The current issue and full text archive of this journal is available on Emerald Insight at:

https://www.emerald.com/insight/1755-4179.htm

Systematic
Limit order books: a systematic review of
review of literature literature
Abhinava Tripathi, Vipul and Alok Dixit
Department of Finance and Accounting, Indian Institute of Management,
Lucknow, India
Received 8 July 2019
Revised 6 December 2019
Abstract 31 March 2020
Accepted 7 May 2020
Purpose – This study aims to provide a systematic literature review of the research study in the area of
limit order book (LOB) mechanism of trading and its implications for market efficiency. The study attempts to
document the recent theoretical developments and empirical findings from the literature exhaustively and
identifies the research gaps for future research.
Design/methodology/approach – The study uses seven reputable databases to select 2,514 research
studies spanning over 1981-2018 (finally compressed to a pool of 103 articles, based on relevance and impact). The
study uses bibliometric network visualization and text analytics to categorize and examine the literature. The
chosen articles are compiled and analyzed to provide a comprehensive account of the current research on LOBs.
Findings – The recent LOB literature is summarized on various criteria as follows: sub-areas, the types of
economies and markets, methodologies and the LOB measures. The review identifies a dearth of studies on
the LOBs in emerging markets. It suggests the potential research areas as intraday studies in emerging LOB
markets; application of market indicators based on deeper levels of LOB, beyond the best prices; market
fragmentation, order routing decision and its impact on order execution quality; optimal display of LOB
levels; liquidity dynamics in quote-driven markets vis-à-vis LOB markets; effect of high-frequency trading on
market microstructure; application of advanced techniques (e.g. machine learning models, zero-intelligent
models); relationship between the trading speed, order aggressiveness, shape and resilience of the order book
and informed trading; and information content of the auxiliary order submission strategies, including
cancellation, amendments and hidden orders.
Originality/value – For the past 15 years, to the best of the knowledge, a comprehensive review of the
literature on LOBs has not been published. The financial markets have transformed significantly over this
period, driven by the adoption of LOBs, low latency trading and technological advancements in information
dissemination. This article provides an extensive collection and classification of the literature on LOBs. This
would be useful for the practitioners, future researchers and academics in the area of financial markets.
Keywords Limit order books, Market microstructure, Financial markets,
Systematic literature review, Bibliographic network visualization
Paper type Literature review

1. Introduction
Since 1990s, the majority of financial markets have adopted limit order books (LOB) for
the execution of trades. This includes the conventional quote-driven markets (e.g. the
USA) and new age order-driven emerging markets (e.g. China, India). The most notable
works that provided a comprehensive account of the LOB literature are Coughenour and
Shastri (1999), Madhavan (2000), Biais et al. (2005) and Parlour and Seppi (2008).
However, over the past 15 years, financial markets have undergone major technological

The authors thank Prof Mark Aleksanyan (Associate Editor) and two anonymous reviewers for their
Qualitative Research in Financial
in-depth review of the manuscript and the insightful comments. Markets
The authors received no financial support for the research, authorship, and/or publication of this © Emerald Publishing Limited
1755-4179
article. DOI 10.1108/QRFM-07-2019-0080
QRFM transformations, driven by the algorithmic low latency trading and the rise of digital
information aggregators (e.g. Bloomberg and Reuters). This has major repercussions for
the market microstructure, affecting market efficiency and market quality [1]. During this
period, the empirical research in the area has also witnessed significant but fragmented
growth, driven by the availability of quality high-frequency data, advanced econometric
tools and enhanced computing power. In this backdrop, the understanding of LOB
microstructure is particularly important to academics, practitioners and regulators, for
its role in pre-trade transparency, price discovery, liquidity, market quality and thereby
market efficiency. However, there are no comprehensive reviews of literature on LOB
over this eventful period. This study aims to fill the gap by providing a systematic
account of the fragmented and diverse strands of literature on LOB. In particular, the
study investigates the following research question (RQ):

RQ1. What are the implications of LOB microstructure on market efficiency?

We address this question by examining the important strands of market efficiency
literature, namely- price discovery, pre-trade transparency, information asymmetry,
volatility, liquidity, the shape of order book, order submission strategies and order
aggressiveness, dark trading and iceberg orders and low latency trading. The study covers
a battery of literature documenting insights and characteristic implications for market
efficiency associated with LOBs, over the past 38 years (1981-2018).
The traditional market microstructure used designated market makers (e.g. dealers,
specialists), who provided liquidity and continuity to the market. These market makers were
central to the theoretical underpinnings in the microstructure literature before 1990s (Glosten
and Milgrom, 1985; Ho and Stoll, 1983; Kyle, 1985). Post-1990s, a majority of financial markets
have used LOBs. These LOBs are devoid of designated market makers. The LOBs operate
with two types of orders, namely, limit orders and market orders (Hollifield et al., 2004). A limit
order is a commitment made at a specific time to trade a given volume of a security at a pre-
specified price limit. A market order is executed immediately at the best price(s) available. The
unexecuted limit orders are stored in the LOB until these are matched or canceled. In LOB
microstructure, traders supply liquidity through limit orders and consume liquidity through
market orders (Bloomfield et al., 2005). A limit order faces two major risks [2], namely:
(1) Execution risk because of the probability that it will not be filled.
(2) Pick-off risk because of the presence of informed traders.

A limit order offers more favorable pricing as against a market order (Biais et al., 1995;
Handa et al., 2003; Handa and Schwartz, 1996).
Technological advancements in telecommunication and growth of internet have
transformed the operations of stock exchanges across the world (Jain, 2005; O’Hara, 2015;
Rosu, 2009). The trading volumes and order frequencies have increased drastically. LOBs
facilitate this fast-paced and highly competitive trading environment. The LOB markets
operate through highly automated electronic communication networks (ECNs) with certain
protocols [3] of order matching and trade execution (Madhavan, 2000). The major
advantages of LOBs include:

low trading costs (brokerage, commissions, spreads and fees, etc);

remote access to market participants;

pre- and post-trade transparency;

efficient price discovery through comprehensive information aggregation; and

less information asymmetry and more competitive market microstructure (Anand Systematic
et al., 2005; Back and Baruch, 2007; Ranaldo, 2004). review of
Presently, 16 exchanges across the world comprise about 90% of the global security market literature
capitalization [4]. All these exchanges use LOBs with varying degrees of trading activity. For
example, the USA [New York stock exchange (NYSE) and National Association of Securities
Dealers Automated Quotations (NASDAQ)], the UK [London stock exchange (LSE)] and Toronto
[Toronto stock exchange (TSX)] use LOBs, in the presence of designated market makers. The
major pure LOB markets include China [Shanghai stock exchange (SSE), Shenzhen stock
exchange (SZSE)], Japan [Tokyo stock exchange (TSE)], Australia [Australian securities exchange
(ASX)], Hong-Kong [Honk Kong stock exchange (HKEX)], Germany [Frankfurt stock exchange
(FSE)] and India [National stock exchange (NSE) and Bombay stock exchange (BSE) [5].
The practitioners, researchers and academics in the area may peruse the article as a ready
reference for understanding the LOB market microstructure and its implications for market
efficiency. The article encompasses a large number of LOB markets from various economies,
both developed and emerging. For the systematic review and objective interpretation of the
literature, the study uses bibliometric network visualization (VOSviewer) methodology and
word-cloud analysis. The study summarizes the major findings and approaches in the extant
LOB literature. It documents a plethora of order book related parameters [6] and econometric
methodologies (for relationship estimation), as used in the literature. For future research, the
study identifies the following important research gaps: the relative dearth of studies in
emerging LOB markets; market fragmentation, order routing decision and its impact on order
execution quality; optimal display of LOB levels; liquidity dynamics in quote-driven markets
vis-à-vis LOB markets; market indicators based on LOB that capture the market
characteristics better than those based on the best prices; low latency trading; auxiliary order
submission strategies (cancellation, amendment and hidden orders); and shape of the order
book (slope and resilience). For practitioners, it identifies LOB attributes that may help in
predicting short-term (ST) returns and formulating trading strategies. Finally, the study also
documents the insights developed by researchers into price instability and episodic failures of
trading activity associated with LOBs; this may be useful to policymakers and regulators in
minimizing the impact of crisis events.
The remaining article is structured into four sections. Section 2 elaborates the data and
methodology. Section 3 provides a qualitative discussion of the literature reviewed. Section 4
discusses the insights, research gaps and the future scope and Section 5 concludes.

2. Data and methodology

2.1 Data
This study peruses 103 research articles on the LOBs in order-driven markets, spanning the
period 1981-2018. It includes studies from highly ranked journals and working papers from
globally recognized and reputable institutions.

2.2 Methodology
The article uses the systematic literature review (SLR) methodology to compile and examine
the extant literature on LOBs (Kitchenham, 2004; Tranfield et al., 2003). The methodology
offers a well-accepted protocol to identify, select, review and synthesize the relevant literature.
The steps followed in the study, in accordance with the SLR methodology, are provided in
Figure 1. We adopt this framework and customize the steps for this article, as follows.
2.2.1 Purpose and scope definition (Step 1). To establish the contours of the literature
review, we define our scope through the following main RQ: what are the implications of
QRFM

Figure 1.
A brief indicative
illustration of the
steps involved in the
SLR methodology

LOB microstructure on market efficiency? We explore various sub-strands of the

fragmented LOB literature and tie them together within our central theme, i.e. market
efficiency. More specifically, the following research issues guide our journey around the
central theme. (a) What are the important characteristics of LOB microstructure? (b) How
does the LOB microstructure differ from the conventional microstructure (e.g. formation of
spread, role of market makers)? (c) Empirical evidence on the relation of LOB with various
dimensions of market efficiency (i.e. price discovery, information asymmetry, transparency,
volatility and liquidity), (d) New developments in market microstructure literature (i.e.
algorithmic low latency trading, iceberg orders, dark trading, applications of models from
allied fields such as physics and mathematics) and (e) What are the potential research gaps
that may determine the future avenues of research?
2.2.2 Explore, identify and compile research articles (Step 2). To aggregate high-quality
research articles on LOBs, the study uses seven important databases, namely, Emerald,
Wiley Inter-Science, EBSCO, Elsevier Science Direct, web of science, Scopus and JSTOR. Systematic
Three broad keywords are chosen to filter the relevant research articles: Order-driven review of
markets, LOBs and Market microstructure. The search leads to a large pool of 2,514 research
articles. It is not feasible to review such a large corpus of research articles. To make the task
literature
of review feasible, the study uses VOSviewer (Van Eck and Waltman, 2009) for bibliometric
mapping and network visualization [7] of the literature. The methodology ensures a wide
coverage of literature and inclusion of all the important contributions. VOSviewer uses the
meta-data from the corpus of 2,514 articles to perform the co-citation, co-occurrence and
bibliographic coupling of the articles across sub-areas, authors and journals. It also
performs these analyses over time, which indicates the evolution pattern of literature. The
co-citation analyses and bibliometric coupling is performed across authors, documents and
sources. Similarly, the evolution pattern over time is examined across keywords, documents,
authors and sources.
Figure 2 provides an illustrative sample of bibliometric network visualization. The size of
the node (circle) indicates the weight of the sub-area (in terms of the number of articles). The
articles, authors and journals with the same color are clustered together and their proximity
reflects the similarity of their subject matter. VOSviewer provides a pictorial representation
of the major sub-areas, their relative place in the literature, the associations between the
strands of literature, significant contributions and their sources, most prolific authors and
evolution of the subject area over time [8].
The pool of articles is narrowed down from 2,514 articles to 500 articles, using the
bibliometric analyses facilitated by VOSviewer (keyword co-occurrence analysis, co-citation
and bibliographic coupling analysis) [9]. While the study covers significant contributions in
the area, it is difficult to encompass the entire ground in a limited space. There is a
possibility that some relevant articles may have been left out, which is an inherent risk in
any literature survey of this kind (Grant and Booth, 2009). However, the bibliometric
network visualization ensures that no major sub-area is left out completely and the other
related articles would largely compensate for specific exclusions.
2.2.3 Evaluation and review of the selected articles (Step 3). In the final stage of filtration,
we aim to identify the most significant and diverse contributions that cover the major
interest areas over time and across different markets. An important objective of SLR is to
minimize various subjective biases in the selection of articles and to ensure the quality,
relevance, diversity and appropriateness of these articles. We use the following conditions
for the final stage of filtration:

Studies that are not related to market efficiency are excluded.

Studies on order-driven markets that do not examine the LOB data specifically are
excluded.

