Moody's Investors Service Hong Kong Ltd v. Securities and Futures Commission

CACV 103/2016

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF APPEAL

CIVIL APPEAL NO 103 OF 2016

(ON APPEAL FROM THE SECURITIES AND FUTURES APPEALS

TRIBUNAL APPLICATION NO 4 OF 2014)

________________________

________________________

BETWEEN

_______________________

_________________

JUDGMENT

_________________

Hon Lam VP (giving the Judgment of the Court):

Background

1.Since 2011, regulated activities under the Securities and Futures Ordinance [“SFO’] Cap 571 as specified in Schedule 5 Part 1 of the Ordinance include providing credit rating services (listed as Type 10 activities in Part 1).  “Credit ratings” is defined in Part 2 of Schedule 5 as follows:

“ Credit ratings means opinions, expressed using a defined ranking system, primarily regarding the creditworthiness of -

(a) a person other than an individual;

(b) debt securities;

(c) preferred securities;

(d) an agreement to provide credit.”

2.There is also a definition for “providing credit rating services” in Part 2:

“ Providing credit rating services means –

(a) preparing credit ratings-

(i) for dissemination to the public, whether in Hong Kong or elsewhere; or

(ii) with a reasonable expectation that they will be so disseminated; or

(b) preparing credit ratings-

(i) for distribution by subscription, whether in Hong Kong or elsewhere; or

(ii) with a reasonable expectation that they will be so distributed,

but does not include-

(c) preparing, pursuant to a request made by a person, a credit rating which is exclusively prepared for, and provided to, the person and that is neither intended for dissemination to the public or distribution by subscription, whether in Hong Kong or elsewhere, nor reasonably expected to be so disseminated or distributed; or

(d)   gathering, collating, disseminating or distributing information concerning the indebtedness or credit history of any person.”

3.Moody’s Investors Service Hong Kong Limited [“Moody’s”] is part of the global Moody’s credit rating agency network.  It is licensed under the SFO to carry on Type 10 activity and has been so since June 2011.  On 11 July 2011, it published a document which it called a special comment with this heading: “Red Flags for Emerging-Market Companies: A focus on China” [“the Report”].  It was a Report of 25 pages which was distributed to subscribers.  It was also available for sale to the general public.

4.At the time of the publication of the Report, there was concern in the market about the accounting standards and corporate governance of some Mainland corporations which had issued debt securities. Moody’s issued a press release when it published the Report.  The press release introduced the Report in these terms:

“ [Moody’s] highlights … governance and accounting risks prevalent when investing in fixed-income securities in the emerging markets.

Through a framework of ‘Red Flags’, the report focuses on transparency concerns and the general complexities associated with rapidly developing markets.”

5.The red flag framework was explained in the Report as follows:

“ Red flags as a screen. In rapidly developing emerging markets, the use of frameworks to assess elements of credit risk provides consistency in identifying relative strengths and weaknesses across a growing pool of rated issuers. In this report, we look at 20 red flags, grouped into five categories, that highlight issues meriting scrutiny to identify possible governance or accounting risks for non-financial corporate issuers in emerging markets. These categories are as follows: (1) possible weaknesses in corporate governance, (2) riskier or more opaque business models, (3) fast-growing-business strategies (4) poorer quality of earnings or cash flow, and (5) concerns over auditors and quality of financial statements.”

6.The Report described the relationship between the red flag framework and ratings:

“ Limited correlation in China between lower ratings and larger numbers of red flags. In China, property companies with lower ratings do not tend to have a greater number of red flags. However, a degree of correlation exists for non-property firms, where on average, investment-grade issuers of Baa and above trip 6 red flags, while high-yield Ba issuers trip 7, B issuers trip 8, and Caa issuers 12. These findings show that screens for governance or accounting risks can help identify areas to investigate but cannot serve as mechanisms to rank order credit risk.”

