Supreme Court allows insider trading claim where undisclosed price-sensitive info materially affected share prices.

THE SUPREME COURT

No 144/06

Denham J.

Geoghegan J.

Fennelly J.

Macken J.

Finnegan J.

BETWEEN/

FYFFES PLC

PLAINTIFF/APPELLANT

AND

DCC PLC

S & L INVESTMENTS LIMITED

JAMES FLAVIN AND

LOTUS GREEN LIMITED

DEFENDANTS/RESPONDENTS

JUDGMENT of MR JUSTICE FENNELLY delivered the 27th day of July, 2007.

Introduction

It used not to be considered any sort of sin to profit financially from the use of secret, private or privileged information. That was

how fortunes were made. Now things are different. To trade on the use of inside information is recognised for what it is. It is a fraud

on the market. The insider who exploits his access to the special knowledge he enjoys for the purposes of the company in his

capacity as executive or director of a company commits a crime. He may be made, additionally, to answer for the profits he has

made.

The Appellant (whom I will call “Fyffes”) has sued to recover the very large profit some of the defendants made from the sale of

shares in Fyffes. It claims that the third named defendant, Mr James Flavin, a very well-known businessman, and Chief Executive of

DCC had access, in his capacity as a director of Fyffes, to inside information, which has come to be called, if inaccurately, price

sensitive information, when he traded a very large volume of Fyffes shares owned by the other defendants in 2000.

The High Court (Laffoy J), in a lengthy and meticulously detailed judgment, dismissed Fyffes’ claim after a hearing lasting 88 days. It

is a tribute to the extraordinary patience and care of the learned trial judge that none of her findings of primary fact are challenged

on this appeal. Nor does either party question her conclusions on several major legal and factual issues.

A major issue at trial was whether Mr Flavin, as distinct from the fourth named defendant had “dealt” in the shares. The learned trial

judge determined that issue against the defendants and there is no appeal. Another was whether the defendants could be made to

account in equity for any profit made from the sales. She determined that issue against Fyffes and there is no appeal.

The only issue on this appeal is the issue which was dispositive of the claim in the High Court, namely whether the information

admittedly available to Mr Flavin at the time of the share sales was in truth price sensitive. The learned trial judge held that it was

not. For that reason, she dismissed Fyffes’ claim.

The learned trial judge was also deeply concerned by what she repeatedly described as a “fundamental incongruity” in the Fyffes

claim. It is that Fyffes was totally aware, being the source and author of it, of the information it now alleges to be price sensitive.

Yet it acted in a manner utterly inconsistent with the holding of any such belief. The learned trial judge did not find it necessary to

draw any conclusions from the incongruity she identified. Fyffes believe, however, that she was mistakenly influenced by this

perception in reaching her decision to dismiss the claim.

The appeal turns on the assessment of the likely price effects of a limited body of comparatively simple facts. In large measure, as I

will be saying later, they are the sort of facts upon which common-sense judgments and opinions can be formed without the input of

an extraordinary degree of expertise. It is difficult to escape the expression that the length of the trial was the product of the large

amounts of money at stake and the depth of the respective corporate pockets rather than of the complexity of the issues.

The statutory provisions are brief. The Court was led upon expeditions into comparative jurisprudence from the United States and the

orient, not always based on similar legislation. There is no Irish case-law. This is the first insider-trading case, apart from one criminal

prosecution, in this jurisdiction.

The claim of Fyffes, so far as relevant to the appeal, is for:

A. a declaration that the sale by the first and second named defendants of in excess of 31 million ordinary shares in the

plaintiff between 3rd February, 2000 and 15th February, 2000 constituted an unlawful dealing within the meaning of Part

V of the Companies Act 1990 (the Act of 1990);

B. an order pursuant to section 109(1)(b) of the Act of 1990 requiring the defendants and each of them to account to

the plaintiff for any profit accruing to the defendants from those sales.

The statutory claim is based on the contention that the sales were unlawful because, the plaintiff alleges, they were effected by Mr.

Flavin at a time when he was in possession of price-sensitive information by reason of his directorship of Fyffes.

The Facts: The Companies and the Shares

The following is an abbreviated version of the basic facts found by the High Court.

Fyffes and DCC are public companies. Their shares are listed on both the London Stock Exchange and the Irish Stock Exchange.

Fyffes is involved in the fresh produce trade. The trade in bananas has long been the core of its business. From its flotation in 1981

to 2003 Neil McCann was its chairman. Other members of the board and management of Fyffes in 1999 and 2000 were: Carl McCann,

who was deputy chairman from 1988 until 2003, when he became chairman; David McCann, who became chief executive in 1995;

Frank Gernon, group finance director; John Ellis, who was based in the United Kingdom. Each of these was an executive director at

the relevant times. The non-executive directors included Mr Flavin and Mr Gerry Scanlan. The company secretary was Philip Halpenny.

Following the example of the learned trial judge, I will refer to the first named defendant as DCC, to the second named defendant as

S&L, to the third defendant as Mr. Flavin and to the fourth named defendant as Lotus Green.

Mr. Flavin has been chief executive of DCC since 1976 and is deputy chairman.

S&L and Lotus Green are Irish registered companies which are part of the DCC Group. S&L is a wholly-owned subsidiary of DCC. Mr.

Flavin at the relevant time was a director of S&L. He has never been a director of Lotus Green. That company established tax

residence in the Netherlands.

The DCC Group’s shareholding in Fyffes dated back to January, 1981, when it acquired a 9.46% stake, which grew to approximately

10.5%. Mr. Flavin joined the board in 1981.

The learned trial judge summarised subsequent events as follows:

“Following the flotation of DCC in May 1994 its holding in the plaintiff, another public company, was perceived as being

anomalous. Its strategy from 1996, and probably from 1995, was to exit from Fyffes when a suitable opportunity arose.

With a view to mitigating its tax liability, in particular, liability for capital gains tax, on a future disposal of the holding, in

August, 1995 DCC and S&L, which held part of the holding, agreed to sell the shares in Fyffes of which they were

respectively the owners to Lotus Green. The agreed purchase price was paid by Lotus Green, but legal title was not

transferred, so that DCC and S&L remained the registered owners of the shares. Just over a fortnight after the

agreement, Lotus Green became resident for tax purposes in the Netherlands.”

Further shares, issued by way of scrip dividend, increased the DCC Group holding in Fyffes, which consisted of both ordinary and

preference shares. It is not necessary, for the purposes of the appeal, to make any distinction between the three corporate entities

in the DCC Group. All issues related to their separate interests in the shares were disposed of by the High Court. DCC can be treated

as the owner of the shares.

All of DCC’s ordinary shares were sold in three tranches – on 3rd, 8th and 14th February 2000 at prices of €3.20, €3.60 and €3.90 (an

average of €3.42) per share. The three sales grossed in excess of €106 million, resulting in the accrual of a substantial profit to the

DCC Group. Mr. Flavin resigned from the board of Fyffes with effect from 9th February, 2000.