Priority is given to highly cited research articles from peer-reviewed academic
journals of international repute [10], where all the details (including complete text)
are available. Articles that are not from rated journals or with less than 5-citations
are omitted.

For selection from a similar set of studies, preference is given to the articles that use
novel databases (particularly high-frequency LOB data from emerging economies)
and advanced methodologies. This ensures diversity and adequate coverage of
emerging markets, as these markets have attracted considerable attention for the
past five years.

Finally, we also conduct a backward iterative review of the references from the final
set of articles (Al-Emran et al., 2018; Busalim, 2016; Gutiérrez-Nieto and Serrano-
QRFM

Figure 2.
A sample illustration
of keyword co-
occurrence and their
evolution over time,
co-citation of
references and
bibliographic
coupling of authors
using VOSviewer-
from the meta-data of
2,514 articles
Systematic
review of
literature

Figure 2.
QRFM Cinca, 2019; Henriksen-Bulmer and Jeary, 2016) to ensure that all the significant
contributions and important areas are covered (Backward-snowballing). Seven
important studies have been identified using this method.

The filtration criteria lead to 103 articles spanning a period of 38 years (1981-2018) [11]. A
detailed review of these articles is provided in Section 3. Step 4 in Figure 1 pertains to the
summary, key insights and future avenues of research. It is discussed in detail in Section 4.
2.2.4 Classification of the articles. We perform a descriptive classification analysis of the
final set of articles to assess their quality. These articles are classified based on the sub-
areas being studied, the nature of economy, the methodology applied, year of publication,
key journals, sample years, country-wise distribution and LOB measures. This helps in
understanding the evolution of the literature, trend and other related issues. The figures
showing the analysis are provided in Appendix 2 (Figures A1-A6).
2.2.4.1 Analysis by year of publication. The analysis by year of publication (Figure A1)
suggests that the area has gained traction only after 2005, although the articles were being
published regularly since 1990s. This increased interest may be ascribed to the availability
of high-frequency data, which has generally led to the growth in empirical investigations.
2.2.4.2 Analysis by journal frequency, number of citations and journal rankings. The
analysis by journal frequency (Figure A2) indicates that a large number of articles are from
the most reputable journals in the area, namely, Journal of Financial Economics, Journal of
Finance, Journal of Financial Markets, Journal of Banking and Finance and Journal of
Financial and Quantitative Analysis. Citation analysis indicates that more than 60 articles
have received citations in the range of 100-500 and 23 of these articles have more than 500
citations each. From the selected articles, 74% come from A* Australian Business Deans
Council (ABDC) and 61% come from 4* Chartered Association of Business School (CABS)
journal rankings. This reveals the attention and importance received by the topic and also
the relevance and impact of the research articles selected in the current study.
2.2.4.3 Distribution of research articles based on the Sub-areas of limit order book. The
majority of the articles on LOB have explored its relationship with liquidity, asset pricing,
information asymmetry, trader behavior, volatility and high-frequency trading (HFT).
These sub-areas (Figure A3) are integral to the topic of market efficiency and remain the
most explored areas over the years. Some of the other notable areas include market design,
market quality, order book shape and aggressiveness, transparency and hidden (iceberg)
orders.
2.2.4.4 Distribution of research articles based on the geographical distribution and sam-
pling frequency of the data. The research articles selected for this study cover diverse
geographical regions (Figure A4); this includes the markets of America, Asia, Europe and
other emerging economies. A vast majority of the articles, however, have examined the
developed markets (80%), especially the USA This is in order, as the majority of good
quality studies are conducted in the developed economies given the early development of the
financial markets and availability of good quality data. The databases such as trade
reporting and compliance engine (TRACE), center for research in security prices (CRSP),
institute for the study of security markets (ISSM) and Data stream have made the quality
data available, including the time-stamped intraday data. Notably, the sample periods over
the years have shortened owing to the availability of high-frequency intraday data; almost
75% of the articles use intraday data.
2.2.4.5 Analysis by methodologies and order-book measures used. The selected articles
are a mix of empirical estimation, theoretical model building, event study and literature
survey (Figure A5). As regards statistical/econometric techniques, most of the articles have
applied the Ordinary least square (OLS) regression. Some of the more frequently used Systematic
regression techniques include the market model, the Fama-Factor models (with 2, 3, 4 review of
factors), the Fama-Macbeth model, pooled sample analysis, time series analysis and cross-
sectional analysis. Some articles have also applied a wide variety of methodologies, viz., the
literature
seemingly unrelated regression, the vector auto-regression, the granger causality analysis
and the panel regression. For the estimation of these models, the studies use the maximum
likelihood estimation, the generalized method of moments and the generalized least square
procedures. The state of order book (breadth, tightness, resilience, depth) in a financial
market is ascertained using a number of measures (such as spread, volume, slope and
impact cost). As evident from Figure A5, the spread, depth and mid-quote appear to be the
most commonly used measures, followed by the order imbalance, order aggressiveness,
slope and impact cost measures. The formulae for some of the important measures are
provided in Appendix 3.
2.2.4.6 Trends in the limit order book research: a word-cloud analysis of the literature.
One potential concern with the SLR methodology is that it may be affected by the subjective
judgment of authors. This may result in a biased selection of key areas of research and
subsequent classification, categorization and analysis of the literature. To review the articles
thoroughly and to ensure an exhaustive coverage of different sub-areas, a word-cloud
analysis has been carried out for the complete text from the selected 103 research articles
(Figure A6). This text-analysis offers the readers an exhaustive and objective picture of the
literature under review. The word-cloud (presented here) includes the top 100 words in the
order of their frequency; here, the words with relatively larger sizes occur more frequently in
the selected literature.

3. Literature review of limit order books

This section examines the major strands of literature on LOBs in relation to market
efficiency theory of market microstructure. These strands (sub-areas) of literature and
significant contributions are identified from prior literature and by implementing
bibliometric analysis (VOSviewer) with subjective judgment of the authors, where needed.
Our initial classification of the sub-areas of literature included the union of different
categorizations suggested by Biais et al. (2005), Coughenour and Shastri (1999), Madhavan
(2000) and Parlour and Seppi (2008) in their literature surveys of market microstructure
(Classifications 3.1-3.4, 3.7 and 3.9). We augment this union of categories to include the
topics of more recent interest, based on bibliometric network analysis (Categories 3.5, 3.6, 3.8
and 3.10). In this process, we ensure the inclusion of all the important topics, represented by
the dark blue colored large-sized nodes in the bibliometric networks and the topics that are
getting considerable attention in the more recent literature (yellow colored nodes) [12]. The
categories of sub-areas are appropriately named and renamed (those from the prior
literature) to represent their constituent topics. The robustness of sub-area-classification is
validated using “bibliometrix” R package and Backward-Snowballing technique.

3.1 Price discovery and transparency

Heterogeneous investors [13] trade asynchronously with different sets of information.
Therefore, efficient price discovery, transparency and market clearing become the important
objectives in organizing exchanges. The contribution of LOBs toward these objectives is
well-documented (Baruch, 2005; Harris, 1998, 1990; Harris and Panchapagesan, 2005). Harris
(1990) observed that pre-committed traders and value-motivated traders operate in LOBs.
Value-motivated traders exhibit their information with limit orders. In contrast, the pre-
QRFM committed traders provide market orders, when their limit orders fail to execute. Overall, the
order aggregation from both kinds of traders leads to efficient price discovery.
LOBs, especially the open books, are more efficient in order aggregation and in
facilitating a more competitive market microstructure, as compared to dealer markets (Cao
et al., 2009). LOBs offer pre- and post-trade [14] transparency through the dissemination of
real-time information. The information is used by traders, academics, corporates and
policymakers. This leads to timely information aggregation and hence, a lower information
asymmetry. Moreover, exchanges also use microstructure related rules (e.g. pre-trade
auctions, circuit breakers, trading halts, block-trading sessions) to maintain orderly trading
and contain spurious volatility in prices. These rules of microstructure support LOBs in the
discovery of informationally efficient prices (O’hara, 1995). Overall, LOBs offer a significant
improvement in information dissemination, transparency, competition, efficient price
discovery, fairness and social welfare (Baruch, 2005) – the most important dimensions of
market efficiency and market quality.
The rise of decentralized LOBs has led to considerable market fragmentation in the past
10 years. These LOBs provide alternative trading venues (BATS, Chi-X, EdgeX, Turquoise,
etc.) that compete for market share with the conventional exchanges (NYSE, NASDAQ, etc.).
As a consequence, the issue of order routing across different venues has gained a greater
interest among investors (Battalio et al., 2016; Colliard and Foucault, 2012; Maglaras et al.,
2015) [15]. This decision entails investor welfare as the key consideration, which, in turn, is
affected by market design (e.g. make and take fee schedules across trading platforms) and
the fiduciary responsibility of brokers [16]. The extant literature predominantly considers
the trading platform as given (i.e. single platforms such as NYSE for the execution of limit
orders). Thus, future works can substantially contribute to the literature by empirically
examining the issue of order routing and its impact on the quality of limit order execution.
During the past 15 years, with the availability of high-frequency intraday data and
advancements in low latency trading, the debate on price discovery and transparency has
shifted to intraday horizons (Brogaard et al., 2014, 2019; Hendershott and Riordan, 2013;
O’Hara, 2015). While there are a number of studies in this domain in the developed markets
(e.g. the USA), the emerging order-driven markets are less explored (Anagnostidis et al.,
2016; Arjoon, 2016; Bai and Qin, 2015; Bekaert et al., 2007; Griffin et al., 2010; Kim and
Shamsuddin, 2008; Lesmond, 2005; Rejeb and Boughrara, 2013; Syamala et al., 2014). The
exploration of emerging markets in the context of price discovery and transparency, using
high-frequency data, may significantly contribute to the existing state of knowledge in this
domain.

3.2 Liquidity
In LOB markets, market orders consume liquidity and limit orders supply it. Traders
monitor and compete to consume and provide liquidity through these orders. Thus, a
trader’s decision to supply or demand liquidity is endogenous - aimed at optimal execution
of trades within a given set of market parameters (Chung et al., 2012; Foucault, 1999; Harris,
1998; Parlour, 1998). Traders formulate complex trading strategies to gain price-time
priority in LOBs (Kumar and Seppi, 1994). For example, to improve upon the existing best
bid-ask [17] quotes, traders hit between the quotes, i.e. undercutting. These actions reflect
the competition in the demand and supply of liquidity (Biais et al., 1998).
The impatient [18] traders use market orders against limit orders to consume liquidity. In
contrast, the limit order traders are more patient and they supply liquidity. Hence, the limit
order traders receive compensation [19] for their patience and for the service of liquidity
provision. In the process, they are also exposed to execution risk and pick-off risk (trading
against informed investors). Overall, the LOBs serve the useful purpose of matching the Systematic
liquidity suppliers and demanders (Baruch, 2005; Seppi, 1997). review of
The proportion of limit orders increases significantly when the spread and other liquidity
measures indicate low liquidity and vice-versa (Chung et al., 1999; Griffiths et al., 2000;
literature
Harris and Hasbrouck, 1996; Hollifield et al., 2004; Ranaldo, 2004). Investors supply liquidity
when it is scarce and costly and consume it when it is cheap and plentiful. In summary, thin
books attract limit orders and thick books attract market orders. This also causes the
stationary and mean-reverting bid-ask spreads. LOBs, through the efficient aggregation of
these orders, maintain the ecological equilibrium in liquidity supply and demand (Handa
et al., 1998, 2003; Handa and Schwartz, 1996).
The nature of liquidity provisioning in LOB markets has an important difference from
that in quote-driven markets. The limit order suppliers – i.e. the de-facto market makers –
have extremely low entry and exit barriers. They can freely provide or withdraw liquidity
from the market depending on the market conditions. This is in contrast to the designated
market makers in quote-driven markets, who have a commitment to provide liquidity. This
important aspect of liquidity provisioning in LOB markets is captured by “free-entry and
free-exit” hypothesis (Brockman and Chung, 2002). However, there are very few studies that
examine this essential property of LOB markets (Anand and Venkataraman, 2016; Tripathi
et al., 2020). Future works may examine this aspect in more detail to address an important
question pertaining to market design, i.e. which of these two market microstructures offers a
more reliable mechanism of liquidity provisioning, particularly during crises?
Furthermore, prior research on the liquidity of LOBs has not addressed the difference
between bid and ask sides of liquidity. More recent evidence indicates that the bid and ask
sides of the book behave asymmetrically, especially during crisis periods (Cenesizoglu and
Grass, 2018; Sensoy, 2019). For example, Cenesizoglu and Grass (2018) found that the ask
side deteriorates and contributes to the crisis events more significantly than the bid side.
Future works may separately examine various dimensions of bid and ask sides of liquidity.
Most of the literature on liquidity, including that on LOBs, measures it in terms of spread
and depth computed with price and volume information from the best bid-ask orders
[Chordia et al. (CRS) [20] 2000, 2002, 2008]. More recent evidence from the developed markets
suggests that these orders may be noisy and informationally less efficient (Amaya et al.,
2018; Cao et al., 2009; Cenesizoglu and Grass, 2018; Deuskar and Johnson, 2011; Dierker et al.,
2015; Domowitz et al., 2005; Kempf and Mayston, 2008) [21]. Therefore, the price and volume
information from the deeper levels of order book may be required to assess the state of
liquidity accurately. There is a deficiency of similar studies in the emerging markets, more
particularly due to the non-availability of quality high-frequency data (Bekaert et al., 2007;
Lesmond, 2005). The prior studies on emerging markets have used low-frequency proxies of
liquidity [22]. The area offers significant potential for future research.