7.Twenty three Chinese non-property issuers and 26 Chinese property issuers were examined and the number of red flags tripped in respect of each issuer was set out in Appendix 3 Figures 9 to 18 of the Report.  Amongst these companies, 5 non-property issuers tripping highest numbers of red flags were identified as negative outliers.  In addition, 1 property issuer was also mentioned specifically in the context of negative outliers.  The Report was cautious as to the implications arising from these observations:

“ An issuer’s tripping of many red flags does not represent an immediate rating concern because our ratings already reflect many of the issues highlighted by the relevant red flags, and the ratings also incorporate more than just the potential concerns that the flags capture. Moreover, as indicated, there is only limited correlation between lower ratings and a higher number of red flags tripped. Nevertheless, we provide some comments below on the red flags tripped by the negative outliers, and what risks these flags are highlighting.”

The SFC decision and the appeal to the Securities and Futures Appeals Tribunal

8.The Securities and Futures Commission [“the SFC”] conducted an enquiry in respect of the Report.  In a Decision Notice of 3 November 2014, the SFC informed Moody’s that it had determined that Moody’s had failed to meet the standards and comply with the practices expected of a licensed corporation in the publication of the Report.  The SFC found that Moody’s had failed to have the required procedural safeguards in place to ensure the integrity of the Report and the Report was materially misleading, confusing and inaccurate in several respects.  After the publication of the Report, the share prices of more than half of the companies red-flagged dropped substantially.  The SFC took the view that though the Report was not the sole contributing factor to the drop in the share prices, it was a significant contributor.  The SFC concluded that Moody’s acted in breach of the Code of Conduct issued under section 169 of the SFO in these respects: General Principle 1- Honesty and Fairness; General Principle 2 – Diligence and para 4.3- internal control, financial and operational resources. 

9.For such misconduct, the SFC imposed a penalty of public reprimand and a fine of $23 million against Moody’s.

10.Moody’s sought a review of the SFC’s decision pursuant to section 217(1) of the SFO.  The review was heard by the Securities and Futures Appeals Tribunal[1] in September 2015.  The Tribunal handed down its determination on 31 March 2016.  The Tribunal found as follows:

(a) in the preparation and publication of the Report, Moody’s was carrying on Type 10 activities, thus regulated under the SFO;

(b) there were substantive breaches of General Principles 1 and 2 of the Code of Conduct;

(c) Moody’s should be subject to a public reprimand together with a pecuniary penalty of $11 million.

The appeal to the Court of Appeal

11.Moody’s appealed against that determination. Pursuant to section 229 of the SFO, this court can only entertain an appeal from the Tribunal on a point of law.  We heard the appeal on 11 January 2017. In the Notice of Appeal of 25 April 2015, three grounds of appeal were advanced:

(a) The Tribunal was wrong in law in holding that the preparation and publication of the Report constituted the carrying on of Type 10 activity;

(b) The Tribunal was wrong in law in affirming that the SFC had jurisdiction to exercise its disciplinary powers under section 194 of the SFO;

(c) The Tribunal erred by determining the jurisdiction issue by reference to whether the Report would have been likely to convey to the readership the message that it contained ‘credit ratings’ rather than by reference to an objective evaluation of whether the Report did actually contain ‘credit ratings’ as defined in the Ordinance. 

12.With respect, ground (c) is obviously without merit.  In the Reasons for Determination, the Tribunal carefully analysed the statutory scheme in the SFO [2]and explained why the preparation and publication of the Report constituted Type 10 activity by reference to the contents of the Report and the press release issued by Moody’s introducing the Report[3].  In this respect, the key conclusions and findings of the Tribunal were set out in the Reasons for Determination as follows:

(a) Credit ratings are opinions on credit worthiness.  Such opinions, dealing not only with historical fact, must also deal with current and future uncertainties on a forward-looking basis and such opinions may have to be on-going with clarifications and supplements from time to time.  There is no statutory limitation as to the form of such opinions, the only qualification is that they have to be expressed using a ‘defined ranking system’.  The ranking system needs not gear towards any particular integration of mathematical and accounting data and may come in many forms.  It may be self-contained or it may be complementary to an existing system.  There is no statutory limitation on the factors that may be taken into account in the assessment, it is up to the provider to determine what factors are to be incorporated and the weight to be given to them, see §§81-86.  

(b) The Report expressed opinions on credit worthiness of the corporate issuers.  On an objective assessment the red flag framework constitutes a system aimed at identifying possible governance or accounting risks which in turn was put together as a system for assessing potential credit risk, and as such constituted a credit rating, see §§87-93. 

