Insider Trading, Liquidity, and The Role of The Monopolist Specialist

Insider Trading, Liquidity, and the Role of the Monopolist Specialist
Author(s): Lawrence R. Glosten

Source: The Journal of Business, Vol. 62, No. 2 (Apr., 1989), pp. 211-235
Published by: The University of Chicago Press
Stable URL: http://www.jstor.org/stable/2353227 .
Accessed: 08/11/2013 04:07
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to The
Journal of Business.
http://www.jstor.org
This content downloaded from 133.30.18.89 on Fri, 8 Nov 2013 04:07:39 AM

All use subject to JSTOR Terms and Conditions
Lawrence R. Glosten
NorthwesternUniversity
Insider Trading, Liquidity, and

the Role of the Monopolist
Specialist*
The purpose of this article is to investigate a rationalefor the specialist system on the New York
Stock Exchange (NYSE). A key characteristicof
the specialist system is that there is one individual in each stock, the specialist, that has sole
access to informationabout the tradingprocess,
and this information provides some monopoly
power. It is shown that this monopoly power
may, in some environments,reduce some of the
social costs associated with trading on private
information.
Promotionalliteraturepublished by the New
York Stock Exchange points to the "specialist
system" as a superiorform of marketorganization, citing the "increased liquidity" and the
maintenanceof a "fair and orderly market" attributableto the specialist. On the face of it, the
claim seems preposterous. One's intuitionmight
suggest that any model of rationalmarketmaking
* Many of the ideas in this article were developed in discussions with Paul Milgrom.I also benefittedfrom discussions with Anat Admati,DouglasDiamond,MichaelDothan,
MichaelFishman,KathleenHagerty, Pat Hess, Ravi Jagannathan, Juan Ketterrer, Ananth Madhavan, Steven Matthews, Ailsa Roell, WilliamRogerson, and Chester Spatt. I
also appreciatethe commentsof seminarparticipantsat the
University of Minnesota,the London School of Economics
conference on market making, Standford University, and
Yale University. The usual disclaimerapplies. Fundingfrom
the BankingResearch Centerat NorthwesternUniversityis
gratefullyacknowledged.Partof this researchwas performed
while visiting the CurtisL. CarlsonSchool of Management,
Universityof Minnesota.
Tradingon private informationcreates inefficiencies because

there is less than optimal risk sharing.This
occurs because the response of marketmakers to the existence of
traderswith private informationis to reduce
the liquidityof the
market.The institution
of the monopolistspecialist may ease this
inefficiencysomewhat
by increasingthe liquidity of the market.
While competingmarket makerswill expect
a zero profiton every
trade, the monopolist
will average his profits
across trades. This implies a more liquid
marketwhen there is
extensive tradingon
private information.
(Journal of Business, 1989, vol. 62, no. 2)

? 1989 by The University of Chicago. All rights reserved.
0021-9398/89/6202-0001$O1.50
211

212
Journal of Business
would show that a monopolistwill optimallyprovidea less liquidmarket than competing marketmakers. In fact, Ho and Stoll (1981) show
that competingmarketmakerswill lead to a more liquidmarketin the
sense that the average bid/ask spread will be smallerwith competing
marketmakers.
A possible response is that the NYSE monitorsthe performanceof
specialists, and therefore, in fear of retribution,specialists provide a
more liquid marketthan they would if unregulated.However, if this is
the only argument,the obvious rebuttalis, If a regulatedmonopolistcan
supply what would be suppliedby competitors,why shouldregulatory
resources be expended when the seemingly more efficient alternative
wouldbe to replacethe specialistsystemwith competingmarketmakers?
The possible advantage of the specialist system was suggested in
Glosten and Milgrom(1985),and the intuitionis providedby the observation that a monopolist specialist can average profitsover time. For
example, the analysis of Glosten and Milgromsuggests that there may
be instances in which competingmarketmakerswill not quote, bid, or
ask prices. If the adverse selection problem is too extreme, then no
matterwhat bid and ask are set, each marketmakerwill expect to lose
money on a trade. The consequence is that the marketshuts down until
enough public information arrives to relieve the adverse selection
problem. Now imagine a situation in which no public informationarrives until the informationalevent occurs. In this case, after the informedtradersreceive their informationthe marketwill remainclosed
until the informationis revealed.
A monopolist specialist may also close the marketin such a situation, but he need not. By keepingthe marketopen, the specialistlearns
some of the informationof the informed.This may reduce the adverse
selection problem, making subsequent trades more profitable.With a
protected position, the specialist is able to realize the greater profits
that offset the losses earned early on in the trading.The result is that
both liquiditytradersand informedtradersare made better off relative
to the competing marketmaker system.
This article extends this intuitionby consideringan environmentin
which trades are not restrictedto unit amountsas in Glosten and Milgrom(1985). In this case, the marketmaker(monopolistor competitor)
quotes a price schedule that specifies the price per share as a function
of the quantity to be traded. While competing market makers are
forced to set a price schedule that leads to a conditionalexpected profit
of zero (conditionedon the quantitytraded),the monopolistspecialist
maximizes expected profitsover all possible quantities.This extra degree of freedom, the abilityto averageprofitsacross tradesat a point in
time, reinforces the above observationthat the specialist can average
his profits across time. The result is that in some environmentsthe
monopolistwill provide a more liquid market.This reinforcementmay

Insider Trading, Liquidity
213
be important.It is reasonableto assume that there are times in which

the specialist enjoys a monopoly position and times in which this monopoly power is curtailed by other traders. Thus, the ability of the
specialist to average profits over time may be restricted.
The model considered bears some resemblanceto other models of
trade with asymmetricinformation.Like Glosten and Milgrom(1985),
an environment is considered where traders arrive one at a time,
motivated by liquidity and/or informationconsiderations. As noted
above, however, trades may be of differentquantities,and hence there
is a relation between this model and the one of Easely and O'Hara
(1987).1This feature also recalls the model of Kyle (1985). In contrast
to Kyle's model, liquiditydemandis not perfectly inelastic and only in
the analysis of competingmarketmakersis the zero-profitassumption
made. Furthermore,all tradersare assumedto know the price they will
pay when they submit their demands. The problem of the optimal
dynamic strategy of an informedtraderis not considered as in Kyle.
Furthermore,the model ignores the "manipulation"strategiesconsidered by Ketterrer(1987).
Like Gould and Verrecchia (1985), part of this article considers the
behavior of a monopolist specialist facing informedagents. There are
several importantdifferencesin the analysis. In the Gouldand Verrecchia model, the specialist has some private informationand faces
agents with private information. Furthermore,the specialist is constrainedto quotinga single price. Despite this lack of a "spread"in the
Gouldand Verrecchiamodel, it is still possible to talk aboutthe liquidity of the market.Since the risk-sharingmotive of the traderis assumed
to be common knowledge and since the specialist is risk neutral, the
amount of risk sharingmeasures the liquidity of the market.
ThoughGould and Verrecchia(1985)do not work out the zero-profit
solution for comparisonwith the monopolist solution, it can be shown
(at least in the uninformedspecialist case) that the monopolistspecialist will provide less risk sharingthan a specialist who sets the single
price that will yield zero-expectedprofits.Essentially, given thatonly a
single price is quoted, both a zero-profitspecialist and a monopolist
specialist are able to average profitsacross trades. Thus, in the Gould
and Verrecchiamodel there is no advantageassociatedwith a monopolist. However, it is unlikely that competing market makers would
charge a single price-rather they would establish a price schedule, in
which case the Gould and Verrecchiamodel is not easily extended to
the case of competing specialists.
In the context of insurance markets, Rothschild and Stiglitz (1978)
raise the possibility that-but state that it is impossibleto say whether
1. The model of competitive market makers is identical to the independently derived
model in Madhavan (1987). He uses the model to examine continuous markets and call
markets.