3.3 Components of bid-ask spread – dealer markets vs limit order books

The presence of bid-ask spread in dealer markets is well-documented. These bid-ask quotes
in dealer markets reflect the conscious judgment of designated specialists (market makers).
The primary objective of these specialists is market making and ensuring orderly and
smooth market operations (Chung et al., 2004). The bid-ask quotes reflect the compensation
to these market makers for inventory holding, information asymmetry and order processing
costs (Amihud and Mendelson, 1986; Copeland and Galai, 1983; Demsetz, 1968; Easley
and O’hara, 1987; Glosten and Milgrom, 1985; Ho and Stoll, 1983; Stoll, 1978). More
importantly, the information asymmetry component has a high persistence. In contrast, the
other costs (e.g. order processing costs, inventory holding costs) reflect the transitory
QRFM components of the bid-ask spread (Glosten and Harris, 1988). The evidence in quote-driven
markets ascribes about 60%80% of the bid-ask spread to transitory components (George
et al., 1991; Huang and Stoll, 1997; Lin et al., 1995; Madhavan et al., 1997; Stoll, 1989).
The presence of positive bid-ask spread in LOBs is also well-documented (Ahn et al.,
2002; Biais et al., 1995; Cohen et al., 1981; Glosten, 1994; Handa et al., 1998). However, the
motivations of limit order suppliers are quite contrasting to those in dealer markets (Handa
et al., 2003; Handa and Schwartz, 1996). Unlike the designated market makers in quote-
driven markets, the major concern of limit order suppliers is the profitability of their limit
orders (Chung et al., 2004). Moreover, these limit order suppliers are not obliged to provide
continuous and both-sided quotes. In LOBs, the lowest sell limit order is the best ask price
and the highest buy limit order is the best bid price at any instant. The difference between
these best bid-ask prices is the bid-ask spread at that instant (Ahn et al., 2002; Brockman
and Chung, 1999). The components of bid-ask spread in LOBs include the transaction and
order processing costs, adverse selection costs and waiting costs. In order-driven markets,
the major component (almost 70%) of the spread is highly persistent (Brockman and Chung,
1999; Chan, 2000; Chung et al., 1999, 2004; De Jong et al., 1996). This indicates that a sizable
portion of the bid-ask spread in order-driven markets is related to adverse selection costs.
Overall, the information asymmetry and inventory considerations of designated market
makers play an important role in determining the spread in the quote-driven markets
(Admati and Pfleiderer, 1988; Amihud and Mendelson, 1980; Ho and Stoll, 1983; Kyle, 1985;
O’Hara and Oldfield, 1986; Stoll, 1978). While the implications of these hypotheses on spread
formation are expected to be contrasting [23], they are difficult to disentangle. In LOB
markets, the inventory considerations weaken significantly due to the absence of designated
market makers. In the past 20 years, the LOB market microstructure, that operates in the
absence of designated market makers, has come up significantly, particularly in the
emerging markets. These pure LOB markets (e.g. India) present a natural experimental set-
up to examine the effect of information asymmetry on spread formation (Brogaard et al.,
u, 2009).
2019; Goettler et al., 2009; Rosu, 2016; Ros

3.4 Information asymmetry and limit order books

A major risk faced by limit order suppliers is the pick-off risk; that is, the counterparty may
be better informed and therefore, may get a price advantage. The earlier literature
predominantly suggested that the uninformed traders use limit orders and provide liquidity
to LOBs, while informed traders use market orders and consume liquidity (Copeland and
Galai, 1983; Glosten, 1994). Subsequently, the rational equilibrium models were introduced
(Back and Baruch, 2007; Foucault et al., 2005; Ros u, 2009). These models argued that the
decision to demand or supply liquidity is endogenous. That is, both the informed and
uninformed traders use a mix of market and limit orders (Bloomfield et al., 2005; Harris,
1998; Kaniel and Liu, 2006). Interestingly, the studies based on rational equilibrium models
suggested that informed traders conceal themselves among the uninformed traders. These
informed traders use a mix of both types (market and limit) of orders to avoid detection.
Moreover, Kaniel and Liu (2006) observed that informed traders increasingly use limit
orders, when private information is persistent. In contrast, they prefer market orders to
exploit the private information immediately, when they apprehend that their private
information is short-lived.
Literature has widely documented the informativeness of LOBs and their contribution in
decreasing the information asymmetry (Cao et al., 2009; Goettler et al., 2009; Harris and
Panchapagesan, 2005; Kozhan and Salmon, 2012; Menkhoff et al., 2010). Various measures such
as order book imbalances, depth and spread [24] are used to demonstrate the return-
predictability and information content of the LOBs. Moreover, the widening of bid-ask quotes Systematic
on account of large trades is consistent with the information effects on LOB (Goettler et al., review of
2005, 2009; Parlour, 1998; Parlour and Seppi, 2003). The implications of information asymmetry
for market efficiency are well-documented for the traditional dealer markets (CRS, 2008; Fama,
literature
1998). Similar evidence on LOBs is also available on developed markets (Biais et al., 1995; Cao
et al., 2009; Kempf and Mayston, 2008). The emerging LOB markets are substantially different
from their developed counterparts and offer an exciting research opportunity in this field.
With rising investments in the fast trading technologies (e.g. smart routers, fast access to
exchange data feed, colocation rights and high-speed connections), the ability of market
participants to observe and process market information has improved radically (Biais et al.,
2015; Hoffmann, 2014; Jovanovic and Menkveld, 2016; O’Hara, 2015). This has added a new
dimension to the debate on information asymmetry hypothesis. Fast traders who are
equipped with these advance technologies are associated with informational superiority.
Slow traders, who have not invested much in these technologies, bear high adverse selection
costs. The theoretical underpinnings to the proposition of excess returns available to these
fast traders in efficient markets are provided by Grossman and Stiglitz (1980). Grossman
and Stiglitz argued that those who invest in information acquisition are rewarded in the
form of excess returns. The proponents in the debate argue in favor of banning these fast-
traders or limiting their capacity. It is suggested that the fast trading essentially transfers
the surplus to the technologically advanced market participants, without necessarily
bringing any social benefit. The opponents, on the other hand, contend that the fast traders
bring prices back to informationally efficient values, using their superior information
processing capabilities. Future works may examine the socially desirable level of
investment in these technologies, in the context of LOB markets.

3.5 Effect of short-term price volatility on limit order book s

ST price volatility arises due to the bid-ask bounce related to liquidity trading and
information asymmetry (Handa and Schwartz, 1996). Its effect on LOBs in the context of
market efficiency is well-documented (Goldstein and Kavajecz, 2004; Handa and Schwartz,
1996; Hasbrouck and Saar, 2002) [25]. The ST price volatility because of information
asymmetry is associated with the higher pick-off risk faced by limit orders, due to adverse
selection (Ahn et al., 2001; Foucault et al., 2007; Pascual and Veredas, 2009; Pham et al.,
2018). An increase in the bid-ask spread reflects a premium to bear this risk. Moreover, the
ST price volatility caused by the bid-ask bounce attributable to liquidity and noise-driven
trading is expected to improve the execution probability of limit orders (Bae et al., 2003;
Biais et al., 1995; Chung et al., 1999).
In summary, during the periods of high ST volatility, supplying liquidity becomes a
remunerative proposition and consuming liquidity becomes costly (Handa and Schwartz,
1996). Therefore, this ST volatility also affects the composition of limit and market orders.
More limit orders are submitted when the ST volatility is high and more market orders are
submitted when the ST volatility is low (Bae et al., 2003; Foucault, 1999). These limit orders
increase the depth and, in turn, decrease the ST volatility. Thus, there exists a mean-
reverting equilibrium in LOB liquidity, LOB depth and ST volatility. This mechanism is
instrumental in enhancing the efficiency of LOB markets. The strength of this mechanism is
expected to vary across different microstructure designs (namely, quote-driven, order-driven
and hybrid). Future works may examine this mechanism across different markets,
particularly around the introduction of LOBs, to understand the contribution of different
microstructure designs to market quality and efficiency.
QRFM 3.6 Shape of limit order book
The important parameters indicating the shape of order book include the depth, height,
slope, resilience and elasticity of buy and sell sides (Cao et al., 2009; Naes and Skjeltorp,
2006). The formulae to compute these measures are provided in Appendix 3. The current
state (indicated by shape) of a LOB substantially affects the price formation, liquidity
provisioning, ST price volatility and order choice of traders (Ahn et al., 2001; Cao et al., 2008;
Obizhaeva and Wang, 2013; Pascual and Veredas, 2009; Pham et al., 2018; Rosu, 2012, 2016;
Valenzuela et al., 2015). For example, the distribution of liquidity across the order book
reflects the consensus or the lack of it, about the fundamental price. A greater depth (thick
order book) around the best quotes reflects a consensus on the true price (Goettler et al.,
2005; Valenzuela et al., 2015). In contrast, a high preponderance of limit orders away from
the best quotes (thin order book) reflects significant mispricing. This mispricing may lead to
increasing ST price volatility and may improve the execution probability of limit orders.
The disagreement on fundamental price results in a lower (gentle) slope of LOB and causes a
strong volume-volatility relation. A higher slope reflects a broad consensus about the
fundamental value and is negatively related to trading activity, price volatility and volume-
volatility relation. A skewed order book (excess of one side over the other, buy or sell)
indicates higher trading costs (Biais et al., 1995). A thick book with small spread indicates a
low probability of order execution (Foucault, 1999; Naes and Skjeltorp, 2006). It induces
investors to place the orders that are more aggressive (e.g. market orders, undercutting).
Overall, the shape of order book has significant implications for market efficiency.
A review of the literature on the shape of LOBs indicates that an important measure of
market efficiency, i.e. the resilience of the order book to order flow shocks, remains less
explored (Degryse et al., 2005; Large, 2007; Lo and Hall, 2015). This may be so because the
computation of this measure is data intensive and requires intraday high-frequency order
book and trade book data (e.g. order book snapshots at 1-min, 5-min frequencies). These
kinds of quality high-frequency LOB data have become available only during the past 5-
10 years. Researchers may focus on this potential area of research in the near future,
particularly examining this property around crisis events.