The Facts: Figures, Trading, Accounts

Fyffes’ financial year, prior to 2000, ran from 1st November to 31st October. In common with other Irish public companies it

announced its results at half-yearly intervals.

Fyffes reported its preliminary results for the financial year ended 31st October, 1999 on 14th December, 1999. It reported that profit

before tax and exceptional items in that year had increased over the previous financial year by 5.1% to €82.9 million. It also reported

that, while turnover for the period decreased marginally, the total operating profit was €80 million, up 3.8%. In an outlook section,

having recorded that the results for the year had maintained the Group’s record of continuous growth, it was stated that the board

believed that 2000 would be “another year of further growth for Fyffes”. (my emphasis).

On 29th October 1999 a Fyffes board meeting, which Mr Flavin attended, considered the budget for the financial year 2000.

Comparison with the expected or budgeted figures for the year 1999/2000 and later management figures is at the heart of the debate

about price sensitive information. Those figures sufficiently appear from the comparative table displayed later.

The allegedly price sensitive information consists of the management reports for the months of November and December 1999.

The November report stated:

“Losses of €2.6 million are €2.3 million worse than budget, and €4.1 worse than last year, and are due primarily to weaker

Banana prices.”

Commentary on the banana division was as follows:

“The loss for the month was €2.7 million compared to a budget profit of €1.2 million, and last years profit of €2.8 million,

was due to weaker banana pricing in the UK and European markets, low market value for licence, and the weaker

exchange rate of the Euro vis a vis the US Dollar.

Prices for UK multiple customers have been consistently below prior years for many months now, but the

November/December prices are also well below the budget levels. 130,000 boxes of Suriname fruit were diverted to

Portugal to ease the pressure on the market.”

Looking forward to December the November report stated:

“The banana market has improved a little in December although the UK was affected by snow this week. December losses

are still expected to be €1.6 million which would be €1.6 million worse than budget, and €4.9 million worse than last year."

Cumulative losses at end December are expected to be €4.2 million which is €3.9 million worse than budget, and €9 million

worse than last year.”

Some optimism was expressed about January, with mention of possible “improved sentiment.”

The management accounts for December, 1999 were circulated to board members on 25th January, 2000. The introductory

commentary was as follows:

“Sales:

Total sales of €149 million were €11.3 million lower than budget due to the non consolidation of Capespan and €6.1 lower

than last year due to lower UK produce throughput and the disposal of Swithenbanks.

Net Profit:

The pre-goodwill loss for the month is €1.3 million compared to a breakeven budget and last year’s profit of €3.3 million,

bringing the cumulative pre-goodwill loss to €4 million compared to last year’s profit of €4.8 million…….”

The commentary on the banana division was as follows:

“December showed very little improvement in market prices or sentiment with a multiple price war effecting an already

over supplied market. The UK result was very disappointing even after the diversion of fruit to Europe.

Urgent contract renegotiations are required with ACP suppliers to correct the imbalance in the supply/demand curve which

occurs in November/December in the absence of autumnal natural disasters.

The European market has improved quickly in January with prices at budget levels going into February but the UK market

continues to be affected by competitors seeking to increase their multiple market share.”

The December management report concluded with the following commentary:

“Current forecast is for a profit of €1.3 million compared to a budget of €5 million and €6.3 million last year. The shortfall is

mainly due to weaker UK banana pricing.”

The learned trial judge set out a composite table in order to illustrate the information in relation to Fyffes’ pre-tax profit which can be

extrapolated from the November and December trading reports in combination.

(1) (2) (3) (4) (5)

Actual Budget Actual Variance Gernon Forecast

FY 1999 FY 2000 FY 2000 Actual

€’m €’m €’m €’m

(a) (b)

Prior Year Budget

€’m €’m

November 1.5 (0.3) (2.6) (4.1) (2.3)

December 3.3 - (1.3) (4.6) (1.3)

January 6.3 5.0 1.3 (5.0) (3.7)

(3.43)

(2.92)

2.54

Total Q1 11.1 4.7 (2.6) (13.7) (7.3)

(3.81)

Capespan (2.5) (3.0) (2.5)

Capespan (9.8)

Adjustment

It may be noted at once that the learned trial judge accepted that the figures shown or to be deduced from the November and

December trading figures, “taken on their own,” represented “unquestionably bad news about Fyffes’ trading and earnings

performance in the first quarter.”

Before any consideration of the significance of these figures, it is, nonetheless, essential to draw attention to aspects of the market

background.

Worldoffruit.com

In November 1999, Fyffes launched a new venture under the name of worldoffruit.com, described as “an Internet portal for the fresh

produce industry including FruitXchange, a business-to-business (B2B) on-line trading system.” It would “combine B2B on-line trading

with trade news…” It was claimed that “through this dedicated industry website, businesses would have access to on-line trading and

other services which will add value to their operations.”

It was fully accepted by Fyffes at the hearing of the appeal that this worldoffruit.com venture had a very significant effect on the

Fyffes share price in the ensuing months. It became, as the learned trial judge found, the “driver” of the share price. It is,

consequently, unnecessary to describe the venture, its contents, its vigorous promotion or its prospects in greater detail. It is

common case that, even if it had been successful, worldoffruit.com would not have contributed significantly or at all to the earnings

of Fyffes in the short term. The share-price effect is sufficiently demonstrated by a table set out in the judgment. It will be

necessary to bear this table in mind when assessing the likely price effect of the alleged price sensitive information:

December January February March April

1999 2000 2000 2000 2000

Date Price Date Price Date Price Date Price Date Price

€ € € € €

01 1.60 04 1.95 01 3.26 01 3.35 03 2.25

02 1.63 05 1.98 02 3.32 02 3.55 04 2.00

03 1.62 06 1.98 03 3.20 03 3.60 05 1.95

06 1.62 07 1.97 04 3.50 06 3.65 06 1.98

07 1.60 10 2.25 07 3.70 07 3.65 07 1.97

08 1.60 11 2.23 08 3.60 08 3.50 10 2.00

09 1.60 12 2.25 09 3.68 09 3.20 11 2.12

10 1.62 13 2.58 10 3.85 10 3.45 12 2.15

13 1.65 14 2.55 11 3.85 13 3.55 13 2.08

14 1.80 17 2.63 14 3.90 14 3.43 14 2.05

15 1.99 18 2.72 15 3.90 15 3.35 17 1.93

16 2.00 19 2.63 16 3.93 16 3.30 18 1.97

17 2.00 20 2.60 17 3.95 20 2.70 19 1.97

20 2.00 21 2.60 18 3.98 21 2.46 20 2.00

21 1.97 24 2.70 21 3.95 22 2.60 25 1.93

22 1.97 25 2.62 22 3.87 23 2.50 26 1.87

23 1.98 26 2.80 23 3.87 24 2.62 27 1.85

27 2.95 24 3.80 27 2.55 28 1.85

28 3.14 25 3.75 28 2.55

31 3.30 28 3.55 29 2.55

29 3.40 30 2.40

31 2.20

A major element in the case, which remained central to the contentions on the appeal was that, in the early months of 2000,

surrounding the time of the relevant share sales, there was a world-wide phenomenon of fascination with stocks and shares based on

businesses with prospects of prodigious profits from newly emerging internet ventures. The learned trial judge said that this was a

“phenomenon variously described in the evidence by the colloquialisms “dotcom mania”, “irrational euphoria” and “irrational

exuberance.”

