• Have any questions?
  • info.zbook.org@gmail.com

Market Developments: High Frequency Trading

3m ago
42 Views
11 Downloads
951.15 KB
31 Pages
Last View : Today
Last Download : Today
Upload by : Warren Adams
Share:
Transcription

Market Developments: HighFrequency TradingBy Werner BijkerkHead of the ResearchDepartment, IOSCO

IntroductionMarkets have changed considerably over thepast decade: Automation Globalization Dispersion – multiple trading venues Internalization Etc.

IntroductionChanges have brought benefits Increased efficiency More use of markets and challenges: Scattered trading venues and information High speed trading Increased difficulty of oversight Systemic risk?

High frequency tradingTOTAL TRADINGALGORITHMICaLa TRADINGAlgo tradingHIGH FREQUENCY TRADING

High frequency tradingAlgorithmic trading: In simple words an algorithmic trading strategy isa step-by-step instruction for trading actionstaken by computers (automated systems)High frequency trading: High frequency trading is a subset of algorithmictrading – use of high speed/technology/colocation is part of it5

High frequency tradingImpact of algorithmic trading:1. Automated trading – less likelihood of emotionalattachment to trading2. Perhaps more efficient3. Well thought out positions4. Odd/smaller trade sizeIs this really different from traditional strategies butnow using computers?6

High frequency tradingChallenges of algorithmic trading:1. If similar strategies employed than largedirectional movements2. Increase in volatility3. Human error – algo design4. There is no time for human intervention5. Upper hand (unfair) advantage employed byinstitutions – but this has always been the case?7

High frequency trading

High frequency tradingHFT is a subset of algorithmic trading: Using speed and reduced latencyProprietary tradingLimited capital commitmentWide range of strategies – combining“quasi‐market making”, arbitrage, directional,etc.9

High frequency tradingImpact of HFT on the markets: Quicker price discovery Tighter spread– Narrower effective spreads– Other dimensions (depth, resiliency, etc.) not fullyassessed Increase in liquidity In normal times – reduces market volatility Promotes competition among market centersand liquidity providers10

High frequency tradingConsider the following situation: DJIA falls sharply in 20 mins, with some stocksplummeting 9% in under 12 mins. The SEC launchedan investigation but its cause remains unknown. During the crash:––––Liquidity disappeared and sell orders flooded the marketSpreads widenedOrderly price discovery ceasedConfidence evaporatedWHEN DID THIS TAKE PLACE?11

High frequency tradingConcerns with HFT:1. May 6th event (or other) – Could it happen again?2. During times of market stress– Increased volatility– Unstable liquidity3.4.5.6.7.Can quickly add momentum to pricesLeverage/risk managementManipulation (transitory volatility, spoofing)Unfair advantages in information and latencyTechnological “arms race”12

High frequency tradingCan HFT add to systemic risk? Transmitting shocks through markets and other market segments– With increased global market interconnectedness– Fuel contagion effects Effect on price formation, cost of capital and confidence in marketas a whole Increased speed provides limited opportunities for regulators tointervene during high volatility/uncertainty– Too much data to track (regulator don’t have the technology) Size– Broad use of HFT – equities, commodities, FX– 70% of US trade in HFT, around 30% in Europe, 10-20% in ROW13

High frequency tradingNew IOSCO recommendations for trading venueoperators and trading participants Recommendation 1: Regulators should requirethat trading venue operators provide fair,transparent and non-discriminatory access totheir markets and to associated products andservices;14

High frequency tradingNew IOSCO recommendations for trading venueoperators and trading participants Recommendation 2: Regulators should seek to ensurethat trading venues have in place suitable tradingcontrol mechanisms (such as trading halts, volatilityinterruptions, limit-up-limit-down controls, etc.) to dealwith volatile market conditions. Trading systems andalgorithms should be robust and flexible such that theyare capable of dealing with, and adjusting to, evolvingmarket conditions. In the case of trading systems, thisshould include the ability to adjust to changes(including sudden increases) in message traffic;15

High frequency tradingNew IOSCO recommendations for trading venueoperators and trading participants Recommendation 3: All order flow of tradingparticipants, irrespective of whether they are directvenue members or otherwise, must be subject toappropriate controls, including automated pre-tradecontrols. These controls should be subject to theregulatory requirements of a suitable market authorityor authorities. In addition, regulators should identifyany risks arising from currently unregulated directmembers/participants of trading venues and, whereany are identified, take concrete steps to address them;16

High frequency tradingNew IOSCO recommendations for Regulators: Recommendation 4: Regulators should continueto assess the impact on market integrity andefficiency of technological developments andmarket structure changes, including algorithmicand high frequency trading. Based on this,regulators should seek to ensure that suitablemeasures are taken to mitigate any related risksto market integrity and efficiency, including anyrisks to price formation or to the resiliency andstability of markets, to which such developmentsgive rise.17

