COURT FILE NO.: 06-CV-307599CP
DATE: 20230213
ONTARIO
SUPERIOR COURT OF JUSTICE
BETWEEN:
DENNIS FISCHER, SHEILA SNYDER, LAWRENCE DYKUN, RAY SHUGAR and WAYNE DZEOBA
Applicants
– and –
IG INVESTMENT MANAGEMENT LTD., CI MUTUAL FUNDS INC., FRANKLIN TEMPLETON INVESTMENTS CORP., AGF FUNDS INC. and AIC LIMITED
Respondents
Joel P. Rochon, Peter R. Jervis, Ron Podolny, Matthew W. Taylor, Sarah Fiddes, for the plaintiffs
James D.G. Douglas, David Di Paolo, Graham Splawski, Julia Webster, Brianne Taylor, for the defendant AIC Limited
David D. Conklin, Tamryn Jacobson, Caitlin Woodford, for the defendant CI Mutual Funds Inc.
HEARD: February 8 - 11, 14 - 17, 22, 23, 25, 28, March 1- 4, 7 – 11, June 13 – 15, 2022.
KOEHNEN J.
REASONS FOR JUDGMENT
OVERVIEW
[1] In September 2003, the Attorney General of New York announced the results of an investigation into what was referred to as “market timing” trading in American mutual funds. That investigation led to a series of settlements requiring the offending funds to pay billions of dollars.
[2] Shortly after the results of that investigation were announced, the Ontario Securities Commission (the “OSC”) launched its own investigation into similar practices in Ontario. It resulted in five fund managers, including both defendants, entering into settlement agreements with the OSC pursuant to which they paid over $200 million to their respective funds.
[3] This class action was commenced against the five funds that settled with the OSC. Three of those funds have since settled with the plaintiffs. As a result, the trial proceeded only against AIC Limited, now known as AIC Global Holdings Inc. (“AIC”) and CI Mutual Funds Inc., now known as CI Investments Inc. (“CI”). Both are sizeable mutual funds in the Canadian market. At the end of the Class Period AIC had approximately $12 billion in assets under management. CI had approximately $5 billion in assets under management.
[4] The plaintiff Wayne Dzeoba is the representative plaintiff for the unitholders who invested in AIC funds. The plaintiff Sheila Snyder is the representative plaintiff for unitholders who invested in CI funds.
[5] The Class Period for AIC runs from January 1, 1999 to September 30, 2003. The Class Period for CI runs from September 1, 1998 to September 30, 2003. On the facts as I have found them, nothing turns on the slightly different starting points for the two Class Periods.
[6] Although many of the materials leading up to trial, including the certification order, refer to “market timing,” the real conduct that is at issue is a practice by some mutual funds to allow certain sophisticated investors to move in and out of their funds very quickly. That practice is more accurately known as frequent trading, short-term trading or switching. As is explained later in these reasons, that practice can dilute the returns available to long-term unitholders.
[7] This action arises out of conduct by which AIC and CI allowed certain large, sophisticated investors to engage in frequent trading in their funds. The plaintiffs allege that this frequent trading diluted the returns of long-term investors in the funds. The precise nature of the conduct that is the subject of the trial is disputed between the parties and will be defined later in these reasons.
[8] This trial has been referred to as the liability trial because it addresses the four certified common issues, all of which relate to liability. Any damages are to be assessed in a separate proceeding if liability is found.
[9] The application of the common issues is limited to trading activity in certain funds of each defendant which are set out in the certification order.
[10] For the reasons set out below I have found that both defendants breached their duty of care to prevent “market timing” in their funds. The defendants admit they owed a duty of care to prevent harm to their funds and to the unitholders of their funds as a whole. While the defendants deny that they breached any standard of care arising out of that duty, I view the evidence differently. There was ample evidence before me to demonstrate that the standard of care during the Class Period required the defendants to be aware of the dangers of frequent trading in and out of their funds and take reasonable steps to prevent it. The harm that frequent trading causes to long-term unitholders has been known for decades. Mutual fund prospectuses (including those of AIC and CI) warned that frequent trading caused harm to funds and could result in fees of up to 2% being charged to frequent traders. It was generally agreed that a 2% fee would have stopped switching or frequent trading immediately. Despite the contents of their prospectuses, the defendants not only failed to take steps to prevent frequent trading or charge the fees set out in their prospectuses when it occurred, they facilitated frequent trading by entering into “Switch Agreements” which allowed certain investors to switch in and out of funds for a fee of only 0.2%. “Switching” in the context of this case refers to a practice whereby market timers invest funds in one of the defendants’ money market funds, transfer or “switch” some or all of those funds into an equity fund and then switch the investment back into the money market fund, all within the course of a few days.
[11] The defendants focussed on the fact that they were not aware, during the Class Period, of a practice known as “time zone arbitrage”. The specific practice of time zone arbitrage is what the “market timers” engaged in here. While time zone arbitrage was not as well known during the Class Period as frequent trading, time zone arbitrage is simply a type of frequent trading. Although the defendants may not have known that the “market timers” were engaged in time zone arbitrage, they did know they were engaged in frequent trading. It is the frequent trading that causes harm, not the specific form of frequent trading that the “market timers” conducted here. Had the defendants taken steps to prevent or prohibit frequent trading, they would have prevented time zone arbitrage as well.
[12] The defendants also agreed that they owed fiduciary duties to act honestly, in good faith and in the best interests of the funds. I do not find that the defendants breached those fiduciary duties. A breach of fiduciary duty requires a subjective element of breach of loyalty or good faith. While I find that the defendants were negligent, I do not find that their negligence rises to a breach of honesty or good faith. As a result of my findings, I direct this action to proceed to a trial to determine damages.
I. The Scope of This Proceeding
[13] The first issue to address is the specific scope of this proceeding. Although both sides agree that the underlying issue is whether the defendants owed a duty of care or fiduciary duty to prevent “market timing,” the parties do not agree on the meaning of “market timing” for purposes of this trial.
[14] To frame the issue, let me start with three over-simplified examples of what has been referred to as market timing in this proceeding and the source of the harm at issue.
A. The Source of the Harm
[15] The first example of market timing is a simple strategy of trying to time the market by buying low and selling high. If the movement out of a fund occurs over a longer period, say more than 90 days after the investment, this sort of market timing causes no harm to longer-term investors.
[16] The second and third examples of market timing are examples of arbitrage opportunities. An arbitrage opportunity is one which seeks to extract profit out of an inefficiency in the market.
[17] The second example refers to a situation that arose in the early days of mutual funds. In those days, if an investor bought mutual fund units on Tuesday, the price the investor paid was the Net Asset Value (“NAV”)[^1] per unit that was calculated at the close of trading on the previous day (i.e., Monday at 4 PM). This differs from the pricing of individual securities which are priced on a real time basis throughout the trading day based on the price of the most recent trade.
[18] The early pricing methodology for mutual funds created arbitrage opportunities for astute investors. Take an oversimplified example of a mutual fund that tracked the TSX index. Assume it had a NAV of $100 as of 4 PM on Monday. Assume also that the fund had 10 investors, each of which held 1 unit which, as of 4 PM Monday, had a unit value of $10.00. Next, assume the index rose by 10% on Tuesday. When the fund re-calculated it’s NAV per unit at 4 PM on Tuesday, the new unit value would be $11.
[19] Now assume that an astute investor, having seen the market rise by 10% on Tuesday, bought a unit on Tuesday at 3:45 PM. He would have paid $10 per unit, being the NAV that had been calculated on Monday at 4 PM. Given that the investor purchased at the end of the day Tuesday, its $10 could not be invested into any securities but would remain in the mutual fund in cash. On Tuesday at 4 PM the new NAV of the fund would be $120 ($100 + 10% increase in the index + $10 the frequent trader invested). This now translates into a unit value of $10.90 ($120 divided by 11 unitholders). On Wednesday morning the investor could sell its unit for $10.90 (the price calculated at 4 PM on Tuesday) thereby making a 9% risk-free profit. The difference in the NAV per unit of $11 without the frequent trader and the $10.90 with the frequent trader is referred to as dilution vis-à-vis the long-term unitholders.
[20] This practice was thought to be unfair to longer-term unitholders and was made impractical in 1968 by a change to the rules of the United States Securities and Exchange Commission (the “SEC”). The rule change required that the price at which a mutual fund unit is bought or sold be determined using the NAV calculated at the end of the day on which the purchase or sale occurred. Canadian exchanges followed suit. As a result of this change, a frequent trader who invested on Tuesday at 3:45 PM would be charged $11 per unit, being the NAV calculated at 4 PM on Tuesday. This change eliminated the arbitrage opportunity and prevented dilution to long-term unitholders.
[21] The third example of “market timing” shows that arbitrage opportunities remained with respect to mutual funds that were exclusively devoted to or heavily weighted towards foreign shares that traded on stock exchanges in a different time zone.
[22] In this third example let us assume that a Canadian mutual fund reflects the Tokyo stock exchange index commonly known as the Nikkei index. Tokyo is 11 hours ahead of Toronto. The trading day on Monday in Tokyo closes at 1 AM Monday morning EST. Again, assume the NAV at the end of Monday’s trading in Tokyo is $100, divided between 10 unitholders with one unit each for a NAV per unit of $10.
[23] By the 1980s, world markets had become increasingly correlated to each other which meant that if North American markets rose on Monday it was increasingly likely that the Tokyo market would also rise when it opened on Tuesday morning in Japan.[^2] Japanese markets could also be positively influenced by business news that emerged during North America’s trading day on Monday which news would not be reflected in Japanese stocks until the Tokyo exchange opened on Tuesday morning Japanese time.
[24] For this third example, let us assume that the North American markets rose by 10% on Monday. That could be expected to result in an increase in the Nikkei index on Tuesday. For purposes of our example, we will assume that the correlation between the Nikkei index and North American markets is perfect. As a result, a 10% increase in North American markets will result in a 10% increase in the Nikkei index. In those circumstances, the frequent trader would buy a unit of the Nikkei index mutual fund for $10 shortly before the end of trading on Monday in Toronto. Although the precise value of that unit will technically not be determined until 4 PM on Monday in Toronto, as a practical matter, the unit value is already known because it is based on the closing price of the Tokyo exchange which closed at 1 AM Monday morning EST.
[25] The Nikkei index would follow suit on Tuesday with a 10% increase. By the end of the day on Tuesday, the Nikkei index fund would have a NAV of $120 ($100 from Monday’s NAV plus 10% market increase plus $10 of the frequent trader’s money).
[26] When markets open in Toronto on Tuesday, the frequent trader would sell its units. Once again, although the precise value of that unit will technically not be known until 4 PM on Tuesday in Toronto, as a practical matter the unit value is already locked in because it will be based on the closing price of the Tokyo exchange which closed at 1 AM Tuesday EST. As with the earlier example, the NAV per unit would be $10.90 for each of the 11 unitholders with the frequent trader’s investment instead of $11 per unit for each of 10 unitholders without the frequent trader’s investment.
[27] This third example of market timing is referred to as time zone arbitrage.
[28] While the above examples are oversimplified because, among other things, they assume a perfect correlation between North American markets and the Nikkei index, they provide a directional illustration of how frequent traders can earn profit at the expense of long-term unitholders. Even though the correlation between North American markets and the Nikkei index will never be perfect, a short-term trader who traded with sufficient frequency and in sufficiently large amounts could potentially earn significant profit by employing a time zone arbitrage strategy. The strategy would work not only with Japanese stocks but also with Asian and European stocks more generally.
[29] What is also clear from the foregoing examples is that the profit that the frequent trader earns and the harm of dilution to longer-term unitholders is caused not by the concept of “market timing” as such but by the short-term or frequent nature of the “market timer’s” investment. In the second and third examples set out above, the profit arises from the fact that the “market timer” pulls his money out of the fund before it can actually be invested into any securities. It is that very short-term duration of the investment that causes the harm. Had the “market timer” taken an approach whereby he invested in a fund for, let us say 60 days and then pulled his investment out, it is highly unlikely that any dilution would have occurred to long-term unitholders because the funds would have been invested in securities. That said, even a short-term investment of 60 days can cause harm to the fund in that it could cause fund managers to buy and sell shares more frequently, or hold more cash than they otherwise may, all in order to fund the switch outs of the frequent trader.
[30] Given that the real harm here is caused by short-term or frequent trading rather than market timing per se, I will refer to the issue in the balance of these reasons as short-term trading, frequent trading, or frequent short-term trading.
[31] A short note on terminology before proceeding. I have so far referred to the defendants as fund managers. That they are. A fund manager is the overall corporate entity that manages a family of funds. The parties referred to the manager of an individual mutual fund as a portfolio manager. I will adopt that distinction in these reasons.
[32] Returning now to the scope of this action, two issues arise: (a) whether it includes only time zone arbitrage or includes other forms of short-term trading; and (b) whether it is restricted to the activity of the market timers identified by the OSC or whether it includes other market timers.
B. Is the Action Limited to Time Zone Arbitrage
[33] The scenarios above provide certain examples of dilution caused by frequent trading. Frequent trading can potentially cause harm in other scenarios as well. For example, if the short-term trader switched money into a domestic mutual fund and withdrew it before the fund manager could invest it in securities, similar dilution of long-term unitholders’ interests would occur if the NAV of the mutual fund increased during the time the frequent trader’s money was in the fund.
[34] The plaintiffs submit that this action is not limited to time zone arbitrage but includes all forms of frequent trading because “market timing” is not defined in the certification order and because the statement of claim refers to market timing broadly and without limitation. The plaintiffs propose the following definition of market timing for this action:
Market timing is: (1) short-term trading, (2) executed in a frequent or excessive manner repeatedly in those funds, (3) which extracts profits from the funds.
[35] In my view, the recovery of damages in this action is limited to harm caused by time zone arbitrage.
[36] The scope of an action is defined by the pleadings. On my reading, the Amended Statement of Claim complains about time zone arbitrage, not other forms of frequent trading.
[37] By way of example, paragraph 25 of the Amended Statement of Claim refers to the fact that NAVs were calculated once a day as of 4 PM. Paragraph 26 refers to quick turnaround trades taking advantage of “inefficiencies in the way the funds set their NAVs.” Pausing there, the form of frequent trading in which one invests in a domestic fund and pulls out before the portfolio manager invests the funds in securities does not amount to an efficiency in the way the funds set their NAV but involves another form of frequent trading that could potentially cause dilution.
[38] Paragraph 27 of the Amended Statement of Claim then goes on to say:
One means by which these structural inefficiencies can be exploited is by employing a strategy known as “timing” or “time zone arbitrage” of mutual funds. Timing is an investment technique involving short-term “in and out” trading of mutual fund shares which capitalize on the fact that some funds use “stale” prices to calculate the value of securities held in their portfolios, prices which do not necessarily reflect the “fair value” of such securities as of the time the NAV is calculated. [Emphasis added.]
Significantly, this paragraph defines timing and time zone arbitrage as being one and the same.
[39] Paragraphs 28 through 30 describe time zone arbitrage with reference to Asian markets in more detail.
[40] The only means referred to in the Amended Statement of Claim of exploiting “structural inefficiencies” is time zone arbitrage.
[41] The Amended Statement of Claim notes that CI was the fund manager for “over 100 mutual funds” including a specific list of funds. Only funds that invested in foreign markets are identified. Of these funds, 24 were included in the certification order.
