One advantage of having a lot of money is that it makes it easier to make even more money. Money scales nicely. For instance, if you have $4.1 trillion, smart people will be happy to give you good advice about how to invest it. All they'll ask in return is that you pay them some of your money, but since you have so much of it this is not too big a burden.
Some of this advice comes from people whom you hire and pay salaries and bonuses -- sometimes bonuses in the tens of millions of dollars -- to work full-time on thinking about how to invest your money. But, financial services being what they are, some of these people work for brokerage firms and help lots of people invest their money, in exchange for "commissions" or "fees" or "trading profits" or whatever. These people take pride in providing exactly the same level of service to all of their fee-paying clients, no matter how big or how small.
No of course they don't! That would be dumb. Some of their clients have only a million dollars and so can't pay them very much (not more than a million dollars anyway!). Meanwhile you have $4.1 trillion and could pay them a million dollars without even noticing it. They are happy to take you out to lunch any time you want. Seriously you could call them right now and be like "lunch at Le Cirque at one, you're buying?" and they'll say "yes I certainly am." Their clients with only a million dollars to invest cannot do this.
Eric T. Schneiderman is the Attorney General of the State of New York and his thing is that he wants Wall Street firms to treat everyone equally. This is a weird thing to want but it is the thing he wants, and he's had some successes. Last night, for instance, he reached a settlement -- technically an "Assurance of Discontinuance" -- with BlackRock Inc., which has $4.1 trillion to invest and so got certain perks. Lunches, probably, but also a thing where BlackRock could call up research analysts at a bunch of brokerage firms and say "so, how're things," and the analysts would tell them what they thought about companies and so forth. This is a possibly useful perk, depending on whether the analysts were good at knowing and thinking things about companies, which I guess is their job though that proves nothing.
(This is a thing that analysts do, by the way. I suppose there is a popular belief that sell-side research analysts spend their days in caves writing lengthy reports that they release once a quarter or so, but that would be a weirdly easy job. They meet with clients! And talk to them on the phone! About sports, maybe? No, probably mostly about the companies they cover. One former analyst told me, "there's a reason it's called sell-side research." Your job is mostly to sell your research. Over the phone.)
In 2009, BlackRock acquired a group called Scientific Active Equities, or SAE, which currently runs about $80 billion of BlackRock's $4.1 trillion. Scientific Active Equities applied the scientific method to the process of calling up analysts and saying "so, how're things," and realized that they could save some time by sending analysts quarterly or monthly electronic surveys saying "so, how're things, on a scale of 1 to 9?" (Specifically, the survey asked analysts for nuance that couldn't be found in public reports: how likely a company was to be acquired, or whether earnings were more likely to beat the analyst's expectations or miss them.) SAE would average the analysts' responses and use them as a trading signal; "survey based signals were generally weighted approximately 5% in SAE's quantitive trading models."
So far so good, really, but there were some problems. For one thing, the analysts would sometimes tell SAE things they hadn't published yet, and according to Schneiderman, "Brokerage customers expect that the information contained in analyst research reports will not be disseminated to a few select brokerage customers prior to those reports being broadly disseminated to all brokerage customers that are entitled to receive those reports." That is a sentence that repays close study! There is a lot going on in it. One question you might ask is, is it true? It seems unlikely, if for no other reason than because brokerage customers are constantly reading stories ( here, here) about how information contained in analyst research reports is being disseminated to a few select customers prior to those reports being broadly disseminated, and at some point you have to stop pretending to be surprised.
The other problem is that BlackRock's internal memos contained a variety of bad words. For instance, here is how SAE characterized what it was up to: "we're agnostic as to whether rec[ommendations] themselves are useful investment info. We are trying to front-run rec[ommendations]." That is a no-no. Just trying to get analysts' already published insights, but in quantified scale-of-1-to-9 form: That sounds pretty much okay. Not caring about the analysts' insights but taking an opportunity to CHEAT and FRONT-RUN and TAKE UNFAIR ADVANTAGES: bad.
BlackRock has agreed to discontinue the survey and pay the $400,000 cost of Schneiderman's investigation, the sort of no-harm-no-foul solution that seems to be Schneiderman's Wall Street standard: This, like the early disclosure of the Michigan consumer survey, isn't exactly illegal, so nobody agrees to pay any fines, but it's creepy enough that they'll agree to stop doing it when pressed. The "Assurance of Discontinuance" is interesting too in what BlackRock agrees to: Not just to stop this specific survey, but also to "cease and desist from"
conducting any systematic survey of analysts in order to gather non-public issuer specific analyst sentiments.
