PrefBlog

On February 18 – sorry, I’ve been busy – the SEC released the RECOMMENDATIONS REGARDING REGULATORY RESPONSES
TO THE MARKET EVENTS OF MAY 6, 2010, which is the Summary Report of the Joint CFTC-SEC Advisory Committee on Emerging Regulatory Issues.

I found it rather disappointing, but this is unsurprising. On February 7 I passed on a Reuters report in which it was stated:

While “Sunshine” laws have prevented the committee from regularly meeting, [Nobel Prize-winning finance professor at New York University Robert] Engle said the subcommittee has discussed a bevy of sometimes esoteric market structure issues

In other words, they knew that their process wouldn’t withstand scrutiny, so they hashed it all out in the back rooms instead.

The report in Bloomberg harshly criticized the recommendation for the “trade-at” rule:

Individual investors could be hurt should regulators alter an equities-trading rule limiting the prices at which brokers can execute orders away from public markets, an executive at TD Ameritrade Holding Corp. said.

An eight-member committee urged the Securities and Exchange Commission in a report yesterday to adopt a restriction called a trade-at rule. It would prevent venues and brokerages from executing orders within their walls unless they improve pricing by a specified amount versus the market’s best level.
…
“I was disappointed,” Nagy, a managing director for order routing, sales and strategy at the third-largest retail brokerage by client assets, said in an interview. “The report appears to be a politically motivated stalking horse to implement the trade-at rule. A trade-at would serve to increase costs for retail investors by creating an inconsistent trading experience.”

We never see TD criticizing the regulators so heartily in Canada. The regulatory-industry complex in Canada is way too cosy, as discussed on March 15.

Be that as it may, the committee was good enough to state the purpose of the report quite clearly:

One additional, specific point of background is appropriate to mention at the outset. The broad, visible, and often controversial, topic of High Frequency Trading (HFT)— including the definition of the practice, its impact on May 6, and potentially systemic benefits and problems that arise from the growing volume of HFT participants in all of our markets—has been pervasive in our discussions and in comments received from others. Rather than detail specific recommendations about HFT in this report, steps to address issues associated with this practice are evident throughout our report.

In other words, the committee was set up to do a hatchet job on HFT and eagerly went about its task of pleasing the established players, who are most upset that arrivistes are introducing competition to their comfortable lives. This serves – unsurprisingly – as a continuation of the intellectually dishonest Flash Crash report.

Anyway, back to the report – while skipping over the recommendations that don’t interest me! – which makes a point of telling the committee’s paymasters what an excellent job they’re doing:

The Committee supports the SEC’s “naked access” rulemaking and urges the SEC to work closely with FINRA and other Exchanges with examination responsibilities to develop effective testing of sponsoring broker-dealer risk management controls and supervisory procedures.

So what’s wrong with naked access? I mean, really? The official line:

According to the SEC: “The new rule prohibits broker-dealers from providing customers with ‘unfiltered’ or ‘naked’ access to an Exchange or ATS. It also requires brokers with market access – including those who sponsor customers’ access to an Exchange or ATS – to put in place risk management controls and supervisory procedures to help prevent erroneous orders, ensure compliance with regulatory requirements, and enforce pre-set credit and capital thresholds.”

I would like to see a lot more discussion of access as an economic transaction. Say we’ve got a small firm trading its own capital (call it $10-million) as principal. Why can’t the exchanges and marketplaces offer them direct access themselves? I have no idea what the requirements are for gaining such access, but I’ll bet it involves a lot of regulatory expense and rigamarole that is completely unnecessary in such a case.

Why isn’t it happening? What are the risks? What controls can be justified? And how would it interact with the rest of regulation?

Say, for instance, I am the risk manager at Very Big Brokerage Corp. (and I mean the real risk manager, not the clown with the title). And say, there is a marketplace (“Sleazy Trading Inc.”) that I’m not happy with, in terms of counterparty risk. I’ve looked at their controls and their access requirements and come to the conclusion that if I execute a big trade that makes a lot of money for me prior to settlement, I’m taking on too much exposure to the notion that the trade won’t settle. Or maybe I have made a decision on how much exposure I’m willing to take with Sleazy, and they’re currently over the limit.

