February 05, 2013
Why Increasing Tick Sizes May Lead to More Dark Pools Instead of More IPOs
Posted by Bobby Bartlett

Today, the SEC will convene a much-anticipated roundtable examining the current regime of penny-priced tick sizes on U.S. stock markets. A principal purpose of the roundtable is to explore whether the transition to penny-priced quotations in 2001 (known as “decimalization”) has harmed liquidity in the securities of small and middle-sized companies. The general theory, initially advanced in this Grant Thornton white paper, is that when securities were quoted in sixteenths of a dollar, trading spreads were kept artificially wide given the fact that bid-ask spreads could be no less than $0.0625 per share, creating large profit margins for dealers making markets in U.S. equities. Such market-making profits, so the argument goes, were then used to support trading operations and analyst research in thinly-traded securities and the securities of newly-public firms. In this fashion, the argument continues, the higher trading costs associated with fractional-quoting were actually part of a healthy ecosystem for nurturing the market for IPO stocks and smaller company securities more generally.

According to this theory, one way to bring back the IPO market is to undo the harm decimalization caused this ecosystem by increasing the tick size, or minimum price variation (MPV), when quoting the securities of smaller issuers. It is an argument that has gained considerable support over the past year, as reflected in both Section 106(b) of the JOBS Act (which required the SEC to study the effects of decimalization on the liquidity of smaller firms) and the draft recommendations of the SEC Advisory Committee on Small and Emerging Companies (which recommends a “meaningful increase in tick size as a necessary step toward encouraging the reestablishment of an infrastructure designed to increase liquidity for small public companies.”) And based on the agenda for today’s roundtable, a reasonable bet would be to see some form of pilot study being implemented in which the securities of certain firms must be quoted in an increment greater than $0.01.

While it is heartening to see the SEC take an empirically-driven approach to capital market reform, new research by Justin McCrary and myself underscores the need for the SEC to assess this issue especially carefully and for any policy changes to take place within an incremental framework. In our working paper, Shall We Haggle in Pennies at the Speed of Light or in Nickels in the Dark? How Minimum Price Variation Regulates High Frequency Trading and Dark Liquidity, we document how modification of the penny-based system of stock trading will likely have simultaneous and opposite effects on the incidence of both high frequency trading (HFT) and the trading of undisplayed (or “dark”) liquidity (what we refer to as “trading hidden liquidity” or THL). Specifically, in the event of an increase in the MPV, our research strongly suggests we can expect to see both an increase in off-exchange trading in venues such as dark pools and a decrease in HFT.

Although often conflated within the popular press, HFT and THL reflect two distinct types of trading strategies that have distinct consequences for price discovery and market liquidity. In terms of strategy, traders focusing on HFT typically seek to profit from discrete, short-lived pricing inefficiencies by rapidly bidding on and selling securities, customarily through pre-programmed algorithms. The emergence of so-called “maker/taker” fee structures at stock exchanges—whereby limit order providers are paid a “maker” rebate and traders using market orders are assessed a “taker” fee—creates an additional profit opportunity for such traders provided they can position their limit orders at the top of exchanges’ order books. For firms engaged in HFT, minimizing the latency of processing information and entering orders is therefore of paramount importance to profitability. In contrast, a firm focusing on THL will generally seek to profit by providing liquidity to investors without the necessity of publishing public bids or paying exchange access fees, thus minimizing the price impact and cost of the transaction. Access to investors looking for liquidity—rather than speed of trading per se—is accordingly a primary goal of those engaged in THL.

Despite the recent focus on changing tick sizes, there has been remarkably little focus on how each of these strategies is intimately connected with the rules governing the MPV. As was revealed following decimalization, smaller tick sizes have led to both a surge in market message traffic as prices dispersed across more price points as well as a dramatic reduction in quoted spreads. Both developments favor algorithmic trading strategies capable of processing quickly large flows of order messages, while reducing the costs of rapidly trading in and out of positions. With respect to THL, larger tick sizes create the opportunity for larger spreads and, consequently, larger profits for those firms that can capture them by trading against marketable orders from individuals and institutions seeking immediate liquidity. In this regard, dark pools and broker-dealer internalizers are aided by a technical rule concerning how the penny-pricing requirement is actually implemented: Although it is prohibited for anyone to quote (i.e., post an order) at other than a penny-increment, it is perfectly fine to execute a trade in subpenny increments. Using this flexibility to execute subpenny trades, a dark pool or internalizer can thus offer price improvement over the National Best Bid or Offer (NBBO) available at conventional stock exchanges (even if the improvement is as little as $0.0001 per share). In this fashion, dark pools and broker-dealer internalizers have both the incentive and the means to trade directly with incoming marketable orders rather than route them to exchanges.