214
Journal of Business
-asymmetric informationand the noncompetitivenatureof insurance

markets are related. However, subsequent work by Stiglitz (1977)
would seem to provide some evidence that, in certainenvironments,an
equilibriumamongcompetinginsurerswill not exist while a monopolist
will provide insurancein such environments.This result is in the same
spirit as the liquidity argumentspresented here.
The intuition that a monopolist specialist could average profits
across time is centralto the analysis of Gammill(1986).He proposes a
specific market structure in which there is a "large-trade"trading
mechanism and a "small-trade" trading mechanism. Small trades
transact at exogenously specified bid and ask prices, whereas large
trades are accommodatedby a monopolist specialist. In certain environments, the informedchoose to trade via the large-trademechanism
and reveal their information.The specialist loses on these trades, but
makes up his losses in the subsequent small-trademarket.
While this model requires a monopolist specialist to achieve the
temporal averagingof profits, it is not clear that a monopolist would
choose the tradingmechanismposited. Furthermore,the exogeneously
specified "small-trade"market seems somewhat ad hoc. This makes
difficulta comparisonof the proposed tradingmechanismwith a competitive marketmakersystem, and hence Gammill'sarticle(1986)does
not (indeed was not designed to) directly address the impact of a monopolist specialist on the performanceof the market.
This article is organized as follows. In the next section, modeling
issues are discussed, and the model itself is displayed. It is shown that
there will be instances in which competing marketmakers will close
down the market. The intuition for this is as follows: start with the
hypothesis that small trades are not generatedby extremeinformation.
In such a case, the spreadfor smalltradeswill be nearzero. The breakeven spread for large trades, however, will be large if there is a high
probability of extreme information. If this spread is too large, then
those with extreme informationwill in fact wish to make small trades,
violatingthe initialhypothesis. Extendingthis argument,every market
maker's hypothesis about the behavior of traders is refuted and the
marketcloses down.
The equilibriumof the competitivemodel is used to show that investors would vote, ex ante, for a law restrictinginsider trading. Intuitively, the fact that the zero-profit price schedule must be upward
sloping implies that asymmetricinformationleads to a reductionin the
liquidityof the market.Ex ante, an investor does not know whetherhe
will profit from informationor pay a premiumto escape a risky position. Hence he sees the consequence of allowing insider tradingas
being a reduction in the liquidityof the market.
The subsequent subsection examines the behavior of a monopolist
specialist. It is shown that a monopolist optimally earns a positive

215
profiton frequentsmalltrades, but loses on the infrequentlargetrades.

Essentially, the monopolistkeeps the price of large-investorbuys relatively low to prevent pooling of extreme and nonextremeinformation
at low quantities. Consequently,he subsidizes the large trades so that
the small, profitabletrades remain.
There are two conclusions of this section. First, the monopolistwill
always keep the marketopen since he can achieve this cross-subsidization of trades. Consequently, in situationsin which asymmetricinformationis perceived to be a significantproblem,the monopolistspecialist is able to maintain a more liquid market. Second, even in some
environments in which competitive market makers would keep the
marketopen, marketparticipantswould vote for a monopolistspecialist. This is proved by showingthat the benefitof a monopolistspecialist
is related to a parameterthat measures the severity of the asymmetry
of information.The benefit starts out positive (when the marketcloses
down) and ends up negative (when there is no asymmetryof information). The article concludes with some speculationon the meaningof
the results, extensions of the analysis, and a summaryof the major
results.
I. Model
As stated in the introduction,the purposeof this articleis to examinea
rationalefor the specialist's existence. Of course, the specialistis not a
total monopolist-he faces competition from limit order submitters,
other floor traders, and specialists in other securities (see Hagerty
1986).In fact, some of the advantagethat the specialisthas is informational. He, and he alone, knows what is in the "specialist's book."
Furthermore,he is in a position to know more about marketactivity
since most trades will be crossed with him. Indeed, many formal and
informal discussions of the position of the specialist assert that his
monopoly position derives from this information(see Leffler and Farwell 1963;Ho and Stoll 1981;and Stoll 1984).
This observationcan be interpretedseveral ways. Some have taken
it as a motivationfor analyses of a monopolistwith privateinformation
(see Gould and Verrecchia 1985;and Altug 1984).This articletakes the
point of view that the informationavailablefrom the specialist's book,
and the fact thathe sees the result of all trades, is informationnot about
the futureprofitabilityof the firmor the futurerealizationsof its stock
price, but rather about uncertaintyin the tradingprocess. Thus, this
informationgives the specialist a monopolyposition, but it is not information that tradersoff the floor of the exchange care about (except to
the extent that it might affect their order placement strategy, i.e., the
choice between limit orderand marketorder).Furthermore,the observation that specialists do not typically engage in security analysis

216
Journal of Business
strengthensthe position that a more appropriatemodel of the specialist

assumes that he has monopoly power but no privateinformationabout
the cash flows of the security.
A model embodying private informationand rational expectations
should provide for "liquidity motivated" trade-trade unrelated to
information-and the inabilityof the marketmakersto distinguishinformed and uninformedtraders. Withoutthe former, there will be no
trade. Without the latter, marketmakers will refuse to trade with the
informedand always trade with the uninformed,in which case there
will be no informed trade. The indistinguishabilityof informed and
uninformedis modeled here by assumingthat every traderhas both an
informationalas well as liquiditymotivationfor trading.To the extent
that a trader's endowment is, ex ante, optimal and his informationis
extreme, he is tradingfor informationreasons. To the extent that his
informationis not extreme and his portfolio is out of balance, he is
tradingfor liquidityreasons.
Marketmakers (monopolistor competitor)are presumedto be risk
neutral. Indeed, the analysis of Gould and Verrecchia (1985) would
suggest that market makers would be relatively less risk averse than
the typical investor. However, this implies that inventory costs induced by risk aversion are ignored in this model. This may bias the
case in favor of the monopolist system since presumablythe ability to
handlea given "market"inventoryis increasedby increasingthe number of marketmakers. This bias should be recognized, but removingit
would not alter the basic results. It will merely reduce the set of environments in which the monopolist organizationis optimal.
Finally, the dynamic order-placementstrategy of the traders is ignored. It is easiest to interpretthe model in the following way. Each
agent becomes informed and comes to market immediatelywith an
endowmentof shares known only to him. He plans to trade only once
and chooses the quantity to maximize his expected utility given the
posted price schedule. Subsequenttradershave endowmentsthat any
marketmaker considers to be independentof the past.
The analysis of a marketmaker's behaviorwill consider a response
to a single proposed trade withoutregardto the dynamicstrategiesthe
market maker could adopt. Given the assumed behavior of traders,
there is no loss in generality in this for the case of competingmarket
makers. Free entry of marketmakersand a risk-neutralityassumption
imply that the equilibriumin the competing market-makersetting is
independentof expectations of future trades. A monopolistspecialist,
on the other hand, will adopt a dynamic strategyand considerationof
the single trade case ignores this. The model shouldbe interpretedas a
descriptionof what occurs at a particularpoint in time. Then, except
for revisions in the relevant distributions,the situationis repeated.
Though the analysis will be carried out with a specific example of