(c) Whilst acknowledging that there were qualifications in the Report, on the whole the red flag framework was clearly a system that assessed elements of credit risk by a form of ranking.  The qualifications only set out limitations in the system.  See §§94-97.

(d) In the alternative, even if the red flag framework did not of itself constitute a form of credit rating, it was intended to be read as amplifying and supplementing Moody’s ratings.  As such, the Report was part and parcel of Moody’s ratings, §§102-105. 

13.In his submissions, Mr Huggins SC referred to two sentences at §96 and §151 of the Reasons for Determination to support Ground (c).  We have read those paragraphs with care and we cannot accept that the isolated statements should be read as indicating that the Tribunal adopted the wrong approach suggested by Mr Huggins.  With respect, the way counsel urged us to read those statements is simply to read those statements out of context with a complete disregard of the rest of the Reasons for Determination which clearly shows that the Tribunal did not commit the mistake as submitted by counsel.  We have no hesitation in rejecting the submission.

14.In his written submissions, Mr Huggins made an alternative point not canvassed in the Notice of Appeal.  He submitted that if it was permissible to refer to market reaction in assessing whether the preparation and publication of the Report constituted Type 10 activity, account should be taken of published reviews of the Report which appreciated that a line had been drawn by Moody’s between the Report and its credit ratings and that its warnings (about the companies which tripped red flags) were short of credit downgrades.  No leave had been sought to argue the point and we did not grant leave for the point to be canvassed.  It was based on the false premise that the Tribunal adopted the incorrect approach that market reaction determined whether the Report was a credit rating exercise.  As we said above, the Tribunal did not adopt that erroneous approach.

15.Coming now to ground (b), this court is hearing an appeal from the Tribunal, not an appeal from the decision of the SFC.  Thus, whatever defects there may be in the decision of the SFC, those should not be the focus of this appeal.  It is obvious that the Tribunal reached its determination in some respects differently from the reasoning of the SFC. Notably, the Tribunal did not find any contravention of para 4.3 in terms of internal control.  The sanction imposed by the Tribunal was therefore less severe than that originally imposed by the SFC. 

16.Under Section 218 of the SFO, in a review the Tribunal may confirm, vary or set aside the decision of the SFO and substitute for the decision any other decision which the Tribunal considers appropriate, see section 218(2)(a).  The Tribunal is specifically given the power to make any decision that the relevant authority had power to make, whether or not under the same provision as that under which the original decision was made, see section 218(3).  Thus, the Tribunal was clearly correct in holding that it should conduct a full merits review (see footnote 12 at p.33 of the Reasons for Determination).  Before us, Mr Huggins did not dispute this proposition.  On that analysis, whatever defects were contained in the SFC decision were water under the bridge.

17.Mr Huggins however argued that the Tribunal should not have ignored the fact that the SFC’s own position was that the Report did not contain credit ratings, and there was only misconduct on the basis that it was “in close proximity” to the carrying out of Type 10 activity. 

18.We do not accept this submission.  The Tribunal is either correct or incorrect in its legal analysis as to the preparation and publication of the Report being a Type 10 activity.  This is the issue we shall determine under Ground (a).  The way in which the SFC originally proceeded against Moody’s would not matter unless Moody’s suffered any procedural prejudice leading to injustice in terms of relevant evidence not being tendered before the Tribunal.  As Mr Yu SC submitted, Mr Huggins did not suggest that there was any unfairness arising from the Tribunal’s adoption of a line of reasoning different from the SFC’s approach.  Hence, SFC’s reasoning cannot be germane to our determination of Ground (b).   

19.In our judgment, Ground (b) is not a relevant issue for the resolution of this appeal and it is not profitable for us to engage in a discussion of the same.

20.The only ground which merits discussion at some length in this appeal is Ground (a).  As we have seen, the Tribunal came to its conclusion that Type 10 activity is engaged on two alternative bases: (i) the Report itself constituted credit ratings; or (ii) the Report amplified or supplemented Moody’s credit ratings, and as such it was part and parcel of the same.

Did the Report amplify or supplement Moody’s credit ratings?

21.Though the Tribunal devoted a greater part of its Reasons for Determination in explaining its conclusion on (i), we prefer to examine (ii) first.

