O v. S

HCCT 82/2026

[2026] HKCFI 4910

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

CONSTRUCTION AND ARBITRATION PROCEEDINGS NO 82 OF 2026

________________

________________

________________

________________

D E C I S I O N

________________

I.  INTRODUCTION

1.  This is the Originating Summons filed by the Plaintiff (“Party P”) on 23 March 2026 (the “Originating Summons”) against the Defendant (“Party D”) seeking to:-

(1)  set aside the whole partial award issued on 23 December 2025 (the “Partial Award”) by a 3-member tribunal (the “Tribunal”) under the Hong Kong International Arbitration Centre (the “HKIAC”) in an arbitration (the “Subject Arbitration”) commenced by Party D against Party P, on the ground of apparent bias in that that a member (“Member R”) also the presiding arbitrator of the Tribunal had in an earlier arbitration (the “Prior Arbitration”) ruled against one of the Party P’s witnesses (“Witness Z”); or

(2)  alternatively, set aside paragraphs 289 to 291 and 391.5 of the Partial Award on the ground that these paragraphs concern issues which the parties had agreed should be dealt with in the first phase (“Phase 1”) of the Subject Arbitration (resulting in the Partial Award) rather than leaving these issues to be dealt with in the second phase (“Phase 2”) of the Subject Arbitration.

II.  UNDERLYING DISPUTE IN THE SUBJECT ARBITRATION

2.  To understand Party P’s complaint, it is necessary to briefly understand the underlying dispute in the Subject Arbitration.

3.  By an investment management agreement dated 25 September 2019 (the “IMA”), Party P agreed to act as Party D’s investment manager in respect of US$117 million (being part of Party D’s proceeds obtained in Hong Kong) deposited into Party D’s account opened with Party P (the “Managed Assets”).

4.  By an investment instructions given by Party D on 25 September 2019 (the “Investment Instructions”) to Party P, Party D instructed Party P to invest US$117 million in a bond (the “LRC Bond”) issued by a company I shall refer to as “LRC”. Witness Z was and is the founder and chairman of the parent group of LRC.

5.  On 14 October 2020, Party D withdrew, or purported to withdraw, US$1.34 million of the LRC, and accordingly, on 23 October 2020, a company related to LRC, which I shall refer to as “LRC China”, paid Party D US$1.34 million.

6.  On 30 December 2020, all the remaining the LRC Bond was withdrawn from Party D’s portfolio with Party P. In replacement, three promissory notes (“Note G”, “Note H” and “Note S” respectively and collectively “Three Notes”) issued by three companies (“Company G”, “Company H” and “Company S” respectively) were deposited into the portfolio.

7.  Between October 2020 and May 2021, Party D recovered US$30.64 million of its fund (including funds in some of Note H) from Party P.

8.  On 25 September 2021, Party D sent Party P a notice (the “Termination Notice”) to terminate the IMA on the basis that the IMA’s initial fixed term of two years had expired. By the Termination Notice, Party D also demanded the immediate return of the Managed Assets.

9.  Party D declined on the basis that on a proper construction of the IMA (in particular Clauses 1.3 and 9.1), the IMA would continue indefinitely until both parties agreed to terminate.

10.  On 21 February 2023, Note G were redeemed and Party D received US$20 million.

11.  On 27 June 2023, Party D caused the remaining funds in Note H and Note S, in the sum of US$9.7 million and US$39 million respectively (totalling US$48.7 million), to be withdrawn, and replaced with a US$48.7 million perpetual bond (the “Fund Note”) issued by a fund which I shall referred to as “Fund” was deposited.

III.  THE SUBJECT ARBITRATION

12.  On 29 August 2023, Party D commenced the Subject Arbitration by filing a Notice of Arbitration with the HKIAC, seeking, among others:-

(1)  A declaration that the IMA had expired;

(2)  An order that Party P is liable to Party D for breach of the IMA and its fiduciary duties, and for converting Party D’s funds to its own benefit; and

(3)  An order that Party P return or pay Party D the Managed Assets and interest, and other damages.

13.  On 15 March 2024, Party D filed its Statement of Claim.

14.  On 24 June 2024, Party P filed its Statement of Defence. In the Statement of Defence, LRC and LRC Bond were mentioned for the first time in the Subject Arbitration.

15.  On 7 December 2024, Party D filed its Statement of Reply and Defence to Counterclaims. In its Reply, Witness Z was mentioned for the first time in the Subject Arbitration.

16.  On 5 February 2025, Party P filed its Rejoinder, together with, among others, the Witness Statement of Witness Z.

17.  On 8 March 2025, upon the parties’ agreement (the “Bifurcation Agreement”) evidenced in Party D’s letter of 7 March 2025, the Tribunal issued an order (the “Bifurcation Order”) that the arbitration be bifurcated into two phases: Phase 1 for “liability and already-quantified main claims/counterclaims damages”, and Phase 2 for “remaining yet-to-be-quantified other claims/counterclaims damages”. In the Bifurcation Order was a table listing out items for Phase 1 and items for Phase 2. The following items in the table are salient to the present application:-

(1)  One of the items for Phase 1 listed out in the said table was Item 1 thereunder:-

“[Party D’s] claim for US$66,360,666.11 [that is, the Managed Assets] to be returned, repaid, accounted for, and/or restored to [Party D] (inclusive of [Party D’s] claim for an account of all necessary inquiries to be taken).”

(2)  Relatedly, one of the items for Phase 2 listed out in the said table was Item 2 thereunder:-

“[Party D]’s claim for additional losses and damages, including those for lost opportunity costs, interference with ability to pursue alternative investment opportunities, and reputational damage, suffered as a result of [Party P]’s failure to return [the Managed Assets]…”

18.  From 14 to 16 April 2025, the oral hearing (the “Subject Arbitration Oral Hearing”) of Phase 1 took place before the Tribunal. On 6 October 2025, the Tribunal emailed the parties:-

“In accordance with Article 31.1 of the 2018 HKIAC Administered Arbitration Rules, the Tribunal hereby declares that the proceedings of this phase of the arbitration are now closed. The Tribunal will proceed to finalise and issue its award as soon as possible.”

IV.  THE PARTIAL AWARD

19.  I am not concerned here with the substantive merits of the Partial Award. My task here, mainly concerned with whether there is any apparent bias as shall be explained later and whether the Tribunal went beyond the Bifurcation Agreement and the Bifurcation Order, does not require me to review the substantive merits of the Partial Award. That said, for these questions, the Partial Award itself is an important document to consider. The Partial Award is worth some details.

20.  At §108 of the Partial Award, the Tribunal listed out the issues for determination, among which were:-

“108.1 First, whether the IMA was terminated by [Party D’s] Termination Notice or automatically expired on 25 September 2021, under Clauses 1.3 or 9.1 of the IMA.

108.2 Second, if the answer to the first issue is in the negative…

108.3 Third, if the answer to the first or second issue is positive, is [Party D] entitled to the return of the Managed Assets in the investment portfolio or the equivalent value thereof, and if so (a) what are the Managed Assets? (b) what is the proper date of calculation of their value? (c) what is the amount for which [Party P] is liable to [Party D]?

108.4 Fourth, is [Party P] liable for breach of its duty to exercise reasonable care and skill in managing [Party D’s] IMA portfolio?”

21.  The structure of the Partial Award is that after listing out the issues, a section was devoted to each issue, and under each section, the Tribunal summarised the parties’ contentions under the subheading “Parties’ Contentions”, and set out the Tribunal’s own analysis and decisions under the subheading “Tribunal’s Analysis and Decisions”.

22.  For the first issue, at §§114-140, under the subheading “Parties’ Contentions”, the Tribunal set out Party D’s submissions and Party P’s submissions. Party P’s position was that the IMA was to continue perpetually until the parties agreed to terminate, while Party D’s position was that it was not perpetual. One of the arguments Party P raised in support of its position was, briefly, that LRC Bond was a perpetual bond with no maturity date, and thus the IMA, whereunder investment in the LRC Bond was made, should also be construed to mean perpetually. In this regard, from §142 onwards, the Tribunal summarised Party P’s submissions as follows:-

“142. [Party P] submits that it is important to note that the [LRC Bond] is a perpetual bond with no maturity date, in other words that it would continue to be effective unless both Parties agreed to terminate. In this regard, [Party P] relies on the one-age undated [LRC Bond] (Exhibit R-120) that it produced shortly after the conclusion of the Hearing on 16 April 2025. [Party P] also relies on the oral testimony of [Witness Z] (who had provided the copy of Exhibit R-120 to [Party P]), who noted that Exhibit R-120 was a bond in the amount of US$117 million with an interest rate of 0.25%, with [LRC] having a right of early redemption, but that it would need to pay interest at 6% if it redeemed early, before the term was over, and that there would be one payment every ten years.

143…Accordingly, submits [Party P], ‘[t]he fact that both the IMA and the [LRC Bond] were on their apparent wording perpetual further suggests that they were to serve the same purpose’, and notably to assist [Party D] with its purpose ‘to park [the proceeds it obtained in Hong Kong] for which it had no immediate use in Hong Kong, and be in a position to ward off any demand by SAFE [that is, the State Administration of Foreign Exchange] to repatriate the funds to the Mainland.”

23.  This argument raised by Party D, as said above, was only one of the arguments in support of Party D’s position in relation to the first issue, and it was only here at §142 of the Partial Award that Witness Z’s evidence was mentioned.

