تنوع معامله گران با فراوانی بالا

The regulatory debate concerning high-frequency trading (HFT) emphasizes the importance of distinguishing different HFT strategies and their influence on market quality. Using data from NASDAQ-OMX Stockholm, we compare market-making HFTs to opportunistic HFTs. We find that market makers constitute the lion's share of HFT trading volume (63–72%) and limit order traffic (81–86%). Furthermore, market makers have higher order-to-trade ratios and lower latency than opportunistic HFTs. In a natural experiment based on tick size changes, we find that the activity of market-making HFTs mitigates intraday price volatility.

Recent advances in information technology employed in equity markets allow traders to process information and submit orders at lightning speed. With typical holding periods measured in seconds or even fractions of seconds, resulting in large trading volumes, algorithmic strategies are now major forces in equity markets. Since the arrival of such strategies has coincided with massively increased limit order submissions and cancellations, high intraday price volatility (including flash crashes), and fragmentation of volumes across marketplaces, many voices have been raised, calling for the regulation of algorithmic trading (AT). To understand the influence of AT on the quality of markets, it is important to realize that investors use algorithms for many different reasons and strategies. The mere fact that a strategy is algorithmic or based on fast trading does not determine its market impact. To distinguish the main strategies and their individual market impacts is thus pivotal for effective regulatory design. A useful distinction to make is to separate agency algorithms from proprietary algorithms. The former is typically a service provided to clients to minimize the price impact of trading. Proprietary algorithms, on the other hand, are used by technologically sophisticated firms aiming to profit from the trading process itself. Firms using proprietary algorithms are referred to as high-frequency traders (HFTs).1 HFT strategies may be further subdivided into market-making strategies, on the one hand, and opportunistic trading such as arbitrage and directional (order anticipation and momentum ignition) trading, on the other (SEC, 2010). As limit order book markets in general preserve trader anonymity, empirical analysis of distinct AT and HFT strategies is difficult. Academic research in this field is based on market-wide proxies of algorithmic activities (e.g., Hendershott et al., 2011 and Boehmer et al., 2012) or proprietary data sets (e.g., Hirschey, 2013, Kirilenko et al., 2011 and Brogaard et al., 2012). None of these papers has been able to distinguish different HFT strategies in equity markets. In this paper, we use a proprietary data set that allows us to observe all limit order submissions, cancellations, and executions, complete with the identities of the traders.2 Our sample includes 30 Swedish large-cap stocks traded on the NASDAQ-OMX Stockholm exchange (henceforth NOMX-St). A key contribution of our paper is that, by using detailed trading and quoting information, we are able to subcategorize HFTs into market-making strategies and opportunistic (e.g., arbitrage and directional) strategies. Our categorization methodology, combined with rich order book information, allows us to conduct a more detailed analysis of HFT in equity markets than previous studies managed. We carry out the analysis for one highly volatile month (August, 2011) and one relatively calm month (February, 2012). We define market-making HFTs as proprietary-only firms that use algorithms in their order submission, and that have a continuous presence at the best bid and offer prices in the limit order book. We find that, within the group of HFTs, such market makers represent around 71.5% of the trading volume in August 2011, and 62.8% in February 2012. During both months, more than 80% of the HFT limit order submissions originate from the market-making strategies. The market-making HFTs in our study conform to the market makers described by Jovanovic and Menkveld (2012) and Menkveld (in this issue), in that they trade large volumes but keep inventories close to zero, and in that they are on the passive side in a majority of their trades. The result that is new to the literature is that market-making HFTs have higher order-to-trade ratios and lower latency than opportunistic HFTs. This reflects modern market making, where anyone who is unable to respond immediately to news about fundamentals or the order flow, by modifying posted orders, will be picked off by competitors. As a second contribution of this paper, we study the influence of market-making HFTs on short-term volatility. To investigate causality between HFT activity and market quality, it is necessary to overcome the endogeneity problem posed by the fact that HFTs are selectively active in order books where market quality is high. In previous literature, this problem has been addressed through the use of exogenous events that influence AT or HFT activity, but do not directly influence market quality. For example, Hendershott, Jones, and Menkveld (2011) use the automation of quotes on the New York Stock Exchange, and Boehmer, Fong, and Wu (2012) use the availability of co-location services to proxy for AT activity. Both find that AT has a positive influence on liquidity, a finding also supported by Brogaard (2011; for HFTs), Hendershott and Riordan (2013), and Hasbrouck and Saar (in this issue). Brogaard, Hendershott, and Riordan (2012) and Hendershott and Riordan (2013) find that HFT and AT, respectively, contribute to price discovery. As pointed out by Boehmer, Fong, and Wu (2012), however, there is less agreement on how AT and HFT influence volatility. Whereas Hasbrouck and Saar (in this issue) find a positive effect, Kirilenko, Kyle, Samadi, Tuzun (2011) find that HFTs may have worsened the flash crash in May 2010, and Boehmer, Fong, and Wu (2012) find that short-term volatility is amplified by AT. We use tick size changes as exogenous events to study HFTs' influence on short-term volatility. On European stock exchanges, the tick size (minimum price increment) depends on the stock price level. For example, when the price of a stock increases from SEK 49 to SEK 51, the tick size increases fivefold from SEK 0.01 to SEK 0.05 (SEK is the abbreviation for the Swedish currency krona). We hypothesize that an increased tick size makes market making more profitable, and other strategies, such as arbitrage trading, more costly. This is because market makers typically earn the spread, whereas opportunistic traders tend to pay the spread. Thus, we predict that a tick size increase would increase the share of trading activity due to market-making HFTs, whereas the share of trading activity due to opportunistic HFTs' activities would decrease. A reduction in the minimum tick size would have the opposite effect. Our results, which in this application are based on more than two years of data (February 8, 2010–March 31, 2012), confirm that this is indeed the case. When using the tick size events to study the influence of HFT on short-term volatility, we find that an increase in market-making HFT activity mitigates short-term volatility. This paper is closely related to that of Brogaard (2011) and Carrion (in press). They both investigate HFT activity in a set of 120 U.S. stocks trading on the NASDAQ, using an indicator provided by the exchange to determine whether or not a transaction is due to HFT. While Brogaard (2011) focuses his analysis of characteristics of HFTs' trading activities, Carrion (in press) relates HFTs' activities to stock market efficiency. A distinct feature of our paper compared to their papers is our ability to distinguish different types of HFT. Our finding that HFTs are indeed a heterogeneous group of traders shows the importance of distinguishing HFT strategies. Hirschey (2013) uses the same HFT classification as Brogaard (2011) and Carrion (in press) and, in addition, observes trader identities. In his study on anticipatory trading, he finds that HFTs differ in their ability to forecast future order flows, but he does not relate such abilities to differences in HFT strategies. Kirilenko, Kyle, Samadi, Tuzun (2011) are able to observe the trades of individual firms and investigate the trading and quoting surrounding the market turbulence of May 6, 2010, known as the flash crash. They define HFTs as the 7% most active intermediaries in the market, and find that HFTs did not cause, but may have amplified, the volatility in the flash crash. Their data concern only one security, the S&P 500 E-mini futures, over three days of extraordinary volatility. Our investigation can be seen as a complement to their paper, as we cover 30 stocks over two months of different volatility levels. Finally, our paper relates to the work of Jovanovic and Menkveld (2012). They develop a model of market-making HFT, showing that the presence of such traders may either increase or reduce adverse selection costs, depending on how well-informed liquidity demanders are. In an empirical application, they show that one HFT that dominated the trading on the Chi-X in 2007–2008 behaved as a middleman in their model, with low net inventories, predominantly passive trades, and fast trading. The HFT market makers investigated in this paper have the same properties. In the next section, we present our empirical setting and data, as well as our HFT categorization methodology. After that, we present estimates of various metrics frequently associated with HFT. In Section 4, we divide the HFT group into subcategories. Specifically, we study how market makers differ from other HFTs. Furthermore, we investigate how the behavior of market makers and opportunistic traders differs across segments of stocks, and how market-making activity is related to market quality measures. In Section 5, we present our event study, where we analyze the causal effects of HFT activity on volatility. Section 6 offers concluding remarks.

