Quest for transparency
In March 2010, the WFE was invited to take part in a conference organized by the Australian market regulator, ASIC. The topic was the transparency: how had or transparent or ‘lit’ markets performed compared to the counter markets or ‘dark’ trading venues.
The crisis : and how exchanges performed
Since 2007, the operators of financial exchanges have withstood not one, but two crises. The first crisis is the one reported on the news: the economic meltdown. This devolved from a ‘credit crunch’ outside the realm of exchanges and regulated products to a stampede for the exits in the liquid markets. Next came the string of bankruptcies, and then things bottomed out (we hope) with the bailouts, the relaxation of fair value accounting, a recovery of the financial sector and a worldwide recession.
The second crisis that some exchanges faced was a rapid shift in the market structure and the way trading firms use markets. These challenges were completely independent from the downturn in the market. This period coincides with the deregulation of European stock exchanges, the growth of high frequency trading and dark pools, and the fragmentation of liquidity,
With regard to the economic meltdown, the regulated exchanges came through unscathed. Nearly all of the stock and derivative exchanges remained open, and managed to complete post trade operations as normal. The contrast with several OTC markets is obvious.
As one of the few liquid markets open, exchanges were used as cash dispensers for firms that needed to sell liquid assets in a hurry. The result was a sharp decline in the market capitalization of nearly 50%, peak to trough. Liquidity, as measured by the trading value, has continued to fall from historic highs in most regions, the one exception being Asia Pacific.
With regard to the second great change of market structure, this has seen the number of trades skyrocket, while the value per trade has plummeted. The amount of data that is transmitted, and the speed of exchange communications, networks has also been vastly upgraded. Last but not least, many of the markets in Europe and North America became fragmented.
What reforms are underway?
The pace of reform has been disappointing to many market observers, and may yet result in a popular backlash against the financial services industry that we would all be better off avoiding.
Concerning the derivative markets, the main question is whether clearinghouses and exchanges can help reduce systemic risk in the OTC markets. The WFE supports efforts to help some OTC products, in particular credit default swaps, be cleared through CCPs. WFE members also recognize that not all OTC products are suitable for central counterparty clearing. A straight transfer of systemic risk to the clearing house would not make the markets safer. But some exchanges and clearinghouses are already having some success with transferring OTC products into a safer environment.
With regard to the stock exchanges I would like to focus more on experiences in Europe where major regulatory change has happened. A formal review of these new rules is beginning, but it is already insightful to hear the reactions from market participants.
According to the CFA Institute, in a recent survey conducted by both buy-side and sell-side firms, 68% responded that the new rules fragmenting the market made trade reporting harder, 64% said that it increased the cost of data access. Concerning best execution for clients, nearly twice as many firms felt this had been impaired compared to those who saw an improvement. By a similar margin it was felt that price discovery on exchanges had been damaged.
According to the AMAFI, the French trade association of investment bankers and institutional investors, while exchange fees decreased, overall transaction costs had increased this year by 12%.
This figure is in line with a report issued 12 February 2010 report by agency broker, CA Chevreux that notes “while the costs of lit markets in basis point terms decreased by 30% and settlement-delivery costs were down 47%, transaction costs rose 24% between 2007 and 2009.
At the same time, implicit costs also skyrocketed. Intra-day volatility, which might have been considered linked to the effect of the financial crisis on markets, has not gone back down and, combined with the reduction in liquidity, has ultimately increased the impact retransmitted to the market. This is how building any position of significant size implicitly became more costly.”
To paraphrase a recent report to the French Ministry of the Economy, Industry and Employment, (http://www.economie.gouv.fr/directions_services/sircom/100217_rap_fleuriot.pdf) if the increased complexity of the financial market was expected due to the multiplication of execution venues, what was not anticipated is the dominant conclusion of market participants concerning the degradation of quality of the transparency pre-trade and post-trade. The increased cost for data and, above all, the bad quality of trade data is truly a failure for the directive. Listed companies are severely critical about the growing difficulty in following what is happening in the trading of their shares. The end investors have not benefitted from lower rates for trade execution and clearing services, which have been more than offset by the cost of connecting to multiple venues, the need for new technology, as well as the implicit costs linked to the smaller trade sizes per transaction.