3.7 Order submission strategies and aggressiveness

A sizable volume of literature on LOBs focuses on order submission strategies [26],
particularly the order aggressiveness (Bessembinder et al., 2009; Biais et al., 1995; Griffiths
et al., 2000). The aggressiveness of an order is indicated by impatience – the desire to attain a
higher probability of execution and the demand for liquidity. For example, a market order is
more aggressive than a limit order. Similarly, a limit order that undercuts an existing quote
is more aggressive than a normal limit order that falls behind in the queue. Empirical
studies have widely used the categorization of orders with varying degrees of
aggressiveness (Biais et al., 1995; Cao et al., 2008; Degryse et al., 2005; Duong et al., 2009;
Griffiths et al., 2000). Impatient investors use aggressive orders (undercutting, jumping the
queue) to move ahead in a long queue and improve the execution probability. This leads to a
crowding-out effect on the limit orders on the same side of the book. The degree of
aggressiveness of limit orders is the result of a trade-off between the certainty of execution
and unfavorable prices because of high impact costs (Biais et al., 1995).
Each side of the book (bid or ask) witnesses aggressive order submissions when the
depth on the same side is greater than that on the opposite side and spreads are tight
(Degryse et al., 2005; Duong et al., 2009; Griffiths et al., 2000), indicating a low probability of
execution. The aggressive orders have an enduring effect on prices; they cause widening in
quotes, increase in depth and increase in the ST price volatility.
 The evidence (Bloomfield et al., 2005; Kaniel and Liu, 2006) also suggests that the Systematic
informed and impatient traders submit aggressive orders, particularly when the private review of
information is high-value and short-lived. These aggressive orders face higher impact costs.
When private information is long-lasting, the informed traders split their orders to minimize
literature
the price impact of trades. Institutional investors submit more aggressive orders, as
compared to individual investors, to exploit their private information (Cao et al., 2008; Chou
and Wang, 2009; Duong et al., 2009).
In short, the order submission strategies and order aggressiveness of market participants
have important implications for various dimensions of market efficiency. However, the prior
research has examined only the order submissions. In contrast, various other strategies such
as cancellations, hidden orders, order splitting and order amendments are also available to
the market participants and are less well-understood. With the increasing volumes of low
latency trading, these auxiliary order strategies have significant implications for various
dimensions of market efficiency, including liquidity and price discovery (Baruch and
Glosten, 2013; Cao et al., 2009; Fleming et al., 2018; Hautsch and Huang, 2012; Menkhoff
u, 2009, 2019).
et al., 2010; Ros

3.8 Iceberg orders and dark trading

A large number of exchanges allow varying degrees of order opaqueness through iceberg
orders. There is evidence (Buti and Rindi, 2013; Hasbrouck and Saar, 2013, 2009) that traders
maintain a sizable portion of their orders as hidden, i.e. iceberg or reserve orders [e.g. more
than 20%, 40%, 25% and 15% on the NYSE (USA), Euronext-Paris (France), ASE
(Australia) and FSE (Germany), respectively]. These reserve orders hide the trading
intentions of market participants and contribute to the non-displayed component of liquidity
(Bloomfield et al., 2015) – known as “dark trading.” These orders involve considerable trade-
off between the price and time priority. The visible portion has the priority of execution over
the invisible portion of the order. Therefore, the invisible portion faces a higher execution
risk and unfavorable pricing. The hidden orders are subjected to lower exposure costs and
adverse selection costs (Aitken et al., 2001; Bessembinder et al., 2009; De Winne and D’hondt,
2007). The large visible orders invite undercutting by aggressive traders (exposure costs). In
contrast, when a sizable portion of an order is hidden, the motivation of predatory traders to
undercut is low. The hidden orders also face lower impact costs from informed traders
(adverse selection costs).
Patient traders use reserve or hidden orders when the order size is large, volatility is
high, spreads are wide and order arrival is less competitive on the same side (Bloomfield
et al., 2015; Moinas, 2010; Pardo and Pascual, 2012). Both the informed and uninformed
traders use reserve orders. The informed traders use reserve orders to hide their private
information, while the uninformed traders use reserve orders to avoid undercutting (Aitken
et al., 2001; Anand and Weaver, 2004; Harris, 1996). Market participants assess the market
conditions, shape of the book and the information from order arrival and past trades to ex-
ante detect the extent of hidden orders. The decision to hide a certain portion of the order has
significant implications for the order submission strategies and pre-trade transparency. The
area has caught the attention of researchers only for the past 3-5 years and there is a
significant scope for future exploration. Its implications for market efficiency are being
unraveled with ongoing investigations.

3.9 Market microstructure architecture design

LOBs are strongly influenced by the operational rules and regulations framed by the
exchanges, collectively called the microstructure architecture. Across markets and
QRFM over time, LOBs exhibit characteristic differences in trading activity, liquidity and
efficiency because of these rules. The rules on auction, order submission, order
matching and execution, tick sizes, trading halts, short-sale restrictions, algo-trading
regulations, block trading and role of market makers are some of the examples. The
following findings illustrate the influence of microstructure on LOBs and market
efficiency.At NYSE, a market order larger than the available depth is executed fully,
even if not at the same price. In contrast, at Euronext-Paris, the order is executed
partially and the balance portion is converted into limit order. This causes the NYSE
market orders to be more aggressive in their liquidity demand as compared to those at
Euronext-Paris (Biais et al., 2015). Over the period 1993-2002, NYSE has witnessed
three tick-regimes (i.e. 8th, 16th and decimals) with decreasing minimum-tick sizes.
There is strong evidence that the decrease in tick sizes has caused significant
improvement in market liquidity and informational efficiency (Chordia et al., 2008).
Short-sale restrictions, to some extent, are present in various markets (e.g. NYSE,
NSE, NASDAQ). These restrictions create an artificial asymmetry between the buy
and sell sides of the LOB. These restrictions reduce the ability of sell-side traders to
convey their private information. Hence, the buy-side becomes informationally more
efficient (Baruch et al., 2017; Kaniel and Liu, 2006; Saar, 2001). The role of market
makers varies across different markets. On one extreme are the markets such as NSE,
which is a pure LOB market. On the other extreme are the markets such as NYSE and
NASDAQ, where designated market makers play a significant role in conjunction
with LOBs (Madhavan, 2000). The presence of designated market makers has a strong
impact on market microstructure. First, the LOBs are expected to facilitate a more
competitive microstructure, as against the designated market makers, who create a
monopolistic microstructure (Foucault, 1999; Glosten, 1994; Parlour, 1998; Parlour
and Seppi, 2003). Second, the market makers are strongly driven by the inventory
considerations. Their absence may significantly diminish the inventory
considerations from the market microstructure (Chordia et al., 2002). Also, a majority
of LOB markets have some form of opening or closing auction mechanisms to make
price discovery more efficient (Friederich and Payne, 2007; Huang and Tsai, 2008;
Madhavan and Panchapagesan, 2000; Theissen, 2000). For example, NSE conducts
pre-trade single auctions and FSE conducts pre-trade and post-trade double auctions.
These differences have a significant effect on transparency and efficient price
discovery (Boehmer et al., 2005; Cao et al., 2009; Madhavan et al., 2005; Eom et al.,
2007; Aitken et al., 2001).
Overall, the microstructure designs appear to influence various dimensions of
market efficiency, including liquidity, price discovery and transparency, volatility,
information asymmetry and the overall social welfare by efficient allocation of
resources. In this backdrop, an important yet less explored aspect pertains to the
optimum display of LOB levels. Unlike quote-driven markets, LOBs have the
flexibility to disclose the order book up to a certain level. For example, exchanges
such as NSE and Euronext-Paris disclose the order book information up to five levels,
whereas German exchange (FSE) discloses the order book up to 10 levels. This has
important implications for market transparency and informational efficiency
(Boehmer et al., 2005). In particular, very few studies have examined the impact of
visible and invisible portions of the order book on market efficiency (Amaya et al.,
2018; Brogaard et al., 2019). The issue of optimal order display is of significant
interest to the regulators and policymakers. Therefore, future works may empirically
investigate this issue in more detail.
3.10 Low latency trading and limit order books Systematic
The trading strategies that operate with a time-frame of less than 1 ms are part of low review of
latency trading environment (also called HFT; Hasbrouck and Saar, 2013; O’Hara, 2015).
These trades have a significant impact on market quality and more so, during the periods of
literature
high uncertainty.
There is contrasting evidence on the role of HFT on the efficiency and quality of a
market. At one end is the evidence to suggest that HFTs improve the market quality
indicators such as spreads, the price impact of trades, depth, ST volatility and informational
efficiency (Brogaard et al., 2014, 2019). HFTs use innovative technologies to reduce the
latency (time-horizons) to their advantage. Moreover, the evidence suggests that the
proportion of limit orders is much higher in the HFTs as compared to the non-HFTs. Thus,
these HFTs also act as the new age liquidity providers. They significantly contribute to the
competitive nature of the LOB microstructure and in turn, lower the transaction costs
(Chaboud et al., 2014; Hendershott and Riordan, 2013). The algo-driven HFTs are very
efficient in assessing the information content of trades and hence, are instrumental in the
price discovery. This suggests that HFTs positively contribute to market efficiency and
market quality.
At the other end is the evidence that, with their speed and order processing capacities, the
HFTs discourage and crowd-out the non-HFT liquidity suppliers. The HFTs supply
liquidity when it is costly and scarce and demand liquidity when it is cheap and plentiful
(Hendershott et al., 2011; Menkveld, 2013), thereby lowering market quality. There is
evidence to suggest that, during uncertain times, these HFTs withdraw liquidity and rapidly
exacerbate market liquidity conditions. This may cause sudden disruptions in liquidity
supply and episodic failures in trading activity.
Overall, the evidence on the impact of HFTs on market efficiency and market quality is
mixed and predominantly based on the USA market. The area holds significant potential for
future exploration, particularly in the context of LOB markets that offer rich data for this
kind of analysis.

4. Insights, research gaps and scope for future research

Based on the review of literature, the following potential areas of research emerge:
The rise of LOBs has significantly contributed to the fragmentation of global markets.
For a given market, various LOB trading platforms compete for trading volume and market
share. In this regard, the decision problem of order routing across different venues and its
impact on the order execution quality has generated considerable interest across market
participants, regulators and academic community. Future works may examine this issue in
more detail to offer new insights.
Most of the literature on LOBs uses the price and volume information associated with the
best bid-ask orders [27], to examine the state of the order book (e.g. computation of spread –
a proxy of illiquidity). The best orders in LOBs, unlike in dealer markets, maybe noisy and
informationally less efficient. There is evidence that the informed traders prefer the deeper
levels of LOB to conceal private information, as evident from the low volumes of the best
orders (Amaya et al., 2018; Cao et al., 2009; Cenesizoglu and Grass, 2018; Deuskar and
Johnson, 2011; Dierker et al., 2015; Domowitz et al., 2005; Kempf and Mayston, 2008) [28].
Therefore, the examination of order book at deeper levels may offer interesting insights,
especially into the functioning of the hitherto less explored emerging markets. This entails
the analysis of the price and volume information related to the deeper levels of LOB, i.e.
beyond best orders. The information may be useful in understanding various important
aspects of the order book dynamics related to market efficiency such as price discovery and
QRFM return predictability, volatility, liquidity provision and order choices of informed traders.
Various measures have been proposed in the literature to examine the state of the order
book, especially at the deeper levels. A list of these measures and their formulae are
provided in Appendix 3. A related issue in this context is the desirable level of visible and
invisible portions of the order book and its impact on transparency and market efficiency
(Amaya et al., 2018; Brogaard et al., 2019). The issue of optimal order display is of significant
interest to the regulators and policymakers and research in this area may fulfill this need.
Moreover, given the low entry and exit barriers in the LOB markets, we may expect their
liquidity dynamics to differ considerably from that in quote-driven markets. Future works
may examine the implications of this microstructure aspect on the liquidity dynamics in
LOB markets vis-à-vis quote-driven markets.
The progress in computational resources and electronic communication has reduced the
market reaction time to 2-5 min (Brogaard et al., 2014, 2019; Hendershott and Riordan, 2013;
O’Hara, 2015). Consequently, the short-horizon price movements and order flows have
generated new interest among policymakers, academics and investors. However, emerging
market studies on LOBs are lacking owing to non-availability of good high-frequency data
(Anagnostidis et al., 2016; Arjoon, 2016; Bai and Qin, 2015; Bekaert et al., 2007; Griffin et al.,
2010; Kim and Shamsuddin, 2008; Lesmond, 2005; Rejeb and Boughrara, 2013; Syamala
et al., 2014) [29]. This is an interesting and unexplored area for future research in emerging
markets, as the tick-by-tick data have become available only recently (past 5-10 years). Such
data offers a detailed microscopic view of the order flows and price discovery, over
extremely short time horizons. The analysis of this data may help in re-examining the
traditional notions of market efficiency and market microstructure in the context of
emerging LOB markets (O’Hara, 2015).
For the quote-driven markets, the inventory concerns of designated market makers,
along with the information asymmetry, are central to the traditional microstructure models
(Admati and Pfleiderer, 1988; Amihud and Mendelson, 1980; Ho and Stoll, 1983; Kyle, 1985;
O’Hara and Oldfield, 1986; Stoll, 1978). However, the inventory concerns weaken in pure
LOB markets (e.g. India) due to the absence of designated market makers. The limit order
suppliers provide liquidity in these markets and act as the de-facto market makers. Hence,
these pure LOB markets offer a natural experimental set-up to examine the effect of
information asymmetry on various aspects of market efficiency, including the spread
formation.
Moreover, many emerging markets have come up in the past 10 years that are pure LOB
markets. Due to the non-availability of quality high-frequency bid-ask data in the past, the
dynamics of liquidity provisioning is less explored in these emerging LOB markets (Bekaert
et al., 2007; Lesmond, 2005). This includes important phenomena such as the liquidity
commonality, flight-to-quality and flight-to-liquidity, intraday liquidity dynamics, market
liquidity and funding liquidity, role of liquidity risk in asset pricing and liquidity risk
contagion that are well-documented for the developed markets (Acharya and Pedersen,
2005; Anagnostidis et al., 2016; Beber et al., 2008; Brunnermeier and Pedersen, 2008; Kempf
and Mayston, 2008). The examination of the underlying mechanisms that drive liquidity
provisioning in LOB markets and a comparative empirical investigation across different
market microstructures would develop new insights.
The growing presence of HFTs and low latency trading have significantly changed the
market microstructure dynamics related to market efficiency (Biais et al., 2015; Chaboud et al.,
2014; Hendershott and Riordan, 2013; Menkveld, 2013; O’Hara, 2015; Ros u, 2019). The extant
literature predominantly covers the USA market and offers mixed evidence. There is much scope
for exploration across various important LOB markets (Nawn and Banerjee, 2019a, 2019b). With
the growing attention of global investor community toward algorithmic low latency trading, this Systematic
is a potential area of interest to the academics, practitioners, policymakers and regulators. review of
More recently, the econometric techniques from other areas (e.g. physics, operations
research and information technology) have been adopted to model LOBs (Cont et al., 2010;
literature
Kercheval and Zhang, 2015; Moro et al., 2009; Passalis et al., 2018; Toth et al., 2015; Tran
et al., 2019). For example, the machine learning models (neural networks, support vector
machines, bag-of-features) and zero-intelligence models are widely acknowledged to
improve the forecastability in LOB modeling. These approaches may open new avenues of
research in LOBs.
Trading speed (e.g. order intervals, wait-time, duration, etc.) is an important dimension of
market quality. With the availability of time-stamps in LOB data, showing order intervals
smaller than 1 ms, it has become possible to examine the relation between trading speed,
order aggressiveness, shape of the order book and informed trading with more robustness
and higher degree of reliability (Biais et al., 2015; Foucault et al., 2016; Hoffmann, 2014; Rosu,
2016, 2019). The area has witnessed burgeoning interest among researchers.
A sizable volume of research exists on order submissions related to limit and market
orders. Besides order submissions, order cancellations (fleeting orders), amendments and
hidden orders form an integral part of order strategies. With the growing presence of HFTs,
these actions (cancellation, amendment and hidden orders) significantly contribute to the
liquidity, price discovery and thereby to market efficiency (Baruch and Glosten, 2013; Cao
et al., 2009; Fleming et al., 2018; Hautsch and Huang, 2012; Menkhoff et al., 2010; Ros u, 2009,
2019). The information effects of these order choices are less well-explored and documented.
A related strand of literature has emerged recently, documenting the impact of lit and dark
markets, dark pools of trading and crossing networks [30] on market quality, transparency,
liquidity and price discovery (Bloomfield et al., 2015). The examination of these order
choices has significant potential for future research.
The resilience of order book reflects its ability to recover to normal shape after order flow
shocks (Degryse et al., 2005; Large, 2007; Lo and Hall, 2015). Resilience is an important yet
less explored parameter of market quality. This is because the investigation of LOB
resilience requires the application of high-frequency LOB data. The literature on this area is
sparse and leaves much ground for future works, particularly for emerging markets as these
markets are prone to episodic failures in trading activity and sudden disruptions in liquidity
during crisis events.

