Last week, the New York Stock Exchange filed to decommission its Liquidity Replenishment Point functionality, which it introduced in 2006 to help dampen volatility in NYSE-listed stocks. According to the filing, the NYSE is taking this action at the behest of the Securities and Exchange Commission, in connection with the new Limit-Up-Limit-Down (LULD) system for preventing erroneous trades that began its first phase of implementation on all US execution venues today. LULD is replacing the single-stock circuit breakers that were introduced following the May 6, 2010 “flash crash,” during which major US equity indexes dropped several percentage points within minutes, only to recover most of the lost ground just as swiftly.
The disappearance of LRPs is of concern for two reasons.
First, it leaves a significant segment of the market less protected from volatility shocks until later this year. Like the single-stock circuit breakers, LULD during Phase I will not be in effect during the first 15 minutes and the last 30 minutes of each trading day. LRPs, on the other hand, take effect immediately after the opening print in each NYSE-listed stock, and remain active until (but do not apply to) the closing print. As a result, starting today, the NYSE-listed stocks subject to Phase I of LULD will not have volatility protections in place from 9:30-9:45 and 3:30-4:00.
Additionally, forcing LRPs to die raises the question of whether exchanges have sufficient regulatory flexibility to offer innovations that improve the quality and efficiency of US equity markets.
In approving the LULD plan, for example, the SEC expressed concern about “the potential for unnecessary complexity” should volatility-dampening tools at individual exchanges overlap with LULD. It further noted that it expected exchanges to end such plans once LULD went live.
But consider how LRPs performed on the day of the flash crash, which triggered the development of the single-stock circuit breakers and LULD in the first place. In a word, they worked. Unlike other, purely electronic exchanges that broke trades in hundreds of stocks, NYSE did not cancel a single transaction executed during the flash crash. That’s because LRPs slowed down the market when prices started swinging wildly. So what was the problem? Under Regulation NMS, the NYSE is considered a “slow” market when an LRP is triggered and human beings on the floor attempt to discover the appropriate price for the security in question. That meant its quotes were no longer subject to trade-through protection, and brokers were free to bypass NYSE and burn through the already thin books on all-electronic exchanges, despite this action obviously running counter to their client best-execution obligations. This is one reason why such blue-chip stocks as ACN traded at absurd prices, as low as $0.01, on May 6, 2010.
Clearly, the lack of a market-wide system of volatility protection did not serve the market well on May 6, 2010. But LRPs are narrower than both the single-stock circuit breakers and the LULD bands, and have coexisted with the single-stock breakers for several months. In its filing, the NYSE voices disappointment that it is being cajoled to remove LRPs without any analysis of how they perform alongside the LULD bands. Allowing them to remain in place temporarily and studying how they behave in conjunction with LULD, for example, could have been one alternative course.
This raises larger, more existential questions regarding modern market structure and regulation. Today, price discovery in the US equity market is split among 13 licensed exchanges and a couple of small ECNs. Liquidity is further fragmented across about two dozen major dark pools, most of which are licensed ATSs, and a handful of off-exchange market makers that execute virtually all of the country’s marketable retail orders. Why allow such competition in principle, but then prevent exchanges from distinguishing themselves to traders and issuers by offering additional protections against violent market swings beyond the market-wide LULD system? To be sure, we need some base level of protection that applies across all exchanges, but if one exchange wants to slap on a pair of suspenders in addition to wearing the industry-wide belt, why not let it?
A common lament among market-structure critics is that exchanges mostly offer the same structures and rule sets, with fee schedules being the only area of distinction among them. Such a system, this thinking goes, creates undue fragmentation and complexity that benefits brokers and high-frequency traders, who have the time and resources to understand and exploit the Byzantine structure, at the expense of end investors, who don’t. Is forcing LRPs to disappear merely one step by regulators in a broader effort to encourage uniformity at the expense of the vigorous competition that yields innovation and efficiency and improves the market as a whole?
At the very least, the way this transition is being managed opens the industry to a greater risk that a disruptive event will occur in the NYSE-listed stocks subject to Phase I during either the first 15 minutes or the last 30 minutes of a trading day. It’s impossible to precisely quantify the magnitude of that risk, and we certainly hope it does not occur. But the open and close are already the highest-volume segments of the day, and particularly active days in which the closing trade is especially large, such as the upcoming annual Russell indexes rebalancing in June, may be susceptible to disruption.
Disclaimer: The information contained in this communication is not intended as an offer or solicitation for the purchase or sale of any securities, futures, options, or any other investment product. This communication is not research, and does not contain enough information on which to make an investment decision. The information herein has been obtained from various sources. We do not guarantee its accuracy. Any opinion offered herein reflects Rosenblatt Securities Inc. current judgment and may change without notice. Member NYSE, SIPC, FINRA.