217
preferences and information,it is easier to set up the nature of the

model somewhat more generally. The paper concentrateson the trading in one security, and for tractabilityit is assumedthat this securityis
the only risky security in a typical investor's portfolio. An investor
arrivingat a point in time is presumedto have preferencesover distributions of future randomwealth representedby the utility function of
wealth, U( ), which is increasing and concave. There are two securities, one risky and one risk free. The traderhas an endowmentof WO
dollars in cash and W shares of the risky asset. The risky asset will
have a payoff of X(w) in state w while the risk-freeasset has payoff R.
In additionto the endowment, the investor has received information,
S, about the payoff, X. When the investor comes to market,he faces a
pricingschedule, P( ), with the interpretationthat a (signed)tradeof Q
will lead to a transferof QP(Q)from the traderto the specialist. Thus,
if Qis positive, the investor pays a price of P(Q) per share, and if Qis
negative he receives P(Q) per share. Given his position in the other
securities and the price function P(-), the investor chooses Qto solve
max E[U(X(W + Q) + WoR - P(Q)QR)IS,W0,W]I (1)
Solution to this maximizationleads to a trade:
Q = Qp(S,Wo,W),
(2)
where the subscriptedP indicates that the chosen quantity depends

upon the pricing function P.
A typical market maker is assumed to be risk neutral, facing no
inventory costs or constraints and ignorantof the endowment vector
and the realization of the information. He does know the crosssectional distributionof these variables.This specificationleads to the
solution among competing market makers, Pc, which, as long as it
exists, must satisfy
(3)
E[XIQ] = PJ(Q).
That is, any trade of Qshares must lead to a zero conditionalexpected
profit. If PCwere set so that for some trades a profit were expected,
then a competing market maker would have an incentive to beat the
price for those trades.2
There may be many solutions to (3) since the conditionalexpected
value function depends upon the price schedule that is established.
Intuitively, if Pc is one schedule satisfying(3) and P' is anotherschedule that is above PCfor Q> 0, then an investor facing P' will trade a
smallerquantitythan if PCwere the schedule. Hence a given Qleads to
2. It is assumedthat either (1) traderscannot split theirordersamongmarketmakers
or (2) they can split their orders, but such splittingis observedby all marketmakersso
that N orders of size QIN placed by one traderare informationallyequivalentto one
orderof size Q.

218
Journal of Business
a greaterrevision in expectations when P' is the schedule. This greater

revision in expectations may just match the higherprice schedule. It
seems natural to assume that competition among competing market
makerswill determinethe smallest (largest)price schedulefor Q>
(>)
Osatisfying (3).3
If there is only one specialist, then it is assumed that the pricing
function, Pm, must satisfy (again, as long as it exists)
Pm(-) = arg max E[P(Qp(S,W0,W))Qp(S,W0,W) - XQp(S,Wo0W)].
(4)
The specification of the objective function in (4) deserves further
comment. It is conceivable (indeedin the exampleconsideredit will be
shown to be the case) that for some Q the expected profit, conditional
on that Q, will be negative. In specifying the objective function as the
expected profit across possible quantities,the tacit assumptionis that
the specialist quotes a price schedule and then is requiredto honor
those quotes for the trade, no matterwhat Q is requested.Of course, in
reality there is no posting of a price schedule-the only prices posted
are a bid and ask for small trades. If a traderwishes to trade a larger
quantity, then quotes must be obtained for larger quantities. In any
event, once the specialist supplies quotes for various quantities,he is
in fact bound to honor those quotes, whereas the trader is free to
choose the quantityhe wishes to tradebased on those quotes. The fact
that the specialist "moves first," in the sense that he must honor his
quotes, is the motivationfor choosing the objective function in (4).
Price functions satisfying(2) and (3) or (4) need not exist. In fact, the
subsequent analysis will show that there are situations in which the
competitivepricingfunctiondoes not exist. This happenswhen there is
very little uncertaintysurroundingthe endowmentsandrelativelygreat
uncertaintysurroundingthe informationaboutX. In this instance, the
adverse selection problem that market makers face is so severe that
there is marketbreakdown-all marketmakersrefuse to make a market.
This may or may not happenin the case of the monopoly specialist,
and in the example considered here it does not. Since the specialist
maximizes expected profits, he may expect to lose for some Q and
profit on other trades and still expect a positive profit. His monopoly
position allows him to average his profits cross-sectionally, that is,
across possible trades.
The specificationof the competitive and monopoly environmentsin
(1)-(4) are so general as to preclude all but the weakest analysis summarized in the above two paragraphs.To obtain more results, more
3. Of course, schedules satisfying(3) may not be ordered,in which case there is no
basis for choosing among the equilibria.This will not be the case with the example
considered.

219
structuremust be imposed, and the time-honoredexample from the

rationalexpectations literatureis chosen. Assume that X is normally
distributed,S is a noisy observationof X, and the utility function of a
typical traderis exponential. Furthermore,W is, from a marketmaker's point of view, normallydistributedwith zero mean.4
Formally, the model is specified as follows:
U(y) = - exp(- py), p > 0,
X - N(m, I/,rr),
W - N(O, 1/Irr,), independent of X,
S = X+
R=
?, ,
- N(O, lIiTrs),independentof X,
1.
(5a)
(5b)
(5c)
(5d)
(5e)
Thus, an arrivinginvestor has exponentialutility with risk-aversion

parameterp. The payoff on the security is normallydistributedwith
mean m and precision (one over the variance) of Tr, From a market
maker's point of view, W is normallydistributedwith mean zero and
precision Tr, The informationreceived by the traderconsists of a noisy
observationof X, the precision of the signalconditionalon X being 'as
Finally, the returnon cash is normalizedto zero.
A. CompetitiveMarket-MakerEquilibrium
In order to derive the equilibriumwith competingmarketmakers, we
have either to find a price schedule P(Q) satisfying E[XIQ]= P(Q),
when Q is the optimalQ chosen by an arrivingtrader,or show that no
such schedule exists. We begin by assuming that a pricing schedule
exists, and that it is differentiableand defined for all Q. The terminal
wealth of a traderfacing price schedule PQ) and choosing quantityQ
with initial endowment W is thus
-P(Q)Q + (W + Q)X.
(6)
The conditional normalityof X given S and exponentialutility imply

that expected utility maximizationis equivalentto maximizationof the
certaintyequivalent
CE(W,S,Q,P(Q)) = -P(Q)Q
+ (W + Q)E[XIS]
- .5p(W + Q)2var(XIS).
4. This specificationof the liquidity motivationfor trade is also used in Diamond
(1985).Also notice that the specificationof a zero meanimpliesthatthe riskrepresented
by the randomvariableX is in zero net supply. Thus, the effects of the risk-neutral
marketmakersas insurersis minimized.Thatis, the marketmakersfacilitaterisksharing
between agents and they do not expect to end up with a positive (or negative)inventory
of shares. Furthermore,the risk neutralityof the marketmakersdoes not affect the
averageprice of the security.

220
Journal of Business
After noting that E[XIS]and var(XIS)are given by

+ -rr); var(XJS) = 1/(rrx + rrs),
E[XIS] = (wrrm+ TrT.S)I(rrx
(8)
and assuming that the price function is twice differentiablewith first

and second derivatives given by P'({) and P"(), respectively, the
first-and second-ordernecessary conditionsfor the investor's optimal
Q are given by
Q)Q + P(Q)
S
=
-
P"(Q)Q
mK(rx + rrs)Tsr}+ pQITrs

(pW/I-rs) = Z-
2P'(Q)
m,
rsT)]<
[P/('ITx+
0.
(10)
Notice that if an arrivingtraderannouncesQ, then the marketmaker

can calculatethe value of the left side of (9), and hence he knows Z = S
- (pWIs-r)
X +
- (pWIrs), a noisy observation of X. Let
ii,
denote the precisionof this observationconditionalon X (i.e., one over

the variance of ? - [pWIrrs]). Then, a market maker can calculate
E[X|Z]:
+ -r,).
E[XIZ] = m + TrZ/Q(rrx
(11)
Substitutingfor Z from (9) in (11) provides an expression for E[XIQI.

Equating this to P(Q) leads to a differentialequation that P( ) must
satisfy. The family of solutions is presented in the following lemma,
which is proven in the Appendix.
1. Any differentiablepricingfunction that satisfies (9) and
LEMMA
(3) is given by
P(Q) = m + (Prr'rrIs/N)[QI(2ot- 1)] + K[sign (Q)IIQI`, (12)
where
N = p2 Xr +
a
rwrs(rs
rx),
P2 rrxIN # .5, y = a/(1
a),
K unrestricted,or
P(Q)
(x =
m + koQ - k1Q log(iQI),

.5,
ko unrestricted,
ki = p/(lTx + ws).
The price schedules in lemma I were constructedassumingthat (9)

characterized the decision of an investor. In fact, not all the price
schedules satisfy the second-orderconditionin (10). It can be verified
that in order for the second-order condition to be satisfied, (x must
exceed .5, and K, the unspecifiedconstant, must be nonnegative.The

221
assumed competition among market makers will thus determine the

linear price schedule as the only equilibrium.The results concerning
the equilibriumamong competingmarketmakers are presented in the
following proposition.
1.
PROPOSITION
If
OX= P2ITr/[P2Trr +
Tr1Trs(Trr +
Trs)] > .5,
then there is a unique competitive equilibriumpricing shcedule given

by
PJ(Q) = m + {PiTWiTs/[(2ot-
= P2iTx + 1TwITs(TS5+
1)N]}Q,
(13)
Tx).
The Q chosen by an arrivinginvestor is derived from (9). The equilibrium Q, QJ(Z)is given by
QJ(Z) = (Z - m)irs(2a - 1)/p.
(14)
If (xs .5, then there is no equilibriumpricingschedule and the market

closes down. The proof for proposition 1 is in the Appendix.
Whether or not the market is open is determinedby x, which is a
function of the parameters p,ITx,ITs,Trw.Each of these determines how
extreme a position the investor wants to take in response to a given

signal. The more extreme a position the investor wishes to take, the
more the market makers have to protect themselves by making the
price schedule steeper. At some point (when ox= .5), marketmakers
are unable to protect themselves and the marketcloses down.
Some discussion of the equilibriumconcept is in order here. The
price schedules that marketmakerscan pick have been limitedto those
satisfyingthe second-orderconditionin (10). This insuresthat the equilibrium,when it exists, is a separatingequilibrium.It has been pointed
out to me by Ailsa Roell that, if the schedules are not so restricted,the
linearprice schedule that has been derivedabove is not an equilibrium.
Specifically,if the linear schedule above were offered, a marketmaker
could offer a competing schedule that would cause the investors with
Z's in some intervalto all choose one quantity.If the competingschedule is chosen so that this one quantityis large enough, the alternative
schedule will yield nonnegativeexpected profits. Based on the results
in Rothschildand Stiglitz(1978),it is probablytrue that an unrestricted
Nash equilibriumdoes not exist for a model with a continuumof types
of investors.
A continuumof types of investors is used here so that the inference
problem of market makers can be solved easily. I have verified that
there is a discrete model with the property that a Nash equilibrium
exists as long as the informationalasymmetryis not too great,but does
not exist when there is too much private information.The restriction

Journal of Business
222
on the allowableprice schedules in the continuummodel is imposed so

that the results of the continuummodel mirrorthe results of the much
more complex discrete model.
The nonexistence of equilibriumwhen the informationalmotive for
trade outweighs the liquiditymotive for trade is due to the Nash equilibrium assumption. The results of Spence (1978) suggest that (in a
discrete model), even when the informationasymmetriesare large, a
"reactive equilibrium" may exist. Given that market makers must
committo a schedule (i.e., are not able to back away fromunprofitable
trades) I do not find the reactive equilibriumconcept attractive for
modelingcompetingmarketmakers.For example, supposethat a Nash
equilibriumdoes not exist, but a reactive one does. This reactive equilibriuminvolves expected losses on some trades and profitson others.
Suppose that a market maker posts the reactive equilibriumprice
schedule. Then the market maker is committed to this schedule and
hence cannot back away from the unprofitabletradesif anothermarket
makerposts.a price schedule which takes the profitabletrades. Consequently, it seems reasonable that such a competing price schedule
would appear and hence the reactive equilibriumprice schedule does
not seem reasonable in this environment.
Investors with relatively extreme endowments will be tradingfor
portfolio rebalancingreasons, while investors with endowments near
zero will be trading for informationreasons. Intuitively, one would
expect that, taking an ex post point of view, the market makers, on
average, gain from investors with relatively extreme endowmentsand
lose, on average, from investors with endowments near zero. This is
verifiedin the following lemma, the result of which will also be useful
in the subsequentproposition.
LEMMA
2. A trader'sequilibrium,expected profit(equivalently,the
negative of the marketmakers' profit)conditionalon the endowment,
but ex ante to the signal, is given by
= A(1
E(QC(z){X - PC[QC(z)I}IW)
W2),
Trw
(15)
where A is a nonnegative constant. See Appendixfor proof.

The fact that the price schedule is upwardsloping suggests that the
existence of informedtradersimposes a cost on the market.The reduction in "liquidity" means that there will be inefficientrisk sharing.In
fact, if the marketcloses down there will be no risk sharing.An interesting question at this point is whetheror not, ex ante, investorswould
vote for a law limiting the extent of informed trading. If informed
tradingwere outlawed, then competitivemarketmakerswould post the
efficient(in terms of risk sharing)price schedule that is independentof
quantitytraded.
Before receiving a signal, and without knowingwhat his endowment
will be when he wishes to trade, an investor does not know whetherhe

223
will be able to profit from his information(have an endowment near

zero) or end up limitinghis trade because of the positive slope of the
price schedule. Since the marketmakerswill, on average,breakeven,
one would think that, ex ante, traderswould ratherhave a prohibition
on tradingbased on information.This is shown in the followingproposition. The intuitionfor this result would appearto be somewhatgeneral, and the proof is divorced as much as possible from the specific
example considered here.
PROPOSITION
2. Assume that U"'() < 0, that e(W) = E[QC{X PC(QC)}I
W] is concave and symmetricabout W = 0. Furtherassume
thatfj( ), the density of W, is symmetric. Then, the ex ante utility with
asymmetricinformation(expectation is taken over W,S,X),

E[U(-PC(QC(Z))QC(Z) + {W + QC(Z),X)],
is strictly less than the ex ante maximizedutility withoutprivateinformation,E[U(Wm)].The Appendixcontainsthe prooffor proposition2.
The proposition is true, not because of resources wasted in the acquisitionof information(informationis costless in this model) but because of the imperfect risk sharing due to the existence of private
information.The result is thus analogousto the results of Hirschleiffer
(1971) and Marshall(1974) that private informationcan frustrateoptimal risk-sharing.If informationwere costly, the result would continue to hold and the interpretationwould be similarto thatin Diamond
(1985). The advantageof reconsideringthis result in this frameworkis
that the source of the inefficiencyis obvious and (at least in principle)
observable. Private informationdecreases the liquidityof the market,
and it is this lack of liquiditythat provides the result.
B. Monopolist Specialist Equilibrium
Given that any law restrictinginformationacquisitioncan only be imperfectly enforced, the next question to ask is whetheror not the costs
of informedtradingare a functionof institutionaldesign. In this section
it is shown that, in certain circumstances,a specialist with an institutionally protected position can increase the liquidity of the market.
Given the past literatureon the topic, the environmentsin which this
should hold must be those in which asymmetricinformationis thought
to be a significantproblem.Otherwise,the usual inefficiencyattributed
to a monopolist will hold. A monopolist specialist in an environment
with symmetric informationwill restrict trade, leading to inefficient
risk sharing.
It is convenient to recast the investor's choice and the monopolist's
maximizationproblemin the followingway. Notice from(9) that, given
any price schedule, the investor's optimaltradewill be a functionof Z
and observables, say Qp(Z).The specialist'sproblemis to finda sched-

224
Journal of Business
ule P() that will maximizehis expected profits.For a particularschedule, P(), define the functions Q() and R() by
Q(z) = Qp(urzz + m),
R(z) = P(Qp(Uzz + m))Qp(Uzz + m), P(Qp(Uzz + m))Qp(Uzz + m)

(16)
where ar is the variance of Z,
aZ
[p2 Sx
wws(x
2
2lS)]iT"iT
and z is Z normalizedto have mean zero and unit variance. Then the
specialist's optimalchoice of P() amountsto a choice of functionsQ()
and R('), which maximize his expected profit subject to an incentive
compatibilityconstraint. Formally, his problemis
max E[R(z) - XQ(z)],
(17)
subject to
CE[W,S,Q(z),R(z)]- CE[W,S,Q(z'),R(z')],
for all z', z = [S - (pWI-rrs)- m]/uz, and CE() is given in (7).
The strategy for solving this problem is to first assume that the
constraintcan be summarizedby the local conditionfor maximization.
This leads to functions R() and Q() that are optimal given the firstordercondition. However, this solutionis not feasible because there is
a discontinuity that implies that the local representationof the constraintin (17) does not, in fact, characterizethe constraint.A modification of the solution to the first problem is optimal and feasible, however.
Assuming that R(') and Q( ) are differentiablewith derivativesR'(Q)
and Q'(.), and definingD(z) by D(z) = R(z) - mQ(z), the local representation of the constraintin (17) is
D'(z) - Q'(z)[Isruzz - pQ(z)]I(as +
(18)
- Q'(z)V[z,Q(z)].
Note that V(z,Q(z))is the marginalvaluationof an investor of type z,

optimallytradingQ(z). It is independentof D(z) due to the absence of
wealth effects in exponential utility.
Under the distributionalassumptions of the model, z is normally
distributedwith zero mean and unit variance.Furthermore,X andz are
correlatedand
E[XIzI = m + 7rz(oz/(1Tr+ Trz)= m + e(z);
(19)
where rrzis given by -rr = rrswr

Let f(t) indicate the
2(p2 + rrwrrs).
standard normal density and let F() be the associated distribution

225
function.Then the maximizationproblemcan be writtenin termsof the

state variables Q(z) and D(z) and the control variableu(z) = VWz5
max ff(z)[D(z)
Q(z)e(z)],
u(z)V[z,Q(z)].
(20)
subject to
Q'(z) = u(z); D'(z)
The proof of proposition3, in the Appendix, establishes that the solution to this problem is
Q(z) = (asTz/p)(otz - {[1 - F(z)]If(z)}) for z
=
Ox =
OrrUzIp)[az + F(z)If(z)]
p2 rr/X[p22Trx + Trw
Trsf(x +
>
for z < 0,
(21)
s)] .
While Q() given in (21) has a positive derivativealmosteverywhere,at

z = 0 there is a discontinuity:
Q+(0) = -asr1[2pf(0)] < rrs[2pf(0)] = Q_(0).
It can be verified that, in this case, (18) does not completely characterize the constraintin (17). In fact, if z > 0 is the true characterization
of an arrivinginvestor and Q(z) < 0, then the investor will preferto act
given by
as if - z were his characteristic.However, Q*(Q)
Q*(z) = Q(z) for lzl> z*
(22)
= 0 otherwise,
z* satisfyingz* > 0 and Q(z*) = 0, is optimalfor the specialistanddoes
satisfy the constraintin (17). The excess revenue functionD(z) can be
found by integrating(18) to get
D(z)
+ mrs)- pQ(z)I[2(QTx
+ ws)]
Q(z){JiTuzzI(iTx
(23)
ls(rzu(irx+ Trs)fQ(t)dtIQ(z)},
where the integralis from z* to lZl.

3. When the arrivinginvestor has signalS and endowPROPOSITION
ment W, then the equilibriumtrade will be Q*(z) at price R(z)/Q(z),
where z is given by z = [S - m - (pWIrss)]I(z,and Q*(z)is given by
(21) and (22), and R(z) = D(z) + mQ*(z),where D(z) is given by (23).
See Appendix for proof.
The first thing to note from proposition3 is that the monopolistwill
keep the marketopen as long as z* is finite;that is, atis positive. Given
the results of the analysis of competitive market makers, this might
suggest that there are tradesfor which the specialistwill expect to lose
5. The maximizationin (21) is correctas long as Q(z)is monotoneand strictlymonotonic on any intervalwhere Q(z) is nonzero. This will be shown to be the case.

Journal of Business
226
money. In fact, the specialist will always lose money on extreme

trades. This can be seen by consideringthe behaviorof R(z) for large
z.6 For large z, [1 - F(z)]If(z) is near zero and hence, for largez, Q*(z)
is approximately 1TrszazIp. Furthermore,
J'Q*(t)IQ*(Z) =-{.5z
- .5z*2/Z
{1- [1
- F(t)]Iflt)}
F(z)]I[oazf(z)]}
(1/aZ)f[I
is approximatelyz/2 for large z. Thus, for largez, R(z)IQ(z)is approximately

m
-
[Trrsuz(l -
ot)z]/2(Tr
Tr,)
m + .5(ua1TrwTs2z)/[p2Trx
+ TwWs(,Tx+
IrS)].
In contrast, E[XIz]is given by

E[XIz] = m +
24zz)/(p21Tx
(1TwT
TrwTrsTrx +
ITiT 2),
and hence for large z, R(z)/Q(z) < E[Xlz], and the specialist loses
money on large trades.
Extreme z's are relatively unlikely, and hence the specialist still
expects to profiton average. In fact, the specialist's expected profitis
given by
{[P2ITx+
Tw1Tss(1Tx+ 1Ts)]/1TrwTx(1Tx
1TS)}
f(t){at
[1
F(t)]If(t)}2.
That the specialist chooses to make expected losses on extreme

trades, while perhapssurprising,can be understoodby examinationof
the proof of proposition3. The expected benefitof marginallyincreasing the quantitytradedby a traderof type z is the marginalrevenue at
that z times the density at z less the expected reductionin the marginal
valuations of traders with types largerthan z. The expected marginal
cost is the conditional expected value of X times the density at z. At
large z's, the probabilityof an investor arrivingwith a higher type is
small, and hence the marginalbenefitat z is approximatelythe marginal
revenue at z. Since the conditionalexpected value is increasingin z, for
large z's the marginalrevenue must be increasing.But this implies that
for large z's the marginalrevenue is above the price, and, hence, the
price is below the conditional expected value; that is, the specialist
optimallyexpects to make negative profits on large trades.
Anotherway to see the intuitionis to consideran alternativestrategy
for the specialist. Suppose the specialist were to eliminateall the nonprofitabletrades by specifying an arbitrarilyhigh price for quantities
above some level. This would cause all the high z traders to pool at
lower quantities, reducingthe profitabilityof the low-quantitytrades.
6. Largepositive z will be considered,z negativeand largein absolutevalue will lead
to identical results since it is easy to check that Q*(z) = - Q*(- z) and D( - z) = - D(z).