[42] On my reading of the Amended Statement of Claim, it equates market timing with time zone arbitrage. There is no other form of market timing described in the Amended Claim.
[43] The plaintiffs’ expert evidence at trial was similarly focused on time zone arbitrage.
[44] Samuel London was called by the plaintiffs as an expert on the mutual fund industry in Canada and in particular on duties that mutual fund managers owed to investors. He defined a market timer as:
An institution or individual that engages in Frequent Short-Term Trading Activity using specific strategies intended to generate profits by exploiting the fact that (during the Class Period), European, Asian, International and Global equity mutual funds included “Stale Value” securities when the net asset value (“NAV”) of such equity mutual funds were priced at the end of the trading day in North America. Market Timers are also sometimes referred to as arbitrageurs.
[45] Mr. London was asked about the definition during cross-examination and explained that:
[T]his definition is intended to be as consistent as possible with the action at hand, so yes, that's why I've defined it that way. It's frequent trading intending to make profits by using those strategies. So yes, that's -- that’s in my definition.
[46] Professor Zitzewitz was also called as an expert for the plaintiffs. Under the heading “What Is Market Timing,” his 2019 report states:
In the context of this case, market timing means the trading of shares of open-ended mutual funds at high frequency. As I discuss in more detail in my 2003 paper, attached hereto as Appendix E, during the time period of the case, the motivation for this trading was typically inefficiencies in the valuation of mutual fund shares.
For the international and global funds that are the subject of this case, the valuation inefficiency is stale pricing, arising primarily from time zone differences.
The market timing of mutual funds is usually motivated by an inefficiency in the pricing of mutual fund shares known as stale pricing. In the international and global funds that are the subject of this case, stale pricing arises primarily from time zone differences.
[47] Professor Zitzewitz then goes on in paragraph 9 to explain how NAV is determined and how time zone differences can be used to take advantage of the way in which the NAV is set.
[48] Professor Zitzewitz also explained that academics tend to use the term “stale pricing”
rather than market timing because it is more precise.
[49] Professor Zitzewitz’s evidence dealt solely with foreign funds. His evidence was not about short-term trading generally but was about taking advantage of pricing inefficiencies in North American mutual funds by way of time zone arbitrage.
[50] Any finding of a breach of a duty of care or fiduciary duty must be based on the statement of claim. The practice put at issue in the statement of claim is time zone arbitrage. In my view it would be unfair to the defendants to extend the scope of the inquiry 19 years after the end of the Class Period and after the conclusion of the liability trial when other forms of frequent trading were not put at issue in the statement of claim.
[51] Although the damages arising out of the claim will be limited to those caused by time zone arbitrage, the duty of care and standard of care will be assessed in light of duties and standards vis-à-vis frequent trading as a whole. This is because, as I find later in these reasons, the standard of care during the Class Period called on mutual fund managers to prevent and prohibit frequent trading. Had the defendants prohibited frequent trading in their funds, time zone arbitrage would have been prohibited as well.
C. Is the Action Limited to Market Timers Identified by the OSC?
[52] The plaintiffs submit that they should have the right to include in the damages trial the conduct of frequent traders in addition to those identified by the OSC.
[53] The defendants submit that there was no such provision in the certification order and that my jurisdiction is limited to the issues as defined in that order. They further submit that the certification order excludes from the plaintiff class only those market timers who were identified in the OSC settlement agreements and therefore limits this trial and any ensuing damages trial to the conduct of those limited market timers. To enlarge the conduct beyond the market timers identified by the OSC 17 years after the claim was issued and after the trial of the action would, the defendants say, be unfairly prejudicial.
[54] The definition of the class has been a contentious issue from the outset of these proceedings. The focus of that contention has been on who is excluded from the class. Those who are excluded are the market timers. It is therefore the excluded market timers whose activities become subject to the damages assessment.
[55] The plaintiffs initially proposed that the class be defined to include all investors who purchased or redeemed units except any individual or entity that, to the defendants’ knowledge, engaged in market timing activities in the defendants’ funds. When the defendants objected to this definition, the plaintiffs proposed a revised definition which excluded only those market timers that the OSC identified in the settlement agreement with each defendant. The certification order ultimately adopted that definition. At the same time, however, the plaintiffs asked for the right to identify additional market timers as the action proceeded.[^3] In his reasons on the certification motion, Perell J. held that this was inappropriate and that the class definition should expressly specify that there would be no other exclusion of investors from the class going forward and thus implicitly, no expansion in the number of market timers in respect of whom the defendants could be found liable.[^4]
[56] Perell J. ultimately decided not to certify the proceeding as a class action because he was of the view that the OSC proceeding was the preferable procedure by which to resolve the issues in dispute. The plaintiffs appealed. The Divisional Court reversed and certified the proceeding as a class action.[^5] The Ontario Court of Appeal[^6] and the Supreme Court of Canada[^7] upheld the Divisional Court’s decision to certify.
[57] The Divisional Court saw no error in Perell J.’s refusal to build into the class definition the possibility that it would be expanded in the future.[^8] At the same time, the Divisional Court also stated:
However, I do not see the need to build into the definition a proviso that no further exclusions will be permitted. There is no basis for concluding that the only market timers who can ever be excluded are those already identified by the OSC, provided that the same definition of market timer is accepted throughout. It is possible, for example, that the OSC missed some market timers who fit the already established criteria. There is no logical basis for continuing to include those traders as part of the class. As long as the definition of those excluded is settled as part of the definition of the class, there is no logical reason to restrict further amendment of the class as the action evolves. That unduly ties the hands of a future motion judge if grounds emerge which either of the parties believe warrant amendment to the class definition. There is substantial precedent for maintaining flexibility in these situations, and forbidding future amendment of the class, regardless of the reason, does not comply with those general principles.[^9]
[58] This suggests that it may be possible for the plaintiffs to expand the number of the market timers on a proper motion to do so. The defendants note that no such motion has been brought.
[59] A number of issues arise about why no such motion was brought. The defendants point out that the plaintiffs tried to seek production of data that would identify additional market timers, but that Perell J. refused production of the data as being irrelevant to the common issues trial.[^10] The defendants also note that the plaintiffs filed an appeal from that decision but then abandoned the appeal. While the identification of additional market timers may have been irrelevant to the common issues trial, it is not necessarily irrelevant to the damages trial.
[60] The plaintiffs note that they have been caught in a Catch-22 situation where they cannot bring a motion without evidence but that the defendants have refused to produce market data that the plaintiffs’ experts say would identify additional market timers. The plaintiffs also note that additional evidence has emerged which suggests that there are additional market timers who have caused damage to long-term unitholders.
[61] By way of example, Joint Document Book tab 181 is an internal email from Shameena Khan (the Manager of AIC’s Dealer Relations Group during the Class Period) to Miles Radoja (AIC’s VP of Sales). Ms. Khan’s email responds to an email that Mr. Radoja had originally sent which identified several accounts that “switch excessively”. Mr. Radoja then asked whether Ms. Khan can call her contacts at AGF and AIM to find out how they are administering switch fees. Ms. Khan responds, saying:
Thanks Miles. I will find out how this is being done. However we have to take into consideration that we have made agreements (verbal or otherwise) with the switchers to allow this type of activity at the PM level. My understanding is that Neil/ MLC[^11] had given the Ok to allow the daily switching of funds, therefore we need to understand the basis of these agreements before applying a switch fee. Should we apply a fee I am sure will result in withdrawal of the assets (sic). I do know that AGF, CI, Fidelity and AIM did not allow the 'switchers' to do business with them and as such can apply a fee.
[62] Mr. Radoja replies, saying:
Only a few have agreements. These other switchers have no agreements.
[63] The email is significant because the traders in AIC funds that the OSC identified had written switch agreements. In that context, Mr. Radoja’s email means that there are frequent traders who did not have written agreements and who were not identified by the OSC.
[64] There is also a parallel class action in Quebec: Ravary v. CI Mutual Funds Inc. and AIC Global Holdings Inc. On May 11, 2022, the Quebec Superior Court released reasons in that action which broadened the number of market timers with respect to whom the defendants could be found liable by revising the description of class members that were excluded because of frequent trading.[^12]
[65] Originally, the Quebec class action had given the plaintiff access to accounts that traded more than $10,000 in and out of a fund within five days. The OSC investigation had set the threshold for market timing at $50,000 traded in and out within five days. The Quebec plaintiff successfully amended the scope of the Quebec action to refer to $10,000 being traded in and out within a 14 day period.
[66] The amendment came about after the plaintiff received production from the defendants which showed 120 particularly active accounts.[^13] The data from those additional accounts demonstrated that, in the case of AIC, although 96.2% of transactions had a holding period of five days or less and 100% of thirteen days or less, the 5 day trades represented only 78.3% of the cumulative returns in the accounts analyzed.[^14] In the case of CI, although 96.5% of transactions had a holding period of five days or less, they represented only 24% of the cumulative returns of transactions in the accounts analyzed.[^15]
[67] In expanding the scope of the trades at issue from five days to 14 days, the court took note of the principles that the Quebec Court of Appeal had established on issues like this as follows:
As for the Framework Judgment, the Court of Appeal mentions that it is not a final judgment, but rather “an interlocutory judgment which can be modified when the circumstances justify it”. Since such a judgment arises from the "highly evolving" context in which it was rendered, the analysis of the framework "lends itself poorly to rigid decisions" and "requires […] a flexible and open approach to allow the parties to present their point of view and to prove the basis of their allegations. It points out that the Framework Judgment was rendered on the basis of an expert report in which “the expert states that he did not have access to the financial data.”[^16]
[68] In these circumstances I am not prepared to hold that the plaintiffs have no right to bring a motion to expand the identity of market timers beyond those identified by the OSC. That is not to say that I am widening the identity of the relevant market timers but merely that the plaintiffs may bring a motion before me to do so on a proper record. I understand as well that any such motion may be preceded by a request for broader documentary production. I make no finding here on either of those two requests but merely hold that the plaintiffs are not precluded from bringing such motions. Given my finding that the conduct that gives rise to potential damages here is time zone arbitrage, any motions will be limited to identifying data about time zone arbitrageurs as well.
[69] Returning then to the matters at issue in this liability trial, the four common issues that were certified for trial are:
(i) Did the Defendants owe a fiduciary duty to the respective Class Members to take steps to prevent “market timing” activities in their funds?
(ii) If so, did the Defendants, or any of them, breach such a fiduciary duty and, if so, what was the nature of the breach?
(iii) Did the Defendants owe a duty of care to the respective Class Members to take steps to prevent “market timing” activities in their funds?
(iv) If so, did the Defendants, or any of them, breach such a duty of care and, if so, what was the nature of the breach?
[70] Since the bulk of the analysis below deals with issues of negligence, I will address the last two common issues first and leave the issues of fiduciary duty until later in these reasons.
II. Were the Defendants Negligent?
A. General Principles of Negligence
[71] To demonstrate negligence by the defendants, the plaintiffs must establish three core elements: First, that the defendants owed the plaintiffs a duty of care. Second, that the defendants breached the standard of care that arises out of the duty of care. Third, that the loss suffered by the plaintiffs was a foreseeable consequence of the breach of the standard of care.
i. Duty of Care
[72] It has long been the law that one must take reasonable care to avoid acts or omissions that one can reasonably foresee would be likely to injure your neighbour.[^17]
[73] A duty of care arises “where the circumstances of time, place, and person would create in the mind of a reasonable man in those circumstances such a probability of harm resulting in other persons as to require him to take care to avert that probable result.”[^18]
[74] The defendants do not contest that they owed a duty of care to their funds, that is to say to the unitholders as a whole.
ii. Standard of Care
[75] The general standard of care that the defendants owe the plaintiffs derives from three sources, all of which reflect the same general principle.
[76] First, the common law standard of care requires that a person exercise the degree of care, diligence, and skill that a reasonably prudent person would exercise in similar circumstances.
[77] Second, Section 116 (b) of the Ontario Securities Act[^19] provides that:
Every investment fund manager,
(b) shall exercise the degree of care, diligence and skill that a reasonably prudent person would exercise in the circumstances.
[78] Although no cause of action is available for breach of a statutory duty unless the statute provides a specific remedy, the statutory duty can inform the common law standard of care.
[79] Third, the defendants’ constating documents acknowledge their standard of care in language almost identical to the common law and Securities Act formulations.
[80] CI’s Code of Conduct during the Class Period stated:
A.4 STANDARD OF CARE
All employees, officers and directors of CI have a duty to CI's investors to act honestly, in good faith and in their best interests and to exercise the degree of care, diligence and skill that a reasonably prudent person would exercise in the circumstances. This standard of care extends to each area of CI's business operations. [Emphasis added.]
[81] AIC’s Code of Conduct was similar and stated:
- STANDARDS
2.1 Obligations to Clients: It is our policy to continue to maintain the highest standards of service to our clients. AIC has a fiduciary duty to its clients to act honestly, in good faith and in the best interests of our mutual funds and their investors and to exercise the degree of care, diligence and skill that a reasonably prudent manager would exercise in the circumstances. This standard of care extends to the service provided by all employees, officers and directors of AIC in each facet of our business operations. [Emphasis added.]
[82] The Declarations of Trust and Management Agreements created for each of the funds within both defendants contain directionally similar standards of care.
[83] The general principles underlying the duty to act with the degree of care, diligence, and skill that a reasonably prudent person would exercise in similar circumstances are not in dispute. The standard is that of a reasonably prudent mutual fund manager. The standard is not one of perfection. It does not require the defendants to have been omniscient.[^20] Adherence to general practice generally equates to acting with the care, diligence, and skill of a reasonably prudent person in the circumstances.[^21] The standard also entails a duty to take reasonable steps to prevent harm.
[84] A defendant will be expected to exercise the standard of care that prevailed at the time of the events in question. It is important not to view events with the benefit of hindsight and not to apply standards of care that may have developed after the events in question or perhaps even as a result of the learning acquired from the events in question. The court must not, as one earlier case put it, look at a 1947 accident with 1954 spectacles.[^22]
[85] Where the standard of care has evolved, the court must look at the facts with that in mind. In cases of torts occurring over a longer period of time, it may be that different standards of care applied at different times depending on the state of knowledge over the course of the relevant period. In that scenario, for example, it might be that the defendants’ conduct breached the prevailing standard of care at certain times during the Class Period but not at others.[^23]
[86] The defendants lay particular emphasis on this factor. They say they were unaware of the time zone arbitrage strategy during the Class Period and became aware of it only after the Attorney General of New York announced the results of his investigation in September 2003.
[87] A final contextual factor is that the defendants held themselves out as possessing particular expertise. The funds at issue were designed for retail investors who lacked investment expertise and who lacked knowledge of the potential harms that could face their investments. The defendants held themselves out as having such knowledge. In its closing submissions, AIC described the benefit of a mutual fund as giving investors “access to a skilled professional portfolio manager” whose “investment expertise was a valuable commodity.”
[88] CI’s counsel agreed in opening submissions that for a mutual fund manager to knowingly permit time zone arbitrage would amount to a breach of a duty of care. The issue for this trial though is when the defendants knew or ought to have known that permitting the conduct of the market timers to occur amounted to a breach of the standard of care.
[89] The defendants say they did not breach the standard of care because the proper standard of care during the Class Period did not contemplate time zone arbitrage.
iii. Foreseeability
[90] The defendants submit that the plaintiffs’ damages were not reasonably foreseeable consequences of the defendants conduct. The defendants tie the lack of foreseeability of damage to their lack of knowledge about time zone arbitrage.
[91] The defendants rely heavily on the decision of the Supreme Court of Canada in Rankin (Rankin's Garage & Sales) v. J.J.[^24] In that case, the Court dealt with a claim against a garage that had left one of its customers’ cars outside, unlocked, with the keys in the car. One night, two unlicensed teenagers took the car and drove it along a highway. An accident occurred in which the passenger suffered catastrophic injuries. The passenger claimed negligence against the garage arguing that it should have foreseen the possibility of teenagers taking the car and getting into an accident. In dismissing the claim, the majority of the Court stated:
While the risk of theft was reasonably foreseeable, the evidence did not establish that it was foreseeable that someone could be injured by the stolen vehicle. Here, there was no evidence to support the inference that the stolen vehicle might be operated in an unsafe manner, causing injury.[^25] When considering the security of the automobiles stored at the garage, there was no reason upon this record for someone in the position of the defendant garage owner to foresee the risk of injury.
[92] The defendants extract from this the principle that, if the harm that arose is a subset of a broader form of harm that could arise, it is the specific form of harm that actually arose that must be foreseeable. Applying that principle to this case, the defendants say that even if time zone arbitrage is a subset of frequent trading, it is the harm from time zone arbitrage, not just frequent trading, that must be foreseeable. Given that they were unaware of time zone arbitrage, loss arising from it was not foreseeable as a result of which they should not be held liable.
[93] I do not read Rankin that broadly. In leading up to its conclusion, the court made the following comments:
[21] … Reasonable foreseeability of harm and proximity operate as crucial limiting principles in the law of negligence. They ensure that liability will only be found when the defendant ought reasonably to have contemplated the type of harm the plaintiff suffered.
[22] The rationale underlying this approach is self-evident. It would simply not be just to impose liability in cases where there was no reason for defendants to have contemplated that their conduct could result in the harm complained of. …
[24] When determining whether reasonable foreseeability is established, the proper question to ask is whether the plaintiff has “offer[ed] facts to persuade the court that the risk of the type of damage that occurred was reasonably foreseeable to the class of plaintiff that was damaged”: A. M. Linden and B. Feldthusen, Canadian Tort Law (10th ed. 2015), at p. 322. This approach ensures that the inquiry considers both the defendant who committed the act as well as the plaintiff, whose harm allegedly makes the act wrongful. As Professor Weinrib notes, the duty of care analysis is a search for the connection between the wrong and the injury suffered by the plaintiff: p. 150; see also Anns, at pp. 751-52; Childs, at para. 25. [Emphasis added.]
[25] The facts of this case highlight the importance of framing the question of whether harm is foreseeable with sufficient analytical rigour to connect the failure to take care to the type of harm caused to persons in the plaintiff’s situation. Here, the claim is brought by an individual who was physically injured following the theft of the car from Rankin’s Garage. The foreseeability question must therefore be framed in a way that links the impugned act (leaving the vehicle unsecured) to the harm suffered by the plaintiff (physical injury). [Emphasis added.]
[94] Other cases have made it clear that the defendant need not foresee the very specific harm that the plaintiff suffered if the defendant ought to have known of more general harm that could not have been brushed aside as “far-fetched.”[^26]
[95] By way of example, in Hacopian-Armen Estate v. Mahmoud,[^27] the Ontario Court of Appeal dealt with a medical negligence claim by the estate and dependents of a woman who died of uLMS, a rare form of uterine cancer. The physician argued that the harm could not reasonably have been foreseen because the cancer was so rare. In dismissing the physician’s appeal, the Court of Appeal stated:
The appellant’s proposed foreseeability analysis is flawed because in focusing on the presence of uLMS, he inappropriately narrows the scope of the risk that he ought to have foreseen. The appellant was not required to foresee the presence of uLMS or the “precise concatenation of events”: R. v. Coté et al., 1974 CanLII 31 (SCC), [1976] 1 S.C.R. 595, at p. 604. It is sufficient that “the harm suffered must be of a kind, type or class that was reasonably foreseeable as a result of the defendant’s negligence”: Frazer v. Haukioja, 2010 ONCA 249, 101 O.R. (3d) 528, at para. 51. In failing to conduct a test that would have detected the presence of cancers of the “same class” or character as uLMS, including uLMS, it was foreseeable that uLMS or other malignancies would go undetected, with consequent injury to Ms. Hacopian-Armen.[^28] [Emphasis added.]
[96] Almost all of the defendants’ witnesses stressed that they were not aware of time zone arbitrage. The concept of reasonable foreseeability, however, involves an objective, not subjective, test. The question is not whether the specific defendant actually foresaw the harm but whether someone in the defendant’s position ought to have foreseen the harm,[^29] and not the specific harm that occurred but the general class of harm. [^30]
[97] Applying these principles to the present case, as in Hacopian, the defendants here were not required to foresee the “precise concactenation” of time zone arbitrage. They were required to foresee the harm that frequent short-term trading could cause to long-term unitholders. To use the words of Rankin, the record before me establishes a link between the impugned act (allowing frequent short-term trading) and the dilutive harm suffered by the plaintiffs.
[98] It is frequent short-term trading that causes dilution, not the particular strategy of time zone arbitrage. Had the market timers engaged in the same frequent short-term trading in the pursuit of a strategy other than time zone arbitrage, the same harm would have resulted. As noted earlier, frequent short-term trading can result in dilution even in domestic mutual funds if the frequent traders’ money is withdrawn before the portfolio manager can invest it in securities. As set out later in these reasons, the plaintiffs have introduced considerable evidence to demonstrate that dilution was a reasonably foreseeable consequence of the frequent short-term trading that the defendants permitted. On the record before me there is sufficient evidence that links the impugned act of frequent short-term trading to potential harm to the class to warrant a damages trial.
[99] One final question to determine when assessing the negligence of a defendant is for the court to ask itself whether there are any residual policy considerations which would negate a duty of care.[^31] The defendants concede that, if the Court finds proximity and foreseeability, there are no residual policy considerations that would negate a duty of care in this case given the narrow role the Supreme Court has ascribed to residual policy considerations.[^32]
B. General Evidence of Foreseeability and Standard of Care
[100] Both sides introduced considerable evidence about foreseeability and the standard of care from the industry as a whole. That evidence can be grouped into the following categories:
(i) General market knowledge of dilution.
(ii) Prospectus extracts from other mutual funds.
(iii) The defendants’ public disclosures.
(iv) The OSC investigation.
(v) The OSC settlement.
(vi) The Stromberg report.
(vii) Academic papers.
(viii) Fair value methodology.
(ix) American Case Law
[101] I turn now to assess the evidence of foreseeability and standard of care under the category headings set out above.
i. General Market Knowledge of Dilution
[102] Professor Lawrence Harris was called by the plaintiffs and was admitted as an expert qualified to testify about, among other things, the impact of excessive short-term trading in mutual funds.
[103] According to Professor Harris, the concepts of stale pricing and dilution have been known in the mutual fund industry for decades. In Professor Harris’ view, the dilutive harm caused by short-term trading was one about which mutual fund managers were or should have been aware during the Class Period. He and other plaintiffs’ witnesses described various sources of public information that spoke to dilution and stale pricing at the time.
[104] I have described above the dilution that occurred in the early days of mutual fund trading, to which Professor Harris testified and which was effectively eliminated in 1968 by the SEC’s amendment of Rule 22c-1 under the Investment Company Act which required prices of mutual fund units to be set based on the NAV calculated at the end of the trading day on which the units were purchased and sold, instead of the NAV calculated at the end of the day before the units were purchased or sold. As a result, when mutual fund investors buy or sell mutual fund units, they do not know the precise price of the transaction until the end of the day on which it occurred.
[105] When implementing this rule in 1968, the SEC stated its purpose to be:
to eliminate or reduce so far as reasonably practicable any dilution of the value of outstanding redeemable securities of registered investment companies through (i) the sale of such securities at a price below their net asset value or (ii) the redemption or repurchase of such securities at a price above their net asset value.[^33]
[106] The SEC went on to describe a further purpose of the rule change as being:
to eliminate or reduce so far as reasonably practicable other results, aside from dilution, which arise from the sale, redemption, or repurchase of securities of registered investment companies and which are unfair to the holders of such outstanding securities.[^34]
[107] The SEC described the mischief against which the rule is aimed as concerning speculators who:
buy large blocks of such securities under circumstances where the net asset value of the securities has increased but where the increase in value is not reflected in the price. The speculators hold such securities until the next net asset value is determined and then redeem them at large profits. These speculative trading practices can seriously interfere with the management of registered investment companies to the extent that (i) management may hesitate to invest what it believes to be speculators’ money and (ii) management may have to effect untimely liquidations when speculators redeem their securities. Based upon its experience in the administration of the Investment Company Act, the Commission believes that such practices cause unfair results to the holders of outstanding securities of registered investment companies.[^35]
[108] As noted earlier, Canadian regulators followed suit.
[109] The SEC published further Releases IC-10691 and IC-10827, dated May 15, 1979, and August 13, 1979 to further address stale pricing and dilution. In a release dated September 17, 1993,[^36] the SEC further amended Rule 22c-1. In doing so, it described the purpose of rule 22c-1 as follows:
Rule 22c-1 seeks to address the problem of “dilution” and to curb certain speculative trading practices.[^37]
[110] The directional thrust of the evidence at trial was that there was a North American market for securities and that regulation between Canada and the United States was substantially similar. This rule change demonstrates that the concepts of stale pricing and dilution have been known to the securities industry since at least 1968, 30 years before the start of the Class Period. While there is no evidence that the defendants or their officers specifically read the SEC publications referred to above, the concept of how dilution works in connection with short-term trading is one that led to a fundamental rule change in the securities industry. The concepts of stale pricing, dilution, frequent trading, and the reasons for calculating NAV as mutual funds are required to are all sufficiently central to the mechanics of the mutual fund industry that mutual fund managers can fairly be impressed with that knowledge even if they did not read the specific SEC releases that made the changes. It is the sort of general industry knowledge about one’s business that mutual fund managers ought to be aware of, especially when, as noted above, they present themselves to the public and the court as “skilled professional portfolio managers” whose “investment expertise is a valuable commodity.”
[111] In addition to regulations and the reasons for them, several American newspapers published articles about frequent trading and time zone arbitrage shortly before the start of the Class Period.
[112] On June 15, 1997 the New York Times published an article entitled “Timing the Market, and Proud of the Results.” The article described the investment technique of Gerald Appel, whom it described as a manager of $300 who “zips in and out of” mutual funds using mathematical models.
[113] On October 31, 1997 the New York Times published an article that provided specific evidence about time zone arbitrage and dilution. After a decline in US markets, the Hong Kong Stock exchange had fallen 14%. The US market rallied the day following the loss. Toward the end of that same day, $20 million flowed into the Fidelity Hong Kong and China fund. The following day, the Hong Kong market rose. Fidelity, however, refused to apply the Hong Kong closing prices on the day that the investors put in their $20 million because it feared that doing so would cause dilution to existing unitholders. Instead, Fidelity applied fair value pricing[^38] to deprive the switchers of their anticipated benefit. A second mutual fund company, T. Rowe did the same thing for its Japan fund. Investors complained to the SEC who ruled that what Fidelity and T. Rowe had done was acceptable practice.
[114] The Chicago Tribune published a similar article on November 23, 1997 in which it outlined the way in which time zone arbitrageurs could make money off of stock movements like those in respect of which Fidelity and T. Rowe had applied a fair value process:
The recent stock market turmoil in Southeast Asia has exposed a unique situation in which investors--big and small--took advantage of the more than 10-hour time gap between the close of Asian markets and the close of the New York Stock Exchange. The resulting arbitrage meant quick and virtually riskless profits, well above whatever penalties may be imposed on investors by fund sponsors to discourage short-term investing.
One effect was to change radically the character of Asian mutual funds and subject long-term investors in those funds to costly trading volatility. More volatility in Asia is a virtual certainty, as more investors are now aware of the arbitrage play or profiting from differences in price in the same securities in different markets.
It works like this: The stock market craters overnight in Hong Kong amid a speculative panic. As trading follows the sun, Wall Street rebounds sharply, creating a near certainty that Hong Kong will rally in its next trading day.
[115] The Wall Street Journal published a similar article on December 5, 1997.
[116] These articles suggest that by late 1997, the concept of time zone arbitrage was fairly broadly known in investment circles, at least in the United States. Mr. Spatt, an expert called by the defendants, admitted in cross-examination that market participants appear to have understood the concept that Asian markets were likely to follow North American markets.
[117] The defendants and their experts generally acknowledged that frequent trading could be harmful to mutual funds. The parties differ on two points: knowledge of time zone arbitrage and knowledge of dilution.
[118] With respect to time zone arbitrage, the defendants say they were not aware of it as a strategy during the Class Period. The defendants say they did not see these press articles cited above. I accept that evidence and do not find that they should have been aware of the specific information contained in the articles.
[119] There was no evidence tendered by either side to demonstrate discussion of time zone arbitrage as a known trading strategy in Canada. That said, the articles do suggest that concepts of time zone arbitrage, frequent trading and dilution were more broadly known and less arcane than the defendants make out. If the popular press is writing about investment techniques, one might expect that those who market themselves as experts whose knowledge is a “valuable commodity” would also be aware of those techniques. Moreover, as noted above, time zone arbitrage is the “precise concactenation” of events that caused harm. As the Ontario Court of Appeal held in Hacopian, the plaintiff need not establish the “precise concactenation” of events to demonstrate foreseeability. Time zone arbitrage is simply one reason for which market timers may engage in frequent short-term trading. Harold Sharon was one of the defendants’ experts on the mutual fund industry. He testified that during the Class Period short-term trading could be motivated by a number of other strategies such as “tactical asset allocation” and a “momentum strategy”. The former involved an investor attempting to capture excess returns by systematically trading in and out of positions with perceived short-term undervaluation using some sort of economic or market signal.[^39] The latter involved investing in securities that were rising.[^40] What causes harm is not the particular strategy or reason for which the trader engages in short-term trading, but the frequent short-term trading itself, and in particular, the withdrawal of the short-term trader’s funds before they were invested in any securities.
[120] With respect to dilution, the defendants suggest that they knew during the Class Period that frequent trading could cause certain harms to mutual funds such as higher transaction and administrative costs and a need to hold a higher level of cash in the funds thereby creating the potential of a cash drag on fund performance,[^41] but deny that they knew of dilution. If the defendants did not know of dilution caused by frequent short-term trading, I find that they ought to have been aware of it. As noted earlier, it was a concern about dilution that prompted the change to the way mutual funds were valued in 1968, a change that Ontario adopted.
[121] While the evidence discussed in this sub-heading does not demonstrate that the defendants were specifically aware of time zone arbitrage, it does demonstrate that they ought to have been aware of the concept of dilution and its exacerbation by frequent short-term trading.
ii. Prospectus Extracts
[122] The record discloses a broad range of prospectuses that are relevant to the issues of standard of care and foreseeability. They disclose that a wide variety of funds were aware of the harm that frequent short-term trading could inflict upon long-term unitholders. They specifically refer to speculators, market timers, the use of stale pricing as a trading strategy and state that short-term trading causes harm to the fund and other unitholders. Some impose mandatory short-term trading fees; others refer to discretionary short-term trading fees. This demonstrates that the harm that frequent trading does to long-term unitholders was well known.