5. Summary and conclusions

This article provides a comprehensive account of the literature on LOBs based on 103 high-
quality studies, covering the period 1981-2018. The studies are chosen from a collection of
2,514 articles, using the SLR methodology. The article uses bibliometric mapping and word-
cloud analysis to identify significant contributions and classify these into sub-areas. The
methodology is further tested for its robustness (bibiliometrix “R” package and backward
snowballing technique) to ensure that the article exhaustively covers the major sub-areas
and different financial markets.
The study documents the influence of LOBs on different aspects of market efficiency
such as price discovery, transparency, order aggregation, immediacy, liquidity, volatility,
HFT, low latency trading and dark pool trading. The study also discusses the implications
of LOB-based trading for different kinds of market participants. The shape of LOB is a key
determinant of the trading activity, price formation and order flows. The study explores
various measures that are used to capture the shape of LOB. These include depth, height,
order imbalance, slope, spread, resilience and the elasticity of bid and ask sides. These
QRFM measures and their proxies can assist the practitioners and academics in predicting the
order flows, price formation and formulating profitable trading strategies.
A substantial body of literature on LOBs has developed over the study period. However,
there are certain under-explored areas, which have been identified for future research. LOBs
in emerging markets is one such area. Moreover, the examination of the following issues
may provide new insights into the functioning of diverse financial markets: examination of
LOBs beyond the best bid-ask quotes; short-horizon price movement with order flows;
market fragmentation and its implications for order routing and execution quality; optimal
display of LOB levels and its impact on market efficiency; liquidity dynamics in quote-
driven markets vis-à-vis that of LOB markets; resilience of LOB; and the impact of low
latency trading and HFT on market efficiency.

Notes
1. For a detailed discussion on market efficiency and market quality, interested readers may refer to
the following studies (Anand, Tanggaard and Weaver, 2009; Bennett and Wei, 2006; Chakrabarti
and Roll, 1999; Chordia, Roll and Subrahmanyam, 2008; Fama, 1998).
2. Literature recognizes these trade-offs in optimal trade execution (Bae et al., 2003; Cohen et al.,
1981; Glosten, 1994; Harris, 1998; Hollifield et al., 2004; Parlour, 1998; Seppi, 1997).
3. These protocols include price time priority rules, minimum tick size, trading halts, circuit
breakers, etc.
4. Appendix 1 provides the key statistics pertaining to these exchanges.
5. SZSE, SSE, TSE, FSE, NSE, BSE, TSX, LSE, HKEX, ASX
6. Appendix C provides the formulaic representation and related literature pertaining to the key
LOB measures. For brevity, a detailed discussion on the individual measures is not provided here
and the same is available upon request from the authors.
7. Bibliometrics is an emerging field that uses statistical methods to examine the evolution of a
given stream of literature. It involves assessment of studies, mapping of their relative place
in the literature and overall area dynamics and application of graphical methods to
summarize this information (Koseoglu, 2016; McBurney and Novak, 2002; Zupic and Cater,