Her assessment of the influence of this phenomenon was as follows:

“The overall conclusion I have reached is that it is incontrovertible that there was very considerable enthusiasm for

dotcom stocks or stocks with a dotcom component in January and February, 2000, but whether that enthusiasm merits a

diagnosis of mania is not something on which I need express a view. It is incontrovertible that that enthusiasm affected

Fyffes' share price, in that it is incontrovertible that it was the wof.com venture which was the main driver of Fyffes'

share price at the time. However, in my opinion, the evidence does not establish that the enthusiasm for the internet

component of Fyffes' business at the time was of an order that investors regarded the underlying fundamentals of the

core business to be wholly irrelevant.”

At the hearing of the appeal, a consensus between counsel was that approximately €2 of the share price was attributable to the core

fruit business and the excess related to worldoffruit.com. The table shows that the price rose from below €2 at the start of the year

and that it rose to a peak of€3.98 on 18th February only to drop back to its starting point or below in mid to late April.

Conduct of Fyffes

I have mentioned the concern of the learned trial judge regarding what she perceived as the fundamental incongruity or inconsistency

between Fyffes’ knowledge of the alleged price sensitive information and its failure to take relevant action either vis-à-vis Mr Flavin or

the market.

Fyffes made no announcement to the market following the circulation of either the November or December trading figures. Fyffes

claims, in these proceedings, that Mr Flavin unlawfully profited from the same information in dealing in the shares on behalf of the DCC

Group in early February. The information, Fyffes says, would have been, if released, likely materially to affect price. The Listing Rules

of both the London and Irish Stock Exchanges impose obligations on companies to make disclosure of certain types of information

which would be “likely to lead to substantial movement in the price of listed securities.”

Curiously, however, and perhaps paradoxically both parties agreed at the hearing --- and there was much evidence at trial related to

the point --- that Fyffes behaved properly in not making any announcement. It is easy to see why. The defendants could not accuse

Fyffes of disrespect of the Listing Rules while defending the position that the relevant information was not price sensitive. Fyffes, on

the other hand, were compelled necessarily, however uncomfortably, to justify their own conduct. As a corollary, they needed to

argue that the November/December material was radically different from that which became available and was notified in March,

which I will discuss shortly.

To similar effect, the defendants pointed to personal behaviour of Mr Neil McCann in relation to Mr Flavin, which appears, at least at

first sight, utterly inconsistent with any belief that the latter had behaved unlawfully in selling the shares in early February. This

behaviour extended to the purchase of a celebratory bottle of champagne, upon which the learned trial judge remarked: “That seems

to me to have been a generous and gracious gesture, and I attach no weight to it.” She attached more significance, however, to a

letter of 4th February in which Mr McCann stated to Mr Flavin:

“Further to our meeting last evening it is encouraging to note this morning that the share price has stood up but I think,

in all our interests, it would be helpful if the remainder of the shares are disposed of, so that they will not be overhanging

the market. It is quite an achievement to have disposed of such a volume and get such a good reaction. Hopefully it

augurs well for the balance.”

The letter also discussed the timing of Mr Flavin’s intended resignation from the board of Fyffes. Naturally, the defendants saw this

letter as representing encouragement by Mr. McCann of the second and third sales. Its legal significance, if any, will be considered

later.

The defendants also placed particular reliance on notification to the stock exchange of two share transactions.

On 27th January, 2000 Fyffes announced the grant on 25th January, 2000 of options to Mr. Halpenny, the company secretary, in

relation to 50,000 ordinary shares in Fyffes. The learned trial judge rejected a submission by Fyffes that the grant of options to

employees under the Fyffes’ share option scheme was not captured by Part V of the Companies Act, 1990. She held that as secretary

of the company, Mr. Halpenny came within the definition of officer in Part V and, as such, was a connected person within the meaning

of s. 108(1) of the Act. That ruling is not contested.

On 28th January, 2000, Fyffes announced that on 26th January, 2000 Mr. Ellis, an executive director of Fyffes, had sold 45,000

ordinary shares in Fyffes. The learned trial judge held: “From an outside perspective, the defendants were correct in suggesting that

the impression that announcement created was that Mr. Ellis did not have price-sensitive information on 26th January, 2000.”

These latter transactions, combined with the fact that no announcement had been made under the Listing Rules, played a material

part in the judge’s conclusion that the information held by Mr Flavin was not price sensitive.

The Profit Warning: 20th March

A second major issue of contention concerned the effects of an announcement made by the company to the market on 20th March

concerning its unfavourable trading experience in the early months of the financial year 1999/2000 and its admitted significant effect

on the share price. In brief terms, the learned trial judge found that it was the poor trading results and not incipient disillusion with

dotcom stocks that affected the share price on that day. Nonetheless, she decided to exclude this evidence in its entirety from her

consideration of the effect of the alleged price sensitive information. She did so because she found that there was a lack of parity

between the two sets of information and, because, she considered the market conditions for Fyffes to be very different on 20th

March from the dates of the three share sales in early February. Fyffes criticise this ruling and I will consider the issue. First, I will

refer to the facts.

On 20th March 2000 at the AGM of Fyffes, Mr Neil McCann, as Chairman, made a statement which was notified contemporaneously to

the stock exchange (the “March 2000 Trading Statement”), which was described by the learned trial judge as being, in effect, “a

profit warning.” The announcement was headed “Fresh produce trading” and stated as follows:

“The trading environment in the early part of the current financial year has been difficult. In particular, market conditions

in the last two months of calendar 1999 were significantly below expectations. The usual recovery in the first month of

calendar 2000 has been slower than anticipated, particularly because of the continuing weakness of the Euro against the

dollar. As a result, we expect that the performance for the first half of the year, on a like for like basis, will be below that

achieved during the same period last year. Present trading is slightly improved but, at this stage, it is too early to predict

whether the shortfall can be recovered in the second half.

Despite the exceptional market conditions so far this year, we remain confident about the future prospects of the fresh

produce sector and of the Fyffes business in particular. The Group’s strategy remains the active pursuit of further

opportunities for consolidation in our industry.”

Mr David McCann, Chief Executive of Fyffes reported at the AGM on the progress of worldoffruit.com. He stated that up to €20 million

was “to be expensed as incurred, will be invested during 2000………….” He stated that the Fyffes Group was working closely with its

financial advisers to determine the optimal ownership and funding structures that would maximise shareholder value from its traditional

fresh produce business and from its new e-commerce division. He also stated that €100 million was earmarked for new e-commerce

businesses.

The “profit warning” reached the market on 20th March. The learned trial judge made a clear finding regarding its effect on the price

of Fyffes shares. They closed on the London Stock Exchange at €3.16 on 17th March. The Irish Stock Exchange was closed on that

day. Following the profit warning in the March 2000 Trading Statement they closed on 20th March at €2.70. There was evidence from

the defendants’ experts to the effect that the revelations in relation to trading performance made on 20th March, 2000 were not the

cause of the share price decline on that day. The learned trial judge found that at 12.15 p.m. on 20th March, Regulatory News

Service reported that Fyffes had issued a profit warning (the March 2000 Trading Statement) and almost immediately there was a

significant decline in Fyffes’ share price of approximately 15% on the London Stock Exchange. She concluded that there was

“coercive evidence that it was the March 2000 Trading Statement, and not the contemporaneous announcements in relation to the

wof.com venture or the vagaries of rumours in relation to a possible merger, that provoked the adverse reaction.” Further information

later that afternoon, first rumours of a possible merger between Fyffes and Dole, and later a statement by Fyffes playing down talk of

the rumoured Dole deal, failed to generate a reaction.

At trial the issue became whether the March 2000 Trading statement constituted, what came to be called “a valid proxy” for the

November/December trading reports. The learned trial judge found that it did not. I will need to revert to that question, which is

clearly of great potential importance.

The High Court Judgment

Three parts of the High Court judgment are of particular importance. They correspond with the three central grounds of appeal

advanced by Fyffes:

That the learned trial judge was inconsistent in failing to draw the obvious conclusions from her own findings of fact: in

effect that the identified information was price sensitive;

That she applied an inappropriate legal test based in the notion of the reasonable investor;

That she denied any weight to the fall in the share price on 20th March 2000, following the communication of a profit

warning regarding the first quarter of 1999/2000.

The first relevant part of the judgment contains the learned trial judge’s initial assessment of the November/December trading figures.

The second contains her ultimate conclusion, including reference to the reasonable investor, that those figures did not represent price

sensitive information. The third closely related section of her judgment contains her assessment of the March 2000 Trading

Statement. It is logical to set out these aspects of the judgment in that order.

High Court Assessment of November/December figures

The learned trial judge commented as follows on the November/December figures:

“Being management accounts, they were authoritative. As regards quality and provenance, the information contained in

them differed from the general information about trading difficulties available in the market. They were intended for

internal use only and, as such, were confidential, which emphasises their sensitivity, although, in my view, not in any

technical sense. They contained detailed and precise information: precise figures for actual losses in the months of

November and December, 1999 and a forecast for January, which on the evidence, was made some time after the middle

of January, and which would be assumed to be very accurate. From the figures given for deviation from prior year and

budget, by a simple arithmetic calculation, it was possible to deduce what profits were achieved in each of the three

months of the first quarter of financial year 1999 and the profits for which Fyffes had budgeted for each of the three

months in the first quarter of the financial year 2000. The profit level which had to be achieved, as it was put, the hill

which had to be climbed, in the last three quarters of financial year 2000 even to reach the profit achieved in the

financial year 1999 was obvious: it was at least €85.5m, being the aggregate of the loss of €2.6m in the first quarter of

financial year 2000, which had to be made up, together with the actual figure achieved in financial year 1999, €82.9m.”

She referred and appeared to accept the Fyffes contention:

“…the market knew from the Preliminary Announcement that Fyffes’ expectations for financial year 2000 exceeded what

was achieved in the previous year: in the outlook statement the market had been told of Fyffes’ belief that the year 2000

would be “another year of further growth” for Fyffes. Therefore, a reasonable deduction was that the hill to be climbed to

meet budget in the final three quarters was higher than €85.5m.”

She continued in the same vein:

“…at the time, analysts’ profit forecasts for financial year 2000 were in the range of €88m to €92m, from which a

reasonable deduction could be made that the hill to be climbed to match those forecasts was very high indeed, between

€90.6m and €94.6m.”

She encapsulated her own judgment on the November/December figures in a passage to which Fyffes attach the greatest importance.

Fyffes complains that she failed to draw the appropriate conclusions from her own findings as here expressed. That passage reads:

“The information contained in the November and December Trading Reports was unquestionably bad news about Fyffes’

trading and earnings performance in the first quarter. Taken on their own, a reasonable inference could be drawn from the

figures that there was a real risk that Fyffes’ own expectations and analysts’ expectations for the first half and for the full

year would not be met. It is true that the market did not know what Fyffes’ expectations for the first half were, although

it did know what was achieved in the first half of the previous year. Nonetheless, given the size of the loss in the first

quarter against the modest budget for that quarter, that there was a risk was a reasonable deduction. The absence of

interpretation or a narrative explanation from management did not hinder the arithmetical exercise which enabled

deductions to be made in relation to the likely adverse impact of the first quarter results on the half year or the full year

and, to that extent, was immaterial. That the fresh produce business, including the banana business, was, and was

known in the market to be, seasonal with the first quarter being the least profitable quarter, was also immaterial, to the

extent that seasonality was factored into the prior year figures and the budget figures contained in the documents and

was also built into analysts’ forecasts. The plaintiff suggested that a reader of the information would discern a worsening

negative trend. The figures in the above table for variances from prior year and budget bear this out, in that they were

getting larger in absolute terms, but whether this should be regarded of any significance, in my view, is doubtful. In

general, the narrative gave very little inkling that matters would improve, except the statement that the European market

had improved quickly in January with prices at budget level going into February. Conversely, there was nothing in the

narrative to indicate that matters were getting worse.”

The learned trial judge immediately qualified the views expressed in that passage as follows:

“On the basis of the foregoing, the court was invited to conclude that intuitively the information appeared on its face to

price-sensitive. The statutory test does not warrant drawing any conclusion in relation to the information contained in the

trading reports standing alone. The furthest the court can go is that the information was of a type and quality that was

potentially price-sensitive. Whether it was or was not, falls to be judged by applying the principles which I have

adumbrated earlier.”

At a later point, she accepted:

“Although there was some controversy on this point, on the evidence, I consider that investors would have been

concerned about the outcome, not only for Fyffes’ full financial year, but also the outcome for the first half.”