[123] Fidelity Canada’s February 2, 1998 prospectus stated:
Sometimes, however, [closing prices reported on stock exchanges] aren’t the most accurate measure of value at the time of pricing. For example, values may be materially affected by events occurring after the closing of a foreign market. In these cases, Fidelity may take extra steps to properly determine the fair value of securities that have been affected by these events.
… This process also helps to protect our long-term investors against speculators and aggressive traders who try to take advantage of pricing inefficiencies in the market. These inefficiencies can happen with foreign securities, whose closing prices may no longer reflect their true value if there are major changes after overseas markets close. Hong Kong, for example, ends its trading day half a day before the TSE closes, and much can change during that time.
[124] Fidelity Canada’s September 1998, simplified mutual fund prospectus stated:
VALUING SECURITIES IN A FUND [V]alues may be materially affected by events occurring after the closing of a foreign market. In these cases, Fidelity may take extra steps to properly determine the fair value of securities that have been affected by these events. The number of securities valued in this way may be limited to a few stocks or may extend to an entire market.
We call this process fair value pricing because it serves the best interests of fund investors by helping to ensure that the prices at which they buy and sell units are fair and accurate, reflecting all information available at the time of pricing.
This process also helps to protect our long-term investors against speculators and aggressive traders who try to take advantage of pricing inefficiencies in the market. These inefficiencies can happen with foreign securities, whose closing prices may no longer reflect their true value if there are major changes after overseas markets close. Hong Kong, for example, ends its trading day half a day before the TSE closes, and much can change during that time.
[125] Samuel London is an expert called by the plaintiffs. He held senior management and executive roles at two Canadian mutual fund companies, Global Strategy Financial Inc., and Mackenzie Financial Corporation between 1990 and 2002. Since 2002 he has provided advisory services to mutual fund managers through his own firm. Mr. London testified that the potential for the exploitation of stale priced foreign securities by short-term traders was obvious to him in 1995:
So again, to emphasize, this [Fidelity prospectus] is from September '98, and it contemplates that speculators and aggressive traders can cause harm if they take advantage of stale pricing, so – and from my own personal experience at Global Strategy, we launched a Japan fund in 1995, and this -- this was, frankly, an obvious issue due to the valuation methodology. …[A]nd I'm not saying, nor do I say in my report, that fair value pricing was commonly used. What I am saying is the potential if you ever have a speculator or aggressive trader in your funds to use techniques like this was certainly known by me and certainly known by Fidelity, and since Fidelity would have been a leader in the industry, people would have looked to this type of disclosure and been aware of this as a potential issue.[^42]
[126] The defendants note that Fidelity removed the language about “speculators and aggressive traders who try to take advantage of pricing inefficiencies in the market” in its September 2000 prospectus.[^43] While that is correct, Fidelity continued to include language to the effect that:
If you redeem or transfer within 30 days of purchase units of any series of Fidelity Far East Fund, Fidelity RSP Far East Fund, Fidelity Japanese Growth Fund, Fidelity RSP Japanese Growth Fund and Fidelity Focus Funds, you will be charged a ShortTerm Trading Fee of 1% of the value of the units. If you redeem or transfer units of these funds between 31 and 90 days of purchase, you may be charged a Short-Term Trading Fee of 1% of the value of the units.
This fee is designed to protect unitholders from other investors moving quickly in and out of the funds. Frequent trading can hurt a fund’s performance by forcing the portfolio manager to keep more cash in the fund than would otherwise be needed or to sell investments at an inappropriate time. It may also increase the fund’s transaction costs.
[127] Between 1998 and 2003, the Fidelity Advisors Overseas Fund prospectus stated:
Short-term or excessive trading into and out of a fund may harm performance by disrupting portfolio management strategies and by increasing expenses. [The fund] may reject any purchase orders, including exchanges, particularly from market timers or investors who, in FMR's opinion, have a pattern of short-term or excessive trading or whose trading has been or may be disruptive to that fund.
[128] The prospectuses of the Fidelity Diversified International Fund and the Fidelity Overseas Fund have similar statements.
[129] A broad range of funds also imposed short-term trading fees to protect the interests of long-term unitholders.
[130] The Warburg Pincus prospectus for 1998-2000 stated:
The Portfolios reserve the right to refuse exchange purchases...when a pattern of exchanges within a short period of time (often associated with a marketing timing strategy) is discerned. The Portfolios reserve the right to terminate or modify the exchange privilege.
[131] The Lazard International Equity Portfolio prospectuses contained the following statements
2000:
[The fund reserves the right to] close an account due to excessive trading after prior notification”
2001-2003:
The Fund's exchange privilege is not designed to provide investors with a means of speculating on short-term market movements. A pattern of frequent exchanges can be disruptive to efficient portfolio management and, consequently, can be detrimental to the Portfolio's performance and shareholders. Accordingly, if the Fund's management determines that an investor is following a market-timing strategy or is otherwise engaging in excessive trading, the Fund, with or without prior notice, may temporarily or permanently terminate the availability of the exchange privilege, or reject in whole or part any exchange request, with respect to such investor's account.
[132] The BMO International Equity fund had the following statement in its prospectus from 1999 onward:
[The fund] discourages investors from excessive trading because it can harm a fund's performance and the value of other investors' holdings in the fund [and may exercise the right to reject an order.]
[133] The simplified prospectus of BMO dated February 22, 2003 provided:
We discourage investors from excessive trading because it generates significant costs for a fund. This can harm the fund’s performance and the value of other investors’ holdings in the fund. If, in our sole discretion, we determine that you are trading excessively, we may charge a short-term trading fee [ ] or refuse your order to buy or switch securities, and/or suspend or close your account. Excessive trading can include:
• buying and then selling a large number of securities of a fund • redeeming securities of a fund within 90 days of buying them.
[134] In 1998 – 99 the McKenzie IV International Fund prospectus contained the following language:
Because excessive trading (including short-term “market-timing” trading) represents an expense to all investors, the Funds may charge a fee.
[135] As of 1998, Templeton began including the following language in its prospectuses for its Emerging Markets fund, Global Smaller Companies Fund and International Stock Fund:
The Funds reserve the right to reject any order for the purchase of units within two business days of receipt of the order. This right applies without limitation to orders placed by short-term traders.
[136] The Scotia Mutual Funds Prospectus dated September 20, 1998 provided:
Because excessive trading (including short-term “market-timing” trading) represents an expense to all investors, the Funds may charge a fee of 2% of the amount exchanged or redeemed from a Fund or switches between Funds within 90 days of a purchase (or when the Asset Allocation Service is terminated and restarted within a 90 day period), other than for programmatic or portfolio manager rebalancing. This short-term trading fee is retained by the Fund as compensation for the costs associated with the exchange or redemption request. This fee does not apply to investments in cash equivalent funds, and may be waived by the Manager of the Funds, at its discretion.
[137] The Royal Mutual Funds simplified prospectus of August 25, 2000 provided:
Excessive trading
We may refuse your order to buy or switch units or any future orders if you trade excessively, which we explain below. If we refuse your order to buy, we’ll immediately return all the money we received with your order.
Mutual funds are considered long-term investments, so we discourage investors from buying, redeeming or switching units frequently. A switch is the redemption of units of one Royal Mutual Fund to purchase units of another.
We discourage investors from excessive trading because it generates significant costs for a fund. This can reduce a fund’s returns, which affects all unitholders. As a result, we may refuse your order if:
· you try to buy units of a fund within 90 days of redeeming units of the same fund
· you try to switch into units of any fund within 90 days of making a switch from any other fund
· your order to buy or switch would disrupt the efficient and cost-effective management of the funds.
[138] Professor Harris notes that the people who wrote the prospectuses knew that switching in and out of funds was problematic because they placed restrictions on it and gave mutual fund managers the right to impose penalty fees for frequent trading. The fact that someone imposed a restriction on trading in the prospectus is further evidence that fund managers knew or ought to have known that frequent trading in and out of funds caused harm to long-term unitholders.
[139] The plaintiffs’ expert, Samuel London testified that he expected people in the industry to read other prospectuses. The defendants’ expert, Michael Butler who was the President and CEO of Northwest Mutual Funds Inc. during the Class Period testified that it was not industry standard for a mutual fund company to review other companies’ prospectuses.[^44]
[140] Whether the standard in the industry was to read or not read other mutual fund company prospectuses is somewhat beside the point. What I draw from the prospectuses quoted above is that it appears to have been widely known within the mutual fund industry that frequent short-term trading caused harm to long-term unitholders. Whether a particular mutual fund acquired that knowledge from reading other prospectuses, its understanding of the investment business or other sources matters little. What matters is that mutual fund managers during the Class Period ought to have known that frequent trading causes harm to long-term unitholders.
[141] The prospectuses demonstrate that the standard of care during the Class Period required mutual funds to be aware of the risk of frequent short-term trading. The prospectuses also demonstrate that harm from frequent short-term trading was foreseeable.
iii. Public Disclosures of the Defendants
[142] Courts have recognized that the public disclosures of defendants are relevant to the scope and nature of their duties.[^45]
[143] The defendants’ public disclosures present their funds as: (i) being suitable for the long-term investor; (ii) discouraging short-term trading; and (iii) being managed by individuals who put the interest of investors above those of the fund manager.
[144] AIC’s motto during the Class Period was “buy, hold and prosper.” Its prospectuses describe its objective as buying quality businesses for long-term growth and its philosophy as that of a disciplined buy and hold strategy. The prospectuses describe investors who should invest in AIC funds as those seeking capital growth over the long-term and as having an investment horizon of at least five years.
[145] AIC’s investment philosophy was set out in its Simplified Prospectus dated August 16, 2000 which applied to all of its funds as follows:
At AIC, our goal is to create long-term wealth for our unitholders.
We try to achieve this goal by following a well-proven and disciplined investment philosophy: We aim to buy excellent businesses in strong, long-term growth industries and we hold these investments for the long run.
Using this "buy-and-hold" investment strategy, AIC strives to achieve our goals of:
• Capital preservation
• Capital growth at a superior rate of return
• Tax minimization
[146] CI’s prospectuses throughout the Class Period described its objective as obtaining “maximum long-term capital growth.”[^46] They describe their funds as being suitable for investors who are investing for the long-term.
[147] The prospectuses of both defendants also discouraged short-term trading.
[148] AIC’s Simplified Prospectus dated June 30, 1999, August 16, 2000, August 23, 2001, August 21, 2002, and July 24, 2003 stated:
Short-term trading fee
You may be charged a short-term trading fee if you redeem your Mutual Fund Units or Class F Units (other than Units of the Money Market Funds) within ninety days of purchase and you paid a front end sales charge commission of less than 2%. The fee equals the amount necessary to increase the commission you paid to 2% of the purchase price of the units redeemed. This fee is paid to the Fund.
[149] The CI Simplified Prospectus dated August 5, 1999 stated:
Frequent trading can hurt a Fund's performance by forcing a Fund to maintain higher levels of cash in its portfolio than would otherwise be needed. Frequent trading may also increase a Fund's transaction costs. Therefore, you may be charged a fee of up to 2% of the net asset value of the Securities being traded if you sell or transfer Securities of any Fund, other than the C.I. Money Market Fund, C.I. Short-Term Sector Shares or C.I. US Money Market Fund, within 60 days of the date of purchase of those Securities. C.I. will deduct the fee from the amount you sell or transfer and pay it to the Fund from which you transferred or redeemed your Securities. A transfer fee payable to your Financial Adviser or a redemption fee may also apply.[^47]
[150] CI repeated similar statements in its prospectuses throughout the Class Period. [^48]
[151] At trial, the defendants made efforts to qualify these statements. The defendants submit that the prospectus statements about pursuing long-term buy-and-hold strategies referred only to the fund managers, not the personal strategies of individual investors. While that is correct, the point here is that the defendants knowingly managed the funds in a way that facilitated frequent short-term trading by certain sophisticated investors. While the defendants may not have been able to control the investment strategies of the market timers, they were able to control the degree to which frequent short-term trading could occur within the defendants’ funds. Instead of prohibiting the activity, the defendants facilitated it through the switch agreements. That is managing the funds in a manner that is inconsistent with the representations contained in the prospectuses. The representations clearly send a directional message to retail investors to the effect that the funds would be managed in a way that promoted long-term buy-and-hold strategies and that maximized capital returns for long-term investors. Facilitating frequent short-term trading within the funds contradicted those representations.
[152] The defendants further submit that the prospectuses actually permitted frequent short-term trading by charging a fee for it. They also underscore that the fee is discretionary, meaning that frequent trading could occur without any fee being charged.
[153] I do not accept that argument. While a skilled advocate may be able to parse the statements in the prospectuses to render them close to meaningless, the overall tenor of the defendants’ prospectuses is to discourage short-term trading because it harms the fund and other unitholders.
[154] The prospectuses must be read purposively. They are consumer protection documents. They are not intended to be parsed by investors the way a court might deconstruct language of the income tax act before imposing a significant liability on a taxpayer. The overall thrust of the message delivered in the prospectuses is that short-term trading is frowned upon, is undesirable and is not wanted. Having set that message out clearly in their prospectuses, the defendants then went out and took proactive steps to design a program that permitted short-term trading. Consumers should be entitled to rely on the overall thrust of the message set out in the prospectus and should not be required to start asking themselves about the myriad of scenarios that the prospectus does not refer to. The courts should be looking to enforce the spirit of the prospectus and not search them for every possible technical out that might be available to the defendants.
[155] The defendants agreed that charging a 2% fee would have put an immediate end to the practice. Instead, the frequent traders were charged a fee of 0.2% under the Switch Agreements which facilitated the very practice their prospectuses discouraged.
[156] When long-term investors withdraw funds they have only recently invested, they tend to do so because of an unexpected need for cash. The withdrawals are likely to be small and few in number compared to the size and frequency of those in which the defendants allowed frequent traders to engage. As a result, even if withdrawals by long-term investors caused dilution, the harm would be minimal.
[157] A fair reading of the discretionary element in the frequent trading fee is not that the funds would actively encourage frequent trading through switch agreements but that they may allow an investor to remove money from a fund without a fee if the removal shortly after the investment was not a regular occurrence and was required because of genuine, unexpected need.
[158] AIC suggests in its closing argument that it disclosed short-term trading because the annual reports for each fund appended the financial statements for that fund. The financial statements disclosed in note 5 the total number of units issued and redeemed during the year. This indicated the volume of trading in each fund’s units. For example, the note disclosed that in 2001 the World Equity Fund had 38,216,587 units outstanding at the start of the year, that 172,825,453 new units were issued during the year, and that 177,405,471 units were redeemed during the year. This, argues AIC, disclosed that units equal to over four times the total units in the fund were issued and redeemed over the course of the year, which Professor Zitzewitz[^49] and Mr. Scanlan[^50] testified is an indicator of short-term trading. Compare this to the AIC Advantage Fund; it had 35,516,714 units outstanding at the start of the year, 9,702,216 new units issued during the year, and 10,559,758 units redeemed during the year. This implies that at least two-thirds of the units outstanding at the start of the year were not redeemed.[^51]
[159] In my view, this does not amount to disclosure of frequent trading in the funds. The funds were aimed at retail investors. The purpose of disclosure is to bring information home primarily to that retail investor audience. It is unrealistic to expect retail investors to parse through notes to financial statements and infer from the number of units traded that the fund permits frequent trading which dilutes the value of the units of long-term investors. Disclosure must be made in a way that makes clear to the reader the risks involved. Burying information in cryptic footnotes to financial statements does not amount to disclosure to retail investors.