227
Since the specialistis operatingin the elastic partof the demandcurve,

the reduction in the expected revenue accompanyingthis pooling is
greaterthan the reductionin the expected cost, and the expected profit
is lower.
Since Q*(z) is incentive compatible, it must be that Q*(z) is nondecreasing. It was shown in the proof of proposition 3 that for z > z*,
Q*'(z) > 0 and Q*"(z) < 0. Since Q*'(z) > 0 for z > z*, there is a function Q `() such that if Q*(z) = q, Q l(q) = z. Thus, we can define the
price function P,,(q) analogous to the competitive price schedule by
Pm(q) = R[Q-1(q)]Iq.
There is a discontinuityin Pm(m)at q = 0. The limit as q decreases

to zero is m + z*,rrszl((rx + rs), whereas the limit from below is
m - z* Trs(u(/r( + rs). The difference, 2z*irsuzl(/rx + ars), can be
termed the "zero-quantityspread." This spreadis of interestbecause
it is the calculablequantityin this model that is closest to the observed
quoted spread. This spread will depend upon as, a measure of how
much private informationthere is, but the spreadis positive no matter
what irs is, includingirs = 0. Interestingly,the zero-quantityspreadis
not, in general, a monotone function of ars.This suggests that examining quoted spreads for evidence of insider tradingmay be misguided.
Whatis requiredis an examinationof how much large tradesmove the
price. Using the results above for the behaviorof R(z) and Q*(z)for z
large, it is easy to see that Pm(q) for q large is approximatelym +
q(ar,as1pira), which is clearly increasingin irs.
II. Monopolist/Competitor
Comparison
The results of the analysis of the competingmarket-makersmodel and
the monopolist-specialistmodellead immediatelyto the centralwelfare
result: in some environments,the monopolist-specialistsystem will be
weakly preferredby all tradersand strictlypreferredby some traders.
The set of environmentsfor which this is true is the set of environments for which ot = p2Trx/[p2ir2+ ir1/r(, + rs)] < .5. If a < .5, then
the competing market-makermarket will not open, whereas the specialist will not suspendtrading.Since the zero tradeis feasible in either
market, traders who trade a nonzero amount with the specialist will
strictlypreferthat there be a specialistmarket.This proves the following proposition.
PROPOSITION4. If a c .5, then all potentialtradersweakly preferthe
monopolist-specialistsystem to competing marketmakers, and some
potential traders strictly prefer the monopolist specialist.
As the analysis of competingmarketmakerssuggested, the parameter a essentially measures the severity of the adverse selection problem. When a is one, there is no privateinformation,whereaswhen axis

Journal of Business
228
less than or equal to .5, the adverse selection problemis so severe that
the competitive market closes down. Of course, the monopolist specialist will not dominatein all environments.In particular,if there is no
private information, then everyone would prefer competing market
makers. This can be seen by noting that when there is no private
information, competitive market makers will set a zero spread,
whereas the monopolistwill set a positive spread.Thus, in the competitive case, there will be complete risk sharing,whereasit will be limited
in the monopolist-specialistcase.
The next proposition shows that after a reparameterizationof the
model, there is an cx> .5 such that, ex ante, the monopolistspecialistis
preferred when a < 'a, whereas competing market makers are preferred if a > 'a. This verifies the intuition that the superiorityof the
monopolist position derives solely from informationalproblems. If
those problems are not great then the monopolist-specialistsystem is
inferiorto a system with competing marketmakers.
5. Let Vm denote the ex ante expected utility of a
PROPOSITION
traderwhen there is a monopolist specialist, and let V, denote the ex

ante expected utility when there are competingmarketmakers. There
is a reparameterizationof the model with the parameters arW,
Irx, Irs,
and p replaced by independent parameters Yi,i = 1 - 3 and ax = p2Trx/
N, N = P2Ix + 'rr(TrriQ,x+ nrs).Considerchanges in axkeepingthe y's
constant. Then there is an a > .5 such that Vm> Vc when a ac and Vm
> V, when a > a. A sketch of the prooffor this propositionis provided
in the Appendix.
III. ConcludingRemarks
The focus of this articlehas been the inefficienciescreatedby allowing
tradingon private informationand the institutionalresponse to such
inefficiencies. Tradingon privateinformationcreates inefficienciesbecause it leads to less than optimalrisk sharing.This occurs because the
response of market makers to the existence of traders with private
informationis to reduce the liquidityof the market.This reductionin
the liquidityof the marketreduces the amountof trade and hence the
amount of risk sharing.
The institutionof the monopolistspecialistmay ease this inefficiency
somewhatby increasingthe liquidityof the market.Since the monopolist can averagehis profits, he need not make a profiton every trade. In
fact, it is shown in this example that optimallyhe will not. This implies
a more liquid market as long as there is extensive tradingon private
information.If there is not, then the result is that the monopolistspecialist does worse from a welfare perspective than do competingmarket makers.
If this analysis is to be believed as an explanationfor the existence of
the specialist system, there should be some evidence that tradingon

229
privateinformationis a significantproblem. Glosten and Harris(1988)

present evidence that part of the spread can be attributedto private
information.One can also interprettradingsuspensionsas a reactionto
the existence of private information.Finally, the Securities and Exchange Commission(SEC) has been successful in prosecutinga number of insider tradingcases.
If this analysis is correct, one of the sources of institutionaldesign
differences must be the extent of asymmetricinformation.In the case
of options and futures markets, one can make a reasonableargument
that asymmetricinformationis not a largeproblem.Tradingin financial
futures is probablynot substantiallymotivatedby speculationon private information.While there may be privateinformationabout future
commodity spot prices, the individualslikely to have that information
are the producersof the commodities, and it is probablytrue that their
insurancemotive is large enough to swamp any information-motivated
trade.
For options markets, the argumentis not so obvious, particularly
because it is often asserted that traderswith privateinformationmake
extensive use of the options market. However, given that there is a
specialistin the stock market,there may be no gain to havinga monopolist specialist in the option market. Market makers on the options
exchanges can typically hedge using the underlyingstock. Thus, if a
trader has informationthat a stock is undervalued,he may buy the
undervaluedcall from a marketmaker.If the option marketmakercan
buy the (undervalued)stock before word of the option transaction
reaches the floorof the stock exchange, then the risk that the tradewas
information motivated can be transferredfrom the option market
maker to the stock marketmaker.7
Of course, the specialist can also use options to hedge in the other
direction.The point of this discussion is to arguethat given a monopolist in one market, there is no gain and a probable loss to having a
monopolistin the other market.On the other hand, if both marketshad
competitive marketmakers then, accordingto the above model there
would be times in which both markets would shut down. In such an
instance there may be a gain to having a monopolistin one of the two
markets.
The largertask is to explain the coexistence of the over-the-counter
(OTC) market, with competing dealers, and the NYSE, with the specialist system. At firstglance it mightbe thoughtthat asymmetricinformationwould be more of a problemfor the smallerfirmstradedon the
7. Of course, following the analysis of this model, the price may move against the
option marketmakermakingthe hedgingtrade, and rationaloption marketmakerwill
take this into accountin his quotes. However, both a monopolistand competingmarket
makercan do this equally well, so there is no benefit to having a monopolistmarket
makerin the options market.

230
Journal of Business
OTC market. However, it may be that insider tradingis not,the most

serious source of informationalasymmetries.Given that marketmakers do not typically spend time analyzingthe security they deal with
and given the monitoringof insider tradingby the SEC, superioranalytical ability may be the largerday-to-day source of private information. In this case, it is not obvious that the smaller stocks would be
more prone to information-motivatedtrade. This is particularlytrue
given that the dispersion of ownership is probablyless with the OTC
stocks, and hence the gains to doing security analysis may be smaller.
No doubt, one reason that the specialist system began was historical
accident, and this article does not claim to supply the only reason for
its existence. Certainly,currentowners of seats on the exchangemight
object to a change in the system and hence any changecould be costly.
What the analysis in this article suggests is that changingis not obviously socially optimal even if the changes were not costly.
There are a numberof areas of research suggestedby the results and
the limitations of this analysis. It has been argued (see Manne 1966)
that there are benefits to allowing informedtrading,and the result of
proposition 1 completely ignores these. What the propositionsays is
that these benefits, if any, are bought at a cost-the liquidity of the
market suffers. This suggests that a general equilibriumanalysis that
considers both the lijquidityissues as well as the benefits of insider
tradingmightdo much to answerthe questionof what restraintsshould
be placed on informedtrading. A hypothesis is that tradingon inside
informationthat will be revealed in a short period of time hurts the
liquidity of the market without a corresponding increase in the
efficiency of firms' productive decisions. On the other hand, information due to superioranalysis may increase the productiveefficiency of
firmsby makingstock prices reflect more of the informationavailable
to the firm. The increase in efficiency may be worth the concomitant
decrease in the liquidity of the market.
This articletakes a fairly simple view of the role of the specialistas a
dealer and ignores the brokeragefunction. In fact, limit order submitters may be a nontrivialsource of competitionfor the specialist. The
extent to which limit order submittersand floor traderscompete with
the specialist and what effect that has on the results here would be of
interest. Along these same lines, it was arguedin the initialdiscussion
of the model that knowledge of the specialist's book conferredknowledge about tradinguncertainty,not knowledge of future prices of the
stock. This depends on the assumptionthat informedtraders do not
submitlimit ordersbut ratheruse marketorders. The order-placement
strategy of informed traders would be another interestingline of research.
The model in this article does not considerthe dynamicorder-placement strategiesof investors. The intuitionfor the results of this article
suggests that ignoringthis feature would not have a substantialeffect

231
on the generalresult, but it certainlydoes have an effect on the specific

results. The simultaneousconsiderationof the optimalorder strategy
as well as the optimal dynamic strategy of the specialist would be
interesting.
Witha more detailedanalysis of the strategiesopen to marketparticipants, it would be possible to consider the question of optimalmarket
design. While the analysis presented here suggests a feature that an
optimalinstitutionwould exhibit (averagingof profitsacross trades)it
is prematureto use this model to address the issue of optimality.
To sum up, this articlehas two majorresults. First, informedtrading
imposes a welfare cost in that it reduces the liquidityof the market.
Second, the existence of the specialist system may be in partexplained
by this cost. An unregulatedspecialist may providea more liquidmarket than competingmarketmakers,and hence some of the welfareloss
due to informedtradingmay be negated. In analyzingthese questions,
a model was developed that may hold some methodologicalinterest. In
particularit was shown that examinationof the small-tradespreadmay
not supplymuch informationaboutthe perceivedpresence of informed
traders.
Appendix
Proof of Lemma I
Substitutingfor Z from (9) into (3) yields
. {[P'(Q)Q + RN
P(Q)]
P(Q) =
7rx
7rz
7rx
7rs
7s
7rz
Define g(Q) by
g(Q) = P(Q) - m -
Qpw,wsID,
where
D = p27rX
rwwws(n,+ ws)
if D
7O O.
After substitutingfor
s =
112/(p2
nwgs),
we have
p27T,/Q = iwTs(w.( + 1TS)g'(Q)Ig(Q).
ForQ > 0,
g(Q) = KQ, -y = P21T./[1Tw,TS(1Tx + 1TS)]I
with K unrestricted.A similaranalysis holds for Q < 0. If D
expression above can be written
(7x + 7S)[g'(Q)Q
_ g(Q)]1Q2 + pIQ = 0;
0, then the
g(Q) = p(Q) - m.
This has the solution given in the statementof the lemma. Q.E.D.

Journalof Business
232
Proof of Proposition I
First, the uniquenessof the linearpricefunctionis shownwhenD > 0. Assume
that a pricing schedule is given as in lemma 1 with K > 0. Now consider a
continuous,differentiableprice schedulethat is equalto this price functionfor
Q C 1, and linearfor Q> 1. To ensure differentiabilitymake the slope of the
linearportionequal to the slope of the originalscheduleat Q = 1. Evaluation
of the expected profit will show that this second price schedule will yield
positive profitsfor trades largerthan 1. On the other hand, K cannotbe taken
to be negative, for then the second-orderconditionwill be violated. Thus, the
only undominatedpricing schedule when D > 0 is linear.
Suppose D < 0. Then the derivativesof the solutiongiven in lemma 1 are
P(Q) =
(p'n?1TnID) + KylQlj-1,
P"(Q)Q = Ky(y - 1)IQIl-1.

Note that, if D < 0, y < 1. The second-orderconditionis
IQI-1 >
{P[P2ITx + lTwlTS(lTx +
1TS)]}I[Ky(y+ 1)(-D)].
(A1)
Since D < 0, and -y< 1, the second-orderconditionis satisfiedfor - Q' < Q<
Q' where Q' satisfies (Al) with equality. Considerthe expected profitto the
marketmakerson trades of Q'. A traderwill trade Q' if Z is largerthan some
Z'. But then, E[X|Q'] > E[XIZ'] - P(Q'), and the marketmakers'profit is
negativeon tradesof Q contraryto the hypothesis.Thus, if D < 0, thereis no
schedule where the market maker breaks even. The analysis of D = 0 is
similar.Q.E.D.
Proof of Lemma 2
If the marketcloses down (D s 0), then Q, = 0 andA = 0 in the statementof
analysisof
the lemma.If D > 0, then the resultis obtainedby a straightforward
the expressionsin proposition1, assumingthatS and Ware independentandX
and W are independent.Q.E.D.
Proof of Proposition 2
First note that if there is no private information(i.e., ws = 0), competing
marketmakerswill set P(Q) = m for all Q. In that case, the optimaltrade is
- W and the ex ante expected utility is E[U(Wm)].Assume this expectation
exists. When there is privateinformation,the optimaltradeis Qand the price
scheduleis PC(-).The realizedwealthof a typicaltraderis Wm+ W(X - m) +
Q[X - P,(Q)], and the ex ante utility is E[U(Wm + W[X - m] + Q[XP'(Q)])]. This expectation may not exist. If it does not, the proof is complete.
Suppose it does. A sufficientconditionfor the ex ante utility without private
informationto exceed the ex ante utility with privateinformationis
E(U'(Wm){W(X- m) + Q[X- PC(Q)]})' O.
Given the independenceof X and W,this can be rewrittenE[U'(Wm)e(W)]' 0.
Note that
E[e(W)] = E[Q{X - P(Q)}] = E[Q{E[XIQ] - Pj(Q)}] = 0.
Since e(W) is presumedto be symmetricand concave and since E[e(W)] = 0,

thereis w' such that e(w) < 0 for lwl> w' and e(w) > 0 for Iwi< w'. Lettingfw()