2015).
8. A detailed exposition of VOSviewer tool is available with the authors and can be provided upon
request.
9. For robustness checks, we re-conduct the bibliometric analyses using “R” package (Bibliometrix).
The results (not reported here for brevity) yield a similar set of papers and highlight similar
interest areas of research.
10. For ranking purposes, ABDC and CABS rankings/listing were used.
11. The manuscript also refers to studies from the diverse areas that are not strictly related to market
efficiency and hence not part of the main sample-set of 103 studies.
12. This kind of analysis is conducted by using the time-evolution charts from VOSviewer similar to
that shown in Panel B, Figure 2.
13. Investors are heterogeneous in the sense that they have different sets of utility functions, degrees
of risk aversion and consumption horizons.
14. Pre-trade transparency comprises the dissemination of real-time bid/ask quote related
information. Post-trade transparency includes information after trade execution (volume, price,
execution time, identity of the buyer and seller, etc.)
15. Multiple venues post the same best prices, the differences in fee schedules become the Systematic
determining factor for routing orders. review of
16. Two important factors affect the investor welfare. First, the differences in market microstructure literature
rules (e.g. fee schedules) may cause significant variation in the execution quality of orders
(likelihood of order fill, speed of order fill, realized spread and other transaction costs) and market
efficiency in general. This, in turn, may affect the levels of investor welfare. A simple example of
this decision-making is as follows: a broker may first break a large order (Parent) into multiple
small orders (Children) using various algorithms available. These multiple orders are then routed
to different venues by the following optimization scheme: lower the transaction costs and
increase the execution probability of the order. Second, the fiduciary responsibility of brokers to
their trader clients results in the classical principal-agent problem. Order routing decisions are
often delegated to broker firms. These broker firms, in addition to client welfare, also have other
considerations such as preferencing arrangements, payment for order flows and cost
internalization (Boehmer et al., 2007).
17. The lowest ask and highest bid are called the best ask and bid orders.
18. The impatient traders can either be the informed traders who wish to exploit the mispricing or
the traders induced by their liquidity needs.
19. Limit order suppliers are compensated through bid-ask spread in trades against market orders of
noise traders and liquidity motivated traders. In contrast, they lose against information driven
traders (Glosten, 1994).
20. This article refers to T. Chordia, R. Roll and A. Subrahmanyam, together, as “CRS.”
21. These studies are part of the limited literature that offers examination of LOBs beyond the best
prices.
22. These measures require only daily price and volume data (e.g. Amihud measure).
23. The effects of inventory concerns are transient in nature while those of information asymmetry
are enduring and permanent (Easley and O’hara, 1987; Hasbrouck, 1988; Ho and Stoll, 1983;
Huang and Stoll, 1997; O’Hara and Oldfield, 1986).
24. Spread measures include absolute bid-ask spread, proportional spread and proportional quoted
spread.
25. In this discussion, we do not account for the ST volatility caused by the random stochastic price
movements.
26. In addition, there are strategies such as order-splitting, cancellation and amendment (Lee, Liu,
Roll and Subrahmanyam, 2004). In view of the weight of the existing literature, we primarily
focus on the order aggressiveness.
27. The lowest ask and highest bid are the best ask and bid orders.
28. These studies are part of the limited literature that offers examination of LOBs beyond best prices.
29. A brief summary of the literature for emerging markets in this area is available with the authors.
The same can be provided upon request. The literature survey for emerging markets suggests
that the level of regulatory reforms, strength of institutions, economic integration and financial
liberalization are significantly different for these markets from that found in developed markets.
Moreover, these markets are afflicted with the issues like thin trading, which impact the
efficiency of conventional microstructural models. Examination of these markets may offer
interesting insights into the factors affecting market efficiency and liquidity.
30. Crossing networks are part of the alternative trading system that is hidden from exchanges.
Market participants (broker-dealers) carry block trades through these channels to avoid the price
impact of their trades.
QRFM References
Acharya, V.V. and Pedersen, L.H. (2005), “Asset pricing with liquidity risk”, Journal of Financial
Economics, Vol. 77 No. 2, pp. 375-410.
Admati, A.R. and Pfleiderer, P. (1988), “A theory of intraday patterns: volume and price variability”,
Review of Financial Studies, Vol. 1 No. 1, pp. 3-40.
Ahn, H.J., Bae, K.H. and Chan, K. (2001), “Limit orders, depth, and volatility: evidence from the stock
exchange of Hong Kong”, The Journal of Finance, Vol. 56 No. 2, pp. 767-788.
Ahn, H.J., Cai, J., Hamao, Y. and Ho, R.Y. (2002), “The components of the bid – ask spread in a limit-
order market: evidence from the Tokyo stock exchange”, Journal of Empirical Finance, Vol. 9
No. 4, pp. 399-430.
Aitken, M.J., Berkman, H. and Mak, D. (2001), “The use of undisclosed limit orders on the Australian
stock exchange”, Journal of Banking and Finance, Vol. 25 No. 8, pp. 1589-1603.
Al-Emran, M., Mezhuyev, V., Kamaludin, A. and Shaalan, K. (2018), “The impact of knowledge
management processes on information systems: a systematic review”, International Journal of
Information Management, Vol. 43, pp. 173-187.
Amaya, D., Filbien, J.Y., Okou, C. and Roch, A.F. (2018), “Distilling liquidity costs from limit order
books forthcoming in journal of banking and finance”, Journal of Banking and Finance, Vol. 94,
pp. 16-34.
Amihud, Y. (2002), “Illiquidity and stock returns: cross-section and time-series effects”, Journal of
Financial Markets, Vol. 5 No. 1, pp. 31-56.
Amihud, Y. and Mendelson, H. (1980), “Dealership market: market-making with inventory”, Journal of
Financial Economics, Vol. 8 No. 1, pp. 31-53.
Amihud, Y. and Mendelson, H. (1986), “Asset pricing and the bid-ask spread”, Journal of Financial
Economics, Vol. 17 No. 2, pp. 223-249.
Anagnostidis, P., Papachristou, G. and Thomaidis, N.S. (2016), “Liquidity commonality in order-driven
trading: evidence from the Athens stock exchange”, Applied Economics, Vol. 48 No. 22,
pp. 2007-2021.
Anand, A., Chakravarty, S. and Martell, T. (2005), “Empirical evidence on the evolution of liquidity:
choice of market versus limit orders by informed and uninformed traders”, Journal of Financial
Markets, Vol. 8 No. 3, pp. 288-308.
Anand, A., Tanggaard, C. and Weaver, D.G. (2009), “Paying for market quality”, Journal of Financial
and Quantitative Analysis, Vol. 44 No. 6, pp. 1427-1457.
Anand, A. and Venkataraman, K. (2016), “Market conditions, fragility, and the economics of market
making”, Journal of Financial Economics, Vol. 121 No. 2, pp. 327-349.
Anand, A. and Weaver, D.G. (2004), “Can order exposure be mandated?”, Journal of Financial Markets,
Vol. 7 No. 4, pp. 405-426.
Arjoon, V. (2016), “Microstructures, financial reforms and informational efficiency in an emerging
market”, Research in International Business and Finance, Vol. 36, pp. 112-126.
Back, K. and Baruch, S. (2007), “Working orders in limit order markets and floor exchanges”, The
Journal of Finance, Vol. 62 No. 4, pp. 1589-1621.
Bae, K.H., Jang, H. and Park, K.S. (2003), “Traders’ choice between limit and market orders: evidence
from NYSE stocks”, Journal of Financial Markets, Vol. 6 No. 4, pp. 517-538.
Bai, M. and Qin, Y. (2015), “Commonality in liquidity in emerging markets: another supply-side
explanation”, International Review of Economics and Finance, Vol. 39, pp. 90-106.
Baruch, S. (2005), “Who benefits from an open limit-order book?”, The Journal of Business, Vol. 78 No. 4,
pp. 1267-1306.
Baruch, S. and Glosten, L.R. (2013), “Fleeting orders”, Unpublished Working Paper, Columbia Business
School, New York, NY, 11 June, pp. 13-43, available at: http://dx.doi.org/10.2139/ssrn.2278457
Baruch, S., Panayides, M. and Venkataraman, K. (2017), “Informed trading and price discovery before Systematic
corporate events”, Journal of Financial Economics, Vol. 125 No. 3, pp. 561-588.
review of
Battalio, R., Corwin, S.A. and Jennings, R. (2016), “Can brokers have it all? On the relation between
make-take fees and limit order execution quality”, The Journal of Finance, Vol. 71 No. 5, literature
pp. 2193-2238.
Beber, A., Brandt, M.W. and Kavajecz, K.A. (2008), “Flight-to-quality or flight-to-liquidity? Evidence
from the euro-area bond market”, Review of Financial Studies, Vol. 22 No. 3, pp. 925-957.
Bekaert, G., Harvey, C.R. and Lundblad, C. (2007), “Liquidity and expected returns: lessons from
emerging markets”, Review of Financial Studies, Vol. 20 No. 6, pp. 1783-1831.
Bennett, P. and Wei, L. (2006), “Market structure, fragmentation, and market quality”, Journal of
Financial Markets, Vol. 9 No. 1, pp. 49-78.
Bessembinder, H., Panayides, M. and Venkataraman, K. (2009), “Hidden liquidity: an analysis of order
exposure strategies in electronic stock markets”, Journal of Financial Economics, Vol. 94 No. 3,
pp. 361-383.
Biais, B., Foucault, T. and Moinas, S. (2015), “Equilibrium fast trading”, Journal of Financial Economics,
Vol. 116 No. 2, pp. 292-313.
Biais, B., Foucault, T. and Salanié, F. (1998), “Floors, dealer markets and limit order markets”, Journal of
Financial Markets, Vol. 1 Nos 3/4, pp. 253-284.
Biais, B., Glosten, L. and Spatt, C. (2005), “Market microstructure: a survey of microfoundations,
empirical results, and policy implications”, Journal of Financial Markets, Vol. 8 No. 2,
pp. 217-264.
Biais, B., Hillion, P. and Spatt, C. (1995), “An empirical analysis of the limit order book and the order
flow in the Paris bourse”, The Journal of Finance, Vol. 50 No. 5, pp. 1655-1689.
Bloomfield, R., O’hara, M. and Saar, G. (2005), “The ‘make or take’ decision in an electronic market:
evidence on the evolution of liquidity”, Journal of Financial Economics, Vol. 75 No. 1, pp. 165-199.
Bloomfield, R., O’hara, M. and Saar, G. (2015), “Hidden liquidity: some new light on dark trading”, The
Journal of Finance, Vol. 70 No. 5, pp. 2227-2274.
Boehmer, E., Jennings, R. and Wei, L. (2007), “Public disclosure and private decisions: equity market
execution quality and order routing”, Review of Financial Studies, Vol. 20 No. 2, pp. 315-358.
Boehmer, E., Saar, G. and Yu, L. (2005), “Lifting the veil: an analysis of pre-trade transparency at the
NYSE”, The Journal of Finance, Vol. 60 No. 2, pp. 783-815.
Brockman, P. and Chung, D.Y. (1999), “Bid-ask spread components in an order-driven environment”,
Journal of Financial Research, Vol. 22 No. 2, pp. 227-246.
Brockman, P. and Chung, D.Y. (2002), “Commonality in liquidity: evidence from an order-driven market
structure”, Journal of Financial Research, Vol. 25 No. 4, pp. 521-539.
Brogaard, J., Hendershott, T. and Riordan, R. (2014), “High-frequency trading and price discovery”,
Review of Financial Studies, Vol. 27 No. 8, pp. 2267-2306.
Brogaard, J., Hendershott, T. and Riordan, R. (2019), “Price discovery without trading: evidence from
limit orders”, The Journal of Finance, Vol. 74 No. 4, pp. 1621-1658.
Brunnermeier, M.K. and Pedersen, L.H. (2008), “Market liquidity and funding liquidity”, Review of
Financial Studies, Vol. 22 No. 6, pp. 2201-2238.
Busalim, A.H. (2016), “Understanding social commerce: a systematic literature review and directions
for further research”, International Journal of Information Management, Vol. 36 No. 6,
pp. 1075-1088.
Buti, S. and Rindi, B. (2013), “Undisclosed orders and optimal submission strategies in a limit order
market”, Journal of Financial Economics, Vol. 109 No. 3, pp. 797-812.
Cao, C., Hansch, O. and Wang, X. (2008), “Order placement strategies in a pure limit order book
market”, Journal of Financial Research, Vol. 31 No. 2, pp. 113-140.
QRFM Cao, C., Hansch, O. and Wang, X. (2009), “The information content of an open limit-order book”, Journal
of Futures Markets, Vol. 29 No. 1, pp. 16-41.
Cenesizoglu, T. and Grass, G. (2018), “Bid- and ask-side liquidity in the NYSE limit order book”, Journal
of Financial Markets, Vol. 38, pp. 14-38.
Chaboud, A.P., Chiquoine, B., Hjalmarsson, E. and Vega, C. (2014), “Rise of the machines:
algorithmic trading in the foreign exchange market”, The Journal of Finance, Vol. 69
No. 5, pp. 2045-2084.
Chakrabarti, R. and Roll, R. (1999), “Learning from others, reacting, and market quality1”, Journal of
Financial Markets, Vol. 2 No. 2, pp. 153-178.
Chan, Y.C. (2000), “The price impact of trading on the stock exchange of Hong Kong”, Journal of
Financial Markets, Vol. 3 No. 1, pp. 1-16.
Chordia, T., Roll, R. and Subrahmanyam, A. (2000), “Commonality in liquidity”, Journal of Financial
Economics, Vol. 56 No. 1, pp. 3-28.
Chordia, T., Roll, R. and Subrahmanyam, A. (2002), “Order imbalance, liquidity, and market returns”,
Journal of Financial Economics, Vol. 65 No. 1, pp. 111-130.
Chordia, T., Roll, R. and Subrahmanyam, A. (2008), “Liquidity and market efficiency”, Journal of
Financial Economics, Vol. 87 No. 2, pp. 249-268.
Chou, R.K. and Wang, Y.Y. (2009), “Strategic order splitting, order choice, and aggressiveness:
evidence from the Taiwan futures exchange”, Journal of Futures Markets, Vol. 29 No. 12,
pp. 1102-1129.
Chung, K.H., Van Ness, B.F. and Van Ness, R.A. (1999), “Limit orders and the bid–ask spread”, Journal
of Financial Economics, Vol. 53 No. 2, pp. 255-287.
Chung, K.H., Van Ness, B.F. and Van Ness, R.A. (2004), “Specialists, limit-order traders, and the
components of the bid-ask spread”, The Financial Review, Vol. 39 No. 2, pp. 255-270.
Chung, S.L., Hung, C.H. and Yeh, C.Y. (2012), “When does investor sentiment predict stock returns?”,
Journal of Empirical Finance, Vol. 19 No. 2, pp. 217-240.
Cohen, K.J., Maier, S.F., Schwartz, R.A. and Whitcomb, D.K. (1981), “Transaction costs, order placement
strategy, and existence of the bid-ask spread”, Journal of Political Economy, Vol. 89 No. 2,
pp. 287-305.
Colliard, J.-E. and Foucault, T. (2012), “Trading fees and efficiency in limit order markets”, Review of
Financial Studies, Vol. 25 No. 11, pp. 3389-3421.
Cont, R., Stoikov, S. and Talreja, R. (2010), “A stochastic model for order book dynamics”, Operations
Research, Vol. 58 No. 3, pp. 549-563.
Copeland, T.E. and Galai, D. (1983), “Information effects on the bid-ask spread”, The Journal of Finance,
Vol. 38 No. 5, pp. 1457-1469.
Coughenour, J. and Shastri, K. (1999), “Symposium on market microstructure: a review of empirical
research”, The Financial Review, Vol. 34 No. 4, pp. 1-27.
De Jong, F., Nijman, T. and Roell, A. (1996), “Price effects of trading and components of the bid-ask
spread on the Paris bourse”, Journal of Empirical Finance, Vol. 3 No. 2, pp. 193-213.
De Winne, R. and D’hondt, C. (2007), “Hide-and-seek in the market: placing and detecting hidden
orders”, Review of Finance, Vol. 11 No. 4, pp. 663-692.
Degryse, H., Jong, F.D., Ravenswaaij, M.V. and Wuyts, G. (2005), “Aggressive orders and the resiliency
of a limit order market”, Review of Finance, Vol. 9 No. 2, pp. 201-242.
Demsetz, H. (1968), “The cost of transacting”, The Quarterly Journal of Economics, Vol. 82 No. 1,
pp. 33-53.
Deuskar, P. and Johnson, T.C. (2011), “Market liquidity and flow-driven risk”, Review of Financial
Studies, Vol. 24 No. 3, pp. 721-753.
Dierker, M., Kim, J.W., Lee, J. and Morck, R. (2015), “Investors’ interacting demand and supply curves Systematic
for common stocks”, Review of Finance, Vol. 20 No. 4, pp. 1517-1547.
review of
Domowitz, I., Hansch, O. and Wang, X. (2005), Journal of Financial Markets, Vol. 8 No. 4, pp. 351-376.
literature
Duong, H.N., Kalev, P.S. and Krishnamurti, C. (2009), “Order aggressiveness of institutional and
individual investors”, Pacific-Basin Finance Journal, Vol. 17 No. 5, pp. 533-546.
Easley, D. and O’hara, M. (1987), “Price, trade size, and information in securities markets”, Journal of
Financial Economics, Vol. 19 No. 1, pp. 69-90.
Eom, K.S., Ok, J. and Park, J.H. (2007), “Pre-trade transparency and market quality”, Journal of
Financial Markets, Vol. 10 No. 4, pp. 319-341.