24.  From §145 onwards, the Tribunal summarised Party D’s submissions in response. Naturally, Party D made a number of submissions attacking Witness Z, and the Tribunal summarised its submissions in the following:-

“149. [Party D] argues that [Party P’s] evidence with respect to the [LRC Bond] and the purpose of the IMA was an ‘endlessly changing narrative’ and that [Party P’s] reliance on the [LRC Bond] was contradicted by objective and contemporaneous evidence. Moreover, [Party D] notes that while [Witness Z’s] witness statements only state that in 2019 [Mr G, in control of Party D] allegedly ‘indicated interest in acquiring investment products issued by the [LRC] Group…[Witness Z] under cross-examination claimed that he suddenly ‘recalled’ for the first time that [Mr G] allegedly needed ‘a custom-made product that could help satisfy the need to meet certain requirements of foreign exchange regulations’. [Party D] argues that ‘[t]hese claims in [Party P’s] own shifting case are clearly and demonstrably false’.

150. [Party D] also submits that [Party P] has never explained why, if [Party D’s] real purpose was to avoid repatriating its [proceeds] to Mainland China, it would have needed to enter into the IMA in addition to acquiring the [LRC Bond]…[Party D] contends that the only explanation is that [Party P’s] narrative based on [Witness Z’s] oral testimony is false. Therefore, submits [Party D], if the [LRC Bond] actually ever existed, it was [Party P] that decided to acquire the [LRC Bond] for [Party D] under the IMA, and [Party D] did not instruct [Party P] to do so.

…

152. Furthermore, [Party D] contends that [Party P’s] argument, based on [Witness Z’s] oral testimony, to the effect that ‘[the LRC Bond] is a perpetual bond’, is contradicted by [Party P’s] own documentary evidence. In this regard, [Party D] submits that ‘the heavily-watermarked one-page [LRC Bond] belatedly submitted as R-120 is also highly questionable and likely not authentic’. In [Party D’s] submission, even [Party P’s] own submitted records…all prove that the [LRC bond], if it ever existed, was stated to mature on 12 December 2023.

153. [Party D] then submits that ‘the evidence clearly shows the true story: [Party P] took advantage of the IMA and transferred [Party D’s] funds to its own affiliates including [LRC], and later [Company G] and [Company S] (and [Company H]) and/or the [LRC Bond] never even existed…”

25.  Then came the section “Tribunal’s Analysis and Decisions”:-

“168…drafting history and negotiations, as distinct from the full factual context in which a contract is made, are excluded from consideration in the exercise of contractual construction. Accordingly, barring a mistake, both Parties are to be held to have agreed to the terms of Clause 1.3 as finally drafted…

169. Turning to the interpretation of Clause 1.3, the principles of contractual interpretation to be applied are well-known and have been set out fully by the Parties in their submissions. The starting point for the Tribunal is to identify the intention of the contracting parties. This is an objective test; the Tribunal must identify the intention of the Parties by reference to what a reasonable person having all the background knowledge that would have been available to the Parties would have understood them to be using the language in the contract to mean. ln ascertaining the objective meaning of Clause 1.3, the Tribunal should first consider the natural and ordinary meaning of the clause. At the same time, and particularly if the drafting of the clause is not clear, the Tribunal may also seek to infer the Parties’ intention from other relevant provisions of the IMA, the overall purpose of Clause 1.3 and the IMA as a whole, and to a certain extent commercial common sense. ln relation to commercial common sense, where there is ambiguity and more than one possible construction, the Tribunal should select the interpretation that makes the most common sense, the presumption being that the Parties would not have intended an uncommercial result.

171. In the Tribunal’s opinion, [Party P]’s interpretation does not fit with the natural and ordinary meaning of the words of Clause 1.3…

172. In this regard, [Dr H, a witness for Party P] submits that the ‘peculiar wording’ of Clause 1.3 was specifically adopted at [Party D’s] request to satisfy [Party D’s] ‘own initial requirements’ and more particularly to provide an effective means for [Party D] to circumvent Mainland China’s capital control policies and to keep its Hong Kong [proceeds] under investment outside of Mainland China rather than having to repatriate those proceeds.

173. [Party P’s] proposed interpretation therefore goes beyond the actual wording of Clause 1.3…

174. The idea that Part A [of the IMA] refers merely to a ‘notional termination’ is in fact at the core of [Party P’s] proposed interpretation of Clause 1.3. As noted above, [Party P’s] interpretation is predicated on an argument that the notional term and expiry date were added into Part A to show to SAFE or other Mainland China regulatory authorities if necessary, and that the real substance and intention of the parties was contained in Part B, in other words that the IMA would continue to last until both parties agreed to terminate it. [Party P] in advancing this argument relies heavily on the witness statement of [Dr H], who states that [Party D] (through [Mr L] and [Mr Z]) had explained this purpose to [Party P], and on the expert report of Mr Wu, who explains the underlying regulatory environment in Mainland China that might have led [Party D] to want to amend Clause 1.3 in the way it has proposed.

176. In the Tribunal’s opinion, whether [Party P’s] proposed interpretation of Clause 1.3 is credible and tenable will depend first on the answer to the following two questions: (1) whether there was a real and serious risk that [Party D] would be obliged to repatriate its [proceeds] back to Mainland China; and (b) whether the finally agreed wording of Clause 1.3 was likely to be effective in reducing or eliminating that risk.

177. With respect to question (1), the Tribunal is not convinced by [Party P’s] argument with respect to the seriousness of the risk of [Party D’s] being obliged to repatriate its [proceeds] back to Mainland China…

178. The Tribunal does not consider that the documents referred to by Mr Wu support a conclusion that there was a strong risk that SAFE or other Mainland Chinese authorities would oblige [Party D] to repatriate all the proceeds…back to Mainland China…In other words, in the Tribunal’s opinion, the SAFE Notice is inconsistent with [Party P’s] position and rather matches Mr Li’s evidence that there was a discretion whether to repatriate the [proceeds].

…

180. Accordingly…the Tribunal is not convinced that [Party D] is or was exposed to a serious risk of having to repatriate its [proceeds] back to Mainland China.”

26.  At §181, the Tribunal then assumed “for the sake of argument” there was such a serious risk of having to repatriate the proceeds back to Mainland China, and went on to consider “whether the final wording of Clause 1.3 would assist in alleviating that risk”.

27.  The Tribunal continued:-

“185. According to [Party P], the need to reconcile these various different requirements in one provision explains the ‘peculiar wording’ of Clause 1.3, with its expiry date that was notional only, coupled with an automatic renewal clause and mutual termination requirement that would avoid giving too clear an impression that the IMA would last indefinitely but would in fact have that legal effect. The Tribunal finds this explanation of [Party P] convoluted and of doubtful credibility. Even if the Mainland authorities were to examine the terms of the IMA to see whether [Party D] had a legitimate reason to maintain some of its [proceeds] overseas, the Tribunal is not convinced that the particular wording of Clause 1.3 would do anything to persuade those authorities of the existence of such a legitimate reason. If anything, the Tribunal considers that the ‘peculiar wording’ of Clause 1.3, with a fixed term and termination of the IMA after the first two years in Part A, followed by an automatic renewal of the IMA immediately thereafter in Part B, would be just as likely to ‘attract [the] regulatory scrutiny’ that [Party P] says the Parties (and more particularly [Party D]) were seeking to avoid.

186. Moreover, if those Mainland authorities were really minded to require [Party D] to repatriate the assets under the IMA upon the expiry of the initial two-year term referred to in Part A of Clause 1.3, the Tribunal does not consider it likely that they would be dissuaded from doing so merely by the provisions of Part B…

187. The Tribunal is therefore not convinced that the final wording of Clause 1.3 would be interpreted by the relevant Mainland authorities (or indeed by any reasonable person having all the background knowledge that would have been available to the Parties) in the way [Party P] contends for…

…

189. In this regard, the Tribunal also notes that [Party P’s] evidence on the above points was sometimes inconsistent. For example, [Dr H]…

190. For all of the above reasons, the Tribunal determines that [Party P’s] interpretation of Clause 1.3 is unconvincing and should not be accepted.”

28.  When rejecting Party P’s interpretation, in the “Tribunal’s Analysis and Decisions”, nothing was mentioned about Witness Z’s evidence.

29.  Having rejected Party P’s interpretation, the Tribunal went on to consider Party D’s interpretations “to see if they are more persuasive than [Party P’s] interpretation” (§191 of the Partial Award):-

“193. As with [Party P’s] proposed interpretation of Clause 1.3, [Party D] is not seriously arguing that its alternative interpretations reflect the natural and ordinary meaning of the words of Clause 1.3. Rather, [Party D] is in effect stating that Clause 1.3 must be interpreted in a way that makes the clause work, taking into account the relevant factual background of the IMA, the overall purpose of the IMA and commercial common sense…

196. The Tribunal has already determined above that it does not agree with [Party P’s] interpretation that the reference in Part A of Clause 1.3 to termination of the IMA after two years was ‘notional’ only. In other words, the Tribunal considers that the Parties intended that the IMA (or at least its initial term) would expire two years after the Effective Date, or alternatively that either Party would have the right to terminate the IMA at that date. In this regard, the Tribunal agrees with [Party D] that Part A cannot simply be disregarded, which in effect is what [Party P’s] interpretation seeks to do.