22.Mr Huggins submitted that the Tribunal was wrong in concluding that the Report was part and parcel of Moody’s credit ratings because:

(a) The evidence did not establish Moody’s actually used the red flag framework in preparing and distributing its credit ratings;

(b) The SFC’s case was that the Report was misleading in conveying that the red flag framework had been used in Moody’s credit ratings when in fact it had not;

(c) In the Report, Moody’s was not explaining the methodologies it had used in providing its credit ratings.  Moody’s clearly stated that the red flag framework was in addition to the methodologies used in providing the credit ratings, and not part of those methodologies;

(d) The Report being supplemental to and/or amplification of Moody’s credit ratings must have been distinct from the actual preparation and distribution of the credit ratings themselves;

(e) The red flag framework was not a necessary activity to enable Moody’s to provide a credit rating.

23.With respect, this line of argument cannot assist Moody’s.  We accept that in an appeal on point of law, the Court is not confined to instances in which it is apparent on the face of the record that the determination appealed against resulted from a specifically identifiable error of law.  In Runa Begum v Tower Hamlets LBC [2003] 2 AC 430 at p.462G-H, Lord Millett, referring to Edwards v Bairstow [1956] AC 14, summarized the permissible scope of appellate intervention in an appeal on questions of law as follows:

“ A decision may be quashed if it is based on a finding of fact or inference from the facts which is perverse or irrational; or there was no evidence to support it; or it was made by reference to irrelevant factors or without regard to relevant factors. It is not necessary to identify a specific error of law; if the decision cannot be supported the court will infer that the decision-making authority misunderstood or overlooked relevant evidence or misdirected itself in law.”

24.This approach was adopted by the Court of Final Appeal in Kwong Mile Services Ltd v Commissioner of Inland Revenue (2004) 7 HKCFAR 275 in which Bokhary PJ further observed at [37]:

“ In an appeal on law only the appellate court must bear in mind what scope the circumstances provide for reasonable minds to differ as to the conclusion to be drawn from the primary facts found. If the fact-finding tribunal’s conclusion is a reasonable one, the appellate court cannot disturb that conclusion even if its own preference is for a contrary conclusion. But if the appellate court regards the contrary conclusion as the true and only reasonable one, the appellate court is duty-bound to substitute the contrary conclusion for the one reached by the fact-finding tribunal. The correct approach for the appellate court is composed essentially of the foregoing three propositions.”

25.In the present case, at §102 of its determination, the Tribunal made this finding:

“ … it is evident that [the Report] was intended to be read as amplifying and supplementing Moody’s ratings, as being so intimately attendant upon them that it constituted more than mere comment and became part and parcel of Moody’s ratings themselves …”

26.The Tribunal explained why it came to that conclusion at §§103 to 107.  At §105, the Tribunal said:

“ On the basis that credit ratings are ‘probabilistic opinions about future credit worthiness’, it must be integral to Moody’s rating activities that, in times of concern, it should seek to assist the market by adding to or clarifying ratings (of the ‘classic kind’) earlier made. Certainly, a reading of the Report shows that the ‘red flag’ warning signs are to be read in conjunction with existing ratings…”

27.At §§162 and 163, the Tribunal further observed:

“ In the judgment of the Tribunal, educated readers -- they being the great majority of subscribers to Moody’s publications and therefore the ordinary readers – would have treated the red flag framework as a ‘screen’ to enable them to be given extra insight into Moody’s established systems …

On this basis, the Tribunal is satisfied that the red flag framework would have been seen as an extension to Moody’s existing methodologies -- a means for ‘fine-tuning’ – its ratings (of the classic kind) …”

28.Mr Huggins could not and did not challenge these findings of the Tribunal.  No submission was made to the effect that the Tribunal fell into any of the Edwards v Bairstow errors in coming to such conclusions.

29.The core proposition in counsel’s submission is that as the Report was not part of the process in the preparation of Moody’s credit ratings and the red flag framework was not adopted as part of the methodology in working out such credit ratings, the Report should not be taken as part and parcel of the credit ratings.   

30.We cannot accept this proposition.  The Tribunal did not consider the notion of the Report being part and parcel of the credit ratings in a vacuum.  As shown at §106, the Tribunal applied that notion in the context of section 193 of the SFO which defines misconduct as including “an act or omission relating to the carrying on of any regulated activities”.  On the facts of the case, the Tribunal concluded at §106:

“ … In the opinion of the Tribunal, at the very least, the red flag framework contained in the Report, so clearly clarified and/or expanded upon the ratings of a classic kind already made by Moody’s in respect of the companies under review in the Report that, in giving a purposive construction to the relevant statutory provisions, the act of publication must be held to be an act relating to those earlier ratings.”

31.Reading §105 together with §106, the Tribunal was clearly entitled to hold that even though the red flag framework was not part of the methodology in arriving at Moody’s credit ratings of a classic kind (a point which the Tribunal fully acknowledged at §164), the Report did constitute additions and clarifications which were meant to be read together with such classic ratings and as such the publication of the Report was an activity relating to the ratings within the meaning of section 193. 

32.Mr Huggins submitted that it is not enough for the purpose of section 193 that an activity is found to be related to or connected with credit ratings.  He submitted that as the phrase “relating to” refers to activities in the carrying on of the business of credit ratings, only steps taken in the preparation of the credit ratings could fall within the scope of that section.

33.For the reason given by the Tribunal at §105, the business of credit ratings encompasses clarifications or additions to existing ratings on an ongoing basis.  As explained by the Tribunal at §§108 to 119 of its determination, this construction is consistent with the purposive interpretation of the statute and the proportionate interference with the freedom of expression.  We do not accept the narrow construction of that phrase put forward by Mr Huggins.

34.In his written submissions, Mr Huggins further raised the point that the charge against Moody’s referred to the Code of Conduct published by the SFC.  Under section 169(1), the Code could only apply to the carrying on of the regulated activities.  Hence, there could not be a breach of the Code in respect of non-regulated activities.  Counsel submitted that the publication of the Report is not the carrying on of the business of credit ratings. 

35.This is not a ground in the Notice of Appeal. In any event, once we reject the narrow construction of section 193 as to the scope of the business of credit ratings, there is no merit in the argument. Further, having proper regard to the Notice of Proposed Disciplinary Action of 14 February 2013, the inquiry was instigated in respect of misconduct pursuant to section 194(2) of the SFO arising out of the preparation and publication of the Report and as explained earlier the reasoning of the SFC did not bind the Tribunal.

36.We therefore agree with the Tribunal that misconduct can be established on the basis that the preparation and publication of the Report was part and parcel of the carrying on of the business of credit ratings by Moody’s.

Did the Report itself constitute credit rating?

37.Turning now to the holding by the Tribunal that the Report itself constituted credit ratings, it is useful to remind ourselves of the definition of “credit ratings” in Part 2 of Schedule 5 of the SFO. There are two essential components in the definition:

(a) Opinions primarily regarding the creditworthiness of the specified subjects;

(b) Such opinions are expressed using a defined rankings system.

38.We have highlighted the adverb “primarily” in the first element because, as we shall explain, it has a crucial importance in the context of the arguments before us.

39.It is common ground that a defined ranking system is not necessarily a credit rating.  The SFC accepted that a “buy/sell/hold” ranking would not constitute a credit rating.  Instead, coming from a broker, it would be advice to a client concerning the acquisition or disposal of securities and regulated as Type 4 regulated activity[4]. Mr Yu acknowledged this position and explained that as creditworthiness is only a factor impacting upon such advice, a “buy/sell/hold” ranking does not fall within the definition of credit rating.

40.Mr Huggins submitted that likewise to rank companies for creditworthiness is not the same thing as to use a ranking system only in relation to some elements of credit risk. The Report only addressed corporate governance and accounting risks[5], which are only two elements of credit risk. Other relevant elements include the country and sector in which the company operates, the general macroeconomic outlook in respect of the relevant country and sector, the ratio of a company’s earnings to its debt payment obligations, the terms governing and timeframe for repayment of its debts, the company’s long-term level and predictability of cash flow, and the regulatory environment for the sector concerned. Thus, counsel said, it is not correct to assume that an assessment of some elements of credit risk is equivalent to an assessment of creditworthiness. In the present context, one cannot simply say that a high level of corporate governance and/or accounting risk (without considering the other relevant factors) means a high risk of the company defaulting on its debt obligations. Counsel stressed that the two aspects discussed in the Report (corporate governance and accounting risks) form but an extremely limited subset of a whole basket of factors in the consideration of creditworthiness. And these two aspects cannot be determinative of creditworthiness.