The learned trial judge did not, at that point in her judgment, propound any test for reaching the required statutory conclusion. She

developed, at a later point, a standard of the “reasonable investor” to which it will be necessary to return. It is more helpful, at this

stage, to quote the judge’s ultimate conclusion that the information was not price sensitive together with her reasons for it. It was as

follows:

“The key question is whether, as a matter of probability, on 3rd February, 2000, the reasonable investor, having assessed

the negative news about Fyffes’ performance in the first quarter in the context of the total mix of information available

about Fyffes’ prospects, would have concluded that the information indicated a lowering of expectations about Fyffes’

earnings in the first half of financial year 2000 and in the full year of an order of magnitude that would probably impact on

Fyffes’ share price to a substantial or significant degree. I think the answer to that question is that he would not. I think

he would conclude that it was too early in the financial year to make a judgment about the outcome for Fyffes’ existing

business in the first half. That conclusion is consistent with what Fyffes told the disgruntled shareholder in the letter of

11th May, 2000, which I have quoted earlier. But more importantly, I think he would have concluded that prospects for a

merger or a major acquisition, and the potential of the wof.com venture, which were the main focus of the interest of the

market at the time, would offset the impact of the current trading problems. In particular, the strength of the sentiment

for Fyffes’ wof.com venture at the time, as evidenced by what was happening to the share price, was such that he would

have concluded that an adverse share price reaction was not likely. Moreover, he had no reason to suppose that Fyffes’

own expectations about future earnings had changed. On the contrary, the announcements in relation to the sale of Mr.

Ellis’s shares and the grant of options to Mr. Halpenny suggested they had not, as did the fact that no announcement had

been made under the Listing Rules.”

High Court view of March 20th Trading Statement

As already noted, the learned trial judge found that there was coercive evidence that the March 20th profit warning was the cause of

the sharp drop in Fyffes share price which occurred on that day. She rejected evidence from the defendants to the effect that the

reason for the fall was that sentiment for internet stocks had soured earlier in the month. She said that “the information released

contemporaneously in relation to wof.com had no bearing on the share price movement.” She found as follows:

“On the totality of the evidence, it is quite clear that it was the bad news in relation to earnings which caused the share

price decline on 20th and 21st March, 2000.”

Nonetheless, she declined to accept that that event was of any value in establishing the hypothetical proposition that the

November/December trading figures would, if generally available, have had a material effect on the price. She based her conclusion on

two comparisons: firstly, between the contents of the November/December information and the March 2000 Trading Statement;

secondly, between the market conditions prevailing on 3rd, 8th and 14th February (the dates of the share sales) and on 20th March.

She held:

“…what happened on and after 20th March, 2000 is not a matter which is of evidential value in applying the hypothesis

posited in s. 108(1).”

Her entire approach was based on her acceptance of the defendants’ proposition “that a post-market event is only of evidential value

for the purpose of resolving the statutory hypothesis if two factors are present: that there is parity of information between the

alleged price-sensitive information and the information which precipitated the putative proxy share price effect; and that the market

conditions were the same on the date the hypothesis is being applied and the date of the putative proxy event.”

In making the first of the comparisons envisaged by that proposition, she thought it important not to lose sight of the ultimate

objective of the plaintiff in these proceedings, which is to procure an account of profits under that provision. She thus thought it

relevant to examine “whether the evidence to date suggests that, assuming for the purposes of the argument that there would have

been an adverse share price effect if the information contained in the November and December Trading Reports was generally

available at the dates of the Share Sales, that share price effect is measurable in such a manner that one would have confidence in

the fairness of the outcome.” Thus she found it necessary to consider “the prospect of quantifying the share price effect, if the

information alleged to be price-sensitive had been generally available at the date of the Share Sales.”

In respect of the first of these comparisons, the learned trial judge decided that:

“…the information contained in the March 2000 Trading Statement was fundamentally different because, by that

information, Fyffes was formally telling the market that it had revised downwards its expectations in relation to the first

half and was uncertain whether it would meet its target for the full year.”

In respect of the second, she found that market conditions were not the same as they had been on the earlier dates. This was based

on the fact that o 3rd February Fyffes’ share price was rising, that it continued to rise up to and beyond 14th February, but that by

20th March the share price had come off its peak on 18th February and was in decline.

High Court: the Reasonable Investor

As already noted, the learned trial judge, in her determining and crucial passage, decided that the reasonable investor would not have

concluded that the November/December trading information indicated a lowering of expectations about Fyffes’ earnings in the

judgment. The learned trial judge had formulated a test for determining the price sensitivity of the relevant information based on the

postulate of a reasonable investor. This was proposed by the defendants on the basis of case law from the United States.

The court, the learned trial judge held, in assessing the significance of the relevant information and its hypothetical impact on the

share price, “does so from the perspective of the reasonable investor making an investment decision, that is to say, a decision to

buy, sell or hold the shares in the company.” The test is objective and the “question to be addressed is whether it is probable that

the impact of the information would have a substantial or significant effect on the share price.” She drew attention to the words,

“would be likely” in section 108(1).

She then restated that test as follows:

“…the hypothetical test is whether on 3rd February, 2000, 8th February, 2000 and 14th February, 2000, had it been

available to him, the information contained in the November and December Trading Reports, viewed by him against the

“total mix” of information about Fyffes’ trading and earnings available on those dates, would have impacted on the

judgment of the reasonable investor in relation to an investment decision about Fyffes’ shares to the extent that he

would have concluded that the information probably would have a substantial effect on Fyffes’ shares.”

The Appeal

Three principal grounds of appeal are put forward on behalf of Fyffes. It will avoid unnecessary repetition if I reserve more detailed

consideration of these grounds, of the respondents’ replies to them and of the facts to the point in this judgment when I come to

express my own views.

Fyffes’ accept all findings of primary fact without exception. Accepting the facts as found, they challenge the High Court judgment on

three grounds, which I paraphrase as follows:

1. That the learned trial judge failed to draw the correct inferences from the facts as found; in particular, her conclusions

in rejecting the Fyffes allegation of price sensitivity were inconsistent with her own earlier findings;

2. That the learned trial judge adopted an incorrect and inappropriate “reasonable investor” test; moreover, she failed

correctly to apply the test as so adopted;

3. The learned trial judge erroneously failed to pay any regard whatever to the March 20th Trading Statement.

The respondents fully support the findings and conclusions of the learned trial judge. They argue strongly that they were based on

the assessment of evidence, including very extensive expert evidence, heard and carefully analysed by the learned trial judge.

The respondents also defend the grounds upon which the learned trial judge expressed concern about what she called the

“fundamental incongruity” in the fact that Fyffes acted inconsistently with the proposition that information of which they were fully

aware was price sensitive. They lay particular emphasis on the fact that Fyffes did not even consider that the information was such

as to require an announcement to the stock exchange. They suggest that the learned trial judge might have gone further than she

did.

Applicable Legal Principles

The sole issue on the appeal is whether the November/December trading report constituted price sensitive information. The question

is whether it came within the scope of section 108 of the Companies Act 1990 (the Act of 1990), since repealed, which provided:

“It shall not be lawful for a person who is, or at any time in the preceding 6 months has been, connected with a company

to deal in any securities of that company if by reason of his so being, or having been, connected with that company he is

in possession of information that is not generally available, but, if it were, would be likely materially to affect the price of

those securities.”

There is no appeal from the determination of the learned trial judge that Mr Flavin both caused and procured the dealing which

resulted in the share sales. He was connected with Fyffes. The learned trial judge held that for the purposes of affording an effective

remedy to Fyffes under section 109 of the Act of 1990, the three corporate defendants should be treated as a single entity.