197. The Tribunal also considers that an IMA with a perpetual term would not make commercial sense…The Tribunal accepts in this regard [Party D’s] explanation that its purpose in entering [into] the IMA was simply to engage a reputable investment manager to manage part of the proceeds…There seems to the Tribunal no plausible commercial explanation why [Party D] would have wanted an IMA with an indefinite term…Rather, it seems far more likely that [Party D] wanted to have an IMA with a fixed initial period…

204. The Tribunal is conscious of the need to avoid rewriting a contractual term simply because it later might appear to be imprudent for a party, in this case [Party D]. At the same time, the Tribunal considers that where, as here, there are two rival constructions of a contractual provision, the construction that is consistent with business common sense should be preferred and the other rejected. This is because a commercial construction is likely to give effect to the intention of the Parties. ln the Tribunal’s opinion, [Party D’s] construction is more consistent with business common sense and therefore is more likely to give effect to the intention of the Parties when entering into the IMA. The Tribunal considers that this approach is all the more pertinent where, as here, neither Party’s construction of the provision in question fits perfectly with its literal wording.

205. With respect to the [LRC Bond], it seems to the Tribunal that it could be potentially relevant to Issue 1 and the purpose of the IMA, in particular since the bond was contemporaneous with the IMA. Nevertheless, the exact terms of the bond, who had control of the bond, and whether it even actually existed, are heavily contested by the Parties. It is therefore difficult to come to any firm conclusions about the nature and relevance of the [LRC Bond] to Issue 1.

207. As for Exhibit R-120, this is a one-page document produced by [Party P] just after the Hearing. It provides very limited information, except to say that the bond is perpetual and has a coupon rate fixed at 0.25% per annum. [Party P] therefore states that the [LRC Bond], as a perpetual bond with no maturity date, corresponds with and supports [Party P’s] interpretation of Clause 1.3, because the bond, like the IMA itself ‘would continue to be effective unless both parties agreed to terminate’. Specifically, [Party P] submits that the only sensible explanation was that the [LRC Bond], like the IMA itself, both with an apparently indefinite duration, were designed to allow [Party D] to ‘park’ its [proceeds] for which it had no immediate use in Hong Kong.

208. The documents for the partial redemption of the [LRC Bond] in the amount of US$1.34 million do tend to suggest that the [LRC Bond] at least existed. [Witness Z’s] evidence was that the bond certificate had been provided directly to [Party D], and that the bond was still outstanding, although [Party P] could not claim on it as it was perpetual. Given [Witness Z’s] apparent familiarity with the terms of the [LRC Bond], the Tribunal considers it puzzling that neither he nor [Party P] was able to provide more details of the [LRC Bond], for example a sale and purchase agreement or any further evidence of the bond’s detailed terms.

209. If the terms of the [LRC Bond] were as indicated by Exhibit R-120, in other words perpetual with a coupon of 0.25% per annum, the question also arises as to why [Party D] would have agreed to invest in an instrument on such unfavourable terms, especially as [Party D] was obliged under the IMA to pay a management fee of 1.5% of the Managed Assets. As [Party D] has submitted, if [Party D’s] purpose was really to invest its [proceeds] in Hong Kong, it could just as well have achieved that by investing directly in investments such as the [LRC Bond] without needing to enter into the IMA and to incur significant management fees.

210. The matter is complicated by the existence of other documents suggesting that the [LRC Bond] may have had different terms than those shown on Exhibit R-120. For example, Exhibit R-3 suggests that it was a 6% bond due on 12 December 2023. Other documents such as Exhibits R-45 and R-69 are suggestive of different maturity dates for the [LRC Bond]. If the [LRC Bond] in fact had a coupon of 6%, then the investment seems more reasonable and may explain why [Party D] signed the investment instructions set out in Exhibit R-44.

211. Given all the above uncertainties surrounding the [LRC Bond], the Tribunal does not consider that it can be relied upon by [Party P] to support its case that the parties intended the IMA to be automatically renewed after two years and to continue until both Parties agreed to terminate the IMA.

212. For all the above reasons, the Tribunal considers that [Party P’s] proposed interpretation of Clause 1.3 should be rejected and that one or more of [Party D’s] proposals for resolving the potential difficulties of the wording of Clause 1.3 should be adopted. Accordingly, the Tribunal agrees that Part B (the Renewal Clause) should be read subject to Part A (the Automatic Expiry Clause), so that [Party D’s] decision to terminate the IMA would have primacy over any proposal or right to renew the IMA.” (emphasis added)

30.  A fair reading of the Tribunal’s detailed analysis above shows that in coming to the view at §205 of the Partial Award that no firm conclusion could be made upon the LRC Bond, the Tribunal considered various factors like the inherent improbability of the LRC Bond. Witness Z’s evidence was only one of the various factors. Further, in my view, the Tribunal’s comment – “puzzling” – on Witness Z in the emphasised part above was a fair comment any reasonable tribunal would make because Witness Z was supposed to be familiar with the terms of the LRC Bond but could not provide any more details of the LRC Bond. This in itself is sufficient to ground any opinion of “puzzling”.

31.  Having come to the view that the Party D may terminate the IMA, the next live issue was Issue 3, namely, “is [Party D] entitled to the return of the Managed Assets? (b) what is the proper date of calculation of their value? (c) what is the amount for which [Party P] is liable to [Party D]?”  The following under “Parties’ Contentions” of the Partial Award in relation to Issue 3 touched Witness Z:-

“241. The essence of [Party P’s] case in relation to Issue 3 is that (apart from [the Fund Note]) [Party P] in actuality played a very limited ‘bookkeeper’ role under the IMA…Therefore, no matter what the Tribunal’s determination is of the Managed Assets currently still in the portfolio, [Party P] argues that it is not open to [Party D] to claim the return of a sum of money, since such money was never in [Party P’s] hands.

242…The Tribunal will below set out a brief summary of the relevant chronology as gleaned from the Parties’ submissions, as far as it is relevant to the Tribunal’s determination of Issue 3.

243. On signing of the IMA…US$117 million of the [proceeds] were placed by [Party D] into the account ‘01C’ with [Party P]. The money was then transferred to account ‘02C’, from which the whole amount was used to acquire the [LRC Bond]. [Party P] relies on the one-page Investment Instructions, also dated 25 September 2019, signed by [Mr Z] on behalf of [Party D] and addressed to [Party P], as indicating that from the beginning it was [Party D] that was driving the choice of investments. In addition, [Party P’s] case, based on the evidence of [Witness Z], founder of the [LRC] Group, is that the certificate for the [LRC Bond] was issued directly to [Party D].

246. [Party P] says that its quarterly statement does not show what happened to the [LRC Bond], in that it only shows that it had been withdrawn according to [Party P’s] books, but does not show whether it had been redeemed. [Witness Z’] evidence was that the [LRC Bond] has not been redeemed, and remains outstanding to this day. [Party P’s] position appears to be that it is unclear precisely what happened in respect of the [LRC Bond] and the three new notes: whilst [Witness Z] said that the [LRC Bond] had not been redeemed or swapped for the three new bonds, in terms of the portfolio operated as between [Party D] and [Party P], that had occurred.

254. On 27 June 2023, there was a withdrawal of the remining [Note H] (US$9.7 million) and [Note S] (US$39 million) from the portfolio, and a deposit of the [Fund Note] (US$48.7 million) into the portfolio. [Party P’s] case is that as a result of its various concerns over the investments in [Note H] and [Note S], [Mr H of Party P] contacted [Witness Z] (who was an active client of [Party P] and was pursuing a substantial debt claim against [Company S]) to see if he would consider taking over the debt under [Note S] and [Note H]. [Party P] says that [Witness Z] was receptive to the proposal and confirmed his agreement…

255. [Witness Z] then caused [the Fund] to issue the [Fund Note] in favour of [Party D], to replace [Note S] and [Note H], with the certificate of the [Fund Note] being held in [Party P’s] custody. [Party P] contends that no money passed through [Party P] for the purchase of the [Fund Note] and that there is therefore ‘no question of accounting for the same’.

256. As a result of the above fund movements and investment activities, [Party P] argues that the only asset now remaining in the portfolio is the [Fund Note]…

264. With respect to the question of who directed the investments and dealt with the issuers, [Party D] submits that [Party P’s] argument that, apart from the [Fund Note], [Party P] was not involved in the IMA investments, does not withstand scrutiny. [Party D] notes that the bond issuers related to the IMA investments…were all inextricably related to [Party P]. For example, [LRC] was part of the group that used to own [Party P], and [Company S] was owned by [Party P’s principal’s mother], and may also have been directly related to [Witness Z]…

32.  Then came to the “Tribunal’s Analysis and Decisions” on Issue 3:-

“273. [Party P] insists that, before the [Fund Note], it neither participated in nor knew about the purchases of the bond and promissory notes in the [Party D’s] IMA portfolio and that it was [Party D] who had direct dealings with the relevant issuers…

274. [Party D] rejects the argument that [Party P] was only a ‘bookkeeper’, saying that this argument was raised only in the Rejoinder and was inconsistent with the factual matrix and documentary evidence, which [Party D] says clearly showed that [Party P] was in the driving seat with respect to the management of the assets in the IMA portfolio.