For exchanges, the opinions of listed companies, end investors, buy-side asset owners and smaller brokers, and governments raise serious concerns. But it would miss the point to ask if the new rules in Europe were flawed. As the report to the Ministry of Economy notes, the rules intended to bring into existence a complex, fragmented market structure. And for some financial firms the results are beneficial as this makes intermediation more valuable and more necessary in order to find liquidity.
Explaining how fragmentation decreases liquidity, Professor Robert Schwartz, in a paper co-authored with Asani Sarkar and Nick Klagge, writes in his 2009 study, “Fragmenting one large pool into multiple smaller pools can undercut the natural two-sidedness. One small pool, by chance, may receive orders predominantly from sellers, while another small pool, also by chance, is receiving orders predominantly from buyers. Together, the pools would be two-sided; separately, they are not.
This is a law of large numbers result: flip a fair coin many times and the proportions of heads will be very close to 0.5; flip it just a few times and either heads or tails may, by chance, predominate.”
What is the future?
It is fitting that we should be discussing the future of trading in Melbourne, as the Australian market is now in the frontline of the battle to bring fragmented markets to Asia. The Asian markets are where the growth is for stock trading today. Asia is the future for financial markets generally. Several Asian markets are designed to serve much larger retail markets than exist in Europe or in most markets in the Americas.
It will be a hard choice between the transatlantic market consensus for fragmentation and forging an independent path. But it does seem that in several Asian countries, such as Japan and India, there are the benefits of competition between national exchanges, without the issues raised by fragmentation.
Furthermore, jurisdictions will be in competition with each other to draw liquidity out of each other’s markets. As financial centers look to compete, a dark pool from outside the jurisdiction can add pressure on regulators to voluntarily fragment their national market to avoid trading going off-shore.
For the US and European markets, it is unlikely that market fragmentation will reverse soon. Just as Napster changed attitudes about buying music, exchange’s price discovery is now practically considered to be in the public domain, available for free-riders.
As Larry Tabb, Founder and CEO of the Tabb Group, noted at the WFE’s Exchange Technology Workshop in November 2009, US exchanges had lost 20% of market share to dark pools and ECNs in the past two years. The share of liquidity pools (by which I mean regulated exchanges, dark pools, and internalized trading by banks) has begun to converge at 13% for each.
Also speaking at the WFE workshop, Paul Pickup, Director of Trading Technology, referred to this convergence of market share as ‘the Starbucks phenomenon’. This observation describes how commoditized products and services end up with equal market share: if mergers reduce the number of outlets, the market share then reapportions itself to the remaining number of players. To the same degree that the number of coffee shops attract equal market share, the number of trading venues tend to find an equal proportion of the execution business.
The response from exchanges in the most fragmented markets has been to radically overhaul their systems and pricing structures to attract new sources of liquidity. In these markets, it is the algo traders and the high frequency trading firms that are at the center of transparent pre-trade and post-trade markets. These firms are also the most fervent defenders of market transparency.
Not all market participants are pleased by this development. This is driving the development of dark pools, which is the final question before us today.
What do we do about dark pools?
There will always be a need for trades to be negotiated outside of the market, and there have always been mechanisms for working block orders outside of the lit market, going right back to the specialists on the floor of the NYSE.
On the other hand, why should a retail order for 300 – 400 shares be sent to a dark pool? As mentioned above, the retail component in some Asian markets is significantly higher than elsewhere in the world. These investors are better off sending their orders to an exchange where the price can be improved or a mid-point order would interact with them. Retail orders are often considered to be ‘uninformed’, and this flow is the most interesting for all market participants.
It is important to point out the difference between dark pools that are used for block trading and the broker/ market-maker pools. As Joseph Gawronski, President/Chief Operating Officer, of Rosenblatt Securities Inc. reported to the WFE Annual Assembly in October 2009, in the US there are only half a dozen venues specialized in executing block trades, of which three have achieved significant presence : Liquidnet, Pipeline and Posit.
