Wall Street… Compliance… Profits…. Could better analytics help?

Yesterday, the WSJ reported that the SEC is looking into Merrill Lynch’s trading activity for potential front-running violations. To quote the article

The trading probe is a broad look at the relationship between big, institutional investors and the brokerage house. Specifically, one area of inquiry involves whether certain Merrill employees improperly stepped in front of orders placed by Fidelity Investments, the large mutual-fund operator, these people said. The period under scrutiny covers 2002 through 2005.
[...]
The practice is known as “front-running,” and previous regulatory scrutiny has resulted in regulatory fines and changes in industry practice. It gives an unfair advantage to traders because orders from big investment houses such as Fidelity often move stock prices.

For example, if a trader received an order from an institutional investor to buy stock, the trader could then step ahead of the order to buy shares for the house account that could then be sold to Fidelity at a higher price, locking in a profit.

While I have no insight in this case in particular, and know that all firms watch for front-running as part of their standard surveillance activities, I do wonder if better business analytics could be useful in finding and rooting out potential regulatory issues before the regulator finds them. Main Street firms such as FedEx, Apple, Dell, and Wal-Mart routinely use business analytics to make business decisions, to improve business processes, to streamline logistics, to target customers, to influence customer behavior, and ultimately to generate income and profits.

Why doesn’t Wall Street follow Main Street’s example in their use of business analytics? Why doesn’t Wall Street use business simulations to model the behavior of traders and trading floors? As the business of Wall Street is more and more automated, Wall Street firms are not following Main Street’s example and building analytics and simulations. Why not? The tools are there. The data is there. The technology is mature. Is it that Wall Street trading houses are still focused on the day-to-day trading P&L and are unwilling to take a step back to look at ways to incorporate best practices from Main Street to their processes? Is it ego? Wall Street is not some mysterious dark place where the lessons of business analytics, simulation, supply chain, and customer experience do not apply. It is a place where these lessons have not been applied.

What a shame. An NGE would look to use business analytics and simulation as part of its standard day-to-day work.

FINRA fined BD’s for advertised trade violations

// Rant on

FINRA announced this week that 19 firms were being fined a total of $2.8 million for inaccurate advertising of trade volumes.

So what were these firms doing? As a matter of background, sell-side equity trading firms routinely use advertising services from Thomson Financial, Bloomberg, Reuters, and a few others to inform their institutional customers what activity is going on at that sell-side firm. Specifically, there a 3 types of advertising:

(1) Advertised Trades (“AT”) – AT messages tell the buy-side what volume that sell-side firm has traded today in a given security. The theory is if you are a big player in a stock today, then you are more likely to be in contact with buyers and sellers, so the buy-side might be more inclined to send order flow in that stock to you.

(2) Generic Indications of Interest (“IOI”) – IOI messages state the posture the sell-side firm is representing to their customers. These are message like “large buyer of GOOG”, “medium seller of DIS”, “two-way large KO”. IMHO, Generic IOI messages are of limited value beyond potentially putting your name on the customers screen.

(3) Specific Indications of Interest (“Super Messages” and “Natural IOIs”) – Supers and Natural IOIs are messages with a specific price and share amount. For example, buyer of 50,000 FITB. Supers and Natural IOIs are identical in structure with the only difference being by sending Natural IOIs you are representing to the customers that you have a bona-fide customer order behind this advertisement.

Who sees these advertisements? Institutional firms subscribe to the vendors receive these advertisements from the sell-side firms. They used these messages as part of the mix in making order routing decisions during each trading day.

Are these any other uses for this data? For IOIs, no. These messages are for the trade date on which they were sent. For ATs, yes. The vendors aggregate all of the volume sent by the firms and use this data to produce ranking reports. These reports are purchased by both buy-side and sell-side firms to understand who are the consistently large players in given securities.

This is a business, and a big one at that… The sending and receiving of ATs and IOIs is a big business and it is not uncommon for sell-side and buy-side firms each to pay over $100,000 per month in fees to the various advertising vendors. John Q. Public does not see these advertisements (and the share quantities of the advertisements are in terms of thousands of shares, and John Q. Public is not trading in these increments).

What were the firms fined this week doing? (as a disclaimer before I continue, I have not worked at any of the firms fined, so my comments in this section are from my general understanding of practices around the street). The firms cited this week were inflating their volume when sending ATs. For example, some traders who were usually big players in a stock, but in a given day were not, would advertise that they were… To put some numbers to this practice, if these traders normally were 5% of the volume a given stock, but had not traded in that name at all today, might send an AT out that they were 5% of the volume just to appear to be involved. Other firms built systems that kept track of the volume traded and would automatically send ATs for the traders. In some cases, firms would have these systems round up volume sent during the day to appear larger in that stock that they really were.

What rule did these firms violate? There are no regulations directly regarding sending of ATs and IOIs. The rule that the regulators used to justify the fines was that firms must conduct themselves in a just and equitable manner. The regulators started looking at this in August 2006 and released Notice to Members 06-50 which describes the supervisory responsibilities firms have with respect to the sending of ATs and IOIs. From this NTM the regulators state:

The communication of untruthful, inaccurate or misleading information would be
considered conduct inconsistent with high standards of commercial honor and just and
equitable principles of trade.1 In addition, depending on the nature and content of the
communication, such communications may also violate NASD Rule 3310 (Publication of
Transactions and Quotations) and IM-3310 (Manipulative and Deceptive Quotations), as
well as Rule 2120 (Use of Manipulative, Deceptive or Other Fraudulent Devices), Rule
2210 (Communications with the Public) and the anti-fraud provisions of the federal
securities laws.

So who was harmed? The fact that ATs were inflated by many firms was common knowledge by both sell-side and buy-side firms. The buyers of these advertisements (the clients of the sell-side firms) would penalize sell-side firms who were egregious in their inflating their AT numbers. How? They would not trade with them for a period of time, if ever. That hurt immediately where it counts.

So why is my dander up a bit and why put this post out there? I cannot argue that some ATs were inflated. I cannot argue that some firms were worse in this practice than others. I cannot argue there were some traders who had patterns of sending ATs that were more inflated. What I wonder is why the regulators needed to get involved and fine 19 firms. Why wasn’t the option by the buy-side to put firms in the “penalty box” enough as it had been up until September 2006? Where was John Q. Public harmed? Professional, institutional firms were paying for this data. If they felt the data was useless, they could choose to stop paying for it! Lastly, why fine the firms for their advertising in August 2006 when the guidance on how to supervise the sending of ATs was not issued until September 2006.

In the end, the regulators did clean up a practice that was a dirty, and that is clearly a good thing (despite my dander being up!). Firms have put more automation, supervision, and controls around the sending ATs. That makes sense. Some firms have gone so far as to study the true effectiveness of this kind of advertising versus the costs, and in the cases I was involved in this analysis have found the results enlightening to say the least.

// Rant off