275. From the evidence, it is not absolutely clear to the Tribunal who had actual control of the Managed Assets at all relevant times, and what certificates and agreements existed to evidence those assets. Certain documents seem to suggest that [Party D] may have had some measure of control over some of the investment decisions and/or some direct contact with the issuers, for example Exhibit R-44 in relation to the [LRC Bond], or the communications between [Party D] and [Company G] in relation to redemption of [Note G]. Nevertheless, the Tribunal considers that the preponderance of the evidence does not support [Party D’s] assertion that it was merely a bookkeeper making occasional ‘notional’ recordings in the accounts and with no responsibility to account to [Party D] for the Managed Assets on termination or expiry of the IMA. In that regard, the Tribunal agrees with [Party D’s] argument that its repeated and unanswered requests for information from [Party P] regarding the Managed Assets is not consistent with [Party D’s] being in overall charge of investing in and managing the assets in [Party D’s] IMA portfolio.”

33.  Then, at §§276-277, the Tribunal referred to certain evidence “by way of example to illustrate the point”. The evidence was not from Witness Z’s evidence.

34.  In relation to Party P’s contention that Party D would only be entitled to the return of the Fund Note which replaced the original investments in the IMA portfolio, at §283, the Tribunal rejected such contention. At §284, the Tribunal concluded that Party P should be ordered to return the Managed Assets or their liquidated equivalent to Party D, and given that uncertainties with respect to the existence and the validity of the certificates and other documents evidencing the Managed Assets, the Tribunal ordered Party P to pay a liquidated sum representing the value of the Managed Assets. At §288, the Tribunal left the assessment of this sum to Phase 2.

35.  Then at §§289-291, the Tribunal said:-

“289. [Party D] has also claimed for additional damages to be assessed for [Party P’s] failure to return [Party D’s] Managed Assets, in other words damages for losses flowing from being kept out of its Managed Assets or money equivalent since 25 September 2021. In its Reply, [Party D] refers to those damages as ‘including but not limited to those for lost opportunity costs, interference with ability to pursue alternative investment opportunities and reputational damage’.

290. The Tribunal notes that in the [Bifurcation Agreement] set out in [Party D’s] letter of 7 March 2025, the question of [Party P’s] liability for any of the above additional damages was intended to be determined in this first phase, with the quantum of any such damages to be determined in the second phase. The Tribunal is of the preliminary view that the various categories of additional losses identified above could potentially flow from [Party P’s] breach in not returning the Managed Assets on 25 September 2021. Nevertheless, the Tribunal does not consider that it has been sufficiently addressed on the question of liability for any of those categories of additional losses for it to make any determination on liability at this stage. For example, the Tribunal would need to be persuaded that losses for reputational damage are recoverable as a matter of law before it could proceed, if necessary, to consider quantum. Similarly, with respect to damages for lost opportunity costs, the Tribunal would need to be addressed on issues such as whether those losses could be dealt with by an award of pre-award interest rather than by an assessment of damages for [Party D’s] being kept out of money.

291. For the above reasons, the Tribunal will defer determination of the additional damages claimed by [Party D], both in terms of liability and quantum, until the second phase of this arbitration.”

36.  I pause here to note that the Tribunal’s such sentiment was understandable because from the Bifurcation Agreement and the parties’ submissions, in my view, the parties assumed that Party P would be liable to pay these additional losses, and/or the same would be recoverable, if there would be breach of the IMA by failing to return the Managed Assets:-

(1)  Under the Bifurcation Agreement, for Phase 1, the issue in relation to the return of the Managed Assets was whether there was any liability to return, and for Phase 2, the issue was the quantum of any additional losses arising from the failure to return the Managed Assets. Nothing was mentioned in Phase 1 (let alone Phase 2) about Party P’s liability to pay, and/or the recoverability of, these losses in the first place.

(2)  Consistently, in the Agreed List of Issues for Phase 1, neither party put it as an issue about the liability and/or recoverability of the additional losses. Instead, the parties’ focus was on when the IMA could be terminated.

(3)  Similarly, in the Parties’ 1^st Post-Hearing Written Submissions (that is, closing submissions), both parties, including Party P, assumed without argument that those heads would be recoverable.

(a)  In §83 of Party D’s submissions:-

“a. [Party D] asks that the Tribunal determine that the IMA was terminated or had expired based on any of the following four submissions…So long as the Tribunal accepts any one of these four submissions, [Party D’s] quantified claim for US$66,360,666.11 [that is, the Managed Assets] to be returned, repaid, accounted for, and/or restored (and an account of all necessary inquires), as well as [Party D’s] claim for additional damages to be assessed for failing to return funds and assets, should be granted.

c. [Party D] further submits that [Party P] is liable for breach of trust and duty…[Party D’s] quantified claim for US$66,360,666.11 to be restored, reconstituted, and/or compensated, [Party D’s] claim for accrued interest to be assessed…should be granted.”

(b)  In §68 of Party P’s submissions, Party P’s counsel dealt with Party D’s claim for the return of the Managed Assets. Only at §68.5 (a 4-lined paragraph) did Party P’s counsel deal with the additional losses, in the following way:-

“[Party D] also has a claim for compensation arising from [Party P’s] refusal to return the assets when it was demanded to do so from September 2021, which is to be assessed. It should be noted that this should be assessed by reference to the notes being returned to [Party D], not by reference to [Party D] having a sum of money and it could make alternative investments.”

37.  Therefore, it is fair to say that the parties, assuming that Party P would be liable to pay the additional losses and/or that such additional losses would be recoverable, did not put it as an issue for determination, whether in Phase 1 or Phase 2, on the liability or recoverability of the additional losses, and going along with the parties’ agreement, the Tribunal would simply have to assume that the additional losses were recoverable and Party P was liable to pay the additional losses, and proceed direct to Phase 2 for assessing the quantum of these additional losses. However, the liability to pay these additional losses, or the recoverability of such losses, would be a matter of law, and it is well established that parties’ agreement to law does not bind the tribunal (and the court). Thus, I can see why the Tribunal would of its own motion raise the issue of liability for these additional damages for submissions (and I would not say these would be “further” submissions given that no submissions on this issue had been made in the first place), despite the Bifurcation Agreement by which the Tribunal may simply proceed with the assessment of these additional losses on the assumption or the parties’ agreement that the additional losses were recoverable.

38.  The next relevant parts of the Partial Award are on Issue 4, namely, “Is [Party P] liable for breach of its duty to exercise reasonable care and skill in managing [Party D’s] IMA portfolio?”. Under “Parties’ Contentions”:-

“296. With respect to the [Fund Note], [Party D] argues that [Party P’s] submissions, and [Witness Z’s] evidence that the [Fund Note] was a ‘good deal’, do not stand up to scrutiny. In this regard, [Party D] submits that the [Fund Note] document produced by [Party P] shared similar problems with the [LRC Bond] document, including that it consisted of only one page, even though, as conceded by [Witness Z] in response to questions from the Tribunal, ‘[u]sually there would be standard terms and conditions attached at the back, but those do not seem to be signed’.

300. As already mentioned above, [Party P] says it was concerned about the performance of [Note H] and [Note S], and also about the reported financial difficulties of [Company S], and therefore agreed with [Witness Z] that he would take over those notes and arrange for the issuance of the [Fund Note] in the same investment value as the sum of the other two notes…”

39.  In the “Tribunal’s Analysis and Decisions”, the Tribunal found some in favour of Party P and some in favour of Party D:-

“314. For [Party D] to succeed in this claim, therefore, the Tribunal would have expected to see more detailed particulars and expert evidence to support [Party D’s] case that the investments made by [Party P], such as the [LRC Bond] and the subsequent Three Notes, were in fact negligent or unsuitable. Another potential obstacle to a finding of negligence on [Party P’s] part would be that, at least as far as the initial investment in the [LRC Bond] is concerned, [Party D] appears to have consented to the investment, as indicated for example by the investment instructions in Exhibit R-44.

315. With respect to the specific example of the [Fund Note], however, the Tribunal considers that the balance of the evidence tends to support [Party D’s] argument that this investment was carried out without due regard to the interests of [Party D]…

316. Given [Party P’s] own position that this was its discretionary investment decision involving US$48.7 million of assets, the Tribunal finds it puzzling that [Party P] was unable to provide more details of the alleged swap transaction. From the testimony of [Mr H], who was the Managing Director of [Party P], the basis for the decision to invest in the [Fund Note] seems to the Tribunal rather thin. There was, for example, no indication of any assessment of the standing of the issuer [the Fund] beyond the fact that its Chairman [Witness Z] was considered to be ‘a reputable entrepreneur and investment manager’.

317. Furthermore, the only details provided in relation to the [Fund Note] are on Exhibit C-48. The coupon is 0.25% per annum, much lower than the management fees under the IMA. On the face of it, the [Fund Note] appears to be an investment that is almost guaranteed to lose money, and if it is really ‘perpetual’ [Party D] would be effective stuck with the investment forever at a very low rate of return. The Tribunal also notes that, as already mentioned above, the existence of the [Fund Note] investment was only disclosed to [Party D] one year after it was made, in the [Statement of Defence].

318. Given the above circumstances, the Tribunal has some sympathy with [Party D’s] submission that the [Fund Note] was a dissipation of [Party D’s] funds at a time when disputes had already arisen between the Parties.