Now, I’ve got a client order to sell 10-million shares of IBM and wouldn’t you know it, there’s a good bid – the best bid – for 10-million shares at Sleazy Trading. Will the regulations allow me to ignore it? I don’t believe so. And that is a problem.

In Canada, there are strong inducements that say that each “protected marketplace” is as good as any other protected marketplace.

We also applaud the CFTC requesting comment regarding whether it is appropriate to restrict large order execution design that results in disruptive trading. In particular, we believe there are questions whether it is ever appropriate to permit large order algorithms that employ unlimited use of market orders or that permit executions at prices which are a dramatic percentage below the present market price without a pause for human review.

Accordingly:
7. The Committee recommends that the CFTC use its rulemaking authority to impose strict supervisory requirements on DCMs or FCMs that employ or sponsor firms implementing algorithmic order routing strategies and that the CFTC and the SEC carefully review the benefits and costs of directly restricting “disruptive trading activities “with respect to extremely large orders or strategies.

Note how careful they are in restricting their concerns to “large orders”. In other words Stop-Loss orders, beloved of the brokerage community because they’re so insanely profitable, are not in the scope of this recommendation.

But, as I argued in the November, 2010, edition of PrefLetter, Stop-Loss orders appear to have been the exacerbating cause that turned a sharp decline into a rout. But the sheer size of the Stop-Loss avalanche only made it into one insignificant speech – never into any official report of any kind. Ms. Schapiro’s speech was reported on PrefBlog in an update to the post The Flash Crash: The Impact of High Frequency Trading on an Electronic Market.

Perhaps the committee’s most laughable recommendation is:

We therefore believe that the Commission should consider encouraging, through incentives or regulation, persons who regularly implement marker maker strategies to maintain best buy and sell quotations which are “reasonably related to the market.”
…
We recognize that many High Frequency Traders are not even broker-dealers and therefore their compliance with quoting requirements would have to be addressed primarily through pricing incentives. We note that these incentives might be effectively interconnected with the peak load pricing discussed above.
Accordingly:
9. The Committee recommends that the SEC evaluate whether incentives or regulations can be developed to encourage persons who engage in market making strategies to regularly provide buy and sell quotations that are “reasonably related to the market.”

Earth to Committee: Market Making loses money in a directional market. There is no amount of exchange pricing incentive that can possibly counteract this fact.

The original Flash Crash report makes some useful classifications of market participants:

In order to examine what may have triggered the dynamics in the E-Mini on May 6, over 15,000 trading accounts that participated in transactions on that day were classified into six categories: Intermediaries, HFTs, Fundamental Buyers, Fundamental Sellers, Noise Traders, and Opportunistic Traders.
…
Opportunistic Traders are defined as those traders who do not fall in the other five categories. Traders in this category sometimes behave like the intermediaries (both buying and selling around a target position) and at other times behave like fundamental traders (accumulating a directional long or short position). This trading behavior is consistent with a number of trading strategies, including momentum trading, cross-market arbitrage, and other arbitrage strategies.

It seems to me that if you want to encourage tranquility of market prices, you should be concentrating on the potential for getting contra-flow orders from Opportunistic Traders, rather than market makers; and the only way I can see that being done by regulators is encouraging the development and execution of opportunistic algorithms by “real money” accounts, rather than discouraging this process.

The committee has another recommendation good for not much more than a laugh:

Accordingly:
10. The Committee recommends that the SEC and CFTC explore ways to fairly allocate the costs imposed by high levels of order cancellations, including perhaps requiring a uniform fee across all Exchange markets that is assessed based on the average of order cancellations to actual transactions effected by a market participant.

Central planning at its finest, complete with the implicit assertion that prices can only be fair if they are both uniform and approved by the central planners. If data flow from order cancellations gets to be a problem, it’s easy enough to sever connection with the offending marketplace – which should be sufficient to ensure that fees are put in place to charge the cancellers and rebate the other participants. But, oops, sorry, not possible to sever connections. One market’s as good as any other – just ask the regulators.