Fama, E.F. (1998), “Market efficiency, long-term returns, and behavioral finance1”, Journal of Financial
Economics, Vol. 49 No. 3, pp. 283-306.
Fleming, M.J., Mizrach, B. and Nguyen, G. (2018), “The microstructure of a US treasury ECN: the
BrokerTec platform”, Journal of Financial Markets, Vol. 40, pp. 2-22.
Foucault, T. (1999), “Order flow composition and trading costs in a dynamic limit order market”,
Journal of Financial Markets, Vol. 2 No. 2, pp. 99-134.
Foucault, T., Hombert, J. and Ros u, I. (2016), “News trading and speed”, The Journal of Finance, Vol. 71
No. 1, pp. 335-382.
Foucault, T., Kadan, O. and Kandel, E. (2005), “Limit order book as a market for liquidity”, Review of
Financial Studies, Vol. 18 No. 4, pp. 1171-1217.
Foucault, T., Moinas, S. and Theissen, E. (2007), “Does anonymity matter in electronic limit order
markets?”, Review of Financial Studies, Vol. 20 No. 5, pp. 1707-1747.
Friederich, S. and Payne, R. (2007), “Dealer liquidity in an auction market: evidence from the London
stock exchange”, The Economic Journal, Vol. 117 No. 522, pp. 1168-1191.
George, T.J., Kaul, G. and Nimalendran, M. (1991), “Estimation of the bid–ask spread and its
components: a new approach”, Review of Financial Studies, Vol. 4 No. 4, pp. 623-656.
Glosten, L.R. (1994), “Is the electronic open limit order book inevitable?”, The Journal of Finance, Vol. 49
No. 4, pp. 1127-1161.
Glosten, L.R. and Harris, L.E. (1988), “Estimating the components of the bid/ask spread”, Journal of
Financial Economics, Vol. 21 No. 1, pp. 123-142.
Glosten, L.R. and Milgrom, P.R. (1985), “Bid, ask and transaction prices in a specialist market
with heterogeneously informed traders”, Journal of Financial Economics, Vol. 14 No. 1,
pp. 71-100.
Goettler, R.L., Parlour, C.A. and Rajan, U. (2005), “Equilibrium in a dynamic limit order market”, The
Journal of Finance, Vol. 60 No. 5, pp. 2149-2192.
Goettler, R.L., Parlour, C.A. and Rajan, U. (2009), “Informed traders and limit order markets”, Journal of
Financial Economics, Vol. 93 No. 1, pp. 67-87.
Goldstein, M.A. and Kavajecz, K.A. (2004), “Trading strategies during circuit breakers and extreme
market movements”, Journal of Financial Markets, Vol. 7 No. 3, pp. 301-333.
Grant, M.J. and Booth, A. (2009), “A typology of reviews: an analysis of 14 review types and associated
methodologies”, Health Information and Libraries Journal, Vol. 26 No. 2, pp. 91-108.
Griffin, J.M., Kelly, P.J. and Nardari, F. (2010), “Do market efficiency measures yield correct inferences?
A comparison of developed and emerging markets”, Review of Financial Studies, Vol. 23 No. 8,
pp. 3225-3277.
Griffiths, M.D., Smith, B.F., Turnbull, D.A.S. and White, R.W. (2000), “The costs and determinants of
order aggressiveness”, Journal of Financial Economics, Vol. 56 No. 1, pp. 65-88.
Grossman, S.J. and Stiglitz, J.E. (1980), “On the impossibility of informationally efficient markets”, The
American Economic Review, Vol. 70 No. 3, pp. 393-408.
QRFM Gutiérrez-Nieto, B. and Serrano-Cinca, C. (2019), “20 Years of research in microfinance: an information
management approach”, International Journal of Information Management, Vol. 47, pp. 183-197.
Handa, P., Schwartz, R. and Tiwari, A. (2003), “Quote setting and price formation in an order driven
market”, Journal of Financial Markets, Vol. 6 No. 4, pp. 461-489.
Handa, P. and Schwartz, R.A. (1996), “Limit order trading”, The Journal of Finance, Vol. 51 No. 5,
pp. 1835-1861.
Handa, P., Schwartz, R.A. and Tiwari, A. (1998), “The ecology of an order-driven market”, The Journal
of Portfolio Management, Vol. 24 No. 2, pp. 47-55.
Harris, L. (1990), “Liquidity, trading rules and electronic trading systems”, Papers 91-8, Southern
California - School of Business Administration, available at: https://ideas.repec.org/p/fth/socabu/
91-8.html
Harris, L. (1998), “Optimal dynamic order submission strategies in some stylized trading problems”,
Financial Markets, Institutions and Instruments, Vol. 7 No. 2, pp. 1-76.
Harris, L. and Hasbrouck, J. (1996), “Market vs limit orders: the SuperDOT evidence on order
submission strategy”, The Journal of Financial and Quantitative Analysis, Vol. 31 No. 2,
pp. 213-231.
Harris, L.E. and Panchapagesan, V. (2005), “The information content of the limit order book:
evidence from NYSE specialist trading decisions”, Journal of Financial Markets, Vol. 8 No. 1,
pp. 25-67.
Hasbrouck, J. (1988), “Trades, quotes, inventories, and information”, Journal of Financial Economics,
Vol. 22 No. 2, pp. 229-252.
Hasbrouck, J. (2009), “Trading costs and returns for US equities: estimating effective costs from daily
data”, The Journal of Finance, Vol. 64 No. 3, pp. 1445-1477.
Hasbrouck, J. and Saar, G. (2002), “Limit orders and volatility in a hybrid market: the island ECN”, NYU
Working Paper No. FIN-01-025, New York University, Stern School of Business, New York, NY,
26 July, available at: https://doi.org/10.2139/ssrn.310940
Hasbrouck, J. and Saar, G. (2013), “Low-latency trading”, Journal of Financial Markets, Vol. 16 No. 4,
pp. 646-679.
Hautsch, N. and Huang, R. (2012), “The market impact of a limit order”, Journal of Economic Dynamics
and Control, Vol. 36 No. 4, pp. 501-522.
Hendershott, T., Jones, C.M. and Menkveld, A.J. (2011), “Does algorithmic trading improve liquidity?”,
The Journal of Finance, Vol. 66 No. 1, pp. 1-33.
Hendershott, T. and Riordan, R. (2013), “Algorithmic trading and the market for liquidity”, Journal of
Financial and Quantitative Analysis, Vol. 48 No. 4, pp. 1001-1024.
Henriksen-Bulmer, J. and Jeary, S. (2016), “Re-identification attacks – a systematic literature review”,
International Journal of Information Management, Vol. 36 No. 6, pp. 1184-1192.
Ho, T.S. and Stoll, H.R. (1983), “The dynamics of dealer markets under competition”, The Journal of
Finance, Vol. 38 No. 4, pp. 1053-1074.
Hoffmann, P. (2014), “A dynamic limit order market with fast and slow traders”, Journal of Financial
Economics, Vol. 113 No. 1, pp. 156-169.
Hollifield, B., Miller, R.A. and Sandas, P. (2004), “Empirical analysis of limit order markets”, Review of
Economic Studies, Vol. 71 No. 4, pp. 1027-1063.
Huang, R.D. and Stoll, H.R. (1997), “The components of the bid-ask spread: a general approach”, Review
of Financial Studies, Vol. 10 No. 4, pp. 995-1034.
Huang, Y.C. and Tsai, P.L. (2008), “Effectiveness of closing call auctions: evidence from the Taiwan
stock exchange”, Emerging Markets Finance and Trade, Vol. 44 No. 3, pp. 5-20.
Jain, P.K. (2005), “Financial market design and the equity premium: electronic versus floor trading”,
The Journal of Finance, Vol. 60 No. 6, pp. 2955-2985.
Jovanovic, B. and Menkveld, A.J. (2016), “Middlemen in limit order markets”, available at: http://dx.doi. Systematic
org/10.2139/ssrn.1624329
review of
Kaniel, R. and Liu, H. (2006), “So what orders do informed traders use?”, The Journal of Business,
Vol. 79 No. 4, pp. 1867-1913. literature
Kempf, A. and Mayston, D. (2008), “Liquidity commonality beyond best prices”, Journal of Financial
Research, Vol. 31 No. 1, pp. 25-40.
Kercheval, A.N. and Zhang, Y. (2015), “Modelling high-frequency limit order book dynamics with
support vector machines”, Quantitative Finance, Vol. 15 No. 8, pp. 1315-1329.
Kim, J.H. and Shamsuddin, A. (2008), “Are Asian stock markets efficient? Evidence from new multiple
variance ratio tests”, Journal of Empirical Finance, Vol. 15 No. 3, pp. 518-532.
Kitchenham, B. (2004), “Procedures for performing systematic reviews”, Technical Report [TR/SE-0401],
Keele University, Keele, July 2004, available at: www.inf.ufsc.br/aldo.vw/kitchenham.pdf
Koseoglu, M.A. (2016), “Growth and structure of authorship and co-authorship network in the strategic
management realm: evidence from the strategic management journal”, BRQ Business Research
Quarterly, Vol. 19 No. 3, pp. 153-170.
Kozhan, R. and Salmon, M. (2012), “The information content of a limit order book: the case of an FX
market”, Journal of Financial Markets, Vol. 15 No. 1, pp. 1-28.
Kumar, P. and Seppi, D.J. (1994), “Information and index arbitrage”, The Journal of Business, Vol. 67
No. 4, pp. 481-509.
Kyle, A.S. (1985), “Continuous auctions and insider trading”, Econometrica, Vol. 53 No. 6, pp. 1315-1335.
Large, J. (2007), “Measuring the resiliency of an electronic limit order book”, Journal of Financial
Markets, Vol. 10 No. 1, pp. 1-25.
Lee, Y.T., Liu, Y.J., Roll, R. and Subrahmanyam, A. (2004), “Order imbalances and market efficiency:
evidence from the Taiwan stock exchange”, Journal of Financial and Quantitative Analysis,
Vol. 39 No. 2, pp. 327-341.
Lesmond, D.A. (2005), “Liquidity of emerging markets”, Journal of Financial Economics, Vol. 77 No. 2,
pp. 411-452.
Lin, J.C., Sanger, G.C. and Booth, G.G. (1995), “Trade size and components of the bid-ask spread”,
Review of Financial Studies, Vol. 8 No. 4, pp. 1153-1183.
Lo, D.K. and Hall, A.D. (2015), “Resiliency of the limit order book”, Journal of Economic Dynamics and
Control, Vol. 61, pp. 222-244.
McBurney, M.K. and Novak, P.L. (2002), “What is bibliometrics and why should you care?”,
Proceedings of IEEE International Professional Communication Conference, IEEE, Portland,
OR, USA, 20 September, 2002, pp. 108-114.
Madhavan, A. (2000), “Market microstructure: a survey”, Journal of Financial Markets, Vol. 3 No. 3,
pp. 205-258.
Madhavan, A. and Panchapagesan, V. (2000), “Price discovery in auction markets: a look inside the
black box”, Review of Financial Studies, Vol. 13 No. 3, pp. 627-658.
Madhavan, A., Porter, D. and Weaver, D. (2005), “Should securities markets be transparent?”, Journal of
Financial Markets, Vol. 8 No. 3, pp. 265-287.
Madhavan, A., Richardson, M. and Roomans, M. (1997), “Why do security prices change? A
transaction-level analysis of NYSE stocks”, Review of Financial Studies, Vol. 10 No. 4,
pp. 1035-1064.
Maglaras, C., Moallemi, C.C. and Zheng, H. (2015), “Optimal execution in a limit order book and an
associated microstructure market impact model”, Columbia Business School Research Paper,
No. 15-60, available at: http://dx.doi.org/10.2139/ssrn.2610808.
Menkhoff, L., Osler, C.L. and Schmeling, M. (2010), “Limit-order submission strategies under
asymmetric information”, Journal of Banking and Finance, Vol. 34 No. 11, pp. 2665-2677.
QRFM Menkveld, A.J. (2013), “High frequency trading and the new market makers”, Journal of Financial
Markets, Vol. 16 No. 4, pp. 712-740.
Moinas, S. (2010), “Hidden limit orders and liquidity in order driven markets”, TSE Working Paper, n.
10-147, March 2010, available at: www.tse-fr.eu/publications/hidden-limit-orders-and-liquidity-
order-driven-markets
Moro, E., Vicente, J., Moyano, L.G., Gerig, A., Farmer, J.D., Vaglica, G., Lillo, F. and Mantegna, R.N.
(2009), “Market impact and trading profile of hidden orders in stock markets”, Physical Review E,
Vol. 80 No. 6, pp. 066102(1–8).
Naes, R. and Skjeltorp, J.A. (2006), “Order book characteristics and the volume–volatility relation:
empirical evidence from a limit order market”, Journal of Financial Markets, Vol. 9 No. 4,
pp. 408-432.
Nawn, S. and Banerjee, A. (2019a), “Do the limit orders of proprietary and agency algorithmic traders
discover or obscure security prices?”, Journal of Empirical Finance, Vol. 53, pp. 109-125.
Nawn, S. and Banerjee, A. (2019b), “Do proprietary algorithmic traders withdraw liquidity during
market stress?”, Financial Management, Vol. 48 No. 2, pp. 641-676.
Obizhaeva, A.A. and Wang, J. (2013), “Optimal trading strategy and supply/demand dynamics”,
Journal of Financial Markets, Vol. 16 No. 1, pp. 1-32.
O’hara, M. (1995), Market Microstructure Theory, Vol. 108, Blackwell Publishers, Cambridge, MA.
O’Hara, M. (2015), “High frequency market microstructure”, Journal of Financial Economics, Vol. 116
No. 2, pp. 257-270.
O’Hara, M. and Oldfield, G.S. (1986), “The microeconomics of market making”, The Journal of Financial
and Quantitative Analysis, Vol. 21 No. 4, pp. 361-376.
Pardo, A. and Pascual, R. (2012), “On the hidden side of liquidity”, The European Journal of Finance,
Vol. 18 No. 10, pp. 949-967.
Parlour, C.A. (1998), “Price dynamics in limit order markets”, Review of Financial Studies, Vol. 11 No. 4,
pp. 789-816.
Parlour, C.A. and Seppi, D.J. (2003), “Liquidity-based competition for order flow”, Review of Financial
Studies, Vol. 16 No. 2, pp. 301-343.
Parlour, C.A. and Seppi, D.J. (2008), “Limit order markets: a survey”, Handbook of Financial
Intermediation and Banking, Vol. 5, pp. 63-95.
Pascual, R. and Veredas, D. (2009), “Does the open limit order book matter in explaining informational
volatility?”, Journal of Financial Econometrics, Vol. 8 No. 1, pp. 57-87.
Passalis, N. Tefas, A. Kanniainen, J. Gabbouj, M. and Iosifidis, A. (2018), “Temporal bag-of-features
learning for predicting mid price movements using high frequency limit order book data”, IEEE
Transactions on Emerging Topics in Computational Intelligence, available at: https://doi.org/
10.1109/TETCI.2018.2872598.
Pham, M.C., Anderson, H.M., Duong, H.N. and Lajbcygier, P. (2018), “Dynamics of the limit order book
and the volume-volatility relation”, FMA Conference, San Diego, available at: http://
Fmaconferences.Org/SanDiego/Papers/Pham-et-al.(2018)-Volume-Volatility-Relation.Pdf
Ranaldo, A. (2004), “Order aggressiveness in limit order book markets”, Journal of Financial Markets,
Vol. 7 No. 1, pp. 53-74.
Rejeb, A.B. and Boughrara, A. (2013), “Financial liberalization and stock markets efficiency: new
evidence from emerging economies”, Emerging Markets Review, Vol. 17, pp. 186-208.
u, I. (2009), “A dynamic model of the limit order book”, Review of Financial Studies, Vol. 22 No. 11,
Ros
pp. 4601-4641.
Rosu, I. (2012), “Order choice and information in limit order markets”, Chapter 2, in Abergel, F.,
Bouchaud, J., Foucault, T., Lehalle, C.A. and Rosenbaum, M. (Eds), John Wiley and Sons, West
Sussex, available at: http://webhost.hec.fr/rosu/research/oc_lom.pdf
Rosu, I. (2016), “Liquidity and information in order driven markets”, Chicago Booth School of Business, Systematic
Chicago (Forthcoming in Journal of Financial and Quantitative Analysis), available at: http://dx.
doi.org/10.2139/ssrn.1286193 review of
u, I. (2019), “Fast and slow informed trading”, Journal of Financial Markets, Vol. 43, pp. 1-30.
Ros literature
Saar, G. (2001), “Price impact asymmetry of block trades: an institutional trading explanation”, Review
of Financial Studies, Vol. 14 No. 4, pp. 1153-1181.
Sensoy, A. (2019), “Commonality in ask-side vs. bid-side liquidity”, Finance Research Letters, Vol. 28,
pp. 198-207.
Seppi, D.J. (1997), “Liquidity provision with limit orders and a strategic specialist”, Review of Financial
Studies, Vol. 10 No. 1, pp. 103-150.
Stoll, H.R. (1978), “The supply of dealer services in securities markets”, The Journal of Finance, Vol. 33
No. 4, pp. 1133-1151.
Stoll, H.R. (1989), “Inferring the components of the bid-ask spread: theory and empirical tests”, The
Journal of Finance, Vol. 44 No. 1, pp. 115-134.
Stoll, H.R. (2000), “Presidential address: friction”, The Journal of Finance, Vol. 55 No. 4, pp. 1479-1514.
Syamala, S.R., Reddy, V.N. and Goyal, A. (2014), “Commonality in liquidity: an empirical examination
of emerging order-driven equity and derivatives market”, Journal of International Financial
Markets, Institutions and Money, Vol. 33, pp. 317-334.
Theissen, E. (2000), “Market structure, informational efficiency and liquidity: an experimental
comparison of auction and dealer markets”, Journal of Financial Markets, Vol. 3 No. 4,
pp. 333-363.
Toth, B., Palit, I., Lillo, F. and Farmer, J.D. (2015), “Why is equity order flow so persistent?”, Journal of
Economic Dynamics and Control, Vol. 51, pp. 218-239.
Tran, D.T., Iosifidis, A., Kanniainen, J. and Gabbouj, M. (2019), “Temporal attention-augmented bilinear
network for financial time-series data analysis”, IEEE Transactions on Neural Networks and
Learning Systems, Vol. 30 No. 5, pp. 1407-1418.
Tranfield, D., Denyer, D. and Smart, P. (2003), “Towards a methodology for developing evidence-
informed management knowledge by means of systematic review”, British Journal of
Management, Vol. 14 No. 3, pp. 207-222.
Tripathi, A., Dixit, A. and Vipul (2020), “Liquidity commonality in extreme quantiles: Indian evidence”,
Finance Research Letters, p. 101448. doi: 10.1016/j.frl.2020.101448.
Valenzuela, M., Zer, I., Fryzlewicz, P. and Rheinländer, T. (2015), “Relative liquidity and future
volatility”, Journal of Financial Markets, Vol. 24, pp. 25-48.
Van Eck, N. and Waltman, L. (2009), “Software survey: VOSviewer, a computer program for
bibliometric mapping”, Scientometrics, Vol. 84 No. 2, pp. 523-538.