Part V of that Act implements the State’s obligations under Council Directive 89/592/EEC of 13 November 1989 coordinating

regulations on insider dealing.

The purpose of the Directive is to provide assurance to investors that they are placed on equal footing and that they will be

protected against improper use of inside information, thus obviating the undermining of confidence in, and the smooth running of, the

market. Member States were required to prohibit the misuse of inside information by those involved in the management of companies.

They were not required to provide the type of civil remedy contained in Part V, though they were permitted to adopt more stringent

measures than those required. The Directive contains the following definition of inside information:

“... 'inside information' shall mean information which has not been made public of a precise nature relating to one or

several issuers of transferable securities or to one or several transferable securities, which, if it were made public, would

be likely to have a significant effect on the price of the transferable security or securities in question.”

The Act does not use a definition of “inside information.” It is obvious, however, that section 108 must be interpreted in the light of

the Directive and so as to be in conformity with it—the principle of conforming interpretation. It is clear—indeed it was common case-

-- that the expression, “likely materially to affect the price of those securities,” in the section must have the same meaning as “likely

to have a significant effect on the price of…” securities in the Directive. The term, in fact used throughout the trial and the appeal

was “price sensitive information.”

Where it is established that a connected person uses information to which he has access by virtue of his position in the company to

deal in shares, that dealing is unlawful only if, assuing it were to be generally available, that information would be likely materially to

affect the price of those. A court in applying that provision must address a hypothetical question. In fact, the information has not

been available. Thus its likely effect has to be established by evaluation of the quality of the information in relation to the company

and the state of the market.

The parties were agreed that the statutory hypothesis does not envisage the information in issue finding its way into the market in

any particular manner.

The learned trial judge adopted an objective test for the purpose of arriving at the statutory hypothesis. As already indicated, she

used the notion of a reasonable investor to enable her to answer the hypothetical question.

The section contains no reference to a reasonable investor. It appears to envisage a simple question: what is likely to have been the

effect of the availability of the information on the price of the shares? The learned trial judge, following a caeeful review of authorities

form a number of jurisdictions, decided to look at the matter “from the perspective of the reasonable investor making an investment

decision, that is to say, a decision to buy, sell or hold the shares in the company.” She was influenced in coming to this view by the

similarity of the legislative purpose which underlies federal security law in the United States in particular.

She identified as a “seminal decision” that of the United States Court of Appeals, Second Circuit in SEC v. Texas Gulf Sulphur

Company 401 F.2d 833 [1968]. That case concerned alleged violations of the provisions of s. 10(b) of the Securities Exchange Act of

1934 and Rule 10(b)-5 promulgated pursuant to it. Rule 10(b)-5 provided that it should be unlawful for any person, inter alia, “to

make any untrue statement of a material fact or to omit to state a material fact necessary in order to make the statements made, in

the light of the circumstances under which they were made, not misleading” in connection with the purchase or sale of any security.

(emphasis added). Although the legislative purpose of that provision may have much in common with section 108(1), it is structured

differently. In this and other cases, the courts of the United States were required to commence with a threshold determination of the

“materiality” of certain facts. The court explained that the test of materiality must necessarily be a conservative one, since many

actions are “brought with the benefit of hindsight.” The court cited earlier authority to the effect that: “The basic test of materiality

is whether a reasonable man would attach importance…in determining his choice of action in the transaction in question.” (page 849).

Marshall J, writing for the Supreme Court of the United States in TSC Industries Inc. v. Northway Inc. 426 U.S. 438. stated:

“The question of materiality, it is universally agreed, is an objective one, involving the significance of an omitted or

misrepresented fact to a reasonable investor. Variations in the formulation of a general test of materiality occur in the

articulation of just how significant a fact must be or, put another way, how certain it must be that the fact would affect

a reasonable investor’s judgment.”

Here it can be seen that it was considered necessary to postulate the reasonable investor to test materiality. The statutory provision

was different.

It is possible to trawl through these and many other US authorities and to find useful pronouncements on the effects of information

on markets. However, I have come to the conclusion that this body is more likely to confuse than to illuminate. In particular, there is

no warrant and, more importantly, no need to insert the reasonable investor into the section. The section lays down a single test.

That is whether the information alleged to be price sensitive would, if it were generally available, be likely materially to affect the

price of those securities. The price is the price on the market. The market is comprised of many investors, wise and unwise,

reasonable and unreasonable. At the time relevant to the present proceedings, it was generally agreed that some parts at least of

the markets for shares and securities world-wide were in the grip of what was described by some as “dot.com mania.” Mr Cush, Senior

Counsel, for the respondents argued at the hearing that fund managers and other major market operators necessarily had to take

account of this phenomenon and to include dot.com stocks in their portfolios. That is undoubtedly true. Nonetheless, the problem

remains. Can one apply the statutory test by reference to a reasonable investor? I do not think so. The court is required to make a

judgment as to whether, on the statutory assumption, the information in question would have had a material effect on price.

It should be added that the test ultimately adopted by the learned trial judge was a modified version of the reasonable investor. She

said that the “key question” was “whether, as a matter of probability…… the reasonable investor………would have concluded that the

information indicated a lowering of expectations about Fyffes’ earnings in the first half of financial year 2000 and in the full year of an

order of magnitude that would probably impact on Fyffes’ share price to a substantial or significant degree.” Thus she used the

reasonable investor not as a representative of all investors in the market whose response to the information might or might not lead

to a material effect on the share price, but rather as a test of opinion as to how the market would respond. The respondents did not

try to defend this latter formulation. Rather they treated it as a sort of aberration from the true reasonable investor test which the

learned trial judge had formulated earlier in her judgment.

In the result, I believe that the learned trial judge was mistaken in superimposing a test based on the reasonable investor on the

single statutory test: would the price have been materially affected.

Conclusions of the High Court on the Issue of Price Sensitivity

I have already expressed the view that the learned trial judge was mistaken in adopting the criterion of the reasonable investor for

the purpose of assessing the statutory hypothesis. Thus, her approach was flawed. She should have asked herself the simpler

question, namely whether the November/December trading figures, if the information contained in them had been generally available,

would have been “likely materially to affect the price” of the shares.

With this important qualification in mind, I turn to consider the way in which the learned trial judge assessed the evidence. It is also

necessary to decide whether she was right to exclude from consideration the market effects of the March 2000 Trading Statement.

The passage of the High Court judgment, which describes the November/December trading report as “unquestionably bad news about

Fyffes’ trading and earnings performance in the first quarter,” represented the major plank of Fyffes’s case on the appeal. She

qualified this expression by the phrase “taken on their own.” She added that the “statutory test does not warrant drawing any

conclusion in relation to the information contained in the trading reports standing alone.” The furthest she could go was to say “that

the information was of a type and quality that was potentially price-sensitive.”

Fyffes start from the position that the information was “intuitively” price sensitive. Intuition could not, of course, be a sound basis for

reaching a legal conclusion on such an important issue. At the hearing, “common sense” was accepted as an appropriate substitute.