To examine empirically how changes in the MPV might have these effects on the incidence of HFT and THL we turned to a peculiar quirk in the ban on sub-penny pricing described in the previous paragraph. In particular, the ban on sub-penny quotations (Rule 612 of Regulation NMS) only applies to equity orders priced at or above $1.00 per share, thus creating a sharp distinction in tick size regulation between those orders priced just above $1.00 per share and those priced just below it. Using a regression discontinuity (RD) research design, we can therefore identify in a clean, parsimonious way how changes in tick size regulations can affect the incidence of these two forms of trading. For our data, we used the NYSE’s Trade and Quote (TAQ) database, focusing on the 300 million trades and the 3 billion updates to exchanges’ best published bids and asks made during 2011 for securities that traded below $2.00 per share at some point in 2011.

Overall, our results are strongly consistent with the hypothesized effect of MPV on both THL and HFT. To measure THL, we examined for each completed trade the market center at which it occurred, using trades reported to a FINRA Trade Reporting Facility (TRF) as our measure for THL. The figure below presents our RD estimates of the effect of subpenny quoting on the incidence of such off-exchange trading. In the figure, the x-axis represents a transaction’s reported trade price truncated to two-decimal places, and the circles represent the fraction of all reported trades reported to a FINRA TRF at each such price. The solid line plots fitted values from a regression of the fraction of TRF trades on a fourth-order polynomial in two-decimal price (the point estimate and standard error are in the legend). As the figure indicates, trades executed at prices immediately above $1.00 per share revealed a sharp increase of 8 percentage points in the percent of trades reported to a TRF facility. Because all other market centers reflect stock exchanges, this translates to a corresponding decrease of 8 percentage points in the incidence of transactions on the public exchanges.

Fig1

With respect to HFT, we examined (among other things) the incidence of “strategic runs” within the quotation data at each price point truncated to two-decimal places. Notably, the TAQ data does not permit tracking individual orders since it covers only updates to each exchange’s best bid or offer (BBO), but evidence of such strategic runs nevertheless appears in the TAQ data to the extent they affect an exchange’s BBO, which is continually updated by the exchanges to reflect the new orders that change it. Accordingly, we measure for each second of the trading day the rate of BBO updates for each security in our sample (a “security-second”). As might be expected in the (for modern financial markets) relatively quiet corner of penny stocks, the vast majority of security-seconds experienced no update of an exchange’s BBO. In particular, over 90% of the security-seconds in the sample showed no BBO updates, with higher-priced orders generally being more likely to have at least one BBO update per second. As shown in the figure below, RD analysis of security-seconds having at least one BBO update by two-decimal order price reveals that this trend was generally continuous at the $1.00 cut-off.

Fig2

In contrast, analysis of those security-seconds where a BBO was updated with significant frequency reveals a sharp increase in the incidence of such strategic runs below the cut-off. The next figure, for instance, provides our RD estimates for the incidence of security-seconds where the BBO was updated at least fifty times per second. Consistent with the previous figure, the rate of these strategic runs generally declines from $2.00 to $1.00 where it reveals a discontinuous upward jump from .02% of all security-seconds to .1% of all security seconds, highlighting the negative relationship between the size of the MPV and the incidence of HFT.

Fig3

In sum, these findings suggest that current proposals to increase the MPV may very well entail significant, unanticipated structural changes in the nature of how equity trading occurs on U.S. markets. To be sure, many of these changes in trading such as the higher incidence of THL would actually be consistent with a core objective of Section 106(b) of the JOBS Act insofar as they would increase the profitability of market-making in affected stocks. However, our finding that these market-making profits are generally captured by dark pools and internalizers causes us to question how these enhanced profits will translate into additional analyst coverage and sales support for emerging growth companies. For instance, most dark pools and the two largest internalizers by volume—Citadel Investments and Knight Capital—do not offer sell-side analysis or advisory services. Moreover, the new retail price improvement (RPI) programs at major U.S. stock exchanges—which seek to allow exchanges to compete with internalizers through establishing de facto dark pools to capture trading spreads—only further undermine the theorized benefits for IPO firms of larger tick sizes given that the beneficiaries of such programs (i.e., stock exchanges and RPI participants) are also not known to provide market support for emerging growth companies. To the extent the SEC chooses to implement a pilot program modifying tick sizes, coupling such a program with increased disclosures concerning which broker-dealers are reporting trades to a FINRA TRF could help ascertain whether the appropriate market participants are benefiting from the wider spreads.

Finance, JOBS Act, Securities | Bookmark

TrackBacks (0)

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8345157d569e2017d40c7b9e2970c

Links to weblogs that reference Why Increasing Tick Sizes May Lead to More Dark Pools Instead of More IPOs:

Bloggers
Papers
Posts
Recent Comments
Popular Threads
Search The Glom
The Glom on Twitter
Archives by Topic
Archives by Date
August 2014
Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31            
Miscellaneous Links