Zupic, I. and Cater, T. (2015), “Bibliometric methods in management and organization”, Organizational
Research Methods, Vol. 18 No. 3, pp. 429-472.
QRFM Appendix 1

Details of major stock exchanges

S. No. of Market
no. Country Exchange Established ECN Symbol listings cap. Turnover

1 US NYSE 1792 2000 NYSE 2,258 20.68 19.34

2 US NASDAQ stock 1971 1971 NASDAQ 3,058 9.76 16.79
exchange
3 Japan TSE 1878 1982 JPX 3,657 5.30 6.30
4 China SSE 1990 1990 SSE 1,450 3.92 6.12
5 Hong Kong HKEX 1891 1986 HKEX 2,315 3.82 2.34
6 UK LSE 1773 1997 LSE 2,479 3.64 2.55
7 France Euronext 1826 1986 Euronext 1,208 3.73 2.20
Amsterdam
stock exchange
8 China SZSE 1990 1990 SZSE 2,134 2.41 7.56
9 Canada TSX 1878 1977 TSX 1,489 2.65 1.88
10 Germany FSE 1585 1991 FSX 514 1.76 1.82
11 India BSE 1875 1995 BSE 5,066 2.08 0.12
12 India NSE 1992 1994 NSE 1,923 2.03 1.17
13 Australia Australian 1859 1987 ASX 2,146 1.26 0.46
securities
exchange
14 Sweden Stockholm stock 1863 1989 OMX 1,863 1.52 0.55
exchange
15 South Korea South Korea 1956 1988 KRX 2,207 1.41 2.52
stock exchange
16 Switzerland Swiss stock 1938 1996 SIX 270 1.44 0.97
exchange

Notes: The table shows the year of establishment; the year of start of electronic trading (ECN); symbol of
the exchange; number of companies listed during the month of December, 2018; average market
Table A1. capitalization of the exchange for the month of December, 2018 (USD trillion); and the total annual trade
volume of the exchange (USD trillion) for the year 2018. These exchanges are part of the trillion-dollar
Key statistics of 16 (market capitalization) club. The aggregate capitalization of these exchanges comprise nearly 90% of the
major exchanges total market capitalization of all exchanges across the world. For details, refer- https://www.world-
across the world exchanges.org/our-work/statistics. All of these exchanges use LOBs
Appendix 2 Systematic
review of
Classification of articles
literature

Figure A1.
Annual distribution
of the research
articles on LOBs for
the period 1981-2018
QRFM

Figure A2.
Categorization of
articles across the
Journal frequency,
number of citations
and Journal rankings

Figure A3.
Classification of the
research articles on
LOBs based on its
major sub-areas
 Systematic
review of
literature

Figure A4.
Classification of the
research articles
based on the
geographical
distribution and
sampling frequency
of the data
QRFM

Figure A5.
Summary of the
order-book measures
and methodologies
used by the research
articles
 Systematic
review of
literature

Figure A6.
Word-cloud analysis
of the one hundred
three selected
research articles on
LOB for the most
frequent terms
QRFM Appendix 3

Limit order book measures

Order book measure Formulae Research article

Mid-quote (Pmid) (Pask þ Pbid)/2 Chordia et al. (2002,

2008)
Spread Pask  Pbid Amihud (2002)
Depth (Qask þ Qbid)/2 Ahn et al. (2001),
Ahn et al. (2002),
Amihud and
Mendelson (1986)
Dollar depth (Pask * Qask þ Pbid * Qbid)/2 Brockman and
Chung (1999, 2002)
Bid/ask VWAP P
N P
N
PnVWAP ¼ QL *PL = QL ) Cao et al. (2008,
L¼1 L¼1 2009), Stoll (2000)
VWAP-based Spread VWAP
Pask; n  VWAP
Pbid; n
PN
Order book volume Qbuy;L  QSell;L Friederich and
imbalance L¼1 Payne (2007),
PN
Harris and
Order book dollar Qbuy;L Pbuy;L  QSell;L PSell;L
imbalance L¼1 Panchapagesan
  (2005), Pardo and
PN Qbuy;L Pbuy;L  QSell;L PSell;L
Normalized order   Pascual (2012),
book imbalance L ¼ 1 Qbuy;L Pbuy;L þ QSell;L PSell;L Pascual and
Veredas (2009)
PAsk;N  Pmid PBid;N  Pmid
Ask- and bid-side  ;   Biais et al. (1995),
elasticities QAsk;N  QAsk;middepth QBid;N  QBid;middepth Naes and Skjeltorp
(2006), Ranaldo
Slope a) RAW measure (2004)
0 1
QAsk;N  QAsk;middepth 1
B  QBid;N  QBid;middepth C
*
@ A 2
PAsk;N  Pmid þ  
PAsk; N  Pmid
b) NOSH measure and c) OV measure
Relative spread/ratio (Pask  Pbid)/ Pmid Chordia et al. (2000,
spread/proportional 2002, 2008)
quoted spread
Proportional effective 2*jPtrade  Pmid j=Pmid
spread
Order aggressiveness Ratio measure: market orders/limit orders Degryse et al.
Ordinal scales to measure aggressiveness: (2005), Griffiths
Extremely aggressive (large market order) to et al. (2000),
Extremely passive (order cancellation) Ranaldo (2004)
1 X T
jrt j
Amihud measure * Amihud (2002),
T t¼1 Vt
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Hasbrouck (2009),
Table A2. Roll measure Roll = 2* CovðDPt ; DPt1 Þ, where CovðDPt ; DPt1 Þ < 0 Syamala et al.
Formulae of LOB Else, roll = 0 (2014)
measures (continued)
 Systematic
Order book measure Formulae Research article
review of
Order exposure Hidden order volume/total order volume Bessembinder et al. literature
(2009), Bloomfield
et al. (2015),
Brogaard et al.
(2017)
Order cancellation Order cancellation frequency, order cancellation volume, Large (2007), Lo
proportion of orders cancelled; and Hall (2015)
Sequence of order submission and cancellation
Wait time and Time interval between two limit orders; time interval between Biais et al. (2015),
duration submission and execution of a limit order; limit order Bloomfield et al.
submission frequency (2005)
Trader type HFT vs Non-HFT traders, institutional vs Individual traders, Bloomfield et al.
specialists vs limit orders (2015), Cao et al.
(2008), Hendershott
et al. (2011)

Notes: (1) L = 1, 2, . . ., N, where N is the selected level of LOB. Pask,L, Qask,L, Pbid,L, Qbid,L represent the price
and volume information of the ask and bid prices, for level “L” of the LOB. (2) VWAP: volume weighted
average price. (3) A detailed discussion on these measures is available with the authors (to be provided
upon request) and not produced here for brevity Table A2.

Corresponding author
Abhinava Tripathi can be contacted at: [email protected]

For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: [email protected]