In truth what Fyffes are saying is that it is that the November/December figures were so bad that they would obviously have had an

adverse effect on the market. There is much force in that view. At the risk of being repetitious, it is worth identifying what the

learned trial judge herself identified as the “bad news.”

To begin with, she had already characterised the November/December management figures as authoritative, noting that: “As regards

quality and provenance, the information contained in them differed from the general information about trading difficulties available in

the market.” She described the information in them as “detailed and precise.”

She considered that “a reasonable inference could be drawn from the figures that there was a real risk that Fyffes’ own expectations

and analysts’ expectations for the first half and for the full year would not be met.” Here the market would be learning, firstly, that

the two months’ figures fell significantly short of Fyffes’ own expectations of which the market had not, prior to then, been aware.

Those disappointed expectations would have related not only to the first half but to the full year. It will be recalled that the first

reason given by the learned trial judge for ultimately ruling against the price sensitivity of the November/December figures was that it

was “too early” for the reasonable investor to form an opinion.

She also pointed to “the size of the loss in the first quarter against the modest budget for that quarter…”

She considered the “absence of interpretation or a narrative explanation from management” to be immaterial, since an “arithmetical

exercise [would have] enabled deductions to be made in relation to the likely adverse impact of the first quarter results on the half

year or the full year…”

She made the important observation that the fact that “the fresh produce business, including the banana business, was, and was

known in the market to be, seasonal with the first quarter being the least profitable quarter, was also immaterial, to the extent that

seasonality was factored into the prior year figures and the budget figures contained in the documents and was also built into

analysts’ forecasts.”

She accepted Fyffes’ suggestion that a reader of the table (set out in her judgment and earlier in this judgment) “would discern a

worsening negative trend” from the “figures e for variances from prior year and budget bear this out, in that they were getting larger

in absolute terms……”

She analysed in some detail what was referred to as “the hill which had to be climbed in the last three quarters of financial year 2000

even to reach the profit achieved in the financial year 1999.” That hill was very steep. She calculated that, in the last three quarters

a profit of “at least €85.5m” would have been necessary. That was made up of: the loss of €2.6m in the first quarter of financial year

2000 which had to be made up, together with the actual figure achieved in financial year1999, €82.9m.” All of this was accentuated if

increased profit was to be assumed for the year 1999/2000. She referred to the 14th December prediction of another “another year of

further growth” and, as already mentioned summarised current analysts forecasts of much higher figures.

It is unnecessary to cite further detail. The basic point is that the figures represented “unquestionably bad news.”

The learned trial judge was, of course, right to emphasise the danger of taking such figures on their own. It is necessary to consider

the context. The context is the market in Fyffes share at the relevant time and the market’s expectations in respect of those shares.

It was observed at the hearing that good results may lead to a fall in the share price; conversely, poor figures may lead to a rise. All

depends on expectations. In the first case, the market may have been expecting better figures than the good ones published; in the

other, the figures will not have been as bad as had been feared.

Having made that important qualification, however, one would look for something to counter the normal expectation that bad news

about a company’s earnings and prospects will case a fall in the share price. Other things being equal, the November/December

figures were such as to be likely to disappoint the market. There was nothing current that other things were not equal. There was no

built-in expectation of bad news. Expectations were to the contrary.

Thus, I find the approach of the learned trial judge surprisingly cautious. To my mind, the November/December figures were such as

to cast serious doubt on the prospects of the company, not only for the first quarter, but also, as the learned trial judge herself

observed, for the first half and even for the full year.

The learned trial judge gave three reasons for ruling that the information was not price sensitive. Before expressing my views on those

reasons, it is appropriate to form a view about the exclusion of the March 2000 Trading Statement.

The learned trial judge accepted as being correct in principle submissions made by the defendants that there were “two prerequisites

to a post-market event being of evidential value in applying the price-sensitivity hypothesis are that:

(a) the information alleged to be price-sensitive should be substantially the same as the information which gave rise to

the share price movement which is proffered as a proxy, and

(b) the market conditions on the date at which the hypothesis is being applied are identical with market conditions on the

date on which the supposed proxy event occurred.”

These propositions were based on a number of American authorities. I find them extraordinarily rigid. The learned judge herself rightly

observed that this was “an exercise akin to the use of comparators for valuation purposes in litigation in a wide range of contexts.”

Yet no such fixed set of rules is applied in such cases. Courts hear evidence of valuers offering suggested comparisons. The

properties differ in size, nature and location. The court adjusts for all relevant factors.

I do not think the US cases support such an approach. Principal among them was that of SEC. v. Bausch & Lomb Inc. 565 F. 2d8

[1977]. There the court criticized what it perceived as SEC’s “facile inference” based on fall of stock on one day that all information in

possession of an individual was material. The court identified a number of specific convincing matters of fact to demonstrate the

unreliability of the particular evidence. The second case cited by the defendants was Elkind v. Liggett & Myers Inc. 635 F2d 156

[1980], a decision of United States Court of Appeal, Second Circuit, a case concerned with the quantification of damages. The

Attorney General emphasised that the need for parity of information arose in that context. Mansfield J discussed the point at page

170 of the report as follows:

“The ‘value’ of the stock traded during the period of non-disclosure of the tipped information (i.e. the price at which the

market would have valued the stock if there had been a disclosure) is hypothetical. Expert testimony regarding that

‘value’ may, as the district court found in the present case, be entirely speculative. This has led some courts to conclude

that the drop in price of the stock after actual disclosure and after allowing a period of time to elapse for the market to

absorb the news may sometimes approximate the drop which would have occurred earlier had the tip been disclosed. …

The court below adopted this approach of using post-public disclosure market price as nunc pro tunc evidence of the

‘value’ of the stock during the period of non-disclosure.

Whatever may be the reasonableness of the nunc pro tunc ‘value’ method of calculating damages in other contexts, it has

serious vulnerabilities here. It rests on the fundamental assumptions (1) that the tipped information is substantially the

same as that later disclosed publicly, and (2) that one can determine how the market would have reacted to the public

release of the tipped information at an earlier time by its reaction to that information at a later, proximate time. This

theory depends on the parity of the ‘tip’ and the ‘disclosure’. When they differ, the basis of the damage calculation

evaporates.”

The need for “parity” there discussed arose from the need to have a reliable basis for calculating damages. It is, of course, dangerous

to draw inferences from any piece of evidence relating to events on different dates. The court must be astute to see that the

evidence is genuinely of value. The American cases do not, as I see it, establish the general propositions advanced by the defendants

and opted by the learned trial judge. A simpler approach was adopted by an English court in Chase Manhattan Equities v. Goodman

[1991] B.C.L.C. 897. There the shares had been suspended for more than five months before the relevant information had its market

effect. Applying a test remarkably similar to that in section 108(1), Knox J (at page 931 of the report) said:

“I am satisfied that all three of the matters listed above would if generally known have been likely materially to affect the

price of the company’s shares. The proof of that pudding is in the eating in that when the suspension which followed

almost immediately, was lifted the price of the company’s shares was very sharply down.””