This papercontributestotheHFTliteraturebysubcategorizingHFTstrategiesintomarket- making HFTandopportunisticHFT,byprovidingadetailedanalysisoftheirtrading characteristics,andassessingtheirinfluence onvolatility.We find thatamajorityoftheHFTs' trading volumeandmorethan80%ofHFTlimitordersubmissionsontheNOMX-Stare associated withmarket-makingstrategies.Market-makingHFTsthusconstituteanimportant group oftradersinmodernequitymarkets,whichinvestorsandregulatorsneedtounderstand when choosingtheirpolicies.We find thatmarket-makingHFTshavehigherorder-to-traderatios and lowerlatencythanopportunisticHFTs.Wethinkthisisduetomarketmakerscontinuously monitoringorder flows andrespondingtonewsfasterthanothertraders,therebyreducingtheir exposuretoadverseselection[asarguedby JovanovicandMenkveld(2012)]. Interestingly opportunisticHFTs,includingthoseusingarbitrageanddirectionalstrategies,haveorder-to- trade ratiosandlatencyonaparwithnon-HFTs. Using aneventstudybasedonchangesinminimumticksize,we find thatmarket-making HFT activitiesaregoodfortheoverallmarketqualityinthesensethattheyreduceshort-term volatility.This finding isimportant,aspreviousempiricalresearchhasshownmixedconclusions on theATandHFTimpactsonvolatility(whereasmostagreethattheactivityisbeneficial for liquidityandpricediscovery).Inpractice,volatilityaffectsbothriskmanagementpoliciesand liquiditysupplydecisionsamongtraders.Givenourrichdatasetandexogenouseventsetting based onticksizechanges,ourevidencebringsdeeperinsighttothepolicydebateonHFT. Finally, oureventstudyshowsthatahigherticksizedecreasesboththetradingandquoting activitiesofalltradergroups,inparticularHFTs.Anincreasedminimumticksizemakes strategiesbasedonliquiditysupply(suchasmarketmaking)moreattractive,whereasstrategies that consumeliquidity(suchasarbitragestrategies)becomemoreexpensive.Thus,ticksize regulationmaybeaninterestingsolutionforlimitingquotingtraffic withoutthreateningthe advantagesofHFTactivity.Weconsiderthisaninterestingdirectionforfutureresearch.