41.Counsel drew our attention to the statements in the Report that the correlation between the number of flags and creditworthiness is limited and the use of the flags as screens for governance or accounting risks can help identify areas to investigate but cannot serve as mechanisms to rank order credit risk.

42.Thus, Mr Huggins submitted, the Report did not express any opinion primarily on the creditworthiness of the companies concerned.  He urged us to bear in mind the requirement of primacy of the opinion on creditworthiness in the definition. 

43.It is fair to say that the adverb “primarily” in the definition did not receive much attention until the last set of submissions from Mr Huggins on 16 March 2017.  Thus, in the discussion on this issue in the determination of the Tribunal[6]:

(a) reference was first made to the broad definition for credit rating and the need to meet exigencies of changing circumstances to allow for different forms of credit ratings (§§79 to 82);

(b) then the Tribunal considered if the Report could be regarded as a defined ranking system (§§83 to 84 and 88 to 89);

(c) at the same time, the Tribunal explained why the system constituted a mechanism for judging levels of credit risk and creditworthiness (§§85 to 87 and 90 to 93);

(d) the Tribunal addressed the qualifications in the Report that the red flag framework could not per se serve to rank credit risk (§§94 to 101).   

44.It did not appear that the argument based on the emphasis on the adverb “primarily” was advanced and the Tribunal therefore did not address the same.  As we recall, it was not put forward in the submissions before us at the hearing of the appeal on 11 January 2017. 

45.But a main element of the argument, viz the distinction between elements of credit risk and a credit rating, had been advanced.  In any event, since it is a point of law and one cannot construe the statutory definition without regard to that adverb, we are of the view that we should address the same in this judgment.

46.In principle, we are persuaded that Mr Huggins is correct that there is a distinction between assessment of one or two elements of credit risk and the credit rating itself.  Though, as the Tribunal observed at §86, the statutory definition does not limit the factors that may be taken into account in an assessment of creditworthiness, the whole exercise must be the expression of opinion “primarily regarding the creditworthiness” of the subjects concerned.  The use of the adverb, in our judgment, highlights the distinction made by Mr Huggins.

47.Different credit rating agencies may have different baskets of factors.  In determining whether an agency did provide credit rating services, the Commission or the Tribunal need not assess whether the factors or elements that such agency had taken into account were scientifically or statistically sound in terms of the final assessment.  So long as the product was presented by way of expression of opinions primarily regarding creditworthiness (using a defined ranking system), it would be caught by the statutory definition.  We agree with Mr Yu that a badly done credit rating is still a credit rating, and as such subject to regulation.

48.But it is quite a different matter if the product was not presented as an expression of opinions primarily regarding creditworthiness and instead only presented as a discussion limited to one or two elements without expressing any opinion on the overall assessment of the creditworthiness.  In this instance, the Report only focused on corporate governance and accounting risks.  Mr Huggins quite rightly reminded us that there are many other factors apart from these two elements in the overall assessment. 

49.Take an example of one of the other factors mentioned by Mr Huggins, the country sector.  A report on assessment of risks pertaining to different country sectors, albeit expressed through some defined ranking system, would not be an expression of opinion primarily regarding credit worthiness.  Thus, even if specific securities or debts issuers were included in such a table, that by itself would not be a credit rating.

50.The same is true if one refers to an example of a report on risk pertaining to different industry sectors. 

51.In our judgment, the same reasoning is applicable with regard to corporate governance risk or accounting risk. 

52.The crucial question is whether the Report, read as a whole, went beyond the expression of opinions confined to these two elements only and expressed an opinion primarily regarding credit worthiness. The Tribunal placed reliance on the following matters in concluding that the Report was by itself a credit rating:

(a) The tripping or triggering of red flags was presented as warning signs as to governance and/or accounting risks and those with highest numbers are ranked as “negative outliers” that should attract greater scrutiny (§89);

(b) The Red Flag system was a method for identifying relative strengths and weaknesses among rated issuers and it could be applied more generally (§90);

(c) Moody’s recognized that it had put together a system for assessing potential credit risk with the greater number of flags allocated the higher the potential credit risk (§§91 to 92);

(d) The qualifications in the Report were not sufficient to militate against such conclusion.  Read as a whole, they cannot alter the very clear structure of the framework and no matter how blunt and/or unsophisticated it was, it suggested that the greater the accumulation of red flags, the greater the relative weakness of a company.  Also, a senior executive suggested that the Report should be “actionable”, which the Tribunal found to be a recommendation that the Report should be capable of being acted upon (§§94-101).