There are over 50 dark pools operating the US markets today and 36 in European markets, according to the Tabb Group, accounting for about 12% market share in the US and about 10% in Europe. However, it is in the broker pools where most of the growth has occurred. There, trade sizes are similar to those on exchanges, but traders do not have the same reporting and transparency requirements.
The CFA survey, mentioned above, asked three questions about dark pools: how they impacted price discovery, liquidity, and volatility. The responses were that 71% felt that dark pools were “somewhat problematic” to “very problematic” for price discovery, 58% had the same opinion concerning liquidity, and 69% were concerned about volatility.
These concerns reflect the way that brokers or other dark pool operators execute the best transactions within the pool before showing the remaining interest to the market. In these cases, the exchange or the ATS, becomes the execution venue of last resort.
Dark pools or dark liquidity?
Another distinction worth making is between dark pools and dark liquidity. Dark liquidity, which describes trades internalized by banks and through broker crossing networks, has taken a large slice of the European market share.
In the United States, the percentage of trading done off exchange is generally assumed to be around 25%, of which about 10% is done in dark pools. The rest (15% of overall US market share) is internalized by banks. According to some industry observers, while dark pools often execute trades at the mid-point, the internalized trades are more often executed at the bid or ask. As noted above, this ‘uniformed flow’ would likely be better served by having its order shown to the market.
In addition to the impact on price discovery, liquidity and volatility, there are also concerns about market integrity. As the Credit Agricole Chevreux report underlines: “dark market transactions are sharply affected by the price formation process in lit markets and any deterioration in lit markets has a detrimental effect. Moreover, as we have seen, the quality of the lit markets has declined and they can be manipulated far more easily.”
Some exchanges in the WFE offer dark pool services in response to clients needs and in the face of competition. A survey of 27 WFE members in October 2009, found that 7 exchanges currently have dark pools, 9 are considering opening such facilities and 8 have decided not to open a dark pool. Exchanges operating dark pools demonstrate that they are able to avoid fragmentation, provide for more complete surveillance, and ensure better price discovery by knitting the dark orders back into their main exchanges and lit pools.
Meanwhile, in a letter to the G20 in September 2009, the WFE Board of Directors noted in its remarks on dark pools “the combination of the absence of a level playing field between execution venues and decreased market transparency is an unsettling development. The policies and practices that exchanges have developed to ensure fair, orderly markets are at risk of becoming less meaningful and less available to investors and listed companies.”
In the aftermath of the financial crisis, regulators who have not already authorized their markets to fragment may be reassessing the trend to deregulation. While these benefits of fragmentation have been difficult to ensure in markets that are heavily institutionalized, that is with a strong market making culture and institutional investors, it will be all the more problematic for equity markets that are essentially retail.
The quest for transparency must recognize the legitimate needs for liquidity providers and market makers to execute block trades outside the central market. However, regulators may want to consider why retail orders should be executed away from lit venues when exposing these kinds of order to the market could likely improve both execution and price discovery. A restraint on retail orders being sent to venues other than lit markets is one option to safeguard market integrity.
Furthermore, transparency should not be confused with an avalanche of orders submitted and cancelled in microseconds. The problematic nature of many recent market techniques – flash orders, high frequency trading, dark pools, and lack of depth at MTFs and ECNs – are symptoms of the weakening of the central market due to fragmentation. It will be difficult to significantly improve the market quality without addressing the fundamental cause of its decline.
About Peter Clifford
Peter Clifford is Deputy Secretary General of the World Federation of Exchanges.
Since 2000, the Deputy Secretary General has lead several projects including:
· Re-launching the WFE Regulation Committee
· Integrating the derivative markets and clearinghouses network (IOMA) into the agenda of the WFE cash markets
· Developing the WFE brand strategy; upgrading the WFE communications on-line and off-line.
· Overseeing the inspections of new members, including exchanges in China and India.
His background includes employment as Head of International Market Development at Euronext Paris, Interest Rate Swaps broker at Viel & Cie., the Head of Product Development for Dow Jones Telerate in Paris and London. He also worked in transfer agency with State Street Bank & Trust in Boston.