[160] Moreover, Michael Butler, an expert in mutual funds called by the defendants and a person who ran a smaller mutual fund during the Class Period, testified that he would not know if someone was conducting frequent trading in his funds unless a portfolio manager brought it to his attention. It is somewhat ironic that the defendants expect retail investors to dissect footnotes in financial statements to detect frequent trading when their own expert says he would have required a portfolio manager to bring frequent trading to his attention.
[161] Like the prospectuses of other fund managers, those of the defendants demonstrate that the harms of frequent short-term trading were known during the Class Period. Their prospectuses indicated that fees of up to 2% were warranted in response to that harm. This demonstrates a standard of care during the Class Period that required the defendants to be aware of frequent short-term trading and to take reasonable steps to protect long-term unitholders from its harms. The defendants’ own reference to penalties of up to 2% for short-term trading demonstrates that the defendants knew that harm to others was foreseeable.
iv. The OSC Investigation
[162] When the OSC learned of the possibility of market timing, it launched the largest investigation in its history. The investigation culminated in a report published in March 2005, entitled The Ontario Securities Commission Report on Mutual Fund Trading Practices Probe.
[163] The plaintiffs sought to admit the report as further evidence of the standard of care during the Class Period. The defendants objected to its admissibility.
a. Admissibility of the OSC Report
[164] The defendants submit that the OSC Report was only admitted as a description of the OSC’s investigation and not for the truth of its contents. As a result, any statements about the OSC’s findings amount to inadmissible hearsay. I am not able to accept that submission. The OSC report is admissible both under the public records exception and the principled approach to hearsay.
[165] At common law, statements made in public documents are admissible as an exception to the rule against hearsay evidence. In R. v. A.P., (1996)[^52] the Court of Appeal said the following about the public records exception:
… This exception is “founded upon the belief that public officers will perform their tasks properly, carefully, and honestly.” Public documents are admissible without proof because of their inherent reliability or trustworthiness and because of the inconvenience of requiring public officials to be present in court to prove them. ..
[166] The Court of Appeal went on to note that public documents are generally admitted without proof if they meet the following four criteria:
(i) the document must have been made by a public official, that is a person on whom a duty has been imposed by the public;
(ii) the public official must have made the document in the discharge of a public duty or function;
(iii) the document must have been made with the intention that it serve as a permanent record; and
(iv) the document must be available for public inspection.
[167] The OSC Report clearly meets all four criteria and should therefore be admitted into evidence for the truth of its contents.
[168] The Report is also admissible under the principled exception to hearsay.[^53] That exception admits hearsay evidence if the necessary and reliable.[^54]
[169] The Report meets the necessity test because it is unsigned and its exact author(s) is unknown. At the time, it was “the biggest investigation in OSC history”[^55] and is likely to reflect the work product of numerous OSC staff. The attendance of all of its authors to identify the particular passages they wrote and the evidence they relied on in doing so would needlessly add to the length and complexity of the trial, without any benefit. This is especially true given that the report is being admitted only to demonstrate that other mutual funds prohibited frequent trading and took a variety of steps to do so. Evidence that is corroborated by other sources in the record before me.
[170] The Report also bears numerous indicia of reliability. The circumstances surrounding its creation render it inherently trustworthy.[^56] It is the result of a year-long investigation by a regulatory body.[^57] The investigation, the report and the settlements that arose out of it involved the participation of the defendants. The investigation underpinning the report proceeded in several phases of information gathering and included on-site reviews of 20 fund managers.[^58] It had the attention of the highest levels of the OSC and the mutual fund industry in Ontario.
[171] The Report is similar to other documents that have been admitted under the principled exception to hearsay such as: reports of Toronto Public Health investigators on an outbreak of infectious disease;[^59] an audit report prepared by Ernst and Young for the Senate;[^60] reports related to immunization prepared by the Public Health Agency of Canada and Ministry of Health and Long-term Care;[^61] reports of the British Columbia Utilities Commission relating to gas prices;[^62] and various governmental reports related to an Indigenous group’s land claim.[^63]
[172] As a result of the foregoing, I admit the report for the truth of its contents.
b. Contents of OSC Report
[173] The Report describes the scope of its investigation as “market timing” which it describes as:
The market timing which was the focus of our probe involved short-term trading (i.e. the rapid trading in and out) of mutual fund securities to take advantage of short-term discrepancies between the stale values of securities within the fund's portfolio and the current market values of those securities.
Stale values can occur in mutual fund portfolios when the prices of securities upon which a fund's price is based do not take account of the most recently available market information. This is most common in mutual funds whose portfolios have a material component of foreign securities traded on markets which closed many hours before the close of North American markets (e.g. European, Asian, International and Global funds).
[174] Thus, although the report observed that market timing was most common in funds with a material foreign component, the Report did not necessarily limit its definition of Market Timing to those funds.
[175] The OSC investigation proceeded in three phases. In phase one, the OSC wrote to all 105 mutual fund managers in Ontario who offered open-ended retail mutual funds to ask for information. In phase two, the OSC narrowed its probe to 36 of the 105 fund managers. Some of the 36 made it to the list because of frequent trading activity, others were included through random selection. These 36 funds were asked to submit trading data for the two years preceding December 31, 2003. This request included information about all “round-trip” trades exceeding $50,000 within five business days.[^64] A “round trip” is the combination of a switch in and switch out. The list of 36 managers was then reduced to 20 who were examined during phase three. During phase three the OSC concluded that five fund management companies stood out based on the degree and impact of frequent market timing activity in their funds. Those five firms were referred to enforcement and were the original defendants in this action. The OSC entered into settlement agreements with all five which resulted in a settlement of $205 million being distributed to investors. At the time, this was the largest settlement in OSC history.[^65]
[176] The report noted that some mutual fund managers looked for and prohibited market timing in their funds well before 2003. Many funds produced daily reports of market timing in their funds. Those managers also took active steps to discourage and ultimately stop the practice. The steps they took included:
Putting the account in which the short-term trading was being conducted on a watch list for further monitoring;
Issuing warning letters to the dealer and/or the client engaging in the trading;
Imposing a short-term trading fee;
Prohibiting further trading, except for redemptions, by the account holder; and
Closing the account.
[177] In his evidence at trial, Professor Kirzner testified that these controls were available to mutual fund managers throughout the Class Period.
[178] The OSC Report concluded:
Our case against the five fund managers referred to enforcement was based on their failure to protect fully the best interests of the affected funds. These fund managers had a duty to have regard to the potentially harmful impact of frequent trading market timing on a fund and its investors, and take reasonable steps to protect the fund from harm, to the extent that a reasonably prudent person would have done in the circumstances.[^66]
[179] The OSC probe is relevant because it discloses that many fund managers took steps to detect and prohibit frequent trading in their funds. The report sets out a number of specific steps that other funds took to detect and prohibit frequent trading. As noted, only 5 of 105 funds were reported to enforcement. This provides further evidence of a standard of care that the defendants failed to live up to.
[180] The OSC report is also consistent with other evidence in the record about the refusal of other funds to permit frequent trading.
[181] Matthew Scanlon, an expert called by the plaintiffs and the head of the Barclays Bank mutual fund division, testified that Barclays refused to permit frequent trading. The internal emails within AIC in July 2003 stated that AGF, CI, Fidelity and AIM did not permit frequent trading. Although the email appears to have been incorrect insofar as CI is concerned, it indicates that AIC knew that other funds prohibited frequent trading.
[182] It also appears that BMO, where one of the switch traders with whom the defendants dealt worked,[^67] was asked on 21 occasions by 15 different mutual fund managers to stop its frequent trading practices.[^68] The evidence before me indicated that Fidelity, Mackenzie, Royal Bank, Barclays, and Talvest all rejected frequent traders. Finally, in the United States, which has between 700 and 800 mutual fund complexes, settlements were negotiated with only 21 fund companies.[^69] This too suggests that the phenomenon was relatively limited which further suggests that the prevailing industry practice was to prevent frequent trading.[^70]
[183] The defendants submit that the OSC Report indicates that short-term trading was a broader phenomenon which occurred in more than just the five funds that were reported to enforcement. They rely on the following passage from the OSC Report in support of that submission:
Some fund managers became aware of frequent trading market timing activities as a result of conducting investigative procedures to respond to the OSC's Phase One letter, while others had already detected such trading activities through existing monitoring procedures prior to the OSC probe;
Many fund managers indicated that they made significant enhancements to their monitoring procedures in light of the OSC's Phase One letter.
Our analysis of the trading data reviewed in Phase Two revealed that all or a majority of the following indicators were present at some of the 36 fund managers:
significant volumes of trading activity in global and international funds;
a number of accounts with high volumes of transactions;
a number of accounts with trading activity in large dollar volumes over a majority of the two-year period reviewed;
large volumes of backdated trades, manual trades or error corrections;
the existence of agreements with certain clients permitting them to engage in short-term trading within certain parameters. These indicators, together with our objective of ensuring that the bulk of mutual fund assets under management in Canada would be reviewed, led us to determine that 20 fund managers warranted an on-site review;
[184] This suggests some ambiguity about which fund managers are being referred to, when which fund managers became aware of frequent short-term trading and when which managers took steps to prevent it.
[185] The Report goes on, however, to say the following about the fund managers apart from the five that were referred to enforcement:
However, none of the factors indicating risk of harm to investors were found to be present in a material way. In addition, our consideration of other relevant factors led us to conclude that these fund managers had taken reasonable steps to identify and prevent harm to their funds and their investors. As a result, and as illustrated in Chart 2, the impact of the frequent trading market timing activity to investors in those funds was found to be minimal on a relative basis.
[186] The Report also explained how the OSC had ascribed an adjusted risk rating to the funds in Phase 3. Those not referred to enforcement had an average risk rating of approximately 4.5. The five referred to enforcement had an average risk rating of 13.4.
[187] The final point to note in relation to the OSC Report is that when publishing the report, the then Chair of the OSC, David Brown, made the following published comments to the Economic Club of Toronto:
Mutual fund managers must exercise the degree of care, diligence, and skill that a reasonably prudent person would exercise in the circumstances.
Compliance with this duty requires that a mutual fund manager have regard for the potential harm to a fund from investors seeking to employ trading strategies that may be harmful or disruptive to the fund and its other investors. We believe it is a fund manager’s responsibility to put in place policies, procedures and other mechanisms to monitor trading that could be disruptive or harmful to the funds and take reasonable steps to protect the fund. The market timers owe no such duty to the other investors in the funds. They broke no laws.[^71]
[188] It is important to note that in citing Mr. Brown’s comments I am not suggesting that a speech by the Chair of a regulatory tribunal amounts to a determination of the duty or standard of care. It is merely one piece of evidence. I am also mindful that when acting here, the OSC was exercising its public interest jurisdiction which differs from and is broader than the law of negligence. That said, Mr. Brown’s comments were phrased in terms of the standard of care and were based on what he and his staff observed about industry practice. Standing alone, those comments would not satisfy the burden of proof on any issue relevant to this trial. When the comments corroborate other evidence in the record, which they do, they at least give me some comfort that I am not imposing an unrealistically high standard of care on the mutual fund industry.
[189] In short, the contents of the OSC Report provide further evidence that the standard of care during the Class Period required mutual funds to be mindful of market timing as defined in the Report and take steps to prevent it.
v. The OSC Settlement
[190] As noted earlier, the defendants entered into settlement agreements with the OSC. The AIC settlement sets out that AIC had three market timers who made profits of $127 million in respect of which AIC paid $58.8 million as part of the settlement; and that CI had five market timers who made profits of $90.2 million in respect of which CI paid $49.3 million as part of the settlement. The settlement funds were paid directly to investors in the AIC and CI funds.
[191] The settlement agreements contain admissions of fact. The plaintiffs tendered the settlement agreements at trial as proof of the facts admitted in them. The plaintiffs submit that, in light of those factual admissions, there is no real issue about liability.
[192] The defendants contest the admissibility of the settlement agreements because section 4 of each agreement provides:
Staff and the Respondent agree with the facts set out in Part IV herein for the purposes of this Settlement Agreement only and further agree that this agreement of facts is without prejudice to the Respondent or Staff in any other proceeding of any kind including, but without limiting the generality of the foregoing, any proceedings brought by the Commission under the Act (subject to paragraph 28) or any civil or other proceedings which may be brought by any other person or agency. No other person or agency may raise or rely upon the terms of this Settlement Agreement or any agreement to the facts stated herein whether or not this Settlement Agreement is approved by the Commission.
[193] In addition, the defendants note that the purpose of the OSC proceedings was not to assess liability for negligence or breach of fiduciary duty but to determine whether certain conduct was consistent with the public interest pursuant to s. 127 of the Ontario Securities Act.[^72] As a result, the contents of the settlement agreements are irrelevant to the determination of negligence or breach of fiduciary duty in this action.
[194] At the moment, it is settled law that any factual admissions in settlement agreements with a securities regulator are admissible in subsequent civil proceedings despite language of the sort contained in s. 4 of the settlement agreements. The Divisional Court made express findings to that effect in Hill v. Gordon-Daly Grenadier Securities.[^73] That conclusion has been applied in several other cases. I note, however, that s. 4 of the settlement agreements in this case is substantially more detailed than was the clause at issue in Gordon-Daly. More recent cases have observed that, as with any evidence, it is up to the trial judge to determine the weight the evidence should be given.[^74]
[195] Although I find that the settlement agreements are admissible for the truth of the factual admissions contained in them, they played no material role in my findings. The findings of fact necessary to establish liability were amply demonstrated by other evidence led at trial.
vi. Stromberg Report
[196] In 1995, Glorianne Stromberg, a leading Canadian securities lawyer and a former Commissioner of the OSC, published a report on the mutual fund industry in Canada. Its purpose was to consider changes in how mutual funds were regulated. Section 27.03 of the report is entitled “Short-term Trading by Investors” and notes, among other things, that:
One of the changes made at the time that National Policy No. 39 was adopted was the elimination of the requirement that there be a redemption fee payable by investors to an investment fund if an investor redeemed the securities that he or she had purchased within 90 days of such purchase. The maximum fee was 2% of the purchase price and the purpose of the fee was to discourage short-term trading by investors which negatively impacted on the investment fund by reason of there being increased administrative and transaction costs as well as opportunity costs arising out of the need to retain larger than normal cash positions to fund short-term redemptions.
People have mentioned that the practices that led up to the imposition of the short-term redemption fee are again occurring and that it would be in the interests of investors to discourage such practices by re-instating the requirement for a short-term redemption fee that would be retained by the investment fund. It is my recommendation that this be done.
[197] The defendants ask me to infer from this that there was no duty of care with respect to frequent trading because there had been controls on it in the past, those controls were removed, Stromberg recommended their reintroduction, but no regulatory controls were reintroduced. In those circumstances, the defendants submit that there can be no duty to prevent or prohibit frequent trading.
[198] I do not agree. The absence of a specific regulatory requirement does not negate a duty or standard of care. It may well be that a regulator believes it preferable for market participants to have a degree of flexibility and be regulated by common-law duties rather than absolute rules that require a fee to be applied regardless of the reason for the short-term redemption or trade.
[199] What I draw from the Stromberg Report is that it is further evidence, although not especially forceful evidence, that it was known in the industry that permitting frequent trading in mutual funds could cause harm to long-term unitholders.
vii. Academic Papers
[200] I was directed to 10 papers published during the Class Period that referred to time zone arbitrage. All were from American journals. The weight of the evidence suggests that the journals were academic in nature and were not widely read by investors or mutual fund managers. There was also no evidence before me about the extent to which there is crossover between academic research and the conduct of mutual fund managers or investors as there is for example in law or medicine. In both law and medicine, practitioners read academic journals either as part of regularly keeping up to date in their field (as in medicine) or as part of research for cases (as in law). In medicine, academic research is usually ahead of practitioners who gradually implement the research in day-to-day practice. In law the situation is more nuanced. Research both follows practice in the sense that it often synthesizes a complex body of caselaw into more digestible principles, but it also leads practice in its analysis and critique of caselaw which is then sometimes implemented by practitioners and courts. No one explained whether in finance, academics lead practitioners, whether the inverse is true or whether both lead on different aspects as in law.
[201] In these circumstances, I do not consider these papers as part of my negligence analysis.
viii. Fair Value Valuation
[202] In the ordinary course, mutual funds calculate their NAV with reference to the last traded price of each security in the fund. In circumstances where a mutual fund believes that the last traded market price does not accurately reflect value, they are permitted to use their own fair value calculation to determine NAV. During the Class Period, regulators in both the United States and Canada issued certain commentaries about the use of fair value trading. The plaintiffs submit that these commentaries constitute further evidence of a standard of care that requires the defendants to have been aware of time zone market timing.
[203] On April 30, 2001 the Securities and Exchange Commission sent a letter to the Investment Company Institute (“ICI”), the primary American industry association for mutual funds and the principal point of contact between the SEC and the mutual fund industry. The letter clarified when fund managers may consider using fair value to calculate NAV and stated:
Funds generally calculate their NAVs by using the closing prices of portfolio securities on the exchange or market (whether foreign or domestic) on which the securities principally trade. Many foreign markets, however, operate at times that do not coincide with those of the major U.S. markets. For example, Asian markets generally operate during the evening and nighttime in the United States and close before the opening of the major U.S. markets. As a result, the closing prices of securities that principally trade on foreign exchanges or markets (“foreign securities”) may be as much as 12-15 hours old by the time of the funds' NAV calculation, and may not reflect the current market values of those securities at that time. In particular, the closing prices of foreign securities may not reflect their market values at a fund's NAV calculation if an event that will affect the value of those securities (“significant event”) has occurred since the closing prices were established on the foreign exchange or market, but before the fund's NAV calculation.
[204] The letter continued to observe under the heading “The Failure to Determine the Fair Value of Portfolio Securities Following Significant Events May Result in Dilution”:
Funds may dilute the value of their shareholders' interests if they calculate their NAVs using closing prices that were established before a significant event has occurred. Dilution generally may occur, for example, if fund shares are overpriced because redeeming shareholders will receive a windfall at the expense of the shareholders that remain in the fund. Similarly, dilution may occur when a fund sells its shares at a price lower than its NAV. The risk of dilution increases when significant events occur because such events attract investors who are drawn to the possibility of arbitrage opportunities. In such situations, short-term investors may attempt to exploit the discrepancies between market prices that are no longer current, and the values of a fund's portfolio securities.
[205] In early 2000, The Investment Funds Institute of Canada (“IFIC”) formed a Fair Value Working Group at the request of its Regulatory Committee to develop guidelines for the use of fair value calculations in Canada.
[206] IFIC’s role was to advocate for the industry to regulators, provide industry perspective to regulators, study and formulate best practices in fund management, educate the industry on best practices, and monitor developments in investment fund regulation, governance, and management.[^75] The members of the IFIC board during the Class Period included representatives of mutual fund companies, investment dealers, and regulators.
[207] Although there were no specific legal requirements in Canada to apply fair value principles, IFIC recognized that there may be circumstances in which a fund should consider fair valuing a security. The working group was struck to examine the issue and provide guidance about what processes a fund should follow to determine a fair value and when it should do so. The ultimate IFIC bulletin that arose out of this exercise was published in March 2002. It provided as examples of when to use fair value situations like earthquakes, the closure of securities markets, the September 11 attack, or unexpected news after the closing of the foreign market such as a surprise interest rate decline.
[208] That bulletin provided, among other things:
Foreign securities that trade in markets that close earlier than the Canadian markets are normally valued at their most recent closing price on the principal exchange, even if that is earlier than the time when the fund's NAV is struck.
The fund should have policies and procedures in place to ensure timely identification of events or information between the closing of the principal exchange and the striking of the fund's NAV that may impact the value of a foreign security or securities.
[209] Several of the plaintiffs’ experts also pointed to the use of fair value pricing as a further option to control market timing. The defendants’ experts contest this and highlight the rare use of fair value by market participants, the limited situations in which regulators refer to it and the frailties of fair value as an appraisal mechanism to calculate NAV of mutual funds. By way of example, fair value calculations could lead mutual fund managers to value securities at more than they may be objectively worth in order to increase their management fees. I accept all of these frailties and limitations in connection with fair value calculations.
[210] In my view, the evidence relating to the use or development of fair value methods does not demonstrate a standard during the Class Period to apply fair value calculations in the circumstances at issue in this case. There was no evidence before me about how widely read SEC publications were in the Canadian mutual fund industry. The IFIC publication came relatively late in the Class Period. Although both the SEC and IFIC publications provide further evidence about the knowledge of dilution, they do not constitute sufficient evidence to require mutual funds to apply fair value methodologies to frequent short-term trading during the Class Period.
[211] Fair value considerations therefore do not play any role in my decision.
ix. American Case Law
[212] The plaintiffs rely on two American decisions to support their arguments about the standard of care during the Class Period: The 1993 decision in Windsor Securities v. Hartford Life Ins. Co.,[^76] and the 2001 decision in First Lincoln Holdings v. Equitable Life Assurance.[^77]
[213] Both cases involve situations in which American mutual funds took steps to restrict frequent short-term trading in their funds. The traders sued the mutual funds. In both cases the courts noted that the mutual funds were acting to protect their long-term unitholders against the harm that frequent short-term trading imposed on the funds.
[214] There is no evidence that any Canadian mutual fund manager was or should have been aware of either decision during the Class Period. The plaintiffs did not put either decision to any witness at trial. There was also no evidence at trial about how American legal decisions made their way into standards of practice in the Canadian mutual fund industry.
[215] In those circumstances, the Windsor Securities and First Lincoln decisions have no bearing on my findings.
C. Negligence and the Defendants’ Specific Conduct
[216] In this section I turn to the specific conduct of AIC and CI to examine whether the circumstances of the frequent short-term trading within their funds ought to have caught their attention and whether permitting it to occur breached the standard of care during the Class Period. On my view of the record and as set out below, both AIC and CI breached the standard of care they owed to their funds by permitting frequent short-term trading.
[217] Both AIC and CI entered into Switch Agreements with frequent traders to allow them to conduct frequent short-term trading.
[218] For Professor Laurence Harris, the Switch Agreements were a “particularly egregious” example of the Defendants’ breach of the standard of care, because they represent “active efforts” to facilitate a behaviour that is contrary to the interest of long-term unitholders and contrary to the provisions of the prospectus.[^78] This is not only a failure to recognize a problem the defendants should have recognized, it is active facilitation and exacerbation of the problem.[^79] I accept this evidence.
[219] Both defendants knew that the parties with whom they were entering into Switch Agreements were sophisticated hedge funds. That should have been seen as the first red flag.
[220] Professor Harris underscored that hedge funds usually do not invest in mutual funds. Hedge funds, generally speaking, manage their own money and do not want to pay the high management fees associated with mutual funds. Hedge funds themselves charge their own clients a management fee. Their more sophisticated clients would ask why they are paying the hedge fund a management fee if the hedge fund is in turn paying someone else to manage their investment.[^80] Moreover, large institutional traders such as hedge funds could always ask a mutual fund manager to manage their money for them in a separate account, not within the mutual fund and usually at a lower management fee than retail-oriented mutual funds charge. A frequent trading strategy of the sort the switch traders wanted would not work, however, if they had their own private pools that the defendants managed. As a result, for Professor Harris, the very fact that a hedge fund was asking to invest in a mutual fund should have raised a large red flag.
[221] Professor Harris also noted that people in the finance industry are generally sharp, intelligent individuals who are always looking for the motives behind someone’s conduct. It is those motives that alert them to profit possibilities. As a result, when a hedge fund invests in a mutual fund and engages in a pattern of large volume switching, the first question the mutual fund should ask is whether the switcher is making money on the strategy. According to Professor Harris, mutual fund managers are intelligent people and it is hard to imagine that they lacked any degree of curiosity about the activities of the frequent traders.
[222] The defendants submit that I should discount Professor Harris’ evidence because he has little, if any, Canadian experience. In my view that is not a significant factor. The evidence was largely consistent that investment practices and duties of mutual fund managers were largely similar in Canada and the United States.[^81] The contextual factors about which Professor Harris testified were based either on common sense[^82] or were uncontested historical facts.[^83] In addition, the defendants suggested that Professor Harris was testifying beyond his expertise as an economist. I do not agree. A large branch of economics focuses on how people behave in relation to information and economic facts. That was the thrust of Professor Harris’s evidence.
[223] Moreover, the evidence at trial demonstrated that the mutual fund industry is highly integrated across North America and is governed by similar professional standards.
[224] The discussions surrounding these switch agreements were conducted through Larry Ullman, a stockbroker at RBC and Devon Yuill, a stockbroker at BMO and later at TD. Messrs. Ullman and Yuill were well known within the Canadian mutual fund industry. Mr. Butler, one of the defendants’ experts, confirmed even though he was at a smaller mutual fund, he was aware of Mr. Ullman before speaking with him.[^84] Mr. Butler described Mr. Ullman as being “a very large and influential RBC stockbroker” and noted that for his smaller fund company “it would be a really big coup if all of a sudden a gentleman like Larry Ullman was promoting our funds”.[^85] Mr. London, one of the plaintiffs’ experts, testified about the notoriety that Messrs. Yuill and Ullman had among Canadian mutual fund industry members as frequent traders to the point that McKenzie funds, where Mr. London worked, had flagged their accounts for closer monitoring for frequent trading.[^86]
i. AIC Conduct
a. Entry into Switch Agreements
[225] In AIC’s internal correspondence during the Class Period, the frequent traders were known as “flippers” or “switchers”, which referred to their practice of moving their investments between funds.[^87]
[226] Although AIC did not enter into switch agreements until 2001, it had been doing business with Mr. Ullman’s client Reliable Capital since at least 1999. Mr. Murdoch knew Reliable to be a Bermuda based institutional investor. The account statements of Reliable from 1999 onward show a pattern of switching in a certain amount and switching out the principal plus profit within a few days. That would then be followed by the further switch in of the original principal amount and so on. When viewed across AIC’s various funds, Reliable’s frequent trading occurred on an almost daily basis.
[227] Those account statements alone demonstrate that AIC had institutional knowledge of frequent short-term trading and the profits the switchers earned from it. To the extent AIC claims that institutional knowledge would require knowledge by a senior director or officer of AIC, I would decline to accept that approach here. As noted earlier, it was evident that other mutual funds were alive to the harms of frequent trading and took steps to identify and prevent it. AIC should not be able to avoid liability because of an alleged lack of knowledge when that ignorance was brought about by its failure to take steps to identify conduct that it had described as harmful in its prospectuses.
[228] Between November 2001 and November 2002, AIC entered into Switch Agreements with each of Reliable, SII and Pentagon that contemplated those investors conducting frequent short-term trading in AIC funds. AIC’s first Switch Agreement in the record before me was entered into in November 2001 with Reliable Capital through Mr. Ullman.
[229] The first record of discussions concerning that agreement arose on September 17, 2001. The limited documentation surrounding the Switch Agreement between AIC and Reliable suggests that AIC entered into the agreement because of a concern that Mr. Ullman/Reliable would pull their business out of AIC if AIC did not allow them to engage in frequent trading.
[230] Two documents are critical in this regard: an email and a letter from Mr. Ullman, both dated September 17, 2001, both addressed to Neil Murdoch, AIC’s Chief Financial Officer and Executive Vice-President during the Class Period. The email in its entirety says:
Hi Neil,
I hope this opens up a productive dialogue so that we can continue to invest with AIC. I know that this method of investing works for all stakeholders (unitholder, fund company, my client), since we have been successfully implementing it for close to five years.
The key to maintaining the viability is the amount of assets we have invested relative to the overall fund asset levels. Under no circumstances should we be in a position to effect (sic) the funds in a manner you are not comfortable with.
Unfortunately fund companies cannot accomodate (sic) everyone who employs this strategy, but we would like to create the conditions necessary for a long-term relationship that is in the interest of all stakeholders. I have made these type of fund relationships a cornerstone of my practice because I am confident in its viability. Please review the enclosed documentation as an example of how we believe we can hopefully proceed going forward.
[231] The opening line is telling:
I hope this opens up a productive dialogue so that we can continue to invest with AIC.
[232] It contains a polite but clear threat to withdraw business from AIC unless AIC and Mr. Ullman can come to an agreement on frequent trading. The email does not refer only to losing Reliable’s business but all of Ullman’s business. The purpose of the email is to determine whether “we can continue to invest.” Both parties had a decision to make: continue to allow short-term trading or have Mr. Ullman withdraw his clients’ money from AIC. The letter Mr. Ullman sent Mr. Murdoch the same day speaks of Mr. Ullman using this strategy with approximately 25 mutual fund companies in North America. In other words, Mr. Ullman had options and could easily remove his money from AIC.
[233] The second critical point in the email is that Mr. Ullman says he has been successfully implementing this strategy for close to five years. This made it very easy for Mr. Murdoch to determine whether Reliable’s past activity had created a threat of dilution. There can be no doubt that Mr. Murdoch knew that the strategy being referred to was one of frequent trading which should have alerted him to the threat of dilution. The separate letter that Mr. Ullman sent Mr. Murdoch on the same day as the email referred to a “shorter-term strategy” and set out the parameters for the proposed frequent trading. That letter also indicated that the client has had a relationship with AIC since 1998 and identified the client by account number. That gave Mr. Murdoch the ability to determine Reliable’s trading history with the push of a button. Reliable’s monthly account statements with AIC identify the date of each switch into and each switch out of each of AIC’s funds and calculate Reliable’s profit or loss on each transaction. By looking at the time during which Reliable’s money was in the fund and looking at the profit/loss column AIC would have been able to determine at a glance whether the trading posed a threat of dilution that warranted further investigation. Neither Mr. Murdoch nor anyone else at AIC analysed Reliable’s account statements to assess the threat or fact of dilution. Nor was that exercise ever performed for any of AIC’s other frequent short-term traders.
[234] The letter that Mr. Ullman sent Mr. Murdoch contains a further red flag. After describing himself as a person “who established this business concept in the Canadian market,” Mr. Ullman sets out the essential components of the proposal. The second component of the proposal reads:
- Client's assets become part of the equity fund's cash position - i.e. no extra trading activity is necessary by the fund manager.
