Description
The continuous market has been widely studied by academic researchers. Much of the microstructure literature has focused on the dealer market. Early analyses include Garman (1976), HO and Stoll(1980), Amihud and Mendelsohn (1980), and Mildenstein and Schleef (1983).
Volume 4, Number 4
Equity Trading Practices and
Market Structure: Assessing Asset
Managers' Demand for Immediacy
NICHOLAS ECONOMIDES
AND
ROBERT A. SCHMlARTZ
NEW YORK UNIVERSITY SALOMON CENTER
Director INW WALTER
Leading members of the financial community, both knowledgeable and supportive
of the activities and publications of the New York University Salomon Center,
work with us to further the interaction of financial executives and research faculty,
and to ensure stability of funding for research at the Center.
Bankers Trust Company
Brinson Partners, Inc.
Canadian Imperial Bank of Commerce
Capital Markets Assurance Corporation
Chemical Bank
Chicago Board Options Exchange. Inc.
Citicorp
Continental Asset Management
cornerstone Research, Inc.
Deutsche Bank North America
Duff & Phelps, Inc.
The First National Bank of Chicago
Fitch Investors Service. Inc.
General Motors Corporation
Global Advanced Technology Corporation
Goldman, Sachs & Co.
The Guardian Life Insurance Company
Heine Securities Corporation
Henry Kaufman & Co., Inc.
.ING Bank, N.V.
Investment Technology Group
Investors Research Corporation
KPMG Peat Marwick
Loan Pricing Corporation
Merck & Co., Inc.
Memll Lynch & Co.. Inc.
Metropolitan Life Insurance Company
Milliman & Robertson, Inc.
Moody's Investors Service
Morgan Guaranty Trust Company of
New York
Morgan Stanley & Co., Inc.
Municipal Bond Investors Assurance
Corporation
National Association of Securities Dealers,
Inc.
National Economic Research Association
New York Stock Exchange, Inc.
Nomura Research Institute (America) Inc.
Pfizer, Inc.
Price Waterhouse
Prudential Foundation
Reuters America Inc.
Salomon Brothers Inc
Sanford C. Bernstein & Co.
Standard & Poor's Corporation
TIAA-CREF
Ti g h a s t
LEONARD N. STERN SCHOOL OF BUSINESS
I Dean: George G. Daly
NEW YORK UNIVER!XC'Y
Resident: L. Jay O l h
FINANCIAL MARKETS, INSTITUTIONS & INSTRUMENTS (ISSN 0963-8008)
is published five times a year in February, May, August, November, and December by
Blackwell Publishers with offices at 238 Main Street, Cambridge, MA 02142, USA, and
108 Cowley Road, Oxford OX4 IJF, UK.
INFORMATION FOR SUBSCRIBERS New orders, renewals, sample copy requests.
claims, change-of-address information and all other correspondence should be sent to the
Subscriber Services Coordinator at the publisher's Cambridge office.
SUBSCRIPTION RATES FOR VOLUME 4,1995*
N America Rest of the World
Institutions $115.50 $135.00
Individuals $61.00 $73.00
Single issue rates:
Institutions $26.00 $31.00
Individuals $15.00 $18.00
'Checks should be made payable to Blackwell Publishers; Canadian residents please add
7% GST.
BACK ISSUES Back and single issues of the journal from Volume 1 onwards are avail-
able from the publisher's Cambridge office at the current single-issue rate. For back issues
of The Monograph Series in Finance and Economics, please see information at the back of
the journal.
MICROFORM The journal is available on microfilm. For microfilm service address
inquiries directly to University MicrofiIms Inc., 300 North Zeeb Road, Ann Arbor, MI
48106-1346, USA.
MAILING The journal is mailed via secondclass in N. America and by Virgin Mailing
and Distribution to the rest of the world.
POSTMASTER Second-class postage paid at Boston, MA, and additional offices. Send
'all address corrections to Journals Subscription Department. 238 Main Street, Cambridge,
MA 02142.
ADVERTISING For details please contact the Journals Marketing Manager at the pub-
lisher's Cambridge office.
COPYRIGET All rights reserved. Reproduction or translation of any part of this work
beyond that permitted by Section 107 and 108 of the US Copyright Law without permission
of the publishers is unlawful. Authorization to photocopy items for internal or personal
use, or the internal or personal use of specific clients, is granted by New York University
Salomon Center, provided that the base fee of US $8.00 per copy, plus $15 per page is
paid directly to Copyright Clearance Center (CCC), 222 Rosewood Drive, Danvers, MA
01923, USA. Rates for educational photocopying for classroom use are S.15 per page per
copy also payable directly to CCC or the National Association of College Stores. For those
organizations that have been granted a photocopy license by CCC, a separate system of
payment has been arranged. The fee code for users of the Transactional Reporting Service
is: 0963-8008194 $8.00 + $.15.
For all other permission requests or inquiries, please contact the Journals Permissions
Manager at the publisher.
'Qpeset by Archetype, Monticello, IL. Printed and bound by Edwards Brothers, Ann Arbor,
MI.
'@ 1995 New York University Salomon Center
Equity Trading Practices and Market Structure:
Assessing Asset Managers' Demand for Immediacy
CONTENTS
II. MARKET STRUCTURE AND THE DEMAND FOR IMMEDI-
. . . . . . . . . . . . ACY A REVIEW OF THE ISSUES
IV. TRADING PRACTICES, ORDER SIZE, AND TRANSACTION
COSTS . . . . . . . . . . . . . . . . . . . . . . . . .
V. THE USE OF A L T E R N ~ ELECTRONIC WING SYS-
TEMS . . . . . . . . . . . . . . . . . . . . . . . . .
VI. SAMPLE OF RESPONDENTS AND THEIR REASONS FOR
TRADING . . . . . . . . . . . . . . . . . . . . . . .
IX. REFERENCES . . . . . . . . . . . . . . . . . . . . .
. XI. N m ON CONTRIBUTORS/ACKNOWLEDGMENTS . . .
Equity Trading Practices and Market Structure:
Assessing Asset Managers' Demand for Immediacy
BY NICHOLAS ECONOMIDES AND ROBERT A. SCHWARTZ
This paper summarizes the responses to a questionnaire sent to equity traders through
TraderForum of the Institutional Investor. The respondents manage in total avery significant
percentage of equity assets under management-in the ~nitedStates. The focus of the
questions was the extent of the demand for immediate execution of orders. We found that
the majority of traders an willing to trade patiently if this reduces execution costs. Many
traders indicate that they frequently delay mdes to obtain better prices. Most respondents
indicate that they are typically given more than a day to implement a large order. that they
typically break up more than 20% of their large orders for execution over time. and that
they regularly take more than a day for a large order that has been broken into lots to be
executed completely. There is a generally positive view of alternative electronic trading
systems, such as Instinet and Investment Technology Group's POSIT. The key motives for
trading on these systems arc reduced market impact, lower spreads. better liquidity, and
anonymity. The respondents indicate that the key changes that would make alternative
elecmnic systems more attractive are an increase in execution rates and more convenient
times of trading. The responses to the survey also show that alternative electronic systems
would be wd more if the traders did not have soft dollar arrangements.
I. INTRODUCTION
Practitioners and students of the securities markets widely assume that traders
demand immediate execution of their orders. Indeed, a major function of tradi-
tional brokerldealer firms is to provide the services that result in trades being made
huickly. In volatile markets, an advantage of trading quickly is that opportunity
costs (i.e., the risk of an asset's price "getting away" before a portfolio decision
is implemented) are reduced. However, higher direct costs (i.e., market impact,
bid-ask spreads, commissions, and other transaction costs) are generally incurred
when fast executions are obtained. Little information exists about the relative
importance professional asset managers attach to these two types of costs, and
about the tradeoffs they are willing to make between them. The current survey is
motivated by this lack of information.
The results show that experienced participants often do not trade with maximum
possible speed so as to "nail down" a price, and that they do commonly work their
orders patiently over time. However, the very dynamics of the continuous market
appear to induce a demand to trade quickly. Based on the survey responses, we
conclude that traders would be even more willing to forgo immediacy of execution
if, by so doing, their direct costs of transacting could be further reduced.
Consequently, the survey findings have a major implication for market struc-
ture. Asset managers should be given the opportunity to delay their orders until
predetermined points in time at which they may trade with each other at reduced
Finandal Markets, Institutions & Insmunenu. V. 4. N. 4. November 1995. @ 1995 New Yo& University Salomon
Center. Published by Blackwell Publishers. 238 Main Saeet. Cambridge, MA 02142, USA and 108 Cowley R o d
Oxford. OX4 IJF. UK
2 Nicholas Economides and Robert A. Schwartz
trading cost. The incorporation of an electronic call market would provide the
requisite environment. A call market is an environment that enables buyers and
sellers to meet at pre-determined points in time. We have elsewhere considered
the desirability of holding an electronic call three times a day, along with contin-
uous trading.' The call environment would provide a useful pricing device for
the broad market, while resulting in lower transaction costs (bid-ask spread and
market impact) for individual participants.
In recent years, increasing numbers of institutional investors are breaking out
of traditional molds to explore various proprietary trading systems (PTS). With
the exception of Instinet's continuous market, the PTSs are crossing networks
(e.g., Instinet's after hours cross and Investment Technology Group Inc.'s POSIT
system) and call markets that are capable of independent price discovery (e.g.,
the Arizona Stock Exchange's AZX system). Nevertheless, immediacy continues
to be a major service provided by market centers such as the New York Stock
Exchange (NYSE) and Nasdaq. These market centers operate on the assumption
that participants want instant access to the market, and that they are willing to pay
the price for trading with immediacy.
However, little empirical evidence exists on asset managers' demand for im-
mediacy. To assess this demand, 825 questionnaires were mailed to traders of
managed equity funds, and 150 responses were received. These respondents r ep
resent approximately $1.5 trillion in equity assets under management. In broad
sweep, the responses to the survey suggest that buy-side participants do trade
patiently in an attempt to control execution costs. The key results include:
Two-thirds of the respondents indicated that they are willing to trade pa-
tiently to reduce execution costs (Table 2).
Nearly half say they frequently do delay trades in an effort to obtain better
prices (Table 3).
One-third would regularly or frequently accept a trading delay of one hour
for a $50 stock if they could save 256 per share in trading costs (Table 4).
Nearly a quarter would regularly or frequently accept a trading delay of
three hours for a $50 stock if they could save 259 per share in trading costs
(Table 5).
About one in five would regularly or frequently accept a trading delay of
one hour for a $50 stock if they could gain anonymity on a trade of 10,000
shares or more (Table 6).
Nearly two-thirds regularly or frequently use limit orders (Table 8).
One-third report that 20% or more of their orders for a stock are larger than
the stock's average daily trading volume (Table 15).
More than two-thirds typically give more than one day to implement a large
order for a small cap stock (Table 16).
'See. Economides and Schwartz (1995).
Equio Trading Practices and Marker Structure 3
More than half typically give more than one day to implement a large order
for a large cap stock (Table 16).
Approximately three out of five break up at least 20% of their orders for
100,000 shares or more for execution over a series of trades (Table 17).
Close to half report that they regularly or frequently take more than one
day to completely execute a large order broken into lots (Table 18).
The picture that emerges is that immediacy is not commonly demanded by buy-
side participants, and that executions for large orders are generally not realized
within brief periods of time (a few hours or less). Respondents appear to be less
concerned about trading quickly than about controlling execution costs, the loss
of anonymity, and the information leakage that occurs when an intermediary is
contacted. Understanding this is essential for making proper decisions with regard
to the structure and regulation of our security markets.
We do not claim that immediacy is never demanded. For specific institutions and
specific situations, the advantages of rapid trading may indeed outweigh the costs
involved. Our objective, however, is not to assess the intensity with which most
asset managers, or even the representative asset manager, demands immediacy.
Rather, we wish to determine whether or not a meaningful number do handle a
substantial proportion of their orders patiently because immediacy is costly.
The paper is organized as follows. Section I1 discusses the relationship between
market structure and the demand for immediacy. In this section we also present
our reasons for believing that immediacy is not universally demanded. Section ID
-discusses the respondents willingness to accept trading delays. In section N, we
present findings regarding trading practices, order size, and transaction costs. In
section V, we discuss attitudes towards the use of alternative electronic trading
systems. Section VI discusses the sample of respondents and their reasons for
trading. Section W discusses the differences between active and passive traders
in their responses to the questionnaire. Section VIII contains our concluding
remarks. The Appendix presents the distributed questionnaire.
II. MARKET STRUCTURE AND THE DEMAND FOR IMMEDIACY:
A REVIEW OF THE ISSUES
An understanding of participants demand for immediacy is key to designing the
trading structure of a securities market. In this sect i o~ we review alternative market
structures, consider the relationship between market structure and the demand for
immediacy, and briefly review the literature. A major choice in trading design is
between a continuous market and a call market A continuous market is open for
an extended span of time; e.g., at the New York Stock Exchange trading begins at
9:30 and continues until the 4 p.m. close. During this period, trades are made any
-time two contra-side orders cross in price. The continuous market can be a dealer
market (quote-driven) or an agencylauction market (order-driven). Nasdaq in the
U.S. and SEAQ in the U.K. are dealer markets. Examples of the agencylauction
4 Nicholas Economides and Robert A. Schwartz
market include the New York Stock Exchange, the Paris Bourse, the Toronto Stock
Exchange, and the Tokyo Stock Exchange?
The continuous market has been widely studied by academic researchers. Much
of the microstructure literature has focused on the dealer market. Early analyses
include Garman (1976), HO and Stoll(1980), Amihud and Mendelsohn (1980),
and Mildenstein and Schleef (1983): An analysis of the agencylauction market is
provided in Cohen, Maier, Schwartz, and Whitcomb (1986) and Schwartz (1$1).~
In contrast with a continuous market, orders are batched in a call market for
simultaneous, multilateral execution at a single price, the value that maximizes
the number of shares that trade at the call. Examples of call markets include the
opening procedure on most electronic exchanges (e.g., Toronto's CATS, Paris's
CAC, and Tokyo's CORES), and on non-electronic exchanges such as the NYSE.
Pure electronic call markets include the Tel Aviv Stock Exchange, the Bolsa Mexi-
cans's Intermediate Market, the Arizona Stock Exchange, and the Paris Borse (for
\e;ss Qpu;\d issues), Preuious\y, n&ecton'~c cd\s ex'lsted in%\ Aviu, Pasis, and
roughly 100 years ago at the NYSE. Call markets have received significantly less
'attention than continuous markets in the academic literature. Studies of the call
market include Cohen and Schwartz (1989), Economides and Schwartz (1995),
Schwartz (1996), and Amihud and Mendelson (1985).
In comparison with continuous trading, the call market has distinct advantages
as a trading environment. These include enhanced price discovery, el i nat i on of
the bid-ask spread, reduced market impact of large orders, superior handling of
limit orders and, in general, easier order handling and better market surveillance?
An often-noted disadvantage of call market trading is that it does not provide
immediate access to the market over an extended period of time.6 However, this
i s not a problem if call market trading is integrated with continuous trading?
Nevertheless, if multiple calls are held during a trading day, one might question
whether or not a sufficient number of participants will postpone their orders so that
the intraday calls may be viable. Traders will postpone orders or not depending
on their demand for immediacy and on the price of immediacy in a continuous
market.
Clearly the selI side of the market has a vested interest in supplying immediacy.
It is difficult in a continuous market for ultimate buyers and sellers to find each
2~ontinuous markets may also include a dealer, such as the specialist on the New York Stock
Exchange.
'For further discussion and references, see Schwartz (1991).
'Furthu refmncu are provided in both of these books.
%r funher discussion of call markets, see Economides and Schwarrz (1995). The role of liquidity
in financial exchange is discussed in Economides (1993). (1995). For an analysis of positive size
effects of financial and other networks see Economides (1996).
6Econornides and Siow (1988) and Garbade and Siber (1976). (1979) analyze the cost of longer
waiting until execution and balance it with the benefit of participating in a more liquid market.
. 'For further discussion of the integration of call and continuous trading see Handa and Schwartz
(1996). Economides and Heisler (1995).
Equity Trading Practices and Market Structure
5
other quickly without the services of brokerldealers. However, when a meeting
point in time is pre-specified (i.e., the time of a call), buyers and sellers can more
easily find each other without the services of intermediaries. The key question is,
"are buy-side traders willing to wait?"
We anticipate that an appreciable number of them will answer "yes". Cer-
tainly, the pace with which trading progresses in a continuous market is not in
harmony with the pace with which the underlying investment decisions are com-
monly made.8 Institutional decision making with respect to fundamental infor-
mation takes time. Investment decisions commonIy involve information gathering
and analysis, and the entire process can take place over a period of several days.
But once a decision has been made, an order is typically brought to a continuous
trading environment that accentuates the importance of minutes and even seconds.
T i e is suddenly of the essence. Is it likely that the value of a decision made
over a period of a day or more can decay within the span of an hour or less?
Or, is the demand for immediacy generated endogenously by the dynamics of the
continuous market? Certainly part of the demand for immediacy comes from the
price dynamics of the continuous market Rapid trading may be motivated by
knowledge of the order flow and by charting signals.
Regarding the fundamental determinants of share value, we distinguish two
types of information release: natural (e.g., an earthquake or fire) and managed
(e-g., an unemployment or earnings report). The introduction of a call market
would enable the pace at which managed information is released and portfolio
decisions are made to be better harmonized with the pace at which trading is
pursued. That is, both news releases and institutional investor portfolio decisions
could be timed with reference to the schedule of the calls.
The conventional wisdom is that immediacy is provided by acontinuous market.
On the contrary, the continuous market may actually make it more difficult for
institutional investors to execute large orders at reasonable cost by the end of a
-trading day. Data collected by the Plexus Group indicate that roughly 67% of the
orders given to buy-side trading desks are for more than half of the stocks' average
daily trading volumes, and 40% of the orders are for more than the total average
daily trading volume for the stocks? Orders of this size cannot be traded quickly
in the continuous market at acceptable levels of cost. The reality is that immediacy
is not always obtained in a continuous market.
It is also conventionally believed that intermediaries provide buy-side partic-
ipants with anonymity vis-a-vis each other. And they do. However, buy-side
participants are increasingly concerned about the loss of anonymity to sell-side
t~rokerldealers. It is also becoming apparent that anonymity can be provided by
an electronic trading system, and it certainly is characteristic of call market trad-
8 ~ h e pace of trading in a continuous market has accelerated with the application of information
technology.
9 ~e e Wagner and Edwards (1993).
6 Nicholas Economides and Robert A. Schwartz
ing. We expect that buy-side participants will be willing to forsake immediacy for
disintermediation and anonymity.
