MARKET LIQUIDITY INDICATORS

Transcription

1 Time of Mixed Blessings for Consolidation While the European Price Formation Process Continues to Move to MTFs CONTENTS THE EUROPEAN COMMISSION PREVENTS DB-NYSE MERGER...1 BATS AND CHI-X HAVE NEVERTHELESS PAVED THE WAY FOR CONSOLIDATION...1 TWO RECENT OUES: MTFS HOST A LARGER PART OF THE PRICE FORMATION PROCESS THAN EXPECTED...2 FRAGMENTATION...3 VOLUMES, AVERAGE DAILY NUMBER OF TRADES AND AVERAGE TRADE SIZE...4 DARK MARKET SHARES AND AUCTION LIQUIDITY...5 VOLATILITY, BID-ASK SPREAD AND TIME AT EBBO FIGURES...6 MARKET INDICATORS: 1 January to 31 January MARKET INDICATORS: 1 December to 31 December MARKET INDICATORS: 1 November to 30 November DEFINITIONS...10 THE EUROPEAN COMMISSION PREVENTS DB-NYSE MERGER: Deutsche Börse and NYSE Euronext did not receive approval for their merger from the European Commission. The latter considered that the loss of competition on the derivative market would have been too high otherwise. The decision was made public at the very beginning of February, after months of negotiations. The two groups had revised their merger submission in order to gain approval. Among other things, they had told the European Commission that, if they could merge, they would cap trading fees and offer open access to clearing. This latter point is in accordance with EMIR and MiFIR, in which vertically-siloed exchange groups will be required to open up their clearing services to third-parties. They had also committed to divesting their single equity derivatives business. By doing so, they intended to address the remaining regulatory concerns in derivatives and clearing. In January, European competition officials already expressed their intention to block the merger unless one of the groups' main derivatives business, not only single equity ones, were sold. The US Department of Justice had given its approval, on the condition that Deutsche Börse would sell its stake in Direct Edge. BATS AND CHI-X HAVE NEVERTHELESS PAVED THE WAY FOR CONSOLIDATION Conversely, BATS and Chi-X have closed their deal, after receiving final approval from regulators on 24 November. Their combined market share on European LMIs reached 31.8% in January. It has been on a constant uptrend from 29.1% in June Once again, this trend is due to improvement for Chi-X. This can be seen in the chart on the right, showing monthly market shares on FTSE 100 stocks since the beginning of Following the merger, the combined group reaches a market share of almost 40%. The remaining question is how long they will be able to offer two separate order books with almost the same features. Market shares on FTSE 100 stocks since January % 60% 50% 40% 30% 20% 10% 0% j f m a m j j a s o n d j Primary market Chi-X Turquoise Bats Bats-Chi-X _ Mergers are also going ahead outside Europe, as the Osaka Stock Exchange has agreed to a merger with the Tokyo Stock Exchange. Finally, TMX's board gave its official backing to Maple's proposal at end October. Given the competition issues at stake, the regulators' final decision is expected in early

2 TWO RECENT OUES: MTFS HOST A LARGER PART OF THE PRICE FORMATION PROCESS THAN EXPECTED BATS suffered two outages over the last two months. On 6 December it was forced to halt trading as of 9.36am (UK time). There were system issues but no further information was given. On the same day, BATS' market share on FTSE 100 stocks fell from the usual 9% to less than 2%. The remaining market share went partly to Chi-X (an additional 2% market share vs. the usual level), but mainly to the primary market (up 4%). Note that BATS also faced a very short (1-min) technical issue on 2 February. Chi-X also faced technical issues on 2 January, and started trading at 8.36am (UK time) only, as shown on the left-hand chart below. The next section will concentrate on how market shares are re-distributed during such events. Does the primary market recover the majority of volumes when a MTF is halted or is it the contrary? Furthermore, do people stop their routing or concentrate volumes on other MTFs? MAIN CHI-X / BATS (where trading remains) Market share rebalancing during outages on CAC40 stocks (1-min data) Less than expected More than expected In order to study these mechanisms, the first assumption is that the volume that cannot take place on the halted venue is redistributed uniformly to the three other venues the primary market and the two other MTFs. This means that, in theory, each venue should recover a certain amount of market share, which is proportional to its usual market share. Using 1-minute volumes on CAC40 stocks during the two first outages: BATS on 6 December and Chi-X on 2 January, the charts to the left show how real market shares were different from a uniform redistribution of volumes. What these charts bring to light is that the primary market recovers less than it should do (the distribution is centred to the left, which is the negative side). On the contrary, the two other MTFs recover more than expected (be this Chi- X when BATS is halted, or BATS when Chi-X is halted, and Turquoise either way: distributions are centred on the positive side). TURQUOISE More than expected BATS' outage on 2011/12/06 Chi-X' outage on 2012/01/02 _ - 2 -