BlackRock will not use any other systematic effort to gather material non-public information in order to make investment decisions or recommendations based upon non-public issuer specific analyst sentiment.
So the big question: Can they still call up analysts and say, "so, how're things?" Or are they limited to reading analysts' published research reports? I'm going to assume that the word "systematic" there means that BlackRock can still do what every other big investment firm does, and talk directly to research analysts to try to get their insights. (And on that, BlackRock being BlackRock, the analysts will be happy to oblige.) They just can't automate the process. They're still allowed to talk to analysts, and take advantage of the privilege of being one of the biggest investors in the world. There are just some limits now to their economies of scale.
First notice the artful "information contained in analyst research reports." That's not "customers expect analysts will never tell favored clients anything before publishing it"; it's just "customers expect analysts will never tell favored clients anything that goes into a report before publishing it." So analysts can have general conversations about their coverage companies with favored clients; they just can't preview their reports. This is an unbelievably murky area. Goldman got in trouble for its "trading huddles," which the SEC thought were unfair dealing but which Goldman said were just short-term trading ideas totally separate from its analyst research reports.
Second, notice the "all brokerage customers that are entitled to receive those reports." What entitles you to receive those reports? I think the answer is, "the brokerage firm has decided that you are entitled to receive those reports." Brokerage firms don't have to disclose their research broadly! For one thing, most brokerages disclose most of their research only to brokerage customers. So if you're BlackRock, you're a customer of every brokerage (or at least they want you to be), and you can get every firm's research. If you're some schlub, you have at most, what, one or two brokerage accounts? Even those accounts don't guarantee you all of a bank's research: If you've got a small equities account you can't necessarily get their credit research, even if it's relevant to your stock trading. There's a lot of wiggle room in that notion of customers "entitled to receive those reports." Schneiderman's thesis seems to be that brokerages can decide who gets their research in their sole discretion, but they can only have one tier: You can either get research or not get research, but you can't get research faster than anyone else who gets research. This is similar to his views on Reuters, but in both cases it seems arbitrary.
Finally, notice that Schneiderman says "Brokerage customers expect ...", not "It is the law that ..." And there's no citation to any rule that forbids disseminating reports "to a few select brokerage customers prior to those reports being broadly disseminated." Is there such a rule? Is the law actually what Schneiderman says it is? I ... I do not know. When Goldman got in trouble for its huddles, the SEC cited Section 15(g) of the Exchange Act, which requires brokers to have policies "to prevent the misuse in violation of this chapter, or the rules or regulations thereunder, of material, nonpublic information by such broker or dealer or any person associated with such broker or dealer." But it didn't cite any rules or cases saying that selective disclosure of trade ideas was such a "misuse" of "material, nonpublic information." (The Wall Street Journal mentioned, in this connection, rules requiring "fair dealing with customers," but again that's not the same as a specific rule saying "you gotta tell all your customers the same thing at the same time.") Or when Morgan Stanley got in trouble for helping Facebook selectively disclose changing estimates to favored research customers, the insane Massachusetts consent order (it's nuts, trust me, more here) cited mostly generalities about "unethical or dishonest conduct or practices," not specific rules against selective disclosure.
Perhaps someone can put me wise to regulations or case law -- as opposed to settlements or SEC/attorney general positions -- against selective disclosure of brokerage research, but I have to say the law does not seem as clear to me as it does to Schneiderman and friends. (Don't say "insider trading," either: It's only insider trading if the information is disclosed in violation of a duty of confidentiality, and what we're discussing here is whether that duty exists.)
That's also pretty much how BlackRock described the survey publicly, and in the survey itself. So "The introduction to the surveys provided: 'We would like to highlight the fact that we are only interested in public information. Please only share with us information that you publish through your research notes, investor calls, and/or disclose in client meetings.' "
But, again, notice that "and/or disclose in client meetings"! The work of research analysts is not disseminated only in published research notes! Notice that Schneiderman is not going after BlackRock for going to meetings with analysts. He's only going after them for the survey.
To contact the author on this story:
Matt Levine at firstname.lastname@example.org