INTERNATIONAL EVIDENCE ON ALGORITHMIC TRADING

1
INTERNATIONAL EVIDENCE ON ALGORITHMIC TRADING

• Ekkehart Boehmer, Singapore Management University
• Kingsley Fong, University of New South Wales
• Julie Wu, University of Georgia

2
Todays markets

• High frequency trading (HFT) activity of
  algorithms that submit and cancel orders,
  reacting within milliseconds to market updates .
• HFT is for real between 60 and 80 of trading
  volume.
• HFT strategies are not transparent

3
Issues

• Regulators and academics are interested in the
  consequences of HFT for
• market quality
• welfare (of traders, society, )
• systemic risk
• Our study focuses on market quality.

4
Priors on HFTs effect on market quality

• Depends on strategies
• Passive market making should improve liquidity
• Stat arb should improve efficiency
• Structural and directional strategies could be
  wealth transfers
• Algorithmic trading (AT) is a precondition for
  all HFT strategies.
• Since actual strategies are not known to
  researchers, most research studies the AGGREGATE
  EFFECT of AT/HFT.

5
Prior studies

• Data used
• Transactions data where AT/HFT is inferred
  indirectly from the rate of electronic message
  traffic
• cost and speed consideration - gt electronic
  orders
• commonly used as a proxy by consultants,
  exchanges and other market venues
• Transactions data with trader category
  information that have a set of transactions
  attributed to AT/HFT
• Transactions data with trader account information

6
Prior studies message counts

• Hendershott, Jones, and Menkveld (JF2011)
  electronic message counts from NYSEs System
  Order Data (SOD) database as a proxy for AT
• concentrate on 2003 NYSE autoquote event
• algo trading improves spreads and price
  discovery, reduces information asymmetry
• Hasbrouck and Saar (JFM 2013) similar findings
  with HFT inferred from ITCH millisecond episodes.
• Eggington at al. (WP2014) liquidity worsens on
  extremely high-volume days

7
Prior studies trader categories

• Brogaard (and his coauthors), in several recent
  papers, uses a 2008-2009 random sample of 120
  Nasdaq stocks with 26 HFT firms
• HFT activity is associated with better liquidity,
  mixed effect on volatility, better price
  discovery
• Potential selection issue with exchange-selected
  HFT firms
• nature of order flow, fraction of order flow, no
  large proprietary trading desks

8
Prior studies trader IDs

• Kirilenko, Kyle, Samadi, and Tuzun (WP 2014)
• see individual strategies in SP500 E-minis
• find that HFT may have worsened (but did not
  cause) the Flash Crash on May 6, 2010.
• Baron, Brogaard, and Kirilenko (WP 2014)
• find large returns in E-minis for top performing
  HFT firms.

9
Our objectives

• Broaden the scope of evidence on AT/HFT to an
  international sample over a long period and
  assess effects on
• liquidity
• price efficiency
• volatility
• Examine differences in the cross section of firms
• Size, price level
• Study AT/HFT liquidity provision in different
  market conditions

10
Data and variables
11
Data sources

• Intraday quote and trade data from
  Thomson-Reuters Tick History (TRTH) and Trades
  andQuotes (TAQ)
• 42 stock exchanges, 37 countries, 2001-2011
• on average about 21,552 common stocks per year
• Daily data on returns, volume, price from
  Datastream and Center for Research in Security
  Prices (CRSP)
• Buy-side transaction costs data from the Ancerno
  database
• Information about trading protocols from Reuters
  Speedguide, Exchange Handbook, World Federation
  of Exchanges

12
(No Transcript)
13
Proxy for algorithmic trading

• AT - trading volume / messages
• volume per message times (-1), (US100)
• follows Hendershott, Jones, and Menkveld (JF
  2011)
• normalize raw message traffic with trading volume
• messages include trades and quote updates
• for the US, TAQ and System Order Data (SOD) based
  measures are highly correlated

14
Liquidity measures

• Spreads RQS(Ask-Bid)/M, RES2P-M/M
• P is transaction price, M is bid-ask midpoint
• Amihud daily return/dollar volume
• Short Dk ( XP RP ) / RP
• execution shortfall Ancerno actual (buy side)
  institutional price impacts.
• XP is the volume weighted average price across
  component trades of a daily order RP is the
  reference price, defined as the opening price on
  the day of the order D is buy(sell) indicator

15
Informational efficiency

• ARn for various time intervals
• for each stock, compute mid-quote returns for
  various intervals
• then compute autocorrelation of returns
• ignore overnight returns
• no bid-ask bounce in this measure
• The more efficient the stock price (the closer it
  is to a random walk), the smaller is ARn

16
Volatility

• Use several standard volatility measures
• Ret, Ret2 for stock raw daily return
• MktadjRet, MktadjRet2, for market-adjusted
  daily return
• Log (Ret10_Var), Log (Ret30_Var)
• Intraday return variances computed from 10-min
  and 30-min mid-quote returns
• daily relative price range (High-Low)/Close

17
Econometric methodology
18
Main Method

• Have three-dimensional unbalanced panel
• 42 markets, about 2770 days, about 550 stocks per
  market
• Main method estimate firm and day fixed effects
  panel regression for each market, then aggregate
  across markets
• All variables are winsorized (99.5 and 0.5) and
  standardized daily within a market so
  coefficients are comparable across markets

19
Main Method

• Regress market quality measures on AT proxy and
  controls
• All regressions control for volume, volatility,
  inverse price, firm size
• volatility regressions exclude volatility control
  and add controls for RES and AR
• Inference is based on means across markets (42
  coefficients, use cross-sectional t-test for
  inference)

