March 31, 2022 To Deal or Not to Deal, That is the Question George Bollenbacher G. M. Bollenbacher& Co., Ltd. TabbFORUM Editor at Large Follow | Profile | More Share The SEC's proposed rule redefining the term "securities dealer" raises more questions than it answers. Market Structure Analyst George Bollenbacher examines the language of the rule and the evolution of the securities markets, and looks at some unintended possible outcomes. On March 28th the SEC released its long-expected proposed rule on redefining the definition of “dealer” in the securities markets, and in the Treasury securities markets in particular. Even before it was issued, this proposed rule has generated considerable speculation, and it has already generated some discussion, especially around a possible applicability in the area known as DeFi. As usual, understanding its impact, as proposed, necessitates understanding both the detailed language and the ever-morphing nature of the fixed-income markets. First, let’s look at the evolution of the securities markets, which have all followed the same path, albeit at different speeds. All the markets, even fixed income, started as physical exchanges, where liquidity makers and liquidity takers were easily distinguishable. In many cases the makers had monopolies on that position, which advantage was regarded as compensation for their commitment to provide liquidity in all market conditions. Either very quickly or over many years, most markets migrated to a more free-form structure, where liquidity takers arrived at a venue looking for makers to transact with. Those venues could be telephonic, electronic or even a commodity pit, but it was always clear to everyone who was acting as the maker and who was the taker. In these environments the information advantage that made liquidity-making profitable was not mandated, but it existed nonetheless. Two modern developments have succeeded in muddying the maker-taker waters considerably. The first is the evolution of algorithmic trading and the second is the proliferation of electronic venues. Looked at from the taker’s viewpoint, many trades begin with takers’ orders that are generated internally, and handled externally, entirely by algos, sent to an electronically-chosen venue(s) where they encounter either another taker’s order or a maker’s standing bid or offer. The trade is executed electronically and booked on both sides without any human intervention at all. In particular, the other side of the taker’s order may not be acting as a liquidity maker at that moment – it may be another buy-side order at just the right time, or an algorithmic trader that isn’t providing liquidity at all. As a result, in many markets and many instruments, we may not be able to distinguish the intentions or business models of most of the participants in any particular venue at any point in time. A firm that is in the market to take advantage of very short term opportunities (i.e., short term volatility), and prepared to exit the market if it sees those opportunities curtailed, even temporarily, might be acting to absorb liquidity one nanosecond, and to supply it the next. If we define a dealer in terms of a liquidity maker, firms can jump into and out of that category hundreds of times a second. Into this confusing and still changing picture comes the SEC’s intention to require registration of firms which they say, “play an increasingly significant liquidity providing role in overall trading and market activity—a role that has traditionally been performed by entities regulated as dealers.” They go on to say that, “The proliferation of fully electronic trading venues has been accompanied by the rise of certain market participants who are not registered as dealers and who today account for a majority of trading in the Treasury interdealer market.” So they make the distinction that any firm trading for its own account as a line of business is a dealer, and must register with them, without regard to whether that firm is making or taking liquidity. In keeping with this approach, the SEC specifically defines a dealer as, “Any person that meets the activity-based standards identified in the Proposed Rules…As the Proposed Rules focus on activity rather than label or status, they [the Rules] would potentially scope in other market participants as discussed below in Section V, thereby triggering a registration requirement and subjecting those entities to dealer regulation and oversight.” As a precaution, the rule does restrict itself to firms (or individuals) that control more than $50 million of assets, but they do acknowledge that, “under certain circumstances, a registered investment adviser could trigger application of the Proposed Rules because of aggregating trading in its own account with client accounts it controls.” To be clear, an important word throughout the rule is “routine,” indicating that the candidate needs to do this trading routinely. Except that there’s no definition that I can find for “routine.” Which raises some interesting questions. If the trigger for registering as a dealer is activity-based, what level or pattern of activity will be considered routine and generate both a dealer status and a registration requirement? Is it expressed in absolute volume, or a percent of total volume? Is it across all venues or only one? For how long and how continuously? Does posting bids and offers trigger the status, even if trading volumes are modest? Will the trigger volume or activity pattern stay the same over time, or could the SEC change it whenever they want? And there are also questions about what dealer status and registration means. If an asset manager does enough trades in customer accounts to trigger the status, does that make its clients, who are the actual parties to the trades, dealers? If a firm does sufficient business to trigger the requirements for, say, two hours one day a month, is it now a dealer? If so, for how long does the status last? If not, for how long and how frequently would such trading trigger the dreaded “dealer trap”? Since the proposed rule does not have any hard and fast criteria, including the length of time dealer status is in place, there may be a very strong incentive for high frequency traders to alter their trading patterns to obviate the rule’s criteria. If trading algos can be designed to incorporate large amounts of market data very rapidly and optimize trading strategies in real time, it won’t be hard to write algos designed to skate very close to the SEC dealer trigger without tripping it. And if we have multiple HFTs utilizing the same algos to avoid the “dealer trap,” will that actually exacerbate market volatility? Having five HFTs pull out of all the venues in a market at the same instant because their algos picked up the same signal will not lead to what the SEC thinks it is searching for. In the end, we may be past the point of any such dealer distinction, particularly because the language of the rule itself ignores the function of, and long-term importance of, the liquidity maker. If we pile a cost onto the function of providing liquidity without adding any information advantage, we may work our way back to a market where liquidity takers meet at the post in the hopes that some other taker will want the other side of their trade. I’m not sure that’s what the SEC had in mind. Photo Credit: “New York Stock Exchange” by BlatantWorld.com is marked with CC BY 2.0. • • • • George Bollenbacher is President of G. M. Bollenbacher & Co., Ltd. He spent 20 years as a bond trader and 10 years in the technology business. For the last 20 years he has assisted banks, asset managers, and custodians in implementing process and technology changes. He is the author of The Professional’s Guide to the US Government Securities Market and The New Business of Banking, as well as many articles on the evolution of the financial markets. TabbFORUM is an open community that provides a platform for capital markets professionals to share their ideas and thought leadership with their peers. The views and opinions expressed are solely those of the author(s). They do not necessarily reflect the opinions of TABB Group, its analysts, TabbFORUM and its editors, or their employees, affiliates and partners.
Oh gosh, I wished more financial journalists worked at least for a few years at Bloomberg news. Then they would learn how to summarize the story in 2 or 3 sentences and present the key takeaways first before blabbering about the history of the maker taker model. Horrible journalism.
The meaning of RULE is a prescribed guide for conduct or action. define rule - https://www.dogpile.com/serp?q=define+rule This article describes a situation rife with hornswaggle and balderdash. define hornswaggle - https://www.dogpile.com/serp?q=define+hornswaggle define balderdash - https://www.dogpile.com/serp?q=define+balderdash define devious scheme - https://www.dogpile.com/serp?q=define+devious+scheme define corruption - https://www.dogpile.com/serp?q=define+corruption define influence peddling - https://www.dogpile.com/serp?q=define+influence+peddling
Taken from that article up there, would you say this quote below is a true statement? "the proposed rule does not have any hard and fast criteria,"