233
be the density of W and assumingf,() is symmetricimplies

E[U'(Wm)e(W)] =
e(w)fw(w)[U'(wm)+ U'(-wm)] +
{
_00
e(w)f(w)[U'(wm) + U'(-wm)].
Since U"'() 2 0, U"(x)is increasingin x, and hence U'(wm) + U'(- wm) is
decreasing in w. Thus, for w c -w', U'(wm) + U'(-wm) 2 U'(w'm) +
U'(-w'm)
< w
and for -w'
0, U'(wm) + U'(-wm)
'
U'(w'm) +
U'(-w'm), and hence

e(w)fw(w)= 0.
E[U'(Wm)e(W)]' [U'(w'm) + U'(-w'm)]{

Q.E.D.
Proof of Proposition 3
The Hamiltonianassociated with (20) is-Q(z) and D(z) are state variables,
Q'(z) = u(z) is the controlf(z)[D(z) - e(z)Q(z)] + X1(z)u(z)+ X2(z)u(z)V[z,Q(z)].
The conditions the maximummust satisfy are
i. X1(z) + X2(z)V[z,Q(z)]
= 0;
ii. -XA(z) = -e(z)f(z) + X2(z)u(z)V2[z,Q(z)];

iii. -A2(z) = f(z).
The transversalityconditionsimply X2(Z)= 1 - F(z) for z positive and - F(z)

for z negative. Differentiatingcondition(i) after substitutingfor X2(z)andusing
(ii),
f(z)V[z,Q(z)] - [1 - F(z)]Vl[z,Q(z)] = e(z)f(z).
Substituting for V[z,Q(z)] yields Q(z) = (Tsuzlp) {az - [1 - F(z)]lf(z)} for z
positive. The procedurefor z negative is similar.This Q(z) is not monotonic.

However, if we add the constraintz[Q(z) - Q(0)] - 0, the resultingQ(z) is
monotonic. The symmetryof the problemimplies that Q(O)= 0. Q.E.D.
Sketch of the Proof of Proposition 5
Since all functions of z are symmetricaround0, the discussion will proceed
with z > 0. The quantitiestradedin the competitiveand monopolycases are of
the form
Q(z)
= (ursquIp)q(z),
where the competitive and monopolist q's, qc and qm, respectively, are given
by qc(z) = (2a - 1)z, qm(z) = oz - [1 - F(z)]If(z) for z > z*, 0 otherwise. The
certaintyequivalentis given by:

CEi = wm + iTsazwzI(iT + ws)
.5pw2/(IQ
'Ts)
IT2oMz/[p(w
ws)]
+
qi(t)dt,

Journal of Business
234
where i = c,m. Denote the coefficient on the integralby Yi/Pand the integral
by Gj(z,a). Let CEc and CEmdenote, respectively, the equilibriumcertainty
equivalentsfor the competitiveand monopolycases. Ourintentionis to evaluate.
E[exp(- pCEm) - exp(- pCEc)],
the gain in utilityfromusing the competitivesystem. Note that the expectation

is taken over both z and w. We know that for a = .5 this gain is negative, and
for a = 1 it is positive. By continuitythere is an a > .5 such that the difference
in expected utility is 0.
After some tedious calculationsit can be verifiedthat
E[exp(- pCEi)lz] = Kexp(- yoz _- Y2Z2- y1Gj(z,ot)) = KH(z,y,a),
where K dependson all the parameters,and the y's can be chosen independent
of a. The differencein expected utilities can thus be written
KE[H(z,y,a)exp{jy1[Gc(z,6) - GC(z,o)]} (exp{jy[Gc(z,a) - Gm(z,aY)]}- 1)].
Note that Gj(z,o) - Gm(z,o)= A(z,o) is increasingin a. Furthermore,there is

a t(a) such that for z < t(a), A(z,o) is positive while it is negativefor z > t(a).
Also note that Gc(z,a^)- G(z,ot) = (a^- a)z2. For a > ai, A(z,o) > A(z,&3).By
is positive [z < t(&)]and
integratingseparatelyover the regionsin which A(z,&3)
negative[z > t(&)], one can verify that the differencein utilitiesis at least as big
as
KE{H(z,,y,o)exp[yj(U
a)t( )2](exp[ylA(z,^)] - 1)} = 0.
A similaranalysis holds for a < a^.Q.E.D.

References
Altug, Sumru. 1985. The effect of insider tradingby a dominanttrader in a simple
securitiesmarketmodel. Workingpaper.Minneapolis:Universityof Minnesota,Center for Economic Research.
Diamond,Douglas. 1985.Optimalrelease of informationby firms.Journal of Finance 40
(September):1071-94.
Easely, David, and O'Hara,Maureen.1987.Price, tradesize and informationin securities markets.Journal of Financial Economics 19 (September):69-90.
Gammill,JamesF., Jr. 1986.Financialmarketdesignwhen tradershave privateinformation. New York: ColumbiaUniversity, GraduateSchool of Business, Centerfor the
Study of FuturesMarkets.
Glosten, LawrenceR., and Harris,LawrenceE. 1988.Estimatingthe componentsof the
bid/ask spread.Journal of Financial Economics 21 (May): 123-42.
Glosten, LawrenceR., and Milgrom,Paul R. 1985.Bid, ask and transactionprices in a
specialist marketwith heterogeneouslyinformedtraders.Journal of Financial Economics 14 (March):71-100.
Gould, John P., and Verrecchia, Robert E. 1985. The specialist as economic agent.
Journal of Political Economy 93 (February):66-83.
Hagerty, Kathleen. 1986. Equilibriumbid-ask spreadsin marketswith multipleassets.
Workingpaper. Evanston, Ill.: Northwestern University, J. L. Kellogg Graduate
School of Management.
Hirschleifer,Jack. 1971.The privateand social value of informationand the rewardto
inventive activity. American Economic Review 61 (September):561-74.
Ho, Thomas, and Stoll, Hans R. 1981. Optimaldealer pricingunder transactionsand
returnuncertainty.Journal of Financial Economics 9 (March): 47-73.

235
Ketterrer,Juan. 1987. Market games with inside information.Workingpaper. Minneapolis, Minn.: Universityof Minnesota,Economics Department.
Kyle, Albert. 1985.Continuousauctionsand insidertrading.Econometrica53 (November): 1315-35.
Leffler,George L., and Farwell, LoringC. 1963. The Stock Market.3d ed. New York:
RonaldPress.
Madhavan,AnanthN. 1987.Tradingmechanismsin securitiesmarkets.Workingpaper.
Ithaca, N.Y.: CornellUniversity.
Manne,H. 1966.Insider Tradingand the Stock Market.New York: Free Press.
Marshall,John M. 1974.Privateincentivesand publicinformation.AmericanEconomic
Review 64 (June):373-90.
Rothschild,M., and Stiglitz, J. 1978.Equilibriumin competitiveinsurancemarkets:An
essay on the economics of imperfectinformation.In P. Diamondand M. Rothschild
(eds.), Uncertaintyin Economics. New York: AcademicPress.
Spence, Michael. 1978. Productdifferentiationand performancein insurancemarkets.
Journalof Public Economics 10 (December):427-47.
Stiglitz,J. 1977.Monopoly,non-linearpricingand imperfectinformation:The insurance
market.Review of Economic Studies 44 (October)407-30.
Stoll, Hans R. 1984. Alternativeviews of market-making.Workingpaper. Nashville,
Tenn.: VanderbiltUniversity, Owen GraduateSchool of Management.


Insider Trading, Liquidity, and The Role of The Monopolist Specialist

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Insider Trading, Liquidity, and The Role of The Monopolist Specialist

Încărcat de

Drepturi de autor:

Formate disponibile

Insider Trading, Liquidity, and the Role of the Monopolist Specialist

Author(s): Lawrence R. Glosten

This content downloaded from 133.30.18.89 on Fri, 8 Nov 2013 04:07:39 AM