3 FRAGMENTATION Fragmentation Efficiency Index on European LMIs over one year (%) / / / / / / / / / / / / /02 EURO STOXX 50 AEX BEL20 CAC40 DAX FTSE 100 SWISS LEADER PR INDEX "The FEI shows how much attention should be paid to smart order routing (FEI=100%: no attention is required, FEI close to 0%: routing is very important)." Turquoise's market shares fell sharply in January, as did BATS's. Chi-X continued to increase numbers while primary markets remained stable vs. December. The latter were also stable or increasing in December vs. November. However, the overall market share of MTFs had decreased during the last month of the year to 38% (vs. 38.9% in November). This was exclusively due to 16 December, which was a triple witching day, and when most European indices rebalanced, leading to extremely high volumes at the closing auction. Notwithstanding 16 December, which was a particular day, Chi-X remained the winner in December, while Turquoise and BATS showed mixed results. In terms of November market shares, Chi-X once again confirmed its upward trend, reaching 26.2% market share on European LMIs (versus 25.6% in October). More precisely, it posted a significant improvement on the SLI30. Conversely, BATS and Turquoise showed decreasing figures. There were mixed results for primary markets: NYSE Euronext lost ground slightly, while Xetra remained stable and the London Stock Exchange increased its market share. Note that the latter managed to stay above the 50% threshold despite the sharp rise of Chi-X. Equiduct's share decreased substantially in November to 1.61% on CAC40 stocks vs. 1.86% in October, but it started increasing once again in December and January (1.63% and 1.70% respectively). In terms of 's Fragmentation Efficiency Index (see the chart above), it is worth noting the strong decrease in fragmentation for FTSE 100 and DAX30 stocks, while the other indices remained quite stable in the past few months. Get instant access with the Fragmentation Efficiency Index _ - 3 -

4 VOLUMES, AVERAGE DAILY NUMBER OF TRADES AND AVERAGE TRADE SIZE Average daily volumes on European Local Main Indices (LMIs) increased in January vs. December (EUR19.8bn in January vs. EUR17bn in December), but remained lower than in November (EUR21.4bn). Average numbers of trades increased consequently, while Average Trade Sizes were heterogeneous. In fact, December showed very weak daily volumes on European LMIs (EUR17bn vs. EUR22.4bn in December), as usual. However, total volume did not reach last year's level over the same period. The average number of trades and average trade size both decreased versus November. Most markets were closed on 26 December, but there was no trading on UK stocks on 27 December either. Average daily volumes on European Local Main Indices decreased slightly in November vs. October as well (by 4.6% from EUR22.4bn in October to EUR21.4bn). This was partially due to the several bank holidays that month. 11 November saw weak volumes, and levels were even weaker on the following Monday. Thanksgiving in the US implied a long weekend for American investors (24 and 25 November), and European volumes fell accordingly. Note that, conversely, there were unusually high volumes on All Saints' Day (1 November) due to European economic announcements in the wake of the G20 summit. As a result, the average daily number of trades decreased on most indices and venues. The average trade size also decreased in November, continuing on its long-term path. Average daily number of trades on AEX stocks since April / /05 NYSE EURONEXT CHI-X TURQUOISE BATS 2009/ / / / / / / / / / / / / / / /01 This type of phenomenon can be found in the trend of the average daily number of trades. The chart to the left shows the average daily number of trades on AEX stocks, for each month since April It decreases at the end of each year, and 2011 did not break the rule. Note that the number of trades remained very high after August 2011, and January came back to November levels. At the same time, we can notice the strong upward trend of Chi-X, which follows the primary market. _ - 4 -