20
Regression results
21
The relation between AT and liquidity

RQS RES Amihud
Mean coef. on AT -0.0093 -0.0097 -0.0110
t-stat -6.69 -3.52 -6.81
positive 2 26 5
More AT activity is associated with higher
liquidity (i.e. lower spreads and smaller price
impact).
22
Cross-sectional tests

• How does the relation between AT and liquidity
  differ for stocks with different characteristics?
• Sort within each market according to firm
  characteristic (e.g. SIZE)
• Create dummies for Small and Large tercile
• Include interactions with AT in regression models

23
AT-liquidity relation for different firm sizes

• Solid colors indicate significance at the 5
  level
• More AT is associated with better liquidity in
  medium and large stocks

24
The relation between AT and informational
efficiency
AR10 AR30
Mean coef. on AT -0.0126 -0.0042
t-stat -7.23 -4.01
positive 7 14

More AT activity is associated with better
informational efficiency.
25
AT-efficiency relation for different firm sizes

• More AT is associated with consistently better
  price efficiency in large stocks.

26
The relation between AT and volatility
ret Ret2 PriceRange Ln(Ret10_Var) Ln(Ret30_Var)
Mean coef. of AT 0.027 0.0182 0.0401 0.0216 0.0295
t-stat 7.65 6.67 8.52 4.25 5.17
positive 81 81 83 71 79

• More AT activity is associated with higher
  volatility.
• control for efficiency and liquidity.

27
AT-volatility relation for different firm sizes

• The positive AT-volatility relation decreases
  with firm size.

28
Is AT-related volatility associated with improved
liquidity?

• A two-step procedure
• 1. estimate a cross-sectional regression within
  each market each day, using liquidity,
  efficiency, and volatility as dependent
  variables, and record the AT coefficients.
• 2. compute Spearman rank correlations between AT
  coef on liquidity and AT coef on volatility.
• The correlations are positive (0.02 -
  0.14).
• On an average day, when AT is associated with
  higher volatility, AT also is associated with
  wider spreads.

29
Is the AT-related volatility good volatility?

• Probably not.
• Controlling for the efficiency of prices in
  regressions produces the same result.
• This implies that higher volatility cannot easily
  be attributed to greater price efficiency
  accompanied with higher AT.

30
How resilient is AT liquidity supply in different
market conditions?

• Identify days when market making is difficult
• MMs dislike one-sided order flow that moves
  price.
• E.g., consider sell imbalance price moves down,
  MM is long, faces inventory losses
• Tend to cut back on liquidity provision on such
  one-sided trading days
• Tend to cut back more when imbalance continues
  through the next day

31
A proxy for difficult MM days

• Select all days when the daily return has the
  same sign as the previous days return
• Set HARD dummy to one on these days if the 2-day
  cumulative return exceeds the 20-day historical
  mean by at least one standard deviation
• Then interact with AT as before

32
How does AT change on difficult market-making
days?

• More AT is still associated with higher
  liquidity, but this is significantly less than on
  regular days
• Greater information content of trades (RPI)
• Smaller reward for providing liquidity (RRS)
• More AT, higher volatility and efficiency
• If AT use MM strategies on average, they tend to
  resort to other strategies when market making is
  unusually difficult.

33
How does this compare to traditional market
makers?

• The importance of traditional vs. HFT market
  making should increase with AT.
• Compare low-AT tercile (traditional MM) with
  high-AT tercile (new MMs) by estimating the
  HARD interactions separately

34
Results on new vs. traditional market makers

• AT benefits are concentrated in traditional MM
  stocks, especially on HARD days.
• Negative AT association (higher volatility, no
  liquidity improvement) are concentrated in new-MM
  stocks where traditional MMs are less important.
• Not clear that HFT MM are substitutes for
  traditional MMs, consistent with Anand and
  Venkataraman (2012).

35
Which way does causality go?

• Use co-location within each market as an
  instrument for AT
• Estimate IV regression at the market level
• compute value-weighted daily averages for each
  market
• estimate first stage regressions of AT on
  co-location dummy variable with market and day
  fixed effects
• estimate second-stage IV model using predicted
  values from 2

36
IV estimation using co-location as an exogenous
shock to AT
Dependent variable AT coefficient t
RQS -0.023 -4.06
RES -0.045 -7.12
Amihud -0.003 -0.32
Shortfall -0.024 -1.95
   
AR10 -0.041 -4.00
   
PriceRange 0.060 9.99
ln(Ret10_Var) 0.076 15.62
Ret 0.066 9.16
Results are largely unchanged with IV.
37
Robustness checks

• Results maintain if we
• control for news announcements
• exclude financial crisis period
• use Fama-McBeth regression for weekly or monthly
  aggregation periods
• run time-series regression at firm level first,
  then aggregate across firms

38
Conclusions

• Algo trading
• improves liquidity and informational efficiency
• increases volatility, even when controlling for
  efficiency and liquidity
• But
• Little liquidity effect in the smallest third of
  firms in each market
• AT increases volatility the most for small firms
  that are small.
• On days when market making is more difficult, AT
  provides less liquidity, increases information
  content of trades, and increases volatility more.

39
Conclusions

• Volatility increases with more AT what exactly
  are the implications?
• In assessing the current market structure, market
  observers should take into account that
• the effects of AT are not uniform across markets,
  across stocks, and over time.

40
Thank you for your attention.