319. Having said that, given the Tribunal’s decision in favour of [Party D’s] claim with respect to termination and the recovery of the value of the Managed Assets at the time of termination, which should male [Party D] whole, the Tribunal does not find it necessary or appropriate to come to any formal finding against [Party P] in respect of the claims of breach of trust or breach of duty of reasonable care and skill in relation to the [Fund Note] investment, particularly as that investment was made after the termination or expiry of the IMA.” (emphasis added)

V.  DISCOVERY OF MEMBER R BEING AN ARBITRATOR IN A PRIOR ARBITRATION WHERE WITNESS Z’S EVIDENCE WAS NOT ACCEPTED

40.  On 7 January 2026, Mr H, Party P’s managing director, attended business meetings with his business partners including Witness Z. At one of the private meetings between them, Mr H discussed the Partial Award with Witness Z. At that meeting, Witness Z told Mr H that Witness Z had also been dealing with post-award enforcement issues arising from the Prior Arbitration commenced back in May 2023. Witness Z further told Mr H that:-

(1)  The final award (the “Prior Award”) in the Prior Arbitration was issued in September 2024 by a tribunal consisting of, among others, Member R.

(2)  The Prior Award contained findings adverse to the respondents therein including two LRC Group entities.

(3)  Witness Z testified in the Prior Arbitration as the sole witness for the respondents therein, and he was cross-examined during the hearing in the Prior Arbitration in May 2024 before the tribunal chaired by Member R.

(4)  The dispute in the Prior Arbitration centred around an investment product issued by entities of the LRC Group.

(5)  Witness Z explained that he did not raise these around the time of the Subject Arbitration Oral Hearing because he genuinely had not recognised Member R’s face or name at the oral hearing. This was especially so when (a) the hearing in the Prior Arbitration took place in May 2024, almost 1 year earlier; (b) Witness Z had not personally reviewed the Prior Award but had relied upon his lawyers; (c) he only participated in the Subject Arbitration as a factual witness and not as a party and so he did not pay close attention to the identity of the Tribunal members; (d) he testified in the Subject Arbitration via videoconference from a laptop and it was hard to see anyone’s faces; and (e) he relied on, and therefore focused on, the interpreter to understand questions and testify at the Subject Arbitration Oral Hearing.

41.  It is not in dispute that in the Subject Arbitration, Member R did not disclose anything about the Prior Arbitration.

VI.  CHALLENGES ARISING FROM MEMBER R’S BEING AN ABITRATOR IN THE PRIOR ARBITRATION

42.  Party P thus issued the present Originating Summons to set aside the Partial Award. In parallel, Party P filed a Notice of Challenge with the HKIAC on 22 January 2026 seeking removal of Member R pursuant to articles 11.6 – 11.8 of the HKIAC Rules.

43.  In response, on 24 February 2026, Member R filed his Answer to the Notice of Challenge (“Member R’s Answer”).

44.  The HKIAC convened a three-member panel (the “Panel”) to consider Party P’s challenge. By a written recommendation dated 14 May 2026 (the “Panel’s Recommendation”), the Panel recommended the rejection of the challenge. On 22 May 2026, the Proceedings Committee of the HKIAC accepted the Panel’s Recommendation and dismissed Party P’s challenge. I pause to note that the Originating Summons herein is not about any review of the Panel’s Recommendation or the HKIAC’s decision to adopt the Panel’s Recommendation, and that I am in no way bound by the Panel’s Recommendation. The determination of the issues underlying the Originating Summons herein is entirely a matter for me to determine de novo and on my own.

VII.  PARTY P’S GROUNDS FOR SETTING ASIDE THE PARTIAL AWARD

45.  In the Originating Summons, Party P relies on article 34(2)(a)(ii), article 34(2)(a)(iv) and article 34(2)(b)(ii) of the UNCITRAL Model Law (given effect to by section 81 of the Arbitration Ordinance (Cap. 609)). In the submissions made by Mr Charles Manzoni SC, leading Mr Keith Lam and Mr William Wong, counsel for Party P, the grounds are refined as follows:-

(1)  The entirety of the Partial Award was tainted by apparent bias on the part of the presiding arbitrator being Member R; and

(2)  Further or alternatively, §§289-291 and 391.5 of the Partial Award should be set aside as these holdings gave Party D a second chance at establishing its case on Party P’s alleged liability for the claim for the additional damages by deferring it to Phase 2, despite (1) the Bifurcation Agreement; and (2) the Tribunal having formed the view that Party D had failed to substantiate its case on liability for this claim at the Subject Arbitration Oral Hearing.

46.  I shall deal with Ground (1) (apparent bias) first.

VIII.  APPARENT BIAS

A.  Legal principles

47.  The test for apparent bias applicable to a challenge to an arbitrator is whether “an objective fair-minded and informed observer, having considered the relevant facts, would conclude that there was a real possibility that the tribunal was biased”: see CNG v G (No 2) [2025] 4 HKLRD 781 at §31 per Mimmie Chan J. It has been described as a two-stage test in that the first stage is to ascertain all the relevant circumstances, and the second stage is to ask whether the fair-minded and informed observer would conclude that there was a real possibility that the judge was biased: see §40 quoting §17 of Bubbles & Wine Limited v Reshat Lusha [2018] EWCA Civ 468. In CNG v G (No 2), supra, Mimmie Chan J went on to elaborate on this test:-

(1)  A fair-minded observer “does not reach a judgment on any point before acquiring a full understanding of both sides of the argument” (§35);

(2)  A fair-minded and informed observer “will appreciate that the context forms an important part of the material which she must consider before passing judgment” (§35);

(3)  “The conclusions which the observer reaches must be justified objectively” (§35);

(4)  “The ‘real possibility’ test ensures the exercise of a detached judgment” (§35), and the fair-minded observer is not to be confused with the person making the complaint of apparent bias (§41);

(5)  The fair-minded and informed observer is “not a lawyer”, although she is not “wholly uninformed and uninstructed about the law in general or the issue to be decided” (§36);

(6)  The fair-minded and informed observer is “neither complacent nor unduly sensitive or suspicious” (§36);

(7)  “The fictitious bystander will also be aware of the strong professional pressures on adjudicators (reinforced by the facilities of appeal and review) to uphold traditions of integrity and impartiality” (§36);

(8)  “[T]he objective observer is alive to the possibility of opportunistic or tactical challenges…and this pursuit [to win] can include making tactical objections or challenges in the hope of having their dispute determined by a tribunal which might, without any question of bias, be more predisposed towards their view or simply to delay an arbitral determination” (§38);

(9)  Therefore, “a court, when asked to remove an arbitrator, needs to be astute to see whether the ground of real possibility of bias is made out” (§38); and

(10)  “[I]n applying the test to arbitrators, the differences in nature and circumstances between judicial determination of disputes and arbitral determination may have to be considered.” 

48.  Mr Ambrose Ho SC, leading Mr Benny Lo and Mr Jiang Zixin, counsel for Party D, emphasises that not any possibility of bias is sufficient, and that the possibility must be “real”. I see no qualm from Mr Manzoni. This must be right – a fair-minded and objective person living in the real world, the possibility must be real.

49.  Further, in determining whether there is “a real possibility that the tribunal was biased”, “[t]here must be articulation of a logical connection between the conduct giving rise to the apprehension of bias and the possibility of departure from impartial decision-making”: see Re Li Xiaoming [2021] HKCA 779 at §19 per Kwan VP.

50.  For §47(10) above concerning the differences between judicial officers and arbitrators, one feature is “the professional reputation and experience of an individual arbitrator is a relevant consideration for the objective observer when assessing whether there is apparent bias, as an established reputation for integrity and wide experience in arbitration may make any doubts harder to justify” (§39).

51.  More features were noted in Halliburton Co v Chubb Bermuda Insurance Ltd [2021] AC 1083, consistently endorsed in Hong Kong. In that case, Lord Hodge DPSC pointed out, among others, the following to which Mr Manzoni specifically draws my attention to and with which I agree as a matter of common sense:-

“56. First, judges resolve civil disputes in courts which are, as a general rule, open to the public; by contrast arbitration is a consensual form of dispute resolution which is generally conducted in private and of which there is very limited public oversight. A person who is not a party to an arbitration may know nothing about the arbitration and may have no ready means of discovering its existence, the evidence adduced and the legal arguments advanced at it, or the award made. Arbitrators and the parties to an arbitration are generally under a duty of privacy and confidentiality which militates against such discovery, in the absence of disclosure. That puts a premium on frank disclosure.

…

58. Secondly, unlike a judge who decides issues of fact and law at first instance and from whose decisions the parties usually have a right of appeal, an arbitrator is not subject to appeals on issues of fact and often not on issues of law. By contrast with a first instance judge, there are very limited powers of review of the decision of an arbitral tribunal.

59. Thirdly, a judge is the holder of a public office, is funded by general taxation and has a high degree of security of tenure of office and therefore of remuneration. An arbitrator is nominated to act by one or both of the parties to the arbitration either directly or by submitting names to the appointing body, whether an institution or the court, for appointment. The arbitrator is remunerated by the parties to the arbitration in accordance with the terms set out in the reference, and often is ultimately funded by the losing party. He or she is appointed only for the particular reference and, if arbitral work is a significant part of the arbitrator’s professional practice, he or she has a financial interest in obtaining further appointments as arbitrator. Nomination as an arbitrator gives the arbitrator a financial benefit. There are many practitioners whose livelihood depends to a significant degree on acting as arbitrators. This may give an arbitrator an interest in avoiding action which would alienate the parties to an arbitration, for example by assertive case management against the wishes of the legal teams who are presenting their clients’ cases. It also may give those legal teams an incentive to be more assertive of their side’s interests in the conduct of the arbitration than might be the case in a commercial court.