5 DARK MARKET SHARES AND AUCTION LIQUIDITY Dark market shares on CAC40 stocks since December % 2.5% 2.0% 1.5% 1.0% 0.5% 0.0% 2010/ / / / / / / /07 BATS DARK TURQUOISE DARK CHI-DELTA SMARTPOOL 2011/ / / / / /01 Simultaneously, our Auction Liquidity indicator decreased sharply between December and January. December's figures were particularly high because of the triple witching day and the rebalancing of many European indices. This is visible on the chart to the right, indicating the auction liquidity indicators since December 2010 for each index (the last three months are in green). December's exception is particularly visible both in 2010 and In January, dark market shares increased significantly on our universe since December, reaching a better number than in October and November (2.41% vs. 2.15% and 2.06% on European LMIs). Note that December showed weak dark market shares for almost all venues and indices (1.90% on European LMIs). The chart to the left represents dark market shares per venue on the CAC40 since December As a reminder, dark market shares represent dark turnover (turnover on mid-point type venues: Smartpool, Chi-Delta, Turquoise Dark and BATS Dark) over total visible turnover. Volumes occurring in Broker Crossing Networks or Over The Counter are not taken into account Auction liquidity indicators since December 2010 AEX BEL20 CAC40 DAX30 FTSE100 SLI 30 The chart also indicates that there is no general trend in the weight of fixings: it was stable in 2011 for CAC40 stocks, decreasing for DAX30 stocks, and slightly increasing for FTSE100 stocks. _ - 5 -

6 VOLATILITY, BID-ASK SPREAD AND TIME AT EBBO FIGURES Intraday volatility on DAX30 over one year % Daily Daily Intraday Jan /02/12 09/01/12 08/12/11 08/11/11 10/10/11 08/09/11 10/08/11 12/07/11 10/06/11 12/05/11 08/04/11 10/03/11 09/02/11 In January, intraday volatility stayed at a relatively low level compared to the last quarter of 2011, but did not keep the comfortable level reached at the very end of December. Still, bid-ask spreads were consequently reduced vs. the whole month of December (-7% on average), and EBBO figures increased accordingly (+7% on average). In December, intraday volatility decreased vs. November, but remained a little higher than at the beginning of 2011 for most indices. Spreads were reduced accordingly, except for Turquoise on BEL20 stocks (spreads were around 21bps). This movement was due to one stock in particular, and depends strongly on the method by which the VWAS is calculated. The chart above shows intraday volatility, calculated over the whole day and at 5-min intervals, on the DAX30 since the beginning of June The August peak in volatility and the period since that date is shown in the dark green shading. Volatility is decreasing slowly but was only back to its usual levels (light green) during the last week of the year (as highlighted by the orange line), and started higher at the beginning of Intraday volatility remained high in November (as shown on the chart above), despite relatively weak volumes. It led to consistently wider bid-ask spreads and a lower percentage of time spent at the European Best Bid and Offer for most venues and indices. _ - 6 -