The question then is whether the learned trial judge was correct, on the facts, to exclude the March 2000 Trading Statement from

her consideration. I do not think that she was. Both that statement and the November/December trading figures have been set out in

full earlier in this judgment. There is remarkable similarity between their contents. Insofar as they differ, the earlier figures are more

detailed. The learned trial judge described them as authoritative, detailed and precise. The March 2000 Trading Statement is

expressed in general terms. I am satisfied that they conveyed broadly similar information about the trading experience of Fyffes in the

first quarter. The time interval between them was some six weeks. Thus, the later information may be said to be more reliable, but

that is all.

Nor do I see that there was any such enormous change in market conditions as would justify excluding the effect of the March 2000

Trading Statement entirely from consideration. Fyffes submit that it provides real and palpable proof of the market reaction to such

information. Its release led to an immediate fall of some 15% in the share price. Nobody has suggested that such a price movement

was other than material. The trial judge was influenced by the fact that, in early February the share price was rising, whereas, by

20th March it was in decline. Fyffes point out, however, that the price on the two dates was approximately the same, with a peak in

the interval at €3.95.

In my view, the events of March 20th were a useful pointer to the likely market effect of the release of trading information of as

substantially kind, if it had occurred on the various dates of the share sales in early February.

Since the learned trial judge had decided, as a matter of principle, to exclude the March 20th events from her consideration, she did

not have the benefit of using it in her ultimate assessment, to which I now turn.

I have already remarked upon the view of the trial judge that the reasonable investor “would conclude that it was too early in the

financial year to make a judgment about the outcome for Fyffes’ existing business in the first half.” This seems to be in clear

contradiction of her earlier assessment, under the heading of “bad news,” to which I have referred. Mr Cush made a distinction in

argument between “risk” and “probability:” in the earlier passage the judge was referring to a “real risk,” whereas, at the end of the

day she had to consider whether an effect on the share price was “likely.” The latter proposition is, of course, correct. However,

news, good or bad, about a company may be likely to affect the share price of a company even though the disclosure relates only to

a risk. It is a question of balancing the importance of the event against the degree of probability or risk of its occurring.

The second reason was that the reasonable investor “would have concluded that prospects for a merger or a major acquisition, and

the potential of the wof.com venture, which were the main focus of the interest of the market at the time, would offset the impact of

the current trading problems.”

The notion of “offset was the subject of much argument at the hearing.” The Attorney General argued on behalf of Fyffes that it was

inadmissible in principle. To the extent that they existed at all, both of the elements mentioned were taken into account in the share

price prior to the hypothetical information becoming available on the dates of the share sales. The evidence was that the share price

started out at just below €2 at the start of 2000. It then climbed to a peak of some €3.95 in mid-February. At least fifty per cent of

the price related to the core business. The judge found that “the evidence does not establish that the enthusiasm for the internet

component of Fyffes' business at the time was of an order that investors regarded the underlying fundamentals of the core business

to be wholly irrelevant.”

The Attorney General also submitted that there was no new information at all in relation to mergers.

It is clear, on the evidence that Fyffes’ core business of fresh produce, especially bananas continued to represent a major component

of its share price. It follows that bad news of the sort contained in the November/December trading report would be likely to affect

the market in the share adversely. There is no doubt that, as the judge correctly found, the worldoffruit.com venture had become the

driver of the share price in January and February 2000. It does not follow that the trading report would not have had an effect on the

market. The very notion of offset supposes that there something to “offset.” It suggests a balancing of one influence against

another. If the effect of the worldoffruit.com venture was to counteract or to cancel the effect of the “bad news,” it would still have

materially affected the price. The statutory test does not postulate a fall in price.

The third reason given by the judge for concluding that the information was not price sensitive was that Fyffes had made

announcements regarding the share transactions involving Mr Ellis and Mr Halpenny and that the company itself had made no

announcement. Fyffes does not contest the proposition that it was possible to infer from these events that the company did not

have price sensitive information which it had not released. However, as was submitted by the Attorney General, it did not follow that

these facts would have confused the hypothetical reasonable investor postulated by the judge. The existence of such facts could

not, in principle undermine the statutory test. To do so would be to undermine the effect of the statutory provision. The statutory

hypothesis would be automatically negatived in every case where the company itself had made no announcement. It must not be

forgotten that the focus of Part V and of the Directive is on insider dealing. Insiders cannot be allowed to use inside information

merely because—perhaps especially because---the company does not itself disclose it.

A court, asking itself whether a piece of information would have had a material effect on prices on one date, is entitled to have

regard, for reasons I have given, to the effect of the release of broadly similar information on another date not too far removed in

time. In my view, the events of March 20th provide compelling evidence that the November/trading reports, if generally available on

the dates of the share sales, have materially affected the share price.

The ultimate conclusion of the learned trial judge was necessarily affected by what I have described as two errors: she should have

adopted a straightforward test of market effect, without the interposition of the reasonable investor; she should not have excluded

the effect of the release March 2000 Trading Statement. Were it not for these two aspects of her judgment, it seems very likely that

her own compelling findings of fact would have led her to infer that the price would have been materially affected.

On this appeal, I would find as a fact that the information contained in the November/December trading figures would, if it had been

generally available on the dates of the share sales, have been likely materially to affect the price of Fyffes shares.

Fundamental Incongruity

The foregoing conclusion is not affected by the circumstances considered by the learned trial judge to give rise to a fundamental

incongruity. Those facts have been set out earlier in this judgment. Perhaps the most striking of all the facts, though one rightly

discounted as being devoid of legal weight, was the purchase of a celebratory bottle of champagne. That fact coupled with the

ensuing letter written by the Chairman sits ill with the allegation of illegality which is the basis of the present claim.

However, the only matter which could conceivably have carried legal weight was the fact that the company made no announcement

to the market pursuant to the Listing Rules. Rule 9.2, which had only been introduced in the January, 2000, obliged companies to

communicate to the market “without delay of all relevant information which is not public knowledge…” concerning specified matters,

including “the company’s expectation as to its performance…” which “if made public, would be likely to lead to substantial

movement…” in the share price.

This topic occupied a great deal of time in the High Court. In the end, it seems to be without legal consequence. Neither party

contended that Fyffes was in breach of Rule 9.2. The defendants abandoned any claim based on estoppel.

One can only conclude that the “fundamental incongruity,” if there is one, is just that. It may arise from the remedy provided by the

legislature to the company itself. The Directive does not require Member States to provide a civil remedy to a company against those

trading in its shares in breach of section 108(1). The incongruity would not exist if the plaintiff were another party to a share

transaction, not in possession of the relevant information, and suing by virtue of section 109(1)(a) for loss sustained by him.

Part V provides a statutory remedy, which is not defeated by incongruity.