…

61. Fifthly, it follows from the private nature of most arbitrations that where there are multiple references concerning the same or overlapping subject matter in which the same arbitrator is a member of the tribunal, the party which is not common to the various arbitrations has no means of informing itself of the evidence led before and legal submissions made to the tribunal (including the common arbitrator) or of that arbitrator’s response to that evidence and those submissions in the arbitrations in which it is not a party. It is not unusual in commercial litigation for an interested party to instruct its lawyer to sit in on a court case involving other parties which may have a bearing on its interests in a separate action. Such an expedient is generally not available in arbitration.” (emphasis added)

52.  Among the relevant circumstances a fair-minded and informed observer would consider is the adjudicator’s duty to disclose and how it has been discharged. Section 25 of the Arbitration Ordinance (giving effect to Article 12 of the UNCITRAL Model Law) provides:-

“(1) When a person is approached in connection with his possible appointment as an arbitrator, he shall disclose any circumstances likely to give rise to justifiable doubts as to his impartiality or independence. An arbitrator, from the time of his appointment and throughout the arbitral proceedings, shall without delay disclose any such circumstances to the parties unless they have already been informed of them by him.”

53.  As DHCJ Lisa Wong SC (as she then was) held in Jung Science Information Technology Co Ltd v ZTE Corp [2008] 4 HKLRD 776:-

“57…A failure to disclose, of itself, can be one of the circumstances which together with others may give rise to a reasonable apprehension of bias as a party or the public may well be left with the impression that there was intentional concealment or non-disclosure, or that something was ‘wrong about it all’. A failure to disclose, no matter how unwitting, can undermine public confidence in the integrity of, and the administration of justice by, the judicial officer or the tribunal concerned.

58. The facts to be disclosed are not confined to those warranting or perceiving to be warranting disqualification but those that might found or warrant a bona fide application for disqualification…” (emphasis added)

54.  In the context of arbitration, as already pointed out above, in Halliburton Co v Chubb Bermuda Insurance Ltd, supra, at §56, Lord Hodge pointed out that there is a “premium on frank disclosure” by arbitrators. His Lordship further observed that:-

“78. Unless there is disclosure, the parties may often be unaware of matters which could give rise to justifiable doubts about an arbitrator’s impartiality and entitle them to a remedy from the court…Those remedies are necessary in the public interest. A legal obligation to disclose such matters is encompassed within the statutory obligation of fairness. It is also an essential corollary of the statutory obligation of impartiality: an arbitrator who knowingly fails to act in a way which fairness requires to the potential detriment of a party is guilty of partiality…such disclosure is not a just a question of best practice but is a matter of legal obligation.

…

88. Where the information which must be disclosed is subject to an arbitrator’s duty of privacy and confidentiality, disclosure can be made only if the parties to whom the obligations are owed give their consent. In such a circumstance, if a person seeking appointment as an arbitrator in a later arbitration does not obtain the consent of the parties to a prior related arbitration to make a necessary disclosure about it, or the parties to the later arbitration do not consent to the arbitrator’s disclosure of confidential matters relating to that prospective appointment to the parties to the earlier arbitration, the arbitrator will have to decline the second appointment…” (emphasis added)

55.  That said, a failure to disclose or breach of duty to disclose does not automatically lead to a conclusion of a real possibility of bias. Such a failure or breach is, as said above, a factor for a fair-minded and informed observer to consider.

56.  On this note about declining the appointment, in my view, it does not follow that an arbitral award would then automatically be set aside. As explained in §65 below, setting aside an award is different from declining an appointment or removing an arbitrator before an award is made.

57.  In relation to witnesses, Lord Bingham’s dicta in Locabail (UK) Ltd v Bayfield Properties Ltd [2000] QB 451 at §25 has been endorsed by Lam VP, Kwan and Barma JJA in TCWF v LKKS (No 1) [2014] 1 HKLRD 896 at §53:-

“The mere fact that a judge, earlier in the same case or in a previous case, had commented adversely on a party or witness, or found the evidence of a party or witness to be unreliable, would not without more found a sustainable objection…The greater the passage of time between the event relied on as showing a danger of bias and the case in which the objection is raised, the weaker (other things being equal) the objection will be.”

58.  In Locabail (UK) Ltd v Bayfield Properties Ltd, supra, at §25, Lord Bingham gave a few examples which may be regarded as more than this “mere fact” and which may give rise to a real possibility of bias:-

(1)  “[T]here were personal friendship or animosity between the judge and any member of the public involved in the case”;

(2)  “[T]he judge [was] closely acquainted with any member of the public involved in the case, particularly if the credibility of that individual could be significant in the decision of the case”;

(3)  “[I]n a case where the credibility of any individual [was] an issue to be decided by the judge, he had in a previous case rejected the evidence of that person in such outspoken terms as to throw doubt on his ability to approach such person’s evidence with an open mind on any later occasion”; and

(4)  The judge had expressed views in such extreme and unbalanced terms on certain issues as to throw doubt on his ability to try the issues in the proceedings before him.

59.  Similarly, in HKSAR v Md Emran Hossain (2016) 19 HKCFAR 679, Fok PJ (delivering the unanimous judgment of the Court of Final Appeal) had the following to say:-

“43. That the participation of the same judge in a subsequent hearing involving the same parties and subject matter does not necessarily give rise to an appearance of bias is illustrated by the case of Sengupta v Holmes [2002] EWCA Civ 1104…Laws LJ considered what would be known to the fair-minded and informed observer, which included the following features. He regarded it as material that, in the ordinary case, a single judge having initially refused leave to appeal will have done so ‘in the knowledge that, at the option of the applicant, his view may be reconsidered at an oral hearing’ and that there was, in the circumstances, no reasonable basis for supposing he might not bring an open mind to bear on the substantive appeal if he subsequently sat on the Full court. He regarded a readiness to change one’s mind on an issue to be ‘a commonplace of all the professions, indeed of the experience of all thinking men and women’ and the fact that judges do change their minds under the influence of oral argument to be a central feature of the adversarial system. Another important feature of the English legal culture which Laws LJ considered important was the expectation that a judge would be positively expected to alter an earlier expressed view if objectively persuaded by his colleagues that it should be altered.

44. In my view, all these features of the English common law system described by Laws LJ apply equally to Hong Kong and inform the view of the fair-minded observer in this jurisdiction…

45. However, that is not to say that there might never be circumstances in which a single judge who has refused leave to appeal would be in a position in which the fair-minded and informed observer would consider there to be a real possibility that the judge might be biased on a subsequent hearing of the underlying merits of the case. In Sengupta v Holmes, Laws LJ gave as examples: (i) where a judge of first instance has committed himself to a view of the facts which he had the responsibility to decide…(ii) where he had decided that a party or witness was a crook or a rogue…and (iii) where he had expressed himself in such vituperative language that any reasonable person would think him unable to take a fair view of the case if called on to revisit it…this list is not closed…” (emphasis added)

60.  The same, in my view, applies equally to arbitrators, subject to the different features pointed above.

61.  In relation to disclosure about witnesses, Mr Ho draws my attention to §12.05[K] of Gary B. Born, International Commercial Arbitration (3^rd edition, 2021):-

“the fact that an arbitrator has previously considered evidence (as an arbitrator in an earlier, unrelated arbitration) from a fact or expert witness should not be grounds for disqualification or a subject of disclosure. Similarly, the fact that an arbitrator or his or her law firm has previously retained an expert witness, or is presently retaining that witness, in an unrelated matter should not be grounds for disqualification.”

62.  Mr Ho also highlights that the International Bar Association Guidelines on Conflicts of Interest in International Arbitration (2024) (the “IBA Guidelines”) does not mention that disclosure would be required for such fact. However, with respect, I do not regard the view expressed in the learned textbook and/or absence of any view on the matter as an absolute rule on what to disclose and what not. This can be taken as a starting point, but it depends on the circumstances. For example, I would tend to think that disclosure would be required if the adjudicator in the previous case made extreme, adverse comment on the witness’s integrity, and if he is prevented by confidentiality to make such disclosure, he may have to consider the option of not accepting the reference. That said, a fair-minded and informed observer would also consider the absence of mention of disclosure for such fact in the IBA Guidelines.

63.  Sometimes, the adjudicator himself may give some statements about the circumstances that were said to constitute a real possibility of bias. The approach to such statements has been set out by Lord Bingham in Locabail (UK) Ltd v Bayfield Properties Ltd, supra, at §19, with which I agree:-

“… the court is not necessarily bound to accept such statement at its face value. Much will depend on the nature of the fact of which ignorance is asserted, the source of the statement, the effect of any corroborative or contradictory statement, the inherent probabilities and all the circumstances of the case in question. Often the court will have no hesitation in accepting the reliability of such a statement; occasionally, if rarely, it may doubt the reliability of the statement; sometimes, although inclined to accept the statement, it may recognise the possibility of doubt and the likelihood of public scepticism. All will turn on the facts of the particular case. There can, however, be no question of cross-examining or seeking disclosure from the judge. Nor will the reviewing court pay attention to any statement by the judge concerning the impact of any knowledge on his mind or his decision: the insidious nature of bias makes such a statement of little value, and it is for the reviewing court and not the judge whose impartiality is challenged to assess the risk that some illegitimate extraneous consideration may have influenced the decision.”

64.  Given that the test is fact-sensitive, it is not necessary for me to deal with each and every case (including Sphere Drake Insurance v American Reliable Insurance Co [2004] ArbLR 54, Livesey v The New South Wales Bar Association (1983) 151 CLR 288 and Esquire (Electronics) Limited v The Hongkong and Shanghai Banking Corporation Limited and Another, HCA 11077/1994, 14 December 2005) cited by the parties as illustrations of how the test was applied. Suffice to say that the facts in those cases are not materially the same as the present, and it is the application of the correct legal principles to the facts of the present case that matters, but not the outcomes of the previous cases that matter.

65.  Lastly, for setting aside an arbitral award, proof of apparent bias is in itself insufficient. The complaint has the burden of proof that the outcome would have been different. Thus, in Grand Pacific Holdings Ltd v Pacific China Holdings Ltd (in liq) (No 1) [2012] 4 HKLRD 1, an application to set aside an arbitral award on the ground that the award contained errors and the parties had been prevented from presenting their cases was allowed at the first instance, but the Court of Appeal allowed the appeal and reinstated the award, with Tang VP (with whom Kwan and Fok JJA agreed) saying at §106:-

“Lastly, [counsel] submitted that the burden is on the party resisting an application to set aside to show that the result could not have been different. With respect, I agree with Saunders J that the burden is on an applicant to show that he had or might have been prejudiced. In some cases, the prejudice is obvious and it matters little who has the burden. Generally speaking, an applicant who complains of a violation is best placed to show that it has been prejudiced and thus, the burden to show prejudice should be on the applicant.” (emphasis added)

66.  Relying on ASM Shipping Ltd of India v TTMI Ltd of England [2006] 2 All ER (Comm) 122 at §39(3), Mr Manzoni appears to make some faint suggestion that this requirement of prejudice is not necessary and is not consistent with the international practice. All that said, he has to accept that the above ratio is binding on me. Further, with respect, I agree with Mr Ho for Party D, that this requirement in fact is consistent with the international practice:-

(1)  In DLS v DLT [2025] SGHC 61, the General Division of the High Court of Singapore explained at §177 that:-

“the presence or absence of prejudice remained relevant, for the court retains a discretion to refuse to set aside an award if no prejudice has been suffered by the challenging party: Bloomberry Resorts and Hotels Inc and another v Global Gaming Philippines LLC and another [2021] 2 SLR 1279 at [72].” (emphasis added)

(2)  In Aiteo Eastern E & P Company Limited v Shell Western Supply and Trading Limited and Others [2024] EWHC 1993 (Comm) at §203, Jacobs J mentioned the statutory requirement of substantial injustice under section 68(2) of the Arbitration Act 1996. Mr Manzoni rightly points out that Hong Kong legislation has no such requirement. But I think that the case shows that the international practice has been that substantial injustice is required, so much so that in the UK, it has become a statutory requirement. As DLS v DLT, supra, shows, though not a statutory requirement, this can be a factor for the Court’s consideration in determining whether to set aside an arbitral award.

(3)  In Gary B. Born, International Commercial Arbitration, supra at §25.04[E][3], the learned authors state:-

“Moreover, in contrast to interlocutory challenges to arbitrators, annulment of an award requires a showing of materiality and prejudicial effects of the arbitrator’s bias on the arbitral process – which can provide a substantial obstacle to annulment of an award, when based on one arbitrator’s asserted lack of impartiality, made after a lengthy and otherwise satisfactory arbitral process.” (emphasis added)

67.  Insofar as there is any suggestion that requiring such substantial injustice would mean actual bias rather than apparent bias, I disagree.

(1)  As an illustration, there is apparent (or even actual) bias towards a witness of the losing party. However, in the award, the winning party wins the case purely on the point of law. There is every reason why the award should be upheld.

(2)  Another illustration is that there is apparent bias towards a witness of the losing party. However, the witness is not relied upon because his evidence is peripheral or even irrelevant. Again, there is every reason why the award should be upheld.

B. Analysis

68.  Mr Manzoni’s arguments, in support of the application to set aside the Partial Award, can be summarised as follows:-

(1)  Witness Z’s evidence was put at issue for the Tribunal to determine. According to Party D’s opening submissions made in the Subject Arbitration, Witness Z’s evidence was crucial and therefore he had to be cross-examined. Witness Z was cross-examined, and Party D criticised his evidence (see §24 above) as, for example, “endlessly changing narrative”, “shifting case”, “demonstrably false”, “incredible”, and appeared to suggest that the one-page LRC Bond was not authentic. In that regard, Member R made a comment that “it would be interesting to see” the LRC Bond document. Further, during the cross-examination, it was put to Witness Z that the investment in the Fund Note was a bad faith dissipation.

(2)  The Tribunal made findings on the credibility and reliability of Witness Z’s evidence and the documents produced by him. The Tribunal found it “puzzling” that Witness Z could not provide more details of the LRC Bond (see §27 above), implicitly rejected Witness Z’s evidence about the terms and purpose of the LRC Bond (see §22 above), including his evidence on the discussion with Mr G about the acquisition of the LRC Bond, which was crucial to Party P’s case in the Subject Arbitration that the purpose of the IMA was to deal with certain foreign exchange regulations. Further, the Tribunal expressed sympathy towards Party D’s claim that the Fund Note was a dissipation of Party D’s fund (see §39 above). Though the Tribunal did not expressly make finding on the credibility and reliability of Witness Z’s evidence, the above show that it did so implicitly.

(3)  Member R should have disclosed to the parties his involvement as an arbitrator in the Prior Arbitration, having regard to the temporal proximity of the two Arbitrations. The hearing in the Prior Arbitration took place in May 2024, and the Prior Award in the Prior Arbitration (to which two LRC Group entities were parties) was issued on 19 September 2024:-

(a)  1 month after the hearing and 3 months before the issuance of the Award, in the Statement of Defence was filed on 24 June 2024, where LRC Bond was mentioned;

(b)  More than 3 months after the hearing and just 2 weeks before that, on 4 September 2024, the Tribunal issued Procedural Order No. 2 which contained references to the LRC Bond and the “Offshore Issuers”; and

(c)  More than 6 months after the hearing and less than 2.5 months after the issuance of the award in the Prior Arbitration, the Statement of Reply and Defence to Counterclaims was filed on 7 December 2024, where Witness Z was mentioned.

(4)  Member R must have received confidential information in the Prior Arbitration relating to Witness Z, LRC Group, the nature and probity of their business. However, Member R, bound by confidentiality, could not disclose all these, and Party P’s “predicament lies precisely in its inability to further verify and investigate” the matter.

(5)  Member R was argumentative in Member R’s Answers, appearing defensive and thus giving a fair-minded and informed observer an impression that he had something to hide.

(6)  The Tribunal’s failure to observe the Bifurcation Agreement and the Bifurcation Order adds to the view formed by a fair-minded and informed observer of a real possibility of bias.

(7)  Taking the above circumstances cumulatively, a fair-minded and informed observer would come to a view that there was a real possibility of bias.

69.  The question to ask in the first stage of the test (see §47 above) is, are the circumstances identified by Mr Manzoni correct?

70.  A fair-minded and informed observer would also read the Partial Award. In my view, such an observer, having read the Partial Award, would come to the following view about the Partial Award:-

(1)  For Issue 1 determined in the Partial Award (see §§22-30 above), in rejecting Party P’s contention, the Tribunal did not deal with Witness Z’s evidence at all. Rather, it dealt with the relevant documents and the text of the documents, and dealt with Dr H’s evidence (see §§25-27 above). Then, in accepting Party D’s contention (see §29 above), the Tribunal noted Witness Z’s evidence that the bond certificate was provided to Party D, but did not make any definitive finding on this. The Tribunal expressed a puzzling view on Witness Z’s failure to provide more details about the terms, and this failure alone was sufficient to sustain such a puzzling view (see §30 above). Since there was no evidence on more details about the terms, the Tribunal did not make any finding on the details. The Tribunal then looked at the inherent probability on the assumption that the terms of the LRC Bond produced by Witness Z were as indicated (see §209 of the Partial Award), and the contrast with other documents to come to the view that the LRC Bond could not be relied upon. Witness Z’s evidence did not appear to be material at all. In fact, the Tribunal even went so far as to analyse the LRC Bond produced by Witness Z as it was, without any finding on its authenticity despite Party D’s attacks.

(2)  For Issue 3 determined in the Partial Award (see §§31-37 above), although Witness Z was mentioned (see §31 above), his involvement was so much about his agreement to the Party P’s proposal to invest in the Fund Note. It must be remembered that for Issue 3, the Tribunal had to determine who had the control of the Managed Assets before the Fund Note (see §32 above), and it was Party P’s position that before the Fund Note, Party D already had control, accepting that Party P had control over the Fund Note. The Tribunal’s finding over control over the Managed Assets was in relation to the Managed Assets before the Fund Note came into picture. While the Managed Assets included the LRC Bond and thus the LRC Bond document produced by Witness Z was considered (see §275 of the Partial Award), it was considered on the assumption that it was true (see §275 of the Partial Award), and thus nothing was said against Witness Z here. It was only that despite the terms of the LRC Bond, and despite that the terms showed that Party D had some control over the Managed Assets, that the Tribunal, having considered other evidence like Party P’s failure to respond to Party D’s repeated requests and demands, came to the view that Party P still had sufficient control to be more than a bookkeeper.

