] and [36]–[37] of this judgment.
[131] From my point of view, the compelling argument is the reality of what happened. Whatever the educated best guess of the experts as to how long CBLI could continue to trade, the reality is that at the time of its liquidation it did not, and could not, pay its due debts. For every dollar a creditor is owed it is unlikely they will fully recover, in fact as I understand it, they are likely to recover 50 cents to the dollar or even less. That tells its own clear and unarguable story.
[132] I accept that the interim liquidation would have caused a loss of public confidence and that would have made ongoing trading difficult. I also accept I have not heard detailed evidence as to the accounts and debts due. However, on what was before me, Alpha has not rebutted the statutory presumption that at the time of the transactions CBLI could not pay its due debts.
Result
[133]The transactions are set aside as insolvent transactions under s 292 of the Act.
[134] Counsel informed me that if this was the result, they would be able to agree on the precise form of appropriate orders. I will await their joint memorandum. Ten working days from the date of this decision would seem an appropriate time frame.
[135] There is no reason why costs should not follow the result, on a 2B basis. If counsel cannot agree on costs and disbursements, then memoranda (no more than three pages) should be filed. CBLI is to file within 15 working days from the date of this decision, with Alpha to file within a further seven working days.
Becroft J
0
6
1