[235] In essence, Mr. Ullman is candidly setting out the fact that his proposal makes money through dilution. Dilution only occurs because the frequent trader’s money is not invested in securities. He switches money in one day and switches it out a day or two later before his funds are invested. If the frequent trader’s money were actually invested in securities, there would be no dilution because the profit earned within the fund would include profit earned on the short-term trader’s capital.
[236] Finally, the third paragraph of Mr. Ullman’s email starts by saying:
Unfortunately fund companies cannot accommodate everyone who employs this strategy…
[237] This should have been seen as yet another red flag in the reader’s mind. Mr. Ullman is saying that not only can fund companies not allow all of their investors to pursue the proposed strategy; fund companies cannot allow even the smaller number of investors who know of and want to pursue the strategy to do so. Mr. Ullman is openly stating that he wants to be given special treatment. Treatment certainly better than the run-of-the-mill investor. And treatment even better than the smaller subset of investors who are aware of and would like to engage in frequent short-term trading.
[238] To summarize, the email and letter sent on September 17, 2001:
(i) threatened to remove business;
(ii) provided a roadmap to an easily accessible record of Reliable’s trading history;
(iii) specified that Reliable’s money would never be invested in securities; and
(iv) advised that Ullman wanted special treatment compared to other investors.
[239] That should have brought home the risk of dilution to any reasonably prudent mutual fund manager.
[240] In addition to the frequent trading conducted by Reliable, SII, and Pentagon, it was known within AIC that it had made oral agreements with other switchers to allow frequent trading. This is evident from an email from Shameena Khan to Miles Radoja dated July 11, 2003. After receiving a complaint about frequent trading, Ms. Khan replied:
Thanks Miles. I will find out how this is being done.
However we have to take into consideration that we have made agreements (verbal or otherwise) with the switchers to allow this type of activity at the PM level. My understanding is that Neil/ MLC had given the Ok to allow the daily switching of funds, therefore we need to understand the basis of these agreements before applying a switch fee. Should we apply a fee I am sure will result in withdrawal of the assets. I do know that AGF, Cl, Fidelity and AIM did not allow the 'switchers' to do business with them and as such can apply a fee.[^88]
[241] Once again, the concern is with the withdrawal of investments by the frequent traders if fees are applied. At the same time, there appears to be internal knowledge that other mutual fund companies do not allow switch trades.
b. Terms of the Switch Agreements
[242] The terms of the three Switch Agreements were substantially similar:
(i) They set out between 2 and 7 AIC Funds that the frequent trader could switch in and out of.
(ii) They limited the investment into any one fund to between $20 and $50 million or 4% of the fund per switch.
(iii) They limited switches to between 4 and 8 round trips per fund per month.
(iv) They charged a fee of 2 basis points[^89] (0.02%) on all switches into a fund with no fee being payable on a switch out into the AIC money market funds.
(v) They allowed AIC to terminate the agreements on 3 - 10 days’ notice.
(vi) They required the frequent trader to keep the terms and conditions of the switching program confidential and not to disclose them to any third party without the prior written consent of AIC unless required to do so by law.
(vi)
[243] The plaintiffs argue that the confidentiality clauses were included because the agreements violated the prospectuses and because the agreements were evidence of the fund manager putting its own interest above those of its long-term investors. Given that the defendants’ prospectuses had already said that short-term trading can cause harm, the plaintiffs say it would have been problematic to have disclosed publicly that the fund manager was permitting short-term trading by certain investors.
[244] Mr. Murdoch testified that AIC did not think disclosure was necessary because, in management’s view, the Switch Agreements did not meet the materiality test for disclosure. I find that explanation difficult to accept.
[245] AIC went beyond deciding whether it was required to disclose. It took the additional step of prohibiting the frequent traders from disclosing. Whether the frequent traders could disclose the Switch Agreements has nothing to do with materiality requirements for AIC, it has to do with AIC wanting to take proactive steps to keep the agreements confidential.
[246] Moreover, the question is not what level of materiality securities legislation requires for disclosure in the abstract; the question is whether the prospectuses leave a misimpression on investors that would be important for an investor’s decision to invest or continue to hold their units. The suggestion in the prospectus is that frequent trading is discouraged and would be subject to a 2% fee. In this context, the right question is whether it would have been important for an investor to know that the mutual fund was allowing frequent trading by certain sophisticated investors for a fee of 0.2% and that frequent trading could cause dilution of long-term unitholders’ returns. In my view, that would have been important for investors to know.
[247] Moreover, AIC appears to have been careful to ensure that whatever fee the market timer was charged was one that allowed the market timer’s strategy to remain profitable. By way of example, in an email exchange between AIC and Yuill on November 5, 2002 in connection with a revised Switch Agreement, Yuill writes:
Thank you for the revised agreement. The client is okay with the agreement excluding the fee charge on the Advantage fund.
The fee on the Advantage fund would render their strategy for this fund unprofitable. [^90]
[248] In other words, it appears AIC was prepared to eliminate the switch fee if the client asked.
c. Advice
[249] AIC defends the legitimacy of the Switch Agreements based on the fact that Mr. Murdoch says he received advice from his legal and financial teams before entering into them.
[250] Nothing about the legal advice has been disclosed. Without knowing specifically what advice was sought and given, the general allegation that a party received legal advice does not provide a defence.
[251] The advice from the finance team came from Victoria Ringelberg. Ms. Ringelberg began working in AIC’s accounting department in 1988. She became accounting manager in 1990, Comptroller in 1992, Vice President of Finance and Accounting in 1998 and CFO in 2002. As it came out at trial, Ms. Ringelberg’s advice provides no defence for AIC but only strengthens the case against it.
[252] Before entering the Reliable Switch Agreement, Mr. Murdoch asked Ms. Ringelberg to assess the impact on a mutual fund of uninvested cash that was being switched in and out of the fund. She came up with a number of hypothetical scenarios that she discussed with Mr. Murdoch. The directional conclusion of each scenario was that if the market rose between the switch in and the switch out, the uninvested cash would have a negative impact on the fund’s NAV and dilute its long-term unitholders. If the market declined between the switch in and switch out, the uninvested cash would have a positive impact on NAV and long-term unitholders.
[253] In the face of this analysis, AIC nevertheless continued to permit frequent trading both with and without Switch Agreements. AIC justifies its decision by asserting that market returns are a “random walk” that cannot be predicted, as a result of which the dilution on days when markets rose would be balanced out by the positive effect of days when markets declined.
[254] AIC did not test its random walk assertion against the account statements of Reliable or any other short-term traders. Had AIC done so, it would have discovered that the short-term traders’ “walk” was not random, but highly profitable. AIC failed to check its random walk theory against Reliable’s statements even though Mr. Ullman told Mr. Murdoch in his email of September 17, 2001 that he had been successfully implementing the frequent trading strategy for close to five years and that Mr. Ullman had made frequent trading a cornerstone of his practice because he was confident of its viability.
[255] Indeed, the “random walk” theory would rarely apply to a frequent trader whose cash is not invested in securities. All the frequent trader need do is wait until the market increases above the level at which the frequent trader invested. Adding the element of time zone arbitrage only shortens the period during which uninvested cash remains in the fund because it improves the probability of the frequent trader’s prediction of market increases being correct.
[256] Despite the Ringelberg analysis, Mr. Murdoch and others at AIC state that they first learned of market timing and its dilutive impact after the announcement of the Spitzer investigation. While that may have been true specifically of time zone arbitrage, AIC had more than ample evidence to demonstrate that frequent trading by Mr. Ullman’s clients created at least a strong risk of dilution. Enough of a risk to warrant at least a brief review of Reliable’s trading statements.
[257] As noted earlier, the Switch Agreement with Reliable charged a fee of 2 basis points (0.02%) per switch. Ms. Ringelberg’s analysis showed that an increase of 0.5% in the market diluted the net asset value of the fund by 89 basis points. Larger increases led to larger dilutive effects.
[258] When Mr. Murdoch was asked at trial about the meaning of certain conclusions in Ms. Ringelberg’s analysis, he replied that Ms. Ringelberg would be better able to answer the question because it was her analysis. When Ms. Ringelberg testified, she tried to distance herself and AIC from the analysis in several ways.
[259] First, she described her task as merely crunching numbers. She denied that AIC’s role was to protect the interests of investors in the fund. She did not recall Mr. Murdoch asking her whether the proposal from Mr. Ullman might be harmful to the fund. She did not know how frequent the trading in and out would be nor did she know the amounts that would be traded.
[260] Second, she argued that there were so many permutations that no one analysis could be taken as reflecting the true impact of frequent trading on the funds, that her calculation was flawed and that the numbers don’t mean anything because the analysis was flawed. I agree that there are many permutations and that a single analysis may not be conclusive. The point I take from the analysis is that every permutation showed damage to long-term unitholders if the market rose between a switch in and switch out. That warranted significant caution and should have led AIC to refuse the proposal or, at a minimum, analyse actual past results by reviewing Reliable’s monthly statements. Ms. Ringelberg agreed that she could have but did not do so.
[261] If, as Ms. Ringelberg said in cross-examination, the analysis was flawed and meaningless, the situation only worsens for AIC. It means that the analysis Mr. Murdoch relied on to justify the Switch Agreements provided no such justification at all.
[262] The only thing that appeared flawed in the analysis during cross-examination was that it seemed to underestimate the damage to the fund. The Ringelberg scenario produced as Joint Document Book Tab 170A showed that a frequent trader’s activity would result in a price difference of four cents per unit in the price of 93 units held by long-term unitholders. She calculated that as a total of $0.88 of damage to the fund. During cross-examination, Mr. Jervis tried to get her to agree that the proper calculation of harm in her scenario would require the 4 cent difference to be multiplied by the 93 units which would arrive at damage of $3.72 and not $0.88. Ms. Ringelberg would not agree with that analysis because she had not carried it out herself and was “not comfortable thinking about things on the fly.”
[263] Professor Christoffersen, an expert tendered by the defendants, confirmed that Mr. Jervis’ approach was the notionally correct method to calculate harm to the fund. I say notionally because that calculation may be subject to a number of other nuances that I do not address here but which will be left for the damages trial.
[264] That said, the precise amount of harm that Ms. Ringelberg’s scenarios demonstrate is somewhat beside the point. The real point is that they demonstrate harm in excess of 2 basis points.
[265] Ms. Ringelberg agreed on cross-examination that the 2 basis point fee was not intended to compensate for dilution but was intended to compensate only for the increased transaction costs that switch trading caused the fund to incur. That becomes even more problematic for AIC because it means that they had evidence of dilution in the Ringelberg analysis which they simply ignored based on their theory that markets were a random walk. This despite Mr. Ullman’s statement to the contrary and despite evidence to the contrary in Reliable’s monthly statements.
[266] AIC submits that Professor Christoffersen found that the “success rate” of the parties with Switch Agreements in the certified AIC Funds “was barely more than a coin flip” at 58%.[^91] That is not quite the case. The report is limited to analysing the next day NAV. It may well be that the true measure is whether there was an increase in the NAV before the switches in were invested in securities. If the funds were never invested as Ullman suggested, the true measure may be the time between the switch in and the switch out. In addition, Professor Christoffersen’s report also shows that the next-day NAV decreased only 39% of the time and remained the same 5% of the time. Even on Christoffersen’s analysis, a win loss ratio of 58 to 39 is something more than a coin flip. Those, however, are all issues for the damages trial.
[267] Mr. Murdoch admitted that Ms. Ringelberg’s analysis led him to conclude that that he would have to hold more cash in the fund than he otherwise would, but he did not think it would change how he managed the fund. This misses the point. The issue is not so much how Mr. Murdoch manages the fund; the issue is whether the interests of long-term unitholders are being diluted. If the additional cash that Mr. Murdoch concluded he would have to hold reflects the frequent traders’ uninvested cash, as Mr. Ullman wanted, it is clearly diluting the interests of long-term unitholders. Moreover, even if the additional cash is not the uninvested money of the frequent traders, it still means that the fund is holding more cash than it otherwise would and has less money invested for the benefit of long-term unitholders. While that might not affect Mr. Murdoch’s management of the fund in the sense of influencing what securities he purchased, it clearly did affect his management of the fund in terms of the proportion of fund assets that were invested in profit generating securities.
d. Due Diligence
[268] Section 2.1 of AIC’s compliance manual provides:
Obligations to Clients.
It is our policy to continue to maintain the highest standards of service to our clients. AIC has a fiduciary duty to its clients to act honestly, in good faith and in the best interests of our mutual funds and their investors and to exercise the degree of care, diligence and skill that a reasonably prudent manager would exercise in the circumstances. This standard of care extends to the service provided by all employees, officers and directors of AIC in each facet of our business operations.
[269] Mr. Murdoch understood that this included a duty to protect unitholders from harm, a duty to be diligent and vigilant and a duty to ensure against activities that could cause harm to unitholders. Mr. Murdoch agreed that this imposed a duty to review carefully what frequent traders were proposing.
[270] Even though Mr. Murdoch believed that Mr. Ullman’s form of frequent short-term trading was unique, he disagreed that reviewing Reliable’s history was a form of due diligence AIC was required to undertake. He agreed, however, that if they had reviewed Reliable’s statements, AIC could have discovered what Reliable had been doing and what the result was for both the client and the fund.
[271] AIC also seems to have had a blinkered view of its duties to protect clients. When asked whether he made inquiries with other fund managers to see how they were dealing with frequent traders, Mr. Murdoch replied that that was not within the scope of duties he performed at AIC. Mr. Murdoch was AIC’s, Chief Financial Officer and Executive Vice President during the Class Period. Those are senior executive officer positions that do not allow someone to say “that’s not my job”. In roles that senior, one is expected to put a halt to something that is amiss or at least vigorously bring it to the attention of those with authority to put a halt to it rather than turning a blind eye. He then added that AIC was based in Hamilton which meant it had less contact with firms in Toronto. I do not accept that as a basis for ignorance. AIC and its employees were members of various fund organizations such as IFIC. It had a seat on the IFIC board. As a CFO and Executive Vice President, one might reasonably expect Mr. Murdoch to have contacts with whom he could raise issues.
e. Trading Patterns
[272] The scale of frequent trading within AIC was significant. In June 2002, Reliable alone switched a total of $343 million in and out of various AIC funds. In August 2002 the number was $387 million. In November 2001, Pentagon, SII and Reliable switched approximately $60.8 million into the Global Advantage Fund amounting to approximately 62.2% of its total NAV. Exhibit 3A to Professor Christoffersen’s report shows approximately $6.596 billion being switched in and out of AIC funds during the Class Period. Professor Christoffersen disagrees that these were all time zone arbitrage trades because some of the funds had large American components. That too will be an issue for the damages trial.
[273] AIC submitted in its closing argument that it prudently monitored the trading conducted pursuant to the Switch Agreements to ensure compliance with their terms. I disagree. Instead, I see a consistent pattern of limits being exceeded in response to which AIC simply increased the limits after the fact.
[274] By way of example, the Reliable Switch Agreement limited the maximum investment in the Global Advantage Fund to $3.7 million, or just less than 4% of the fund’s total net asset value. Yet Reliable switched in $9.7 million on November 12, 2001, $10 million on November 26, and a further $10 million on November 29.