(3)  For Issue 4 determined in the Partial Award (see §§38-39 above), while Party D suggested that the Fund Note was a way deployed by Party P to dissipate Party D’s assets, it must, however, be remembered that for Issue 4, the question was whether there was any breach of duty of care, and Party D’s case was that the investment was a bad investment or even so bad that it was a deliberate way to dissipate Party D’s assets. At §317 of the Partial Award, the way to dissipate would be by way of management fees received by Party P higher than the coupon rate of the Fund Note, thus guaranteeing a loss of money, even perpetually if Party P’s case that the IMA could not be terminated. Thus, Party D’s case was that by such an investment guaranteeing a loss, there was breach of duty of care. It was in this context that the Tribunal expressed at §318 of the Partial Award its sympathy with Party D’s argument that the investment in the Fund Note was dissipation. While Party D might have suggested in cross-examination that Witness Z was a party knowingly participating in the dissipation, §318 did not indicate anything of that sort at all.

(4)  In gist, although Party D put the credibility and reliability of Witness Z’s evidence in issue before the Tribunal, a fair-minded and informed observer would observe that the Tribunal did not deal with this issue, explicitly or implicitly, in the Partial Award and/or that the Tribunal did not find it necessary to deal with this issue in the Partial Award. Further and in any event, having considered the Partial Award on its own, a fair-minded and informed observer would not, in my view, see any dishonesty or unfavourable finding against Witness Z at all. It would only be when there was such a finding against Witness Z would a fair-minded and informed observer think further why there were such a finding (for example, some possibility of bias). From the analysis of the Tribunal, a fair-minded and informed observer would also observe that the Tribunal relied upon the evidence in the Subject Arbitration only. Overall, on the Partial Award itself, a fair-minded and informed observer would not see any real possibility of bias.

71.  But a fair-minded and informed observer does not only read the Partial Award. A fair-minded and informed observer would also consider the following:-

(1)  The temporal proximity as submitted by Mr Manzoni, and relatedly, whether there was any failure to make proper disclosure;

(2)  Given the temporal proximity, it must have rung the bell to Member R that LRC and Witness Z had appeared in the Prior Arbitration;

(3)  As gleaned from the IBA Guidelines at the very least, if not the relevant case law, the starting point in such a situation was that Member R had no duty to disclose;

(4)  It may be that no disclosure was made because Member R had not made any extreme comments on Witness Z;

(5)  It may be that Member R did make some adverse comments on Witness Z but had the financial interest in staying as the arbitrator and thus did not make the disclosure;

(6)  Professional pressures on Member R in relation to his integrity and impartiality;

(7)  Nothing has been suggested negatively about Member R’s professional integrity and his experience, and/or that Member R would put aside his professional integrity for his financial interest;

(8)  Member R may appear to be argumentative in Member R’s Answers because he did care about his professional integrity and his experience, or because he did really do something wrong that he had to hide, and/or because any professional would tend to be defensive when accusations are made against him, but this does not necessarily mean, retrospectively, a real possibility that he had been biased; and

(9)  The Bifurcation Agreement and the Bifurcation Order gave room for difference between the parties and the Tribunal itself as to what they exactly meant (and in my view to be explained below, there was no breach of the Bifurcation Agreement and the Bifurcation Order).

72.  The biggest grievance of Party P, or the most attractive argument by Mr Manzoni, lies in Party P’s predicament in its inability to know what happened in the Prior Arbitration. This, in my view, should also be considered by a fair-minded and informed observer, who may well feel stronger about the matter in §71(5) above. However, a real possibility of bias does not necessarily follow from such inability and such feeling.

73.  In my view, a fair-minded and informed observer may have some doubt as a result of the matter set out in §71(5) that Member R may have financial interest in staying as the arbitrator. That said, having considered all the relevant circumstances above including the Partial Award, this would remain a doubt only to a fair-minded and informed observer, who would, in my view, conclude that there was no real possibility that the Tribunal or Member R was biased. Therefore, the Partial Award should not be set aside.

74.  By way of footnote, Mr Ho urges me to read Member R’s Answers in some detail, while Mr Manzoni urges me to pay little regard to Member R’s Answers except that the Answers appeared to be argumentative thus suggesting apparent bias. As my analysis above shows, it is not necessary for me to go into the details of Member R’s Answers, but I do note my acceptance of Mr Manzoni’s submissions that they appeared argumentative (see §70(8) above).

75.  Further and in any event, from the Tribunal’s analyses and decisions in the Partial Award, in my opinion, the outcome of the Partial Award would not be affected by Witness Z or his credibility and reliability. As canvassed above, the Tribunal mainly, if not only, relied on the inherent probability and documentary evidence, and perhaps also Dr H’s oral evidence, to come to its findings. In other words, irrespective of whether there was any apparent bias, the outcome would, in my view, be the same. This is another ground for not setting aside the Partial Award, even if there were apparent bias.

IX.  BIFURCATION AGREEMENT AND BIFURCATION ORDER

76.  I now deal with Ground (2) concerning the Bifurcation Agreement and the Bifurcation Order.

A.  Legal principles

77.  Article 34 of the UNCITRAL Model Law (given effect to by section 81 of the Arbitration Ordinance) provides that:-

“(2) An arbitral award may be set aside by the court… only if:

(a) the party making the application furnishes proof that:

(iv) the composition of the arbitral tribunal or the arbitral procedure was not in accordance with the agreement of the parties, unless such agreement was in conflict with a provision of this Law from which the parties cannot derogate, or, failing such agreement, was not in accordance with this Law…”

78.  As Mimmie Chan J put it in P v M [2018] HKCFI 2280 at §18, “[p]arties to an arbitration have the right to expect that it will be conducted fairly, in accordance with the procedure they had agreed under the rules to which they had submitted as governing the arbitration, the conduct of the arbitration, and the rules of pleading of their case”.

79.  Further, despite the parties’ agreement, the arbitral tribunal is not bound by the same if to give effect to such agreement would undermine a party’s opportunity to present his case fairly, relying on Brunswick Bowling & Billiards Corp v Shanghai Zhonglu Industrial Co Ltd [2011] 1 HKLRD 707 at §§85-86. That is a high-level, general principle. In my view, the Court should be slow to come to find that such an opportunity has been undermined. One consideration is whether the parties themselves have agreed to limit such opportunity. If so, one may argue that the opportunity to present his case fairly has not been undermined, because that is what the parties have agreed to. All depend on the context and the facts.

80.  Lastly, again, to set aside an arbitral award, the applicant bears the burden to show that the outcome would have been difference: see §§65-67 above.

B.  Analysis

81.  As I have explained in §§36-37 above, the parties did not put it as an issue for determination, whether in Phase 1 or Phase 2, on the liability or recoverability of the additional losses. Therefore, there is no question of infra petita as submitted by Mr Manzoni.

82.  Mr Manzoni submits that in any event, by the Tribunal’s email sent to the parties on 6 October 2025 (§18 above), the Tribunal closed Phase 1 in accordance with Article 31.1 of the 2018 HKIAC Administered Arbitration Rules and thus the issue could not be reserved and the matter cannot be re-opened by virtue of Article 31.4. These two articles provide that:-

“31.1 When it is satisfied that the parties have had a reasonable opportunity to present their case, whether in relation to the entire proceedings or a discrete phase of the proceedings, the arbitral tribunal shall declare the proceedings or the relevant phase of the proceedings closed. Thereafter, no further submissions or arguments may be made, or evidence produced in respect of the entire proceedings or the discrete phase, as applicable, unless the arbitral tribunal reopens the proceedings or the relevant phase of the proceedings in accordance with Article 31.4.

…

31.4 The arbitral tribunal may, if it considers it necessary, decide, on its own initiative or upon application of a party, to reopen the proceedings at any time before the award is made.” (emphasis added)

83.  In my view, these articles apply only to issues already put in issue for determination. When an issue has not been put to determination, one cannot say “further submissions” or “reopen” the proceedings on that issue, since that issue has never been a live issue for the tribunal and the parties. Given that the issue on the liability or recoverability of the additional losses was not an issue for Phase 1 (nor Phase 2), the Tribunal, concerned with the question of law on such additional losses, was, in my view, entitled to raise this issue of its own motion for determination. I add that the Tribunal has done so in favour of Party P, because but for this, the parties’ agreement implicit in the Bifurcation Agreement and the Tribunal’s approach implicit in the Bifurcation Order, as explained above, was to assume that the additional losses were recoverable, and the only question left would be for assessment in Phase 2 only.

84.  Having raised this issue on its own, it would be a matter of case-management of how to deal with it. It was reasonable as proper case-management for the Tribunal to deal with it together with the issues in Phase 2.

85.  In the circumstances, I find no breach of the Bifurcation Agreement and the Bifurcation Order. I also find no injustice caused to Party P, and on the contrary, the Tribunal’s approach has done a favour to Party P.

86.  Therefore, there is no ground to set aside the specific paragraphs of the Partial Award concerning these additional losses.

X.  CONCLUSION

87.  In the circumstances, I make the following order:-

(1)  The Originating Summons be dismissed;

(2)  Party P shall pay Party D costs (including costs of Party P’s application for recusal of the Recorder originally assigned to hear the Originating Summons[1]) to be summarily assessed on paper, with certificate for two counsel; and

(3)  For the summary assessment, Party D shall lodge a statement of costs within 3 days from today and Party P shall lodge a list of objections 3 days thereafter.

88.  It remains for me to thank Mr Manzoni, Mr Lam and Mr Wong (for Party P) and Mr Ho, Mr Lo and Mr Jiang (for Party D) for their able and thorough assistance.