All things considered, picture an institutional investor who makes a portfolio
decision at 2 p.m. when a market call is scheduled for the 4 p.m. close. The
investor could avoid paying the price of intermediation and immediacy by waiting
two hours and trading at lower cost at the 4 p.m. close. By waiting, he or she
has effectively unbundled the act of "trading" from the "immediacy" of the trade.
If enough participants do this, they will naturally meet without the assistance of
intermediaries, and the intra-day calls will be viable.
The archetypal role of a dealer is to provide the liquidity that enables investors to
trade with immediacy. "Immediacy," however, is a vague concept. For retail-sized
orders, it could mean the ability to trade within a few minutes. Large institutional
orders, however, would incur unacceptably large execution costs (bid-ask spread
plus market impact) if executed so quickly. An asset manager seeking to buy
100,000 shares of a stock that on average trades 200,000 shares a day, might
consider an execution obtained within an hour or even a day to be immediate. This
section of the paper contains our findings with regard to various issues concerning
the patience with which a respondent is willing to seek a trade.
The first issue we address concerns the meaning of immediacy itself: how
quickly must a trade be made to be considered immediate? Respondents were also
asked what they would be willing to pay for immediacy, and how frequently they
do in fact delay a trade in an attempt to obtain a better price. Respondents were also
asked about their willingness to accept a trading delay to reduce their trading costs
or to gain anonymity. The extent to which index options andtor futures are used
so that shares may be traded more patiently in the cash market, and the frequency
with which limit orders are used also are reported here.
With regard to the meaning of immediacy, we asked respondents if they would
consider a trade to be immediate if it executed within a stated period of time
(Table 1). The majority (71%) answered that a trade must be realized in under 10
minutes to be considered immediate. Only 3% answered %thin 2 hours," and
6% said "within one day."
Having established a sense of what the respondents consider "immediacy" to be,
the questionnaire asked about the respondents' willingness to trade patiently if
their execution costs could be reduced by doing so ('Table 2). A total of 67%
Equity Trading Practices and Market Structure 7
Table 1: Time In Which You Consider A Trade To Be Immediate
Table 2: Willingness To Trade Patiently To Reduce Execution Costs
Under 10 minutes
1 Hour
2 Hours
1 Day
Other
No Answer
indicated that they would be willing or very willing to delay a trade if it reduced
their costs. Only 8% said they would not be willing.
Number of Respondents
107
1
4
9
7
1
5 (Very Willing)
4
3
2
1 (Not at All Willing)
No Answer
Willingness is one thing; the perception of how frequently a trade is delayed is
another. Therefore, the questionnaire asked how frequently traders in fact delay a
trade in an attempt to obtain a price that is more favorable than the price currently
prevailing on the market (TabIe 3). The vast majority (77%) of respondents said
they delay trades in hopes of finding a better price for 25-75% of their trades.
Only 11% said they "never" or "rarely" delay a trade for a better price. The
preponderance of the respondents perceive it desirable to trade patiently.
Percentage of Respondents
71.3
3
2.7
6.0
4.7
0.7
WILLINGNESS TO ACCEFT A TRADING DELAY OF ONE HOUR FOR A $50 STOCK
-IF You COULD SAVE 254 PER SHARE IN TRADING COSTS
Number of
Respondents
5 1
50
34
6
6
3
Evidence on the demand for immediacy was also obtained by asking respondents
whether they would be willing to accept a trading delay of one hour if, by so doing,
Percentage of
Respondents
34.0
33.3
22.7
4.0
4.0
2.0
8 Nicholas Economides and Roben A. Schwartz
Table 3: The Frequency With Which A Trade Is Delayed To Obtain A Price More
Favorable Than The Current Market Price
Table 4: Willingness to Accept A Trading Delay Of One Hour For A $50 Stock If
You Could Save 25# Per Share In Trading Costs
they could decrease trading costs by 25# a share for a $50 stock (Table 4). One
in four respondents said they would "rarely" or "never" delay a trade for an hour
t o reduce costs. On the other hand, more than half said they would be willing to
delay a trade to reduce costs on some or all of their trades.
Equity Trading Practices and Market Structure
9
Table 5: Willingness to Accept A Trading Delay Of 3 Hours For A $50 Stock If
You Could Save 25$ Per Share In Trading Costs
WILLINGNESS TO ACCEPT A TRADING DELAY OF 3 HOURS FOR A $50 STOCK
You COULD SAVE 2% PER SHARE IN TRADING COSTS
Traders were then asked if they would accept a delay of three hours for the same
-cost savings. One in three respondents said they would rarely, if ever, delay a
trade three hours. On the other hand, an appreciable subset (23%) said they would
accept a three-hour delay regularly or frequently to save 25 cents per share for a
$50 stock.
Jtespondents were also asked if they would delay a trade of 10,000 shares or more
for one hour if, by so doing, they could gain anonymity (Table 6). Slightly less
than half of the respondents said they would "rarely" or "never" delay a trade
for an hour to gain anonymity. But 19% said they would do so "regularly" or
"frequently." This indicates that an appreciable subset of participants commonly
do not seek to trade immediately to preserve anonymity.
How FREQUENTLY YOU WAIT MORE THAN ONE DAY BEFORE ACQUIRING OR
SELLING SHARES IN THE CASH MARKET IF YOU HAVE USED B ~ E X O ~ O N S
AND/OR FUTURES TO TRADE QUICKLY
One way to delay trading in the cash market is to trade a derivative contract to
establish a position that is then converted into shares over time. Table 7 shows the
responses from those that indicated that they use index options andlor futures to
reduce their need to execute trades quickly in the cash market. These respondents
were asked the frequency with which they would wait more than a day before
10 Nicholas Economides and Robert A. Schwartz
Table 6: Willingness To Accept A One Hour Trading Delay To Gain Anonymity
On A Trade Of 10,000 Shares Or More
Table 7: How Frequently You Wait More Than One Day Before Acquiring Or
Selling Shares In The Cash Market If You Have Used Index Options And/or
Futures To Trade Quickly
acquiring or selling the desired shares in the cash market. A total of 46% said
they rarely or never wait. On the other hand, a significant number (27%) said they
"ffequently" waited.
Never
Rarely (1-2496 Trades)
Sometimes ( 2549% of Trades)
Regularly (50-74% of Trades)
Frequently (75-100% of Trades)
No Answer
How O m DO YOU USE LIMIT ORDERS, MARKET ORDERS, AND MORE
In Table 8, we report on the frequency of the use of limit orders, market orders,
percentage orders, and basket or&rs. As expected, market, limit, and not held
orders are all widely used, and in roughly similar amounts. Basket orders and
'index optiondfutures are not used very much.
Perhaps the most interesting finding is the extent to which l i t orders are
used: 52% of the respondents said they used them on at least half of their trades.
The use of limit orders is essential to an order driven market. The dynamics of
price behavior apparently compensates traders sufficiently for placing limit orders.
Number of
Respondents
12
3
4
2
9
3
Percentage of
Respondents
36.4
9.1
12.1
6.1
27.3
9.1
Equity Trading Practices and Market Structure 11
Traders are explicitly not demanding or paying for immediacy when they use limit
orders in seeking to trade.
IV. TRADING PRACTICES, ORDER SIZE, AND TRANSACTION COSTS
Trading practices, order size, and costs shed further light on the willingness of
the respondents to trade patiently. The first question raised is the importance of
the major costs of these three categories: the opportunity cost of missing a price,
market impact, and commissions. Reasons to execute a trade quickly include the
volatility of prices, the possible mispricing of stocks, and the prevention of front
running. The costs associated with limit orders (e.g., the risk of non-execution
and the difficulty of withdrawing limit orders quickly) also impact the decision
to trade quickly. Respondents were further questioned as to their concern about
information leakage when a broker is called, and about the frequency with which
they demand capital from a broker for a block order. Attitudes toward costs are
also reflected in the respondents' answer to one other question, "How frequently
do you decide not to adjust your portfolio because the market is too illiquid?"
Concerning the size of their orders, respondents were asked about the frequency
with which an order for a stock is larger than the stock's average daily trading
volume, the time commonly given by portfolio managers to implement large orders,
and the frequency with which large orders are broken up for execution over time.
They were further questioned about the frequency with which it takes more than
one day for a large order broken into lots to execute completely. Lastly, the
questionnaire asked about the times of the day when the traders most prefer and
least prefer to place their orders.
How IMPORTANT ARE THE FOLLOWING COSTS?
Regarding the costs of trading, our findings on the importance of three major
components (the opportunity costs of missing a price, market impact, and com-
missions) are summarized in Table 9. The opportunity costs of missing a price are
rated the most important cost by 55% of traders, followed by market impact, which
i s rated the most important cost by 41% of traders. Commissions are important to
only 3% of the respondents.
Table 10 summarizes the most important and second most important factors that
may cause traders to want to execute a trade quickly. The most important factor
is "because prices are volatile and the risk of waiting is too greats' 48% of re-
spondents said this was the most important factor, and 32% said it was the second
most important factor. Fewer indicated that the prevention of front-running was a
12 Nicholas Economides and Robert A. Schwartz
Equity Trading Practices and Market Structure
13
Table 9: How Important Are The Following costs?
factor: 11% indicated it was the most important factor, and another 26% indicated
it was the second most important factor. Interestingly, only 23% said that the most
important factor was that other traders will realize that the stock is overpriced or
underpriced, and 21% indicated this was the second most important factor. To the
extent that trading is motivated by news and not just the assessment of existing
information, this number would be expected to be higher.
Opportunity Costs
of Missing A Price
Market Impact
Commissions
WHAT DO YOU CONSIDER THE MOST IMPORTANT AND THE SECOND MOST IM-
PORTANT DRAWBACKS OF USING LIMIT ORDERS?
When asked what they consider the most important drawback of using limit orders
(Table 1 I), most of the respondents (70%) stated that the most important factor is
the risk of non-execution. An additional 22% checked a closely related factor: limit
orders may cause you to miss a favorable market movement. Only 4% indicated
that the drawback is that the limit orders may be difficult to withdraw quickly. This
i s not surprising, given that the professional buy-side traders keep current about
market events, and that order handling procedures are rapid. The response here
is consistent with the previously discussed finding that price volatility is the most
important motivation for trading quickly (see Table 10).
Percentage of Respondents
Indicating Particular Cost as
A willingness to delay a trade on the part of roughly a third of the respondents
.in order to achieve anonymity is evidenced by the responses reported in Table 6.
Anonymity may be valued by buy-side participants because of the adverse price
impact that can occur when news gets out that they are seeking to trade. To assess
this, respondents were asked how concerned they are about information leakage
after they have called a broker to make a trade. The results are reported in Table 12.
A total of 45% indicated they were concerned about information leakage.
Most
Important
54.7
40.7
3.3
Neutral
Neutral
36.0
51.3
9.3
Least
Important
6.0
6.7
84.7
No Answer
3.3
1.3
2.7
14 Nicholas Economides and Robert A. Schwartz
Table 10: What Are The Most Important And Second Most Important Factors
That May Cause You To Want To Execute A Trade Quickly?
I I SecondMost I
and the Risk of Waiting is Too I I I
Because Prices Are Volatile
- - - - . . - - - . - -
Most Important Factor I Important Factor
72 I 48
Traders Will Realize the Stock I I I
Great
Becuase you Think Other
48.0%
35
is Overpriced or Underpriced
No Answer
0.7% 5.3%
32.0%
32
Opportunity Costs
To Prevent Other Traders
From Front-Running Your
Order
ONote that columns in this table add to more than 100% becuase some repondents have checked
more than one category.
23.3%
34
%o primary functions of intermediaries are (i) to help a customer find the coun-
terpart to a trade (i.e., act as a broker), and (ii) to provide capital as the counterpart
in a trade (i.e., act as a dealer). For a customer who is concerned about infor-
mation leakage, a strong motive must exist for contacting the intermediary in the
first place. Accordingly, the survey asked about the frequency with which the
respondents demand capital from their brokers for transactions of 10,000 shares
or more. The results are reported in Table 13. Approximately three out of four
respondents said they rareIy, if ever, demand capital fiom their brokers. Only 7%
said they regularly or frequently demand broker capital. Presumably this means
that the role of intermediaries in finding the other side of a trade is more important
than their role in providing capital.
21.3%
47
11.3%
17
11.3%
26.0%
39
26.0%
Equity Trading Practices and Market Strucrure 15
Table 11: What Do You Consider The Most Important And The Second Most
Important Drawbacks Of Using Limit Orders?
'Note that columns add to more than 100% because some respondents have checked more than
one category.
-
Table 12: Concern About Information Leakage When A Broker Is CaIIed
How FREQUE~ Y DO YOU DECIDE NOT TO ADJUST YOUR P o r n o m BECAUSE
THE hhiWET IS TOOUQUI D?
5 (Very Concerned)
4
3 (Neutral)
2
1 (Not Concerned)
No Answer
A total of 16% of the traders do not adjust their portfolio 10-19% of the time
because the market is too illiquid (Table 14). Almost twice as many do not adjust
a e i r portfolio for the same reason 1-9% of the time. In both cases, active traders
are more likely not to adjust their portfolios than passive traders.
Number of Respondents
46
22
47
15
18
2
Percentage of Respondents
30.7
14.7
31.3
10.0
12.0
1.3
16 Nicholas Economides and Robert A. Schwartz
Table 13: Frequency With Which Capital Is Demanded From A Broker For Trans-
actions of 10,000 Shares or More
Table 14: How Frequently Do You Decide Not To Adjust Your Portfolio Because
The Market Is Too Illiquid?
FREQUENCY WITH WHICH YOUR ORDER FOR A STOCK IS LARGER THAN THE
STOCK'S AVERAGE DAILY TRADING VOLUME
If the order is large relative to average daily trading volume, it may not be possible
10 execute the order entirely in a very short period of time without incurring an
unacceptably high execution cost. As noted above, Wayne Wagner and Mark
Edwards found that 66% of the orders in Plexus Group's data set exceed half of
the stock's average daily trading volume, and that 40% of the orders exceed the
stock's total average daily trading volume.
The respondents were asked the frequency withwhich their orders for a stock
are larger than the stock's average daily trading volume (Table 15). Almost a third
of the respondents answered that 20% or more of their orders are this large.
.
Never
27
22.7%
9
37.5%
37
24.7%
Active Traders
Passive Traders
All Respondents
1-9%
40
33.6%
6
25.0%
47
31.3%
10-19%
17
14.3%
1
4.2%
24
16.0%
20% or
more
5
4.2%
1
4.2%
6
4.0%
Don't
Know
24
20.2%
7
29.2%
34
22.7%
No
Answer
6
5.0%
0
0.0%
8
5.3%
Equity Trading Practices and Market Structure 17
Table 15: Frequency With Which Your Order For A Stock Is Larger Than The
Stock's Average Daily Trading Volume
In light of the size of institutional orders relative to average daily trading volume,
the questionnaire asked about the time a portfolio manager might typically give
a trader to implement an order (Table 16). For small cap stocks, less than 1%
answered "one hour or less:' and 69% answered "one day" or longer. For large
cap stocks, 5% answered "one hour or less," and 59% answered "one day" or
ionger. This finding reinforces the impression that asset managers do not attempt
to implement their trading decisions within brief intervals of time.
Never
1-9% of Orders
10-19% of Orders
7
20% or More of Orders
Don't Knowmot Sure
No Answer
FREQUENCY WITH WHICH LARGE ORDERS (100,000 SHARES OR MORE) ARE
BROKEN INTO SMALLERLOTS FOR SEPARATE EXECUTIONS OVER AN EXTENDED
PERIOD OF TIME
If an order is given time to be executed, it might be broken up for execution in
smaller pieces over a series of trades. The questionnaire asked the frequency with
which large orders of 100,000 shares or more are in fact broken up for this purpose
(Table 17). More than three out of five respondents indicated that 20% or more of
their orders are broken into smaller lots. Only 5% indicated that they never break
up their large orders.
Number of
Respondents
5
5 1
33
49
8
4
FREQUENCY WITH WHICH IT TAKES MORE THAN ONE DAY FOR A LARGE ORDER
B ~ o m INTO LOTS TO BE EXECUTED COMPLETELY
Percentage of
Respondents
3.3
34.0
22.0
32.7
5.3
2.7
The length of time typically taken to implement an investment decision in the
marketplace more directly reveals a willingness to trade patiently. The survey
questioned the frequency with which more than one day is taken to execute an
order completely when the order is broken into smaller lots to be executed over
'1 8 Nicholas Economides and Robert A. Schwartz
Table 16: Time Qpically Given By Portfolio Manager To Trader To Implement
A Large Order (25% Of Average Daily Trading Value Or More)
Other 5 3.3 5 3.3
No Answer 10 6.6 8 5.3
Table 17: Frequency With Which Large Orders (100,000 Shares Or More) Are
Broken Into Smaller Lots For Separate Executions Over An Extended Period Of
Time.
time (Table 18). While only 3% answered "never," 44% said that they frequently
or regularly broke their orders into smaller lots for execution over time.
Never
1-9% of Orders
10-19% of Orders
20% or More of Orders
Don't Know/Not Sure
No Answer
WHEN DO YOU PREFER TO PLACE YOUR ORDWS?
Turning to the question of when orders are placed, the respondents expressed
clear preferences for trading at different times during the day mbl e 19). Traders
Number of
Respondents
8
19
2 1
93
5
4
Percentage of
Respondents
5.3
12.7
14.0
62.0
3.3
2.7
Equity Trading Practices and Market Structure 19
Table 18: Frequency With Which It Takes More Than One Day For A Large Order
Broken Into Lots To Be Executed Completely
preferred the half hourjust following market opening to the actual market opening:
44% said that 9:31-10:00 am. was their most preferred time to place an order,
compared to 27% who most preferred the actual market opening. Traders also
preferred the half-hour period immediately prior to market close as compared to
the actual closing time: 23% said the 3:31-3:59 period was "most preferred,"
compared to 8% who most preferred the actual closing time to place their orders.
The survey did not ask for the reasons behind these preferences. Presumably, the
uncertainty concerning price determination at the open lead many to prefer the
9:31-10:00 a.m. period; and the uncertainty concerning price, and perhaps the
ability to trade at all, caused many of them to find the close least preferable, and
the 3:31-359 period less preferable than the 9:31-10:OO period.
Recognizing that the periods are not of equal length, one might expect from the
responses that the pattern of trading over the day would be "U" shaped, as indeed
it has been observed to be by, for instance, McInish and Wood (1990).1° The
questionnaire did not ask, however, the frequency with which orders were delayed
so that their arrival might be harmonized with the time of the day the respondent
felt to be most desirable.
V. THE USE OF ALTERNATIVE ELECTRONIC TRADING SYSTEMS
.The emergence of alternative electronic markets in recent years has given buy-
side traders new opportunities to receive timely executions at reasonable cost.
Respondents were asked about the frequency with which they use these systems
(e.g., NYSE after hours Sessions 1 and 2, Instinet's crossing session and continuous
market, POSITS crossing sessions and AZX's call market), and their motives
for using them (e.g., lower trading costs, the ability to trade anonymously). The
'Osee McInish and Wood (1990).
20 Nicholas Economides and Robert A. Schwartz
Table 19: When Do You Prefer To Place Your Orders?
respondents also were asked whether or not they felt the benefits of electronic trade
execution outweigh the disadvantages, how satisfied they are with the alternative
systems, and what would get them to use the alternative systems more (e.g., if they
gave higher execution rates, if they allowed trading at more convenient times, and
if the respondents' did not have soft dollar arrangements).
Table 20 shows that use of the alternative systems is limited. Use of these systems
is similar for Listed and for NASD stocks, except for Instinet's continuous market
which is used more for NASD stocks.
M m FOR TRADING ON THE ELECTRONIC SYSTEMS
Of particular interest are the respondents' motives for trading on electronic systems
(Table 21). The considerations that were rated "important" are: reduced market
impact (47%), lower bid-ask spreads (47%), better liquidity (41 %), lower general
transaction cost (39%), the ability to trade anonymously (38%). and the ability to
have greater control of the negotiation process (33%).
WHAT Emm DOES THE ANONYMITY OFFERED BY ELECTRONIC TRADING SYS-
. m s HAVE ON YOUR F~CU?ION ABILITY?
Nearly half of the traders expressed the opinion that the anonymity offered by the
alternative electronic trading systems improves their execution ability (Table 22).
Less than 1% think that it worsens it. On the other hand, 34.0% of the respondents
said they did not know what the effect would be or they did not answer the question.
Table 20: How Often Do You Use The Following Alternative Electronic Trading
Systems?
'The indication NA stands for "No Answer".
Listed Stocks
NYSE Session 1
NYSE Session 2
Instinet Crossing
POSIT
AZX
Instinet Continuous
NASD Stocks
Instinet Crossing
POSIT
AZX
Instinet Continuous
Never
86.7
90.0
63.3
60.0
78.7
52.7
62.7
62.7
78.0
49.7
1-9%
6.7
4.7
22.7
22.7
13.3
28.0
16.7
20.7
12.0
14.7
10-19%
0.7
0
2.7
4.7
0
4.7
6.7
3.3
0
11.3
30% ormore
1.3
0
5.3
5.3
2.0
4.0
5.3
4.7
2.0
14.7
20-2996
0.7
0
2.0
3.3
1.3
6.7
4.7
3.3
1.3
6.0
Don't Know
1.3
2.0
0.7
1.3
0.7
0.7
0
0.7
0.7
0
NAa
2.7
3.3
3.3
2.7
4.0
3.3
4.0
4.7
6.0
4.0
22 Nicholas Economides and Robert A. Schware
Table 21: Motives For Trading On The Electronic Systems
Table 22: What Effect Does The Anonymity Offered By Electronic Trading Sys-
tems Have On Your Execution Ability?
Do You B m THAT THE BENEFITS OF ELECI-RONIC TRADE EXECUTION OUT-
WEIGH THE DISADVANTAGES OR THAT THE DISADVANTAGES OUTWEIGH THE
BENEFITS?
Improves it
Has no Effect
Worsens it
Don't Know
No Answer
Table 23 shows that a large majority of respondents (67%) believe that the benefits
of electronic trade execution outweigh the disadvantages. Only 16% of the traders
expressed the opinion that the disadvantages of electronic execution outweigh its
benefits.
Number of
Respondents
63
35
1
27
24
Percentage of
Respondents
42.0
23.3
0.7
18.0
16.0
Equity Trading Practices and Market Structure 23
Table 23: Do You Believe That The Benefits Of Electronic Trade Execution Out-
weigh The Disadvantages Or That The Disadvantages Outweigh The Benefits?
SATISFACTION OF USERS WITH THE DIFFERENT ALTERNATIVE WING SYSTEMS
Benefits Outweigh the
Disadvantages
Disadvantages Outweigh the
Benefits
No answer
Table 24 shows the degree of satisfaction with the alternative trading systems.
Among these, the least satisfaction was expressed for AZX (29%). It is not sur-
prising that 46% of AZX's customers are not satisfied: the AZX market is called
at 5 p.m., liquidity is insufficient, and execution rates are low because aggregate
-order flow is sparse. One would expect that satisfaction with these alternative
markets would be considerably greater if they were integrated better with the ma-
jor trading systems. We also note that there was widespread dissatisfaction with
NYSE's Sessions 1 and 2. However, a substantial majority (79%) were satisfied
with Instinet's continuous market.
WHAT WOULD GET YOU TO USE THE ALTERNATIVE TRADING SYSTEMS MORE
If clear motives exist for trading on electronic systems, why aren't the systems
used more heavily? ITG's POSIT, Instinet's crossing network, and the Arizona
StockExchange's AZX each batch orders for multilateral execution at a single time
at a single price; if institutional asset managers are willing to forgo immediacy,
why aren't these systems particularly attractive to them? The survey asked the
question, What would get you to use the alternative trading systems more?" The
results are in Table 25.
Not surprisingly, 55% said that they would use the alternative systems more if
-they gave higher execution rates. This is consistent with the reality that a lack of
order flow is a major impediment to the success of any trading system (and with
the adage, "order flow attracts order flow"). Second on the list, 35% indicated that
they would use the systems more if they did not have soft dollar arrangements (that
is, soft dollar arrangements appear to be an impediment to change). A total of 3 1 %
claimed they would use the alternative markets more if they allowed trading at more
Number of
Respondents
101
24
25
Percentage of
Respondents
67.3
16.0
16.7
24 Nicholas Economides and Robert A. Schwartz
Table 24: Satisfaction Of Users With The Following Alternative Trading Systems
Table 25: What Would Get You To Use The Alternative Trading Systems Morea
Instinet Continuous
"Note that columns add to more than 100% because some mpondena have checked more than
one category.
29.2%
56
78.9%
They Gave Higher Execution Rates
You Didn't Have Soft Dollar
Arrangements
They Allowed Trading at More
Convenient Times
You Knew More About Them
Other
None of the Above
convenient times (presumably during the day rather than after hours), and 20%
responded that they would use them more if they knew more about them (which
suggests some continuing lethargy on the part of some institutional investors).
25.0%
9
12.7%
Number of
Respondents
82
53
47
30
7
47
45.8%
6
8.5%
Percent of
Respondents
54.7
35.3
31.3
20.0
4.7
31.3
Equity Trading Practices and Market Structure 25
Table 26: Distribution of Respondents by Institution
VI. THE SAMPLE OF RESPONDENTS
AND THEIR REASONS FOR TRADING
Independent Investment Management F i
Subsidiary of Bank or Brokerage Firm
Mutual Fund
Internally Managed Pension Fund
Other
Total
A total of 825 questionnaires were mailed to 125 members of the TraderForum
and to 700 non-members. A total of 150 responded. These include approxi-
mately 90 TraderForum members and 60 non-members." Thus, the response was
72% of TraderForum members and 8.6% of non-members. In terns of our total
respondents, 60% were TraderForum members and 40% were non-members.
The questionnaires were filled out by the equity trader at each institution. In
some of the smaller institutions the trader may also be an asset manager. Re-
spondents were asked the total value of their organization's equity assets under
management. A total of 135 out of the 150 answered. The estimated amount of
equity under management was $1.54 trillion.12 This represents approximately half
the managed equity assets13 The distribution of the respondents, according to the
type of institution, is shown in Table 26.
Table 27 shows the reasons for trading stated by the respondents in descending
order. The primary reasons are stock specific fundamental issues (79%), internally-
generated research (68%), reassessment of portfolio structure (47%), bargain-
hunting (37%). and profit taking (36%).
l l ~e mbe r s automatically receive the =port while non-members must send in a card to receive the
report. We expect that virtually all non-members who took the time to fill out the questionnaire would
want to receive the repon Thus the number of non-membezs is inferred from the number of cards that
wen received.
120ut of the 150 respondents. 128 respondents repoxted a total of $1.316.42 billion of equity under
management and 22 gave no answer. Extrapolating to the total of 150. we estimate the total assets
under management of the mpondents of the questionnaire at $1.54 trilIion.
13~otal equity assets in the US at the end of 1992 were $5.5 trillio- ( F h ofFundt Codcd Tnbk s.
Board of Governors of the Federal Resuve System. Washington. D.C.). It is estimated that 60% of
these an managed, so that managed equity assets are $3.3 trillion. Thus. our survey covers about 50%
of all managed equity assets.
Number of
Respondents
69
53
14
9
5
150
Percentage of
Respondents
46.0
35.3
9.3
6.0
3.3
100
26 Nicholas Economides and Robert A. Schwartz
Table 27: Why Do You Trade?
"The indication NA in the top nght hand comer stands for "No Answef.
MI. DIFFERENCES BETWEEN ACTIVE AND PASSIVE TRADERS
One of the interesting questions for which the responses to our questionnaire pro-
vide an answer is whether significant differences exist in the trading behavior of
active traders in comparison with passive traders. In particular, we are interested to
see if these two groups have reported differences in what they consider an immedi-
ate trade, on the willingness to trade patiently, on motives for trading in electronic
?ystems, on the effects of anonymity, or in their reasons for trading in general.
The answers of active and passive traders to many questions were similar.
However, in some questions their answers could easily be differentiated. We
summarize below the responses in which active and passive traders showed clear
difference^.'^
l4Nl Tables in this section an numbered Nb where N is the number of the comsponding table
presented earlier that summarized responses to the same question by al l participants.
Equity Trading Practices and Marker Structure 27
Table lb: Time In Which You Consider A Trade To Be Immediate: Differences
Between Active And Passive Traders
Table l b shows that the time horizon appears to be a bit shorter for active than
for passive traders, as one might expect. At the short end of the scale, 76% of the
active traders checked 10 minutes or less, versus 50% of the passive traders. At
.the long end of the scale, 3% of the active traders checked one day versus 25% of
the passive traders.
TabIe 2b shows a tendency for passive traders to be more willing to trade
patiently: 46% of the passive traders said they would be "very willing:' versus
31% of active traders. No passive traders indicated they would be not willing
at all or not very willing versus 10% of active traders. This is consistent with
expectations. Trading on news implies a need for immediacy on the part of active
traders, and seeking to minimize transaction costs implies patient trading on the
part of passive traders. But again, the difference between the two groups is not
.large.
Table 12b distinguishes between active and passive traders on the issue of
concern about information leakage. Despite the general similarity, a substantial
percentage (33%) of passive traders are not concerned at all about information
leakage compared to 8% of active traders.
Table 13b shows that passive traders are much more likeIy never to demand
capital from a broker (33% versus 17% for active traders). Further, none of the
passive traders regularly or frequently demand capital from a broker, while 8% of
the active traders do.
. Table 21b shows the differences between active and passive traders in their mo-
tives for trading on electronic systems. Among passive traders, lower transaction
costs are the primary motivation (75.0%); this motive was indicated by 42.9% of
active traders. Active traders are motivated by a variety of other reasons. Reduced
market impact and lower spread costs are the primary reasons for active traders,
28 Nicholas Economides and Robert A. Schwam
Table 2b: Willingness To Trade Patiently To Reduce Execution Costs: Differences
Between Active And Passive Traders
[ No Answer I 2 I 1.7 0 0
Table 12b: Concern About Information Leakage When A Broker Is Called: Dif-
ferences Between Active And Passive Traders
indicated by 52.9% and 52.1% respectively. Each of these motives was indicated
by only 25% of the passive traders as a primary motive. 45.4% of the active traders
indicated better liquidity as a primary motive, in contrast with 20.8% of the passive
traders. Similarly, anonymity was indicated by 42.9% of active traders as a motive,
and only by 16.7% of passive traders. Finally, greater control of the negotiation
process was indicated by 36.1% of the active and 20.8% of the passive traders.
-
Table 22b shows that more active than passive traders believe that the anonymity
offered by electronic trading systems improves their execution ability. This opinion
is expressed by 46% of the active traders and 25% of the passive traders.
Table 13b: Frequency With Which Capital Is Demanded From A Broker For
Transactions Of 10,000 Shares Or More: Differences Between Active And Passive
Traders
Never
Rarely (1-24% of
Trades)
Sometimes (25-49%
of Trades)
Regularly (50-74%
of Trades)
Frequently (75-
100% of Trades)
Don't Know1
Not Sure
No Answer
Active Traders
Number of
Respondents
20
65
19
5
5
4
1
Passive Traders
Percentage of
Active Traders
16.8
54.6
16.0
4.2
4.2
3.4
0.8
Number of
Respondents
8
13
3
0
0
0
0
Percentage of
Passive Traders
33.3
54.2
12.5
0
0
0
0
2
30 Nicholas Economides and Robert A. Schwartz
Table 21 b: Motive For Trading On The Electronic Systems: Differences Between
Active And Passive Traders
Table 23b shows that a larger percentage of active than passive traders believe
that the benefits of electronic trade execution outweigh its disadvantages. This
was expressed by 71% of active traders, compared with 46% of passive traders.
Table 25b shows significant differences between active and passive traders in
the reasons that would make them use the alternative trading systems more. Active
traders (61%) would use the electronic trading systems more if the systems gave
higher execution rates. Only 21% of passive traders indicated this. Similarly, 36%
-of active traders, in contrast with 25% of passive traders, indicate that they would
use more such systems if they did not have soft dollar arrangements. If systems
were available at convenient times, 35% of active traders, in contrast with 13% of
passive traders, indicate that they would use such systems more. Finally, 22% of
active traders, in contrast with 8% of passive traders, indicate that they would use
these systems more if they knew more about them.
Reduced
Market
Impact
Lower
Spread
Costs
Better
Active Traders
Number of
Respondents
Who Rated
Motive
Important
63
62
Passive Traders
Percentage
of Active
Traders
52.9
52.1
Number of
Respondents
Who Rated
Motive
Important
6
6
Percentage
of Passive
Traders
25 .O
25.0
Equity Trading Practices and Market Structure 3 1
Table 22b: What Effect Does The Anonymity Offered By Electronic Trading
Systems Have On Your Execution Ability?
Table 23b: Do You Believe That The Benefits Of Electronic Trade Execution
Outweigh The Disadvantages Or That The Disadvantages Outweigh The Benefits?
Table 26b distinguishes between active and passive traders in the distribution
of the respondents according to the type of institution where they are more likely
l o trade. Among active traders, the majority (5 1%) are in independent investment
management firms, while 34% trade for a subsidiary of a bank or a brokerage h.
The roles are reversed among passive traders: the majority (54%) of passive trader
respondents trade for a subsidiary of bank or a brokerage firm, while 29% is in
independent investment management firms.
Table 27b below shows that the most important reasons to trade for active
traders concern their evaluation of fundamental information concerning individual
stocks. Most respondents (87%) indicated that they frequentIy trade because of
stock specific, fundamental issues; and 74% indicated that they frequently trade
because of internally generated research. Interestingly, passive traders also trade
for these reasons, though not as much as the active traders. Only 58% of passive
-
Benefits
Outweigh the
Disadvantages
Disadvantages
Outweigh the
Benefits
No Answer
Active Traders
Number of
Respondents
84
19
16
Passive Traders
Percentage
of Active
Traders
70.6
16.0
13.4
Number of
Respondents
11
5
8
Percentage
of Passive
Traders
45.8
20.8
33.3
32 Nicholas Economides and Robert A. Schwartz
Table 25b: What Would Get You To Use The Alternative Trading Systems More:
Differences Between Active and Passive Tradersa
'Note that columns add to more than 100% because some respondents have checked more than
one category.
traders frequently traded because of stock specific fundamental issues, and 21%
-because of internally generated research. On the other hand, 33% of the passive
traders traded to trace a market index versus 8% of active traders. And 33% of
passive traders traded because of fund redemptions or other cash flow reasons
versus 19% of active traders.
In some respects the two groups are quite similar: 39% of active traders fre-
quently trade for profit taking compared to 25% for passive traders; 38% of active
traders trade for bargain-hunting purposes compared to 38% of passive traders; and
74% of active traders infrequently trade because of chartist signals compared to
75% for passive traders. Overall, the active and passive traders differ in emphasis,
'but not by as much as expected.
VIII. CONCLUSION
This paper has presented an assessment of the demand for immediacy by buy-side
institutional equity traders that we surveyed. The 150 surveys that were returned
clearly indicate that an appreciable number of the respondents do trade patiently.
This is not surprising in light of the size of trades that the institutions commonly
seek to make, and of the costs to them of obtaining immediacy in a continuous
Equity Trading Practices and Market Structure 33
Table 26b: Distribution of Respondents by Institution: Differences Between Ac-
tive and Passive Traded
ONote that columns add to more than 100% because some respondents have checked more than
one category.
Independent
Investment
Management
Firm
Subsidiary
of Bank or
Brokerage
Firm
Mutual
market. Their orders commonly exceed the average daily trading volume for a
stock, the large orders are commonly broken into smaller pieces, and the smaller
pieces often take a day or more to be executed completely. The respondents were
also concerned about losing anonymity to sell-side brokerfdealers.
These findings have a major implication for market structure. Increasingly,
electronic technology is making it possible for institutional buy-side participants to
meet each other directly in a disintermediated environment. This can be done most
effectively with batched (i.e., call market) trading arrangements, which establish
place and time meeting points. A perceived limitation of call market trading is that
it does not supply immediacy to participants. This is true, however, only if call
markets are used in place of continuous trading, rather than along with continuous
trading, as we recommend.
However, immediacy per se does not appear to be urgently sought by many
buy-side asset traders. This suggests that, if both call and continuous markets
were readily available to participants, the order flow directed to the calls would,
indeed, be appreciable. The bottom line is that providing electronic call market
trading would be desirable for an appreciable number of institutional investors.
Internally
Managed
Pension
Fund
Other 16.7
Active Traders
Number of
Respondents
61
40
Passive Traders
Percentage
of Active
Traders
51.3
33.6
Number of
Respondents
7
13
Percentage
of Passive
Traders
29.2
54.2
34 Nicholas Economides and Robert A. Schwartz
Table 27b: Why Do You Trade? Differences Between Active and Passive Traders
'The indication NA stands for "No Answer".
M. REFERENCES
Amihud, Yakov, and Haim Mendelson. 1980. "Dealership Market: Market-Making with
,
Stock Specific
Fundamental
Issues
Internally-
generated
Research (from
Portfolio
Manager)
Reassessment
of Portfolio
Structure
Bargain-hunting
Profit Taking
Market-wide
News
Fund Redemp-
tions or Other
Cash Flow Reasons
Trading
Information
(i.e., Knowledge
of an Order
on the Floor)
Desire to
Cut Losses
Chartist Signals
Need to Track
a Market
Index
Derivatives-
motivated
Trading
Other Factors
Frequ
ently
86.6
74.8
47.9
37.8
39.4
34.5
19.3
18.5
18.5
14.3
7.6
4.2
1.7
Frequ
ently
58.3
41.7
54.2
37.5
25.0
20.8
33.3
8.3
4.2
4.2
33.3
8.3
0
Active
Neu
tral
8.4
17.6
31.9
29.4
30.3
33.6
24.4
31.9
34.5
11.8
10.1
5.9
0.8
Passive
Neu
tral
16.7
20.8
20.8
12.5
33.3
29.2
29.2
29.2
29.1
20.8
12.5
0
0
Traders
Infreq
uently
3.4
6.7
17.6
32.8
30.3
30.3
54.6
48.7
46.2
73.9
80.7
88.2
10.1
NA
1.6
0.9
2.6
0
0
1.6
1.7
0.7
0.8
0
1.6
1.7
87.4
Traders
Infreq
uently
25.0
33.3
25.0
50.0
41.7
50.0
37.5
62.5
66.7
75.0
54.2
91.7
12.5
NAa
0
4.2
0
0
0
0
0
0
0
0
0
0
87.5
Equity Trading Practices and Market Structure
35
Inventory." Journal of Financial Economics. (March 1980).
Amihud. Yakov, and Haim Mendelson. 1985. "An Integrated Computerized Trading Sys-
tem," in Market Making and the Changing Sttucture of the Securities Indurtr~ Ed.
Y. Amihud, T. Ho and R. Schwartz Lexington Books Lexington. MA 1985.
Cohen, Kalrnan J., S. Maier, Robert A. Schwartz, and D. Whitcomb. 1986. The Micmstruc-
rure of Securities Markets. Englewood Cliis. NJ: Prentice Hall.
Cohen. Kalman J., and Robert A. Schwartz. 1989. "An Electronic Call Market: Its De-
sign and Desirability," in The Challenge of Information Technology for the Securities
Markets: Liquidity, Volatility, and Global Trading. H. Lucas and R. Schwartz Editors.
Economides. Nicholas. 1993. "Network Economics with Application to Finance." Financial
Markets, Institutions & Instnunents 2: 5: 89-97.
Economides, Nicholas. 1995. "How to Enhance Market Liquidity," chapter 6 in
Robert A. Schwartz (ed.) Global Equity Markets, Irwin Professional. New York: 1995.
Economides, Nicholas. 1996. 'The Economics of Networks." International Journal of In-
dwtrial Organization. Forthcoming.
Economides. Nicholas and Jeff Heisler. 1994. "Co-existence of Call and Continuous Mar-
kets:' mimeo.
Economides. Nicholas and Robert A. Schwartz. 1995. "Electronic Call Market Trading."
Journal of Portfolio Management 21: 3: 1&18.
Economides, Nicholas and Robert A. Schwartz. 1994. Making the T d e : Equity T d n g
Practices and Marker Structure-1994, TraderFonun, Institutional Investor. New York.
Economides. Nicholas and Aloysius Siow. 1988. "The Division of Markets is Limited by
the Extent of Liquidity: Spatial Competition with Externalities." American Economic
Review 78: 1: 108-121.
Garbade, Kenneth, and William Silber. 1976. "Price Dispersion in the Government Securi-
ties Market." Journal of Political Economy 84.
Garbade. Kenneth, and William Silber. 1979. "Structural Organization of Secondary Mar-
kets: Clearing Frequency, Dealer Activity and Liquidity Risk." Journal of Finance 34:
577-93.
Garman, M. 1976. "Market Microstructure." Journal of Financial Economics (June).
Handa. Puneet, andRobert A. Schwartz. 1996. "How Best to Supply Liquidity to a Securities
Market:' Journal of Portfolio Management. Forthcoming.
Ho, T.. and Stoll, H. (1981). "On Dealer Markets Under Competition." Journal of Finance
(May).
McInish, Thomas H., and Robert A. Wood. 1990. "An Analysis of Transactions Data for
- the Toronto Stock Exchange: Return Patterns and End of the Day Effect" Journal of
Banking and F i i c e 14: 441458.
Mildenstein, E., and 'schleef, H. 1983. 'The Optimal Pricing Policy of a Monopolistic
Marketmaker in the Equity Market." Journal of Finance (March).
Schwartz, Robert A. 1983. Reshaping the Equity Markets: A Guide for the 1990s. ~ & r -
Business, 1991 (reissued by Business One Irwin. 1993).
Schwartz, Robert A.. 1996. (editor) The Electmnic Call Market, Irwin Professional, forth-
coming, 1996.
Wagner, Wayne, and Mark Edwards. 1993. "Best Ekecution." Financial Analysts Journal
(JanuaryIFebruary).
36 Nicholas Economides and Robert A. Schwam
X. APPENDIX: THE QUESTIONNAIRE
WuUonal Imwtor CARD 1
ZAaxHonr W E : E m T w x m p w m wo YUm mRvcTm
HOWANDWm YOU TRAM
Listed below pc r nmnber of rwta th.1 might motivate s oc k wades Pl ase indicate how Itequcmly yom nada
uemachr l t al byt hef ol l opr i ngk~( Chr ckI boxh t t os r her es - ' l t egant l y- d 1 -'wcatrIl'.)
b. Market-vide news 5 0 ~ L I 1 0 -2 3 0 .J 2 0 4 I 0 .I
d. Fund redemptions or
ochnuhflowrcaums50 at 4 0 .I 3 0 .J 2 0 4 1 0 .I
e. Need to track I
market index 5 0 rrl 4 0 a 3 0 .J 2 0 4 I 0 .I
h. Desire to cut losses 5 0 ~rc 4 0 .r 3 0 a 2 0 .I 1 0 .I
i. Chutin signals 5 0 ln 4 0 . a 3 0 a 2 0 4 I 0 J
j. lnternrlly-:-ted
reseuch ('om portfolio
manager) S 0 14.1 4 0 d 30.1 ' '204 1 0 :I
k. Tnding informmion (i.e. -
knowled~e o f u adn
cnlhc[bor) S 0 crc 4 0 -z 30.1 2 0 4 1 0 .I
m. 0th- (please speci&) I
I
I
5 0 17-28 . I 4 0 .I ' 3 0 .I 2 0 4 10 4
How milling at you to trade pmiently in m attempt to redwe execution corn?
Vc y willing or m tit rillm:
5 0 21-1 4 0 . 1 3 0 .J 2 0 4 I 0 -I
CARD 1
Equity Trading Practices and Market Structure
3 Whsl do you cauida an trrmaliate &&? A t d c a d mthk
Beuwprices am d d k md lhc risk of waiting is tw 0 .I b -1
$. Plcue indicate, ar a sea* of 1 lo J rvhen 1- lan p r c f d nd II - *most prcled,' how you f d fib& plums
orden a~ each of the lhna listed below. ( k k cne box fa each h e paiod l i d )
38
Nicholas Economides and Robert A. Schwanz
7. Ona) oobndai dcd( o~horol l cnrol ddyoowccp( addayrl rncbwr I b a a o J I ~ g a S J Oi f y o o
mdd m he frdkwinr in carmdrrian* bid-mk wmd md mrLct bout PLEASE INDICATE A RESPONSE
sniap or:
r e p e r h e 0n.1 0.1 0.1 0 4 0 .I 0 4
b. l l Cpqrbra 0-1 0 . 1 0 .I 0 4 [3 .I 04
LlllCparbra OM 0 4 0 .I 0 4 t ) .I 6 4
d S W p e r h Ont 0 . 1 0 .I 0 4 0 .r 04
C7scpCrhra 0 I b t 0.1 0 .I 0 4 0 4 0 4
c l l p e r h r a 0 ~1 0 . 1 0 4 0 4 0 .I 0 4
I. Ona y a , h&d&k dt o~hov ol t a r r oul dyar r apl addsyof t hr nbol r nf aar t ocf ~g~S5Oi f you
d save che fokwhg in oarnioiom. bid-ak spmd md mrLa i mpat: (PLEASE INDICATE A RESPONSE
FOR EACH OF THE USTED SAVINGS PER SHAM OPTIONS.)
Don'tbund
S a me r i ma R r ml r l v - - ,
d (2549??d (S&74%d ( 7S- I W?of
-) -1 -)
Saving 0T:
r6cperha-e O w t 0.1 0 .I 0 4 0 J 0 4
b.lZdpahare Omt 0.1 0 .I 0 4 0 4 04
c211~paha-e 0-1 0.1 0 .I 0 4 0 -I 0 4
d SW pa thrre 0 41.1 0.1 0 .I 0 4 0 4 0 4
L 7% padwe 0 a 1 0.1 0 -1 0 4 0 .I 0 4
f . $ l p ~ h t 0-1 0 -1 0 4 0 -1 0 4
- -
mdini at S O ifp muld raw che fdkr;inl in c a mk h , bid.& &d md h e f
(PL~WE
INDICATE A RESPONSE FOR EACH OF THE USTED SAVINGS PER SHARE OPTIONS.)
Equity Trading Practices and Market Structure
0nChoodtl.y O r a l 0 . 2 0 .I 0 4 0 4 0 4
n Kc hoo dehy 0 11.1 a .I 0 .1 0 4 a 4 0 4
bby ddg O m r 0 . 2 0 4 0 4 0 .1 0 4
Owrm~byddy 0 a r 0 .I 0 .1 0 4 0 .l 0 4
It. W N G YOUR ORDERS
I How 0 t h do yar me theLllmm~:
Dear h I
&!a BmIY Sanai mal l rml rf v- -
(1.24% of 0549%0f (5674% of ClJ-lW/. d
?a. Do yar believe k are drawbacks to usin8 limit adco? y a 0 -1 m 0.1
26. U~wmw+rcd~toQuacMo2a,rhrcdoyacrmaidathe ' -,
. . .
7
P l u w Ldiatc by checkin8 it in C o l m A belor.
? c . I n Cd u n n B . p l a r e c h c d r h ~ ?' ' ~ t o l i i t a d a .
(Pk.pcchcctaJyaaboxinachdm)
REASONS FQB Not USW-
CdmraA s ! h n B
MarCirnpatal Secadmortimportal
fmI
Risk oCnon-cx& 0 a.r 0 a 1
on-i0lmcdi.l~ cxauti~l 0 4 , 0 r
MycrrtlccanpticiwdimdvQUgc 0 -3 0 4
M y be di f f dt lo withdm qtkkly 0 4
Myaurcya,tomiuabvurbknwkumovarrn(
0,
0 -1 0 1-3
Gikophnstohdcrla 0 4 0 4
olhcr (piur~ spaiv): 0 -74 0 .,a
40 Nicholas Economides and Robert A. Schwanz
i W h e n y w d . n a d e r o f 1 0 . 0 0 0 ~ h ~ a m m b a L m f i n ~ ~ h ~ ~ & ~ ~ i c ~ f l ~ ~ ~ a ~ ~
fobtin# execution?
Don't know/
Equity Trading Practices and Marker Structure 41
8. Howmuch time doa yar pcdol i i mrupr lypially lire p 0 imp- r ! r ~ adcr (25% of avenge daily
t n d i n g v . h r a m m ) t o r r k ( c unpmy ( mnpy r muk c t ~ of Sl M) mi i or a m) B;
(cunpmy wi& r nmka api ul i i o f b dun SlO0 dllim). (PLEASE CHECK ONLY ONE
BOX W EACH COLUMN.)
9 Pkuci ndi acc~i mpaul ( I hcf dl oumga~r r ebyml dngl hani nader 0t ' ~h I (modhqmmnt)
to 3 (kut impacrm). (PLEASE USE EACH NUMBER ONLY ONfX.)
Mda impact - -.I
costs ofmissins r pria
- I
Ctmnniams - I
10. When you 4 r brdrcr, how c o n d are you aboul infam~ia, leakage?
11. Haw~tly&yw~wmmIh~anbmksbdiuliaurbatlaadsvhichislu~dativetoIhcmnrge
daily adhrg volume (ADTV)?
2 5 ~ 0 % of ADW 0 at 0.3 0 .s 0~ 0 J
Ma e lhm JogA of
ADTV b ISI 0.3 0 -I o a 0 J
42 Nicholas Economides and Robert A. Schwartz
Up to l hour 0 I 0 .I 0 .I 0 4 0 .s .- 0 4
BrrwcenIhour
rd3hcm OILI 0.l 0 .t 0 4 0 .s Or
&c~ar 3homr
md l by 0 15.1 0 .I 0 .I 0 4 0 .s Or
M a c h
Iby 0 lCl 0 .1 0 .I 0 4 0 -3 0 4
Equity Trading Practices and Market Structure
43
II. ALTERNATIVE TRAMWQ SYSTEMS
russsmh
Cnnsmg 0u.1 0 .1 0 -1 0 4 0 .9 O r
POSK C]%I 0 .I 0 4 0 A 0 .s 04
AU[ 0 n.1 U .I 0 -1 0 4 0 4 Or
I n r c i n a r mt h mn Un ~ 0.1 0 .I 0 4 0 .s Or
2. If you w my dlhe folloumg rltanrtirc systems. wlut is yar gmml kvd dmisfwtion with durn oo I sale of I to
5. u h 5 ' vay salisrt. ad I= .nac a dl satisft.
NYSE Sasion I S OS I 40. 1 30. 3 2 0 4 1 0. 3 or
NYSE Sasion 2 SC]mt 4 0 ~ 3 0 4 2 0 4 l o . ? 04
InshtCrossing 3031. 1 4 0. 1 30 . 3 2 0 4 1 0 - s 04
POSK S o n - I 4 0 . a 3 0 . 1 2 0 4 I 0 . s or
IW S 0 a 1 4 0 . 1 3 0 4 2 0 4 1 0 4 O r
br ut ~~~~ndkmw 5 0 - 1 4 0 . 2 30. 8 2 0 4 , 1 0 4 04
I
Bette liquidity 5fJas1 4 0 4 3 0 . 1 2 0 4 1 0 . 1 04
TRde ma y mdy 5 Ow1 4 0 . 3 3 0 - s 2 0 4 1 0 . 1 or
Reduccnddi mpl d 5 o m 1 4 0 4 3 0 4 2 0 4 1 0 . 1 or
~~~ S o n 4 0 . 1 30. 3 2 0 4 1 0 . 1 04
LmuWa uc l i ooc os u S O W 4 0 . a 3 0 . 1 2 0 4 1 0 . s $0 4
Greaaeorardof
&ItMopocar 5 0 - 1 4 0 4 3 0 . ~ 2 0 4 1 0 . 1 or
*(pk-&fy):
~ On r r . 1 ' 4 O- r 3 0 . s 2 0 . 1 1 0 . 1 04
Nicholas Economides and Robert A. Schwartt
G. One the whole, do you Wic Bu the W~S of clecbon* lndc c~ccutia! outrucigh the didvmuges ar h t
thediudvantaga outweigh thebrncfits?
Benefits ouweigh IIIC disdvmugcs 0 w
Disldvantlgcs ouh\righ the benefits 0 .l
Equity Trading Practices and Market Structure
CARDS 2n
N. CWSlFICATWN DATA ( l o bo used for afutyrlt, nd for )dmlllkr(bn)
C. What is he toul value of your uganbtion's assets llndcr mmugmmt?
D. Please sate the puunup of .v& funds that are h a t e d b the fdlowing
Equitia: %OM
Fixed inuxne: % mu
Crrh ./. ~r t r
olha @leare &&I:
% nd
% s n
-
100.3 (CaIegolia dlcdd Id loo./.)
Value ofsmwitia badd Vduc of cumkial dourn: (41-79-2 I
S . , 000. 000 sr 5 . , 000. 000 sa
El .
bnh yw fw campkthq hh qurr(knluh. Pkna nd nrmd Ihd al yaw mponm m conM~&l andanolc)m~n. Pkeae
durn I l n h. pnplid burhas8 nply anvrkpa pdded Wf U RW
Irrld R8armhOeprrku
h ~ J h v 8 . k
UI Yadhon Annu8
w.rY*Nv lMP
Fa: UI I ) ms%4!
NOTTIS ON SUBMISSION OF MANUSCRIPTS
Send two copies of the manuscript to the editor at the address on the back cover of
the journal. The contents must represent original and unpublished work and the
manuscript should not be considered for publication elsewhere. Any necessary
permissions or copyright clearance must be obtained by the author. There is no
submissions charge or page fee.
Please follow journal style when word processing or typing the manuscript.
Manuscript should be submitted on standard paper with text on one side only,
and on diskette if possible. The manuscript should be approximately 75 journal
pages. Double-space all lines, including footnotes and referenceshibliography.
Footnotes should be numbered in the text and grouped at the end of the manuscript.
The position of diagrams and figures should be clearly marked in the manuscript
and, where possible, camera-ready copy should be provided by the author.
Each manuscript should include a title page, a table of contents, and a short
abstract. A brief biography for each author should also be included. Acknowl-
edgements, if any, should be supplied on a separate page or on the same page as
the author's biography.
References should be cited in text by author and (in parentheses) year of pub-
lication. References andlor bibliographies at the end of the manuscript should be
arranged alphabetically by author and follow the style of these examples:
Belsley, D. A., E. Kuh, and R. E. Welsch. 1980. Regression Diagnostics. New
York: John Wiley and Sons, Inc.
Dimson, P. and P. Marsh. 1984. "An Analysis of Brokers' and Analysts' Unpub-
lished Forecasts of U.K. Stock Returns." Journal of Finance 39:4: 1257-92.
Golec, J. 1988. "Empirical Tests of Principal-Agent Model of Investor-Investment
Advisor Relationship." Unpublished paper, Clark University.
Nakansson, N. 1979. "A Characterization of Optimal Multiperiod Portfolio Poli-
cies." Pp. 169-77 in Portfolio Theory, 25 Years AAfter: Essays in Honor of
Harry Markowitz, eds. E. Elton and M. Gruber. Amsterdam: North-Holland.
FINANCIAL MARKETS, INSTITUTIONS &
INSTRUMENTS
Editor Anthony Saunders
Managing Editor: Mary Jaffier
Editorial Policy Created out of the Monograph Series in Wnance and
Economics, this unique new journal specializes in the publication of
original studies in the fields of finance and economics, specifically in the areas
of financial markets, institutions and instruments. It provides an
outlet for manuscripts which are too long and broad for typical scholarly
journals and business periodicals.
Articles for review should be sent to the editor at the Leonard N. Stem
School of Business, New York University, 44 West 4th Street, Suite 9-160,
New York, NY 10012. For derails on manuscript submission. please see the
inside back cover of the journal.
doc_150669243.pdf
The continuous market has been widely studied by academic researchers. Much of the microstructure literature has focused on the dealer market. Early analyses include Garman (1976), HO and Stoll(1980), Amihud and Mendelsohn (1980), and Mildenstein and Schleef (1983).
Volume 4, Number 4
Equity Trading Practices and
Market Structure: Assessing Asset
Managers' Demand for Immediacy
NICHOLAS ECONOMIDES
AND
ROBERT A. SCHMlARTZ
NEW YORK UNIVERSITY SALOMON CENTER
Director INW WALTER
Leading members of the financial community, both knowledgeable and supportive
of the activities and publications of the New York University Salomon Center,
work with us to further the interaction of financial executives and research faculty,
and to ensure stability of funding for research at the Center.
Bankers Trust Company
Brinson Partners, Inc.
Canadian Imperial Bank of Commerce
Capital Markets Assurance Corporation
Chemical Bank
Chicago Board Options Exchange. Inc.
Citicorp
Continental Asset Management
cornerstone Research, Inc.
Deutsche Bank North America
Duff & Phelps, Inc.
The First National Bank of Chicago
Fitch Investors Service. Inc.
General Motors Corporation
Global Advanced Technology Corporation
Goldman, Sachs & Co.
The Guardian Life Insurance Company
Heine Securities Corporation
Henry Kaufman & Co., Inc.
.ING Bank, N.V.
Investment Technology Group
Investors Research Corporation
KPMG Peat Marwick
Loan Pricing Corporation
Merck & Co., Inc.
Memll Lynch & Co.. Inc.
Metropolitan Life Insurance Company
Milliman & Robertson, Inc.
Moody's Investors Service
Morgan Guaranty Trust Company of
New York
Morgan Stanley & Co., Inc.
Municipal Bond Investors Assurance
Corporation
National Association of Securities Dealers,
Inc.
National Economic Research Association
New York Stock Exchange, Inc.
Nomura Research Institute (America) Inc.
Pfizer, Inc.
Price Waterhouse
Prudential Foundation
Reuters America Inc.
Salomon Brothers Inc
Sanford C. Bernstein & Co.
Standard & Poor's Corporation
TIAA-CREF
Ti g h a s t
LEONARD N. STERN SCHOOL OF BUSINESS
I Dean: George G. Daly
NEW YORK UNIVER!XC'Y
Resident: L. Jay O l h
FINANCIAL MARKETS, INSTITUTIONS & INSTRUMENTS (ISSN 0963-8008)
is published five times a year in February, May, August, November, and December by
Blackwell Publishers with offices at 238 Main Street, Cambridge, MA 02142, USA, and
108 Cowley Road, Oxford OX4 IJF, UK.
INFORMATION FOR SUBSCRIBERS New orders, renewals, sample copy requests.
claims, change-of-address information and all other correspondence should be sent to the
Subscriber Services Coordinator at the publisher's Cambridge office.
SUBSCRIPTION RATES FOR VOLUME 4,1995*
N America Rest of the World
Institutions $115.50 $135.00
Individuals $61.00 $73.00
Single issue rates:
Institutions $26.00 $31.00
Individuals $15.00 $18.00
'Checks should be made payable to Blackwell Publishers; Canadian residents please add
7% GST.
BACK ISSUES Back and single issues of the journal from Volume 1 onwards are avail-
able from the publisher's Cambridge office at the current single-issue rate. For back issues
of The Monograph Series in Finance and Economics, please see information at the back of
the journal.
MICROFORM The journal is available on microfilm. For microfilm service address
inquiries directly to University MicrofiIms Inc., 300 North Zeeb Road, Ann Arbor, MI
48106-1346, USA.
MAILING The journal is mailed via secondclass in N. America and by Virgin Mailing
and Distribution to the rest of the world.
POSTMASTER Second-class postage paid at Boston, MA, and additional offices. Send
'all address corrections to Journals Subscription Department. 238 Main Street, Cambridge,
MA 02142.
ADVERTISING For details please contact the Journals Marketing Manager at the pub-
lisher's Cambridge office.
COPYRIGET All rights reserved. Reproduction or translation of any part of this work
beyond that permitted by Section 107 and 108 of the US Copyright Law without permission
of the publishers is unlawful. Authorization to photocopy items for internal or personal
use, or the internal or personal use of specific clients, is granted by New York University
Salomon Center, provided that the base fee of US $8.00 per copy, plus $15 per page is
paid directly to Copyright Clearance Center (CCC), 222 Rosewood Drive, Danvers, MA
01923, USA. Rates for educational photocopying for classroom use are S.15 per page per
copy also payable directly to CCC or the National Association of College Stores. For those
organizations that have been granted a photocopy license by CCC, a separate system of
payment has been arranged. The fee code for users of the Transactional Reporting Service
is: 0963-8008194 $8.00 + $.15.
For all other permission requests or inquiries, please contact the Journals Permissions
Manager at the publisher.
'Qpeset by Archetype, Monticello, IL. Printed and bound by Edwards Brothers, Ann Arbor,
MI.
'@ 1995 New York University Salomon Center
Equity Trading Practices and Market Structure:
Assessing Asset Managers' Demand for Immediacy
CONTENTS
II. MARKET STRUCTURE AND THE DEMAND FOR IMMEDI-
. . . . . . . . . . . . ACY A REVIEW OF THE ISSUES
IV. TRADING PRACTICES, ORDER SIZE, AND TRANSACTION
COSTS . . . . . . . . . . . . . . . . . . . . . . . . .
V. THE USE OF A L T E R N ~ ELECTRONIC WING SYS-
TEMS . . . . . . . . . . . . . . . . . . . . . . . . .
VI. SAMPLE OF RESPONDENTS AND THEIR REASONS FOR
TRADING . . . . . . . . . . . . . . . . . . . . . . .
IX. REFERENCES . . . . . . . . . . . . . . . . . . . . .
. XI. N m ON CONTRIBUTORS/ACKNOWLEDGMENTS . . .
Equity Trading Practices and Market Structure:
Assessing Asset Managers' Demand for Immediacy
BY NICHOLAS ECONOMIDES AND ROBERT A. SCHWARTZ
This paper summarizes the responses to a questionnaire sent to equity traders through
TraderForum of the Institutional Investor. The respondents manage in total avery significant
percentage of equity assets under management-in the ~nitedStates. The focus of the
questions was the extent of the demand for immediate execution of orders. We found that
the majority of traders an willing to trade patiently if this reduces execution costs. Many
traders indicate that they frequently delay mdes to obtain better prices. Most respondents
indicate that they are typically given more than a day to implement a large order. that they
typically break up more than 20% of their large orders for execution over time. and that
they regularly take more than a day for a large order that has been broken into lots to be
executed completely. There is a generally positive view of alternative electronic trading
systems, such as Instinet and Investment Technology Group's POSIT. The key motives for
trading on these systems arc reduced market impact, lower spreads. better liquidity, and
anonymity. The respondents indicate that the key changes that would make alternative
elecmnic systems more attractive are an increase in execution rates and more convenient
times of trading. The responses to the survey also show that alternative electronic systems
would be wd more if the traders did not have soft dollar arrangements.
I. INTRODUCTION
Practitioners and students of the securities markets widely assume that traders
demand immediate execution of their orders. Indeed, a major function of tradi-
tional brokerldealer firms is to provide the services that result in trades being made
huickly. In volatile markets, an advantage of trading quickly is that opportunity
costs (i.e., the risk of an asset's price "getting away" before a portfolio decision
is implemented) are reduced. However, higher direct costs (i.e., market impact,
bid-ask spreads, commissions, and other transaction costs) are generally incurred
when fast executions are obtained. Little information exists about the relative
importance professional asset managers attach to these two types of costs, and
about the tradeoffs they are willing to make between them. The current survey is
motivated by this lack of information.
The results show that experienced participants often do not trade with maximum
possible speed so as to "nail down" a price, and that they do commonly work their
orders patiently over time. However, the very dynamics of the continuous market
appear to induce a demand to trade quickly. Based on the survey responses, we
conclude that traders would be even more willing to forgo immediacy of execution
if, by so doing, their direct costs of transacting could be further reduced.
Consequently, the survey findings have a major implication for market struc-
ture. Asset managers should be given the opportunity to delay their orders until
predetermined points in time at which they may trade with each other at reduced
Finandal Markets, Institutions & Insmunenu. V. 4. N. 4. November 1995. @ 1995 New Yo& University Salomon
Center. Published by Blackwell Publishers. 238 Main Saeet. Cambridge, MA 02142, USA and 108 Cowley R o d
Oxford. OX4 IJF. UK
2 Nicholas Economides and Robert A. Schwartz
trading cost. The incorporation of an electronic call market would provide the
requisite environment. A call market is an environment that enables buyers and
sellers to meet at pre-determined points in time. We have elsewhere considered
the desirability of holding an electronic call three times a day, along with contin-
uous trading.' The call environment would provide a useful pricing device for
the broad market, while resulting in lower transaction costs (bid-ask spread and
market impact) for individual participants.
In recent years, increasing numbers of institutional investors are breaking out
of traditional molds to explore various proprietary trading systems (PTS). With
the exception of Instinet's continuous market, the PTSs are crossing networks
(e.g., Instinet's after hours cross and Investment Technology Group Inc.'s POSIT
system) and call markets that are capable of independent price discovery (e.g.,
the Arizona Stock Exchange's AZX system). Nevertheless, immediacy continues
to be a major service provided by market centers such as the New York Stock
Exchange (NYSE) and Nasdaq. These market centers operate on the assumption
that participants want instant access to the market, and that they are willing to pay
the price for trading with immediacy.
However, little empirical evidence exists on asset managers' demand for im-
mediacy. To assess this demand, 825 questionnaires were mailed to traders of
managed equity funds, and 150 responses were received. These respondents r ep
resent approximately $1.5 trillion in equity assets under management. In broad
sweep, the responses to the survey suggest that buy-side participants do trade
patiently in an attempt to control execution costs. The key results include:
Two-thirds of the respondents indicated that they are willing to trade pa-
tiently to reduce execution costs (Table 2).
Nearly half say they frequently do delay trades in an effort to obtain better
prices (Table 3).
One-third would regularly or frequently accept a trading delay of one hour
for a $50 stock if they could save 256 per share in trading costs (Table 4).
Nearly a quarter would regularly or frequently accept a trading delay of
three hours for a $50 stock if they could save 259 per share in trading costs
(Table 5).
About one in five would regularly or frequently accept a trading delay of
one hour for a $50 stock if they could gain anonymity on a trade of 10,000
shares or more (Table 6).
Nearly two-thirds regularly or frequently use limit orders (Table 8).
One-third report that 20% or more of their orders for a stock are larger than
the stock's average daily trading volume (Table 15).
More than two-thirds typically give more than one day to implement a large
order for a small cap stock (Table 16).
'See. Economides and Schwartz (1995).
Equio Trading Practices and Marker Structure 3
More than half typically give more than one day to implement a large order
for a large cap stock (Table 16).
Approximately three out of five break up at least 20% of their orders for
100,000 shares or more for execution over a series of trades (Table 17).
Close to half report that they regularly or frequently take more than one
day to completely execute a large order broken into lots (Table 18).
The picture that emerges is that immediacy is not commonly demanded by buy-
side participants, and that executions for large orders are generally not realized
within brief periods of time (a few hours or less). Respondents appear to be less
concerned about trading quickly than about controlling execution costs, the loss
of anonymity, and the information leakage that occurs when an intermediary is
contacted. Understanding this is essential for making proper decisions with regard
to the structure and regulation of our security markets.
We do not claim that immediacy is never demanded. For specific institutions and
specific situations, the advantages of rapid trading may indeed outweigh the costs
involved. Our objective, however, is not to assess the intensity with which most
asset managers, or even the representative asset manager, demands immediacy.
Rather, we wish to determine whether or not a meaningful number do handle a
substantial proportion of their orders patiently because immediacy is costly.
The paper is organized as follows. Section I1 discusses the relationship between
market structure and the demand for immediacy. In this section we also present
our reasons for believing that immediacy is not universally demanded. Section ID
-discusses the respondents willingness to accept trading delays. In section N, we
present findings regarding trading practices, order size, and transaction costs. In
section V, we discuss attitudes towards the use of alternative electronic trading
systems. Section VI discusses the sample of respondents and their reasons for
trading. Section W discusses the differences between active and passive traders
in their responses to the questionnaire. Section VIII contains our concluding
remarks. The Appendix presents the distributed questionnaire.
II. MARKET STRUCTURE AND THE DEMAND FOR IMMEDIACY:
A REVIEW OF THE ISSUES
An understanding of participants demand for immediacy is key to designing the
trading structure of a securities market. In this sect i o~ we review alternative market
structures, consider the relationship between market structure and the demand for
immediacy, and briefly review the literature. A major choice in trading design is
between a continuous market and a call market A continuous market is open for
an extended span of time; e.g., at the New York Stock Exchange trading begins at
9:30 and continues until the 4 p.m. close. During this period, trades are made any
-time two contra-side orders cross in price. The continuous market can be a dealer
market (quote-driven) or an agencylauction market (order-driven). Nasdaq in the
U.S. and SEAQ in the U.K. are dealer markets. Examples of the agencylauction
4 Nicholas Economides and Robert A. Schwartz
market include the New York Stock Exchange, the Paris Bourse, the Toronto Stock
Exchange, and the Tokyo Stock Exchange?
The continuous market has been widely studied by academic researchers. Much
of the microstructure literature has focused on the dealer market. Early analyses
include Garman (1976), HO and Stoll(1980), Amihud and Mendelsohn (1980),
and Mildenstein and Schleef (1983): An analysis of the agencylauction market is
provided in Cohen, Maier, Schwartz, and Whitcomb (1986) and Schwartz (1$1).~
In contrast with a continuous market, orders are batched in a call market for
simultaneous, multilateral execution at a single price, the value that maximizes
the number of shares that trade at the call. Examples of call markets include the
opening procedure on most electronic exchanges (e.g., Toronto's CATS, Paris's
CAC, and Tokyo's CORES), and on non-electronic exchanges such as the NYSE.
Pure electronic call markets include the Tel Aviv Stock Exchange, the Bolsa Mexi-
cans's Intermediate Market, the Arizona Stock Exchange, and the Paris Borse (for
\e;ss Qpu;\d issues), Preuious\y, n&ecton'~c cd\s ex'lsted in%\ Aviu, Pasis, and
roughly 100 years ago at the NYSE. Call markets have received significantly less
'attention than continuous markets in the academic literature. Studies of the call
market include Cohen and Schwartz (1989), Economides and Schwartz (1995),
Schwartz (1996), and Amihud and Mendelson (1985).
In comparison with continuous trading, the call market has distinct advantages
as a trading environment. These include enhanced price discovery, el i nat i on of
the bid-ask spread, reduced market impact of large orders, superior handling of
limit orders and, in general, easier order handling and better market surveillance?
An often-noted disadvantage of call market trading is that it does not provide
immediate access to the market over an extended period of time.6 However, this
i s not a problem if call market trading is integrated with continuous trading?
Nevertheless, if multiple calls are held during a trading day, one might question
whether or not a sufficient number of participants will postpone their orders so that
the intraday calls may be viable. Traders will postpone orders or not depending
on their demand for immediacy and on the price of immediacy in a continuous
market.
Clearly the selI side of the market has a vested interest in supplying immediacy.
It is difficult in a continuous market for ultimate buyers and sellers to find each
2~ontinuous markets may also include a dealer, such as the specialist on the New York Stock
Exchange.
'For further discussion and references, see Schwartz (1991).
'Furthu refmncu are provided in both of these books.
%r funher discussion of call markets, see Economides and Schwarrz (1995). The role of liquidity
in financial exchange is discussed in Economides (1993). (1995). For an analysis of positive size
effects of financial and other networks see Economides (1996).
6Econornides and Siow (1988) and Garbade and Siber (1976). (1979) analyze the cost of longer
waiting until execution and balance it with the benefit of participating in a more liquid market.
. 'For further discussion of the integration of call and continuous trading see Handa and Schwartz
(1996). Economides and Heisler (1995).
Equity Trading Practices and Market Structure
5
other quickly without the services of brokerldealers. However, when a meeting
point in time is pre-specified (i.e., the time of a call), buyers and sellers can more
easily find each other without the services of intermediaries. The key question is,
"are buy-side traders willing to wait?"
We anticipate that an appreciable number of them will answer "yes". Cer-
tainly, the pace with which trading progresses in a continuous market is not in
harmony with the pace with which the underlying investment decisions are com-
monly made.8 Institutional decision making with respect to fundamental infor-
mation takes time. Investment decisions commonIy involve information gathering
and analysis, and the entire process can take place over a period of several days.
But once a decision has been made, an order is typically brought to a continuous
trading environment that accentuates the importance of minutes and even seconds.
T i e is suddenly of the essence. Is it likely that the value of a decision made
over a period of a day or more can decay within the span of an hour or less?
Or, is the demand for immediacy generated endogenously by the dynamics of the
continuous market? Certainly part of the demand for immediacy comes from the
price dynamics of the continuous market Rapid trading may be motivated by
knowledge of the order flow and by charting signals.
Regarding the fundamental determinants of share value, we distinguish two
types of information release: natural (e.g., an earthquake or fire) and managed
(e-g., an unemployment or earnings report). The introduction of a call market
would enable the pace at which managed information is released and portfolio
decisions are made to be better harmonized with the pace at which trading is
pursued. That is, both news releases and institutional investor portfolio decisions
could be timed with reference to the schedule of the calls.
The conventional wisdom is that immediacy is provided by acontinuous market.
On the contrary, the continuous market may actually make it more difficult for
institutional investors to execute large orders at reasonable cost by the end of a
-trading day. Data collected by the Plexus Group indicate that roughly 67% of the
orders given to buy-side trading desks are for more than half of the stocks' average
daily trading volumes, and 40% of the orders are for more than the total average
daily trading volume for the stocks? Orders of this size cannot be traded quickly
in the continuous market at acceptable levels of cost. The reality is that immediacy
is not always obtained in a continuous market.
It is also conventionally believed that intermediaries provide buy-side partic-
ipants with anonymity vis-a-vis each other. And they do. However, buy-side
participants are increasingly concerned about the loss of anonymity to sell-side
t~rokerldealers. It is also becoming apparent that anonymity can be provided by
an electronic trading system, and it certainly is characteristic of call market trad-
8 ~ h e pace of trading in a continuous market has accelerated with the application of information
technology.
9 ~e e Wagner and Edwards (1993).
6 Nicholas Economides and Robert A. Schwartz
ing. We expect that buy-side participants will be willing to forsake immediacy for
disintermediation and anonymity.
All things considered, picture an institutional investor who makes a portfolio
decision at 2 p.m. when a market call is scheduled for the 4 p.m. close. The
investor could avoid paying the price of intermediation and immediacy by waiting
two hours and trading at lower cost at the 4 p.m. close. By waiting, he or she
has effectively unbundled the act of "trading" from the "immediacy" of the trade.
If enough participants do this, they will naturally meet without the assistance of
intermediaries, and the intra-day calls will be viable.
The archetypal role of a dealer is to provide the liquidity that enables investors to
trade with immediacy. "Immediacy," however, is a vague concept. For retail-sized
orders, it could mean the ability to trade within a few minutes. Large institutional
orders, however, would incur unacceptably large execution costs (bid-ask spread
plus market impact) if executed so quickly. An asset manager seeking to buy
100,000 shares of a stock that on average trades 200,000 shares a day, might
consider an execution obtained within an hour or even a day to be immediate. This
section of the paper contains our findings with regard to various issues concerning
the patience with which a respondent is willing to seek a trade.
The first issue we address concerns the meaning of immediacy itself: how
quickly must a trade be made to be considered immediate? Respondents were also
asked what they would be willing to pay for immediacy, and how frequently they
do in fact delay a trade in an attempt to obtain a better price. Respondents were also
asked about their willingness to accept a trading delay to reduce their trading costs
or to gain anonymity. The extent to which index options andtor futures are used
so that shares may be traded more patiently in the cash market, and the frequency
with which limit orders are used also are reported here.
With regard to the meaning of immediacy, we asked respondents if they would
consider a trade to be immediate if it executed within a stated period of time
(Table 1). The majority (71%) answered that a trade must be realized in under 10
minutes to be considered immediate. Only 3% answered %thin 2 hours," and
6% said "within one day."
Having established a sense of what the respondents consider "immediacy" to be,
the questionnaire asked about the respondents' willingness to trade patiently if
their execution costs could be reduced by doing so ('Table 2). A total of 67%
Equity Trading Practices and Market Structure 7
Table 1: Time In Which You Consider A Trade To Be Immediate
Table 2: Willingness To Trade Patiently To Reduce Execution Costs
Under 10 minutes
1 Hour
2 Hours
1 Day
Other
No Answer
indicated that they would be willing or very willing to delay a trade if it reduced
their costs. Only 8% said they would not be willing.
Number of Respondents
107
1
4
9
7
1
5 (Very Willing)
4
3
2
1 (Not at All Willing)
No Answer
Willingness is one thing; the perception of how frequently a trade is delayed is
another. Therefore, the questionnaire asked how frequently traders in fact delay a
trade in an attempt to obtain a price that is more favorable than the price currently
prevailing on the market (TabIe 3). The vast majority (77%) of respondents said
they delay trades in hopes of finding a better price for 25-75% of their trades.
Only 11% said they "never" or "rarely" delay a trade for a better price. The
preponderance of the respondents perceive it desirable to trade patiently.
Percentage of Respondents
71.3
3
2.7
6.0
4.7
0.7
WILLINGNESS TO ACCEFT A TRADING DELAY OF ONE HOUR FOR A $50 STOCK
-IF You COULD SAVE 254 PER SHARE IN TRADING COSTS
Number of
Respondents
5 1
50
34
6
6
3
Evidence on the demand for immediacy was also obtained by asking respondents
whether they would be willing to accept a trading delay of one hour if, by so doing,
Percentage of
Respondents
34.0
33.3
22.7
4.0
4.0
2.0
8 Nicholas Economides and Roben A. Schwartz
Table 3: The Frequency With Which A Trade Is Delayed To Obtain A Price More
Favorable Than The Current Market Price
Table 4: Willingness to Accept A Trading Delay Of One Hour For A $50 Stock If
You Could Save 25# Per Share In Trading Costs
they could decrease trading costs by 25# a share for a $50 stock (Table 4). One
in four respondents said they would "rarely" or "never" delay a trade for an hour
t o reduce costs. On the other hand, more than half said they would be willing to
delay a trade to reduce costs on some or all of their trades.
Equity Trading Practices and Market Structure
9
Table 5: Willingness to Accept A Trading Delay Of 3 Hours For A $50 Stock If
You Could Save 25$ Per Share In Trading Costs
WILLINGNESS TO ACCEPT A TRADING DELAY OF 3 HOURS FOR A $50 STOCK
You COULD SAVE 2% PER SHARE IN TRADING COSTS
Traders were then asked if they would accept a delay of three hours for the same
-cost savings. One in three respondents said they would rarely, if ever, delay a
trade three hours. On the other hand, an appreciable subset (23%) said they would
accept a three-hour delay regularly or frequently to save 25 cents per share for a
$50 stock.
Jtespondents were also asked if they would delay a trade of 10,000 shares or more
for one hour if, by so doing, they could gain anonymity (Table 6). Slightly less
than half of the respondents said they would "rarely" or "never" delay a trade
for an hour to gain anonymity. But 19% said they would do so "regularly" or
"frequently." This indicates that an appreciable subset of participants commonly
do not seek to trade immediately to preserve anonymity.
How FREQUENTLY YOU WAIT MORE THAN ONE DAY BEFORE ACQUIRING OR
SELLING SHARES IN THE CASH MARKET IF YOU HAVE USED B ~ E X O ~ O N S
AND/OR FUTURES TO TRADE QUICKLY
One way to delay trading in the cash market is to trade a derivative contract to
establish a position that is then converted into shares over time. Table 7 shows the
responses from those that indicated that they use index options andlor futures to
reduce their need to execute trades quickly in the cash market. These respondents
were asked the frequency with which they would wait more than a day before
10 Nicholas Economides and Robert A. Schwartz
Table 6: Willingness To Accept A One Hour Trading Delay To Gain Anonymity
On A Trade Of 10,000 Shares Or More
Table 7: How Frequently You Wait More Than One Day Before Acquiring Or
Selling Shares In The Cash Market If You Have Used Index Options And/or
Futures To Trade Quickly
acquiring or selling the desired shares in the cash market. A total of 46% said
they rarely or never wait. On the other hand, a significant number (27%) said they
"ffequently" waited.
Never
Rarely (1-2496 Trades)
Sometimes ( 2549% of Trades)
Regularly (50-74% of Trades)
Frequently (75-100% of Trades)
No Answer
How O m DO YOU USE LIMIT ORDERS, MARKET ORDERS, AND MORE
In Table 8, we report on the frequency of the use of limit orders, market orders,
percentage orders, and basket or&rs. As expected, market, limit, and not held
orders are all widely used, and in roughly similar amounts. Basket orders and
'index optiondfutures are not used very much.
Perhaps the most interesting finding is the extent to which l i t orders are
used: 52% of the respondents said they used them on at least half of their trades.
The use of limit orders is essential to an order driven market. The dynamics of
price behavior apparently compensates traders sufficiently for placing limit orders.
Number of
Respondents
12
3
4
2
9
3
Percentage of
Respondents
36.4
9.1
12.1
6.1
27.3
9.1
Equity Trading Practices and Market Structure 11
Traders are explicitly not demanding or paying for immediacy when they use limit
orders in seeking to trade.
IV. TRADING PRACTICES, ORDER SIZE, AND TRANSACTION COSTS
Trading practices, order size, and costs shed further light on the willingness of
the respondents to trade patiently. The first question raised is the importance of
the major costs of these three categories: the opportunity cost of missing a price,
market impact, and commissions. Reasons to execute a trade quickly include the
volatility of prices, the possible mispricing of stocks, and the prevention of front
running. The costs associated with limit orders (e.g., the risk of non-execution
and the difficulty of withdrawing limit orders quickly) also impact the decision
to trade quickly. Respondents were further questioned as to their concern about
information leakage when a broker is called, and about the frequency with which
they demand capital from a broker for a block order. Attitudes toward costs are
also reflected in the respondents' answer to one other question, "How frequently
do you decide not to adjust your portfolio because the market is too illiquid?"
Concerning the size of their orders, respondents were asked about the frequency
with which an order for a stock is larger than the stock's average daily trading
volume, the time commonly given by portfolio managers to implement large orders,
and the frequency with which large orders are broken up for execution over time.
They were further questioned about the frequency with which it takes more than
one day for a large order broken into lots to execute completely. Lastly, the
questionnaire asked about the times of the day when the traders most prefer and
least prefer to place their orders.
How IMPORTANT ARE THE FOLLOWING COSTS?
Regarding the costs of trading, our findings on the importance of three major
components (the opportunity costs of missing a price, market impact, and com-
missions) are summarized in Table 9. The opportunity costs of missing a price are
rated the most important cost by 55% of traders, followed by market impact, which
i s rated the most important cost by 41% of traders. Commissions are important to
only 3% of the respondents.
Table 10 summarizes the most important and second most important factors that
may cause traders to want to execute a trade quickly. The most important factor
is "because prices are volatile and the risk of waiting is too greats' 48% of re-
spondents said this was the most important factor, and 32% said it was the second
most important factor. Fewer indicated that the prevention of front-running was a
12 Nicholas Economides and Robert A. Schwartz
Equity Trading Practices and Market Structure
13
Table 9: How Important Are The Following costs?
factor: 11% indicated it was the most important factor, and another 26% indicated
it was the second most important factor. Interestingly, only 23% said that the most
important factor was that other traders will realize that the stock is overpriced or
underpriced, and 21% indicated this was the second most important factor. To the
extent that trading is motivated by news and not just the assessment of existing
information, this number would be expected to be higher.
Opportunity Costs
of Missing A Price
Market Impact
Commissions
WHAT DO YOU CONSIDER THE MOST IMPORTANT AND THE SECOND MOST IM-
PORTANT DRAWBACKS OF USING LIMIT ORDERS?
When asked what they consider the most important drawback of using limit orders
(Table 1 I), most of the respondents (70%) stated that the most important factor is
the risk of non-execution. An additional 22% checked a closely related factor: limit
orders may cause you to miss a favorable market movement. Only 4% indicated
that the drawback is that the limit orders may be difficult to withdraw quickly. This
i s not surprising, given that the professional buy-side traders keep current about
market events, and that order handling procedures are rapid. The response here
is consistent with the previously discussed finding that price volatility is the most
important motivation for trading quickly (see Table 10).
Percentage of Respondents
Indicating Particular Cost as
A willingness to delay a trade on the part of roughly a third of the respondents
.in order to achieve anonymity is evidenced by the responses reported in Table 6.
Anonymity may be valued by buy-side participants because of the adverse price
impact that can occur when news gets out that they are seeking to trade. To assess
this, respondents were asked how concerned they are about information leakage
after they have called a broker to make a trade. The results are reported in Table 12.
A total of 45% indicated they were concerned about information leakage.
Most
Important
54.7
40.7
3.3
Neutral
Neutral
36.0
51.3
9.3
Least
Important
6.0
6.7
84.7
No Answer
3.3
1.3
2.7
14 Nicholas Economides and Robert A. Schwartz
Table 10: What Are The Most Important And Second Most Important Factors
That May Cause You To Want To Execute A Trade Quickly?
I I SecondMost I
and the Risk of Waiting is Too I I I
Because Prices Are Volatile
- - - - . . - - - . - -
Most Important Factor I Important Factor
72 I 48
Traders Will Realize the Stock I I I
Great
Becuase you Think Other
48.0%
35
is Overpriced or Underpriced
No Answer
0.7% 5.3%
32.0%
32
Opportunity Costs
To Prevent Other Traders
From Front-Running Your
Order
ONote that columns in this table add to more than 100% becuase some repondents have checked
more than one category.
23.3%
34
%o primary functions of intermediaries are (i) to help a customer find the coun-
terpart to a trade (i.e., act as a broker), and (ii) to provide capital as the counterpart
in a trade (i.e., act as a dealer). For a customer who is concerned about infor-
mation leakage, a strong motive must exist for contacting the intermediary in the
first place. Accordingly, the survey asked about the frequency with which the
respondents demand capital from their brokers for transactions of 10,000 shares
or more. The results are reported in Table 13. Approximately three out of four
respondents said they rareIy, if ever, demand capital fiom their brokers. Only 7%
said they regularly or frequently demand broker capital. Presumably this means
that the role of intermediaries in finding the other side of a trade is more important
than their role in providing capital.
21.3%
47
11.3%
17
11.3%
26.0%
39
26.0%
Equity Trading Practices and Market Strucrure 15
Table 11: What Do You Consider The Most Important And The Second Most
Important Drawbacks Of Using Limit Orders?
'Note that columns add to more than 100% because some respondents have checked more than
one category.
-
Table 12: Concern About Information Leakage When A Broker Is CaIIed
How FREQUE~ Y DO YOU DECIDE NOT TO ADJUST YOUR P o r n o m BECAUSE
THE hhiWET IS TOOUQUI D?
5 (Very Concerned)
4
3 (Neutral)
2
1 (Not Concerned)
No Answer
A total of 16% of the traders do not adjust their portfolio 10-19% of the time
because the market is too illiquid (Table 14). Almost twice as many do not adjust
a e i r portfolio for the same reason 1-9% of the time. In both cases, active traders
are more likely not to adjust their portfolios than passive traders.
Number of Respondents
46
22
47
15
18
2
Percentage of Respondents
30.7
14.7
31.3
10.0
12.0
1.3
16 Nicholas Economides and Robert A. Schwartz
Table 13: Frequency With Which Capital Is Demanded From A Broker For Trans-
actions of 10,000 Shares or More
Table 14: How Frequently Do You Decide Not To Adjust Your Portfolio Because
The Market Is Too Illiquid?
FREQUENCY WITH WHICH YOUR ORDER FOR A STOCK IS LARGER THAN THE
STOCK'S AVERAGE DAILY TRADING VOLUME
If the order is large relative to average daily trading volume, it may not be possible
10 execute the order entirely in a very short period of time without incurring an
unacceptably high execution cost. As noted above, Wayne Wagner and Mark
Edwards found that 66% of the orders in Plexus Group's data set exceed half of
the stock's average daily trading volume, and that 40% of the orders exceed the
stock's total average daily trading volume.
The respondents were asked the frequency withwhich their orders for a stock
are larger than the stock's average daily trading volume (Table 15). Almost a third
of the respondents answered that 20% or more of their orders are this large.
.
Never
27
22.7%
9
37.5%
37
24.7%
Active Traders
Passive Traders
All Respondents
1-9%
40
33.6%
6
25.0%
47
31.3%
10-19%
17
14.3%
1
4.2%
24
16.0%
20% or
more
5
4.2%
1
4.2%
6
4.0%
Don't
Know
24
20.2%
7
29.2%
34
22.7%
No
Answer
6
5.0%
0
0.0%
8
5.3%
Equity Trading Practices and Market Structure 17
Table 15: Frequency With Which Your Order For A Stock Is Larger Than The
Stock's Average Daily Trading Volume
In light of the size of institutional orders relative to average daily trading volume,
the questionnaire asked about the time a portfolio manager might typically give
a trader to implement an order (Table 16). For small cap stocks, less than 1%
answered "one hour or less:' and 69% answered "one day" or longer. For large
cap stocks, 5% answered "one hour or less," and 59% answered "one day" or
ionger. This finding reinforces the impression that asset managers do not attempt
to implement their trading decisions within brief intervals of time.
Never
1-9% of Orders
10-19% of Orders
7
20% or More of Orders
Don't Knowmot Sure
No Answer
FREQUENCY WITH WHICH LARGE ORDERS (100,000 SHARES OR MORE) ARE
BROKEN INTO SMALLERLOTS FOR SEPARATE EXECUTIONS OVER AN EXTENDED
PERIOD OF TIME
If an order is given time to be executed, it might be broken up for execution in
smaller pieces over a series of trades. The questionnaire asked the frequency with
which large orders of 100,000 shares or more are in fact broken up for this purpose
(Table 17). More than three out of five respondents indicated that 20% or more of
their orders are broken into smaller lots. Only 5% indicated that they never break
up their large orders.
Number of
Respondents
5
5 1
33
49
8
4
FREQUENCY WITH WHICH IT TAKES MORE THAN ONE DAY FOR A LARGE ORDER
B ~ o m INTO LOTS TO BE EXECUTED COMPLETELY
Percentage of
Respondents
3.3
34.0
22.0
32.7
5.3
2.7
The length of time typically taken to implement an investment decision in the
marketplace more directly reveals a willingness to trade patiently. The survey
questioned the frequency with which more than one day is taken to execute an
order completely when the order is broken into smaller lots to be executed over
'1 8 Nicholas Economides and Robert A. Schwartz
Table 16: Time Qpically Given By Portfolio Manager To Trader To Implement
A Large Order (25% Of Average Daily Trading Value Or More)
Other 5 3.3 5 3.3
No Answer 10 6.6 8 5.3
Table 17: Frequency With Which Large Orders (100,000 Shares Or More) Are
Broken Into Smaller Lots For Separate Executions Over An Extended Period Of
Time.
time (Table 18). While only 3% answered "never," 44% said that they frequently
or regularly broke their orders into smaller lots for execution over time.
Never
1-9% of Orders
10-19% of Orders
20% or More of Orders
Don't Know/Not Sure
No Answer
WHEN DO YOU PREFER TO PLACE YOUR ORDWS?
Turning to the question of when orders are placed, the respondents expressed
clear preferences for trading at different times during the day mbl e 19). Traders
Number of
Respondents
8
19
2 1
93
5
4
Percentage of
Respondents
5.3
12.7
14.0
62.0
3.3
2.7
Equity Trading Practices and Market Structure 19
Table 18: Frequency With Which It Takes More Than One Day For A Large Order
Broken Into Lots To Be Executed Completely
preferred the half hourjust following market opening to the actual market opening:
44% said that 9:31-10:00 am. was their most preferred time to place an order,
compared to 27% who most preferred the actual market opening. Traders also
preferred the half-hour period immediately prior to market close as compared to
the actual closing time: 23% said the 3:31-3:59 period was "most preferred,"
compared to 8% who most preferred the actual closing time to place their orders.
The survey did not ask for the reasons behind these preferences. Presumably, the
uncertainty concerning price determination at the open lead many to prefer the
9:31-10:00 a.m. period; and the uncertainty concerning price, and perhaps the
ability to trade at all, caused many of them to find the close least preferable, and
the 3:31-359 period less preferable than the 9:31-10:OO period.
Recognizing that the periods are not of equal length, one might expect from the
responses that the pattern of trading over the day would be "U" shaped, as indeed
it has been observed to be by, for instance, McInish and Wood (1990).1° The
questionnaire did not ask, however, the frequency with which orders were delayed
so that their arrival might be harmonized with the time of the day the respondent
felt to be most desirable.
V. THE USE OF ALTERNATIVE ELECTRONIC TRADING SYSTEMS
.The emergence of alternative electronic markets in recent years has given buy-
side traders new opportunities to receive timely executions at reasonable cost.