The “Trade-At” recommendation is number 11; the committee’s justification is:

We believe, however, that the impact of the substantial growth of internalizing and preferencing activity on the incentives to submit priced order flow to public exchange limit order books deserves further examination. While the SEC has properly concluded in the past that permitting internalization and preferencing, even accompanied by payment for order flow agreements, increases competition and potentially reduces transaction costs, we believe the dramatic growth argues for further analysis. Notable in the trading activity of May 6 was the redirection of order flow by internalizing and preferencing firms to Exchange markets during the most volatile periods of trading. While these firms provide significant liquidity during normal trading periods, they provided little to none at the peak of volatility.

The last sentence is simply so much horseshit. The original Flash Crash report makes it quite clear that orders were routed to the public exchanges only when the internalizers has provided so much liquidity that they were up to their position limits.

The recommendation simply shows the committee’s total lack of comprehension of the market maker’s role; additionally, they didn’t waste their precious Nobel Prize-winning brain power discussing – or even considering – the possibility that such requirements will make internalization less profitable, therefore (surprise!) leading to a lower allocation of capital and therefore (surprise!) increasing the odds that another market break will exhaust that capital.

Update: Public comments are available.

Update: The CME comment letter recommends that regulators keep their cotton-picking hands off algorithms:

Large orders represent demand for liquidity and that demand necessarily informs price discovery. Participants typically rely on algorithms to execute large orders today precisely because sophisticated algorithms can employ intelligent real time analytics that allow traders to significantly reduce the market impact of their orders and enhance the quality of their execution. As discussed in our previously referenced letter on this topic, we do not believe the Commissions are equipped or should be involved in regulating the design of algorithms, and should instead focus on regulating conduct that is shown to be harmful to the market.

It also points out:

CME Group does not believe that high frequency traders, however such traders are in fact defined, should be required by third parties to put their own capital at risk when it is unprofitable to do so. High frequency traders, like other independent traders who are uncompensated by the trading venue, should quote responsibly based upon their ability to responsibly manage the risks associated with the orders they place. It would be extremely irresponsible for a high frequency trader, or any other trader, to continue to operate an algorithm under conditions in which it was not designed to operate or when the inputs to the algorithm are not reliable. Doing so could potentially put the firm itself at risk and arguably subject the firm to regulatory exposure if their algorithm malfunctioned and created or exacerbated a disruption in the market.
…
Rules that would undermine a trading firm’s own risk management processes by creating affirmative trading obligations in highly volatile periods are misguided. Assuming participants in fact complied with such obligations, which they likely would not, this “cure” would simply lead to the depletion of market making capital and result in less liquid and more volatile markets.

With respect to cancellation fees:

As an initial matter, the Committee has not identified how the market will be served by this proposal or how it will enhance the stability of markets. Other than apparently seeking to impose a tax on a high frequency trading, the objective is unclear.
…
CME Group additionally employs a CME Globex Messaging Policy that is broadly designed to encourage responsible messaging practices and ensure that the trading system maintains the responsiveness and reliability that supports efficient trading. Under this policy, CME Group establishes messaging benchmarks based on a per-product volume ratio which measures the number of messages submitted to the volume executed in a given product. These benchmarks are tailored to the liquidity profile of the contract to ensure that contract liquidity is not compromised. CME Group works with firms who exceed the benchmarks to refine their messaging practices and failure to correct excessive messaging results in a surcharge billed to the clearing firm.

Knight Capital’s letter commits lese majeste by asking for data:

Many have posed the following question time and again:

“What is the quantitative and qualitative justification for taking steps to change or slow internalization?”

To date, there has been no answer offered and no credible data presented to support such a dramatic shift in market structure.

and with respect to Trade-At:

In short, there has been no qualitative or quantitative data offered to suggest that such shift in market structure is warranted. Rather, the evidence offered in support has been anecdotal at best. As a result, we strongly encourage the SEC to proceed with the same thoughtful consideration that has guided its decisions in the past. It should demand empirical data, and thoroughly vet that data before making any determination to propose such a rule.