[275] Joint Document Book production 103 is an email exchange within AIC in November 2002. It indicates that the switch trade limits that were placed on Reliable were being exceeded. There does not appear to have been any mechanism within AIC to limit the trading to the agreed-upon caps. The solution to the issue was not to rein Reliable in but to increase its limits. An email from Angela Burlock dated November 19, 2002 set out the new limits and added:
The trading limits have been changed for this account. As per Shameena's instructions if the limits increase again, we are to accept the increase and send an e-mail and voicemail to Neil Murdoch.
[276] In other words, even if these new higher limits are exceeded, AIC will not reject the trade but will accept it and advise Mr. Murdoch.[^92]
ii. CI Conduct
[277] CI’s conduct in relation to the switch agreements follows a pattern similar to that of AIC.
a. Entry into Switch Agreements
[278] Beginning in 1999, CI entered into switch agreements with three short-term traders: Reliable, Triangle/Credit Lyonnais/ SII and Nesbitt Burns. In addition, it allowed at least two other parties, Cambridge and TIE to trade frequently without switch agreements.
[279] Like AIC, CI submits that the switch agreements were a method of controlling short-term trading. There was no explanation at trial for why controlling frequent short-term trading was preferable to banning it altogether. CI admits that the imposition of a 2% short-term trading fee would have put an end to the practice.
[280] David Pauli, CI’s Senior Vice President, Fund Operations during the Class Period, understood that the parties with whom CI had switch agreements were frequent traders. Peter Anderson, CI’s Executive Vice President during the Class Period had the same understanding. By way of example, on October 28, 1999, Mr. Anderson wrote an email to his general counsel stating:
I spoke directly with John Platt at NB[^93] to review the market time (Triangle Investments). He told me that this was a $1.8 billion Bermuda based off-shore hedge fund. NB had completed significant investigation on the account and were satisfied that the business was legitimate.
John Platt questioned why they were buying retail mutual funds, but said that the client was aware of the higher fees than an institutional pool.
John had received the copy of the agreement and was prepared to have it signed by the appropriate person at NB. He thought there was no problem with the fee charge.
He also said that we were the toughest of all the firms to get this approved.
[281] The email discloses that the highest level of management at CI was aware of several red flags. They knew the client was a hedge fund. Mr. Anderson admitted that, at the time this email was sent, he knew that Reliable wished to engage in “frequent, short-term trading” of up to five round trips per month. CI knew an institutional pool would have given the client lower fees, but the client preferred the higher fees of a retail mutual fund.
[282] The answer that a hedge fund was aware of higher fees in a mutual fund than in an institutional pool should come as no particular surprise to an Executive Vice President of a mutual fund. What is more significant is that CI did not appear to have pursued the original question about why a hedge fund was investing in retail mutual funds to begin with.
b. Terms of the Switch Agreements
[283] The terms of CI’s Switch Agreements follow the same general pattern as those of AIC and contained terms to the following effect:
(i) They limited frequent trading to between eight and all of CI’s funds.
(ii) They limited the size of the investment to between $5 and $150 million and imposed a percentage limit of between 0.75% and 1.25% of the fund at issue.[^94]
(iii) They limited the number of round-trip switches to 5 per fund per month.
(iv) They charged a fee of 3 or 4 basis points (0.03% or 0.04%) on all switches.
(v) Where redemptions (as distinct from switches) occurred, a fee of up to 2% of the NAV of the units being redeemed could be imposed. That fee would be paid to the fund.
(vi) They allowed CI to terminate the agreements on 10 days’ notice, if CI deemed it necessary to do so to protect the best interests of the unitholders of the applicable fund.
(vii) They contained a confidentiality provision.[^95]
[284] Like the confidentiality provisions of the AIC Switch Agreements, those of the CI agreements provided:
You agree to keep the terms and conditions of the Program confidential and not to disclose them to any third party without the prior written consent of CI.
Again, the drafting suggests that confidentiality came at the request of CI. Mr. Anderson and Mr. Pauli said that including confidentiality provisions in agreements was a standard operating procedure at CI.
c. Advice
[285] CI’s executives explained at trial that they believed throughout the Class Period that the limitations and switch fees set out in the Switch Agreements adequately compensated the funds for any potential costs of the trading activity. CI did not, however, conduct any analysis to compare the cost of frequent trading to the funds with the switch fees they charged. Nor did it ever calculate dilution within any of the funds during the Class Period. Nor did it ever analyse the profits being made by the frequent traders, although as with AIC, that could have been done simply by glancing at their monthly statements.
[286] Stephen McPhail was CI’s CFO and Chief Operating Officer during the Class Period. He testified that he satisfied himself through discussions with Mr. Pauli and Douglas Jamieson (another of CI’s senior officers during the Class Period) that the switch fees more than compensated the funds for the costs of frequent trading.
[287] Mr. McPhail did not, however, see any written analysis or presentation. His view was based solely on discussions.
[288] Mr. Pauli testified that he conducted an analysis of the cost of the activities proposed in the Switch Agreements before the first one was signed. He says he satisfied himself that the Switch Agreements charged a fee that was sufficient to cover the ongoing costs to the funds. That analysis has been lost. Mr. Pauli does not know when it was lost. Nor does he remember the details of the analysis.
[289] The plaintiffs submit that an analysis of any depth would have been developed with Excel or a similar program which should have remained accessible electronically. Mr. Pauli says the analysis was communicated to Mr. Anderson and to the legal department. Had that analysis been communicated electronically, there should also be some sort of record of those communications.
[290] Whatever the reason for the analysis being inaccessible, the fact remains that the only proof at trial of any analysis within CI is the general allegation that there were discussions in which Mr. McPhail was told that the switch fees compensated the funds for the cost of frequent trading. The degree to which that analysis would exculpate CI turns on the reason for and the content of the analysis. By way of example, was the analysis restricted solely to increased transaction costs or did it also include dilution? It appears that it would not have included dilution since CI witnesses conceded that no dilution analysis was ever conducted.
[291] In addition, CI admits that whatever the content of the initial analysis, CI never compared the actual data of the frequent traders against the analysis, either before or after entering into the Switch Agreements.
[292] CI denies that it had any reason to analyse the frequent traders’ profits because it believed their trading was short-term trading market timing in the speculative, gambling sense of trying to “time the market.” CI submits that the plaintiffs’ suggestion that CI should have looked at the investors’ account statements or activity to determine whether they were earning abnormally high profits assumes that CI knew that these investors had an ability to arbitrage the CI Funds, or an ability to correctly “time” the market in a way that caused harm to the CI Funds. No one at CI knew this during the Class Period.
[293] Like AIC, CI frames the issue too narrowly. The issue is not whether CI knew or ought to have known about time zone arbitrage, the issue is whether it knew or ought to have known about dilution caused by short-term trading. CI knew or ought to have known about the concept of dilution. The fact that a hedge fund was willing to pay the higher fees associated with a retail fund in order to engage in frequent short-term trading should have led CI to raise concerns about dilution or, at the very least, investigate further.
d. Trading Patterns
[294] The amount of switching in and out of CI funds was significant. Exhibit 3A to Professor Christoffersen’s report shows approximately $35.5 billion being switched in and out of CI funds during the Class Period. Professor Christoffersen disagrees that these were all time zone market timing trades because some of the funds had large American components. As with AIC, that will be an issue for the damages trial.
[295] In January 2003, Reliable switched in and out of the CI Global Fund 5 times in amounts of between $53 and 54.8 million. On the same days, SII Limited switched amounts in and out ranging from $67.6 to $73.7 million. The Reliable and SII switches in January 2003 amounted to more than 40% of CI Global Fund’s total assets.
[296] Reliable alone switched approximately $3.9 billion in and out of the CI Global Fund, $2.6 billion in and out of the BPI Global Equity Fund, $2.7 billion in and out of the CI international balanced fund and $2.9 billion in and out of the CI Global Balanced Corporate Class fund between January 1, 2001 and December 23, 2003.
[297] The frequent traders appear to have exceeded the limits imposed on them through a variety of methods.
[298] First, at least one opened accounts under different names. SII and Reliable both belonged to Trout Trading Management Company, a single large, sophisticated, Bermudan hedge fund. SII and Reliable had virtually the same Bermuda address recorded on their client account records at CI and had begun executing identical large dollar switch-in and switch-out trades in several CI Funds on the same days.
[299] Second, the frequent traders simply ignored the limits and invested more than what was permitted under the switch agreements. The CI Switch Agreements limited switches to 1.25% of a fund’s NAV. By the end of December 2022 Trout was switching amounts equal to 8.54% of the CI Global Fund. On January 2, 2003, Reliable Capital switched into the CI Global Fund an amount equal to 3.54% of its total NAV.[^96] On January 2, 2003 SII Limited switched into the CI Global Fund an amount equal to 4.54% of its NAV.[^97]
[300] Mr. Pauli agrees that the frequent traders should not have been allowed to exceed the percentage limits provided for in the Switch Agreements.
[301] Third, as the frequent traders exceeded their limits, CI entered into new, more permissive agreements. For example, Triangle’s first switch agreement of October 26, 1999 limited frequent trades to: an aggregate of five per month across all of CI’s funds, an aggregate monetary limit of $20 million across all of CI’s funds and a limit of a 1% investment in any one fund. A further agreement dated December 8, 1999 increased the number of eligible funds to eight, allowed an aggregate investment of $40 million, and increased the number of switches from five per month across the entire CI fund family to five switches per fund per month. In November 2002 the agreement was revised again to acknowledge that Triangle had actually invested an aggregate of $150 million ($110 million in excess of previously authorized limits) and authorized this higher aggregate limit.
[302] CI’s agreements with Reliable followed a similar trend. Its initial agreement of April 6, 2000 provided for an investment of no more than 0.75% in any one of 14 named funds with a cap of $85.5 million. A further letter of June 25, 2002 increased the limits to 1.25% of the total fund assets with an overall cap of $150 million.
[303] By June 2002, agreements between CI, SII, and Reliable permitted an aggregate of $300 million to be switched into CI Funds at any one time. However, since these agreements each allowed five switches per month per fund, this meant that CI had committed to permitting up to $18 billion annually in frequent short-term trading activity with SII and Reliable alone.
[304] The accounts of Cambridge Investments Inc. and Tie Limited with whom CI had no written switch agreements show respective switches in amounts as high as $42 million and $62 million at a time.
[305] During the Class Period, CI’s Compliance Department raised the issue of overdrafts in several funds on numerous occasions. In all cases of such notices introduced at trial, the overdraft was attributable to a market timer. By way of example, internal memos on May 28, 2002 noted cash deficiencies in three different funds of $4.4 million, $47.8 million, and $81.3 million all attributed to market timers. On June 13 and 14, 2002 cash deficiencies in the International Balanced Fund reached almost 5%. Fund managers refer to the “material impact” and the “significant costs to investors” of such overdrafts. In July 2002, internal emails at CI disclose that:
The CI Pacific Fund in particular goes into overdraft solely as a result of the “market timer”. We cannot control this activity, all we can do is attempt to minimize the costs to our investors and to maximize their returns. One of the reasons why we are slow to cover these overdrafts through stock sales is that on a number of occasions we have liquidated positions at great expense to meet the overdrafts, only to be told the next day that the cash is back and that we need to reinvest it. This is a costly exercise.
This would appear to interfere with the long-term buy-and-hold strategy that CI’s prospectuses promoted as being key to the management of its funds.
[306] An email dated October 25, 2002 from Felipe Gonzalez in CI’s compliance department to Mike Gramegna, CI’s Sales Manager and Vice President of Sales and Marketing during the Class Period notes:
To cover the significant costs for unitholders of the Fund of these constant switches, and according to the prospectus, CI can charge up to 2% of the total amount switched in these cases. Please advise.
Mr. Gonzalez at least appeared not to be aware that the market timing trading had been expressly permitted in exchange for a fee of three basis points.
[307] For the reasons set out above, the entry into the Switch Agreements amounted to a further breach of the standard of care. They amounted to facilitation of conduct that the prospectuses discouraged. They were entered into either without proper analysis or, as in the case of AIC, without paying attention to the analysis that was done. They were entered into based on a gut feeling that markets were a random walk. Although the defendants may have believed the random walk theory, they never tested it against real data. Had they done so they would have seen a profitable strategy that, at a very minimum, posed serious risk of dilution to long-term unitholders. All of that fell short of a standard of care that called on the defendants to be alert to harms to long-term unitholders, be alert to frequent short-term trading and to prevent the harms it caused. If the defendants were not aware of dilutive harm, the consequences of that ignorance lie on them. If they were not aware of time zone arbitrage, the consequences of that should also lie on them. They entered into agreements that knowingly contravened the directional thrust of their prospectuses. If they did so on a theory that was wrong, that was a risk they took. In doing so they miscalculated the risks even though they were clearly aware of the risk of frequent short-term trading. Willingly incurring a risk by miscalculating the degree of danger is a form of negligence. As between the innocent long-term unitholders who played no role in the frequent trading and the defendants who held themselves out as experts and facilitated the frequent trading, the risk of miscalculation and loss must fall on the defendants.
D. General Defences
[308] The defendants raise the following defences to the claim:
(i) They complied with the standard of care.
(ii) The alleged losses were not foreseeable.
(iii) There was a lack of regulatory action.
(iv) Liquidity is an essential element of mutual funds.
(v) The defendants cannot be expected to know their unitholders’ strategy.
(vi) The defendants limited frequent trading.
(vii) The fees earned were not material.
(viii) Frequent traders earned profits like other unitholders.
(ix) Frequent trading is difficult to detect.
(x) The defendants depended on portfolio managers to detect frequent trading.
(xi) Class members were obliged but failed to commence a derivative action.
(xii) The defendants are protected by the business judgment rule.
(xiii) Auditors raised no objections to frequent trading.
I will address each defence in turn.
i. Standard of Care
[309] The defendants submit that they satisfied the standard of care during the Class Period. They argue that this is an entirely novel claim and that mutual fund managers have never been found to owe a duty of care or meet a standard of care that would require them to prevent frequent trading or time zone arbitrage. They submit that imposing liability here would be viewing the issues before me with the benefit of hindsight.
[310] I find myself unable to agree with that proposition.
[311] The fact that the defendants characterize this as a novel claim is of little importance. The point of having general principles of negligence is that they are meant to be flexible and to be adapted to new and evolving circumstances. The fact that no court before has found mutual fund managers liable for permitting time zone arbitrage or frequent trading is no barrier to doing so now. The fact that courts have never been required to do so in the past may well reflect the fact that mutual fund managers never permitted it before the Class Period or that mutual fund unitholders have never recognized it before.
[312] I agree, however, that it is important for the court to ensure that it is not finding liability with the benefit of hindsight. I am satisfied that the various sources of standards of care set out earlier in these reasons demonstrate that there were ample indicators during the Class Period which demonstrated that prudent mutual fund managers were alert to the dangers of frequent short-term trading and took steps to prevent it. The defendants themselves alerted their investors to its dangers in their prospectuses. Having alerted their own investors to the dangers of frequent short-term trading, it is difficult to understand how the court would be acting with hindsight by holding the defendants to account for the dangers they consistently pointed out during the Class Period.
[313] The defendants focus their defence on the concept of time zone arbitrage. They argue that it was not sufficiently well known