The Evolution of Informed Liquidity Provision: Evidence from an Order‐driven Market

Date01 November 2016
DOIhttp://doi.org/10.1111/eufm.12082
AuthorJoey W. Yang,Marvin Wee
Published date01 November 2016
The Evolution of Informed Liquidity
Provision: Evidence from an
Order-driven Market
Marvin Wee and Joey W. Yang
UWA Business School, Discipline of Accounting and Finance, University of Western Australia,
35 Stirling Hwy, Crawley, WA 6009, Australia
E-mails: marvin.wee@uwa.edu.au; joeywenling.yang@uwa.edu.au
Abstract
The liquidity provision strategies by institutional traders on the ASX have changed
over the period 2006 to 2012. Besides using smaller-sized orders more frequently
than their retail counterparts, they have increased the use of passive limit orders.
Institutional traders are found to be more sensitive and responsive to changes in
market conditions. Analyses on order placement and price impact suggest that
institutional traders are better informed. However, their limit orders are found to
have a lower price impact at the intra-day level in the 2012 subsample period. We
show evidence this is associated with the proliferation of algorithmic and high
frequency trading.
Keywords: liquidity provision, informed trading, limit orders, information
asymmetry
JEL classification: G10, G12, G13
1. Introduction
The issue of liquidity provision and consumption in securities markets is a signicant
source of concern for market participants, regulators, and stock exchanges. This was
highlighted in the ash crash that occurred on 6 May 2010 on the New York Stock
Exchange (NYSE), for which blame was apportioned to the withdrawal of high-
frequency liquidity providers from the markets (Buchanan, 2012).
1
Some studies have
The authors would like to thank two anonymous referees for their useful suggestions. The
authors would like to acknowledge Sirca for kindly providing the data. The authors are
grateful for comments from Tom Smith, Alfonso Dufour, and audience members at the
European Financial Management Association 2013 Conference in Reading.
1
The dynamics of liquidity in an electronic system in the context of market reaction to
liquidity shocks is analysed by Gomber et al. (2015).
European Financial Management, Vol. 22, No. 5, 2016, 882915
doi: 10.1111/eufm.12082
© 2015 John Wiley & Sons, Ltd.
challenged our understanding of the roles that different participants play in liquidity
provision (Menkhoff et al., 2010) and the assumption that market liquidity is provided by
the uninformed traders (e.g., Glosten, 1994; Rock, 1990, Handa et al., 2003). For
instance, Bloomeld et al. (2005) nd that informed traders play an important role in
liquidity provision. More recently, Rosu (2015) proposes that having a higher share of
informed traders improves liquidity. The 2010 ash crash further highlighted the
importance of understanding the dynamics of liquidity provision and consumption in
securities markets and spurred research on the effects of algorithmic and high-frequency
trading on liquidity provision (e.g., Hendershott et al., 2011; Hendershott and Riordan,
2013; Hoffmann, 2014).
We investigate the provision of liquidity by different traders over an extended period
spanning 2006 to 2012 in a limit order book setting. The choice of the sample period
allows us to provide insight into the evolution of liquidity provision over the past decade
given the substantial advancement and inuence of technology on order and trader
composition.
2
In examining the provision of liquidity, we also study the contribution by
institutional and retail traders over the trade day and how their provision of liquidity
varies with market conditions. In the second half of our analysis, we examine the
informativeness of the market and limit orders placed by institutional and retail investors.
Kaniel and Liu (2006) argue that the limit orders placed by informed traders are more
informative than are their market orders, while Rosu (2015) documents that the
price impact of limit orders is one fourth of that of market orders. Our study differs in that
we examine, empirically, whether institutional and retail traders are informed when they
use limit orders. We also examine whether the informativeness of the different order
types has changed over our sample period.
Understanding the sources of liquidity provision in an order-driven market such as the
Australian Securities Exchange
3
(ASX) is particularly pertinent, as the viability of this
type of market depends on public traders providing liquidity. With algorithmic trading
(AT) and high frequency trading (HFT) having become more prevalent (for example,
Hendershott et al., 2011; Hendershott and Riordan, 2013), it is unclear whether
institutional traders, faced with the increased risk of being picked off by traders using
computerised trading, continue to provide liquidity in the market place. Our data span a
sample period characterised by a substantial change in the prevalence of algorithmic and
high-frequency traders. In view of the recent debate on the effects of AT and HFT, our
paper provides insights into the impact of these developments.