53.As we have mentioned, the Tribunal did not address the requirement of primacy in the statutory definition and the distinction between assessment of some (but not all) elements of credit risk and assessment of creditworthiness.  At §87, the Tribunal seems to have equated an assessment of some elements with the assessment of overall creditworthiness. With respect, the Tribunal failed to have regard to such distinction when it elided the assessment of strengths and weaknesses in terms of governance and accounting risks into overall assessment of creditworthiness.  Having fallen into that error of law, the Tribunal’s assessment of the significance of the qualifications is flawed. 

54.In the Report, Moody’s made it clear that though there was a degree of correlation for some non-property firms between lower ratings and larger numbers of red flags, the red flags do not represent a change in Moody’s rating methodologies (which encompassed other factors). There were references to warning signs and negative outliers, but they by no means represented that such signs or labels should be read as overall assessment of creditworthiness relating to the companies in question.  The focus of the Report was clearly confined to governance and accounting risks. The passage cited at [7] above is a clear statement that the red flags system was not to be read as ratings. 

55.We have considered the argument that if every other relevant factor was equal, the outcomes in the red flag system could be perceived as a ranking on overall creditworthiness.  However, it is quite plain that in the real world, the other relevant factors would not be equal and Moody’s did not represent in the Report that one should proceed as if they would be equal for the purpose of overall assessment of creditworthiness.

56.Further, Moody’s also highlighted that the weighting for these red flags may vary in respect of the circumstances of each company[7].

57.In other words, Moody’s did not represent in the Report that it had adopted an alternative credit risk assessment approach based on governance and accounting risks alone and that the Report expressed an opinion primarily on overall assessment of creditworthiness of the companies in the chart.

58.Even on the Tribunal’s finding as to the meaning of the suggestion that the Report should be actionable, we do not regard that suggestion as being relevant for the purpose of determining if the Report itself constitutes a credit rating.  A “buy/hold/sell” type of opinion would also be actionable.  It would not render such an opinion, even if it is expressed by way of a defined ranking system, an opinion on creditworthiness. No doubt the Report covered topics of general interest to those who participated in the financial markets and some investors might have acted upon the same. But that does not answer the question whether the Report expressed an opinion primarily on creditworthiness. 

59.In our judgment, Mr Huggins was correct in submitting that the Report expressed an opinion primarily on corporate governance and accounting risks which are relevant but far from determinative of creditworthiness.

60.We therefore respectfully differ from the determination of the Tribunal in this respect.

61.Having arrived at this conclusion, it is not strictly necessary for us to address Mr Huggins’s further submission that the red flag system did not provide a defined ranking system.  Essentially, Mr Huggins made two points: (1) the ordinality required for a ranking system is lacking as each flag potentially carried different weight in relation to each company; (2) there is no definition provided alongside the ranking system (as in the traditional alphanumeric credit ratings) to enable a reader to understand what was being said about a company’s creditworthiness.

62.We shall only comment on these submissions briefly.  In substance, point (2) is in a sense the logical extension of the argument that the Report did not express primarily an opinion on creditworthiness.  There is no need to consider it separately.  In relation to point (1), we agree with Mr Yu that given the broad statutory definition, credit ratings do not have to be expressed on an ordinal scale.  This is acknowledged in the article by Mr Johnstone on which Mr Huggins placed some reliance[8].

Disposition

63.For these reasons, we uphold the actual decision of the Tribunal though we respectfully disagree with its conclusion that the Report itself constituted credit rating.

64.The appeal is therefore dismissed with costs, such costs are to be taxed if not agreed.

Mr Adrian Huggins SC, instructed by Linklaters, for the appellant

Mr Benjamin Yu SC and Mr Laurence Li, instructed by Securities and Futures Commission, for the respondent