7 _ REFERENCE PERIOD: 1 January to 31 January 2012 THREE-MONTH TREND ONE-MONTH TREND TWO-WEEK TREND MARKET SHARE DARK (MID-POINT) MARKET SHARE AUCTION LIQUIDITY* AVG DAILY NO. of TRADES Average Trade Size (ATS) Bid-ASK Spread (VWAS in bp) % time at BEST BID &OFFER* % time at BEST BID & OFFER with greatest size* AEX EURONEXT 61.1% 0.1% % 20% CHI-X 27.6% 1.0% % 7% TURQUOISE 6.6% 0.4% % 0% BATS 4.7% 0.5% % 0% BEL20 EURONEXT 64.0% 0.1% % 24% CHI-X 22.3% 0.8% % 7% TURQUOISE 7.8% 0.5% % 3% BATS 5.8% 0.6% % 1% CAC40 EURONEXT 64.0% 0.2% % 25% CHI-X 25.2% 1.1% % 11% TURQUOISE 6.1% 0.4% % 1% BATS 4.6% 0.8% % 0% DAX XETRA 67.6% % 26% CHI-X 23.9% 0.7% % 13% TURQUOISE 5.1% 0.3% % 1% BATS 3.5% 0.3% % 0% FTSE 100 LSE 54.1% % 23% CHI-X 31.2% 1.6% % 17% TURQUOISE 7.7% 0.8% % 1% BATS 6.9% 1.0% % 0% SLI (30) SIX 66.6% % 47% CHI-X 23.7% 0.8% % 9% TURQUOISE 5.0% 0.3% % 0% BATS 4.7% 0.6% % 0% EUROSTOXX 50 CHI-X 19.8% 0.7% % 10% TURQUOISE 4.2% 0.3% % 0% BATS 3.2% 0.4% % 0% * Exclusive - 7 -

8 _ REFERENCE PERIOD: 1 December to 31 December 2011 THREE-MONTH TREND ONE-MONTH TREND TWO-WEEK TREND MARKET SHARE DARK (MID-POINT) MARKET SHARE AUCTION LIQUIDITY* AVG DAILY NO. of TRADES Average Trade Size (ATS) Bid-ASK Spread (VWAS in bp) % time at BEST BID &OFFER* % time at BEST BID & OFFER with greatest size* AEX EURONEXT 61.8% 0.1% % 21% CHI-X 26.9% 0.7% % 8% TURQUOISE 6.6% 0.5% % 0% BATS 4.7% 0.4% % 1% BEL20 EURONEXT 63.7% 0.1% % 22% CHI-X 22.3% 0.5% % 7% TURQUOISE 7.9% 0.3% % 3% BATS 6.1% 0.4% % 1% CAC40 EURONEXT 63.3% 0.2% % 26% CHI-X 24.9% 0.7% % 11% TURQUOISE 6.9% 0.4% % 1% BATS 4.8% 0.6% % 1% DAX XETRA 67.0% % 29% CHI-X 24.3% 0.8% % 12% TURQUOISE 5.1% 0.2% % 1% BATS 3.6% 0.3% % 0% FTSE 100 LSE 54.2% % 22% CHI-X 30.4% 1.2% % 15% TURQUOISE 8.1% 0.7% % 1% BATS 7.3% 0.7% % 1% SLI (30) SIX 67.1% % 41% CHI-X 23.2% 0.7% % 8% TURQUOISE 5.6% 0.3% % 1% BATS 4.2% 0.5% % 1% EUROSTOXX 50 CHI-X 20.1% 0.6% % 9% TURQUOISE 4.7% 0.3% % 1% BATS 3.3% 0.3% % 1% * Exclusive - 8 -