Although earlier theoretical models propose that informed traders use market orders
due to the short-lived nature of their information, this proposition is challenged by recent
empirical evidence that nds informed traders also using limit orders (Anand et al., 2005;
Bloomeld et al., 2005; Chakravarty and Holden, 1995).
4
Consistent with those studies,
we propose that informed traders are more likely to provide liquidity than are uninformed
2
Our initial sample period includes data from 2003. However, the results are not tabulated, as
they are similar to those for 2006.
3
With a total market capitalisation of approximately A$1.3 trillion (December 2011), the
ASX equity market is currently ranked sixth largest in the world in terms of free-oat market
capitalisation.
4
Also see Harris and Hasbrouck (1996), Biais et al. (1995) and Grifths et al. (2000) for
implicit evidence.
© 2015 John Wiley & Sons, Ltd.
The Evolution of Informed Liquidity Provision 883
traders, as the former are less likely to face the risk of being picked off. We also propose
that market conditions readily affect informed tradersprovision of liquidity. Boulatov
and George (2013) model the equilibrium liquidity provision and demand of informed
traders and postulate that a market mechanism that encourages informed liquidity
provision will improve market quality. We investigate the effects of electronic trading on
liquidity provision by institutional traders.
Our paper contributes to the existing literature in three ways. Firstly, our study
complements the work by Anand et al. (2005) who test the experimental study of
Bloomeld et al. (2005) in a hybrid market setting, and provides empirical evidence on
Rosus (2015) predictions of the effects of informed trading on liquidity provision. While
we also examine the changes in strategy of traders over the course of the trading day, our
study differs from that of Anand et al. (2005) in that we use an electronic limit order
market setting and directly investigate the hypotheses proposed by Bloomeld et al.
(2005) and Rosu (2015). Cai et al. (2013) compared and contrasted the market reaction to
news arrival for a dealer and an electronic market. They found that automated electronic
markets minimise adverse selection costs hence total trading costs due to intermediated
trading characteristics. Secondly, we extend Anand et al. (2005) by verifying their
assumption that institutional traders are informed. Specically, we test whether
institutional tradersliquidity provision is driven by information. Further, we explore the
effects of the advancement in technology for trading on the informativeness of limit
orders.
Theory suggests that prices impound information in the trading process (Kyle, 1985;
Glosten and Milgrom, 1985, among others) such that orders placed by informed traders
would cause more prolonged price impact, or permanent price movements towards its
fundamental value. Therefore, we examine the informativeness of the limit orders
through their price impact in each aggressiveness category. We then examine the
evolution of liquidity provision by different trader types over an extended period,
allowing us to document any changes in order submission that may have been brought by
the widespread use of algorithmic and high-frequency trading (AT and HFT) by
institutional traders.
Using a dataset comprising stocks in the ASX50 from the years 2006, 2009 a nd
2012, we nd the familiar intra-day U-sh aped pattern in order volum e and frequency
in which the midday is associated with a lo wer level of order placement activi ty. We
also nd that institutional trader s place more limit orders than market order s, and that
institutional traders ar e more likely to use limit orders t han market orders compared to
retail traders. This suggests that in stitutional traders play a vital role in the provision of
liquidity on the ASX. The subst antial increase in the use of smaller-sized limit orders
by institutional traders in 2009 and 2012 is likely to reect the growth in alg orithmic
and high-frequency trading. We do not nd a similar increase in the use of limit orders
by retail traders. We also show that, compared with retail traders, instituti onal traders
are more sensitive and respons ive to changes in market conditio ns when formulating
their limit orders placeme nt strategy. In the study of price impact of ord ers using
5-minute, 60-minute and d ay-end horizons, we nd evi dence to suggest that
institutional tradersmarket orders and limit orders placed at the be st bid and ask
prices are better informed th an are those of retail traders. We nd the increase in AT
and HFT partially accounts for the different ndings in 2012. In particular, the resu lts
suggest that AT has weakened the shor t-run informativeness of the li mit orders placed
by institutional traders.
© 2015 John Wiley & Sons, Ltd.
884 Marvin Wee and Joey W. Yang

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT