The Ontario Securities Commission's Mar. 25 decision in its case against former Bay Street lawyer Mitchell Finkelstein, his friend Paul Azeff and associates Korin Bobrow, Howard Miller and Francis Cheng, was the kind of clear, convincing and cogent victory in an insider trading hearing that has been desperately hard to come by recently.
The standard for insider trading is a simple one: did the accused person make a trade in a reporting issuer, based on non-public information, obtained from someone that he knew or ought to have known was in a special relationship with the reporting issuer? Similarly, the standard for tipping requires that the person in a special relationship communicate material non-public information to another person outside of the necessary course of business (for those nerds here, note that neither the tipping nor insider trading standard in Canada requires a personal benefit to the tipper like in the U.S. Hence, Mr. Finkelstein's tin boxes jammed with twenties or bank deposits didn't play a major role).
The problem faced by Securities Commission staff members attempting to prosecute insider trading is set out by the Alberta Court of Appeal's decision in Walton v. Alberta Securities Commission. Commissions must rely on inferences to find out what happens at the moment of the tip; due to limitations on the investigative powers of securities commissions, tipping and insider trading cases in Canada rarely pan out like a legal version of The Jinx, with Robert Durst in a bathroom, muttering possible confessions. In Walton, which was recently given tacit approval as the law of the land when the Supreme Court of Canada declined to review it, the ACA found that, while circumstantial evidence was permissible, the inferences the ASC drew in Walton were too speculative to sustain a conviction. This seemed to some, including me, to suggest that insider trading convictions were being held to a slightly higher standard of proof than the traditional civil burden of a "balance of probabilities." In order to garner an insider trading conviction, commission staff must navigate Walton's evidentiary thicket.
OSC commissioner Alan Lenczner and the panel were well aware of this when crafting their decision last week. They frame their discussion of circumstantial evidence as one of logic versus speculation. Proper inferences drawn from circumstantial evidence must "flow naturally and logically from the established facts," but "inferences drawn from speculated facts are legally unacceptable." Convictions require that the decision makers find that a "speculative fact" – namely, that a ninety second phone call contained within it a tip about a pending transaction – is more likely than not to be true. Convictions further down the line hinge on drawing an inference from a speculative fact: assuming that it is true that subsequent traders received a tip, the existence of which is a speculative fact, can we infer that subsequent traders "knew or should have known" that such a fact came from someone in a special relationship with the reporting issuer? Despite what the panel says, insider trading convictions almost always require drawing inferences from speculative facts.
The panel's statement that conclusions from circumstantial evidence must flow naturally and logically from the established facts isn't about formal logic – this is not some undergraduate philosophy seminar discussing What the Tortoise said to Achilles (hint: it was not a hot stock tip) – it's about good sense. The way out of the finger trap the ACA set in Walton is close, factual analysis. That is where the rubber hits the road. So, instead of spending much time with the law of insider trading, the panel dispenses with the purely legal questions in a brisk six pages, and gets on with the business of close, factual analysis in light of what they see as a pretty solid legal standard (though, I would say, for those so inclined, the discussion of the "special relationship" standard is illuminating).
Mr. Lenczner and the panel find evidence of insider trading where the factual leaps are small and the incriminating assumptions are the assumptions that make the most sense given the facts. For example, the panel finds that in the Masonite International part of the investigation, where Mr. Finkelstein was accused of tipping Mr. Azeff that U.S. private equity firm Kohlberg Kravis Roberts & Co. was planning on acquiring Masonite, the most likely set of assumed facts were unfavourable to Mr. Finkelstein, Mr. Azeff and all. The (likely) acquisition of material non-public information by Mr. Finkelstein was followed by a spurt of communications between Mr. Finkelstein and Mr. Azeff, which was followed by a spurt of trading in a stock that Mr. Azeff and his associates had not been involved in before, which was followed by further trading down the line and communications that suggested that the information was based on a tip.
Contrast that with another part of the case, the IPC U.S. REIT transaction, where the OSC's staff's allegations hinged on a single phone call from Mr. Azeff to Mr. Finkelstein when Mr. Finkelstein claimed to be in a board meeting. Even with the subsequent purchase of shares on behalf of Mr. Azeff and Mr. Bobrow's family and friends the day after the call, the panel found that the inferential leap was a bit too far. Yes, the transaction does look a bit fishy. But no, they aren't straightforward assumptions; they require some mental gymnastics.
This is the theme of the decision. Each time the panel finds insider trading, the impugned events are packed tightly together and the easiest explanation is that there was insider trading going on. Where they fail to find wrongdoing, it would require some relatively large assumptions to get to liability. All factual assumptions are necessarily speculative; to insulate the decision against the almost inevitable appeal, and to stay on the right side of the Walton decision, the panel required a tight factual matrix to find liability.
So, was the panel applying a higher standard? I don't think so. The test – clear, convincing and cogent evidence that the events were more likely than not to have happened – is a different test from "more likely than not" insofar as it has two parts. So, even if the panel had a feeling that some instances were "more likely than not" insider trading, if the evidence wasn't good enough, they can't find it. Law is not just about rules but also about facts, and those facts have to be good ones. And, at least if you believe the panel, there were plenty of facts in this case that were good enough.
Adrian Myers is a lawyer at Torkin Manes LLP.