Respondents were asked about the frequency with which they use these systems
(e.g., NYSE after hours Sessions 1 and 2, Instinet's crossing session and continuous
market, POSITS crossing sessions and AZX's call market), and their motives
for using them (e.g., lower trading costs, the ability to trade anonymously). The
'Osee McInish and Wood (1990).
20 Nicholas Economides and Robert A. Schwartz
Table 19: When Do You Prefer To Place Your Orders?
respondents also were asked whether or not they felt the benefits of electronic trade
execution outweigh the disadvantages, how satisfied they are with the alternative
systems, and what would get them to use the alternative systems more (e.g., if they
gave higher execution rates, if they allowed trading at more convenient times, and
if the respondents' did not have soft dollar arrangements).
Table 20 shows that use of the alternative systems is limited. Use of these systems
is similar for Listed and for NASD stocks, except for Instinet's continuous market
which is used more for NASD stocks.
M m FOR TRADING ON THE ELECTRONIC SYSTEMS
Of particular interest are the respondents' motives for trading on electronic systems
(Table 21). The considerations that were rated "important" are: reduced market
impact (47%), lower bid-ask spreads (47%), better liquidity (41 %), lower general
transaction cost (39%), the ability to trade anonymously (38%). and the ability to
have greater control of the negotiation process (33%).
WHAT Emm DOES THE ANONYMITY OFFERED BY ELECTRONIC TRADING SYS-
. m s HAVE ON YOUR F~CU?ION ABILITY?
Nearly half of the traders expressed the opinion that the anonymity offered by the
alternative electronic trading systems improves their execution ability (Table 22).
Less than 1% think that it worsens it. On the other hand, 34.0% of the respondents
said they did not know what the effect would be or they did not answer the question.
Table 20: How Often Do You Use The Following Alternative Electronic Trading
Systems?
'The indication NA stands for "No Answer".
Listed Stocks
NYSE Session 1
NYSE Session 2
Instinet Crossing
POSIT
AZX
Instinet Continuous
NASD Stocks
Instinet Crossing
POSIT
AZX
Instinet Continuous
Never
86.7
90.0
63.3
60.0
78.7
52.7
62.7
62.7
78.0
49.7
1-9%
6.7
4.7
22.7
22.7
13.3
28.0
16.7
20.7
12.0
14.7
10-19%
0.7
0
2.7
4.7
0
4.7
6.7
3.3
0
11.3
30% ormore
1.3
0
5.3
5.3
2.0
4.0
5.3
4.7
2.0
14.7
20-2996
0.7
0
2.0
3.3
1.3
6.7
4.7
3.3
1.3
6.0
Don't Know
1.3
2.0
0.7
1.3
0.7
0.7
0
0.7
0.7
0
NAa
2.7
3.3
3.3
2.7
4.0
3.3
4.0
4.7
6.0
4.0
22 Nicholas Economides and Robert A. Schware
Table 21: Motives For Trading On The Electronic Systems
Table 22: What Effect Does The Anonymity Offered By Electronic Trading Sys-
tems Have On Your Execution Ability?
Do You B m THAT THE BENEFITS OF ELECI-RONIC TRADE EXECUTION OUT-
WEIGH THE DISADVANTAGES OR THAT THE DISADVANTAGES OUTWEIGH THE
BENEFITS?
Improves it
Has no Effect
Worsens it
Don't Know
No Answer
Table 23 shows that a large majority of respondents (67%) believe that the benefits
of electronic trade execution outweigh the disadvantages. Only 16% of the traders
expressed the opinion that the disadvantages of electronic execution outweigh its
benefits.
Number of
Respondents
63
35
1
27
24
Percentage of
Respondents
42.0
23.3
0.7
18.0
16.0
Equity Trading Practices and Market Structure 23
Table 23: Do You Believe That The Benefits Of Electronic Trade Execution Out-
weigh The Disadvantages Or That The Disadvantages Outweigh The Benefits?
SATISFACTION OF USERS WITH THE DIFFERENT ALTERNATIVE WING SYSTEMS
Benefits Outweigh the
Disadvantages
Disadvantages Outweigh the
Benefits
No answer
Table 24 shows the degree of satisfaction with the alternative trading systems.
Among these, the least satisfaction was expressed for AZX (29%). It is not sur-
prising that 46% of AZX's customers are not satisfied: the AZX market is called
at 5 p.m., liquidity is insufficient, and execution rates are low because aggregate
-order flow is sparse. One would expect that satisfaction with these alternative
markets would be considerably greater if they were integrated better with the ma-
jor trading systems. We also note that there was widespread dissatisfaction with
NYSE's Sessions 1 and 2. However, a substantial majority (79%) were satisfied
with Instinet's continuous market.
WHAT WOULD GET YOU TO USE THE ALTERNATIVE TRADING SYSTEMS MORE
If clear motives exist for trading on electronic systems, why aren't the systems
used more heavily? ITG's POSIT, Instinet's crossing network, and the Arizona
StockExchange's AZX each batch orders for multilateral execution at a single time
at a single price; if institutional asset managers are willing to forgo immediacy,
why aren't these systems particularly attractive to them? The survey asked the
question, What would get you to use the alternative trading systems more?" The
results are in Table 25.
Not surprisingly, 55% said that they would use the alternative systems more if
-they gave higher execution rates. This is consistent with the reality that a lack of
order flow is a major impediment to the success of any trading system (and with
the adage, "order flow attracts order flow"). Second on the list, 35% indicated that
they would use the systems more if they did not have soft dollar arrangements (that
is, soft dollar arrangements appear to be an impediment to change). A total of 3 1 %
claimed they would use the alternative markets more if they allowed trading at more
Number of
Respondents
101
24
25
Percentage of
Respondents
67.3
16.0
16.7
24 Nicholas Economides and Robert A. Schwartz
Table 24: Satisfaction Of Users With The Following Alternative Trading Systems
Table 25: What Would Get You To Use The Alternative Trading Systems Morea
Instinet Continuous
"Note that columns add to more than 100% because some mpondena have checked more than
one category.
29.2%
56
78.9%
They Gave Higher Execution Rates
You Didn't Have Soft Dollar
Arrangements
They Allowed Trading at More
Convenient Times
You Knew More About Them
Other
None of the Above
convenient times (presumably during the day rather than after hours), and 20%
responded that they would use them more if they knew more about them (which
suggests some continuing lethargy on the part of some institutional investors).
25.0%
9
12.7%
Number of
Respondents
82
53
47
30
7
47
45.8%
6
8.5%
Percent of
Respondents
54.7
35.3
31.3
20.0
4.7
31.3
Equity Trading Practices and Market Structure 25
Table 26: Distribution of Respondents by Institution
VI. THE SAMPLE OF RESPONDENTS
AND THEIR REASONS FOR TRADING
Independent Investment Management F i
Subsidiary of Bank or Brokerage Firm
Mutual Fund
Internally Managed Pension Fund
Other
Total
A total of 825 questionnaires were mailed to 125 members of the TraderForum
and to 700 non-members. A total of 150 responded. These include approxi-
mately 90 TraderForum members and 60 non-members." Thus, the response was
72% of TraderForum members and 8.6% of non-members. In terns of our total
respondents, 60% were TraderForum members and 40% were non-members.
The questionnaires were filled out by the equity trader at each institution. In
some of the smaller institutions the trader may also be an asset manager. Re-
spondents were asked the total value of their organization's equity assets under
management. A total of 135 out of the 150 answered. The estimated amount of
equity under management was $1.54 trillion.12 This represents approximately half
the managed equity assets13 The distribution of the respondents, according to the
type of institution, is shown in Table 26.
Table 27 shows the reasons for trading stated by the respondents in descending
order. The primary reasons are stock specific fundamental issues (79%), internally-
generated research (68%), reassessment of portfolio structure (47%), bargain-
hunting (37%). and profit taking (36%).
l l ~e mbe r s automatically receive the =port while non-members must send in a card to receive the
report. We expect that virtually all non-members who took the time to fill out the questionnaire would
want to receive the repon Thus the number of non-membezs is inferred from the number of cards that
wen received.
120ut of the 150 respondents. 128 respondents repoxted a total of $1.316.42 billion of equity under
management and 22 gave no answer. Extrapolating to the total of 150. we estimate the total assets
under management of the mpondents of the questionnaire at $1.54 trilIion.
13~otal equity assets in the US at the end of 1992 were $5.5 trillio- ( F h ofFundt Codcd Tnbk s.
Board of Governors of the Federal Resuve System. Washington. D.C.). It is estimated that 60% of
these an managed, so that managed equity assets are $3.3 trillion. Thus. our survey covers about 50%
of all managed equity assets.
Number of
Respondents
69
53
14
9
5
150
Percentage of
Respondents
46.0
35.3
9.3
6.0
3.3
100
26 Nicholas Economides and Robert A. Schwartz
Table 27: Why Do You Trade?
"The indication NA in the top nght hand comer stands for "No Answef.
MI. DIFFERENCES BETWEEN ACTIVE AND PASSIVE TRADERS
One of the interesting questions for which the responses to our questionnaire pro-
vide an answer is whether significant differences exist in the trading behavior of
active traders in comparison with passive traders. In particular, we are interested to
see if these two groups have reported differences in what they consider an immedi-
ate trade, on the willingness to trade patiently, on motives for trading in electronic
?ystems, on the effects of anonymity, or in their reasons for trading in general.
The answers of active and passive traders to many questions were similar.
However, in some questions their answers could easily be differentiated. We
summarize below the responses in which active and passive traders showed clear
difference^.'^
l4Nl Tables in this section an numbered Nb where N is the number of the comsponding table
presented earlier that summarized responses to the same question by al l participants.
Equity Trading Practices and Marker Structure 27
Table lb: Time In Which You Consider A Trade To Be Immediate: Differences
Between Active And Passive Traders
Table l b shows that the time horizon appears to be a bit shorter for active than
for passive traders, as one might expect. At the short end of the scale, 76% of the
active traders checked 10 minutes or less, versus 50% of the passive traders. At
.the long end of the scale, 3% of the active traders checked one day versus 25% of
the passive traders.
TabIe 2b shows a tendency for passive traders to be more willing to trade
patiently: 46% of the passive traders said they would be "very willing:' versus
31% of active traders. No passive traders indicated they would be not willing
at all or not very willing versus 10% of active traders. This is consistent with
expectations. Trading on news implies a need for immediacy on the part of active
traders, and seeking to minimize transaction costs implies patient trading on the
part of passive traders. But again, the difference between the two groups is not
.large.
Table 12b distinguishes between active and passive traders on the issue of
concern about information leakage. Despite the general similarity, a substantial
percentage (33%) of passive traders are not concerned at all about information
leakage compared to 8% of active traders.
Table 13b shows that passive traders are much more likeIy never to demand
capital from a broker (33% versus 17% for active traders). Further, none of the
passive traders regularly or frequently demand capital from a broker, while 8% of
the active traders do.
. Table 21b shows the differences between active and passive traders in their mo-
tives for trading on electronic systems. Among passive traders, lower transaction
costs are the primary motivation (75.0%); this motive was indicated by 42.9% of
active traders. Active traders are motivated by a variety of other reasons. Reduced
market impact and lower spread costs are the primary reasons for active traders,
28 Nicholas Economides and Robert A. Schwam
Table 2b: Willingness To Trade Patiently To Reduce Execution Costs: Differences
Between Active And Passive Traders
[ No Answer I 2 I 1.7 0 0
Table 12b: Concern About Information Leakage When A Broker Is Called: Dif-
ferences Between Active And Passive Traders
indicated by 52.9% and 52.1% respectively. Each of these motives was indicated
by only 25% of the passive traders as a primary motive. 45.4% of the active traders
indicated better liquidity as a primary motive, in contrast with 20.8% of the passive
traders. Similarly, anonymity was indicated by 42.9% of active traders as a motive,
and only by 16.7% of passive traders. Finally, greater control of the negotiation
process was indicated by 36.1% of the active and 20.8% of the passive traders.
-
Table 22b shows that more active than passive traders believe that the anonymity
offered by electronic trading systems improves their execution ability. This opinion
is expressed by 46% of the active traders and 25% of the passive traders.
Table 13b: Frequency With Which Capital Is Demanded From A Broker For
Transactions Of 10,000 Shares Or More: Differences Between Active And Passive
Traders
Never
Rarely (1-24% of
Trades)
Sometimes (25-49%
of Trades)
Regularly (50-74%
of Trades)
Frequently (75-
100% of Trades)
Don't Know1
Not Sure
No Answer
Active Traders
Number of
Respondents
20
65
19
5
5
4
1
Passive Traders
Percentage of
Active Traders
16.8
54.6
16.0
4.2
4.2
3.4
0.8
Number of
Respondents
8
13
3
0
0
0
0
Percentage of
Passive Traders
33.3
54.2
12.5
0
0
0
0
2
30 Nicholas Economides and Robert A. Schwartz
Table 21 b: Motive For Trading On The Electronic Systems: Differences Between
Active And Passive Traders
Table 23b shows that a larger percentage of active than passive traders believe
that the benefits of electronic trade execution outweigh its disadvantages. This
was expressed by 71% of active traders, compared with 46% of passive traders.
Table 25b shows significant differences between active and passive traders in
the reasons that would make them use the alternative trading systems more. Active
traders (61%) would use the electronic trading systems more if the systems gave
higher execution rates. Only 21% of passive traders indicated this. Similarly, 36%
-of active traders, in contrast with 25% of passive traders, indicate that they would
use more such systems if they did not have soft dollar arrangements. If systems
were available at convenient times, 35% of active traders, in contrast with 13% of
passive traders, indicate that they would use such systems more. Finally, 22% of
active traders, in contrast with 8% of passive traders, indicate that they would use
these systems more if they knew more about them.
Reduced
Market
Impact
Lower
Spread
Costs
Better
Active Traders
Number of
Respondents
Who Rated
Motive
Important
63
62
Passive Traders
Percentage
of Active
Traders
52.9
52.1
Number of
Respondents
Who Rated
Motive
Important
6
6
Percentage
of Passive
Traders
25 .O
25.0
Equity Trading Practices and Market Structure 3 1
Table 22b: What Effect Does The Anonymity Offered By Electronic Trading
Systems Have On Your Execution Ability?
Table 23b: Do You Believe That The Benefits Of Electronic Trade Execution
Outweigh The Disadvantages Or That The Disadvantages Outweigh The Benefits?
Table 26b distinguishes between active and passive traders in the distribution
of the respondents according to the type of institution where they are more likely
l o trade. Among active traders, the majority (5 1%) are in independent investment
management firms, while 34% trade for a subsidiary of a bank or a brokerage h.
The roles are reversed among passive traders: the majority (54%) of passive trader
respondents trade for a subsidiary of bank or a brokerage firm, while 29% is in
independent investment management firms.
Table 27b below shows that the most important reasons to trade for active
traders concern their evaluation of fundamental information concerning individual
stocks. Most respondents (87%) indicated that they frequentIy trade because of
stock specific, fundamental issues; and 74% indicated that they frequently trade
because of internally generated research. Interestingly, passive traders also trade
for these reasons, though not as much as the active traders. Only 58% of passive
-
Benefits
Outweigh the
Disadvantages
Disadvantages
Outweigh the
Benefits
No Answer
Active Traders
Number of
Respondents
84
19
16
Passive Traders
Percentage
of Active
Traders
70.6
16.0
13.4
Number of
Respondents
11
5
8
Percentage
of Passive
Traders
45.8
20.8
33.3
32 Nicholas Economides and Robert A. Schwartz
Table 25b: What Would Get You To Use The Alternative Trading Systems More:
Differences Between Active and Passive Tradersa
'Note that columns add to more than 100% because some respondents have checked more than
one category.
traders frequently traded because of stock specific fundamental issues, and 21%
-because of internally generated research. On the other hand, 33% of the passive
traders traded to trace a market index versus 8% of active traders. And 33% of
passive traders traded because of fund redemptions or other cash flow reasons
versus 19% of active traders.
In some respects the two groups are quite similar: 39% of active traders fre-
quently trade for profit taking compared to 25% for passive traders; 38% of active
traders trade for bargain-hunting purposes compared to 38% of passive traders; and
74% of active traders infrequently trade because of chartist signals compared to
75% for passive traders. Overall, the active and passive traders differ in emphasis,
'but not by as much as expected.
VIII. CONCLUSION
This paper has presented an assessment of the demand for immediacy by buy-side
institutional equity traders that we surveyed. The 150 surveys that were returned
clearly indicate that an appreciable number of the respondents do trade patiently.
This is not surprising in light of the size of trades that the institutions commonly
seek to make, and of the costs to them of obtaining immediacy in a continuous
Equity Trading Practices and Market Structure 33
Table 26b: Distribution of Respondents by Institution: Differences Between Ac-
tive and Passive Traded
ONote that columns add to more than 100% because some respondents have checked more than
one category.
Independent
Investment
Management
Firm
Subsidiary
of Bank or
Brokerage
Firm
Mutual
market. Their orders commonly exceed the average daily trading volume for a
stock, the large orders are commonly broken into smaller pieces, and the smaller
pieces often take a day or more to be executed completely. The respondents were
also concerned about losing anonymity to sell-side brokerfdealers.
These findings have a major implication for market structure. Increasingly,
electronic technology is making it possible for institutional buy-side participants to
meet each other directly in a disintermediated environment. This can be done most
effectively with batched (i.e., call market) trading arrangements, which establish
place and time meeting points. A perceived limitation of call market trading is that
it does not supply immediacy to participants. This is true, however, only if call
markets are used in place of continuous trading, rather than along with continuous
trading, as we recommend.
However, immediacy per se does not appear to be urgently sought by many
buy-side asset traders. This suggests that, if both call and continuous markets
were readily available to participants, the order flow directed to the calls would,
indeed, be appreciable. The bottom line is that providing electronic call market
trading would be desirable for an appreciable number of institutional investors.
Internally
Managed
Pension
Fund
Other 16.7
Active Traders
Number of
Respondents
61
40
Passive Traders
Percentage
of Active
Traders
51.3
33.6
Number of
Respondents
7
13
Percentage
of Passive
Traders
29.2
54.2
34 Nicholas Economides and Robert A. Schwartz
Table 27b: Why Do You Trade? Differences Between Active and Passive Traders
'The indication NA stands for "No Answer".