High frequency tradingNew IOSCO recommendations for Regulators: Recommendation 5: Market authorities shouldmonitor for novel forms or variations of marketabuse that may arise as a result of technologicaldevelopments and take action as necessary. Theyshould also review their arrangements (includingcross-border information sharing arrangements)and capabilities for the continuous monitoring oftrading (including transactions, orders entered ororders cancelled) to help ensure that they remaineffective.18

High frequency tradingFurther work: Study on additional tools to deal with the challenges arising frommarket surveillance, some of which may include:– additional audit trail or surveillance data consisting of all orders andtrades by market participants in a given instrument;– a single reporting point for all orders and for all transactions, byjurisdiction or geographical zone and across asset classes; and– unique legal entity identifiers. In addition, IOSCO will build on its work on supervisorycooperation to consider how best to enhance operational contactsand cooperation between regulators in order to more flexibly shareinformation and views for day-to-day market supervision.19

High frequency tradingFurther work:An analysis of the new market structure and itsimpact on market efficiency/integrityIOSCO will analyse the evolving markets’ structure,in order to assess what specific issues suchstructural developments raise with regard tomarket efficiency and integrity; and considerwhether and what recommendations may beneeded to address any risks.20

High frequency tradingConclusions: Speed combined with automation is newStretches marketsSome old wine in new bottlesCan be of systemic risk?Challenges to regulators21

High frequency tradingThe High Frequency Trader versus the Regulator

High frequency tradingSome recent studies: IOSCO Market Efficiency and Integrity paper:Regulatory Issues Raised by the Impact ofTechnological Changes on Market Integrity andEfficiency. ESMA Public Consultation on HFT Kirilenko et al. The Flash Crash: The Impact of HighFrequency Trading on an Electronic Market; Staff reports CFTC/SEC Report of individual regulators (the Netherlands AFM)23

Dark liquidityDark liquidity (IOSCO report of 27 October 2010):Dark liquidity and the use of dark pools can have: Impact on the price discovery process where there is asubstantial number of dark orders and/or orderssubmitted into dark pools which may or may not bepublished; Impact of potential fragmentation on information andliquidity searches; and Impact on market integrity due to possible differencesin access to markets and information.24

Dark liquidityIOSCO report of 27 October 2010 on dark liquidity launched newprinciples that are designed to: minimise the adverse impact of the increased use of dark pools anddark orders in transparent markets on the price discovery process; mitigate the effect of any potential fragmentation of informationand liquidity; help to ensure that regulators have access to adequate informationto monitor the use of dark pools and dark orders; help to ensure that investors have sufficient information so thatthey are able to understand the manner in which orders will behandled and executed; and increase the monitoring of dark orders and dark pools in order tofacilitate an appropriate regulatory response.25

Dark liquidityPrinciples that are designed to:Transparency to Market Participants and IssuersPrinciple 1: The price and volume of firm bids and offersshould generally be transparent to the public. However,where regulators consider permitting different marketstructures or order types that do not provide pre-tradetransparency, they should consider the impact of doingso on price discovery, fragmentation, fairness andoverall market quality.26

Dark liquidityPrinciple 2: Information regarding trades,including those executed in dark pools or as aresult of dark orders entered in transparentmarkets, should be transparent to the public.With respect to the specific information thatshould be made transparent, regulatorsshould consider both the positive and negativeimpact of identifying a dark venue and/or thefact that the trade resulted from a dark order.27

Dark liquidityPriority of Transparent OrdersPrinciple 3: In those jurisdictions where dark tradingis generally permitted, regulators should takesteps to support the use of transparent ordersrather than dark orders executed on transparentmarkets or orders submitted into dark pools.Transparent orders should have priority over darkorders at the same price within a trading venue.28

Dark liquidityReporting to RegulatorsPrinciple 4: Regulators should have a reportingregime and/or means of accessing informationregarding orders and trade information invenues that offer trading in dark pools or darkorders.29

Dark liquidityInformation Available to Market Participantsabout Dark Pools and Dark OrdersPrinciple 5: Dark pools and transparent marketsthat offer dark orders should provide marketparticipants with sufficient information so thatthey are able to understand the manner inwhich their orders are handled and executed.30

Dark liquidityRegulation of the Development of Dark Pools andDark OrdersPrinciple 6: Regulators should periodically monitorthe development of dark pools and dark orders intheir jurisdictions to seek to ensure that suchdevelopments do not adversely affect theefficiency of the price formation process ondisplayed markets, and take appropriate action asneeded.31

Algo trading TOTAL TRADING ALGORITHMIC TRADING HIGH FREQUENCY TRADING . Algorithmic trading: In simple words an algorithmic trading strategy is a step-by-step instruction for trading actions taken by computers (au