9 _ REFERENCE PERIOD: 1 November to 30 November 2011 THREE-MONTH TREND ONE-MONTH TREND TWO-WEEK TREND MARKET SHARE DARK (MID-POINT) MARKET SHARE AUCTION LIQUIDITY* AVG DAILY NO. of TRADES Average Trade Size (ATS) Bid-ASK Spread (VWAS in bp) % time at BEST BID &OFFER* % time at BEST BID & OFFER with greatest size* AEX EURONEXT 62.3% 0.1% % 17% CHI-X 26.3% 0.8% % 9% TURQUOISE 6.5% 0.4% % 0% BATS 4.9% 0.4% % 1% BEL20 EURONEXT 63.8% 0.1% % 20% CHI-X 22.2% 0.7% % 7% TURQUOISE 8.3% 0.5% % 3% BATS 5.8% 0.6% % 1% CAC40 EURONEXT 63.7% 0.3% % 25% CHI-X 23.7% 0.7% % 10% TURQUOISE 7.6% 0.4% % 1% BATS 5.0% 0.5% % 0% DAX XETRA 65.1% % 22% CHI-X 24.7% 1.0% % 11% TURQUOISE 6.0% 0.2% % 2% BATS 4.1% 0.3% % 1% FTSE 100 LSE 53.5% % 22% CHI-X 30.7% 1.2% % 14% TURQUOISE 8.3% 0.7% % 1% BATS 7.6% 0.7% % 1% SLI (30) SIX 66.4% % 40% CHI-X 23.1% 0.8% % 7% TURQUOISE 5.5% 0.4% % 1% BATS 5.1% 0.6% % 1% EUROSTOXX 50 CHI-X 20.2% 0.7% % 9% TURQUOISE 5.2% 0.3% % 1% BATS 3.8% 0.3% % 1% * Exclusive - 9 -

10 MONTHLY UPDATE DEFINITIONS: Market Share: For local main indices, this is the ratio of the venue's turnover over the sum of the turnover of all the four considered venues. For the Eurostoxx 50, it is the ratio of the venue's turnover over the sum of the turnover of all venues trading Eurostoxx 50 components. Dark (mid-point) Market Share: For local main indices, this is the ratio of the venue's dark turnover (midpoint type) over the sum of the visible turnover of all the four considered venues. Euronext lines correspond to Smartpool turnover, whereas Chi-X represents Chi-Delta figures, Turquoise Dark and Bats is Bats Dark. Volumes occurring in Broker Crossing Networks or Over The Counter are not taken into account. Auction Liquidity: This is the ratio of the proportion of the turnover traded during auctions by the typical time ratio between the continuous auction phase (generally 8½ hours) and the fixing auctions (usually two 5- minute phases), which is less than 2%. It is only available for primary exchanges. The higher it is, the more turnover one will find during auctions. Average Daily Number of Trades: The higher it is, the more chance an investor has to be present when a trade happens. Contrary to block venues, the granularity of a venue is a must to attract different order flows. Average Trade Size (ATS): Mean trading size in euros. This can be considered as the "natural size" of orders on the venue Bid-Ask Spread (Volume Weighted Average Spread): The average spread when a trade occurs, weighted by the turnover of the trade. It is the bid-ask spread from the liquidity provider's point of view. % Time at EBBO at Best Bid and Offer: This is the proportion of the day during which the venue offers a spread equal to the EBBO. % Time at EBBO at Best Bid and Offer with Greatest Size: This is the proportion of the day during which the venue offers the greatest sizes at a spread equal to the EBBO. OTHER CA CHEUVREUX PUBLICATIONS: _ This document is provided for general information purposes only (i) and does not constitute investment advice and/or investment research and financial analysis or other forms of general recommendation relating to transactions in financial instruments (ii) and is based on objective market data for a quantitative market liquidity analysis only. This text is exclusively directed to Crédit Agricole Cheuvreux's ("Cheuvreux") Eligible Counterparties and Professional Clients or prospects. Services referred to herein are not available to Retail Clients. It is not to be construed as a solicitation or an offer to buy or sell any financial instruments. The information contained in this publication is prevailing at the date of publication and is subject to change, without notice, to reflect subsequent developments. No liability is accepted by Cheuvreux that may arise for any direct or consequential damages arising from the use of this document or its contents. This document and its contents are proprietary to Cheuvreux, and and no part of this document or its subject matter may be reproduced, disseminated or disclosed without the prior written approval of Cheuvreux and. The distribution of this document in other jurisdictions outside of EEA (European Economic Area) may be restricted by law, and persons into whose possession this document comes should inform themselves about, and observe, any such restrictions Crédit Agricole Cheuvreux. All rights reserved.