M. REFERENCES
Amihud, Yakov, and Haim Mendelson. 1980. "Dealership Market: Market-Making with
,
Stock Specific
Fundamental
Issues
Internally-
generated
Research (from
Portfolio
Manager)
Reassessment
of Portfolio
Structure
Bargain-hunting
Profit Taking
Market-wide
News
Fund Redemp-
tions or Other
Cash Flow Reasons
Trading
Information
(i.e., Knowledge
of an Order
on the Floor)
Desire to
Cut Losses
Chartist Signals
Need to Track
a Market
Index
Derivatives-
motivated
Trading
Other Factors
Frequ
ently
86.6
74.8
47.9
37.8
39.4
34.5
19.3
18.5
18.5
14.3
7.6
4.2
1.7
Frequ
ently
58.3
41.7
54.2
37.5
25.0
20.8
33.3
8.3
4.2
4.2
33.3
8.3
0
Active
Neu
tral
8.4
17.6
31.9
29.4
30.3
33.6
24.4
31.9
34.5
11.8
10.1
5.9
0.8
Passive
Neu
tral
16.7
20.8
20.8
12.5
33.3
29.2
29.2
29.2
29.1
20.8
12.5
0
0
Traders
Infreq
uently
3.4
6.7
17.6
32.8
30.3
30.3
54.6
48.7
46.2
73.9
80.7
88.2
10.1
NA
1.6
0.9
2.6
0
0
1.6
1.7
0.7
0.8
0
1.6
1.7
87.4
Traders
Infreq
uently
25.0
33.3
25.0
50.0
41.7
50.0
37.5
62.5
66.7
75.0
54.2
91.7
12.5
NAa
0
4.2
0
0
0
0
0
0
0
0
0
0
87.5
Equity Trading Practices and Market Structure
35
Inventory." Journal of Financial Economics. (March 1980).
Amihud. Yakov, and Haim Mendelson. 1985. "An Integrated Computerized Trading Sys-
tem," in Market Making and the Changing Sttucture of the Securities Indurtr~ Ed.
Y. Amihud, T. Ho and R. Schwartz Lexington Books Lexington. MA 1985.
Cohen, Kalrnan J., S. Maier, Robert A. Schwartz, and D. Whitcomb. 1986. The Micmstruc-
rure of Securities Markets. Englewood Cliis. NJ: Prentice Hall.
Cohen. Kalman J., and Robert A. Schwartz. 1989. "An Electronic Call Market: Its De-
sign and Desirability," in The Challenge of Information Technology for the Securities
Markets: Liquidity, Volatility, and Global Trading. H. Lucas and R. Schwartz Editors.
Economides. Nicholas. 1993. "Network Economics with Application to Finance." Financial
Markets, Institutions & Instnunents 2: 5: 89-97.
Economides, Nicholas. 1995. "How to Enhance Market Liquidity," chapter 6 in
Robert A. Schwartz (ed.) Global Equity Markets, Irwin Professional. New York: 1995.
Economides, Nicholas. 1996. 'The Economics of Networks." International Journal of In-
dwtrial Organization. Forthcoming.
Economides. Nicholas and Jeff Heisler. 1994. "Co-existence of Call and Continuous Mar-
kets:' mimeo.
Economides. Nicholas and Robert A. Schwartz. 1995. "Electronic Call Market Trading."
Journal of Portfolio Management 21: 3: 1&18.
Economides, Nicholas and Robert A. Schwartz. 1994. Making the T d e : Equity T d n g
Practices and Marker Structure-1994, TraderFonun, Institutional Investor. New York.
Economides. Nicholas and Aloysius Siow. 1988. "The Division of Markets is Limited by
the Extent of Liquidity: Spatial Competition with Externalities." American Economic
Review 78: 1: 108-121.
Garbade, Kenneth, and William Silber. 1976. "Price Dispersion in the Government Securi-
ties Market." Journal of Political Economy 84.
Garbade. Kenneth, and William Silber. 1979. "Structural Organization of Secondary Mar-
kets: Clearing Frequency, Dealer Activity and Liquidity Risk." Journal of Finance 34:
577-93.
Garman, M. 1976. "Market Microstructure." Journal of Financial Economics (June).
Handa. Puneet, andRobert A. Schwartz. 1996. "How Best to Supply Liquidity to a Securities
Market:' Journal of Portfolio Management. Forthcoming.
Ho, T.. and Stoll, H. (1981). "On Dealer Markets Under Competition." Journal of Finance
(May).
McInish, Thomas H., and Robert A. Wood. 1990. "An Analysis of Transactions Data for
- the Toronto Stock Exchange: Return Patterns and End of the Day Effect" Journal of
Banking and F i i c e 14: 441458.
Mildenstein, E., and 'schleef, H. 1983. 'The Optimal Pricing Policy of a Monopolistic
Marketmaker in the Equity Market." Journal of Finance (March).
Schwartz, Robert A. 1983. Reshaping the Equity Markets: A Guide for the 1990s. ~ & r -
Business, 1991 (reissued by Business One Irwin. 1993).
Schwartz, Robert A.. 1996. (editor) The Electmnic Call Market, Irwin Professional, forth-
coming, 1996.
Wagner, Wayne, and Mark Edwards. 1993. "Best Ekecution." Financial Analysts Journal
(JanuaryIFebruary).
36 Nicholas Economides and Robert A. Schwam
X. APPENDIX: THE QUESTIONNAIRE
WuUonal Imwtor CARD 1
ZAaxHonr W E : E m T w x m p w m wo YUm mRvcTm
HOWANDWm YOU TRAM
Listed below pc r nmnber of rwta th.1 might motivate s oc k wades Pl ase indicate how Itequcmly yom nada
uemachr l t al byt hef ol l opr i ngk~( Chr ckI boxh t t os r her es - ' l t egant l y- d 1 -'wcatrIl'.)
b. Market-vide news 5 0 ~ L I 1 0 -2 3 0 .J 2 0 4 I 0 .I
d. Fund redemptions or
ochnuhflowrcaums50 at 4 0 .I 3 0 .J 2 0 4 1 0 .I
e. Need to track I
market index 5 0 rrl 4 0 a 3 0 .J 2 0 4 I 0 .I
h. Desire to cut losses 5 0 ~rc 4 0 .r 3 0 a 2 0 .I 1 0 .I
i. Chutin signals 5 0 ln 4 0 . a 3 0 a 2 0 4 I 0 J
j. lnternrlly-:-ted
reseuch ('om portfolio
manager) S 0 14.1 4 0 d 30.1 ' '204 1 0 :I
k. Tnding informmion (i.e. -
knowled~e o f u adn
cnlhc[bor) S 0 crc 4 0 -z 30.1 2 0 4 1 0 .I
m. 0th- (please speci&) I
I
I
5 0 17-28 . I 4 0 .I ' 3 0 .I 2 0 4 10 4
How milling at you to trade pmiently in m attempt to redwe execution corn?
Vc y willing or m tit rillm:
5 0 21-1 4 0 . 1 3 0 .J 2 0 4 I 0 -I
CARD 1
Equity Trading Practices and Market Structure
3 Whsl do you cauida an trrmaliate &&? A t d c a d mthk
Beuwprices am d d k md lhc risk of waiting is tw 0 .I b -1
$. Plcue indicate, ar a sea* of 1 lo J rvhen 1- lan p r c f d nd II - *most prcled,' how you f d fib& plums
orden a~ each of the lhna listed below. ( k k cne box fa each h e paiod l i d )
38
Nicholas Economides and Robert A. Schwanz
7. Ona) oobndai dcd( o~horol l cnrol ddyoowccp( addayrl rncbwr I b a a o J I ~ g a S J Oi f y o o
mdd m he frdkwinr in carmdrrian* bid-mk wmd md mrLct bout PLEASE INDICATE A RESPONSE
sniap or:
r e p e r h e 0n.1 0.1 0.1 0 4 0 .I 0 4
b. l l Cpqrbra 0-1 0 . 1 0 .I 0 4 [3 .I 04
LlllCparbra OM 0 4 0 .I 0 4 t ) .I 6 4
d S W p e r h Ont 0 . 1 0 .I 0 4 0 .r 04
C7scpCrhra 0 I b t 0.1 0 .I 0 4 0 4 0 4
c l l p e r h r a 0 ~1 0 . 1 0 4 0 4 0 .I 0 4
I. Ona y a , h&d&k dt o~hov ol t a r r oul dyar r apl addsyof t hr nbol r nf aar t ocf ~g~S5Oi f you
d save che fokwhg in oarnioiom. bid-ak spmd md mrLa i mpat: (PLEASE INDICATE A RESPONSE
FOR EACH OF THE USTED SAVINGS PER SHAM OPTIONS.)
Don'tbund
S a me r i ma R r ml r l v - - ,
d (2549??d (S&74%d ( 7S- I W?of
-) -1 -)
Saving 0T:
r6cperha-e O w t 0.1 0 .I 0 4 0 J 0 4
b.lZdpahare Omt 0.1 0 .I 0 4 0 4 04
c211~paha-e 0-1 0.1 0 .I 0 4 0 -I 0 4
d SW pa thrre 0 41.1 0.1 0 .I 0 4 0 4 0 4
L 7% padwe 0 a 1 0.1 0 -1 0 4 0 .I 0 4
f . $ l p ~ h t 0-1 0 -1 0 4 0 -1 0 4
- -
mdini at S O ifp muld raw che fdkr;inl in c a mk h , bid.& &d md h e f
(PL~WE
INDICATE A RESPONSE FOR EACH OF THE USTED SAVINGS PER SHARE OPTIONS.)
Equity Trading Practices and Market Structure
0nChoodtl.y O r a l 0 . 2 0 .I 0 4 0 4 0 4
n Kc hoo dehy 0 11.1 a .I 0 .1 0 4 a 4 0 4
bby ddg O m r 0 . 2 0 4 0 4 0 .1 0 4
Owrm~byddy 0 a r 0 .I 0 .1 0 4 0 .l 0 4
It. W N G YOUR ORDERS
I How 0 t h do yar me theLllmm~:
Dear h I
&!a BmIY Sanai mal l rml rf v- -
(1.24% of 0549%0f (5674% of ClJ-lW/. d
?a. Do yar believe k are drawbacks to usin8 limit adco? y a 0 -1 m 0.1
26. U~wmw+rcd~toQuacMo2a,rhrcdoyacrmaidathe ' -,
. . .
7
P l u w Ldiatc by checkin8 it in C o l m A belor.
? c . I n Cd u n n B . p l a r e c h c d r h ~ ?' ' ~ t o l i i t a d a .
(Pk.pcchcctaJyaaboxinachdm)
REASONS FQB Not USW-
CdmraA s ! h n B
MarCirnpatal Secadmortimportal
fmI
Risk oCnon-cx& 0 a.r 0 a 1
on-i0lmcdi.l~ cxauti~l 0 4 , 0 r
MycrrtlccanpticiwdimdvQUgc 0 -3 0 4
M y be di f f dt lo withdm qtkkly 0 4
Myaurcya,tomiuabvurbknwkumovarrn(
0,
0 -1 0 1-3
Gikophnstohdcrla 0 4 0 4
olhcr (piur~ spaiv): 0 -74 0 .,a
40 Nicholas Economides and Robert A. Schwanz
i W h e n y w d . n a d e r o f 1 0 . 0 0 0 ~ h ~ a m m b a L m f i n ~ ~ h ~ ~ & ~ ~ i c ~ f l ~ ~ ~ a ~ ~
fobtin# execution?
Don't know/
Equity Trading Practices and Marker Structure 41
8. Howmuch time doa yar pcdol i i mrupr lypially lire p 0 imp- r ! r ~ adcr (25% of avenge daily
t n d i n g v . h r a m m ) t o r r k ( c unpmy ( mnpy r muk c t ~ of Sl M) mi i or a m) B;
(cunpmy wi& r nmka api ul i i o f b dun SlO0 dllim). (PLEASE CHECK ONLY ONE
BOX W EACH COLUMN.)
9 Pkuci ndi acc~i mpaul ( I hcf dl oumga~r r ebyml dngl hani nader 0t ' ~h I (modhqmmnt)
to 3 (kut impacrm). (PLEASE USE EACH NUMBER ONLY ONfX.)
Mda impact - -.I
costs ofmissins r pria
- I
Ctmnniams - I
10. When you 4 r brdrcr, how c o n d are you aboul infam~ia, leakage?
11. Haw~tly&yw~wmmIh~anbmksbdiuliaurbatlaadsvhichislu~dativetoIhcmnrge
daily adhrg volume (ADTV)?
2 5 ~ 0 % of ADW 0 at 0.3 0 .s 0~ 0 J
Ma e lhm JogA of
ADTV b ISI 0.3 0 -I o a 0 J
42 Nicholas Economides and Robert A. Schwartz
Up to l hour 0 I 0 .I 0 .I 0 4 0 .s .- 0 4
BrrwcenIhour
rd3hcm OILI 0.l 0 .t 0 4 0 .s Or
&c~ar 3homr
md l by 0 15.1 0 .I 0 .I 0 4 0 .s Or
M a c h
Iby 0 lCl 0 .1 0 .I 0 4 0 -3 0 4
Equity Trading Practices and Market Structure
43
II. ALTERNATIVE TRAMWQ SYSTEMS
russsmh
Cnnsmg 0u.1 0 .1 0 -1 0 4 0 .9 O r
POSK C]%I 0 .I 0 4 0 A 0 .s 04
AU[ 0 n.1 U .I 0 -1 0 4 0 4 Or
I n r c i n a r mt h mn Un ~ 0.1 0 .I 0 4 0 .s Or
2. If you w my dlhe folloumg rltanrtirc systems. wlut is yar gmml kvd dmisfwtion with durn oo I sale of I to
5. u h 5 ' vay salisrt. ad I= .nac a dl satisft.
NYSE Sasion I S OS I 40. 1 30. 3 2 0 4 1 0. 3 or
NYSE Sasion 2 SC]mt 4 0 ~ 3 0 4 2 0 4 l o . ? 04
InshtCrossing 3031. 1 4 0. 1 30 . 3 2 0 4 1 0 - s 04
POSK S o n - I 4 0 . a 3 0 . 1 2 0 4 I 0 . s or
IW S 0 a 1 4 0 . 1 3 0 4 2 0 4 1 0 4 O r
br ut ~~~~ndkmw 5 0 - 1 4 0 . 2 30. 8 2 0 4 , 1 0 4 04
I
Bette liquidity 5fJas1 4 0 4 3 0 . 1 2 0 4 1 0 . 1 04
TRde ma y mdy 5 Ow1 4 0 . 3 3 0 - s 2 0 4 1 0 . 1 or
Reduccnddi mpl d 5 o m 1 4 0 4 3 0 4 2 0 4 1 0 . 1 or
~~~ S o n 4 0 . 1 30. 3 2 0 4 1 0 . 1 04
LmuWa uc l i ooc os u S O W 4 0 . a 3 0 . 1 2 0 4 1 0 . s $0 4
Greaaeorardof
&ItMopocar 5 0 - 1 4 0 4 3 0 . ~ 2 0 4 1 0 . 1 or
*(pk-&fy):
~ On r r . 1 ' 4 O- r 3 0 . s 2 0 . 1 1 0 . 1 04
Nicholas Economides and Robert A. Schwartt
G. One the whole, do you Wic Bu the W~S of clecbon* lndc c~ccutia! outrucigh the didvmuges ar h t
thediudvantaga outweigh thebrncfits?
Benefits ouweigh IIIC disdvmugcs 0 w
Disldvantlgcs ouh\righ the benefits 0 .l
Equity Trading Practices and Market Structure
CARDS 2n
N. CWSlFICATWN DATA ( l o bo used for afutyrlt, nd for )dmlllkr(bn)
C. What is he toul value of your uganbtion's assets llndcr mmugmmt?
D. Please sate the puunup of .v& funds that are h a t e d b the fdlowing
Equitia: %OM
Fixed inuxne: % mu
Crrh ./. ~r t r
olha @leare &&I:
% nd
% s n
-
100.3 (CaIegolia dlcdd Id loo./.)
Value ofsmwitia badd Vduc of cumkial dourn: (41-79-2 I
S . , 000. 000 sr 5 . , 000. 000 sa
El .
bnh yw fw campkthq hh qurr(knluh. Pkna nd nrmd Ihd al yaw mponm m conM~&l andanolc)m~n. Pkeae
durn I l n h. pnplid burhas8 nply anvrkpa pdded Wf U RW
Irrld R8armhOeprrku
h ~ J h v 8 . k
UI Yadhon Annu8
w.rY*Nv lMP
Fa: UI I ) ms%4!
NOTTIS ON SUBMISSION OF MANUSCRIPTS
Send two copies of the manuscript to the editor at the address on the back cover of
the journal. The contents must represent original and unpublished work and the
manuscript should not be considered for publication elsewhere. Any necessary
permissions or copyright clearance must be obtained by the author. There is no
submissions charge or page fee.
Please follow journal style when word processing or typing the manuscript.
Manuscript should be submitted on standard paper with text on one side only,
and on diskette if possible. The manuscript should be approximately 75 journal
pages. Double-space all lines, including footnotes and referenceshibliography.
Footnotes should be numbered in the text and grouped at the end of the manuscript.
The position of diagrams and figures should be clearly marked in the manuscript
and, where possible, camera-ready copy should be provided by the author.
Each manuscript should include a title page, a table of contents, and a short
abstract. A brief biography for each author should also be included. Acknowl-
edgements, if any, should be supplied on a separate page or on the same page as
the author's biography.
References should be cited in text by author and (in parentheses) year of pub-
lication. References andlor bibliographies at the end of the manuscript should be
arranged alphabetically by author and follow the style of these examples:
Belsley, D. A., E. Kuh, and R. E. Welsch. 1980. Regression Diagnostics. New
York: John Wiley and Sons, Inc.
Dimson, P. and P. Marsh. 1984. "An Analysis of Brokers' and Analysts' Unpub-
lished Forecasts of U.K. Stock Returns." Journal of Finance 39:4: 1257-92.
Golec, J. 1988. "Empirical Tests of Principal-Agent Model of Investor-Investment
Advisor Relationship." Unpublished paper, Clark University.
Nakansson, N. 1979. "A Characterization of Optimal Multiperiod Portfolio Poli-
cies." Pp. 169-77 in Portfolio Theory, 25 Years AAfter: Essays in Honor of
Harry Markowitz, eds. E. Elton and M. Gruber. Amsterdam: North-Holland.
FINANCIAL MARKETS, INSTITUTIONS &
INSTRUMENTS
Editor Anthony Saunders
Managing Editor: Mary Jaffier
Editorial Policy Created out of the Monograph Series in Wnance and
Economics, this unique new journal specializes in the publication of
original studies in the fields of finance and economics, specifically in the areas
of financial markets, institutions and instruments. It provides an
outlet for manuscripts which are too long and broad for typical scholarly
journals and business periodicals.
Articles for review should be sent to the editor at the Leonard N. Stem
School of Business, New York University, 44 West 4th Street, Suite 9-160,
New York, NY 10012. For derails on manuscript submission. please see the
inside back cover of the journal.
doc_150669243.pdf